Professional Documents
Culture Documents
(Surgerubook - Net) Shackelford's Surgery of The Alimentary Tract 8th
(Surgerubook - Net) Shackelford's Surgery of The Alimentary Tract 8th
(Surgerubook - Net) Shackelford's Surgery of The Alimentary Tract 8th
SURGERY of the
ALIMENTARY
TRACT
SECTION
EDITORS
Volume Volume
1 2
Steven R. DeMeester, MD, FACS Jeffrey B. Matthews, MD, FACS
Division of Foregut and Minimally Invasive Surgery Dallas B. Phemister Professor and Chairman of Surgery
The Oregon Clinic The University of Chicago
Portland, Oregon Chicago, Illinois
Section I Esophagus and Hernia Section III Pancreas, Biliary Tract, Liver, and Spleen
Shackelford’s
SURGERY of the
ALIMENTARY
TRACT Volume
No part of this publication may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or any information storage and
retrieval system, without permission in writing from the publisher. Details on how to seek permission,
further information about the Publisher’s permissions policies and our arrangements with
organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be
found at our website: www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by the
Publisher (other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and
experience broaden our understanding, changes in research methods, professional practices, or
medical treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in
evaluating and using any information, methods, compounds, or experiments described herein. In
using such information or methods they should be mindful of their own safety and the safety of
others, including parties for whom they have a professional responsibility.
With respect to any drug or pharmaceutical products identified, readers are advised to check
the most current information provided (i) on procedures featured or (ii) by the manufacturer of
each product to be administered, to verify the recommended dose or formula, the method and
duration of administration, and contraindications. It is the responsibility of practitioners, relying
on their own experience and knowledge of their patients, to make diagnoses, to determine
dosages and the best treatment for each individual patient, and to take all appropriate safety
precautions.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors,
assume any liability for any injury and/or damage to persons or property as a matter of products
liability, negligence or otherwise, or from any use or operation of any methods, products,
instructions, or ideas contained in the material herein.
Previous editions copyrighted 2013, 2007, 2002, 1996, 1991, 1986, 1983, 1982, 1981, 1978, 1955.
Mayo Foundation retains copyright to their original artwork.
Printed in China
CHARLES J. YEO
STEVEN R. DEMEESTER
I would like to dedicate this book to all the past residents and fellows with whom I have
worked and to thank them for making education such a wonderful part of my life; I hope
this will help us to remember the times we’ve spent together taking care of complex patients
and will encourage you to continue to pass on your knowledge to those whom you mentor.
JAMES W. FLESHMAN
To Dr. William Silen and my late grandfather, Dr. Benjamin M. Banks, for their wisdom;
to the surgical residents and students, for their thirst for knowledge; and to my wife, Joan,
and our boys, Jonathan, David, and Adam, for their love and support.
JEFFREY B. MATTHEWS
To my wife, Nancy, and my children, William, Hunter, and Nora; and to all of my
mentors, colleagues, and patients who challenge and inspire me every day.
DAVID W. MCFADDEN
Contributors
Abbas E. Abbas, MD, MS, FACS Emanuele Asti, MD, FACS
Professor and Chief, Division of Thoracic Surgery, Assistant Professor, General and Emergency Surgery, IRCCS
Department of Thoracic Medicine and Surgery; Director, Policlinico San Donato, University of Milano, Milan, Italy
Thoracic and Foregut Surgery, Temple University School of
Medicine, Philadelphia, Pennsylvania Hugh G. Auchincloss, MD, MPH
Cardiothoracic Fellow, Massachusetts General Hospital,
David B. Adams, MD Boston, Massachusetts
Professor of Surgery, Medical University of South Carolina,
Charleston, South Carolina Benjamin Babic, MD
Department of Surgery, Agaplesion Markus Hospital,
Piyush Aggarwal, MBBS Frankfurt, Germany
Fellow, Division of Colorectal Surgery, Mayo Clinic, Phoenix,
Arizona Talia B. Baker, MD
Associate Professor of Surgery, Transplantation Institute,
Bestoun H. Ahmed, MD, FRCS, FACS, FASMBS The University of Chicago Medicine, Chicago, Illinois
Associate Professor of Surgery, University of Pittsburgh
School of Medicine, Pittsburgh, Pennsylvania Chad G. Ball, MD, MSC, FRCSC, FACS
Associate Professor of Surgery, University of Calgary, Foothills
Craig Albanese, MD, MBA Medical Center, Calgary, Alberta, Canada
Division of Pediatric Surgery, Department of Surgery,
Stanford University School of Medicine, Stanford, California Arianna Barbetta, MD
Research Fellow, General Surgery Department, Thoracic
Matthew R. Albert, MD, FACS, FASCRS Surgery Service, Memorial Sloan Kettering Cancer Center,
Program Director, Florida Hospital Colorectal Fellowship, New York, New York
Department of Colon and Rectal Surgery, Center for Colon
and Rectal Surgery, Florida Hospital, Orlando, Florida John M. Barlow, MD
Assistant Professor, Department of Radiology, Mayo Clinic
Abubaker Ali, MD College of Medicine, Rochester, Minnesota
Assistant Professor of Surgery, Wayne State University, Detroit,
Michigan Justin Barr, MD, PhD
Department of Surgery, Duke University Medical Center,
Evan Alicuben, MD Durham, North Carolina
General Surgery Resident, Keck School of Medicine of the
University of Southern California, Los Angeles, California Juan Camilo Barreto, MD
Assistant Professor of Surgery, Division of Surgical Oncology,
Marco E. Allaix, MD, PhD University of Arkansas for Medical Sciences, Little Rock,
Department of Surgical Sciences, University of Torino, Arkansas
Torino, Italy
Linda Barry, MD, FACS
Ashley Altman, MD Associate Professor of Surgery, University of Connecticut
Department of Radiology, The University of Chicago School of Medicine; Chief Operating Officer, Connecticut
Medicine, Chicago, Illinois Institute for Clinical and Translational Science, Farmington,
Connecticut
Hisami Ando, MD
President, Aichi Prefectural Colony; Emeritus Professor, Eliza W. Beal, MD
Department of Pediatric Surgery, Nagoya University Graduate Department of Surgery, The Ohio State University Wexner
School of Medicine, Nagoya-city, Aichi, Japan Medical Center, Columbus, Ohio
T
he time has come to release the eighth edition of the eighth edition has been carefully planned by me and the
classic textbook Shackelford’s Surgery of the Alimentary four expert section editors to represent the current state
Tract. This publication has served as an important of alimentary tract surgery as practiced throughout the
resource for surgeons, internists, gastroenterologists, world. This edition has been completed with an enormous
residents, medical students, and other medical profession- amount of assistance from my four colleagues, who have
als over the past 61 years. I hope that you will find the served as editors for the four major sections of the book.
eighth edition brimming with new information, beautifully These section editors have worked tirelessly, planning,
illustrated, up to date, and educationally fulfilling. organizing, and developing this massive textbook. They
have incorporated numerous changes in surgical practice,
operative techniques, and noninvasive therapies within the
BRIEF HISTORY text. Although each area does retain sections on anatomy
The first edition of Surgery of the Alimentary Tract was and physiology, the numerous advances in genomics,
written solely by Dr. Richard T. Shackelford, a Baltimore proteomics, laparoscopic techniques, and robotics are
surgeon, and was published in 1955. Following the success included. The eighth edition includes the contributions
of the first edition, the book’s publisher, W.B. Saunders of two new and two retained section editors from the
Company, urged Dr. Shackelford to produce a second seventh edition, providing both innovation and stability.
edition. Considerable time passed. A second edition was Section I, Esophagus and Hernia, is now edited by
released, as separate five-volume tomes, between 1978 and Dr. Steven R. DeMeester, a Professor of Surgery at the
1986, with Dr. Shackelford enlisting the assistance of Dr. University of Southern California, in Los Angeles. Dr.
George D. Zuidema, the Chairman of the Department of DeMeester is a nationally and internationally known expert
Surgery at Johns Hopkins, as co-editor. It was the second who brings his detailed knowledge of the esophagus and
edition that served as my “bible” for alimentary tract esophageal diseases to the textbook. Dr. DeMeester is the
diseases during my surgical residency and early faculty son of Dr. Tom DeMeester, a legendary individual who
appointment. has made numerous contributions to the field. Steve has
The third edition, edited by Dr. Zuidema, was published enlisted new authors for his chapters and has put together
as a five-volume set in 1991 and proved to be a major a spectacular section on esophageal diseases, including
tour de force. The field of alimentary tract surgery had pathology and ambulatory diagnostics, extensive sections
advanced, new research findings were included in the on gastroesophageal reflux disease, esophageal motility
edition, and emerging techniques were illustrated. For disorders, and esophageal neoplasia.
the third edition, Dr. Zuidema enlisted the help of a guest Dr. David W. McFadden remains as the editor for Section
editor for each of the five volumes. II, Stomach and Small Intestine. Dr. McFadden works at
The fourth edition, again headed by Dr. Zuidema, was the University of Connecticut in Farmington, Connecticut,
published in 1996 and remained encyclopedic in scope, serving as the Professor and Chairman of the Department
breadth, and depth of coverage. The textbook had become of Surgery and the Surgeon in Chief of their health system.
a classic reference source for surgeons, internists, gastroen- Dr. McFadden is an expert in alimentary tract diseases,
terologists, and other health care professionals involved in surgical research, and education. He served for many years
the care of patients with alimentary tract diseases. as the co-editor-in-chief of the Journal of Surgical Research,
The fifth edition was published in 2002. At that time, and he has served as the president of the Society for
Dr. Zuidema asked me to join him as a co-editor. The fifth Surgery of the Alimentary Tract. He has done a superb
edition remained a five-volume set, and it was filled with job of enlisting many new chapter authors so as to present
new operative techniques, advances in molecular biology, an updated section regarding the luminal structures of
and noninvasive therapies. It marked progress in the the upper gastrointestinal system. Dr. McFadden’s section
co-management of patients by open surgical, laparoscopic is an outstanding contribution to this area.
surgical, and endoscopic techniques. For Section III, Pancreas, Biliary Tract, Liver, and Spleen,
In 2007, the sixth edition of Shackelford’s Surgery of we have retained our section editor, Dr. Jeffrey B. Matthews,
the Alimentary Tract was published. The look of the sixth the Dallas B. Phemister Professor and Chairman of the
edition was changed. The book went from five volumes to Department of Surgery at the University of Chicago, in
two volumes with the deletion of outdated material, and Chicago, Illinois. Dr. Matthews’ surgical career has focused
it included a four-color production scheme, emphasized on diseases of the nonluminal structures of the alimentary
new procedures, and focused on advances in technology. tract, and he has done a superb job in enlisting new
In 2012, the seventh edition was published. contributors and reorganizing this section. Dr. Matthews’
credentials include serving as the editor-in-chief of the
Journal of Gastrointestinal Surgery, and he is also a recent
THE EIGHTH EDITION past president of the Society for Surgery of the Alimentary
The eighth edition maintains the exterior changes and Tract. This section serves as an outstanding contribution
look of the sixth and seventh editions. However, the to the field.
xxi
Finally, Section IV, Colon, Rectum, and Anus, has been
designed and reorganized by Dr. James W. Fleshman, the
Seeger Professor and Chairman of the Department of
Surgery at Baylor University Medical Center. Dr. Fleshman
is an internationally known figure in his field, and his
section has been totally reorganized. Included are various
new developments in the field, updates on pelvic floor
anatomy and physiology, new therapies for inflamma-
tory bowel disease, increased emphasis on laparoscopic
interventions, and new chapters dealing with the topics
not previously presented.
xxii PREFACE
T
he esophagus is a muscular tube connecting the swallowing can be described in three to six different
mouth with the stomach. The main function of the phases.2-6 Three main phases are often differentiated: the
esophagus is transport of fluids and food to ensure oral, pharyngeal, and esophageal phase.
regular nutrition of the body. At the proximal and distal In the oral phase, the tongue and the surrounding
end of the tube, special boundaries are necessary to fulfill structures of the pharynx, the soft and hard palate, and
complex functional tasks such as swallowing, belching, the closed glossopalatal area form a bolus (Fig. 1.1). This
and vomiting and allowing breathing and coughing, while process is started by will of the individual. The first
preventing substantial reflux of gastric contents into the step of bolus formation is performed by the soft palate
esophagus. The proximal part of the esophagus and its and the posterior aspect of the tongue. The second
upper esophageal sphincter (UES) comprise mainly striated component of the oral phase is the upward movement
musculature.1 In the esophageal corpus, a few centimeters of the soft palate toward the hard palate, closing the
below the UES, the structure switches through a zone of nasopharynx, and the upward swing of the hyoid bone
mixed striated-smooth muscle to the distal esophagus more ventrally.
including the lower esophageal sphincter (LES) with When the tongue creates more pressure on the bolus
complete smooth muscle structures. The complex function toward the soft and hard palates, the bolus is propelled
of the two sphincters is regulated and influenced by nerval through the glossopalatal opening and enters the pha-
innervations, pressure systems, hormonal and chemical ryngeal area, thus representing the transition to the
influences, and external and possible psychological factors. pharyngeal phase (Fig. 1.2). The sequences of the pha-
ryngeal phase happen involuntarily. In health, the most
important part of the process is the well-coordinated
UPPER ESOPHAGEAL SPHINCTER IN HEALTH closure of the airway during the passage of the bolus. The
nasopharyngeal area is closed by the soft palate and the
SWALLOWING PROCESS posterior pharyngeal wall. The muscles of the floor of
The UES is involved in the swallowing process. This process the mouth pull the hyoid bone, and subsequently the
is complex and requires a number of important features larynx and cricoid, anteriorly and upward, thus allowing
to temporarily change the pathway of breathing air between the epiglottis to flip downward and close off the airways.
the mouth, nose, pharynx, and the trachea into the channel The constrictor pharyngeal muscle contracts, and the
for passing liquid and food farther into the esophagus. UES is opened by approximately 1 second of relaxation.
The main anatomic structures of the upper boundary The bolus leaves the pharyngeal area, moves through the
of the esophagus are the cricopharyngeal muscle connected hypopharynx, and into the esophagus. The esophageal
to the cricoid cartilage. Because there is no circular sym- phase begins as the bolus enters the proximal esophagus
metric muscle structure, but instead a ventral cartilage through the opened cricopharyngeal sphincter (Fig. 1.3).
with lateral and dorsal muscular parts, the resulting The peristaltic contractions take over propulsion of the
pressure profile of the UES is not symmetric. It is bolus and assist gravity in transporting the bolus through
important to note that the anatomic relations functionally the esophageal corpus. Finally, all structures return to
change during swallowing. The UES moves upward their resting positions, and the airway is opened again as
approximately 1 cm because of the laryngeal upward the UES closes.
swing. The innervation of the UES is mainly achieved by
the vagus nerve and, to a lesser extent, by cranial nerves ASSESSMENT OF THE UPPER
IX and XII.1 ESOPHAGEAL SPHINCTER
Swallowing is a complex process and the UES is involved The UES comprises cricoid cartilage at the ventral border
in the pharyngeal phase, which occurs later in the swal- and cricopharyngeal muscle toward its lateral and posterior
lowing process. Depending on the respective author, borders.1 This structure creates the asymmetric pressure
2
Esophageal Sphincters in Health and Disease CHAPTER 1 2.e1
ABSTRACT
Introduction: The esophagus is a muscular tube connecting
the mouth with the stomach. The main function of the
esophagus is the transport of fluids and food to ensure
regular nutrition of the body. At the proximal and distal
end of the tube, special boundaries are necessary to fulfill
complex functional tasks such as swallowing, belching,
and vomiting and allowing breathing and coughing, while
preventing substantial reflux of gastric contents into the
esophagus.
Structures and Function in Health and Disease: This
chapter focuses on the upper esophageal sphincter (UES)
and the lower esophageal sphincter (LES). The complex
functions of the two sphincters are regulated and influ-
enced by nerval innervations, pressure systems, hormonal
and chemical influences, and external and possible psy-
chological factors. The functional assessment is usually
performed by radiographic studies and manometric
techniques. There are many disorders that can cause
cervical or oropharyngeal dysphagia. This includes myo-
genic, neurogenic, iatrogenic, mechanical, and psychogenic
causes, as well as idiopathic dysfunctions of the UES.
The LES structure is incorporated in the distal esopha-
gus. The major function is the proximal closure of the
gastric reservoir to prevent reflux of substantial amounts
of gastric contents back into the esophagus. This is
important because gastric juice can be toxic for the
esophageal mucosa. The major functional finding at the
distal esophagus is a high-pressure zone. The most frequent
failure of the LES is a weakening causing increased gas-
troesophageal reflux.
KEYWORDS
Esophagus, upper esophageal sphincter, lower esophageal
sphincter, dysphagia, myotomy, achalasia, gastroesophageal
reflux disease.
Esophageal Sphincters in Health and Disease CHAPTER 1 3
FIGURE 1.1 The swallowing process: The tongue has formed a FIGURE 1.3 The swallowing process: The upper esophageal
bolus between the ventral part of the tongue and its posterior part sphincter is fully relaxed for a short time in which the bolus
against the soft palate. passes into the proximal esophagus, where it is pushed farther by
gravity and esophageal peristalsis. The larynx and epiglottis then
resume their resting positions.
FIGURE 1.9 High-resolution manometry of the esophagus: demonstration of the esophageal body and lower esophageal sphincter in a
healthy individual, showing swallowing and relaxation of the lower esophageal sphincter.
years, which could be an argument for earlier surgical result of simultaneous and/or spastic contractions and/
therapy.26 or retention esophagitis, caused by long-standing remain-
ders of food in the esophageal body because of obstruction.
LOWER ESOPHAGEAL SPHINCTER FUNCTION IN Interestingly, the resting pressure of the nonrelaxing LES
can vary in achalasia from physiologic values up to massive
PATIENTS WITH ACHALASIA AND OTHER hypertension. An unfavorable relationship between persist-
ESOPHAGEAL MOTILITY DISORDERS ing pressure in the LES and the absence of esophageal
Dysphagia related to solids and liquids is the most common peristalsis is important when considering obstruction. As
complaint in patients with esophageal motility disorders. a result, only gravity and/or a rise in intraluminal hydro-
Some patients also experience chest pain. Although most static pressure because of the food and fluid column can
of the esophageal motility disorders are rather nonspecific overcome the persistent pressure in the LES.
changes of peristaltic contractions, achalasia is the best- The diagnostic work-up for patients with achalasia and
defined and probably best-described esophageal motility all other esophageal motility disorders should consist of
disorder. a combination of investigations to describe the alterations
One major sign of achalasia is a failure of the LES to at the LES.21,27,28 Endoscopy should be done to rule out
relax during swallowing. This leads to obstruction of the any malignant disease, scarring, or a peptic stricture. An
bolus transport. In addition to relaxation failure, the upper gastrointestinal (GI) radiographic study performed
esophageal body loses its ability to develop a normal as a timed barium swallow can demonstrate and quantify
peristalsis leading to a total failure of transport.22,27,28 the degree of esophageal outflow obstruction at the LES.
Retrosternal pain or burning can occur in achalasia as a HRM will confirm the diagnosis and can be used to
Esophageal Sphincters in Health and Disease CHAPTER 1 9
FIGURE 1.11 High-resolution manometry in a patient with gastroesophageal reflux disease and hiatal hernia. The weak lower esophageal
sphincter can be noted as well as the separation of the remaining pressure level of the lower esophageal sphincter and the
diaphragmatic structures.
10 SECTION I Esophagus and Hernia
contents in the stomach once they reach it. This simple 15. Mittal RK, Holloway RH, Penagini R, Blackshaw A, Dent J. Transient
requirement involves complex interactions involving lower esophageal sphincter relaxations. Gastroenterology. 1995;109:
601-610.
pharyngeal muscles, the diaphragm, the UES and LES, 16. Stein HJ, Barlow AP, DeMeester TR, Hinder RA. Complications of
and coordinated peristalsis in the esophageal body. These gastroesophageal reflux disease. Role of the lower esophageal
complex physiologic mechanisms allow for a number of sphincter, esophageal acid and acid/alkaline exposure, and duo-
critical points where malfunction can create symptoms. denogastric reflux. Ann Surg. 1992;216(1):35-43.
17. Fein M, Ritter M, DeMeester TR, et al. Role of lower esophageal
sphincter and hiatal hernia in the pathogenesis of GERD. J Gastrointest
Surg. 1999;3(4):405-410.
REFERENCES 18. Lord RV, DeMeester SR, Peters JH, et al. Hiatal hernia, lower
esophageal sphincter incompetence, and effectiveness of Nissen
1. Liebermann-Meffert D, Duranceau A. Anatomy and embryology of fundoplication in the spectrum of gastroesophageal reflux disease.
the esophagus. In: Orringer MB, Zuidema GD, eds. Shackelford’s J Gastrointest Surg. 2009;13(4):602-610.
Surgery of the Alimentary Tract, Vol 1, The Esophagus. 4th ed. Philadephia: 19. Mittal RK, Holloway R, Dent J. Effect of atropine on the frequency
Saunders; 1996:3-38. of reflux and transient lower esophageal sphincter relaxation in
2. Dodds WJ, Hogan WJ, Lyndon SB. Quantification of pharyngeal normal subjects. Gastroenterology. 1995;109(5):1547-1554.
motor function in human subjects. J Appl Physiol. 1975;39:960. 20. Van Herwaarden MA, Samson M, Smout AJP. Excess gastroesopha-
3. Donner MW, Bosma JF, Robertson DL. Neuromuscular disorders geal reflux in patients with hiatal hernia is caused by mechanisms
of the pharynx. Gastrointest Radiol. 1985;10:196. other than transient LES relaxations. Gastroenterology. 2000;119:
4. Hannig C, Wuttge-Hannig A. Status of high-frequency roentgen 1439.
cinematography in the diagnosis of the pharynx and esophagus. 21. Kahrilas PJ, Sifrim D. High-resolution manometry and impedance-pH/
Rontgenpraxis. 1987;40(10):358-377. manometry: valuable tools in clinical and investigational esophagology.
5. Kahrilas PJ, Dodds WJ, Dent J, Logemann JA, Shaker R. Upper Gastroenterology. 2008;135(3):756-769.
esophageal sphincter function during deglutition. Gastroenterology. 22. Pandolfino JE, Kwiatek MA, Nealis T, Bulsiewicz W, Post J, Kahrilas
1988;95(1):52-62. PJ. Achalasia: a new clinically relevant classification by high-resolution
6. Duranceau A. Disorders of the pharyngoesophageal junction. In: manometry. Gastroenterology. 2008;135(5):1526-1533.
Yeo CJ, ed. Shackelford’s Surgery of the Alimentary Tract. Vol. 1. 6th ed. 23. Kahrilas PJ. Esophageal motor disorders in terms of high-resolution
Philadelphia: Saunders; 2007:374-390. esophageal pressure topography: what has changed? Am J Gastroenterol.
7. Castell JA, Dalton CB, Castell DO. Pharyngeal and upper esophageal 2010;105(5):981-987.
sphincter manometry in humans. Am J Physiol. 1990;21:G73. 24. Stefanidis D, Hope WW, Kohn GP, et al. Guidelines for surgical
8. Cook IJ, Gabb M, Panagopoulos V, et al. Pharyngeal (Zenker’s) treatment of GERD. Surg Endosc. 2010;24(11):2647-2669.
diverticulum is a disorder of upper esophageal sphincter opening. 25. Fuchs KH, Babic B, Breithaupt W, et al. EAES recommendations
Gastroenterology. 1992;103(4):1229-1235. for the management of gastroesophageal reflux disease. Surg Endosc.
9. Goyal RK, Martin SB, Shapiro J, Spechler SJ. The role of cricopha- 2014;28(6):1753-1773.
ryngeus muscle in pharyngoesophageal disorders. Dysphagia. 26. Kuster E, Ros E, Toledo-Pimentel V, et al. Predictive factors of the
1993;8(3):252-258. long term outcome in gastro-oesophageal reflux disease: six year
10. Ghosh SK, Pandolfino JE, Zhang Q, Jarosz A, Kahrilas PJ. Deglutitive follow up of 107 patients. Gut. 1994;35(1):8-14.
upper esophageal sphincter relaxation: a study of 75 volunteer 27. Zaninotto G, Costantini M, Rizetto C, et al. Four hundred laparoscopic
subjects using solid-state high-resolution manometry. Am J Physiol myotomies for esopahegal achalasia—a single center experience.
Gastrointest Liver Physiol. 2006;291(3):G525-G531. Ann Surg. 2008;248:986-993.
11. DeMeester TR. Definition, detection and pathophysiology of gas- 28. Campos GM, Vittinghoff E, Rabel C, et al. Endoscopic and surgical
troesophageal reflux disease. In: DeMeester TR, Matthews HR, eds. treatments of achalasia: a system review and meta-analysis. Ann Surg.
International Trends in General Thoracic Surgery, Vol 3, Benign Esophageal 2009;249:45-57.
Disease. St. Louis: Mosby; 1987:99-127. 29. Kahrilas PJ, Bredenoord AJ, Fox M, et al. The Chicago Classification
12. Zaninotto G, DeMeester TR, Schwizer W, Johansson KE, Cheng SC. of esophageal motility disorders, v3.0. Neurogastroenterol Motil.
The lower esophageal sphincter in health and disease. Am J Surg. 2015;27:160-174.
1988;155(1):104-111. 30. Gockel I, Lord RV, Bremner CG, Crookes PF, Hamrah P, DeMeester
13. Fuchs KH, Freys SM, Heimbucher J, Fein M, Thiede A. Pathophysi- TR. The hypertensive lower esophageal sphincter: a motility disorder
ologic spectrum in patients with gastroesophageal reflux disease in with manometric features of outflow obstruction. J Gastrointest Surg.
a surgical GI function laboratory. Dis Esophagus. 1995;8:211-217. 2003;7(5):692-700.
14. Dent J, Holloway RH, Toouli J, Dodds WJ. Mechanisms of lower 31. Tamhankar AP, Almogy G, Arain MA, et al. Surgical management
oesophageal sphincter incompetence in patients with symptomatic of hypertensive lower esophageal sphincter with dysphagia or chest
gastro-oesophageal reflux. Gut. 1988;29:1020-1028. pain. J Gastrointest Surg. 2003;7(8):990-996.
CHAPTER
Esophageal Body in Health and Disease
Marco E. Allaix
| Marco G. Patti
2
H
igh-resolution manometry (HRM) is a well- on the HRM device used: the upper limit value in normal
established diagnostic tool that evaluates esopha- subjects varies between the transducer used, ranging from
geal motility. It dynamically measures intraluminal 15 mm Hg in the Sierra design transducers to 28 mm Hg
pressure changes in the esophagus by using closely spaced in the Unisensor design transducers in the supine position.5
pressure sensors. Data are acquired, displayed, and inter- The EGJ can be classified into three subtypes based on
preted by esophageal pressure topography plots.1 the axial relationship between the LES and CD8:
The Chicago Classification aims to define esophageal • Type I: LES and CD are completely overlapped.
motility disorders according to HRM findings. This clas- • Type II: LES and CD are separated, with the separation
sification was first proposed in 20082 and then updated between the pressure peaks being 2 cm or less.
in 20113 and in 2014.4,5 The latest version, the Chicago • Type III: the separation between LES and CD is greater
Classification, version 3.0,4,5 evaluates new parameters than 2 cm.
when compared with the previous versions, including • Type IIIa: the pressure inversion point remains at
esophagogastric junction (EGJ) morphology and contractil- the level of CD.
ity at rest, “fragmented” contractions, and ineffective • Type IIIb: the pressure inversion point is localized at
esophageal motility (IEM). This classification categorizes the level of LES.
(1) disorders of the EGJ outflow, such as achalasia and When food passes through the UES, a contraction is
EGJ outflow obstruction, (2) major peristalsis disorders initiated in the upper esophagus, which progresses distally
such as absent contractility, distal esophageal spasm and toward the stomach (Fig. 2.2). The wave initiated by
hyper contractile esophagus, and (3) minor disorders swallowing is referred as primary peristalsis. It travels at
characterized by impaired bolus transit. a speed of 3 to 4 cm/s and reaches amplitudes of 60 to
The aim of this chapter is to review the esophageal 140 mm Hg in the distal esophagus. Local stimulation by
function in the healthy esophagus and in the most frequent distention at any point in the body of the esophagus will
esophageal motility disorders. elicit a peristaltic wave from the point of stimulus. This is
called secondary peristalsis and aids esophageal emptying
when the primary wave has failed to clear the lumen of
ESOPHAGEAL BODY IN HEALTH ingested food, or when gastric contents reflux from the
The evaluation of esophageal motility by HRM is based stomach. Tertiary waves are considered abnormal, but
on the assessment of ten 5-mL water swallows performed they are frequently seen in elderly people who have no
in supine position (Fig. 2.1). During each swallow the symptoms of esophageal disease.
following features are evaluated: Postdeglutitive esophageal contraction is evaluated by
• EGJ relaxation using the following metrics:
• esophageal contractile activity • CDP represents the inflection point in the contractile
• esophageal pressurization front propagation velocity in the distal esophagus.9–11
The pressure topographic measurements used are • DL represents the interval between UES relaxation
• integrated relaxation pressure (IRP) and the CDP. It is considered an important metric
• distal contractile integral (DCI) indicating the integrity of the inhibitory pathway in the
• contractile deceleration point (CDP) distal esophagus. A value less than 4.5 seconds defines
• distal latency (DL) a premature contraction, indicative of spasm.12
• DCI describes the vigor of the distal esophageal contrac-
tion.13 It is measured as the “volume” of the esophageal
EVALUATION OF THE ESOPHAGOGASTRIC contraction spanning from the transition zone to the
JUNCTION MORPHOLOGY AND DEGLUTITIVE EGJ. The DCI is the product of the integral of the
ESOPHAGEAL CONTRACTION amplitude exceeding 20 mm Hg, the duration, and
During swallowing, the pressure detected at the level of the the length of the contractile segment between the
EGJ is defined by lower esophageal sphincter (LES) pres- transition zone and the EGJ. Cutoff values defining
sure, contraction of crural diaphragm (CD), and intrabolus different diagnostic categories depend on the type
pressure as the swallowed bolus passes through the EGJ.6 of HRM hardware and software used. DCI in normal
IRP, defined as the mean pressure for the 4 seconds of subjects ranges between 450 and 8000 mm Hg-s-cm.
maximal deglutitive relaxation in the 10-second window Hypercontractility is defined by a DCI greater than
starting with deglutitive upper esophageal sphincter (UES) 8000 mm Hg-s-cm. A DCI ranging between 100 and
relaxation, is the best metric to differentiate between 450 mm Hg-s-cm defines weak peristalsis, whereas a DCI
normal and impaired EGJ relaxation.7 IRP is influenced lower than 100 mm Hg-s-cm identifies failed peristalsis.
not only by LES relaxation but also by CD contraction and Both failed and weak contractions are considered
intrabolus pressure. In addition, normal values depend ineffective.4,5,14–16
11
Esophageal Body in Health and Disease CHAPTER 2 11.e1
ABSTRACT
Esophageal high-resolution manometry (HRM) is currently
the gold standard for the evaluation of esophageal motility.
The Chicago Classification categorizes the esophageal
motility disorders based on the HRM findings. Achalasia
and esophagogastric junction outflow obstruction are
characterized by a median integrated relaxation pressure
that is higher than in healthy subjects. Major motility
disorders are aperistalsis, distal esophageal spasm, and
hypercontractile esophagus. Minor motility disorders
include ineffective esophageal motility and fragmented
peristalsis.
KEYWORDS
High-resolution manometry, Esophageal motility, Peristalsis,
Upper esophageal sphincter, Esophageal body, Lower
esophageal sphincter, Achalasia, Distal esophageal spasm,
Ineffective esophageal motility, Hypercontractility
12 SECTION I Esophagus and Hernia
mm Hg mm Hg
150.0 150.0
140 140
120 120
100 100
80 80
60 60
40 40
30 30
20 20
10 10
0 0
–10.0 –12.0
Wet swallow
Swallow9
mm Hg 11:10.7
150.0
140
120
Pharynx
100
UES
80
60
Esophageal
body 40
30
20
10
LES
0
–10.0
sphincter.
FIGURE 2.4 Achalasia type II.
Contraction integrity, contraction pattern, and intrabolus
pressure pattern characterize each swallow.6 Contraction
integrity is defined based on the integrity of the 20-mm Hg are subdivided based on the patterns of contractions and
isobaric contour. Small (2 to 5 cm in length) interruptions, pressurization in the esophageal body into
or breaks, in the 20-mm Hg isobaric contour between • achalasia subtypes
the UES and EGJ are considered normal, whereas large • EGJ outflow obstruction.4,5
(>5 cm in length) breaks define weak contractions. The Although EGJ outflow obstruction is defined by the
associated intrabolus pressure pattern is assessed using presence of high median IRP in association with normal or
the 30-mm Hg isobaric contour. Intrabolus pressure is weak peristalsis, the presence of impaired EGJ relaxation
qualified as panesophageal pressurization if it spans from in the absence of peristalsis defines achalasia. Achalasia
UES to EGJ and as compartmentalized pressurization if is further subdivided in three subtypes4,5,17:
it is restricted to the segment between the deglutitive • Type I achalasia (classic): It is characterized by 100%
contractile front and the EGJ.4,5 failed contractions (DCI <100 mm Hg-s-cm) and no
esophageal pressurization (Fig. 2.3);
• Type II achalasia (with esophageal compression): It is defined
ESOPHAGEAL BODY IN DISEASE as 100% failed contraction and panesophageal pres-
surization for at least 20% of swallows (Fig. 2.4);
ACHALASIA AND ESOPHAGOGASTRIC JUNCTION • Type III achalasia (spastic): It is defined as the presence
OUTFLOW OBSTRUCTION of preserved fragments of distal peristalsis or premature
EGJ outflow obstruction is defined by a median IRP greater contractions (DCI >450 mm Hg-s-cm) for at least 20%
than 15 mm Hg.3 Disorders of EGJ outflow obstruction of the swallows.
Esophageal Body in Health and Disease CHAPTER 2 13
patients with heartburn (19%). Distal esophageal spasm manometry: a quantitative analysis of 400 patients and 75 controls.
and nutcracker esophagus were rarely seen. Incidence of Am J Physiol. 2007;293:G878-G885.
8. Pandolfino JE, Kim H, Ghosh SK, Clarke JO, Zhang Q, Kahrilas PJ.
hypertensive or hypotensive LES was similar in the two High-resolution manometry of the EGJ: an analysis of crural dia-
groups. Total esophageal acid clearance time was longer phragm function in GERD. Am J Gastroenterol. 2007;102:1056-1063.
in patients with GERD-associated respiratory symptoms 9. Pandolfino JE, Leslie E, Luger D, Mitchell B, Kwiatek MA, Kahrilas
than in patients with heartburn. PJ. The contractile deceleration point: an important physiologic
landmark on oesophageal pressure topography. Neurogastroenterol
Diener et al.23 evaluated 1006 consecutive patients Motil. 2010;22:395-400.
with GERD, who were divided into three groups based 10. Pandolfino JE, Roman S, Carlson D, et al. Distal esophageal spasm
on the character of esophageal peristalsis as shown by in high-resolution esophageal pressure topography: defining clinical
esophageal manometry: (1) normal peristalsis; (2) IEM; phenotypes. Gastroenterology. 2011;141:469-475.
(3) nonspecific esophageal motility disorder (NSEMD; 11. Lin Z, Pandolfino JE, Xiao Y, et al. Localizing the contractile decelera-
tion point (CDP) in patients with abnormal esophageal pressure
motor dysfunction intermediate between the other two topography. Neurogastroenterol Motil. 2012;24:972-975.
groups). Peristalsis was normal in 563 patients (56%), 12. Roman S, Lin Z, Pandolfino JE, Kahrilas PJ. Distal contraction
IEM was detected in 216 patients (21%), and NSEMD was latency: a measure of propagation velocity optimized for esophageal
observed in 227 patients (23%). Patients with abnormal pressure topography studies. Am J Gastroenterol. 2011;106:443-451.
13. Ghosh SK, Pandolfino JE, Zhang Q, Jarosz A, Shah N, Kahrilas PJ.
peristalsis had worse reflux and slower esophageal acid Quantifying esophageal peristalsis with high-resolution manometry:
clearance. Heartburn, respiratory symptoms, and mucosal a study of 75 asymptomatic volunteers. Am J Physiol. 2006;290(5):G988-
injury were all more severe in patients with IEM. G997.
14. Roman S, Pandolfino JE, Chen J, Boris L, Luger D, Kahrilas PJ.
Fragmented Peristalsis Phenotypes and clinical context of hypercontractility in high resolu-
tion pressure topography (EPT). Am J Gastroenterol. 2012;107:37-
This motility disorder is defined as the presence of 45.
fragmented contractions in more than 50% of swallows 15. Xiao Y, Kahrilas PJ, Kwasny MJ, et al. High-resolution manometry
without meeting the IEM criteria.4,5 correlates of ineffective esophageal motility. Am J Gastroenterol.
2012;107(11):1647-1654.
16. Xiao Y, Kahrilas PJ, Nicodème F, Lin Z, Roman S, Pandolfino JE.
REFERENCES Lack of correlation between HRM metrics and symptoms during
the manometric protocol. Am J Gastroenterol. 2014;109:521-526.
1. Soudagar AS, Sayuk GS, Gyawali CP. Learners favour high resolution 17. Pandolfino JE, Kwiatek MA, Nealis T, Bulsiewicz W, Post J, Kahrilas
oesophageal manometry with better diagnostic accuracy over con- PJ. Achalasia: a new clinically relevant classification by high-resolution
ventional line tracings. Gut. 2012;61(6):798-803. manometry. Gastroenterology. 2008;135:1526-1533.
2. Pandolfino JE, Fox MR, Bredenoord AJ, Kahrilas PJ. High-resolution 18. Pratap N, Kalapala R, Darisetty S, et al. Achalasia cardia subtyping
manometry in clinical practice: utilizing pressure topography to by high-resolution manometry predicts the therapeutic outcome of
classify oesophageal motility abnormalities. Neurogastroenterol Motil. pneumatic balloon dilatation. J Neurogastroenterol Motil. 2011;17:48-53.
2009;21:796-806. 19. Salvador R, Costantini M, Zaninotto G, et al. The preoperative
3. Bredenoord AJ, Fox M, Kahrilas PJ, et al. Chicago classification manometric pattern predicts the outcome of surgical treatment for
criteria of esophageal motility disorders defined in high resolution esophageal achalasia. J Gastrointest Surg. 2010;14:1635-1645.
esophageal pressure topography (EPT). Neurogastroenterol Motil. 20. Roman S, Kahrilas PJ, Mion F, et al. Partial recovery of peristalsis
2012;24(suppl 1):57-65. after myotomy for achalasia; more the rule than the exception.
4. Roman S, Gyawali CP, Xiao Y, Pandolfino JE, Kahrilas PJ. The JAMA Surg. 2013;148:157-164.
Chicago classification of motility disorders: an update. Gastrointest 21. Blonski W, Vela M, Safder A, Hila A, Castell DO. Revised criterion
Endosc Clin N Am. 2014;24:545-561. for diagnosis of ineffective esophageal motility is associated with
5. Kahrilas PJ, Bredeboord AJ, Fox M, et al. The Chicago Classification more frequent dysphagia and greater bolus transit abnormalities.
of esophageal motility disorders, v3.0. Neurogastroneterol Motil. 2015; Am J Gastroenterol. 2008;103:699-704.
27:160-174. 22. Fouad YM, Katz PO, Hatlebakk JG, Castell DO. Ineffective esophageal
6. Kahrilas PJ, Roman S, Pandolfino JE. The Chicago Classification of motility: the most common motility abnormality in patients with
esophageal motility disorders. In: Fisichella PM, Soper NJ, Pellegrini GERD-associated respiratory symptoms. Am J Gastroenterol. 1999;94:
CA, Patti MG, eds. Surgical Management of Benign Esophageal Disorders. 1464-1467.
The “Chicago Approach”. London: Springer-Verlag; 2014:25-38. 23. Diener U, Patti MG, Molena D, Fisichella PM, Way LW. Esophageal
7. Ghosh SK, Pandolfino JE, Rice J, Clarke JO, Kwiatek M, Kahrilas dysmotility and gastroesophageal reflux disease. J Gastrointest Surg.
PJ. Impaired deglutitive EGJ relaxation in clinical esophageal 2001;5:260-265.
CHAPTER
Esophageal Mucosa in Health and Disease
Parakrama Chandrasoma
| Yanling Ma
| Evan E. Yung
3
P
athology has no clinical value at the present time in From the different perspective of endoscopy, GERD
the diagnosis and management of gastroesophageal progresses from no visible endoscopic change to erosive
reflux disease (GERD) before the occurrence of esophagitis of increasing severity (Los Angeles [LA] grade
visible columnar lined esophagus (vCLE). Its only value A to D), to vCLE, Barrett esophagus (defined as vCLE with
is in the diagnosis of intestinal metaplasia, increasing intestinal metaplasia in the United States), and through
dysplasia and adenocarcinoma in the patient with Barrett increasing dysplasia, to adenocarcinoma.
esophagus. Biopsy is currently not recommended by most societies
We will only consider GERD in this chapter. We will in patients who do not have an endoscopic abnormality at
explore the pathophysiology of GERD through its entire the GEJ.2 Biopsy of the endoscopically normal squamous
progression from the normal state to severe GERD. This epithelium may show histologic changes of reflux, but
will lead to the proposal of a new pathologic test for these are not sufficiently sensitive or specific to have
lower esophageal sphincter (LES) damage that is based practical value. Biopsy of the “normal” squamocolumnar
on mucosal changes defined by histology. The new ability junction (SCJ) is not recommended, although it is known
to measure LES damage has the potential to open the that a small but significant number of patients will have
door to a new method of diagnosis and management of intestinal metaplasia if biopsies are taken, particularly if
GERD that has the potential to eradicate GERD-induced the SCJ is slightly irregular.3
esophageal adenocarcinoma. Endoscopy in the patient who has failed PPI therapy
The evidence base in support of the new test is solid, changes management only in the patient with Barrett
albeit small. Its acceptance requires the removal of two esophagus, who enters an endoscopic surveillance program
long-held and powerful dogmas that presently preclude aimed at detecting early neoplastic changes (see Fig. 3.1).
acceptance of the new method. One is a histology dogma In patients without Barrett esophagus, endoscopy provides
and the other an endoscopic dogma. The histologic dogma little if any useful information that impacts symptom
that must be discarded is that cardiac epithelium normally control with PPIs. Often symptoms are diminished in
lines the proximal stomach and is present at the normal patients with Barrett esophagus, and the efficacy of medical
gastroesophageal junction (GEJ). The endoscopic dogma therapy to prevent Barrett’s progression remains unproven.
that must be discarded is that the GEJ is accurately defined Progression to dysplasia and adenocarcinoma in all patients
by the proximal limit of the rugal folds and/or the end cannot be effectively prevented.4
of the tubular esophagus. The evidence shows clearly that Symptoms of GERD and endoscopic findings are often
these are both false even as they continue to be accepted. not concordant. A person without symptoms of GERD can
What is being proposed is revolutionary. have long segment Barrett esophagus or present with an
advanced GERD-induced adenocarcinoma. Conversely,
a patient with symptoms of GERD can be endoscopically
PRESENT STATUS OF GASTROESOPHAGEAL normal (nonerosive reflux disease [NERD]). Treatment
REFLUX DISEASE of GERD with PPIs can heal erosive esophagitis without
completely resolving GERD symptoms.2 Patients with
GERD is regarded as a chronic progressive disease. When NERD are more resistant to symptom control with PPIs
defined by the presence of symptoms that reach a point than those with erosive esophagitis.2
where they are considered troublesome,1 20% to 40% of GERD is generally diagnosed when typical reflux symp-
the population has GERD. Approximately 70% of these toms such as heartburn and regurgitation are present.
patients are well controlled throughout life with proton Objective testing with ambulatory pH monitoring can
pump inhibitors (PPIs). Their disease does not seem to confirm the diagnosis, but is infrequently done when
be progressive, although some dose escalation may be patients first present with symptoms. Instead, most patients
needed for control. receive empiric acid suppressive treatment with the sole
From this perspective, progression of GERD is limited objective of symptom control. A positive empiric test of PPI
to the approximately 30% of GERD patients in whom PPI therapy is commonly used to confirm the symptom-based
therapy fails to control symptoms (Fig. 3.1). There is no diagnosis of GERD.2
ability or attempt to prevent the progression of 30% of There is no symptom complex or test at present that
GERD patients into the stage of refractory GERD defined can accurately predict which GERD patient under empiric
by treatment failure. Patients who fail to be controlled with treatment will progress to failure of PPI therapy in the
PPIs live a life whose quality is compromised to varying future. Failure is recognized only when maximum PPI
degrees by their symptoms. It is only when they reach this therapy fails to control symptoms. There is no symptom
stage defined by failure of PPIs to control symptoms, or complex or endoscopic finding short of Barrett esopha-
when they develop alarm symptoms such as dysphagia, gus that can predict with sufficient accuracy to warrant
that endoscopy is indicated.2 surveillance endoscopy that a GERD patient will develop
15
Esophageal Mucosa in Health and Disease CHAPTER 3 15.e1
ABSTRACT
The area distal to the end of the tubular esophagus is
confusing and controversial. Existing viewpoints that
cardiac epithelium normally lines the proximal stomach
and that the gastroesophageal junction (GEJ) is accurately
defined at endoscopy by the proximal limit of rugal folds
are not evidence based. Correction of these errors results
in the recognition that the normal state is an esophagus
that is entirely tubular and lined by squamous epithelium
while the stomach is lined by gastric oxyntic epithelium.
Cardiac epithelium always results from metaplasia of
esophageal squamous epithelium when it is exposed to
gastric juice. Cardiac metaplasia of esophageal squamous
epithelium results in loss of lower esophageal sphincter
(LES) pressure and dilatation of the esophagus. Measure-
ment of cardiac epithelium distal to the endoscopic GEJ is
an exact histologic measure of damage to the abdominal
segment of the LES. Gastroesophageal reflux disease,
when viewed from the perspective of LES damage, is an
inexorably progressive disease whose progression varies
with the rate of progression of LES damage. This permits
recognition of the root cause of gastroesophageal reflux
disease (GERD) at its earliest preclinical stage, providing a
new method of early diagnosis and intervention designed
to prevent complications of GERD. This has the potential
to prevent esophageal adenocarcinoma.
KEYWORDS
Gastroesophageal reflux disease, lower esophageal sphinc-
ter, histopathology, diagnosis, dilated distal esophagus
16 SECTION I Esophagus and Hernia
adenocarcinoma in the future. Screening for Barrett A revolution is essential if there is to be any control of
esophagus is not recommended.3 the ever-increasing incidence of adenocarcinoma.6 This
This treatment algorithm therefore precludes any is the first attempt at such a revolution.
method that can prevent the progression of GERD to its
severe end points of treatment failure and adenocarcinoma.
When the end point of severe GERD is compromised PROGRESSION OF GASTROESOPHAGEAL
quality of life, antireflux surgery offers the only hope of REFLUX DISEASE WITH EMPIRIC PROTON
control. However, surgery has its own problems and is PUMP INHIBITOR THERAPY
performed relatively rarely. Many patients who opt to not
have surgery continue to live a life that is disrupted by The best available scientific prospective study of long-term
fear of eating, sleep deprivation, and loss of productivity outcomes associated with treating symptomatic GERD with
at work.5 acid suppressive medical therapy is the Pro-GERD study.7
When the end point is advanced adenocarcinoma, hope A total of 6215 patients older than 18 years old with the
exists for very few patients and too commonly for a very primary symptom of heartburn were enrolled into this
short period of time (see Fig. 3.1). Only 10% of patients prospective multicenter open cohort study in Europe.
developing adenocarcinoma have ever had a diagnosis of The study was largely conducted under the auspices of
Barrett esophagus. If carefully followed with surveillance Astra-Zeneca, makers of esomeprazole, which makes any
endoscopy, these patients can be detected with early-stage result that suggests a negative effect of PPI therapy highly
cancer that is amenable to endoscopic therapy, which is credible.
often curative and obviates the need for esophagectomy, All patients underwent an index endoscopy done in
chemotherapy, and radiation. Unfortunately, most patients selected centers by endoscopists who received special
have a dismal outcome, with a 5-year survival of around training. Endoscopic findings were recorded and the
15%. patients given 4 to 8 weeks of PPI therapy with assessment
This is a sad commentary of our present management of symptom control and repeat endoscopy to assess healing.
of GERD. We have abandoned the hallowed principles They were then sent back to their primary care physicians
of early diagnosis and prevention in favor of an illogical for continuation of empiric acid suppressive treatment
and unrealized hope that PPIs will cure the disease. We at their discretion. Treatment used during follow-up and
simply permit the development severe GERD and then symptom control was monitored by questionnaires, and
struggle with few good answers to prevent the inevitable 2721 of this cohort of patients reported to the study centers
impaired quality of life and progression to adenocarcinoma for repeat endoscopic assessment at 5 years.
in a highly significant minority of patients with GERD. At the initial endoscopy, the distribution of endoscopic
There is no attempt to control progression of GERD. changes of these 2721 patients was as follows: nonerosive
There is no attempt to prevent adenocarcinoma or its disease, 1224; erosive disease LA A/B, 1044; erosive disease
premalignant state, Barrett esophagus. There is no attempt LA C/D, 213; and 240 (8.8%) patients with vCLE. (Note:
to prevent the state of misery associated with GERD that vCLE was reported as “Barrett oesophagus, endoscopic”
becomes refractory to PPI therapy. and “Barrett oesophagus with histologic confirmation,” the
Esophageal Mucosa in Health and Disease CHAPTER 3 17
latter with intestinal metaplasia.) The patients with vCLE We will also explore how this simple histologic test for a-LES
at the initial endoscopy were not included in this study. damage can transform the future management of GERD.
Reversal and prevention of progression of erosive
esophagitis at 5 years was impressive. Of the 1041 patients
with nonerosive disease at baseline, 784 remained nonero- A PROPOSED NEW OBJECTIVE IN THE
sive, 248 progressed to LA A/B, and 9 to LA C/D erosive MANAGEMENT OF GASTROESOPHAGEAL
disease. Of the 918 patients with LA A/B erosive disease REFLUX DISEASE
at baseline, 578 had reversed to nonerosive disease, 331
remained LA A/B, and 9 had progressed to LA C/D The present treatment algorithm for GERD (see Fig. 3.1)
erosive disease. Of the 188 patients with LA C/D erosive can be described as totally reactive. There is no defined
disease at baseline, 94 now had nonerosive disease, 78 objective aimed at detecting or preventing any cellular
had LA A/B, and 16 stayed at LA C/D erosive disease. change that may be a harbinger of adenocarcinoma.
Over a period of 5 years, the number of patients with We simply wait for symptoms that are “troublesome” to
severe erosive esophagitis had decreased from 188 to 34. begin empiric PPI therapy1; then wait for failure of PPI
Regular intake of PPI reduced the likelihood of progres- therapy to perform endoscopy2; and then wait for the
sion compared with on-demand PPI or other therapy. occurrence of high-grade dysplasia and adenocarcinoma.
The severity of symptoms at baseline was not a predictor The only proactive event in this algorithm that improves
of progression to severe erosive esophagitis. It could outcomes is Barrett esophagus surveillance, but current
reasonably be concluded that PPI therapy was highly recommendations for surveillance intervals often render
effective in healing erosive esophagitis. this effort a failure too.
In contrast, 241 (9.7%) patients who did not have vCLE Even worse, most physicians convince themselves that
initially had developed this at 5 years. These patients who PPI therapy is a wonderful method of treating GERD that
progressed included 72 of 1224 (5.9%) who originally brings comfort to millions of GERD sufferers. This is true.
had NERD, 127 of 1044 (12.1%) with LA grade A/B, and However, we hide and ignore the greatest increase of a
42 of 213 (19.7%) with LA grade C/D erosive esophagitis. specific cancer type in the history of medicine that has
The factors significantly associated with progression to concurrently occurred while patients are being treated
vCLE at 5 years were (1) female gender, which had a with increasingly effective acid reducing drugs.6
negative association (P = .041); (2) alcohol intake (P = In this chapter, we will attempt to change the present
.033); (3) erosive esophagitis compared with NERD (P < outcomes of GERD with a new approach based on the
.001); and (4) regular PPI use (P = .019). development of a new understanding of GERD based on
These data show that empiric PPI therapy titrated the pathogenesis of progression of a-LES damage.
to control symptoms in the primary care setting heals It is well known that GERD is the result of LES damage.
erosive esophagitis effectively but simultaneously results As such, focus on LES damage attacks the problem at its
in an endoscopic progression to vCLE with and without root. The primary objective of the new approach is to
intestinal metaplasia. Whether PPI therapy causes this turn the curve of increasing incidence of adenocarcinoma
conversion is unproven. However, the study data prove downward all the way to zero. A secondary objective is to
that nearly 10% of the GERD population under empiric prevent failure of medical therapy.
acid reducing treatment will progress from not having
vCLE to vCLE within 5 years. DEFINING A CRITERION OF IRREVERSIBILITY:
When one considers that 20% to 40% of the population VISIBLE COLUMNAR LINED ESOPHAGUS
have symptomatic GERD, 10% translates to an absolute The first step in preventing adenocarcinoma is to recognize
number that easily explains why GERD-induced adeno- the point of irreversibility that signals the inability to
carcinoma has increased sevenfold in the past 4 decades.6 prevent progression to adenocarcinoma. In GERD, at
this point in time, that point of irreversibility is the occur-
rence of vCLE. In the United Kingdom, vCLE defines
VALUE OF PATHOLOGY IN THE DIAGNOSIS Barrett esophagus.8 In the United States and Europe,
OF GASTROESOPHAGEAL REFLUX DISEASE intestinal metaplasia is required for the diagnosis of Barrett
esophagus.
Pathologic criteria for diagnosis of GERD are presently Medical treatment does not reverse vCLE or prevent its
limited to changes in the squamous epithelium of the progression to intestinal metaplasia, increasing dysplasia,
esophagus that result from exposure to gastric contents. and adenocarcinoma. Present medical treatment of GERD
Reflux esophagitis is characterized by intercellular edema therefore commits 10% of all patients to irreversibility
(dilated intercellular spaces), basal cell hyperplasia, every 5 years.7
papillary elongation, and infiltration by eosinophils and The advantage with defining irreversibility in GERD
neutrophils. These changes do not have the necessary by the presence of vCLE is that there is no evidence
sensitivity or specificity for the diagnosis of GERD. As that any patient who does not have vCLE progresses to
such, histologic examination of biopsies has no practical adenocarcinoma. If we prevent vCLE, we will prevent
value in the diagnosis of GERD. adenocarcinoma.
Pathologic criteria do not exist at present for assessment It can be reasonably argued that the person who is endo-
of the LES. In this chapter, we will develop a new set of scopically normal with intestinal metaplasia at the normal
pathologic criteria that can define the presence and extent SCJ is at risk for adenocarcinoma of the “gastric cardia.”
of damage to the abdominal segment of the LES (a-LES). However, present management guidelines recommend
18 SECTION I Esophagus and Hernia
that such patients with GERD should not undergo biopsies These reasons for early endoscopy are presently not
because the risk of cancer in patients who have intestinal justified because of the cost associated with increasing the
metaplasia is unknown.2 The argument, therefore, has no number of endoscopies. However, it emphasizes the fact
practical merit at this time. It may change in the future if that any push to prevent adenocarcinoma must change
an increased cancer risk is defined in this group. If and the indications for endoscopy to an earlier stage in the
when that happens, preventing intestinal metaplasia at progression of GERD. This will only happen if a new and
the SCJ in the endoscopically normal person will become more accurate method of predicting progression of GERD
necessary. to vCLE becomes available. The new histologic measure
The detection of vCLE requires endoscopy. The present of LES damage that we propose can be that test.
management guidelines delay endoscopy to the point of
treatment failure. At this point a significant number of CAUSE OF VISIBLE COLUMNAR LINED ESOPHAGUS
patients will already have vCLE. If endoscopy is performed To be effective in preventing vCLE, we must identify its
proactively without waiting for treatment failure, as was cause. It is certain that vCLE is the result of exposure
done in the Pro-GERD study, 240 of 2721 (8.8%) patients of the esophagus above the endoscopic GEJ to gastric
would already have vCLE.7 In addition, the following contents as a result of reflux. As such, it is also certain
endoscopic findings were predictive of progression to vCLE that if reflux can be prevented, vCLE will not occur.
in the next 5 years: (1) presence of erosive esophagitis with There is strong evidence that the risk of vCLE increases
risk increasing to 19.7% in patients with severe erosive with increasing severity of reflux (demonstrated by objec-
esophagitis7; and (2) presence of intestinal metaplasia tive evidence of acid exposure by a pH test), increasing
in a biopsy taken from the SCJ of an endoscopically duration of reflux, male gender, regular PPI therapy, and
normal patient, with such patients having a 25% risk of possibly alcoholism and smoking. Nason et al.10 showed that
progression to vCLE within 5 years.9 The patients in the the prevalence of Barrett esophagus was higher in patients
Pro-GERD study had well-established GERD, often with whose symptoms were controlled with PPI therapy. They
severe symptoms and a long duration.7 It is probable that suggested that the present practice of waiting for treatment
endoscopy performed at the onset of GERD would have failure was irrational if the objective for endoscopy was
a lower prevalence of vCLE. the detection of Barrett esophagus.
In the Pro-GERD study, the nonendoscopic findings that The most dominant factors in the etiology of vCLE are
were significantly associated with progression to vCLE in the severity and duration of reflux. Patients with Barrett
GERD patients under medical therapy were male gender, esophagus are known to have a higher prevalence of an
alcohol use, and regular PPI use. abnormal pH test than any other category of GERD. There
None of the nonendoscopic criteria that are predictive is no specifically defined level of abnormality in the pH
for development of vCLE within 5 years listed previously test or a specific number of years of reflux that correlates
are indications for endoscopy in the GERD patient. The with the occurrence of vCLE. If the objective is preventing
indication remains the occurrence of treatment failure. vCLE, success will demand intervention at the earliest
The main reason for this is the lack of any desire to prevent practical time after the onset of reflux. For prevention
vCLE in the minds of the medical community. To them, of vCLE to be certain, intervention must occur before
vCLE is simply another inevitable event in the course of any significant reflux occurs into the thoracic esophagus.
GERD that occurs in a minority of GERD patients. The From a practical standpoint, it is necessary to identify
fact that it is a cellular change whose end point is a lethal criteria that separate very low and high risk of impending
malignancy is ignored. and future vCLE by some defined severity and/or duration
This is a nihilistic attitude that permits conversion of reflux. This is not possible by presently available tests.
of the patient without risk to one whose progression to We will propose that the new test of LES damage provides
adenocarcinoma becomes inevitable. The only excuse for accurate criteria for predicting future vCLE.
this attitude is that cancer is rare in GERD patients. With It is certain that the severity of reflux into the thoracic
the sevenfold increase in the incidence of GERD-induced esophagus correlates with the frequency of LES failure,
adenocarcinoma over the past 4 decades,6 this excuse has which in turn correlates with the severity of LES damage.
become increasingly lame and unacceptable. In relation to our objective of preventing vCLE, this
If the presence of vCLE is recognized as the point recognition establishes a new more practical objective:
of irreversibility in GERD, there can be a new objective prevention of reflux into the thoracic esophagus that is severe
of management of the GERD patient: the prevention of enough to cause vCLE.
progression to vCLE.
This would then provide an incentive and demand for
earlier endoscopy before failure with empiric treatment
LOWER ESOPHAGEAL SPHINCTER
with PPI in the patient with GERD. Early endoscopy One of the great obstacles to the study of GERD is the
presently has the ability only to recognize the presence of absence of a pathologic method of assessing the LES by
vCLE and predict its occurrence within the next 5 years pathology at autopsy and resection specimens. Careful study
by the presence of severe erosive esophagitis (19.7%) and of the region has identified complicated arrangement of
intestinal metaplasia at the normal SCJ (25%). Successful the muscle fibers that may represent the LES,11 but these
repair of the damaged LES in the patient with a high risk cannot be translated into routine pathology practice. The
of vCLE in 5 years has a high probability of preventing LES can only be defined and measured by manometry
vCLE. (Fig. 3.2).
Esophageal Mucosa in Health and Disease CHAPTER 3 19
The LES acts as a beautifully designed barrier that segment. Manometric studies of asymptomatic subjects
prevents reflux of gastric contents into the esophagus.12,13 indicate that the LES pressure is greater than 15 mm Hg,
The LES pressure is normally greater than 15 mm Hg, the total LES length is 40 to 50 mm, and the length of
exceeding the baseline luminal pressure in the esophagus the abdominal segment (a-LES) is 30 to 35 mm.
(normally around −5 mm Hg) proximally, and the base- The criteria that define a defective LES that correlates
line luminal pressure in the stomach (normally around with the presence of sufficient reflux into the esophagus
+5 mm Hg) distally. The LES therefore acts as a valve to produce clinical GERD are13 (1) a decrease in the mean
that effectively prevents reflux along the natural pressure LES pressure to less than 6 mm Hg, (2) a decrease in
gradient that exists from the stomach into the esophagus total LES length to less than 20 mm, and (3) a decrease
(Figs. 3.3 and 3.4). in a-LES length to less than 10 mm. At these levels of
LES damage, sphincter failure occurs so frequently that
DEFINING THE NORMAL AND DEFECTIVE LOWER it results in an abnormal pH test and significant exposure
ESOPHAGEAL SPHINCTER BY MANOMETRY of the squamous epithelium in the body of the esophagus
The functional state of the LES can be defined mano-
metrically by three separate components12,13: its mean
pressure, its total length, and the length of its abdominal
Esophagus
Shortened LES
taken up into gastric
contour (temporary)
Exposed
squamous Stomach
epithelium (overdistended by
a heavy meal)
ORIGINAL GEJ
THORAX
Shortened
– 5 mm Hg LES
LES
15–20 mm Hg
CLE in dilated
distal esophagus
ABDOMEN
No LES pressure
+ 5 mm Hg + 5 mm Hg
FIGURE 3.3 Effect of loss of pressure in the abdominal segment of the lower esophageal sphincter (LES). The normal resting pressure
of the abdominal LES overcomes the positive intraluminal pressure in the abdominal esophagus and maintains the tubal shape of the
esophagus. When the LES pressure is lost, the intraluminal pressure causes this part of the distal esophagus to dilate. CLE, Columnar
lined esophagus.
20 SECTION I Esophagus and Hernia
TABLE 3.1 Length of Abdominal Segment of Lower Esophageal Sphincter (a-LES) Damage (Measured by the New Test),
Length of the Residual Functional a-LES, and Their Correlation With Lower Esophageal Sphincter Failure and Severity of Reflux*
Residual Postprandial Severity of Reflux
a-LES Damage a-LES Length a-LES Length Probability of LES Failure (% time pH < 4)
Zero 35 mm 25 mm Zero Zero
>0–<5 mm 30–35 mm 20–25 mm Zero Zero
5–<10 mm 25–30 mm 15–20 mm Zero Zero
10–<15 mm 20–25 mm 10–15 mm Postprandial—rare >zero–4.5%
15–<20 mm 15–20 mm 5–10 mm Postprandial—frequent >zero–4.5%
20–25 mm 10–15 mm 0–5 mm Postprandial—very frequent >4.5%
25–30 mm 5–10 mm Zero Incessant >>4.5%
30–35 mm Zero–5 mm Zero Incessant >>>4.5%
*We assume that the patient has an initial a-LES length of 35 mm, that a heavy meal causes 10 mm of dynamic shortening of the a-LES in the post-
prandial phase, and that LES failure occurs at an a-LES length of less than 10 mm.
GERD, Gastroesophageal reflux disease; LES, lower esophageal sphincter. Green areas, The LES is competent with damage that is within its reserve
capacity. Orange areas, Clinical GERD from onset of symptoms to point of transition from postprandial reflux to incessant reflux and an increasing
prevalence of visible columnar-lined esophagus (vCLE). Red areas, The LES is incompetent with severe reflux and a high prevalence of vCLE.
to reflux.13 LES damage defined by these criteria correlate This only assumes that the anatomic part of the abdomi-
with an increased probability of symptoms of GERD, severe nal esophagus that contains the a-LES does not disappear
grades of erosive esophagitis, and vCLE. into thin air when LES pressure is lost.
There is a significant gap between the previously listed Manometrically, LES damage is equivalent to loss of
criteria that define a normal LES and a defective LES that pressure. When this occurs at the distal end, it results in
is associated with abnormal reflux into the esophagus, shortening of the manometric a-LES. The damaged LES
as defined by an abnormal pH test and the presence of is distal to the end of the residual LES at manometry and
clinical GERD. The mean LES pressure must decrease from therefore identical in its pressure characteristics to the
a normal of greater than 15 mm Hg to less than 6 mm Hg; proximal stomach.
the total LES length must decrease from a normal of 40 In a patient with LES damage, the distal limit of the
to 50 mm to less than 20 mm; and the a-LES length must manometric LES is not the end of the esophagus (see
decrease from 30 to 35 mm to less than 10 mm before it Fig. 3.2). The true end of the esophagus includes the
becomes a criterion of LES failure. damaged a-LES. Any manometric interpretation that
At least part of this gap between a normal and defective makes the assumption that the esophagus ends at the
LES represents the reserve capacity of the LES. As LES distal end of the manometric LES is potentially wrong
damage increases, its reserve capacity is progressively by as much as 35 mm (the entire initial a-LES length).
reduced. However, as long as it is not exhausted, the For example, if the distal limit of the manometric LES is
LES maintains its competence (green zone in Table 3.1). above the diaphragmatic pressure impression, this is not
This early LES damage cannot be recognized by any necessarily a hiatal hernia because the true end of the
present criterion for the diagnosis of GERD: the patient has esophagus cannot be defined by manometry.
no symptoms, no endoscopic abnormality, no manometric At present, the previous formula that defines the LES
criteria of a defective LES, and no abnormal pH test. This cannot be applied because two elements, LES damage
state where the LES is damaged within its reserve functional and the initial LES length, are unknown. As a result,
capacity can be called the phase of compensated LES damage. manometry has no practical value in the diagnosis of
We will show that histologic examination with new criteria GERD. However, it illustrates the critically important and
can define and measure this early LES damage. misunderstood concept that the manometric definition of
Before the onset of LES damage, all persons have an the distal end of the a-LES is not the end of the esophagus.
initial a-LES length that is equal to the length of the The true end of the esophagus must include the damaged
abdominal esophagus. Zaninotto et al.13 reported that a-LES that is present distal to the manometric end of
the manometric length of the a-LES in 49 asymptomatic the LES in virtually all people This cannot be measured
volunteers had the following distribution (I have taken the at present.
liberty of removing one outlier that had an a-LES length In Zaninotto et al.13, therefore, the measured manomet-
of >50 mm): less than 10 mm in 1; 10 to 15 mm in 6; 15 ric a-LES does not necessarily represent individual variation
to 20 mm in 10; 20 to 25 mm in 17; 25 to 30 mm in 11; of the length of the normal a-LES, simply because the
and 30 to 35 mm in 5 persons. subjects had no symptoms of GERD. It could be the result
The manometric measurement of the a-LES at any of shortening of the a-LES by progressive a-LES damage.
given point in a person’s life after the LES has developed The data in the study can be explained by assuming that
completely can be expressed by the following formula: the initial a-LES length was 35 mm (the highest length)
in all patients, and the distribution represents different
Initial a-LES length = manometric a-LES length degrees of a-LES damage. For example, an asymptomatic
+ LES damage person with a measured manometric a-LES length of
Esophageal Mucosa in Health and Disease CHAPTER 3 21
22 mm (the median a-LES length in the study) could have meals when the stomach distends and the intragastric
an initial length of 35 mm with 13 mm of a-LES damage pressure increases. The distal abdominal esophagus that
(see Table 3.1). That person is asymptomatic because the has lost LES pressure will therefore dilate to form the
LES, though damaged, is still sufficiently competent to dilated distal esophagus (see Fig. 3.3).14
prevent reflux. With LES damage, the tubular abdominal esophagus
The distribution of acid exposure in these volunteers shortens, the damaged esophagus dilates15 and takes up
in Zaninotto et al.13 showed a pH less than 4 for a mean the gastric contour and becomes part of the reservoir, and
of 1.57%, a median of 1.1%, and a range of 0% to 6% of the angle of His becomes more obtuse.16 Mucosal rugal
the 24-hour period. This shows that these asymptomatic folds, which are a feature of all reservoir organs, develop
persons had evidence of mild reflux with 5% reaching the in this dilated distal esophagus that results from loss of
pH test definition for abnormal reflux. This was objective abdominal LES function (as discussed later).
evidence of LES failure, despite the fact that they did not The dilated distal esophagus has a variable length that
have symptoms. is equal to the amount of shortening of the a-LES due
The data in Zaninotto et al.13 raise the intriguing but to damage. The equation that defines the a-LES now
obvious probability that the LES has a reserve capacity. resolves as follows:
It can shorten significantly from its initial length while Initial length of a-LES = length of residual a-LES
remaining competent—that is, there is a phase of com- ( = tubular abdominal esophagus )
pensated LES damage where patients have LES damage + length of LES damage ( = dilated distal esophagus)
within its reserve capacity without significant LES failure
and reflux into the thoracic esophagus. A person without The end of the tubular esophagus, which has been
significant reflux into the thoracic esophagus will be at used by pathologists to define the GEJ since Hayward in
zero risk for developing vCLE. 196117, is proximal to the true GEJ by the length of the
Based on this understanding, we can divide the severity dilated distal esophagus.
of LES damage into (1) compensated, (i.e., LES damage This “gastricization” of the abdominal esophagus that
is such that it does not produce LES failure, where the has lost LES tone occurs at a manometric, endoscopic,
pH test is zero; green zone in Table 3.1); (2) LES damage and gross anatomic level. This has led to confusion that
that causes infrequent LES failure and mild reflux (i.e., has created error in this region from the beginning of
pH test is greater than zero but pH test normal; <4.5% of time and continues to the present.18
time pH < 4 or DeMeester score < 14), noting that vCLE We will show that it is only the correct interpretation
is extremely unlikely in such patients (orange zone in of the histology of this region that can resolve this error.
Table 3.1); and (3) severe LES damage with LES failure
sufficient to produce an abnormal pH test and a high MECHANISM OF ABDOMINAL LOWER ESOPHAGEAL
prevalence of vCLE (red zone in Table 3.1). SPHINCTER DAMAGE
This further refines our objective into an LES-based LES damage is the result of pressure exerted from below, as
objective to prevent vCLE: prevention of a-LES damage beyond a result of a heavy meal that causes gastric overdistention.
the point where reflux is sufficiently severe to cause vCLE. In Ayazi et al.19 and Robertson et al.20 showed elegantly that
Table 3.1, this corresponds to preventing a-LES damage gastric overdistention causes “effacement” of the distal
from reaching 25 mm. When there is a measure of a-LES part of the LES, resulting in a temporary decrease in LES
damage, there is a range of zero to 25 mm of LES damage length. The squamous epithelium lining the effaced LES
that is available for intervention to prevent progression of is exposed to gastric juice because the pH transition point
LES damage. Prevention of vCLE becomes theoretically has moved proximally (see Fig. 3.4).
very feasible in this method. The phenomenon of effacement of the distal end of the
This new objective clearly shows the futility of present LES can be demonstrated at endoscopy. In a person with
management of GERD. The presently accepted criteria that normal endoscopy, the SCJ is the GEJ. In retroflex view,
define GERD (troublesome symptoms, erosive esophagitis, when the stomach is insufflated with air, the SCJ moves
an abnormal pH test, and a defective LES on manometry downward and becomes visible. When the same thing
where the a-LES is <10 mm) are the very things that happens during a heavy meal, the squamous epithelium
must be prevented if we hope to prevent esophageal is in the stomach, below the pH transition point.
adenocarcinoma. There is a pocket of strong acid at the height of the
food column during a meal.21 Repeated and frequent
exposure of the squamous epithelium to this acid pocket
RESULT OF ABDOMINAL LOWER ESOPHAGEAL during gastric overdistention during heavy meals results
SPHINCTER DAMAGE: THE DILATED first in reversible injury to the distal esophageal squamous
DISTAL ESOPHAGUS epithelium, followed by permanent columnar metaplasia
A largely unappreciated normal function of the a-LES is to of the squamous epithelium.
maintain the tubular shape of the abdominal esophagus. If LES damage occurs because of pressure from below,
The high resting pressure of the a-LES continually opposes it must follow that LES damage begins at its distal end
the dilatory tendency of the positive (around +5 mm Hg) and progresses upward. Loss of length therefore begins in
intraluminal pressure of the abdominal esophagus. the distal a-LES. Robertson et al.20 showed that early LES
When the a-LES is damaged, the protection provided shortening produced by a heavy meal in asymptomatic
by the tonic contraction of the LES is lost. The dilatory volunteers was entirely in the abdominal segment and
positive intraluminal pressure will be accentuated during did not affect the thoracic LES.
22 SECTION I Esophagus and Hernia
TABLE 3.2 Changes With Age of the Functional Residual Length of the Abdominal Lower Esophageal Sphincter*
LES damage can therefore be considered to be basically patients with GERD without a hiatal hernia, and patients
the result of an eating disorder. Viewed in this light, each with GERD who had a hiatal hernia.
person can be regarded as having a unique relationship The baseline LES length in the fasting state was progres-
between his/her eating habit, the response of the LES to sively less in normal persons compared to nonhernia
this overeating, and the damage caused to the esophageal GERD with hernia GERD. This correlated with an increase
squamous epithelium by exposure to gastric juice. in baseline reflux as measured by a pH electrode placed
At one extreme, the patient’s LES is not damaged by 5 cm above the upper border of the LES.
the effect of his/her eating habit on the LES. This patient In this study, Kahrilas et al.22 infused air into the stomach
never has LES failure and reflux, the pH test is zero, and at 15 mL/min, causing progressive gastric distention. This
this person never gets GERD. At the other extreme, the caused an additional shortening of the LES of 5 to 7 mm
patient’s LES is damaged early in life by an excessive from baseline in all three groups as distention increased.
eating habit and/or an LES susceptible to damage and The additional temporary shortening of the LES was similar
progresses rapidly to LES incompetence and severe reflux in the three groups, suggesting that gastric overdistention
into the thoracic esophagus at a relatively young age. caused LES exposure to gastric contents in a linear manner.
This damage includes erosive esophagitis and becomes During the temporary shortening of the LES with gastric
irreversible when vCLE occurs. distention, the number of reflux episodes and total acid
Between these two extremes is the entire clinico- exposure in the esophagus increased significantly and most
pathologic spectrum of GERD. Progression of GERD can prominently in the hernia-GERD group. This showed that
therefore be defined by the rate of progression of LES a damaged LES with a shorter baseline length was more
damage resulting from a person’s eating habit (Table 3.2). susceptible to failure when exposed to gastric distention.
This study confirms that a-LES length is a critical deter-
minant of the severity of reflux (objectively measured in
RELATIONSHIP BETWEEN ABDOMINAL LOWER a pH test). It also shows that significant reflux can occur
ESOPHAGEAL SPHINCTER LENGTH AND LOWER during the postprandial phase in asymptomatic persons.
ESOPHAGEAL SPHINCTER FAILURE
LES damage is a progressive phenomenon. When looked
at from the perspective of LES damage, progression is HISTOLOGIC MEASUREMENT OF ABDOMINAL
inexorable from the onset of LES damage to the end LOWER ESOPHAGEAL SPHINCTER DAMAGE
point defined as the end of life, or some intervention that
stops the progression, such as an antireflux procedure. The objective of preventing vCLE is not possible at the
PPI therapy has no positive impact on the rate of pro- present time, because there is no test that has the ability to
gression of LES damage. By removing the pain associated predict with sufficient accuracy those patients at high risk
with reflux and allowing the patient to eat excessively, PPIs of progressing to vCLE. The use of symptom severity has
may actually prevent the body’s natural defense against no value; some patients with vCLE are asymptomatic. The
progression of LES damage. control of symptoms with PPIs has a negative correlation
In general, the greater the LES damage, the greater with prevalence of vCLE.10 There is no defined value in
is the severity of reflux. Kahrilas et al.22 beautifully dem- the pH test or manometry that can predict impending or
onstrated the close relationship between decreasing LES future vCLE. The presence of severe erosive esophagitis
length and LES failure. They measured baseline total LES and intestinal metaplasia in a biopsy of the SCJ in the
length in three groups with increasing severity of GERD: endoscopically normal GERD patient has a 20% to 25%
patients who had no symptoms of GERD (“normal”), known progression to vCLE within 5 years, but this has
Esophageal Mucosa in Health and Disease CHAPTER 3 23
IM
Esophagus Esophagus CM
OCM
Diaphragm Diaphragm
GEJ GEJ
Stomach Stomach
A B
Esophagus SUBMUCOSAL
GLAND DUCT
Diaphragm
FIGURE 3.7 The histologic composition of the dilated distal
esophagus, showing the three metaplastic columnar epithelial
types. Intestinal metaplasia with goblet cells (IM) is proximal,
GEJ cardiac epithelium (CM) is in the middle, and oxyntocardiac
epithelium with parietal cells (CCM) is distal (on the left). Note the
Stomach presence of submucosal gland ducts. Ducts of submucosal
glands are specific for the esophagus; their presence proves that
the location of this tissue is esophageal.
C
FIGURE 3.6 Progression of the gap between the squamocolumnar TABLE 3.3 Histologic Criteria for Diagnosis of Four
junction and gastric oxyntic epithelium with increasing severity of Columnar Epithelial Types Encountered in the Esophagus and
gastroesophageal reflux disease. (A) Normal state with no gap; Proximal Stomach*
the squamous epithelium (gray) transitions directly to gastric
oxyntic epithelium (blue); note rugal folds (lines). (B) Metaplastic
columnar epithelium limited to the dilated distal esophagus. This is Mucous Cells Parietal Goblet
depicted with intestinal metaplasia (yellow), cardiac epithelium in Glands† Cells Cells
(green), and oxyntocardiac epithelium (purple). Note that these Gastric oxyntic epithelium − + −‡
epithelia have replaced squamous epithelium. The proximal limit of Cardiac epithelium + − −
gastric oxyntic epithelium has not moved. The area of columnar Oxyntocardiac epithelium + + −
metaplasia of squamous epithelium is dilated and has developed Intestinal epithelium + − +
rugal folds. This is the dilated distal esophagus resulting from *Gastric oxyntic epithelium lined the entire proximal stomach. Cardiac,
abdominal lower esophageal sphincter (LES) damage. This is oxyntocardiac, and intestinal epithelia are, when present, interposed
presently mistaken for proximal stomach (gastric cardia), because between the squamous epithelium and gastric oxyntic epithelium (i.e., form
it is distal to the end of the tubular esophagus and the proximal the squamo-oxyntic gap). Note: There is no epithelium defined in this
limit of rugal folds. (C) Final phase of progression where LES scheme that has both parietal and goblet cells in one foveolar-gland
damage has led to sufficient reflux into the esophageal body complex. This is an extremely rare finding; when found, goblet cells take
precedence and the epithelium is designated as intestinal.
to cause visible columnar-lined esophagus (vCLE). (Note: The †
Mucous cells are present at the surface and foveolar pit in all epithelial
damaged LES is shown as a white wall that has replaced the red types; it is the presence of mucous cells in glands below the foveolar pit
wall where the LES is intact.) GEJ, Gastroesophageal junction. that are relevant to the definitions.
‡
Gastric oxyntic epithelium with atrophic gastritis can have goblet
cells. This is intestinal metaplasia in gastric oxyntic epithelium, which is
different than cardiac (metaplastic esophageal) epithelium with intestinal
for which evidence is lacking, this definition of the GEJ metaplasia.
has incredible universal acceptance.
Chandrasoma et al.27 have shown conclusively that
this endoscopic definition of the GEJ is incorrect (Fig. its position. In the person with LES damage, whether
3.8A and B). They showed that the area distal to the symptoms or GERD are present or not, the true GEJ is
endoscopic GEJ lined by cardiac epithelium (with and separated from the endoscopic GEJ by cardiac epithelium
without parietal and/or goblet cells) was the esophagus, (with and without parietal and/or goblet cells; see Fig. 3.6).
by virtue that submucosal glands that are specific to the The present use of the endoscopic GEJ results in an
esophagus were present in and concordant with the length error that is equal to the length of the dilated distal
of the dilated distal esophagus (Fig. 3.9). esophagus. Ironically, the greater the amount of LES
The correct definition of the true GEJ is the proximal damage (i.e., the more severe the GERD), the greater the
limit of gastric oxyntic epithelium. This never changes error. This error is made at endoscopy, manometry, and
Esophageal Mucosa in Health and Disease CHAPTER 3 25
A B
FIGURE 3.8 Present incorrect and correct interpretation of an esophagectomy specimen. (A) This specimen shows a tubular esophagus
lined by 5.5 cm of visible columnar-lined esophagus (vCLE) above the proximal limit of rugal folds. There is an ulcerated adenocarcinoma
immediately distal to the squamocolumnar junction. The area distal to the end of the tubular esophagus is lined by rugal folds. This area
will be interpreted as proximal stomach by present criteria for defining the gastroesophageal junction (GEJ) at endoscopy and gross
dissection. (B) Histologic findings show that 20.5 mm of the area distal to the end of the tubular esophagus and containing rugal folds is
lined by cardiac epithelium with intestinal metaplasia proximally and oxyntocardiac epithelium distally. The red line is the squamocolumnar
junction; the yellow line is the distal limit of intestinal metaplasia; the black line is the proximal limit of gastric oxyntic epithelium, which is
the true GEJ. The dilated distal esophagus between the end of the tubular esophagus and the true GEJ contains submucosal glands
(black dots), whose extent is concordant with the length of cardiac epithelium (with parietal and goblet cells). This is proof of the dilated
distal esophagus.
A B C D
A
FIGURE 3.10 Multilevel biopsy protocol for measuring the length of
cardiac epithelium in the dilated distal esophagus (blue) between
the distal limit of squamous epithelium (pink) and the proximal limit
of gastric oxyntic epithelium (green). Three biopsies (black circles)
are taken at 5 mm intervals. (A) Normal state with no cardiac
epithelium. (B) Cardiac epithelium present in the zero to 5 mm
biopsy. (C) Cardiac epithelium present in the zero to 10 mm
biopsies. (D) Cardiac epithelium present in all three biopsies
up to 15 mm distal to the squamocolumnar junction.
epithelium (with and without parietal and/or goblet cells) With a suitable specimen, this can be measured
was present distal to the end of the tubular esophagus to a with an accuracy within 1 µm. The measurement is made
length that varied from a minimum of 4 mm (median) to on a standard histologic slide with a standard micro-
a maximum length of 11 mm (median). In eight patients scope that has an ocular micrometer. These are available
(25%), cardiac and/or oxyntocardiac epithelium was in every pathology laboratory the world over. The test is
situated over submucosal glands. inexpensive.
The true GEJ cannot be seen at endoscopy because the
dilated distal esophagus and proximal stomach both have CLASSIFICATION OF GASTROESOPHAGEAL REFLUX
rugal folds. With standard endoscopy, it is not possible to DISEASE BY THE RESULTS OF THE NEW TEST
differentiate cardiac and gastric oxyntic epithelium. It is The ability to measure a-LES damage opens a new dimen-
possible that newer endoscopic modalities, such as confocal sion in the diagnosis and management of GERD. The
microscopy and optical coherence tomography, can do entire spectrum of the disease from the normal state
this. At present, though, only histologic examination is to the most severe disease can be understood by the
capable of identifying the true GEJ. extent of damage to the 35 mm of the a-LES (see Table
3.1). Correlations between severity of a-LES damage and
frequency of LES failure, severity of reflux, and severity
NEW PATHOLOGIC TEST OF LOWER of cellular changes in the esophagus are likely to be more
ESOPHAGEAL SPHINCTER DAMAGE accurate than with any other measure.
Theoretically, we can divide GERD into four stages
We have proposed a new test that can accurately measure based on the amount of a-LES damage. To do this, we
the presence and severity of a-LES damage in any person, will make the following assumptions: (1) the initial length
whether or not there are symptoms of GERD. LES damage of the a-LES is 35 mm; (2) LES failure correlates with a
is equal to the measured length of the dilated distal functional a-LES length of less than 10 mm; (3) a-LES
esophagus. This is the length of cardiac epithelium (with damage has a linear progression with a variable rate in
and without parietal and/or goblet cells) between the SCJ any person; and (4) dynamic shortening of the a-LES with
and the proximal limit of gastric oxyntic epithelium in a meal has a maximum of 10 mm.
persons without a visible CLE at endoscopy. Four stages of GERD emerge in this new method:
1. Normal. There is no LES damage. The residual a-LES
length is 35 mm. Defined by the absence of a dilated
distal esophagus. This is rare in adults. However, I
GEJ have encountered a 67-year-old male without cardiac
Columnar Lined Esophagus epithelium in an esophagectomy done for squamous
SCJ carcinoma (Fig. 3.11).
2. The phase of compensated a-LES damage. There is
a-LES damage less than 15 mm defined by a dilated
distal esophagus of less than 15 mm. Residual a-LES
length is greater than 20 mm. This is the finding in 70%
of the population at large who do not have symptoms
of GERD. Their LES is competent at all times, and
there is no significant reflux on a pH test (zero to well
GASTRIC
REFLUX CARDITIS below normal).
OXYNTIC
MUCOSA 3. Mild GERD. There is a-LES damage of 15 to 25 mm,
defined by a dilated distal esophagus of 15 to 25 mm.
Residual a-LES length is 10 to 20 mm. This is the finding
in 70% of patients with GERD. Their symptoms are
FIGURE 3.12 The histologic gap composed of cardiac and controlled with PPIs, and there is a low prevalence of
oxyntocardiac epithelia between the distal end of the squamous vCLE. Their LES tends to fail during the postprandial
epithelium and proximal limit of gastric oxyntic epithelium (2 mm period when dynamic shortening decreases the a-LES
long in this section). This is the histologic definition of the dilated length to less than 10 mm. At the low end of this range,
distal esophagus. This patient has 2 mm of abdominal lower the patients are at the onset of disease with infrequent
esophageal sphincter damage. GEJ, Gastroesophageal junction; postprandial reflux. At the high end, they are at the cusp
SCJ, squamocolumnar junction. of severe GERD. Somewhere in the higher end of this
range, they may develop vCLE. They have significant The ability to predict future a-LES damage permits the
reflux with a high normal or abnormal pH test. identification of persons who are at risk of progressing to
4. Severe GERD. There is a-LES damage of greater severe GERD defined by vCLE in the future long before
than 25 mm. The residual a-LES in the fasting state the point at which the person is in danger. This permits
is below the threshold at which LES failure occurs. intervention to slow the progression of LES damage. If
Reflux is severe and unrelated to meals. There is a a successful intervention can be developed, progression
high prevalence of refractory GERD and vCLE. to severe GERD and vCLE can be prevented. The objective
of preventing vCLE is achieved. Adenocarcinoma will not occur
EVIDENCE BASE SUPPORTING THE NEW without vCLE.
DIAGNOSTIC METHOD
We have presented the data that show strong support that
the length of the dilated distal esophagus, measured by the POTENTIAL VALUE OF THE NEW TEST IN
length of cardiac epithelium (with and without parietal THE MANAGEMENT OF GASTROESOPHAGEAL
and/or goblet cells) between the endoscopic GEJ and REFLUX DISEASE
proximal limit of gastric oxyntic epithelium, accurately
represents LES damage. We have shown that the length The new test has obvious value in the management of
of a-LES damage can be measured accurately within 1 µm GERD.
in autopsy and resected specimens.
Unfortunately, there is almost no data relating to EXCLUSION OF GASTROESOPHAGEAL REFLUX
the measured a-LES damage in either asymptomatic DISEASE AS A CAUSE OF SYMPTOMS
people or patients with GERD. This is the result of At the present time, there is no diagnostic test that can
the erroneous beliefs that cardiac epithelium lines the accurately determine whether symptoms that could possibly
normal proximal stomach and the GEJ is defined by the be caused by GERD are actually caused by GERD. Diagnosis
proximal limit of rugal folds. These false dogmas have commonly depends on an empiric PPI test that is known
resulted in a general recommendation by all gastroen- to have a significant false-positive rate.2 The consequence
terology societies that biopsies should not be taken in of a false-positive empiric PPI test is that many people
the person who is endoscopically normal, which has are unnecessarily placed on long-term PPI therapy who
inhibited data being accumulated.2,8 As a result, there do not have GERD. The new test provides a definitive
has never been a systematic study of the dilated distal answer: If the measured a-LES damage is less than 15 mm (or
esophagus. a number based on new data), the symptoms are not caused
We have made several assumptions in using the new test by GERD.
to classify patients into four stages based on a-LES damage.
These assumptions are based on a careful examination of STRATIFICATION OF GASTROESOPHAGEAL REFLUX
the best available evidence. However, the evidence base DISEASE TREATMENT ACCORDING TO RISK
is scanty. There is an opportunity to study the dilated At the present time, all GERD patients are treated with a
distal esophagus at every upper endoscopy. Our hope is one-size-fits-all regimen of acid reducing drugs as needed
that knowledge of this new diagnostic test will stimulate to control symptoms. The ability of the new test to identify
esophagologists to produce data that will refine the criteria the minority of people who are at risk to progress to vCLE
for defining these various stages of GERD. allows a concentrated attack on this group. The interval
between the test and the predicted time of occurrence
PREDICTION OF PROGRESSION OF LOWER of a-LES damage sufficient to cause vCLE is likely to be
ESOPHAGEAL SPHINCTER DAMAGE many decades. This allows time to watch the patient, repeat
LES damage is irreversible. There is evidence that a-LES testing to verify findings, try a test of dietary control, and
damage progresses in an inexorably relentless manner. intervene to prevent progression of LES damage before
Progression is not impacted in any way by medical therapy. vCLE develops.
The cause of LES damage is an eating disorder that can
be described as LES-unfriendly. Once this eating habit is
established, the rate of progression of a-LES damage is WHAT NEEDS TO HAPPEN FOR THE NEW
likely to be linear over the long term. TEST TO WORK
The new test of a-LES damage provides a unique ability
to predict the status of the a-LES in the future if it is Like any new scientific test, there is much research,
assumed that progression of LES damage is linear over development, and testing that needs to be done to bring
the long term. Theoretically, if a-LES damage is measured the test to fruition. We can conceive some of these at
on two occasions separated by a significant interval, a this time. However, when the test is recognized as being
simple straight-line slope can be drawn that extrapolates valuable, novel developments yet not conceived are likely
the extent of damage back into the past and forward to emerge.
into the future.
The prediction is also possible with one measurement NEED TO REMOVE ERRORS IN INTERPRETATION
with the assumption that a-LES damage begins at an early The evidence base that cardiac epithelium is always a
age, say 15 years old, when a person’s adult eating habit metaplastic esophageal epithelium is powerful. The present
is established. There are now two points that permit the dogma that cardiac epithelium is the normal lining of the
slope of future a-LES damage to be drawn. proximal stomach must be eliminated.
Esophageal Mucosa in Health and Disease CHAPTER 3 29
The evidence that the true GEJ is the proximal limit less than perfect and variable effectiveness with significant
of gastric oxyntic epithelium and cannot be seen at complications.
endoscopy is powerful. The opinion-based definitions The new need is simpler. The objective is not to repair
that are universally used to define the GEJ at endoscopy or augment a defective LES; it is to prevent progression
and gross dissection (proximal limit or rugal folds and of a damaged LES with significant residual function that
end of the tubular esophagus) must be eliminated. is predicted to progress to severe damage in the future.
This is a far easier surgical problem. It is very likely that
NEED FOR A NEW BIOPSY DEVICE many of the techniques that currently have low success rates
Accurate measurement of the dilated distal esophagus is in augmenting a defective LES will have more success in
not possible with present biopsy forceps. A critical length preventing progression of damage in a partially damaged
of between 15 and 25 mm of a-LES damage differentiates LES.
mild GERD at its onset to severe GERD. Accuracy is vital.
With present biopsy forceps, multiple level biopsies distal
to the endoscopic GEJ need to be performed. These
CONCLUSION
are likely to be extremely difficult and time consuming, We envision the management of GERD in the future to
and therefore fraught with error. A simple new biopsy be very different with the availability of the new test. At
device that can remove a piece of mucosa measuring its ultimate end point of development, a biopsy device
25 mm long, 2 mm wide, and 1 mm deep should be easy will be used in a doctor’s office as a screening test in the
to produce. This will provide a mucosal biopsy sample population at around age 30 to 35 years, unless symptoms
that is equivalent to a section taken from a resected appear earlier. The test will be simple, inexpensive, and
specimen that we know can produce a measurement of relatively painless. It will identify those at risk in the
the dilated distal esophagus that is accurate to within future, with a timeline that shows the exact status of
1 µm. the a-LES at specific times in the life of the patient. We
can equate it to a Pap smear for cancer of the uterine
NEED FOR DATA ON ASYMPTOMATIC cervix.
The patient will have choices depending on the results
PERSONS AND GASTROESOPHAGEAL REFLUX of the test:
DISEASE PATIENTS 1. The test predicts that a-LES damage will remain within
A large database is the essential requirement for defining the reserve capacity, indicating that there will be no
criteria for length of a-LES damage that is associated with GERD. The person can be confident that esophageal
vCLE. This is the critical value, because preventing vCLE adenocarcinoma will not occur. No treatment will ever
is the objective and basis of preventing adenocarcinoma. be necessary. The senior author (PC) has had this
We have used the best available evidence at this time to assessment. He knows that at age 56 years, he had 4 mm
suggest that greater than 25 mm of a-LES damage is the of a-LES damage, measured in an endoscopic biopsy of
earliest point at which vCLE occurs. This may be optimistic; the normal SCJ. Biopsies greater than 5 mm, greater
it is possible that this number is closer to 20 mm. The than 10 mm, greater than 15 mm, and greater than
important thing is that examination of a sufficient number 20 mm distal to the SCJ consisted of normal gastric
of people will provide that number. oxyntic epithelium. By the algorithm that predicts
linear progression, he is predicted to have 8 mm of
NONENDOSCOPIC MEASUREMENT OF A-LOWER a-LES damage at age 97 years. He will always stay in
ESOPHAGEAL SPHINCTER DAMAGE the green zone of Table 3.1 with LES damage that is
At present, all biopsy methods are designed for use with within its reserve capacity. He will never develop GERD,
an endoscope. The need for endoscopy to perform the Barrett esophagus, or esophageal adenocarcinoma. No
new test will seriously limit its usage. Endoscopy will one else in the world has that certainty.
have low value in the assessment of GERD if the new 2. The test predicts the occurrence of LES damage that
test becomes a stand-alone diagnostic test. In that event, is likely to cause mild GERD during the expected life
it will be important to develop nonendoscopic methods span that can be easily controlled with acid reducing
of inserting the biopsy device to the appropriate location therapy without progression to severe GERD, vCLE,
and orientation to allow the required biopsy to be taken. or cancer. This person can opt to let GERD arise and
If such a method can be developed that is safe, quick, and be treated with PPIs for the entire lifetime or undergo
inexpensive, with the ability to be done in the doctor’s the procedure to protect the LES and prevent GERD.
office without the need for sedation, the scope of the test 3. The test predicts future severe LES damage with a high
can be expanded dramatically to the general population, risk of refractory GERD, vCLE, and adenocarcinoma.
allowing screening and early diagnosis. These persons can develop a long-term treatment
protocol with repeat testing, dietary control to slow
progression of a-LES damage, and timely intervention
NEW EFFECTIVE METHOD FOR PREVENTING with a simple procedure well before the predicted time
PROGRESSION OF A-LOWER ESOPHAGEAL of LES damage sufficient to cause severe GERD and
SPHINCTER DAMAGE vCLE.
Many procedures are presently available to repair or In all scenarios, knowledge of future a-LES damage
augment a defective LES. Some are done by endoscopy allows for stratification of treatment according to the risk
and others require laparoscopic surgery. These all have of future cellular complications of GERD.
30 SECTION I Esophagus and Hernia
We can only guess whether this vision of a world without 14. Chandrasoma P, Wijetunge S, Ma Y, DeMeester S, Hagen J, DeMeester
GERD and esophageal adenocarcinoma will come true. T. The dilated distal esophagus: a new entity that is the pathologic
basis of early gastroesophageal reflux disease. Am J Surg Pathol.
What is important is that this new method provides us 2011;35:1873-1881.
with the possibility that there may be a way to this goal. 15. Korn O, Csendes A, Burdiles P, Braghetto I, Stein HJ. Anatomic
It will change the perspective toward a belief that GERD, dilatation of the cardia and competence of the lower esophageal
Barrett esophagus, and esophageal adenocarcinoma are sphincter: a clinical and experimental study. J Gastrointest Surg.
2000;4:398-406.
preventable. This is better than the present nihilistic 16. Curcic J, Roy S, Tech M, et al. Abnormal structure and function of
attitude, where we do nothing and simply hope that our the esophagogastric junction and proximal stomach in gastroesopha-
patients will not progress to severe GERD that is refractory geal reflux disease. Am J Gastroenterol. 2014;109:658-666.
to therapy or, worse, be complicated by adenocarcinoma. 17. Hayward J. The lower end of the oesophagus. Thorax. 1961;16:36-41.
18. Chandrasoma PT. Histologic definition of gastro-esophageal reflux
disease. Curr Opin Gastroenterol. 2013;29:460-467.
REFERENCES 19. Ayazi S, Tamhankar A, DeMeester SR, et al. The impact of gastric
distension on the lower esophageal sphincter and its exposure to
1. Vakil N, van Zanten SV, Kahrilas P, Dent J, Jones B, The Global acid gastric juice. Ann Surg. 2010;252:57-62.
Consensus Group. The Montreal definition and classification of 20. Robertson EV, Derakhshan MH, Wirz AA, et al. Central obesity in
gastroesophageal reflux disease: a global evidence-based consensus. asymptomatic volunteers is associated with increased intrasphincteric
Am J Gastroenterol. 2006;101:1900-1920. acid reflux and lengthening of the cardiac mucosa. Gastroenterology.
2. Kahrilas PJ, Shaheen NJ, Vaezi MF. American Gastroenterological 2013;145:730-739.
Association medical position statement on the management of 21. Clarke AT, Wirz AA, Manning JJ, Ballantyne SA, Alcorn DJ, McColl
gastroesophageal reflux disease. Gastroenterology. 2008;135:1383-1391. KE. Severe reflux disease is associated with an enlarged unbuffered
3. Chandrasoma PT, Der R, Ma Y, Peters J, DeMeester T. Histologic proximal gastric acid pocket. Gut. 2008;57:292-297.
classification of patients based on mapping biopsies of the gastro- 22. Kahrilas PJ, Shi G, Manka M, Joehl RJ. Increased frequency of
esophageal junction. Am J Surg Pathol. 2003;27:929-936. transient lower esophageal sphincter relaxation induced by gastric
4. Spechler SJ, Sharma P, Souza RF, Inadomi JM, Shaheen NJ. American distension in reflux patients with hiatal hernia. Gastroenterology.
Gastroenterological Association medical position statement on the 2000;118:688-695.
management of Barrett’s esophagus. Gastroenterology. 2011;140:1084- 23. Chandrasoma P. Controversies of the cardiac mucosa and Barrett’s
1091. esophagus. Histopathology. 2005;46:361-373.
5. Toghanian S, Wahlqvist P, Johnson DA, Bolge SC, Liljas B. The 24. Chandrasoma PT, Der R, Ma Y, Dalton P, Taira M. Histology of the
burden of disrupting gastro-esophageal disease: a database study in gastroesophageal junction: an autopsy study. Am J Surg Pathol.
US and European cohorts. Clin Drug Investig. 2010;30:167-178. 2000;24:402-409.
6. Pohl H, Sirovich B, Welch HG. Esophageal adenocarcinoma incidence: 25. McClave SA, Boyce HW Jr, Gottfried MR. Early diagnosis of columnar
are we reaching the peak? Cancer Epidemiol Biomarkers Prev. lined esophagus: a new endoscopic diagnostic criterion. Gastrointest
2010;19:1468-1470. Endosc. 1987;33:413-416.
7. Malfertheiner P, Nocon M, Vieth M, et al. Evolution of gastro- 26. Sharma P, McQuaid K, Dent J, et al. A critical review of the diagnosis
oesophageal reflux disease over 5 years under routine medical and management of Barrett’s esophagus: the AGA Chicago Workshop.
care—the ProGERD study. Aliment Pharmacol Ther. 2012;35:154-164. Gastroenterology. 2004;127:310-330.
8. Fitzgerald RC, di Pietro M, Ragunath K, et al. British Society of 27. Chandrasoma P, Makarewicz K, Wickramasinghe K, Ma YL, DeMeester
Gastroenterology guidelines on the diagnosis and management of TR. A proposal for a new validated histologic definition of the
Barrett’s oesophagus. Gut. 2014;63:7-42. gastroesophageal junction. Hum Pathol. 2006;37:40-47.
9. Leodolter A, Nocon M, Vieth M, et al. Progression of specialized 28. Sarbia M, Donner A, Gabbert HE. Histopathology of the gastro-
intestinal metaplasia at the cardia to macroscopically evident Barrett’s esophageal junction. A study on 36 operation specimens. Am J Surg
esophagus: an entity of concern in the Pro-GERD study. Scand J Pathol. 2002;26:1207-1212.
Gastroenterol. 2012;47:1429-1435. 29. Park YS, Park HJ, Kang GH, Kim CJ, Chi JG. Histology of gastro-
10. Nason KS, Wichienkuer PP, Awais O, et al. Gastroesophageal reflux esophageal junction in fetal and pediatric autopsy. Arch Pathol Lab
disease symptom severity, proton pump inhibitor use, and esophageal Med. 2003;127:451-455.
carcinogenesis. Arch Surg. 2011;146:851-858. 30. Kilgore SP, Ormsby AH, Gramlich TL, et al. The gastric cardia: fact
11. Stein HJ, Liebermann-Meffert D, DeMeester TR, Siewert JR. Three- or fiction? Am J Gastroenterol. 2000;95:921-924.
dimensional pressure image and muscular structure of the human 31. Ringhofer C, Lenglinger J, Izay B, et al. Histopathology of the
lower esophageal sphincter. Surgery. 1995;117:692-698. endoscopic esophagogastric junction in patients with gastroesophageal
12. Bonavina L, Evander A, DeMeester TR, et al. Length of the distal reflux disease. Wien Klin Wochenschr. 2008;120:350-359.
esophageal sphincter and competency of the cardia. Am J Surg.
1986;151:25-34.
13. Zaninotto G, DeMeester TR, Schwizer W, Johansson KE, Cheng SC.
The lower esophageal sphincter in health and disease. Am J Surg.
1988;155:104-111.
CHAPTER
Relevant Anatomic Relations of the Esophagus
Jules Lin
4
T
he primary purpose of the esophagus is to transport cricoid cartilage as two muscle bundles that converge 3 cm
food from the mouth to the stomach, and the distal to the cricopharyngeus muscle, leaving a V-shaped
esophagus has no digestive or absorptive function. weakened area posteriorly covered only with circular
The esophagus is a muscular tube that starts at the inferior muscle fibers, known as Laimer triangle (Fig. 4.4). The
border of the cricoid cartilage in the neck and traverses cricopharyngeal muscle and up to 2 cm of the cervical
the chest, ending as it enters the stomach in the upper esophagus are composed mostly of striated muscle.4,5 There
abdomen. Esophageal surgeons need to be familiar with is a gradual transition from striated muscle, which is only
the anatomy and clinical importance of the relationship present in the upper 40% of the esophagus, to smooth
of the esophagus to surrounding structures in all three muscle (Fig. 4.5). As a result, most dysmotility disorders,
areas of the body. which involve smooth muscle, predominantly affect the
The esophagus is anchored superiorly to the cricoid distal two-thirds of the esophagus.
cartilage and inferiorly to the diaphragm. Its length from
the cricoid cartilage at C6 to the gastric orifice at T11 CERVICAL ESOPHAGUS
ranges from 22 to 28 cm (Fig. 4.1).1 The length of the
esophagus varies with the height of the patient. The The cervical esophagus shifts to the left at the base of the
esophageal bed lies in the posterior mediastinum along neck, so the best approach to the cervical esophagus is a
the ventral surface of the vertebra. This is the shortest left neck incision (e.g., during the cervical anastomosis
distance for reconstruction after esophagectomy and for a transhiatal esophagectomy or resection of a Zenker
measures approximately 30 cm. When the orthotopic route diverticulum). The cervical esophagus is 5 cm in length
is not available due to previous surgery, the retrosternal and extends from the cricoid cartilage at the C6 verte-
(32 cm) or subcutaneous (34 cm) routes are alternatives.2,3 bra to the suprasternal notch anteriorly and the T1-T2
On endoscopy, the upper esophageal sphincter is interspace posteriorly. The esophagus is surrounded by
approximately 15 cm from the incisors, the carina at loose fibroareolar tissue, and unlike the remainder of
25 cm, and the lower esophageal sphincter at 38 to 40 cm the gastrointestinal tract, there is no serosal lining or
in men and 36 to 38 cm in women. When planning the mesentery.
surgical approach to the esophagus, it is useful to consider The pretracheal fascia anteriorly, prevertebral fascia
the location of the tumor relative to the carina at 25 cm in posteriorly, and carotid sheaths laterally create two cervical
deciding whether to perform a left or a right thoracotomy compartments, the paraesophageal space anteriorly, and
to avoid the aortic arch, which blocks access to the upper the retroesophageal space posteriorly. The pretracheal
thoracic esophagus on the left. fascia extends to the pericardium, while the prevertebral
While the cervical esophagus is flattened by surrounding fascia extends to the diaphragm, connecting compart-
structures, the thoracic esophagus is more round due ments in the neck and chest, allowing infections from
to negative intrathoracic pressure. Once the esophagus tonsillar or dental abscesses to spread quickly into the
enters the abdomen, it becomes more flattened again as mediastinum. A descending infection can rapidly lead
a result of positive intraabdominal pressure. The overall to potentially fatal necrotizing mediastinitis, emphasizing
diameter of the esophagus is approximately 2.5 cm.1 the importance of early surgical drainage. Iatrogenic
There are three anatomic narrowings of the esophagus perforations most commonly occur above the narrowing
(Fig. 4.2). The narrowest part of the gastrointestinal of the cricopharyngeal muscle, resulting in a posterior
tract is at the cricopharyngeal constriction at the upper pharyngeal perforation (see Fig. 4.4).6
esophageal sphincter, where the esophagus is 1.5 cm in The arterial blood supply for the cervical esophagus
diameter. In the superior mediastinum, the aortic arch, originates from the right and left superior and inferior
left atrium, and left mainstem bronchus compress the thyroid arteries (Fig. 4.6). There are also smaller branches
left anterolateral aspect of the esophagus around 22 cm from the common carotid and subclavian arteries. Venous
from the incisors. The third area of narrowing is at the drainage is through the inferior thyroid vein. The middle
lower esophageal sphincter. These areas are important thyroid vein crosses from the internal jugular vein to the
since swallowed foreign bodies often lodge in these areas. thyroid gland, and often is divided to improve exposure
The esophageal wall is composed of four layers, of the cervical esophagus. Innervation of the cervical
including an inner squamous mucosal layer, submucosa, esophagus is from the recurrent laryngeal nerves and the
muscularis propria, and adventitia. The mucosal layer sympathetic chains.
consists of the squamous epithelium, lamina propria, and
muscularis mucosa, while the muscularis propria consists UPPER ESOPHAGEAL SPHINCTER
of an inner circular and outer longitudinal layer of muscle The posterior pharynx consists of three bilateral constric-
(Fig. 4.3). The longitudinal layer originates from the tor muscles originating from the base of the skull, the
31
Relevant Anatomic Relations of the Esophagus CHAPTER 4 31.e1
ABSTRACT
The primary purpose of the esophagus is to transport
food from the mouth to the stomach; the esophagus
has no digestive or absorptive function. The esophagus
is a muscular tube that starts at the inferior border of
the cricoid cartilage in the neck and traverses the chest,
ending as it enters the stomach in the upper abdomen.
While the esophagus seems like a simple organ at first
glance, esophageal surgeons must be familiar with the
complex anatomy and function of the upper and lower
esophageal sphincters, the extensive lymphatic and vascular
submucosal networks, and the clinical importance of the
relationship of the esophagus with surrounding structures
in all three areas of the body for effective surgical planning
and the prevention of potential complications.
KEYWORDS
Esophageal anatomy, upper esophageal sphincter, recurrent
laryngeal nerves, esophageal lymphatic drainage, lower
esophageal sphincter
32 SECTION I Esophagus and Hernia
15 cm
Incisors
Anatomy Function Surgery
Cricoid cartilage
Vertebra
C VI–Th I 1 1 Cervical UES Cervical
Incisura Thoracic
Jugularis sterni proximal
Thoracic Tubular
18 – 22 cm
2 2 Thoracic
Th I–X distal
3
Diaphragm 3
3 – 6 cm
Total length
38–48 cm
FIGURE 4.1 Division of the esophagus into cervical, thoracic, and abdominal segments (Anatomy). The approximate length of each
segment is shown with its relationship to the cervical (C) and thoracic (Th) vertebrae. The esophagus can also be subdivided according
to functions of the upper or lower esophageal sphincters (Function) or based on cranial or caudal lymphatic drainage (Surgery).
LES, Lower esophageal sphincter; UES, upper esophageal sphincter.
Muscularis
mucosae
Lumen
Circular
muscle
Longitudinal
muscle
Type of muscle
Striated
100 %
Smooth
A
FIGURE 4.5 Transverse (A) and longitudinal (B) histologic sections of the esophageal muscle 4 cm superior to the tracheal bifurcation,
showing the transition between striated and smooth muscle. There are individual striated muscle fibers among the smooth muscle fibers.
The diagram illustrates the gradual transition from striated to smooth muscle in the proximal to distal esophagus. (Specimen and photo
courtesy D. Liebermann-Meffert, Geissdorfer, and Winter, Munich.)
esophagus can be completely mobilized from the stomach aortic arch are to the left of the esophagus (Fig. 4.11).
to the aortic arch, yet continue to be well perfused.11–13 The descending aorta turns posteriorly and continues to
However, care should be taken if the inferior thyroid the left of the esophagus until T8, where the esophagus
artery has been ligated during a previous thyroidectomy. moves anterior to the aorta.
The arterial vessels supplying the esophagus branch into In the lower thorax, the pericardium and left atrium
small, fine vessels at a distance from the esophageal wall, lie anterior to the esophagus (Fig. 4.12). Invasion of the
and bleeding can be controlled by packing the medias- heart or the aorta in locally advanced esophageal cancer
tinum, allowing for blunt dissection during a transhiatal is unresectable and can be evaluated on a chest computed
esophagectomy.14 Venous drainage is through the azygos tomography with contrast or endoscopic ultrasound. Direct
vein, with some drainage through the hemiazygos and invasion of the aorta can also lead to an aortoesophageal
bronchial veins. fistula. Hematemesis in a patient with esophageal cancer
The thoracic esophagus is bordered by the mediastinal can represent sentinel bleeding from an aortoesophageal
pleura on either side. The esophagus is surrounded by fistula, which must be recognized early to prevent fatal
loose fibroareolar tissue. This loose investing tissue, with exsanguinating hemorrhage. However, if the tumor
no dense fibrous attachments in the chest, allows the just abuts these structures, often there is no invasion at
esophagus to be bluntly dissected during a transhiatal intraoperative exploration. Posteriorly, the esophagus is
esophagectomy. During transhiatal or laparoscopic dis- adjacent to the spine from the thoracic inlet to T8, where
section of the esophagus, it is important to recognize the esophagus moves anteriorly to enter the esophageal
entry into either pleural space. A chest tube is placed for hiatus. The left lateral distal esophagus is covered only
drainage of blood during an esophagectomy, and a tension with mediastinal pleura. As a result of the relative weakness
pneumothorax must be recognized during laparoscopy. in this area, spontaneous Boerhaave perforation occurs
In the upper thorax, the trachea is anterior, the azygos is most commonly along the left lateral aspect of the distal
on the right (Fig. 4.10), and the left subclavian artery and esophagus.
Relevant Anatomic Relations of the Esophagus CHAPTER 4 35
5 3 87 12 44 3 5 6
5 3 8 1 2 3 5
1 Esophagus
2 Trachea
3 Common carotid artery
4 Thyroid artery
5 Jugular vein
6 Phrenic nerve
7 Recurrent nerve
T1 8 Thyroid gland
T1 T1 1st thoracic vertebra
FIGURE 4.7 Topographic anatomy of the cervical esophagus at the level of the thyroid gland with a corresponding computed tomography
image. (Modified from Koritké H, Sick J. Atlas of Sectional Human Anatomy. 2nd ed. Baltimore: Urban & Schwarzenberg; 1988.)
36 SECTION I Esophagus and Hernia
A. bronchialis
with Bronchial
esophageal artery
branches
Previous site of
10 cm diaphragm
A. gastr. sin.
AORTA
Aorta
FIGURE 4.9 Arterial cast showing the vascular supply to the middle and lower esophagus. There is a rich intramural network of small
vessels, allowing the mobilized esophagus to remain well perfused, even after extensive mobilization. (Photo courtesy D. Liebermann-
Meffert, Munich.)
10 7 5 7 7
3 6 9 2 5 1 4 3 8 69 2 1 4 8
1 Esophagus
2 Trachea
3 Lung
4 Aortic arch
5 Brachiocephalic vein
6 Azygos vein
7 Vagus nerve
8 Phrenic nerve
9 Lymph nodes
T4
T4 T4 4th thoracic vertebra
FIGURE 4.11 Topographic anatomy of the upper thoracic esophagus at the level of the aortic arch, along with a corresponding computed
tomography image. (Modified from Koritké H, Sick J. Atlas of Sectional Human Anatomy. 2nd ed. Baltimore: Urban & Schwarzenberg;
1988.)
11 8 4 5 19 10 8
5 1 10
1 Esophagus
2 Lung
3 Aorta
4 Coronary artery
5 Inferior vena cava
6 Azygos vein
7 Hemiazygos vein
T9 8 Phrenic nerve
9 Vagus nerve
10 Heart, left ventricle
T9 11 Diaphragmatic pleura
T9 9th thoracic vertebra
2 6 3 2
2 9 6 73 2
FIGURE 4.12 Topographic anatomy of the lower thoracic esophagus at the level of the left atrium, along with a corresponding computed
tomography image. (Modified from Koritké H, Sick J. Atlas of Sectional Human Anatomy. 2nd ed. Baltimore: Urban & Schwarzenberg;
1988.)
during an Ivor-Lewis esophagectomy to improve exposure become blocked by tumor, and the pattern of metastatic
of the upper thoracic esophagus. spread may not follow normal lymphatic drainage, allowing
the cephalad spread of distal tumors. Collecting lymphatic
LYMPHATIC DRAINAGE AND THORACIC DUCT ducts also originate intermittently from the submucosal
Understanding esophageal lymphatic drainage is important network. These collecting ducts connect directly to the
in tumor staging and treatment planning. A rich submu- thoracic duct in 40% of patients.17
cosal lymphatic network transports lymph to subadventitial The importance of the submucosal lymphatic network
lymphatics. The vast network of lymphatics not only drains in nodal metastases is highlighted by the association of
fluid, debris, and bacteria, but also allows the spread the depth of tumor invasion (T stage) with the incidence
of tumor cells. While there are few lymphatics in the of nodal disease.18,19 T1a tumors localized to the mucosa
superficial mucosa, once tumors invade the submucosa, the with limited lymphatics have a low incidence of nodal
submucosal lymphatic plexus allows tumor cells to spread spread of 2.6%, while T1b tumors crossing the muscularis
widely (Fig. 4.13). Sakata first noted that the esophageal mucosa into the submucosa have a 22.2% incidence.20
lymphatic drainage was longitudinal and intramural.15 Nodal involvement increases to 43.2% in T2, 77.2% in
Lehnert believed that these longitudinal lymphatic chan- T3, and 66.7% in T4 esophageal tumors.
nels within the submucosa allow lymph to flow most easily The main lymphatic vessel is the thoracic duct, which
up and down the esophagus, which is consistent with the begins in the abdomen as the cisterna chyli at T12 and
longitudinal spread of esophageal cancer sometimes far passes through the aortic hiatus along with the aorta
away from the primary tumor.16 and the azygos and hemiazygos veins. The duct lies on
While lymph from the esophagus above the tracheal the anterior surface of the vertebra and is posterior to
bifurcation drains superiorly to paratracheal, subcarinal, the esophagus between the azygos on the right and the
and paraesophageal lymph nodes, lymph below the carina descending aorta on the left. To perform a mass ligation
drains inferiorly toward the lower mediastinal, celiac, and for a chylothorax, the tissue between the azygos vein and
left gastric lymph nodes (Fig. 4.14).16 However, the direc- aorta is ligated just above the diaphragm. In greater than
tion of lymphatic flow may change when the lymphatics 50% of patients, the duct turns behind the left mainstem
L. mucosa
T. submucosa
with lymph channels
L. muscularis
TUMOR
L. mucosa
T. submucosa
L. muscularis
FIGURE 4.13 The diagram illustrates the extensive submucosal lymphatics that allow tumor cells to travel longitudinally a distance away
from the primary tumor. When lymphatics become blocked, tumor cells can travel retrograde, demonstrating how cells from a distal
esophageal tumor can be found in the proximal esophagus. L, Lamina; T, tunica. (From Liebermann-Meffert D, Duranceau A, Stein HJ.
Anatomy and embryology. In: Orringer MB, Heitmiller R, eds. The Esophagus. Vol 1. In: Zuidema GD, Yeo CJ, series eds. Shackelford’s
Surgery of the Alimentary Tract. 5th ed. Philadelphia: Saunders; 2002:3.)
Relevant Anatomic Relations of the Esophagus CHAPTER 4 39
Concept of lymphatic drainage and esophageal cancer.23,24 The proximity of the esophagus
to the left atrium also allows optimal transesophageal
echocardiographic imaging of the mitral valve.
Compartment Flow direction
ABDOMINAL ESOPHAGUS
Cervical esophagus The abdominal esophagus and proximal stomach are
Neck
approached through a laparotomy or laparoscopy. Nor-
mally, the abdominal esophagus is 3 to 6 cm in length.1
The abdominal esophagus angulates to the left after
Thoracic esophagus
passing through the diaphragmatic hiatus, and care must
be taken to carefully guide an esophageal bougie during
Partition Line
a Nissen fundoplication or a rigid esophagoscope through
this area to avoid perforation (see Fig. 4.1). The distal
Thorax
esophagus is the second most common site of perforation
after the pharynx, just proximal to the cricopharyngeus.
The esophagus ends as it enters the cardia along the
lesser curvature of the stomach. The abdominal esophagus
receives blood supply from small arteries derived from
the left gastric artery and the inferior phrenic artery (see
Terminal esophagus Fig. 4.6). Venous drainage is through the left gastric vein.
Intraesophageal veins are located in a subepithelial plexus
located in the lamina propria. In the distal esophagus, this
venous plexus connects with the portal system, leading
Abdomen to esophageal varices with elevated portal pressures.25
Regional lymph nodes in the abdomen include stations
15 (diaphragmatic), 16 (paracardial), 17 (left gastric),
Celiac lymph nodes 18 (common hepatic), 19 (splenic), and 20 (celiac).26
cysterna chyli The esophagus passes through the esophageal hiatus at
the level of the 10th thoracic vertebra (Fig. 4.15). Distal
esophageal tumors can directly invade the crura, which
FIGURE 4.14 Lymph from the esophagus above the tracheal
can be resected en bloc with the esophageal tumor. The
bifurcation (partition line) drains mostly toward the neck, while
crura originate from the anterior aspect of the first three
lymph below the tracheal bifurcation drains toward the lower to four lumbar vertebrae. As the esophagus passes through
mediastinum and the celiac axis. Lymph at the tracheal bifurcation the hiatus, the left lobe of the liver lies anteriorly, and
can flow in either direction. When lymphatics become blocked, the triangular ligament can be divided to help expose the
tumor cells can travel in the opposite direction. (From Liebermann- hiatus (Fig. 4.16). The aorta is posterior to the esophagus
Meffert D, Duranceau A, Stein HJ. Anatomy and embryology. In: and passes through its own hiatus in the diaphragm. The
Orringer MB, Heitmiller R, eds. The Esophagus. Vol 1. In: Zuidema aortic hiatus is bordered by the median arcuate ligament
GD, Yeo CJ, series eds. Shackelford’s Surgery of the Alimentary anteriorly, which is used to anchor sutures in performing
Tract. 5th ed. Philadelphia: Saunders; 2002:3.) a Hill repair for gastroesophageal reflux.27,28 The inferior
vena cava lies lateral and inferior to the right crus.
VAGUS NERVES
bronchus, at the level of the fifth and sixth thoracic The anterior and posterior vagal nerves run along the
vertebrae, to eventually empty into the junction of the left distal esophagus through the esophageal hiatus (see
subclavian and jugular veins.21 There are also a number of Fig. 4.8). The anterior nerve is usually located adjacent
possible anatomic variations.22 The thoracic duct is adjacent to the esophageal wall, while the posterior nerve can
to the thoracic esophagus and is damaged in 1% to 2% of be farther away. The anterior nerve should be carefully
esophagectomies, resulting in a chylothorax.10,14 If there identified after the phrenoesophageal membrane is opened
is suspicion for a thoracic duct injury postoperatively, and preserved during a Heller myotomy, continuing the
with high pink, clear chest tube output, which becomes myotomy underneath the nerve when necessary. Both
milky once fat is added to the patient’s diet, the thoracic nerves should be identified during a Nissen fundoplication,
duct should either be surgically ligated or embolized by especially with a hiatal hernia repair where the nerves
interventional radiology to prevent complications associ- can become involved in the hernia sac and inadvertently
ated with the loss of the protein and leukocyte-rich lymph. divided. If the nerves are injured, approximately 15% of
The thoracic esophagus is in contact with lymph nodes patients will develop delayed gastric emptying.29
from several stations, allowing endoscopic ultrasound
imaging and fine-needle aspiration of stations 2 (upper DIAPHRAGM
paratracheal), 3P (posterior mediastinal), 4 (lower paratra- The left and right edges of the esophageal hiatus most
cheal), 7 (subcarinal), 8 (paraesophageal), and 9 (inferior commonly originate from the deep and superficial layers
pulmonary ligament) lymph nodes in the staging of lung of the right crus, while the left crus forms one edge of
40 SECTION I Esophagus and Hernia
Traumatic laceration
Central tendon
Site of esophageal
hiatus hernia
Site of Bochdalek hernia
Aortic hiatus
FIGURE 4.15 The esophagus passes through the diaphragm at the esophageal hiatus, while the aorta passes through its own hiatus
posteriorly. The inferior vena cava is lateral to the right crus. The location of various diaphragmatic hernias is shown. (Diagram courtesy
D. Liebermann-Meffert. In: Shackelford’s Surgery of the Alimentary Tract. 3rd ed. Philadelphia: Saunders; 1991.)
11 10 12
6 6 10 1 7 11 8 10 2
4 1 11 2
1 Esophagus
2 Lung
3 Aorta
4 Azygos vein
5 Inferior vena cava
6 Hepatic vein
7 Vagus nerve
8 Heart, ventricles
9 Stomach
10 Diaphragm
11 Liver
T10 T10 T10 10th thoracic vertebra
5 3 9
6 5 4 3 13 10 10 9
FIGURE 4.16 Topographic anatomy of the abdominal esophagus along with a corresponding computed tomography image. (Modified
from Koritké H, Sick J. Atlas of Sectional Human Anatomy. 2nd ed. Baltimore: Urban & Schwarzenberg; 1988.)
the aortic hiatus. The phrenoesophageal membrane, phrenoesophageal membrane allows the esophagus to
also known as Laimer or Allison membrane, arises from move dynamically relative to the diaphragm.
the subdiaphragmatic fascia and attaches the esophagus Abnormalities in the phrenoesophageal membrane may
to the diaphragm (Fig. 4.17). The phrenoesophageal lead to hiatal hernia formation. The elastic fibers of the
membrane is composed of two sheaths. One sheath passes phrenoesophageal membrane are replaced by inelastic
superior from the diaphragm for 2 to 4 cm along the distal fibers with aging.30 Eliska et al. believe that abnormal
esophagus, and its fibers insert into the submucosa of the anchorage of the phrenoesophageal membrane, along with
esophagus.4,30 The other sheath extends inferiorly across increasing adipose tissue between the membrane and the
the cardia and blends into the gastric serosa, muscle, cardia, may lead to the development of a hiatal hernia.30
dorsal mesentery, and the gastrohepatic ligament. The Mittal reported that the phrenoesophageal membrane
Relevant Anatomic Relations of the Esophagus CHAPTER 4 41
Diaphragm
Fundus GEJ
Clasps
of LES
Stomach
FIGURE 4.18 The dry muscle fiber specimen shows the muscle orientation at the gastroesophageal junction (GEJ) with the outer,
longitudinal muscle (left) and the inner, circular fibers (right). Along the lesser curvature, the circular fibers become short muscular clasps,
while the circular fibers become oblique gastric sling fibers along the gastric curve at the angle of His, forming a functional and anatomic
lower esophageal sphincter (LES). (Specimens from the collection of D. Liebermann-Meffert.)
42 SECTION I Esophagus and Hernia
mm
Constrictor muscle
20
UES GC
50 0
SM-M
MP
30
mm
Longitudinal Circular
30 20 20
esophageal esophageal
muscle muscle 20
30 0
0
20 20
20 mm
LES mm
0
Diaphragm Gastric LC
sling AW 20
fibers
FIGURE 4.20 The illustration shows the correlation between the
mean (n = 15) radial muscle thickness at the posterior gastric
Clasps
of LES wall (PW), greater curvature (GC), anterior gastric wall (AW), and
lesser curvature (LC) (left) and a three-dimensional manometric
pressure image (right) at the gastroesophageal junction. There
is asymmetry of the radial and axial muscular thickness and
manometric pressure profiles. (From Liebermann-Meffert D,
Allgöwer M, Schmid P, Blum AL. Muscular equivalent of the lower
esophageal sphincter. Gastroenterology. 1979;76:31; and Stein
HJ, Liebermann-Meffert D, DeMeester TR, Siewert JR. Three-
Longitudinal Circular
dimensional pressure image and muscular structure of the human
gastric muscle gastric muscle
lower esophageal sphincter. Surgery. 1995;117:692.)
FIGURE 4.19 The outer, longitudinal (left) and inner, circular (right)
muscle layers are shown. The upper (UES) and lower (LES)
esophageal sphincters are composed of inner circular fibers. REFERENCES
(Diagram courtesy D. Liebermann-Meffert, Munich.)
1. Liebermann-Meffert D. Anatomy, embryology, and histology. In:
Pearson FG, Cooper JD, Delauriers J, et al., eds. Esophageal Surgery.
Philadelphia: WB Saunders; 2000.
to the diaphragmatic hiatus are also felt to contribute to 2. Goldsmith HS, Akiyama H. A comparative study of Japanese and
the antireflux mechanism at the gastroesophageal junc- American gastric dimensions. Ann Surg. 1979;190:690-693.
tion. The lower esophageal sphincter provides a barrier 3. Ngan SY, Wong J. Lengths of different routes for esophageal replace-
to gastroesophageal reflux and relaxes 3 seconds after ment. J Thorac Cardiovasc Surg. 1986;91:790-792.
4. Liebermann-Meffert D, Duranceau A, Stein HJ. Anatomy and
swallowing is initiated allowing the food bolus to enter embryology. In: Orringer MB, Heitmiller R, eds. The Esophagus. Vol.
the stomach.36,37 Some have suggested that an esophago- 1. 5th ed. Philadelphia: Saunders; 2002.
myotomy for achalasia should be performed between the 5. Liebermann-Meffert D, Geissdörfer K. Is the transition of striated
short clasp fibers and the oblique gastric sling fibers to into smooth muscle precisely known? In: Giuli R, McCallum RW,
Skinner DB, eds. Primary Motility Disorders of the Esophagus: 450
maintain the strength of the sling mechanism.38–40 Questions—450 Answers. Paris: Libbey Eurotext; 1991.
6. Savary M, Miller G. The Esophagus: Handbook and Atlas of Endoscopy.
CONCLUSION Solothurn, Switzerland: Gassman; 1978.
7. Winans CS. The pharyngoesophageal closure mechanism: a mano-
The esophagus is a muscular tube that starts at the cricoid metric study. Gastroenterology. 1972;63:768-777.
8. Lerche W. The Esophagus and Pharynx in Action. Springfield, IL:
cartilage in the neck and traverses the chest before ending Charles C Thomas; 1950.
in the stomach in the upper abdomen. The primary 9. Toniato A, Rubello D, Merante Boschin I. Nonrecurrent and ipsilateral
purpose of the esophagus is to transport food from the recurrent inferior laryngeal nerves. Minerva Chir. 2012;67:286-287.
mouth to the stomach. While the esophagus seems like 10. Orringer MB, Marshall B, Chang AC, et al. Two thousand transhiatal
esophagectomies: changing trends, lessons learned. Ann Surg.
a simple organ at first glance, esophageal surgeons must 2007;246:363-372; discussion 72-74.
be familiar with the complex anatomy and function of 11. Liebermann-Meffert D, Siewert JR. Arterial anatomy of the esophagus:
the upper and lower esophageal sphincters, the extensive a review of literature with brief comments on clinical aspects. Gullet.
lymphatic and vascular submucosal networks, and the 1992;2:3.
clinical importance of the relationship of the esophagus 12. Liebermann-Meffert DM, Meier R, Siewert JR. Vascular anatomy
of the gastric tube used for esophageal reconstruction. Ann Thorac
to surrounding structures in all three areas of the body Surg. 1992;54:1110-1115.
for effective surgical planning and the prevention of 13. Liebermann-Meffert DM, Luescher U, Neff U, Rüedi TP, All-
potential complications. göwer M. Esophagectomy without thoracotomy: is there a risk of
Relevant Anatomic Relations of the Esophagus CHAPTER 4 43
intramediastinal bleeding? A study on blood supply of the esophagus. 27. Aye RW, Rehse D, Blitz M, Kraemer SJ, Hill LD. The Hill antireflux
Ann Surg. 1987;206:184-192. repair at 5 institutions over 25 years. Am J Surg. 2011;201:599-604.
14. Orringer MB, Orringer JS. Esophagectomy without thoracotomy: a 28. Lindner HH, Kemprud E. A clinicoanatomical study of the arcuate
dangerous operation? J Thorac Cardiovasc Surg. 1983;85:72-80. ligament of the diaphragm. Arch Surg. 1971;103:600-605.
15. Sakata K. Ueber die Lymphgefasse des Oesophagus und uber seine 29. Swanson EW, Swanson SJ, Swanson RS. Endoscopic pyloric balloon
regionaren Lymphdrusen mit Berucksichtigung der Verbrietung des dilatation obviates the need for pyloroplasty at esophagectomy. Surg
Carcinoms. Mitt Grenzgeb Med. 1903;11:629-656. Endosc. 2012;26:2023-2028.
16. Akiyama H. Surgery for carcinoma of the esophagus. Curr Probl Surg. 30. Eliska O. Phreno-oesophageal membrane and its role in the develop-
1980;17:53-120. ment of hiatal hernia. Acta Anat (Basel). 1973;86:137-150.
17. Murakami G, Sato I, Shimada K, Dong C, Kato Y, Imazeki T. Direct 31. Mittal RK. Current concepts of the antireflux barrier. Gastroenterol
lymphatic drainage from the esophagus into the thoracic duct. Surg Clin North Am. 1990;19:501-516.
Radiol Anat. 1994;16:399-407. 32. Liebermann-Meffert D, Heberer M. Allgöwer M. The muscular
18. Matsubara T, Ueda M, Abe T, Akimori T, Kokudo N, Takahashi T. counterpart of the lower esophageal sphincter. In: DeMeester TR,
Unique distribution patterns of metastatic lymph nodes in patients Skinner DB, eds. Esophageal Disorders: Pathology and Therapy. New
with superficial carcinoma of the thoracic oesophagus. Br J Surg. York: Raven Press; 1985.
1999;86:669-673. 33. Liebermann-Meffert D, Allgower M, Schmid P, Blum AL. Muscular
19. Nigro JJ, DeMeester SR, Hagen JA, et al. Node status in transmural equivalent of the lower esophageal sphincter. Gastroenterology.
esophageal adenocarcinoma and outcome after en bloc esophagec- 1979;76:31-38.
tomy. J Thorac Cardiovasc Surg. 1999;117:960-968. 34. Code CF, Fyke FE Jr, Schlegel JF. The gastroesophageal sphincter
20. Rice TW, Zuccaro G Jr, Adelstein DJ, Rybicki LA, Blackstone EH, in healthy human beings. Gastroenterologia. 1956;86:135-150.
Goldblum JR. Esophageal carcinoma: depth of tumor invasion is 35. Winans CS. Manometric asymmetry of the lower-esophageal high-
predictive of regional lymph node status. Ann Thorac Surg. 1998;65: pressure zone. Am J Dig Dis. 1977;22:348-354.
787-792. 36. Friedland GW, Melcher DH, Berridge FR, Gresham GA. Debatable
21. Wirth W, Frommhold H. Thoracic duct and its variations. Lymphog- points in the anatomy of the lower oesophagus. Thorax. 1966;21:
raphy studies. Fortschr Geb Rontgenstr Nuklearmed. 1970;112:450-459. 487-498.
22. Netter FH. The Ciba Collection of Medical Illustrations. Digestive System. 37. Pope CE 2nd. The esophagus: 1967 to 1969. I. Gastroenterology. 1970;59:
Part 1: Upper Digestive Tract. Vol. 3. New York: Ciba Pharmaceutical 460-476.
Embassy; 1971. 38. Bombeck CT, Nyhus LM, Donahue PE. How far should the myotomy
23. Fritscher-Ravens A, Sriram PV, Bobrowski C, et al. Mediastinal extend on the stomach? In: Giuli R, McCallum RW, Skinner DB,
lymphadenopathy in patients with or without previous malignancy: eds. Primary Motility Disorders of the Esophagus. Paris: Libbey Eurotext;
EUS-FNA-based differential cytodiagnosis in 153 patients. Am J 1991:455.
Gastroenterol. 2000;95:2278-2284. 39. Korn O, Stein HJ, Richter TH, Liebermann-Meffert D. Gastroesopha-
24. Hyer JD, Silvestri G. Diagnosis and staging of lung cancer. Clin Chest geal sphincter: a model. Dis Esophagus. 1997;10:105-109.
Med. 2000;21:95-106, viii-ix. 40. Preiksaitis HG, Tremblay L, Diamant NE. Regional differences
25. Vianna A, Hayes PC, Moscoso G, et al. Normal venous circulation in the in vitro behaviour of muscle fibers from the human lower
of the gastroesophageal junction. A route to understanding varices. esophageal sphincter. J Gastrointest Motility. 1991;3:195.
Gastroenterology. 1987;93:876-889.
26. Casson AG, Rusch VW, Ginsberg RJ, Zankowicz N, Finley RJ. Lymph
node mapping of esophageal cancer. Ann Thorac Surg. 1994;58:
1569-1570.
PART TWO
Diagnostic Evaluation
of the Esophagus
CHAPTER
A
surgeon evaluating a patient prior to a possible vagus nerve and in the spinal nerves. The vagus nerve
esophageal operation must spend enough time afferents have been described as tension-sensitive fibers
in conversation with that patient to ensure a clear with low thresholds for response. They are thought to
understanding of their symptoms. Such dialogue not contribute to physiologic reflexes. The spinal afferents
only illuminates potential causes of those symptoms, but likely provide a nociceptor function and convey noxious
more importantly, it indicates whether a given surgical intensity of various stimuli. Stimuli that excite these
procedure may resolve them. The spectrum of symptoms in afferents include distention and exposure to acid. The
esophageal disease is wide, ranging from common typical afferent nerves have been classified as muscle-tension-
symptoms (e.g., heartburn, regurgitation, and dysphagia) sensitive, mucosal mechano/chemosensitive, and tension/
to atypical symptoms (e.g., cough, voice changes, chest mucosal receptors; however, most afferent fibers respond
pain, and globus sensation). Esophageal symptoms are to both mechanical and chemical stimuli.2,3 Acid excites
often vague in nature, which should prompt further primary sensory neurons in the esophagus by activating
questioning of the patient. Surgeons should be familiar two proton-gated channels: transient receptor potential
with the concept of “functional heartburn” so that surgery vanilloid-1 (TRPV1) and acid-sensing ion channels. The
is not performed for patients who would not benefit from latter is thought to mediate heartburn in nonerosive
surgical intervention. reflux disease (NERD).4 The activation of the TRPV1
The tests used to evaluate symptoms should be selected channel by acid can initiate neurogenic inflammation
in accordance with the suspected underlying pathology. and the release of proinflammatory substances from the
Functional testing is critical when a functional operation surrounding cellular matrix, with a resulting increase in
is planned, and in the present era of justified cost contain- noxious stimuli. Visceral hypersensitivity in patients with
ment, exhaustive testing is not always necessary. Nonethe- NERD appears to involve neurogenic inflammation, with
less, shortcuts should be avoided, as most functional testing an increase in both release of substance P and expression
is additive in the assessment of a patient’s pathology and of neurokinin 1 receptor, which in turn activates TRPV1
generally carries meaning. An operation performed for the and protease-activated receptor 2.4
incorrect indication can be disastrous, particularly because A number of tests have historically been used to under-
the esophagus is a relatively unforgiving organ. The most stand esophageal symptomatology, including acid infusion
common cause of failure after antireflux surgery is poor (i.e., the Bernstein test) and balloon distention. These
patient selection1; therefore, thoughtful analysis of present- proved to be useful for understanding the stimulus-symptom
ing symptoms and a complete work-up to identify the axis of the esophagus, but they are seldom used in today’s
cause of these symptoms are necessary before the surgeon clinical practice. The sensory neurons from the heart and
advises a surgical procedure for a functional disorder. the esophagus converge on the dorsal horns of the spinal
Surgeons whose patients experience dysphagia should cord, which likely explains the overlap of pain from either
be aware of the possibility of malignancy, as dysphagia is organ and often prompts suspicion of myocardial infarction
the most common presenting symptom associated with in patients with severe esophageal spastic syndromes. A
esophageal cancer. gender difference in esophageal symptom perception
has also been described, with males having both lower
tolerance to acid perfusion and greater sensitivity to balloon
ORIGIN OF ESOPHAGEAL SENSATION distention after acid sensitization.5 Hyperalgesia (i.e., the
Sensory nerve fibers in the esophagus are present in heightened perception of painful stimuli) appears to occur
both the muscle and mucosa, and are carried in the over a greater anatomic region in females.5
44
Esophageal Symptoms and Selection of Diagnostic Tests CHAPTER 5 44.e1
ABSTRACT
It is important for surgeons to understand the nature of
esophageal symptoms and to be aware of the various tests
available to elucidate these symptoms before recommend-
ing surgical intervention to a given patient. This chapter
describes a host of esophageal symptoms patients may
experience and explains different testing modalities that
should be carried out to ensure good surgical outcomes.
KEYWORDS
Chest pain, dysphagia, endoscopy, esophageal cancer,
gastroesophageal reflux disease, globus sensation, heart-
burn, impedance testing, manometry, pH monitoring,
symptoms
Esophageal Symptoms and Selection of Diagnostic Tests CHAPTER 5 45
1 8
7 pH channel
* 6
5 5 cm
4 above LES
3
2 Reflux
1
0
Press
175 Pressure
4 155 channels
135
115 Cough
95
75
55
35 15 cm
15
Press above LES
5 175
155
135
115
95
75
55 10 cm
35
15 above LES
Press
6 155
135
115 Esophageal
95 contractions
75
55
35 5 cm
15
-5 above LES
pH
1 8
7 pH channel
* 6
5
4
3
2 Reflux
1
0
Press
160 Pressure
4 140 channels
* 120 Cough
100
80
60
40
20 15 cm
0
Press above LES
5 160
140
* 120
Contractions
100
80
60
40 10 cm
20
0 above LES FIGURE 5.1 Cause-and-effect relationship between
Press cough and esophageal acid exposure. (A) Coughing
6 145
125 precipitated by a reflux episode may be the result
* 105
85 of occult aspiration of refluxed gastric juice or a
65
45 reflex brought on by esophageal acidification.
25 5 cm (B) Conversely, increased intraabdominal pressure,
5
above LES
-15 as occurs with coughing, may overcome antireflux
mechanisms and result in a gastroesophageal reflux
B episode. LES, Lower esophageal sphincter.
Hoarseness and Dental Caries sphincter function, esophageal motor disorders, or local
Reflux of gastric juice to the laryngeal aditus or the mouth laryngopharyngeal causes should be considered. The
is called laryngopharyngeal reflux (LPR), and has been association of this symptom with psychological factors
associated with laryngeal symptoms and dental caries. In is not well documented, but more than 90% of patients
the work-up of patients thought to have significant reflux, report symptom exacerbation during periods of high
catheters with two or more probes have been used to emotional intensity.33
assess reflux into the more proximal esophagus or even
the pharynx.23,24 Furthermore, the addition of impedance
catheter monitoring has shown that reflux into the pharynx RELEVANT TESTING BEFORE
is more frequent than previously thought, even with pH CONSIDERATION OF ANTIREFLUX SURGERY
monitoring. Kawamura et al.25 have shown that gaseous
reflux with weak acidity is more common in patients with It is important to understand the underlying physiologic
reflux-related laryngeal lesions. Again, objective testing defect associated with the presenting symptoms before
with pH or pH impedance is needed to diagnose proximal recommending antireflux surgery. Long-term symptom
reflux as a possible cause of LPR. relief depends on sound patient selection and avoidance
of intraoperative complications. An operation performed
Nausea and Vomiting for the wrong diagnosis will likely yield a poor outcome,
Although nausea and vomiting may accompany other and reoperations are less successful than initial procedures.
foregut symptoms such as heartburn and regurgitation, Patients with spastic motor disorders and achalasia may
they should alert the surgeon to proceed with caution present with heartburn34; clearly, performing an antireflux
when considering an antireflux procedure. Early satiety and procedure in these patients or in patients with functional
bloating may indicate delayed gastric emptying or gastro- heartburn would produce undesirable outcomes.
paresis, especially in diabetic patients. Severe preoperative Although a defective lower esophageal sphincter is
nausea and vomiting should be relative contraindications the most common underlying pathology of GERD, other
to antireflux surgery, as most antireflux operations inhibit defects may lead to increased exposure of acid within the
the ability to vomit, and postoperative heaving may disrupt esophagus. Xerostomia results in increased esophageal
the fundoplication or result in recurrent herniation. exposure to refluxed acid, because it diminishes the
Similarly, patients with concomitant symptoms of irritable bicarbonate buffering associated with swallowed saliva.
bowel syndrome, postcholecystectomy patients, or patients Likewise, hypomotility of the esophagus interferes with
with other prominent bowel symptoms such as diarrhea the clearance mechanism of the esophagus of refluxed
should be treated conservatively, if possible. gastric juice, thereby prolonging esophageal acid exposure.
Delayed gastric emptying may result in increased gastric
Globus Sensation volumes for both reflux of gastric juice and for regurgita-
The sensation of a lump or fullness in the back of the tion of partly digested food. Duodenogastric reflux may
throat, or of having a foreign sensation in the cervical result from a gastroduodenal pathology or may result
esophageal region, has been termed globus sensation. This from a previous surgical procedure (that could restrict
symptom was recognized 2500 years ago by Hippocrates, available options for a future surgical procedure). For all
and in the 19th century it was referred to as “globus of these reasons, a thorough history is the cornerstone of
hystericus” due to its frequent appearance in individuals understanding the underlying pathology and of guiding
with psychological disorders. Although the sensation future surgical interventions. Further diagnostic testing
is generally associated with benign disorders, it is still is usually needed before a surgical procedure is offered.
extremely irritating to most patients. It usually presents
in the fifth and sixth decades of life, and up to 46% of pH MONITORING
healthy individuals experience this sensation at some Prolonged ambulatory intraesophageal pH testing allows
point in their lives.26,27 Globus sensation has a significant for objective documentation of abnormal esophageal
association with GERD, and some have suggested a trial of exposure to refluxed acid, and can provide a link between
high-dose PPIs for 3 to 6 months prior to further work-up.28 reflux events and presenting symptoms. A 24-hour pH test
It may require a more prolonged and intensive trial of is the most accurate means of confirming suspected GERD
PPIs than is required for typical heartburn.29 and provides details on the extent of the esophageal acid
Objective pH testing may reveal abnormal esophageal exposure (see Table 5.1). A pH test may be performed
acid exposure in many patients with globus sensation, but with either a 24-hour transnasal probe or a small capsule
some have suggested that pH impedance studies are more (Bravo capsule, Medtronic, Dublin, Ireland) that is affixed
sensitive, as some patients are found to have symptoms to the esophageal mucosa and transmits pH data via a
associated with weak acid reflux.30–32 Patients who respond radiotelemetry device (Figs. 5.2 to 5.4). Both catheter- and
to PPIs only partially or not at all should not expect good capsule-based technology have been shown to be valid and
outcomes after antireflux surgery, and antireflux surgery reproducible,35,36 although the capsule allows for prolonged
should not be considered as treatment for globus sensation recordings (up to 48 or 72 hours). This may add additional
alone. If the disorder is associated with other symptoms, diagnostic information and can provide a greater database
such as dysphagia or weight loss, immediate work-up should for correlation of symptoms. The catheter system, on the
be carried out to rule out a malignant cause. The origin other hand, allows for a proximal and a distal pH sensor,
of the globus sensation is poorly understood, but if it is which may aid in identifying patients with LPR and increase
persistent, local causes such as abnormal upper esophageal understanding of the effect of reflux on chronic cough.
Esophageal Symptoms and Selection of Diagnostic Tests CHAPTER 5 49
pH probe Battery
Transmitter
Reference electrode
26 mm
5.5 mm
Well is
4 mm x 4 mm deep
6.3 mm
FIGURE 5.2 Dimensions and electronics of the Bravo pH capsule (Medtronic, Dublin, Ireland). The capsule is oblong (6.3 × 5.5 × 26 mm).
A well (diameter, 4 mm; depth, 4 mm) is located on the superior-lateral aspect of the probe. The well is connected to a custom-made
vacuum unit capable of generating 600 mm Hg vacuum pressure to the well via the delivery system. An antimony pH electrode and
reference electrode are located on the distal tip of the capsule, and an internal battery and transmitter are contained within the capsule.
A B
FIGURE 5.4 Bilitec probe (A) and electronic data logger (B) for 24-hour esophageal bilirubin monitoring.
since an antireflux procedure usually accompanies repair patients with normal pH studies. Impedance studies are
of the paraesophageal hernia, regardless of whether the relatively new; therefore, they are still being evaluated
patient experienced heartburn preoperatively. in the surgical arena. The test has the potential to add
Antireflux surgery performed in patients with normal substantially more information, although the testing does
24-hour pH tests is associated with significantly worse require a transnasal catheter for 24 hours.
outcomes than when it is performed in patients with abnor-
mal tests; therefore, routine pH testing is recommended MANOMETRY
prior to antireflux surgery.16 As a corollary, 24-hour pH Manometry provides information on the tone and coor-
testing or pH impedance testing is very useful in selecting dination of the upper and lower esophageal sphincters
those patients with functional heartburn or with visceral and on the propulsive function of the esophageal body.
hyperesthesia for whom surgery may prove ineffective. It allows for identification of motility disorders and also
provides information that may preclude some surgical
IMPEDANCE TESTING approaches.
Multichannel intraluminal impedance testing relies Most historical, water-perfused manometry systems
on the change in resistance to an electrical current have been replaced with high-resolution manometry that
when a bolus, liquid, or gas passes between the metal- uses sensitive, solid-state pressure transducers. The data
lic rings on a catheter. When the esophagus is empty, that can be collected with this test have allowed for more
the electrical current is conducted by the ions on the precise evaluation of esophageal motility disorders, and a
mucosa. Liquids with more ions increase conductivity new classification system (i.e., the Chicago Classification
and decrease resistance, which can be measured and System) has been proposed to capture this new level of
recorded in much the same way acid is measured in precision. This classification system uses the pressure
24-hour monitoring scenarios. By measuring impedance tomography plots from high-resolution manometry to
at many sites along a catheter, the direction of the bolus’s categorize various disorders by prioritizing (1) disorders
movement can be determined. In this way, a swallow of a of esophagogastric junction outflow, (2) major peristaltic
liquid will be antegrade, whereas a reflux episode will be disorders (i.e., aperistalsis, esophageal spasm, and jackham-
retrograde. mer or nutcracker esophagus), and (3) minor peristaltic
The benefit of this type of testing is that the fluid bolus disorders (i.e., those that impede bolus transport; namely,
(i.e., refluxate) can be measured regardless of its acidity. ineffective esophageal motility and fragmented peristalsis).
By combining pH testing with impedance testing, all Already this system has facilitated clarification of subtypes
types of reflux can be measured. This allows practitioners I, II, and III for achalasia, and implications of therapy
to determine acid episodes, weak acid episodes, and and potential surgical outcomes for each subtype have
nonacid episodes, and to correlate symptoms to each type.39 been delineated.40,41
Although most patients undergoing antireflux surgery The success of the Nissen fundoplication in patients
will have a positive acid pH test, a subgroup of patients with hypomotility of the esophageal body has been the
with weakly acid or nonacid reflux who may benefit from subject of much debate. Some have shown that motility
surgery has been described.39 Impedance testing may also of the esophageal body may improve slightly in some
have value in identifying the cause of chronic cough in patients after fundoplication.42–47 Although it is likely
Esophageal Symptoms and Selection of Diagnostic Tests CHAPTER 5 51
true that mildly depressed contraction amplitudes do functional esophageal disorders, as is well described later
not significantly impact postoperative dysphagia in expert in this book.
hands,48–50 severely depressed contractions and the so- Perhaps the most important aspect of performing the
called aperistaltic esophagus (as seen in connective tissue barium esophagram is effective communication between
disorders) will likely result in poor esophageal clearance the surgeon and the radiologist with regard to the patient’s
after a 360° fundoplication. Manometric evaluation of symptoms and suspected diagnoses. The study can be
the esophagus is the only objective way to provide a clear modified to concentrate on mucosal relief (for patients
evaluation of the esophageal pressure profiles that guide a with ulcerative esophagitis, infections, or early tumors),
surgeon performing functional surgery. Data are emerging propulsive action (recumbent swallows with different
on the benefits of the Chicago Classification System in barium consistencies), emptying of the esophagus in the
predicting outcomes after surgery depending on various upright position in patients with achalasia,54,55 or evaluation
motility abnormalities.41 High-resolution manometry of strictures. Certain provocative maneuvers, such as leg
may identify abnormalities that contraindicate surgery; raising or a Valsalva maneuver while the patient is supine,
others may help a surgeon counsel patients and manage can elicit free reflux, which—although not a sensitive
expectations of surgical outcomes. test—is usually quite specific for GERD.56 The radiolo-
Spastic motor disorders (e.g., diffuse esophageal spasm gist should have some understanding of the concepts
or achalasia) may present with heartburn-like discomfort.51 of deglutitive inhibition (i.e., rapid swallowing followed
In a large retrospective study of 1081 patients referred for only by a terminal peristaltic wave), and the refractory
antireflux surgery, 1 in 14 were found to have abnormalities period of the esophagus, as swallows closer together than
on high-resolution manometry that were considered abso- 20 seconds may have poorer quality propulsive function.
lute or relative contraindications for antireflux surgery.51 Overall, the barium study—when performed correctly—is
The researchers further noted that almost 25% of patients a wonderful screening tool in patients with esophageal
would have undergone antireflux surgery without clear symptoms.
evidence of GERD, had pH monitoring and manometry
not been performed during work-up. Of all 1081 patients, ENDOSCOPY
2.5% were found to have abnormalities on the achalasia Endoscopy is considered a critical preoperative study before
spectrum. An antireflux operation in this context would recommending antireflux surgery and provides necessary
obviously have a poor outcome that would likely require biopsy material in patients with Barrett esophagus, sus-
a second, corrective operation. In patients for whom pected malignancy, or eosinophilic esophagitis. Endoscopy
manometry is impossible (e.g., the patient refuses or the can identify esophagitis, strictures, hiatal hernia, the
catheter cannot be passed through the nose), a partial presence of Barrett esophagus, or malignancy, and provide
fundoplication should be performed to minimize the a qualitative look at the function of the lower esophageal
risk of postoperative dysphagia. In this situation, a well- sphincter. This analysis has been well described by Hill.57
performed barium study may provide some insight on Endoscopy facilitates categorization of esophagitis; the
peristaltic function. most commonly used scale is the Los Angeles classification
system (Table 5.2).58 Endoscopy also allows diagnosis of
IMPEDANCE MANOMETRY Barrett esophagus via biopsy, as the presence of high-
The recent addition of impedance to esophageal manom- grade dysplasia or intramucosal carcinoma in a segment
etry may provide further information on esophageal of the esophagus with Barrett esophagus may alter the
clearance in real time, but the role this technology plays in surgeon’s approach. In that situation, further studies
surgical planning is presently unclear. This modality may may be indicated, and ablation or resection may be the
help evaluate the propulsive efficiency of the esophagus
for different bolus consistencies and for different body
positions.52 This method may be useful for evaluation of
esophageal function both before and after fundoplication
in the future. TABLE 5.2 The Los Angeles Classification of Esophagitis
BARIUM ESOPHAGRAM Grade A One (or more) mucosal break no longer than
5 mm that does not extend between the tops
Continuous digital or video barium esophagram can
of two mucosal folds
identify structural abnormalities and can provide functional
Grade B One (or more) mucosal break more than 5 mm
information about the esophagus. If the patient is unable
long that does not extend between the tops
to tolerate a manometric study, the barium esophagram can of two mucosal folds
offer insight into the peristaltic activity of the esophagus, Grade C One (or more) mucosal break that is continuous
especially if the test is done with both solids and liquids. between the tops of two or more mucosal
A barium esophagram can also help identify the nature folds, but which involves less than 75% of
and position of a hiatal hernia. Interestingly, the standard the circumference
barium esophagram performed in many institutions today Grade D One (or more) mucosal break that involves at
has changed little over the 30 years it has been in use.53 least 75% of the esophageal circumference
Effervescent crystals are still used to provide information
Reprinted from Sami SS, Ragunath K. The Los Angeles classification of
on mucosal relief, and prone oblique views are still used, gastroesophageal reflux disease. Video J Encyclopedia GI Endoscopy.
even when achalasia is suspected. Some specialized centers 2013;1:103–104. Used with permission under Creative Commons license
have modified the barium study to better understand CC BY-NC-ND 4.0.
52 SECTION I Esophagus and Hernia
first course of action. The length of a Barrett segment, accordion-like shortening of the esophagus, with the
the size of a hiatal hernia, the presence of esophagitis or external pressure of the paraesophageal stomach pressing
stricture, and the ease of eructation in the retroflexed on the esophagus itself, or from a coexisting motility
position on insufflation should all be noted. Patients disorder. Chest discomfort and bloating may be related
with Barrett esophagus should undergo biopsies at four to distention of the intrathoracic stomach and the slow
quadrants every one to two centimeters, and should be emptying of this compartment. Dyspnea as a symptom is
placed on a surveillance program regardless of whether poorly understood, but is usually improved after surgical
they undergo fundoplication. intervention. Not infrequently, patients with giant hernias
or large paraesophageal hernias describe only a remote
GASTRIC EMPTYING STUDIES history of heartburn, but it is either no longer an issue
For patients with prominent symptoms of nausea, vomit- or is easily controlled with PPIs.
ing, postprandial fullness, or bloating, an evaluation of Large hernias occasionally present acutely with torsion,
gastric emptying should be considered. A bezoar visible which usually mandates immediate surgical intervention
after an overnight fast is usually a good indication of (although decompression and de-torsion with endoscopy
delayed gastric emptying, but the best objective study is a has been described). Severe abdominal and chest pain is
scintigraphic radionucleotide study using both solids and usually attributed to ischemia of the herniated stomach,
liquids.59 Unfortunately, the methods and normal values and patients with these presenting symptoms often pro-
vary significantly between centers, and the results are not gress rapidly to being critically ill with hemodynamic
always reliable.60,61 A recent consensus report suggested instability.
using a standardized meal of radiolabeled low-fat egg For the most part, patients’ symptoms progress slowly,
whites with imaging up to 4 hours after the meal.59 A and an outpatient work-up can be performed. This work-up
Nissen fundoplication may improve gastric emptying,62 should include barium studies, endoscopy, and manometry.
but a significantly delayed gastric emptying study should pH studies are likely unnecessary, as the indication for
alert the surgeon to the need for other procedures, such surgery is the large hernia itself, and the surgical repair
as pyloric injection of botulinum toxin type A, pyloroplasty, of the hernia is usually accompanied by some sort of
or a Roux-en-Y procedure. fundoplication. Barium studies will show the size and
position of the hernia, and if performed with a 13-mm
RELEVANT PREOPERATIVE TESTING FOR barium tablet, will reveal any delay during passage through
the esophagus. Endoscopy will identify Barrett esophagus,
SPECIFIC INDICATIONS dysplasia, and Cameron ulcers in the stomach. Manometry
will identify any marked motility abnormalities—which
PREOPERATIVE TESTING FOR BARRETT ESOPHAGUS is especially important, as the incidence of spastic motor
Barrett esophagus can only be diagnosed on endoscopy disorders increases in the older population.
with confirmation of intestinal metaplasia on biopsy. The
finding of goblet cells intermingled in a columnar lining PREOPERATIVE TESTING FOR MYOTOMY FOR
of the tubular esophagus is pathognomonic of Barrett ACHALASIA OR DIVERTICULA
esophagus. The disease is considered to be a metaplastic The diagnosis of spastic motor disorders and achalasia
response to severe GERD,63 and its clinical significance is depends on a well-performed manometric study, and
the premalignant potential of the unstable epithelium. manometry is critical before surgery is considered for
The definition has changed slightly over the years, but these disorders. Barium studies usually show areas of
biopsy is required for confirmation. corkscrewing or the tapered bird-beak appearance of
The finding of Barrett esophagus has significance the distal esophagus associated with achalasia. Endoscopy
outside its potential for malignancy, as patients with Barrett is recommended to rule out pseudoachalasia, in which
esophagus often have associated esophageal hypomotility the motility disorder of the esophagus is secondary to
and generally have severely abnormal esophageal exposure an infiltrating process (usually adenocarcinoma) at the
to both refluxed acid and duodenal contents. Patients with gastroesophageal junction.64,65 Chagas disease may give
Barrett esophagus may also have a shorter-than-normal a similar picture.66
esophagus, which may have implications for a planned Barium esophagram and a manometric study can
surgical procedure. indicate how far the myotomy should extend, especially
in patients with a diverticulum that is situated more than
PREOPERATIVE TESTING FOR GIANT HERNIAS OR a few centimeters above the gastroesophageal junction.
LARGE PARAESOPHAGEAL HERNIAS Manometry can also help counsel patients with different
The aging population, coupled with the ubiquitous use variants of achalasia regarding expected outcomes.40,41
of PPIs for heartburn control (often of decades’ dura-
tion), has led to an epidemic of large paraesophageal PREOPERATIVE TESTING FOR END-STAGE LUNG
hernias or the so-called intrathoracic stomach. Presum- DISEASE AND TRANSPLANTATION
ably, heartburn has been adequately controlled in these The effect of GERD in patients with end-stage lung disease
patients, but the hernia has progressively enlarged to and in patients who have undergone lung transplantation
the point that the hernia itself is responsible for the has historically been underestimated. A high proportion
presenting symptoms. These symptoms may include chest of patients with end-stage lung disease have pathologic
discomfort, postprandial fullness or discomfort, dyspnea, GERD, and it has been suggested that silent aspiration
dysphagia, or anemia. Dysphagia may be related to the contributes to pulmonary injury in many of these patients.
Esophageal Symptoms and Selection of Diagnostic Tests CHAPTER 5 53
Similarly, the chronic cough associated with many end- are indicated. pH impedance studies are diagnostic, but
stage lung diseases is thought to promote reflux, as it manometry is also important, as abnormalities in peristalsis
increases intraabdominal and transsphincteric pressure or function of the lower esophageal sphincter may occur
(see Fig. 5.1). Recently, GERD has also been implicated in up to 66% of patients73 and may contribute to poor
as a significant adverse contributor to the development of esophageal clearance of refluxed fluid. Endoscopy is
bronchiolitis obliterans syndrome after lung transplanta- important, as Barrett esophagus may be more prevalent
tion.67,68 Davis and colleagues67 have shown that 73% of in the lung transplant population. 74 Early antireflux
patients who have undergone lung transplantation have surgery should be considered in patients with GERD; if
GERD (diagnosed by pH monitoring). This may be due the patient can tolerate such a procedure, this is optimal,
to the significant number of patients with unrecognized because the patient can recover when they are not taking
GERD before transplantation, to vagal damage at the steroids or other immunosuppressive medications, as they
time of surgery, or to the reflux-promoting side effects will be posttransplant. If antireflux surgery cannot be
of postoperative immunosuppressive medications. None- performed before transplantation, it should be considered
theless, Davis et al.67 have shown that fundoplication soon after transplant to protect the transplanted organ.
in lung transplant recipients with GERD is associated By the time the organ shows evidence of decline (i.e.,
with significantly improved lung function, particularly decreased forced expiratory volume in 1 second [FEV1]),
if the fundoplication is performed before bronchiolitis it may be too late to prevent the progression of organ
obliterans syndrome reaches advanced stages. Others have dysfunction.
suggested that performing a fundoplication more than
6 months after lung transplantation is associated with PREOPERATIVE TESTING FOR REOPERATION AFTER
worse long-term allograft function.69,70 Many patients with FAILURE OF PREVIOUS ANTIREFLUX SURGERY
progressive deterioration in lung function and the diagnosis A very detailed patient history is critical when evaluating a
of pulmonary fibrosis referred for transplantation have patient who has undergone an antireflux procedure that
experienced stabilization of their pulmonary disease after has presumably failed. It is useful to review any preopera-
fundoplication, again emphasizing the negative effects tive studies carried out before the initial operation, if
that GERD and silent aspiration have on pulmonary the results of those studies are available. An attempt
function. should be made to prioritize the patient’s symptoms to
It is now the standard of care to have a solid under- understand the nature of the failure (Box 5.3). When
standing of a patient’s reflux history before considering considering a reoperation, it is crucial to perform all
lung transplantation, and in many centers all patients necessary testing to optimize the outcome, because the
considered for lung transplantation undergo manometry, second and third procedures are usually less successful
videoesophagraphy, and 24-hour pH testing to assess for than the initial procedure, especially if dysphagia is a
GERD. Patients with severe reflux and a nonprohibitive prominent symptom.75–77
surgical risk will undergo fundoplication before trans- Manometry is particularly important because it provides
plantation. Antireflux surgery in the early posttransplant the surgeon with useful information about the motor
period is considered for patients who cannot undergo function of the esophagus, as it may differ from its status
surgery before transplantation, or who develop GERD before the first operation. A gastric emptying study is
after transplantation. also important, particularly if bloating and early satiety
pH monitoring has provided tremendous insight into are prominent symptoms. Operative options include
the significance of GERD in this complex group of patients, takedown of the previous wrap, redo fundoplication, pos-
and continues to provide important information that sible pyloroplasty, and gastric diversion such as Roux-en-Y
directs therapy. A mere symptom assessment is completely esophagoenterostomy or gastroenterostomy, all of which
inadequate when evaluating patients slated to undergo are discussed in greater detail later in this book.
lung transplantation, and pH or pH impedance testing is
required to identify patients who may be at risk of aspira-
tion as a result of GERD. Soresi and colleagues71 found
BOX 5.3 Questions to Ask Your Patient Prior to
almost 60% of lung transplantation patients suffered from
Redo Surgery
postoperative GERD. Similarly, Tamhankar et al.72 showed
that PPIs only affect the pH of the refluxate—not the What were your symptoms before your first operation?
occurrence or frequency of the reflux episodes. As such, What tests did you undergo before your first operation?
it may be prudent to have all transplant patients undergo Did you have a hiatal hernia or a paraesophageal hernia? If
multichannel pH impedance testing posttransplantation, so, did your surgeon use mesh to repair the hernia?
especially as patients with abnormal proximal esophageal How long were you in the hospital?
reflux are more likely to be readmitted to the hospital. Did you have any side effects of the operation?
This group of patients should be treated aggressively Did your symptoms improve or resolve after your first
(i.e., early antireflux surgery) to protect the implanted operation?
What are your main symptoms currently?
allograft, especially in the face of immunosuppression.69
When did these start? Are they progressively getting worse?
The problematic effects of reflux in the patients who
Did anything precipitate the onset of these symptoms?
have undergone lung transplantation are complex. Clearly, Were any of these present prior to your initial operation?
aspiration occurs more frequently than previously thought, Please describe any symptoms of dysphagia, heartburn,
as bile salts and pepsin have been found in bronchoalveolar regurgitation, bloating, nausea, or early satiety in detail.
lavage fluid, and thorough testing pre- and posttransplant
54 SECTION I Esophagus and Hernia
25. Kawamura O, Aslam M, Rittmann T, Hofmann C, Shaker R. Physical 49. Patti MG, Perretta S, Fisichella PM, et al. Laparoscopic antireflux
and pH properties of gastroesophagopharyngeal refluxate: a 24-hour surgery: preoperative lower esophageal sphincter pressure does not
simultaneous ambulatory impedance and pH monitoring study. Am affect outcome. Surg Endosc. 2003;17(3):386-389.
J Gastroenterol. 2004;99(6):1000-1010. 50. Zornig C, Strate U, Fibbe C, Emmermann A, Layer P. Nissen vs
26. Drossman DA, Li Z, Andruzzi E, et al. U.S. householder survey of Toupet laparoscopic fundoplication. Surg Endosc. 2002;16(5):758-766.
functional gastrointestinal disorders. Prevalence, sociodemography, 51. Chan WW, Haroian LR, Gyawali CP. Value of preoperative esophageal
and health impact. Dig Dis Sci. 1993;38(9):1569-1580. function studies before laparoscopic antireflux surgery. Surg Endosc.
27. Moloy PJ, Charter R. The globus symptom. Incidence, thera- 2011;25(9):2943-2949.
peutic response, and age and sex relationships. Arch Otolaryngol. 52. Gao F, Gao Y, Hobson AR, Huang WN, Shang ZM. Normal esophageal
1982;108(11):740-744. high-resolution manometry and impedance values in the supine and
28. Lee BE, Kim GH. Globus pharyngeus: a review of its etiology, diagnosis sitting positions in the population of Northern China. Dis Esophagus.
and treatment. World J Gastroenterol. 2012;18(20):2462-2471. 2016;29(3):267-272.
29. Remacle M. The diagnosis and management of globus: a perspective 53. Levine MS, Rubesin SE, Herlinger H, Laufer I. Double-contrast
from Belgium. Curr Opin Otolaryngol Head Neck Surg. 2008;16(6): upper gastrointestinal examination: technique and interpretation.
511-515. Radiology. 1988;168(3):593-602.
30. Bajbouj M, Becker V, Neuber M, Schmid RM, Meining A. Combined 54. Kostic S, Andersson M, Hellstrom M, Lonroth H, Lundell L.
pH-metry/impedance monitoring increases the diagnostic yield in Timed barium esophagogram in the assessment of patients with
patients with atypical gastroesophageal reflux symptoms. Digestion. achalasia: reproducibility and observer variation. Dis Esophagus.
2007;76(3-4):223-228. 2005;18(2):96-103.
31. Lee BE, Kim GH, Ryu DY, et al. Combined dual channel impedance/ 55. Vaezi MF, Baker ME, Achkar E, Richter JE. Timed barium oesopha-
pH-metry in patients with suspected laryngopharyngeal reflux. gram: better predictor of long term success after pneumatic dilation
J Neurogastroenterol Motil. 2010;16(2):157-165. in achalasia than symptom assessment. Gut. 2002;50(6):765-770.
32. Malhotra A, Freston JW, Aziz K. Use of pH-impedance testing to 56. Christiansen T, Funch-Jensen P, Jacobsen NO, Thommesen P.
evaluate patients with suspected extraesophageal manifestations of Radiologic quantitation of gastro-oesophageal reflux. Correla-
gastroesophageal reflux disease. J Clin Gastroenterol. 2008;42(3):271- tion between height of food stimulated gastro-oesophageal reflux
278. and level of histologic changes in reflux oesophagitis. Acta Radiol.
33. Harris MB, Deary IJ, Wilson JA. Life events and difficulties in relation 1987;28(6):731-734.
to the onset of globus pharyngis. J Psychosom Res. 1996;40(6):603-615. 57. Hill LD, Kozarek RA, Kraemer SJ, et al. The gastroesophageal
34. Crookes PF, Corkill S, DeMeester TR. Gastroesophageal reflux in acha- flap valve: in vitro and in vivo observations. Gastrointest Endosc.
lasia. When is reflux really reflux? Dig Dis Sci. 1997;42(7):1354-1361. 1996;44(5):541-547.
35. Ayazi S, Lipham JC, Portale G, et al. Bravo catheter-free pH monitor- 58. Lundell LR, Dent J, Bennett JR, et al. Endoscopic assessment of
ing: normal values, concordance, optimal diagnostic thresholds, and oesophagitis: clinical and functional correlates and further validation
accuracy. Clin Gastroenterol Hepatol. 2009;7(1):60-67. of the Los Angeles classification. Gut. 1999;45(2):172-180.
36. Johnson LF, DeMeester TR. Twenty-four-hour pH monitoring of 59. Abell TL, Camilleri M, Donohoe K, et al. Consensus recommendations
the distal esophagus. A quantitative measure of gastroesophageal for gastric emptying scintigraphy: a joint report of the American
reflux. Am J Gastroenterol. 1974;62(4):325-332. Neurogastroenterology and Motility Society and the Society of
37. Bytzer P, Havelund T, Hansen JM. Interobserver variation in the Nuclear Medicine. Am J Gastroenterol. 2008;103(3):753-763.
endoscopic diagnosis of reflux esophagitis. Scand J Gastroenterol. 60. Guo JP, Maurer AH, Fisher RS, Parkman HP. Extending gastric
1993;28(2):119-125. emptying scintigraphy from two to four hours detects more patients
38. Tefera L, Fein M, Ritter MP, et al. Can the combination of symptoms with gastroparesis. Dig Dis Sci. 2001;46(1):24-29.
and endoscopy confirm the presence of gastroesophageal reflux 61. Ziessman HA, Bonta DV, Goetze S, Ravich WJ. Experience with a
disease? Am Surg. 1997;63(10):933-936. simplified, standardized 4-hour gastric-emptying protocol. J Nucl
39. Hila A, Agrawal A, Castell DO. Combined multichannel intraluminal Med. 2007;48(4):568-572.
impedance and pH esophageal testing compared to pH alone for 62. Hinder RA, Stein HJ, Bremner CG, DeMeester TR. Relationship of
diagnosing both acid and weakly acidic gastroesophageal reflux. a satisfactory outcome to normalization of delayed gastric emptying
Clin Gastroenterol Hepatol. 2007;5(2):172-177. after Nissen fundoplication. Ann Surg. 1989;210(4):458-464; discussion
40. Pandolfino JE, Fox MR, Bredenoord AJ, Kahrilas PJ. High-resolution 464–455.
manometry in clinical practice: utilizing pressure topography to 63. Bremner CG, Lynch VP, Ellis FH Jr. Barrett’s esophagus: congenital or
classify oesophageal motility abnormalities. Neurogastroenterol Motil. acquired? An experimental study of esophageal mucosal regeneration
2009;21(8):796-806. in the dog. Surgery. 1970;68(1):209-216.
41. Pandolfino JE, Kwiatek MA, Nealis T, Bulsiewicz W, Post J, Kahrilas 64. Rosenzweig S, Traube M. The diagnosis and misdiagnosis of
PJ. Achalasia: a new clinically relevant classification by high-resolution achalasia. A study of 25 consecutive patients. J Clin Gastroenterol.
manometry. Gastroenterology. 2008;135(5):1526-1533. 1989;11(2):147-153.
42. Fibbe C, Layer P, Keller J, Strate U, Emmermann A, Zornig C. 65. Tucker HJ, Snape WJ Jr, Cohen S. Achalasia secondary to carcinoma:
Esophageal motility in reflux disease before and after fundoplica- manometric and clinical features. Ann Intern Med. 1978;89(3):315-318.
tion: a prospective, randomized, clinical, and manometric study. 66. de Oliveira RB, Rezende Filho J, Dantas RO, Iazigi N. The spectrum
Gastroenterology. 2001;121(1):5-14. of esophageal motor disorders in Chagas’ disease. Am J Gastroenterol.
43. Herbella FA, Tedesco P, Nipomnick I, Fisichella PM, Patti MG. Effect 1995;90(7):1119-1124.
of partial and total laparoscopic fundoplication on esophageal body 67. Davis RD Jr, Lau CL, Eubanks S, et al. Improved lung allograft
motility. Surg Endosc. 2007;21(2):285-288. function after fundoplication in patients with gastroesophageal
44. Lund RJ, Wetcher GJ, Raiser F, et al. Laparoscopic Toupet fundoplica- reflux disease undergoing lung transplantation. J Thorac Cardiovasc
tion for gastroesophageal reflux disease with poor esophageal body Surg. 2003;125(3):533-542.
motility. J Gastrointest Surg. 1997;1(4):301-308; discussion 308. 68. Verleden GM, Dupont LJ, Van Raemdonck DE. Is it bronchiolitis
45. Pizza F, Rossetti G, Del Genio G, Maffettone V, Brusciano L, Del obliterans syndrome or is it chronic rejection: a reappraisal? Eur
Genio A. Influence of esophageal motility on the outcome of Respir J. 2005;25(2):221-224.
laparoscopic total fundoplication. Dis Esophagus. 2008;21(1):78-85. 69. Lo WK, Goldberg HJ, Burakoff R, Feldman N, Chan WW. Increased
46. Rydberg L, Ruth M, Abrahamsson H, Lundell L. Tailoring antireflux proximal acid reflux is associated with early readmission following
surgery: a randomized clinical trial. World J Surg. 1999;23(6):612- lung transplantation. Neurogastroenterol Motil. 2016;28(2):251-259.
618. 70. Lo WK, Goldberg HJ, Wee J, Fisichella PM, Chan WW. Both pre-
47. Tsereteli Z, Sporn E, Astudillo JA, Miedema B, Eubanks WS, Thaler K. transplant and early post-transplant antireflux surgery prevent
Laparoscopic Nissen fundoplication is a good option in patients with development of early allograft injury after lung transplantation. J
abnormal esophageal motility. Surg Endosc. 2009;23(10):2292-2295. Gastrointest Surg. 2016;20(1):111-118; discussion 118.
48. Booth M, Stratford J, Dehn TC. Preoperative esophageal body 71. Soresi S, Zeriouh M, Sabashnikov A, et al. GORD symptoms in
motility does not influence the outcome of laparoscopic Nissen lung transplantation: how efficient is the reflux symptom index
fundoplication for gastroesophageal reflux disease. Dis Esophagus. questionnaire compared to the esophageal impedance test? Clin
2002;15(1):57-60. Transplant. 2016;30(1):44-51.
56 SECTION I Esophagus and Hernia
72. Tamhankar AP, Peters JH, Portale G, et al. Omeprazole does not 75. Khajanchee YS, O’Rourke R, Cassera MA, Gatta P, Hansen PD,
reduce gastroesophageal reflux: new insights using multichannel intra- Swanstrom LL. Laparoscopic reintervention for failed antireflux
luminal impedance technology. J Gastrointest Surg. 2004;8(7):890-897; surgery: subjective and objective outcomes in 176 consecutive
discussion 897–898. patients. Arch Surg. 2007;142(8):785-901; discussion 791–782.
73. Basseri B, Conklin JL, Pimentel M, et al. Esophageal motor dysfunc- 76. Papasavas PK, Yeaney WW, Landreneau RJ, et al. Reoperative lapa-
tion and gastroesophageal reflux are prevalent in lung transplant roscopic fundoplication for the treatment of failed fundoplication.
candidates. Ann Thorac Surg. 2010;90(5):1630-1636. J Thorac Cardiovasc Surg. 2004;128(4):509-516.
74. Davis CS, Shankaran V, Kovacs EJ, et al. Gastroesophageal reflux 77. Wee JO. Redo laparoscopic repair of benign esophageal disease.
disease after lung transplantation: pathophysiology and implications J Thorac Cardiovasc Surg. 2012;144(3):S71-S73.
for treatment. Surgery. 2010;148(4):737-744; discussion 744–735.
CHAPTER
Radiology of the Esophagus: Barium, Computed
Tomography Scan, Positron Emission Tomography 6
Scan, Magnetic Resonance Imaging
John M. Barlow
| Daniel A. Craig
| Val J. Lowe
| Robert L. MacCarty
R
adiologic evaluation of esophageal diseases can be discussed in the settings of GERD, esophageal motility
performed by fluoroscopic barium examination disorders, esophageal neoplasms, and the postoperative
(esophagram), computed tomography (CT) scan, esophagus. The discussion of esophageal neoplasms
positron emission tomography (PET) scan, fused CT includes a discussion of the role of cross-sectional imaging,
and PET images (PET/CT), and magnetic resonance especially PET/CT, for tumor staging. Subsequently, miscel-
imaging (MRI). These five imaging modalities constitute laneous conditions such as hiatal hernias, esophageal rings
the radiologic options currently available for esophageal and webs, less common types of esophageal strictures,
evaluation. Which test is best for the diagnosis of esopha- caustic injury, and esophageal perforation, diverticula,
geal disease? It depends. The choice of esophageal imaging and varices are discussed and illustrated.
modality depends on the goal of esophageal imaging.
Indeed, the choice of esophageal testing, in general, ESOPHAGRAM
depends on the goal of the testing. When the diagnostic
questions are esophageal obstruction, characterization An understanding of the imaging techniques of the
of hiatal hernia, esophageal perforation, or achalasia, esophagram helps the referring physician evaluate the
the esophagram is the best test to perform. When the quality of images, the completeness of the study, and
diagnostic questions are gastroesophageal reflux disease the validity of the conclusions reached by the radiolo-
(GERD), other esophagitis, mass, or motility disorder other gist. The essential techniques of the esophagram are not
than achalasia, the esophagram can be useful, although it complex. The performance of the radiologist performing
is not the best test available for these entities. When the esophagrams improves when referring physicians and
diagnostic question is the staging of esophageal tumors, surgeons provide follow-up regarding the accuracy of his
esophagram is not indicated because it demonstrates the or her diagnoses. Indeed, a team approach to the care of
luminal margin of the esophagus only. Tumor staging esophageal diseases not only benefits the patient; it also
questions are best answered by cross-sectional imaging benefits the physicians caring for the patient. They learn
modalities. from each other.
The barium esophagram remains a test of definite utility A complete esophagram is a multiphasic examination
in the 21st century, along with cross-sectional imaging performed in upright and recumbent positions with
(CT, MRI, and PET), endoscopy, endoscopic ultrasound variable consistencies of barium resulting in air-contrast,
(EUS), high-resolution manometry, pH monitoring, and single-contrast, and mucosal relief images.1 Unfortunately,
multichannel intraluminal impedance. In fact, all of these radiologists typically refer to this complete esophagram
tests are complementary for the diagnosis of esophageal as an air-contrast esophagram; nevertheless, they know that
diseases. single-contrast, recumbent images are necessary, in addi-
In our practice over the last decade, the volume of tion to air-contrast images, to consider an esophagram
radiographic contrast studies of the stomach, small bowel, complete. The images obtained during air-contrast, single-
and colon has continued to decline; however, the volume contrast, and mucosal relief phases of the examination
of esophagrams has increased each year. This increasing are complementary.
utilization of a nearly 100-year-old test by gastroenterolo- The air-contrast technique is performed in the upright
gists and other referring physicians and surgeons indicates position, typically with the patient in the left posterior
its continuing relevance for the evaluation of dysphagia. oblique position with respect to the vertical fluoroscopy
The utility of the esophagram is increased further when, table. This technique allows for detailed evaluation of
in the appropriate clinical context of dysphagia, it is the esophageal mucosa. Maximum gaseous distention of
paired with a fluoroscopic video swallow. Between them, the esophagus is achieved when the patient ingests an
these two examinations can demonstrate the structure effervescent agent that releases carbon dioxide when mixed
and function of the oropharynx, hypopharynx, cervical with water. Immediately after the esophagus is distended
esophagus, and thoracic esophagus. Their relatively low by gas, the patient drinks high-density barium as quickly
cost and almost universal availability make them a logical as possible in the upright position. Uniform distribution
starting point for the evaluation of dysphagia. of this thick barium on the luminal surfaces of the gas-
This chapter begins by describing the essential techni- distended esophagus allows demonstration of the entire
cal elements, normal findings, and common artifacts of mucosa of the esophagus. Normally, the esophageal mucosa
the esophagram. Then, the utility of the esophagram is is featureless on these air-contrast images. Occasionally,
57
Radiology of the Esophagus: Barium, Computed Tomography Scan, Positron Emission Tomography Scan, Magnetic Resonance Imaging CHAPTER 6 57.e1
ABSTRACT
Although multimodality imaging plays an essential role in
the staging of esophageal neoplasia, it is less helpful for
the evaluation of common esophageal symptoms, such as
dysphagia, regurgitation, and heartburn, than the barium
esophagram. Therefore this chapter begins by describing
the essential techniques, normal findings, and common
artifacts of the esophagram. Subsequently, the utility of the
esophagram is discussed in the settings of gastroesophageal
reflux disease, esophageal motility disorders, postoperative
esophagus, and esophageal neoplasms (a discussion of
the role of cross-sectional imaging, especially positron
emission tomography, for esophageal tumor staging is
included in this section). Finally, miscellaneous conditions
such as hiatal hernias, esophageal rings and webs, less
common types of esophageal strictures, caustic injury,
and esophageal perforation, diverticula, and varices are
discussed and illustrated.
KEYWORDS
Radiology of the esophagus, esophagram, esophageal
motility disorders, esophageal neoplasms, postoperative
esophagus, gastroesophageal reflux disease
58 SECTION I Esophagus and Hernia
FIGURE 6.1 Normal air-contrast esophagram. The mucosa is FIGURE 6.2 Normal mucosal relief esophagram. The longitudinal
featureless, except for the occasional tiny filling defect caused by mucosal folds (arrows) appear smooth and continuous. They
undissolved effervescent crystals (arrows). measure less than 3 mm in thickness.
tiny filling defects, representing undissolved effervescent When the patient indicates that food sticks at the level
crystals, are present on the mucosal surface (Fig. 6.1). of the neck or thoracic inlet, the oropharynx and cervical
These artifactual filling defects usually change position esophagus should be evaluated in the upright position. For
on sequential upright images. example, the radiologist puts the patient in the upright
Because the esophagus remains distended by gas for only lateral position and observes a swallow fluoroscopically.
a short time in the upright position, the radiologist needs Structural causes of oropharyngeal dysphagia such as
to obtain air-contrast images quickly. As the esophageal cricopharyngeal bar, cervical esophageal stricture, or
lumen collapses, residual barium is trapped between cervical esophageal web are readily identified in the lateral
redundant longitudinal mucosal folds of the esophagus, position. When a neuromuscular cause of dysphagia is
resulting in a mucosal relief image of the esophagus. suggested by this single upright view, it is best evaluated by
These longitudinal folds should appear as smooth and a speech pathologist trained in the evaluation of dysphagia.
linear structures less than 3 mm thick on mucosal relief The speech pathologist interviews and examines the patient
images (Fig. 6.2). Mild thickening and irregularity of the before performing a video swallow study with multiple
distal longitudinal folds may be a subtle sign of reflux consistencies of barium. In fact, we often perform the
esophagitis. video swallow, in consultation with speech pathologist,
Before tilting the table into the horizontal position, the and the esophagram during a single patient visit to the
radiologist should determine the need for upright evalua- fluoroscopy suite. This patient convenience is facilitated
tion of swallowing and the cervical esophagus. Anatomically, by our referring physicians and surgeons, who are aware
the two main sites of abnormalities resulting in dysphagia of the variable etiology of so-called high dysphagia (when
are the oropharynx and the esophagus. When patients the patient indicates that food sticks at the level of neck
report food sticking, and point to the neck or thoracic or thoracic inlet). Consequently, they can order both
inlet as the level of obstruction, their symptoms may result the video swallow study and the esophagram initially, or
from oropharyngeal dysphagia or esophageal dysphagia. they can give us permission to evaluate the esophagus
However, when patients report food sticking, and point as necessary based on the result of the video swallow.
to the substernal region as the level of obstruction, their A motion-recording device that captures 30 frames per
symptoms typically result from an esophageal dysphagia. second is very helpful for the evaluation of swallowing.
Therefore, referring physicians and surgeons, as well as This type of continuous recording captures the dynamic
radiologists, should ask the patient two questions: (1) events of swallowing far better than rapid sequential
Does food stick? (2) Where does it stick? radiographic images captured at 4 to 8 frames per second.
Radiology of the Esophagus: Barium, Computed Tomography Scan, Positron Emission Tomography Scan, Magnetic Resonance Imaging CHAPTER 6 59
FIGURE 6.5 Cricopharyngeal bar. A smooth posterior impression at FIGURE 6.6 Esophageal ampulla (vestibule). The caliber of the
the pharyngoesophageal junction (arrow) caused by failure of normal esophagus increases slightly just superior to the level of
cricopharyngeus muscle to relax. the gastroesophageal junction (arrows).
Preoperative Planning
FIGURE 6.10 Midesophageal scarring secondary to gastroesophageal The presence of a large hiatal hernia (>5 cm) or evidence
reflux disease. Transverse scars (arrow) are typical for a benign of a shortened esophagus can influence the type of surgical
stricture caused by gastroesophageal reflux disease. However, repair and operative approach during surgery for GERD.
the location of this stricture, many centimeters proximal to the In fact, failure to recognize these conditions may lead to
gastroesophageal junction, suggests the presence of Barrett surgical failure as a result of an inappropriate surgical
metaplasia between gastroesophageal junction and this stricture. approach or type of repair. The size of a hiatal hernia
Radiology of the Esophagus: Barium, Computed Tomography Scan, Positron Emission Tomography Scan, Magnetic Resonance Imaging CHAPTER 6 63
ESOPHAGEAL NEOPLASMS
Patients with malignant esophageal neoplasms usually
present with dysphagia. Conversely, benign esophageal
tumors tend to be incidental radiographic or endoscopic
findings. However, when these tumors are symptomatic,
excision is usually curative. CT can occasionally suggest
the diagnosis of esophageal neoplasm, but it is more
useful in staging esophageal malignancies, along with
more specific modalities such as PET and EUS.
FIGURE 6.16 Diffuse esophageal spasm. Multiple tertiary
contractions produce a “corkscrew” appearance of the CARCINOMA
esophagus. The symptoms causing patients with esophageal malignancy
to seek medical care are typically dysphagia of recent
onset (1 to 4 months) and weight loss. The prognosis for
of concentric distal esophageal wall thickening by CT, symptomatic patients is dismal. Historically, more than
along with infectious, inflammatory, and neoplastic causes. 95% of esophageal cancers have been squamous cell
NEMD is a “waste basket” category used that includes carcinomas. However, in recent decades, the incidence of
motility disorders that do not meet established mano- esophageal adenocarcinoma has increased dramatically.31
metric criteria for other motility disorders. Manometric Radiographically, these two types of carcinoma are indis-
abnormalities include failed peristalsis, low-amplitude tinguishable. However, adenocarcinoma predominantly
contractions, prolonged duration of peristalsis, simultane- occurs in the distal esophagus within regions of the Barrett
ous contractions, tertiary contractions, and incomplete esophagus. Squamous cell carcinoma, by comparison, tends
relaxation of the LES. Symptoms are nonspecific and to occur in the upper two-thirds of the esophagus. Other
include chest pain and dysphagia. Radiographic findings primary malignancies of the esophagus, such as sarcomas,
are frequently normal. When present, radiographic findings gastrointestinal stromal tumors (GISTs), melanoma, and
are nonspecific, and include ineffective peristalsis and lymphoma, are rare.
tertiary contractions causing stasis of esophageal barium.
Recently, a subgroup of patients with NEMD defined Radiologic Appearance
manometric criteria demonstrating hypocontraction of The esophagram can contribute to the initial diagnosis of
the distal esophagus. GERD is common in these patients. esophageal cancer. It can characterize the size, location,
However, radiographic findings in these patients are and morphology of the mass. It can demonstrate complica-
nonspecific, and are similar to those of other patients tions that make the cancer unresectable, such as a fistula
with NEMD.30 to the tracheobronchial tree. It can also demonstrate
coexisting esophageal disorders, such as benign strictures,
SECONDARY MOTILITY DISORDERS hiatal hernias, motility disorders, and rare synchronous
Secondary motility disorders are systemic disorders that second tumors.
secondarily affect the esophagus. With few exceptions, the Early resectable esophageal carcinomas can be suggested
radiographic appearance is nonspecific. Of the collagen on air-contrast images of the esophagus performed with
vascular diseases, scleroderma most often involves the careful technique. Early disease has a variety of subtle radio-
esophagus. The pathologic changes of scleroderma result graphic appearances, including fixed mucosal irregularity,
in hypomotility of the distal esophagus and a hypotensive irregular strictures, polypoid filling defects, or plaque-like
LES. The combination of these two motor abnormalities filling defects (Fig. 6.17). When radiographic findings of a
sets the stage for severe GERD, secondary to recumbent smooth benign-appearing stricture are demonstrated, they
GER and poor acid clearance. can reliably be considered benign.17 However, endoscopy
The radiographic changes of scleroderma include poor may still be useful to search for signs of acute esophagitis
distal esophageal peristalsis, stasis, and esophageal injury or Barrett esophagus. When radiographically equivocal
secondary to GERD. Eventually, the severe GERD caused or malignant-appearing strictures are demonstrated,
66 SECTION I Esophagus and Hernia
A B
FIGURE 6.17 Adenocarcinoma arising from a region of Barrett esophagus in a 71-year-old man. (A) An air-contrast image in the left
posterior oblique projection shows barium outlining subtle areas of mucosal irregularity (arrows) of this subtle, round 1-cm cancer
demonstrated en face. (B) A single-contrast, prone image shows a plaque-like lesion (arrow) in profile along the left side of the distal
esophagus.
FIGURE 6.19 Adenocarcinoma in the upper thoracic esophagus FIGURE 6.20 A 77-year-old man with grade 4 adenocarcinoma
in an 83-year-old man. This is a broad-based, eccentric mucosal within Barrett esophagus located at the esophagogastric junction
mass (arrowheads) with ulceration (arrow). Note the aspirated causing the appearance of secondary achalasia. Note the retained
barium in the trachea (curved arrow) related to the dysphagia barium in the dilated esophagus (curved arrow) above the fixed,
caused by the stricture resulting from this mass. narrowed esophagogastric junction from this nearly obstructive
carcinoma (arrows).
FIGURE 6.21 Squamous cell carcinoma of the upper thoracic esophagus in a 60-year-old man. (A) Upright esophagram image
demonstrates irregular luminal narrowing secondary to ulcerated mass (arrows) and a tracheoesophageal fistula (curved arrow). (B) Axial
computed tomography image with intravenous contrast material confirms the tracheoesophageal fistula (curved arrow) traversing the
malignant esophageal mass (arrows).
68 SECTION I Esophagus and Hernia
FIGURE 6.23 Adenocarcinoma of the distal esophagus within an area of Barrett esophagus in a 65-year-old man. (A) A single-contrast
esophagram shows an abruptly marginated (shouldered), ulcerated mass causing asymmetric, nearly occlusive, luminal narrowing (arrow).
(B) An axial computed tomography image with intravenous contrast material shows asymmetric esophageal wall thickening, confirmed by
the presence of oral contrast (arrow) in the eccentrically narrowed lumen. (C) An axial computed tomography image with intravenous
contrast material demonstrates celiac lymphadenopathy (arrows). Fine-needle aspiration of these nodes, with endoscopic ultrasound
guidance, confirmed metastatic adenocarcinoma.
45% (13/29) for CT.34 However, the PET sensitivity for not be as important but can help in locating metastases
detection of nodal metastasis has also been reported to be (e.g., in bone versus soft tissue). These issues make the
as low as 33% for local nodal disease in other studies.35–37 use of PET with CT fusion of significant importance when
In our experience, local nodal staging has been roughly performing PET imaging for esophageal cancer.
equivalent between EUS, CT, and PET when all referred PET can improve the staging of distant metastases. In
patients are included. However, in about 10% of cases, one one study of seven patients who did not undergo surgery,
imaging method does identify disease not demonstrated PET detected distant metastases that were not identified by
by the other. CT in five cases, and another patient had an unsuspected
Identifying distant metastatic disease has some important concomitant primary lung tumor discovered by PET alone.
caveats for PET. Relative to distant nodal disease, identifica- In another study of 35 patients with potentially resectable
tion of M1a disease can be difficult without the use of CT esophageal cancer by CT, PET identified distant metastatic
fusion imaging to provide anatomic guidance on location disease in 20%. The accuracy of PET in determining
of the celiac axis. For M1b disease, CT fusion with PET may distant metastatic disease in this group was 91%.38 Other
70 SECTION I Esophagus and Hernia
FIGURE 6.25 Adenocarcinoma of the right lung in a 78-year-old woman. (A) An esophagram shows a long extrinsic impression on the
right side of the midesophagus (arrows). (B) Axial computed tomography image with intravenous contrast material demonstrates bulky,
metastatic, mediastinal lymphadenopathy (arrows), displacing the esophagus (curved arrow) to the left and compressing it against the
aorta, is the cause of the esophagram finding.
http://surgerybook.net
72 SECTION I Esophagus and Hernia
Leaks can occur after any esophageal surgery, but they irregularity and mild protrusion of the pharyngoesophageal
are most common after esophagectomy. The development segment posteriorly are not worrisome findings.48
of pain and fever after esophagectomy warrants emergency Because the major complication of cricopharyngeal
esophagraphy43 performed initially with water-soluble myotomy is leak, the postoperative esophagram should be
contrast material. If this initial esophagram is negative, the performed initially with water-soluble contrast material.
examination should be immediately repeated with thick This contrast material needs to be administered cautiously,
barium. As a result of the greater radiographic density of because transient postoperative pharyngeal dysfunction
barium, small leaks may be diagnosed only with barium. predisposes these patients to aspiration (low-osmolar
In a retrospective study of 24 esophagectomy patients water-soluble contrast material can be considered for
with postoperative leaks, 16 (67%) of these leaks were these examinations). If the water-soluble contrast study
demonstrated only with the use of high-density (250% is negative, reexamination with high-density barium
weight per volume [w/v]) barium.46 The benefit of dem- should be performed. Leaks often appear as blind-ending
onstrating a leak often outweighs the risk for mediastinitis tracts extending from the esophagus posteriorly into the
secondary to mediastinal barium.47 Nevertheless, each prevertebral space.48
patient presents unique challenges. Therefore, these
difficult cases require close communication between Cardiomyotomy
radiologist and referring surgeon. After cardiomyotomy, the esophagram usually demonstrates
The risk for pulmonary edema after the aspiration of prompt esophageal emptying and a widely patent GEJ.42
water-soluble contrast material depends on the volume Eccentric ballooning of the esophageal mucosa through the
and osmolarity of the material aspirated. Aspiration of myotomy defect is a common finding (Fig. 6.29) and occurs
high-osmolar water-soluble contrast material, such as in 50% of patients after cardiomyotomy.49 Frequently, an
diatrizoate meglumine or diatrizoate sodium, is more likely antireflux procedure (usually partial fundoplication) is
to cause pulmonary edema than aspiration of a similar performed in conjunction with the cardiomyotomy, and
amount of low-osmolar water-soluble contrast material, such radiographic evidence of this procedure may also be
as iohexol. Therefore the use of low-osmolar water-soluble demonstrated on the postoperative esophagram.
contrast material should be considered in postoperative An early complication of cardiomyotomy is leak. Radio-
patients at risk for aspiration whose evaluation requires graphic evaluation for leak should begin with water-soluble
the use of water-soluble contrast material.47 esophagram. If this study is negative, it should be followed
by barium esophagram to more confidently exclude a
perforation. Late complications include dysphagia second-
SPECIFIC POSTOPERATIVE FINDINGS ary to inadequate myotomy or tight fundoplication.
Cricopharyngeal Myotomy
Cricopharyngeal myotomy is typically combined with Antireflux Procedures
Zenker diverticulectomy or diverticulopexy. Postopera- The esophagram after antireflux procedures demonstrates
tive esophagrams in successfully treated patients show reduction of esophageal hiatal hernia, restoration of an
resolution of the prominent cricopharyngeus muscle and intraabdominal esophageal segment, and gastric fundal
nonfilling of the diverticulum (Fig. 6.28). Mild mucosal wrap. Common antireflux surgeries include the Nissen,
Belsey Mark IV, and Hill procedures.42 The Nissen pro- Esophageal Resection
cedure results in a 360-degree wrap of the gastric fundus The radiographic appearance after esophagectomy depends
around the esophagus. Radiographically, the Nissen wrap on the bowel segment used as an esophageal substitute.
creates a smooth, symmetric, fundal soft-tissue pseudomass. Stomach, colon, and jejunum are used as esophageal sub-
The esophagus passes through the center of this pseudo- stitutes, with gastric substitution being the most common.
mass (Fig. 6.30). The Belsey Mark IV procedure uses a Gastric substitution requires resection of the esophagus
240-degree fundal wrap with suturing of the esophagus and cardia, mobilization of the stomach, and anastomosis
to the gastric fundus to recreate an acute angle (of His) of the esophagus to the stomach. Pyloromyotomy, or
at the GEJ. By esophagram, this procedure results in a pyloroplasty, and partial resection of the gastric fundus may
smaller soft tissue pseudomass in the fundus and angula- also be performed to facilitate drainage of the denervated
tion of the intraabdominal esophagus. During the Hill stomach.42 A normal postoperative esophagram should
procedure, the GEJ is sutured to the median arcuate demonstrate patency of the esophagogastrostomy (Fig.
ligament posteriorly. No fundoplication is performed. By 6.31), patency of the stomach as it passes through the
esophagraphy, this procedure results in lengthening of the esophageal hiatus, and patency of the pylorus.
intraabdominal esophagus and exaggeration of the angle Leak is the most feared early postoperative complication
of His. Regardless of the specific antireflux procedure, of esophagectomy and esophagogastrostomy. The leak
the esophagram should not demonstrate a hiatal hernia may occur at the esophagogastric anastomosis, at the
or evidence of reflux esophagitis.42 pyloroplasty or pyloromyotomy, or along the gastric staple
The most common early complication of fundoplica- line resulting from partial gastric resection.42 Pain and fever
tion demonstrated by esophagram is obstruction of the after esophagectomy warrant emergency esophagram47
distal esophagus secondary to edema of the fundal wrap. with water-soluble contrast material and, if necessary,
This process usually resolves in a matter of weeks. Late barium. High-density barium has been reported to be
complications include (1) esophageal obstruction caused more effective in demonstrating leaks.46
by a tight fundal wrap or tight esophageal hiatus, (2) Early postoperative obstruction may result from edema
recurrent hiatal hernia and GER caused by disruption at the esophagogastrostomy or pyloroplasty/pyloromyotomy
of fundoplication sutures (fundal pseudomass no longer sites. Obstruction may also result from diaphragmatic
visible), and (3) recurrent hiatal hernia (fundal pseudomass compression of the distal part of the stomach or from
remains visible) caused by dehiscence of diaphragmatic gastric volvulus.42 Gastric atony causes similar obstructive
sutures.42 symptoms.
Radiology of the Esophagus: Barium, Computed Tomography Scan, Positron Emission Tomography Scan, Magnetic Resonance Imaging CHAPTER 6 75
A B
FIGURE 6.31 Esophagogastrostomy. Upright, frontal (magnified) (A) and lateral air-contrast images (B) from a barium esophagram
performed 1 month after esophagectomy for T1N0 adenocarcinoma demonstrate an esophagogastric anastomosis (large arrows in both
images). A possible ulcerated mass along the left posterior margin of the gastrostomy, just distal to the anastomosis (small arrows in
both images), should represent a benign postoperative finding since the patient had no evidence of recurrent disease 10 months after
this esophagram.
MISCELLANEOUS CONDITIONS
HIATAL HERNIAS
Although the correlation between GER and sliding hiatal
hernias is far from perfect, such hernias are thought to
predispose to GER.50 Hiatal hernias can be classified
into several types, depending on their esophagraphic
appearance. By far, the most common type is a sliding
hiatal hernia, characterized by superior migration, often
transient, of the GEJ into the chest (Fig. 6.33).
The second major type of hiatal hernia (Fig. 6.34) is
a paraesophageal hernia, in which the GEJ remains near
the esophageal hiatus, and all or part of the stomach
herniates through the esophageal hiatus to lie adjacent FIGURE 6.32 Stricture of esophagogastrostomy. An upright, frontal
to the esophagus (paraesophageal). Such hernias are air-contrast view from a barium esophagram was performed 6
important to recognize because they are more likely than weeks after esophagectomy for T2N0 adenocarcinoma of the
sliding hernias to be associated with symptomatic gas esophagus. A partially obstructing anastomotic stricture (large
entrapment, obstruction, incarceration, and strangula- arrows) secondary to chronic reflux esophagitis is causing
tion. These complications are more common with large aspiration of barium into the trachea (small arrows).
76 SECTION I Esophagus and Hernia
FIGURE 6.39 Single-contrast esophagram (frontal view) shows FIGURE 6.40 Single-contrast esophagram, left oblique view, from
ectopic gastric mucosa. Two indentations of the right lateral an 86-year-old man with cicatricial pemphigoid causing a lengthy
aspect of the cervical esophageal lumen (arrows) result from a stricture of moderate severity of the hypopharynx and cervical
patch of ectopic gastric mucosa confirmed by endoscopic biopsy. esophagus.
FIGURE 6.42 Single-contrast esophagram in a 26-year-old man FIGURE 6.43 Single-contrast esophagram in a 38-year-old man
with mediastinal histoplasmosis. Extrinsic compression of the with a long history of dysphagia and multiple food impactions. A
midesophagus by mediastinal lymphadenopathy mimics an midesophageal stricture with corrugated margins results from
intrinsic esophageal stricture. biopsy-proven eosinophilic esophagitis.
CAUSTIC INJURY
Caustic esophagitis usually results from ingestion of a liquid
solution of concentrated lye. The severity of esophageal
injury depends on the volume and concentration of the
caustic agent and the duration of mucosal contact.62 Mild
injuries may be confined to the mucosa and heal without
sequelae. Severe injuries may result in esophageal perfora-
tion, mediastinitis, and death. Patients who survive severe
injury are typically left with long, irregular strictures of the
midesophagus. The entire esophagus may be affected, with
marked narrowing of the lumen producing a threadlike
appearance.63
Less severe injury to the esophagus may result from
the ingestion of other household products, including
ammonium chloride. Also, a variety of medications,64 such FIGURE 6.44 Single-contrast, recumbent image demonstrating
as tetracycline, doxycycline, potassium chloride, quinidine, smooth, diffuse narrowing of the midesophagus, with maximum
nonsteroidal antiinflammatory drugs, and alendronate diameter of 14 mm (arrows), also results from biopsy-proven
sodium (Fosamax) can cause esophageal strictures. eosinophilic esophagitis.
80 SECTION I Esophagus and Hernia
ESOPHAGEAL PERFORATION
Esophageal perforation may result from blunt or penetrat-
ing chest trauma, foreign body ingestion, instrumentation,
or caustic ingestion. Spontaneous esophageal perforation
(Boerhaave syndrome) results from a sudden increase
in intraluminal pressure, usually from extreme vomiting
or retching, classically occurring after alcoholic binge
drinking.
Regardless of the cause, esophageal perforations are
potentially life threatening and require immediate atten-
tion. Localized perforations, especially of the cervical
esophagus, may be managed nonoperatively, but perfora-
tions of the thoracic esophagus almost always require
surgical intervention.65
Plain film findings of esophageal rupture include retro-
pharyngeal gas, cervical subcutaneous emphysema, widen-
ing of the mediastinum, pneumomediastinum, pleural
effusion, and hydropneumothorax (more commonly on
the left; Fig. 6.47A). Because plain films are relatively
insensitive and nonspecific, contrast esophagrams should
be used early in the investigation of clinically suspected
esophageal perforations (see Fig. 6.47B). Water-soluble
contrast agents, swallowed or injected through a nasogastric
tube, are the agents of first choice. For patients at risk for
aspiration (or fistula to the airway), low-osmolar agents
are less likely to cause pulmonary edema, if aspirated,
FIGURE 6.45 Double-contrast esophagram, in left posterior oblique than high-osmolar agents.
projection, in a 25-year-old man. Irregularity and ulceration of the When water-soluble agents leak into the mediastinum,
lower esophagus are nonspecific. These findings could result from they are rapidly absorbed and do not incite an inflamma-
an esophageal mass such as an adenocarcinoma. However, in tory response, conferring a margin of safety over barium
this young male, they resulted from esophageal Crohn disease. contrast agents, which are nonabsorbable and may incite
foreign body granuloma formation.66,67 Nevertheless, a
negative esophagram with water-soluble contrast should
be immediately followed by a barium esophagram, which
because of its higher density, has been reported to increase
the sensitivity for detecting esophageal leaks by 15% to
25%.68,69 The low risk for mediastinal complications from
barium extravasation is more than offset by the benefits
of earlier diagnosis.
Chest CT is more sensitive in detecting pneumo-
mediastinum than plain films are, and is useful after a
negative contrast esophagram in high-risk patients or
when contrast esophagrams are difficult to perform in
seriously ill patients. With modern scanners, chest CT can
be combined with contrast esophagraphy to expedite the
diagnosis of esophageal rupture.70
DIVERTICULA
Esophageal diverticula vary greatly in size, shape, location,
cause, and significance. Even incidentally discovered
diverticula are important to document, because they may
predispose the patient to injury during instrumentation.71
Traditionally, esophageal diverticula are classified as
either traction diverticula, occurring primarily in the
midesophagus, or pulsion diverticula, occurring primar-
ily in the upper or lower esophagus. In reality, many
midesophageal diverticula are pulsion type,72 caused by
increased intraluminal pressure resulting in “ballooning”
FIGURE 6.46 Upright image from single-contrast esophagram in a of mucosal and submucosal layers through localized weak
78-year-old woman with Crohn disease and a midesophageal areas of the esophageal wall. True traction diverticula are
stricture causing complete obstruction. recognized by elongation, or “tenting,” of the diverticulum
Radiology of the Esophagus: Barium, Computed Tomography Scan, Positron Emission Tomography Scan, Magnetic Resonance Imaging CHAPTER 6 81
A B
FIGURE 6.47 (A) Anteroposterior portable chest radiograph in a patient with Boerhaave syndrome and left apical pneumothorax (arrows).
Diffusely increased density of the left hemothorax results from a pleural effusion. (B) Supine image from water-soluble contrast upper
gastrointestinal examination shows a left pleural effusion (closed arrows). A retrocardiac, mediastinal collection of gas and leaked contrast
material (open arrows) indicates esophageal rupture.
VARICES
Although less sensitive than endoscopy, barium esopha-
gram may demonstrate esophageal varices as undulating,
serpentine, and sometimes nodular filling defects, often FIGURE 6.48 Double-contrast esophagram demonstrating a
transient. They are more commonly demonstrated in the midesophageal traction diverticulum. Note the elongated, or
lower esophagus as the result of portal venous hypertension “tented,” appearance of the diverticulum.
82 SECTION I Esophagus and Hernia
SUMMARY
This chapter has described the essential technical elements
of the esophagram and its normal structural and functional
findings, as well as common artifacts and normal variants;
the efficiency of the esophagram for the investigation of
structural and functional causes of dysphagia (in conjunc-
tion with a video swallow study when an oropharyngeal
cause of dysphagia is suspected); the appropriateness of the
esophagram for evaluating the postoperative esophagus; the
detection of benign and malignant esophageal neoplasms;
and the role of cross-sectional imaging, especially PET/CT,
for staging of esophageal cancer. Specific esophagraphic
FIGURE 6.50 Lateral, upright, single-contrast image of
hypopharynx and cervical esophagus shows a Zenker diverticulum
findings such as hiatal hernias, esophageal rings and webs,
immediately superior to prominent cricopharyngeus muscle less common types of esophageal strictures, caustic injury,
causing extrinsic compression of the pharyngoesophageal lumen. and esophageal perforation, diverticula, and varices have
also been described and illustrated.
The barium esophagram remains a valuable test in the
secondary to hepatic cirrhosis. These varices represent 21st century. The esophagram is complementary to CT,
portosystemic venous shunts that allow portal venous PET, MRI, endoscopy, EUS, high-resolution manometry,
blood to return to the right heart, despite increased pH monitoring, and multichannel intraluminal imped-
resistance to intrahepatic portal venous blood flow second- ance. These complementary esophageal examinations are
ary to cirrhosis. These distal esophageal varices are often especially beneficial to patients with esophageal disease
referred to as uphill varices because of the inferior to when the radiologists and other physicians and surgeons
superior direction of blood flow inside them (Fig. 6.51). ordering and performing them collaborate. This team
Rarely, varices may be demonstrated in the upper part approach to the diagnosis and management of esophageal
of the esophagus; these varices represent venous shunts diseases also increases the understanding of these diseases
between the superior and inferior vena cava secondary to by all collaborating physicians.
Radiology of the Esophagus: Barium, Computed Tomography Scan, Positron Emission Tomography Scan, Magnetic Resonance Imaging CHAPTER 6 83
40. Flamen P, Lerut A, Van Cutsem E, et al. The utility of positron 57. Naylor MF, MacCarty RL, Rogers RS 3rd. Barium studies in esophageal
emission tomography for the diagnosis and staging of recurrent cicatricial pemphigoid. Abdom Imaging. 1995;20:97.
esophageal cancer. J Thorac Cardiovasc Surg. 2000;120:1085. 58. Dominiquez R, Zarabi M, Oh KS, Bender TM, Girdany BR. Congenital
41. Miettinen M, Sarlomo-Rikala M, Sobin LH, Lasota J. Esophageal esophageal stenosis. Clin Radiol. 1985;36:263.
stromal tumors: a clinicopathologic, immunohistochemical, and 59. Pokieser P, Schima W, Schober E, et al. Congenital esophageal
molecular genetic study of 17 cases and comparison with esophageal stenosis in a 21-year-old man: clinical and radiographic findings.
leiomyomas and leiomyosarcomas. Am J Surg Pathol. 2000;24:211. AJR Am J Roentgenol. 1998;170:147.
42. Rubesin S, Williams N. Postoperative esophagus. In: Gore RM, Levine 60. Croese J, Fairley SK, Masson JW, et al. Clinical and endoscopic
MS, eds. Textbook of Gastrointestinal Radiology. 2nd ed. Philadelphia: features of eosinophilic esophagitis in adults. Gastrointest Endosc.
WB Saunders; 2000:495. 2003;58:516.
43. Orringer M. Complications of esophageal surgery. In: Orringer M, 61. Vasilopoulos S, Murphy P, Auerbach A, et al. The small-caliber
Heitmiller R, eds. Shackelford’s Surgery of the Alimentary Tract. 5th ed. esophagus: an unappreciated cause of dysphagia for solids in patients
Philadelphia: WB Saunders; 2002:443. with eosinophilic esophagitis. Gastrointest Endosc. 2002;55:99.
44. Kim SH, Lee KS, Shim YM, Kim K, Yang PS, Kim TS. Esophageal 62. Goldman LP, Weigert JM. Corrosive substance ingestion: a review.
resection: indications, techniques, and radiologic assessment. Am J Gastroenterol. 1984;79:85.
Radiographics. 2001;21:1119; discussion 1138. 63. Franken EA. Caustic damage of the gastrointestinal tract: Roentgen
45. Maher M, Lucey BC, Boland G, Gervais DA, Mueller PR. The features. AJR Am J Roentgenol. 1973;118:77.
role of interventional radiology in the treatment of mediastinal 64. Bova JG, Dutton NE, Goldstein HM, Hoberman LJ. Medication-
fluid collections caused by esophageal anastomotic leaks. AJR Am J induced esophagitis: diagnosis by double-contrast esophagography.
Roentgenol. 2002;178:649. AJR Am J Roentgenol. 1987;148:731.
46. Swanson JO, Levine MS, Redfern RO, et al. Usefulness of high- 65. Port JL, Kent MS, Korst RJ, Bacchetta M, Altorki NK. Thoracic
density barium for detection of leaks after esophagogastrectomy, esophageal perforations: a decade of experience. Ann Thorac Surg.
total gastrectomy, and total laryngectomy. AJR Am J Roentgenol. 2003;75:1071.
2003;181:415. 66. James AE, Montali RJ, Chaffee V, Strecker EP, Vessal K. Barium
47. Levine MS. Miscellaneous Abnormalities of the Esophagus. Philadelphia: or Gastrografin: which contrast media for diagnosis of esophageal
WB Saunders; 2000:465. tears? Gastroenterology. 1975;68:1103.
48. Sydow BD, Levine MS, Rubesin SE, Laufer I. Radiographic findings 67. Vessal K, Montali RJ, Larson SM, Chaffee V, James AE Jr. Evaluation
and complications after surgical or endoscopic repair of Zenker’s of barium and Gastrografin as contrast media for the diagnosis
diverticulum in 16 patients. AJR Am J Roentgenol. 2001;177:1067. of esophageal ruptures or complications. AJR Am J Roentgenol.
49. Rubesin SE, Kennedy M, Levine MS, Rosato EF, Laufer I. Distal 1975;123:307.
esophageal ballooning following Heller myotomy. Radiology. 68. Buecker A, Wein BB, Neuerburg JM, Guenther RW. Esophageal
1988;167:345. perforation: comparison of use of aqueous and barium-containing
50. Ott DJ, Gelfand DW, Chen YM, Wu WC, Munitz HA. Predictive contrast media. Radiology. 1997;202:683.
relationship of hiatal hernia to reflux esophagitis. Gastrointest Radiol. 69. Foley MJ, Ghahremani GG, Rogers LF. Reappraisal of contrast
1985;10:317. media used to detect upper gastrointestinal perforations. Radiology.
51. Hill LD. Incarcerated paraesophageal hernia: a surgical emergency. 1982;144:213.
Am J Surg. 1973;126:286. 70. Fadoo F, Ruiz DE, Dawn SK, Webb WR, Gotway MB. Helical CT
52. Dunn DB, Quick G. Incarcerated paraesophageal hernia. Am J Emerg esophagography for the evaluation of suspected esophageal perfora-
Med. 1990;8:36. tion or rupture. AJR Am J Roentgenol. 2004;182:1177.
53. Schatzki RGJ. Dysphagia due to diaphragm-like localized narrowing in 71. Nutter KM, Ball OG. Esophageal diverticula: current classification
the lower esophagus (lower esophageal ring). Radiology. 1953;70:911. and important complications. J Miss State Med Assoc. 2004;45:131.
54. Waldenström J, Kjeulberg SR. The roentgenological diagnosis of 72. Schima W, Schober E, Stacher G, et al. Association of mid esophageal
sideropenic dysphagia (Plummer-Vinson’s syndrome). Acta Radiol. diverticula with esophageal motor disorders: videofluoroscopy and
1939;20:618. manometry. Acta Radiol. 1997;38:108.
55. Chisholm M. The association between webs, iron and postcricoid 73. Fasano NC, Levine MS, Rubesin SE, Redfern RO, Laufer I. Epiphrenic
carcinoma. Postgrad Med J. 1974;50:215. diverticulum: clinical and radiographic findings in 27 patients.
56. Mauro MA, Parker LA, Hartley WS, Renner JB, Mauro PM. Epi- Dysphagia. 2003;18:9.
dermolysis bullosa: radiographic findings in 16 cases. AJR Am J 74. Perrott JW. Anatomical aspects of hypopharyngeal diverticula. Aust
Roentgenol. 1987;149:925. N Z J Surg. 1962;31:307.
CHAPTER
Endoscopic Evaluation of the Esophagus and
Endoscopic Ultrasonography of the Esophagus 7
Daniel S. Oh
| Stuart Jon Spechler
| Jacques Bergman
|
Thomas W. Rice
| Gregory Zuccaro Jr.
T
he endoscopist who examines the esophagus evalu- and rare instances of adenocarcinoma have been described
ates a muscular tube whose primary function is developing from an inlet patch.
to convey swallowed material from the mouth to Within the chest at about the T4 level, the esophagus is
the stomach. The esophagus is approximately 25 cm in indented on its left side by the aortic arch. This pulsating
length measured from its origin in the neck just below indentation can be noted during endoscopic examination
the cricoid cartilage (C6 level, approximately 15 cm at a distance of approximately 23 cm from the incisor teeth
from the incisor teeth as measured by the endoscopist) (Fig. 7.2).5 Just below the arch at approximately 25 cm,
to its termination in the abdomen at the gastric cardia the left main bronchus causes a subtle indentation on
(T10 to T11 level, approximately 40 cm from the incisor the left anterior aspect of the esophagus (see Fig. 7.2).
teeth).1 Proximally, the upper esophageal sphincter (UES) Below the bronchus, the esophagus abuts the left atrium.
separates the pharynx from the esophagus. The UES The heart normally causes no prominent indentation of
extends approximately 3 cm in length and comprises the esophageal lumen, but atrial pulsations often can
three skeletal muscle groups including the distal portion be visualized at a level approximately 30 cm from the
of the inferior pharyngeal constrictor, the cricopharyngeus, incisor teeth.
and the circular muscle of the proximal esophagus. 2
Introduction of the endoscope into the UES often causes
gagging, and the muscles relax only briefly during a ENDOSCOPIC EVALUATION OF THE
swallow. Consequently, the endoscope typically is passed GASTROESOPHAGEAL JUNCTION
quickly through the UES, and endoscopic visualization
of its mucosal lining often is limited. The gastroesophageal junction (GEJ) is the level at which
The esophagus passes from the chest into the abdomen the esophagus ends and the stomach begins. Unfortunately,
through the diaphragmatic hiatus, a canal-shaped opening there are no universally accepted landmarks that clearly
in the right crus of the diaphragm. Approximately 2 cm delimit the distal esophagus and the proximal stomach,
of the distal esophagus normally lie within the abdomen.3 and the GEJ has been defined differently by anatomists,
The lower esophageal sphincter (LES) comprises both radiologists, physiologists, and endoscopists.6 Landmarks
the skeletal muscle of the crural diaphragm (external suggested by anatomists, such as the peritoneal reflection
LES muscle) and the circular smooth muscle of the or the character of the muscle bundles in the esophageal
distal esophagus itself (internal LES muscle), although wall, are not useful for endoscopists. Radiologists refer to
endoscopists often refer to only the latter when describing the region of the GEJ as the vestibule, and they seldom
the LES. Unlike the UES, endoscopic examination of the attempt to localize the precise point at which the esophagus
LES region generally is not limited either by sustained joins the stomach.7 Physiologists have used the distal border
sphincter muscle contraction or by patient discomfort. of the LES (determined manometrically) to define the
The esophageal lumen is collapsed at rest and must be GEJ,8 but it is not feasible to identify this border precisely
distended with air during endoscopy so that the stratified by endoscopic techniques. Indeed, one study has shown
squamous epithelial lining can be visualized well. When so that manometric and endoscopic localizations of the LES
distended, the squamous epithelium appears pale, glossy, often differ by several centimeters.9 This has implications
and relatively featureless. In the proximal esophagus, for placement of a wireless pH capsule endoscopically
within a few centimeters of the UES, it is common to based on the GEJ versus based on manometry because
find patches of columnar epithelium that have a reddish the location and consequently the findings may differ
color and velvetlike texture similar to the epithelium of based on technique (see later).
the stomach (Fig. 7.1).4 These so-called inlet patches When considering any proposed landmark for the GEJ,
are believed to be congenital rests of heterotopic gastric it is important to appreciate that there is no clear-cut
epithelium. They are often overlooked during routine “gold standard” for the structure and, consequently, all
endoscopic examinations, but, if sought specifically, they of the suggested landmarks can be considered arbitrary.
can be found in up to 11% of patients who have endoscopic For most disorders of the esophagus and stomach that are
examinations. Inlet patches usually are of no clinical diagnosed endoscopically, furthermore, it is not important
importance, but they can produce acid and, in rare cases, that the GEJ be identified with great precision. For some
can cause peptic ulcerations in the proximal esophagus. In disorders, most notably Barrett esophagus, for which the
addition, they occasionally contain intestinal metaplasia, endoscopist must determine the extent of esophageal
85
Endoscopic Evaluation of the Esophagus and Endoscopic Ultrasonography of the Esophagus CHAPTER 7 85.e1
ABSTRACT
In this chapter the fundamentals of esophagogastroduo-
denoscopy (EGD) will be explained with an emphasis on
the assessment of patients with gastroesophageal reflux
disease (GERD). It includes the fundamentals of endo-
scopic ultrasound (EUS) in the assessment of esophageal
cancer. Numerous color photographs demonstrate different
findings of interest on EGD through the spectrum of
GERD, from erosive esophagitis to cancer. Photographs
of EUS images will help demonstrate findings of interest
in cancer patients.
KEYWORDS
EGD, EUS, GERD, esophageal cancer
86 SECTION I Esophagus and Hernia
Esophagogastric
junction
Squamocolumnar
junction
Columnar-lined
esophagus
FIGURE 7.4 Endoscopic photograph of the gastroesophageal FIGURE 7.6 Palisade vessels in the distal esophagus are fine,
junction region in a patient who has a hiatal hernia. The longitudinal veins in the lamina propria. The distal end of the
squamocolumnar junction is located above some of the gastric palisade vessels has been proposed as an endoscopic landmark
folds (i.e., there is a columnar-lined segment of esophagus), for the gastroesophageal junction.
whereas for others the squamocolumnar junction seems to
coincide with the proximal extent of the folds.
vessels pierce the muscularis mucosae distally to join the
submucosal vessels of the gastric zone and proximally to
join the submucosal vessels of the perforating zone. The
palisade vessels can be difficult to visualize by conventional
endoscopy, especially if there is inflammation in the
distal esophagus. The appearance of these vessels can be
enhanced by narrow band imaging endoscopy, which uses
primarily blue light that penetrates only the superficial
layers of the mucosa (where the palisade vessels are found)
and that is absorbed by the hemoglobin within the vessels.
Furthermore, even in autopsy studies in which blood vessels
of the GEJ region are injected with resins that provide
exquisite detail of the venous structures, it is difficult to
identify precisely the termination of the palisade vessels.15
Finally, it is not clear conceptually why the distal end of
the palisade vessels should be considered the precise end
of the esophagus.
Few studies have addressed specifically the problem of
endoscopic localization of the GEJ and, even in those that
have done so, the accuracy of the criteria used cannot be
assessed meaningfully in the absence of a gold standard.
It is not clear which is the best diagnostic criterion for
FIGURE 7.5 Endoscopic photograph of the gastroesophageal the GEJ, and the reproducibility of the various criteria
junction region in a patient with long-segment Barrett esophagus. have not been established. If one cannot determine with
Columnar epithelium extends above the tops of the gastric folds certainty where the esophagus ends and the stomach
to involve the distal esophagus in a circumferential fashion. begins, then any assessment of the extent of esophagus
lined by columnar epithelium will be inherently imprecise.
This unresolved problem continues to confound clinicians
A number of Asian investigators use the end of the and investigators who deal with Barrett esophagus.
esophageal palisade vessels as their landmark for the
GEJ (Fig. 7.6).13 Elegant anatomic studies of the GEJ
have revealed four distinct zones of venous drainage, CONVENTIONAL ENDOSCOPIC DIAGNOSIS
including a gastric zone, a palisade zone, a perforating OF BARRETT ESOPHAGUS
zone, and a truncal zone.15 The palisade zone comprises
a group of fine, longitudinal veins located largely within Barrett esophagus is the condition in which metaplas-
the lamina propria of the distal esophagus. The palisade tic columnar epithelium that predisposes to cancer
88 SECTION I Esophagus and Hernia
FIGURE 7.7 In this drawing, the gastroesophageal junction and the FIGURE 7.8 In this patient with long-segment Barrett esophagus,
Z-line coincide, and there is no columnar-lined segment of the Z-line is relatively smooth.
esophagus. (Reprinted with permission from Spechler SJ. The role
of gastric carditis in metaplasia and neoplasia at the
gastroesophageal junction. Gastroenterology. 1999;117:218.)
FIGURE 7.11 Magnification endoscopy of mucosa sprayed with FIGURE 7.13 Magnification endoscopy of the region shown in Fig.
acetic acid showing the pit pattern of columnar epithelium at the 7.11 combined with narrow band imaging.
squamocolumnar junction. The relatively featureless squamous
epithelium is seen adjacent to the columnar epithelium in the
upper left corner of the slide. make a diagnosis of reflux esophagitis. However, more
than 50% of patients who have typical GERD symptoms
have normal endoscopic examinations.33,34 Thus it appears
that GERD usually does not cause visible damage to the
esophageal mucosa in most patients.
Mild changes of GERD that may be visible to the
endoscopist include mucosal erythema, edema, hyper-
vascularity, friability, and blurring of the SCJ. However,
identification of those changes is a subjective skill, and
agreement among endoscopists regarding the presence
of such minimal signs of reflux esophagitis can be very
poor.35,36 More severe GERD can result in esophageal
erosions and ulcerations. Histologically, erosions are
defined as superficial necrotic defects that do not penetrate
the muscularis mucosae, whereas ulcerations are deeper
defects that extend through the muscularis mucosae into
the submucosa.37 Endoscopically, these peptic esophageal
lesions are identified on the basis of their gross features,
and clinicians seldom have histologic confirmation that the
lesions they call “esophageal ulcers” in fact have breached
the muscularis mucosae. Thus the distinction between
an esophageal ulceration and an erosion usually is based
FIGURE 7.12 Magnification endoscopy of the region shown in Fig.
on a subjective assessment of the depth of the necrotic
7.11 after application of indigo carmine dye. lesion. One modern system for grading the severity of
reflux esophagitis, the Los Angeles classification, avoids
the problem of distinguishing erosions from ulcerations
by referring to both as “mucosal breaks.”38
ENDOSCOPIC DIAGNOSIS OF More than 30 systems for the classification of reflux
REFLUX ESOPHAGITIS esophagitis have been proposed over the past few decades.38
The endoscopic criteria for three of the most widely used
Gastroesophageal reflux disease (GERD) has been defined systems are listed in Table 7.1.36,38,39 All of the proposed
as the condition that develops when the reflux of stomach systems have limitations, and no one system has been
contents causes troublesome symptoms and/or complica- shown to be clearly superior to another for establishing
tions.32 Heartburn is the most common symptom of GERD, the diagnosis of GERD or for predicting the response to
and tissue injury results when esophageal epithelial cells treatment. Arguably the best validated and most widely
succumb to the damaging effects of the refluxed acid used system now is the Los Angeles classification that
and pepsin. When these caustic agents cause macroscopic was proposed at the World Congress of Gastroenterol-
injury to the esophageal epithelium, the endoscopist can ogy meeting in Los Angeles in 1994.38 In this system a
Endoscopic Evaluation of the Esophagus and Endoscopic Ultrasonography of the Esophagus CHAPTER 7 91
A B
FIGURE 7.14 (A) Endoscopic photograph of Los Angeles grade B esophagitis. There is a mucosal break defined as “an area of slough or
erythema with a discrete line of demarcation from the adjacent, more normal-looking mucosa.” Notice the whitish exudates covering the
mucosal break, which is greater than 5 mm in length. In addition, there is scarring of the distal esophagus, indicated by the fibrous
strands that run perpendicular to the mucosal break at the 12- and 5-o’clock positions. (B) Same area as shown in (A) after the whitish
exudates have been washed off. The mucosal break is still visible but less prominent.
EOSINOPHILIC ESOPHAGITIS
Eosinophilic esophagitis (EoE) is a modern esophageal
disorder that has become recognized widely only within
the past decade.41,42 EoE appears to be a manifestation of
food allergy in which eosinophils infiltrate the esophageal
epithelium, causing symptoms and tissue damage mediated
by cytokines released from the eosinophils and surrounding
tissues. The disorder commonly is diagnosed in men in
the fourth and fifth decades of life who describe a long
history of dysphagia for solid foods, often with hospital
visits for food impactions. Heartburn is also a common
FIGURE 7.18 Endoscopic photograph of a paraesophageal hernia, complaint, and it can sometimes be difficult to distinguish
retroflexed view. The herniated pouch of stomach is located next EoE from GERD. Patients frequently have a personal
to the fundoplication folds. (Reproduced with permission from and family history of allergic disorders such as asthma,
Spechler SJ. The management of patients who have “failed” atopic dermatitis, eczema, hay fever, and food allergies.
antireflux surgery. Am J Gastroenterol. 2004;99:552.) Children with EoE may have symptoms of abdominal
94 SECTION I Esophagus and Hernia
A B
FIGURE 7.21 Adenocarcinoma of the gastroesophageal junction photographed from the esophageal side (A) and from the gastric side (B).
If there is no Barrett epithelium seen in the esophagus, it is not possible to determine whether such a tumor originated from the distal
esophagus or from the gastric cardia.
a rate at which the eye cannot detect single images (12/ A radial mechanical blind probe (Fig. 7.25) is avail-
second). This fast-frame display is called real-time ultrasound able for the evaluation of esophageal strictures. This
and allows the ultrasonographer to study tissue temporally echoendoscope provides images similar to those of larger-
as well as spatially. diameter radial mechanical echoendoscopes, but it has
no endoscopic optical capabilities and is less than 8 mm
in diameter. More commonly used in current practice
INSTRUMENTS AND TECHNIQUES are higher-frequency miniprobes passed through the
Because EUS does not provide adequate endoscopic inspec- operating channel of standard endoscopes (Fig. 7.26);
tion of the upper gastrointestinal tract, every ultrasound these miniprobes provide radial images from 12 to 30 MHz.
study should be preceded by a standard flexible endoscopic These three instruments are used in conjunction with
upper gastrointestinal examination. This provides precise an image processor (Fig. 7.27). The image processor
location and mucosal definition (including biopsy) of the allows for adjustment of gain, contrast, and sensitivity time
esophageal lesion and guides the ultrasound examiner. control to regulate the strength of the returning echo at
The ultrasound endoscope is generally passed blindly different depths. Onscreen calibration and labeling can
through the oropharynx and hypopharynx. Care must be be done with the image processor. The image may be
taken because the distal tip containing the transducer is displayed on a video monitor or stored digitally or on
rigid. For complete examination, the endoscope must be videotape. The image processor has been refined and
passed beyond the esophagus into the stomach.
In the past, the radial mechanical ultrasound endoscope
(Fig. 7.24) was the principal instrument used for EUS.
The ultrasound transducer is housed in the tip of the
endoscope. It produces up to a 360-degree sector scan
perpendicular to the transducer tip. Because the transducer
is adjacent to tissues to be examined, higher frequencies
than those used in extracorporeal ultrasound can be used.
In the newest models, a range of transducer frequencies,
from 5 to 20 MHz, are available. These transducers allow
adequate visualization of anatomic structures to a depth
of 3 to 12 cm. An acceptable acoustic interface between
the transducer and the tissue being examined must be
obtained to ensure good-quality ultrasound images. This
is most commonly accomplished by covering the tip of the
endoscope with a latex balloon, which can be filled with
water to provide an excellent acoustic interface (see Fig.
7.24). A less commonly used technique is rapid insufflation
of the esophageal lumen with water. This provides an
excellent, but transient acoustic interface without the tissue
compression that may occur with the latex balloon. Current
echoendoscopes also provide a video endoscopic image,
albeit with a somewhat limited view in a forward oblique
direction. The control section contains the deflection FIGURE 7.24 The Olympus GF-UM130 ultrasound endoscope.
controls and air/water and suction valves, similar to those Upper left inset, The control section contains the deflection
on a standard endoscope (see Fig. 7.24). A water inflation/ controls and air/water and suction valves similar to those on a
deflation system for the balloon is incorporated into the standard endoscope. Upper right inset, The ultrasound transducer
air/water and suction valve mechanisms. A direct-current is housed in the tip of the endoscope. The forward oblique
motor and drive mechanism that rotates the ultrasound viewing endoscope and suction channel are proximal to the
transducer are housed in the control section. Current ultrasound transducer. Lower right inset, The distal tip of the
ultrasound endoscopes are totally immersible in liquids. ultrasound endoscope with the water-inflated contact balloon,
which covers the ultrasound transducer.
A B
FIGURE 7.26 (A) High-frequency (12 to 30 MHz) miniprobe passed through the operating channel of a standard endoscope. (B) Miniprobe
ultrasound image of a normal esophagus. The probe is not centered in the nondistended esophageal lumen. The mucosa and
submucosa are the inner hyperechoic layer. The muscularis propria (arrows) is the inner hypoechoic layer.
FIGURE 7.27 (A) The Olympus EU-M20 image processor (lower arrowhead) is rack-mounted in a standard cart, which includes the other
essential endoscopic equipment. The keyboard (upper arrowhead) can be used to measure and mark ultrasound findings. (B) The
complete system includes the light source rack, image processor, and ultrasound endoscope.
pierce the muscularis propria and drain into regional wall may be apparent because the superficial mucosa,
lymphatics or directly into the thoracic duct. deep mucosa, and submucosa compose one hyperechoic
The esophagus has no investing adventitia. The para- layer. The thickness of each ultrasound layer is about
esophageal tissue is composed of fibrofatty tissue that lies equal and does not represent the thickness of the tissue
directly against the outer fibers of the muscularis propria. layer but, instead, the time that it takes the ultrasound
The normal esophagus is usually viewed as five discrete wave to traverse this layer.
layers by EUS (Fig. 7.32). These layers are seen as alternat-
ing hyperechoic (white) and hypoechoic (black) rings.
Studies demonstrate that the five layers seen by EUS
ESOPHAGEAL CARCINOMA
correspond to the balloon-mucosa interface, the mucosa Staging of cancer of the esophagus and esophagogastric
deep to this interface, the submucosa and the acoustic junction (EGJ) has been extensively changed and improved
interface between the submucosa and muscularis propria, in the seventh edition of the American Joint Committee
the muscularis propria minus the acoustic interface on Cancer/International Union Against Cancer (AJCC/
between the submucosa and the muscularis propria, and UICC) Cancer Staging Manual (Box 7.1).47 Changes
the periesophageal tissue.45,46 For clinical purposes, these address problems of empiric stage grouping and lack of
layers represent the superficial mucosa, deep mucosa, harmonization with stomach cancer. This was accomplished
submucosa, muscularis propria, and periesophageal tissue. by assembling worldwide data and using modern machine
In the upper part of the esophagus, with overdistention of learning techniques for data-driven staging.48–51 Improve-
the examining balloon or if the transducer is too close to ments include new definitions of Tis, T4, regional lymph
the esophageal wall, only three layers of the esophageal node, N classification, and M classification, and addition
Endoscopic Evaluation of the Esophagus and Endoscopic Ultrasonography of the Esophagus CHAPTER 7 99
FIGURE 7.28 The Olympus GF-UE160 electronic radial echoendoscope. The electronic radial echoendoscope provides an enhanced
image as a result of use of tissue harmonic echo and can provide color and power Doppler, not available in mechanical radial design.
Inset, The tip of the Olympus GF-UE160 electronic radial echoendoscope with water-filled balloon. This tip is easier to maneuver
endoscopically compared with prior models.
FIGURE 7.29 The Olympus GF-UC140P convex scanning linear echoendoscope. This endoscope has a high-resolution CCD (charge-
coupled device) chip that provides outstanding optics and four imaging frequencies (5 to 10 MHz). It is shown with the Olympus EZ Shot
aspiration needle. Inset, The tip of the Olympus GF-UC140P echoendoscope. Like the electronic radial echoendoscope, insertion and
maneuverability are improved compared to previous iterations.
of the nonanatomic cancer characteristics: histopathologic includes all noninvasive neoplastic epithelium that was
cell type, histologic grade, and tumor location. Stage previously called carcinoma in situ. T1 tumors breach
groupings were constructed by adherence to principles the basement membrane to invade the lamina propria,
of staging, including monotonic decreasing survival with muscularis mucosae, or the submucosa but do not invade
increasing stage group, distinct survival between groups, beyond the submucosa. T1 tumors may be subclassified
and homogeneous survival within groups. into T1a, tumors that invade only the mucosa, and T1b,
Depth of tumor invasion classifies the primary tumor tumors that invade the submucosa.52 T2 tumors invade
(T). Tis tumors are intraepithelial malignancies confined into but not beyond the muscularis propria. T3 tumors
to the epithelium without invasion of the basement invade beyond the esophageal wall into the periesophageal
membrane and are now termed high-grade dysplasia. Tis tissue but do not invade adjacent structures. T4 tumors
100 SECTION I Esophagus and Hernia
Epithelium
Basement membrane
Lamina propria
Muscularis mucosae
Lymphatics Submucosal
Submucosal gland
Submucosal Adventitia
Muscularis
propria
Regional
Thoracic duct
FIGURE 7.31 The esophageal wall is composed of mucosa, submucosa, and muscularis propria. The mucosa is composed of epithelium,
lamina propria, and muscularis mucosae.
Endoscopic Evaluation of the Esophagus and Endoscopic Ultrasonography of the Esophagus CHAPTER 7 101
Epithelium
Basement membrane
Lamina propria
Muscularis mucosae
Submucosa
1
2
Muscularis
3 propria
4 Periesophageal
tissue
FIGURE 7.32 The esophageal wall is visualized as five alternating layers of differing echogenicity by esophageal ultrasound. The first layer,
which is hyperechoic (white), represents the superficial mucosa (epithelium and lamina propria). The second layer, which is hypoechoic
(black), represents the deep mucosa (muscularis mucosae). The third layer, which is hyperechoic (white), represents the submucosa. The
fourth layer, which is hypoechoic (black), represents the muscularis propria. The fifth layer, which is hyperechoic (white) is the
periesophageal tissue.
differentiated; G3, poorly differentiated; or G4, undif- or esophagus (Siewert III) are staged as adenocarcinoma
ferentiated. Because the data indicate that squamous cell of the esophagus. All other cancers with an epicenter in
carcinoma has a poorer prognosis than adenocarcinoma, the stomach greater than 5 cm distal to the EGJ, or those
G4, undifferentiated cancers, are staged similar to G3 within 5 cm of the EGJ but not extending into it or the
squamous cell carcinoma. esophagus, are stage grouped using the gastric (non-EGJ)
Tumor location is defined by position of the upper cancer staging system.47
end of the cancer in the esophagus (Fig. 7.33). It is best TNM descriptors are grouped into stages to assemble
expressed as distance from incisors to proximal edge of subgroups with similar behavior and prognosis (see Box
the tumor, and conventionally by its location within broad 7.1). Stages 0 and IV are by definition (not data driven)
regions of the esophagus. Typical esophagoscopy measure- TisN0M0 and Tany Nany M1, respectively. The difference
ments of cervical esophageal cancer measured from the in survival between adenocarcinoma and squamous cell
incisors are from 15 to less than 20 cm. If esophagoscopy carcinoma was best managed by separate stage groupings
is not available, location can be assessed by computed for stages I and II. For T1N0M0 and T2N0M0 adenocarci-
tomography (CT). If thickening of the esophageal wall noma, subgrouping is by histologic grade, G1 and G2 (not
begins above the sternal notch, location is cervical. Typical G3) versus G3. For T1N0M0 squamous cell carcinoma,
esophagoscopy measurements of upper thoracic esophageal subgrouping is by histologic grade: G1 versus all other G.
cancer from the incisors is from 20 to less than 25 cm. For T2N0M0 and T3N0M0 squamous cell carcinoma, stage
CT location of an upper thoracic cancer is esophageal grouping is by histologic grade and location. The four
wall thickening that begins between the sternal notch combinations range from G1 lower thoracic squamous cell
and azygos vein. Typical esophagoscopy measurements carcinoma (stage IB), which has the best survival, to G2–G4
of middle thoracic esophageal cancer from the incisors upper and middle thoracic squamous cell carcinomas
is from 25 to less than 30 cm. CT location is wall thicken- (stage IIB), which have the worst. G2–G4 lower thoracic
ing that begins between the azygos vein and inferior squamous cell carcinomas and G1 upper and middle
pulmonary vein. Typical esophagoscopy measurements thoracic squamous cell carcinomas are grouped together
of lower thoracic esophageal cancer from the incisors (stage IIA) with intermediate survival.
are from 30 to 40 cm (see Fig. 7.33). CT location is wall Stages 0, III, and IV adenocarcinoma and squamous cell
thickening that begins below the inferior pulmonary carcinoma are stage grouped identically. Adenosquamous
vein. The abdominal esophagus is included in the lower carcinomas are staged as squamous cell carcinoma.
thoracic esophagus. Cancers whose epicenter is in the EUS may be used at different periods in the course
lower thoracic esophagus, EGJ, or within the proximal of esophageal carcinoma. The principal times are at the
5 cm of the stomach (cardia) that extend into the EGJ initial staging examination (cStage) and after induction or
102 SECTION I Esophagus and Hernia
BOX 7.1 American Joint Committee on Cancer Staging of Cancer of the Esophagus and Esophagogastric Junction
definitive chemotherapy/chemoradiotherapy (ycStage). It specific for esophageal carcinoma and lacks the defini-
may also be used to diagnose and stage cancer recurrence tion required to distinguish T1, T2, and T3 tumors.59 In
(rStage), also referred to as retreatment stage. differentiation of T3 from T4 tumors, EUS is superior to
CT. Evaluation of the fat planes is used to define local
invasion at CT examination. The obliteration or lack of
CLINICAL STAGE (cTNM) fat planes is not sensitive in predicting local invasion, but
preservation of these planes is specific for the absence of
DETERMINATION OF COMPUTED T4 disease.60–67 When compared with CT, EUS provides
TOMOGRAPHY CLASSIFICATION a more sensitive and reliable determination of vascular
Detailed images of the esophageal wall by EUS make it the involvement.68
most accurate modality available for clinical determination Experience with both examination technique and
of the depth of tumor invasion (T) before treatment (Figs. ultrasound interpretation is critical to accurately deter-
7.34 to 7.37).53–58 The same definition of the esophageal mine the clinical depth of tumor invasion. Seventy-five
wall is not offered by CT. A thickened esophageal wall, to 100 examinations are required before competence is
the principal CT finding in esophageal carcinoma, is not obtained.69,70 A center that does a high volume of EUS
Endoscopic Evaluation of the Esophagus and Endoscopic Ultrasonography of the Esophagus CHAPTER 7 103
Incisors
UES 15 cm
Cervical
Sternal Esophagus
notch 20 cm
Upper
Thoracic
Azygos
vein 25 cm
Middle
Thoracic
Inferior
pulmonary 30 cm
vein
A
Lower
Thoracic
EGJ 40 cm
CCF
©2007
A A
A B
FIGURE 7.40 (A) Hepatic metastasis (upper arrowhead) in the left lateral segment of the liver. A perigastric lymph node metastasis is
shown (lower arrowhead). The esophageal ultrasound probe is seen in the gastric cardia. (B) Hepatic metastasis (upper arrow) as seen
from the gastric cardia by esophageal ultrasound. The metastasis was imaged only by esophageal ultrasound. A perigastric lymph node
metastasis is shown (lower arrow). (From Rice TW, Boyce GA, Sivak MV, et al. Esophageal carcinoma: esophageal ultrasound
assessment of preoperative chemotherapy. Ann Thorac Surg. 1992;53:972.)
ESOPHAGEAL VARICES
Esophageal varices have the typical appearance of blood
vessels at EUS. Appearing as tubular, round, or serpigi-
nous echo-free structures, they may be visualized within
the submucosa or in tissues adjacent to the esophagus
(Fig. 7.43). These EUS patterns change after sclerosis.165
Intravariceal sclerosis fills the varix with echogenic material
representing thrombus. Paravariceal injection leads to
obliteration of the varix with hypoechoic extravariceal
thickening. EUS in combination with color-flow Doppler
may be useful in the hemodynamic assessment of the
portal venous system and treatment effects upon hepatic
blood flow.166
A
ACHALASIA
EUS findings in achalasia are controversial. Some authors
report thickened esophageal wall in most patients exam-
ined.167,168 However, this excessive thickening may be
artifactual. In a dilated and convoluted esophagus, the
ultrasound transducer may orient at an angle oblique to
the esophageal wall and give a false appearance of wall
thickening.169 The main role of EUS in achalasia is to
exclude other mural abnormalities.170–172
PARAESOPHAGEAL DISEASES
EUS has been used to examine the mediastinal lymph
nodes in patients with bronchogenic carcinoma.140–142,173–175
In this setting, EUS has a reported positive predictive value
of 77%, a negative predictive value of 93%, and an overall
B accuracy of 92% when using criteria similar to regional
lymph node evaluation in esophageal carcinoma141,174
Anatomic constraints limit its usefulness for evaluation of
FIGURE 7.42 Foregut cyst. (A) Esophageal ultrasonography lymph nodes in proximity to the airway. EUS-FNA provides
demonstrates a mass (arrows) adjacent to the trachea and
cytologic differentiation between benign and malignant
esophagus. The cyst has two components, one hyperechoic
lymphadenopathy143,176 and has successfully diagnosed
(white), representing proteinaceous material, and one hypoechoic
(black), representing fluid. (B) A foregut cyst in close proximity to
solid lesions of the mediastinum and lung.118,144–146,177–179
the esophagus and trachea.
Esophagus
VV
VV
Heart VV
Wall
Aorta
VV
A B
FIGURE 7.43 Paraesophageal varices. (A) At endoscopy, small varices are not visible. (B) On esophageal ultrasound, the varices (VV) are
prominent anechoic, tubular, and rounded structures outside the esophageal wall.
Endoscopic Evaluation of the Esophagus and Endoscopic Ultrasonography of the Esophagus CHAPTER 7 111
21. Spechler SJ, Goyal RK. The columnar lined esophagus, intestinal
CONCLUSION metaplasia, and Norman Barrett. Gastroenterology. 1996;110:614.
22. Canto MIF, Setrakian S, Willis J, et al. Methylene blue-directed
EUS and EUS-FNA are essential in determining the clinical biopsies improve detection of intestinal metaplasia and dysplasia
stage and directing treatment of esophageal cancer. The in Barrett’s esophagus. Gastrointest Endosc. 2000;51:560.
diagnosis of benign esophageal tumors requires EUS 23. Scotiniotis IA, Kochman ML, Lewis JD, Furth EE, Rosato EF, Ginsberg
examination, which determines both the layer of origin in GG. Accuracy of EUS in the evaluation of Barrett’s esophagus and
high-grade dysplasia or intramucosal carcinoma. Gastrointest Endosc.
the esophageal wall and the ultrasound characteristics of 2001;54:689.
the tumor. Because many of these tumors are asymptomatic, 24. Kobayashi K, Izatt JA, Kulkarni MD, Willis J, Sivak MV Jr. High-
EUS affords simple follow-up and avoids unnecessary resolution cross-sectional imaging of the gastrointestinal tract using
excision. EUS is a useful adjuvant for the diagnosis and optical coherence tomography: preliminary results. Gastrointest
treatment of paraesophageal disease. Endosc. 1998;47:515.
25. Georgakoudi I, Jacobson BC, Van Dam J, et al. Fluorescence,
reflectance, and light-scattering spectroscopy for evaluating dysplasia
REFERENCES in patients with Barrett’s esophagus. Gastroenterology. 2001;120:1620.
26. Kendall C, Stone N, Shepherd N, et al. Raman spectroscopy, a
1. Netter FH. Anatomy of the esophagus. In: Oppenheimer E, ed. potential tool for the objective identification and classification of
The CIBA Collection of Medical Illustrations. Vol. 3. Digestive System. neoplasia in Barrett’s oesophagus. J Pathol. 2003;200:602.
Part I. Upper Digestive Tract. New York: CIBA Pharmaceutical 27. Bergman JJ, Tytgat GN. New developments in the endoscopic
Company; 1959:34. surveillance of Barrett’s oesophagus. Gut. 2005;54:i38.
2. Goyal RK, Martin SB, Shapiro J, Spechler SJ. The role of crico- 28. Olliver JR, Wild CP, Sahay P, Dexter S, Hardie LJ. Chromoendoscopy
pharyngeal muscle in pharyngoesophageal disorders. Dysphagia. with methylene blue and associated DNA damage in Barrett’s
1993;8:252. oesophagus. Lancet. 2003;362:373.
3. Mittal RK, Balaban DH. The esophagogastric junction. N Engl J 29. Amano Y, Kushiyama Y, Ishihara S, et al. Crystal violet chromoendos-
Med. 1997;336:924. copy with mucosal pit pattern diagnosis is useful for surveillance of
4. Weickert U, Wolf A, Schröder C, Autschbach F, Vollmer H. short-segment Barrett’s esophagus. Am J Gastroenterol. 2005;100:21.
Frequency, histopathological findings, and clinical significance 30. Toyoda H, Rubio C, Befrits R, Hamamoto N, Adachi Y, Jaramillo
of cervical heterotopic gastric mucosa (gastric inlet patch): a E. Detection of intestinal metaplasia in distal esophagus and
prospective study in 300 patients. Dis Esophagus. 2011;24:63. esophagogastric junction by enhanced-magnification endoscopy.
5. Johnson LF, Moses FM. Endoscopic evaluation of esophageal disease. Gastrointest Endosc. 2004;59:15.
In: Castell DO, Johnson LF, eds. Esophageal Function in Health and 31. Endo T, Awakawa T, Takahashi H, et al. Classification of Barrett’s
Disease. New York: Elsevier Science Publishing Co., Inc.; 1983:237. epithelium by magnifying endoscopy. Gastrointest Endosc. 2002;55:641.
6. Goyal RK, Bauer J, Spiro HM. The nature and location of the lower 32. Vakil N, van Zanten SV, Kahrilas P, et al.; Global Consensus Group.
esophageal ring. N Engl J Med. 1971;284:1175. The Montreal definition and classification of gastroesophageal
7. Ott DJ. Radiology of the oropharynx and esophagus. In: Castell DO, reflux disease: a global evidence-based consensus. Am J Gastroenterol.
ed. The Esophagus. Boston: Little, Brown and Company; 1995:41. 2006;101:1900.
8. Paull A, Trier JS, Dalton MD, Camp RC, Loeb P, Goyal RK. The his- 33. Armstrong D. Endoscopic evaluation of gastro-esophageal reflux
tologic spectrum of Barrett’s esophagus. N Engl J Med. 1976;295:476. disease. Yale J Biol Med. 1999;72:93.
9. Kim SL, Waring PJ, Spechler SJ, et al.; the Department of Veterans 34. Richter JE, Peura D, Benjamin SB, Joelsson B, Whipple J. Efficacy
Affairs Gastroesophageal Reflux Study Group. Diagnostic inconsisten- of omeprazole for the treatment of symptomatic acid reflux disease
cies in Barrett’s esophagus. Gastroenterology. 1994;107:945. without esophagitis. Arch Intern Med. 2000;160:1810.
10. Bozymski EM. Barrett’s esophagus: endoscopic characteristics. 35. Bytzer P, Havelund T, Hansen JM. Interobserver variation in the
In: Spechler SJ, Goyal RK, eds. Barrett’s Esophagus: Pathophysiology, endoscopic diagnosis of reflux esophagitis. Scand J Gastroenterol.
Diagnosis, and Management. New York: Elsevier Science Publishing 1993;28:119.
Co., Inc.; 1985:113. 36. Lundell LR, Dent J, Bennett JR, et al. Endoscopic assessment of
11. McClave SA, Boyce HW Jr, Gottfried MR. Early diagnosis of columnar- oesophagitis: clinical and functional correlates and further validation
lined esophagus: a new endoscopic criterion. Gastrointest Endosc. of the Los Angeles classification. Gut. 1999;45:172.
1987;33:413. 37. Grossman MI, ed. Peptic Ulcer: A Guide for the Practicing Physician.
12. De Carvalho CA. Sur l’angio-architecture veineuse de la zone de Chicago: Year Book Medical Publishers, Inc.; 1981.
transition esophago-gasgtrique et son interpretation fonctionnelle. 38. Armstrong D, Bennett JR, Blum AL, et al. The endoscopic assess-
Acta Anat. 1966;64:125. ment of esophagitis: a progress report on observer agreement.
13. Choi DW, Oh SN, Baek SJ, et al. Endoscopically observed lower Gastroenterology. 1996;111:85.
esophageal capillary patterns. Korean J Intern Med. 2002;17:245. 39. Savary M, Miller G. The Esophagus. Handbook and Atlas of Endoscopy.
14. Spechler SJ. The role of gastric carditis in metaplasia and neoplasia Solothurn, Switzerland: Verlag Gassman; 1978.
at the gastroesophageal junction. Gastroenterology. 1999;117:218. 40. Spechler SJ. The management of patients who have “failed”
15. Vianna A, Hayes PC, Moscoso G, et al. Normal venous circulation antireflux surgery. Am J Gastroenterol. 2004;99:552.
of the gastroesophageal junction. A route to understanding varices. 41. Furuta GT, Liacouras CA, Collins MH, et al.; First International
Gastroenterology. 1987;93:876. Gastrointestinal Eosinophil Research Symposium (FIGERS)
16. Spechler SJ, Fitzgerald RC, Prasad GA, Wang KK. History, molecular Subcommittees. Eosinophilic esophagitis in children and adults:
mechanisms, and endoscopic treatment of Barrett’s esophagus. a systematic review and consensus recommendations for diagnosis
Gastroenterology. 2010;138:854. and treatment. Gastroenterology. 2007;133:1342.
17. Sharma P, Morales TG, Sampliner RE. Short segment Barrett’s 42. Rothenberg ME. Biology and treatment of eosinophilic esophagitis.
esophagus. The need for standardization of the definition and of Gastroenterology. 2009;137:1238.
endoscopic criteria. Am J Gastroenterol. 1998;93:1033. 43. Kimmey MB, Martin RW. Fundamentals of endosonography.
18. Wallner B, Sylvan A, Stenling R, Janunger KG. The esophageal Z-line Gastrointest Endosc Clin N Am. 1992;2:557.
appearance correlates to the prevalence of intestinal metaplasia. 44. Vilmann P, Khattar S, Hancke S. Endoscopic ultrasound examination
Scand J Gastroenterol. 2000;35:17. of the upper gastrointestinal tract using a curved-array transducer.
19. Wallner B, Sylvan A, Janunger KG. Endoscopic assessment of the A preliminary report. Surg Endosc. 1991;5:79.
“Z-line” (squamocolumnar junction) appearance: reproducibility 45. Bolondi L, Casanova P, Santi V, Caletti G, Barbara L, Labò G. The
of the ZAP classification among endoscopists. Gastrointest Endosc. sonographic appearance of the normal gastric wall: an in vitro
2002;55:65. study. Ultrasound Med Biol. 1986;12:991.
20. Sharma P, Dent J, Armstrong D, et al. The development and 46. Kimmey MB, Martin RW, Haggitt RC, Wang KY, Franklin DW,
validation of an endoscopic grading system for Barrett’s esophagus: Silverstein FE. Histologic correlates of gastrointestinal ultrasound
the Prague C & M criteria. Gastroenterology. 2006;131:1392. images. Gastroenterology. 1989;96:433.
112 SECTION I Esophagus and Hernia
47. Edge SB, Byrd DR, Compton CC, Fritz AG, Greene F, Trotti A. 73. Kelly S, Harris SM, Berry E, et al. A systematic review of the staging
AJCC Cancer Staging Manual. 7th ed. New York: Springer; 2009. performance of endoscopic ultrasound in gastroesophageal carci-
48. Rice TW, Rusch VW, Apperson-Hansen C, et al. Worldwide esopha- noma. Gut. 2001;49:534.
geal cancer collaboration. Dis Esophagus. 2009;22:1. 74. Puli SR, Reddy JB, Bechtold ML, Antillon D, Ibdah JA, Antillon MR.
49. Ishwaran H, Blackstone EH, Apperson-Hansen C, Rice TW. A novel Staging accuracy of esophageal cancer by endoscopic ultrasound:
approach to cancer staging: application to esophageal cancer. a meta-analysis and systematic review. World J Surg. 2008;14:1479.
Biostatistics. 2009;10:603. 75. Rice TW, Blackstone EH, Adelstein DJ, et al. Role of clinically
50. Rice TW, Rusch VW, Ishwaran H, et al. Cancer of the esophagus determined depth of tumor invasion in the treatment of esophageal
and esophagogastric junction: data-driven staging for the 7th carcinoma. J Thorac Cardiovasc Surg. 2003;125:1091.
edition of the AJCC cancer staging manual. Cancer. 2010;16:3763. 76. Heidemann J, Schilling MK, Schmassmann A, Maurer CA, Büchler
51. Rice TW, Blackstone EH, Rusch VW. A cancer staging primer: MW. Accuracy of endoscopic ultrasonography in preoperative
esophagus and esophagogastric junction. J Thorac Cardiovasc Surg. staging of esophageal carcinoma. Dig Surg. 2000;17:219.
2010;139:527. 77. Rice TW, Mason DP, Murthy SC, et al. T2N0M0 esophageal cancer.
52. Rice TW, Blackstone EH, Rybicki LA, et al. Refining esophageal J Thorac Cardiovasc Surg. 2007;133:317.
cancer staging. J Thorac Cardiovasc Surg. 2003;125:1103. 78. Pech O, Günter E, Dusemund F, Origer J, Lorenz D, Ell C. Accuracy
53. Botet JF, Lightdale CJ, Zauber AG, Gerdes H, Urmacher C, Brennan of endoscopic ultrasound in preoperative staging of esophageal
MF. Preoperative staging of esophageal cancer: comparison of cancer: results from a referral center for early esophageal cancer.
endoscopic US and dynamic CT. Radiology. 1991;181:419. Endoscopy. 2010;42:456.
54. Date H, Miyashita M, Sasajima K, et al. Assessment of adventitial 79. Rice TW, Blackstone EH, Adelstein DJ, et al. Role of clinically
involvement of esophageal carcinoma by endoscopic ultrasonog- determined depth of tumor invasion in the treatment of esophageal
raphy. Surg Endosc. 1990;4:195. carcinoma. J Thorac Cardiovasc Surg. 2003;125:1091.
55. Heintz A, Hohne U, Schweden F, Junginger T. Preoperative detection 80. Kelly S, Harris KM, Berry E, et al. A systematic review of the staging
of intrathoracic tumor spread of esophageal cancer: endosonography performance of endoscopic ultrasound in gastro-oesophageal
versus computed tomography. Surg Endosc. 1991;5:75. carcinoma. Gut. 2001;49:534.
56. Tio TL, Cohen P, Coene PP, Udding J, den Hartog Jager FC, Tytgat 81. Yusuf TE, Harewood GC, Clain JE, Levy MJ, Topazian MD, Rajan
GN. Endosonography and computed tomography of esophageal E. Clinical implications of the extent of invasion of T3 esophageal
carcinoma. Preoperative classification compared to the new (1987) cancer by endoscopic ultrasound. J Gastroenterol Hepatol. 2005;20:
TNM system. Gastroenterology. 1989;96:1478. 1880.
57. Vilgrain V, Mompoint D, Palazzo L, et al. Staging of esophageal 82. Meining A, Dittler HJ, Wolf A, et al. You get what you expect? A
carcinoma: comparison of results with endoscopic sonography and critical appraisal of imaging methodology in endosonographic
CT. AJR Am J Roentgenol. 1990;155:277. cancer staging. Gut. 2002;50:599.
58. Ziegler K, Sanft C, Zeitz M, et al. Evaluation of endosonography 83. Bhutani MS, Barde CJ, Markert RJ, Gopalswamy N. Length of
in TN staging of oesophageal cancer. Gut. 1991;32:16. esophageal cancer and degree of luminal stenosis during upper
59. Reinig JW, Stanley JH, Schabel SI. CT evaluation of thickened endoscopy predict T stage by endoscopic ultrasound. Endoscopy.
esophageal walls. AJR Am J Roentgenol. 1983;140:931. 2002;34:461.
60. Consigliere D, Chua CL, Hui F, Yu CS, Low CH. Computed 84. Twine CP, Roberts SA, Lewis WG, et al. Prognostic significance of
tomography for oesophageal carcinoma: its value to the surgeon. endoluminal ultrasound-defined disease length and tumor volume
J R Coll Surg Edinb. 1992;37:113. (EDTV) for patients with the diagnosis of esophageal cancer. Surg
61. Duignan JP, McEntee GP, O’Connell DJ, Bouchier-Hayes DJ, Endosc. 2010;24:870.
O’Malley E. The role of CT in the management of carcinoma of 85. Dancygier H, Classen M. Endoscopic ultrasonography in esophageal
the oesophagus and cardia. Ann R Coll Surg Engl. 1987;69:286. diseases. Gastrointest Endosc. 1989;35:220.
62. Kasbarian M, Fuentes P, Brichon PY. Usefulness of Computed Tomog- 86. Hordijk ML, Zander H, van Blankenstein M, Tilanus HW. Influence
raphy in Assessing the Extension of Carcinoma of the Esophagus and of tumor stenosis on the accuracy of endosonography in preoperative
Gastroesophageal Junction. Berlin: Springer-Verlag; 1988. T staging of esophageal cancer. Endoscopy. 1993;25:171.
63. Kirk SJ, Moorehead RJ, McIlrath E, Gibbons JP, Spence RA. Does 87. Morgan MA, Twine CP, Lewis WG, et al. Prognostic significance
preoperative computed tomography scanning aid assessment of of failure to cross esophageal tumors by endoluminal ultrasound.
oesophageal carcinoma? Postgrad Med J. 1990;66:191. Dis Esophagus. 2008;21:508.
64. Markland CG, Manhire A, Davies P, Beggs D, Morgan WE, Salama 88. Catalano MF, Van Dam J, Sivak JMV. Malignant esophageal strictures:
FD. The role of computed tomography in assessing the operability staging accuracy of endoscopic ultrasonography. Gastrointest Endosc.
of oesophageal carcinoma. Eur J Cardiothorac Surg. 1989;3:33. 1995;41:535.
65. Rice TW, Boyce GA, Sivak MV. Esophageal ultrasound and the 89. Van Dam J, Rice TW, Catalano MF, Kirby T, Sivak MV Jr. High-grade
preoperative staging of carcinoma of the esophagus. J Thorac malignant stricture is predictive of esophageal tumor stage: risks
Cardiovasc Surg. 1991;101:536, [discussion 543–544]. of endosonographic evaluation. Cancer. 1993;71:2910.
66. Ruol A, Rossi M, Ruffatto A. Reevaluation of Computed Tomography 90. Kallemanis GE, Gupta PK, al-Kawas FH, et al. Endoscopic ultrasound
in Preoperative Staging of Esophageal and Cardial Cancers: A Prospective for staging esophageal cancer, with and without dilation, is clinically
Study. New York: Springer-Verlag; 1987. important and safe. Gastrointest Endosc. 1995;41:540.
67. Sondenaa K, Skaane P, Nygaard K, Skjennald A. Value of computed 91. Pfau PR, Ginsberg GG, Lew RJ, Faigel DO, Smith DB, Kochman
tomography in preoperative evaluation of resectability and staging ML. Esophageal dilation for endosonographic evaluation of malig-
in oesophageal carcinoma. Eur J Surg. 1992;158:537. nant esophageal strictures is safe and effective. Am J Gastroenterol.
68. Ginsberg GG, Al-Kawas EH, Nguyen CC. Endoscopic ultrasound 2000;95:2813.
evaluation of vascular involvement in esophageal cancer: a com- 92. Wallace MB, Hawes RH, Sahai AV, Van Velse A, Hoffman BJ. Dilation
parison with computed tomography [abstract]. Gastrointest Endosc. of malignant esophageal stenosis to allow EUS guided fine-needle
1993;39:A276. aspiration: safety and effect on patient management. Gastrointest
69. Fockens P, Van den Brande JH, van Dullemen HM, van Lanschot Endosc. 2000;51:309.
JJ, Tytgat GN. Endosonographic T-staging of esophageal carcinoma: 93. Binmoeller KF, Seifert H, Seitz U, Izbicki JR, Kida M, Soehendra
a learning curve. Gastrointest Endosc. 1996;44:58. N. Ultrasonic esophagoprobe for TNM staging of highly stenosing
70. Schlick T, Heintz A, Junginger T. The examiner’s learning effect esophageal carcinoma. Gastrointest Endosc. 1995;41:547.
and its influence on the quality of endoscopic ultrasonography 94. Hunerbein M, Ghadimi BM, Haensch W, Schlag PM. Transendoscopic
in carcinoma of the esophagus and gastric cardia. Surg Endosc. ultrasound of esophageal and gastric cancer using miniaturized
1999;13:894. ultrasound catheter probes. Gastrointest Endosc. 1998;48:371.
71. van Vliet EP, Eijkemans MJ, Poley JW, Steyerberg EW, Kuipers EJ, 95. McLoughlin RF, Cooperberg PL, Mathieson JR, Stordy SN, Halparin
Siersema PD. Staging of esophageal carcinoma in low-volume EUS LS. High resolution endoluminal ultrasonography in the staging
center compared with reported results from high-volume centers. of esophageal carcinoma. J Ultrasound Med. 1995;14:725.
Gastrointest Endosc. 2006;63:938. 96. Menzel J, Hoepffner N, Nottberg H, Schulz C, Senninger N,
72. Rosch T. Endosonographic staging of esophageal cancer: a review Domschke W. Preoperative staging of esophageal carcinoma:
of literature results. Gastrointest Endosc Clin N Am. 1995;5:537. miniprobe sonography versus conventional endoscopic ultrasound
Endoscopic Evaluation of the Esophagus and Endoscopic Ultrasonography of the Esophagus CHAPTER 7 113
in a prospective histopathologically verified study. Endoscopy. 117. Wiersema MJ, Hawes RH, Tao LC, et al. Endoscopic ultrasonography
1999;31:291. as an adjunct to fine needle aspiration cytology of the upper and
97. Worrell SG, Oh DS, Greene CL, Demeester SR, Hagen JA. Endo- lower gastrointestinal tract. Gastrointest Endosc. 1992;38:35.
scopic ultrasound staging of stenotic esophageal cancers may be 118. Wiersema MJ, Kochman ML, Chak A, Cramer HM, Kesler KA.
unnecessary to determine the need for neoadjuvant therapy. J Real-time endoscopic ultrasound-guided fine-needle aspiration of
Gastrointest Surg. 2014;18:318-320. a mediastinal lymph node. Gastrointest Endosc. 1993;39:429.
98. Buskens CJ, Westerterp M, Lagarde SM, Bergman JJ, ten Kate FJ, 119. Mortensen MB, Pless T, Durup J, Ainsworth AP, Plagborg GJ,
van Lanschot JJ. Prediction of appropriateness of local endoscopic Hovendal C. Clinical impact of endoscopic ultrasound-guided fine
treatment for high-grade dysplasia and early adenocarcinoma by needle aspiration biopsy in patients with upper gastrointestinal
EUS and histopathologic features. Gastrointest Endosc. 2004;60:703. tract malignancies. A prospective study. Endoscopy. 2001;33:478.
99. Scotiniotis IA, Kochman ML, Lewis JD, Furth EE, Rosato EF, Ginsberg 120. Vazquez-Sequeiros E, Norton ID, Clain JE, et al. Impact of EUS-
GG. Accuracy of EUS in the evaluation of Barrett’s esophagus and guided fine-needle aspiration on lymph node staging in patients
high-grade dysplasia or intramucosal carcinoma. Gastrointest Endosc. with esophageal carcinoma. Gastrointest Endosc. 2001;53:751.
2001;54:689. 121. Vazquez-Sequeiros E, Wiersema MJ, Clain JE, et al. Impact of lymph
100. Pech O, May A, Günter E, Gossner L, Ell C. The impact of endoscopic node staging on therapy of esophageal carcinoma. Gastroenterology.
ultrasound and computed tomography on TNM stage of early 2003;125:1626.
cancer in Barrett’s esophagus. Am J Gastroenterol. 2006;101:2223. 122. Wiersema MJ, Vilmann P, Giovannini M, Chang KJ, Wiersema
101. Rampado S, Bocus P, Battaglia G, Ruol A, Portale G, Ancona E. LM. Endosonography-guided fine-needle aspiration biopsy:
Endoscopic ultrasound: accuracy in staging superficial carcinomas diagnostic accuracy and complication assessment. Gastroenterology.
of the esophagus. Ann Thorac Surg. 2008;85:251. 1997;112:1087.
102. Mandal RV, Forcione DG, Brugge WR, Nishioka NS, Mino-Kenudson 123. O’Toole D, Palazzo L, Arotcarena R, et al. Assessment of complica-
M, Lauwers GY. Effect of tumor characteristics and duplication of tions of EUS-guided fine-needle aspiration. Gastrointest Endosc.
the muscularis mucosae on the endoscopic staging of superficial 2001;53:470.
Barrett esophagus-related neoplasia. Am J Surg Pathol. 2009;33:620. 124. Eloubeidi MA, Wallace MB, Reed CE, et al. The utility of EUS
103. May A, Gunter E, Roth F, et al. Accuracy of staging in early oesopha- and EUS-guided fine needle aspiration in detecting celiac lymph
geal cancer using high resolution endoscopy and high resolution node metastasis in patients with esophageal cancer: a single-center
endosonography: a comparative, prospective, and blinded trial. experience. Gastrointest Endosc. 2001;54:714.
Gut. 2004;53:634. 125. Harewood GC, Wiersema MJ. A cost analysis of endoscopic ultra-
104. Murata Y, Napoleon B, Odegaard S. High-frequency endoscopic sound in the evaluation of esophageal cancer. Am J Gastroenterol.
ultrasonography in the evaluation of superficial esophageal cancer. 2002;97:452.
Endoscopy. 2003;35:429, [discussion 436]. 126. Pfau PR, Ginsberg GG, Lew RJ, Brensinger CM, Kochman ML.
105. Waxman I, Raju GS, Critchlow J, Antonioli DA, Spechler SJ. High- EUS predictors of long-term survival in esophageal carcinoma.
frequency probe ultrasonography has limited accuracy for detecting Gastrointest Endosc. 2001;53:463.
invasive adenocarcinoma in patients with Barrett’s esophagus and 127. Harewood GC, Kumar KS. Assessment of clinical impact of endo-
high-grade dysplasia or intramucosal carcinoma: a case series. Am scopic ultrasound on esophageal cancer. J Gastroenterol Hepatol.
J Gastroenterol. 2006;101:1773. 2004;19:433.
106. Thomas T, Gilbert D, Kaye PV, Penman I, Aithal GP, Ragunath 128. McGrath K, Brody D, Luketich J, Khalid A. Detection of unsuspected
K. High-resolution endoscopy and endoscopic ultrasound for left hepatic lobe metastases during EUS staging of cancer of the
evaluation of early neoplasia in Barrett’s esophagus. Surg Endosc. esophagus and cardia. Am J Gastroenterol. 2006;101:1742.
2010;24:1110. 129. Sultan J, Robinson S, Hayes N, Griffin SM, Richardson DL, Preston
107. Catalano MF, Sivak MV Jr, Rice T, Gragg LA, Van Dam J. Endosono- SR. Endoscopic ultrasonography-detected low-volume ascites as a
graphic features predictive of lymph node metastasis. Gastrointest predictor of inoperability for oesophagogastric cancer. Br J Surg.
Endosc. 1994;40:442. 2008;95:1127.
108. Bhutani MS, Hawes RH, Hoffman BJ. A comparison of the accuracy 130. Adelstein DJ, Rice TW, Boyce GA, et al. Adenocarcinoma of the
of echo features during endoscopic ultrasound (EUS) and EUS- esophagus and gastroesophageal junction. Clinical and pathologic
guided fine-needle aspiration for diagnosis of malignant lymph assessment of response to induction chemotherapy. Am J Clin Oncol.
node invasion. Gastrointest Endosc. 1997;45:474. 1994;17:14.
109. Catalano MF, Alcocer E, Chak A, et al. Evaluation of metastatic 131. Hordijk ML, Kok TC, Wilson JH, Mulder AH. Assessment of response
celiac axis lymph nodes in patients with esophageal carcinoma: of esophageal carcinoma to induction chemotherapy. Endoscopy.
accuracy of EUS. Gastrointest Endosc. 1999;50:352. 1993;25:592.
110. Wiersema MJ, Vilmann P, Giovannini M, Chang KJ, Wiersema 132. Roubein LD, DuBrow R, David C, et al. Endoscopic ultrasonography
LM. Endosonography-guided fine-needle aspiration biopsy: in the quantitative assessment of response to chemotherapy in
diagnostic accuracy and complication assessment. Gastroenterology. patients with adenocarcinoma of the esophagus and esophagogastric
1997;112:1087. junction. Endoscopy. 1993;25:587.
111. Reed CE, Mishra G, Sahai AV, Hoffman BJ, Hawes RH. Esophageal 133. Dittler HJ, Fink U, Siewert GR. Response to chemotherapy in
cancer staging: improved accuracy by endoscopic ultrasound of esophageal cancer. Endoscopy. 1994;26:769.
celiac lymph nodes. Ann Thorac Surg. 1999;67:319. 134. Giovannini M, Seitz JF, Thomas P, et al. Endoscopic ultrasonography
112. Eloubeidi MA, Wallace MB, Reed CE, et al. The utility of EUS for assessment of the response to combined radiation therapy
and EUS-guided fine needle aspiration in detecting celiac lymph and chemotherapy in patients with esophageal cancer. Endoscopy.
node metastasis in patients with esophageal cancer: a single-center 1997;29:4.
experience. Gastrointest Endosc. 2001;54:714. 135. Zuccaro G Jr, Rice TW, Goldblum J, et al. Endoscopic ultrasound
113. Natsugoe S, Yoshinaka H, Shimada M, et al. Number of lymph node cannot determine suitability for esophagectomy after aggressive
metastases determined by presurgical ultrasound and endoscopic chemoradiotherapy for esophageal cancer. Am J Gastroenterol.
ultrasound is related to prognosis in patients with esophageal 1999;94:906.
carcinoma. Ann Surg. 2001;234:613. 136. Isenberg G, Chak A, Canto MI, et al. Endoscopic ultrasound in
114. Chen J, Xu R, Hunt GC, Krinsky ML, Savides TJ. Influence of the restaging of esophageal cancer after neoadjuvant chemoradiation.
number of malignant regional lymph nodes detected by endoscopic Gastrointest Endosc. 1998;48:158.
ultrasonography on survival stratification in esophageal adenocar- 137. Laterza E, de Manzoni G, Guglielmi A, Rodella L, Tedesco P,
cinoma. Clin Gastroenterol Hepatol. 2006;4:573. Cordiano C. Endoscopic ultrasonography in the staging of esopha-
115. Twine CP, Roberts SA, Rawlinson CE, et al. Prognostic significance geal carcinoma after preoperative radiotherapy and chemotherapy.
of the endoscopic ultrasound defined lymph node metastasis count Ann Thorac Surg. 1999;67:1466.
in esophageal cancer. Dis Esophagus. 2010;23:652. 138. Kalha I, Kaw M, Fukami N, et al. The accuracy of endoscopic
116. Rice TW, Zuccaro G Jr, Adelstein DJ, Rybicki LA, Blackstone EH, ultrasound for restaging esophageal carcinoma after chemoradiation
Goldblum JR. Esophageal carcinoma: depth of tumor invasion therapy. Cancer. 2004;101:940.
is predictive of regional lymph node status. Ann Thorac Surg. 139. Beseth BD, Bedford R, Isacoff WH, Holmes EC, Cameron RB.
1998;65:787. Endoscopic ultrasound does not accurately assess pathologic stage
114 SECTION I Esophagus and Hernia
of esophageal cancer after neoadjuvant chemoradiotherapy. Am 159. Blum MG, Bilimoria KY, Wayne JD, de Hoyos AL, Talamonti MS,
Surg. 2000;66:827. Adley B. Surgical considerations for the management and resection
140. Fleshman JW, Myerson RJ, Fry RD, Kodner IJ. Accuracy of transrectal of esophageal gastrointestinal stromal tumors. Ann Thorac Surg.
ultrasound in predicting pathologic stage of rectal cancer before and 2007;84:1717.
after preoperative radiation therapy. Dis Colon Rectum. 1992;35:823. 160. Bhutani MS, Hoffman BJ, Reed C. Endosonographic diagnosis of
141. Chak A, Canto MI, Cooper GS, et al. Endosonographic assessment an esophageal duplication cyst. Endoscopy. 1996;28:396.
of multimodality therapy predicts survival of esophageal carcinoma 161. Faigel DO, Burke A, Ginsberg GG, Stotland BR, Kadish SL, Kochman
patients. Cancer. 2000;88:1788. ML. The role of endoscopic ultrasound in the evaluation and
142. Agarwal B, Swisher S, Ajani J, et al. Endoscopic ultrasound after management of foregut duplications. Gastrointest Endosc. 1997;45:99.
preoperative chemoradiation can help identify patients who benefit 162. Lim LL, Ho KY, Goh PM. Preoperative diagnosis of a paraesopha-
maximally after surgical esophageal resection. Am J Gastroenterol. geal bronchogenic cyst using endosonography. Ann Thorac Surg.
2004;99:1258. 2002;73:633.
143. Ngamruengphong S, Sharma VK, Nguyen B, Das A. Assessment of 163. Massari M, De Simone M, Cioffi U, Gabrielli F, Boccasanta P,
response to neoadjuvant therapy in esophageal cancer: an updated Bonavina L. Endoscopic ultrasonography in the evaluation of
systematic review of diagnostic accuracy of endoscopic ultrasonog- leiomyoma and extramucosal cysts of the esophagus. Hepatogas-
raphy and fluorodeoxyglucose positron emission tomography. Dis troenterology. 1998;45:938.
Esophagus. 2010;3:216. 164. Van Dam J, Rice TW, Sivak MV Jr. Endoscopic ultrasonography and
144. Catalano MF, Sivak MV Jr, Rice TW, Van Dam J. Postoperative endoscopically guided needle aspiration for the diagnosis of upper
screening for anastomotic recurrence of esophageal carcinoma gastrointestinal tract foregut cysts. Am J Gastroenterol. 1992;87:762.
by endoscopic ultrasonography. Gastrointest Endosc. 1995;42:540. 165. Yasuda K, Cho E, Nakajima M, Kawai K. Diagnosis of submucosal
145. Lightdale CJ, Botet JF, Kelsen DP, Turnbull AD, Brennan MF. lesions of the upper gastrointestinal tract by endoscopic ultraso-
Diagnosis of recurrent upper gastrointestinal cancer at the surgi- nography. Gastrointest Endosc. 1990;36(suppl 2):S17.
cal anastomosis by endoscopic ultrasound. Gastrointest Endosc. 166. El-Saadany M, Jalil S, Irisawa A, Shibukawa G, Ohira H, Bhutani MS.
1989;35:407. EUS for portal hypertension: a comprehensive and critical aaprasal
146. Kawamoto K, Yamada Y, Utsunomiya T, et al. Gastrointestinal of clinical and experimental indications. Endoscopy. 2008;40:690.
submucosal tumors: evaluation with endoscopic US. Radiology. 167. Bergami GL, Fruhwirth R, Di Mario M, Fasanelli S. Contribution of
1997;205:733. ultrasonography in the diagnosis of achalasia. J Pediatr Gastroenterol
147. Lewin KJ, Appelman HD. Mesenchymal tumors and tumor-like Nutr. 1992;14:92.
proliferations of the esophagus. In: Rosai J, Sobin LH, eds. Tumors of 168. Deviere J, Dunham F, Rickaert F, Bourgeois N, Cremer M. Endoscopic
the Esophagus and Stomach. Washington, DC: Armed Forces Institute ultrasonography in achalasia. Gastroenterology. 1989;96:1210.
of Pathology; 1996:145 Atlas of Tumor Pathology; 3rd series, fascicle 169. Falk GW, Van Dam J, Sivak MV. Endoscopic ultrasonography (EUS)
18. in achalasia. Gastrointest Endosc. 1991;37:241.
148. Schuhmacher C, Becker K, Dittler HJ, Höfler H, Siewert JR, Stein 170. Barthet M, Mambrini P, Audibert P, et al. Relationships between
HJ. Fibrovascular esophageal polyp as a diagnostic challenge. Dis endosonographic appearance and clinical or manometric features
Esophagus. 2000;13:324. in patients with achalasia. Eur J Gastroenterol Hepatol. 1998;10:559.
149. Devereaux BM, LeBlanc JK, Kesler K, Burttet LM, Kruel CD, da 171. Ponsot P, Chaussade S, Palazzo L, et al. Endoscopic ultrasonography
Rosa AP. Giant fibrovascular polyp of the esophagus. Endoscopy. in achalasia. Gastroenterology. 1990;98:253.
2003;35:970. 172. Ziegler K, Sanft C, Friedrich M, Gregor M, Riecken EO. Endo-
150. Palazzo L, Landi B, Cellier C, et al. Endosonographic features of sonographic appearance of the esophagus in achalasia. Endoscopy.
esophageal granular cell tumors. Endoscopy. 1997;29:850. 1990;22:1.
151. Love MH, Glaser M, Edmunds SE, Mendelson RM. Granular cell 173. Kobayashi H, Danabara T, Sugama Y, Saito T, Kitamura S, Kira S.
tumour of the oesophagus: endoscopic ultrasound appearances. Observation of lymph nodes and great vessels in the mediastinum
Australas Radiol. 1999;43:253. by endoscopic ultrasonography. Jpn J Med. 1987;26:353.
152. Goldblum JR, Rice TW, Zuccaro G, Richter JE. Granular cell tumors 174. Kondo D, Imaizumi M, Abe T, Naruke T, Suemasu K. Endoscopic
of the esophagus: a clinical and pathologic study of 13 cases. Ann ultrasound examination for mediastinal lymph node metastases
Thorac Surg. 1996;62:860. of lung cancer. Chest. 1990;98:586.
153. Araki K, Ohno S, Egashira A, et al. Esophageal hemangioma: 175. Potepan P, Meroni E, Spagnoli I, et al. Non–small-cell lung cancer:
a case report and review of the literature. Hepatogastroenterology. detection of mediastinal lymph node metastases by endoscopic
1999;46:3148. ultrasound and CT. Eur Radiol. 1996;6:19.
154. Maluf-Filho F, Sakai P, Amico EC, Pinotti HW. Giant cavernous 176. Mishra G, Sahai AV, Penman ID, et al. Endoscopic ultrasonography
hemangioma of the esophagus: endoscopic and echo-endoscopic with fine-needle aspiration: an accurate and simple diagnostic
appearance. Endoscopy. 1999;31:S32. modality for sarcoidosis. Endoscopy. 1999;31:377.
155. Takada N, Higashino M, Osugi H, Tokuhara T, Kinoshita H. Utility 177. Fritscher-Ravens A, Petrasch S, Reinacher-Schick A, Graeven U, König
of endoscopic ultrasonography in assessing the indications for M, Schmiegel W. Diagnostic value of endoscopic ultrasonography-
endoscopic surgery of submucosal esophageal tumors. Surg Endosc. guided fine-needle aspiration cytology of mediastinal masses in
1999;13:228. patients with intrapulmonary lesions and nondiagnostic bronchos-
156. Xu GM, Niu YL, Zou XP, Jin ZD, Li ZS. The diagnostic value of copy. Respiration. 1999;66:150.
transendoscopic miniature ultrasonic probe for esophageal diseases. 178. Hunerbein M, Ghadimi BM, Haensch W, Schlag PM. Transesophageal
Endoscopy. 1998;30(suppl 1):A28. biopsy of mediastinal and pulmonary tumors by means of endoscopic
157. Stelow EB, Jones DR, Shami VM. Esophageal leiomyosarcoma ultrasound guidance. J Thorac Cardiovasc Surg. 1998;116:554.
diagnosed by endoscopic ultrasound guided fine-needle aspiration. 179. Pedersen BH, Vilmann P, Folke K, et al. Endoscopic ultrasonography
Diagn Cytopathol. 2007;35:167. and real-time guided fine-needle aspiration biopsy of solid lesions
158. Gouveia AM, Pimenta AP, Lopes JM, et al. Esophageal GIST: of the mediastinum suspected of malignancy. Chest. 1996;110:539.
therapeutic implications of an uncommon presentation of a rare
tumor. Dis Esophagus. 2005;18:7.
CHAPTER
High-Resolution Esophageal Manometry: Techniques
and Use in the Diagnosis of Esophageal Motility 8
Disorders and for Surgical Decision Making
Ezra N. Teitelbaum
| Christy M. Dunst
E
sophageal motility disorders may be implicated as that patients undergo routine preoperative HRM. HRM
an explanation for dysphagia and noncardiac chest can diagnose previously unrecognized major disorders of
pain after exclusion of esophageal structural lesions esophageal motility, such as achalasia, and identify patients
by endoscopy, with the caveat that eosinophilic esophagitis with partially impaired esophageal body function, or IEM.
has been ruled out with histology. Gastroesophageal reflux How to approach patients with IEM who are undergoing
disease (GERD) must also be carefully considered, and most fundoplication is an area of considerable controversy. Some
patients will be given a course of proton pump inhibitor surgeons preferred a strategy of “tailored” fundoplication in
therapy or evaluated with a 24- or 48-hour pH monitoring which a partial wrap is constructed in such patients, whereas
study to exclude that possibility even in the absence of others do not feel this is necessary and perform a complete,
endoscopic lesions. The best-defined esophageal motor or Nissen, fundoplication regardless of preoperative HRM
disorder is achalasia; however, other motility disorders such results. This chapter will focus on describing esophageal
as distal esophageal spasm (DES), hypercontractile (or motor disorders using HRM and EPT interpretation and
jackhammer) esophagus, absent peristalsis, and ineffective will illustrate how these techniques are used in the manage-
esophageal motility (IEM) have also been reported to be ment of esophageal motility disorders. In addition, it will
associated with dysphagia and/or chest pain.1 discuss the use of HRM in the perioperative evaluation
Esophageal manometry is the clinical test that defines of patients undergoing foregut operations, with a focus
the contractile characteristics of the esophagus to identify on antireflux surgery.
and classify motility disorders. Manometric evaluation
of the tubular esophagus assesses the integrity, rate of
progression, and morphology of the contractile complex
TECHNIQUES OF ESOPHAGEAL MANOMETRY
(amplitude, duration, repetitive contractions). Classifica- The utility of esophageal manometry in clinical practice
tion strategies grounded in conventional manometry have resides in two domains: (1) to accurately define esophageal
characterized esophageal motor patterns with three to eight motor function and (2) to delineate a treatment plan
pressure sensors spaced 3 to 5 cm apart, using pressure based on motor abnormalities.
displayed along a time axis. However, with recent advances
in pressure transduction hardware, computer processing, TECHNICAL ASPECTS
and analysis software, conventional manometry has been Esophageal manometry is a test in which intraluminal
rapidly supplanted by high-resolution manometry (HRM) pressure sensors, either water perfused or solid state, are
and esophageal pressure topography (EPT) analysis as positioned axially within the esophagus to quantify the
the methodology of choice. HRM and EPT were initially contractile characteristics of the esophagus and segregate
described experimentally by Clouse in the 1990s2 and have it into functional regions. The probe/catheter is inserted
now become widely available for clinical practice through transnasally and connected to a recording unit (via a
his initiatives with industry partners. Using EPT, pressure hydraulic pump in case of perfused pressure sensors).
data are presented as a seamless dynamic not only in Whereas conventional technique used probes with 3 to
time but also along the length of the esophagus. A key 8 pressure sensors spaced 3 to 5 cm apart, HRM typically
advantage is in the ability to assess pressure profiles along uses 36 solid-state pressure sensors spaced at 1-cm intervals.
the vertical (length) axis of the esophagus (spatial-pressure The concept of HRM is to use a sufficient number of pres-
variation plots) improving both the accuracy and detail sure sensors within the esophagus such that intraluminal
of the study compared with the conventional techniques pressure can be monitored as a continuum along the
that it replaces. entire length of the esophagus, much as time is viewed as
In addition to its use as a tool to diagnose esophageal a continuum in line tracings of conventional manometry.
motility disorders, HRM plays an important role in the Fig. 8.1 superimposes representative conventional and
evaluation of esophageal function in patients before HRM recordings with the HRM displayed in EPT format.
and after foregut operations, particularly laparoscopic The most common currently available HRM catheters
antireflux surgery. Long-term functional complications consist of 36 pressure sensors spaced 1 cm apart. These
such as dysphagia and gas bloat syndrome are concerning devices provide sufficient recording length (35 cm) for
sequelae that occur in some patients after laparoscopic the recording to span from the hypopharynx to the
fundoplication. To identify patients who may be at risk for stomach (with several intragastric sensors) without need
developing postoperative dysphagia, it is recommended for probe repositioning during the course of a study. HRM
115
High-Resolution Esophageal Manometry: Techniques and Use in the Diagnosis of Esophageal Motility Disorders and for Surgical Decision Making CHAPTER 8 115.e1
ABSTRACT
High-resolution manometry is the gold standard test
for evaluation of esophageal motility, which includes
both the efficacy of esophageal body contractility and
esophagogastric junction relaxation in response to swal-
lowing. It is the best study for the work-up of patients
with dysphagia and noncardiac chest pain, in whom a
mechanical obstruction has been ruled out via upper
endoscopy. As such, manometric parameters form the
basis for diagnosis of major and minor esophageal motility
disorders, including achalasia. This chapter will describe
in detail the technical aspects of performing and analyzing
a high-resolution manometry study. The current Chicago
Classification of esophageal motility disorders serves as a
basis for interpretation of these studies and classification of
patients into diagnostic categories. Based on manometric
findings, the resulting major and minor motility disorders
and their treatments are discussed. Lastly, the use of
manometry in the preoperative evaluation of patients with
gastroesophageal reflux disease is explored, as well as the
use of manometry for selective “tailoring” of fundoplication
for patients undergoing antireflux surgery.
KEYWORDS
high-resolution manometry; esophageal manometry;
esophageal motility; achalasia; gastroesophageal reflux
disease; antireflux surgery
116 SECTION I Esophagus and Hernia
40
mm Hg
0
40
mm Hg
mm Hg 150
0
100
mm Hg 100
0
100
mm Hg
0 50
40
mm Hg
0 0
Manometric port
20
mm Hg
Manometric sleeve 0
Times (s)
FIGURE 8.1 Depiction of normal esophageal motility by fluoroscopy (pink tracings), conventional manometry (white line tracings), and
high-resolution manometry with esophageal pressure topography (EPT). The positions of pressure sensors in the esophagus for the
conventional manometry are indicated in the anatomic drawing on the left with a sleeve device across the esophagogastric junction
(EGJ). On the right the conventional manometry tracings, bolus distribution, and EPT are overlaid to illustrate how they correspond with
each other. The peristaltic wave strips the bolus from the esophagus, with the pressure upstroke (conventional) or isobaric contour (EPT)
demarcating the bolus/no bolus interface. Note the tremendously enhanced detail provided by EPT, especially in the area of the EGJ.
offers several theoretical advantages over conventional the subsequent swallow. The manometric diagnosis is based
manometry: (1) the technique lends itself to standardized on the analysis of the 10 5-mL test swallows. Increased
objective metrics of interpretation, (2) it is easier to water volume (10, 20 mL) can be given to stress peristalsis,
perform studies of uniform high quality, (3) movement and multiple rapid swallows may be used to assess degluti-
artifact attributable to a relative change in the position tive inhibition,4 but these challenges have not yet been
of a sensor and contractile zone (especially sphincters) is sufficiently standardized to serve as diagnostic criteria.
minimized, and (4) the process of interpretation is more However, multiple rapid swallows are a simple method
intuitive and more easily learned by trainees naïve to either for assessing the integrity of deglutitive inhibition, defects
conventional or high-resolution manometric formats.3 of which are thought to be responsible for some motility
disorders. Finally, the consistency of the bolus can be varied
MANOMETRIC PROTOCOL using viscous solutions or solids such as marshmallow or
Esophageal manometry is usually performed in the supine bread. However, again, these challenges have not yet been
position. This position allows the testing of peristaltic sufficiently standardized to serve as diagnostic criteria.
function without the effect of gravity on bolus transit and
esophageal contractile pressures are augmented when ESOPHAGEAL PRESSURE TOPOGRAPHY
supine. Historically, using perfused pressure sensors, the When HRM is coupled with sophisticated algorithms to
supine position was mandatory to have all the sensors at display the manometric data as pressure topography plots,
the same height as the external pressure transducers, esophageal contractility is visualized with isobaric condi-
thereby avoiding pressure offsets secondary to hydrostatic tions among sensors indicated by isocoloric regions on the
pressure. With solid-state transducers, this is no longer an pressure topography plots. In EPT plots (or Clouse plots)
issue and studies can be performed in the upright position, the y-axis represents the axial length of the esophageal
which some argue to be more physiologic. However, all body, with the pharynx and upper esophageal sphincter
currently available normative data have been established (UES) at the top of the graph and the esophagogastric
in the supine position. junction (EGJ) and proximal stomach at the bottom. The
A typical manometry protocol consists of a 30-second x-axis represents time, so that peristaltic pressure waves
basal period without swallowing followed by 10 test 5-mL can be seen propagating to the right during swallows.
water swallows. Test swallows are separated by at least Pressure is represented as color, with “hot” colors (red,
20 seconds to reestablish basal activity and avoid having orange) representing higher pressures and “cool” colors
deglutitive inhibition from the prior swallow modulating (green, blue) depicting lower pressures.
High-Resolution Esophageal Manometry: Techniques and Use in the Diagnosis of Esophageal Motility Disorders and for Surgical Decision Making CHAPTER 8 117
Swallow
UES
5
P mm Hg
Length along the esophagus (cm)
10 150
15 100
30-mm Hg isobaric contour
20
50
30
25
D
CDP 0
30
EGJ relaxation
Deglutitive window 5s
35
Times (s)
FIGURE 8.2 A normal swallow in an esophageal pressure topography plot. Before and after the swallow, two high-pressure zones are
visualized: the upper esophageal sphincter (UES) and the esophagogastric junction (EGJ). The highlighted black line is the 30-mm Hg
isobaric contour circumscribing areas on the plot with intraluminal pressure greater than 30 mm Hg. The peristaltic esophageal
contraction is characterized by two troughs, one proximal (P) and one distal (D). The contractile deceleration point (CDP) represents the
inflexion point in the contractile front propagation. It is localized by fitting two tangential lines to the initial and terminal portions of the
30-mm Hg isobaric contours and noting intersection of the lines (white dot). The EGJ relaxation window, extending for 10 seconds after
UES relaxation, is the area in which deglutitive EGJ relaxation is assessed.
Fig. 8.2 depicts a normal swallow in a high-resolution and terminal portions of the 30-mm Hg isobaric contour
EPT plot encompassing both sphincters and the interven- and noting intersection of the lines as illustrated in Fig. 8.2.
ing esophagus; the relative timing of sphincter relaxation
and segmental contraction as well as the position of ALGORITHM OF ANALYSIS USING PRESSURE
the transition zone are all readily demonstrated. The TOPOGRAPHY PARAMETERS
swallowing sequence is described as follows. The UES The algorithm for classifying esophageal motor disorders
relaxation induced by swallowing is followed by a peristaltic using EPT is based on a systematic analysis that begins
contraction in the esophageal body that is dependent on by separating the plot into two functional domains: the
the regional gradient of inhibitory neurons within the EGJ and esophageal body. This system for analysis was
myenteric plexus. The peristaltic esophageal contraction developed after the advent of HRM and relies on the use
is preceded by a period of latency or quiescence in the of computer calculations to measuring metrics specific
esophageal body and contractile activity is not generated to EPT. The resulting algorithm for HRM measurement
until the period of inhibition or latency is supplanted and subsequent classification and diagnosis of esophageal
by excitatory activity at that particular location. Swallow- contractility disorders has been termed the Chicago Clas-
induced EGJ relaxation also begins just after UES relaxation sification.5 Drawing on an initial experience using the
and ends when the propagated esophageal contraction system, the Chicago Classification has subsequently been
reaches the EGJ. The peristaltic contraction is character- updated by international working groups to the current
ized by two major pressure troughs, one proximal and version 3.0 (v3.0).6 The remainder of the chapter will be
one distal (see P and D in Fig. 8.2). A middle pressure based on this current iteration.
trough is sometimes evident, but this is variable among Chicago Classification analysis begins with an assess-
individuals. Another notable feature of peristalsis is an ment of EGJ function because abnormal EGJ pressure
inflexion point in propagation velocity as the contraction morphology or impaired deglutitive EGJ relaxation can
nears the EGJ. This inflexion point, termed the contractile profoundly affect peristalsis and pressure topography
deceleration point (CDP), demarcates the initial segment within the esophageal body. EGJ abnormalities are also
of the esophageal contraction dominated by esophageal of important clinical significance because bolus transport
peristalsis from the later portion of the contraction during depends on the balance among resistance through the
which ampullary emptying transpires. The CDP can be EGJ, intrabolus pressure (IBP), and esophageal closure
localized objectively by fitting tangential lines to the initial pressure behind the bolus.7 Consequently, the first step
118 SECTION I Esophagus and Hernia
in analyzing esophageal motility should focus on the occurred). However, subsequent studies demonstrated
EGJ. Consistent with this, a stepwise analysis algorithm that significant breaks in peristalsis frequently occur in
first characterizes EGJ pressure morphology (presence healthy individuals, especially at the transition zone in
of hiatus hernia) and the adequacy of deglutitive EGJ the proximal esophagus between striated and smooth
relaxation. The implications of abnormal EGJ pressure muscle, and that these breaks are not a reliable measure
morphology on clinical classification have yet to be fully for defining clinically relevant diagnostic categories. 8
defined, but physiologic data support the concept that As a result, the Chicago Classification v3.0 focuses on
there is a strong interaction between EGJ structure and assessing the effectiveness of individual swallows based
esophageal function, as well as competence of the EGJ on contractile vigor, or the summed pressure front of the
valve mechanism in preventing gastroesophageal reflux peristaltic wave of each swallow, irrespective of breaks in
(GER).7 The consequences of impaired deglutitive EGJ the wave pattern. This is done by measuring the distal
relaxation are more obvious, leading to increased distal contractile integral (DCI), which can be conceptualized
esophageal (or panesophageal) IBP. Hence, although as the volume of the pressure topography graph in the
EGJ pressure morphology will likely be incorporated into peristaltic wave distal to the transition zone. Swallows are
future diagnostic categories, the first branch point in the classified as failed, weak, normal, or hypercontractile,
current scheme is of normal or impaired EGJ relaxation based on DCI. This method of analysis both simplifies the
because this consistently affects esophageal function. classification of swallows and makes it a more clinically
After defining EGJ anatomy and deglutitive relaxation, relevant assessment of bolus clearance. Swallows are then
the next step in analysis is focused on the esophageal further characterized by distal latency (DL) and peristaltic
peristalsis. The topography pattern of individual swallows breaks to identify instances of premature contraction (i.e.,
are each classified according to the Chicago Classifica- spasm) or fragmented peristalsis (currently defined as a
tion parameters shown in Table 8.1. Earlier versions of minor disorder of esophageal motility of unclear clinical
the Chicago Classification focused on the integrity of significance).
peristalsis (i.e., whether breaks in the peristaltic waves In addition to the contractile pattern, each swallow
is examined for an abnormal esophageal pressurization
with the contractile activity. This is a unique feature of
EPT as these patterns are much more evident in pressure
TABLE 8.1 Esophageal Pressure Topography Scoring of topography compared with the conventional line-tracing
Individual Swallows format. After all test swallows are characterized, the study
results are summarized using the classification algorithm,
CONTRACTION VIGOR as presented in Fig. 8.3 and Table 8.2.
Failed DCI <100 mm Hg-s-cm
Weak DCI >100 mm Hg-s-cm, but Esophagogastric Junction Morphology
<450 mm Hg-s-cm Both the lower esophageal sphincter (LES) and the sur-
Ineffective Failed or weak rounding crural diaphragm (CD) contribute to measured
Normal DCI ≥450 mm Hg-s-cm, but intraluminal EGJ pressure. The CD component is most
<8000 mm Hg-s-cm
evident during inspiration but probably also contributes a
Hypercontractile DCI ≥8000 mm Hg-s-cm
minor component to EGJ pressure during expiration. Thus
CONTRACTION PATTERN there are two major confounding variables in describing
Premature DL <4.5 s EGJ intraluminal pressure: phase of the respiratory cycle
Fragmented Large break (>5 cm length) in the and the relative positions of the LES and the CD. No
20 mm Hg isobaric contour with consensus was ever achieved with conventional manometry
DCI >450 mm Hg-s-cm on how to deal with either of these variables. Indeed,
Normal contraction Not achieving any of the above there was generally little recognition of the EGJ as a
diagnostic criteria complex sphincter, instead simply referring to it as the
INTRABOLUS PRESSURE PATTERN (30-mm Hg ISOBARIC LES on manometry studies. With HRM, the sphincteric
CONTOUR) contributions of the CD and LES become somewhat
Panesophageal Uniform pressurization extending obvious and the relative localization of the LES and CD
pressurization from the UES to the EGJ elements defines EGJ morphologic subtypes (Fig. 8.4). The
Compartmentalized Pressurization extending from the magnitude of CD augmentation of EGJ pressure during
esophageal contractile front to the EGJ normal respiration is readily quantified. A retrospective
pressurization analysis of the relationship between these attributes of
EGJ pressurization Pressurization restricted to zone EGJ pressure topography and GERD found that GERD
between the LES and CD with patients had significantly greater CD-LES separation (i.e.,
LES-CD separation (i.e., presence hiatal hernia) compared with either controls or non-
of a hiatal hernia)
GERD patients.9 GERD patients also had significantly less
Normal pressurization No bolus pressurization >30 mm Hg
inspiratory CD augmentation compared with controls or
CD, Crural diaphragm; DCI, distal contractile integral; DL, distal latency; non-GERD patients. Furthermore, in a logistic regression
EGJ, esophagogastric junction; LES, lower esophageal sphincter; UES, model, only inspiratory augmentation was found to have a
upper esophageal sphincter.
From Kahrilas PJ, Bredenoord AJ, Fox M, et al. The Chicago Classification
significant independent association with GERD, suggesting
of esophageal motility disorders, v3.0. Neurogastroenterol Motil. that CD impairment was the mediator of both the hiatus
2015;27:160–174. hernia and LES hypotension effects.
High-Resolution Esophageal Manometry: Techniques and Use in the Diagnosis of Esophageal Motility Disorders and for Surgical Decision Making CHAPTER 8 119
Achalasia
IRP ≥ ULN and 100% failed Yes Type I: No contractility
1 peristalsis or spasm Type II: ≥20% PEP
Type III: ≥20% Spasm (DL <4.5 s) Disorders with EGJ
No outflow
EGJ outflow obstruction obstruction
IRP ≥ ULN and not Yes • Incompletely expressed
2
type I-III achalasia achalasia
• Mechanical obstruction
No
DES
• ≥20% Premature (DL <4.5 s) Major disorders
IRP normal and Yes Jackhammer esophagus of peristalsis
3 short DL or high DCI or ≥20% DCI >8000 mm Hg-s-cm • Entities not seen in
100% failed peristalsis Absent contractility normal subjects
• No scorable contraction
No • Consider achalasia
FIGURE 8.3 Algorithm for analysis of esophageal pressure topography studies according to the Chicago Classification v3.0. Note that
motility disorders should be considered as a cause of dysphagia and/or chest pain only after first evaluating for structural disorders,
eosinophilic esophagitis, and, where appropriate, cardiac disease. The first branch point is to identify patients meeting criteria for
achalasia (elevated integrated relaxation pressure [IRP] and absent peristalsis), which is then subclassified. If IRP is normal, then other
peristaltic abnormalities such as spasm and hypercontractility are identified if present. Major disorders of peristalsis are shown above the
dotted line, with minor disorders and normal peristalsis below it. DCI, Distal contractile integral; DES, distal esophageal spasm; DL, distal
latency; EGJ, esophagogastric junction; PEP, panesophageal pressurization; ULN, upper limit of normal.
Length along the esophagus (cm)
mm Hg
0 UES UES 150
UES
5
5s 5s 5s
10 100
15
20 LES 50
LES
25 5 cm
1.5 cm
30 LES CD 0
CD CD
CD CD
35 CD CD
Time (s) Time (s) Time (s)
Type I Type II Type III
FIGURE 8.4 Esophagogastric junction (EGJ) morphology characterized in esophageal pressure topography. The two main EGJ
components are lower esophageal sphincter (LES) and crural diaphragm (CD), which cannot be independently quantified when
superimposed, classified as type I EGJ. With a type II EGJ, the LES and CD are separated by 1 to 2 cm, and in a type III the LES and
CD are separated by more than 2 cm. A type III EGJ is the manometric criterion for hiatal hernia. UES, Upper esophageal sphincter.
Finally, dynamic HRM studies during reflux monitoring morphology, it is less clear that any measure of basal EGJ
revealed that this is not a static situation. Rather, GERD pressure has much significance.
patients oscillated between types I and II EGJ conforma-
tions. Reflux events preferentially occurred during the Esophagogastric Junction Relaxation
periods of type II conformation characterized by a small Incomplete deglutitive EGJ relaxation is an essential feature
separation of the two high-pressure areas.10 Paradoxically, in the diagnosis of achalasia, and achalasia is not only the
in contrast to the findings related to the CD and EGJ best-defined esophageal motor disorder but also the one
120 SECTION I Esophagus and Hernia
Diagnostic Criteria
ACHALASIA AND EGJ OUTFLOW OBSTRUCTION
Type I achalasia (classic achalasia) Elevated median IRP (>15 mm Hg), 100% failed peristalsis (DCI <100 mm Hg-s-cm)
Type II achalasia (with esophageal Elevated median IRP (>15 mm Hg), 100% failed peristalsis, panesophageal pressurization
compression) with ≥20% of swallows
Contractions may be masked by esophageal pressurization and DCI should not be calculated
Type III achalasia (spastic Elevated median IRP (>15 mm Hg*), no normal peristalsis, premature (spastic) contractions
achalasia) with DCI >450 mm Hg-s-cm with ≥20% of swallows
May be mixed with panesophageal pressurization
EGJ outflow obstruction Elevated median IRP (>15 mm Hg), sufficient evidence of peristalsis such that the criteria for
types I–III achalasia are not met†
OTHER MAJOR MOTILITY DISORDERS
Absent contractility Normal mean IRP, 100% of swallows with failed peristalsis
Achalasia should be considered when IRP values are borderline and when there is evidence
of esophageal pressurization
Premature contractions with DCI values <450 mm Hg-s-cm meet criteria for failed
peristalsis
Distal esophageal spasm Normal mean IRP, ≥20% premature contractions with DCI >450 mm Hg-s-cm. Some normal
peristalsis may be present.
Hypercontractile esophagus At least two swallows DCI >8000 mm Hg-s-cm
(jackhammer) Hypercontractility may involve, or even be localized to, the LES
MINOR DISORDERS OF PERISTALSIS
Ineffective esophageal motility ≥50% ineffective swallows
(IEM) Ineffective swallows may be failed or weak (DCI <450 mm Hg-s-cm)
Multiple repetitive swallow assessment may be helpful in determining peristaltic reserve
Fragmented peristalsis ≥50% fragmented contractions with DCI >450 mm Hg-s-cm
Normal esophageal motility Not fulfilling any of the above classifications
*Cutoff values dependent on the manometric hardware; these are the cutoffs for the Sierra device.
†
Potential etiologies: early achalasia, mechanical obstruction, esophageal wall stiffness, or manifestations of hiatal hernia.
DCI, Distal contractile integral; DL, distal latency; EGJ, esophagogastric junction; IRP, integrated relaxation pressure; LES, lower esophageal sphincter.
From Kahrilas PJ, Bredenoord AJ, Fox M, et al. The Chicago Classification of esophageal motility disorders, v3.0. Neurogastroenterol Motil. 2015;27:
160–174.
with the most specific treatments. These features impart for distinguishing well-defined achalasia patients from
important clinical relevance on the accurate detection of control subjects and patients with other diagnoses.11 It
incomplete deglutitive EGJ relaxation. Despite this cardinal should be noted that the measurement of IRP is variable
significance, there has never been a unified convention depending on the manometry probe and analysis software
for defining incomplete deglutitive EGJ relaxation with that is being used. The Chicago Classification v3.0 set
conventional manometry. Furthermore, numerous poten- the cutoff for normal at less than 15 mm Hg for the
tial confounding factors exist, including CD contraction Sierra design catheters and less than 28 mm Hg for the
during respiration, esophageal shortening, hiatal hernia, Unisensor design. A model-specific cutoff value should
IBP through the EGJ, sphincter radial asymmetry, and be used when categorizing EGJ relaxation.
movement of the recording sensor relative to the EGJ.
With HRM, the ease and reliability of measurement of Contraction Vigor
EGJ relaxation is greatly improved. Pressure topography The vigor of the distal esophageal contraction between
plotting facilitates accurate localization of the EGJ and the the major pressure nodes P and D is quantified
deglutitive relaxation window, as illustrated in Fig. 8.2. An using the DCI. Conceptually the DCI corresponds to
exploratory study comparing criteria for detecting impaired the volume of the distal contraction in dimensions of
deglutitive EGJ relaxation within that relaxation window in time, length, and amplitude between the proximal and
a large group of patients and control subjects concluded the distal troughs using the 20-mm Hg isobaric contour
that the optimal measure for quantifying deglutitive relax- at the base and expressed as (mm Hg-s-cm) (Fig. 8.6B).
ation was the integrated relaxation pressure (IRP), with It is calculated by multiplying the mean pressure of the
normal being defined as less than 15 mm Hg. The IRP is contraction (less 20 mm Hg), duration of the contraction,
amenable to automated calculation, and conceptually it is and the length of the esophageal segment between the
the lowest average pressure for 4 seconds (either contigu- proximal and distal troughs. Based on this value, swallows
ous or noncontiguous) within the 10-second relaxation are classified as failed (DCI < 100 mm Hg-s-cm), weak
window (Fig. 8.5). This single measure of deglutitive EGJ (DCI > 100, but < 450 mm Hg-s-cm), normal (DCI from
relaxation exhibited 98% sensitivity and 96% specificity 450 to 8000 mm Hg-s-cm) or hypercontractile (DCI >
High-Resolution Esophageal Manometry: Techniques and Use in the Diagnosis of Esophageal Motility Disorders and for Surgical Decision Making CHAPTER 8 121
5
mm Hg
50
Length along the esophagus (cm) 10
15 25
20
0
25
10
30
FIGURE 8.5 Concomitant esophageal pressure topography (EPT) and fluoroscopy during esophageal emptying illustrating the transition
from peristaltic transport to ampullary emptying. The fluoroscopic images in the windows are synchronized with the EPT plot. The white
and blue dots indicate areas of intrabolus pressure and the onset of luminal closure, respectively. The second image (at about time 8
seconds) is near the contractile deceleration point, evident both by the transition of the fluoroscopic image to ampullary conformation and
slowing of the luminal closure front. The colored rectangles within the deglutitive relaxation window indicate the time fragments used to
compute the integrated relaxation pressure, whereas the pink rectangles above the esophagogastric junction show the times used to
compute intrabolus pressure.
0 mm Hg
150
Length along the esophagus (cm)
10 Proximal defect 8 cm
100
15
20
50
25
20
30
0
5s 5s 5s
35
A Time (s) B Time (s) C Time (s)
FIGURE 8.6 Varying degrees of peristaltic integrity. In each panel, the black line represents the 20-mm Hg isobaric contour. The swallow
in (A) is intact (no disruption in the 20 mm Hg isobaric contour). (B) A swallow with a large break in the 20 mm Hg isobaric contour.
(C) A failed swallow (absence of 20 mm Hg integrity in the distal two-thirds of the esophagus resulting in a DCI <100 mm Hg-s-cm).
8000 mm Hg-s-cm). Failed and weak swallows are together concept of latency to quantify this period of quiescence
classified as “ineffective” swallows. and suggested that patients with spasm had a substantial
reduction in contractile latency.12 Distal contractile latency,
Distal Contractile Latency measured from the onset of the swallow to the onset of
The esophageal deglutitive response is initiated with the the contraction, was shorter in patients with simultaneous
oropharyngeal swallow. However, the subsequent peristaltic contractions than in those with normal peristaltic velocity.
contraction in the distal esophagus is preceded by a In EPT terms, distal contractile latency (DL) is defined as
period of quiescence. Behar and Biancani introduced the the duration of the interval between UES relaxation and
122 SECTION I Esophagus and Hernia
0
mm Hg
5 150
Length along the esophagus (cm) P P
10
15 100
DCI 3220
CFV 30 mm Hg 20 mm Hg
mm Hg-s-cm
20 3.6 cm/s isobaric contour isobaric contour
25 50
CDP
DL 7.1 s
30
D D
5s 5s 0
35
Time (s) Time (s)
A B
FIGURE 8.7 Metrics used in the analysis of esophageal pressure topography. The distal latency is measured from the onset of the UES
relaxation (dashed vertical line) to the contractile deceleration point (A). The distal contractile integral (DCI) corresponds to the entire
volume (amplitude × time × duration) of the distal contraction spanning from the proximal to the distal troughs (pink box) above
20 mm Hg (B). The DCI is calculated by multiplying the average pressure × the duration × the length of the contractile segment contained
in the pink box. CFV, Contractile front velocity; D, distal; DCI, distal contractile integral; DL, distal latency; P, proximal.
0 mm Hg
150
Length along the esophagus (cm)
5
30 mm Hg
10
isobaric
contour 100
15
Pan
No pressurization esophageal
20 pressurization CFV
(30 mm Hg 42 cm/s 50
25 isobaric
contour)
DL 3.0 s
30
Absence of EGJ relaxation Absence of EGJ relaxation Absence of EGJ relaxation 0
(IRP 18.4 mm Hg) 5s (IRP 33.7 mm Hg) 5s (IRP 26.4 mm Hg) 5s
35
Time (s) Time (s) Time (s)
A Type 1 B Type 2 C Type 3
FIGURE 8.8 Achalasia subtypes. All three subtypes are characterized by impaired esophagogastric junction (EGJ) relaxation (integrated
relaxation pressure [IRP] >15 mm Hg) and absent peristalsis. In type 1 (A), there is negligible pressurization in the esophageal body,
evident by the absence of any area circumscribed by the 30-mm Hg isobaric contour (black line). In type 2 (B), panesophageal
pressurization occurs evident by the banding pattern of the 30-mm Hg isobaric contour spanning from the upper esophageal sphincter
to the EGJ. This represents elevated intrabolus pressure and is associated with contraction of the longitudinal muscle on the muscularis
propria. Type 3 achalasia (C) is characterized by spastic contractions (short distal latency [DL]) in the esophageal body.
temporizing maneuvers than as definitive therapies. The no data exist on botulinum toxin as a long-term treat-
definitive treatments of achalasia are disruption of the ment strategy for achalasia. Studies comparing botulinum
LES either surgically (laparoscopic Heller myotomy or toxin injection with pneumatic dilation suggest that the
peroral endoscopic myotomy [POEM]) or with endoscopic expense of repeated injection outweighs the potential
pneumatic dilation. economic benefits of added safety, unless the patient’s
life expectancy is minimal. Thus this option is mainly
Pharmacologic Therapy reserved for elderly or frail individuals who are poor risks
Smooth muscle relaxants such as nitrates (isosorbide for definitive treatments.
dinitrate) or sildenafil and calcium channel blockers
(diltiazem, nifedipine, verapamil) have been proposed as Pneumatic Dilation
pharmacologic treatment of achalasia. They are admin- Therapeutic dilation for achalasia requires distention of
istered immediately before eating, and they can relieve the LES to a diameter of at least 30 mm to effect a lasting
dysphagia in achalasia by reducing the LES pressure. reduction of LES pressure, presumably by partially disrupt-
However, placebo-controlled crossover trials have found ing the circular muscle of the sphincter. Dilation with an
only minimal benefit.16 Side effects also limit the use of endoscope, standard bougies (up to 60 French), or with
these drugs (headache, hypotension for nitrates; flushing, through-the-scope balloon dilators (up to 2 cm) provides
dizziness, headache, peripheral edema, and orthostasis very temporary benefit at best, and these are not consid-
for nifedipine). ered to be definitive treatments. Only dilators specifically
designed to treat achalasia are able to achieve adequate
Botulinum Toxin Injection sphincter disruption for lasting effectiveness. The basic
Because LES tone is partially mediated via a cholinergic element of an achalasia dilator is a long, noncompliant,
pathway, blockade of acetylcholine release from excitatory cylindrical balloon that can be positioned fluoroscopically
motor neurons should partially eliminate the neurogenic (Rigiflex dilator) or endoscopically (Witzel dilator) across
component of LES pressure, thereby decreasing LES the LES and then inflated to a characteristic diameter in a
pressure. Botulinum toxin (Botox) irreversibly inhibits controlled fashion using a hand-held manometer. Failure
the release of acetylcholine from presynaptic choliner- of response with a 30-mm dilator can be retreated with
gic terminals. However, because this effect is eventually greater-diameter balloons, sequentially progressing to 35
reversed by the growth of new axons, botulinum toxin is to 40 mm diameters if needed.
not long-lasting therapy. The initial landmark study of The reported success rates for pneumatic dilation vary
botulinum toxin in achalasia reported that intrasphincteric from 70% to 90%.18 The need for further dilations is
injection of 80 units of botulinum toxin decreased LES determined by the persistence of symptoms approximately
pressure by 33% and improved dysphagia in 66% of 4 weeks after treatment. In this case, a larger-diameter
patients for a 6-month period.17 Although some efficacy balloon (35 mm, sometimes 40 mm) may be used. The
is noted in a majority of achalasia patients, the effects are major complication of pneumatic dilation is esophageal
temporary. Success rates drop from 80% to 90% after 1 perforation; however, mortality is very rare. The reported
month to 53% to 54% after 1 year. Although there are incidence of esophageal perforation from pneumatic
data to suggest that repeat treatments can be effective, dilation ranges between 0.4% and 5%.18
124 SECTION I Esophagus and Hernia
Laparoscopic Heller Myotomy 15 mm Hg arguing for further therapy targeting the EGJ.
Although a more detailed description of surgical therapy Fluoroscopy is useful both to identify anatomic problems,
for achalasia is included elsewhere in this textbook, a brief as well as to evaluate esophageal emptying.25
description of the surgery and outcomes are described In the case of a patient who failed therapy and was not
here. Modern surgical procedures for treating achalasia are previously operated on, further treatment could potentially
variations on the esophagomyotomy originally described be repeat dilation, Heller myotomy, or POEM. In patients
by Heller in 1913, consisting of an anterior and posterior who have already undergone myotomy, manometric demon-
myotomy performed through either a laparotomy or stration of an inadequate myotomy or functional esophageal
a thoracotomy. The Heller myotomy has subsequently obstruction from the antireflux component of the surgery
evolved to the current standard in which a single, anterior usually requires reoperation, but pneumatic dilation can
myotomy is performed laparoscopically and a partial be pursued as an alternative. POEM provides an attractive
fundoplication is added to prevent postoperative iatrogenic option for reoperation after prior Heller myotomy because
GER. Published series of the efficacy of Heller myotomy it avoids the need for extensive adhesiolysis, reopening
in treating achalasia report good to excellent results in of the hiatus, and takedown of the prior fundoplication.
around 90% of patients, with persistent dysphagia troubling In extremely advanced or refractory cases of achalasia,
fewer than 10% of patients.19 esophageal resection may be the only surgical option.
0
mm Hg
150
5
Length along the esophagus (cm)
10
100
15 CFV
CFV 17 cm/s
17 cm/s
20
DL 7.6 s 50
25 DL 3.0 s
30
30
Normal EGJ relaxation
Normal EGJ relaxation 5s (IRP 10.4 mm Hg) 5s 0
35 (IRP 10.7 mm Hg)
Time (s) Time (s)
A B
FIGURE 8.9 Examples of spasm and rapidly propagated peristalsis. In the context of normal esophagogastric junction (EGJ) relaxation,
a rapidly propagated contraction with short latency (distal latency [DL] < 4.5 seconds) defines esophageal spasm (A). However, this
example of a rapidly propagated contraction occurring with normal DL was encountered in an asymptomatic control (B). CFV, Contractile
front velocity; IRP, integrated relaxation pressure.
with dilation. However, an important caveat is that it peristaltic amplitude greater than 180 mm Hg between
is completely uncertain as to whether or not patients 3 and 8 cm above the LES.35 Subsequently, the defining
who benefited by pneumatic dilation would not be more peristaltic amplitude has been debated, and more recent
properly categorized as spastic achalasia or achalasia work suggests that this should be increased to 260 mm Hg,
with esophageal compression, emphasizing the need for a value more likely to be associated with chest pain and
accurate manometric classification. dysphagia.33
Botulinum toxin injection is a pathophysiologically The introduction of HRM and EPT has allowed further
attractive approach to treating patients with spastic stratification of hypertensive peristalsis to account for
disorders. Therapeutic trials suggest it can reduce chest both excessive amplitude and abnormal morphology
pain.30 The technique has not been standardized in this of the peristaltic contraction. The summary metric for
application with some reports injecting botulinum toxin contractile vigor in the entire distal segment is the DCI
only at the level of the EGJ and others also injecting the with a value of 5000 mm Hg-s-cm being the 95th percentile
distal esophagus. No trial has yet compared botulinum of normal. DCI values greater than 5000 mm Hg-s-cm but
toxin injection with other treatments. less than 8000 mm Hg-s-cm are found in individuals with
hypertensive peristalsis akin to nutcracker esophagus in
Surgical Treatment conventional terms. However, because values in this range
Long myotomy extending from the LES proximally onto are also encountered in normal individuals, they have now
the esophageal body has been used to treat patients with been classified as normal in the Chicago Classification v3.0.
spastic disorders. The extent of the myotomy may be guided In contrast, DCI values greater than 8000 mm Hg-s-cm
by manometric findings. In an uncontrolled study, surgical are almost universally associated with chest pain and
treatment seemed more effective than the medical treat- dysphagia, and these patients appear to have a more
ment.31 Due to its ability to create an extended proximal exaggerated pattern of hypercontractility that is repetitive
myotomy, POEM also offers a potentially superior surgical and more akin to a jackhammer than a nutcracker (Fig.
option for the treatment of DES. Only a handful of small 8.10). The current version of the Chicago Classification
case series have reported results after POEM performed refers to this condition as jackhammer esophagus to fit
for DES and other spastic esophageal motility disorders, better with the contractile morphology. Having said that,
so far showing excellent outcomes with a meta-analysis the clinical relevance of these conditions is still unclear.
reporting symptomatic success in 88% of patients.32 Nonetheless, it seems likely that focusing future trials
on patients with a DCI greater than 8000 mm Hg-s-cm
JACKHAMMER ESOPHAGUS rather than a lower threshold is more likely to identify a
Vigorous esophageal contractions with normal propagation homogeneous population potentially amenable to specific
have been reported in association with chest pain.33 The pharmacologic treatment.
pathophysiology of hypertensive peristalsis is unclear but
is believed to be related to either excessive excitation, Pharmacologic Treatment
reactive compensation for increased EGJ obstruction, or The same therapeutic options used for DES have also
myocyte hypertrophy.34 been advocated for patients with hypertensive peristalsis.
The conventional manometric definition of hypertensive Smooth muscle relaxants, such as calcium channel blockers
peristalsis used the term nutcracker esophagus and peak and nitrates, have been used for these disorders because
126 SECTION I Esophagus and Hernia
0
mm Hg
150
Length along the esophagus (cm) 5
30 20
Normal EGJ relaxation Normal EGJ relaxation 0
35 (IRP 5.0 mm Hg) 5s (IRP 3.9 mm Hg) 5s
FIGURE 8.10 Hypercontractile (jackhammer) esophagus. Hypercontractile swallows are defined by a distal contractile integral (DCI) greater
than 8000 mm Hg-s-cm (A). This example is associated with a hypercontractile lower esophageal sphincter but normal esophagogastric
junction (EGJ) relaxation. (B) An extremely abnormal contraction in a jackhammer esophagus patient with repetitive prolonged
contractions evoking the action of the jackhammer. CFV, Contractile front velocity; IRP, integrated relaxation pressure.
they reduce peristaltic amplitude even though neither has preoperative evaluation of such patients: first, it confirms
been shown to relieve chest pain or dysphagia in clinical conclusively that the patient does not have a previously
trials. Alternatively, sildenafil is appealing because of its undiagnosed major disorder of esophageal motility, such as
profound effect of reducing contraction amplitude and achalasia. This is important because a significant number
potentially reducing the occurrence of repetitive contrac- of patients with achalasia will have chest pain that they
tions.36 Again, there are no supportive clinical trial data describe as “heartburn” and are initially misdiagnosed
yet. Finally, botulinum toxin injection in the esophageal with GERD. Conversely, a number of patients with true
muscle with or without endoscopic ultrasound guidance GERD will complain of dysphagia as a result of esophageal
may be an option for patients with refractory symptoms. hypersensitivity and partially impaired motility resulting
Because of the potential overlap of hypertensive from prolonged reflux. Second, some surgeons choose to
peristalsis with GERD and the observation that many of “tailor” their fundoplication (i.e., decide to perform either
these patients have coexistent psychologic distress, therapy a partial or complete fundoplication) based on the results
focused on modulating acid secretion, visceral sensitivity, of preoperative HRM, in conjunction with the patient’s
and stress has been attempted. Proton pump inhibitors symptoms. Lastly, having a preoperative HRM serves as
have been proposed to treat hypertensive peristalsis based an important baseline if a patient develops troublesome
on the hypothesis that GERD can induce chest pain and symptoms postoperatively, such as dysphagia. Being able
hypertensive peristalsis.37 Similarly, treatment with low-dose to compare preoperative and postoperative HRM studies is
tricyclic antidepressants may reduce contractions via the invaluable in determining whether such patients’ symptoms
anticholinergic effect and may alter visceral sensitivity. Akin are due to elevated EGJ pressure or impaired esophageal
to DES, POEM has been proposed as an ideal surgical peristalsis as a result of the fundoplication.
option for treatment of patients with jackhammer esopha-
gus who are refractory to medical therapy. With POEM, FUNDOPLICATION TAILORING
the myotomy can be extended proximally to encompass Long-term functional problems, including dysphagia,
the entire hypercontractile segment of the smooth muscle gas bloat, diarrhea, and increased flatulence, are feared
esophagus, from the transition zone through the EGJ. complications of antireflux surgery. Some surgeons use
Again, due to the scarcity of the condition, reports of a tailored approach in which they perform a partial
POEM performed for jackhammer esophagus are just fundoplication (either anterior or posterior) in patients
beginning to appear and further data regarding its efficacy determined to be at high risk of postoperative dysphagia,
are needed.32 as shown in Fig. 8.11. Several studies have shown partial
fundoplication to result in lower rates of dysphagia as
compared with Nissen40; therefore if at-risk patients could
ROLE OF MANOMETRY IN be identified by preoperative HRM and other factors,
ANTIREFLUX SURGERY they could potentially benefit from such a modification.
However, the precise means for identifying such patients
HRM is an important component of the physiologic and the role of HRM measurements in this algorithm
evaluation of patients prior to antireflux surgery and is remain matters of considerable controversy.
now recommended by many as a routine preoperative The presence of preoperative dysphagia is an important
test, although this has not be universally accepted as risk factor for worsening, or persistence, of dysphagia
standard of care.38,39 HRM serves several roles in the after fundoplication.41 In addition, patients with IEM
High-Resolution Esophageal Manometry: Techniques and Use in the Diagnosis of Esophageal Motility Disorders and for Surgical Decision Making CHAPTER 8 127
CONCLUSION
Esophageal motility disorders are diagnosed and classified
using esophageal manometry, with HRM now being the
gold standard. The Chicago Classification is an evolv-
ing classification scheme based on the consensus of an
B international working group that periodically meets for
updates and clarifications, with v3.0 serving as the most
current edition.6 The major recognized disorders are the
three subtypes of achalasia, EGJ outflow obstruction, DES,
jackhammer esophagus, absent contractility, with IEM and
fragmented peristalsis now categorized as minor motility
disorders. HRM serves an important role, not only in the
diagnosis of these disorders, but also in prognostication
and guiding the most appropriate treatment option. In
addition, HRM forms an important component in the
preoperative evaluation of patients undergoing lapa-
roscopic antireflux surgery and can help to guide the
choice of fundoplication in a patient-specific manner. For
these reasons, it is important that surgeons have a firm
C understanding of the principles of HRM diagnosis and
its application in surgical decision making.
FIGURE 8.11 Examples of complete and partial fundoplications
performed during laparoscopic antireflux surgery. A Nissen (A) is ACKNOWLEDGMENTS
a complete 360-degree fundoplication that provides the most
effective and durable antireflux barrier but may result in increased The authors thank and acknowledge Sabine Roman, MD, Peter J.
rates of postoperative dysphagia. Some surgeons choose to tailor Kahrilas, MD, and John E. Pandolfino, MD, for their contribu-
their fundoplications based on manometry results and other tions to the prior edition of this chapter.
patient factors by performing a partial wrap, either a posterior
Toupet fundoplication (B) or an anterior Dor fundoplication (C). REFERENCES
1. Pandolfino JE, Kahrilas PJ, American Gastroenterological Associa-
tion. American Gastroenterological Association medical position
statement: clinical use of esophageal manometry. Gastroenterology.
on HRM are more likely to have dysphagia at baseline.42 2005;128:207-208.
Therefore it would seem that preoperative IEM would be 2. Clouse RE, Staiano A, Alrakawi A. Development of a topographic
a significant risk factor for dysphagia after fundoplication analysis system for manometric studies in the gastrointestinal tract.
Gastrointest Endosc. 1998;48:395-401.
and that such patients would be better served with a 3. Grubel C, Hiscock R, Hebbard G. Value of spatiotemporal repre-
partial wrap. However, most randomized trials comparing sentation of manometric data. Clin Gastroenterol Hepatol. 2008;6:525-
partial and complete fundoplication have failed to show 530.
128 SECTION I Esophagus and Hernia
4. Fornari F, Bravi I, Penagini R, Tack J, Sifrim D. Multiple rapid 25. Vaezi MF, Baker ME, Achkar E, Richter JE. Timed barium oesopha-
swallowing: a complementary test during standard oesophageal gram: better predictor of long term success after pneumatic dilation
manometry. Neurogastroenterol Motil. 2009;21:718-e41. in achalasia than symptom assessment. Gut. 2002;50:765-770.
5. Pandolfino JE, Fox MR, Bredenoord AJ, Kahrilas PJ. High-resolution 26. Grubel C, Borovicka J, Schwizer W, Fox M, Hebbard G. Diffuse
manometry in clinical practice: utilizing pressure topography to esophageal spasm. Am J Gastroenterol. 2008;103:450-457.
classify oesophageal motility abnormalities. Neurogastroenterol Motil. 27. Pandolfino JE, Ghosh SK, Rice J, Clarke JO, Kwiatek MA, Kahrilas
2009;21:796-806. PJ. Classifying esophageal motility by pressure topography charac-
6. Kahrilas PJ, Bredenoord AJ, Fox M, et al. The Chicago Classifica- teristics: a study of 400 patients and 75 controls. Am J Gastroenterol.
tion of esophageal motility disorders, v3.0. Neurogastroenterol Motil. 2008;103:27-37.
2015;27:160-174. 28. Eherer AJ, Schwetz I, Hammer HF, et al. Effect of sildenafil on
7. Mittal RK, Fisher M, McCallum RW, Rochester DF, Dent J, Sluss J. oesophageal motor function in healthy subjects and patients with
Human lower esophageal sphincter pressure response to increased oesophageal motor disorders. Gut. 2002;50:758-764.
intra-abdominal pressure. Am J Physiol. 1990;258:G624-G630. 29. Clouse RE, Lustman PJ, Eckert TC, Ferney DM, Griffith LS. Low-dose
8. Kumar N, Porter RF, Chanin JM, Gyawali CP. Analysis of interseg- trazodone for symptomatic patients with esophageal contraction
mental trough and proximal latency of smooth muscle contraction abnormalities. A double-blind, placebo-controlled trial. Gastroenterology.
using high-resolution esophageal manometry. J Clin Gastroenterol. 1987;92:1027-1036.
2012;46:375-381. 30. Storr M, Allescher HD, Rosch T, Born P, Weigert N, Classen M.
9. Pandolfino JE, Kim H, Ghosh SK, Clarke JO, Zhang Q, Kahrilas Treatment of symptomatic diffuse esophageal spasm by endoscopic
PJ. High-resolution manometry of the EGJ: an analysis of crural injections of botulinum toxin: a prospective study with long-term
diaphragm function in GERD. Am J Gastroenterol. 2007;102:1056-1063. follow-up. Gastrointest Endosc. 2001;54:754-759.
10. Bredenoord AJ, Weusten BL, Timmer R, Smout AJ. Intermittent 31. Patti MG, Pellegrini CA, Arcerito M, Tong J, Mulvihill SJ, Way LW.
spatial separation of diaphragm and lower esophageal sphincter favors Comparison of medical and minimally invasive surgical therapy for
acidic and weakly acidic reflux. Gastroenterology. 2006;130:334-340. primary esophageal motility disorders. Arch Surg. 1995;130:609-615,
11. Ghosh SK, Pandolfino JE, Rice J, Clarke JO, Kwiatek M, Kahrilas PJ. discussion 615–616.
Impaired deglutitive EGJ relaxation in clinical esophageal manometry: 32. Khan MA, Kumbhari V, Ngamruengphong S, et al. Is POEM the
a quantitative analysis of 400 patients and 75 controls. Am J Physiol answer for management of spastic esophageal disorders? A systematic
Gastrointest Liver Physiol. 2007;293:G878-G885. review and meta-analysis. Dig Dis Sci. 2017;62:35-44.
12. Behar J, Biancani P. Pathogenesis of simultaneous esophageal 33. Agrawal A, Hila A, Tutuian R, Mainie I, Castell DO. Clinical relevance
contractions in patients with motility disorders. Gastroenterology. of the nutcracker esophagus: suggested revision of criteria for
1993;105:111-118. diagnosis. J Clin Gastroenterol. 2006;40:504-509.
13. Roman S, Lin Z, Kwiatek MA, Pandolfino JE, Kahrilas PJ. Weak 34. Dogan I, Puckett JL, Padda BS, Mittal RK. Prevalence of increased
peristalsis in esophageal pressure topography: classification and esophageal muscle thickness in patients with esophageal symptoms.
association with dysphagia. Am J Gastroenterol. 2011;106:349-356. Am J Gastroenterol. 2007;102:137-145.
14. Pandolfino JE, Kwiatek MA, Nealis T, Bulsiewicz W, Post J, Kahrilas 35. Spechler SJ, Castell DO. Classification of oesophageal motility
PJ. Achalasia: a new clinically relevant classification by high-resolution abnormalities. Gut. 2001;49:145-151.
manometry. Gastroenterology. 2008;135:1526-1533. 36. Fox M, Sweis R, Wong T, Anggiansah A. Sildenafil relieves symptoms
15. Rohof WO, Salvador R, Annese V, et al. Outcomes of treatment and normalizes motility in patients with oesophageal spasm: a report
for achalasia depend on manometric subtype. Gastroenterology. of two cases. Neurogastroenterol Motil. 2007;19:798-803.
2013;144:718-725; quiz e13–e14. 37. Pehlivanov N, Liu J, Mittal RK. Sustained esophageal contraction:
16. Traube M, Hongo M, Magyar L, McCallum RW. Effects of nifedipine a motor correlate of heartburn symptom. Am J Physiol Gastrointest
in achalasia and in patients with high-amplitude peristaltic esophageal Liver Physiol. 2001;281:G743-G751.
contractions. JAMA. 1984;252:1733-1736. 38. Stefanidis D, Hope WW, Kohn GP, et al. Guidelines for surgi-
17. Pasricha PJ, Ravich WJ, Hendrix TR, Sostre S, Jones B, Kalloo AN. cal treatment of gastroesophageal reflux disease. Surg Endosc.
Intrasphincteric botulinum toxin for the treatment of achalasia. 2010;24:2647-2669.
N Engl J Med. 1995;332:774-778. 39. Kahrilas PJ, Shaheen NJ, Vaezi MF, et al. American Gastroenterological
18. Boeckxstaens GE. Achalasia. Best Pract Res Clin Gastroenterol. Association medical position statement on the management of
2007;21:595-608. gastroesophageal reflux disease. Gastroenterology. 2008;135:1383-1391,
19. Schoenberg MB, Marx S, Kersten JF, et al. Laparoscopic Heller 1391.e1–e5.
myotomy versus endoscopic balloon dilatation for the treatment 40. Varin O, Velstra B, De Sutter S, Ceelen W. Total vs partial fundo-
of achalasia: a network meta-analysis. Ann Surg. 2013;258:943- plication in the treatment of gastroesophageal reflux disease: a
952. meta-analysis. Arch Surg. 2009;144:273-278.
20. Inoue H, Minami H, Kobayashi Y, et al. Peroral endoscopic myotomy 41. Tsuboi K, Lee TH, Legner A, Yano F, Dworak T, Mittal SK. Iden-
(POEM) for esophageal achalasia. Endoscopy. 2010;42:265-271. tification of risk factors for postoperative dysphagia after primary
21. Hungness ES, Sternbach JM, Teitelbaum EN, Kahrilas PJ, Pandolfino anti-reflux surgery. Surg Endosc. 2011;25:923-929.
JE, Soper NJ. Per-oral endoscopic myotomy (POEM) after the 42. Xiao Y, Kahrilas PJ, Kwasny MJ, et al. High-resolution manometry
learning curve: durable long-term results with a low complication correlates of ineffective esophageal motility. Am J Gastroenterol.
rate. Ann Surg. 2016;264:508-517. 2012;107:1647-1654.
22. Boeckxstaens GE, Annese V, des Varannes SB, et al. Pneumatic 43. Fibbe C, Layer P, Keller J, Strate U, Emmermann A, Zornig C.
dilation versus laparoscopic Heller’s myotomy for idiopathic achalasia. Esophageal motility in reflux disease before and after fundoplica-
N Engl J Med. 2011;364:1807-1816. tion: a prospective, randomized, clinical, and manometric study.
23. Bhayani NH, Kurian AA, Dunst CM, Sharata AM, Rieder E, Swanstrom Gastroenterology. 2001;121:5-14.
LL. A comparative study on comprehensive, objective outcomes of 44. Chrysos E, Tsiaoussis J, Zoras OJ, et al. Laparoscopic surgery for
laparoscopic Heller myotomy with per-oral endoscopic myotomy gastroesophageal reflux disease patients with impaired esopha-
(POEM) for achalasia. Ann Surg. 2014;259:1098-1103. geal peristalsis: total or partial fundoplication? J Am Coll Surg.
24. Kumbhari V, Tieu AH, Onimaru M, et al. Peroral endoscopic myotomy 2003;197:8-15.
(POEM) vs laparoscopic Heller myotomy (LHM) for the treatment 45. Biertho L, Sebajang H, Anvari M. Effects of laparoscopic Nissen
of Type III achalasia in 75 patients: a multicenter comparative study. fundoplication on esophageal motility: long-term results. Surg Endosc.
Endosc Int Open. 2015;3:E195-E201. 2006;20:619-623.
CHAPTER
pH and Impedance Evaluation of the Esophagus
Geoffrey P. Kohn
9
G
astroesophageal reflux disease (GERD) is a very remain attached to large equipment, making the procedure
common global disorder that poses a significant an inpatient system. In 1974 Johnson and DeMeester per-
public health burden. GERD affects approximately formed landmark studies in controls and GERD patients,
20% of the population of the Western world1–5 and has establishing the technique and normal values for 24-hour
been ranked as the fourth most prevalent gastrointestinal ambulatory pH monitoring, and subsequently ambulatory
disease and the most expensive disease of the alimentary pH monitoring has been widely applied in both clinical
tract.6 and research settings.15
There can be no standard diagnostic criteria for the
definition of GERD because the threshold distinction
between physiologic reflux and reflux disease is arbitrary.7
INDICATIONS FOR pH TESTING
A consensus panel of experts, the so-called Montreal For patients with the typical symptoms of reflux, lifestyle
consensus, has defined GERD as “a condition which changes of weight loss, avoidance of refluxogenic foods,
develops when the reflux of stomach contents causes and avoidance of the full recumbent position, particularly
troublesome symptoms and/or complications.”8 in the postprandial period, relieve symptoms. For others a
GERD patients report “typical” symptoms of heartburn therapeutic trial of acid suppressant medications, such as
and regurgitation. The regurgitation is often acidic in proton pump inhibitors (PPIs) and histamine H2 channel
nature, leading to the sour taste of water brash. “Atypical” antagonists, will suffice, after excluding alternative patholo-
or so-called extraesophageal symptoms include chest pain gies. However, it is prudent to objectively document GER
and respiratory symptoms, including cough, wheeze, and before considering invasive procedures such as antireflux
dysphonia. Extraesophageal manifestations of reflux disease surgery. Objective endoscopic evidence of GERD includes
may result from direct action of the primary refluxate or “mucosal breaks” of esophagitis in the distal esophagus,
may be due to reflex neurologic arcs. Barrett metaplasia, and the reflux-related complication of
It is problematic to define a disease based on symptoms, peptic esophageal stricture. In the absence of endoscopic
especially when symptoms are not specific to the disease. evidence of reflux, objective evidence can be obtained by
Therapy directed at GERD will be ineffective if the etiology 24-hour ambulatory esophageal pH monitoring.16 The
of the symptoms is from a nonreflux condition. Therefore presence of an abnormal 24-hour pH score has been
objective assessment of GERD is often required, with shown to be the strongest predictor of a good outcome
ambulatory pH evaluation the most widely used technique. after antireflux surgery.17 In addition, pH monitoring is
useful in the evaluation of patients with persisting reflux
symptoms despite medical or surgical therapy.10
HISTORY OF ESOPHAGEAL pH MONITORING
The first attempt to objectively detect gastroesophageal
reflux (GER) was accomplished by Reichman in 1884.9,10 ELECTROCHEMICAL PROPERTIES OF
He lowered gelatin-coated sponges into the esophagus of pH ELECTRODES
patients with heartburn and showed that they contained
acid when retrieved. Several decades later Aylwin found Esophageal pH monitoring techniques are designed to
acid and pepsin in esophageal juice retrieved with a tube measure the intraluminal hydrogen ion concentration.10
from a patient with esophagitis.11 Bernstein and Baker The negative logarithm of the hydrogen ion concentration
recognized the relationship between the presence of defines the pH value (pH = +log 1/[H+]). The hydrogen
acid in the esophagus and symptoms of heartburn and activity can be detected by a number of different electrodes
regurgitation.12 They developed the acid infusion test, with different characteristics depending on the electrode
which is based on reproducing esophageal symptoms material. The electrical potential difference, generated
by the instillation of 0.1 N hydrochloric acid in the by a concentration gradient of hydrogen ions between
esophagus. The first in situ measurement of acid reflux two electrodes, can be extrapolated to give a pH value.
in the esophagus was achieved by Tuttle and Grossman in Esophageal pH systems consist of glass, antimony, or
195813 using pH-metry equipment previously described for ion-sensitive field effect transistor (ISFET) pH sensors
studies of gastric acidity. Although this approach greatly and a reference electrode.18 The reference electrode is
improved the detection of acid reflux, it was limited by either external, placed on the patients’ skin, or built into
its failure to differentiate normal subjects from abnormal the catheter. A disadvantage of using external cutaneous
subjects. Prolonged esophageal pH measurements were reference electrodes is the risk of disturbed skin contact
first described by Spencer in 1969.14 This technique soon during the pH recording, which can lead to artifactual pH
became the standard method to quantify GER. Initially, the values. External electrodes are also associated with a risk
recording machines were not portable, forcing patients to of erroneous results as a consequence of influence of the
129
pH and Impedance Evaluation of the Esophagus CHAPTER 9 129.e1
ABSTRACT
Gastroesophageal reflux disease (GERD) is globally a
very common disorder that poses a significant public
health burden. GERD affects approximately 20% of the
population of the Western world and has been ranked as
the fourth most prevalent gastrointestinal disease and the
most expensive disease of the alimentary tract.
GERD is a condition that develops when the reflux of
stomach contents causes troublesome symptoms and/or
complications. However, the symptoms of GERD are not
specific to the disease. Therapy directed at GERD will
be ineffective if the etiology of the symptoms is from a
nonreflux condition. Therefore objective assessment of
GERD is often required, with pH and impedance evaluation
of the esophagus being the most widely used techniques.
KEYWORDS
gastroesophageal reflux disease; GERD; GORD; pH testing;
multichannel intraluminal impedance; lower esophageal
sphincter
130 SECTION I Esophagus and Hernia
mucosal potential difference. Based on these observations, products, wine, and carbonated beverages, although the
pH catheters with an internal reference electrode are intake of these products has a rather short-lived effect on
considered superior19 and are much more common in esophageal pH. Meal periods can be excluded from the
modern clinical practice. Glass electrodes measure the analysis to avoid potential artifacts produced by acidic meal
electrical potential difference across a thin glass membrane. ingestion, and this may improve the clinical reliability of
The monocrystalline antimony electrode is a metal/metal the test.26 The activity and the diet of the patients during
oxide electrode that measures the corrosion potential at the study should ideally be identical to that of the control
the hydrogen ion and antimony interface.20 The ISFET population from which the normative values for esophageal
is a modification of the normal field effect transistor and acid exposure were developed. Some centers encourage
combines in one device the sensing surface and a signal the patients to eat their usual meals and may include one
amplifier.21 Laboratory studies suggest that the more meal likely to precipitate their symptoms, the so-called
expensive glass electrodes are superior to monocrystal- refluxogenic or challenge meal, which often is a burger,
line antimony electrodes because they respond much fries, and shake at a fast-food restaurant. This challenge
quicker to changes in pH and have less drift and a better meal is useful because the process of esophageal pH
linear response.22 However, in a clinical setting, the less monitoring has been shown to reduce reflux-provoking
expensive antimony electrodes provide similar results activities and patients tend to be more sedentary.27 In part
and better patient comfort as compared with the larger this is related to social embarrassment and discomfort
glass electrodes.23,24 There is some evidence that ISFET related to having the catheter coming out of the nose.
electrodes produce the most accurate in vivo measurements When patients change their routines during pH testing,
of acid exposure, but they are not in widespread use.18 the degree of esophageal acid reflux may theoretically
Calibration of the pH electrodes is performed in all be underestimated, which could potentially decrease
pH systems prior to each study, using reference buffer the sensitivity of the pH test. Therefore patients should
solutions, usually either nitrate or phthalate based. The pH be strongly encouraged to return to work and to engage
value of the calibration solution varies among manufactur- in all normal daily activities. The catheter itself does not
ers, but most systems calibrate the pH sensor at room induce reflux.28 The test is not tolerated by all patients,
temperature to an acidic pH (range, 1.0 to 4.0) and a and adverse symptoms may lead to interference with daily
more neutral pH (range, 6.0 to 7.0). When the patient activities and eating and interruption of normal sleep,29
returns after completion of the test, the calibration should possibly leading to underestimation of reflux related to
be repeated to rule out electrode failure and to allow physical activity and meals and hampering the evaluation
correction for slow pH drift. of nocturnal reflux.10
FIGURE 9.3 A gastroesophageal reflux event shown by an abrupt decline in intraesophageal pH.
using the worst day data as compared with either 24-hour study. Placement on the basis of the pH profile recorded
data or averaged 48-hour data.25 on withdrawal of the electrode from the stomach has been
Shorter recording periods have also been investigated found to be inferior to placement based on manometric
with the catheter-based study so as to improve patient LES localization.47
comfort and tolerance. Sensitivity of a 3-hour catheter-based In wireless esophageal pH monitoring the pH sensor is
study has been reported to have 88% of the sensitivity and most commonly placed according to endoscopic landmarks.
98% of the specificity in reflux detection of a 24-hour By convention, the electrode is placed 6 cm above the
catheter-based study.41 However, studies that contain less GEJ (or squamocolumnar junction if positioned normally
than 16 hours of data are generally considered to be at the GEJ). This position has been derived from the
inadequate. findings of concurrent manometry and fluoroscopy studies
that show the upper border of the LES high-pressure
zone typically extends 1 to 1.5 cm above the SCJ.48 Posi-
pH ELECTRODE PLACEMENT tioning the pH electrode 6 cm above the SCJ therefore
Consistent positioning of the pH electrode is vital for approximates the standard 5 cm above the upper border
obtaining reliable esophageal pH data and for comparison of the manometrically defined LES electrode position
with normal values. Studies performed using catheters of the catheter-based technique. Transnasal placement of
with multiple pH sensors for simultaneous pH recording the pH capsule requires prior manometry and provides
at different levels in the esophagus have shown, as would the most accurate placement of the capsule.32 If placed
be expected, greater acid exposure in the distal esopha- transnasally using measurements obtained via endoscopy
gus compared with more proximally.42,43 Consequently, or transoral manometry, a conversion factor of 4 cm is
esophageal acid detection will be significantly reduced typically applied,49 approximating the distance between
when the distance from the LES to the recording level the nares and oropharynx. However, there is significant
increases, and therefore accurate pH probe placement patient variability in this number and this technique is
is essential for a reliable diagnosis of GERD. Typically, not recommended.
the pH electrode should be placed close enough to the
stomach to sample the region of esophageal mucosa most
affected by GER, without displacing into the stomach INTERPRETATION OF ESOPHAGEAL
during the course of the study. During swallowing, the pH STUDIES
GEJ migrates approximately 2 to 4 cm,44,45 and so position-
ing of the pH probe must consider this. By convention, Due to swallowed saliva and esophageal bicarbonate secre-
the catheter-based pH electrode is placed 5 cm above tion, esophageal pH is normally maintained between pH 5
the manometrically defined upper border of the LES.46 and pH 7. Gastric acid secretion generates a highly acidic
This position should avoid inadvertent dislocation of the environment within the stomach, with a pH of 1 to 2 and
electrode into the stomach during breathing, movements rarely more than 3. During esophageal pH monitoring
induced by changes in body position, or swallow-induced (Fig. 9.3), GER events are detected as abrupt declines in
esophageal shortening. Thus to determine this location, intraesophageal pH. Episodes in which pH falls below 4 are
esophageal manometry must be performed prior to the pH counted as reflux events. The choice of the cutoff of pH 4 is
pH and Impedance Evaluation of the Esophagus CHAPTER 9 133
supported by the observation that patients with symptomatic pattern typically remains as upright, postprandial reflux.
reflux usually report heartburn at an intraesophageal pH With increasing severity of GERD the reflux may become
below this threshold.50 When the pH recording is complete, bipositional, with increased acid exposure in both the
the software generates a report that includes a graphical pH upright and supine postures.51–53 Often both the duration
tracing, reflux parameters, and symptom-reflux correlations and number of acid reflux episodes increase, resulting in
(Fig. 9.4). In the analysis of esophageal pH tracings the prolonged esophageal acid exposure times. Isolated supine
pattern of acid reflux can provide important information. reflux is rare and can reflect stasis or pooling of refluxed
Physiologic acid reflux seen in healthy subjects typically gastric juice in the esophagus, leading to very prolonged
occurs in the upright position after a meal and is rapidly reflux episodes.
cleared from the esophagus. In patients with early reflux Several discrete parameters are routinely calculated by
disease the number of reflux events increases, but the the software of the pH system, including the frequency of
134 SECTION I Esophagus and Hernia
TABLE 9.1 Reports on Normal Values for 24-Hour Catheter-Based and 48-Hour Wireless pH Monitoring
and duration of reflux events, the total number of reflux the Symptom Index (SI), the Symptom Sensitivity Index
episodes and those that last longer than 5 minutes, and (SSI), and the Symptom Association Probability (SAP).10
duration of the longest episode. These parameters are First described by Wiener et al.58 the SI is the percentage
presented as totals for the entire study and separately of symptoms preceded by a drop in esophageal pH below
for the upright and supine periods of the recording. The 4.0 within a 5-minute time window divided by the total
total percentage of time the pH is less than 4 is a useful number of symptoms. The SI can be calculated for each
discriminator between physiologic and pathologic reflux symptom attributable to reflux, including heartburn,
but is gender specific.46 An abnormal test is defined by regurgitation, or an atypical symptom, such as chest pain
a value greater than an established threshold, which is or respiratory symptoms. A positive symptom association is
typically the 95th percentile of normal controls. The declared if the SI is greater than or equal to 50% (i.e., at
normal values for ambulatory pH monitoring in adults least half of the reported symptoms are preceded within
reported in the literature vary widely because of differences a 5-minute time window by an intraesophageal pH below
in the selection and the age and gender distribution of 4.0).59 Over the years the definition and use of the SI have
the control population. Table 9.1 shows reports on normal been challenged. Based on a sensitivity analysis in patients
data for catheter-based and capsule-based pH monitoring.10 with chest pain, it has been proposed to use a shorter,
A more precise method of presenting esophageal acid 2-minute time window after the onset of a reflux episode
exposure data is calculation of a composite score, originally in which a symptom has to occur for it to be considered
proposed by Johnson and DeMeester.15 The DeMeester associated with reflux.60 The SI does not take into account
Score includes six parameters: (1) total percent time the total number of reflux events, and there currently is
pH less than 4.0; (2) percent time pH less than 4.0 in insufficient evidence to use SI as a reliable determination
the upright period; (3) percent time pH less than 4.0 in of the presence or absence of reflux disease.
the recumbent period; (4) the total number of reflux The SSI is defined as the percentage of symptoms
episodes; (5) the total number of reflux episodes longer associated with reflux events.58,61 The SI does not consider
than 5 minutes; and (6) the duration of the longest the total number of reflux episodes, and the SSI does
reflux episode. The score is automatically calculated and not include the total number of symptom events. As a
reported in most commercially available pH software. consequence, the probability that the SI becomes positive
Unlike total time pH less than 4, the score is gender increases with an increasingly high number of reflux
independent. The most referenced value for an abnormal episodes and SSI is more likely to be positive when the
DeMeester composite score is a value larger than 14.7.15 number of symptom episodes is high. An SSI greater
The DeMeester composite score was also calculated based than 10% is usually considered positive and suggests an
on normals for use with the Bravo pH system and are association between symptom and reflux.
slightly different than the catheter normals (day 1 = 14, The SAP is a statistical method to determine the rela-
day 2 = 14, combined = 16). Regardless of whether the tionship between symptoms and reflux episodes. The
composite score or individual acid exposure time is used, SAP is calculated by dividing the entire study’s pH data
a detailed evaluation of the pH tracing is of fundamental into consecutive 2-minute segments. For each of these
importance to recognize and exclude artifacts and to segments, it is determined whether reflux occurred in the
assess symptom association. segment, allowing for calculation of the total number of
2-minute segments with and without reflux. Subsequently,
SYMPTOMS ASSOCIATION it is determined whether or not a reflux episode occurred
Reflux symptoms such as heartburn and regurgitation are in the 2-minute period before each symptom. A 2 × 2
very common, but because these symptoms are not very contingency table is constructed in which the number
specific for GERD, it can be useful to determine if there of 2-minute segments with and without symptoms and
is a temporal relationship between symptoms and reflux with and without reflux are tabulated. Using the Fisher
events. The relationship between symptoms and reflux exact test, a P value is calculated, and the SAP index is
episodes can be expressed numerically using symptom calculated as (1 − P) × 100%.62 The cutoff value for a
association analysis.57 The most frequently used indices are positive test is often defined as SAP greater than or equal
pH and Impedance Evaluation of the Esophagus CHAPTER 9 135
to 95%. However, even a statistically significant relationship there is a paucity of data showing a correlation between
between reflux events and symptoms does not necessarily intragastric pH and GER.65,66 Alternatively, combining
imply causality. The yield of the SI and SAP is greater when intraluminal impedance with pH testing can provide
performed off rather than on acid-suppressant therapy.25 additional information on the nature of refluxed material
while on medications, but these studies are limited by the
lack of clearly defined normal values for the impedance
pH TESTING ON VERSUS OFF ACID portion of the test, unreliable automated reading of the
SUPPRESSIVE MEDICATION study, and poor interobserver agreement particularly for
proximal reflux events.
For patients complaining of reflux symptoms, current
clinical practice guidelines recommend empirical trials
of PPIs instead of pH testing.63 The favorable side effect LIMITATIONS OF ESOPHAGEAL
profile of PPIs has encouraged this initial step, although pH MONITORING
the sensitivity and specificity of this approach are only
78% and 54%, respectively.64 If symptoms persist, then Ambulatory esophageal pH monitoring is not without
esophageal pH testing is performed. Studies suggest that some limitations. The sensitivity and specificity of catheter-
a more cost-effective approach is to study all patients based pH monitoring have traditionally been reported
with pH testing as an initial approach because many to be in the range of 87% to 96% and 97% to 100%,
patients are put on PPI medication for nonreflux symptoms respectively.55,56 As there is a relationship between the
and stay on the medications for years inappropriately. If severity of the disease and the discriminatory power of the
patients are started on PPIs for reflux symptoms without test,67 the published data on sensitivity and specificity reflect
confirmation of the disease and present with persistent the severity of reflux disease in the populations tested,
symptoms, objective testing is necessary. At this point an and patients with mild reflux may be missed.10 In studies
important decision has to be made whether to perform of patients with typical reflux symptoms and esophagitis,
testing on or off medications. Esophageal pH testing a sensitivity of 76% to 78% and a specificity of 93% to
without acid-suppressant medication is more accurate, 95% were reported for the capsule-based technique for
and there are established normal values. A test showing esophageal pH monitoring.30,67 In contrast, other studies
normal distal esophageal acid exposure is very helpful in suggest increased sensitivity with capsule pH monitoring
suggesting that the symptoms are not caused by reflux. based on the longer duration of monitoring, up to 96
A positive pH test while off acid-suppressant therapy hours.
confirms the diagnosis of reflux disease but may not explain In patients with typical symptoms and endoscopy-
why the patient is still having symptoms while taking PPIs. If proven Los Angeles Classification of Gastroesophageal
the symptoms are regurgitation, it can be reliably assumed Reflux Disease (LA) grade C or D esophagitis or Barrett
that these symptoms are reflux related because it is well esophagus, the diagnosis of GERD is confirmed and
established that reflux continues on PPI therapy; only the additional investigation with esophageal pH monitoring is
nature of the refluxed material is changed from acid to unnecessary.16 However, with lesser degrees of esophagitis
weak or nonacid. These persistent regurgitation symptoms and in patients without endoscopic evidence of GERD, who
are well addressed with antireflux procedures, including constitute up to two-thirds of all patients with typical reflux
transoral incisionless fundoplication, the LINX device, or symptoms, objective GERD testing is recommended.68 A
a fundoplication. In contrast, if the symptoms on PPI are negative pH test in most cases excluded the presence of
persistent heartburn, testing on therapy may allow assess- significant reflux disease. However, because normal values
ment of the efficacy of PPI therapy, with continued acid are based on the 95th percentile, there by definition
reflux, suggesting that additional medication or a change are 5% false-negative studies. In addition, patients with
in medication may be effective to improve symptomatic typical reflux symptoms and no esophagitis (so-called
control of the GERD. In these patients the use of dual pH NERD or nonerosive reflux disease patients) are likely a
electrodes to monitor both distal esophageal and gastric heterogeneous group with different etiologies for their
pH are sometimes recommended (Fig. 9.5).25 Although symptoms, including esophageal motor events, reflux
intragastric pH measurement can help to determine the of nonacid contents, acid hypersensitivity, functional
efficacy of acid-suppressive medications or suggest poor heartburn, or emotional or psychologic abnormalities that
patient compliance, its clinical relevance is unclear because cannot be reliably detected using pH testing alone.69,70
FIGURE 9.5 Double-channel pH study with sensors in the stomach (“pH2”) and 5 cm above the upper border of the lower esophageal
sphincter (“pH1”), showing intragastric acidification.
136 SECTION I Esophagus and Hernia
Perhaps one of the most important roles of pH testing swallowing, without the use of radiation and for monitoring
is in the diagnosis of a patient considering antireflux GER independent of its pH.
surgery. Particularly with typical heartburn and regur- The presence and movement of an intraesophageal
gitation symptoms of reflux, a good patient outcome bolus is detected on MII based on measuring differences in
from antireflux surgery can be predicted by an abnormal electrical conductivity affected by the presence of various
preoperative pH study, and unsatisfactory outcomes are materials within the esophagus.81 An MII catheter requires
much more likely if surgery is performed after a normal an alternating current source connected to a series of metal
pH study. Atypical esophageal symptoms, such as cough, rings located along segments of the esophageal lumen. An
globus sensation, dysphonia, throat pain, and respiratory isolator separates the rings so that the electrical circuit is
symptoms, are often multifactorial, but if due to GERD, closed by the electrical charges (i.e., ions) surrounding
they may arise from proximal extension of the refluxate. the catheter. The impedance within a given segment is
The diagnosis of reflux in these patients can be challenging, then determined by measuring the electrical resistance
and abnormal proximal acid exposure can occur in some as a substance passes through the current established
patients despite normal distal esophageal acid exposure by the rings. Once placed in the esophagus, the ions of
related to proximal jetting of refluxed gastric juice and the esophageal mucosa close the circuit and the system
the lower thresholds for abnormal acid exposure in the measures a relatively stable resistance of approximately
proximal esophagus. Dual probe monitoring with a distal 2000 to 3000 ohms.
and a proximal probe can be useful in these patients. In Liquid boluses conduct better than the empty esopha-
addition, new technologies for proximal esophageal pH gus, leading to a rapid decline in intraluminal impedance
assessment using a pharyngeal pH probe (Restech pH, when the bolus enters the impedance measuring segment.81
California) and multichannel intraluminal impedance Impedance returns to baseline once the bolus has exited
(MII) studies may be of use. the segment. Multiple impedance measuring segments
mounted on the same catheter allow determination of
the direction of bolus movement based on the timing
PROXIMAL ESOPHAGEAL pH ASSESSMENT of changes in impedance at individual levels. Proximal
The association between reflux of acidic gastric contents to distal (antegrade) progression of impedance changes
into the larynx and laryngeal symptoms was initially pro- is indicative of swallowing (Fig. 9.6), whereas a distal to
posed in 1968.71 There are multiple potential causes for proximal (retrograde) progression indicates a reflux
these respiratory and laryngeal symptoms, and establishing episode (Fig. 9.7).82
reflux as part of or the major cause based on symptoms When combined with pH, MII can evaluate the pres-
alone is unreliable.72,73 Distal esophageal acid measure- ence of a reflux event and its pH, thereby allowing the
ment can confirm increased esophageal acid exposure in detection and differentiation of both acid and nonacid
patients with extraesophageal symptoms, but that does not reflux (see Fig. 9.6).
confirm that the symptoms are caused by reflux.74 Proximal
esophageal pH abnormalities have been identified in
patients with normal distal esophageal acid exposure,75 and COMBINED MULTICHANNEL INTRALUMINAL
standardization of placement of the proximal esophageal IMPEDANCE AND pH
pH probe has been proposed using the upper esophageal
sphincter as the anatomic landmark.76 Normative values Esophageal pH monitoring quantifies the amount of distal
(95th percentile) for upper esophageal acid exposure have esophageal acid exposure but requires the refluxed gastric
been defined (24 episodes and 0.9 for % time pH < 4).73 juice to have a pH of less than 4 to be considered a reflux
These values are lower than those in the distal esophagus. event. Reflux with a pH greater than 4 in a patient off
Probes for the measurement of pH require liquid to be medications may be related to gastric achlorhydria or to
presented to the sensor for accurate measurement. For excessive bile in the gastric juice. Different approaches
example, drying of the pH sensor when the pH sensor is (e.g., bilirubin monitoring and scintigraphy) have been
placed in the hypopharynx or proximal esophagus can proposed to evaluate these patients. Adding impedance
cause artifacts in the pH recording. Certain sensors, such assessment to pH measurements is a new method to
as the Dx-pH probe (Restech, California) are specifically evaluate the function of the antireflux barrier.83
designed not to dry out in the pharynx and also allow for Because impedance can detect the presence of refluxate
measurement of aerosolized pH. This probe is designed in the esophagus independent of its composition and can
to help to diagnose reflux as the cause of respiratory be mounted on a regular pH catheter, MII has several
and laryngeal symptoms. Pharyngeal pH monitoring is advantages in monitoring gastroesophageal reflux. MII-pH
probably superior to proximal esophageal pH measure- is the best test to detect reflux of all types, whether acid or
ments using a dual probe when predicting resolution of nonacid.83 However, the implications of weak or nonacid
extraesophageal atypical GERD symptoms after antireflux reflux events and normal thresholds for these events have
surgery,77 although the clinical utility of this system remains not been clearly defined.
under investigation.78,79 For monitoring of gastroesophageal reflux via MII-pH,
multiple impedance-measuring segments are added to a
regular pH probe. Gastroesophageal reflux episodes are
MULTICHANNEL INTRALUMINAL IMPEDANCE detected by retrograde declines in intraluminal impedance
First described by Silny80 in 1991 MII is a relatively new produced by increased conductivity of the liquid reflux,
technique for evaluating esophageal bolus transit during and data from the esophageal pH sensor are used to
pH and Impedance Evaluation of the Esophagus CHAPTER 9 137
FIGURE 9.7 An impedance trace of retrograde bolus movement—an acidic reflux event.
categorize the reflux as acid or nonacid. The refluxate (Table 9.2), as have various population-specific norms.88–90
is considered acidic if the pH drops from greater than However, of all reflux parameters the number of reflux
to less than 4, weakly acidic if the pH is between 4 and events has the most variability in normals and appears to
7, and nonacidic if the intraesophageal pH during an be affected by body mass index (BMI) and other factors.
MII-detected reflux episode remains greater than 7.83 This problem limits the utility of MII for the reliable
In addition to the chemical properties of the gastro- diagnosis of GERD, unless the pH component confirms
esophageal refluxate, MII has the ability to determine its increased esophageal acid exposure.
physical state; MII can differentiate between liquid only, Esophageal mucosal healing rates of up to 90% have
gas only, and mixed gas-liquid reflux episodes based on been documented in patients taking potent acid–suppressive
changes in intraluminal impedance. Gas or air has very therapy,91 and therefore nonacid reflux is presumed less
poor electrical conductivity and, when present between likely to cause esophageal lesions. Nonacid reflux does
impedance-measuring rings, will produce a rise in imped- appear to have a role in causing persistent symptoms
ance to greater than 7000 ohms; in contrast, liquid, which in patients taking acid-suppressive therapy, particularly
has better electrical conductivity, will produce a decline regurgitation symptoms. There are both direct evidence
in impedance.81 of postprandial symptoms being associated with nonacid
Nonacid reflux (i.e., a reflux episode with a pH greater reflux86 and indirect data from a large PPI trial indicating
than 4) is relatively infrequent in subjects not taking acid- that 35% to 40% of patients receiving acid-suppressive
suppressive therapy; it occurs primarily in the postprandial therapy continue to have symptoms.91
periods84 and rarely at night.85 However, in subjects taking Quantifying the type and proximal extent of reflux may
acid-suppressive therapy, the medications typically change be of interest in patients with extraesophageal atypical
the composition of the gastroesophageal refluxate without symptoms. Furthermore, gas-containing reflux episodes
affecting the total number of reflux episodes.86,87 Normal may participate in proximal extension of reflux and extra-
values for the number of acid and nonacid reflux events esophageal symptoms.92,93 Combined MII-pH monitoring
on and off acid-suppressive therapy85 have been published of patients with refractory symptoms has been shown to
138 SECTION I Esophagus and Hernia
be able to clarify the symptom association with reflux 2. Locke GR 3rd, Talley NJ, Fett SL, Zinsmeister AR, Melton LJ 3rd.
events and has revealed that about 50% of patients with Prevalence and clinical spectrum of gastroesophageal reflux: a
population-based study in Olmsted County, Minnesota. Gastroenterol-
persistent symptoms on therapy do not have a temporal ogy. 1997;112(5):1448-1456.
correlation between their symptoms and any type of 3. Mohammed I, Cherkas LF, Riley SA, Spector TD, Trudgill NJ.
reflux, whereas 40% have temporal association between Genetic influences in gastro-oesophageal reflux disease: a twin
their symptoms and nonacid reflux.94–96 A poor symptom study. Gut. 2003;52(8):1085-1089.
correlation may suggest that reflux is not the cause for 4. Isolauri J, Laippala P. Prevalence of symptoms suggestive of gastro-
oesophageal reflux disease in an adult population. Ann Med.
a patient’s symptoms. There may be a subset of patients 1995;27(1):67-70.
with normal pH studies on medications but abnormal 5. Knox SA, Harrison CM, Britt HC, Henderson JV. Estimating
MII tests that will have good short-term outcomes after prevalence of common chronic morbidities in Australia. Med J
fundoplicaiton,97 but these patients have to be approached Aust. 2008;189(2):66-70.
6. Sandler RS, Everhart JE, Donowitz M, et al. The burden of
cautiously because the outcomes with surgery are less selected digestive diseases in the United States. Gastroenterology.
predictable. It is advised that these patients have pH 2002;122(5):1500-1511.
testing off medications to confirm the diagnosis of GERD 7. Kahrilas PJ, Shaheen NJ, Vaezi MF, et al. American Gastroen-
before offering an antireflux procedure given the absence terological Association Medical Position Statement on the
of clearly defined normal values for impendence-detected management of gastroesophageal reflux disease. Gastroenterology.
2008;135(4):1383-1391, 1391.e1–e5.
number of reflux events. Nonetheless, MII-pH monitoring 8. Vakil N, van Zanten SV, Kahrilas P, Dent J, Jones R, Global
is being increasingly used in the assessment of patients Consensus Group. The Montreal definition and classification of
with atypical symptoms of GERD.98–100 With studies showing gastroesophageal reflux disease: a global evidence-based consensus.
a postfundoplication improvement in extraesophageal Am J Gastroenterol. 2006;101(8):1900-1920; quiz 1943.
9. Babey A. Observations on the nature of heartburn. Am J Dig Dis.
symptoms after diagnosis based on pH-only criteria of 15% 1937;4:600.
to 95%,101 comparison with outcomes of reflux diagnosed 10. Öberg S. Esophageal pH monitoring. In: Yeo CJ, ed. Shackelford’s
by impedance or pH-impedance will be challenging due Surgery of the Alimentary Tract. 7th ed. Philadelphia: Saunders;
to the wide range of the results, and these studies are 2013:147-153.
currently lacking. 11. Aylwin JA. The physiological basis of reflux oesophagitis in sliding
hiatal diaphragmatic hernia. Thorax. 1953;8(1):38-45.
12. Bernstein LM, Baker LA. A clinical test for esophagitis. Gastroenterol-
CONCLUSION ogy. 1958;34(5):760-781.
13. Tuttle SG, Grossman MI. Detection of gastro-esophageal reflux by
Ambulatory pH monitoring has become widely available simultaneous measurement of intraluminal pressure and pH. Proc
Soc Exp Biol Med. 1958;98(2):225-227.
and well-established normal thresholds allow confirmation 14. Spencer J. Prolonged pH recording in the study of gastro-oesophageal
of the presence of increased esophageal acid exposure reflux. Br J Surg. 1969;56(12):912-914.
in patients with GERD symptoms to allow selection of 15. Johnson LF, DeMeester TR. Twenty-four-hour pH monitoring of
an appropriate therapy. New techniques to add further the distal esophagus. A quantitative measure of gastroesophageal
information on pharyngeal reflux and to assess weak and reflux. Am J Gastroenterol. 1974;62(4):325-332.
16. Stefanidis D, Hope WW, Kohn GP, et al. Guidelines for surgi-
nonacid reflux events can be useful in some patients. A cal treatment of gastroesophageal reflux disease. Surg Endosc.
complete understanding of the strengths and weaknesses 2010;24(11):2647-2669.
of all these studies is beneficial to ensure cost-effective 17. Campos GM, Peters JH, DeMeester TR, et al. Multivariate analysis of
and appropriate evaluation and therapy for patients with factors predicting outcome after laparoscopic Nissen fundoplication.
J Gastrointest Surg. 1999;3(3):292-300.
both typical and atypical reflux symptoms. 18. Hemmink GJ, Weusten BL, Oors J, Bredenoord AJ, Timmer R,
Smout AJ. Ambulatory oesophageal pH monitoring: a comparison
REFERENCES between antimony, ISFET, and glass pH electrodes. Eur J Gastroenterol
Hepatol. 2010;22(5):572-577.
1. Fedorak RN, Veldhuyzen van Zanten S, Bridges R. Canadian Digestive 19. Opekun AR, Graham DY. Antimony electrodes for in vivo pH
Health Foundation Public Impact Series: gastroesophageal reflux monitoring. Dig Dis Sci. 1991;36(8):1180-1181.
disease in Canada: incidence, prevalence, and direct and indirect 20. Emde C. Electrochemical aspects of pH electrodes. Dig Dis.
economic impact. Can J Gastroenterol. 2010;24(7):431-434. 1990;8(suppl 1):18-22.
pH and Impedance Evaluation of the Esophagus CHAPTER 9 139
21. Duroux P, Emde C, Bauerfeind P, et al. The ion sensitive field relaxations and swallows in healthy subjects. Neurogastroenterol
effect transistor (ISFET) pH electrode: a new sensor for long term Motil. 2012;24(11):990-e539.
ambulatory pH monitoring. Gut. 1991;32(3):240-245. 45. Edmundowicz SA, Clouse RE. Shortening of the esophagus in
22. McLauchlan G, Rawlings JM, Lucas ML, McCloy RF, Crean GP, response to swallowing. Am J Physiol. 1991;260(3 Pt 1):G512-G516.
McColl KE. Electrodes for 24 hours pH monitoring—a comparative 46. Kahrilas PJ, Quigley EM. Clinical esophageal pH recording: a
study. Gut. 1987;28(8):935-939. technical review for practice guideline development. Gastroenterology.
23. Ward BW, Wu WC, Richter JE, Lui KW, Castell DO. Ambulatory 1996;110(6):1982-1996.
24-hour esophageal pH monitoring. Technology searching for a 47. Mattox HE 3rd, Richter JE, Sinclair JW, Price JE, Case LD. Gas-
clinical application. J Clin Gastroenterol. 1986;8(suppl 1):59-67. troesophageal pH step-up inaccurately locates proximal border of
24. Pandolfino JE, Ghosh S, Zhang Q, Heath M, Bombeck T, Kahrilas lower esophageal sphincter. Dig Dis Sci. 1992;37(8):1185-1191.
PJ. Slimline vs. glass pH electrodes: what degree of accuracy should 48. Kahrilas PJ, Lin S, Chen J, Manka M. The effect of hiatus hernia
we expect? Aliment Pharmacol Ther. 2006;23(2):331-340. on gastro-oesophageal junction pressure. Gut. 1999;44(4):476-482.
25. Hirano I, Richter JE; Practice Parameters Committee of the American 49. Lacy BE, O’Shana T, Hynes M, et al. Safety and tolerability of
College of Gastroenterology. ACG practice guidelines: esophageal transoral Bravo capsule placement after transnasal manometry using
reflux testing. Am J Gastroenterol. 2007;102(3):668-685. a validated conversion factor. Am J Gastroenterol. 2007;102(1):24-32.
26. Ter RB, Johnston BT, Castell DO. Exclusion of the meal period 50. Tuttle SG, Rufin F, Bettarello A. The physiology of heartburn. Ann
improves the clinical reliability of esophageal pH monitoring. Intern Med. 1961;55:292-300.
J Clin Gastroenterol. 1997;25(1):314-316. 51. Ouatu-Lascar R, Lin OS, Fitzgerald RC, Triadafilopoulos G. Upright
27. Fass R, Hell R, Sampliner RE, et al. Effect of ambulatory 24-hour versus supine reflux in gastroesophageal reflux disease. J Gastroenterol
esophageal pH monitoring on reflux-provoking activities. Dig Dis Hepatol. 2001;16(11):1184-1190.
Sci. 1999;44(11):2263-2269. 52. Campos GM, Peters JH, DeMeester TR, Oberg S, Crookes PF, Mason
28. Decktor DL, Krawet SH, Rodriguez SL, Robinson M, Castell DO. RJ. The pattern of esophageal acid exposure in gastroesophageal
Dual site ambulatory pH monitoring: a probe across the lower reflux disease influences the severity of the disease. Arch Surg.
esophageal sphincter does not induce gastroesophageal reflux. 1999;134(8):882-887; discussion 887–888.
Am J Gastroenterol. 1996;91(6):1162-1166. 53. Saraswat VA, Dhiman RK, Mishra A, Naik SR. Correlation of 24-hr
29. Wong WM, Bautista J, Dekel R, et al. Feasibility and tolerability esophageal pH patterns with clinical features and endoscopy in
of transnasal/per-oral placement of the wireless pH capsule vs. gastroesophageal reflux disease. Dig Dis Sci. 1994;39(1):199-205.
traditional 24-h oesophageal pH monitoring—a randomized trial. 54. Richter JE, Bradley LA, DeMeester TR, Wu WC. Normal 24-hr
Aliment Pharmacol Ther. 2005;21(2):155-163. ambulatory esophageal pH values. Influence of study center, pH
30. Pandolfino JE, Richter JE, Ours T, Guardino JM, Chapman J, electrode, age, and gender. Dig Dis Sci. 1992;37(6):849-856.
Kahrilas PJ. Ambulatory esophageal pH monitoring using a wireless 55. Jamieson JR, Stein HJ, DeMeester TR, et al. Ambulatory 24-h
system. Am J Gastroenterol. 2003;98(4):740-749. esophageal pH monitoring: normal values, optimal thresholds,
31. Wenner J, Johnsson F, Johansson J, Oberg S. Wireless oesophageal specificity, sensitivity, and reproducibility. Am J Gastroenterol.
pH monitoring: feasibility, safety and normal values in healthy 1992;87(9):1102-1111.
subjects. Scand J Gastroenterol. 2005;40(7):768-774. 56. Johnsson F, Joelsson B, Isberg PE. Ambulatory 24 hour intraesopha-
32. Ayazi S, Lipham JC, Portale G, et al. Bravo catheter-free pH monitor- geal pH-monitoring in the diagnosis of gastroesophageal reflux
ing: normal values, concordance, optimal diagnostic thresholds, disease. Gut. 1987;28(9):1145-1150.
and accuracy. Clin Gastroenterol Hepatol. 2009;7(1):60-67. 57. Bredenoord AJ, Weusten BL, Smout AJ. Symptom association
33. Emde C, Garner A, Blum AL. Technical aspects of intraluminal analysis in ambulatory gastro-oesophageal reflux monitoring. Gut.
pH-metry in man: current status and recommendations. Gut. 2005;54(12):1810-1817.
1987;28(9):1177-1188. 58. Wiener GJ, Richter JE, Copper JB, Wu WC, Castell DO. The symptom
34. Wenner J, Johnsson F, Johansson J, Oberg S. Wireless esophageal pH index: a clinically important parameter of ambulatory 24-hour
monitoring is better tolerated than the catheter-based technique: esophageal pH monitoring. Am J Gastroenterol. 1988;83(4):358-361.
results from a randomized cross-over trial. Am J Gastroenterol. 59. Tutuian R, Castell DO. Gastroesophageal reflux monitoring: pH
2007;102(2):239-245. and impedance. GI Motility online. Published May 16, 2006.
35. Prakash C, Clouse RE. Value of extended recording time with 60. Lam HG, Breumelhof R, Roelofs JM, Van Berge Henegouwen
wireless pH monitoring in evaluating gastroesophageal reflux GP, Smout AJ. What is the optimal time window in symptom
disease. Clin Gastroenterol Hepatol. 2005;3(4):329-334. analysis of 24-hour esophageal pressure and pH data? Dig Dis Sci.
36. Ahlawat SK, Novak DJ, Williams DC, Maher KA, Barton F, Benjamin 1994;39(2):402-409.
SB. Day-to-day variability in acid reflux patterns using the BRAVO 61. Breumelhof R, Smout AJ. The symptom sensitivity index: a valuable
pH monitoring system. J Clin Gastroenterol. 2006;40(1):20-24. additional parameter in 24-hour esophageal pH recording. Am J
37. Bhat YM, McGrath KM, Bielefeldt K. Wireless esophageal pH Gastroenterol. 1991;86(2):160-164.
monitoring: new technique means new questions. J Clin Gastroenterol. 62. Weusten BL, Roelofs JM, Akkermans LM, Van Berge-Henegouwen
2006;40(2):116-121. GP, Smout AJ. The symptom-association probability: an improved
38. Hirano I, Zhang Q, Pandolfino JE, Kahrilas PJ. Four-day Bravo method for symptom analysis of 24-hour esophageal pH data.
pH capsule monitoring with and without proton pump inhibitor Gastroenterology. 1994;107(6):1741-1745.
therapy. Clin Gastroenterol Hepatol. 2005;3(11):1083-1088. 63. Katz PO, Gerson LB, Vela MF. Guidelines for the diagnosis and
39. Lee YC, Wang HP, Chiu HM, et al. Patients with functional heartburn management of gastroesophageal reflux disease. Am J Gastroenterol.
are more likely to report retrosternal discomfort during wireless 2013;108(3):308-328; quiz 329.
pH monitoring. Gastrointest Endosc. 2005;62(6):834-841. 64. Numans ME, Lau J, de Wit NJ, Bonis PA. Short-term treatment
40. Remes-Troche JM, Ibarra-Palomino J, Carmona-Sanchez RI, with proton-pump inhibitors as a test for gastroesophageal reflux
Valdovinos MA. Performance, tolerability, and symptoms related disease: a meta-analysis of diagnostic test characteristics. Ann Intern
to prolonged pH monitoring using the Bravo system in Mexico. Med. 2004;140(7):518-527.
Am J Gastroenterol. 2005;100(11):2382-2386. 65. Fackler WK, Ours TM, Vaezi MF, Richter JE. Long-term effect of
41. Arora AS, Murray JA. Streamlining 24-hour pH study for GERD: H2RA therapy on nocturnal gastric acid breakthrough. Gastroenterol-
use of a 3-hour postprandial test. Dig Dis Sci. 2003;48(1):10-15. ogy. 2002;122(3):625-632.
42. Anggiansah A, Sumboonnanonda K, Wang J, Linsell J, Hale P, 66. Bell NJ, Burget D, Howden CW, Wilkinson J, Hunt RH. Appropriate
Owen WJ. Significantly reduced acid detection at 10 centimeters acid suppression for the management of gastro-oesophageal reflux
compared to 5 centimeters above lower esophageal sphincter in disease. Digestion. 1992;51(suppl 1):59-67.
patients with acid reflux. Am J Gastroenterol. 1993;88(6):842-846. 67. Wenner J, Johansson J, Johnsson F, Oberg S. Optimal thresholds
43. Wenner J, Johnsson F, Johansson J, Oberg S. Acid reflux immediately and discriminatory power of 48-h wireless esophageal pH monitoring
above the squamocolumnar junction and in the distal esophagus: in the diagnosisof GERD. Am J Gastroenterol. 2007;102(9):1862-
simultaneous pH monitoring using the wireless capsule pH system. 1869.
Am J Gastroenterol. 2006;101(8):1734-1741. 68. Venables TL, Newland RD, Patel AC, Hole J, Wilcock C, Turbitt ML.
44. Lee YY, Whiting JG, Robertson EV, et al. Kinetics of transient Omeprazole 10 milligrams once daily, omeprazole 20 milligrams
hiatus hernia during transient lower esophageal sphincter once daily, or ranitidine 150 milligrams twice daily, evaluated as
140 SECTION I Esophagus and Hernia
initial therapy for the relief of symptoms of gastro-oesophageal reflux 86. Vela MF, Camacho-Lobato L, Srinivasan R, Tutuian R, Katz PO,
disease in general practice. Scand J Gastroenterol. 1997;32(10):965- Castell DO. Simultaneous intraesophageal impedance and pH
973. measurement of acid and nonacid gastroesophageal reflux: effect
69. Bredenoord AJ, Weusten BL, Curvers WL, Timmer R, Smout AJ. of omeprazole. Gastroenterology. 2001;120(7):1599-1606.
Determinants of perception of heartburn and regurgitation. Gut. 87. Tamhankar AP, Peters JH, Portale G, et al. Omeprazole does
2006;55(3):313-318. not reduce gastroesophageal reflux: new insights using multi-
70. Tack J, Fass R. Review article: approaches to endoscopic-negative channel intraluminal impedance technology. J Gastrointest Surg.
reflux disease: part of the GERD spectrum or a unique acid-related 2004;8(7):890-897; discussion 897–898.
disorder? Aliment Pharmacol Ther. 2004;19(suppl 1):28-34. 88. Ndebia EJ, Sammon AM, Umapathy E, Iputo JE. Normal values
71. Cherry J, Margulies SI. Contact ulcer of the larynx. Laryngoscope. of 24-hour ambulatory esophageal impedance-pH monitoring in a
1968;78(11):1937-1940. rural South African cohort of healthy participants. Dis Esophagus.
72. Mahieu HF. Review article: the laryngological manifestations of reflux 2016;29(4):385-391.
disease: why the scepticism? Aliment Pharmacol Ther. 2007;26(suppl 89. Xiao YL, Lin JK, Cheung TK, et al. Normal values of 24-hour
2):17-24. combined esophageal multichannel intraluminal impedance and pH
73. Ayazi S, Hagen JA, Zehetner J, et al. Proximal esophageal pH monitoring in the Chinese population. Digestion. 2009;79(2):109-114.
monitoring: improved definition of normal values and determination 90. Zerbib F, des Varannes SB, Roman S, et al. Normal values and
of a composite pH score. J Am Coll Surg. 2010;210(3):345-350. day-to-day variability of 24-h ambulatory oesophageal impedance-pH
74. Vaezi MF, Hicks DM, Abelson TI, Richter JE. Laryngeal signs and monitoring in a Belgian-French cohort of healthy subjects. Aliment
symptoms and gastroesophageal reflux disease (GERD): a critical Pharmacol Ther. 2005;22(10):1011-1021.
assessment of cause and effect association. Clin Gastroenterol Hepatol. 91. Castell DO, Kahrilas PJ, Richter JE, et al. Esomeprazole (40 mg)
2003;1(5):333-344. compared with lansoprazole (30 mg) in the treatment of erosive
75. Schnatz PF, Castell JA, Castell DO. Pulmonary symptoms esophagitis. Am J Gastroenterol. 2002;97(3):575-583.
associated with gastroesophageal reflux: use of ambulatory pH 92. Kawamura O, Aslam M, Rittmann T, Hofmann C, Shaker R. Physical
monitoring to diagnose and to direct therapy. Am J Gastroenterol. and pH properties of gastroesophagopharyngeal refluxate: a 24-hour
1996;91(9):1715-1718. simultaneous ambulatory impedance and pH monitoring study.
76. McCollough M, Jabbar A, Cacchione R, Allen JW, Harrell S, Wo Am J Gastroenterol. 2004;99(6):1000-1010.
JM. Proximal sensor data from routine dual-sensor esophageal 93. Ribolsi M, Savarino E, De Bortoli N, et al. Reflux pattern and role
pH monitoring is often inaccurate. Dig Dis Sci. 2004;49(10):1607- of impedance-pH variables in predicting PPI response in patients
1611. with suspected GERD-related chronic cough. Aliment Pharmacol
77. Worrell SG, DeMeester SR, Greene CL, Oh DS, Hagen JA. Pha- Ther. 2014;40(8):966-973.
ryngeal pH monitoring better predicts a successful outcome for 94. Mainie I, Tutuian R, Shay S, et al. Acid and non-acid reflux in
extraesophageal reflux symptoms after antireflux surgery. Surg patients with persistent symptoms despite acid suppressive therapy:
Endosc. 2013;27(11):4113-4118. a multicentre study using combined ambulatory impedance-pH
78. Ummarino D, Vandermeulen L, Roosens B, Urbain D, Hauser monitoring. Gut. 2006;55(10):1398-1402.
B, Vandenplas Y. Gastroesophageal reflux evaluation in patients 95. Triadafilopoulos G. A closure without a closure: impedance pH
affected by chronic cough: Restech versus multichannel intraluminal monitoring expanding the indications for antireflux surgery.
impedance/pH metry. Laryngoscope. 2013;123(4):980-984. Gastroenterology. 2010;138(1):390-392; discussion 392.
79. Patel DA, Harb AH, Vaezi MF. Oropharyngeal reflux monitoring 96. Zerbib F, Roman S, Ropert A, et al. Esophageal pH-impedance
and atypical gastroesophageal reflux disease. Curr Gastroenterol Rep. monitoring and symptom analysis in GERD: a study in patients
2016;18(3):12. off and on therapy. Am J Gastroenterol. 2006;101(9):1956-1963.
80. Silny J. Intraluminal multiple electric impedance procedure for 97. del Genio G, Tolone S, del Genio F, et al. Prospective assessment
measurement of gastrointestinal motility. Neurogastroenterol Motil. of patient selection for antireflux surgery by combined multi-
1991;3(3):151-162. channel intraluminal impedance pH monitoring. J Gastrointest
81. Tutuian R, Castell DO. Multichannel intraluminal impedance. Surg. 2008;12(9):1491-1496.
In: Yeo CJ, ed. Shackelford’s Surgery of the Alimentary Tract. 7th ed. 98. Malhotra A, Freston JW, Aziz K. Use of pH-impedance testing
Philadelphia: Saunders; 2013:154-161. to evaluate patients with suspected extraesophageal manifesta-
82. Mainie I, Tutuian R, Agrawal A, Adams D, Castell DO. Combined tions of gastroesophageal reflux disease. J Clin Gastroenterol.
multichannel intraluminal impedance-pH monitoring to select 2008;42(3):271-278.
patients with persistent gastro-oesophageal reflux for laparoscopic 99. Bajbouj M, Becker V, Neuber M, Schmid RM, Meining A. Combined
Nissen fundoplication. Br J Surg. 2006;93(12):1483-1487. pH-metry/impedance monitoring increases the diagnostic yield in
83. Sifrim D, Castell D, Dent J, Kahrilas PJ. Gastro-oesophageal reflux patients with atypical gastroesophageal reflux symptoms. Digestion.
monitoring: review and consensus report on detection and defini- 2007;76(3-4):223-228.
tions of acid, non-acid, and gas reflux. Gut. 2004;53(7):1024-1031. 100. Forootan M, Ardeshiri M, Etemadi N, Maghsoodi N, Poorsaadati
84. Wildi SM, Tutuian R, Castell DO. The influence of rapid food S. Findings of impedance pH-monitoring in patients with atypical
intake on postprandial reflux: studies in healthy volunteers. Am J gastroesophageal reflux symptoms. Gastroenterol Hepatol Bed Bench.
Gastroenterol. 2004;99(9):1645-1651. 2013;6(suppl 1):S117-S121.
85. Shay S, Tutuian R, Sifrim D, et al. Twenty-four hour ambulatory 101. Iqbal M, Batch AJ, Spychal RT, Cooper BT. Outcome of surgical
simultaneous impedance and pH monitoring: a multicenter report fundoplication for extraesophageal (atypical) manifestations of
of normal values from 60 healthy volunteers. Am J Gastroenterol. gastroesophageal reflux disease in adults: a systematic review. J
2004;99(6):1037-1043. Laparoendosc Adv Surg Tech A. 2008;18(6):789-796.
CHAPTER
Novel Diagnostic Technologies: Mucosal Impedance,
Optical Coherence Tomography, Endomicroscopy 10
Fahim Habib
| Blair A. Jobe
E
ndoscopy remains the fundamental diagnostic determine if the flow is anterograde or retrograde. When
technique for the evaluation of a wide spectrum there is no measurable movement of liquid or gas through
of esophageal disease. Its cardinal purposes are the esophagus, the catheter is in direct contact with the
to perform a comprehensive visual assessment of the esophageal mucosa. The mucosal impedance measured
esophagus, and to obtain tissue for histologic evaluation. under such conditions is termed the esophageal baseline
The current standard for endoscopic evaluation, white-light impedance value. This baseline impedance correlates with
endoscopy (WLE), suffers from several limitations. Key transepithelial resistance (TER), which is reflective of the
among these is its inability to identify subtle mucosal structural integrity of the esophageal mucosa. The TER is
abnormalities and to detect pathology beneath the surface significantly influenced by the structural integrity of the
of mucosa, thereby resulting in a failure to establish the intercellular junction complex,9,10 a critical element in the
diagnosis or an inability to make the diagnosis at an early maintenance of epithelial barrier function. These tight
stage when therapy is more likely to be potentially curative. junctions function to seal off paracellular pathways, form
This is especially important in the surveillance of Barrett paracellular ion channels, and act as transporters.11 In
esophagus, a rising problem in Western countries due to the presence of reflux, both acid and nonacid, structural
the significant increase in incidence of gastroesophageal alterations in the mucosa occur, critical among which is
reflux disease (GERD). Several advanced imaging options the appearance of dilated intercellular spaces (DISs) with
have been developed to overcome the limitations of resultant impairment in esophageal mucosal integrity12
conventional WLE. These include high-definition WLE, and a decrease in TER, and as a consequence, a lowered
magnification endoscopy,1 chemical chromoendoscopy,2–5 baseline impedance value.13
and electronic chromoendoscopy including narrow band Currently, preliminary studies demonstrate a potential
imaging.6,7 These advanced imaging techniques have for using measures of baseline mucosal impedance in
been shown to significantly increase the detection of the clinical evaluation of patients with GERD, assessing
dysplasia in Barrett esophagus, and reduce the number adequacy of acid suppression therapy in patients with
of biopsies needed to adequately screen patients.8 In spite Barrett esophagus, and in the diagnosis and response to
of these advances, our ability to accurately detect the treatment with fluticasone for eosinophilic esophagitis.
presence of dysplasia and early carcinoma remains at best Baseline impedance can be measured using a special-
suboptimal, and the need for the development of novel ized multichannel intraluminal impedance/pH catheter
diagnostic technologies remains critical. Among the novel (ComfortTec, Sandhill Scientific, Inc, Highlands Ranch,
diagnostic technologies under development, those with Colorado). The length of the catheter selected for use
current clinical applicability include mucosal impedance, is based on the height of the patient; each catheter has
optical coherence tomography, and endomicroscopy. specialized circumferential electrodes located at 3, 5, 7, 9,
In addition, several other novel technologies are in the 15, and 17 cm from the tip. The pH probe is placed 5 cm
process of development but have yet to gain clinical utility. above the upper margin of the manometrically defined
These include optical coherence tomography angiography lower esophageal sphincter (LES). Impedance and pH
(OCTA), capsule volumetric laser endomicroscopy (VLE), signals are collected at a sampling rate resolution of 50 Hz.
and Raman spectroscopy, among others. Data recorded over a 24-hour period are analyzed using
a dedicated software program (BioView analysis; Sandhill
Scientific, Inc, Highlands Ranch, Colorado). Distal baseline
MUCOSAL IMPEDANCE impedance is calculated using data obtained from the
Impedance is a measure of resistance to flow of current. sensor at 3 cm, while proximal baseline impedance is
Impedance of the esophageal mucosa can be measured by calculated from data obtained at the sensor 17 cm above
placing a specialized multichannel catheter with a series of the LES. Measurements are made during four different
conducting rings into the lumen of the esophagus. When periods of time over the 24 hours of recording. The first
there is movement of intraluminal material between these interval is the time between breakfast and lunch, the
conducting rings, changes in impedance occur, and can second interval is between lunch and dinner, the third
be recorded in terms of amplitude of impedance change interval is between dinner and going to sleep, and the
over time. Presence of intraluminal liquids as a result of fourth interval is during sleep. During each of these four
oral intake or reflux results in a drop in impedance from time intervals, three different time periods are selected,
the baseline, while gas due to burping or swallowing of each of 1-minute duration, where a constantly stable
air results in a rise in impedance above the baseline. impedance tracing is noted without alterations due to
The directionality of change in impedance allows us to swallowing or reflux events. Impedance during each
141
Novel Diagnostic Technologies: Mucosal Impedance, Optical Coherence Tomography, Endomicroscopy CHAPTER 10 141.e1
ABSTRACT
Visual evaluation and random or targeted biopsies during
conventional white-light endoscopy is the current standard
for the evaluation of esophageal disease. This approach has
a limited ability to detect disease in the early stages, where
the visual changes may not be evident, when the disease
is predominantly located beneath the visible surface, and
when the disease process has no pathognomonic visual
features. There is, therefore, a great need for the develop-
ment of novel diagnostic technologies for the diagnosis
of esophageal disease. Of the various technologies in
development, mucosal impedance, the second-generation
optical coherence tomography technology of volumetric
laser endomicroscopy (VLE), and confocal laser microen-
doscopy are the best developed and have current clinical
applications. Additional technologies such as capsule VLE,
optical coherence tomography angiography (OCTA), and
Raman spectroscopy are under development.
KEYWORDS
Mucosal impedance, optical coherence tomography, volu-
metric laser microendoscopy, Gastroesophageal reflux
disease, Barrett esophagus, eosinophilic esophagitis, peroral
endoscopic myotomy
142 SECTION I Esophagus and Hernia
17 cm
15 cm
9 cm 30 s
7 cm
5 cm
3 cm
FIGURE 10.1 Protocol for measuring baseline impedance. The 24-hour impedance recording is divided into four periods: two periods
between meals, one period before sleep, and one period during sleep. During each of these periods, three different 1-minute
measurements are made at times when the impedance is at a stable level. The average of these three measurements is calculated. The
final baseline impedance is calculated by averaging the average impedance of these four periods. (From Min YW. Impaired esophageal
mucosal integrity may play a causative role in patients with nongastroesophageal reflux disease-related noncardiac chest pain. Medicine
[Baltimore]. 2015;94:1–6.)
of these intervals is calculated by averaging the three median baseline impedance values being lower at 2 cm
impedance readings obtained, and then overall baseline above the squamocolumnar junction in patients with
impedance at a particular site is calculated by averaging the GERD as compared to those without GERD, baseline
measurements made at the four time intervals (Fig. 10.1). impedance is decreased even further at sites of erosive
Farré and colleagues placed impedance catheters in esophagitis compared to those without erosion. Further, a
rabbits, and after obtaining baseline values, perfused a graded increase in baseline impedance along the axis of
control solution with a pH of 7.2. During perfusion of the esophagus from distal to proximal is seen in patients
the liquid solution, impedance dropped dramatically, with GERD. A similar gradient is not seen in individuals
and on cessation of perfusion the impedance recovered without reflux.16
immediately. In contrast, when perfusion was performed Wright et al. performed a prospective observational
using saline with a pH of 1.5 and 1.0, impedance values study to determine if mucosal impedance values could
persistently remained lower at 39.1 ± 7.0% and 63.9 ± help determine if patients with Barrett esophagus were
6.5% (P < .05) respectively. The in vivo basal impedance compliant with their acid suppression regimen. All patients
correlated positively with the TER obtained in vitro (r had histologically confirmed intestinal metaplasia. Mea-
= 0.72; P = .0021). Histologic evaluation of the mucosa surements of mucosal impedance were made using a
demonstrated induction of the DIS even though the customized single-channel mucosal impedance catheter
tissue showed no gross evidence of erosions. Similarly, with unique sensors that was passed through the working
in healthy volunteers, infusion with saline of pH 2.0 and channel of a standard endoscope. Mucosal impedance was
1.0 resulted in the baseline impedance remaining lower measured at the site of Barrett epithelium as well as 2, 5,
at 21.9 ± 6.5% and 52.7 ± 5.0% (P < .0001) respectively.14 and 10 cm above the squamocolumnar junction. Mucosal
Kessing et al. studied esophageal baseline impedance impedance values in patients with Barrett esophagus who
levels in patients with GERD both on and off therapy, were not compliant with their acid suppression therapy
and in normal controls. They found a negative correla- were lower and similar to the values in patients with GERD.
tion between esophageal acid exposure time and distal In contrast, mucosal impedance values in patients who
baseline impedance, suggesting that acid reflux lowers were compliant with the acid suppression therapy were
baseline impedance levels. Further, use of proton pump higher and similar to values found in patients without
inhibitors resulted in increases in baseline impedance, GERD (Fig. 10.2).17
suggesting the role of acid exposure in altering mucosal In addition, a number of small studies have investi-
integrity in lowering baseline impedance.15 In addition to gated the role of changes in baseline impedance of the
Novel Diagnostic Technologies: Mucosal Impedance, Optical Coherence Tomography, Endomicroscopy CHAPTER 10 143
EoO Control
Proximal Proximal
5000 Ω 5000 Ω
impedance impedance
Mid Mid
impedance impedance
Distal Distal
impedance impedance
pH channel pH 1–10 pH channel pH 1–10
FIGURE 10.3 pH impedance showing significantly lower baseline impedance levels in the proximal, mid, and distal esophagus in patients
with eosinophilic esophagitis as compared with normal controls. (From Van Rhijn BD, Kessing BF, Smout AJ, Bredenoord AJ.
Oesophageal baseline impedance values are decreased in patients with eosinophilic oesophagitis. United Eur Gastroenterol J.
2013;1:242–248.)
eosinophilic esophagitis (1909 Ohms) than subjects with the potential toxicities resulting from the long-term use
inactive disease (4349 Ohms) or in controls (5530 Ohms). of proton pump inhibitors.
There was a significant inverse correlation between mucosal
impedance and the number of eosinophils per HPF on
histology (RS = −0.584). When a mucosal impedance
OPTICAL COHERENCE TOMOGRAPHY
cutoff value of 2300 Ohms was used to define active OCT represents a clinically useful “red flag technology”
eosinophilic esophagitis, test characteristics demonstrated that allows for the evaluation of large mucosal surface
90% sensitivity and 91% specificity.25 The low baseline areas at high resolution,28 based on the principles of echo
impedance levels are apparent over the entire esophagus time delay and back scattering of light. Using a rotating
including the proximal, mid, and distal portions (Fig. optical laser probe that passes through the working channel
10.3). Serial changes in baseline impedance can also of the endoscope and contacts the esophageal mucosa,
be used to determine response to therapy. Van Rhijn near-infrared light is transmitted onto the surface of the
and colleagues in a prospective study of 15 patients with esophagus and the light reflectivity of esophageal tissue
eosinophilic esophagitis, assessed esophageal mucosal is measured. This allows generation of real-time images
barrier integrity before and after the 8-week course of swal- that permit detection of mucosal surface changes, and
lowed fluticasone propionate (500 µg twice daily [BID]). changes in glandular architecture on a microscopic scale
Substantial increases in esophageal mucosal integrity with an axial resolution of 10 µm. Although this degree of
were noted following treatment with increase in mucosal resolution is 10 times that of high-frequency endoscopic
impedance and transepithelial electrical resistance, and ultrasound, there is less depth of penetration. Changes in
decrease in transit epithelial molecular flux.26 architecture facilitate identification of targets suspicious for
The role of acid suppression therapy with proton pump intestinal metaplasia, which can subsequently be subjected
inhibitors in patients with the eosinophilic esophagitis to targeted biopsy.29,30
is unclear. Although some patients clearly benefit from VLE is a second-generation frequency domain optical
being placed on acid suppression therapy, others do not. coherence tomography imaging technique that is currently
Studies of mucosal impedance suggest that eosinophilic available for clinical use in the United States (NinePoint
esophagitis may not represent a single disease entity but Medical Inc., Bedford, Massachusetts). An integrated
rather represent a heterogeneous group of disorders. imaging system consisting of a console, monitor, and
One such subgroup has been termed proton pump a near-infrared-light optical probe centered within a
inhibitor responsive eosinophilia (PPI-REE). Although transparent balloon is used to deliver a 1350-nm wavelength
parameters of mucosal integrity (electrical tissue imped- laser that emanates from the probe in a helical manner
ance, transepithelial electrical resistance, transit epithelial with automated pullback. The system allows fast real-time
molecule flux) are reduced at baseline in both classic imaging with a 6-cm length of esophagus being scanned
eosinophilic esophagitis and PPI-REE, treatment with in approximately 90 seconds and provides wide-field,
high-dose acid suppression therapy using proton pump cross-sectional imaging of the surface and subsurface of
inhibitors partially restores mucosal integrity in patients the esophagus with an axial resolution of up to 7 µm and
with PPI-REE.27 Measurement of mucosal impedance down to a depth of 3 mm (Fig. 10.4).
before and after an 8-week course of acid suppression Prior to performing VLE, a diagnostic endoscopy is
using proton pump inhibitors may therefore help identify performed with WLE using a high-definition diagnostic
the subset of patients who may benefit from continued endoscope (HQ 190, Olympus, Tokyo, Japan) to identify
treatment, and allow the discontinuation of these medica- the location of the gastroesophageal junction, and evaluate
tions in patients who do not show any improvement in the esophagus for conditions that would preclude safe
mucosal impedance, thereby reducing the exposure to passage and inflation of the optical probe; these include
Novel Diagnostic Technologies: Mucosal Impedance, Optical Coherence Tomography, Endomicroscopy CHAPTER 10 145
Physician display
monitor
System
control
monitor and
keyboard
Longitudinal
recut of VLE
Cross-section of dataset
esophagus
Region of
Region of interest 2
interest 1
Magnified
ROI 1 Magnified
A C ROI 2
Esophagus
Probe scanner
Balloon
Imaging beam
Optics core
FIGURE 10.4 The Nvision volumetric laser endomicroscopy (VLE) system. (A) Imaging system console. (B) Optical probe with centering
balloon. (C) The Nvision VLE display shows a complete circumferential cross-sectional view of the esophagus, with longitudinal recuts of
the 3-dimensional VLE dataset displayed alongside, and magnified regions of interest in the circumferential and longitudinal views shown
in separate insets. (From Wolfson HC. Safety and feasibility of volumetric laser endomicroscopy in patients with Barrett’s esophagus [with
videos]. Gastrointest Endosc. 2015;82:631–640.)
146 SECTION I Esophagus and Hernia
the presence of mass lesions or severe strictures. Caution 20 psi may be necessary to center the balloon. On comple-
should be exercised in patients with a tortuous esophagus, tion of the scan, the balloon is deflated, and the location
and in those with suspected eosinophilic esophagitis. of the registration line as described according to the face
The extent of the Barrett segment is recorded using the of the clock is noted. This then allows orientation of the
Prague criteria.31 During endoscopy the appropriate size cross-sectional imaging on the monitor with its location
of the optical probe to be used can also be determined. in the esophagus allowing for targeted biopsies. When
Although the standard length of the optical probe is 6 cm, the segment of interest is greater than 6 cm, multiple
it is available in diameters of 14 mm, 17 mm, and 20 mm. scans may be performed. Here it is critical to identify
In our experience, in the absence of major pathology the a unique anatomic characteristic that is visualized on
20 mm balloon can be used in virtually all cases. After both scans and to overlap them accurately such that the
connecting the optical probe to the console, the probe resultant scan represents a seamless continuum. In the
is passed through the working channel of the endoscope absence of such a unique anatomic characteristic, cautery
until the entire balloon extrudes from the scope and the marks may be made, and will be visible on the scan and
opaque marker at the proximal end of the probe can be can then be used to align multiple scans. Localization
visualized. The balloon is then positioned to lie with its of lesions is achieved by a process of triangulation based
distal tip 1 cm beyond the gastroesophageal junction. on the location of the gastroesophageal junction and the
This is achieved by positioning the scope 7 cm above the registration line. Superficial abnormalities can be subjected
location of the gastroesophageal junction as identified to precisely targeted biopsies, while submucosal abnormali-
during the screening endoscopy. Once the balloon has ties can be biopsied using endoscopic mucosal resection
been positioned, it is important to hold the probe firmly techniques.
between the fingers at the level of the bite block, and as it On VLE the normal gastric mucosa demonstrates vertical
emerges from the working channel, to prevent movement pit and crypt architecture, has high surface reactivity, shows
of the balloon as it is inflated. If the probe is left lax, the poor image penetration, and thick characteristic presence
balloon tends to drop into the stomach. The custom syringe of rugal folds. Normal esophageal squamous mucosa is
is then used to inflate the balloon with air to a pressure visualized as was a layered horizontal architecture without
of 10 psi. A scout scan is then obtained. The primary the presence of glands in the epithelium (Fig. 10.5). In
purpose of the scout scan is to ensure that the obtained contrast, esophageal mucosa with intestinal metaplasia
scan includes the gastric cardia, the gastroesophageal shows a loss of layered architecture and the presence of
junction, and the distal portion of the esophagus. It is also glands in the epithelium (Fig. 10.6).
used to confirm that the balloon is optimally centered. If In the most recent version of the VLE, which has only
the balloon has migrated, it can be deflated, repositioned, just become available, a 1470-nm wavelength laser is
reinflated, and a repeat scout scan obtained to confirm used. A hand control attached to the endoscope allows
ideal placement. The balloon is then inflated to 15 psi, the operator to navigate the optical probe in real time
and a full scan obtained. The full scan is generated by the through the esophagus. On identification of areas of
automatic helical pullback of the probe from the distal interest, the laser can be fired, creating a cautery mark
to the proximal end of the balloon. This occurs over 90 on the surface of the esophagus. Subsequent WLE allows
seconds and creates a real-time 360-degree volumetric the endoscopist to identify the marked area and facilitate
image that is the composite of 1200 cross-sectional scans the performance of VLE-directed biopsies (Fig. 10.7).32,33
that are generated over the 6-cm segment. In cases with The most common current clinical indications for the
a dilated esophagus, a higher inflation pressure of up to use of VLE are Barrett esophagus, including surveillance
Squamous epithelium
Lamina propria
muscularis mucosa
Submucosa
Muscularis propria
FIGURE 10.5 Histologic correlates of the layers of normal esophageal wall obtained by volumetric laser endomicroscopy. (From Lightdale
CJ. Optical coherence tomography in Barrett’s esophagus. Gastrointest Endosc Clin N Am. 2013;23:549–563.)
Novel Diagnostic Technologies: Mucosal Impedance, Optical Coherence Tomography, Endomicroscopy CHAPTER 10 147
of high-risk, treatment-naive Barrett esophagus, and for as characterized by surface intensity that is greater than the
postablation surveillance to detect recurrence or persis- intensity of the subsurface tissue denotes a high suspicion
tence.34,35 VLE is also used to guide peroral endoscopic for dysplasia. Then, in cases where effacement is partial,
myotomy (POEM) for the treatment of achalasia. presence of atypical glands is assessed with the detection
of more than five atypical glands, raising suspicion for the
SURVEILLANCE OF TREATMENT-NAIVE presence of dysplasia. Using this algorithm Legget et al.
BARRETT ESOPHAGUS obtained a sensitivity of 86% and a specificity of 88% for
The generated image allows differentiation between the detection of dysplasia.38
normal and abnormal tissue in the stomach and in the Swager et al. recently presented results of her study
esophagus.36 On VLE, normal gastric cardia demonstrates investigating the feasibility of a computer algorithm for
vertical pit and crypt architecture, high surface reflectivity, detection of early Barrett neoplasia using VLE. A total of
poor image penetration, and presence of rugal folds. 60 VLE images were used in the study, of which 30 were
Normal esophageal squamous mucosa demonstrates a dysplastic and 30 nondysplastic. Decision making by the
layered horizontal architecture without the presence of computer-based algorithm relied on features associated
glands in the epithelium. Loss of layered architecture with dysplasia including (1) a higher optical frequency
and presence of glands in the epithelium are suggestive domain imaging (OFDI) surface signal than subsurface
of intestinal metaplasia.37 Presence of dysplasia is assessed signal in tissue and (2) a lack of layering. The algorithm
on the basis of two features of the obtained image: signal had a sensitivity of 93%, specificity of 70%, and accuracy
intensity and glandular architecture. Signal intensity is of 82%. The area under the curve (AUC) on the receiver
assessed by comparing the surface signal intensity with operating characteristic (ROC) curve was 0.91, which
the signal intensity in the subsurface tissue. Normally, compares favorably with a recently developed clinical
surface intensity should be less than subsurface intensity. prediction model that had an AUC of 0.81.39,40 In the latter
Partial effacement is present when this surface intensity is model, Swager et al. found three independent predictors
equivalent to that of subsurface tissue. Full effacement is for neoplasia in Barrett esophagus: (1) lack of layering
present when the surface intensity is greater than that of (6 points); (2) higher surface than subsurface signal
the subsurface tissue. Glandular architecture is assessed (6 points for a surface signal that was equivalent to the
by determining the presence or absence of glands, with subsurface tissue, and 8 points for a higher surface signal
or without atypia. Using these two features, Leggett et al. when compared to subsurface tissue); and (3) presence
developed an OCT scoring index (OCT-SI) (Fig. 10.8). of irregular, dilated glands/ducts (5 points) (Fig. 10.9).
A score of 2 or higher is associated with 83% sensitivity The ROC curve in the validation phase showed an AUC
and 75% specificity for dysplasia. They also used these of 0.81 (95% confidence interval [CI], 0.71 to 0.90).
characteristics to develop a VLE diagnostic algorithm Using a cutoff value of more than 8 points, sensitivity of
based on ex vivo VLE assessment of specimens obtained 83% and specificity of 71% were obtained. As with several
by endoscopic mucosal resection. First, a comparative studies involving VLE, histologic analysis was performed
assessment of surface intensity versus the intensity of on ex vivo specimens. The applicability of these findings
subsurface tissue is performed. Presence of full effacement in vivo has yet to be determined.
148 SECTION I Esophagus and Hernia
A B C
D E F
G H I
FIGURE 10.7 Visibility of laser marks (arrows) in gastric (A, D, G), nondysplastic (B, E, H), and squamous mucosa (C, F, I) on white-light
endoscopy, circumferential volumetric laser endomicroscopy (VLE), and zoom-in views on VLE. (From Swager A, de Groof AJ, Meijer SL,
Weusten BL, Curvers WL, Bergman JJ. Feasibility of laser marking in Barrett’s esophagus with volumetric laser endomicroscopy:
first-in-man pilot study. Gastrointest Endosc. pii:S0016-5107[17]30074-3.)
The ability to mark the area of interest during per- allows direct in vivo marking of suspicious areas as they
formance of VLE with subsequent targeted biopsy of are identified during real-time VLE. One of two marking
the marked regions has potential to further improve the options may be used. In the offset marking mode, two
accuracy of image-guided biopsy for Barrett esophagus. laser marks are placed automatically in the horizontal
Suter et al. first explored this in a pilot feasibility study position 6 mm apart. In the point marking mode, a single
involving 22 patients.41 Burn marks using lasers of different laser mark is placed. The marking laser is activated and
wavelengths were made on either side of the VLE-detected controlled by the operator with a handheld control that
abnormality. The optimal marking parameter was found to is attached to the endoscope. In a recent pilot feasibility
be 2 seconds at 410 mW, with separation of the two marks study, the protocol for laser marking was developed in
by 6 mm. All marks made were visible at endoscopy. There a primary learning phase. Subsequently the visibility on
were no adverse events related to VLE or laser marking. follow-up endoscopy of random laser markings made in
Accuracy for corrected VLE postmarking images was 100% squamous, Barrett, and gastric tissue was assessed; positional
when compared to histologic interpretations. This approach accuracy was tested; and finally, the most suspicious area for
has been further refined, and the newest version of the VLE neoplasia as identified by VLE was targeted. The location
imaging system incorporates a laser marking system. This of laser marks made could be identified on VLE in 100%
Novel Diagnostic Technologies: Mucosal Impedance, Optical Coherence Tomography, Endomicroscopy CHAPTER 10 149
+
Surface intensity < subsurface intensity = 0
++
Signal intensity score
+
Surface intensity = subsurface intensity = 1
++
+
Surface intensity > subsurface intensity = 2
++
No mucosal glands = 0
Glandular architecture score
1 mm
FIGURE 10.8 OCT Scoring Index (OCT-SI). A score of ≥2 has a sensitivity of 83% and specificity of 75% for detecting dysplasia. (From
Leggett CL, Gorospe EC, Chan DK, et al. Comparative diagnostic performance of volumetric laser endomicroscopy and confocal laser
endomicroscopy in the detection of dysplasia associated with Barrett’s esophagus. Gastrointest Endosc. 2016;83:880–888.e2.)
of cases and on WLE in 97%. Areas found to be suspicious incidence of buried Barrett glands following radiofrequency
on VLE were successfully targeted by the laser marks in ablation was further confirmed by Swager and colleagues.
all cases.33 Although VLE identified subsquamous granular structures
following radiofrequency ablation of Barrett esophagus, the
POSTABLATION SURVEILLANCE overwhelming majority of identified subsquamous glands
OF BARRETT ESOPHAGUS were determined to be normal histologic structures such
When endoscopic ablation of Barrett esophagus is per- as dilated glands and blood vessels. Buried glands were
formed, residual metaplastic glands can remain under the identified in only 1 of 17 post-radiofrequency ablation
layer of neosquamous epithelium that forms. In addition, patients. The development of high-grade dysplasia and
buried glands may be present at the junction of Barrett adenocarcinoma in these varied subsquamous metaplastic
epithelium and squamous epithelium. The incidence of tissues has been reported.43,44 Characteristic VLE findings
buried glands is higher following photodynamic therapy include increased signal intensity and atypical appearing
(14.2%) than when radiofrequency ablation is employed as glands with dilated ducts under a layer of squamous
the treatment modality (0.9%).42 However, these reported epithelium.
incidences may underestimate the true incidence of buried Trindade et al.45 recently began to address the issue of
glands because biopsy specimens may not contain suf- reproducibility of the interpretation of VLE images. Eight
ficient subepithelial tissue to make the diagnosis. This low high-frequency VLE users (who have read more than 50
150 SECTION I Esophagus and Hernia
Mucosal layer
+ +
++
++ 1 mm
FIGURE 10.9 The volumetric laser endomicroscopy diagnostic algorithm index has a sensitivity of 86% and a specificity of 88% in
detecting dysplasia. (From Leggett CL, Gorospe EC, Chan DK, et al. Comparative diagnostic performance of volumetric laser
endomicroscopy and confocal laser endomicroscopy in the detection of dysplasia associated with Barrett’s esophagus. Gastrointest
Endosc. 2016;83:880–888.e2.)
VLE examinations) from academic tertiary care centers available at this time. It does, however, seem plausible that
in the United States evaluated 120 identified images replacement of extensive protocol biopsies such as those
classifying them as gastric cardia, esophageal squamous employed by the Seattle protocol with targeted biopsies
mucosa, nonneoplastic Barrett esophagus (nondysplastic and a reduction in the need for multiple subsequent
Barrett esophagus and low-grade dysplasia), and neoplastic endoscopic procedures has the potential to result in an
Barrett esophagus. They were also instructed to distin- overall cost saving in the long term.
guish nonneoplastic Barrett esophagus from neoplastic
Barrett esophagus using the OCT-SI. An OCT-SI score VOLUMETRIC LASER ENDOMICROSCOPY
of 2 or greater was used as the criteria of neoplastic IN ACHALASIA
disease. The overall agreement among users was excellent Desai and colleagues recently described a novel application
(kappa equals 0.81; 95% CI, 0.79 to 0.83). Agreement of optical coherence tomography. They incorporated VLE
was near perfect for esophageal squamous and gastric into the algorithm for assessing the esophagus and planning
cardia (kappa equals 0.95 and 0.86, respectively [95% approach prior to POEM in the treatment of achalasia. VLE
CI, 0.92 to 0.98 and 0.83 to 0.89]). Agreement was strong images were analyzed for regional differences in thickness
for distinguishing nonneoplastic Barrett esophagus and of circular muscle, and for the presence of large vessels at
neoplastic Barrett esophagus (Kappa = 0.66 [95% CI, the location of intended incision. Based on these features,
0.63 to 0.69] and Kappa = 0.79 [95% CI, 0.75 to 0.82], a decision to perform an anterior POEM or a posterior
respectively). Overall median accuracy for identifying POEM was made. Results were compared in patients who
the correct tissue type was 96% (95% CI, 94% to 99%). underwent the procedure prior to introduction of VLE
These findings suggest that with increasing experience for preoperative assessment with those who underwent
and application of uniform criteria, targeted biopsies the procedure after this new diagnostic modality was
can be made a clinical reality. However, one must take adopted into clinical practice. Less bleeding (8% vs. 43%;
into account that these images were analyzed offline. P = .0001) and a shorter procedural time (85.8 vs. 121.7
Interpretation of images in real time as the study is in minutes; P = .000097) were noted when the use of VLE
progress may not be as easily learned. Hence, at the present was introduced.46 An additional use of optical coherence
time, random biopsies obtained using the Seattle protocol tomography is to assess the adequacy of myotomy after
should still remain the gold standard. Targeted biopsies as performance of a POEM procedure. A full scan of the
guided by abnormalities noted on VLE must be taken in esophagus can be performed, and demonstrates a break
addition. in the muscular layers at the site where the POEM was
VLE has only recently been adopted into clinical practice performed on (Fig. 10.10). This technique can therefore
and largely at institutions specializing in the management of potentially be used to evaluate recurrence of symptoms
Barrett esophagus. Cost effectiveness data are therefore not following endoscopic myotomy.47
Novel Diagnostic Technologies: Mucosal Impedance, Optical Coherence Tomography, Endomicroscopy CHAPTER 10 151
FIGURE 10.10 Use of optical coherence tomography to evaluate success of myotomy following peroral endoscopic myotomy (POEM).
(From Parra V, Kedia P, Minami H, Sharaiha RZ, Kahaleh M. Endoscopic optical coherence tomography as a tool to evaluate successful
myotomy after a peroral endoscopic myotomy. Gastroint Endosc. 2015;81;1251.)
A B
C D
FIGURE 10.11 Confocal microscopy Miami criteria. (A) Normal squamous epithelium: Flat cells without crypts or villi. (B) Barrett without
dysplasia: Uniform villiform architecture, columnar cells interspersed with dark goblet cells. (C) High-grade dysplasia: Villiform structure
preserved, dark irregular thickened epithelial borders, dilated irregular vessels. (D) Adenocarcinoma: Villiform structure absent or
significantly disorganized, multiple dark columnar cells with dilated irregular vessels. (From Shahid MW, Wallace MB. Endoscopic imaging
for the detection of esophageal dysplasia and carcinoma. Gastrointest Endosc Clin N Am. 2010;20:17; with permission.)
Novel Diagnostic Technologies: Mucosal Impedance, Optical Coherence Tomography, Endomicroscopy CHAPTER 10 153
AFI
Collection
Excitation fibers WLR/NBI
fiber
Distal tip of
endoscope
Epithelium
stroma
FIGURE 10.12 The rapid fiberoptic confocal Raman spectroscopy system. AFI, Autofluorescence imaging; CCD, charge-coupled device;
WLR/NBI, white light reflectance/narrow band imaging. (From Bergholt MA, Zheng W, Ho KY, et al. Fiberoptic confocal Raman
spectroscopy for real-time in vivo diagnosis of Barrett’s esophagus. Gastroenterology. 2014;146:27–32.)
the necessity for intravenous administration of fluorescent the molecular composition of the tissue being evaluated,
contrast agents, which may temporarily cause the skin providing histopathologic assessments at the biomolecular
and urine to become yellow to orange for up to 24 hours. level.57 Early studies on ex vivo esophageal tissue demon-
strated a sensitivity of 86% and a specificity of 88% in the
detection of high-grade dysplasia and adenocarcinoma.
OTHER EMERGING TECHNOLOGIES The technique was also able to identify the presence of
Capsule VLE imaging: Liang and colleagues have developed intestinal metaplasia and grade the degree of dysplasia.58
an ultra-high-speed OCT tethered capsule that is capable Recently, a novel endoscopic confocal Raman probe was
of obtaining a large field of view, imaging the subsurface, developed that allows rapid acquisition of data from
and obtaining cross-sectional volumetric imaging of the large tissue areas (Fig. 10.12). The utility of this probe is
esophagus. This capsule, which is 30 mm long and 12 mm currently under evaluation.59
in diameter, has a micro motor that scans at 300 frames Endoscopic optical coherence angiography is being
per second. The capsule is attached to a semirigid tether developed to overcome limitations inherent to current
and is introduced into the esophagus of sedated patients, OCT modalities including limited frame rate, narrow
with a pull-push technique to volumetrically map the field of view, sensitivity to motion artifact, and limitation
esophagus. In its current form the technology suffers of analysis to surface vascular patterns. Using a micro
from an inability to image the full circumference of the motor catheter, circumferential cross-sectional scans in a
esophagus. Regions of noncontact appear dark. The need helical pattern are performed at 400 frames per second
for manual pull–push, movements due to respiration, with images from an area greater than 100 mm2 being
cardiac motion, and peristalsis also produce image distor- acquired in about 8 seconds (Fig. 10.13). Detection of
tion. Further refinements in the technology are necessary the intensity correlation generated by moving erythrocytes
to overcome these limitations.56 allows further visualization of the microvasculature without
Raman spectroscopy is based on the principle of inelastic the need for administration of exogenous contrast
scattering of light with frequency shifts that correspond to agents.60
154 SECTION I Esophagus and Hernia
X X
r r
A B
X X
r r
*
*
BE
C D
X X
r r
SE
E F
FIGURE 10.13 En face optical coherence tomography angiography. (A–C) Nondysplastic Barrett esophagus: Regular honeycomb
microvascular pattern (arrows), may be compressed or streaked in the longitudinal direction due to motion artifact (stars). (D–E) High-
grade dysplasia. The dashed line delineates the boundary between abnormal microvasculature and neighboring nondysplastic regions.
(F) Low-grade dysplasia: Abnormal vessel branching with crowding, a corkscrew appearance, and heterogeneous vessel size (arrows).
(From Lee HC, Ahsen OO, Liang K, et al. Endoscopic optical coherence tomography angiography microvascular features associated with
dysplasia in Barrett’s esophagus. Gastrointest Endosc. 2017;pii:S0016-5107[17]30078-0.)
11. Colegio OR, Van Itallie CM, McCrea HJ, Rahner C, Anderson 33. Swager A, de Groof AJ, Meijer SL, Weusten BL, Curvers WL, Bergman
JM. Claudins create charge-selective channels in the paracel- JJ. Feasibility of laser marking in Barrett’s esophagus with volumetric
lular pathway between epithelial cells. Am J Physiol Cell Physiol. laser endomicroscopy: first-in-man pilot study. Gastrointest Endosc.
2002;283(1):C142-C147. 2017;pii: S0016-5107(17)30074-3.
12. Orlando LA, Orlando RC. Dilated intercellular spaces as a marker 34. Leggett CL, Gorospe E, Owens VL, Anderson M, Lutzke L, Wang
of GERD. Curr Gastroenterol Rep. 2009;11(3):190-194. KK. Volumetric laser endomicroscopy detects subsquamous Barrett’s
13. van Malenstein H, Farré R, Sifrim D. Esophageal dilated intercellular adenocarcinoma. Am J Gastroenterol. 2014;109(2):298-299.
spaces (DIS) and nonerosive reflux disease. Am J Gastroenterol. 35. Swager AF, Tearney GJ, Leggett CL, et al. Identification of volumetric
2008;103(4):1021-1028. laser endomicroscopy features predictive for early neoplasia in
14. Farré R, Blondeau K, Clement D, et al. Evaluation of oesophageal Barrett’s esophagus using high-quality histological correlation.
mucosa integrity by the intraluminal impedance technique. Gut. Gastrointest Endosc. 2016;pii: S0016-5107(16)30581-8.
2011;60(7):885-892. 36. Sauk J, Coron E, Kava L, et al. Interobserver agreement for the
15. Kessing BF, Bredenoord AJ, Weijenborg PW, Hemmink GJ, Loots detection of Barrett’s esophagus with optical frequency domain
CM, Smout AJ. Esophageal acid exposure decreases intraluminal imaging. Dig Dis Sci. 2013;58(8):2261-2265.
baseline impedance level. Am J Gastroenterol. 2011;106(12):2093-2097. 37. Evans JA, Bouma BE, Bressner J, et al. Identifying intestinal meta-
16. Yuksel ES, Higginbotham T, Slaughter JC, et al. Use of direct, plasia at the squamocolumnar junction by using optical coherence
endoscopic-guided measurements of mucosal impedance in diag- tomography. Gastrointest Endosc. 2007;65(1):50-56.
nosis of gastroesophageal reflux disease. Clin Gastroenterol Hepatol. 38. Leggett CL, Gorospe EC, Chan DK, et al. Comparative diagnostic
2012;10(10):1110-1116. performance of volumetric laser endomicroscopy and confocal
17. Wright MR, Higginbotham T, Slaughter JC, Ates F, Yuksel ES, Vaezi laser endomicroscopy in the detection of dysplasia associated with
M. Sa1260 mucosal impedance in Barrett’s esophagus: can it assess Barrett’s esophagus. Gastrointest Endosc. 2016;83(5):880-888.
compliance with medication? Gastroenterology. 2016;150(Suppl 1, 39. Swager A, van der Sommen F, Klomp SR, et al. Computer-aided
April):S260. detection of early Barrett’s neoplasia using volumetric laser endo-
18. Fukahori S, Yagi M, Ishii S, et al. A baseline impedance analysis in microscopy. Gastrointest Endosc. 2017;pii:S0016-5107(17)30191-8.
neurologically impaired children: a potent parameter for estimating 40. Deleted in review.
the condition of the esophageal mucosa. Neurogastroenterol Motil. 41. Deleted in review.
2017;29(6):doi:10.1111/nmo.13012. 42. Gray NA, Odze RD, Spechler SJ. Buried metaplasia after endoscopic
19. Lottrup C, Krarup AL, Gregersen H, Ejstrud P, Drewes AM. Patients ablation of Barrett’s esophagus: a systematic review. Am J Gastroenterol.
with Barrett’s esophagus are hypersensitive to acid but hyposensitive 2011;106(11):1899-1908.
to other stimuli compared with healthy controls. Neurogastroenterol 43. Titi M, Overhiser A, Ulusarac O, et al. Development of subsqua-
Motil. 2017;29(4):doi:10.1111/nmo.12992. mous high-grade dysplasia and adenocarcinoma after successful
20. Weijenborg PW, Smout AJ, Bredenoord AJ. Esophageal acid sensitiv- radiofrequency ablation of Barrett’s esophagus. Gastroenterology.
ity and mucosal integrity in patients with functional heartburn. 2012;143(3):564-566.e1.
Neurogastroenterol Motil. 2016;28(11):1649-1654. 44. Deleted in review.
21. Kandulski A, Weigt J, Caro C, Jechorek D, Wex T, Malfertheiner 45. Trindade AJ, Inamdar S, Smith MS, et al. Volumetric laser endomicros-
P. Esophageal intraluminal baseline impedance differentiates copy in Barrett’s esophagus: interobserver agreement for interpretation
gastroesophageal reflux disease from functional heartburn. Clin of Barrett’s esophagus and associated neoplasia among high-frequency
Gastroenterol Hepatol. 2015;13(6):1075-1081. users. Gastrointest Endosc. 2016;pii:S0016-5107(16)30807-0.
22. Liacouras CA, Furuta GT, Hirano I, et al. Eosinophilic esophagitis: 46. Desai AP, Tyberg A, Kedia P, et al. Optical coherence tomography
updated consensus recommendations for children and adults. (OCT) prior to peroral endoscopic myotomy (POEM) reduces
J Allergy Clin Immunol. 2011;128(1):3-20. procedural time and bleeding: a multicenter international col-
23. Ravi K, Katzka DA, Smyrk TC, et al. Prevalence of esophageal laborative study. Surg Endosc. 2016;30(11):5126-5133.
eosinophils in patients with Barrett’s esophagus. Am J Gastroenterol. 47. Parra V, Kedia P, Minami H, Sharaiha RZ, Kahaleh M. Endoscopic
2011;106(5):851-857. optical coherence tomography as a tool to evaluate successful
24. Mueller S, Neureiter D, Aigner T, Stolte M. Comparison of histological myotomy after a peroral endoscopic myotomy. Gastrointest Endosc.
parameters for the diagnosis of eosinophilic oesophagitis versus 2015;81(5):1251.
gastro-oesophageal reflux disease on oesophageal biopsy material. 48. De Palma GD. Confocal laser endomicroscopy in the “in vivo”
Histopathology. 2008;53(6):676-684. histological diagnosis of the gastrointestinal tract. World J Gastroenterol.
25. Katzka DA, Ravi K, Geno DM, et al. Endoscopic mucosal impedance 2009;15(46):5770-5775.
measurements correlate with eosinophilia and dilation of intercellular 49. Wallace M, Lauwers GY, Chen Y, et al. Miami classification for
spaces in patients with eosinophilic esophagitis. Clin Gastroenterol probe-based confocal laser endomicroscopy. Endoscopy. 2011;43(10):
Hepatol. 2015;13(7):1242-1248. 882-891.
26. van Rhijn BD, Verheij J, van den Bergh Weerman MA, et al. Histologi- 50. Kiesslich R, Gossner L, Goetz M, et al. In vivo histology of Barrett’s
cal response to fluticasone propionate in patients with eosinophilic esophagus and associated neoplasia by confocal laser endomicroscopy.
esophagitis is associated with improved functional esophageal mucosal Clin Gastroenterol Hepatol. 2006;4(8):979-987.
integrity. Am J Gastroenterol. 2015;110(9):1289-1297. 51. Dunbar KB, Okolo P 3rd, Montgomery E, Canto MI. Confocal laser
27. van Rhijn BD, Weijenborg PW, Verheij J, et al. Proton pump inhibitors endomicroscopy in Barrett’s esophagus and endoscopically inappar-
partially restore mucosal integrity in patients with proton pump ent Barrett’s neoplasia: a prospective, randomized, double-blind,
inhibitor-responsive esophageal eosinophilia but not eosinophilic controlled, crossover trial. Gastrointest Endosc. 2009;70(4):645-654.
esophagitis. Clin Gastroenterol Hepatol. 2014;12(11):1815-1823. 52. Pohl H, Rösch T, Vieth M, et al. Miniprobe confocal laser microscopy
28. Anandasabapathy S. Advanced imaging in Barrett’s esophagus: for the detection of invisible neoplasia in patients with Barrett’s
are we ready to relinquish the random? Clin Gastroenterol Hepatol. oesophagus. Gut. 2008;57(12):1648-1653.
2013;11(12):1571-1572. 53. Bajbouj M, Vieth M, Rösch T, et al. Probe-based confocal laser
29. Brand S, Poneros JM, Bouma BE, Tearney GJ, Compton CC, Nishioka endomicroscopy compared with standard four-quadrant biopsy
NS. Optical coherence tomography in the gastrointestinal tract. for evaluation of neoplasia in Barrett’s esophagus. Endoscopy.
Endoscopy. 2000;32(10):796-803. 2010;42(6):435-440.
30. Sivak MV Jr, Kobayashi K, Izatt JA, et al. High-resolution endoscopic 54. Sharma P, Meining AR, Coron E, et al. Real-time increased detec-
imaging of the GI tract using optical coherence tomography. Gas- tion of neoplastic tissue in Barrett’s esophagus with probe-based
trointest Endosc. 2000;51(4 Pt 1):474-479. confocal laser endomicroscopy: final results of an international
31. Sharma P, Dent J, Armstrong D, et al. The development and validation multicenter, prospective, randomized, controlled trial. Gastrointest
of an endoscopic grading system for Barrett’s esophagus: the Prague Endosc. 2011;74(3):465-472.
C NM criteria. Gastroenterology. 2006;131(5):1392-1399. 55. Gupta A, Attar BM, Koduru P, Murali AR, Go BT, Agarwal R. Utility
32. Suter MJ, Gora MJ, Lauwers GY, et al. Esophageal-guided biopsy of confocal laser endomicroscopy in identifying high-grade dysplasia
with volumetric laser endomicroscopy and laser cautery marking: and adenocarcinoma in Barrett’s esophagus: a systematic review and
a pilot clinical study. Gastrointest Endosc. 2014;79(6):886-896. meta-analysis. Eur J Gastroenterol Hepatol. 2014;26(4):369-377.
156 SECTION I Esophagus and Hernia
56. Liang K, Ahsen OO, Lee HC, et al. Volumetric mapping of Barrett’s 59. Bergholt MS, Zheng W, Ho KY, et al. Fiberoptic confocal Raman
esophagus and dysplasia with en face optical coherence tomography spectroscopy for real-time in vivo diagnosis of dysplasia in Barrett’s
tethered capsule. Am J Gastroenterol. 2016;111(11):1664-1666. esophagus. Gastroenterology. 2014;146(1):27-32.
57. Shim MG, Song LM, Marcon NE, Wilson BC. In vivo near- 60. Tsai TH, Ahsen OO, Lee HC, et al. Endoscopic optical coherence
infrared Raman spectroscopy: demonstration of feasibility during angiography enables 3-dimensional visualization of subsurface
clinical gastrointestinal endoscopy. Photochem Photobiol. 2000;72(1): microvasculature. Gastroenterology. 2014;147(6):1219-1221.
146-150.
58. Almond LM, Hutchings J, Lloyd G, et al. Endoscopic Raman
spectroscopy enables objective diagnosis of dysplasia in Barrett’s
esophagus. Gastrointest Endosc. 2014;79(1):37-45.
PART THREE
Esophageal Motility
Disorders and Diverticula
CHAPTER
Cricopharyngeal Dysfunction and
Zenker Diverticulum 11
Giovanni Zaninotto
| Mario Costantini
D ASSESSMENT OF OROPHARYNGEAL
isorders at the cricopharyngeal level lead to oro-
pharyngeal dysphagia, or transfer dysphagia. With this
term, we usually refer to the difficulty in making DYSPHAGIA
the food progress from the oropharynx to the esophagus,
through the upper esophageal sphincter (UES), mostly CLINICAL EVALUATION
but not exclusively constituted by the cricopharyngeal Symptom assessment of a patient with dysphagia is essential
muscle. With this term, we differentiate it from esophageal for a correct diagnosis. In the case of oropharyngeal
dysphagia, which we define as difficulty in the progression dysphagia, the patient clearly describes his/her difficulty
of the bolus from the esophagus to the stomach after in transferring food from the mouth to the pharynx and
the bolus itself has been correctly transferred from the the esophagus and to initiate the involuntary (esophageal)
oropharynx to the esophagus. phase of swallowing. The patient is usually very accurate
Dysphagia has been recognized by the World Health in describing the accumulation of food in the mouth,
Organization (WHO) as a medical disability associated an inability to control the bolus in the oral cavity, the
with increased morbidity, mortality, and cost of care.1 aspiration of food before, during, or after the swallowing
The real incidence is poorly understood, but it is continu- act, and the location where the bolus has gotten stuck.
ously increasing, probably related to the longer overall Often patients with esophageal dysphagia also locate the
survival of the general population, the development of level of food arrest at the cervical region, making the
degenerative and neurologic disease, treatment of other perceived location of food arrest inaccurate for determin-
conditions and, probably, from a better understanding ing the etiology of the process. Associated symptoms of
and assessment of this symptom. In the nonhospitalized oropharyngeal dysphagia include nasal regurgitation and
elderly population, the prevalence of oropharyngeal cough related to the inadequacy of protective mechanisms,
dysphagia is 11% to 16%, whereas it may rise to 55% in and dysarthria or nasal speech related to weakness of the
hospitalized patients or nursing home residents.2 Dysphagia palate. A gurgle sound may suggest a Zenker diverticulum
is reportedly present in greater than 70% of patients (ZD; see later). Other symptoms, when present, may be
after a stroke and remains severe in 15% of patients with useful for the final diagnosis, such as a speech disorder
early swallowing problems.3 Moreover, dysphagia is now or other signs of impairment of cranial nerves. Often
recognized as a symptom of concern in conditions such dysphagia is only part of the symptom complex of the
as acquired brain injury and cervical spine surgery. The patient, and the diagnosis of a neuromuscular disorder
treatment of head and neck cancer with chemoradiation is well evident; however, sometimes dysphagia is the first
or surgery has more pronounced adverse effects on swal- symptom that causes the patient to be seen by a doctor, and
lowing function than on other functions such as breathing. only the subsequent physical and neurologic examination
Dysphagia may complicate an inflammatory myopathy in may reveal the underlying disease. Finally, weight loss may
about half the cases and may represent the presenting be the only sign of a swallowing disorder, since patients may
symptom in different neuromuscular diseases, for example, avoid eating because of the difficulties they experience.
inclusion body myositis. Finally, dysphagia is present in
up to 57% of patients with established dementia.3 The ENDOSCOPIC EVALUATION
various causes of oropharyngeal dysphagia are listed Since dysphagia is an ominous sign, endoscopy is manda-
in Box 11.1. tory to rule out any anatomic abnormality or organic
157
Cricopharyngeal Dysfunction and Zenker Diverticulum CHAPTER 11 157.e1
ABSTRACT
This chapter reviews the etiology and pathogenesis of
cricopharyngeal dysfunction (CD) and cricopharyngeal or
Zenker diverticula (ZD). The diagnostic modalities and pos-
sible treatments of each condition are examined. Surgery
(myotomy of upper esophageal sphincter [UES]) is rarely
used to treat CD; endoscopic transoral diverticulostomy to
treat ZD is becoming more and more popular, but surgery
(myotomy of UES ± diverticulectomy or diverticulopexis)
still plays a significant role with some specific indications
such as small diverticula.
KEYWORDS
Cricopharyngeal dysphagia, cricopharyngeal diverticula,
Zenker diverticula, upper esophageal sphincter dysfunction,
manometry, videofluoroscopy, upper esophageal sphincter
myotomy, diverticulectomy, diverticulopexis, transoral
diverticulostomy
158 SECTION I Esophagus and Hernia
150
1 Velopharynx
50
A
0
Time (s)
Absent pharyngeal
Postswallow contraction
B
Time (s)
FIGURE 11.3 Lateral videofluoroscopic view and high-resolution manometry color plot of a liquid (10 mL water) swallow. The color panel
indicates the corresponding pressure values along the pharyngoesophageal segment. (A) An individual with a normal swallow. During
normal deglutition, three manometric regions can be identified along the pharyngoesophageal segment: the soft palate (velopharynx), the
tongue base (oropharynx and hypopharynx), and the upper esophageal sphincter (UES) (left). On the high-resolution manometry color
plot (right), the velopharyngeal closure (1), elevation of the larynx (2), initiation of the pharyngeal stripping wave (3), and relaxation of the
UES (4) can be directly visualized. (B) A patient with an oropharyngeal swallowing disorder. Videofluoroscopy shows residue in the
piriform sinuses and inadequate opening of the UES. Manometrically, the swallow is characterized by pharyngeal paresis (absent
pharyngeal peristalsis) and normal UES relaxation. The observed residue in the hypopharynx could be due to ineffective pharyngeal
function or due to inadequate UES relaxation. (From Rommel M, Hamdy S. Oropharyngeal dysphagia: manifestations and diagnosis.
Nat Rev Gastroenterol Hepatol. 2016;13:49–59, Fig. 2.)
35 mL
Participant 1 Participant 1 Participant 2
distension
FLIP
balloon
positioned
in UES 4 mm
UES
at rest 4.3 mm
13.9 mm
Dry
17.7 mm
swallow
Participant 2
FLIP balloon
positioned
in UES
6.2 mm
Head
turn
to rignt
4.2 mm
FIGURE 11.4 Still images from fluoroscopic studies of the functional lumen imaging probe (FLIP) in position in the upper esophageal
sphincter (UES) of two healthy volunteers (left). The volunteer turns the head in the bottom image. The FLIP can clearly be identified by
the electrode arrays. The right panel shows FLIP studies on these participants indicating the profiles at rest, with estimated minimum
diameters identified; the minimum diameter during dry swallows, clearly indicative of opening; and finally, profiles for both with minimum
diameters during right head turns, indicating a shift in either the probe or the narrow region. (From Lottrup C, Reggersene H, Liao D,
et al. Functional lumen imaging of the gastrointestinal tract. J Gastroenterol. 2015;50:1005–1016, Fig. 4.)
or gastric-refluxed material, a tracheostomy with a cuffed The short-term outcome is similar to other more invasive
tube is probably one of the most common procedures used procedures (e.g., cricopharyngeal myotomy) but requires
in such circumstances. This procedure can be performed repetition in about 50% of cases, and further treatment
with the patient under local anesthesia; it is reversible with surgical myotomy is eventually indicated in a consistent
and allows vocalization to be maintained if the cuff can percentage of patients.19,20
be deflated at times. Sometimes, however, more radical
and irreversible procedures may be used, such as injection Injection of Botulinum Toxin
of Teflon or other substances to medialize vocal cords, Injection of botulinum toxin (BoT) in the cricopharyngeal
laryngoplasty, subcricoid cricoidectomy, epiglottic sew- muscle (transcutaneously with electromyography [EMG]
down, or the narrow-field laryngectomy.16 guidance or endoscopically) has gained some popularity
Interventional procedures specifically aimed to treat after its introduction in 1994.21 BoT inhibits the tonic
dysphagia are directed toward the UES region. The current contraction of muscles by inhibiting the release of ace-
operative management of UES dysfunction is quite variable, tylcholine across the neuromuscular junction; therefore
and there are little data to support the relative superiority it will primarily benefit patients with hypertonicity of
in outcomes of one intervention over another.19 the UES and a retained ability to complete pharyngeal
bolus formation. Furthermore, it has distinct appeal in
Endoscopic Dilations patients who are not ideal candidates for surgery. A recent
Endoscopic dilations (with bougies or balloons) have review has been performed.22 Only two of the analyzed
been sometimes used, especially if an obvious organic or studies, however, reported on more than 20 patients,
functional cause (e.g., strictures or webs) is detectable. with the majority of articles being either case reports or
Dilations have the advantages of being simple, relatively limited series of less than 10 patients. The causes of CD in
safe, and typically performed under conscious sedation. these published series encompassed numerous diagnoses,
162 SECTION I Esophagus and Hernia
including neurologic diseases, multiple sclerosis, diabetic dysphagia are rare; other more severe and potentially
neuropathy, radiation treatment, and cerebrovascular lethal complications such as aspiration and pneumonia are
accidents. The dosages and administration techniques of related to the underlying pathology and the severity of the
BoT were also quite variable. The techniques for admin- swallowing impairment.25 The challenge in this category is
istration of BoT to the cricopharyngeal muscle included to identify which patients will benefit from cricopharyngeal
endoscopic injection under general anesthesia or mask myotomy. Intact voluntary oral-phase of deglutition with
ventilation, percutaneous injection with or without EMG good control of the tongue, good control of the laryngeal
guidance, and injection via flexible endoscopy. In general, aditus with normal phonation, and absence of dysarthria
the majority of patients reported improved swallowing are reliable prognostic factors for a successful outcome.26
function: approximately 75% in the combined analysis. Poor coordination in the swallowing mechanism may still
Complications were infrequent and included transient result after surgery in persistent silent aspiration and the
vocal fold paresis, temporary worsening of dysphagia, potential for pulmonary infection. Cumulative analysis of
neck cellulitis, and aspiration pneumonia. There were published data (mostly uncontrolled case series) showed
no reported deaths directly related to the procedure. an overall favorable response rate of 63% with an average
BoT injection may also be used as a test to determine mortality of 1.8%.15 Other reviews report a success rate
whether myotomy would be effective, although success of 78%, with a morbidity of 7%.19 At present, there are
with myotomy has also been reported in patients who no clear and accepted indications for cricopharyngeal
failed BoT treatment.23 myotomy in oropharyngeal dysphagia of neurologic or
myogenic origin, and final clinical decisions need to be
Cricopharyngeal Myotomy made on a case-by-case basis.
Cricopharyngeal myotomy has long been used for treating
patients with cricopharyngeal dysfunction of neurologic or
myogenic origin, with the aim to remove the obstructive
ZENKER DIVERTICULUM
effect of the UES below a pharynx unable to initiate an Pharyngoesophageal diverticula are protrusions of pharyn-
effective contraction to push the bolus into the cervical geal mucosa through a weak zone in the posterior wall of
esophagus. A brief description of the technique adopted the pharynx that are limited inferiorly by the upper border
in our institution follows.24 Under general anesthesia, the of the cricopharyngeal muscle and laterally by the oblique
patient is positioned supine on the operating table with fibers of the thyropharyngeal muscle, the so-called Killian
a small pillow under the shoulders and with the head triangle. These diverticula are named after the German
hyperextended and slightly turned on the right side. pathologist, Frederick Albert von Zenker who published
An incision is performed along the anterior border of a review of 27 patients with this disease.27 Though not
the left sternocleidomastoid muscle (Fig. 11.5A). After the first to describe this condition, Zenker is credited
dividing the subcutaneous tissue and the platysma, the for realizing that the pathogenesis of the diverticula lay
pharynx and esophagus are exposed by retracting the in an increased intrapharyngeal pressure, thus causing a
sternocleidomastoid muscle and vessels (carotid artery and “pulsion” diverticulum.
jugular vein) laterally and the laryngothyroid block medi-
ally (see Fig. 11.5B). Exposure of the region is facilitated PHYSIOLOGY AND PATHOPHYSIOLOGY OF
by dividing the omohyoid muscle and the middle thyroid ZENKER DIVERTICULUM
vein (see Fig. 11.5C). Dividing the inferior thyroid artery After recognition of the central role of the UES in causing
helps in further exposing and preventing the risk of damag- the diverticulum, different theories have been suggested
ing the recurrent laryngeal nerve. This approach is the during the years. An increased resting pressure of the UES,
same as described later for the cricopharyngeal myotomy a lack of complete relaxation (achalasia), or premature or
associated to the treatment of a ZD. The transverse fibers discoordinated UES relaxation with the incoming pharyn-
of the cricopharyngeal muscle are identified and the geal contraction have all been proposed.28 The development
myotomy is performed from its upper border and the of modern manometric techniques has shown, however,
lower pharyngeal wall to the cervical esophagus, over that the UES resting pressure is similar and even decreased
a length of at least 5 cm, using the scalpel and scissors in these patients compared with controls, that the UES
(see Fig. 11.5D). When the myotomy is complete, the appears to properly relax on manometry during deglutition
mucosa bulges free through the muscle edges (see Fig. and, finally, that pharyngosphincteral incoordination was
11.5E). A drain is left for 24 hours to avoid hematomas an inconsistent finding. The demonstration of the real
in the neck region, and after a hydrosoluble contrast role of the UES in the pathogenesis of the diverticulum
study, the patient is allowed to start a liquid and soft diet. was finally demonstrated in the early 1990s by means of
Complications of the procedure may be the development elegant concurrent manometric and videofluorographic
of hematomas of the cervical region that can jeopardize studies, and it was confirmed by histologic examinations.
the airway patency and require immediate surgical revision Cook et al.,29 using a sleeve catheter for manometry and
and perforations of the esophageal mucosa that can be simultaneous video-radiographic recordings, demonstrated
repaired during the operation (a muscle flap from the a significantly reduced radiologic sphincter opening despite
sternocleidomastoid muscle may be used to buttress the a manometrically “complete” relaxation due to the loss
repairing suture) or conservatively treated (with nil per of contact between the sphincter walls and recording
os [NPO; nothing by mouth] and parenteral support) if catheter. Furthermore, in these patients, they found a
discovered with the contrast control study. These surgical greater “intrabolus” pressure, demonstrating therefore that
complications in patients with neurologic or myogenic the pathophysiologic abnormality is indeed a diminished
Cricopharyngeal Dysfunction and Zenker Diverticulum CHAPTER 11 163
Cricoid cartilage
Prethyroid
muscles Omohyoid
muscle
Line of
incision Line of incision
Sternomastoid muscle
B
Thyroid gland
Inferior thyroid
artery covered
by deep
cervical fascia
MAYO
©1999
4 cm
MAYO
©1999
FIGURE 11.5 Surgical technique for the myotomy of the upper esophageal sphincter (cricomyotomy). (A) The patient is positioned supine
on the operating table with a small pillow under the shoulders and with the head hyperextended and slightly turned on the right side; the
incision is performed along the anterior border of the left sternocleidomastoid muscle. (B) The exposure of the pharyngoesophageal tract
is facilitated by dividing the omohyoid muscle. (C) The middle thyroid vein and the inferior thyroid artery are also divided. (D) The
transverse fibers of the cricopharyngeal muscle are identified and the myotomy is performed from its upper border and the lower
pharyngeal wall to the cervical esophagus over a length of at least 4 cm using the scalpel and scissors. (E) When the myotomy is
complete, the mucosa bulges free through the muscle edges. ([D and E] Copyright Mayo Clinic, 1999.)
164 SECTION I Esophagus and Hernia
40
6
Controls Control Amplitude : 3 g
Intrabolus pressure (mm Hg)
5
Zenker (dp/dt)max : 0.06 g/ms
30 4 Latent time : 12 ms Contract time : 123 ms
3 + 1/2 relax. time : 165 ms
Area : 836 gms
2
20 +
1
0
10 0 50 100 150 200 250 300 350 400 450 500
4
Amplitude :9g
Zenker (dp/dt)max
0 3 : 0.01 g/ms
0 10 20 Latent time : +38 ms
Contract time : 342 ms
Bolus volume (mL) 2
1/2 relax. time : 188 ms
Area : 444 gms
FIGURE 11.6 Relationship of swallow volume to intrabolus pressure 1 +
in patients who have lost compliance of the cricopharyngeal and +
cervical esophageal muscle. Higher pressure for increased 0
swallowed volume indicates that patients have sufficient 0 100 200 300 400 500 600 700 800 900 1000
pharyngeal muscle power to create an intrabolus pressure, and
improvement of the compliance with a myotomy of the FIGURE 11.8 Contractility pattern of the cricopharyngeal muscle in
cricopharyngeal and cervical esophageal muscle should result in a control specimen and Zenker diverticulum.
clinical improvement.
PRE POST taken from patients with ZD and controls. Lerut et al.30
Bolus showed a slower and weaker contraction curve with a lower
pressure amplitude, a longer time to peak twitch, and a much longer
half relaxation time in muscle fibers of ZD patients (Fig.
Pharyngeal 11.8). They also found histologic, electron microscopic,
contractions
and immunohistochemical alterations, prompting them to
suggest the hypothesis that both neurogenic and myogenic
Pharynx abnormalities are the underlying cause for dysfunction of
the UES in these patients. Further studies31,32 confirmed
these findings, by showing a significantly higher collagen
content, higher isodesmosine/desmosine and collagen/
UES
elastin ratios in the cricopharyngeal muscle and the cervical
esophagus muscularis propria in patients compared with
Esophagus controls (Fig. 11.9). These findings also highlighted the
1s 20 mm Hg
role of the proximal muscle fibers of the cervical esopha-
gus in the pathogenesis of the diverticulum, supporting
FIGURE 11.7 Pharyngoesophageal manometric tracing in a patient the need to also divide these fibers for 2 to 3 cm when
with Zenker diverticulum before and after diverticulectomy and performing a myotomy (see subsequent text).
myotomy. A nonrelaxing upper esophageal sphincter and a
prominent bolus pressure are evident in the preoperative SYMPTOMS AND DIAGNOSIS
recording. Myotomy increased the compliance of the The main symptom in these patients is dysphagia, present
pharyngoesophageal segment, with complete disappearance of in nearly all the cases. One can distinguish this intrinsic
the “shoulder” of the bolus pressure in the pharyngeal dysphagia from the extrinsic one, caused by the distention
contractions. UES, Upper esophageal sphincter. of the pouch by accumulated food that can further com-
press the esophageal lumen. Regurgitation of undigested
food particles (from previous meals), abnormal noise
UES opening with increased intrapharyngeal pressure that during swallowing, and halitosis are also common. The
probably accounts for the development of the diverticulum most frequent symptoms are listed in Table 11.1. Dysphagia
(Fig. 11.6). They postulated that degenerative changes may lead to weight loss, and penetration and inhalation
cause a lack of elasticity of the sphincter muscle preventing of food particles into the airway can cause coughing and
it from relaxing completely, and introduced the concept of recurrent pulmonary infections. This is particularly relevant
reduced compliance of the pharyngoesophageal segment since the majority of patients are elderly and often frail.
that can be seen manometrically by the appearance of a Other symptoms related to the upper gastrointestinal
“shoulder” before the onset of the pharyngeal contraction, tract are often present. In particular, a high frequency of
representing a higher pressure regimen in the pharynx reflux symptoms and esophagitis has been reported, with
(Fig. 11.7). pH-proven GERD in as high as 44% of patients.33 GERD
This theory has been subsequently endorsed by several has long been postulated as a chronic irritative element
histologic and contractility studies on biopsy specimens in causing damage to the cricopharyngeal muscle that
Cricopharyngeal Dysfunction and Zenker Diverticulum CHAPTER 11 165
FIGURE 11.9 Histologic examination of a cricopharyngeal muscle specimen of a control subject (left) and of a patient with Zenker
diverticula (right). A marked decrease in the amount of muscle fibers is well evident in the patient. Therefore the muscle:connective ratio
is lower in patients than in controls. (From Zaninotto G, Costantini M, Boccù C, et al. Functional and morphological study of the
cricopharyngeal muscle in patients with Zenker’s diverticulum. Br J Surg. 1996;83:1263–1267.)
MAYO
©1999
MAYO
©1999
4 cm
A B
FIGURE 11.11 Treatment of a Zenker diverticulum. (A) After the myotomy is completed, the diverticulum bulges free through the muscle
edges. (B) The diverticulum is then transected with a linear stapler if it is 2 cm or more in size. Smaller diverticula (1 cm or less) can be
safely left in place, since the simple myotomy is sufficient to relieve patients’ symptoms. Diverticula nearing 2 cm in size can be inverted
below the pharyngeal muscles and sutured to the muscle layer, thus performing a sort of intramural suspension. (Copyright Mayo Clinic,
1999.)
transverse fibers of the cricopharyngeal muscle are easy to Endoscopic Stapling Diverticulostomy
see below the neck of the diverticulum. The myotomy is Patients with diverticula larger than 2 cm can be considered
performed from this level to the cervical esophagus over for an endoscopic stapled technique. Smaller diverticula
a length of at least 5 cm, using the scalpel and scissors lead to an incomplete division of the dysfunctional
(Fig. 11.11A). When the myotomy is complete, the mucosa cricopharyngeal muscle and diverticulum or symptom
bulges free through the muscle edges. If the diverticulum recurrence. Under general anesthesia with orotracheal
is 3 cm or more in size, it is transected with a linear stapler intubation, the patient is placed supine on the operat-
(see Fig. 11.11B). Smaller diverticula are best upended ing table with a small pillow under the upper torso and
and sutured to the prevertebral fascia or the pharyngeal the head hyperextended. The surgeon sits behind the
muscles to avoid the potential for leakage from the staple patient’s head. A Weerda diverticuloscope is inserted
or suture line when resecting the pouch.33 Very small in the hypopharynx, positioning its anterior blade in
(<1 cm) pouches can be safely left in place, since the the esophageal lumen and the posterior blade in the
myotomy alone suffices to reduce the pouch and alleviate diverticulum. A 5-mm diameter telescope is passed through
the symptoms. the scope. After visualization of the septum between the
Cricopharyngeal Dysfunction and Zenker Diverticulum CHAPTER 11 167
and cut), prior to dissection, two endoclips are placed on far, there have been no prospective trials comparing the
either side of the septum. A metal endoclip may also be different endoscopic treatment options with surgery.
placed at the apex of the incision to prevent tearing of Information comes from retrospective series or prospec-
the section and microperforation.45 tively recorded case series, using one or more of these
The aim of the procedure is to completely obliterate techniques. Moreover, given the rarity of the disease
the septum, thus achieving a wide opening between the and its prevalence in elderly patients, often with severe
diverticulum and the esophagus (diverticulostomy). Only comorbidities, it is unlike that randomized studies can
one session is usually needed for small diverticula (≤2 cm), be performed in the near future.
whereas repeated sessions may become necessary if they are We reviewed our experience a few years ago, reporting
larger. At any one session, a 1.5- to 2.0-cm incision is per- on 51 patients treated with endoscopic stapling diverticulos-
formed, and it is repeated in 1-weeks’ time. A nasogastric tomy, and 77 patients were treated with surgical myotomy.41
tube is usually placed for nutritional purposes for 2 days. After the operation, there was a statistically significant
improvement in the symptom score in both groups. In
RESULTS AND DISCUSSION the endoscopic group, however, 11 patients (21.5%) still
All authors agree that a fundamental step in therapy for complained of severe dysphagia and were considered as
ZD lies in dividing the cricopharyngeus muscle fibers (i.e., procedure failures, requiring further endoscopic treatment
the UES). As obvious as it may seem today, this notion is a (8 cases) or surgical myotomy (3 cases). On the other hand,
recent acquisition that is based empirically on the finding only 4 patients (5.2%) of the surgical group had recurrent
of much lower complication (i.e., leakage) and recurrence dysphagia (P < .05), successfully treated with pneumatic dila-
rates when myotomy is added to the diverticulectomy.35 As tion in 3 (1 patient refused further treatment). A posterior
said, definitive support for this belief came from studies29 pouch was still evident (Fig. 11.15) in all patients treated
which demonstrated that the muscle pathology of the UES endoscopically and studied with postoperative barium
with inflammation and fibrosis restricts the opening of the swallow; however, most of the patients were asymptomatic.
UES, leading to a higher pressure of the bolus arriving By further dividing the two groups of patients following the
in the hypopharynx and ultimately to the formation of a size of the diverticulum (≤3 cm and >3 cm), we found that
pulsion diverticulum. Surgical cricopharyngeal myotomy, 64% and 92% of asymptomatic patients were in the endo-
alone in the case of small (<2 cm) diverticula, or combined scopic group compared with 94.5% and 96%, respectively,
with either diverticulectomy or diverticulum suspension in the surgical group. Therefore patients with a greater
(diverticulopexy) ensures symptom relief in nearly all than 3-cm diverticulum may have the same probability of
treated patients (Table 11.2; Fig. 11.14). The related a good outcome with the endostapling procedure as the
morbidity may involve some local hematomas and recurrent patients undergoing open surgery. Patients with smaller
nerve palsy in addition to leakages, which occur in about diverticula (≤3 cm) should be offered the open surgical
2% of cases.35 It should be emphasized, however, that the approach or a flexible endoscopic approach whereby the
mortality rate of this procedure in frail, elderly patients, entire cricopharyngeal muscle is divided using peroral
is far from negligible, mostly due to cardiopulmonary myotomy techniques that continue the myotomy distal to
complications. the extent of the diverticulum.
The risks of complications and even death cannot be Similar results are reported by the Brussels group,48
completely overcome even with endoscopic techniques, which first proposed endoscopic stapling for ZD (Table
but they are likely lower than with open surgery. Thus 11.3). They retrospectively compared their experience
RESULTS (%)
Complications Mortality Recurrence
Author, Year Time Period Method N (%) (%) Good Partial (%)
Payne, 1992 1944–78 D, DM, M 888 7.9 2 82 11 3.6
D 184 21 1.5 94 4.9
DM 121 10 4.9
GEEMO, 1995 1960–82 PM 55 12.7 1.8
(n = 390) M 26 0 NA
P 4 0 7.6
Bonafede, 1997 1976–93 M, DM, PM 87 24 3.5 78 13 NA
Zbaren, 1999 1987–97 D, DM 66 15 1.5 77 11 6
Feussner, 1999 1982–98 PM, DM 140 4.2 1 >90 0.8
Leporrier, 2001 1988–98 DM, PM 40 17.5 0 92 8 0
Jougon, 2003 1987–2000 DM 73 4 0 99 1 0
Colombo, 2003 1985–95 D, DM 79 15 0 76 19 2.5
Lerut, 2008 1975–2003 PM, M, DM 289 8.5 0 94.2 3.8 0.03
Total 2119 10.5 1.4 3.5
D, Diverticulectomy; GEEMO, Group Européen d’Etude des Maladies de l’Oesophage; M, myotomy; NA, not announced; P, diverticulopexy.
Modified from Lerut T, Coosemans W, Decaluwé H, et al. Pathophysiology and treatment of Zenker diverticulum. In: Yeo CJ, et al., eds. Shackelford’s
Surgery of the Alimentary Tract. 7th ed. Philadelphia: Saunders; 2013:336–348.
Cricopharyngeal Dysfunction and Zenker Diverticulum CHAPTER 11 169
A B
FIGURE 11.14 (A) Radiologic image of a pharyngoesophageal (Zenker) diverticulum. (B) The same patient, after myotomy of the upper
esophageal sphincter and diverticulectomy. (From Costantini M, Zaninotto G, Rizzetto C, Narne S, Ancona E. Esophageal diverticula.
Best Pract Res Clin Gastroenterol. 2004;18:3–17, Fig. 1.)
with endoscopic treatment against traditional myotomy. at the follow-up (compared with 97% of the patients who
Patients treated endoscopically had shorter hospital stays had open surgery). Furthermore, only 57% of patients
and fasting periods after the operation; they also had with a diverticulum less than 3 cm were satisfied with the
fewer complications (though two cervical abscesses and treatment, whereas this was true of 98% of patients treated
mediastinitis were recorded). Symptom outcome was less with open surgery (P < .05). Also, Bonavina et al.49 in their
favorable than with open surgery, however, because only series of 100 patients treated with endostapler reported a
75% of patients treated endoscopically were symptom free success rate of 88.4% in patients with greater than 3 cm
170 SECTION I Esophagus and Hernia
TABLE 11.3 Outcome of Transoral Rigid Procedures With Stapler for Zenker Diverticulum
RESULTS (%)
Complications Mortality Recurrence
Author Year Time Period N (%) (%) Good Partial (%)
Peracchia 1998 1992–96 95 0 0 92.2 7.8 5.4
Van Eeden 1999 1996–97 18 5.9 0 53 35 NA
Cook 2000 1995–99 74 5 0 71 24 8.7
Luscher 2000 1997–98 23 4.3 0 76 14 4.3
Philippsen 2000 1996–99 14 0 0 57 21 NA
Sood 2000 1992–99 44 4.5 1 70 24 9
Jaramillo 2001 1996–99 32 3.7 0 80 7.4
Stoeckli 2002 1997–2000 30 27 0 96 NA
Counter 2002 1993–97 31 9.7 0 50 44 22
Raut 2002 1994–98 25 8 0 48 32
Chang 2003 1995–2001 150 12.7 0 73.3 22 11.8
Chiari 2003 1997–2001 39 10 0 71 20 10.9
Wasserzug 2010 1997–2001 55 4 0 90 10
Bonavina 2015 2001–13 100 2 0 84 5.4 24
Total 730 7.8 0.02 48–92 10.9
NA, Not announced.
Modified from Lerut T, Coosemans W, Decaluwé H, et al. Pathophysiology and treatment of Zenker diverticulum. In: Yeo CJ, et al., eds. Shackelford’s
Surgery of the Alimentary Tract. 7th ed. Philadelphia: Saunders; 2013:336–348.
TABLE 11.4 Retrospective Studies Comparing Surgical Myotomy and Endoscopic Treatment for Zenker Diverticula
diverticula, as compared with 54.9% in those with smaller, Finally, a systematic review on available literature (28
less than 3 cm diverticula (P < .05). comparative studies and 43 cohort studies), analyzing
Two recent papers reported the outcome of endoscopic results of surgical and the various endoscopic techniques,
and surgical treatment in Zenker patients.50,51 Even with has been recently published.52
all the drawbacks of being retrospective analyses, all these The rate of failure was significantly higher with
papers agree in reporting a significantly better outcome endoscopic techniques compared with external surgical
with traditional surgery (Table 11.4). Furthermore, Lerut approaches (18.4% vs. 4.2%; P < .001). Complications
reports on a prospective randomized study comparing presented a different pattern with the various surgical
endoscopic stapling and open surgery that was initiated approaches, since mediastinitis (1.2% vs. 0.3%; P < .01) and
but terminated after 20 cases (9 surgical, 11 endoscopic) emphysema (3.0% vs. 0.1%; P < .01) occurred significantly
because of a higher number of complications and modest more often with endoscopic treatment, compared with
results in the endoscopic group.33 fistula (3.7% vs. 1.2%; P < .01), recurrent nerve palsy (3.4%
No differences in the failure and complications rate vs. 0.3%; P < .001), and hematoma (2.2% vs. 0.6%; P < .01)
were observed when diverticulostomy was performed with with transcervical treatment. Surgery-related deaths were
a rigid endoscope, and the septum was transected with infrequent in both groups (0.9% for the open approach
cautery or CO2 laser, although two deaths were reported vs. 0.4% for endoscopic techniques). Overall postoperative
(Table 11.5), or with a flexible endoscope compared with complications tended to occur more frequently after
the transoral stapled technique (Table 11.6). transcervical approach (7% vs. 11%).
Cricopharyngeal Dysfunction and Zenker Diverticulum CHAPTER 11 171
TABLE 11.5 Outcome of Transoral Rigid Procedures With Cautery or Laser for Zenker Diverticulum
RESULTS (%)
Complications Mortality Recurrence
Author Year Time Period Method N (%) (%) Good Partial (%)
Van Overbeek 1994 1964–92 Cautery/CO2L 545 6.7 1 90.6 8.6 NA
Ishioka 1995 1982–92 Cautery 42 4.8 0 92.9 7.1 7.1
Von Doersten 1997 1985–94 Cautery 40 25 0 92.5 0
Hashiba 1999 >1978 Cautery 47 14.9 0 96 4.3
Lippert 1999 1984–96 CO2L 60 10 0 73 21 10
Nyrop 2000 1989–99 CO2L 61 13.3 0 70 22 13
Mattinger 2002 1974–98 CO2L 52 13.5 1 84.6 15.4
Krespi 2002 1989–2001 CO2L 83 4.8 0 85.5 11 7.5
Total 930 8.7 0.02 70–96 7.2
CO2L, Carbon dioxide laser; NA, not announced.
Data from Lerut T, Coosemans W, Decaluwé H, et al. Pathophysiology and treatment of Zenker diverticulum. In: Yeo CJ, et al., eds. Shackelford’s Surgery
of the Alimentary Tract. 7th ed. Philadelphia: Saunders; 2013:336–348.
It is difficult to draw final conclusions from these studies. morbidity, which is higher than with endoscopy mainly
Endoscopic treatment is very attractive because it is less due to leakage from the suture line. Although this
invasive and has a lower complication rate, although, when does not normally require further surgery and heals
compared with surgery, it is sometimes less effective in spontaneously (with nasogastric suction, NPO, and
relieving dysphagia completely. Endoscopic stapling also antibiotic therapy), it is nonetheless a potentially severe
has other drawbacks, mainly related to the size of the complication in patients with concurrent respiratory or
diverticulum: in the case of a small diverticulum (≤2 cm), heart disease.
the stapler anvil is too long to be properly accommodated
inside the pouch, and the cricopharyngeal fibers cannot
be transected completely. In this sense, diverticulostomy
CONCLUSION
with the laser or cautery may be more effective. On the ZD can be effectively treated with either endoscopic
other hand, very large diverticula (>5 cm) plunging into diverticulostomy or open surgery. Both approaches have
the mediastinum carry the risk of vascular lesions if they advantages and disadvantages. An individual approach to
are transected blindly. Myotomy of the UES is probably ZD should therefore be recommended: High-risk patients
more effectively achieved with open surgery, when the with medium-size diverticula are probably better served by
muscle fibers are cut under direct vision and the edges of diverticulostomy; open surgery should be recommended
the myotomy are further separated by blunt dissection, for small (<2 cm) or giant ZD or in patients with a low
leaving the submucosa widely exposed. Furthermore, surgical risk. New peroral flexible endoscopic techniques
the inferior pharyngeal constrictor muscle layers offer promise for patients with any size diverticulum but
of the proximal cervical esophagus may be easily divided. will probably be best for small to medium (0 to 5) cm
The major drawback of open surgery is the related diverticula.
172 SECTION I Esophagus and Hernia
D
iverticular diseases of the esophagus consist of esophageal motility disorder.2,6,7 Named motility disorders
variations of outpouchings of one or more layers associated with these diverticula include achalasia, diffuse
of the gut wall that are epithelial lined. These esophageal spasm, nutcracker esophagus, and hyperten-
outpouchings can be found along the entire length of sive lower esophageal sphincter (LES), with the most
the esophagus. They are described by their location along common being achalasia followed by diffuse esophageal
the esophagus: pharyngoesophageal, mid-esophagus, and spasm.2,14,15 One theory suggests that the presence of
epiphrenic. Often these diverticula are asymptomatic, dysmotility causes an uncoordinated contraction between
but when they are symptomatic, they create a significant the distal esophagus and lower esophageal sphincter
constellation of symptoms that reduce the patient’s quality leading to increased intraluminal pressure and subsequent
of life and may lead to life-threatening complications such herniation through a weakened area of the esophagus.16
as aspiration pneumonia. Because these occur often in Another study identified that esophageal diverticulum is
the elderly or patients with comorbidities, careful surgical associated with areas of low peristaltic pressure amplitude,
evaluation needs to be completed before treatment is bizarre peristaltic wave forms, and hypertensive peristaltic
rendered to ensure a good outcome. This chapter focuses pressures.17 Other reported etiologies such as a distal
on the presentation and treatment of mid- and distal stricture, prior fundoplication, and hiatal hernia all act
esophageal diverticula. to create the same outflow pressure dynamics as a motility
disorder.
ABSTRACT
Mid- and distal esophageal diverticula are uncommon
diseases that often present without symptoms. However,
some patients will have incapacitating or life-threatening
symptoms from either the diverticulum, the underlying
motility disorder, or both. After identification by barium
swallow, careful evaluation of the patient followed by
upper endoscopy, high-resolution manometry, and the
occasional CT scan of the chest will provide the key
information to decide upon surgical treatment. Similar to
many esophageal procedures, treatment of these diverticula
has moved toward minimally invasive surgery to perform
the diverticulectomy, distal esophageal myotomy, and, in
most instances, a partial fundoplication. Diverticula near
the inferior pulmonary vein may be better addressed with
a combined video-assisted thoracic surgery (VATS) and
laparoscopy. For most patients, this results in relief of
the presenting symptoms and a low rate of mortality but
a not insignificant rate of morbidity. The staple line leak
remains the key complication that surgeons hope to avoid.
KEYWORDS
Epiphrenic diverticulum; mid-esophageal diverticulum;
myotomy; traction diverticulum; pulsion diverticulum;
fundoplication; motility disorders; achalasia
174 SECTION I Esophagus and Hernia
A B
FIGURE 12.2 (A) Barium swallow demonstrating a right-side diverticulum. (B) Barium swallow demonstrating a left-side diverticulum.
Heart
MAYO
©1997
Diverticulum
Lung
Diaphragm
MAYO
Esophagus ©1997
A Aorta
Aortic arch
MAYO
©1999 Myotomized esophagus
C
FIGURE 12.5 Surgical management of an epiphrenic esophageal diverticulum. A left posterolateral thoracotomy incision is shown in the
inset. Exposure of the diverticulum is obtained when the chest is entered through the bed of the eighth rib. Note that the esophagus has
been delivered from its mediastinal bed, tape has been passed around the esophagus, and the esophagus has been rotated to bring the
diverticulum into view. The neck of the diverticulum has been dissected to identify the defect in the esophageal muscular wall (A). A TA
stapling device is used to transect and close the diverticulum followed by closure of the esophageal musculature over a mucosal suture
line (B). The site of the diverticular incision has been rotated back to the right and is not visible. A long esophagomyotomy extending
from the esophagogastric junction to the aortic arch has been performed. The musculature of the esophagus has been freed from
approximately 50% of the circumference of the esophageal mucosal tube to allow the mucosa to bulge through the muscular incision (C).
(Copyright Mayo Clinic, 1999.)
aspect of diverticulectomy and extended onto the stomach control,22 and in this circumstance, most opt to create a
for 2 cm. Some surgeons will also extend the myotomy Belsey Mark IV fundoplication.23
proximally for a centimeter or two, whereas other surgeons
will extend the myotomy up to the aortic arch as shown VIDEO-ASSISTED THORACIC APPROACH ±
in Fig. 12.5. However, the value of this proximal exten- LAPAROSCOPIC MYOTOMY/FUNDOPLICATION
sion is unclear, since the high-pressure zone is distal to After placement of a double-lumen endotracheal tube,
the diverticulum. Many surgeons favor the addition of a the patient is placed in the left lateral decubitus position
partial antireflux repair to provide an element of reflux with the bed flexed at the top of the iliac crest. A total
Surgical Management of Mid- and Distal Esophageal Diverticula CHAPTER 12 177
of four ports are placed (seventh intercostal space [ICS] The challenge in using a VATS-only approach is per-
posterior axillary line for surgeon’s left hand, stapler, forming a distal esophageal myotomy. Since the majority
ninth ICS in the line of the scapular tip for the camera, of diverticula are found in the right chest and access
fourth ICS posterior axillary line for retraction and suction- to the proximal stomach is limited from this approach,
ing, and seventh ICS just inferior and posterior to the several options have been proposed. First, it has been
scapular tip for the surgeon’s right hand) (Fig. 12.6).The suggested to perform pneumatic dilation of the lower
diverticulum is identified and the muscular layers of the esophageal sphincter prior to diverticulectomy in patients
esophageal wall split to identify the mucosa at the inferior with a documented motility disorder.24 It is likely this will
and superior aspect of the diverticulum. Once the neck need to be repeated to achieve a similar result to surgical
of the diverticulum (Fig. 12.7A) is exposed and dissected myotomy based on a recent trial.25 Second, the patient can
free, the stapler is then applied over the neck of the be repositioned supine or low lithotomy and a laparoscopic
diverticulum with either a bougie or upper endoscope in myotomy with or without partial fundoplication can be
situ and resected with an endoscopic stapler (Fig. 12.7B). performed. To ensure the proper extent of the myotomy,
The muscular layers are closed with interrupted sutures the distal end of the diverticulum is marked with a clip
and occasionally reinforced with pleura. A 24-French chest on the anterior surface of the esophageal wall at the
tube and/or a 24-French drain is placed adjacent to the completion of the VATS portion. The clip can then be
esophagus prior to closure. identified at laparoscopy and the esophagomyotomy
performed from the clip on the left side of the esophagus
onto the gastric cardia.8 Lastly, if the diverticulum is left
sided, the myotomy can be performed similar to the
approach described by Pellegrini et al. though the relief
of dysphagia and ability to create a fundoplication favors
adding a laparoscopic myotomy.26,27
LAPAROSCOPIC APPROACH
A laparoscopic transhiatal approach is becoming
increasingly common.6,9,16,28–30 Most surgeons position
4th ICS
Assistant
the patient as they would for a modified Heller-Dor or
port Nissen fundoplication procedure. We favor the patient in
low lithotomy with placement of five laparoscopic ports.31
7th ICS The initial exposure is to dissect and mobilize the entire
esophageal hiatus for maximal exposure. The vagal nerves
7th ICS are identified as dissection is performed cranially up the
esophagus until the distal aspect of the diverticulum is
identified. Once identified, the dissection proceeds anterior
and posterior along the diverticulum until circumferential
9th ICS - Camera
dissection is complete. The distally identified vagus nerves
are then traced toward the diverticulum; on occasion it is
necessary to separate a nerve from the diverticulum. The
neck of the diverticulum is then identified by separating
FIGURE 12.6 Example of video-assisted thoracic surgery ports. any remaining muscle fibers and then dissecting down
A B
FIGURE 12.7 (A) Operative video-assisted thoracic surgery (VATS) photo showing the diverticulum and the narrow aspect of the neck or
waist. (B) Operative VATS photo showing application of the stapler in line with the esophagus at the level of the neck of the diverticulum.
178 SECTION I Esophagus and Hernia
FIGURE 12.8 (A) Stapled resection of an epiphrenic diverticulum through a laparoscopic approach. (B) Operative laparoscopic photo of
stapler across the neck of the diverticulum. Orange vessel loop around the vagus nerve.
to expose the mucosa. Similar to the open and VATS about the development of gastroesophageal reflux. At
approaches, the neck of the diverticulum is exposed, the present time, an endoscopic approach should be
and an endoscopic stapler is applied over the neck of the considered experimental and while the distal myotomy
diverticulum with either a bougie or upper endoscope in can be performed successfully, endoscopic management of
situ (Fig. 12.8). The staple line is then imbricated using the diverticulum requires further thought and innovation
sutures. to minimize the risk of gastroesophageal reflux and the
The esophageal myotomy is performed along the left potential stricture from mucosal sloughing.
anterior wall of the esophagus just to the left of midline.
The myotomy is extended inferiorly from at least the CURRENT CONTROVERSIES IN
inferior aspect of the diverticulum through the lower SURGICAL APPROACH
esophageal sphincter and extended 2 cm onto the proximal Surgeons caring for patients with an epiphrenic diverticu-
stomach. The hiatus is then reapproximated posteriorly, lum and who wish to approach these minimally invasively
and either a Toupet fundoplication is created posteriorly should be sufficiently skilled in both laparoscopy and
or a Dor fundoplication is created anteriorly (Fig. 12.9). thoracoscopy, because even though the diverticulum is
commonly just above the gastroesophageal junction, it
ENDOSCOPIC APPROACH can be located anywhere in the distal esophagus with the
With the advent of endoscopic surgical techniques such cephalad edge located above the inferior pulmonary veins.
as endoscopic tunneling and endoscopic myotomy,32,33 it is During laparoscopic hiatal hernia repair, access to the
possible to imagine that an epiphrenic diverticulum may level of the inferior pulmonary veins can consistently be
be approached in this fashion. Two recent abstracts report achieved but rarely above this point. In the patient with a
on the use of these techniques to treat an epiphrenic normal hiatus access, visualization into the mediastinum
diverticulum.34,35 In both cases, a submucosal tunnel is can be limited. It is likely then that diverticula that are
created to facilitate a distal esophageal myotomy using greater than 5 cm from the GEJ will be inadequately
the same technique used to perform per oral endoscopic addressed with a transhiatal dissection, with a higher
myotomy for achalasia. However, treatment of the diver- propensity for incomplete resection or a staple line leak
ticulum differs. In one case, the diverticulum is inverted at the superior-most aspect of the diverticulectomy. This
into the lumen, and an endoscopic snare placed around is supported by a recent report by Allaix et al.36 in which
the neck of the diverticulum. The mucosa eventually 13 patients with achalasia and epiphrenic diverticulum
sloughs and the defect is healed over time. In the second underwent attempted resection. The diverticulum was
case, a channel is created between the diverticulum and excised in six patients but left in situ in seven patients.
the gastric body by means of a transdiverticulum-to-gastric In four of the seven patients, the diverticulum was left
puncture and subsequent dilation of the channel and in place because it was too far from the GEJ and could
placement of an endoscopic stent. The authors cautioned not be safely dissected laparoscopically. Postoperatively,
Surgical Management of Mid- and Distal Esophageal Diverticula CHAPTER 12 179
A B
FIGURE 12.9 (A) Heller myotomy performed on the opposite esophageal wall of the stapled line and extending for approximately 2 cm on
the gastric side. (B) A Dor fundoplication is constructed by suturing the anterior fundic wall to the edges of the myotomy.
patient symptoms were controlled as measured by the after diverticulectomy alone; myotomy creates the potential
Eckardt score, but with only 2 years of follow-up. This for GERD, which requires a fundoplication and/or the
contrasts with other reports recommending the addition need for proton pump inhibitor medication, and leaks
of a VATS in these high diverticula.8,30 occur regardless of the presence of a myotomy. There has
Regardless of symptoms, it is unclear what the impact been no clear resolution about the use of a myotomy, but
is of leaving the diverticulum in situ with its long-term the majority of cases reported in the literature include
potential for bleeding, impaction, and stasis with regurgita- a myotomy, and it is suggested to include a myotomy as
tion. Until further data are reported, it is advisable except part of treatment until additional data supporting its
in extenuating circumstances or a small diverticulum that exclusion is provided.
disappears after myotomy to resect the diverticulum. In
patients with high diverticula, a combined VATS and COMPLICATIONS
laparoscopic approach may be a more suitable and effica- Regardless of the surgical approach, the operative risks
cious option. One strategy is to assess the location of the remain significant in a patient population that is often
diverticulum based on the location of the upper border malnourished, elderly, has significant comorbidities,
of the diverticulum in relation to the endoscopically and potentially long-standing pulmonary soilage from
identified GEJ. For diverticula less than 5 cm above the aspiration. A wide variety of postoperative complications
GEJ, a laparoscopic transhiatal approach is likely to be have been reported but complications specific to this
successful, whereas for those greater than 5 cm above the operation include staple line leak, incomplete myotomy,
GEJ or above the inferior pulmonary vein, a combined vagal nerve injury (manifested by delayed gastric empty-
thoracoscopic-laparoscopic minimally invasive approach ing), and pleural effusion. Broader complications that are
is likely necessary.8 more commonly associated with open resection have also
The need for an esophagomyotomy is accepted by been reported and include intraoperative hemorrhage,
most esophageal surgeons based on the fact that most pulmonary complications (acute respiratory distress syn-
diverticula are associated with an underlying motility drome, pneumonia), and cardiac events (atrial fibrillation,
disorder, and theoretically, a distal obstruction or high infarction).
pressure zone potentially increases the risk for staple line The most significant complication is a staple line
dehiscence and subsequent leakage.13,21 However, selective leak (Fig. 12.10). These are best avoided by careful and
use of a myotomy in patients without a motility disorder meticulous dissection, reapproximation of the esophageal
has been proposed based on a small series of patients.37 muscle, and complete myotomy.9 When a staple line leaks
The rationale for abstaining from a myotomy is that occurs, treatment is individualized for the patient and
patients with a normal LES will resolve their symptoms the extent of leak. Patients are made nil per os (NPO,
180 SECTION I Esophagus and Hernia
or nothing by mouth), antibiotics to cover gram-negative One recent alternative is to consider placement of an
enteric organisms are begun, and alternative forms of endoluminal wound vac sponge to manage the staple line
nutrition are tailored to the specific patient and situation. disruption.38 However, if less invasive endoscopic measures
Upper endoscopy is crucial in determining early manage- prove unsuccessful in mitigating leakage, return to the
ment. Whenever feasible, we employ endoscopic stenting, operating room is always available for wide drainage,
clips, or suturing to control leakage, as these procedures control of contamination, and diversion (when necessary).
are far less morbid than traditional open interventions.
OUTCOMES OF DIVERTICULECTOMY AND MYOTOMY
The surgical treatment results for epiphrenic diverticula
in series reporting at least 10 cases are tabulated in Tables
12.1 and 12.2. Regardless of the approach, transthoracic
or minimally invasive, the reported outcomes show good
to excellent relief of symptoms in the majority of patients.
The operative mortality rate has decreased with transition
to a minimally invasive approach, but the rates of morbid-
ity remain significant, likely reflecting the underlying
age and comorbidities that are carried by patients with
this disease. Lastly, the postoperative leak rate remains
the Achilles heel of any operation on the esophageal
mucosa though it appears the prompt treatment has been
able to successfully manage the leak without mortality or
compromise of long-term outcomes.
At present, epiphrenic diverticula remain a rare disease
with most surgeons reporting on 20 to 30 cases over
decades of care. Over the last two decades, there has been
a slow but steady transition away from a transthoracic
approach to the use of minimally invasive techniques that
maintain the surgical principles outlined for the treatment
FIGURE 12.10 Computed tomography scan demonstrating a staple of epiphrenic diverticula. The cumulative experience
line leak with a large abscess cavity in the right chest. to date suggests that a laparoscopic approach is quickly
Mortality Morbidity
Author Years N Approach Operation (%) (%) Leak Outcomes
Allen 1944–53 24 RT D 4 25 10/24 persistent symptoms
et al.21 1954–63 17 LT D/M 6 18 15/27 good to excellent relief of symptoms
Streitz 1960–90 18 LT D (8) 0 33 6 Median follow-up = 7 years
et al.37 D/M (10) • 13/18 good to excellent relief of
symptoms
• Neurogenic
Fékéte 1969–89 27 RT/LT D (10) 11 19 7
et al.43 Lap D/M/F (10)
M/F (4)
Altorki 1970–90 17 LT D/M/F (14) 6 Median follow-up = 7 years
et al.44 • 13/17 symptom free
• 1/17 poor result
Benacci 1975–91 33 LT D (7) 9.1 33 18 Median follow-up = 7 years
et al.11 D/M (22) • 67% good to excellent outcomes
M (1) • 33% poor to fair outcomes
Resection (3)
Castrucci 1983–95 27 LT D (5) 7 11 4 Mean follow-up = 47 months
et al.45 D/M (5) • 92% satisfied
D/M/F (17) • Visick I and II
Nehra 1987–96 21 LT D/M/F 5 5 Median follow-up = 24 months
et al.13 • 88% good to excellent outcomes
Varghese 1976–2005 35 LT D/M/F (33) 2.8 9 6 Median follow-up = 33 months
et al.7 • 76% excellent resolution of symptoms
• 21 mild dysphagia
D’Journo 1977–2008 23 LT D/M/F 0 9 Median follow-up = 61 months
et al.46 • Significant symptom improvement
D, Diverticulectomy; F, fundoplication; Lap, Laparoscopic transhiatal; LT, left thoracotomy; M, myotomy; RT, right thoracotomy.
Surgical Management of Mid- and Distal Esophageal Diverticula CHAPTER 12 181
SUMMARY
Mid- and distal esophageal diverticula are uncommon
diseases that often present without symptoms. However,
some patients will have incapacitating or life-threatening
symptoms from either the diverticulum, the underlying
motility disorder, or both. After identification by barium
swallow, careful evaluation of the patient followed by
FIGURE 12.12 Bronchoscopy photo of methylene blue in the right
upper endoscopy, high-resolution manometry, and the
upper lobe that was injected from the esophageal side of a
diverticulum.
occasional CT scan of the chest will provide the key
information to decide upon surgical treatment. Similar to
many esophageal procedures, treatment of these diverticula
inflammation in the region can lead to the development has moved toward minimally invasive surgery to perform
of a fistula between the diverticulum and the airway41 and the diverticulectomy, distal esophageal myotomy, and in
bleeding secondary to erosion into a bronchial artery, most instances a partial fundoplication. Diverticula near
small esophageal vessel, or even major vessel.42 Bleeding the inferior pulmonary vein may be better addressed with
is also caused by the friable granulation tissue found at a combined VATS and laparoscopy. For most patients, this
the tip of the diverticulum. results in relief of the presenting symptoms and a low rate
The diagnostic approach to the mid-esophageal diver- of mortality but a not insignificant rate of morbidity. The
ticulum involves the same investigations that were used staple line leak remains the key complication surgeons
for an epiphrenic diverticulum. However, the initial test hope to avoid.
that identifies the diverticulum is more likely to be a
computed tomography scan of the chest during evaluation
for mediastinal adenopathy or for chronic cough due to
REFERENCES
a small fistula. A contrast radiographic study should be 1. Mondiere J. Notes sur quelques maladies de l’oesophage. Arch Gen
Med. 1833;3:28-65.
included in the evaluation. In the absence of obvious 2. Soares R, Herbella FA, Prachand VN, Ferguson MK, Patti MG.
mediastinal pathology being the causative agent, high- Epiphrenic diverticulum of the esophagus. From pathophysiology
resolution manometry should be undertaken given that to treatment. J Gastrointest Surg. 2010;14(12):2009-2015.
many of these patients will have an underlying disorder that 3. Bruggeman LL, Seaman WB. Epiphrenic diverticula. An analysis
may impact surgical decision making. During endoscopic of 80 cases. Am J Roentgenol Rad Ther Nucl Med. 1973;119(2):266-
276.
evaluation, the option of bronchoscopy should be available, 4. Postlethwait RW. Diverticula of the esophagus. In: Postlethwait RW,
particularly if there is a suspicion for a fistula. This can ed. Surgery of the Esophagus. 2nd ed. Norwalk, CT: Appleton-Century-
often be difficult to ascertain, and an injection of dilute Crofts; 1986:129.
methylene blue into the esophagus may be visualized in 5. Vinson PP. Diverticula of the thoracic portion of the esophagus.
Arch Otolaryngol Head Neck Surg. 1934;19:508-513.
the proximal airways (Fig. 12.12). 6. Macke RA, Luketich JD, Pennathur A, et al. Thoracic esophageal
diverticula: a 15-year experience of minimally invasive surgical
TREATMENT management. Ann Thorac Surg. 2015;100(5):1795-1802.
The indications for surgical treatment are related to 7. Varghese TK, Marshall B, Chang AC, Pickens A, Lau CL, Orringer MB.
the presence of symptoms from the diverticulum. This Surgical treatment of epiphrenic diverticula: a 30-year experience.
Ann Thorac Surg. 2007;84(6):1801-1809.
is best approached with a right thoracotomy through 8. Achim V, Aye RW, Farivar AS, Vallières E, Louie BE. A combined
the fifth intercostal space given the ease of access to the thoracoscopic and laparoscopic approach for high epiphrenic
carina, mediastinal nodes, and esophagus. The surgeon diverticula and the importance of complete myotomy. Surg Endosc.
should expect extreme scarring and distorted anatomy 2017;31(2):788-794.
9. Bowman TA, Sadowitz BD, Ross SB, Boland A, Luberice K,
due to the ongoing inflammation. The first step will be Rosemurgy AS. Heller myotomy with esophageal diverticulectomy:
to separate the esophagus and diverticulum from the an operation in need of improvement. Surg Endosc. 2016;30(8):3279-
adjoining mediastinal nodes. Once this is accomplished, 3288.
any fistulous tract should be divided and controlled. The 10. Debas HT, Payne WS, Cameron AJ, Carlson HC. Physiopathology of
diverticulum should be isolated, and the mucosa should lower esophageal diverticulum and its implications for treatment.
Surg Gynecol Obstet. 1980;151(5):593-600.
be evaluated and repaired or resected depending on the 11. Benacci JC, Deschamps C, Trastek VF, Allen MS, Daly RC, Pairolero
degree of damage. The overlying muscle layers should be PC. Epiphrenic diverticulum: results of surgical treatment. Ann
reapproximated over top with an interposition graft usually Thorac Surg. 1993;55(5):1109-1114.
Surgical Management of Mid- and Distal Esophageal Diverticula CHAPTER 12 183
12. Fasano NC, Levine MS, Rubesin SE, Redfern RO, Laufer I. Epiphrenic 32. Pasricha PJ, Hawari R, Ahmed I, et al. Submucosal endoscopic
diverticulum: clinical and radiographic findings in 27 patients. esophageal myotomy: a novel experimental approach for the treat-
Dysphagia. 2003;18(1):9-15. ment of achalasia. Endoscopy. 2007;39(9):761-764.
13. Nehra D, Lord R V, DeMeester TR, et al. Physiologic basis for the 33. Inoue H. Peroral endoscopic myotomy (POEM) for esophageal
treatment of epiphrenic diverticulum. Ann Surg. 2002;235(3):346-354. achalasia. Endoscopy. 2010;42:265-271.
14. Nascimento F, Lemme E, Costa M. Esophageal diverticula: patho- 34. Khashab MA. Thoughts on starting a peroral endoscopic myotomy
genesis, clinical aspects, and natural history. Dysphagia. 2006;198-205. program. Gastrointest Endosc. 2013;77(1):109-110.
15. Melman L, Quinlan J, Robertson B, et al. Esophageal manometric 35. Liu B-R, Song J, Fan Q. 899 endoscopic esophageal epiphrenic
characteristics and outcomes for laparoscopic esophageal diver- diverticulum inversion by using the submucosal tunneling technique.
ticulectomy, myotomy, and partial fundoplication for epiphrenic Gastrointest Endosc. 2015;81(5):AB180.
diverticula. Surg Endosc Other Interv Tech. 2009;23(6):1337-1341. 36. Allaix ME, BorraezSegura BA, Herbella FA, Fisichella PM, Patti
16. Fisichella PM, Jalilvand A, Dobrowolsky A. Achalasia and epiphrenic MG. Is resection of an esophageal epiphrenic diverticulum always
diverticulum. World J Surg. 2015;39(7):1614-1649. necessary in the setting of achalasia? World J Surg. 2015;39(1):203-207.
17. Dodds WJ, Stef JJ, Hogan WJ, Hoke SE, Stewart ET, Arndorfer RC. 37. Streitz JM, Glick ME, Ellis H Jr. Selective use of myotomy for treatment
Radial distribution of esophageal peristaltic pressure in normal of epiphrenic diverticula. Arch Surg. 1992;127:585-588.
subjects and patients with esophageal diverticulum. Gastroenterology. 38. Smallwood N, Fleshman J, Leeds S, Burdick J. The use of endoluminal
1975;69:584-590. vacuum (E-Vac) therapy in the management of upper gastrointestinal
18. Orringer MB. Epiphrenic diverticula: fact and fable. Ann Thorac leaks and perforations. Surg Endosc. 2016;30(6):2473-2480.
Surg. 1993;55(5):1067-1068. 39. Dukes RJ, Strimlan CV, Dines DE, Payne WS, MacCarty RL.
19. Kostic SV, Rice TW, Baker ME, et al. Timed barium esophagogram: Esophageal involvement with mediastinal granuloma. JAMA.
a simple physiologic assessment for achalasia. J Thorac Cardiovasc 1976;236(20):2313-2315.
Surg. 2000;120(5):935-943. 40. Kaye MD. Oesophageal motor dysfunction in patients with diverticula
20. Oezcelik A, Hagen JA, Halls JM, et al. An improved method of of the mid-thoracic oesophagus. Thorax. 1974;29(6):666-672.
assessing esophageal emptying using the timed barium study following 41. Wychulis AR, Ellis FH, Andersen HA. Acquired nonmalignant
surgical myotomy for achalasia. J Gastrointest Surg. 2009;13(1):14-18. esophagotracheobronchial fistula. Report of 36 cases. JAMA.
21. Allen TH, Claggett OT. Changing concepts in the surgical treatment 1966;196(2):117-122.
of pulsion diverticula of the lower esophagus. J Thorac Cardiovasc 42. Jonasson OM, Gunn LC. Midesophageal diverticulum with hemor-
Surg. 1965;50(4):455-462. rhage. Arch Surg. 1965;90:713-715.
22. Richards WO, Torquati A, Holzman MD, et al. Heller myotomy versus 43. Fékéte F, Vonns C. Surgical management of esophageal thoracic
Heller myotomy with Dor fundoplication for achalasia: a prospective diverticula. Hepatogastroenterology. 1992;39(2):97-99.
randomized double-blind clinical trial. Ann Surg. 2004;240(3):405-412. 44. Altorki NK, Sunagawa M, Skinner DB. Thoracic esophageal
23. Little AG, Soriano A, Ferguson MK, Winans CS, Skinner DB. Surgical diverticula. Why is operation necessary? J Thorac Cardiovasc Surg.
treatment of achalasia: results with esophagomyotomy and Belsey 1993;105(2):260-264.
repair. Ann Thorac Surg. 1988;45(5):489-494. 45. Castrucci G, Porziella V, Granone PL, Picciocchi A. Tailored surgery
24. Peracchia A, Bonavina L, Rosati R. Thoracoscopic resection of for esophageal body diverticula. Eur J Cardiothorac Surg. 1998;
epiphrenic esophageal diverticula. In: Peters JH, DeMeester T, eds. 14(4):380-387.
Minimally Invasive Surgery of the Foregut. St. Louis: Quality Medical 46. D’Journo XB, Ferraro P, Martin J, Chen LQ, Duranceau A. Lower
Publishing; 1995:110. oesophageal sphincter dysfunction is part of the functional abnormal-
25. Boeckxstaens GE, Annese V, des Varannes SB, et al. Pneumatic ity in epiphrenic diverticulum. Br J Surg. 2009;96(8):892-900.
dilation versus laparoscopic Heller’s myotomy for idiopathic achalasia. 47. Rosati R, Fumagalli U, Elmore U, De Pascale S, Massaron S, Peracchia
N Engl J Med. 2011;364(19):1807-1816. A. Long-term results of minimally invasive surgery for symptomatic
26. Pellegrini C, Wetter L, Patti M, et al. Thoracoscopic esophagomy- epiphrenic diverticulum. Am J Surg. 2011;201(1):132-135.
otomy. Initial experience with a new approach for the treatment 48. Fumagalli Romario U, Ceolin M, Porta M, Rosati R. Laparoscopic
of achalasia. Ann Surg. 1992;216(3):291-296. repair of epiphrenic diverticulum. Semin Thorac Cardiovasc Surg.
27. Stewart KC, Finley RJ, Clifton JC, Graham AJ, Storseth C, Inculet 2012;24(3):213-217.
R. Thoracoscopic versus laparoscopic modified Heller myotomy for 49. Klaus A, Hinder RA, Swain J, Achem SR. Management of epiphrenic
achalasia: efficacy and safety in 87 patients. J Am Coll Surg. 1999;189(2): diverticula. J Gastrointest Surg. 2003;7(7):906-911.
164-170. 50. Del Genio A, Rossetti G, Maffetton V, et al. Laparoscopic approach
28. Rosati R, Fumagalli U, Bona S, Bonavina L, Peracchia A. Diver- in the treatment of epiphrenic diverticula: long-term results. Surg
ticulectomy, myotomy, and fundoplication through laparoscopy: a Endosc. 2004;18(5):741-745.
new option to treat epiphrenic esophageal diverticula? Ann Surg. 51. Zaninotto G, Parise P, Salvador R, et al. Laparoscopic repair of
1998;227(2):174-178. epiphrenic diverticulum. Semin Thorac Cardiovasc Surg. 2012;24(3):
29. Fernando HC, Luketich JD, Samphire J, et al. Minimally invasive 218-222.
operation for esophageal diverticula. Ann Thorac Surg. 2005; 52. Palanivelu C, Rangarajan M, Maheshkumaar GS, Senthilkumar
80(6):2076-2080. R. Minimally invasive surgery combined with peroperative endos-
30. Soares RV, Montenovo M, Pellegrini CA, Oelschlager BK. Laparoscopy copy for symptomatic middle and lower esophageal diverticula:
as the initial approach for epiphrenic diverticula. Surg Endosc Other a single institute’s experience. Surg Laparosc Endosc Percutan Tech.
Interv Tech. 2011;25(12):3740-3746. 2008;18(2):133-138.
31. Bellevue OC, Louie BE. Laparoscopic diverticulectomy and myotomy.
In: Wee J, Linden PA, eds. Thoracic and Esophageal Surgery. Scientific
American; In press.
CHAPTER
A
chalasia is a rare esophageal disorder presenting diagnosis of gastroesophageal reflux disease. In fact, in
primarily with dysphagia, characterized by well- a series of 145 untreated patients with achalasia, 65%
defined esophageal motor abnormalities. This were taking acid-suppressing medications at the time of
chapter addresses the epidemiology, pathophysiology, and referral.13
diagnosis of achalasia, as well as the medical therapeutic Once the diagnosis of achalasia is suspected, anatomic
options for this disorder. evaluation is required to exclude other conditions that
can mimic achalasia. Endoscopy is particularly useful for
diagnosing mechanical obstruction or pseudoachalasia, a
EPIDEMIOLOGY rare entity caused by tumors involving the gastroesophageal
Epidemiologic data originate mostly from Western popula- junction.15 Occasionally tumors may infiltrate the tissue
tions and vary among studies. The incidence rate ranges beneath the mucosa, and in such cases imaging studies
from a low of 0.03/100,000 per year to 1.63/100,000 such as endoscopic ultrasound or computed tomography
per year, with the majority of rates clustering between scan are required for diagnosis. Endoscopic findings in
0.5 and 1.2/100,000 per year, with prevalence rates of the more advanced stage of achalasia include a dilated
approximately 10/100,000.1,2 Hospitalizations for achalasia esophagus containing retained saliva or food residue, stasis
increase in frequency with age, with a peak incidence in changes in the mucosa, and occasionally the presence of
those older than 65 years, reaching a high of 37/100,000 candidiasis. However, early on endoscopic findings may be
in patients older than 85 years.3 The incidence of acha- unremarkable. The puckered appearing gastroesophageal
lasia is comparable among males and females, and it can junction presents with mild resistance when attempt-
affect adults of all age groups.1,4 A genetic etiology is ing to intubate the stomach. The feeling of a stronger
documented in a fraction of patients with achalasia. The resistance should raise suspicion for pseudoachalasia, and
triple-A syndrome (Allgrove disease) is a rare condition the need for further evaluation. Careful inspection of the
presenting with achalasia, alacrima, adrenocorticotropic gastroesophageal junction (GEJ) and cardia, including a
hormone (ACTH)–resistant adrenal insufficiency, and retroflex view, should help exclude infiltrating lesions.
neurologic disturbances.5–7 A few familial cases have been
described.8,9 Polymorphisms in the nitric oxide synthase RADIOLOGY
gene have been investigated, but data are conflicting. Typical findings on barium esophagram in advanced disease
No difference in polymorphisms was found between are a dilated esophagus with food and contrast retention,
patients with achalasia and controls; however, a recent lack of peristaltic stripping waves, and a narrowed GEJ
report described two siblings with infant-onset achalasia (the so-called bird beak). The type III variant can present
that were homozygous for a premature stop codon in radiographically with a corkscrew appearance, the result of
the gene encoding nitric oxide synthase.10,11 Idiopathic spastic contractions. Comparable to endoscopy, radiologic
achalasia was also found to be associated with class II findings in early disease may be nonspecific. Radiology is
human leukocyte antigens (HLAs).12 helpful in assessing esophageal emptying using the time
barium esophagram (TBE). This simple technique involves
drinking a large bolus of barium in an upright position
DIAGNOSIS and obtaining a radiograph after 1 and 5 minutes and
Progressive dysphagia to both solids and liquids is the assessing the height of the barium column in the esophagus
most common presenting symptom (90%), followed by (Fig. 13.1).16 This measure proved useful in predicting
regurgitation of undigested food (76% to 91%), respiratory symptomatic response to therapy, whether by balloon
complications (nocturnal cough [30%] and aspiration dilatation or by myotomy. Lack of adequate reduction in the
[8%]), chest pain (25% to 64%), heartburn (18% to 52%), height of the barium column after therapy was associated
and weight loss (35% to 91%).13,14 However, symptoms are with higher risk of treatment failure during follow-up,
nonspecific, and the diagnosis is commonly delayed by and should prompt careful follow-up or reintervention
a number of years after the onset of symptoms. Patients in such patients.17,18
often accommodate to their dysphagia by changing their
eating habits, avoiding solid food such as meat and bread, MANOMETRY
and drinking liquids with their meals. Unless specifically The diagnosis of achalasia is contingent upon the presence
asked, they may not provide these clues to their condition. of impaired lower esophageal sphincter (LES) relaxation
Heartburn, caused by fermentation of retained food in and apersiatalsis, in the absence of obstructive lesions
the esophagus, and regurgitation can lead to an incorrect involving the gastroesophageal junction.19 While endoscopy
184
Epidemiology, Diagnosis, and Medical Management of Achalasia CHAPTER 13 184.e1
ABSTRACT
Achalasia is a rare esophageal disorder of unknown etiol-
ogy. Its pathophysiology is well understood, and involves
the loss of esophageal myenteric neurons, primarily the
inhibitory neurons. Such loss results in absent peristaltic
activity and impaired relaxation of the lower esophageal
sphincter. Dysphagia for solids and liquids is the most
common symptom. Esophageal manometry is the key
diagnostic procedure, particularly early in the course of
the disease. Recent advances in manometric techniques
established subtypes of achalasia that can be used to predict
response to medical or surgical therapy. Endoscopy and
radiology mostly support the diagnosis and help exclude
other lesions. Treatment modalities are aimed at reducing
the barrier to flow by disrupting lower esophageal sphincter
integrity and function. Pneumatic balloon dilation is the
main mode of medical therapy, while pharmacotherapy
and botulinum toxin injection are reserved for high-risk
patients.
KEYWORDS
Acalasia, pathophysiology, balloon dilation,
pharmacotherapy
Epidemiology, Diagnosis, and Medical Management of Achalasia CHAPTER 13 185
A: 1,350.25 * mm2
39 * mm P: 211.78 * mm
M:828.45
SD:43.63
6.41 * mm
65.22 * mm
FIGURE 13.1 Time barium esophagram in
achalasia. (A) Appearance at 5 minutes after
swallow, with retention of contrast in a
dilated esophagus. (B) Near complete
clearance at 1 minute in a nondilated
A B
esophagus, following successful myotomy.
and imaging studies are complementary tests to the diag- response, those with type III have the poorest response but
nosis, and can be highly suggestive in classic achalasia, tend to do better with myotomy as compared to pneumatic
they are frequently not sensitive enough, and manometry dilation, and type I patients present with intermediate
is required for proper diagnosis. Traditional esophageal response that worsens with increasing dilatation of the
manometry was performed with water-perfused or strain esophagus.22,24
gauge systems, usually with sensors spaced a few centimeters Another entity, recently detected by HRM, EGJ outflow
apart, with a line mode display. The last decade saw the obstruction, is characterized by impaired EGJ relaxation
replacement of the traditional system by high-resolution in the presence of preserved peristaltic contractions (see
manometry (HRM) systems. Originally conceived by Fig. 13.2).25 In some patients it is secondary to obstructive
Clouse, the system includes catheters that incorporate etiologies such as paraesophageal hernia, Schatzki ring,
multiple sensors, spaced 1 cm apart, allowing recording benign, or malignant tumors involving the EGJ, while in
from the upper esophageal sphincter to the stomach, and others, no obvious abnormality is found. Imaging studies
software that provides esophageal pressure topography are required to distinguish the secondary from the primary
(EPT) display, with an added presentation in color mode.20 group. Some of the patients with primary GEJ outlet
HRM simplifies the recording process and facilitates the obstruction may go on to develop achalasia.25,26 Various
identification of pressure events and pattern recognition. treatment modalities for symptomatic patients have been
In the case of achalasia, it has tightened diagnostic criteria reported, including balloon dilation, botulinum toxin
by the introduction of new metrics, and in particular the A (Botox) injection, and Heller myotomy, with varying
development of the integrated relaxation pressure (IRP) degrees of success. In some patients spontaneous resolution
for better assessment of esophagogastric junction (EGJ) has been documented.25,27,28
relaxation.21 Esophageal pressure topography has allowed
for the classification of achalasia into three variants on the MEDICAL MANAGEMENT
basis of the contractile pattern of the esophageal body.22 In
type I achalasia (classic achalasia), no contractile activity Treatment modalities for achalasia, whether surgical,
is detected and no significant pressurization is within the endoscopic, or medical, are aimed at reducing the resis-
esophageal body. In type II achalasia, swallowing water tance to esophageal emptying by reducing LES tone, since
results in panesophageal pressurization, while type III meaningful peristaltic activity cannot be restored by any
achalasia (spastic/vigorous achalasia) is associated with intervention. The purpose of all treatment modalities is
premature contractions (characterized by short distal to relieve symptoms and prevent complications such as
latency; Fig. 13.2). Most recently, a new recording method, weight loss, aspiration pneumonia, and further dilation of
the functional lumen imaging probe (FLIP), has detected the esophagus. Nonsurgical treatments consist primarily
esophageal contractile activity in achalasia not seen with of pharmacologic therapy and pneumatic balloon dilation
the standard thin intraluminal manometric catheters.23 (PD).
The pathophysiologic and prognostic implications of these
patterns require further study. A number of publications PHARMACOLOGIC THERAPY
have highlighted the clinical prognostic value of these This approach uses smooth muscle relaxants to reduce LES
variants in predicting response to medical or surgical tone. Nitrates and calcium channel blockers are the most
therapy, whereby patients with type II variant have the best commonly used. A review of randomized trials concluded
186 SECTION I Esophagus and Hernia
mm Hg mm Hg
150.0 150.0
140 140
120 120
100 100
80
80
60
60
40
40
30
30
20
20
10
10
0
0
−6.0
A −7.0 C
mm Hg mm Hg
150.0 150.0
140 140
120 120
100 100
80 80
60
60
40
40
30
30
20
20
10
10
0
0
−6.0
B D −9.0
FIGURE 13.2 Subtypes of achalasia as they appear on high-resolution manometry. (A) Type I achalasia is characterized by absent
peristalsis. (B) Type II achalasia is characterized by panesophageal pressurization, the result of pressure buildup within a closed cavity.
(C) Type III achalasia is characterized by contractile activity with short latency (spastic) following a swallow. (D) Esophagogastric junction
outflow obstruction in a patient with dysphagia. Incomplete swallow-induced lower esophageal sphincter relaxation (high integrated
relaxation pressure) with normal contractile sequence.
that there was not enough evidence to determine the the use of these agents. Hence these agents are useful in
therapeutic efficacy of nitrates in achalasia.29 When used, patients in whom more definitive interventions are deemed
they should be delivered in sublingual form, such as too risky, or when patients desire a low-risk therapy.
isosorbide dinitrate 5 mg, taken 10 to 15 minutes before
meals. Nifedipne, a calcium channel blocker, is also taken BOTULINUM TOXIN
sublingually, at a dose of 10 to 30 mg) 30 to 45 minutes This agent has been more thoroughly studied, and is more
before meals.30,31,32 Limited symptomatic response and side widely used. Botulinum toxin A is a neurotoxin that inhibits
effects such as hypotension, dizziness, and headaches limit the release of acetylcholine from the nerve terminal, thus
Epidemiology, Diagnosis, and Medical Management of Achalasia CHAPTER 13 187
<45 yr >45 yr
Botulinum toxin
FIGURE 13.3 Suggested approach to the patient with dysphagia. PD, Pneumatic balloon dilation. (Modified with permission from Richter
JE. Esophageal motility disorder achalasia. Curr Opin Otolaryngol Head Neck Surg. 2013;21:535–542.)
blocking the neurogenic but not the myogenic component is suboptimal in patients with adequate reduction in LES
of LES tone. It is injected into the LES endoscopically by pressure.38 Chest pain is common after PD, but the feared
a sclerotherapy needle, at a total dose of 100 U, usually complication is perforation. Previous data suggested rates
divided in four aliquots, injected to the four quadrants of 0% to 16%, but a recent systematic review reported a
of the LES, just above the squamocolumnar junction. rate of about 1%, comparable to the rate of unrecognized
Prospective randomized controlled trials compared its perforation during Heller myotomy.39,40 Fig. 13.3 outlines
therapeutic efficacy with pneumatic dilation and with a suggested algorithm for the treatment of patients with
surgical intervention.33,34 While the response rate early on is achalasia.
quite high and comparable to more invasive interventions,
relapse is observed in approximately 50% of patients ESOPHAGEAL STENTS
by 1 year, and repeat injections are required. Given the Initial experience in a small number of patients dem-
simplicity and good safety profile of this intervention, it onstrated limited success but substantial morbidity.41,42
is an attractive option in patients who are unsuitable for Recently the use of temporary, self-expanding, fully
more risky interventions. covered metal stents or partially covered metal stents
was reported.43,44 Diameter varied between 20 and 30 mm,
BALLOON DILATION and stents were removed after 6 or 30 days, respectively.
The main medical intervention for achalasia is PD. It is Follow-up of up to 36 months showed good remission
currently done with a standard noncompliant balloon rates with low morbidity related to stent migration in a
(with a Rigiflex dilator by Boston Scientific) and comes in few patients. This treatment modality may be attractive
three sizes (30, 35, and 40 mm diameters). It is positioned in high-risk patients, but further data are needed for
across the LES, commonly under fluoroscopic control better assessment of its role in the therapeutic approach
and with the aid of a guidewire, and is inflated with air to achalasia.
to forcefully stretch the LES muscle. The balloon is kept
inflated for about 15 to 60 seconds, while straightening
of the balloon waist at the level of the LES is observed.
REFERENCES
Following dilation, patients are evaluated by a Gastrogra- 1. O’Neill OM, Johnston BT, Coleman HG. Achalasia: a review of
clinical diagnosis, epidemiology, treatment and outcomes. World J
fin study followed by a barium esophagram to exclude Gastroenterol. 2013;19(35):5806-5812.
esophageal perforation.32 Graded dilation, starting with 2. Kim E, Lee H, Jung HK, Lee KJ. Achalasia in Korea: an epidemiologic
the smallest diameter, is the recommended approach, with study using a national healthcare database. J Korean Med Sci. 2014;
symptomatic and radiographic assessment within 4 to 6 29(4):576-580.
weeks after treatment, and repeat dilation with a larger 3. Sonnenberg A. Hospitalization for achalasia in the United States
1997–2006. Dig Dis Sci. 2009;54:1680.
size balloon in case of failed response. Clinical response 4. Podas T, Eaden J, Mayberry M, Mayberry J. Achalasia: a critical review
following PD as the initial therapy for achalasia varies of epidemiological studies. Am J Gastroenterol. 1998;93:2345-2347.
depending on the balloon size and number of dilations. 5. Allgrove J, Claden GS, Grant DB, Macaulay JC. Familial glucocorticoid
A success rate of 28% versus 44% was achieved at 6 years deficiency with achalasia of the cardia and deficient tear production.
Lancet. 1978;1:1284-1286.
with single versus serial dilation, respectively.35 A number 6. Stuckey BG, Mastaglia FL, Reed WD, Pullan PT. Glucocorticoid
of variables have been associated with a favorable response insufficiency, achalasia, alacrima with autonomic motor neuropathy.
to PD, including older age (>45 years), female gender, Ann Intern Med. 1987;106(1):61-63.
LES pressure of less than 10 mm Hg following PD, narrow 7. Handschg K, Sperling S, Yoon SJ, Hennig S, Clark AJ, Huebner A.
esophagus prior to PD, and type II pattern on HRM, and Triple A syndrome is caused by mutations in AAAS, a new WD-repeat
protein gene. Hum Mol Genet. 2001;10:283-290.
may influence the selection of medical versus surgical 8. Gordillo-González G, Guatibonza YP, Zarante I, Roa P, Jacome LA,
treatment approach.22,36,37 Balloon dilation can also be Hani A. Achalasia familiar: report of a family with an autosomal
performed after failed Heller myotomy, though response dominant pattern of inherence. Dis Esophagus. 2011;24(1):E1-E4.
188 SECTION I Esophagus and Hernia
9. Evsyutina YV, Trukhmanov AS, Ivashkin VT. Family case of achalasia cardia: 27. Porter RF, Gyawali CP. Botulinum toxin injection in dysphagia
case report and review of literature. World J Gastroenterol. 2014;20(4): syndromes with preserved esophageal peristalsis and incomplete
1114-1118. lower esophageal sphincter relaxation. Neurogastroenterol Motil.
10. Vigo AG, Martínez A, de la Concha EG, Urcelay E, Ruiz de León 2011;23:139-144, e127–e138.
A. Suggested association of OS2A polymorphism in idiopathic 28. Pérez-Fernández MT, Santander C, Marinero A, Burgos-Santamaría
achalasia: no evidence in a large case-control study. Am J Gastroenterol. D, Chavarría-Herbozo C. Characterization and follow up of esophago-
2009;104(5):1326-1327. gastric junction outflow obstruction detected by high resolution
11. Shteyer E, Edvardson S, Wynia-Smith SL, et al. Truncating mutation manometry. Neurogastroenterol Motil. 2016;28:116-126.
in the nitric oxide synthase 1 gene is associated with infantile 29. Wen ZH, Gardener E, Wang YP. Nitrates for achalasia. Cochrane
achalasia. Gastroenterology. 2015;148(3):533-536.e4. Database Syst Rev. 2004;(1):CD002299.
12. De la Concha EG, Fernandez-Arquero M, Mendoza JL, et al. Con- 30. Bortolotti M, Labo G. Clinical and manometric effects of nifedipine
tribution of HLA class II genes to susceptibility in achalasia. Tissue in patients with esophageal achalasia. Gastroenterology. 1981;80:39-44.
Antigens. 1998;52:381-384. 31. Traube M, Dubovik S, Lange RC, McCallum RW. The role of nife-
13. Fisichella PM, Raz D, Palazzo F, Niponmick I, Patti MG. Clinical, dipine therapy in achalasia: results of a randomized, double-blind,
radiological, and manometric profile in 145 patients with untreated placebo-controlled study. Am J Gastroenterol. 1989;84:1259-1262.
achalasia. World J Surg. 2008;32:1974-1979. 32. Vaezi MF, Pandolfi JE, Vela MF. ACG clinical guideline: diagnosis
14. Boeckxstaens G, Zaninotto G, Richter J. Achalasia. Lancet. 2014;383: and management of achalasia. Am J Gastroenterol. 2013;108:1238-1249.
83-93. 33. Leyden JE, Moss AC, MacMathuna P. Endoscopic pneumatic dilation
15. Kahrilas PJ, Kishk SM, Helm JF, Dodds WJ, Harig JM, Hogan versus botulinum toxin injection in the management of primary
WJ. A comparison of pseudoachalasia and achalasia. Am J Med. achalasia. Cochrane Database Syst Rev. 2006;(4):CD005046.
1987;82:439-446. 34. Campos GM, Vittinghoff E, Rabl C. Endoscopic and surgical treat-
16. De Oliveira JM, Birgisson S, Doinoff C, et al. Timed barium swallow: ments for achalasia: a systematic review and meta-analysis. Ann Surg.
a simple technique for evaluating esophageal emptying in patients 2009;249:45-57.
with achalasia. AJR Am J Roentgenol. 1997;169:473-479. 35. Vela MF, Richter JE, Khandwala F, et al. The long-term efficacy of
17. Vaezi MF, Baker ME, Achkar E, Richter JE. Timed barium oesopha- pneumatic dilatation and Heller myotomy for the treatment of
gram: better predictor of long term success after pneumatic dilation achalasia. Clin Gastroenterol Hepatol. 2006;4:580-587.
in achalasia than symptom assessment. Gut. 2002;50:765-770. 36. Farhoomand K, Connor JT, Richter JE, Achkar E, Vaezi MF. Predictors
18. Andersson M, Lundell L, Kostic S, et al. Evaluation of the response of outcome of pneumatic dilation in achalasia. Clin Gastroenterol
to treatment in patients with idiopathic achalasia by the timed Hepatol. 2004;2:389-394.
barium esophagogram: results from a randomized clinical trial. Dis 37. Pratap N, Kalapala R, Darisetty S, et al. Achalasia cardia subtyping
Esophagus. 2009;22:264-273. by high resolution manometry predicts the therapeutic outcome of
19. Pandolfino JE, Kahrilas PJ. American Gastroenterological Association pneumatic balloon dilatation. J Neurogastroenterol Motil. 2011;17:48-53.
medical position statement: clinical use of esophageal manometry. 38. Guardino JM, Vela MF, Connor JT, Richter JE. Pneumatic dilation
Gastroenterology. 2005;128:207-208. for the treatment of achalasia in untreated patients and patients
20. Clouse RE, Staiano A. Topography of the esophageal peristaltic with failed Heller myotomy. J Clin Gastroenterol. 2004;38:855-860.
pressure wave. Am J Physiol. 1991;261(4 Pt 1):G677-G684. 39. Katzka DA, Castell DO. Review article: an analysis of the efficacy,
21. Ghosh SK, Pandolfino JE, Rice J, Clarke JO, Kwiatek M, Kahrilas P. perforation rates and methods used in pneumatic dilation for
Impaired deglutitive EGJ relaxation in clinical esophageal manometry: achalasia. Aliment Pharmacol Ther. 2011;34:832-839.
a quantitative analysis of 400 patients and 75 controls. Am J Physiol 40. Lynch KL, Pandolfino JE, Howden CW, Kahrilas PJ. Major complica-
Gastrointest Liver Physiol. 2007;293:G878-G885. tions of pneumatic dilation and Heller myotomy for achalasia:
22. Pandolfino JE, Kwiatek MA, Nealis T, Bulsiewicz W, Post J, Kahrilas single-center experience and systematic review of the literature.
PJ. Achalasia: a new clinically relevant classification by high-resolution Am J Gastroenterol. 2012;107(12):1817-1825.
manometry. Gastroenterology. 2008;135:1526-1533. 41. Mukherjee S, Kaplan DS, Parasher G, Sipple MS. Expandable metal
23. Carlson D, Zhiyue L, Kahrilas P, et al. The functional lumen imaging stents in achalasia—is there a role? Am J Gastroenterol. 2000;95:
probe detects esophageal contractility not observed with manometry 2185-2188.
in patients with achalasia. Gastroenterology. 2015;149:1742-1751. 42. De Palma GD, Lovino P, Masone S, Persico M, Persico G. Self-
24. Rohof WO, Salvador R, Annese V, et al. Outcomes of treatment of expanding metal stents for endoscopic treatment of esophageal
achalasia depend on manometric subtype. Gastroenterology. 2013;144: achalasia unresponsive to conventional treatments: long-term results
718-725. in eight patients. Endoscopy. 2001;33:1027-1030.
25. Scherer JR, Kwiatek MA, Soper NJ, Pandolfino JE, Kahrilas PJ. 43. Zeng Y, Dai YM, Wan XJ. Clinical remission following endoscopic
Functional esophagogastric junction obstruction with intact peristalsis: placement of retrievable, fully covered metal stents in patients with
a heterogeneous syndrome sometimes akin to achalasia. J Gastrointest esophageal achalasia. Dis Esophagus. 2014;27:103-108.
Surg. 2009;13:2219-2225. 44. Coppola F, Gaia S, Rolle E, Recchia S. Temporary endoscopic metallic
26. Van Hoeij FB, Smout AJ, Bredenoord AJ. Characterization of idio- stent for idiopathic esophageal achalasia. Surg Innov. 2014;21:11-14.
pathic esophagogastric junction outflow obstruction. Neurogastroenterol
Motil. 2015;27:1310-1316.
CHAPTER
Endoscopic and Surgical Therapies for Achalasia
Paul D. Colavita
| Lee L. Swanstrom
14
W
ith an estimated annual incidence of 1 in Classification.9 All three require incomplete or failed
100,000 and prevalence of 10 in 100,000, 1,2 relaxation of the LES defined by integrated relaxation
achalasia is a rare, primary motility disorder of pressure (IRP) thresholds. Type I is consistent with the
the esophagus, defined by absence of normal peristalsis classical consideration of achalasia: incomplete or failed
and failure of swallow-induced relaxation of the lower relaxation of the LES with no esophageal body contractility.
esophageal sphincter (LES). The first described treatment Type II requires absence of normal peristalsis with at least
for “cardiospasm” in the 1600s was self-dilatation with oral 20% of swallows having panesophageal pressurization. Type
passage of a whale rib-bone with an attached sponge.3 III consists of absent normal peristalsis with secondary
In 1937 Lendrum proposed the pathologic mechanism and tertiary spastic distal esophageal contractions in at
of failed LES relaxation, leading to the new name of least 20% of swallows.10 The response of these subtypes
achalasia.4 The etiology of achalasia remains unclear, but to various treatments will be discussed in the sections
autoimmune, viral, and neurodegenerative factors have that follow.
been implicated in the development of impaired LES
relaxation through loss of ganglion cells in Auerbach TREATMENT OPTIONS
myenteric plexus, leading to loss of neurotransmitters, such
as vasoactive intestinal peptide and nitric oxide.5–7 As the Surgical myotomy was initially described by Heller11 in
underlying mechanism for achalasia development remains 1913, and was performed as a thoracotomy with two
unknown, current treatment options rely on disruption separate 8 cm myotomies, separated by 180 degrees. The
of the LES to relieve outflow obstruction. technique has undergone several modifications and now
Patients typically present with progressive dysphagia, involves a single anterior myotomy with addition of a
volume regurgitation, and sometimes noncardiac chest partial fundoplication to reduce the incidence of iatrogenic
pain. reflux. Pneumatic dilatation became a common treatment
in the 1970s and 1980s. In the early 1990s, laparoscopic
and thoracoscopic approaches to the Heller myotomy were
PREOPERATIVE WORK-UP described.12,13 Peroral endoscopic myotomy (POEM) was
An upper gastrointestinal series or esophagram is the developed in the first decade of this century, but has roots
usual initial test to evaluate dysphagia, although some in the 1980s. Ortega14 described an endoscopic myotomy
clinicians include endoscopy as an initial assessment. in 1980, consisting of division of the mucosa and muscle
An esophagram will often demonstrate a “bird’s beak” of the LES with two short full thickness incisions. Despite
tapering of the distal esophagus, with progressive dilatation satisfactory outcomes, the technique was not adopted,
and tortuosity of the esophagus demonstrated in later due to concern for esophageal perforation. POEM was
stages of the disease. The definitive diagnosis requires first described by Pasricha15,16 in 2007 as a “submucosal
confirmation by manometry, preferably high-resolution endoscopic esophageal myotomy” in a porcine model.
manometry (HRM). Endoscopy is important to rule out Inoue presented the first human experience in 2008 with
pseudoachalasia from an obstruction, such as cancer, and four patients and coined the term “POEM.” He published
endoscopic ultrasound can be considered when suspicion a series on 17 patients 2 years later.17
for malignancy causing obstruction is high. Endoscopy
frequently appears normal in early stages of the disease, LAPAROSCOPIC HELLER MYOTOMY WITH
as dilation of the esophageal body may not have occurred, PARTIAL FUNDOPLICATION
but esophagitis from stasis and/or yeast esophagitis may be By the mid- to late-1990s, laparoscopic Heller myotomy
present. Later stages of disease may demonstrate dilatation had become the primary treatment for achalasia.3 The
of the esophageal body with tortuosity or sigmoidization. authors’ practice is to keep patients on a clear liquid
A timed barium swallow (TBS)8 provides objective data on diet for 24 hours before surgery. The procedure requires
esophageal emptying and is commonly used before and general anesthesia and is most commonly performed in a
as follow-up after treatment for achalasia. TBS involves split leg position. The procedure is performed with both
ingestion of 200 mL of oral contrast, with radiographs arms on arm boards. As with all procedures involving the
taken at baseline and then 1, 2, and 5 minutes after hiatus, steep reverse Trendelenburg positioning is utilized,
ingestion. The height and width of the barium column so patients should be well secured to the operating room
at each time period are recorded. table. The authors’ preferred positioning involves foot
boards at the base of the flat padded and abducted leg
boards with sequential compression devices for deep venous
ACHALASIA MANOMETRIC SUBTYPES thrombosis prophylaxis in place. The legs are covered with
With the advent of HRM, three manometric subtypes a warming device and blankets, and the legs are secured
of achalasia have been described as part of the Chicago to the leg boards with tape above and below the knee.
189
Endoscopic and Surgical Therapies for Achalasia CHAPTER 14 189.e1
ABSTRACT
Achalasia is a rare, primary motility disorder of the esopha-
gus, defined by absence of normal peristalsis and failure
of swallow-induced relaxation of the lower esophageal
sphincter (LES). As the underlying mechanism for acha-
lasia development remains unknown, current treatment
options rely on disruption of the LES to relieve outflow
obstruction. The chapter focuses on laparoscopic Heller
myotomy and peroral endoscopic myotomy: technique,
results, complications, and clinical follow-up.
KEYWORDS
Achalasia; Heller myotomy; peroral endoscopic myotomy
(POEM)
190 SECTION I Esophagus and Hernia
3
2
4 1
5
A B
FIGURE 14.3 (A to C) A partial anterior fundoplication using the gastric fundus is added to the myotomy. The most proximal stitches also
include the corresponding diaphragmatic pillar.
added. Given the aperistaltic nature of the esophageal usually detected at the time of occurrence. These can be
body, a partial fundoplication is typically performed. A repaired laparoscopically with fine interrupted absorbable
posterior partial fundoplication involves loose posterior sutures. Rates of perforation during primary laparoscopic
crural closure and fastening the left and right wrap to the Heller myotomies range from 2.9% to 6.0% in large series,
corresponding crura. Finally, the right and left sides of with increased rates of perforation after prior endoscopic
the wrap are secured to the esophageal myotomy edges treatments,20–26 and can be as high as 31.2% in redo Heller
with three interrupted nonabsorbable sutures. myotomies.20
An anterior partial fundoplication (Dor)19 is used if The outcome after repair of intraoperative mucosal
there is no hiatal hernia and adequate intraabdominal injury is similar to patients who undergo uncomplicated
esophageal length for myotomy. This requires minimal laparoscopic Heller myotomies.24
mobilization of the fundus and minimizes mediastinal
dissection. The fundus is folded over the myotomy and Results
secured to each arch of the hiatus and to each side of Early results for laparoscopic Heller myotomy demon-
the myotomy with three permanent sutures (Fig. 14.3). strate good improvement in dysphagia, generally greater
than 90% with symptom improvement and with 80% to
Postoperative Care 96%21,27–30 reporting no dysphagia in the first year after
Before initiating oral intake, the authors’ standard practice surgery. Dysphagia rates appear to increase with time, with
is to obtain a water-soluble contrast esophagram on post- only 63.4% to 88%21,22,27 free from dysphagia in follow-up
operative day 1 to exclude perforation and obstruction, beyond 1 year. Studies demonstrate similar improvement
followed by 14 days of a pureed diet, and then a regular with regurgitation with 93.1% to 95%,27,28 reporting no
diet. A routine postoperative visit is planned for 3 to 4 regurgitation in the early postoperative period, and 79.1%27
weeks postoperatively with endoscopy, TBS, and pH testing free from symptoms in long-term follow-up. Dilation after
planned for 6 months to evaluate for GERD. Due to the recurrent symptoms has been demonstrated to provide
increased risk of squamous cell carcinoma, screening durable dysphagia relief in up to 75% of patients with
endoscopies are recommended every 5 years. recurrent dysphagia.22
Objective GERD identified with pH testing was found in
Complications 6% to 41.7%, with variance based on type of fundoplication
The most common complication of an esophageal myotomy performed and more reflux noted after Dor fundoplica-
is mucosal perforation. These are infrequent and are tion.22,31–34 Postoperative reflux has been noted in up to
192 SECTION I Esophagus and Hernia
Other Considerations
FIGURE 14.5 Division of circular fibers and exposure of longitudinal Extended proximal myotomy should be considered for
fibers during peroral endoscopic myotomy.
noncardiac chest pain and for type 3 achalasia, which can
be extended from a few centimeters from the cricopha-
ryngeus to the gastric cardia. POEM has been shown to
Circular Myotomy. The myotomy begins 2 to 3 cm distal be both safe and effective for these extended indications.
to the most distal extent of the mucosotomy and proceeds Due to concerns for submucosal fibrosis from dilations
antegrade to allow the dissecting cap to place tension on or injections, previous treatments were initially felt to be
the muscle fibers. The myotomy consists of division of the contraindications to POEM. As experience has increased,
circular muscle fibers using an Endocut current using a POEM has been successfully completed with no increases
triangle tip or hook cautery. Once the circular fibers are in procedure length or perioperative complication rates
divided, the longitudinal muscle fibers are easily visible after prior intervention.38 POEM has also been reported
(Fig. 14.5). The longitudinal fibers are very thin and after prior laparoscopic myotomy; it is technically difficult
splitting often occurs under the pressure of the dissecting but safe and effective with published efficacy of 91.7% to
cap, which will reveal mediastinal structures beyond the 100%.39–41 In the setting of failed Heller myotomy, some
intact esophageal adventitia. This “mediastinal exposure” is recommend creation of the submucosal tunnel in the
common and does not indicate a full-thickness perforation. posterior aspect of the esophagus42 or the lateral wall.43
The plane between the longitudinal and circular muscle End-stage achalasia or “sigmoid” esophagus is more dif-
fibers is followed distally with continued myotomy. The ficult technically to treat and has somewhat poorer clinical
myotomy plane often becomes tight at the GEJ due to results. As with dilation or Heller myotomy, many end-stage
the thickened musculature, the curve from the esophagus patients will have continued functional deterioration,
194 SECTION I Esophagus and Hernia
8. de Oliveira JM, Birgisson S, Doinoff C, et al. Timed barium swallow: 33. Khajanchee YS, Kanneganti S, Leatherwood AE, Hansen PD,
a simple technique for evaluating esophageal emptying in patients Swanstrom LL. Laparoscopic Heller myotomy with Toupet fundo-
with achalasia. AJR Am J Roentgenol. 1997;169(2):473-479. plication: outcomes predictors in 121 consecutive patients. Arch
9. Bredenoord AJ, Fox M, Kahrilas PJ, et al. Chicago classification criteria Surg. 2005;140(9):827-833; discussion 833–824.
of esophageal motility disorders defined in high resolution esophageal 34. Richards WO, Torquati A, Holzman MD, et al. Heller myotomy versus
pressure topography. Neurogastroenterol Motil. 2012;24(suppl 1): Heller myotomy with Dor fundoplication for achalasia: a prospective
57-65. randomized double-blind clinical trial. Ann Surg. 2004;240(3):405-412;
10. Kahrilas PJ, Bredenoord AJ, Fox M, et al. The Chicago Classifica- discussion 412–405.
tion of esophageal motility disorders, v3.0. Neurogastroenterol Motil. 35. Falkenback D, Johansson J, Oberg S, et al. Heller’s esophagomyotomy
2015;27(2):160-174. with or without a 360 degrees floppy Nissen fundoplication for
11. Heller E. Extramuköse Cardiaplastik beim chronischen Cardio- achalasia. Long-term results from a prospective randomized study.
spasmus mit Dilatation des Oesophagus. Mitt Grenzgeb Med Chir. Dis Esophagus. 2003;16(4):284-290.
1913;27:141-149. 36. Kurian AA, Dunst CM, Sharata A, Bhayani NH, Reavis KM, Swanstrom
12. Shimi S, Nathanson LK, Cuschieri A. Laparoscopic cardiomyotomy LL. Peroral endoscopic esophageal myotomy: defining the learning
for achalasia. J R Coll Surg Edinb. 1991;36(3):152-154. curve. Gastrointest Endosc. 2013;77(5):719-725.
13. Pellegrini C, Wetter LA, Patti M, et al. Thoracoscopic esophagomy- 37. Dunst CM, Kurian AA, Swanstrom LL. Endoscopic myotomy for
otomy. Initial experience with a new approach for the treatment of achalasia. Adv Surg. 2014;48:27-41.
achalasia. Ann Surg. 1992;216(3):291-296; discussion 296–299. 38. Sharata A, Kurian AA, Dunst CM, Bhayani NH, Reavis KM, Swanstrom
14. Ortega JA, Madureri V, Perez L. Endoscopic myotomy in the treatment LL. Peroral endoscopic myotomy (POEM) is safe and effective in the
of achalasia. Gastrointest Endosc. 1980;26(1):8-10. setting of prior endoscopic intervention. J Gastrointest Surg. 2013;17(7):
15. Pasricha PJ, Hawari R, Ahmed I, et al. Submucosal endoscopic 1188-1192.
esophageal myotomy: a novel experimental approach for the treat- 39. Onimaru M, Inoue H, Ikeda H, et al. Peroral endoscopic myotomy
ment of achalasia. Endoscopy. 2007;39(9):761-764. is a viable option for failed surgical esophagocardiomyotomy instead
16. Rajan E, Gostout CJ, Feitoza AB, et al. Widespread EMR: a new of redo surgical Heller myotomy: a single center prospective study.
technique for removal of large areas of mucosa. Gastrointest Endosc. J Am Coll Surg. 2013;217(4):598-605.
2004;60(4):623-627. 40. Zhou PH, Li QL, Yao LQ, et al. Peroral endoscopic remyotomy for
17. Inoue H, Minami H, Kobayashi Y, et al. Peroral endoscopic myotomy failed Heller myotomy: a prospective single-center study. Endoscopy.
(POEM) for esophageal achalasia. Endoscopy. 2010;42(4):265-271. 2013;45(3):161-166.
18. Toupet A. [Technic of esophago-gastroplasty with phrenogastropexy 41. Sharata AM, Dunst CM, Pescarus R, et al. Peroral endoscopic
used in radical treatment of hiatal hernias as a supplement to Heller’s myotomy (POEM) for esophageal primary motility disorders: analysis
operation in cardiospasms]. Mem Acad Chir (Paris). 1963;89:384-389. of 100 consecutive patients. J Gastrointest Surg. 2015;19(1):161-170;
19. Dor J, Humbert P, Paoli JM, Miorclerc M, Aubert J. [Treatment of discussion 170.
reflux by the so-called modified Heller-Nissen technic]. Presse Med. 42. Ren Z, Zhong Y, Zhou P, et al. Perioperative management and
1967;75(50):2563-2565. treatment for complications during and after peroral endoscopic
20. Zhang LP, Chang R, Matthews BD, et al. Incidence, mechanisms, and myotomy (POEM) for esophageal achalasia (EA) (data from 119
outcomes of esophageal and gastric perforation during laparoscopic cases). Surg Endosc. 2012;26(11):3267-3272.
foregut surgery: a retrospective review of 1,223 foregut cases. Surg 43. Chiu PW, Wu JC, Teoh AY, et al. Peroral endoscopic myotomy
Endosc. 2014;28(1):85-90. for treatment of achalasia: from bench to bedside (with video).
21. Patti MG, Pellegrini CA, Horgan S, et al. Minimally invasive surgery Gastrointest Endosc. 2013;77(1):29-38.
for achalasia: an 8-year experience with 168 patients. Ann Surg. 44. Duranceau A, Liberman M, Martin J, Ferraro P. End-stage achalasia.
1999;230(4):587-593; discussion 593–584. Dis Esophagus. 2012;25(4):319-330.
22. Zaninotto G, Costantini M, Rizzetto C, et al. Four hundred laparo- 45. Hu JW, Li QL, Zhou PH, et al. Peroral endoscopic myotomy for
scopic myotomies for esophageal achalasia: a single centre experience. advanced achalasia with sigmoid-shaped esophagus: long-term out-
Ann Surg. 2008;248(6):986-993. comes from a prospective, single-center study. Surg Endosc. 2015;29(9):
23. Rosemurgy A, Villadolid D, Thometz D, et al. Laparoscopic Heller 2841-2850.
myotomy provides durable relief from achalasia and salvages failures 46. von Renteln D, Inoue H, Minami H, et al. Peroral endoscopic
after botox or dilation. Ann Surg. 2005;241(5):725-733; discussion myotomy for the treatment of achalasia: a prospective single center
733–725. study. Am J Gastroenterol. 2012;107(3):411-417.
24. Salvador R, Spadotto L, Capovilla G, et al. Mucosal perforation 47. Zhou P-H, Yao L, Zhang Y-Q, et al. Peroral endoscopic myotomy
during laparoscopic Heller myotomy has no influence on final (POEM) for esophageal achalasia: 205 cases report. Gastrointest
treatment outcome. J Gastrointest Surg. 2016;20(12):1923-1930. Endosc. 2012;75(4):AB132-AB133.
25. Patti MG, Molena D, Fisichella PM, et al. Laparoscopic Heller 48. Stavropoulos SN, Brathwaite CE, Iqbal S, et al. P.O.E.M. (peroral
myotomy and Dor fundoplication for achalasia: analysis of successes endoscopic myotomy), a U.S. gastroenterologist perspective: initial
and failures. Arch Surg. 2001;136(8):870-877. 2 year experience. Gastrointest Endosc. 2012;75(4):AB149.
26. Torquati A, Richards WO, Holzman MD, Sharp KW. Laparoscopic 49. Hungness ES, Sternbach JM, Teitelbaum EN, Kahrilas PJ, Pandolfino
myotomy for achalasia: predictors of successful outcome after 200 JE, Soper NJ. Per-oral endoscopic myotomy (POEM) after the
cases. Ann Surg. 2006;243(5):587-591; discussion 591–583. learning curve: durable long-term results with a low complication
27. Finley CJ, Kondra J, Clifton J, Yee J, Finley R. Factors associated rate. Ann Surg. 2016;264(3):508-517.
with postoperative symptoms after laparoscopic Heller myotomy. 50. Gockel I, Junginger T. The value of scoring achalasia: a comparison
Ann Thorac Surg. 2010;89(2):392-396. of current systems and the impact on treatment—the surgeon’s
28. Hunter JG, Trus TL, Branum GD, Waring JP. Laparoscopic viewpoint. Am Surg. 2007;73(4):327-331.
Heller myotomy and fundoplication for achalasia. Ann Surg. 51. Hungness ES, Teitelbaum EN, Santos BF, et al. Comparison of
1997;225(6):655-664; discussion 664–655. perioperative outcomes between peroral esophageal myotomy
29. Rosati R, Fumagalli U, Bonavina L, et al. Laparoscopic approach (POEM) and laparoscopic Heller myotomy. J Gastrointest Surg.
to esophageal achalasia. Am J Surg. 1995;169(4):424-427. 2013;17(2):228-235.
30. Frantzides CT, Moore RE, Carlson MA, et al. Minimally invasive surgery 52. Swanstrom LL, Kurian A, Dunst CM, Sharata A, Bhayani N, Rieder
for achalasia: a 10-year experience. J Gastrointest Surg. 2004;8(1):18-23. E. Long-term outcomes of an endoscopic myotomy for achalasia:
31. Bhayani NH, Kurian AA, Dunst CM, Sharata AM, Rieder E, Swanstrom the POEM procedure. Ann Surg. 2012;256(4):659-667.
LL. A comparative study on comprehensive, objective outcomes of 53. Schneider AM, Louie BE, Warren HF, Farivar AS, Schembre DB,
laparoscopic Heller myotomy with per-oral endoscopic myotomy Aye RW. A matched comparison of per oral endoscopic myotomy
(POEM) for achalasia. Ann Surg. 2014;259(6):1098-1103. to laparoscopic Heller myotomy in the treatment of achalasia.
32. Rawlings A, Soper NJ, Oelschlager B, et al. Laparoscopic Dor versus J Gastrointest Surg. 2016;20(11):1789-1796.
Toupet fundoplication following Heller myotomy for achalasia: 54. Familiari P, Gigante G, Marchese M, et al. EndoFLIP system for the
results of a multicenter, prospective, randomized-controlled trial. intraoperative evaluation of peroral endoscopic myotomy. United
Surg Endosc. 2012;26(1):18-26. European Gastroenterol J. 2014;2(2):77-83.
196 SECTION I Esophagus and Hernia
55. Tyberg A, Seewald S, Sharaiha RZ, et al. A multicenter 60. Zhang Y, Wang H, Chen X, et al. Per-oral endoscopic myotomy
international registry of redo per-oral endoscopic myotomy versus laparoscopic Heller myotomy for achalasia: a meta-analysis of
(POEM) after failed POEM. Gastrointest Endosc. 2017;85(6): nonrandomized comparative studies. Medicine (Baltimore). 2016;95(6):
1208-1211. e2736.
56. Bechara R, Ikeda H, Inoue H. Peroral endoscopic myotomy: an 61. Miller HJ, Neupane R, Fayezizadeh M, Majumder A, Marks JM. POEM
evolving treatment for achalasia. Nat Rev Gastroenterol Hepatol. is a cost-effective procedure: cost-utility analysis of endoscopic and
2015;12(7):410-426. surgical treatment options in the management of achalasia. Surg
57. Talukdar R, Inoue H, Reddy DN. Efficacy of peroral endoscopic Endosc. 2017;31(4):1636-1642.
myotomy (POEM) in the treatment of achalasia: a systematic review 62. Docimo S Jr, Mathew A, Shope AJ, Winder JS, Haluck RS, Pauli EM.
and meta-analysis. Surg Endosc. 2015;29(11):3030-3046. Reduced postoperative pain scores and narcotic use favor per-oral
58. Familiari P, Gigante G, Marchese M, et al. Peroral endoscopic endoscopic myotomy over laparoscopic Heller myotomy. Surg Endosc.
myotomy for esophageal achalasia: outcomes of the first 100 patients 2017;31(2):795-800.
with short-term follow-up. Ann Surg. 2016;263(1):82-87.
59. Stavropoulos SN, Modayil R, Friedel D. Per oral endoscopic myotomy
for the treatment of achalasia. Curr Opin Gastroenterol. 2015;31(5):
430-440.
PART FOUR
Gastroesophageal
Reflux Disease
CHAPTER
Gastroesophageal Reflux Disease: Definition and
Scope of the Problem in the United States of 15
America and Worldwide
Joshua Sloan
| Philip O. Katz
G
astroesophageal reflux disease (GERD) is among injury is controversial. In combination with acid there is
the most common diseases seen by both primary likely synergy causing esophagitis. A recent study suggests
care and gastrointestinal (GI) specialists worldwide. that reflux injury may be a result of an inflammatory
Its prevalence is increasing, with reflux symptoms ranging reaction.8 Reflux is cleared by secondary peristalsis and
from 10% to 30% of the population of Western countries.1–3 neutralization by salivary bicarbonate. Failure of either
Compared to the numbers reported in North America, of these clearance mechanisms may contribute to injury.
the prevalence of weekly GERD symptoms is slightly lower In some cases, gastric emptying delay may contribute to
in Europe (8.8% to 25.9%) and substantially lower in GERD by increasing gastric volume and precipitating
Asia (approximately 10%).1,2 Overall, the incidence of TLESRs. Other factors such as central obesity can increase
GERD remains somewhere around 5 per 1000 person intraabdominal pressure and result in GERD.9,10
years.1,2 In this chapter we review the disease as it affects Symptoms of GERD can be divided into typical, atypical,
patients in the United States and around the world; discuss and extraesophageal (Fig. 15.1).4,5 The most common
complications including its relation to strictures, Barrett typical symptoms of GERD are heartburn and regurgita-
esophagus, and esophageal adenocarcinoma; and address tion. Many consider dysphagia and chest pain as typical
GERD’s effect on patient quality of life. symptoms as well.4,5 Atypical symptoms include dyspepsia,
epigastric pain, nausea, bloating, and belching.5 The most
common extraesophageal symptoms include chronic
DEFINITION cough, asthma, and throat symptoms including chronic
GERD is a condition that is defined as troublesome voice disturbance and laryngitis.4,5,11 The latter may be
symptoms or complications that result from the reflux associated with typical symptoms or sometimes may be
of gastric contents into the esophagus or beyond into the the presenting symptom of the disease. Direct aspiration
oral cavity or lung.4,5 Troublesome is defined by consensus of gastric contents are believed to be the most common
as mild symptoms that occur at least 2 times per week or cause of these extraesophageal symptoms, although rarely
moderate to severe symptoms at least once per week.4 The they may result from distal esophageal acid exposure
pathogenesis is multifactorial. Overall, GERD is a disorder alone via a reflex event.
caused by abnormal function of the lower esophageal
sphincter (LES). The most frequent abnormality is an
increase in so-called transient lower esophageal sphincter
SCOPE
relaxations (TLESR), a normal physiologic occurrence of GERD is a common disease affecting millions of people
sphincter opening without an antecedent swallow that worldwide. The prevalence is up to 30% in Western
results in reflux of gastric contents into the esophagus.6,7 countries and 10% in Asia (Table 15.1 and Fig. 15.2).2
Basal LES pressure may be low, and transient increases in These data are based on a number of epidemiologic
intraabdominal pressure and a hiatal hernia contribute to studies that have been performed and reviewed in the
LES dysfunction and reflux. Traditional thinking is that literature (see Table 15.1 and Fig. 15.2).1,2
acidic gastric contents cause symptoms and/or injury via In the United States, the prevalence ranges from
direct contact with mucosa resulting in inflammation. 18.1% to 27.8%.2 A study published in 2007 explored
The role of bile acids in producing symptoms and reflux the prevalence of irritable bowel syndrome (IBS) and
197
Gastroesophageal Reflux Disease: Definition and Scope of the Problem in the United States of America and Worldwide CHAPTER 15 197.e1
ABSTRACT
Gastroesophageal reflux disease (GERD) is a common
condition seen worldwide. The prevalence in North
America and Europe ranges from 8.8% to 30% and is
lower in Asia, ranging anywhere from 3.5% to 8.5%.
GERD also can cause esophageal strictures and Barrett
esophagus, and has a significant impact on patients’
quality of life. Additionally, GERD results in millions of
office visits, missed days of work, and lost productivity. In
this chapter we explore the epidemiology of GERD along
with the full scope of the problem, including the effect
it has on patients’ quality of life and economic burden.
KEYWORDS
Prevalence, Epidemiology, GERD, PPI, Barrett esophagus
198 SECTION I Esophagus and Hernia
GERD.12 The researchers sent a questionnaire to 4194 Affairs (VA) employees.13 The study included a survey
people with 2273 returning the questionnaire.12 The followed by endoscopy. No statistically significant dif-
study found the overall prevalence of GERD to be 18.1% ference in GERD prevalence was found between the
in a group representative of the United States Caucasian two groups (27% and 23% with heartburn, and 16%
population.2,12 The upper range of GERD prevalence in the and 15% for regurgitation in African Americans and
United States of 27.8% was established in a cross-sectional Caucasians, respectively). However, there were higher
survey asking questions pertaining to weekly heartburn rates of esophagitis in Caucasians compared with African
or regurgitation symptoms.1,2,12,13 Americans.13 The prevalence of GERD in Hispanics in the
Studies have been performed in the United States to United States is thought to be similar to non-Hispanic
evaluate if the prevalence of GERD changes based on whites.14 A survey-based study performed in Philadelphia
race. A 2004 cross-sectional survey studied the prevalence found that Hispanics had higher monthly heartburn than
of GERD in African American and Caucasian Veterans other racial groups with a prevalence of 50%.15,16 Asians
surveyed in the same study were found to have a 20%
prevalence of monthly heartburn.15,16
In South America, the prevalence of weekly symptoms
GERD symptoms of heartburn or regurgitation was approximately 23% in
a study published in 2005.2,17 This study highlighted the
prevalence of GERD only in an Argentinian population.2,17
More recently, a Brazilian study echoed findings similar to
the 2004 study.17,18 In that study, monthly GERD symptoms
Typical Atypical Extraesophageal were found to be approximately 22.7%.18
Based on eight studies in two separate epidemiologic
reviews, the prevalence of GERD in Europe ranges from
8.8% to 25.9%.2,19–26 This represents an increased preva-
Dyspepsia
Heartburn Epigastric pain Chronic cough lence from the previously presented data of 9.8% to 18%.1
Regurgitation Nausea Asthma Two studies were large surveys sent via mail to Swedish
Dysphagia Bloating Throat clearing populations, which established the increased prevalence.25,26
Chest pain Belching Laryngitis The prevalence of GERD in East Asian populations
has increased from 2.5%–4.8% to 3.5%–7.8%.1,2 In the
epidemiologic review published in 2005, the prevalence
FIGURE 15.1 Gastroesophageal reflux disease (GERD) symptoms. was based on three studies conducted in mainland China
Sweden
UK
Turkey
USA Spain China
Iran
Israel South Korea
Australia
0.1–5.0%
5.1–10.0%
10.1–15.0% Argentina
15.1–30.0%
FIGURE 15.2 Worldwide epidemiology of gastroesophageal reflux disease. (From El-Serag HB, Sweet S, Winchester CC, Dent J. Update
on the epidemiology of gastro-oesophageal reflux disease: a systematic review. Gut. 2014;63[6]:878.)
Gastroesophageal Reflux Disease: Definition and Scope of the Problem in the United States of America and Worldwide CHAPTER 15 199
and Hong Kong.27–29 Two of the studies were telephone weekly, and daily GERD symptoms were 40%, 20%, and
surveys completed by 3858 individuals; the third was 6%, respectively, with no statistical differences between
a survey of 5000 people and was conducted with the females and males.25 EGD showed normal endoscopic
assistance of physicians and medical students.27–29 More findings, esophagitis, and hiatal hernia in 77%, 15.5%, and
recently, the three studies included in the review by El-Serag 23.9% of subjects, respectively. Of the group with monthly
were conducted in South Korea and China.2 Between GERD symptoms, 35%, 63%, and 53% had normal endos-
the three studies, 20,833 people were surveyed leading copy, esophagitis, and hiatal hernia, respectively. When
to the findings of 3.5% to 7.8% prevalence of GERD.30–32 compared to females, men had more esophagitis (odds
An additional review performed in Korea between 2005 ratio: 2.83), especially in the younger age groups (32%).
and 2010 found that the prevalence in eastern Asia may Compared to asymptomatic individuals, those with GERD
range from 5.2% to 8.5%.33 symptoms, esophagitis, or both reported an increased
Middle Eastern populations also have a significant prevalence of medical treatment—that is, antacids, proton
prevalence of GERD ranging from 8.7% to 33.1%.2,34–40 pump inhibitors (PPIs) or any medication—0% to 3% in
When evaluating the data there was one study in Turkey, normal patients versus 8% to 33% in persons with GERD
one in Israel, and five studies in Iran. The Turkish study symptoms.25
evaluated 630 patients and the prevalence of GERD The ProGERD study assessed symptoms and endoscopic
symptoms, heartburn or regurgitation, ranged from 10% findings in a cohort of GERD patients from Germany,
to 15.6%.34 The study performed in Israel evaluated 981 Switzerland, and Austria with the intent to examine the
patients and found a prevalence of 9.3%.35 Finally, the natural history of the disease. A total of 3894 patients
Iranian studies of GERD showed a prevalence of 8.7% to with GERD symptoms participated in a longitudinal study
21.2% with the exception of one study evaluating nomads undergoing EGD at baseline and at 2-year follow-up.48
in Iran.2,36–40 Similar to the Israeli study, this particular After a baseline EGD, all patients received treatment with
study evaluated only a subgroup of the population and may esomeprazole. Thereafter, further medical treatment was
falsely increase the prevalence in the general population.40 given at the discretion of the physician. After 2 years,
A more recent study suggests that the prevalence of GERD 25% of those with non-erosive reflux disease (NERD)
may range from 14% to 34% in Iran.41 If we include a were noted to have grade A or B esophagitis, whereas 6%
study published in 2014, which surveyed a Saudi Arabian had grade C or D esophagitis. In patients with baseline
population, the overall prevalence range in the Middle grade A and B esophagitis, 1.6% progressed to grade
East ranges from 8.7% to 45.4%.42 C and D, and 61% with baseline A and B esophagitis
The current body of data does not show a difference regressed to NERD (no esophagitis; normal endoscopy)
in prevalence of GERD in aging patients.2,5,43,44 Available at 2 years. Among those with grade C and D esophagitis,
data do, however, suggest that esophageal sensitivity to acid 42% regressed to A and B esophagitis, and 50% of those
decreases with age, thus predisposing those older than 55 with grade C and D regressed to NERD. The study tells
to 65 years to an increase in higher grades of esophagitis us that although the endoscopic findings of GERD are
than younger patients.44,45 There does not appear to be not static, the vast majority of patients with esophagitis
a large difference in GERD symptoms between men and heal and those with NERD remain stable. A second
women. However, some studies suggest women tend to study from Germany reproduced the trend of the above
present with nonerosive disease, whereas men tend to ProGERD study in a smaller group of GERD patients.49
have more esophagitis and Barrett esophagus.2,43 Overall GERD symptom prevalence remains stable
GERD is also prevalent in pregnancy and ranges from over time.
30% to 80% at some point during the course of preg- Multiple logistic regression analyses show male gender,
nancy.46,47 A 2012 study evaluated prevalence throughout increased body mass index, regular alcohol consumption,
the course of pregnancy and found a prevalence of 26.1%, history of GERD longer than 1 year, and smoking were
36.1%, and 51.2% in trimesters one through three. 46 risks for erosive esophagitis. A positive Helicobacter pylori
A separate study found that the incidence of GERD is status was associated with a lower risk.50
approximately 25% in each trimester of pregnancy. 47 The ProGERD study also examined the history of GERD
Little information is available on the end organ effects, medication over a 4-year period.51 PPIs were used in 79%,
if any, of GERD in pregnancy, though clinical experience 84%, 85%, and 87% of patients after 1, 2, 3, and 4 years,
suggests they are minimal. Although some feel that GERD respectively. Continuous PPI administration was needed
in pregnancy is a risk for long-term disease, the data in 53%, 49%, 56%, and 56% of patients after 1, 2, 3, and
supporting this are lacking. GERD symptoms commonly 4 years. On-demand PPI therapy was administered in
resolve with delivery, but may start again at a later date 26%, 35%, 29%, and 29% of patients after 1, 2, 3, and 4
in some patients. years, respectively. The need for continuous PPI treatment
Two longitudinal studies are worth reviewing in some increased with advanced grades of esophagitis. After 1, 2,
detail. The first was a representative random sample of 3, and 4 years, 61%, 56%, 60%, and 60% of those with
the normal population of two communities in northern severe esophagitis at baseline, respectively, remained on
Sweden.25 Subjects were surveyed and a large number (n continuous PPI treatment.51
= 1000) underwent endoscopy. GERD symptoms increased The next analysis of the ProGERD study examined
with age, the lowest being the 20- to 34-year-old group the effect of GERD on the quality of life of the patients
and peaking in the 50- to 65-year-old and older groups. over a 5-year period.52 Over the 5-year period, medical
One thousand participants underwent esophagogastro- treatment improved the categories emotional distress,
duodenoscopy (EGD); the prevalence rates for monthly, sleep disturbances, eating problems, and vitality by 60%
200 SECTION I Esophagus and Hernia
to 69%. In 54% of patients, physical/social function of developing Barrett esophagus.59,61–63 The prevalence of
remained unchanged, whereas it improved in 42%. In Barrett esophagus in the US general population without
all dimensions, clinically relevant worsening was observed GERD symptoms may be as high as 5.6%.60,63,64 In a study
in less than 6% of patients. Impairment of the quality of performed in Sweden, out of 3000 study participants, 1.6%
life could largely be attributed to advanced disease with had Barrett esophagus.65 Out of their study population,
a high symptom load and the perception of nighttime 2.3% with reflux symptoms were found to have Barrett
reflux with sleep disturbances. These patients need more esophagus.65 Having erosive esophagitis increases the
than medical treatment and should be offered surgical likelihood of Barrett esophagus. Additional risk factors for
management of GERD.52 the development of Barrett esophagus are male gender,
In addition to typical symptoms, GERD can impair Caucasian, central obesity, and the duration of GERD
the quality of life because of the generation of so-called symptoms.59,66–71 Several studies have evaluated the inci-
extraesophageal or atypical symptoms (chest pain, cough, dence of progression of nondysplastic Barrett esophagus to
laryngeal symptoms, asthma). Jaspersen et al. examined the esophageal adenocarcinoma and have shown the rate to be
frequency of extraesophageal GERD symptoms in persons anywhere from 0.33% to 0.63% annual risk.72,73 A separate
who participated in the ProGERD study (48% NERD, 52% meta-analysis evaluated 41 studies and found an annual
GERD).53 Extraesophageal symptoms were present in 34.9% incidence of esophageal adenocarcinoma progressing
and 30.5% of GERD and NERD patients, respectively, and from low-grade dysplasia of 0.54%, and an annual risk
included chest pain (14.5%), cough (13%), laryngeal of high-grade dysplasia of 1.73%.74 Finally, the annual
complaints (10.4%), and asthma (4.8%). Except for incidence of developing esophageal adenocarcinoma
asthma, all atypical symptoms were more prevalent in from high-grade dysplasia ranges from 7% to 19%.75,76
GERD. After 5 years, the prevalence of the symptoms The 7% annual risk is based on a meta-analysis of four
remained unchanged, except for asthma, which increased studies that included 236 patients.75 The 19% annual risk
from 4.5% to 7.8%. The resolution of the extraesophageal is based on a study that randomized 127 patients into a
symptoms was independent of erosive disease, typical high-grade dysplasia surveillance arm and an ablation
symptoms, disease duration, and PPI medication.53 Risk arm. The surveillance arm was found to have the 19%
factors for the presence and persistence of symptoms annual incidence of esophageal adenocarcinoma.76 In
included female gender, increased age, more severe patients with Barrett esophagus, the use of PPIs resulted
esophagitis (types C and D), GERD history longer than in a significant decrease in high-grade dysplasia and
1 year, and smoking.54 The important epidemiologic data esophageal adenocarcinoma.59,77
indicate that compared to typical symptoms, extraesopha-
geal GERD symptoms may originate from a different or QUALITY OF LIFE
multifactorial pathogenesis, do not correlate with endo- In addition to the physical complications described earlier,
scopic findings, and do not reliably respond to medical GERD patients have a decrease in their quality of life as a
treatment. result of their disease. A 1998 study evaluated 533 adults
using the Medical Outcomes Study Short Form (SF-36)
and found GERD patients had worse health-related quality
COMPLICATIONS of life than the general population. This study also found
Complications associated with GERD include strictures, that the greatest impact was in areas pertaining to pain,
Barrett esophagus, and adenocarcinoma. In addition, social function, and mental health.78 A subsequent study
significant decreases in quality of life and economic impact used the SF-36 questionnaire to evaluate the effects of
are associated with this disease. nocturnal GERD symptoms compared to the general
population and found subjects with nocturnal symptoms,
STRICTURES again, had a worse quality of life.79 Kulig et al. evaluated
Since the era of PPIs, the incidence of peptic strictures has quality of life in GERD patients using the Reflux Disease
decreased.55 Data from the mid-1980s, pre-PPI use, suggest Questionnaire (RDQ), Quality of Life in Reflux and
a prevalence of esophageal stricture of approximately Dyspepsia (QOLRAD), and SF-36. Their study further
0.07% to 0.12%.55,56 Strictures develop from repeated acid supported the claims that quality of life is negatively
exposure resulting in fibrosis and subsequent narrowing of impacted with GERD, and additionally, treatment with
the esophageal lumen.55,57 With adequate acid suppression, a PPI helped to improve quality of life by controlling
namely by PPIs, stricture formation can be reduced.55,58 symptoms.80 A meta-analysis of 19 studies showed that
However, even today with a wide array of PPIs in the GERD led to a decrease in health-related quality of life,
armamentarium, stricture formation has an incidence of a decrease in physical and mental health, and resulted in
1.1 per 10,000 person-years and 11.1 per 100 person-years increased missed days from work.81 An additional meta-
of stricture recurrence.55 analysis of 9 studies further corroborated the previous
findings that mental and physical health are impacted
BARRETT ESOPHAGUS by GERD symptoms.82
Barrett esophagus is defined as a change from the normal
squamous cell esophageal mucosa to columnar mucosa with ECONOMIC BURDEN
intestinal metaplasia and is a premalignant condition (Fig. Not only does GERD affect quality of life, it also results
15.3).59,60 Intestinal metaplasia can subsequently degenerate in large economic consequences. This is partly a result
further to dysplasia—both low and high grade—and of lost wages from missed days at work and lost pro-
adenocarcinoma. GERD patients have a 10% to 15% risk ductivity due to the decreases in quality of life. Suzuki
Gastroesophageal Reflux Disease: Definition and Scope of the Problem in the United States of America and Worldwide CHAPTER 15 201
A C
B D
FIGURE 15.3 (A) Normal-appearing squamocolumnar junction seen on upper endoscopy. (B) Normal-appearing histology of esophageal
squamous mucosa. (C) Barrett esophagus extending proximal from the gastroesophageal junction. (D) Intestinal metaplasia with arrows
pointing to goblet cells. (From Shaheen NJ, Richter JE. Barrett’s oesophagus. Lancet. 2009;373:851, Figure 1.)
8. Dunbar KB, Agoston AT, Odze RD, et al. Association of acute gas- systematic investigation of gastrointestinal diseases in China. BMC
troesophageal reflux disease with esophageal histologic changes. Gastroenterol. 2010;10:94.
JAMA. 2016;315:2104-2112. 32. Cho YS, Choi MG, Jeong JJ, et al. Prevalence and clinical spectrum
9. Corley DA, Kubo A, Zhao W. Abdominal obesity, ethnicity and of gastroesophageal reflux: a population-based study in Asan-si,
gastro-oesophageal reflux symptoms. Gut. 2007;56:756-762. Korea. Am J Gastroenterol. 2005;100:747-753.
10. El-Serag HB, Ergun GA, Pandolfino J, Fitzgerald S, Tran T, Kramer 33. Jung HK. Epidemiology of gastroesophageal reflux disease in Asia:
JR. Obesity increases oesophageal acid exposure. Gut. 2007;56:749-755. a systematic review. J Neurogastroenterol Motil. 2011;17:14-27.
11. Irwin RS, Curley FJ, French CL. Chronic cough. The spectrum and 34. Kitapcioglu G, Mandiracioglu A, Caymaz Bor C, Bor S. Overlap of
frequency of causes, key components of the diagnostic evaluation, symptoms of dyspepsia and gastroesophageal reflux in the community.
and outcome of specific therapy. Am Rev Respir Dis. 1990;141:640-647. Turk J Gastroenterol. 2007;18:14-19.
12. Jung HK, Halder S, McNally M, et al. Overlap of gastro-oesophageal 35. Sperber AD, Halpern Z, Shvartzman P, et al. Prevalence of GERD
reflux disease and irritable bowel syndrome: prevalence and risk symptoms in a representative Israeli adult population. J Clin Gastro-
factors in the general population. Aliment Pharmacol Ther. enterol. 2007;41:457-461.
2007;26:453-461. 36. Solhpour A, Pourhoseingholi MA, Soltani F, et al. Gastro-esophageal
13. El-Serag HB, Petersen NJ, Carter J, et al. Gastroesophageal reflux reflux symptoms and body mass index: no relation among the
among different racial groups in the United States. Gastroenterology. Iranian population. Indian J Gastroenterol. 2008;27:153-155.
2004;126:1692-1699. 37. Nouraie M, Radmard AR, Zaer-Rezaii H, Razjouyan H, Nasseri-
14. Sharma P, Wani S, Romero Y, Johnson D, Hamilton F. Racial and Moghaddam S, Malekzadeh R. Hygiene could affect GERD prevalence
geographic issues in gastroesophageal reflux disease. Am J Gastroenterol. independently: a population-based study in Tehran. Am J Gastroenterol.
2008;103:2669-2680. 2007;102:1353-1360.
15. Yuen E, Romney M, Toner RW, et al. Prevalence, knowledge and 38. Nouraie M, Razjouyan H, Assady M, Malekzadeh R, Nasseri-
care patterns for gastro-oesophageal reflux disease in United States Moghaddam S. Epidemiology of gastroesophageal reflux symptoms
minority populations. Aliment Pharmacol Ther. 2010;32:645-654. in Tehran, Iran: a population-based telephone survey. Arch Iran Med.
16. Friedenberg FK. GERD in minority populations. Gastroenterol Hepatol 2007;10:289-294.
(N Y). 2011;7:411-413. 39. Nasseri-Moghaddam S, Mofid A, Ghotbi MH, et al. Epidemiological
17. Chiocca JC, Olmos JA, Salis GB, et al. Prevalence, clinical spectrum study of gastro-oesophageal reflux disease: reflux in spouse as a risk
and atypical symptoms of gastro-oesophageal reflux in Argentina: factor. Aliment Pharmacol Ther. 2008;28:144-153.
a nationwide population-based study. Aliment Pharmacol Ther. 40. Mostaghni A, Mehrabani D, Khademolhosseini F, et al. Prevalence
2005;22:331-342. and risk factors of gastroesophageal reflux disease in Qashqai
18. do Rosario Dias de Oliveira Latorre M, Medeiros da Silva A, Chinzon migrating nomads, southern Iran. World J Gastroenterol. 2009;15:
D, Eisig JN, Dias-Bastos TR. Epidemiology of upper gastrointestinal 961-965.
symptoms in Brazil (EpiGastro): a population-based study according 41. Delavari A, Moradi G, Elahi E, Moradi-Lakeh M. Gastroesophageal
to sex and age group. World J Gastroenterol. 2014;20:17388-17398. reflux disease burden in Iran. Arch Iran Med. 2015;18:85-88.
19. Thompson WG, Heaton KW. Heartburn and globus in apparently 42. Almadi MA, Almousa MA, Althwainy AF, et al. Prevalence of symptoms
healthy people. Can Med Assoc J. 1982;126:46-48. of gastroesopahgeal reflux in a cohort of Saudi Arabians: a study
20. Isolauri J, Laippala P. Prevalence of symptoms suggestive of gastro- of 1265 subjects. Saudi J Gastroenterol. 2014;20:248-254.
oesophageal reflux disease in an adult population. Ann Med. 43. Locke GR 3rd, Talley NJ, Fett SL, Zinsmeister AR, Melton LJ 3rd.
1995;27:67-70. Prevalence and clinical spectrum of gastroesophageal reflux: a
21. Valle C, Broglia F, Pistorio A, Tinelli C, Perego M. Prevalence and population-based study in Olmsted County, Minnesota. Gastroenterology.
impact of symptoms suggestive of gastroesophageal reflux disease. 1997;112:1448-1456.
Dig Dis Sci. 1999;44:1848-1852. 44. Becher A, Dent J. Systematic review: ageing and gastro-oesophageal
22. Terry P, Lagergren J, Wolk A, Nyrén O. Reflux-inducing dietary reflux disease symptoms, oesophageal function and reflux oesopha-
factors and risk of adenocarcinoma of the esophagus and gastric gitis. Aliment Pharmacol Ther. 2011;33:442-454.
cardia. Nutr Cancer. 2000;38:186-191. 45. Johnson DA, Fennerty MB. Heartburn severity underestimates erosive
23. Mohammed I, Cherkas LF, Riley SA, Spector TD, Trudgill NJ. Genetic esophagitis severity in elderly patients with gastroesophageal reflux
influences in gastro-oesophageal reflux disease: a twin study. Gut. disease. Gastroenterology. 2004;126:660-664.
2003;52:1085-1089. 46. Malfertheiner SF, Malfertheiner MV, Kropf S, Costa SD, Malfertheiner
24. Diaz-Rubio M, Moreno-Elola-Olaso C, Rey E, Locke GR 3rd, Rodriguez- P. A prospective longitudinal cohort study: evolution of GERD
Artalejo F. Symptoms of gastro-oesophageal reflux: prevalence, symptoms during the course of pregnancy. BMC Gastroenterol.
severity, duration and associated factors in a Spanish population. 2012;12:131.
Aliment Pharmacol Ther. 2004;19:95-105. 47. Rey E, Rodriguez-Artalejo F, Herraiz MA, et al. Gastroesophageal
25. Ronkainen J, Aro P, Storskrubb T, et al. High prevalence of gastro- reflux symptoms during and after pregnancy: a longitudinal study.
esophageal reflux symptoms and esophagitis with or without symptoms Am J Gastroenterol. 2007;102:2395-2400.
in the general adult Swedish population: a Kalixanda study report. 48. Labenz J, Nocon M, Lind T, et al. Prospective follow-up data from
Scand J Gastroenterol. 2005;40:275-285. the ProGERD study suggest that GERD is not a categorial disease.
26. Lofdahl HE, Lane A, Lu Y, et al. Increased population prevalence Am J Gastroenterol. 2006;101:2457-2462.
of reflux and obesity in the United Kingdom compared with Sweden: 49. Bajbouj M, Reichenberger J, Neu B, et al. A prospective multicenter
a potential explanation for the difference in incidence of esophageal clinical and endoscopic follow-up study of patients with gastro-
adenocarcinoma. Eur J Gastroenterol Hepatol. 2011;23:128-132. esophageal reflux disease. Z Gastroenterol. 2005;43:1303-1307.
27. Pan G, Xu G, Ke M, et al. Epidemiological study of symptomatic 50. Labenz J, Jaspersen D, Kulig M, et al. Risk factors for erosive
gastroesophageal reflux disease in China: Beijing and Shanghai. esophagitis: a multivariate analysis based on the ProGERD study
Chin J Dig Dis. 2000;1:2-8. initiative. Am J Gastroenterol. 2004;99:1652-1656.
28. Hu WH, Wong WM, Lam CL, et al. Anxiety but not depression 51. Nocon M, Labenz J, Jaspersen D, et al. Long-term treatment of
determines health care-seeking behaviour in Chinese patients with patients with gastro-oesophageal reflux disease in routine care—
dyspepsia and irritable bowel syndrome: a population-based study. results from the ProGERD study. Aliment Pharmacol Ther.
Aliment Pharmacol Ther. 2002;16:2081-2088. 2007;25:715-722.
29. Wong WM, Lai KC, Lam KF, et al. Prevalence, clinical spectrum and 52. Nocon M, Labenz J, Jaspersen D, et al. Health-related quality of life
health care utilization of gastro-oesophageal reflux disease in a in patients with gastro-oesophageal reflux disease under routine
Chinese population: a population-based study. Aliment Pharmacol care: 5-year follow-up results of the ProGERD study. Aliment Pharmacol
Ther. 2003;18:595-604. Ther. 2009;29:662-668.
30. Chen M, Xiong L, Chen H, Xu A, He L, Hu P. Prevalence, risk 53. Jaspersen D, Nocon M, Labenz J, et al. Clinical course of laryngo-
factors and impact of gastroesophageal reflux disease symptoms: a respiratory symptoms in gastro-oesophageal reflux disease during
population-based study in South China. Scand J Gastroenterol. routine care—a 5-year follow-up. Aliment Pharmacol Ther.
2005;40:759-767. 2009;29:1172-1178.
31. He J, Ma X, Zhao Y, et al. A population-based survey of the epidemiol- 54. Jaspersen D, Kulig M, Labenz J, et al. Prevalence of extra-oesophageal
ogy of symptom-defined gastroesophageal reflux disease: the manifestations in gastro-oesophageal reflux disease: an analysis
Gastroesophageal Reflux Disease: Definition and Scope of the Problem in the United States of America and Worldwide CHAPTER 15 203
based on the ProGERD Study. Aliment Pharmacol Ther. 72. Desai TK, Krishnan K, Samala N, et al. The incidence of oesophageal
2003;17:1515-1520. adenocarcinoma in non-dysplastic Barrett’s oesophagus: a meta-
55. Ruigomez A, Garcia Rodriguez LA, Wallander MA, Johansson S, analysis. Gut. 2012;61:970-976.
Eklund S. Esophageal stricture: incidence, treatment patterns, and 73. Sikkema M, de Jonge PJ, Steyerberg EW, Kuipers EJ. Risk of esopha-
recurrence rate. Am J Gastroenterol. 2006;101:2685-2692. geal adenocarcinoma and mortality in patients with Barrett’s
56. Fennerty MB. The continuum of GERD complications. Cleve Clin J esophagus: a systematic review and meta-analysis. Clin Gastroenterol
Med. 2003;70(suppl 5):S33-S50. Hepatol. 2010;8:235-244; quiz e.32.
57. Spechler SJ. Clinical manifestations and esophageal complications 74. Singh S, Manickam P, Amin AV, et al. Incidence of esophageal
of GERD. Am J Med Sci. 2003;326:279-284. adenocarcinoma in Barrett’s esophagus with low-grade dysplasia: a
58. Marks RD, Richter JE, Rizzo J, et al. Omeprazole versus H2-receptor systematic review and meta-analysis. Gastrointest Endosc. 2014;79:897-909.
antagonists in treating patients with peptic stricture and esophagitis. e4; quiz 983.e1, 983.e3.
Gastroenterology. 1994;106:907-915. 75. Rastogi A, Puli S, El-Serag HB, Bansal A, Wani S, Sharma P. Incidence
59. Shaheen NJ, Falk GW, Iyer PG, Gerson LB, American College of of esophageal adenocarcinoma in patients with Barrett’s esophagus
Gastroenterology. ACG Clinical guideline: diagnosis and management and high-grade dysplasia: a meta-analysis. Gastrointest Endosc.
of Barrett’s esophagus. Am J Gastroenterol. 2016;111:30-50. 2008;67:394-398.
60. Shaheen NJ, Richter JE. Barrett’s oesophagus. Lancet. 2009;373: 76. Shaheen NJ, Sharma P, Overholt BF, et al. Radiofrequency ablation
850-861. in Barrett’s esophagus with dysplasia. N Engl J Med. 2009;360:2277-2288.
61. Halland M, Katzka D, Iyer PG. Recent developments in pathogenesis, 77. Singh S, Garg SK, Singh PP, Iyer PG, El-Serag HB. Acid-suppressive
diagnosis and therapy of Barrett’s esophagus. World J Gastroenterol. medications and risk of oesophageal adenocarcinoma in patients
2015;21:6479-6490. with Barrett’s oesophagus: a systematic review and meta-analysis.
62. Dent J. Barrett’s esophagus: a historical perspective, an update on Gut. 2014;63:1229-1237.
core practicalities and predictions on future evolutions of manage- 78. Revicki DA, Wood M, Maton PN, Sorensen S. The impact of gas-
ment. J Gastroenterol Hepatol. 2011;26(suppl 1):11-30. troesophageal reflux disease on health-related quality of life. Am J
63. Sharma P. Clinical practice. Barrett’s esophagus. N Engl J Med. Med. 1998;104:252-258.
2009;361:2548-2556. 79. Farup C, Kleinman L, Sloan S, et al. The impact of nocturnal
64. Rex DK, Cummings OW, Shaw M, et al. Screening for Barrett’s symptoms associated with gastroesophageal reflux disease on health-
esophagus in colonoscopy patients with and without heartburn. related quality of life. Arch Intern Med. 2001;161:45-52.
Gastroenterology. 2003;125:1670-1677. 80. Kulig M, Leodolter A, Vieth M, et al. Quality of life in relation to
65. Ronkainen J, Aro P, Storskrubb T, et al. Prevalence of Barrett’s symptoms in patients with gastro-oesophageal reflux disease—an
esophagus in the general population: an endoscopic study. Gastro- analysis based on the ProGERD initiative. Aliment Pharmacol Ther.
enterology. 2005;129:1825-1831. 2003;18:767-776.
66. Rubenstein JH, Mattek N, Eisen G. Age- and sex-specific yield of 81. Tack J, Becher A, Mulligan C, Johnson DA. Systematic review: the
Barrett’s esophagus by endoscopy indication. Gastrointest Endosc. burden of disruptive gastro-oesophageal reflux disease on health-
2010;71:21-27. related quality of life. Aliment Pharmacol Ther. 2012;35:1257-1266.
67. Kubo A, Cook MB, Shaheen NJ, et al. Sex-specific associations 82. Becher A, El-Serag H. Systematic review: the association between
between body mass index, waist circumference and the risk of symptomatic response to proton pump inhibitors and health-related
Barrett’s oesophagus: a pooled analysis from the international quality of life in patients with gastro-oesophageal reflux disease.
BEACON consortium. Gut. 2013;62:1684-1691. Aliment Pharmacol Ther. 2011;34:618-627.
68. Singh S, Sharma AN, Murad MH, et al. Central adiposity is associated 83. Suzuki H, Matsuzaki J, Masaoka T, Inadomi JM. Greater loss of
with increased risk of esophageal inflammation, metaplasia, and productivity among Japanese workers with gastro-esophageal reflux
adenocarcinoma: a systematic review and meta-analysis. Clin Gastro- disease (GERD) symptoms that persist vs resolve on medical therapy.
enterol Hepatol. 2013;11:1399-1412, e7. Neurogastroenterol Motil. 2014;26:764-771.
69. Cook MB, Wild CP, Forman D. A systematic review and meta-analysis 84. Menees SB, Guentner A, Chey SW, Saad R, Chey WD. How do US
of the sex ratio for Barrett’s esophagus, erosive reflux disease, and gastroenterologists use over-the-counter and prescription medications
nonerosive reflux disease. Am J Epidemiol. 2005;162:1050-1061. in patients with gastroesophageal reflux and chronic constipation?
70. Lieberman DA, Oehlke M, Helfand M. Risk factors for Barrett’s Am J Gastroenterol. 2015;110:1516-1525.
esophagus in community-based practice. GORGE consortium. 85. Peery AF, Crockett SD, Barritt AS, et al. Burden of gastrointestinal,
Gastroenterolgy Outcomes Research Group in Endoscopy. Am J liver, and pancreatic diseases in the United States. Gastroenterology.
Gastroenterol. 1997;92:1293-1297. 2015;149:1731-1741.e3.
71. Thrift AP, Kramer JR, Qureshi Z, Richardson PA, El-Serag HB. Age 86. Peery AF, Dellon ES, Lund J, et al. Burden of gastrointestinal disease
at onset of GERD symptoms predicts risk of Barrett’s esophagus. in the United States: 2012 update. Gastroenterology. 2012;143:1179-1187,
Am J Gastroenterol. 2013;108:915-922. e1-3.
CHAPTER
G
astroesophageal reflux disease (GERD) is the most not the patient’s symptoms come under control with the
common foregut disease in the world and accounts escalation of the PPI dose, the possibility of progressive
for approximately 75% of all esophageal pathology.1 disease remains and must be assessed.
The majority of affected patients have mild disease and
are successfully managed with lifestyle modification and
acid suppression medication.2 Fortunately, progression PROGRESSION OF GASTROESOPHAGEAL
to erosive disease occurs in only 13% of patients over 5 REFLUX DISEASE UNDER THERAPY
years.3 Unfortunately, progression to Barrett esophagus
(BE), the premalignant lesion for adenocarcinoma of Two scenarios have been proposed regarding the natural
the esophagus, occurs in 10% of patients over 5 years.3 history of GERD. The first is that GERD is a categorical
This has led to concerns that proton pump inhibitor disorder, and the patient can be categorized as having
(PPI) therapy does not directly address the underlying either nonerosive reflux disease (NERD) or erosive reflux
cause of the disease. It is estimated that more than 113 disease (ERD) and patients remain in their diagnosed
million PPI prescriptions are filled globally each year.4 category.10 The second is that GERD is a spectrum disorder
A 10% progression to BE every 5 years is therefore an with NERD at one end of the spectrum and BE and
enormous problem. esophageal adenocarcinoma at the other end, with the
GERD is currently defined as a condition that develops ability of the disease to progress through the spectrum
when the reflux of stomach contents into the esophagus over time.11 Current clinical evidence appears to support
causes troublesome symptoms such as heartburn and/or the spectrum concept in that there are several studies
regurgitation.5 This symptomatic definition is not a precise showing progression of patients from their initial category
guide to the presence of disease because there is not to a more advanced category.12
always a clear correlation between reflux symptoms and One of the largest studies of GERD progression was the
objective evidence of the disease, such as esophagitis or ProGERD study involving 2721 patients from Germany,
increased esophageal acid exposure on 24-hour esophageal Switzerland, and Austria.13 Patients were categorized
pH monitoring. Further, patients may have typical reflux endoscopically as having NERD or ERD based on the
symptoms in the absence of endoscopic esophagitis or have Los Angeles (LA) classification. Patients with BE were
endoscopic esophagitis in the absence of typical reflux excluded. The study consisted of an initial endoscopy to
symptoms.6 In both situations a 24-hour esophageal pH categorize the patients, followed by 4 to 8 weeks of PPI
monitoring study is necessary to confirm the presence therapy, followed by maintenance therapy provided by the
of disease. patients’ primary care physician. A follow-up endoscopy was
In practice, it is common for primary care physicians performed at 2 and 5 years. Progression to BE, confirmed
to treat patients with GERD symptoms on their initial by endoscopy and biopsy during the 5 years of therapy,
visit with a trial of PPIs. If symptoms are relieved, they was observed in 5.9% of patients with NERD, 12.1% of
accept that the diagnosis of GERD is confirmed despite patients with ERD (LA grade A/B), and 19.7% of patients
studies showing that the “PPI trial” has a low accuracy for with severe ERD (LA grade C/D). Patients with severe
identifying the patient with GERD.7,8 In the absence of a ERD on initial endoscopy had the highest incidence of
complete response to a trial of PPIs, it is recommended progression to BE. Overall, 10% of patients progressed to
that the dose of PPIs be doubled.9 If this does not lead to BE during the 5-year follow-up period. This study clearly
symptom resolution, a 24-hour esophageal pH monitoring showed that disease progression occurred in a proportion
study should be performed off medication to measure of patients while receiving PPI therapy and established
the esophageal acid exposure and confirm the diagnosis that, although regular and consistent PPI therapy can
of GERD. If the study is positive, it is recommended that improve symptoms and heal erosive esophagitis, it does
the PPI dose be further increased or an alternative PPI not stop progression to BE.
prescribed.9 This approach has popularized the concept The second study consisted of a group of 33 patients
that patients with persistent symptoms while on PPI therapy who were referred to a gastrointestinal clinic in Milan, Italy
are undermedicated. As a consequence, the possibility of with the diagnosis of NERD.14 Patients were endoscopically
progressive disease occurring under PPI therapy is rarely normal but had an abnormal 24-hour esophageal pH
considered. A more prudent conclusion is that whether or monitoring study at baseline. The patients were observed
204
Etiology and Natural History of Gastroesophageal Reflux Disease and Predictors of Progressive Disease CHAPTER 16 204.e1
ABSTRACT
Objective and background: The etiology and natural
history of gastroesophageal reflux disease (GERD) is
discussed in regard to the reported progression of GERD
in patients while on proton pump inhibitor (PPI) therapy.
The probability of this occurring has led to concern that
PPI therapy does not address all aspects of the disease and
suggests that in selected patients a minimally invasive surgi-
cal procedure may be required for complete therapeutic
control of the disease.
Study design and methods: Review the literature on
GERD and its progression under PPI therapy. If verified,
design a plan to prevent progression.
Results: Existing literature supports that GERD can
progress under PPI therapy and is likely caused by inflam-
matory injury to the lower esophageal sphincter (LES).
Current treatment of progression is elevation of the PPI
dose without investigating the condition of the LES.
Further, endoscopic biopsies of the squamocolumnar
junction (SCJ) within the LES can show microscopic
intestinalized metaplastic cardiac mucosa produced
by reflux-induced inflammatory injury of the original
squamous mucosa. This finding is predictive of future
endoscopically visible Barrett esophagus—the precursor
of esophageal adenocarcinoma. A proposed algorithm
to avoid progression of GERD under therapy using LES
augmentation is presented.
Conclusions: Patients who partially respond or fail to
respond to PPI therapy should undergo biopsies of the
SCJ to identify histologic changes predictive of future
development of visible Barrett esophagus. Intervention
using an LES augmentation procedure is a potential form
of effective therapy.
KEYWORDS
GERD
PPI
Progression
Micro-Barrett
LES
Augmentation
Etiology and Natural History of Gastroesophageal Reflux Disease and Predictors of Progressive Disease CHAPTER 16 205
for 10 years. During the first 5 years of follow-up, 18 patients effects of LES incompetency on esophageal exposure to
had a repeat endoscopy and 17 (94.4%) had esophagitis. acid and bile.17 Fifty symptomatic consecutive patients
When active PPI therapy was discontinued after 10 years were identified for each of the four stages of GERD from
of follow-up, symptoms relapsed in 96.6% of the available the preoperative records of those who had a laparoscopic
patients (28/29). This study showed that GERD of all Nissen fundoplication performed by us. The stages were
severities requires long-term medical therapy. It further (1) NERD, (2) mild ERD, defined as “healable esophagitis”
showed the progressive capability of NERD and that the with PPI therapy, (3) severe ERD, defined as “difficult
absence of endoscopic esophagitis at presentation is not to heal esophagitis” that persisted despite PPI therapy,
a positive prognostic factor. and (4) BE. Exclusion criteria were normal preoperative
A third study of 40 Swedish patients with GERD, esophageal acid exposure on pH monitoring, esophageal
confirmed by an abnormal esophageal acid exposure pH monitoring performed elsewhere, previous antireflux
on 24-hour pH monitoring, showed that when progres- surgery, and a named esophageal motility disorder or a
sion occurred during PPI treatment, it was associated low contraction amplitude in distal half of the esophagus.
with the development of manometric abnormalities of Patients who could not be contacted for study approval
the lower esophageal sphincter (LES).15 Patients in the were also excluded. All patients’ records contained a
study underwent endoscopy, esophageal manometry, and detailed preoperative clinical questionnaire, and all
24-hour esophageal pH monitoring at the beginning underwent a preoperative upper gastrointestinal endos-
and end of a 21-year follow-up period. At baseline, 24 copy, esophageal manometry, and distal esophageal pH
patients had NERD and 16 ERD. None had endoscopic or monitoring. Patients who had received PPI therapy prior
histologic evidence of BE. Of the 24 patients with NERD, to their initial endoscopy were excluded from the NERD
a baseline 14 progressed to ERD and 10 (41.7%) to BE. group. BE was diagnosed by the presence of microscopic
Of the 16 patients with ERD at baseline, 8 developed BE intestinal metaplasia on biopsy of an endoscopic visible
(50%). Overall, 18 of 40 (45%) progressed to BE over columnar-lined esophagus of any length. The final studied
the 21-year follow-up period. Progression was associated population consisted of 39 patients with NERD, 42 with
with a significantly shorter LES mean intraabdominal mild ERD (i.e., healable esophagitis), 35 with severe ERD
length (P = .01) and a significantly greater esophageal (i.e., difficult to heal esophagitis), and 44 with BE.
acid exposure on pH monitoring (P = .004) compared The significant anatomic and physiologic differences
with patients who did not progress. Furthermore, the study between the patient groups are shown in Fig. 16.1. The dif-
population showed a trend toward increased use of PPIs ferences between the patient groups “healable esophagitis”
over the 21-year follow-up period, and an increase in the and the “difficult to heal esophagitis” was the status of
number of patients who developed erosive esophagitis. their LES. Preoperative mechanical factors (i.e., altered
These results indicate that patients with a long duration of hiatal anatomy, LES resting pressure, and LES lengths)
GERD are more likely to progress despite PPI treatment, were significantly more impaired in patients with “difficult
likely due to deterioration of the LES during the course to heal esophagitis” and BE compared with those with
of therapy. “healable esophagitis” and NERD. Esophageal acid and
The previous study was the first to introduce the concept bile exposure also was worse in the more severe GERD
that the progression of GERD while on PPI therapy was stages with the most severe in patients with BE. The
likely due to progressive LES damage during therapy. composite pH score, which includes all the acid reflux
In practice the byword became, the greater the LES measures in a weighted calculation of reflux severity,
damage, the less effective the PPI therapy. This concept was discriminated most clearly between the different GERD
examined prospectively in a study of GERD patients with stages. These findings support the importance of LES
different degrees of LES and esophageal body functionality length and resting pressure in the etiology and severity
prior to therapy.16 A damaged LES was defined as a pressure of GERD. The findings also link the extent and severity of
less than 8 mm Hg and/or a LES abdominal length less endoscopic mucosal injury with the extent and severity of
than 1.2 cm. More than 20% ineffective peristaltic contrac- mechanical abnormalities at the gastroesophageal barrier.
tions were used to indicate a compromised esophageal Furthermore, they suggest that progression of GERD in a
body. PPI failure, shown by the recurrence of symptoms patient on PPI therapy commonly requires a concomitant
or esophagitis, occurred in 7.7% (2/26) of patients with a reduction in LES length and pressure and altered hiatal
normal LES and normal esophageal body, 38.1% (24/63) anatomy. These findings are similar to other studies of
of patients with a damaged LES and normal esophageal regression analysis which show that the status of the LES
body, and 79.5% (31/39) of patients with a damaged and size of the hiatal hernia are dominant determinants
LES and a compromised esophageal body. These results of esophagitis and its severity.18,19
strongly indicated that PPI therapy was less effective in Taken together, the above studies show most importantly
patients with a damaged LES than in those with a normal the following: (1) the treatment of GERD with PPIs does
LES and that a compromised esophageal body added to not prevent progression of disease, (2) the stages of GERD
their ineffectiveness. severity correlate well to the altered mechanical features
In an effort to understand further the effects of of the gastroesophageal reflux barrier, (3) damage to the
mechanical factors in the progression of GERD under LES is associated with altered hiatal anatomy, increased
PPI therapy, we studied the existence of mechanical esophageal exposure to refluxed acid and bile, and (4)
abnormalities in the spectrum of GERD. This included PPI therapy does not protect against continuing LES
factors such as anatomic distortion of the hiatal anatomy damage. These findings encourage the concept that to stop
by a hiatal hernia, abnormalities of the LES, and the the progression of GERD requires (1) early recognition
206 SECTION I Esophagus and Hernia
Status of the LES and composition of the refluxed gastric juice in GERD progression
2 to 20 years
FIGURE 16.1 Status of the lower esophageal sphincter (LES), esophageal acid exposure, esophageal bile exposure, and the anatomic
distortion of the hiatal anatomy by a hiatal hernia in the spectrum of gastroesophageal reflux disease (GERD) categories: nonerosive reflux
disease (NERD), “healable esophagitis,” “difficult to heal esophagitis,” and Barrett esophagus (BE). Significant anatomic and physiologic
differences existed between the categories of healable esophagitis and difficult to heal esophagitis. BE differs from difficult to heal
esophagitis only by the degree of esophageal acid and bile exposure. High bile exposure was unique to BE. NS, No significance. (From
Lord RV, DeMeester SR, Peters JH, et al. Hiatal hernia, lower esophageal sphincter incompetence, and effectiveness of Nissen
fundoplication in the spectrum of gastroesophageal reflux disease. J Gastrointest Surg. 2009;13:602–610.)
of the symptoms and signs of progressive disease, (2) was not of primary importance in the prevention of reflux
manometric assessment of the LES, (3) measurements of because there was no direct relationship between its
esophageal acid exposure, (4) endoscopic examination pressure and the existence of reflux.21 This, along with
of the esophagus, and, if indicated, (5) early surgical other similar reports, shelved the LES as a curiosity.
intervention to correct the LES abnormalities. To do so The introduction of 24-hour esophageal pH monitoring
requires an understanding of the pathophysiology and resurfaced an interest to further investigate the LES as
the histopathology of GERD. a barrier to reflux. The renewed interest resulted in
studies of normal subjects and symptomatic GERD patients
with esophageal manometry and 24-hour esophageal pH
PATHOPHYSIOLOGY OF monitoring.22 The results of these studies indicated there
GASTROESOPHAGEAL REFLUX DISEASE were three characteristics of the force, or LES, that worked
together to provide a barrier to challenges of increased
There are two fundamental determinants of GERD: the intragastric and intraabdominal pressure (Fig. 16.2A).
status of the LES and the composition of the gastric juice One of these characteristics was the position of the LES.
refluxed into the esophagus. Of the two, the status of the A portion of the LES length is normally exposed to the
LES is the primary determinant. The LES is a measurable positive intraabdominal pressure environment and is
force located in the distal end of the esophagus where it commonly referred to as the abdominal length of the LES.23
straddles the diaphragm. Anatomically its boundaries are During periods of increased intraabdominal pressure,
unmarked and its exact location can only be determined the resistance of the LES could easily be overcome if the
by using a pressure-sensitive catheter. The LES normally challenge of intraabdominal pressure was unable to be
remains closed with two exceptions: during a swallow, when applied equally to the abdominal esophagus (i.e., synony-
an induced reflex relaxation of the force occurs to allow mous with abdominal length of LES) and stomach.24–26 In
a bolus of food to enter the stomach, or during a belch, situations in which the length of the abdominal esophagus
when the force is disrupted to allow gas to be vented has become permanently inadequate, the abdominal
from a distended stomach. The common denominator esophagus cannot collapse in response to challenges of
for virtually all episodes of gastroesophageal reflux is the intraabdominal pressure. This is called permanent failure
disappearance of the force. When this occurs, resistance of the LES (Fig. 16.2B), and with the encouragement from
to the flow of gastric juice from an environment of higher the negative intrathoracic pressure, results in persistent
pressure—the stomach—to an environment of lower reflux of gastric juice into the esophagus with minimal
pressure—the esophagus—is lost. In early disease the loss abdominal pressure challenges.
of the force is a transient event. In advanced disease, the The remaining two characteristics of the LES, overall
loss or deficiency of the force is permanent. length of the LES and the LES pressure, also provide a
Dr. Charles Code discovered the force (i.e., the “high barrier to reflux. Both function together and depend
pressure zone,” as he referred to it) in 1956.20 The impor- on each other to provide resistance to flow of gastric
tance of its discovery in regard to GERD was nullified by Dr. juice from the stomach into the esophagus unrelated
Campbell McLaurin from Scotland. After a fairly extensive to intraabdominal pressure challenges. Critical to this
study, McLaurin concluded that the “high pressure zone” function is the relationship between the LES overall length
Etiology and Natural History of Gastroesophageal Reflux Disease and Predictors of Progressive Disease CHAPTER 16 207
LES pressure profile FIGURE 16.2 (A) The pressure profile of the lower esophageal
sphincter (LES) in normal subjects. Note the long intraabdominal
length identified by the positive respiratory excursions. The
abdominal length ends at the respiratory inversion point where the
= 3 mm Hg respiratory excursions change from positive excursions with
breathing to negative excursion with breathing. A short
Respiratory intrathoracic portion of the LES is identified by the negative
inversion point respiratory excursions. Adequate overall length, abdominal length,
Gastric baseline and pressure are necessary for the LES to function as a barrier to
reflux. (B) An illustration of permanent failure of the LES. This is
Abdominal length due to the permanent loss of abdominal LES length. This
Overall length compromises the ability of the LES to prevent reflux caused by
Measurements with HRM see: Ayazi S, et al. Gastrointest Surg 2009; 13: 2113–2120
intraabdominal pressure challenges. (C) An illustration of dynamic
failure of the LES. The overall and abdominal length of the LES
A shortens with gastric distention and returns back to its original
Permanent failure of the LES length when the distention is relieved by a burp. As the LES
shortens due to distention or nonpressurized gastric dilation, it
reaches a point where the ratio of length to pressure is insufficient
to keep the LES closed, allowing gastroesophageal reflux to
= 3 mm Hg occur. When the stomach is decompressed or gastric dilation
recedes, the length of the LES returns to normal and competency
Destroyed portion
of LES
Respiratory is reestablished.
inversion point
Gastric baseline
Abdominal length
Overall length
5 4 3 2 1 0 24
LES length in cm
B
LES pressure (mm Hg)
18
Dynamic failure of the LES
Competent
= 3 mm Hg
12
Gastric 6 Incompetent
baseline
0
0 1 2 3 4 5
Gastric distention or dilation shortens the LES LES length (cm)
5 4 3 2 1 0
LES length in cm
C FIGURE 16.3 The relationship between lower esophageal sphincter
(LES) pressure and overall length to its competency. Excessive
eating results in gastric distention and dilation, which can shorten
and the LES pressure.27 The shorter the overall length, the length of the LES to the point where its existing pressure is
ineffective in keeping the circumferential wall of the LES
the higher the pressure must be for the LES to maintain
approximated. When this occurs, the walls of the LES separate,
competency against challenges of intragastric pressure
competency of the LES is lost, and reflux occurs.
above intraabdominal pressure (Fig. 16.3). Consequently,
a normal LES pressure can be nullified by a short overall
LES length, and a normal overall LES length can be
nullified by a low LES pressure. The fundamental principle voluminous gastric distention, the overall length of the LES
is that the overall length of the LES, similar to the LES shortens to a point at which the existing LES pressure is
abdominal length, is important to the function of the LES unable to maintain its closure and belching and/or reflux
as a barrier.28–30 Shortening of the overall length of the occurs. As a consequence the patient’s gastric distention
LES occurs naturally with gastric filling because the distal is relieved, effacement of the LES retreats, LES regains
end of the LES is “taken up” by the expanding fundus. its overall length, and competency of the LES is restored.
This is similar to the shortening of the neck of a balloon This is called dynamic failure of the LES because it is
as it is inflated. During the process the distal portion of reversible after the distention is relieved (see Fig. 16.2C).
the neck of the balloon becomes part of the expanded If the length of the LES becomes permanently short (see
body of the balloon. The process is called effacement Fig. 16.2B), then further shortening caused by normal
and also occurs with gastric distention (Fig. 16.4). With amounts of gastric distention with meals will readily result
208 SECTION I Esophagus and Hernia
in postprandial reflux and gastroesophageal reflux becomes of failure. The latter is the value for a specific component
an ever-constant clinical problem.31 at which esophageal acid exposure becomes abnormal
If esophageal manometry, done at rest, in the recumbent independent from the values of the other components.
position, and after an overnight fast shows the LES to The refocus on the LES also led to the realization
have an abnormally low pressure, a short overall length, that almost half of the patients with confirmed GERD by
or a minimal length exposed to the abdominal pressure 24-hour esophageal pH monitoring have a normal LES
environment, it is called a permanent failure of the LES, on a motility study performed at rest, in the recumbent
and unhampered reflux of the gastric contents into the position, and after an overnight fast. 28 The proposed
esophagus occurs (see Fig. 16.2B). The existence of a etiology of reflux in such patients was described by Dr.
permanently failed LES is identified by one or more Jerry Dodds in 1982.32 He observed transient relaxations
inadequate components: an average pressure of 6 mm of a normal LES that were unrelated to a swallow but were
Hg or less, an average overall length of 2 cm or less, or stimulated by gastric distention or nonpressurized gastric
an average abdominal length of 1 cm or less. Compared dilation.33 The term he used to describe these events was
with normal subjects, these values are less than the 2.5 “transient LES relaxations” (tLESRs).
percentile for each parameter.28 Fig. 16.5 is a schema of There are two proposed explanations for the occurrence
the LES components showing their median normal values of tLSERs. The first is, tLESRs are due to a neuromediated
with the fifth and 95th percentile values and their point reflex initiated by pressurized gastric distention or nonpres-
surized dilation from gastric adaptive relaxation induced
by a meal.34 It was hypothesized that these conditions
stimulate stretch receptors in the gastric fundus, which
Effacement of the LES in turn stimulate vagal afferent nerve fibers that relay
the input from the receptors to the medulla. Medullary
nuclei then orchestrate the efferent limb of the reflex
via the vagal and phrenic nerves to elicit a tLESR, crural
diaphragm inhibition, and distal esophageal shortening.35
This explanation suggests that the basic etiology of GERD
is a neuromuscular abnormality.
The second proposal is, tLESRs are due to transient
shortening of the LES length due to it being taken up
into the stomach during episodes of pressurized gastric
distention or nonpressurized dilation from gastric adaptive
relaxation.36 Normally, in the fasting state and resting
recumbent position, the median overall length of the LES
Distension or dilation
is 3.6 cm.28 With gastric distention or dilation, the length
FIGURE 16.4 The process of lower esophageal sphincter (LES) of the LES shortens as the effaced portion is taken up by
effacement occurs with gastric distention or nonpressurized the expanding fundus. When gastric distention or dilation
dilation. The length of the distal esophagus and the LES within it is excessive, the length of the LES shortens to its point of
shorten as they are effaced into and taken up by the fundus of failure of 2 cm or less. At this length the corresponding
the distended stomach. As a consequence, the squamous pressure of the LES can no longer maintain closure, the
mucosa (red) is exposed to the gastric juice and suffers LES opens, and gastroesophageal reflux occurs.27,30,37 This
inflammatory injury. (From Ayazi S, Tamhankar R, DeMeester TR, occurs predominately during the postprandial period31 and
et al. The impact of gastric distension on the lower esophageal is called “dynamic failure of the LES” in that when the
sphincter and its exposure to acid gastric juice. Ann Surg. gastric distention or dilation is relieved the LES returns
2010;252:52-62.) to its normal initial length.
Understanding the concept of effacement is crucial and mucosal injury.38 The data in Fig. 16.7 further support
to comprehending the pathophysiology of GERD. Not this suggestion by showing the higher the esophageal acid
all effacement events result in the reflux of gastric juice exposure, the more extensive the damage to the LES as
through an open LES into the esophagus. During most assessed by the number of failed LES components per
effacement events, only the distal end of the LES with its patient. In summary, the greater the mucosal injury, the
squamous epithelial covering is taken up by the fundus greater is the LES injury, resulting in greater esophageal
and exposed to gastric juice while the proximal squa- acid exposure and the necessity for a higher dose of
mous epithelium remains protected.36 Inflammation and acid suppression therapy to control symptoms and heal
ulceration does occur in the portion of the squamous epithelial and LES damage. The question is, does PPI
epithelium exposed to gastric juice.36 If the inflammation therapy prevent progressive damage to the LES or does
continues due to repetitive episodes of effacements, it progressive damage occur to the LES despite PPI therapy?
can result in permanent damage with reduction of the The only guidance we have to answer this question is
abdominal length of the LES to 1 cm or less, limiting its the study of Falkenback et al.15 discussed previously, on
ability to respond to intraabdominal pressure challenges the course of GERD in 40 patients treated with PPIs and
and allowing unhampered reflux into the esophagus. followed for 20 years. Over the period of observation
Similarly, persistent inflammation can result in permanent the studied population showed more use of PPI therapy
damage with reduction of the overall length of the LES (P = .007), and an increasing prevalence of esophagitis
to 2 cm or less, limiting its ability to resist intragastric (P = .001) and BE (P = .002). While on PPI therapy,
distention or nonpressurized gastric dilation. In both those who progressed differed from those who did not
situations a “transient failure of the LES” has advanced to progress by a 1-cm shortening of their LES abdominal
a “permanent failure of the LES” due to the permanent length (P < .01) and an increase of 11% in the time the
loss of a crucial amount of LES’s abdominal and/or esophageal pH was less than 4 during the supine period
overall length secondary to inflammatory injury. The (P < .004). This supports the concept that progressive
last component of the LES to go is pressure. The loss disease, despite increases in PPI dose, is associated with
in pressure is due to injury of the complete underlying manometric abnormalities of the LES that develop during
muscle by inflammatory by-products from the overlying the course of PPI treatment. The logical inference is that
inflamed squamous epithelium. “Permanent failure of preventing LES effacement due to gastric distention may be
the LES” is identified when one or more of the following an effective means of preventing damage to the LES and
LES abnormalities are seen on a resting motility study the progression of GERD from early to advanced disease.
performed in the fasted patient: an abdominal length of The second fundamental determinant of GERD is
1 cm or less, an overall length of 2 cm or less, and a resting the composition of the gastric juice refluxed into the
pressure of 6 mm Hg or less.28 When all three components esophagus. Increased esophageal exposure to gastric juice
are abnormal, the LES is completely destroyed and will can cause injury to the esophageal and/or respiratory
require reconstruction.37 Fig. 16.6 shows that the damage epithelium, along with the loss of esophageal and lung
to the LES parallels the severity of esophageal mucosal function. Injury is manifested on endoscopy by linear
injury and corroborates a link between damage to the LES or interlacing ulceration and repair by stricture and/
100%
@ *
80% @ *
26/28
60%
@ *
17/19
40%
FIGURE 16.6 The relationship between various
degrees of mucosal injury and the prevalence of a
@
20% permanently failed lower esophageal sphincter (LES).
43/66
The majority of the patients with mucosal injury have
26/92 a permanently failed LES. No injury vs. injury of any
0% 2/50 type P < .01. GERD, Gastroesophageal reflux
Volunteers GERD, GERD, GERD, GERD, disease. (From Stein HJ, Barlow AP, DeMeester TR,
no esophagitis stricture Barrett’s Hinder RA. Complications of gastroesophageal reflux
complication disease: role of lower esophageal sphincter,
@P < .01 vs volunteers esophageal acid and acid/alkaline exposure, and
*P < .01 vs volunteers and patients with GERD but no complication duodenogastric reflux. Ann Surg. 1992;216:35–43.)
210 SECTION I Esophagus and Hernia
60
42.0
pH score
45 39.5
loss in the absence of mucosal injury, suggest that the conditions, bile acids in the stomach precipitate and
loss is due to the consequences of inflammatory injury. have minimal effect in an acid gastric environment. On
The effect of duodenogastric reflux is the elevation of the other hand, in a more alkaline gastric environment,
the pH of gastric juice. The height of the elevation depends which occurs with excessive duodenogastric reflux or acid
on the baseline level of gastric pH, which varies depending suppression therapy, bile acids remain in solution and are
on whether the patient is taking acid suppression therapy. only partially dissociated. When nondissociated, nonpolar
Elevating the pH of gastric juice has four known effects. bile-acid molecules reflux into the esophagus, they can
Frist, when the pH exceeds 4, heartburn and regurgitation enter the mucosal cells. Once in the cell, where the pH
diminish.44 Second, when the pH enters the 3 to 5 range, is 7, they become completely dissociated into polar ions
it stimulates phenotypic differentiation of the cardiac and are trapped intracellularly in concentrations of up
mucosa toward intestinalization, with proliferation of the to 7 times the luminal concentrations.48 In the cell, bile
mucosal glandular cells.45 Third, when the pH reaches 4.5, acid salts can at low concentrations impair mitochondrial
bacteria normally present in the mouth begin to grow in function49; at high concentrations they become cytotoxic50
the stomach, and bile acids can be deconjugated to release and function as co-mutagens51 or likely direct mutagens.52
the more noxious free bile acid.46,47 Fourth, when the pH If soluble bile acids are to remain innocuous in a patient
enters the 3 to 6 range, bile acids become soluble, and with chronic reflux managed by acid suppression therapy,
a portion dissociates into their ionized salt and free H+ they must remain completely ionized. This requires that a
while the remainder persists as a lipophilic, nonionized gastric pH of 6 to 7 be maintained 24 hours a day, 7 days
acid. As the pH approximates 7, more than 90% of bile a week, while the patient is on acid suppression therapy.
acid becomes soluble and completely ionized. Acidification This is not only impractical but probably impossible without
of bile to a pH below 2 results in an irreversible bile-acid very high doses of medication. Insufficient medication
precipitation. Consequently, under normal physiologic allows the pH to drift down to 4 to 5 and causes cellular
mucosal damage while the patient remains relatively
asymptomatic (Fig. 16.12).53,54
Bile acid concentration (µmol/L)
20
cm
3 2.8 2.8
15 13.6 2.4*
11.6 2 Normal
10 1.8†
7.4* Normal 1
5 5.2†
0 0
No reflux Bile only Acid only Mixed No reflux Bile only Acid only Mixed
(n = 129) (n = 37) (n = 82) (n = 154) (n = 129) (n = 37) (n = 82) (n = 154)
*P ≤ .0004 vs. no reflux and bile only *P = .05 vs. no reflux
† †
P ≤ .002 vs. all groups P ≤ .0004 vs. all groups
mm Hg
cm
2 100
89 92
1.3 1.4 75
1 Normal 65.5* 60.5*
50
0.8* 0.6† Normal
25
0 0
No reflux Bile only Acid only Mixed No reflux Bile only Acid only Mixed
(n = 129) (n = 37) (n = 82) (n = 154) (n = 129) (n = 37) (n = 82) (n = 154)
FIGURE 16.11 Lower esophageal sphincter (LES) pressure, overall length, abdominal length, and contraction amplitude of the distal
esophagus in patients with normal esophageal exposure to acid and bile, increased exposure to bile alone, to acid alone, or both acid
and bile on 24-hour pH and bile monitoring. An increased exposure to acid and bile was associated with the greatest reduction in LES
pressure, overall length, and abdominal length. Increased bile exposure alone had no effect.
60
Insoluble Danger Ionized
(unable to
HISTOPATHOLOGY OF GASTROESOPHAGEAL
zone
40 (ppt) enter cell) REFLUX DISEASE
20 There has been a mystery regarding the pathologic
Acid(RCOOH) changes within the region of the gastroesophageal
0 junction (GEJ) associated with GERD. This was due to
0 1 2 3 4 5 6 7 8
the absence of dependable anatomic, endoscopic, or
pH
histologic landmarks regarding the position of the LES
[Salt] and the fact the LES cannot be recognized either during
pH = 4.3 + log endoscopy, at surgery, or on gross examination of surgi-
[Acid]
cal specimens. To deepen the mystery, physiologic and
FIGURE 16.12 A representative ionization curve for bile acids. At pathologic studies have introduced the concept that
pH 2 or below, bile acids are in their associated form (i.e., GERD begins at the squamocolumnar junction (SCJ).
un-ionized [RCOOH]). In this form and pH the bile acids are In the absence of squamoepithelial injury, the position
insoluble and precipitate. At pH 6 or greater, bile acids are of the SCJ and GEJ are concordant. As squamoepithelial
completely dissociated (i.e., ionized = [H+ + RCOO–]), soluble, and injury occurs, metaplastic cardiac epithelium forms, and
nontoxic to the cells. Between these extremes, bile acids can a new SCJ emerges, separated from the GEJ, and lying
exist in their associated form and remain soluble. In this form, further cephalad in the LES.54a,54b In other words, the SCJ
they can enter the cell with detrimental effects; hence the pH migrates cephalad in patients with GERD. The extent of
range between 2 and 6.5 is called the danger zone. the cephalad migration correlates with the severity of
disease. BE represents the extreme of this migration.55 As
a consequence of these ill-defined and moving landmarks,
Etiology and Natural History of Gastroesophageal Reflux Disease and Predictors of Progressive Disease CHAPTER 16 213
Current
SCJ
Original
SCJ
A B
Normal light imaging Narrow band imaging
FIGURE 16.13 Retroflex endoscopic images of the squamocolumnar junction (SCJ) and lower esophageal sphincter (LES) damage. (A) A
normal white light image and displays a slightly irregular circular SCJ with normal squamous epithelium extending up the esophagus. The
gastric mucosa below the SCJ appears normal with some erythematous streaking. (B) A narrow band image and shows the irregular
circular SCJ and multiple islands of squamous epithelium separated by newly formed metaplastic cardiac epithelium. The original position
of the SCJ is marked by a yellow line. The current irregular SCJ maintains its circular form while the splayed out original SCJ and the
destroyed distal portion of the esophagus take on the appearance of the stomach. This is due to the loss of a portion of the LES muscle
due to inflammatory injury. As a consequence, a portion of the LES’s abdominal length has been destroyed and its ability to function as a
barrier compromised.
ability to withstand damage by refluxed gastric juice, of the squamous mucosa covering the effaced portion of
because biopsy shows little or no inflammation. The devel- the LES.61 Under proper luminal conditions and stimuli,
opment of intestinal metaplasia within metaplastic cardiac intestinalization of the metaplastic cardiac mucosa follows.68
mucosa is considered a detrimental change because this This process may take a period of time. Indeed, in children
mucosa can progress to dysplasia and adenocarcinoma.64 younger than age 5 years who acquire a metaplastic cardiac
In contrast, intestinal metaplasia does not occur within mucosa–lined esophagus, intestinal metaplasia is rare.69
metaplastic cardiac mucosa that has undergone oxyntic Goblet cells begin to appear only later. Clinical studies
mucosal transformation (i.e., oxyntocardiac epithelium) have shown a time lag of 5 to 7 years after the onset of
as a result of exposure to gastric juice. Oxyntic transforma- reflux symptoms for intestinalized cardiac mucosa to
tion of metaplastic cardiac mucosa protects against the appear in adults.70
development of intestinal metaplasia and is a beneficial Oberg et al.71 have shown that the prevalence of intesti-
change.59 nalized metaplastic cardiac mucosa is related to the status
In the 1980s it became evident that the malignant of the LES and the severity of esophageal acid exposure.
transformation of BE occurred in the intestinalized cardiac In 251 patients, who had a normal-appearing SCJ and GEJ
mucosa.65,66 After this association was recognized, the term on endoscopy, biopsy of the SCJ had microscopic cardiac
BE was applied only to an esophagus lined by intestinal- mucosa of which 14% showed intestinal metaplasia. These
ized cardiac mucosa. The length of the intestinalized patients had more LES damage and esophageal acid
cardiac mucosa could be as short as a few millimeters.67 exposure. With time and further deterioration of the LES,
In other words, a patient is considered to have BE if metaplastic cardiac mucosa is endoscopically visible in the
any amount of cardia mucosa in the lower esophagus esophageal body for a distance greater than 3 cm, and 97%
has histologic evidence of goblet cells (i.e., the sign of had intestinal metaplasia on biopsy. These patients had
intestinalization). The length of the intestinalized mucosa the most LES damage and esophageal acid exposure. At
is measured as limited to the GEJ, limited to the lower each step, the patients exhibit more profound hallmarks
3 cm of the esophagus, or involving less than 3 cm of the of GERD, including decreasing length and pressure of the
esophagus. LES, increasing esophageal acid exposure, and eventual
The observation that some patients have cardiac mucosa loss of esophageal contraction amplitude (Table 16.2).
without intestinalization suggests that intestinalization The LES damage results from inflammatory changes in
requires a specific condition or stimulus. In other words, the muscularis propria secondary to increased esophageal
the development of the mucosal change typical of BE is acid exposure. 72 The same process causes the loss of
a stepwise process. In the first step, metaplastic cardiac esophageal body contractility classically seen in patients
mucosa forms from the squamous epithelium. This is with long segments of BE. It is rare that the length of
initiated by an inflammatory response to acid exposure BE exceeds more than 8 to 9 cm up the esophagus from
Etiology and Natural History of Gastroesophageal Reflux Disease and Predictors of Progressive Disease CHAPTER 16 215
TABLE 16.2 Functional Characteristics of Gastroesophageal Reflux Disease in Patients With Varied Lengths of Cardiac-Type
Mucosa With Intestinal Metaplasia
the GEJ. This is due to the normal gradient between the on these studies, H. pylori has no role in the pathogenesis
positive pressure environment in the stomach and the of GERD or its complications.73
negative pressure environment in the thoracic esophagus.
When the LES completely deteriorates, gastric juice flows PREVENTION OF PROGRESSIVE DISEASE
to the point of lowest esophageal pressure, which is in
the mid-thoracic esophagus. Under these conditions, Physicians have developed some concerns after 35 years
intestinalized metaplastic cardiac mucosa could appear of experience with more than 20 million GERD patients
within the lower half of the esophageal body without any taking prescribed PPIs. Despite the introduction and use
apparent subsequent change in length over time. At any of new powerful acid suppression drugs over this time
time during the process, specific luminal conditions or period, the incidence of GERD continues to increase by
stimuli, such as exposure to a specific pH range in the 30% every 10 years, and 30% to 40% of the patients have
presence of a specific solubilized bile acid, intestinalization only partial relief of their symptoms. Every year between
of metaplastic cardiac mucosa can occur and set the stage 2% and 3.5% of the 20 million patients receiving PPI
for malignant degeneration. Studies have suggested that therapy develop BE. Of those who develop BE, 0.5%
intestinalization of the metaplastic cardiac mucosa is to 1% will progress to esophageal adenocarcinoma.74
initiated by exposure to a pH 3 to 5 resulting from the Less than 1% of patients with a partial response and/or
interaction of the reflux gastric juice, duodenal juice, and disease progression on PPI therapy seek surgical therapy.
saliva at the squamocolumnar interface. Exposure to this These concerns have given rise to the thought that the
pH range has been shown to encourage the phenotypic time has come for a change in the treatment strategy of
expression of intestinalization by metaplastic cardiac GERD. Current treatment focuses on only one of the
epithelial cells.45,61 two determinants of the disease, the composition of the
The most controversial issue in the pathology of early gastric juice, whereas the other determinant, failure of
BE is whether intestinalized cardiac epithelium found the LES, is ignored.
below the SCJ is due to Helicobactor pylori. Extensive studies Taken together, the studies on histopathology have
have shown that the presence of inflamed cardiac mucosa introduced the concept that GERD begins at the SCJ. In
at the SCJ was inversely related to H. pylori infection the absence of disease the SCJ and the GEJ are concordant.
and strongly associated with increased esophageal acid As LES damage occurs and progresses, the SCJ separates
exposure. Further, there was no association between the from the GEJ and moves up the esophagus. This results in
presence of intestinal metaplasia and H. pylori. In addition, a squamooxyntic gap between the oxyntic gastric mucosa
the incidence of H. pylori in patients with esophageal in the proximal stomach and the remaining squamous
adenocarcinoma was not different from the prevalence of esophageal mucosa within the LES (Fig. 16.14).75 The
H. pylori in patients with benign esophageal disease. Based gap is filled with metaplastic cardiac mucosa, making the
216 SECTION I Esophagus and Hernia
Loss of LES
Normal New length Original
GEJ + SCJ SCJ GEJ
FIGURE 16.14 The squamooxyntic gap. (A) A photomicrograph of the normal squamocolumnar junction (SCJ) with squamous epithelium
directly apposed to oxyntic gastric epithelium (i.e., normal gastric epithelium). In normal subjects, a biopsy of the SCJ shows a 0- to
2-mm gap between the two epithelia. (B) A photomicrograph of a patient with gastroesophageal reflux disease (GERD) showing a
squamooxyntic gap between the oxyntic gastric mucosa in the proximal stomach and the remaining squamous esophageal mucosa
within the lower esophageal sphincter (LES). The gap is filled with metaplastic cardiac mucosa, which can become intestinalized and give
rise to visible Barrett esophagus in 2 to 5 years. The submucosa and muscularis propria show inflammatory damage resulting in the loss
of LES length. GEJ, Gastroesophageal junction.
determination between esophagus and stomach difficult difficult to reverse after it has become large enough to be
because the gap appears like stomach but is actually visible on endoscopy. The best opportunity to reverse the
esophagus. The prevalence of intestinal metaplasia of process is when intestinal metaplasia of cardiac mucosa
the metaplastic cardiac mucosa is directly proportional to is microscopic and not visible on endoscopy.
the length of the squamooxyntic gap. In a study of 1655 The primary method of identifying early disease in symp-
patients with GERD, intestinal metaplasia was observed tomatic patients is to obtain multiple biopsies of the SCJ.
in 24% of patients with a gap shorter than 1 cm, 86% If one or more biopsies show microscopic intestinalized
of patients with a gap of 1 to 5 cm, and 100% with a cardiac mucosa, the patient has early GERD and is at risk
gap longer than 5 cm. It has been proposed that the for visible BE. The treatment consists of stopping the reflux
squamooxyntic gap could be used as a cellular criterion to of gastric juice into the esophagus by repairing the LES
diagnose GERD. In other words, if you have microscopic and reestablishing gastric acidity by the discontinuation
metaplastic cardiac mucosa on biopsies of the SCJ margin of PPI therapy. This should be done as early as possible in
of the gap you have GERD, the length of the gap provides patients who have an incomplete response to PPI therapy,
an assessment of the severity of GERD, and intestinalized normal endoscopy, no visible BE, increased esophageal
metaplastic cardiac mucosa within the gap is an early acid exposure on 24-hour esophageal pH monitoring, and
indicator of disease progression and risk of esophageal a biopsy of the SCJ that shows microscopic intestinalized
adenocarcinoma. This allows the diagnosis of GERD at a metaplastic cardiac mucosa. The earlier these patients
very early stage when the disease is intrasphincteric and undergo surgical correction of a defective LES, the more
prior to becoming transsphincteric and involving the likely is the success in controlling symptoms and preventing
esophageal body. progressive disease.
These initial histologic changes are subtle, and a patient The grounds for this proposal are based on GERD pro-
with early GERD is likely to have a normal endoscopy. gression in 171 such patients investigated in the ProGERD
Consequently, clinical symptoms and endoscopy alone study.76 These patients had microscopic nonvisible intestinal
are not sufficient to confidently evaluate a patient with metaplasia on biopsy of the SCJ prior to any therapy. After
early disease. This requires a manometric assessment of the 4 to 8 weeks of initial PPI therapy, 128 of the patients
the LES, a measurement of esophageal acid exposure, and had a follow-up endoscopy and biopsy of the SCJ. All
biopsies of and just below the SCJ. Most important in the had persistent microscopic intestinal metaplasia. These
management of such patients is to prevent the develop- patients continued to receive PPI therapy and underwent
ment of visible BE, for this is the premalignant lesion endoscopy and biopsy after 2 and 5 years of follow-up.
of adenocarcinoma of the esophagus and is extremely Endoscopic visible BE was seen in 25.8% (33/128) of the
Etiology and Natural History of Gastroesophageal Reflux Disease and Predictors of Progressive Disease CHAPTER 16 217
Partial or non-
GERD
Responders responders
PPI Double-dose
PPI
PPI
Partial or non-
Endoscopy responders
+ 24-h pH test
patients. The conclusion of the study was that a biopsy shown the efficacy of LES augmentation in eliminating
of the SCJ showing microscopic intestinal metaplasia in a reflux symptoms and healing esophagitis in GERD patients
patient who is otherwise endoscopically normal is a strong suffering from an incomplete response to PPI therapy.83,84
indication of a high risk of progression to visible BE. These procedures avoid the side effects associated with
The problem with medical management of early GERD Nissen fundoplication, are reversible if required, and are
is that the available medications have not shown the ability appropriate for early surgical treatment of patients with
to prevent the development of BE or induce regression of progressive disease and the risk of BE, the premalignant
microscopic intestinal metaplasia that develops at the SCJ lesion of esophageal adenocarcinoma.
and is a precursor of visible BE. In contrast, a competent A proposed treatment algorithm that features early
Nissen fundoplication has been shown to prevent BE if use of LES augmentation procedures to prevent GERD
performed before endoscopically visible BE develops.77,78 progression is shown in Fig. 16.15. The algorithm empha-
Furthermore, a competent Nissen fundoplication can sizes the use of early endoscopy with biopsy of the SCJ,
induce complete regression of microscopic intestinal manometry of the LES, and esophageal pH monitoring
metaplasia at the SCJ and avoid the subsequent develop- to assess patients with the likelihood of progression.
ment of visible BE.79 Despite these proven benefits of the The patients are initially identified by one or several
fundoplication procedure, its side effects (i.e., dysphagia, of the following complaints or observations: (1) partial
postprandial bloating, and the inability to burp or vomit) symptomatic response to PPI therapy, (2) deterioration
have significantly limited its use early in the course of of PPI effectiveness over time, (3) rigid dependency on
GERD. daily PPI therapy, (4) esophagitis that is difficult to heal
Over the past decade, minimally invasive outpatient with PPI therapy, (5) increasing esophageal exposure to
LES augmentation procedures have been developed. acid over time, (6) loss of LES overall and/or abdominal
Examples include the implantation of a collar of magnetic length on esophageal motility study, and (7) a normal
beads around the inferior border of the LES to prevent endoscopy but microscopic intestinalized cardiac mucosa
its effacement into the stomach,80 neuromodulation of on biopsy of the SCJ. The early use of endoscopy in
the LES through electrical stimulation to increase LES patients who are confirmed nonresponders stratifies the
resting pressure,81 and incisionless partial fundoplication patients into four groups: (1) patients with visible BE; (2)
performed using a flexible endoscope introduced into patients with difficult to heal esophagitis; (3) patients with
the stomach through the mouth.82 Clinical studies have a normal endoscopy but have microscopic intestinalization
218 SECTION I Esophagus and Hernia
of metaplastic cardiac mucosa on biopsy of the SCJ; and 11. Pace F, Bianchi Porro G. Gastroesophageal reflux disease: a typical
(4) patients with a normal endoscopy who have carditis spectrum disease (a new conceptual framework is not needed). Am
J Gastroenterol. 2004;99:946-949.
on biopsy of the SCJ. LES augmentation is recommended 12. Labenz J, Nocon M, Lind T, et al. Prospective follow-up data from
for patients in groups 2 and 3 and an option for patients the ProGERD study suggest that GERD is not a categorial disease.
in group 4. In the latter group the patient is counseled Am J Gastroenterol. 2006;101:2457-2462.
toward augmentation if damage to the LES is detected 13. Malfertheiner P, Nocon M, Vieth M, et al. Evolution of gastroesopha-
geal reflux disease over 5 years under routine medical care—the
on manometry. The Nissen fundoplication should be ProGERD study. Aliment Pharmacol Ther. 2012;35:154-164.
considered for patients with extensive damage to the LES 14. Pace F, Bollani S, Molteni P, Bianchi Porro G. Natural history of
after a thorough discussion on the side effects associated gastroesophageal reflux disease without oesophagitis (NERD)—
with the procedure.85 a reappraisal 10 years on. Dig Liver Dis. 2004;36:111-115.
To improve our management of GERD and avoid its 15. Falkenback D, Oberg S, Johnsson F, Johansson J. Is the course of
gastroesophageal reflux disease progressive? A 21-year follow-up.
complication require that patients who have symptoms Scand J Gastroenterol. 2009;44:1277-1287.
of progressive disease be identified early in the course 16. Klaus A, Gadenstaetter M, Muhlmann G, et al. Selection of patients
of their disease. This requires early use of endoscopy in with gastroesophageal reflux disease for antireflux surgery based
patients with a partial or a deteriorating response to PPI on esophageal manometry. Dig Dis Sci. 2003;48:1719-1722.
17. Lord RV, DeMeester SR, Peters JH, et al. Hiatal hernia, lower
therapy, or a rigid dependency on PPI therapy. In contrast esophageal sphincter incompetence, and effectiveness of Nissen
to current management guidelines, we propose routine fundoplication in the spectrum of gastroesophageal reflux disease.
biopsy of the SCJ in such patients if the esophageal body J Gastrointest Surg. 2009;13:602-610.
component of the endoscopic evaluation is normal. If 18. Lord RVN, DeMeester TR. Reflux Disease and Hiatal Hernia.
microscopic intestinal metaplasia of metaplastic cardiac Oxford Textbook of Surgery. Vol. 2. Oxford: Oxford University Press;
2000:1239-1262.
mucosa is identified, a LES augmentation procedure should 19. Bammer T, Freeman M, Shahrian A, Hinder RA, DeVault KR,
be done to prevent progression to endoscopically visible Achem SR. Outcome of laparoscopic antireflux surgery in patients
BE, the premalignant lesion of esophageal adenocarci- with nonerosive reflux disease. J Gastrointest Surg. 2002;6(5):730-
noma. This approach must be first tested in the clinical 737.
20. Fyke FE, Code CF, Schlegel JF. The gastroesophageal sphincter in
setting to confirm that LES augmentation procedures healthy humans. Gastroenterologia. 1956;86:135-150.
can reproduce the encouraging results obtained with 21. McLaurin C. The intrinsic sphincter in the prevention of gastro-
the Nissen fundoplication. 79 It is incumbent on both oesophageal reflux. Lancet. 1963;282:801-805.
gastroenterologists and surgeons to work together to stop 22. DeMeester TR, Johnson LF, Joseph GJ, Toscano MS, Hall AW, Skinner
the rising incidence of BE, the premalignant lesion of DB. Patterns of gastroesophageal reflux in health and disease. Ann
Surg. 1976;184:459-470.
esophageal adenocarcinoma. By 2030, one out of every 23. DeMeester TR, Wernly JA, Bryant GH, Little AG, Skinner DB.
100 European men is projected to be diagnosed with Clinical and in vitro analysis of determinants of gastroesophageal
esophageal adenocarcinoma before the age of 75 years.86 competence. Am J Surg. 1979;137:39-46.
The principle of the solution to this developing crisis is, 24. Johnson LF, Lin YC, Hong SK. Gastroesophageal dynamics during
immersion in water to the neck. J Appl Physiol. 1975;38:449-454.
Where there is no BE, there is no cancer. 25. Pellegrini CA, DeMeester TR, Skinner DB. Response of the distal
esophageal sphincter to respiratory and positional maneuvers in
REFERENCES humans. Surg Forum. 1976;27:380-382.
26. O’Sullivan GC, DeMeester TR, Joelsson BE, et al. The interaction
1. DeMeester TR, Stein HJ. Surgical treatment of gastroesophageal of the lower esophageal sphincter pressure and length of sphincter
reflux disease. In: Castell DO, ed. The Esophagus. Boston: Little, in the abdomen as determinants of gastroesophageal competence.
Brown; 1992:579-626. Am J Surg. 1982;143:40-47.
2. Tutuian R, Castell DO. Management of gastroesophageal reflux 27. Bonavina L, Evander A, DeMeester TR, et al. Length of the distal
disease. Am J Med Sci. 2003;326(5):309-318. esophageal sphincter and competency of the cardia. Am J Surg.
3. Ronkainen J, Talley NJ, Storskrubb T, et al. Erosive esophagitis is a 1986;151:25-34.
risk factor for Barrett’s esophagus: a community-based endoscopic 28. Zaninotto G, DeMeester TR, Schwizer W, Johansson KE, Cheng SC.
follow-up study. Am J Gastroenterol. 2011;106:1946-1952. The lower esophageal sphincter in health and disease. Am J Surg.
4. Shah NH, LePendu P, Bauer-Mehren A, et al. Proton pump inhibitor 1988;155:104-111.
usage and the risk of myocardial infarction in the general population. 29. Pettersson GB, Bombeck CT, Nyhus LM. The lower esophageal
PLoS One. 2015;10:e0124653. sphincter: mechanism of opening and closure. Surgery. 1980;88:307-
5. Vakil N, van Zanten SV, Kahrilas P, Dent J, Jones R. Globale 314.
Konsensusgruppe. The Montreal definition and classification of 30. Bonavina L, DeMeester TR, Evander A. Role of the overall length
gastroesophageal reflux disease: a global evidence-based consensus. of the distal esophageal sphincter in the antireflux mechanism. In:
Am J Gastroenterol. 2006;101:1900-1920. Siewert JR, Holscher AH, eds. Diseases of the Esophagus. New York:
6. Ronkainen J, Aro P, Stoeskrubb T, et al. High prevalence of gas- Springer-Verlag; 1987:1031-1036.
troesophageal reflux symptoms and esophagitis with or without 31. Mason RJ, Oberg S, Bremner CG, et al. Postprandial gastroesophageal
symptoms in the general adult Swedish population: a Kalixanda reflux in normal volunteers and symptomatic patients. J Gastrointest
study report. Scand J Gastroenterol. 2005;40:275-285. Surg. 1998;2:342-349.
7. Bytzer P, Jones R, Vakil N, et al. Limited ability of the proton-pump 32. Dodds WJ, Dent J, Hogan WJ, et al. Mechanics of gastroesopha-
inhibitor test to identify patients with gastroesophageal reflux disease. geal reflux in patients with reflux esophagitis. N Engl J Med.
Clin Gastroenterol Hepatol. 2012;10:1360-1366. 1982;307:1549-1552.
8. Vakil N. Review article: how valuable are proton-pump-inhibitors in 33. Holloway RH, Hong M, Berger K, McCallum RW. Gastric distension: a
establishing a diagnosis of gastroesophageal reflux disease? Aliment mechanism for postprandial gastroesophageal reflux. Gastroenterology.
Pharmacol Ther. 2006;22(suppl 1):64-69. 1985;89:779-784.
9. Katz PO, Gerson LB, Vela MF. Guidelines for the diagnosis and 34. Jahnberg T, Martinson J, Hulten L, Fasth S. Dynamic gastric response
management of gastroesophageal reflux disease. Am J Gastroenterol. to expansion before and after vagotomy. Scand J Gastroenterol.
2013;108:308-328. 1975;10:593-598.
10. Agrawal A, Castell DO. GERD is chronic but not progressive. J Clin 35. Scheffer RC, Akkermans LM, Bais JE, Roelofs JMM, Smout AJPM,
Gastroenterol. 2006;40:374-375. Gooszen HG. Elicitation of transient lower esophageal sphincter
Etiology and Natural History of Gastroesophageal Reflux Disease and Predictors of Progressive Disease CHAPTER 16 219
relaxation in response to gastric distension and meal ingestion. 58. Paull A, Trier JS, Dalton MD, Camp RC, Loeb P, Goyal RK. The
Neurogastroenterol Motil. 2002;14:647-655. histologic spectrum of Barrett’s esophagus. N Engl J Med. 1976;295:476-
36. Ayazi S, Tamhankar R, DeMeester TR, et al. The impact of gastric 480.
distension on the lower esophageal sphincter and its exposure to 59. Chandrasoma PT, Der R, Ma Y, et al. Histology of the gastroesophageal
acid gastric juice. Ann Surg. 2010;252:52-62. junction: an autopsy study. Am J Surg Pathol. 2000;24:402-409.
37. Mason RJ, Lund RJ, DeMeester TR, et al. Nissen fundoplication 60. Chandrasoma P. Pathophysiology of Barrett’s esophagus. Semin
prevents shortening of the sphincter during gastric distension. Arch Thorac Cardiovasc Surg. 1997;9:270-278.
Surg. 1997;132:719-726. 61. Oberg S, Peters JH, DeMeester TR, et al. Inflammation and special-
38. Stein HJ, Barlow AP, DeMeester TR, Hinder RA. Complications of ized intestinal metaplasia of cardiac mucosa is a manifestation of
gastroesophageal reflux disease: role of lower esophageal sphincter, gastroesophageal reflux disease. Ann Surg. 1997;226:522-532.
esophageal acid and acid/alkaline exposure, and duodenogastric 62. Jain R, Aquino D, Harford WV, Lee E, Spechler SJ. Cardiac epi-
reflux. Ann Surg. 1992;216:35-43. thelium is found infrequently in the gastric cardia. Gastroenterology.
39. Kauer WKH, Peters JH, DeMeester TR, Ireland AP, Bremner CG, 1998;114:A160.
Hagen JA. Mixed reflux of gastric and duodenal juice is more 63. Lindahl H, Rintala R, Sariola H. Chronic esophagitis and gastric
harmful to the esophagus than gastric juice alone: the need for metaplasia are frequent late complications of esophageal atresia.
surgical therapy re-emphasized. Ann Surg. 1995;222:525-533. J Pediatr Surg. 1993;28:1178-1180.
40. Stein HJ, Kauer WKH, Feussner H, Siewert JR. Bile reflux in benign 64. Gillen P, Byrne P, West AB, West AB, Hennessy TP. Experimen-
and malignant Barrett’s esophagus: effect of medical acid suppression tal columnar metaplasia in the canine esophagus. Br J Surg.
and Nissen fundoplication. J Gastrointest Surg. 1998;2:333-341. 1988;75:113-115.
41. Nehra D, Howell P, Pye JK, Beynon J. Assessment of combined 65. Skinner DB, Bruno CW, Riddell R, Schmidt H, Iascone C, DeMeester
bile and acid pH profiles using an automated sampling device in TR. Barrett’s esophagus: comparison of benign and malignant cases.
gastro-oesophageal reflux disease. Br J Surg. 1998;85:134-139. Ann Surg. 1983;198:554-566.
42. Kauer WKH, Burdiles P, Ireland AP, et al. Does duodenal juice reflux 66. Reid BJ, Weinstein WM. Barrett’s esophagus and adenocarcinoma.
into the esophagus of patients with complicated GERD? Evaluation Annu Rev Med. 1987;38:477-492.
of a fiberoptic sensor for bilirubin. Am J Surg. 1995;169:98-104. 67. Weinstein WM, Ippoliti AF. Editorial: the diagnosis of Barrett’s
43. Oh DS, Hagen JA, Fein M, et al. The impact of reflux composition esophagus; goblets, goblets, goblets. Gastrointest Endosc. 1996;44:91-95.
on mucosal injury and esophageal function. J Gastrointest Surg. 68. Qualman SJ, Murray RD, McClung J, Lucas J. Intestinal meta-
2006;10:787-797. plasia is age related in Barrett’s esophagus. Arch Pathol Lab Med.
44. Marshall REK, Anggiansah A, Owen WA, Owen WJ. The relationship 1889;114:1236-1240.
between acid and bile reflux and symptoms in gastro-oesophageal 69. Lieberman DA, Oehlke M, Helfand M. Risk factors for Barrett’s
reflux disease. Gut. 1997;40:182-197. esophagus in community-based practice. GORGE consortium.
45. Fitzgerald RC, Omary MB, Triadafilopoulos G. Dynamic effects of Gastroenterology Outcome Research Group in Endoscopy. Gastro-
acid on Barrett’s esophagus. J Clin Invest. 1996;98:2120-2128. enterology. 1997;921:1203-1297.
46. Karmeli Y, Stalnikowitz R, Eliakim R, Rahav G. Conventional dose 70. Qualman ST, Murray RD, McClung J, et al. Intestinal metapla-
of omerprazole alters gastric flora. Dig Dis Sci. 1995;401:2070- sia is age related in Barrett’s esophagus. Arch Pathol Lab Med.
2073. 1990;114:1236-1240.
47. Domeliof L, Reddy BS, Weisburger H. Microflora and deconjugation 71. Oberg S, Peters JH, DeMeester TR, et al. The extent of Barrett’s
of bile acids in alkaline reflux after partial gastrectomy. Am J Surg. esophagus depends on the status of the lower esophageal sphincter
1980;140:291-295. and the degree of esophageal acid exposure. J Thorac Cardiovasc
48. Schweitzer EJ, Bass BL, Batzri S, Harmon JW. Bile acid accumulation Surg. 1999;117:572-580.
by rabbit esophageal mucosa. Dig Dis Sci. 1986;31:1105-1113. 72. Iascone C, DeMeester TR, Little AG, Skinner DB. Barrett’s esophagus:
49. Spirey JR, Bronk SF, Gores GJ. Glycochenodeoxycholate-induced functional assessment, proposed pathogenesis and surgical therapy.
hepatocellular injury in rat hepatocytes: role of ATP depletion and Arch Surg. 1983;118:543-549.
cytosolic free calcium. J Clin Invest. 1993;92:17-24. 73. Oberg S, Peters JH, Nigro JJ, et al. Helicobacter pylori is not associated
50. Latta RK, Fiander H, Ross NW, Simpson C, Schneider H. Toxicity with the manifestation of gastroesophageal reflux disease. Arch Surg.
of bile acids to colon cancer cell lines. Cancer Lett. 1993;70:167- 1999;134:722-726.
173. 74. Lagergren J, Lagergren P. Recent developments in esophageal
51. Silverman SJ, Andrews HW. Bile acids: co-mutagenic activity in the adenocarcinoma. CA Cancer J Clin. 2013;63:232-248.
salmonella-mammalian-microsome mutagenicity test. J Natl Cancer 75. Chandrasoma P, Wijetunge S, DeMeester SR, et al. The histologic
Inst. 1977;59:1557-1559. squamo-oxyntic gap: an accurate and reproducible diagnostic marker
52. Busby WF, Shuker DEG, Charnley G, Newberne PM, Tannenbaum of gastroesophageal reflux disease. Am J Surg Pathol. 2010;34:1574-
SR, Wogan GN. Carcinogenicity in rats of the nitrosated bile acids 1581.
conjugates N-nitroso-glycocholic acid and N-nitrosotaurocholic acid. 76. Leodolter A, Nocon M, Vieth M, et al. Progression of specialized
Cancer Res. 1985;45:1367-1371. intestinal metaplasia at the cardia to macroscopically evident Barrett’s
53. Moersch RN, Ellis FH. Pathologic change occurring in severe reflux esophagus: an entity of concern in the ProGERD study. Scand J
esophagitis. Surg Gyn Obstet. 1959;108:476-484. Gastroenterol. 2012;47:1429-1435.
54. Peghini PL, Katz PO, Bracy NA, Castell DO. Nocturnal recovery 77. Welscher GJ, Gadenstaetter M, Klingler PJ, et al. Efficacy of
of gastric acid secretion with twice-daily dosing of proton pump medical therapy and antireflux surgery to prevent Barrett’s meta-
inhibitors. Am J Gastroenterol. 1998;93:763-767. plasia in patients with gastroesophageal reflux disease. Ann Surg.
54a. Chandrasoma P, Wijetunge S, Ma Y, et al. Histologic classification 2001;234:627-632.
of patients based on mapping biopsies of the gastroesophageal 78. Gutschow CA, Schroder W, Prengel K, et al. Impact of antireflux
junction. Am J Surg Pathol. 2003;27:929-936. surgery on Barrett’s esophagus. Langenbecks Arch Surg. 2002;387:
54b. Chandrasoma P, Wijetunge S, Ma Y, et al. The dilated distal 138-145.
esophagus: a new entity that is the pathological bases of early 79. DeMeester SR, Campos GM, DeMeester TR, et al. The impact of
gastroesophageal reflux disease. Am J Surg Pathol. 2011;35:1873- an antireflux procedure on intestinal metaplasia of the cardia. Ann
1881. Surg. 1998;228:547-556.
55. Csendes A, Maluenda F, Braghetto I, Csendes P, Henriquez A, 80. Ganz RA, Peters JH, Horgan S, et al. Esophageal sphincter device
Quesada MS. Location of the lower esophageal sphincter and the for gastroesophageal reflux disease. N Engl J Med. 2013;368:719-
squamous columnar mucosal junction in 109 healthy controls and 727.
778 patients with different degrees of endoscopic esophagitis. Gut. 81. Rodriguez L, Rodriguez PA, Gomez B, Netto MG, Crowell MD,
1993;94:21-27. Soffer E. Electronic stimulation therapy of the lower esophageal
56. Bremner CJ, Lynch V, Ellis HF. Barrett’s esophagus: congenital or sphincter is successful in treating GERD: long-term 3-year results.
acquired? An experimental study of esophageal mucosal regeneration Surg Endosc. 2016;30:2666-2672.
in dog. Surgery. 1970;68:209-217. 82. Trad KS, Barnes WE, Simoni G, et al. Transoral incisionless fundoplica-
57. Hayward J. The lower end of the esophagus. Thorax. 1961;16:36-41. tion effective in eliminating GERD symptoms in partial responders to
220 SECTION I Esophagus and Hernia
proton pump inhibitor therapy at 6 months: the TEMPO randomized 85. Warren HF, Louie BE, Farivar A, Wilshire C, Aye RW. Manometric
clinical trial. Surg Innov. 2015;22:26-40. changes to the lower esophageal sphincter after magnetic sphincter
83. Bonavina L, Saino G, Lipham JC, Demeester TR. LINX reflux augmentation in patients with chronic gastroesophageal reflux
management system in chronic gastroesophageal reflux: a novel disease. Ann Surg. 2017;266:99-104.
effective technology for restoring the natural barrier to reflux. 86. Arnold M, Laversanne M, Brown LM, Devesa SS, Bray F. Predicting
Therap Adv Gastroenterol. 2013;6:261-268. the future burden of esophageal cancer by histological subtype: inter-
84. Ganz RA, Edmundowicz SA, Taiganides PA, et al. Long-term outcomes national trends in incidence up to 2030. Am J Gastroenterol. 2017;112:
of patients receiving a magnetic sphincter augmentation device 1247-1255.
for gastroesophageal reflux. Clin Gastroenterol Hepatol. 2016;14:671-677.
CHAPTER
Respiratory Complications of Gastroesophageal
Reflux Disease 17
Michael S. Mulvihill
| Shu S. Lin
| Matthew G. Hartwig
G
astroesophageal reflux (GER) refers to the reflux common extraesophageal manifestations of GERD. Avail-
of stomach contents into the esophagus. GER able evidence suggests that GERD is identified in approxi-
can occur physiologically, particularly in the mately 30% to 80% of patients previously diagnosed with
postprandial state.1 When reflux is of small volume for asthma.5 In a longitudinal study, development of asthma
limited durations and limited to the distal esophagus, occurred at a higher rate in patients with longer duration
this retrograde flow of stomach contents is generally of of GERD symptoms. Conversely, the prevalence of asthma
minor medical importance. In this chapter, the repercus- in GERD patients is reported at approximately 4.6%.5
sions of reflux of gastric contents into the airway will be Assessment of the prevalence of GERD-related cough is
reviewed. In the acute setting, the reflux of large volumes influenced heavily by the patient population, diagnostic
of gastric contents proximally into the airway can present testing modalities used, and the decision to attribute
a life-threatening aspiration event, sometimes known as cough to a single or multiple etiologies. As such, estimates
Mendelson syndrome.2 However, for the purposes of this of the prevalence of GERD-related cough have ranged
chapter, a focus will be maintained on the long-term from 10% to 40% in the literature.4
sequelae of small volume reflux proximal to the esophagus. In the following sections, the relationship between
Although the stomach benefits from the secretion of GER and respiratory disease will be reviewed. A growing
protective mucus, the lining of the esophagus lacks these body of evidence supports the finding that, although GER
protective features, and as such the mucosa of the esopha- may not be uniquely causative of respiratory pathology,
gus may become irritated or damaged by the regurgitation it is frequently a contributing factor. Treatment of GERD
of stomach contents. Gastroesophageal reflux disease in the patient with respiratory illness, then, should be
(GERD) describes the presence of excessive reflux of acid tailored to the individual patient based on his or her
or nonacid stomach contents, with unwanted resultant symptoms and the degree to which GER is thought to
manifestations. A host of factors may contribute to the contribute to the respiratory pathology at hand.
change from physiologic to pathologic state.1,3 In particular,
lifestyle factors (use of alcohol, use of cigarettes, obesity),
medications (calcium channel blockade, theophyllines), ROLE OF GASTROESOPHAGEAL REFLUX
diet (fatty food, fried foods, chocolate, caffeine, acidic DISEASE IN CHRONIC COUGH
foods, spicy foods), eating habits (large meals shortly
before sleep), and other medical conditions (hiatus hernia, The American College of Chest Physicians defines chronic
pregnancy, rapid weight gain) are known to contribute cough as one lasting greater than 8 weeks in a nonsmoking,
to GERD. A broad range of the most common symptoms immunocompetent patient who takes no cough-inducing
associated with GERD include the traditional gastro- drugs (such as angiotensin-converting enzyme inhibitors)
intestinal manifestations. However, extraesophageal and has a normal chest radiograph.6 Causative factors of
symptoms such as cough, hoarseness, sore throat, frequent chronic cough represent a vast array of clinical entities.
throat clearing, asthma, bronchitis, and other laryngeal Asthma, postnasal drip, and GER represent the most
and pulmonary manifestations have been noted, and common etiologies of chronic cough. The epidemiologic
increasing attention is being paid to the extraesophageal link between GER and chronic cough was established in
sequelae of GERD. The high prevalence of extraesophageal 1981 and subsequently verified across a range of patient
manifestations of GERD in patients with respiratory illness cohorts.7 Despite a clear association in the literature,
highlights the importance of the consideration of estimates of prevalence have varied widely from 0% to
gastrointestinal causes in the approach to the patient with 73%, driven by differences in populations, methodologies,
respiratory illness. In this chapter, respiratory complications and clinical awareness related to range of specialists
associated with GERD will be reviewed. (pulmonary medicine, gastroenterology, otolaryngology,
Patients with respiratory manifestations of GERD often general surgery) that evaluate these patients in the clinic
do not report the more typical reflux symptoms, with setting.
estimates of 75% of patients with GER-related cough As a theme that will appear in discussion of the interplay
exhibiting otherwise “silent” reflux.4 Because of this, clini- between GER and other respiratory disease states, GER
cians need to maintain a high level of suspicion in patients contributes to the development and sequelae of chronic
with respiratory complaints that may be the only manifesta- cough by way of a variety of mechanisms. The primary
tion of their underlying GERD. Descriptive studies to mechanisms for consideration are microaspiration, the
assess the prevalence of respiratory complications of GERD esophagobronchial reflex, and an increased sensitivity of
have primarily focused on asthma and cough, the most the cough reflex.
221
Respiratory Complications of Gastroesophageal Reflux Disease CHAPTER 17 221.e1
ABSTRACT
Gastroesophageal reflux refers to the reflux of stomach
contents into the esophagus. Gastroesophageal reflux
disease (GERD) describes the presence of excessive reflux
of acid or nonacid stomach contents, with unwanted
resultant manifestations. Although traditionally approached
as a disease with consequences limited to the gastrointestinal
tract, an increasing body of evidence implicates GERD in
injury to the respiratory tract. The high prevalence of
extraesophageal manifestations of GERD in patients with
respiratory illness highlights the importance of the con-
sideration of gastrointestinal causes in the approach to
the patient with respiratory illness. In this chapter, respira-
tory complications associated with GERD will be reviewed.
KEYWORDS
GERD, cough, reflux, asthma, bronchiectasis, pulmonary
fibrosis, lung transplantation
222 SECTION I Esophagus and Hernia
Microaspiration refers to the presence of gastric contents cough subsequently reporting improvements in cough
in the airway as evidenced by detection in bronchoalveolar scores, and freedom from PPI therapy.12 This is most likely
lavage (BAL) fluid. At bronchoscopy, evidence for aspira- due to our inability to better confirm the direct causal
tion includes subglottic stenosis, hemorrhagic tracheo- effect of GERD on chronic cough in all patients.
bronchitis, and erythema of the subsegmental bronchi.
Plain radiographs and axial imaging may also reveal
parenchymal changes consistent with acute or chronic
LARYNGOPHARYNGEAL REFLUX
aspiration. Irritation of the lower respiratory tract by The earliest reports of acid-driven injury to the larynx
aspiration represents the most direct link between GER (manifest by laryngeal ulcerations and granulomas) were
and cough. noted by Cherry in 1968.13 Pellegrini was among the first
Chronic cough may also be triggered indirectly through in 1979 to note that acid-containing reflux of gastric
activation of an esophagobronchial reflex, a means by contents may then be a causative factor in this laryngeal
which cough may be stimulated by GER in the absence pathology.14 Although at the time the potential for a surgi-
of bronchoscopic or radiographic evidence of aspiration. cal cure of extraesophageal symptoms of reflux was noted
Mechanistically, esophageal innervation by sensory-type to be achievable by way of an antireflux procedure, lar-
fibers that express acid-sensitive channel TRPV-1 and the yngitis was not formally recognized as an extraesophageal
subsequent convergence of these sensory afferents with manifestation of GER until 2006 by the Montreal Definition
vagal afferent neurons provides a means by which refluxate and Classification System. Data regarding the prevalence
in the distal esophagus may stimulate a vagal reflex. Ing of laryngopharyngeal reflux (LPR) are relatively limited,
et al. provide evidence of this mechanism in humans by in part stemming from controversy in the professional
showing that instillation of acid into the distal esophagus organizations with respect to the appropriate diagnostic
of patients with both GERD and cough increased the criteria for LPR. In particular, controversy exists with
frequency, duration, and intensity of cough compared respect to the extent to which symptoms related to LPR
with instillation of normal saline.8 Such a response was should be attributed to reflux in the absence of traditional
subsequently attenuated by pretreatment with lidocaine. heartburn symptoms.
Further studies in humans suggest that repetitive exposure A high index of suspicion should be maintained for
can lead to hypersensitivity and a lowering of the cough LPR in the patient with hoarseness, cough, globus, and
threshold.9,10 Dynamic changes in the cough threshold throat clearing. A minority of patients (35%) found to
may cause cough to become relatively stimulus agnostic, have LPR report heartburn symptoms. Scoring systems
such that nonreflux stimuli precipitate symptoms previously such as the Reflux Symptom Index (RSI) or Hull Airway
associated with GER. Reflux Questionnaire (HARQ) may be of utility in the
The approach to the patient with chronic cough screening of these patients for LPR and associated extra-
attributed to GER may be guided by both national guide- esophageal manifestations of GERD as a cause of symptoms.
lines and systematic review of the available evidence. The Patients who are subsequently evaluated by direct laryn-
American College of Chest Physicians recommends empiric goscopy following history and physical exam are assessed
acid suppression in patients thought to experience reflux- for posterior laryngeal edema, true vocal fold edema, and
induced cough.6 A Cochrane systematic review identified pseudosulcus. Although these findings may be sensitive
13 randomized controlled trials examining GERD therapy for the diagnosis of LPR, they often are not specific because
for the treatment of cough in adults without primary lung they are common findings in the general population and
disease.11 Analysis of H2-receptor antagonist, motility are not strictly linked to exposure of the larynx to gastric
agents, and conservative treatment was not possible in contents.
the meta-analysis. The remaining nine trials compared Detection of reflux into the pharynx requires diagnostic
proton pump inhibitor (PPI) with placebo. Although strategies that differ from the detection of reflux in the
there was no difference in the rate of total resolution of distal esophagus, on account of the neutralization of
cough, patients receiving PPI therapy were more likely refluxate as it ascends the esophagus into the more alkaline
to note improvement in cough scores. Thus, even though environment of the pharynx. As such, pharyngeal pH
the primary outcome was not met, there appears to be a monitoring is of low sensitivity in the diagnosis of LPR.
benefit to acid suppression in appropriately selected Currently, pH/impedance monitoring offers a superior
patients. Nonetheless, the long-term consequences of strategy for diagnosis, with a significant benefit of detec-
respiratory complications were not addressed in any of tion of both acid and nonacid reflux. pH/impedance
these studies. monitoring may also identify the proximal extent of reflux.
To date, controlled studies of the efficacy of fundoplica- Reflux into the laryngopharynx is very rare in healthy
tion or other surgical interventions for patients with cough patients, such that any evidence of reflux so far proximal
and GER are lacking. In single-center studies, surgical from the gastroesophageal junction may be considered
intervention consistently demonstrates efficacy in key abnormal and thereby warrant treatment. The use of
metrics. In particular, fundoplication is effective in reduc- a dual-probe (or bifurcated) impedance pH catheter
ing the number of reflux events by esophageal pH monitoring permits detection of both impedance and pH changes
and produces symptomatic improvement. Traditionally, across the upper esophageal sphincter, thereby permitting
patients with extraesophageal GER symptoms such as detection of pharyngeal reflux. More novel techniques
cough have been found to benefit from fundoplication aim to detect the pH of aerosolized reflux, which reveal
at a rate lower than those with typical GER symptoms, the finding that patients with chronic cough may reflux
with estimates of approximately 60% of patients with both liquid and gaseous stomach contents. Patients with
Respiratory Complications of Gastroesophageal Reflux Disease CHAPTER 17 223
predominantly cough-related symptoms may benefit from leads to improvements in patient-related outcomes in
techniques that make use of physiologic changes associated both asthma and COPD. The prevalence and pathophysiol-
with cough (such as pressure changes above and below ogy of GER in each entity are reviewed.
the diaphragm) to demonstrate a temporal link between Asthma is common in the United States, with approxi-
cough and reflux. mately 24 million Americans carrying the diagnosis. Of
Mechanistically, as with exposure of other nonstomach these patients with asthma, an estimated 30% to 90% of
structures to acid, the larynx has relatively little protection patients have GER. 5 Systematic review by Havemann
against both acid and enzymatic activity. The thinner revealed that 59.2% of asthmatics had evidence of GERD,
epithelium and lack of peristalsis to wash away acid whereas the prevalence in controls was 38.1%. Similarly,
contribute to injury seen with relatively brief and short- abnormal esophageal pH, esophagitis, and hiatal hernia
duration exposures in animal models. In addition to the are all more prevalent in patients with asthma.19 GER is
consequences of direct exposure to acid and other gastric also a risk factor for asthma-related hospitalizations in
contents, laryngeal injury may also manifest as an indirect older adults.20
consequence of exposure of the esophagus to refluxate. The prevalence of GER in patients with COPD is more
In the indirect model, irritation of the esophagus can poorly understood. A retrospective study from the Veterans
generate vagally mediated reflexes such as cough and Health Administration found that of 101,366 veterans
bronchoconstriction that can contribute to chronic those with erosive esophagitis and esophageal stricture had
laryngeal injury. higher rates of chronic bronchitis, asthma, and COPD.
Successful treatment of LPR again requires adequacy A smaller study demonstrated that pathologic GER was
of diagnosis. In the patient with modifiable risk factors, identified in 62% of patients with severe COPD. Notably,
behavior modification such as minimization of tobacco episodes of esophageal acid exposure were temporally
and alcohol consumption is recommended. Those patients linked to oxygen desaturations in this patient population.
with LPR and evidence of esophageal symptoms of GERD There is also evidence that GER contributes to detrimental
should be treated with acid suppression. The role of acid outcomes in patients with COPD. The 2010 results of the
suppression in patients with suspected or confirmed LPR ECLIPSE study (Evaluation of COPD Longitudinally to
but without evidence of esophageal manifestations of Identify Predictive Surrogate Endpoints) prospectively
GERD is controversial, and success of therapy likely assessed 2138 patients with stages II–IV COPD to identify
depends on the extent to which acid is a contributing factors associated with increased frequency of COPD
factor in the symptoms associated with LPR. The American exacerbation. In this cohort, GER was independently
Gastroenterological Association recommends against PPI associated with an increased risk of COPD exacerbation.21
and H2-receptor blockade for the treatment of suspected Multiple mechanisms likely contribute to the interaction
LPR in the absence of a concomitant GERD syndrome.15 between esophageal pathology and injury to the respiratory
In contrast, the American Academy of Otolaryngology- system. A shared embryonic origin from the foregut of
Head and Neck Surgery recommends twice-daily PPI for the respiratory and gastrointestinal systems contributes
no less than 6 months for patients with LPR.16 Data to the interplay between asthma, COPD, and GER.
regarding the use of acid suppression in the absence of Shared vagal innervation and resultant converging
esophageal manifestations are weak and are limited by a visceral sensory neural input contributes to pulmonary
lack of standardized diagnostic criteria. Trials have also symptoms at the time of stimulation of esophageal recep-
been limited by the possibility of placebo effect, such that, tors by stimuli such as acid exposure. Nonadrenergic
in small studies, for example, patients with no measurable neurons in the esophageal myenteric plexus communicate
pH response to PPI report symptom improvement.17,18 with the trachea. In animal models, instillation of acid in
Uncontrolled trials of surgery have shown some promise the esophagus subsequently results in the release of
in patients with medication-refractory LPR, but a controlled tachykinin-like substances in the airways and subsequent
trial of antireflux surgery in patients who failed aggressive bronchoconstriction. This finding can be terminated by
acid suppression did not improve laryngeal symptoms vagotomy, suggesting that vagal innervation is required
despite showing successful control of reflux by way of pH for this interaction. This finding may contribute to the
studies. finding of increased airway reactivity common to this
patient population. A single bolus of acid can result in
rapid distribution through the lung and generate a wide
ROLE OF GASTROESOPHAGEAL REFLUX range of histopathologic changes, including neutrophil
DISEASE IN ASTHMA AND CHRONIC sequestration, epithelial damage, pulmonary edema, and
OBSTRUCTIVE PULMONARY DISEASE pulmonary hemorrhage.
Chronic aspiration results in inflammation, an altered
Chronic obstructive pulmonary disease (COPD) and immune response, and worsening of asthma symptoms.
asthma are distinct clinical entities with unique risk factors, In animal studies, instillation of acid increases total lung
pathophysiology, and prognosis. However, they may exhibit resistance and leads to an aspiration pneumonia that is
overlapping clinical presentations. GER represents a characterized by neutrophilic and lymphocytic peribron-
comorbidity experienced by many patients with each chiolar infiltrates, goblet cell hyperplasia, and thickening
disorder. Most broadly, abnormal GER may contribute to of the smooth muscle layer. Barbas et al. demonstrated
worsening asthma symptoms and has been associated with that chronic aspiration of murine gastric fluid produced
an increased risk of COPD exacerbation. Although results an injury pattern characterized by hyperplasia and neu-
have been inconsistent, treatment of symptomatic GER trophil infiltration of the bronchioles that differs from
224 SECTION I Esophagus and Hernia
the response seen in acute aspiration. This chronic aspira- Patients with asthma and extraesophageal reflux symp-
tion model subsequently results in a shift to a Th2 inflam- toms (and without warning signs such as dysphagia or
matory response.22 weight loss) may be considered for an empiric trial of PPI.
Although overlapping mechanisms may contribute to Those with improvements in asthma symptomatology or
the relationship between COPD and GER, additional improvements in pulmonary function testing may warrant
mechanisms are worthy of consideration. Reduction of chronic therapy. Further diagnostic testing is warranted
lower esophageal sphincter (LES) tone may be due to in those patients who do not experience improvement in
multiple factors but regardless can contribute to the reflux symptoms after empiric therapy. Appropriately selected
of gastric contents. In particular, anatomic changes related patients for whom goals of symptom relief and avoid-
to COPD, such as the flattening of the diaphragm, can ance of long-term medication are prioritized may be
result in a decrease in resting LES tone.23 In addition, suitable candidates for surgical antireflux procedures.
oral theophylline, although used less frequently in con- At this time, management strategies for the patient
temporary management of COPD, may reduce LES tone. with COPD and symptomatic GER are not widely gen-
Evidence supports the treatment of symptomatic GER eralizable and should be considered on a case-by-case
in the setting of asthma. Adequate studies in the manage- basis.
ment of GER in the patient with COPD are presently
lacking. To date, three large randomized controlled trials
represent the best evidence that high-dose PPI therapy ROLE OF GASTROESOPHAGEAL REFLUX
improves symptoms of asthma. In the first trial, 207 patients DISEASE IN BRONCHIECTASIS
with moderate to severe asthma and reflux symptoms
experienced improvements in Asthma Quality of Life Patients with bronchiectasis (both due to cystic fibrosis
Questionnaire scores and decreased asthma exacerbations [CF] and non-CF bronchiectasis) experience GER at a
with lansoprazole therapy compared with controls, rate higher than the general population. Symptomatic
although neither the primary outcome (improvement in GER may occur at a rate higher in the CF population
asthma symptoms) nor the secondary outcomes (peak than the general population, with reports ranging from
expiratory flow [PEF] and forced expiratory volume 30% to 40% of CF patients reporting symptoms of GERD,
[FEV]1) were met.24 A second trial randomized 961 patients and up to 90% of patients may have evidence of silent
with moderate to severe asthma and symptoms of GER GER by esophageal monitoring in patients with severe
to esomeprazole, finding that the treatment arm enjoyed disease. Patients with bronchiectasis may frequently have
small but significant improvements in PEF, FEV 1, and a clinical presentation that is atypical with respect to GER
Asthma Quality of Life Questionnaire scores.25 A third symptoms, thereby requiring a higher index of suspicion
trial of patients with moderate to severe asthma and both in the approach to the patient with bronchiectasis and
nocturnal respiratory symptoms and GER found that possible GER.
esomeprazole improved PEF.26 There exist increasing data to suggest that increased
Although a significant fraction (24% to 62% in the GER (with or without aspiration) may lead to functional
available literature) of patients with asthma may have lung impairment in CF. Data from the European Epide-
clinically silent GER, the evidence does not support PPI miologic Registry of Cystic Fibrosis demonstrate that
therapy for patients with silent GER and asthma. Available patients with CF with GER have lower pulmonary function
trial data indicate that the addition of PPI for patients than those without GER.27 Aspiration of gastric contents
with poorly controlled asthma and minimal to no symptoms also appears to be associated in a dose-dependent fashion
of reflux does not improve episodes of poor asthma with an increase in the extent of airway inflammation in
control nor does it improve PEF. these patients.
A Cochrane systematic review synthesizes these findings The mechanisms for GER in the patient with CF or
with 12 randomized controlled trials of medical interven- non-CF bronchiectasis appear similar to the broader
tions (including histamine antagonists, PPIs, and lifestyle patient population, namely relaxation of the LES, but
modification) of GER in patients with asthma. The review additional mechanisms may contribute. In particular, the
concludes that treatment of GER does provide a benefit increased intraabdominal pressure due to chronic cough,
with respect to management of asthma. No consistent wheezing, and lung hyperinflation—all common in the
improvement in secondary outcomes such as lung func- patient with bronchiectasis—may contribute to the develop-
tion, airway responsiveness, or asthma symptoms was ment of GER. In addition, delayed gastric emptying and
identified.11 a decreased basal tone of the LES may be contributing
The role for antireflux surgery in patients with asthma factors. Delayed gastric emptying is seen in approximately
and GER is poorly defined and without high-level evidence. 30% of GER patients and may be higher in patients with
Surgical therapy is associated with improved asthma both CF and GER.28 Approximately 30% of patients with
symptom control, but improvements in pulmonary function CF appear to experience delayed gastric transit. Although
have not been demonstrated in the asthmatic population. increased gastric retention may lead to a higher (more
A meta-analysis of 24 studies estimates that antireflux proximal) extent of reflux, a formal causal relationship
surgery may improve asthma symptoms, suggesting that is yet to be defined. Manometric studies of CF patients
appropriately selected patients may find benefit to surgical reveal a rate of decreased basal LES tone in approximately
therapy.12 To date, the literature has not identified improve- 60% of patients.29
ments in pulmonary function associated with antireflux The sequelae of GER in patients with bronchiectasis
surgery in this patient population. manifest by mechanisms similar to the broader population.
Respiratory Complications of Gastroesophageal Reflux Disease CHAPTER 17 225
In particular, microaspiration, reflex bronchospasm, and oxygen requirements and improved long-term survival in
increased airway inflammation all contribute to pulmonary patients treated with PPI followed by fundoplication if
injury. As mentioned previously, GER may result in required to adequately suppress acid.37,38
functional lung impairment in this patient population. GER is likewise common in patients with other con-
Optimal treatment strategies in this patient population nective tissue disorders, such as scleroderma. In particular,
are poorly defined. To date, there exist no randomized GER (as well as impaired esophageal motility and esopha-
trial data on which to base consensus. Observational geal dilatation) is highly prevalent in those patients with
studies note an improvement in pulmonary function in CTD and evidence of pulmonary fibrosis. Although further
patients with CF after initiation of PPI therapy for symp- mechanistic studies are again warranted, these data do
tomatic GERD. However, acid suppression has been suggest a possibility that patients with CTD exhibit a
demonstrated to increase a rate of bacterial overgrowth disordered response to microaspiration and reflux that
in stomach contents, leading to altered respiratory flora ultimately places them at higher risk for pulmonary fibrosis.
when nonacid but bacterial-laden stomach contents are The patients with CTD and esophageal dysmotility, then,
refluxed. Antireflux procedures should be considered in are subsequently most likely to suffer the pulmonary
patients with GER refractory to medical management and consequences of CTD. Randomized data to guide treatment
in patients with complications of GER, such as erosive for patients with scleroderma or other CTD and GER are
esophagitis, stricture, or failure to thrive. Surgical manage- lacking. Consideration of acid suppression therapy, whether
ment of reflux in the pediatric CF patient population has by lifestyle modification, medical management, or surgical
been shown in observational studies to decrease the rate intervention, should be determined on a case-by-case
of pulmonary function decline.30,31 Outcomes with respect basis.
to GER symptoms have been mixed following surgical
intervention.
ROLE OF GASTROESOPHAGEAL REFLUX
ROLE OF GASTROESOPHAGEAL REFLUX DISEASE IN PATIENTS UNDER
DISEASE IN INTERSTITIAL LUNG DISEASE CONSIDERATION FOR LUNG
The interstitial lung diseases (ILDs) represent a hetero-
TRANSPLANTATION
geneous group of progressive acute and chronic paren- Special attention is warranted in the consideration of the
chymal lung diseases characterized by diffuse infiltrates patient with GER who has undergone or who is under
and progressive fibrosis. The progressive nature of ILD evaluation for lung transplantation. Indeed, many of the
can lead to worsening lung function and respiratory disease states discussed in this chapter (COPD, CF, IPF)
insufficiency. may progress to respiratory insufficiency and end-stage
In a manner similar to bronchiectasis, there is an lung failure. GER is very common in both the pretransplant
increasing recognition of the interplay between GER candidate and the posttransplant recipient, with estimates
and ILD, such as idiopathic pulmonary fibrosis (IPF), in the range of 50% of patients who receive an allograft
scleroderma, and other connective tissue disease (CTD). having evidence of reflux before transplant and upward
Although definitive proof that GER and subsequent micro- of 75% after lung transplantation.39,40 For these patients,
aspiration are the proximal cause of ILD, precipitate strategies for management are extremely limited and lung
exacerbation, or promote disease progression is lacking, transplantation is the gold standard therapy.
there exists a body of evidence that suggests a role The lung is relatively unique among solid-organ
for GER and microaspiration in each of these disease allografts because of its continuous exposure to the outside
states. environment. As such, immunosuppressive strategies and
Development of IPF is associated with a range of the subsequent consequences of immunomodulation must
exposures, viral infections, and inherited genetic factors. be tailored to this need for the allograft to respond to
It is characterized by a histopathologic pattern of usual antigen exposure. Exposure to gastric contents—both
interstitial pneumonia (UIP) on lung biopsy. Classic find- acid and nonacid containing—represent a critical source
ings include areas of distorted parenchymal architecture of environmental exposure in lung transplantation. Survival
interspersed with fibrotic lesions characterized by temporal following lung transplantation is limited by the develop-
heterogeneity. Beginning in the 1970s, studies of prevalence ment of chronic lung allograft dysfunction (CLAD) that
have estimated that in excess of 50% of patients with manifests clinically by a progressive decline in FEV1 from
evidence of pulmonary fibrosis had evidence of GER.32,33 peak posttransplant values. GERD has been recognized
Typical symptoms of GER are poor predictors of GER in as an important risk factor for the development of CLAD
patients with IPF.34 posttransplant. Patients with GERD before transplant have
Although a high fraction of IPF patients have evidence also been identified as being at risk for premature graft
of GER and although animal models of reflux described dysfunction. In addition, patients with GERD before
previously can result in chronic inflammatory changes to transplant are at risk for increasing severity of reflux
the lung tissue, no causal relationship between GER and posttransplant. As such, in retrospective clinical data,
the development of IPF has been established in humans. patients with GERD are at increased risk of negative
A small body of evidence suggests that a role exists for outcomes after lung transplantation.39
acid suppression therapy in the IPF patient population.35,36 In animal models, chronic aspiration of gastric fluid
Three case series of IPF patients report stabilization of led to increased rates of acute rejection and fibrosis after
226 SECTION I Esophagus and Hernia
orthotopic lung transplantation.41–45 In addition, instillation 13. Cherry J, Margulies SI. Contact ulcer of the larynx. Laryngoscope.
of nonacid gastric contents in similar models can produce 1968;78(11):1937-1940. doi:10.1288/00005537-196811000-00007;
PubMed PMID: 5722896.
equivalent aspiration injury, in part due to higher rates 14. Pellegrini CA, DeMeester TR, Johnson LF, Skinner DB. Gastro-
of bacterial overgrowth that might otherwise be suppressed esophageal reflux and pulmonary aspiration: incidence, functional
by low gastric pH.46 abnormality, and results of surgical therapy. Surgery. 1979;86(1):110-119.
Taken together, multiple lines of evidence implicate PubMed PMID: 36677.
15. Kahrilas PJ, Shaheen NJ, Vaezi MF, American Gastroenterological
GERD as a contributing factor to diminished pulmonary Association Institute. Clinical Practice and Quality Management
function posttransplant.47 On account of the critical Committee. American Gastroenterological Association Institute
observation that exposure to nonacid gastric contents can technical review on the management of gastroesophageal reflux
contribute to lung allograft failure in animal models, disease. Gastroenterology. 2008;135(4):1392-1413, 413 e1-5.
there exists minimal role for medical therapy by way of doi:10.1053/j.gastro.2008.08.044; PubMed PMID: 18801365.
16. Koufman JA, Aviv JE, Casiano RR, Shaw GY. Laryngopharyngeal
acid suppression in the treatment of reflux after lung reflux: position statement of the Committee on Speech, Voice, and
transplantation. Use of routine PPI posttransplantation Swallowing Disorders of the American Academy of Otolaryngology-
has not abrogated the relationship between abnormal Head and Neck Surgery. Otolaryngol Head Neck Surg. 2002;127(1):32-35.
esophageal acid contact times and diminished FEV1. Not PubMed PMID: 12161727.
17. Reichel O, Keller J, Rasp G, Hagedorn H, Berghaus A. Efficacy of
surprisingly, then, PPI alone does not prevent aspiration once-daily esomeprazole treatment in patients with laryngopharyngeal
of gastric contents and, as such, is inadequate therapy to reflux evaluated by 24-hour pH monitoring. Otolaryngol Head Neck
minimize the risk of development of bronchiolitis oblit- Surg. 2007;136(2):205-210. doi:10.1016/j.otohns.2006.10.011; PubMed
erans in allograft recipients with evidence of reflux. PMID: 17275540.
Single-center data suggest that surgical intervention by 18. Reichel O, Dressel H, Wiederanders K, Issing WJ. Double-blind,
placebo-controlled trial with esomeprazole for symptoms and signs
way of fundoplication preserves 1-year and peak pulmonary associated with laryngopharyngeal reflux. Otolaryngol Head Neck Surg.
allograft function.48,49 Aggressive work-up and management, 2008;139(3):414-420. doi:10.1016/j.otohns.2008.06.003; PubMed
with consideration of early surgical intervention by way PMID: 18722223.
of fundoplication posttransplant, are warranted in patients 19. Field SK, Underwood M, Brant R, Cowie RL. Prevalence of gastro-
esophageal reflux symptoms in asthma. Chest. 1996;109(2):316-322.
evaluated for lung transplantation. PubMed PMID: 8620699.
20. Diette GB, Krishnan JA, Dominici F, et al. Asthma in older patients:
factors associated with hospitalization. Arch Intern Med.
REFERENCES 2002;162(10):1123-1132. doi:10.1001/archinte.162.10.1123.
21. Hurst JR, Vestbo J, Anzueto A, et al. Susceptibility to exacerbation
1. Dodds WJ, Dent J, Hogan WJ, et al. Mechanisms of gastroesophageal in chronic obstructive pulmonary disease. N Engl J Med.
reflux in patients with reflux esophagitis. N Engl J Med. 1982;307(25): 2010;363(12):1128-1138. doi:10.1056/NEJMoa0909883; PubMed
1547-1552. doi:10.1056/NEJM198212163072503. PMID: 20843247.
2. Mendelson CL. The aspiration of stomach contents into the lungs 22. Barbas AS, Downing TE, Balsara KR, et al. Chronic aspiration shifts
during obstetric anesthesia. Anesthesiology. 1946;7(6):694-695. the immune response from Th1 to Th2 in a murine model of
3. Canning BJ, Mazzone SB. Reflex mechanisms in gastroesophageal asthma. Eur J Clin Invest. 2008;38(8):596-602. doi:10.1111/j.1365-
reflux disease and asthma. Am J Med. 2003;115(3 suppl 1):45-48. 2362.2008.01976.x.
doi:http://dx.doi.org/10.1016/S0002-9343(03)00192-X. 23. Roussos C, Macklem PT. The respiratory muscles. N Engl J Med.
4. Irwin RS. Chronic cough due to gastroesophageal reflux disease: 1982;307(13):786-797. doi:10.1056/NEJM198209233071304.
ACCP evidence-based clinical practice guidelines. Chest. 2006;129 24. Littner MR, Leung FW, Ballard ED 2nd, Huang B, Samra NK,
(1 suppl):80S-94S. doi:10.1378/chest.129.1_suppl.80S. Lansoprazole Asthma Study Group. Effects of 24 weeks of lansoprazole
5. Havemann BD, Henderson CA, El-Serag HB. The association between therapy on asthma symptoms, exacerbations, quality of life, and
gastro-oesophageal reflux disease and asthma: a systematic review. pulmonary function in adult asthmatic patients with acid reflux
Gut. 2007;56(12):1654-1664. doi:10.1136/gut.2007.122465. symptoms. Chest. 2005;128(3):1128-1135. doi:10.1378/chest.128.3.1128;
6. Gibson P, Wang G, McGarvey L, et al. Treatment of unexplained PubMed PMID: 16162697.
chronic cough: CHEST Guideline and Expert Panel Report. Chest. 25. American Lung Association Asthma Clinical Research Centers,
2016;149(1):27-44. doi:http://dx.doi.org/10.1378/chest.15-1496. Mastronarde JG, Anthonisen NR, et al. Efficacy of esomeprazole
7. Christopher KL, Wood RP, Eckert RC, Blager FB, Raney RA, Souhrada for treatment of poorly controlled asthma. N Engl J Med.
JF. Vocal-cord dysfunction presenting as asthma. N Engl J Med. 2009;360(15):1487-1499. doi:10.1056/NEJMoa0806290; PubMed
1983;308(26):1566-1570. doi:10.1056/NEJM198306303082605. PMID: 19357404; PMCID: PMC2974569.
8. Ing AJ, Ngu MC, Breslin AB. Pathogenesis of chronic persistent 26. Kiljander TO, Harding SM, Field SK, et al. Effects of esomeprazole
cough associated with gastroesophageal reflux. Am J Respir Crit Care 40 mg twice daily on asthma: a randomized placebo-controlled trial.
Med. 1994;149(1):160-167. doi:10.1164/ajrccm.149.1.8111576; Am J Respir Crit Care Med. 2006;173(10):1091-1097. doi:10.1164/
PubMed PMID: 8111576. rccm.200507-1167OC; PubMed PMID: 16357331.
9. Javorkova N, Varechova S, Pecova R, et al. Acidification of the 27. Navarro J, Rainisio M, Harms HK, et al. Factors associated with poor
oesophagus acutely increases the cough sensitivity in patients with pulmonary function: cross-sectional analysis of data from the ERCF.
gastro-oesophageal reflux and chronic cough. Neurogastroenterol Motil. Eur Respir J. 2001;18(2):298-305.
2008;20(2):119-124. doi:10.1111/j.1365-2982.2007.01020.x; PubMed 28. Pauwels A, Blondeau K, Dupont LJ, Sifrim D. Mechanisms of increased
PMID: 17999650. gastroesophageal reflux in patients with cystic fibrosis. Am J Gastro-
10. Benini L, Ferrari M, Sembenini C, et al. Cough threshold in reflux enterol. 2012;107(9):1346-1353. doi:10.1038/ajg.2012.213; PubMed
oesophagitis: influence of acid and of laryngeal and oesophageal PMID: 22777342.
damage. Gut. 2000;46(6):762-767. PubMed PMID: 10807885; PMCID: 29. Sabati AA, Kempainen RR, Milla CE, et al. Characteristics of gas-
PMC1756455. troesophageal reflux in adults with cystic fibrosis. J Cyst Fibros.
11. Gibson PG, Henry R, Coughlan JJL. Gastro-oesophageal reflux 2010;9(5):365-370. doi:10.1016/j.jcf.2010.06.004; PubMed PMID:
treatment for asthma in adults and children. Cochrane Database Syst 20674518.
Rev. 2003;(2):CD001496. doi:10.1002/14651858.CD001496; PubMed 30. Fathi H, Moon T, Donaldson J, Jackson W, Sedman P, Morice AH.
PMID: CD001496. Cough in adult cystic fibrosis: diagnosis and response to fundoplica-
12. Garg SK, Gurusamy KS. Laparoscopic fundoplication surgery versus tion. Cough. 2009;5(1):1-6. doi:10.1186/1745-9974-5-1.
medical management for gastro-oesophageal reflux disease (GORD) 31. Boesch RP, Acton JD. Outcomes of fundoplication in children with
in adults. Cochrane Database Syst Rev. 2015;(11):CD003243. cystic fibrosis. J Pediatr Surg. 2007;42(8):1341-1344. doi:http://
doi:10.1002/14651858.CD003243.pub3; PubMed PMID: CD003243. dx.doi.org/10.1016/j.jpedsurg.2007.03.030.
Respiratory Complications of Gastroesophageal Reflux Disease CHAPTER 17 227
32. Sweet MP, Patti MG, Leard LE, et al. Gastroesophageal reflux in 41. Li B, Hartwig MG, Appel JZ, et al. Chronic aspiration of gastric
patients with idiopathic pulmonary fibrosis referred for lung fluid induces the development of obliterative bronchiolitis in rat
transplantation. J Thorac Cardiovasc Surg. 2007;133(4):1078-1084. lung transplants. Am J Transplant. 2008;8(8):1614-1621. doi:10.1111/
doi:10.1016/j.jtcvs.2006.09.085; PubMed PMID: 17382656. j.1600-6143.2008.02298.x; PubMed PMID: 18557728.
33. Raghu G, Freudenberger TD, Yang S, et al. High prevalence of 42. Hartwig MG, Appel JZ, Li B, et al. Chronic aspiration of gastric
abnormal acid gastro-oesophageal reflux in idiopathic pulmonary fluid accelerates pulmonary allograft dysfunction in a rat model of
fibrosis. Eur Respir J. 2006;27(1):136-142. doi:10.1183/09031936.06 lung transplantation. J Thorac Cardiovasc Surg. 2006;131(1):209-217.
.00037005; PubMed PMID: 16387946. doi:10.1016/j.jtcvs.2005.06.054; PubMed PMID: 16399314.
34. Sweet MP, Herbella FA, Leard L, et al. The prevalence of distal and 43. Chang JC, Leung JH, Tang T, et al. In the face of chronic aspiration,
proximal gastroesophageal reflux in patients awaiting lung trans- prolonged ischemic time exacerbates obliterative bronchiolitis in
plantation. Ann Surg. 2006;244(4):491-497. doi:10.1097/01. rat pulmonary allografts. Am J Transplant. 2012;12(11):2930-2937.
sla.0000237757.49687.03; PubMed PMID: 16998357; PMCID: doi:10.1111/j.1600-6143.2012.04215.x; PubMed PMID: 22882880;
PMC1856564. PMCID: PMC4332511.
35. Lee JS, Collard HR, Anstrom KJ, et al. Anti-acid treatment and 44. Meers CM, De Wever W, Verbeken E, et al. A porcine model of acute
disease progression in idiopathic pulmonary fibrosis: an analysis lung injury by instillation of gastric fluid. J Surg Res. 2011;166(2):e195-e204.
of data from three randomised controlled trials. Lancet Respir doi:10.1016/j.jss.2010.10.015; PubMed PMID: 21109258.
Med. 2013;1(5):369-376. doi:http://dx.doi.org/10.1016/S2213- 45. Meltzer AJ, Weiss MJ, Veillette GR, et al. Repetitive gastric aspiration
2600(13)70105-X. leads to augmented indirect allorecognition after lung transplantation
36. Lee JS, Ryu JH, Elicker BM, et al. Gastroesophageal reflux therapy in miniature swine. Transplantation. 2008;86(12):1824-1849.
is associated with longer survival in patients with idiopathic pulmonary doi:10.1097/TP.0b013e318190afe6; PubMed PMID: 19104429; PMCID:
fibrosis. Am J Respir Crit Care Med. 2011;184(12):1390-1394. PMC2717556.
doi:10.1164/rccm.201101-0138OC. 46. Tang T, Chang JC, Xie A, Davis RD, Parker W, Lin SS. Aspiration
37. Linden PA, Gilbert RJ, Yeap BY, et al. Laparoscopic fundoplication of gastric fluid in pulmonary allografts: effect of pH. J Surg Res.
in patients with end-stage lung disease awaiting transplantation. 2013;181(1):e31-e38. doi:10.1016/j.jss.2012.06.036; PubMed PMID:
J Thorac Cardiovasc Surg. 2006;131(2):438-446. doi:10.1016/j. 22765998.
jtcvs.2005.10.014; PubMed PMID: 16434276. 47. Atkins BZ, Petersen RP, Daneshmand MA, Turek JW, Lin SS, Davis
38. Hoppo T, Jarido V, Pennathur A, et al. Antireflux surgery preserves RD Jr. Impact of oropharyngeal dysphagia on long-term outcomes
lung function in patients with gastroesophageal reflux disease and of lung transplantation. Ann Thorac Surg. 2010;90(5):1622-1628.
end-stage lung disease before and after lung transplantation. Arch doi:10.1016/j.athoracsur.2010.06.089; PubMed PMID: 20971276.
Surg. 2011;146(9):1041-1047. doi:10.1001/archsurg.2011.216. 48. Hartwig MG, Anderson DJ, Onaitis MW, et al. Fundoplication after
39. Cantu E 3rd, Appel JZ 3rd, Hartwig MG, et al. Maxwell Chamberlain lung transplantation prevents the allograft dysfunction associated
memorial paper. Early fundoplication prevents chronic allograft with reflux. Ann Thorac Surg. 2011;92(2):462-468. doi:10.1016/j.
dysfunction in patients with gastroesophageal reflux disease. Ann athoracsur.2011.04.035; [discussion 8-9], PubMed PMID: 21801907;
Thorac Surg. 2004;78(4):1142-1151; discussion 1151. doi:10.1016/j. PMCID: PMC3617490.
athoracsur.2004.04.044; PubMed PMID: 15464462. 49. Davis RD Jr, Lau CL, Eubanks S, et al. Improved lung allograft
40. Hartwig MG, Appel JZ, Davis RD. Antireflux surgery in the setting function after fundoplication in patients with gastroesophageal
of lung transplantation: strategies for treating gastroesophageal reflux disease undergoing lung transplantation. J Thorac Cardiovasc
reflux disease in a high-risk population. Thorac Surg Clin. Surg. 2003;125(3):533-542. doi:10.1067/mtc.2003.166; PubMed
2005;15(3):417-427. PubMed PMID: 16104132. PMID: 12658195.
CHAPTER
O
ver the past 40 years, there has been a remark- is multifactorial.4 Anatomic impairment of antireflux
able evolution in our understanding of the defenses, poor clearance of refluxate, compromised
pathophysiology, clinical manifestations, and mucosal defense mechanisms, slow gastric emptying,
treatment strategies underlying gastroesophageal reflux underlying motility disturbances, and altered neural
disease (GERD). Initially synonymous with esophagitis sensitivity pathways can all play a role in the generation
and presence of hiatal hernias, we now understand that of troublesome symptoms.5 Precisely how reflux evolves
reflux disease is a much more complicated and often from a physiologic phenomenon to causing bothersome
nuanced diagnosis to make, wherein a myriad of potential symptoms and mucosal disease is complex and variable
nonspecific symptoms and absence of suitable biomarkers among individuals. At its crux, the presence of a normal
can make the diagnosis challenging.1 In the absence of esophagogastric junction (EGJ) with a coordinated lower
overt GERD manifestations (i.e., erosive esophagitis, Barrett esophageal sphincter (LES) and crural diaphragm (CD)
metaplasia, adenocarcinoma) and in light of imperfect pH maintains a mechanical barrier, which is indispensable
testing and symptom correlation, we find that a diagnosis to preservation of normal physiologic function. As such,
of potential GERD has become a catchall for a constellation reflux occurs exclusively during transient lower esophageal
of gastrointestinal and supraesophageal symptoms. As such, sphincter relaxations (TLESRs), is largely restricted to
proton pump inhibitor (PPI) therapy has become first gas, and refluxed fluid is confined to the distal esophagus
line not only for symptoms attributable to reflux disease and rapidly cleared.6 If there is a disruption to any of the
but also for numerous other complaints potentially associ- components of the mechanical barrier, normal function
ated with reflux disease. This has resulted in substantial may be altered. The presence of a mechanically impaired
overuse of PPIs. Complicating matters further, multiple EGJ, either due to an overt hiatal hernia or disruption of
epidemiologic studies and subsequent follow-up articles in the LES-CD complex, results in an increase in non-TLESR
the lay media have raised concerns over possible adverse reflux events, poor gas/liquid discrimination during gastric
events associated with long-term PPI use.2,3 Hence, patients venting, and increased volume and extent of the refluxate
and clinicians alike are becoming wary of continued PPI leading to regurgitation. Coupled with hypersensitivity,
use and are keen on investigating nonpharmacologic this results in troublesome symptoms. Furthermore, other
alternatives to GERD management. Furthermore, the external factors such as obesity, diet, pregnancy, age,
concept of refractory GERD, which has traditionally referred esophageal neuromuscular dysfunction, and trauma can
to persistence of mucosal disease in spite of PPI therapy, instigate and/or exacerbate these mechanisms.
has now evolved to define failure of symptomatic response It is important to note that despite gastric acid secretion
for potential GERD symptoms, highlighting the need for being the primary target for pharmacologic therapy,
alternative therapeutic approaches. Thus, as we find abnormal gastric acid secretion is not a mechanism for
ourselves witnessing the indiscriminate use of PPIs, we pathogenesis of GERD, except in rare cases of true acid
also notice newly heightened concern for their long-term hypersecretion as seen with some neuroendocrine tumors.
safety and the need for different therapies to bridge this In fact, the level of acid secretion in patients with GERD is
therapeutic gap. The aim of this chapter is to review similar to that of asymptomatic controls.7 This highlights
the pathophysiology of GERD and the indications and one of the main limitations of treatment with PPIs, largely
limitations of acid-suppressive therapy and to highlight that it does not address the root causes of GERD or
other treatment modalities aimed at refractory symptoms the hypersensitivity pathways closely tied to bothersome
and nonpharmacologic approaches. symptoms.
ABSTRACT
Medical management of gastroesophageal reflux disease
(GERD) has hinged on the principle that adequate acid
suppression leads to improvement in symptoms attributable
to acid reflux. In the absence of mucosal manifestations
of GERD (i.e., erosive esophagitis, Barrett metaplasia,
adenocarcinoma) and in light of imperfect pH testing
and symptom correlation, the success of treatment does
not always correlate with improvement in symptoms. As
such, the term refractory GERD, which has traditionally
referred to persistence of mucosal disease in spite of
proton pump inhibitor (PPI) therapy, has now evolved to
define failure of symptomatic response for potential GERD
symptoms. Moreover, with multiple recent epidemiologic
studies raising concerns over possible adverse events
associated with long-term PPI use, both patients and
clinicians are wary of continued PPI use, and interest in
nonpharmacologic alternatives to GERD management has
substantially increased. This review highlights the need for
different therapies to bridge this therapeutic gap, while
providing a comprehensive review of alternative medical
and behavioral treatments, along with novel endoscopic
therapies and surgical approaches to GERD.
KEYWORDS
Gastroesophageal reflux disease, proton pump inhibitor,
fundoplication
Acid-Suppression Therapy for Gastroesophageal Reflux Disease and the Therapeutic Gap CHAPTER 18 229
benefit when compared with placebo to warrant further for GERD other than fundoplication. To be successful,
development.28,29 these interventions have to be conceptually valid, have a
An additional approach to management of refractory valid physiologic proof of concept, demonstrate symptom
GERD symptoms, particularly in patients with predominant reduction, demonstrate pH control, have excellent safety
hypersensitivity, is to use neuromodulators to target pain- profiles, and be reimbursable. However, these have proven
processing pathways and reduce symptom perception. to be difficult hurdles and the development of such novel
Attempts to target esophageal acid-sensing targets mediated therapeutics has been challenging. A number of devices
by transient receptor potential cation channel subfamily and procedures have been developed over the years, but
vanilloid member 1 (TRPV1) receptors, which are activated the majority have been ineffective or lacked durability.
by acid, capsaicin, and heat and have been implicated in Currently there are three main types of interventions
esophageal symptom perception, have not been success- that are in use or in active stages of development: (1)
ful thus far.30 However, the use of psychotropic agents radiofrequency ablation therapy (Stretta, Mederi Thera-
(particularly low-dose antidepressants), which modulate peutics, Greenwich, Connecticut), (2) transoral incisionless
symptom perception by targeting pain pathways, has yielded fundoplication (TIF; Esophyx, Endogastric Solutions,
encouraging results. There is evidence for both tricyclic San Mateo, California); and Medigus Ultrasonic Surgical
antidepressants (TCAs) and selective serotonin reuptake Endostapler (MUSE) (Medigus, Israel), and (3) magnetic
inhibitors (SSRIs) in modulating these pathways and sphincter augmentation device (Linx, Torax Medical,
improving symptoms (and/or quality of life),31,32 and these United Kingdom). A brief review of these interventions
should be considered in patients with a hypersensitivity follows.
phenotype.33–35 However, note that recent randomized
control trials of TCAs have failed to show significant RADIOFREQUENCY ABLATION THERAPY: STRETTA
improvements compared with placebo, whereas studies The Stretta procedure was designed to deliver radiofre-
evaluating SSRIs have yielded positive results, suggest- quency energy to the EGJ via an endoscopically placed
ing that the latter may be a preferred initial treatment catheter, with the goal of mechanically altering the EGJ
choice.36–38 and/or modulating neural pathways using thermal energy.
This process was proposed to promote tissue necrosis,
local inflammation, and collagen deposition, thereby
NONPHARMACOLOGIC TREATMENTS FOR “tightening” the LES. Although an interesting concept, this
GASTROESOPHAGEAL REFLUX DISEASE: effect has not been demonstrated to occur. Alternatively
BRIDGING THE THERAPEUTIC GAP? radiofrequency neurolysis may alter sensitivity and reduce
TLESRs because this is a neuronally mediated reflex.43–45
Lifestyle and pharmacologic interventions, including use Randomized sham-controlled trials evaluating the efficacy
of PPIs, have been effective in improving symptoms in a of Stretta failed to show a reduction in acid exposure
majority of patients with GERD. However, none of these when compared with a sham procedure, and a recent
interventions addresses the associated underlying anatomic meta-analysis evaluating three sham-controlled trials and
defects that are implicated in the pathogenesis of GERD. one trial comparing Stretta with PPI therapy failed to show
Surgical treatment, by way of a fundoplication, is the only any benefit in reduction of PPI use or quality of life.46,47
intervention that targets competency of the EGJ, eliminates However, there has been a resurgence of interest after
hiatus hernia, and creates a barrier for prevention of reflux. studies reported comparable symptom outcomes with
As such, laparoscopic fundoplication has been shown to fundoplication.48–51 However, many of the studies were
effectively reduce episodes of all types of reflux, including limited in their design and selection criteria, and most
acid and nonacid.39 It is most effective in patients who have did not include long-term follow-up data.52 Expert opinion
a favorable response to PPI therapy prior to surgery, good and reviews of the current data have yielded conflicting
compliance with acid-suppressive medications, typical symp- results, with some proposing favorable outcomes, whereas
toms, and objective evidence of acid reflux (esophagitis, others pointing out flaws in methodology and suggesting
or abnormal pH testing).40,41 Interestingly, 5-year results of the technology to be ineffective.53 At this time, the verdict
the LOTUS trial, a randomized controlled trial comparing is still out on Stretta, and high-quality studies are needed
laparoscopic Nissen fundoplication to esomeprazole, to determine its viability as a treatment option.
showed no difference in esophagitis or heartburn control
between treatments.42 However, regurgitation was better TRANSORAL INCISIONLESS FUNDOPLICATION:
controlled by fundoplication and adverse events, such
as flatulence, dysphagia, and an inability to belch, were ESOPHYX AND MEDIGUS ULTRASONIC
also more common in the surgical group. Hence, for SURGICAL ENDOSTAPLER
patients with adequate symptom control on PPI therapy, TIF was envisioned as a less invasive alternative to lapa-
continuation of PPI has fewer side effects, lower morbidity, roscopic fundoplication for restoring competency of the
and similar efficacy compared with surgery and should be EGJ, by way of an endoluminal approach. The EsophyX
recommended. Patients with persistent regurgitation as device allows for the creation of a 210- to 300-degree
their main symptom, with at least partial response to PPI fundoplication at the level of the EGJ by using full-thickness
therapy (and proven abnormal pH testing), are the group fasteners to create a valve. An initial randomized control
most likely to benefit from antireflux surgery. trial comparing EsophyX to standard Nissen fundoplication
There has been considerable interest in developing alter- reported similar benefit between the two therapies with a
native, minimally invasive, nonpharmacologic interventions reduced length of stay in the TIF group.54 More recently,
Acid-Suppression Therapy for Gastroesophageal Reflux Disease and the Therapeutic Gap CHAPTER 18 231
the RESPECT trial, a randomized control trial evaluating may fail to address the spectrum of potential phenotypes.
129 patients with troublesome regurgitation, showed We have also learned that there are a large number of
significant improvement in problematic regurgitation patients with potential GERD symptoms, whose underlying
in the TIF/placebo group compared with the sham/ symptoms are likely due to an alternative diagnosis, and
PPI group (67% vs. 45%) with few adverse events.55 The warrant accurate phenotyping prior to being categorized as
main limitations of the TIF technique have been ques- refractory GERD. Furthermore, we have witnessed a change
tions regarding durability and the inability to address the in patients’ perception of safety and long-term viability
problem of hiatus hernia, making it unclear whether or for pharmacologic therapy, particularly with PPIs, which
not it is a viable alternative for patients seeking to avoid has invigorated investigations into alternative lifestyle and
surgical fundoplication. As such, the ideal target population surgical/endoluminal treatments. At the present time, the
has yet to be elucidated. crux of adequate GERD management rests on a number
The MUSE device, which also creates an incisionless fun- of principles: (1) establishing an accurate diagnosis, (2)
doplication, differs from EsophyX in that it uses ultrasound adequately phenotyping patients to determine the best
guidance and a disposable endoscopic stapling device targets for treatments, (3) educating patients on weight
to place sets of transmural surgical staples to secure the loss and the pros and cons of long-term PPI use, and (4)
fundoplication, which may address the issue of durability. investigating other treatment options such as surgical
Treatment data for the MUSE device are very preliminary and endoluminal therapies for patients with incomplete
but have shown promising results, with reductions in need symptom control on medical therapy and those who are
for PPI and improvements in quality of life that have been PPI intolerant or choose not to take these medications long
sustained at 5-year follow-up.56–58 Recently the company term. We are still working to bridge the “therapeutic gap”
has suspended clinical use of the device. in management of GERD, but new promising therapies
may change the landscape of GERD management if they
MAGNETIC SPHINCTER AUGMENTATION prove successful and safe.
DEVICE: LINX
Linx has emerged as a frontrunner in the realm of novel
therapeutic modalities for minimally invasive antireflux
ACKNOWLEDGMENTS
treatments. The magnetic sphincter augmentation device Supported by R01 DK56033 (PJK) from the Public Health Service.
is sized for each patient and surgically positioned around Peter J. Kahrilas is a consultant for Ironwood Pharmaceuticals.
the LES with the intent of generating a barrier for reflux,
while preserving bolus passage and the ability to belch.59 REFERENCES
Repair of a hiatal hernia can also be done at the time of
the procedure. An initial pivotal study showed improvement 1. Kia L, Pandolfino JE, Kahrilas PJ. Biomarkers of reflux disease. Clin
Gastroenterol Hepatol. 2016;14(6):790-797.
in acid exposure (19.9% to 3.3%), symptoms, and PPI use 2. Kia L, Kahrilas PJ. Therapy: risks associated with chronic PPI
in 100 patients undergoing the procedure but was limited use—signal or noise? Nat Rev Gastroenterol Hepatol. 2016;13(5):253-254.
by lack of a control group. The same group published 3. Forgacs I, Loganayagam A. Overprescribing proton pump inhibitors.
5-year follow-up data on 84 of these patients and found BMJ. 2008;336(7634):2-3.
4. Vakil N, van Zanten SV, Kahrilas P, Dent J, Jones R, Global Consensus
sustained response to treatment. They noted that at 5 Group. The Montreal definition and classification of gastroesophageal
years, only 15.3% of patients were on PPIs (compared reflux disease: a global evidence-based consensus. Am J Gastroenterol.
with 100% prior), 1.2% reported moderate to severe 2006;101(8):1900-1920; quiz 43.
regurgitation (vs. 57% prior), and all patients retained 5. Woodland P, Sifrim D. The refluxate: the impact of its magnitude,
the ability to belch. No device erosions, migrations, or composition and distribution. Best Pract Res Clin Gastroenterol.
2010;24(6):861-871.
malfunctions were reported by this group, although a 6. Boeckxstaens G, El-Serag HB, Smout AJ, Kahrilas PJ. Symptomatic
case report of perforation due to device erosion has been reflux disease: the present, the past and the future. Gut.
described in the literature and a case of dysphagia due to 2014;63(7):1185-1193.
erosion into the esophagus has been reported to the US 7. Hirschowitz BI. A critical analysis, with appropriate controls, of
gastric acid and pepsin secretion in clinical esophagitis. Gastroenterol-
Food and Drug Administration.60,61 Note that dysphagia ogy. 1991;101(5):1149-1158.
was noted to be 6% after the procedure compared with 8. Numans ME, Lau J, de Wit NJ, Bonis PA. Short-term treatment with
5% prior to intervention.62 Overall the data for Linx proton-pump inhibitors as a test for gastroesophageal reflux disease:
have been promising, but further studies are needed. If a meta-analysis of diagnostic test characteristics. Ann Intern Med.
further studies validate these findings, Linx may emerge 2004;140(7):518-527.
9. Kahrilas PJ, Boeckxstaens G. Failure of reflux inhibitors in clinical
as a viable alternative therapeutic option for patients with trials: bad drugs or wrong patients? Gut. 2012;61(10):1501-1509.
reflux disease dissatisfied with medical therapy, particularly 10. Roman S, Keefer L, Imam H, et al. Majority of symptoms in esophageal
in the phenotype of patients with persistent regurgitation reflux PPI non-responders are not related to reflux. Neurogastroenterol
as a predominant symptom. Motil. 2015;27(11):1667-1674.
11. Rohof WO, Bennink RJ, de Jonge H, Boeckxstaens GE. Increased
proximal reflux in a hypersensitive esophagus might explain symptoms
CONCLUSION resistant to proton pump inhibitors in patients with gastroesophageal
reflux disease. Clin Gastroenterol Hepatol. 2014;12(10):1647-1655.
There has been considerable advancement in our under- 12. Hemmink GJ, Bredenoord AJ, Weusten BL, Monkelbaan JF, Timmer
standing, phenotyping, and management of GERD over R, Smout AJ. Esophageal pH-impedance monitoring in patients
with therapy-resistant reflux symptoms: ‘on’ or ‘off’ proton pump
the past two decades. inhibitor? Am J Gastroenterol. 2008;103(10):2446-2453.
We have learned that the pathogenesis is complex and 13. Mainie I, Tutuian R, Shay S, et al. Acid and non-acid reflux in
that treatment modalities that target one disease facet patients with persistent symptoms despite acid suppressive therapy:
232 SECTION I Esophagus and Hernia
a multicentre study using combined ambulatory impedance-pH a randomized, double-blind, placebo-controlled study. Am J Gastro-
monitoring. Gut. 2006;55(10):1398-1402. enterol. 2012;107(11):1662-1667.
14. El-Serag H. Role of obesity in GORD-related disorders. Gut. 35. Peghini PL, Katz PO, Castell DO. Imipramine decreases oesophageal
2008;57(3):281-284. pain perception in human male volunteers. Gut. 1998;42(6):807-813.
15. El-Serag HB, Hashmi A, Garcia J, et al. Visceral abdominal obesity 36. Hershcovici T, Wendel CS, Fass R. Symptom indexes in refractory
measured by CT scan is associated with an increased risk of Barrett’s gastroesophageal reflux disease: overrated or misunderstood? Clin
oesophagus: a case-control study. Gut. 2014;63(2):220-229. Gastroenterol Hepatol. 2011;9(10):816-817.
16. Jacobson BC, Somers SC, Fuchs CS, Kelly CP, Camargo CA Jr. 37. Forcelini CM, Tomiozzo JC Jr, Farre R, et al. Effect of nortriptyline
Body-mass index and symptoms of gastroesophageal reflux in women. on brain responses to painful esophageal acid infusion in patients
N Engl J Med. 2006;354(22):2340-2348. with non-erosive reflux disease. Neurogastroenterol Motil. 2014;26(2):
17. Singh M, Lee J, Gupta N, et al. Weight loss can lead to resolution 187-195.
of gastroesophageal reflux disease symptoms: a prospective interven- 38. Ostovaneh MR, Saeidi B, Hajifathalian K, et al. Comparing omeprazole
tion trial. Obesity (Silver Spring). 2013;21(2):284-290. with fluoxetine for treatment of patients with heartburn and normal
18. Rohof WO, Bennink RJ, Smout AJ, Thomas E, Boeckxstaens GE. endoscopy who failed once daily proton pump inhibitors: double-blind
An alginate-antacid formulation localizes to the acid pocket to placebo-controlled trial. Neurogastroenterol Motil. 2014;26(5):670-678.
reduce acid reflux in patients with gastroesophageal reflux disease. 39. Bredenoord AJ, Draaisma WA, Weusten BL, Gooszen HG, Smout
Clin Gastroenterol Hepatol. 2013;11(12):1585-1591; quiz e90. AJ. Mechanisms of acid, weakly acidic and gas reflux after anti-reflux
19. Reimer C, Lodrup AB, Smith G, Wilkinson J, Bytzer P. Randomised surgery. Gut. 2008;57(2):161-166.
clinical trial: alginate (Gaviscon Advance) vs. placebo as add-on 40. Campos GM, Peters JH, DeMeester TR, et al. Multivariate analysis
therapy in reflux patients with inadequate response to a once daily of factors predicting outcome after laparoscopic Nissen fundoplica-
proton pump inhibitor. Aliment Pharmacol Ther. 2016;Feb 22 [Epub tion. J Gastrointest Surg. 1999;3(3):292-300.
ahead of print]. 41. Hayden J, Jamieson G. Optimization of outcome after laparoscopic
20. Tack J. Prokinetics and fundic relaxants in upper functional GI antireflux surgery. ANZ J Surg. 2006;76(4):258-263.
disorders. Curr Opin Pharmacol. 2008;8(6):690-696. 42. Galmiche JP, Hatlebakk J, Attwood S, et al. Laparoscopic antireflux
21. Scarpellini E, Ang D, Pauwels A, De Santis A, Vanuytsel T, Tack J. surgery vs esomeprazole treatment for chronic GERD: the LOTUS
Management of refractory typical GERD symptoms. Nat Rev Gastro- randomized clinical trial. JAMA. 2011;305(19):1969-1977.
enterol Hepatol. 2016;13(5):281-294. 43. Kim MS, Holloway RH, Dent J, Utley DS. Radiofrequency energy
22. Lidums I, Lehmann A, Checklin H, Dent J, Holloway RH. Control delivery to the gastric cardia inhibits triggering of transient lower
of transient lower esophageal sphincter relaxations and reflux by esophageal sphincter relaxation and gastroesophageal reflux in
the GABA(B) agonist baclofen in normal subjects. Gastroenterology. dogs. Gastrointest Endosc. 2003;57(1):17-22.
2000;118(1):7-13. 44. Triadafilopoulos G. Stretta: an effective, minimally invasive treatment
23. Zhang Q, Lehmann A, Rigda R, Dent J, Holloway RH. Control of for gastroesophageal reflux disease. Am J Med. 2003;115(suppl
transient lower oesophageal sphincter relaxations and reflux by the 3A):192S-200S.
GABA(B) agonist baclofen in patients with gastro-oesophageal reflux 45. Kahrilas PJ. Radiofrequency therapy of the lower esophageal sphincter
disease. Gut. 2002;50(1):19-24. for treatment of GERD. Gastrointest Endosc. 2003;57(6):723-731.
24. Vela MF, Tutuian R, Katz PO, Castell DO. Baclofen decreases acid 46. Arts J, Bisschops R, Blondeau K, et al. A double-blind sham-controlled
and non-acid post-prandial gastro-oesophageal reflux measured by study of the effect of radiofrequency energy on symptoms and
combined multichannel intraluminal impedance and pH. Aliment distensibility of the gastro-esophageal junction in GERD. Am J
Pharmacol Ther. 2003;17(2):243-251. Gastroenterol. 2012;107(2):222-230.
25. Ciccaglione AF, Marzio L. Effect of acute and chronic administration 47. Lipka S, Kumar A, Richter JE. No evidence for efficacy of radiofre-
of the GABA B agonist baclofen on 24 hour pH metry and symptoms quency ablation for treatment of gastroesophageal reflux disease:
in control subjects and in patients with gastro-oesophageal reflux a systematic review and meta-analysis. Clin Gastroenterol Hepatol.
disease. Gut. 2003;52(4):464-470. 2015;13(6):1058-1067.e1.
26. Koek GH, Sifrim D, Lerut T, Janssens J, Tack J. Effect of the GABA(B) 48. Coron E, Sebille V, Cadiot G, et al. Clinical trial: radiofrequency
agonist baclofen in patients with symptoms and duodeno-gastro- energy delivery in proton pump inhibitor-dependent gastro-
oesophageal reflux refractory to proton pump inhibitors. Gut. oesophageal reflux disease patients. Aliment Pharmacol Ther.
2003;52(10):1397-1402. 2008;28(9):1147-1158.
27. Lehmann A, Antonsson M, Holmberg AA, et al. (R)-(3-Amino- 49. Liang WT, Wu JN, Wang F, et al. Five-year follow-up of a prospective
2-fluoropropyl) phosphinic acid (AZD3355), a novel GABAB receptor study comparing laparoscopic Nissen fundoplication with Stretta
agonist, inhibits transient lower esophageal sphincter relaxation radiofrequency for gastroesophageal reflux disease. Minerva Chir.
through a peripheral mode of action. J Pharmacol Exp Ther. 2014;69(4):217-223.
2009;331(2):504-512. 50. Liang WT, Yan C, Wang ZG, et al. Early and midterm outcome after
28. Shaheen NJ, Denison H, Bjorck K, Karlsson M, Silberg DG. Efficacy laparoscopic fundoplication and a minimally invasive endoscopic
and safety of lesogaberan in gastro-oesophageal reflux disease: a procedure in patients with gastroesophageal reflux disease: a prospec-
randomised controlled trial. Gut. 2013;62(9):1248-1255. tive observational study. J Laparoendosc Adv Surg Tech A. 2015;25(8):
29. Vakil NB, Huff FJ, Bian A, Jones DS, Stamler D. Arbaclofen placarbil 657-661.
in GERD: a randomized, double-blind, placebo-controlled study. 51. Yan C, Liang WT, Wang ZG, et al. Comparison of Stretta procedure
Am J Gastroenterol. 2011;106(8):1427-1438. and Toupet fundoplication for gastroesophageal reflux disease-related
30. Krarup AL, Ny L, Gunnarsson J, et al. Randomized clinical trial: extra-esophageal symptoms. World J Gastroenterol. 2015;21(45):12882-
inhibition of the TRPV1 system in patients with nonerosive gastro- 12887.
esophageal reflux disease and a partial response to PPI treatment 52. Das B, Reddy M, Khan OA. Is the Stretta procedure as effective as
is not associated with analgesia to esophageal experimental pain. the best medical and surgical treatments for gastro-oesophageal
Scand J Gastroenterol. 2013;48(3):274-284. reflux disease? A best evidence topic. Int J Surg. 2016;30:19-24.
31. Limsrivilai J, Charatcharoenwitthaya P, Pausawasdi N, Leelakusolvong 53. Hopkins J, Switzer NJ, Karmali S. Update on novel endoscopic
S. Imipramine for treatment of esophageal hypersensitivity and therapies to treat gastroesophageal reflux disease: a review. World J
functional heartburn: a randomized placebo-controlled trial. Am J Gastrointest Endosc. 2015;7(11):1039-1044.
Gastroenterol. 2016;111(2):217-224. 54. Svoboda P, Kantorova I, Kozumplik L, et al. Our experience with
32. Keefer L, Kahrilas PJ. Editorial: low-dose tricyclics for esophageal transoral incisionless plication of gastroesophageal reflux disease:
hypersensitivity: is it all placebo effect? Am J Gastroenterol. NOTES procedure. Hepatogastroenterology. 2011;58(109):1208-1213.
2016;111(2):225-227. 55. Hunter JG, Kahrilas PJ, Bell RC, et al. Efficacy of transoral fundoplica-
33. Broekaert D, Fischler B, Sifrim D, Janssens J, Tack J. Influence of tion vs omeprazole for treatment of regurgitation in a randomized
citalopram, a selective serotonin reuptake inhibitor, on oesophageal controlled trial. Gastroenterology. 2015;148(2):324-333.e5.
hypersensitivity: a double-blind, placebo-controlled study. Aliment 56. Kim HJ, Kwon CI, Kessler WR, et al. Long-term follow-up results
Pharmacol Ther. 2006;23(3):365-370. of endoscopic treatment of gastroesophageal reflux disease with
34. Viazis N, Keyoglou A, Kanellopoulos AK, et al. Selective serotonin the MUSE endoscopic stapling device. Surg Endosc. 2016;30(8):
reuptake inhibitors for the treatment of hypersensitive esophagus: 3402-3408.
Acid-Suppression Therapy for Gastroesophageal Reflux Disease and the Therapeutic Gap CHAPTER 18 233
57. Zacherl J, Roy-Shapira A, Bonavina L, et al. Endoscopic anterior 60. Bauer M, Meining A, Kranzfelder M, et al. Endoluminal perforation
fundoplication with the Medigus Ultrasonic Surgical Endostapler of a magnetic antireflux device. Surg Endosc. 2015;29(12):3806-3810.
(MUSE) for gastroesophageal reflux disease: 6-month results from 61. US Food and Drug Administration. MAUDE adverse event report:
a multi-center prospective trial. Surg Endosc. 2015;29(1):220-229. Torax Medica, Inc. Linx Reflux management system anti-reflux
58. Roy-Shapira A, Bapaye A, Date S, Pujari R, Dorwat S. Trans-oral implant;2016. Available at https://www.accessdata.fda.gov/scripts/
anterior fundoplication: 5-year follow-up of pilot study. Surg Endosc. cdrh/cfdocs/cfmaude/detail.cfm?mdrfoi__id=4208665.
2015;29(12):3717-3721. 62. Ganz RA, Edmundowicz SA, Taiganides PA, et al. Long-term outcomes
59. Ganz RA, Peters JH, Horgan S. Esophageal sphincter device for of patients receiving a magnetic sphincter augmentation device for
gastroesophageal reflux disease. N Engl J Med. 2013;368(21):2039-2040. gastroesophageal reflux. Clin Gastroenterol Hepatol. 2016;14(5):671-677.
CHAPTER
G
astroesophageal reflux disease (GERD) is the closure and the creation of a short, 360-degree “floppy”
most common disorder of the esophagus and wrap around 2 to 3 cm of intraabdominal esophagus. This
gastroesophageal junction. While transient reflux of chapter discusses the indications and technical aspects of
stomach contents into the esophagus occurs physiologically, laparoscopic and open fundoplication for the treatment
according to the Montreal Classification,1 the criteria for of GERD.
GERD are met when reflux causes troublesome symptoms
and/or complications. With 60% of American adults
reporting intermittent heartburn symptoms, GERD is
CLINICAL FEATURES
a serious health concern in the Western world. For the As with all operations, proper patient selection is essential
20% of Americans who describe weekly reflux symptoms, for a successful outcome. A thorough history and physical
GERD increases the risk for esophageal stricture, Barrett examination, as well as appropriate laboratory tests, should
esophagus, and esophageal cancer, while significantly affect- be completed to establish a diagnosis of GERD and elimi-
ing health-related quality of life and work productivity.1–5 nate other potential causes of symptoms. Patients should
The modern era of GERD therapy has brought advances be queried regarding classic GERD symptoms including
in diagnosis and treatment, and, subsequently, a better heartburn, regurgitation, and dysphagia. The frequency
understanding of the pathophysiology of GERD. The single and timing of reflux symptoms, the relationship to meals,
most important factor in the development of GERD is symptom exacerbation in the supine or upright position,
an incompetent lower esophageal sphincter.6 Progressive and difficulty swallowing should be noted. The response
dilation plus deterioration of the gastroesophageal flap- and the duration of medical therapy are also recorded,
valve mechanism results in loss of the anatomic antireflux as this information has prognostic significance following
barrier and allows for acid and bile reflux. The goal of fundoplication.
antireflux surgery is to reestablish the competency of the In addition, patients may have atypical symptoms such
lower esophageal sphincter while preserving the patient’s as chronic cough, asthma, pulmonary disease, odynopha-
normal swallowing capacity.7 gia, hoarseness, and chest pain. These patients should
Acid suppression therapies in the form of histamine H2 undergo cardiac evaluation, including a chest radiograph,
receptor antagonists and proton pump inhibitors (PPIs) electrocardiogram, and, if indicated, pulmonary function
have brought both symptomatic relief and effective resolu- tests, in addition to the standard diagnostic evaluation for
tion of esophageal inflammation. While these approaches gastroesophageal reflux. Patients with atypical symptoms
may ameliorate some of the long-term sequelae of GERD, and those who fail to respond to medical therapy may show
medical therapy must be continued indefinitely and does less improvement in symptoms after Nissen fundoplica-
not address nonacid reflux. Antireflux surgery provides an tion than those with typical GERD symptoms.11 Box 19.1
anatomic and physiologic cure with durable symptomatic includes a list of classic and atypical GERD symptoms.
relief and prevention of the adverse consequences of
ongoing esophageal exposure to caustic refluxate.
The current gold standard for the operative treatment
PREOPERATIVE EVALUATION
of GERD is the laparoscopic Nissen fundoplication. This The preoperative evaluation of patients with GERD should
well-established procedure has proved to be both durable be thorough. At a minimum, patients should undergo
and safe over a period of more than 25 years. Follow- esophagogastroduodenoscopy (EGD). Performance of an
ing the introduction of the laparoscopic approach in esophageal motility study (EMS) is currently a preoperative
1991 by Dallemagne and Geagea, the number of Nissen standard to detect esophageal motility disorders that may
fundoplications performed annually increased threefold. lead to troublesome postoperative dysphagia. Although it
While efficacy of medical therapy and concerns regarding has been dogma that patients with ineffective esophageal
potential complications of fundoplication have led to a motility (IEM) (mean distal peristaltic amplitude <30 mm
recent decline in the number of operations performed Hg or >20% loss of peristalsis) should undergo a partial
annually in the United States, the pendulum may be fundoplication to prevent postoperative dysphagia, recent
swinging back with increasing concerns over the sequelae studies demonstrate that postoperative dysphagia after
of chronic acid suppression.8,9 Nissen fundoplication is no greater with IEM than with
Since Rudolf Nissen’s original fundoplication in 1937, normal esophageal motility.12 We routinely perform a
performed to protect a gastroesophageal anastomosis, the preoperative EMS because it also allows documentation of
Nissen fundoplication has undergone many modifications. a motility “baseline” for comparison should postoperative
The principles of modern Nissen fundoplication are dysphagia develop. In addition, it has been reported that
designed to most closely replicate the normal physiology of up to a third of patients with achalasia report symptoms
the gastroesophageal flap valve,10 including secure crural of heartburn and EMS can help rule out this disease.13
234
Fundoplication for Gastroesophageal Reflux Disease CHAPTER 19 234.e1
ABSTRACT
Gastroesophageal reflux disease (GERD) is the most
common disorder of the esophagus and gastroesophageal
junction. While transient reflux of stomach contents into
the esophagus occurs physiologically, the criteria for GERD
are met when reflux causes troublesome symptoms and/
or complications. With 60% of American adults reporting
intermittent heartburn symptoms, GERD is a serious health
concern in the Western world. For the 20% of Americans
who describe weekly reflux symptoms, GERD increases
the risk for esophageal stricture, Barrett esophagus, and
esophageal cancer, while significantly affecting health-
related quality of life and work productivity. The single
most important factor in the development of GERD is
an incompetent lower esophageal sphincter. Progressive
dilation plus deterioration of the gastroesophageal flap-
valve mechanism results in loss of the anatomic antireflux
barrier and allows for acid and bile reflux. The goal of
antireflux surgery is to reestablish the competency of the
lower esophageal sphincter while preserving the patient’s
normal swallowing capacity. The current gold standard
for the operative treatment of GERD is the laparoscopic
Nissen fundoplication. This well-established procedure has
proved to be both durable and safe over a period of more
than 25 years. This chapter discusses the indications and
technical aspects of laparoscopic and open fundoplication
for the treatment of GERD.
KEYWORDS
Gastroesophageal reflux disease (GERD); Nissen; fundo-
plication; laparoscopic antireflux surgery
Fundoplication for Gastroesophageal Reflux Disease CHAPTER 19 235
taking medications confirms the absence of acid reflux. role of partial fundoplication in the treatment of GERD
However, because elimination of excessive reflux is dif- persists, although the majority of experienced American
ficult to achieve in those patients with complications of surgeons prefer to perform complete fundoplication in
GERD, a marker of disease severity, consideration should most patients.
be given to antireflux surgery. Preoperative endoscopic
or medical treatment of esophageal stricture or peptic OPEN VERSUS LAPAROSCOPIC NISSEN
ulcer disease must be accomplished before surgery. In a FUNDOPLICATION
patient with esophageal stricture, preoperative dilation Laparoscopic Nissen fundoplication was first reported
to at least 16 mm (48 French) is advisable to minimize by Dallemagne et al. in 1991.24 Since then, several large
the chance that the customary postoperative dysphagia clinical series of Nissen fundoplication have been reported,
(a result of edema and early postoperative esophageal including longitudinal studies with long-term follow-up
dysmotility) will be compounded by a tight stricture. If that demonstrate the results of both open and laparoscopic
preoperative dilation to 16 mm is successful—several fundoplication to be equivalent.25–28 Several randomized
sessions are sometimes necessary—it is usually possible clinical trials published in the past decade reached the
to extend the dilation intraoperatively to 18 or 20 mm, same conclusion.25,29–31 The laparoscopic approach is
the standard-size dilators used by surgeons for calibrating associated with shorter hospital stays, less postoperative
the fundoplication. Laparoscopic antireflux surgery has pain, fewer wound-related complications, and earlier
also been increasingly used in the treatment of GERD return to work. Despite these advantages, selection of the
following lung transplantation, as acid suppression alone open versus the laparoscopic approach should depend on
has been shown to be inadequate in preventing allograft surgeon experience and the patient’s previous surgical
injury secondary to aspiration of nonacid refluxate.19 history. The intraoperative steps of surgical repair are
In certain subgroups of patients with severe GERD, relatively similar in both approaches. Laparoscopic Nissen
fundoplication may not be the optimal antireflux surgery. fundoplication, however, requires that the surgeon possess
Medically complicated, morbidly obese (body mass index advanced laparoscopic skills.
>35 kg/m2) patients with significant GERD may be better The approach to reoperative Nissen fundoplication
served by undergoing Roux-en-Y gastric bypass as these should be individualized with the goal being long-term
patients have high failure rates following Nissen fundoplica- success of the procedure, since at the end of the day
tion.20 Patients with Barrett esophagus and high-grade that is what will restore quality of life, independent
dysplasia or adenocarcinoma should be treated by mucosal of the approach. Most first-time redo procedures can
ablation or esophageal resection. Severe strictures that are be done laparoscopically, and several large series have
not responsive to dilation therapy should also be treated demonstrated equivalent results with laparoscopic and
by esophageal resection. Patients with low-grade dysplasia open reoperation.32 Laparoscopic reoperation after open
should be treated with high-dose PPIs for 3 months, after surgery, although feasible, may be tedious because the
which they should undergo repeat biopsy. Fundoplication intraabdominal adhesions associated with open surgery may
may be considered in such patients if subsequent pathology be formidable. It is important when planning a reoperation
shows no progression to high-grade dysplasia or carcinoma. to consider why the prior procedure failed and address
Finally, GERD patients with previous gastric surgery should the cause of failure. An unrecognized short esophagus is
be approached cautiously. GERD in patients after gastric one such potential cause, and a Collis gastroplasty may
bypass, partial or sleeve gastrectomy, and vertical banded be an important addition during the reoperation in some
gastroplasty cannot be treated by fundoplication because patients. In the multiply reoperative foregut, consideration
the fundus has been anatomically disrupted by the prior should be given to an alternative approach, such as an
surgery. open thoracic or thoracoscopic approach or resection of
Once a decision is made to perform a surgical antireflux the stomach or esophagus.
procedure on a patient with GERD, the next step is to
decide which type of fundoplication to perform. Recent
data support the concept that Nissen fundoplication is
PRINCIPLES OF NISSEN FUNDOPLICATION
an effective therapy for GERD and is not associated with Basic surgical principles guide the successful performance
significant long-term dysphagia, even in patients with of Nissen fundoplication, regardless of the approach
IEM.21 These data, combined with data suggesting that (laparoscopic or open). Box 19.3 lists the fundamental
partial fundoplication is associated with high long-term principles of Nissen fundoplication.
failure rates,22 have led to a significant decrease in the
application of partial fundoplication in patients with
GERD, regardless of esophageal peristaltic function. Cur- LAPAROSCOPIC NISSEN FUNDOPLICATION
rently, partial fundoplication is reserved for patients with Position and Port Placement
a “named” esophageal motility disorder, such as achalasia Pneumatic sequential compression devices are applied to
or scleroderma, those with both IEM and significant the calves, and 5000 units of subcutaneous heparin may be
dysphagia, and those patients undergoing revision of a administered before the induction of general anesthesia,
prior 360-degree fundoplication for refractory dysphagia. with placement of a Foley catheter, if indicated. The patient
Despite the recent trend toward complete (Nissen) fun- is placed in a split-leg position with both arms tucked
doplication in most patients, contrasting data suggest that and secured to the operating table. A vacuum beanbag
long-term satisfactory results may be achieved with partial mattress can aid in supporting the arms and perineum
fundoplication.23 Consequently, the debate regarding the to prevent patient migration during repositioning of
Fundoplication for Gastroesophageal Reflux Disease CHAPTER 19 237
the operating table. The surgeon stands between the either an open cut-down approach or alternative Veress
patient’s legs with the primary monitor over the patient’s location can be used.
head. The first assistant stands to the patient’s right, and A five-port (one or two 10-mm ports and three or four
the scrub technician stands to the patient’s left. The 5-mm ports) technique is used (Fig. 19.1). Additional ports
first step involves safe entry into the abdomen, which is may be placed as necessary. A camera port, to accom-
achieved in most patients by inserting a Veress needle modate a 5-mm or 10-mm laparoscope, is placed just
at the umbilicus and establishing pneumoperitoneum. superior and to the left of the umbilicus, approximately
In patients with a history of prior abdominal surgery, 12 cm below the xiphoid and approximately 2 to 3 cm
to the left of midline. A 30- or 45-degree laparoscope is
placed through this port. The laparoscopic camera may
be managed by the first assistant, a dedicated camera
BOX 19.3 Primary Principles of Nissen Fundoplication operator, or with a robotic camera holder. A thorough
abdominal exploration with the laparoscope is routinely
Careful positioning to protect patient and surgeon from performed before initiating dissection. All secondary ports
neuromuscular injury are placed under direct vision. After positioning the patient
Safe entry into the abdomen in steep reverse Trendelenburg, allowing the omentum
Circumferential hiatal dissection near the crura, limiting and abdominal organs to fall away from the diaphragm,
manipulation and avoiding injury to the esophagus a second port (10 mm) is placed approximately 10 cm
Circumferential dissection of the esophagus under direct from the xiphoid process along the left costal margin for
vision, with preservation of the vagus nerves the surgeon’s right hand. The 10-mm port accommodates
Adequate mobilization of the esophagus (or Collis an SH or V-20 size needle to allow for intracorporeal
gastroplasty) to attain 2–3 cm of intraabdominal suturing. The third port, for liver retraction, is a 5-mm
esophagus without inferior traction port placed on the right costal margin 12 to 15 cm from
Complete mobilization of the gastric fundus including the xiphoid (depending on the size of the liver). A 5-mm
division of the short gastric vessels and any posterior articulating liver retractor is placed through the right
attachments to the pancreas lateral port under laparoscopic visualization, and the
Closure of hiatal defect by reapproximation of the
left lateral lobe of the liver is retracted anteriorly and
diaphragmatic crura.
superiorly to expose the hiatus (Fig. 19.2). The right crus
Creation of a short (2 cm), “floppy” (tension-free)
and caudate lobe of the liver should be clearly visible
fundoplication around the distal esophagus only and
anchored to the esophagus
through the gastrohepatic ligament or pars flaccida if
the liver retraction is adequate. Alternatively, the liver
Liver
retractor
cm
10 12 cm
15
cm
12–
Camera
Surgeon Surgeon
A B
FIGURE 19.1 (A) Arrangement of laparoscopic ports during a standard Nissen fundoplication and (B) alternative port site positioning, with
the first assistant to the left of the patient, for cases in which a dedicated camera driver is unavailable. Small circles represent 5-mm
radially dilating trocars and large circles denote 11- to 12-mm trocars.
238 SECTION I Esophagus and Hernia
Liver
Diaphragm
RC
LC
*
Caudate
lobe
Spleen
Stomach
A B
FIGURE 19.2 (A) The liver is reflected anteriorly to expose the esophageal hiatus using a snake-type retractor through the most lateral port
on the patient’s right side. (B) The liver retractor is secured in place to provide exposure of the gastrohepatic ligament, esophagus, and
hiatus. The caudate lobe of the liver is visible through the thin tissue of the pars flaccida. *Hepatic branch of the anterior vagus nerve
running along a replaced left hepatic artery. LC, Left crus; RC, right crus.
can be retracted with a Nathanson retractor placed just of prominent vessels, as well as the hepatic branch of the
inferior and to the right of the xiphoid process. The liver anterior vagus nerve (Fig. 19.2B), can be attempted, but
retractor is stabilized with an endoscopic instrument holder clinically significant liver ischemia has not been reported
attached to the operating table. The fourth port (5 mm), in cases where division is necessary. Dissection of the lesser
for the assistant, is generally placed midway between the curve is extended superiorly, up to the EGJ, to reveal the
liver retractor and the camera port. Finally, based on caudate lobe below and expose the hiatus and the right
the location of the inferior edge of the liver after securing crus of the diaphragm. The peritoneum just medial to the
the retractor, a 5-mm port is placed to the right of midline, right crus is incised and the esophagus carefully dissected
at the level of the 10-mm dissecting port, so that instruments away from its border (Fig. 19.4). The posterior vagus nerve
in the surgeon’s left hand triangulate the hiatus without is identified and preserved as the posterior mediastinal
hitting the falciform ligament or left lobe of the liver. tissue is bluntly swept medially. Using the open jaws of the
In the absence of a dedicated camera holder, the atraumatic grasper to provide anterolateral retraction on
assistant can stand on the left side of the patient and the right crus of the diaphragm, the dissection is continued
control the laparoscope with the left hand. Should this be clockwise up to the phrenoesophageal membrane. The
necessary, the fourth port is moved from the right upper thin anterior leaf of the phrenoesophageal membrane is
quadrant to a location along the left costal margin, at divided along the apex of the hiatus (see Fig. 19.3) and
least 5 cm below the 10 mm port used for the surgeon’s dissection is continued across the top of the crural arch
right hand. until the left crus is exposed. The anterior vagus nerve
runs along the esophagus in this region and should be
Exposure identified and preserved. The dissection is then carried
An atraumatic grasper is used by the assistant to provide down the border of the left crus until the angle of His
retraction of the stomach. The phrenoesophageal fat and the gastric fundus limit further inferior dissection.
pad along the lesser curvature just below the EGJ is used Circumferential hiatal dissection is initiated bluntly and
for inferior retraction to minimize the risk of gastric extended into the mediastinum by introducing the closed
or esophageal injury. The operating surgeon uses an tip of round-nosed atraumatic graspers between the right
atraumatic grasper in the left hand and an ultrasonic crus and the esophagus and opening them vertically
scalpel or advanced energy device in the right hand. (12-o’clock and 6-o’clock positions) (see Fig. 19.4). The
open jaws of the grasper are then used to retract the right
Dissection crus laterally, while the instrument in the right hand gently
Exposure of the Hiatus. The assistant retracts the stomach sweeps the esophagus medially. The use of thermal devices
inferolaterally to the patient’s left, putting the gastrohepatic is limited during mediastinal dissection to minimize injury
ligament on tension. The surgeon’s left hand grasps to the vagus nerves or esophagus.
just above the pars flaccida, which is opened using the Complete Mobilization of the Gastric Fundus. Dissection
ultrasonic scalpel (Fig. 19.3). An aberrant left hepatic of the fundus of the stomach is begun by identifying the
artery may be present in the pars flaccida in up to 13% point on the greater curvature approximately a third
of patients and often requires division as the dissection of the distance from the angle of His to the antrum. A
of the gastrohepatic ligament is carried superiorly toward convenient landmark for this point is the inferior pole
the base of the right crus of the diaphragm. Preservation of the spleen, roughly 10 to 15 cm inferior to the angle
Fundoplication for Gastroesophageal Reflux Disease CHAPTER 19 239
A B
C D
FIGURE 19.3 (A) Opening of the pars flaccida with the ultrasonic scalpel to begin dissection of the gastrohepatic ligament. (B) Dissection
of the lesser curve (extending superiorly) exposes the right crus of the diaphragm. (C) Division of the phrenoesophageal ligament is
continued along the apex of the hiatus, with (D) the assistant providing retraction by grasping the fat pad overlying the gastroesophageal
junction, elevating the tissue away from the underlying esophagus.
of His. The lateral border of the fundus is grasped and 2. If the greater omentum obscures the superior pole of
retracted anteriorly and to the right, while the gastrocolic the spleen, it should be retracted inferiorly. This may
omentum is grasped and retracted ventrally and to the be accomplished by increasing the degree of reverse
left. The lesser sac is entered approximately 5 to 10 mm Trendelenburg, introducing an additional port and
away from the greater curve of the stomach, using the grasper in the left flank or placing a broad-based,
ultrasonic scalpel (Fig. 19.5). Injury to the greater curvature figure-of-eight “reefing” suture in the greater omentum
of the stomach can be avoided by rocking the surgeon’s and retracting the omentum through the left lateral
left hand up and down to visualize the posterior and port with the two long ends of this suture.
anterior surface of the short gastric vessels as they are 3. Layer the dissection of the vascular structures at the
divided. Any visible thermal injury to the stomach may superior pole of the spleen, starting with the visceral
warrant placement of an imbricating suture. The short peritoneal reflection, then the short gastric vessels, and
gastric vessels are divided individually with the ultrasonic then the retroperitoneal gastrophrenic tissues. Dividing
scalpel until the superior pole of the spleen is reached. the pancreaticogastric peritoneal fold and the posterior
As one proceeds superiorly, three strategies may help gastric artery is necessary to fully mobilize the fundus
dissection in this area: and reach the base of the left crus posteriorly.
1. Expose the superior pole of the spleen with “triangular Mediastinal and Posterior Esophagus. After fully mobilizing
retraction.” The three corners of retraction in the the fundus, the base of the left crus is reached. If the
axial plane are the spleen tip, the surgeon’s left-hand earlier dissection reached the base of the right crus, the
instrument retracting anteromedially on the anterior plane behind the esophagus is complete. Passing a Babcock
wall of the fundus, and the first assistant retracting grasper from right to left through the retroesophageal
posteromedially on the posterior wall of the stomach. window and grasping the apex of the fundus near the
240 SECTION I Esophagus and Hernia
A B
C D
FIGURE 19.4 Initiation of the mediastinal dissection. (A) The right crus is grasped and (B) retracted laterally, while the right-handed
instrument is used to bluntly dissect the esophageal tissue medially. (C) As the dissection is carried higher into the mediastinum, the
assistant provides caudal retraction and (D) the left-handed instrument, with jaws opened, is moved inside the hiatus and provides
countertraction.
foreshortened, necessitating further mediastinal dissec- position and confirm the absence of redundant tissue
tion, an esophageal lengthening procedure (i.e., Collis posteriorly (Fig. 19.7). Grasping a point too low on the
gastroplasty), or both (see Acquired Short Esophagus, greater curvature may predispose to the latter error. The
later). fundoplication should not retract when the graspers are
To best assess intraabdominal esophageal length, the released; the presence of such a “spring sign” suggests
fundus or Penrose drain is released and the distance excessive rotational tension and mandates reassessment
from the gastroesophageal junction to the crural closure of the completeness of the dissection.
is measured. At least 2.5 cm of esophagus must be within
the abdomen and under no tension. If the maximal Reconstruction
mediastinal dissection does not adequately reduce more Crural Closure. The crura are closed from the right of
than 2.5 cm of intraabdominal esophagus, a Collis gastro- the esophagus with interrupted nonabsorbable 0-Ethibond
plasty should be performed. The gastric fundus is then sutures placed 8 to 10 mm apart, 5 to 10 mm back from the
passed posteriorly from the patient’s left to right with crural edge. The peritoneal covering of the crura should
atraumatic graspers to assess for adequate mobilization. be incorporated into the repair, and the sutures should be
The “shoeshine maneuver” involves sliding the fundus “staggered” in the anterior–posterior plane on the crura
back and forth behind the esophagus to confirm good to avoid splitting the crural musculature along the length
of the repair. The completed crural closure should be
calibrated such that the crura just touch the walls of the
empty esophagus (Fig. 19.8). To prevent reherniation, a
variety of techniques have been used to reduce tension
at the crural closure including buttressing with 1-cm2
Teflon felt patches, felt strips, or a piece of absorbable
or nonabsorbable mesh. More recently, diaphragmatic
relaxing incisions have allowed for significant reduction
in tension at the crural closure while avoiding the risk
of erosion associated with foreign material surrounding
the esophagus.33,34
Fundoplication. A 56- or 60-French esophageal dilator is
passed transorally into the stomach by the anesthesiologist
under direct laparoscopic vision by the surgeon. Good
communication and slow advancement of the dilator are
essential to minimize the risk of perforation at the EGJ.
The dilator should pass without resistance. If resistance is
encountered, the dilator is removed and a smaller dilator
is passed. An indication that esophageal injury may have
occurred is the appearance of blood on the dilator when
FIGURE 19.6 Following completion of circumferential hiatal it is removed. The size of the dilator is then increased
dissection, an atraumatic instrument is used to grasp the apex of until resistance is noted.
the gastric fundus and pass the tissue through the After dilator placement, the central stitch of the fun-
retroesophageal window from left to right. doplication is placed 1.5 cm proximal to the EGJ with
FIGURE 19.7 The “shoeshine maneuver,” one-to-one movement of the fundus through the retroesophageal window, is performed to
ensure that the fundoplication is in good position without tension or excess tissue posterior to the esophagus.
242 SECTION I Esophagus and Hernia
A B
FIGURE 19.8 (A) Crural closure with nonabsorbable suture starting posteriorly and working anteriorly. (B) Each bite should incorporate
endoabdominal fascia along with the musculature of the left and right crus. (C) Sutures are placed to close the hiatus such that it abuts
the esophageal wall without impinging it or causing anterior angulation.
simple interrupted 2-0 nonabsorbable suture; seromuscular is attained, either through mediastinal dissection or via
bites are taken through each side of the fundoplication, an esophageal lengthening procedure.
without incorporating the esophagus as the first stitch
can be under increased tension as the wrap is being OPEN NISSEN FUNDOPLICATION
constructed. Additional sutures are placed 1 cm above and The principal steps in performing an open Nissen fundo-
below the initial suture, including partial-thickness bites of plication are similar to the laparoscopic approach. Open
the esophageal wall, to create a 2-cm-long fundoplication Nissen fundoplication is indicated if surgeons do not have
that is secured to the esophagus just above the level of adequate laparoscopic experience, patients have dense
the EGJ (Fig. 19.9). The tightness of the fundoplication is adhesions because of previous upper gastrointestinal opera-
tested after placing each suture by gently sliding a blunt- tions, or a complication occurs that requires laparotomy
ended gasper between the esophagus and the wrap. The to address. Despite the improved tactile feedback with
grasper should easily slide along the esophagus, and lateral an open approach, visual exposure of the hiatus may be
retraction of the wrap should visualize the diaphragm less easily achieved than with a laparoscopic approach.
between the wrap and the esophagus. Knots may be tied Because the techniques involved in open and laparoscopic
extracorporeally, but intracorporeal knotting decreases fundoplication are similar, the following sections address
tissue trauma and optimizes knot tension and position. only significant differences.
Some authors advocate infradiaphragmatic fixation of
the fundoplication to the crura to prevent reherniation, Exploration and Exposure
but there is no evidence that this decreases failure rates if An upper midline incision with the use of a self-retaining
adequate, tension-free intraabdominal esophageal length retractor allows adequate exposure. A liver retractor placed
Fundoplication for Gastroesophageal Reflux Disease CHAPTER 19 243
No. 15
Cervical dilator
close to the most posterior part of the left lateral lobe of FIGURE 19.10 The original description of a “floppy” Nissen
the liver permits improved visualization. Optimal exposure fundoplication by Donahue and Bombeck et al. in 1977.
is obtained when the diaphragm is seen to run vertically Fr, French. (From Donahue PE, Larson GM, Stewardson RH,
from the upper end of the incision directly posteriorly Bombeck CT. Floppy Nissen fundoplication. Rev Surg.
to the hiatus. In patients with large left lateral lobe of 1977;34:223.)
the liver, mobilization of the left lateral liver by dividing
the left coronal ligament may be necessary to achieve
adequate exposure. of the fundus, the size of the esophageal dilator, and the
Lesser Curve. The thin gastrohepatic ligaments are sutures placed to create the fundoplication.
incised, extended superiorly, and carried over the anterior
surface of the esophagus as described earlier. Similarly,
an aberrant left hepatic artery and hepatic branch of the
ACQUIRED SHORT ESOPHAGUS
vagus are protected if encountered in the pars flaccida. The presence of a short esophagus increases the difficulty
The anterior vagus is likewise identified and protected. of laparoscopic Nissen fundoplication. Up to 20% of surgi-
Crus. By retracting the lesser curve inferolaterally and to cal failures with Nissen fundoplication may be the result
the right, the left crus is exposed. The right crus is dissected of not recognizing a short esophagus. A short esophagus
bluntly with the left fingers to create a retroesophageal is discovered more frequently in patients with esophageal
space. A Penrose drain is passed around the lower part stricture, Barrett esophagus, and type III paraesophageal
of the esophagus, excluding the posterior vagus nerve, hernia. Esophageal foreshortening is thought to occur as
and used as a retractor to provide better visualization of a result of recurrent transmural inflammation from acid
the retroesophageal space. peptic injury and subsequent fibrosis of the mediastinal
Mediastinal and Posterior Esophagus. With retraction esophagus. Given its pathogenesis, it is not surprising that
on the Penrose drain, the esophagus can be dissected esophageal stricture is often associated with esophageal
circumferentially. Similarly, the mediastinal dissection foreshortening. Large hiatal hernias may also be associated
can be carried superiorly. with a short esophagus as a result of chronic cephalad
Fundus and Greater Curve. The loose attachments displacement of the gastroesophageal junction. Preopera-
between the fundus and the left diaphragm are taken tive results of barium swallow and EGD may provide an
down. The short gastric vessels are sequentially divided indication of a short esophagus, but no combination
with clamps and suture or ligated with the ultrasonic of preoperative clinical variables reliably predicts the
shears, as described earlier (see Fig. 19.5). presence of a short esophagus, and the diagnosis of this
Repair and Fundoplication. Crural repair and fundoplica- entity continues to be made definitively at the time of
tion are performed as described earlier. A “floppy” Nissen operation, where it is defined as failure to achieve 2.5 cm
fundoplication requires that the fundic wrap admit the of intraabdominal esophagus after maximal mediastinal
surgeon’s index finger between the wrap and the esophagus dissection techniques.
with the dilator in place (Fig. 19.10). Factors that influence Collis gastroplasty achieves esophageal lengthening
the tightness of the wrap are the degree of mobilization by using the gastric cardia to create a neoesophagus.
244 SECTION I Esophagus and Hernia
Greater curve
FIGURE 19.11 Laparoscopic stapling of the fundus from the FIGURE 19.12 Lengthening of the esophagus by laparoscopic
greater curve toward the lesser curve to initiate a wedge stapling parallel to the esophagus (with a dilator in place) in the
fundectomy. cranial direction.
same study reported relief of GERD symptoms in 92.4% reflux disease: a global evidence-based consensus. Am J Gastroenterol.
and 90.7% of patients after 10 years for laparoscopic and 2006;101:1900.
5. Friedenburg FK, Hanlon A, Vanar V, et al. Trends in gastroesophageal
open Nissen fundoplications, respectively. Acid-suppression reflux disease as measured by the National Ambulatory Medical
therapy was used in 20% of the patients at 10 years, but 65% Care Survey. Dig Dis Sci. 2010;55:1911.
of these patients did not have objective evidence of reflux 6. Zaninotto G, DeMeester TR, Schwizer W, Johansson KE, Cheng SC.
based on 24-hour pH-impedance studies.41 Eleven-year The lower esophageal sphincter in health and disease. Am J Surg.
1988;155:104.
follow-up of laparoscopic Nissen fundoplication revealed 7. Stein HJ, Crookes PF, DeMeester TR. Three-dimensional manometric
that 92.5% of patients were satisfied with the results of their imaging of the lower esophageal sphincter. Surg Annu. 1995;27:
surgery, 90% had durable improvement of their symptoms, 199.
70% of patients were off acid-suppression therapy, and the 8. Finlayson SR, Laycock WS, Birkmeyer JD. National trends in utilization
revisional surgery rate was 8.3%.42 Surgical failure is most and outcomes of antireflux surgery. Surg Endosc. 2003;17:864.
9. Fuchs KH, DeMeester TR, Albertucci M. Specificity and sensitivity
often attributable to (1) complete disruption of the wrap, of objective diagnosis of gastroesophageal reflux disease. Surgery.
(2) a slipped Nissen fundoplication (in which part of the 1987;102:575.
stomach lies above and part lies below the fundoplication), 10. Hunter JG, Trus TL, Branum GD, Waring JP, Wood WC. A physiologic
or (3) herniation of an intact wrap through the hiatus into approach to laparoscopic fundoplication for gastroesophageal reflux
disease. Ann Surg. 1996;223:673; discussion 685.
the chest.32,43 Surgical failures may require reoperation.27,39 11. Farrell TM, Richardson WS, Trus TL, Smith CD, Hunter JG. Response
Patients should be cautioned that the results of reoperation of atypical symptoms of gastro-oesophageal reflux to antireflux
for GERD are never as favorable as the results after a surgery. Br J Surg. 2001;88:1649.
primary operation and that residual atypical symptoms 12. Rydberg L, Ruth M, Abrahamsson H, et al. Tailoring antireflux
may persist. Laparoscopic reoperative fundoplication is surgery: a randomized clinical trial. World J Surg. 1999;23:612.
13. Spechler SJ, Souza RF, Rosenberg SJ, Ruben RA, Goyal RK. Heartburn
technically feasible by experienced surgeons. in patients with achalasia. Gut. 1995;37:305.
14. Campos GM, Peters JH, DeMeester TR, et al. Multivariate analysis of
CONCLUSION factors predicting outcome after laparoscopic Nissen fundoplication.
J Gastrointest Surg. 1999;3:292.
15. Tseng D, Rizvi AZ, Fennerty MB, et al. Forty-eight-hour pH monitoring
Antireflux surgery is an excellent treatment option for increases sensitivity in detecting abnormal esophageal acid exposure.
patients with symptoms of GERD that are inadequately J Gastrointest Surg. 2005;9:1043; discussion 1051.
treated with medication, for patients who desire to avoid 16. Shay S, Tutuian R, Sifrim D, et al. Twenty-four hour ambulatory
lifelong medical therapy, or for patients with significant simultaneous impedance and pH monitoring: a multicenter report
complications from acid reflux. The effect of antireflux of normal values from 60 healthy volunteers. Am J Gastroenterol.
2004;99:1037.
surgery on progression of Barrett esophagus is not fully 17. Castell DO, Vela M. Combined multichannel intraluminal impedance
understood, and patients with Barrett esophagus who and pH-metry: an evolving technique to measure type and proximal
undergo antireflux surgery still require routine endoscopic extent of gastroesophageal reflux. Am J Med. 2001;111(suppl 8A):157S.
surveillance. The introduction of a laparoscopic approach 18. Kwiatek MA, Pandolfino JE, Hirano I, Kahrilas PJ. Esophagogastric
junction distensibility assessed with an endoscopic functional luminal
to fundoplication should not alter the operative indica- imaging probe (EndoFLIP). Gastrointest Endosc. 2010;72:272.
tions. Finally, to ensure successful surgical outcomes, an 19. Robertson AG, Krishnan A, Ward C, et al. Anti-reflux surgery in
understanding of disease pathophysiology, preoperative lung transplant recipients: outcomes and effects on quality of life.
diagnostic evaluation, appropriate patient selection, and Eur Respir J. 2012;39:691.
complete familiarity with the various types of antireflux 20. Morgenthal CB, Lin E, Shane MD, Hunter JG, Smith CD. Who will
fail laparoscopic Nissen fundoplication? Preoperative prediction of
procedures available are essential. long-term outcomes. Surg Endosc. 2007;21:1978.
The introduction of endoscopic therapies for reflux 21. Pizza F, Rossetti G, Del Genio G, Maffettone V, Brusciano L, Del
control has provided new options for the patient with Genio A. Influence of esophageal motility on the outcome of
GERD. However, the effectiveness and durability of laparoscopic total fundoplication. Dis Esophagus. 2008;21:78.
22. Horvath KD, Jobe BA, Herron DM, Swanstrom LL. Laparoscopic
these endoscopic therapies have yet to approximate the Toupet fundoplication is an inadequate procedure for patients with
outcomes of surgical fundoplication. Regardless of these severe reflux disease. J Gastrointest Surg. 1999;3:583.
advances, surgical therapy for GERD will continue to play 23. Watson DI, Jamieson GG, Lally C, et al. Multicenter, prospective,
an important role in patients with complicated disease, double-blind, randomized trial of laparoscopic Nissen vs anterior
such as those with large hiatal hernias or a shortened 90 degrees partial fundoplication. Arch Surg. 2004;139:1160.
24. Dallemagne B, Weerts JM, Jehaes C, Markiewicz S, Lombard R.
esophagus. Laparoscopic Nissen fundoplication: preliminary report. Surg Laparosc
Endosc. 1991;1:138.
25. Ackroyd R, Watson DI, Majeed AW, Troy G, Treacy PJ, Stoddard CJ.
REFERENCES Randomized clinical trial of laparoscopic versus open fundoplication
for gastro-oesophageal reflux disease. Br J Surg. 2004;91:975.
1. Shaker R, Castell DO, Schoenfeld PS, Spechler SJ. Nighttime 26. Terry M, Smith CD, Branum GD, Galloway K, Waring JP,
heartburn is an under-appreciated clinical problem that impacts Hunter JG. Outcomes of laparoscopic fundoplication for gastro-
sleep and daytime function: the results of a Gallup survey conducted esophageal reflux disease and paraesophageal hernia. Surg Endosc.
on behalf of the American Gastroenterological Association. Am J 2001;15:691.
Gastroenterol. 2003;98:1487. 27. Granderath FA, Kamolz T, Schweiger UM, et al. Long-term results
2. Wu AH, Tseng CC, Bernstein L. Hiatal hernia, reflux symptoms, of laparoscopic antireflux surgery. Surg Endosc. 2002;16:753.
body size, and risk of esophageal and gastric adenocarcinoma. 28. Viljakka MT, Luostarinen ME, Isolauri JO. Complications of open
Cancer. 2003;98:940. and laparoscopic antireflux surgery: 32-year audit at a teaching
3. Srinivasan R, Tutuian R, Schoenfeld P, et al. Profile of GERD in the hospital. J Am Coll Surg. 1997;185:446.
adult population of a northeast urban community. J Clin Gastroenterol. 29. Lundell L, Dalenback J, Hattlebakk J, et al. Outcome of open
2004;38:651. antireflux surgery as assessed in a Nordic multicentre prospective
4. Vakil N, van Zanten SV, Kahrilas P, Dent J, Jones R, Global Consensus clinical trial. Nordic GORD-Study Group. Eur J Surg. 1998;164:751.
Group. The Montreal definition and classification of gastroesophageal Erratum in Eur J Surg 165:1104, 1999.
Fundoplication for Gastroesophageal Reflux Disease CHAPTER 19 247
30. Laine S, Rantala A, Gullichsen R, Ovaska J. Laparoscopic vs con- 36. Johnson AB, Oddsdottir M, Hunter JG. Laparoscopic Collis
ventional Nissen fundoplication. A prospective randomized study. gastroplasty and Nissen fundoplication. A new technique for the
Surg Endosc. 1997;11:441. management of esophageal foreshortening. Surg Endosc. 1998;12:
31. Heikkinen TJ, Haukipuro K, Bringman S, Ramel S, Sorasto A, Hulkko 1055.
A. Comparison of laparoscopic and open Nissen fundoplication 2 37. Terry ML, Vernon A, Hunter JG. Stapled-wedge Collis gastroplasty
years after operation. A prospective randomized trial. Surg Endosc. for the shortened esophagus. Am J Surg. 2004;188:195.
2000;14:1019. 38. Watson DI, Jamieson GG. Antireflux surgery in the laparoscopic
32. Hunter JG, Smith CD, Branum GD, et al. Laparoscopic fundoplication era. Br J Surg. 1998;85:1173.
failures: patterns of failure and response to fundoplication revision. 39. Watson DI, Jamieson GG, Game PA, Williams RS, Devitt PG. Lapa-
Ann Surg. 1999;230:595; discussion 604. roscopic reoperation following failed antireflux surgery. Br J Surg.
33. Crespin OM, Yates RB, Martin AV, Pellegrini CA, Oelschlager BK. 1999;86:98.
The use of crural relaxing incisions with biologic mesh reinforcement 40. Power C, Maguire D, McAnena O. Factors contributing to failure
during laparoscopic repair of complex hiatal hernias. Surg Endosc. of laparoscopic Nissen fundoplication and the predictive value of
2016;30:2179. doi:10.1007/s00464-015-4522-1. preoperative assessment. Am J Surg. 2004;187:457.
34. Alicuben ET, Worrell SG, DeMeester SR. Impact of crural 41. Broeders JA, Jong R, Draaisma W, Bredenoord AJ, Smout AJ, Gooszen
relaxing incisions, Collis gastroplasty, and non-cross-linked HG. Ten-year outcome of laparoscopic and conventional Nissen
human dermal mesh crural reinforcement on early hiatal hernia fundoplication. Ann Surg. 2009;250:698.
recurrence rates. J Am Coll Surg. 2014;219:988. doi:10.1016/ 42. Morgenthal CB, Shane MD, Stival A, et al. The durability of laparo-
j.jamcollsurg.2014.07.937. scopic Nissen fundoplication: 11-year outcomes. J Gastrointest Surg.
35. Swanstrom LL, Marcus DR, Galloway GQ. Laparoscopic Collis 2007;11:693.
gastroplasty is the treatment of choice for the shortened esophagus. 43. Hinder RA, Klingler PJ, Perdikis G, Smith SL. Management of the
Am J Surg. 1996;171:477. failed antireflux operation. Surg Clin North Am. 1997;77:1083.
CHAPTER
T
he continuous search for the ideal antireflux pro- downward trend in the use of surgical fundoplication has
cedure reflects a widely held perception among been noted in the United States over the past decade.12–14
surgeons, gastroenterologists, and patients that The decline in surgical volume has been attributed to the
therapy for gastroesophageal reflux disease (GERD) perceived risk of fundoplication failure, to the availability
remains unsatisfactory. About 30% to 40% of patients of over-the-counter PPI and endoscopic therapies, and to
are resistant or only partial responders to proton pump the rise of bariatric surgery.
inhibitor (PPI) therapy,1,2 and even high-dose escalation The limitations of both PPI therapy and fundoplication
may be inadequate to maintain individuals in a symptom- have left many patients and clinicians in the equivocal
free state with a mechanically defective lower esophageal position of tolerating a lifetime drug dependence with
sphincter (LES). Additionally, there are growing concerns incomplete symptom relief, or undertaking the risk of
over the long-term effects of chronic acid suppression. a surgical procedure that alters gastric anatomy, may
Many patients suffer from persistent nonacid reflux and have considerable side effects, and may deteriorate
nocturnal acid breakthrough, and may progress to serious over time. The Linx Reflux Management System (Torax
complications of the disease, such as volume regurgitation Medical, St. Paul, Minnesota) is a US Food and Drug
with pulmonary aspiration and Barrett metaplasia, the Administration–approved device designed to provide a
leading risk factor for esophageal adenocarcinoma.3 A permanent solution to GERD by augmenting the LES
large European open cohort multicenter study showed barrier with a standardized laparoscopic procedure.
that about 10% of patients under routine medical care
progressed to Barrett esophagus in 5 years of follow-up.4
Recent literature also indicates that chronic acid sup-
MAGNETIC SPHINCTER AUGMENTATION
pression with PPIs can reduce the absorption of vitamin The Linx is a simple mechanical device designed to
B12 and magnesium, the effectiveness of medications augment the physiologic barrier to reflux by magnetic force.
such as clopidogrel, and increase the risk of Clostridium The device consists of a series of biocompatible titanium
difficile infection.5 Other consequences of prolonged PPI beads with magnetic cores hermetically sealed inside. The
therapy include hypergastrinemia, enterochromaffin-like beads are interlinked with independent titanium wires to
cell hyperplasia, and parietal cell hypertrophy, leading form a flexible and expandable ring. At rest, each bead
to rebound acid hypersecretion.6 Finally, there is some is in contact with adjacent beads. The beads can move
evidence suggesting that chronic acid suppression may be independent of the adjacent beads, creating a dynamic
associated with an increased incidence of gastric cancer.7 implant that does not compress the esophagus and does not
The laparoscopic Nissen fundoplication is the current limit its range of motion upon swallowing, belching, and
surgical gold standard for the treatment of GERD. It is vomiting (Fig. 20.1). For reflux to occur, the intragastric
a safe, effective, and durable antireflux procedure when pressure must overcome the resistance to opening of both
performed in specialized centers. A multicenter European the patient’s native LES pressure and the magnetic bonds
trial comparing medical therapy with total or partial of the device. The Linx is manufactured in different sizes
fundoplication performed in selected centers by expert and is capable of nearly doubling its diameter when all
surgeons showed that 92% of medical patients and 85% beads are separated. The magnetic attraction force to be
of surgical patients remained in remission at 5 years of counteracted to allow bead separation is independent of
follow-up.8 However, despite remarkably low morbidity the number of beads contained in the device. The Linx
and mortality rates, the operation is underused due to the device, while augmenting the LES, allows for expansion to
perception of long-term side effects and fear of failure, accommodate a swallowed bolus or the escape of elevated
which impacts referral patterns.9 Also, wide variability gastric pressure associated with belching or vomiting.
in clinical outcomes related to interindividual surgical Once healing is complete after the implant, the device
expertise and/or technical modifications10 have limited is encapsulated in fibrous tissue but is not incorporated
the adoption of this procedure especially in patients with in the esophageal wall15; this makes it possible to remove
early GERD. Patients undergoing a Nissen fundoplica- the device without damage to the esophagus. The Linx
tion are especially at risk for potential side effects of the has recently received magnetic resonance imaging (MRI)
procedure such as bloating, the inability to belch and approval for scanning in systems up to 1.5 Tesla.
vomit, and the occurrence of persistent dysphagia that
may occasionally require revisional surgery.11 These are
the main reasons why gastroenterologists tend to limit
SURGICAL TECHNIQUE
their referrals for fundoplication only to patients with Compared to the current surgical standard of a fundoplica-
long-lasting severe disease and large hiatal hernias. A tion, the Linx procedure requires less dissection and fewer
248
Magnetic Sphincter Augmentation for Gastroesophageal Reflux Disease CHAPTER 20 248.e1
ABSTRACT
The magnetic sphincter augmentation procedure has
been developed as a less disruptive and more standardized
laparoscopic surgical option for patients with early-stage
gastroesophageal reflux disease. This procedure addresses
the limitations of current medical, endoscopic, and surgical
therapies, has a favorable side-effect profile, and is highly
effective in reducing drug dependency and esophageal
acid exposure.
KEYWORDS
Gastroesophageal reflux disease; heartburn; regurgitation;
esophagitis; proton pump inhibitors; lower esophageal
sphincter; sphincter augmentation; Nissen; fundoplication;
Linx; Barrett esophagus; esophageal adenocarcinoma
Magnetic Sphincter Augmentation for Gastroesophageal Reflux Disease CHAPTER 20 249
Closed Open
Titanium
wires
Roman
arch
Titanium
case
Magnetic
core
A B
FIGURE 20.1 An engineering schematic of the magnetic sphincter augmentation device. The device consists of an expansible bracelet of
magnetic beads designed to be placed surgically around the exterior surface of the distal end of the lower esophageal sphincter (LES).
Each bead is composed of a titanium case containing a magnetic core of small disk-shaped magnets. The beads are connected by
titanium wires of specific lengths that limit the distance any two individual beads can move apart. When the device is closed (A), the
magnetic force is sufficient to prevent effacement and opening of the LES yet is weak enough to allow the device to open (B) with the
esophageal peristalsis. When the device is closed, the Roman arch construction prevents compression of the esophageal tissues.
(From DeMeester, TR. New approaches to gastroesophageal reflux disease [LINX]. In: Cameron JL, et al, eds. Current Surgical Therapy.
12th ed. Philadelphia: Elsevier; 2017:19-24.)
steps. The device is implanted with a standard laparoscopic intraoperatively. Operative time is generally less than 1
approach under general anesthesia. Ideally the dissection hour. Patients are discharged the same day of surgery
should be minimal with preservation of the phrenoesopha- or on the first postoperative day and are counseled to
geal ligament. The steps of the procedure are illustrated gradually return to a normal diet and to discontinue use
in Fig. 20.2. The first step is division of the peritoneum on of acid suppression medication.
the anterior surface of the gastroesophageal junction below
the insertion of the inferior leaf of the phrenoesophageal
ligament and above the junction of the hepatic branch to
SUMMARY OF CLINICAL EXPERIENCE
the anterior vagus nerve. The lateral surface of the left crus Since the first human implantation in 2007, all reported
is freed from the posterior fundic wall without dividing studies investigating the long-term clinical outcomes of
any short gastric vessels. The gastrohepatic ligament is the Linx device have confirmed a high rate of symptom
opened above and below the hepatic branch to facilitate relief, discontinuation of PPI therapy, objective reduction
the preparation of the retroesophageal window. Gentle of esophageal acid exposure, and improved quality of life.
dissection from the right side is made toward the left crus Two prospective, single-arm, multicenter, controlled clinical
just above the crural decussation to identify the posterior studies have been conducted to evaluate the Linx System.
vagus nerve. A tunnel is then created between the vagus The feasibility study included 44 patients implanted with
and the posterior esophageal wall, and a Penrose drain is the Linx at four study centers in the United States and
passed in a left-to-right direction. The circumference of in Europe between February 2007 and October 2008; the
the esophagus is measured to determine the proper size short-term, mid-term, 4-year, and final results of this study
of the Linx device to be implanted. The sizing tool is a have been previously published.16–19 In the feasibility study,
laparoscopic instrument with a soft, circular curved tip patients served as their own control to assess the effect
actuated by coaxial tubes through a handset. The handset of treatment on esophageal acid exposure, symptoms,
contains a numerical indicator that corresponds to the and use of PPI. The primary criteria for inclusion in the
size range of the Linx device. The sizing tool is placed feasibility trial were patients between ages 18 and 85 years,
around the esophagus in the dissected space between the typical reflux symptoms at least partially responsive to PPI
esophageal wall and the posterior vagus nerve bundle. therapy, abnormal esophageal acid exposure, and normal
Once the appropriate Linx device has been selected, it is contractile amplitude and wave form in the esophageal
introduced through the posterior tunnel. The opposing body. The primary criteria for exclusion from the trial were
ends are then brought to the anterior surface of the a history of dysphagia, previous upper abdominal surgery,
esophagus and connected together by engaging the two previous endoluminal antireflux procedures, sliding hiatal
clasps. The decision to proceed with a posterior crural hernia greater than 3 cm, esophagitis greater than grade A,
repair depends on the size of the hernia that is found and/or the presence of histologically documented Barrett
250 SECTION I Esophagus and Hernia
A B C
FIGURE 20.2 Surgical steps of the Linx procedure. (A) Establish the area for device placement by preservation of the phrenoesophageal
ligament. A tunnel behind the esophageal wall is made after identification of the posterior vagus nerve. (B) The esophagus is measured
using a special sizing tool. (C) The Linx device is locked in front of the esophagus after alignment and engagement of the two clasps.
esophagus. Patients with abnormal manometric findings seen in GERD-related quality of life, regurgitation, and
(distal esophageal contraction amplitude of less than esophageal acid exposure. Use of PPIs dropped to 13%
35 mm Hg on wet swallows or less than 70% propulsive at 3 years and patient satisfaction with reflux control
peristaltic sequences) were also excluded. Preoperative increased to 94% after implantation. Importantly, these
evaluation consisted of a symptom questionnaire and the positive results were stable and showed no degradation
Gastroesophageal Reflux Disease–Health Related Quality of over the study time period. Although 14% of patients
Life (GERD-HRQL) questionnaire, upper gastrointestinal reported some bloating after implantation, no patients
endoscopy, barium swallow, standard esophageal manom- rated this symptom as severe. Patients retained their
etry, and 24- to 48-hour esophageal pH monitoring. All ability to belch and vomit. Dysphagia proved to be quite
Linx devices were successfully implanted via a standard common, present to some extent in 68% of patients but
laparoscopic approach. The median operative time was decreasing to 4% by 3 years. Five percent of patients rated
40 (19 to 104) minutes. No intraoperative complications the dysphagia as severe and the device was removed in
occurred. Patients were instructed to resume a regular three of them with complete resolution.
diet after a chest film and radiologic assessment of the Two single-center studies have further validated the
esophageal transit were performed. All patients except efficacy of the Linx procedure. In Milan, Italy, 100 consecu-
one were discharged within 48 hours. Thirty-three patients tive patients underwent Linx implantation between 2007
(75%) were followed at 5 years. The mean total GERD- and 2012. The median implant duration was 3 years,
HRQL score off PPIs decreased from 25.7 at baseline to ranging from 378 days to 6 years. There was a significant
2.9 at year 5 (P < .001), and 94% (31/33) patients had a reduction of acid exposure time and improvement of
greater than 50% reduction in the total score compared GERD-HRQL score; freedom from daily dependence on
to baseline; 91% of patients reported being satisfied PPIs was achieved in 85% of the patients.21 Another study
with their current condition. Esophageal pH testing was from the United States, including 66 patients with an
completed in 20 patients at 5 years: 85% of patients average follow-up of 5.8 months, showed similar satisfactory
achieved normal esophageal acid exposure or had at results.22
least a 50% reduction from baseline. Normalization of Three recent case-control studies found comparable
esophageal pH was achieved in 70% of patients. Complete control of reflux symptoms after surgical fundoplication
cessation of PPIs or a reduction of 50% or more of the or Linx implant up to 1-year follow-up. However, the
daily dose at 5 years was achieved by 88% and 94% of fundoplication group showed a higher rate of patients
patients, respectively. who were unable to belch and vomit, along with more
Forty-three percent of patients complained of mild severe gas-bloat symptoms.23–26
dysphagia during the postoperative period; in all individu- Concerns regarding the safety of this operation, espe-
als the symptom resolved by 90 days without treatment. cially the fear of erosions, stem from past adverse experi-
Three patients were explanted: one because of persistent ence with the Angelchik device and, more recently, with
dysphagia, one because of the need to undergo MRI, and the gastric banding device. A recent analysis of the safety
the last one who elected to have a Nissen fundoplication profile of the first 1000 worldwide implants in 82 hospitals
for persisting GERD symptoms. All removals were safely showed 1.3% hospital readmission rate, 5.6% need of
performed by a laparoscopic procedure. postoperative endoscopic dilations, and 3.4% reoperation
Similar rigorous inclusion criteria and perioperative rate.27 All reoperations were performed on a nonemergent
subjective and objective evaluations were used in the basis for device removal. Among the 36 patients who
larger second study involving 100 patients at 14 centers had the device removed, the most common symptoms
in the United States.20 Significant improvements were were dysphagia and recurrence of reflux symptoms. In
Magnetic Sphincter Augmentation for Gastroesophageal Reflux Disease CHAPTER 20 251
addition, 7% of patients enrolled in the US multicenter intense symptoms despite proton pump inhibitor therapy: a post-hoc
single-arm trial had the device removed due to persistent analysis of the 2007 national health and wellness survey. Clin Drug
Investig. 2011;31:703-715.
dysphagia in four, vomiting in one, chest pain in one, and 2. Kahrilas PJ, Howden CW, Hughes N. Response of regurgitation to
reflux in one.28 A recent study focused on reoperations proton pump inhibitor therapy in clinical trials of gastroesophageal
for Linx removal and reported the long-term results of reflux disease. Am J Gastroenterol. 2011;106:1419-1425.
one-stage laparoscopic removal and fundoplication.29 3. Lord R, DeMeester S, Peters J, et al. Hiatal hernia, lower esophageal
sphincter incompetence, and effectiveness of Nissen fundoplication
Eleven (6.7%) out of 164 patients who underwent a in the spectrum of gastroesophageal reflux disease. J Gastrointest
laparoscopic Linx implant with a median follow-up of 48 Surg. 2009;13:602-610.
months were explanted at a later date. The main presenting 4. Malfertheiner P, Nocon M, Vieth M, et al. Evolution of gastro-
symptom requiring device removal was recurrence of oesophageal reflux disease over 5 years under routine medical
heartburn or regurgitation in 46%, dysphagia in 37%, and care—the ProGERD study. Aliment Pharmacol Ther. 2012;35:154-164.
5. Heidelbaugh JJ, Kim AH, Chang R, Walker PC. Overutilization of
chest pain in 18%. In two patients (1.2%) full-thickness proton-pump inhibitors: what the clinician needs to know. Ther Adv
erosion of the esophageal wall with partial endoluminal Gastroenterol. 2012;5(4):219-232.
penetration of the device occurred. Although the course 6. McColl K, Gillen D. Evidence that proton-pump inhibitor therapy
of this complication appeared to be benign and easy to induces the symptoms it is used to treat. Gastroenterology. 2009;137:20-22.
7. Poulsen AH, Christensen S, McLaughlin JK, et al. Proton pump
treat, it is possible that the long-term erosion rate of the inhibitors and risk of gastric cancer: a population-based cohort
Linx device will be higher than has been reported so far. study. Br J Cancer. 2009;100:1503-1507.
The median implant duration was 20 months, with 82% of 8. Galmiche JP, Hatlebakk J, Attwood S, et al. Laparoscopic antireflux
the patients being explanted between 12 and 24 months surgery vs esomeprazole treatment for chronic GERD: the LOTUS
after the implant. Device removal was most commonly randomized clinical trial. JAMA. 2011;305(19):1969-1977.
9. Niebisch S, Fleming FJ, Galey KM, et al. Perioperative risk of lapa-
combined with partial fundoplication. There were no roscopic fundoplication: safer than previously reported—analysis
conversions to laparotomy and the postoperative course of the American College of Surgeons National Surgical Quality
was uneventful in all patients. At the latest follow-up after Improvement Program 2005 to 2009. J Am Coll Surg. 2012;215:
reoperation (12 to 58 months), the GERD-HRQL score 61-69.
10. Richter JE, Dempsey DT. Laparoscopic antireflux surgery: key to
was within normal limits in all patients. success in the community setting. Am J Gastroenterol. 2008;103:289-291.
11. Khajanchee YS, O’Rourke R, Cassera MA, Gatta P, Hansen PD,
CONCLUSION Swanström LL. Laparoscopic reintervention for failed antireflux
surgery: subjective and objective outcomes in 176 consecutive
The Linx procedure was developed as a less disruptive and patients. Arch Surg. 2007;142:785-791.
12. Finks JF, Wei Y, Birkmeyer JD. The rise and fall of antireflux surgery
more standardized surgical option for patients who have in the United States. Surg Endosc. 2006;20:1698-1701.
early evidence of progressive GERD. The Linx addresses 13. Colavita PD, Belyansky I, Walters AL, et al. Nationwide inpatient
the limitations of existing medical, endoscopic, and surgical sample: have antireflux procedures undergone regionalization?
therapies, and provides a more physiologic solution to J Gastrointest Surg. 2013;17:6-13.
14. Khan F, Maradey-Romero C, Ganocy S, Frazier R, Fass R. Utilisation
GERD with a favorable side-effect profile. The results of surgical fundoplication for patients with gastro-oesophageal reflux
of clinical trials have shown that augmentation of the disease in the USA has declined rapidly between 2009 and 2013.
gastroesophageal junction barrier using the Linx procedure Aliment Pharmacol Ther. 2016;43:1124-1131.
is highly effective in decreasing esophageal acid exposure, 15. Ganz R, Gostout C, Grudem J, Swanson W, Berg T, DeMeester TR.
reducing typical GERD symptoms, reducing daily PPI Use of a magnetic sphincter for the treatment of GERD: a feasibility
study. Gastrointest Endosc. 2008;67:287-294.
dependence, and improving patients’ quality of life. Safety 16. Bonavina L, Saino G, Bona D, et al. Magnetic augmentation of the
issues such as device erosions or migrations have been lower esophageal sphincter: results of a feasibility clinical trial.
rare and not associated with mortality. The device can be J Gastrointest Surg. 2008;12:2133-2140.
easily removed if necessary, thereby preserving the option 17. Bonavina L, DeMeester TR, Fockens P, et al. Laparoscopic sphincter
augmentation device eliminates reflux symptoms and normalizes
of fundoplication or other therapies in the future. esophageal acid exposure. Ann Surg. 2010;252:857-862.
The Linx device is mainly intended for use in patients 18. Lipham JC, DeMeester TR, Ganz RA, et al. The Linx reflux manage-
with unsatisfactory response to medical therapy and in those ment system: confirmed safety and efficacy now at 4 years. Surg
with early, uncomplicated GERD who would not usually Endosc. 2012;26:2944-2949.
be considered ideal candidates for fundoplication.30,31 The 19. Saino G, Bonavina L, Lipham J, Dunn D, Ganz RA. Magnetic
sphincter augmentation for gastroesophageal reflux at 5 years: final
potential limitations of this innovative procedure are the results of a pilot study show long-term acid reduction and symptom
untested efficacy in the presence of a large hiatal hernia improvement. J Laparoendosc Adv Surg Tech. 2015;25:787-792.
and Barrett esophagus, the current contraindication to 20. Ganz RA, Peters JH, Horgan S, et al. Esophageal sphincter device
undergo scanning in MRI systems greater than 1.5 Tesla, for gastroesophageal reflux disease. N Engl J Med. 2013;368:719-727.
21. Bonavina L, Saino G, Bona D, Sironi A, Lazzari V. One hundred
and the potential long-term consequences of a permanent consecutive patients treated with magnetic sphincter augmentation
foreign body implant. Future randomized trials comparing for gastroesophageal reflux disease: 6 years of clinical experience
Linx and fundoplication are needed to definitely test the from a single center. J Am Coll Surg. 2013;217:577-585.
effectiveness of the procedure and, possibly, to establish 22. Smith CD, Devault KR, Buchanan M. Introduction of mechanical
the disease level at which lower sphincter augmentation sphincter augmentation for gastroesophageal reflux disease into
practice: early clinical outcomes and keys to successful adoption.
may prove superior to reconstruction. J Am Coll Surg. 2014;218:776-781.
23. Louie BE, Farivar AS, Schultz D, Brennan C, Vallières E, Aye RW.
Short-term outcomes using magnetic sphincter augmentation versus
REFERENCES Nissen fundoplication for medically resistant gastroesophageal reflux
disease. Ann Thorac Surg. 2014;98:498-504.
1. Toghanian S, Johnson DA, Stalhammar NO, Zerbib F. Burden of 24. Riegler M, Schoppman SF, Bonavina L, Ashton D, Horbach T,
gastro-oesophageal reflux disease in patients with persistent and Kemen M. Magnetic sphincter augmentation and fundoplication
252 SECTION I Esophagus and Hernia
for GERD in clinical practice: one-year results of a multicenter, 28. Ganz RA, Edmundowicz SA, Taiganides PA, et al. Long-term outcomes
prospective observational study. Surg Endosc. 2015;29:1123-1129. of patients receiving a magnetic sphincter augmentation device for
25. Reynolds J, Zehetner J, Wu P, Shah S, Bildzukewicz N, Lipham JC. gastroesophageal reflux. Clin Gastroenterol Hepatol. 2016;14:671-
Laparoscopic magnetic sphincter augmentation vs laparoscopic 677.
Nissen fundoplication; a matched-pair analysis of 100 patients. Ann 29. Asti E, Siboni S, Lazzari V, Bonitta G, Sironi A, Bonavina L. Removal
Surg. 2015;221:123-128. of the magnetic sphincter device. Surgical technique and results
26. Asti E, Bonitta G, Lovece A, Lazzari V, Bonavina L. Longitudinal of a single-center cohort study. Ann Surg. 2016;doi:10.1097/
comparison of quality of life in patients undergoing laparoscopic SLA.0000000000001785.
Toupet fundoplication versus magnetic sphincter augmentation: 30. Worrel SG, Greene CL, DeMeester TR. The state of surgical treatment
observational cohort study with propensity score analysis. Medicine of gastroesophageal reflux disease after five decades. J Am Coll Surg.
(Baltimore). 2016;95(30):e4366. 2014;219:819-830.
27. Lipham JC, Taiganides PA, Louie BE, Ganz RA, DeMeester TR. 31. Bonavina L, Attwood S. Laparoscopic alternatives to fundoplication
Safety analysis of first 1000 patients treated with magnetic sphincter for gastroesophageal reflux: the role of magnetic augmentation and
augmentation for gastroesophageal reflux disease. Dis Esophagus. electrical stimulation of the lower esophageal sphincter. Dis Esophagus.
2015;28:305-311. 2015;29(8):996-1001. doi:10.1111/dote.12425.
CHAPTER
Endoscopic Management of Gastroesophageal
Reflux Disease 21
Aaron Richman
| Praveen Sridhar
| Hiran C. Fernando
G
astroesophageal reflux disease (GERD) is defined progressive esophagitis. In addition, there are patients
by the reflux of gastric fluid into the esophagus who want to avoid surgical fundoplication and the associ-
causing troublesome symptoms and/or complica- ated side effects, such as dysphagia, bloating, or meteorism,
tions, such as mucosal inflammation and metaplasia. 1 and who will seek an EAR approach. Endoscopic therapy
Functional disturbances of the lower esophageal sphincter also provides a less invasive option for patients who are
(LES), along with anatomic abnormalities of the esopha- not good candidates for laparoscopic interventions due
gogastric junction (EGJ) such as hiatal hernia, allow the to previous laparotomies, or are not candidates for open
reflux of gastric fluid up into the esophagus. Treatment repair due to medical comorbidities. In addition, there
of GERD is focused on providing relief of symptoms and may be situations when patients with recurrent reflux
reducing reflux-related mucosal injury. The choice of after an operative fundoplication may find benefit from
treatment modality is based on symptom severity, the EAR therapy.
degree of esophageal injury, the presence and severity of Due to the limitations of endoscopic manipulation,
hiatal hernia, and the degree of esophageal dysmotility. patients with hiatal hernias larger than 2 cm and GERD
The mainstay of therapy is medical with gastric acid are not candidates for endoscopic therapy and are best
suppression through the use of proton pump inhibi- managed with a standard laparoscopic repair. Patients
tors (PPIs). PPIs are recommended as first-line therapy with moderate to severe mucosal inflammation (Los
based on their rapid impact on symptoms and efficient Angeles endoscopic classification grade C and D), as well
control of esophageal inflammation. PPIs have a well- as patients with Barrett esophagitis, have been excluded
known safety profile that allows for long-term use. Such from most studies evaluating EAR therapies. Although
additional measures as weight loss, head-of-bed elevation, these characteristics are not an absolute contraindication
and elimination of food triggers are also used but are to an EAR approach, the efficacy for these patients is not
less effective. Because none of the medical therapies yet known.
address the incompetent LES, their efficacy is somewhat
limited.2,3 TECHNIQUES
Surgical therapy is the preferred treatment in patients
with recalcitrant symptoms, especially those with esopha- The primary endoscopic therapies for GERD can be
gitis, Barrett esophagus, or extraesophageal pathology. classified into three major categories: (1) endoscopic
Surgical repair includes two fundamental principles: suturing or plication (full or partial thickness), (2) thermal
restoration of the gastroesophageal flap valve (Fig. 21.1) remodeling and neurolysis of the LES zone by radiofre-
with reinforcement of the LES and placement of the EGJ quency energy delivery, and (3) bulking or reinforcement
below the diaphragm. Laparoscopic fundoplication with of the LES zone by injection of inert material. These
correction of the hiatal hernia, if indicated, is the gold methods, along with a recently described technique using
standard, with well-established efficacy. Postoperative endoscopic mucosal resection (EMR), will be examined
dysphagia and gas-bloat syndrome continue to be trouble- next.
some side effects for surgical fundoplication despite
modifications to the original procedure.4 Over the past ENDOSCOPIC PLICATION
10 to 15 years a number of endoscopic approaches have Endoscopic suturing techniques were first described
been developed to treat GERD and will be described in in the 1980s when Swain and colleagues developed a
this review. miniature sewing machine that attached to the end of a
standard upper gastrointestinal endoscope.5 The first US
Food and Drug Administration (FDA)-approved method
INDICATIONS FOR ENDOSCOPIC THERAPY for treating GERD was an endoscopic suturing method
OF GASTROESOPHAGEAL REFLUX DISEASE introduced by Bard (Murray Hill, New Jersey) under
the name EndoCinch. This method was the commercial
Endoscopic antireflux (EAR) procedures aim to reinforce version of the miniature sewing machine as developed
the mechanical antireflux barrier, thereby reducing by Swain and Mills in London. This technique involved
symptoms and preventing pathologic progression of a partial-thickness plication on the gastric side of the
esophagitis. Indications for EAR intervention are similar EGJ.6 Clinical outcomes with the EndoCinch have gener-
to those for surgical intervention, namely relapsing or ally been marginal.7 As a result, the EndoCinch, like
recalcitrant symptoms despite adequate medical therapy, many endoscopic therapies, has been withdrawn from
PPI intolerance, desire to stop chronic drug therapy, and clinical use.
253
Endoscopic Management of Gastroesophageal Reflux Disease CHAPTER 21 253.e1
ABSTRACT
Gastroesophageal reflux disease is the most common
disorder of the esophagus. The primary medical therapy,
aside from changes in lifestyle, is the use of proton pump
inhibitors. Although patients who are treated initially
with PPIs often experience a significant and prompt
improvement in symptom control, recalcitrant symptoms
may cause patients to pursue surgical therapy. Persistent
nonacid reflux secondary to lower esophageal sphincter
incompetence, esophagitis, and anatomic deficiencies
of the gastroesophageal junction, such as hiatal hernias,
are indications for patients to pursue surgical therapy.
The gold standard for surgical therapy is a laparoscopic
fundoplication; however, over the past decade there has
been increasing interest in endoscopic and incisionless
antireflux procedures. Over this time period there has
been a rise in studies, with long-term follow-up revealing
that endoscopic antireflux procedures may be beneficial
for a subset of patients for whom surgical therapy is
indicated. In this chapter we examine the indications,
techniques, and current data for multiple endoscopic
antireflux therapies.
KEYWORDS
Reflux disease, endoscopic therapy, endoscopic
fundoplication
254 SECTION I Esophagus and Hernia
Anterior
Valve
Lip
Body
Posterior
A C
Scope
Oblique Lip of
section valve
Anterior
groove Crura
fibers
Posterior
B groove D
FIGURE 21.1 Endoscopic appraisal of the gastroesophageal flap valve. (Redrawn from Jobe BA, Kahrilas PJ, Vernon AH, et al.
Endoscopic appraisal of the gastroesophageal valve after antireflux surgery. Am J Gastroenterol. 2004;99:233.).
Currently there are two endoscopic plication systems There are limited long-term data examining the
that are FDA approved for patients with reflux. These are effectiveness of the MUSE system. Zacherl et al. examined
the Medigus Ultrasonic Endostapling system (MUSE; 69 patients prospectively from six centers who underwent
Medigus, Omer, Israel) and the EsophyX device (Endo- transoral fundoplication with the MUSE system using
gastric Solutions, Redmond, Washington). Each is described GERD Health-Related Quality of Life (HRQL) question-
later. naires, PPI dose reduction, and pH monitoring as their
objective measures of outcome, with the most notable
Medigus Ultrasonic Endostapling System exclusion criteria being patients who had hiatal hernias
(MUSE) Method larger than or equal to 3 cm. A reduction of at least 50%
The MUSE system uses an endostapler (titanium 4.8-mm in GERD-HRQL scores was achieved in 73% of patients
staples), ultrasonic range finder, insufflation pump, at 6-month follow-up. In addition, 65% of patients were
suction-irrigation, and an endoscopic console (Fig. 21.2). no longer using PPIs, and 85% of patients were able to
Compared with the EsophyX device, the MUSE is a single halve their dose. Esophageal pH studies showed a signifi-
operator system with the fundoplication completed in cant post-procedure reduction in the number of episodes
the supine position. The stapler is inserted transorally of reflux, as well as a reduction in the total time of pH
through a rigid overtube, advanced under direct vision less than 4.9 Kim et al. then updated results in this same
into the stomach, and retroflexed once the stapler is 5 cm cohort of patients at 4 years. Thirty-seven of the initial
past the EGJ to view the gastric fundus and the EGJ. An cohort of patients were followed for a total of 4 years. At
anchoring staple is placed between the fundus and the follow-up, 69% of patients remained off daily PPIs and
esophagus at the left lateral position, with subsequent mean HRQL scores remained significantly reduced. No
stapling at 60 to 180 degrees along the circumference of patients required repeat procedures or conversion to
the EGJ.8 laparoscopic fundoplication. 10 Similar findings were
Endoscopic Management of Gastroesophageal Reflux Disease CHAPTER 21 255
A B
C D
FIGURE 21.2 The Medigus SRS device. (A to C) This procedure is done by using an adapted endoscope that has two stapler
components; the anvil is located at the tip and the cartridge is located at the shaft. Their accurate alignment is guided by ultrasonic
sights. (D) The endoscope is inserted transorally, and the top of the fundus is caught by the anvil located at the tip of the device and
brought against the shaft, which contains the stapler cartridge that is fi red with standard 4.8-mm staples. The device is then partially
withdrawn, rotated 120 degrees, and the procedure is repeated. (From http://www.medigus.com/AboutGerd/GERD.aspx.)
described by Roy-Shapira et al. in a small cohort of patients transoral incisionless fundoplication (TIF) with maximal
followed out to 5 years.11 The most significant complications medical therapy in a randomized controlled trial, using
encountered were empyema and pneumothorax in one GERD-HRQL as well as postoperative pH studies as end
patient and bleeding requiring transfusion from an points. Patients treated with TIF experienced significantly
endoscopically unidentified source in one patient.9,10 improved quality-of-life (QoL) scores despite showing a
lack of improvement in esophageal acid exposure.15 Most
EsophyX Transoral Incisionless Fundoplication studies demonstrated effectiveness and safety of TIF at
Clinical experience has been greater with the EsophyX 6-, 12-, and 24-month follow-up because 61% to 93% of
system, and currently this is the most widely used endo- patients in these studies were able to discontinue these
scopic approach in the United States. Reports with follow- medications (follow-up endoscopic findings seen in Fig.
up as long as 6 years are now available for the EsophyX 21.3B). Furthermore, symptom response was shown in
technique.12 The device is composed of a handle, a chassis 86% of patients followed up to 6 years in one study.12,16–19
containing a channel for passage of the endoscope, a Notably, factors that negatively impacted outcomes in
tissue invaginator, a tissue mold, a screw, and stylets through patients undergoing TIF were hiatal hernias larger than
which fasteners can be deployed. Patients are positioned 2 cm and preoperative Hill esophageal valve grade of
in left lateral decubitus position, and completion of the III or IV.12
procedure requires two operators—one to operate the Our group previously reported on our early experience
endoscope and one to handle the instrument. After inser- with the EsophyX procedure in 46 patients.20 One patient
tion of the instrument, the stomach is insufflated and the was readmitted with an aspiration pneumonia, and three
instrument retroflexed. The tissue mold creates apposition patients had minor complications. Mean GERD-HRQL
between the stomach and the EsophyX device within the scores improved significantly (23 vs. 7), and 4 (8.6%)
esophageal lumen, and polypropylene fasteners are patients had no improvement in symptoms. More recently
deployed as the greater curve is rotated within the tissue we have looked at longer follow-up (mean 24 months) in
mold and around the EGJ to create a new valve with a 41 patients (unpublished data—currently submitted).
circumference of greater than 240 degrees (Fig. 21.3A).8,13,14 Among these 41 patients, 63% had either stopped or
Although there have been a small number of studies reduced their PPI dose (Table 21.1).
examining the long-term safety and effectiveness of fundo-
plication using the MUSE, there have been considerably RADIOFREQUENCY ABLATION
more data regarding the use of the EsophyX device. The Stretta system was one of the first endoscopic
There have been several prospective studies that have approaches introduced into clinical practice and is still
examined long-term outcomes after transoral fundoplica- currently available. It is believed that radiofrequency
tion with the EsophyX device up to 6 years postoperatively.12 ablation induces a thickened scar and that augments the
Most recently, Witteman et al. prospectively compared LES. This is based on animal data demonstrating that the
256 SECTION I Esophagus and Hernia
FIGURE 21.3 The EsophyX device. (A) The EsophyX device is inserted into the gastric cavity and positioned into the cardia. A retractor is
applied to the gastroesophageal flap valve. A flexible endoscope serves for viewing the procedure (1). The gastroesophageal flap valve is
retracted and brought to the arms of the device (2). The device is closed and the H-shape fasteners are deployed (3 and 4). The device
is retracted (5). Final antireflux valve configuration after finishing the procedure, showing enlargement of the gastroesophageal flap valve
(6). (B) Endoscopic view of the cardia; the top photos show pre-EsophyX procedure; the bottom photos show 12 months after
treatment. ([A] Redrawn from Cadière GB, Buset M, Muls V, et al. Antireflux transoral incisionless fundoplication using EsophyX:
12-month results of a prospective multicenter study. World J Surg. 2008;32:1676; [B] from Cadière GB, Rajan A, Germay O, et al.
Endoluminal fundoplication by a transoral device for the treatment of GERD: a feasibility study. Surg Endosc. 2008;22:333.)
TABLE 21.1 Current Data Regarding the Efficacy and Durability of Transoral Incisionless Fundoplication (TIF)
Percentage
Last Recorded Percentage Reduction Postoperative
Follow-Up Preoperative Postoperative of Postoperative PPI Normalization
Author Year Study Type N (months) GERD-HRQL GERD-HRQL Use at Last Follow-Up of pH
Cadière et al. 2008 Prospective 84 12 24 (median) 7 (median) 68% 37%
Cadière et al. 2009 Prospective 19 24 17 (median) 7 (median) — —
Testoni et al. 2010 Prospective 20 6 45 (mean) 16 (mean) 78% —
Velanovich et al. 2010 Retrospective 24 7 (mean) 25 (median) 5 (median) 79% —
Repici et al. 2010 Prospective 20 12 40 (median) 7 (median) 47% —
Demyttanaere et al. 2010 Retrospective 26 10 (mean) 22 (mean) 10 (mean) 55% —
Hoppo et al. 2010 Retrospective 19 11 (mean) — — 42% —
Barnes et al. 2011 Retrospective 124 7 (median) 28 (median) 2 (median) 93% —
Bell et al 2011 Retrospective 37 6 (median) 16 (median) 4 (median) 82% 61%
Ihde et al. 2011 Retrospective 48 6 (median) 29 (median) 3 (median) 76% —
Trad et al. 2012 Retrospective 34 14 (median) 26 (median) 4 (median) 82% —
Testoni et al. 2012 Retrospective 35 27 (mean) 22 (mean) 18 (mean) 69.2% —
Petersen et al. 2012 Retrospective 23 7 (median) — — 42% 26%
Bell et al. 2012 Prospective 100 6 26 (median) 4 (median) 89% 54%
Muls et al. 2013 Prospective 86 36 25 (median) 6 (median) 61% 82%*
Trad et al.† 2014 Prospective 63 6 19 (median) 2 (median) 95% 54%
Testoni et al. 2015 Prospective 50 53 (mean) 20 (mean) 17 (mean)‡ 86%‡ —
*11 of the initially enrolled 86 patients underwent pH testing at 3-year follow-up, 9 of whom had normalization.
†
Trad et al. recently (2016) published updated results at 30 months of follow-up for patients undergoing TIF in this initial TEMPO trial showing GERD
Health-Related Quality of Life (HRQL) scores improving to 5 (median) and 70% of patients completely off PPIs.
‡
GERD-HRQL scores completed at 36-month follow-up: at 6-year follow-up; 12 of 14 patients had stopped or halved proton pump inhibition (PPI)
use.12,13,16–19,21–31
fibrosis from chemical sclerosants reduces compliance squamocolumnar junction receives treatment.36 Treatment
and relaxation of the gastroesophageal junction and can be repeated if there is not substantial improvement
reduces reflux.32,33 in symptoms.
The Stretta system is an endoscopic instrument that The therapeutic effect appears to occur via two primary
uses radiofrequency ablation to treat GERD. Originally mechanisms. Anatomically, the heating causes collagen
introduced in 2000, the device and its protocols have contraction, yielding a mechanical alteration of the EGJ
been refined over the years to improve its efficacy and and a tightening of the LES. There is an associated increase
safety.34 The device is composed of a flexible balloon in smooth muscle fiber size with larger fibers and more
basket assembly with four electrode needle sheaths smooth muscle cells per bundle. This effect can be seen
introduced orally that delivers radiofrequency energy to immediately after the procedure, and the collagen deposi-
the muscular layer of the EGJ. The device and electrodes tion continues up to 12 months as wound healing ensues.
are positioned in the distal esophagus, the needles are Physiologically, there is ablation of aberrant esophageal
deployed into the circular muscle layer, and radiofrequency nerve pathways leading to an increase in both LES pres-
energy is delivered (Fig. 21.4). The delivery of heat is sure and gastric yield pressure, as well as a decrease in
controlled using thermocouples at the base and tip of transient LES relaxation.37–39 In addition, an improvement
each needle with a target temperature of 85°C. The mucosa in gastroparesis associated with GERD has been noted
is protected from thermal injury by maintaining its after ablation. It is not clear what the mechanism of this
temperature less than 50°C using concurrent irrigation.35 effect is, but likely it is related to vagal neurolysis.40
Serial applications are repeated every 0.5 cm such that Multiple studies, including a multicenter sham-
an area approximately 2 cm above and below the controlled randomized trial, have demonstrated the Stretta
Endoscopic Management of Gastroesophageal Reflux Disease CHAPTER 21 257
1 2 3
4 5 6
B
258 SECTION I Esophagus and Hernia
A B C
D E F
FIGURE 21.4 The Stretta procedure. (A) The gastroesophageal junction is inspected and measured. (B) The radiofrequency catheter is
introduced over the guidewire. (C) The balloon is inflated proximal to the squamocolumnar junction, and the electrode needles are used
for delivering radiofrequency energy for more than 90 seconds for a desired tissue temperature of 85°C. (D) The needles are then
withdrawn, the balloon is deflated, the scope is rotated 45 degrees, and the procedure is repeated. (E) These steps should be serially
repeated every 0.5 cm such that an area covering 2 cm above and 1.5 cm below the squamocolumnar junction receives treatment.
(F) Postprocedure changes such as enlargement of the muscular layer at the esophagogastric junction area can be expected. (From
Hogan WJ. Endoscopic therapy for gastroesophageal reflux disease. Curr Gastroenterol Rep. 2003;5:206.)
CONCLUSION
Over the past decade, EAR therapies have emerged as an
alternative therapy for patients suffering from GERD.
Initially there was great interest in these approaches;
however, several of the techniques introduced earlier into
practice have been abandoned either because of poor
efficacy or adverse outcomes. A few approaches, such as
the EsophyX, are still being used in several centers, and
the introduction of new techniques, such as ARMS, suggests
that physicians and patients have continued interest in
an incisionless approach to treat GERD. Although there
are only a few large studies, the data for these techniques
show promise. The currently available techniques appear
safe, with an acceptable and manageable side effect profile.
Refinements in technology and technique for radiofre-
quency ablation and endoscopic plication appear to have
improved their efficacy. With more experience, further
improvements with EAR instrumentation and technique
may also be expected. The long-term durability of EAR
remains a challenge, and well-controlled studies will need
B
to be performed to ensure that insurance providers are
willing to pay for EAR as a routine therapy. Although use
FIGURE 21.6 EnteryX procedure. (A) Flasks containing injectable of these techniques remains confined mostly to research
solution of 8% ethylene vinyl alcohol copolymer and dimethyl protocols and specialized centers, the large number of
sulfoxide and dedicated endoscopic needles for injection. patients with GERD who fail to see complete resolution
(B) Bulking effect at the esophagogastric junction after with current medical therapies but who are willing to
intramuscular injection. (From Hogan WJ. Endoscopic therapy for
pursue or are not candidates for surgical therapy will
gastroesophageal reflux disease. Curr Gastroenterol Rep. 2003;
drive continued adoption and research.
5:206.)
5. Paul SC, Mills TN. An endoscopic sewing machine. Gastrointest 27. Bell RCW, Freeman KD. Clinical and pH-metric outcomes of transoral
Endosc. 1986;32:36-38. esophagogastric fundoplication for the treatment of gastroesophageal
6. Swain CP, Brown GJ, Mills TN. An endoscopic stapling device: the reflux disease. Surg Endosc. 2011;25:1975-1984.
development of a new flexible endoscopically controlled device for 28. Ihde GM, Besancon K, Deljkich E. Short-term safety and symptomatic
placing multiple transmural staples in gastrointestinal tissue. Gas- outcomes of transoral incisionless fundoplication with or without
trointest Endosc. 1989;35:338-339. hiatal hernia repair in patients with chronic gastroesophageal reflux
7. Filipi CJ, Lehman GA, Rothstein RI, et al. Transoral, flexible disease. Am J Surg. 2011;202:740-747.
endoscopic suturing for treatment of GERD: a multicenter trial. 29. Petersen RP, Filippa L, Wassenaar EB, Martin AV, Tatum R, Oel-
Gastrointest Endosc. 2001;53:416-422. schlager BK. Comprehensive evaluation of endoscopic fundoplication
8. Testoni PA, Mazzoleni G, Testoni SGG. Transoral incisionless fun- using the EsophyXTM device. Surg Endosc. 2012;26:1021-1027.
doplication for gastro-esophageal reflux disease: techniques and 30. Muls V, Eckardt AJ, Marchese M, et al. Three-year results of a
outcomes. World J Gastrointest Pharmacol Ther. 2016;7:179-189. multicenter prospective study of transoral incisionless fundoplication.
9. Zacherl J, Roy-Shapira A, Bonavina L, et al. Endoscopic anterior Surg Innov. 2013;20:321-330.
fundoplication with the Medigus Ultrasonic Surgical Endostapler 31. Trad KS, Barnes WE, Simoni G, et al. Transoral incisionless fundoplica-
(MUSETM) for gastroesophageal reflux disease: 6-month results from tion effective in eliminating GERD symptoms in partial responders
a multi-center prospective trial. Surg Endosc. 2015;29:220-229. to proton pump inhibitor therapy at 6 months: the TEMPO random-
10. Kim HJ, Kwon CI, Kessler WR, et al. Long-term follow-up results of ized clinical trial. Surg Innov. 2015;22:26-40.
endoscopic treatment of gastroesophageal reflux disease with the 32. Carvalho PJ, Donahue PE, Miidla I, et al. Fibrosis of gastric cardia
MUSETM endoscopic stapling device. Surg Endosc. 2015;doi:10.1007/ after endoscopic sclerosis. Mechanism for control of experimental
s00464-015-4622-y. reflux? Am Surg. 1990;56:163-166.
11. Roy-Shapira A, Bapaye A, Date S, Pujari R, Dorwat S. Trans-oral 33. Donahue PE, Carvalho PJ, Davis PE, et al. Endoscopic sclerosis of
anterior fundoplication: 5-year follow-up of pilot study. Surg Endosc. the gastric cardia for prevention of experimental gastroesophageal
2015;29:3717-3721. reflux. Gastrointest Endosc. 1990;36:253-256.
12. Testoni PA, Testoni S, Mazzoleni G, Vailati C, Passaretti S. Long-term 34. Aziz AMA, El-Khayat HR, Sadek A, et al. A prospective randomized
efficacy of transoral incisionless fundoplication with EsophyX (Tif trial of sham, single-dose Stretta, and double-dose Stretta for the
2.0) and factors affecting outcomes in GERD patients followed for treatment of gastroesophageal reflux disease. Surg Endosc. 2010;24:818-825.
up to 6 years: a prospective single-center study. Surg Endosc. 35. Yeh RW, Triadafilopoulos G. Endoscopic antireflux therapy: the
2015;29:2770-2780. Stretta procedure. Thorac Surg Clin. 2005;15:395-403.
13. Testoni PA, Vailati C, Testoni S, Corsetti M. Transoral incisionless 36. Go MR, Dundon JM, Karlowicz DJ, Domingo CB, Muscarella P,
fundoplication (TIF 2.0) with EsophyX for gastroesophageal reflux Melvin WS, et al. Delivery of radiofrequency energy to the lower
disease: long-term results and findings affecting outcome. Surg esophageal sphincter improves symptoms of gastroesophageal reflux.
Endosc. 2012;26:1425-1435. Surgery. 2004;136:786-794.
14. Cadière GB, Rajan A, Rqibate M, et al. Endoluminal fundoplication 37. DiBaise JK, Brand RE, Quigley EMM. Endoluminal delivery of
(ELF)—evolution of EsophyX, a new surgical device for transoral radiofrequency energy to the gastroesophageal junction in uncom-
surgery. Minim Invasive Ther Allied Technol. 2006;15:348-355. plicated GERD: efficacy and potential mechanism of action. Am J
15. Witteman BPL, Conchillo JM, Rinsma NF, et al. Randomized con- Gastroenterol. 2002;97:833-842.
trolled trial of transoral incisionless fundoplication vs. proton pump 38. Tam WCE. Delivery of radiofrequency energy to the lower oesophageal
inhibitors for treatment of gastroesophageal reflux disease. Am J sphincter and gastric cardia inhibits transient lower oesophageal
Gastroenterol. 2015;110:531-542. sphincter relaxations and gastro-oesophageal reflux in patients with
16. Bell RCW, Mavrelis PG, Barnes WE, et al. A prospective multicenter reflux disease. Gut. 2003;52:479-485.
registry of patients with chronic gastroesophageal reflux disease 39. Utley DS, Kim M, Vierra MA, Triadafilopoulos G. Augmentation of
receiving transoral incisionless fundoplication. J Am Coll Surg. lower esophageal sphincter pressure and gastric yield pressure after
2012;215:794-809. radiofrequency energy delivery to the gastroesophageal junction: a
17. Trad KS, Turgeon DG, Deljkich E. Long-term outcomes after transoral porcine model. Gastrointest Endosc. 2000;52:81-86.
incisionless fundoplication in patients with GERD and LPR symptoms. 40. Noar MD, Noar E. Gastroparesis associated with gastroesophageal
Surg Endosc. 2012;26:650-660. reflux disease and corresponding reflux symptoms may be corrected
18. Cadière GB, Buset M, Muls V, et al. Antireflux transoral incisionless by radiofrequency ablation of the cardia and esophagogastric junction.
fundoplication using EsophyX: 12-month results of a prospective Surg Endosc. 2008;22:2440-2444.
multicenter study. World J Surg. 2008;32:1676-1688. 41. Corley DA, Katz P, Wo JM, et al. Improvement of gastroesophageal
19. Repici A, Fumagalli U, Malesci A, Barbera R, Gambaro C, Rosati R. reflux symptoms after radiofrequency energy: a randomized, sham-
Endoluminal fundoplication (ELF) for GERD using EsophyX: a controlled trial. Gastroenterology. 2003;125:668-676.
12-month follow-up in a single-center experience. J Gastrointest Surg. 42. Wolfsen HC, Richards WO. The Stretta procedure for the treatment
2010;14:1-6. of GERD: a registry of 558 patients. J Laparoendosc Adv Surg Tech A.
20. Narsule CK, Burch MA, Ebright MI, et al. Endoscopic fundoplication 2002;12:395-402.
for the treatment of gastroesophageal reflux disease: initial experi- 43. Perry KA, Banerjee A, Melvin WS. Radiofrequency energy delivery
ence. J Thorac Cardiovasc Surg. 2012;143:228-234. to the lower esophageal sphincter reduces esophageal acid exposure
21. Cadière GB, Van Sante N, Graves JE, Gawlicka AK, Rajan A. Two-year and improves GERD symptoms. Surg Laparosc Endosc Percutan Tech.
results of a feasibility study on antireflux transoral incisionless 2012;22:283-288.
fundoplication using EsophyX. Surg Endosc. 2009;23:957-964. 44. Dughera L, Rotondano G, De Cento M, Cassolino P, Cisar F. Durability
22. Testoni PA, Corsetti M, Di Pietro S, et al. Effect of transoral incisionless of Stretta radiofrequency treatment for GERD: results of an 8-year
fundoplication on symptoms, PPI use, and pH-impedance refluxes follow-up. Gastroenterol Res Pract. 2014;2014:531907.
of GERD patients. World J Surg. 2010;34:750-757. 45. Noar M, Squires P, Noar E, Lee M. Long-term maintenance effect
23. Velanovich V. Endoscopic, endoluminal fundoplication for gastro- of radiofrequency energy delivery for refractory GERD: a decade
esophageal reflux disease: initial experience and lessons learned. later. Surg Endosc. 2014;28:2323-2333.
Surgery. 2010;148:646-653. 46. Dughera L, Navino M, Cassolino P, et al. Long-term results of
24. Demyttenaere SV, Bergman S, Pham T, et al. Transoral incisionless radiofrequency energy delivery for the treatment of GERD: results
fundoplication for gastroesophageal reflux disease in an unselected of a prospective 48-month study. Diagn Ther Endosc. 2011;2011:507157.
patient population. Surg Endosc. 2010;24:854-858. 47. Reymunde A, Santiago N. Long-term results of radiofrequency
25. Hoppo T, Immanuel A, Schuchert M, et al. Transoral incisionless energy delivery for the treatment of GERD: sustained improvements
fundoplication 2.0 procedure using EsophyXTM for gastroesophageal in symptoms, quality of life, and drug use at 4-year follow-up. Gas-
reflux disease. J Gastrointest Surg. 2010;14:1895-1901. trointest Endosc. 2007;65:361-366.
26. Barnes WE, Hoddinott KM, Mundy S, Williams M. Transoral incision- 48. Lipka S, Kumar A, Richter JE. No evidence for efficacy of radiofre-
less fundoplication offers high patient satisfaction and relief of quency ablation for treatment of gastroesophageal reflux disease:
therapy-resistant typical and atypical symptoms of GERD in community a systematic review and meta-analysis. Clin Gastroenterol Hepatol.
practice. Surg Innov. 2011;18:119-129. 2015;13:1058-1067.e1.
Endoscopic Management of Gastroesophageal Reflux Disease CHAPTER 21 261
49. Liang WT, Yan C, Wang ZG, et al. Early and midterm outcome after procedure for endoscopic treatment of gastroesophageal reflux
laparoscopic fundoplication and a minimally invasive endoscopic disease. Surg Endosc. 2010;24:1387-1397.
procedure in patients with gastroesophageal reflux disease: a prospec- 53. Wong RF, Davis TV, Peterson KA. Complications involving the
tive observational study. J Laparoendosc Adv Surg Tech A. 2015;25: mediastinum after injection of Enteryx for GERD. Gastrointest Endosc.
657-661. 2003;61:753-756.
50. Liang WT, Wang ZG, Wang F, et al. Long-term outcomes of patients 54. Ota K, Takeuchi T, Harada S, et al. A novel endoscopic submucosal
with refractory gastroesophageal reflux disease following a minimally dissection technique for proton pump inhibitor-refractory gastro-
invasive endoscopic procedure: a prospective observational study. esophageal reflux disease. Scand J Gastroenterol. 2014;49(12):1409-1413.
BMC Gastroenterol. 2014;14:178. 55. Inoue H, Ito H, Ikeda H, et al. Anti-reflux mucosectomy for gastro-
51. Yan C, Liang WT, Wang ZG, et al. Comparison of Stretta procedure esophageal reflux disease in the absence of hiatus hernia: a pilot
and Toupet fundoplication for gastroesophageal reflux disease-related study. Ann Gastroenterol. 2014;27(4):346-351.
extra-esophageal symptoms. World J Gastroenterol. 2015;21:12882-12887. 56. Bechara R, Inoue H. Recent advancement of therapeutic endoscopy
52. Fockens P, Cohen L, Edmundowicz SA, et al. Prospective randomized in the esophageal benign diseases. World J Gastrointest Endosc.
controlled trial of an injectable esophageal prosthesis versus a sham 2015;7:481-495.
CHAPTER
G
astroparesis is described as delayed gastric emptying Several other techniques are being used to investigate
(DGE) without evidence of mechanical outlet gastric emptying. Magnetic resonance imaging (MRI),
obstruction. Common symptoms of gastroparesis which is noninvasive, can evaluate motility in addition to
include chronic nausea, emesis, abdominal pain, early emptying. Ultrasonography, another noninvasive technique,
satiety, and bloating. Abdominal pain can be significant and has also been used to evaluate patients for DGE but is
is associated with narcotic dependence in some patients. dependent on the presence of a skilled technician.
The true prevalence of this potentially debilitating disease Antroduodenal manometry can be used to evaluate
is unknown, but it has been estimated to affect up to 4% of gastric, pyloric, and duodenal motor activity and assess
the population.1 Some affected with this disorder have mild motor dysfunction. This includes antral hypomotility,
symptoms that can be medically controlled. Others have a migrating motor complex (MMC) activity, and focal
chronic debilitating issue that is refractory to conservative dysfunction. This procedure is performed with a per-
management with antiemetic and prokinetic medications. fusion manometry system or a solid-state catheter to
Causes of DGE are varied and include diabetes, gastric measure intraluminal pressure of gastric and duodenal
surgery, central nervous system disorders, and metabolic wall contractions. Antroduodenal manometry findings
and systemic disorders. Approximately one third of cases can be abnormal in multiple disorders including DGE
are idiopathic.1-3 and gastroesophageal reflux disease (GERD).2,4
ABSTRACT
Gastroparesis is delayed gastric emptying (DGE) without
evidence of mechanical outlet obstruction. Symptoms may
include chronic nausea, emesis, abdominal pain, early
satiety, and bloating. Causes are varied and symptoms
can range from mild to debilitating. The latter may be
refractory to conservative management with antiemetic
and prokinetic medications. Gastroesophageal reflux
disease is a failure of the antireflux barrier and is due
to a defective lower esophageal sphincter (LES) with
reduced pressure, transient LES relaxations, and impaired
esophageal peristalsis. It is possible that DGE affects the
occurrence of gastroesophageal reflux. This chapter reviews
the relationship between the rate of gastric emptying and
gastroesophageal reflux and possible surgical interventions
for this complex combination.
KEYWORDS
Delayed gastric emptying, gastroparesis, gastroesopha-
geal reflux disease, prokinetic medication, endoscopic
botulinum toxin injection, radiofrequency ablation,
Stretta, gastric electrical stimulation, laparoscopic subtotal
gastrectomy, laparoscopic total gastrectomy, pyloroplasty,
pyloromyotomy, fundoplication, Nissen, sleeve gastrectomy,
Roux-en-Y gastric bypass, magnetic sphincter augmentation
Options to Address Delayed Gastric Emptying in Gastroesophageal Reflux Disease CHAPTER 22 263
It is possible that DGE affects the occurrence of gastro- but negatively correlated with reflux episodes.14 These
esophageal reflux and has an influence on the refluxate findings of increased amplitude of antral contractions in
composition. In theory it makes sense that DGE may the setting of DGE contrast with results that show a similar
lead to larger quantities of food in the stomach available pattern of distal antral contractions in these two groups.15
to be refluxed. This gastric distention may also cause Carmagnola et al. took a different approach in
increased episodes of reflux by generating TLESRs.6,8-10 attempting to determine whether DGE plays a role in
Yet the relationship between the rate of gastric emptying the number of reflux episodes in GERD. They evaluated
and GERD has been studied in multiple articles without gastric emptying with ultrasonography and esophageal
a consensus on the exact nature or significance of their pH monitoring after patient use of cisapride, a prokinetic
association. medication, and compared these results with placebo.
Maddern et al. assessed solid and liquid gastric emptying Forty percent of their GERD patients were noted to have
simultaneously in 72 patients with symptomatic GERD. Of DGE. Cisapride was seen to increase gastric emptying and
these, 44% were found to have delayed solid emptying and decrease the number of reflux episodes and esophageal
37% delayed liquid emptying. There was no significant acid exposure. However, no correlation was seen between
correlation noted between gastric emptying and the resting changes in gastric emptying and the medication-induced
LES pressure. Symptoms of regurgitation and epigastric changes in reflux variables.16
fullness also did not correlate with gastric emptying.9 As these previous reports found discordant results,
Cunningham et al. also demonstrated DGE of solids in Gourcerol et al. used combined esophageal pH-impedance
46% of patients with GERD.10 The authors noted that this measurements to evaluate whether the occurrence of
may indicate that the role of delayed emptying in GERD gastroesophageal reflux correlated with gastric emptying
was related to gastric distention leading to alteration of rate. Esophageal impedance monitoring was used to help
gastric wall tension in the region of the LES, thereby identify the type of reflux (less acidic or nonacidic).
increasing reflux events.9,10 Gastric emptying was assessed using the 13C-octanoic
Whereas the aforementioned studies noted DGE in a acid breath test. This study found that delay in gastric
significant percentage of patients with GERD, Schwizer emptying increased daily liquid and mixed reflux events
et al. found DGE with similar frequency in GERD patients without affecting esophageal acid exposure. This may be
versus controls. In this study, DGE was associated with attributed to acid buffering due to gastric food retention
a decreased incidence of esophagitis, suggesting that and the production of nonacidic or weakly acidic refluxate
retained food had a buffering effect on the acidity of the or reflux of larger volume. Compared with normal gastric
stomach. This study concluded that DGE was not a major emptying, DGE patients had symptomatic reflux with
contributing factor to GERD.11 higher proximal extension and a longer bolus clearance
The previous studies demonstrate that it is unclear time.6 These patients with DGE and higher proximal
whether there is a significant link between abnormal gastric extension may be more symptomatic despite a similar
emptying and increased gastroesophageal reflux. Yet it has number of reflux episodes.8
been speculated that gastric distention can contribute to All of these studies indicate that gastric function may
reflux episodes by triggering TLESRs. This theory suggests affect GERD pathogenesis, but an unequivocal connection
that although total gastric emptying may be similar in between DGE and the degree of reflux symptoms has
patients with GERD and controls, the difference may lie in not been confirmed. In fact, as evidenced by a few of the
the time taken to empty different sections of the stomach. noted articles, some studies have failed to demonstrate
The gastric fundus and antrum have different functions an association between GERD and DGE. This may be
in gastric emptying, and dysfunctional motor activity at because many of these studies differ in inclusion criteria,
either of these sites could have a role in the production have small sample sizes, and vary in the ingested meal
of reflux episodes. Herculano et al. compared gastric that is studied. Even when studies show a significant
emptying and food retention in the proximal stomach in association between GERD and DGE, it remains to be
these two groups using scintigraphy. This study found that considered whether this has an effect on clinical symptoms
total gastric emptying was similar in GERD patients and or esophageal acid exposure. As noted earlier, it has been
controls but patients had decreased proximal retention described that medication-induced acceleration of gastric
of a liquid meal with an increased number of reflux emptying does not correlate with decreased acid exposure
episodes. This indicated a negative correlation between at the esophagus or other reflux variables.16 At this time
proximal gastric retention and reflux episodes.12 These it appears that the studies on gastric motility and empty-
results are in contrast to the results of Stacher et al., ing in regard to GERD have variable and contradictory
who, using scintigraphy, evaluated total and proximal results.17 DGE may play a larger role in less acidic or
stomach emptying of a semisolid meal in patients with nonacidic reflux events because of buffering effects and
symptoms of DGE and GERD. Their data showed that may not be a major determinant of the number of reflux
delayed proximal gastric emptying was associated with episodes.6,8
increased reflux episodes.13
Attention has also been given to the evaluation of
the distal stomach and to assessing its role in GERD.
TREATMENT OPTIONS
Barbieri et al. used dynamic antral scintigraphy to monitor There is no universally accepted treatment option for
postprandial antral contractions and their relationship patients with DGE in the setting of GERD. Multiple medical
to GERD and DGE. They found that the amplitude of therapies and surgical procedures are available for both of
the contractions was linked to gastric emptying time these disorders, but the number of options for treatment
264 SECTION I Esophagus and Hernia
is indicative of the complex nature of this combination marker to predict which patients are most likely to respond
of disorders. Treatment may oftentimes be geared toward to surgical intervention.
the predominating symptoms.
RADIOFREQUENCY ABLATION
MEDICAL MANAGEMENT Since 2000, GERD has been managed using a noninvasive
Prokinetic medications augment gastrointestinal tract radiofrequency treatment. This option has been used in
contractility and stimulate forward flow and gastric empty- patients whose symptoms are uncontrolled with medication
ing. Dopamine receptor antagonists are often used as but who do not wish to undergo surgical intervention.
treatment for gastroparesis because they have antiemetic The Stretta procedure involves endoluminal delivery of
and prokinetic effects. Metoclopramide has been shown low-level radiofrequency waves to the gastroesophageal
to be effective for short-term treatment of gastroparesis, junction. This therapy leads to inflammation, with collagen
but long-term maintenance of symptoms has not been well accumulation and thickening of the LES muscle. Noar
described. Prolonged use can also cause tardive dyskinesia, and Noar studied 31 patients with GERD and DGE who
and these symptoms sometimes do not improve even underwent the Stretta procedure with ablation of the
after medication use is discontinued. Domperidone, a gastroesophageal junction and cardia.22 Of these 31 patients,
peripheral dopamine receptor antagonist, does not cross 73% had normalization of gastric emptying at 6 months
the blood-brain barrier and therefore has decreased risk postprocedure. Both responders and nonresponders had
of central nervous system side effects. The macrolide significant improvements in GERD–Health Related Quality
antibiotic and motilin receptor agonist erythromycin has of Life (GERD-HRQL) scores, but symptoms of dyspepsia
also been used to treat gastroparesis, but it may cause the were significantly improved only in the responder group.
development of tachyphylaxis.1,2,18 The mechanism of action for this effect is unclear. It may
Prokinetics have been shown to stimulate gastric empty- partly be due to a stronger esophagogastric barrier in the
ing and decrease gastroparesis symptoms, but do they setting of radiofrequency-induced modification of the
have an effect on GERD? Manzotti et al. performed a gastric pacemaker, leading to decreased numbers of TLESRs.
review of randomized control trials (RCTs) to evaluate
the value of prokinetic medication for the treatment of GASTRIC ELECTRICAL STIMULATION
gastroesophageal reflux esophagitis. Prokinetic drugs may As noted earlier in this chapter, the intrinsic gastric
ameliorate symptoms in patients with GERD by improving pacemaker produces antral stimulation at three cycles
gastric emptying and increasing LES pressure. Eighteen per minute. Gastric electrical stimulation (GES) delivers
studies were included in this review, and an increase was high-frequency, low-energy electrical stimulation to the
noted in the probability of symptom and endoscopic gastric smooth muscle cells using surgically implanted
improvement.19 Of note, many of the RCTs used cisapride, a electrodes. GES is used to treat intractable nausea and
5-hydroxytryptamine type 4 (5-HT4) receptor agonist, as the vomiting due to diabetic or idiopathic gastroparesis. The
prokinetic medication. Cisapride has since been removed mechanism for this improvement of symptoms is unclear,
from the market in the United States due to adverse effects and emptying studies after stimulator placement may show
associated with cardiac dysrhythmias. Metoclopramide, a minimal change in gastric emptying despite symptom
dopamine D2 receptor antagonist, has both prokinetic and improvement.1,2 This device can be placed laparoscopically
antiemetic effects and is often used for the treatment of or with an open procedure (Table 22.1). Risks of gastric
gastroparesis.2 Metoclopramide has also been shown to be stimulator placement include erosion of the leads into
significantly more effective than placebo in the treatment the gastric lumen with resultant infection, lead dislodge-
of GERD by significantly improving gastric emptying and ment, intestinal obstruction due to the intraabdominal
LES pressure.20 portions of the wires, and infection at the stimulator site.5
Stimulator placement has been shown to improve quality
ENDOSCOPIC BOTULINUM TOXIN INJECTION of life, reduce hospital visits, improve glycemic control
An imbalance of acetylcholine and nitric oxide (NO) in diabetic patients, and reduce the need for parenteral
neurotransmitters may contribute to pylorospasm, with and enteral nutritional supplementation. Stimulation is
resultant nausea and vomiting in gastroparesis.5 Botulinum
toxin inhibits cholinergic neuromuscular transmission and
may ameliorate the symptoms of DGE. If symptomatic TABLE 22.1 Gastric Electrical Stimulator Implantation
improvement is noted, the treatment may be repeated
at several-month intervals. 1. A small upper midline incision is made in the epigastric area.
Mirbagheri et al. studied the effect of endoscopic pyloric • If laparoscopy is used, a 10-mm camera trocar is placed
botulinum toxin injection in patients with GERD associated above the umbilicus and two working trocars are placed
with DGE.21 Their methodology involved dilution of 200 below the right and left subcostal margins.
units of botulinum toxin A into 4 mL of saline. A 25-gauge 2. The electrodes are placed 10 cm proximal to the pylorus
needle was then used to inject 50 units into each of the approximately 1 cm apart along the greater curvature.
four quadrants of the pyloric sphincter. Of the 11 patients 3. A subcutaneous pocket is created for the GES device using
studied, 8 showed response with improvement of reflux the same midline incision but away from the fascial incision
or gastroparesis symptoms. The period of symptom relief on top of the anterior fascia.
was brief, with a mean duration of a little over 10 weeks. 4. The electrodes are connected to the GES device and the
The authors concluded that although this method of fascia and subcutaneous pocket are closed.
treatment is not a permanent solution, it may act as a GES, Gastric electrical stimulator.
Options to Address Delayed Gastric Emptying in Gastroesophageal Reflux Disease CHAPTER 22 265
TABLE 22.3 Laparoscopic Heineke–Mikulicz Pyloroplasty TABLE 22.4 Laparoscopic Nissen Fundoplication
1. A camera port is placed above and to the left of the 1. The patient is placed in low lithotomy with the surgeon
umbilicus. Two 5-mm ports are placed at the bilateral between the patient’s legs.
subcostal margins in the midclavicular line. A 5-mm port is 2. A 12-mm camera port is placed above and to the left of the
placed above and to the right of the umbilicus. A liver umbilicus. A 12-mm working port is placed at the left
retractor is placed in the subxiphoid position. subcostal margin and a 5-mm port at the right subcostal
2. The pylorus is mobilized from superior and inferior margin in the midclavicular line. A Nathanson liver retractor
attachments to decrease suture line tension. is placed in the subxiphoid position. A 5-mm port in the left
3. A Kocher maneuver is performed, if needed, to improve anterior axillary line is placed as an assistant port.
visualization and decrease tension with incision of the 3. Circumferential dissection of the esophagus is performed
peritoneum at the lateral aspect. with identification of the vagus nerves. A Penrose drain
4. Two stay sutures are placed at the superior and inferior placed around the esophagus can aid in the production of
aspects of the pylorus. an adequately sized retroesophageal window and crural
5. A 5-cm full-thickness antroduodenal transverse incision is closure.
performed. 4. The gastric fundus is mobilized with division of the short
6. Traction is applied at the stay sutures and the incision is gastric vessels.
closed in a single layer transversely using absorbable suture 5. An esophageal bougie (50–60 French) is placed prior to
in either a running or interrupted fashion. passage of the fundus posterior to the esophagus.
6. The fundus is sutured to itself using three interrupted
Data From Hibbard M, Dunst C, Swanstrom L. Laparoscopic and endoscopic
pyloroplasty for gastroparesis results in sustained symptom improvement. sutures to produce an approximately 2-cm loose, floppy
J Gastrointest Surg. 2011;15(9):1513. wrap.
and liquid reflux occurrences in both DGE and normal and gastroesophageal reflux is now being considered. It is
gastric emptying patients, but significant improvement well known that obesity is related to GERD, with overweight
in weakly acidic reflux and mixed gas and liquid reflux patients having up to 3 times the incidence of reflux
was noted in only patients with normal emptying. In symptoms.39,40 Fundoplication for GERD in the obese does
normal emptying patients, esophageal manometry showed not have any effect on the other comorbidities suffered
a significant increase in gastroesophageal pressure at 1 and by this population and has been found to have variable
5 years. In DGE patients, the gastroesophageal pressure results.39,40 Roux-en-Y gastric bypass (RYGB) has been shown
significantly increased at 1 year but deteriorated at 5 to improve reflux symptoms. A study of 58 patients with
years, with a return to baseline in 66.7%. These findings preoperative GERD and morbid obesity who underwent
translated into clinical findings. Postoperative symptom laparoscopic RYGB demonstrated that symptoms improved
scores were significantly improved in the normal emptying or resolved in the majority of patients.41
group, and less than 10% of these patients required proton RYGB can also improve symptoms of gastroparesis.
pump inhibitor (PPI) therapy at 60 months. This was in Seven patients with a mean body mass index (BMI) of
contrast with the DGE patients, who did not have any 39.5 kg/m2 and gastroparesis underwent laparoscopic
meaningful improvement in symptoms, and over 90% RYGB. Mean follow-up at 315 days showed that BMI was
were on medical therapy with PPIs and prokinetics at 5 reduced by a mean of 9.1 units. Patients were also found
years without adequate control of their symptoms. These to have significant improvement in their symptom scores
results indicate that symptoms and reflux are not well with a resultant decrease in prokinetic and antiemetic
controlled in DGE patients who undergo laparoscopic medications.42 These studies indicate that RYGB is a feasible
total fundoplication alone. option for patients with DGE and GERD.
Another bariatric procedure that has widespread use
PARTIAL AND TOTAL GASTRECTOMY and may also have some effect on gastric emptying is the
Subtotal or total gastrectomy is performed for refractory sleeve gastrectomy (SG). An Australian study evaluated
gastroparesis, but its true effectiveness is unclear.3 It has four patients with laparoscopic SG with simultaneous
been shown that laparoscopic subtotal gastrectomy can fundoplication for DGE and GERD.43 Postoperatively,
be considered as a primary surgical treatment for gastro- all of these patients were able to stop PPI usage. Gastric
paresis since it is associated with significant symptomatic emptying improved 67% and GERD-HRQL scores bettered
improvement with acceptable morbidity and mortality.24 in all patients, suggesting this technique should be studied
Gastrectomy has also been evaluated in the treatment further for possible broader applicability.
of recurrent reflux disease after fundoplication. Twelve
patients underwent gastrectomy (near-total, proximal,
or total) and 25 had refundoplication for unsuccessful
CONCLUSION
fundoplication with mean follow-up over 3 years.37 Both The pathophysiology of DGE may include decreased motil-
groups showed improvement in symptom severity scores, ity, compromised gastric tone, impaired antroduodenal
but gastrectomy had a higher resolution of the primary coordination, and pyloric hypercontractility.1-3 There is
symptom (89% vs. 50%). There was a higher morbidity no cure, and symptom control is the mainstay of therapy.
and mortality rate in the gastrectomy patients, but four This disorder becomes even more difficult to manage once
of the patients in the refundoplication group required it is associated with GERD. Medical management with
an additional surgical procedure. This suggests that prokinetics has shown variable results, and the options have
gastrectomy is an option for recurrent reflux in certain become more limited with removal of certain medications
patients, especially those who have had numerous failures with significant side effects from the market. Endoscopic
in the past. pyloric botulinum toxin A injection can give short-term
Clark et al. reviewed nine patients who underwent respite from symptoms of both GERD and DGE but is not
gastrectomy (total, near-total, proximal) for postfun- a viable permanent solution.21 Radiofrequency ablation of
doplication debilitating gastric dysfunction. 38 Patients the gastroesophageal junction and gastric cardia for reflux
may have symptoms due to undiagnosed DGE prior to has been shown to improve gastric emptying and symptoms.
the fundoplication or because of vagal nerve dysfunc- The patient population in this study was limited; hence
tion postoperatively. Three patients had postoperative further study is warranted.22 GES has been shown to have a
complications. Seven of the patients continued to have significant effect on gastroparesis symptoms, and there is a
symptoms despite gastrectomy, and three continued to suggestion that it may also contribute to the improvement
require enteral nutrition via feeding tube. These findings of reflux symptoms. GES also has been used in conjunction
show that although gastrectomy has been shown to improve with gastric fundoplication in a case report but further
gastroparesis and recurrent reflux separately, its effect on evaluation of this technique is required.27 Fundoplication
gastric dysfunction after fundoplication is associated with alone has shown conflicting results in terms of efficacy. As
suboptimal results. noted, one prospective study to 5 years showed that DGE
patients had poor symptom control after fundoplication.35
The addition of gastric drainage procedures such as
BARIATRIC SURGERY FOR GASTROESOPHAGEAL pyloroplasty has been shown to improve symptoms in
REFLUX DISEASE AND DELAYED patients with GERD and DGE.30 Gastric resection, from SG
GASTRIC EMPTYING to total gastrectomy, is the most invasive procedure for the
Bariatric surgery has become more prevalent over time. treatment of DGE in the setting of GERD, but it does not
The influence of bariatric procedures on both gastroparesis always result in symptom improvement and is often used
268 SECTION I Esophagus and Hernia
as the last resort. However, with the growth of minimally 20. McCallum R, Fink S, Winnan G, Avella J, Callachan C. Metoclopramide
invasive techniques, the morbidity of these procedures in gastroesophageal reflux disease: rationale for its use and results
of a double-blind trial. Am J Gastroenterol. 1984;79(3):165.
has become more reasonable.24 Also, surgeons now have 21. Mirbagheri S, Sadeghi A, Amouie M, et al. Pyloric injection of
newer therapies in their armamentarium, many of which botulinum toxin for the treatment of refractory GERD accompanied
can be used in conjunction, to help battle this problematic with gastroparesis: a preliminary report. Dig Dis Sci. 2008;53(10):2621.
issue. Laparoscopic magnetic sphincter augmentation, for 22. Noar MD, Noar E. Gastroparesis associated with gastroesophageal
reflux disease and corresponding reflux symptoms may be corrected
example, has been shown to have similar reflux symptom by radiofrequency ablation of the cardia and esophagogastric junction.
control as Nissen fundoplication but with a lower frequency Surg Endosc. 2008;22(11):2440.
of severe gas-bloat; it has been recommended for patients 23. Lin Z, Sarosiek I, Forster J, McCallum RW. Symptom responses, long-
with mild to moderate GERD.44 It is conceivable that this term outcomes and adverse events beyond 3 years of high-frequency
technology could be used in combination with gastric gastric electrical stimulation for gastroparesis. Neurogastroenterol Motil.
2006;18(1):18.
drainage procedures to also improve DGE symptoms. 24. Zehetner J, Ravari F, Ayazi S, et al. Minimally invasive surgical approach
for the treatment of gastroparesis. Surg Endosc. 2013;27(1):61.
25. Xing J, Felsher J, Soffer E. Gastric electrical stimulation significantly
REFERENCES increases canine lower esophageal sphincter pressure. Dig Dis Sci. 2005;
50(8):1481.
1. Abell T, Bernstein R, Cutts T, et al. Treatment of gastroparesis: a mul- 26. Cigaina V. Long-term follow-up of gastric stimulation for obesity:
tidisciplinary clinical review. Neurogastroenterol Motil. 2006;18(4):263. the Mestre 8-year experience. Obes Surg. 2004;14(suppl 1):S14.
2. Haans J, Masclee A. Review article: the diagnosis and management 27. Filichia LA, Baz M, Cendan JC. Simultaneous fundoplication and
of gastroparesis. Aliment Pharmacol Ther. 2007;26(2):37. gastric stimulation in a lung transplant recipient with gastroparesis
3. Jones M, Maganti K. A systematic review of surgical therapy for and reflux. JSLS. 2008;12(3):303.
gastroparesis. Am J Gastroenterol. 2003;98(10):2122. 28. Okuyama H, Urao M, Starr G, et al. A comparison of the efficacy of
4. Fuchs K-H. Tests of gastric function and their use in the evaluation pyloromyotomy and pyloroplasty in patients with gastroesophageal
of esophageal disease. In: Yeo CJ, Matthews JB, McFadden DW, et al., reflux and delayed gastric emptying. J Pediatr Surg. 1997;32(2):316.
eds. Shackelford’s Surgery of the Alimentary Tract. 7th ed. Philadelphia: 29. Masqusi S, Velanovich V. Pyloroplasty with fundoplication in the
Saunders; 2013:162-173 [Chapter 12]. treatment of combined gastroesophageal reflux disease and bloating.
5. Meilahn J. Motility disorders of the stomach and small intestine. In: World J Surg. 2007;31(2):332.
Yeo CJ, Matthews JB, McFadden DW, et al., eds. Shackelford’s Surgery 30. Farrell T, Richardson W, Halkar R, et al. Nissen fundoplication
of the Alimentary Tract. 7th ed. Philadelphia: Saunders; 2013:781-790 improves gastric motility in patients with delayed gastric emptying.
[Chapter63]. Surg Endosc. 2001;15(3):271.
6. Gourcerol G, Benanni Y, Boueyre E, Leroi AM, Ducrotte P. Influ- 31. Hibbard M, Dunst C, Swanström L. Laparoscopic and endoscopic
ence of gastric emptying on gastro-esophageal reflux: a combined pyloroplasty for gastroparesis results in sustained symptom improve-
pH-impedance study. Neurogastroenterol Motil. 2013;25(10):800. ment. J Gastrointest Surg. 2011;15(9):1513.
7. Stefanidis D, Hope W, Kohn G, et al. Guidelines for surgical treatment 32. Vu M, Straathof J, Schaar P, et al. Motor and sensory function of the
of gastroesophageal reflux disease. Surg Endosc. 2010;24(11):2647. proximal stomach in reflux disease and after laparoscopic Nissen
8. Emerenziani S, Sifrim D. Gastroesophageal reflux and gastric fundoplication. Am J Gastroenterol. 1999;94(6):1481.
emptying, revisited. Curr Gastroenterol Rep. 2005;7(3):190. 33. Bais J, Samsom M, Boudesteijn E, van Rijk PP, Akkermans LM,
9. Maddern G, Chatterton B, Collins P, Horowitz M, Shearman DJ, Gooszen HG. Impact of delayed gastric emptying on the outcome
Jamieson GG. Solid and liquid gastric emptying in patients with of antireflux surgery. Ann Surg. 2001;234(2):139.
gastro-oesophageal reflux. Br J Surg. 1985;72(5):344. 34. Wayman J, Meyers J, Jamieson G. Preoperative gastric emptying
10. Cunningham K, Horowitz M, Riddell P, et al. Relations among and patterns of reflux as predictors of outcome after laparoscopic
autonomic nerve dysfunction, oesophageal motility, and gastric fundoplication. Br J Surg. 2007;94(5):592.
emptying in gastro-oesophageal reflux disease. Gut. 1991;32(12): 35. Rebecchi F, Allaix M, Giaccone C, Morino M. Gastric emptying as a
1436. prognostic factor for long-term results of total laparoscopic fundopli-
11. Schwizer W, Hinder R, DeMeester T. Does delayed gastric emptying con- cation for weakly acidic or mixed reflux. Ann Surg. 2013;258(5):831.
tribute to gastroesophageal reflux disease? Am J Surg. 1989;157(1):74. 36. Broeders J, Mauritz F, Ahmed Ali U, et al. Systematic review and
12. Herculano J, Troncon L, Aprile L, et al. Diminished retention meta-analysis of laparoscopic Nissen (posterior total) versus Toupet
of food in the proximal stomach correlates with increased acidic (posterior partial) fundoplication for gastro-oesophageal reflux
reflux in patients with gastroesophageal reflux disease and dyspeptic disease. Br J Surg. 2010;97(9):1318.
symptoms. Dig Dis Sci. 2004;49(5):750. 37. Williams V, Watson T, Gellersen O, et al. Gastrectomy as a remedial
13. Stacher G, Lenglinger J, Bergmann H, et al. Gastric emptying: a contribu- operation for failed fundoplication. J Gastrointest Surg. 2007;11(1):29.
tory factor in gastro-oesophageal reflux activity? Gut. 2000;47(5):661. 38. Clark C, Sarr M, Arora A, Nichols FC, Reid-Lombardo KM. Does
14. Barbieri C, Troncon L, Herculano J, et al. Postprandial gastric antral gastric resection have a role in the management of severe postfun-
contractions in patients with gastro-oesophageal reflux disease: a doplication gastric dysfunction. World J Surg. 2011;35(9):2045.
scintigraphic study. Neurogastroenterol Motil. 2008;20(5):471. 39. Nadaleto B, Herbella F, Patti M. Gastroesophageal reflux disease in
15. King P, Pryde A, Heading R. Transpyloric fluid movement and the obese: pathophysiology and treatment. Surgery. 2016;159(2):475.
antroduodenal motility in patients with gastro-oesophageal reflux. 40. Khan A, Kim A, Sanossian C, Francois F. Impact of obesity treat-
Gut. 1987;28(5):545. ment on gastroesophageal reflux disease. World J Gastroenterol.
16. Carmagnola S, Fraquelli M, Cantù P, Conte D, Penagini R. Relation- 2016;22(4):1627.
ship between acceleration of gastric emptying and oesophageal acid 41. Schauer P, Ikramuddin S, Gourash W, Ramanathan R, Luketich J.
exposure in patients with endoscopy-negative gastro-oesophageal Outcomes after laparoscopic Roux-en-Y gastric bypass for morbid
reflux disease. Scand J Gastroenterol. 2006;41(7):767. obesity. Ann Surg. 2000;232(4):515.
17. Penagini R, Bravi I. The role of delayed gastric emptying and 42. Papasavas P, Ng J, Stone A, Ajayi OA, Muddasani KP, Tishler DS.
impaired oesophageal body motility. Best Pract Res Clin Gastroenterol. Gastric bypass surgery as treatment of recalcitrant gastroparesis.
2010;24(6):831. Surg Obes Relat Dis. 2014;10(5):795.
18. Stapleton J, Wo J. Current treatment of nausea and vomiting associated 43. Le Page P, Martin D. Laparoscopic partial sleeve gastrectomy with
with gastroparesis: antiemetics, prokinetics, tricyclics. Gastrointest fundoplication for gastroesophageal reflux and delayed gastric
Endosc Clin N Am. 2009;19(1):57. emptying. World J Surg. 2015;39(6):1460.
19. Manzotti M, Catalano H, Serrano F, Di Stilio G, Koch MF, Guyatt G. 44. Reynolds J, Zehetner J, Wu P, Shah S, Bildzukewicz N, Lipham JC.
Prokinetic drug utility in the treatment of gastroesophageal reflux Laparoscopic magnetic sphincter augmentation vs laparoscopic
esophagitis: a systematic review of randomized controlled trials. Nissen fundoplication: a matched-pair analysis of 100 patients.
Open Med. 2007;1(3):e171. J Am Coll Surg. 2015;221(1):123.
CHAPTER
Management of Failed Fundoplications, End-Stage
Gastroesophageal Reflux Disease, and Scleroderma 23
Hugh G. Auchincloss
| David W. Rattner
P
atients with recurrent, persistent, or new symptoms antireflux operations require ongoing medical therapy.4
after antireflux surgery can be a challenging problem Many patients who resume PPIs have nonreflux-related
for the foregut surgeon. Determining who will benefit causes for their symptoms5; it is therefore more accurate
from reoperation and what operation to perform requires to state that between 75% and 80% of patients undergoing
that the surgeon be able to interpret a host of preopera- primary antireflux surgery will have no further pathologic
tive studies and be familiar with the common methods acid reflux, as documented by pH probe testing, for the
of failure associated with antireflux procedures. When remainder of their life span. Many patients who experience
reoperation is contemplated, the anticipated functional mild recurrent heartburn can be managed medically, but
outcome must be balanced against both the efficacy of between 3% and 6% will ultimately require reoperation.6,7
resuming medical therapy and the morbidity of a second, Interestingly, failures do not always occur in the early
third, or fourth procedure. The relative modesty of these postoperative period but rather increase in incidence over
functional outcomes requires that patients’ expectations time; the occasional patient will experience early improve-
be managed carefully. However, an experienced surgeon ment but ultimately experience failure of fundoplication
may reasonably offer many of these patients improvement a decade or more later.2
in alimentary function and quality of life. In patients presenting for reoperation after failed
Patients nowadays rarely present with end-stage gastro- antireflux surgery the most common complaint is recurrent
esophageal reflux disease (GERD) in the absence of prior heartburn or regurgitation. These symptoms are present in
failed antireflux surgery due to the near ubiquitous use of 60% of patients with failed fundoplications. Dysphagia as
proton pump inhibitors (PPIs). However, the occasional a dominant symptom is present in 30% of patients. Other
patient will present with complications related to chronic complaints include hiatal hernia, gas bloat, and atypical
GERD, including profound esophageal dysmotility or symptoms such as chest or abdominal pain.8–10 Often a
long-segment Barrett esophagus with strictures, that combination of symptoms is present, making it difficult
preclude standard antireflux procedures. Such cases to distinguish anatomic and functional reasons for failure.
require thoughtful management by a surgeon and/
or multidisciplinary team that includes expertise in WHY DO FUNDOPLICATIONS FAIL?
performing complex foregut reconstruction. Patients
with scleroderma—otherwise known as systemic sclerosis It is important for the foregut surgeon to understand how
(SSc)—and esophageal involvement represent a particularly and why fundoplications fail. In addition to guiding the
challenging subset of patients. Surgery in these patients surgeon away from similar failures in their own practice,
should be approached with caution given the increased risk this understanding allows the surgeon to put the signs
associated with intervention and the diminished prospects and symptoms of the “failed” patient into context. An
for functional improvement. However, well-selected patients educated assessment about the need and strategy for
with SSc can benefit from surgery to relieve regurgitation, further operations follows. Generally speaking, fundoplica-
heartburn, and occasionally dysphagia. tions fail because of patient factors that existed prior to
surgery, technical problems that lead to compromise of
the operation, or early postoperative coughing or retching.
OUTCOMES AFTER FUNDOPLICATION When failure is attributed to patient factors, it can be
Laparoscopic Nissen fundoplication was introduced in reasonably said that the preoperative assessment of the
19911 and has become the standard approach to the surgi- patient was inadequate, data gathered during that assess-
cal management of GERD. Variations on the 360-degree ment were misinterpreted, or that poor judgment was used
fundoplication—including the Hill repair, 180-degree in developing a surgical strategy. Alternatively, the patient
anterior Dor fundoplication, and the 270-degree posterior may experience progression of a condition that was present
Toupet fundoplication—are also commonly performed. but insignificant prior to surgery, such as deterioration
Transthoracic fundoplication (e.g., the Belsey Mark IV in esophageal peristalsis. The ideal patient for antireflux
fundoplication) has become increasingly uncommon surgery is a nonobese individual with relatively preserved
as a primary antireflux procedure. Several large series esophageal peristalsis, documented abnormal pH testing
have examined long-term outcomes following laparo- with good symptom correlation, typical symptoms of GERD,
scopic antireflux surgery. Although enthusiasts claim and symptoms that are at least partially responsive to PPIs.11
that 90% of patients experience durable symptom relief Patients with concomitant esophageal motility disorder
and improvement in quality of life,2,3 nonsurgical series (such as achalasia or diffuse esophageal spasm), those in
generally report that 20% to 50% of patients undergoing whom obesity contributes to reflux, and those with nonacid
269
Management of Failed Fundoplications, End-Stage Gastroesophageal Reflux Disease, and Scleroderma CHAPTER 23 269.e1
ABSTRACT
Reoperative antireflux surgery is a complex clinical
problem. Foregut surgeons who treat patients with failed
fundoplications must be well versed in the indications
for reoperation, common modes of failure, operative
techniques, and expected outcomes. Scleroderma and end-
stage gastroesophageal reflux disease are encountered less
frequently than failed antireflux surgery but also require
the expertise of an experienced foregut surgeon. Many
of the principles and techniques that apply to reoperative
antireflux surgery are relevant in the management of
these conditions as well.
KEYWORDS
Fundoplication; reflux; dysphagia; herniation; scleroderma
270 SECTION I Esophagus and Hernia
reflux or atypical symptoms including laryngospasm, injury to the vagus nerves probably contributes to some
chest pain, and recurrent aspiration have demonstrably patients’ symptoms postfundoplication, particularly if these
inferior outcomes following fundoplication. Similarly, those symptoms can be attributed to poor gastric emptying,
with overlooked anatomic or functional abnormalities, diarrhea, or perhaps gas bloat syndrome.
such as esophageal strictures, fistula, or delayed gastric The most dangerous cause of early failure is hernia-
emptying, are unlikely to have their symptoms relieved tion of the wrap in the immediate postoperative period.
by fundoplication alone. Advanced age, female gender, This creates an iatrogenic incarcerated hiatal hernia
and the presence of a large hiatal hernia have also been with potential compromise of regional blood flow. An
noted as potential risk factors for failure.12 episode of violent coughing or retching with increase
Technical failures of fundoplication have been well in intraabdominal pressure may precede such event.
described and may compromise the early technical success Without prompt intervention, the patient is at risk for
of the operation or the durability of the repair. A typical necrosis of the stomach leading to significant morbidity
fundoplication involves restoring the gastroesophageal or death. If a portion of the stomach becomes infarcted,
junction and several centimeters of esophagus to an the reconstruction options are limited by the patient’s
intraabdominal position, construction of a tension-free, physiologic condition and the availability of a viable
“floppy” fundoplication wrap composed of the gastric conduit to restore alimentary continuity. A staged repair
fundus around a short segment of the intraabdominal is sometimes necessary in this circumstance.
esophagus, and securing the wrap and closure of the
diaphragmatic crura with permanent suture. Variation
exists with regard to the type of fundoplication performed, APPROACH TO THE PATIENT WITH
the need for division of the short gastric arteries, and A FAILED FUNDOPLICATION
the use of bioprosthetic mesh to reinforce the closure of
the hiatus. Studies by Awais,6 Dallemagne,7 Khajanchee,10 Many patients with recurrent or new symptoms after
Furnée13 found that migration of fundoplication wrap antireflux surgery are reluctant to seek the advice of
was the most common anatomic defect encountered a surgeon because of the perception that surgery has
at the time of reoperation, occurring in approximately contributed to their situation or has little to offer by
two-thirds of patients (Table 23.1). Several types of wrap way of palliation. Such patients may only present after
migration can occur, including transhiatal herniation of years of marginally beneficial medical and minimally
an intact wrap into the mediastinum or herniation of invasive therapy. Symptoms and secondary sequelae of
the proximal stomach through the wrap into a supra- or failed fundoplication may therefore be quite advanced.
infradiaphragmatic position (i.e., the “slipped” Nissen). Evaluation of all patients begins with a careful attempt
Other technical problems encountered at reoperation to understand the circumstances of the prior operation.
include a crural closure or wrap that is too tight or too long, When available, the patient’s preoperative studies and
a malpositioned or twisted wrap, or complete disruption of operative report should be reviewed to look for clues as
the wrap. An important cause of dysfunction after antire- to the nature of the failure. Attention should be paid to
flux surgery stems from a failure to recognize or address similarities and differences between the patient’s previous
a shortened esophagus. The importance of thorough GERD symptoms and their current symptom complex.
mediastinal dissection with restoration of intraabdominal For example, dysphagia that was present prior to the
esophagus to the success of antireflux surgery cannot be first operation and remains speaks to a different etiology
overstated. Inability to create sufficient intraabdominal of failure than dysphagia that developed only after the
esophagus despite this dissection should be addressed with fundoplication.
a Collis gastroplasty. Omission of these steps results in a Even in a patient with a history and symptoms that
wrap that is improperly situated and is subject to tensile strongly suggest a specific reason for failure of fundo-
forces that make it prone to transhiatal herniation. Lastly, plication, it is necessary to obtain objective data prior
to proceeding with reoperation. It is recognized that
symptoms and objective findings in primary and secondary
GERD are imperfectly correlated; however, the informa-
tion gained from such studies—which include upper
TABLE 23.1 Causes of Fundoplication Failure
gastrointestinal contrast imaging, high-resolution manom-
etry, pH testing, multichannel intraluminal impendance
Type of Failure Incidence Symptoms testing, and endoscopy—guides surgical decision making
Hiatal hernia 40%–65% Reflux, dysphagia, and provides a baseline against which the outcome of
asymptomatic reoperation may be measured.
Slipped wrap 4%–16% Reflux, dysphagia,
early satiety, CONTRAST IMAGING
postprandial pain Contrast imaging—traditionally a dynamic barium swal-
Loose or disrupted 3%–23% Reflux lowing study—is essential for all patients with symptoms
wrap after fundoplication. A well-situated, intact fundoplication
Tight or twisted wrap 1%–10% Dysphagia appears as a filling defect of the gastric fundus that is
Underlying esophageal 1%–2% Dysphagia smooth in contour and located mostly anteriorly. The
dysmotility distal esophagus is narrowed slightly as it transverses this
Data from references 6–8, 10, 14. defect.15 A barium swallow in a patient with recurrent
Management of Failed Fundoplications, End-Stage Gastroesophageal Reflux Disease, and Scleroderma CHAPTER 23 271
Procedure Indication
Redo fundoplication First-time failure in a nonobese
patient (consider Toupet
fundoplication for patients with
esophageal dysmotility)
Collis gastroplasty or Any reoperation involving a short
wedge fundectomy esophagus or hiatal hernia (use
with caution in patients with
esophageal dysmotility)
Roux-en-Y gastric First-time failure in an obese
bypass or patient
gastrectomy/ Multiple prior antireflux procedures
esophagojejunostomy Impaired gastric motility
Esophageal or Esophageal stricture
gastroesophageal Barrett esophagus with high-grade
junction resection dysplasia
FIGURE 23.2 An intraoperative photo demonstrating mesh from a Severely scarred gastric fundus
prior hiatal hernia repair eroding into the gastric fundus. from multiple prior operations
Advances in laparoscopy have largely superseded the stomach is necessary only in selected patients with profound
need for a thoracotomy to achieve adequate exposure gastric dysmotility or acid hypersecretion.28 The primary
to the mid- and distal esophagus in first-time redo situ- disadvantages of the procedure are the increased complex-
ations. A Collis gastroplasty may be required to achieve ity and the addition of two gastrointestinal anastomoses.
adequate esophageal length. For patients with significant Roux-en-Y reconstruction should be considered the
dysmotility, a partial fundoplication is probably preferable revision antireflux operation of choice in patients with
to a complete fundoplication. Lastly, although the use obesity or in whom the distal esophagus or gastric fundus
of bioprosthetic mesh to close the hiatus has not been are unusable due to the prior operation or complications
proven beneficial,22 it is a reasonable adjunct in selected of GERD. It should be strongly considered in patients with
patients who have attenuated crural musculature, given that multiple prior antireflux surgeries. However, enthusiasm
recurrent hiatal hernia is a frequent mode of secondary for Roux-en-Y reconstruction in patients who are otherwise
failure. candidates for revision fundoplication should be tempered
by the lack of long-term follow-up. Roux-en-Y reconstruc-
GASTRECTOMY WITH ROUX-EN-Y tion deprives the patient of the reservoir function of
RECONSTRUCTION the stomach. In patients with delayed gastric emptying,
The modest long-term results associated with revision this may be an advantage, but in others it may lead to
fundoplication have led some surgeons to recommend poor oral intake and early satiety. This is particularly
gastrectomy with Roux-en-Y reconstruction as the preferred true in patients who require an esophagojejunostomy.
approach to the patient with a failed antireflux surgery.9,23–28 Roux-en-Y gastric bypass is also known to expose patients
It is already well established that Roux-en-Y gastric bypass to the lifelong consequences of vitamin deficiency and
is the optimal primary antireflux operation for patients malabsorption. This is a particular problem in normal
with GERD and morbid obesity. In these procedures, the weight individuals. Lastly, Roux-en-Y reconstruction can
antireflux mechanism of the lower esophageal sphincter is create conditions for future abdominal surgical problems,
not reconstructed as it is with a fundoplication. The Roux including internal hernia, intussusception, and marginal
limb prevents reflux of biliopancreatic secretions into ulcers. If these problems occur, further reconstructive
the gastric pouch and esophagus. The gastrojejunostomy options may be limited.
should be performed to a small gastric pouch to minimize
inclusion of the acid-producing cells of the gastric body. ESOPHAGEAL REPLACEMENT
A Roux limb of 120 cm is important to facilitate weight Foregut function is seldom normal after resection of the
loss, which in turn decreases intraabdominal pressure proximal stomach and distal esophagus. Most patients
and reflux. experience some degree of reflux, dysphagia, or poor
There are data to show that in patients with obesity and conduit emptying. Esophageal resection for benign disease
a prior failed fundoplication, Roux-en-Y reconstruction is is therefore reserved for patients in whom symptoms are
superior to revision fundoplication in suppression of reflux, intractable and no other good surgical option exists.
avoidance of dysphagia, and overall quality of life.23,29 Indications for esophageal replacement in the patient with
Approximately 80% of these patients are pleased with the prior failed fundoplication include progressive esophageal
result of their operation, despite a higher incidence of ulceration or fistula, a nondilatable esophageal stricture, or
perioperative complications.23 These results have led others severe dysphagia with diffuse dysmotility of the esophageal
to extend this technique to normal weight patients with body. Patients with multiple prior antireflux operations
a prior failed antireflux procedure. Stefanidis27 reported and continued symptoms should also be considered
96% patient satisfaction in 25 obese and nonobese patients for esophageal replacement. Limited series of patients
at 1-year follow-up. There was 100% satisfaction in the undergoing esophagectomy after failed antireflux have
group of patients presenting with dysphagia. Makris,25 found that complications occur in 24% of patients but
Mittal,26 and Kim9 reported similar results, with patient that 88% of patients are ultimately satisfied with their
satisfaction levels of 89%, 72%, and 93%, respectively. operation.30
Awais24 reported “excellent” quality of life in the average Several controversies exist with regard to esophagectomy
normal weight patient undergoing Roux-en-Y reconstruc- for benign disease. These include (1) long- versus short-
tion. Failure of the Roux-en-Y reconstruction was noted segment esophagectomy, (2) optimal operative approach,
in 5% of patients, with the mechanism of failure being and (3) stomach versus colon or jejunum for esophageal
hiatal hernia in all cases.24 substitute.
In some reoperative cases in which the fundus is badly Short-segment resection of the distal esophagus is often
scarred or damaged from prior operations and cannot be possible in patients with reflux-related complications
salvaged, the Roux limb can be brought to the distal or because pathology is focused around the gastroesophageal
mid-esophagus and an esophagojejunostomy constructed. junction. This approach is technically more straightforward
Roux-en-Y reconstruction for failed fundoplication can be and has the theoretical advantage of preserving as much
performed laparoscopically by an experienced minimally normal esophagus as possible. This is true even in patients
invasive foregut surgeon. For normal weight patients with some dysmotility of the proximal esophageal body.
a short Roux limb (~50 cm) should be constructed to These patients may have improvement in esophageal
prevent malabsorption. In selected patients a gastrostomy function after reduction in acid exposure. The drawback
tube can be placed in the remnant stomach to ensure of short-segment resection is that it places the esophageal
adequate enteral intake. A retrocolic retrogastric Roux anastomosis in the chest where the consequences of anas-
limb facilitates this placement. Gastrectomy of the remnant tomotic leak are greater than with a cervical anastomosis.
274 SECTION I Esophagus and Hernia
Of the many techniques for performing esophageal because of the high incidence of gastric emptying
resection—thoracoabdominal, Ivor Lewis, McKeown, dysfunction.
transhiatal, minimally invasive—none is clearly superior in
patients with benign disease, and most surgeons opt for the
technique with which they are most familiar and easiest for APPROACH TO THE PATIENT WITH
the patient to tolerate. Chang31 and Shen32 both reported END-STAGE GASTROESOPHAGEAL
large series of esophagectomy after antireflux surgery and REFLUX DISEASE
found no difference in approach. Luketich33 reported
on outcomes after minimally invasive esophagectomy It is rare to see a patient present with severe complications
and noted no difference in the 111 out of 1011 patients of reflux disease in the absence of prior antireflux surgery.
who had undergone prior gastric or esophageal surgery. These complications include intractable esophagitis,
It seems that the optimal approach should be tailored to dysmotility, ulceration or fistula, or undilatable stricture.
both the patient’s disease and the surgeon’s expertise. In some cases these complications are mild or moder-
For example, a left thoracoabdominal approach provides ate and may be expected to improve or resolve after
excellent exposure to the stomach and gastroesophageal fundoplication as a primary antireflux surgery. However,
junction and is particularly useful in obese patients who beyond this, more extensive resection and reconstruction
have undergone multiple hiatal repairs. A final note should is usually required. Options include Roux-en-Y gastric
be made of vagal-sparing esophagectomy, a technique bypass, esophagojejunostomy, or esophageal replacement
that can be combined with either a gastric or alternative as discussed previously.
conduit and has been shown to result in a decreased
incidence of postoperative dumping syndrome and conduit
dysfunction.34 Unfortunately, in patients with multiple prior APPROACH TO THE PATIENT
antireflux procedures, this technique is seldom possible WITH SCLERODERMA
due to extensive scarring of the hiatus, nor is preservation
of the vagus nerves necessarily desirable. SSc affecting the esophagus is a particularly challenging
The choice of conduit to replace the esophagus is con- problem that leads to significant morbidity in patients
troversial. A gastric conduit is the preferred reconstruction suffering from this progressive disease. SSc is a rare
option following resection of the esophagus for cancer. disorder with 10 new cases per million people each year.
This factors in the relative ease and safety of constructing Women are 4 times more likely than men to be affected.
a gastric conduit and the modest life expectancy of many Symptoms typically begin in the third or fourth decade
patients with esophageal malignancy. In contrast, for of life. Gastrointestinal manifestations are the third most
patients with benign disease the outcome measure of common presenting symptom (behind Raynaud disease
interest is functional performance over a presumably and cutaneous findings) and are seen in 90% of patients.
normal life span. This has led some surgeons to suggest Nearly half of patients with SSc will have symptomatic
that, although the body of the stomach is frequently usable esophageal involvement,39,40 primarily manifested as
as a conduit despite multiple prior surgeries,31,32 a short- or dysphagia or GERD symptoms due to severely impaired
long-segment colon or jejunum interposition graft may esophageal peristalsis.
provide superior quality of life. The theoretical advantages SSc is characterized by atrophy of the smooth muscle
of a colon or jejunum interposition are that it allows the found in the distal two-thirds of the esophagus and
junction of the stomach and the neoesophagus to remain replacement with fibrosis. This transforms the esophagus
in the abdomen and preserves the reservoir function of from propulsive muscular tube into a static, inflexible
the stomach. If the stomach is absent, a Roux limb can be structure. Characteristic manometry findings include
used to complete the distal colonic anastomosis. Colonic absent or diminished peristalsis of the esophageal body
mucosa in particular appears to be reasonably resistant to and weakening of the lower esophageal sphincter. The
changes associated with acid exposure. The blood supply esophagus frequently becomes shortened due to fibrosis,
to the colon and jejunum is also robust and the incidence resulting in a hiatal hernia. Patients experience often
of anastomotic leaks—thought to be a consequence of debilitating GERD with or without dysphagia. Symptoms are
ischemia—is probably less than with a gastric conduit. exacerbated by autonomic nerve dysfunction, gastroparesis
The disadvantages of colon or jejunum interposition are and other gastrointestinal dysmotility, and the frequent
apparent. The operation is more technically challenging coexistence of sicca syndrome (characterized by absent
and requires at least three enteric anastomoses. The or ineffective saliva production). Endoscopy frequently
colon is not always available as a conduit due to intrinsic finds Candida esophagitis (not a feature of typical GERD),
pathology. The jejunum usually does not reach to the neck which is thought to be related to immunosuppressive
for a long-segment esophageal replacement without the drugs and stasis within the distal esophagus. Strictures
addition of a microvascular anastomosis.35 Despite these have been documented in between 15% and 30% of
technical challenges, some centers have found that patient patients. It is unknown if Barrett esophagus and esophageal
satisfaction after colon interposition for benign disease carcinoma are more common in patients with SSc, but
is above 90%,36 and others have reported equivalent or this is presumed to be the case.39
superior outcomes when colon or jejunum interposition The treatment of esophageal dysfunction in patients with
is compared with a gastric conduit.37,38 When a colon or SSc is typically medical. PPIs and other acid-suppressing
jejunum interposition is constructed and anastomosed medications are used along with lifestyle modification to
to the stomach, a pyloromyotomy should be performed reduce GERD symptoms. Dysphagia is more difficult to
Management of Failed Fundoplications, End-Stage Gastroesophageal Reflux Disease, and Scleroderma CHAPTER 23 275
manage. Promotility agents, laxatives, and antibiotics to However, in experienced hands the operation can be
reduce bacterial overgrowth have been used with variable performed laparoscopically with low morbidity. Patients
efficacy. Systemic treatment of SSc may delay progres- with SSc suffering from both esophageal and small bowel
sion of the disease but does not reverse the end-organ dysmotility—particularly those with small bowel bacterial
dysfunction.41,42 overgrowth—are at theoretical risk for worsening dysphagia
Surgery should be avoided in most patients with SSc and esophagitis after Roux-en-Y reconstruction.
because of the poor prospects for functional improvement Kent and colleagues45 at the University of Pittsburgh
and risk of impaired wound healing. Patients with severe have advocated for Roux-en-Y gastric bypass or esophagoje-
symptoms refractory to medical therapy and those with junostomy as the operation of choice in patients with SSc.
complications of GERD including undilatable stricture, This is based on a series demonstrating improvement in
recurrent aspiration from regurgitation, or esophageal reflux symptoms, dysphagia, and overall quality of life at 21
erosion may benefit from carefully planned surgery. months in patients undergoing Roux-en-Y reconstruction
Options include partial fundoplication, gastric resection as opposed to fundoplication. The incidence of bloating
with Roux-en-Y reconstruction, and esophageal replace- and diarrhea were also decreased but not significantly.
ment. Few series have looked at the outcome of antireflux Though small in size, this study establishes credible support
surgery in patients with SSc, and results have been variable. for the hypothesized benefits of Roux-en-Y reconstruction
over standard fundoplication in patients with SSc.
FUNDOPLICATION
Laparoscopic fundoplication is the most common opera- ESOPHAGEAL REPLACEMENT
tion performed in patients with SSc because it is most Esophageal replacement is a morbid procedure in patients
familiar to the majority of foregut surgeons and it avoids with SSc and should be reserved for those patients with
the morbidity of a gastrointestinal anastomosis. However, long-segment undilatable strictures, intractable esophageal
it has significant drawbacks that limit its utility. Chief ulcerations, or esophageal carcinoma. In 1988 Mansour
among these limitations is high incidence of refractory et al.46 recommended that esophagectomy with colon
postoperative dysphagia in the setting of poor under- interposition be the primary operation for SSc; however,
lying esophageal motility. This problem is particularly this thinking does not reflect current advances in medical
pronounced in patients with gastric or diffuse intestinal therapy or laparoscopic surgical techniques. As recently
dysmotility. Also important is the difficulty in achieving an as 2007 Kent45 found that of five patients undergoing
adequate length of intraabdominal esophagus in the setting esophagectomy for SSc, one died and three of the remain-
of severe fibrosis, which may be impossible. Surgeons taking ing four suffered serious morbidity. In rare cases in which
on these cases need to be able to perform esophageal esophagectomy is necessary there is no consensus as to
lengthening procedures, such as a Collis gastroplasty, when which approach is best. However, a colon interposition
indicated. This in turn creates an aperistaltic segment of may provide a superior reflux barrier and alleviate some of
neoesophagus that contributes to postoperative dysphagia. the problems associated with gastric conduit dysfunction
Orringer43 and Poirier44 found that 17 out of 20 and 10 of in patients with preexisting autonomic dysfunction.
14 patients, respectively, had resolution of reflux symptoms
over a short period of follow-up after open Collis-Nissen
fundoplication. The rates of postoperative dysphagia
CONCLUSION
were 39% and 69%, respectively. Objective improvement Surgical management of patients with failed fundoplica-
based on pH monitoring, esophageal manometry, and tions, end-stage GERD, and scleroderma requires a great
endoscopy to assess for esophagitis was marginal. Although deal of judgment and technical skill. Care must be indi-
fundoplication is a potential option for some patients with vidualized based on many of the factors discussed in this
SSc, we would limit its use to those with relatively preserved chapter. For primary reoperations in nonobese patients,
esophageal length and peristalsis. Partial fundoplication a redo fundoplication is usually the best option. However,
is favored over complete fundoplication for the reasons patients who have had two or more prior fundoplications
stated previously. and those who are morbidly obese may be better served by
gastric resection or bypass with Roux-en-Y reconstruction.
GASTRECTOMY WITH ROUX-EN-Y In patients with extensive scarring, complications from
RECONSTRUCTION prior mesh placement, or end-stage damage to the distal
Roux-en-Y reconstruction of the foregut offers several esophagus, resection of the fundus and distal esophagus
advantages over fundoplication in patients with SSc. The may be necessary. In these situations the surgeon must have
reflux barrier is achieved by the addition of Roux limb experience with a variety of options to restore continuity
of sufficient length rather than by restoring the func- of the alimentary tract.
tion of the lower esophageal sphincter, resulting in a
decreased chance of postoperative dysphagia. Resecting
the stomach is beneficial in patients in whom gastroparesis REFERENCES
contributes to reflux disease. Lastly, a Roux limb can be 1. Dallemagne B, Weerts JM, Jehaes C, Markiewicz S, Lombard R.
made to reach to the distal or even mid-esophagus and Laparoscopic Nissen fundoplication: preliminary report. Surg Laparosc
can therefore be used as a reconstructive option in the Endosc. 1991;1(3):138-143.
2. Broeders JA, Rijnhart-de Jong HG, Draaisma WA, Bredenoord AJ,
presence of severe distal esophageal stricture or erosion Smout AJ, Gooszen HG. Ten-year outcome of laparoscopic and
requiring partial resection. Drawbacks include increased conventional Nissen fundoplication: randomized clinical trial. Ann
technical complexity and the addition of two anastomoses. Surg. 2009;250(5):698-706.
276 SECTION I Esophagus and Hernia
3. Morgenthal CB, Shane MD, Stival A, et al. The durability of laparo- 25. Makris KI, Panwar A, Willer BL, et al. The role of short-limb Roux-en-Y
scopic Nissen fundoplication: 11-year outcomes. J Gastrointest Surg. reconstruction for failed antireflux surgery: a single-center 5-year
2007;11(6):693-700. experience. Surg Endosc. 2012;26(5):1279-1286.
4. Spechler SJ, Lee E, Ahnen D, et al. Long-term outcome of medical 26. Mittal SK, Légner A, Tsuboi K, Juhasz A, Bathla L, Lee TH. Roux-en-Y
and surgical therapies for gastroesophageal reflux disease: follow-up reconstruction is superior to redo fundoplication in a subset of
of a randomized controlled trial. JAMA. 2001;285(18):2331-2338. patients with failed antireflux surgery. Surg Endosc. 2013;27(3):927-935.
5. Wijnhoven BP, Lally CJ, Kelly JJ, Myers JC, Watson DI. Use of 27. Stefanidis D, Navarro F, Augenstein VA, Gersin KS, Heniford BT.
antireflux medication after antireflux surgery. J Gastrointest Surg. Laparoscopic fundoplication takedown with conversion to Roux-en-Y
2008;12(3):510-517. gastric bypass leads to excellent reflux control and quality of life
6. Awais O, Luketich JD, Schuchert MJ, et al. Reoperative antireflux after fundoplication failure. Surg Endosc. 2012;26(12):3521-3527.
surgery for failed fundoplication: an analysis of outcomes in 28. Williams VA, Watson TJ, Gellersen O, et al. Gastrectomy as a
275 patients. Ann Thorac Surg. 2011;92(3):1083-1089; discussion remedial operation for failed fundoplication. J Gastrointest Surg.
1089-1090. 2007;11(1):29-35.
7. Dallemagne B, Arenas Sanchez M, Francart D, et al. Long-term results 29. Kellogg TA, Andrade R, Maddaus M, Slusarek B, Buchwald H,
after laparoscopic reoperation for failed antireflux procedures. Br Ikramuddin S. Anatomic findings and outcomes after antireflux proce-
J Surg. 2011;98(11):1581-1587. dures in morbidly obese patients undergoing laparoscopic conversion
8. van Beek DB, Auyang ED, Soper NJ. A comprehensive review of to Roux-en-Y gastric bypass. Surg Obes Relat Dis. 2007;3(1):52-57;
laparoscopic redo fundoplication. Surg Endosc. 2011;25(3):706-712. discussion 58-59.
9. Kim M, Navarro F, Eruchalu CN, Augenstein VA, Heniford BT, 30. Gadenstatter M, Hagen JA, DeMeester TR, et al. Esophagectomy
Stefanidis D. Minimally invasive Roux-en-Y gastric bypass for fun- for unsuccessful antireflux operations. J Thorac Cardiovasc Surg.
doplication failure offers excellent gastroesophageal reflux control. 1998;115(2):296-300, 302; discussion 300-301.
Am Surg. 2014;80(7):696-703. 31. Chang AC, Lee JS, Sawicki KT, Pickens A, Orringer MB. Outcomes
10. Khajanchee YS, O’Rourke R, Cassera MA, Gatta P, Hansen PD, after esophagectomy in patients with prior antireflux or hiatal
Swanström LL. Laparoscopic reintervention for failed antireflux hernia surgery. Ann Thorac Surg. 2010;89(4):1015-1021; discussion
surgery: subjective and objective outcomes in 176 consecutive 1022-1023.
patients. Arch Surg. 2007;142(8):785-901; discussion 791-793. 32. Shen KR, Harrison-Phipps KM, Cassivi SD, et al. Esophagectomy after
11. Campos GM, Peters JH, DeMeester TR, et al. Multivariate analysis of anti-reflux surgery. J Thorac Cardiovasc Surg. 2010;139(4):969-975.
factors predicting outcome after laparoscopic Nissen fundoplication. 33. Luketich JD, Pennathur A, Awais O, et al. Outcomes after minimally
J Gastrointest Surg. 1999;3(3):292-300. invasive esophagectomy: review of over 1000 patients. Ann Surg.
12. Grover BT, Kothari SN. Reoperative antireflux surgery. Surg Clin 2012;256(1):95-103.
North Am. 2015;95(3):629-640. 34. Banki F, Mason RJ, DeMeester SR, et al. Vagal-sparing esophagectomy:
13. Furnée EJ, Draaisma WA, Broeders IA, Smout AJ, Gooszen HG. a more physiologic alternative. Ann Surg. 2002;236(3):324-335;
Surgical reintervention after antireflux surgery for gastroesophageal discussion 335-336.
reflux disease: a prospective cohort study in 130 patients. Arch Surg. 35. Gaur P, Blackmon SH. Jejunal graft conduits after esophagectomy.
2008;143(3):267-274; discussion 274. J Thorac Dis. 2014;6(suppl 3):S333-S340.
14. Furnée EJ, Draaisma WA, Broeders IA, Gooszen HG. Surgical 36. Watson TJ, DeMeester TR, Kauer WK, Peters JH, Hagen JA. Esopha-
reintervention after failed antireflux surgery: a systematic review geal replacement for end-stage benign esophageal disease. J Thorac
of the literature. J Gastrointest Surg. 2009;13(8):1539-1549. Cardiovasc Surg. 1998;115(6):1241-1247; discussion 1247-1249.
15. Carbo AI, Kim RH, Gates T, D’Agostino HR. Imaging findings of suc- 37. Young MM, Deschamps C, Allen MS, et al. Esophageal reconstruction
cessful and failed fundoplication. Radiographics. 2014;34(7):1873-1884. for benign disease: self-assessment of functional outcome and quality
16. Jobe BA, Richter JE, Hoppo T, et al. Preoperative diagnostic workup of life. Ann Thorac Surg. 2000;70(6):1799-1802.
before antireflux surgery: an evidence and experience-based con- 38. Stephens EH, Gaur P, Hotze KO, Correa AM, Kim MP, Blackmon
sensus of the Esophageal Diagnostic Advisory Panel. J Am Coll Surg. SH. Super-charged pedicled jejunal interposition performance
2013;217(4):586-597. compares favorably with a gastric conduit after esophagectomy. Ann
17. Wain JC, Wright CD, Kuo EY, et al. Long-segment colon interposition Thorac Surg. 2015;100(2):407-413.
for acquired esophageal disease. Ann Thorac Surg. 1999;67(2):313-317; 39. Ebert EC. Esophageal disease in scleroderma. J Clin Gastroenterol.
discussion 317-318. 2006;40(9):769-775.
18. Wilshire CL, Louie BE, Shultz D, Jutric Z, Farivar AS, Aye RW. 40. Arif T, Masood Q, Singh J, Hassan I. Assessment of esophageal involve-
Clinical outcomes of reoperation for failed antireflux operations. ment in systemic sclerosis and morphea (localized scleroderma)
Ann Thorac Surg. 2016;101(4):1290-1296. by clinical, endoscopic, manometric and pH metric features: a
19. Pennathur A, Awais O, Luketich JD. Minimally invasive redo antireflux prospective comparative hospital based study. BMC Gastroenterol.
surgery: lessons learned. Ann Thorac Surg. 2010;89(6):S2174-S2179. 2015;15:24.
20. Symons NR, Purkayastha S, Dillemans B, et al. Laparoscopic revi- 41. Nagaraja V, McMahan ZH, Getzug T, Khanna D. Management
sion of failed antireflux surgery: a systematic review. Am J Surg. of gastrointestinal involvement in scleroderma. Curr Treatm Opt
2011;202(3):336-343. Rheumatol. 2015;1(1):82-105.
21. Furnée EJ, Draaisma WA, Broeders IA, Smout AJ, Vlek AL, 42. Carlson DA, Hinchcliff M, Pandolfino JE. Advances in the evaluation
Gooszen HG. Predictors of symptomatic and objective outcomes and management of esophageal disease of systemic sclerosis. Curr
after surgical reintervention for failed antireflux surgery. Br J Surg. Rheumatol Rep. 2015;17(1):475.
2008;95(11):1369-1374. 43. Orringer MB, Orringer JS, Dabich L, Zarafonetis CJ. Combined
22. Oelschlager BK, Pellegrini CA, Hunter JG, et al. Biologic prosthesis to Collis gastroplasty—fundoplication operations for scleroderma
prevent recurrence after laparoscopic paraesophageal hernia repair: reflux esophagitis. Surgery. 1981;90(4):624-630.
long-term follow-up from a multicenter, prospective, randomized 44. Poirier NC, Taillefer R, Topart P, Duranceau A. Antireflux operations
trial. J Am Coll Surg. 2011;213(4):461-468. in patients with scleroderma. Ann Thorac Surg. 1994;58(1):66-72;
23. Awais O, Luketich JD, Tam J, et al. Roux-en-Y near esophagojejunos- discussion 72-73.
tomy for intractable gastroesophageal reflux after antireflux surgery. 45. Kent MS, Luketich JD, Irshad K. Comparison of surgical approaches
Ann Thorac Surg. 2008;85(6):1954-1959; discussion 1959-1961. to recalcitrant gastroesophageal reflux disease in the patient with
24. Awais O, Luketich JD, Reddy N, et al. Roux-en-Y near esophagoje- scleroderma. Ann Thorac Surg. 2007;84(5):1710-1715; discussion
junostomy for failed antireflux operations: outcomes in more than 1715-1716.
100 patients. Ann Thorac Surg. 2014;98(6):1905-1911; discussion 46. Mansour KA, Malone CE. Surgery for scleroderma of the esophagus:
1911-1913. a 12-year experience. Ann Thorac Surg. 1988;46(5):513-514.
CHAPTER
Esophageal Complications of Bariatric Procedures
Joerg Zehetner
24
O
ver the last three decades bariatric surgery moved recurrent and not reversible, the band should be removed
from open to laparoscopic procedures, with all the and conversion to a laparoscopic gastric bypass or gastric
benefits of minimally invasive surgery. All existing sleeve recommended.
restrictive bariatric procedures may affect the esophagus
over time, with the possibility of mild or sometimes severe PSEUDOACHALASIA
complications. The full picture of severe outflow obstruction caused by a
Depending on the procedure performed, the esophagus narrow gastric band can lead to a type of aperistalsis called
can be affected minimally or severely. The gold standard of pseudoachalasia. Early on, it can be a reversible process,
bariatric surgery is still the laparoscopic Roux en-Y gastric as described with the motility disorders, and immediate
bypass. Despite the rise of the gastric sleeve, the gastric emptying of the band is warranted. This complication is
bypass is still one of the most favored bariatric procedures, rare after gastric sleeve or gastric bypass unless a severe
and because of its low side-effect profile, with low mortality stenosis at the incisura of the stomach or at the anastomosis
and morbidity, many bariatric surgeons consider it the is present.
preferred treatment option. Although the gastric bypass has
little or no effect on the esophagus—it is even described REFLUX ESOPHAGITIS
as a protective procedure regarding problems with reflux Although the laparoscopic gastric bypass is a known
and acid exposure—other restrictive procedures have a antireflux procedure, since with a small pouch only a
high potential for leading to esophageal complications. very small amount of acid can reflux into the esophagus,
other procedures—including gastric banding, gastric
sleeve, and duodenal switch—are known for the associated
COMPLICATIONS increased postoperative incidence of reflux esophagitis. If
a patient after gastric bypass develops reflux esophagitis,
ESOPHAGEAL DILATION there should be a high suspicion for a gastrogastric fistula.
With the overuse of the adjustable gastric band in the Diagnosis can be made with a contrast-swallow study of the
late 1990s up to the year 2012, as well as uncontrolled esophagus, a computed tomography (CT) scan with oral
overfilling of the gastric band, esophageal dilation can contrast, or an endoscopy. Even negative endoscopic or
be found at different degrees of severity. Gastric bands contrast studies cannot rule out fistulas. Indirect signs on
should be surveyed and the patient seen once or twice a CT could be air in the remnant stomach or increased acid
year by a physician. Bands were often overfilled to achieve exposure on pH studies. Treatment for reflux esophagitis
rapid weight loss. Nausea and vomiting in combination is aimed at the cause and may include excision of the
with the outlet obstruction of the esophagus may lead fistula, pouch revision with narrowing of the pouch,
to reversible and sometimes irreversible dilation of the or, if the alimentary limb is too short, lengthening the
esophagus. Esophageal dilation can be reversed if the band alimentary limb to 100 cm.
filling is reduced early in the process. Large esophageal In general, reflux esophagitis in bariatric patients is
dilation allows the patient to eat into the esophagus (as a treated with high-dose proton pump inhibitors (PPIs) for
neo-stomach), therefore making the gastric banding of no 4 weeks followed by a repeat upper endoscopy. In patients
use for controlling weight, and removal or conversion to a with reflux esophagitis after gastric banding, the initial
gastric bypass or gastric sleeve are the next necessary steps. treatment is opening the band—by emptying about half
of the amount of fluid in the band—to provide better
MOTILITY DISORDERS emptying of the esophagus. Most of the time a tight band
The adjustable gastric band can cause difficulties for the creates an outflow obstruction, which might still allow
esophagus if the restriction is too high and therefore reflux but can potentially reduce clearance, especially
causing an outflow obstruction. While it will initially cause in patients where the band is high on the stomach and
esophageal dilation and tertiary contractions, it can lead close to the gastroesophageal (GE) junction.
to aperistalsis called pseudoachalasia. The gastric sleeve procedure is associated with the
In cases of a severe motility disorder the gastric band has highest incidence of postoperative reflux and reflux
to be emptied to give the esophagus a chance to recover. If esophagitis. Although different publications report contro-
the overfilled band is left in place with the patient vomiting versial results, it is known that the gastric sleeve facilitates
frequently, progressive esophageal dilation may lead to reflux, especially in patients with a hiatal hernia or a known
worsened motility and potentially permanent damage. weak distal esophageal sphincter and where no concur-
After emptying the band completely for 2 months rent hiatal hernia repair is performed. It is also known
another attempt via increasing the restriction—filling the that patients with prior reflux esophagitis and chronic
band again—can be made. If the motility disorders are gastroesophageal reflux disease (GERD) symptoms are
277
Esophageal Complications of Bariatric Procedures CHAPTER 24 277.e1
ABSTRACT
Over the last three decades bariatric surgery has moved
from open to laparoscopic procedures, with all the benefits
of minimally invasive surgery. All existing restrictive bariat-
ric procedures may affect the esophagus over time, with
the possibility of mild or sometimes severe complications.
Depending on the procedure performed, the esophagus
can be affected minimally or severely. The gold standard of
bariatric surgery is still the laparoscopic Roux-en-Y gastric
bypass. Despite the rise of the gastric sleeve, the gastric
bypass is still one of the most favored bariatric procedures,
and because of its low side-effect profile, with low mortality
and morbidity, many bariatric surgeons consider it the
preferred treatment option. Although the gastric bypass has
little or no effect on the esophagus—it is even described
as a protective procedure regarding problems with reflux
and acid exposure—other restrictive procedures have a
high potential for leading to esophageal complications.
KEYWORDS
Esophageal complications; bariatric surgery; Roux-en-Y
bypass; sleeve gastrectomy
278 SECTION I Esophagus and Hernia
poor candidates for the gastric sleeve procedure because underlying malignancy and esophageal cancer. In
of the higher associated incidence of postoperative GERD. patients with strictures or stenosis, esophageal dilation
Studies have shown that the gastric sleeve—especially when with endoscopic balloon dilation is the first option.
performed with too narrow an angle of His—can lead to Recurrent strictures can be treated with a combination
dysfunction of the lower esophageal sphincter, as shown of high-dose PPIs, balloon or Savary dilation, and/or
by its increased distensibility on planimetry. steroid injections.
An evaluation for GERD symptoms is recommended Patients with persistent ulcers or stenosis after gastric
before sleeve gastrectomy. In patients with GERD symp- sleeve or banding should be converted to a laparoscopic
toms, an upper endoscopy to rule out reflux esophagitis is Roux-en-Y gastric bypass.
mandatory. A contrast-swallow study should be performed
to evaluate for esophageal motility disorders. In patients ESOPHAGEAL PERFORATION
with reflux esophagitis and/or poor esophageal motility, Esophageal perforation is very rare after bariatric surgery
a laparoscopic gastric bypass is the preferred procedure. and reported only in patients with severe vomiting after
In patients with reflux esophagitis after a gastric sleeve gastric banding. Treatment is similar to that for primary
procedure, the initial treatment is high-dose PPIs for 4 esophageal perforation, with immediate surgery and
weeks or, if reflux is recurrent, PPIs as long-term treatment. primary closure if diagnosed within 24 hours and drainage
In patient with persistent symptoms of GERD or persistent and stenting if diagnosed after 24 hours. If the perforation
reflux esophagitis despite medical therapy or side effects is in the distal esophagus under a migrated gastric band,
from long-term PPI treatment and if insufficient weight laparoscopic gastric band removal and fundoplication
loss was achieved with the gastric sleeve, conversion to a over the perforated area as well as drainage would be the
gastric bypass should be implemented. recommended salvage procedure.
In patients with sufficient weight loss and more than 1
year after gastric sleeve surgery, several surgical options are
available. In those with good esophageal motility, the LINX
SUMMARY
reflux management system (Torax Medical, Inc., Shoreview, The frequency of esophageal complications after bariatric
Minnesota) can be implanted. This laparoscopic procedure surgery depends on the type of procedure performed:
should be combined with a hiatal hernia repair in patients Gastric banding: The main complications are motility
with hiatal hernias. Another option is the endoscopic disorder, pseudoachalasia, esophageal dilations, and
Stretta procedure (Mederi Therapeutics, Inc., Norwalk, reflux esophagitis.
Connecticut), which is a radiofrequency application into Gastric sleeve: The main complications are reflux esopha-
the distal esophageal sphincter. Another option is the gitis, possible ulcers, and strictures.
ENDOSTIM stimulator implantation (EndoStim, Inc., Gastric bypass: This is still the gold standard for bariatric
Dallas, Texas), whereby two electrodes are placed close patients with symptomatic GERD.
to the lower esophageal sphincter to improve sphincter
tone. There are currently no evidence-based data at the
level of randomized studies; most of the recommendations
are based on expert opinion. SUGGESTED READINGS
Naef M, Mouton WG, Naef U, van der Weg B, Maddern GJ, Wagner HE.
ULCERS AND STENOSIS Esophageal dysmotility disorders after laparoscopic gastric banding—
Severe acid exposure of the distal esophagus can lead an underestimated complication. Ann Surg. 2011;253(2):285-290.
doi:10.1097/SLA.0b013e318206843e.
to ulcers and strictures, especially in patients with poor Reynolds JL, Zehetner J, Shiraga S, Lipham JC, Katkhouda N.
esophageal motility and hence poor clearance of the Intraoperative assessment of the effects of laparoscopic sleeve gas-
esophagus. In patients with severe reflux esophagitis and trectomy on the distensibility of the lower esophageal sphincter using
ulcers, biopsies should be taken on upper endoscopy to impedance planimetry. Surg Endosc. 2016; Apr 12 [Epub ahead
rule out malignancy. If the biopsies are negative, repeat of print].
Zehetner J, Holzinger F, Triaca H, Klaiber CH. A 6-year experience
studies with biopsies should be performed, as well with the Swedish adjustable gastric band Prospective long-term
as an endoscopic ultrasound of the distal esophagus audit of laparoscopic gastric banding. Surg Endosc. 2005;19(1):
and a CT scan of the chest and abdomen to rule out 21-28.
PART FIVE
Paraesophageal Hernia
CHAPTER
Paraesophageal Hernia: Etiology, Presentation, and
Indications for Repair 25
Jorge A. Vega Jr.
| Vic Velanovich
P
araesophageal hernias are the results of defects in the gastroesophageal (GE) junction into the posterior
diaphragmatic hiatus. Widening of the hiatus between mediastinum, which is usually the result of deterioration
the left and right diaphragmatic crura provides the of the phrenoesophageal ligament.5 The forces exerted
pathway for upward displacement of abdominal contents during swallowing and the negative intrathoracic pres-
into the mediastinum. Paraesophageal hernias are an sure combined with the positive intraabdominal pressure
increasingly common type of hiatal hernia. They can contribute to the stretching of the phrenoesophageal
be associated with life-threatening complications such ligament. Different types of collagen, particularly types I
as gastric volvulus leading to necrosis or perforation of and III, have been found to be reduced in the phreno-
the stomach. Due to these potential complications, it esophageal ligament of patients with GERD and hiatal
was thought that all paraesophageal hernias should be hernia.6 Type I hiatal hernia is also known as the “sliding”
repaired upon diagnosis. Recent evidence, however, has hiatal hernia (Fig. 25.1). Sliding hiatal hernias can be
suggested that a nonsurgical approach is reasonably safe large, but importantly, the GE junction remains above
in asymptomatic patients. Symptoms can be subtle and the herniated stomach.
include chest pain or pressure after meals, dysphagia for Types II, III, and IV hiatal hernias are the paraesophageal
solids, dyspnea on exertion out of proportion to general hernias, where the stomach and esophagus are juxtaposed.
health, early satiety, and the need to eat very small meals A paraesophageal hernia is a true hernia with a hernia
to avoid feeling uncomfortable. In addition, anemia is sac. The key feature that defines a paraesophageal hernia
a common condition in patients with a paraesophageal is that the fundus of the stomach is located above the GE
hernia and typically resolves with correction of the hernia. junction, which can either be in a normal intraabdominal
Surgical intervention is recommended for patients who location or also herniated into the chest. The location of
are exhibiting symptoms or signs associated with a para- the fundus relative to the GE junction defines a sliding
esophageal hernia. The general etiology, presentation, versus a paraesophageal hernia. Type II, or “rolling”
and indications for repair of paraesophageal hernias are hiatal hernias, occur when the gastric fundus herniates
reviewed here. anterior to the esophagus, with a normally positioned
intraabdominal GE junction. Type II is also referred to as
a “true” paraesophageal hernia. Congenital defects in the
ETIOLOGY esophageal hiatus can lead to paraesophageal hernias.7
Hiatal hernias occur when portions of the stomach or Type III hiatal hernias are a combination of types I and
other abdominal contents herniate superiorly into the II, in which both the GE junction and a portion of the
mediastinum through a defect in the esophageal hiatus. stomach—usually the gastric fundus—herniate into the
Hiatal hernias have been associated with gastroesophageal mediastinum. Type IV hiatal hernias contain stomach
reflux disease (GERD), and the prevalence and size of and other abdominal organs such as small bowel, colon,
the hiatal hernia has been described to correlate with the pancreas, or spleen in the mediastinum. The term giant
severity of reflux.1 The presence of a hiatal hernia has also paraesophageal hernia refers to large hiatal hernias where
been identified in nearly 40% of obese patients.2 Some of at least 50% of the stomach is in the mediastinum or the
the causes of hiatal hernias have been attributed to age, hernia measures at least 6 cm on endoscopy.8
stress, and degenerative processes on the diaphragm.3 Most
cases of hiatal hernia are acquired rather than congenital, PREVALENCE
although familial clustering has been reported.4 The actual prevalence of paraesophageal hernias is
not known. The most common hiatal hernia is type I,
CLASSIFICATION which accounts for up to 95% of all hiatal hernias.9 Para-
Four types of hiatal hernias have traditionally been esophageal hernias may account for up to 14% of all hiatal
described. Type I hiatal hernia is a migration of the hernias, and the majority of paraesophageal hernias are
279
Paraesophageal Hernia: Etiology, Presentation, and Indications for Repair CHAPTER 25 279.e1
ABSTRACT
Paraesophageal hernias are the results of defects in the
diaphragmatic hiatus. Types II to IV hiatal hernias are
also known as paraesophageal hernias. These hernias
can be associated with life-threatening complications
such as gastric volvulus leading to necrosis or perforation
of the stomach. Due to these potential complications, it
was thought that all paraesophageal hernias should be
repaired upon diagnosis. Recent evidence, however, has
demonstrated that a nonsurgical approach is a safe option
in many cases. Surgical intervention is usually reserved
for patients who are exhibiting symptoms secondary to
paraesophageal hernias. The general etiology, presentation,
and indications for repair of paraesophageal hernias are
reviewed here.
KEYWORDS
paraesophageal, hiatal, gastric volvulus, Cameron lesions,
epigastric pain
280 SECTION I Esophagus and Hernia
at the diaphragmatic hiatus, and they can be the source appear benign in the majority of cases. Sometimes, however,
of bleeding in patients with hiatal hernias.15 The most chest examination can reveal decreased breath sounds
common location of these ulcerations is on the lesser on the affected side, or the presence of bowel sounds
curve of the stomach, at the level of the diaphragmatic within the chest. Many patients undergo evaluation for
hiatus. In those patients with hiatal hernias, Cameron chest pain that eventually leads to upper gastrointestinal
lesions are visualized in up to 4.7%. 15 Haurani et al. evaluation, and the diagnosis of a paraesophageal hernia.
reported that patients with paraesophageal hernia and Radiographic or endoscopic evaluation for other reasons
anemia were greater than seven times more likely to have may reveal the presence of a paraesophageal hernia in
evidence of linear ulcerations or erosions in gastric mucosa an asymptomatic patient.
compared with patients who had paraesophageal hernias
without evidence of bleeding.16 They also demonstrated RADIOGRAPHIC STUDIES
that anemia secondary to these lesions resolved in the Chest radiographs are obtained many times as part of the
majority of patients undergoing surgical repair of their work-up of chest pain. An upright radiograph of the chest
paraesophageal hernias. may be diagnostic for paraesophageal hernia, revealing
the pathognomic retrocardiac air-fluid level.19 Lateral
RESPIRATORY SYMPTOMS radiographs usually demonstrate retrocardiac opacities
Patients with paraesophageal hernias can also develop or air-fluid levels (Fig. 25.2). A radiograph demonstrating
dyspnea and cough. These patients report a progression coiling of a nasogastric tube in the thorax can be used
of their symptoms throughout the day. GE reflux is an to help demonstrate the presence of an intrathoracic
infrequent complication of type II hiatal hernia and may stomach. Computed tomography (CT) scans can be useful
present in the form of respiratory complaints.17 Evaluation in demonstrating anatomic details in a patient with a hiatal
of paraesophageal hernia should be started in patients hernia, but they are not typically used as part of the work-up
with a history of reflux who present with unexplained for a hiatal hernia. CT scan is useful to differentiate
dyspnea or new-onset bronchospasm. There have been between other types of hernias of the diaphragm such
multiple reports that pulmonary symptoms attributed to as Morgagni hernia or traumatic hernias.
paraesophageal hernias improve after operative repair.
The size of hiatal hernia inversely correlates with total CONTRAST ESOPHAGOGRAPHY
lung capacity and vital capacity, and improvements in Barium esophagram can be useful in the diagnosis of
lung volumes have been reported after surgical repair.18 paraesophageal hernias and often gives the most accurate
information regarding the hernia’s anatomy and location.
An esophagram can help differentiate between type II
DIAGNOSTIC APPROACH and type III hiatal hernias (Fig. 25.3). An esophagram
Evaluation of paraesophageal hernias begins with a history can also help provide functional information regarding
and physical examination. Many patients with paraesopha- esophageal peristalsis and reflux. Diagnostic accuracy of
geal hernias are asymptomatic. Physical examination can hiatal hernias can be improved using the right anterior
A B
FIGURE 25.2 Chest radiographs. Posteroanterior (A) and lateral (B) views of a patient with a paraesophageal hernia. Notice the large
air-fluid level behind the cardiac silhouette because of the intrathoracic stomach.
282 SECTION I Esophagus and Hernia
A B
FIGURE 25.3 Barium swallow of a patient with a paraesophageal hernia (same patient as in Fig. 25.2). (A) The majority of the stomach is
in an intrathoracic position. (B) Esophageal narrowing is seen because of compression from the intrathoracic portion of the stomach.
An analysis of 1005 patients with paraesophageal hernia 11. Oor JE, Wiezeer MJ, Hazebroek EJ. Hiatal hernia after open versus
examined the morbidity and mortality in octogenarians minimally invasive esophagectomy: a systematic review and meta-
analysis. Ann Surg Oncol. 2016;23(8):2690-2698.
after nonelective repair. A six- to sevenfold increase in 12. Chang CC, Tseng CL, Chang YC. A surgical emergency due to
mortality was associated with nonelective repair compared an incarcerated paraesophageal hernia. Am J Emerg Med. 2009;27:
with elective repair.33 This same analysis revealed that 134.
nonelective repairs were associated with a 50% longer 13. Ghosh RK, Fatima K, Ravakhah K, et al. Gastric volvulus: an easily
missed diagnosis of chest pain in the emergency room. BMJ Case
length of stay versus elective repair and were found to Rep. 2016;2016:pii: bcr2015213888. doi:10.1136/bcr-2015-213888;
be the predictor of inpatient mortality in patients over 14. Kabayashi F, Saiki M, Nakamura Y, et al. Aortogastric fistula caused by
the age of 80. It is important to distinguish between a foreign body in a hiatal hernia. Ann Thorac Surg. 2016;101:1976-1978.
asymptomatic patients and those patients with symptoms 15. Gray DM, Kushnir V, Kalra G, et al. Cameron lesions in patients with
attributed to their paraesophageal hernia. Patients who hiatal hernias: prevalence, presentation, and treatment outcome.
Dis Esophagus. 2015;28:448-452.
have symptoms of acute incarceration or strangulation 16. Haurani C, Carlin AM, Hammoud ZT, Velanovich V. Prevalence and
should undergo prompt surgical repair. Patients with resolution of anemia with paraesophageal hernia repair. J Gastrointest
obstructive symptoms, bleeding, or respiratory symptoms Surg. 2012;6:1817-1820.
attributed to their paraesophageal hernia should also 17. Greub G, Liaudet L, Wiesel P, Bettschart V, Schaller MD. Respiratory
complications of gastroesophageal reflux associated with para-
undergo surgical repair. Surgical management of the esophageal hiatal hernia. J Clin Gastroenterol. 2003;37:129.
elderly patient with a paraesophageal hernia should be 18. Naoum C, Kritharides L, Ing A, Falk GL, Yiannikas J. Changes in
individualized. A study of 354 patients who underwent lung volumes and gas trapping in patients with large hiatal hernia.
paraesophageal hernia repair revealed that mortality was Clin Respir J. 2015;11(2):139-150. doi:10.1111/crj.12314.
highest in patients over the age of 75.34 This study also 19. Kahrilas PJ, Kim HC, Pandolfino JE. Approaches to the diagno-
sis and grading of hiatal hernia. Best Pract Res Clin Gastroenterol.
revealed that morbidity was higher in patients with an 2008;22:601-616.
American Society of Anesthesiologists class 3 or 4 and in 20. Heacock L, Parikh M, Jain R, Balthazar E, Hindman N. Improving
patients with a type IV hiatal hernia. the diagnostic accuracy of hiatal hernia in patients undergoing
Another topic of discussion in paraesophageal hernia bariatric surgery. Obes Surg. 2012;22:1730-1733.
21. Skinner DB, Belsey RH. Surgical management of esophageal reflux
repair is the need for an antireflux procedure. Most and hiatus hernia. Long-term results with 1,030 patients. J Thorac
patients develop symptoms of GERD after paraesophageal Cardiovasc Surg. 1967;53:33.
hernia repair, and unless there is a contraindication, these 22. Hill LD. Incarcerated paraesophageal hernia. A surgical emergency.
patients would benefit from a fundoplication procedure in Am J Surg. 1973;126:286-291.
addition to their hiatal hernia repair. Adding a fundoplica- 23. Oddsdottir M, Franco AL, Laycock WS, et al. Laparoscopic repair of
paraesophageal hernia. New access, old technique. Surg Endosc. 1995;
tion to the paraesophageal hernia repair can help prevent 9:164-168.
GERD symptoms in these patients, which can result from 24. Hawasli A, Zonca S. Laparoscopic repair of paraesophageal hiatal
the extensive dissection required during the procedure.35 hernia. Am Surg. 1998;64:703-710.
Surgical management of paraesophageal hernias should 25. Wiechmann RJ, Ferguson MK, Naunheim KS, et al. Laparoscopic
management of giant paraesophageal herniation. Ann Thorac Surg.
be offered to symptomatic patients, but it should be 2001;71:1080-1087.
individualized, especially in the elderly patient with high 26. Asti E, Lovece A, Bonavina L, et al. Laparoscopic management
surgical risk. of large hiatus hernia: five-year cohort study and comparison
of mesh-augmented versus standard crura repair. Surg Endosc.
2016;30(12):5404-5409.
REFERENCES 27. Allen MS, Trastek VF, Deschamps C, et al. Intrathoracic stomach.
Presentation and results of operation. J Thorac Cardiovasc Surg. 1993;
1. Maish MS, DeMeester SR. Paraesophageal hernia. In: Cameron JL, 105:253-259.
ed. Current Surgical Therapy. 8th ed. Philadelphia: Mosby; 2004:38. 28. Pitcher DE, Curet MJ, Martin DT, Vogt DM, Mason J, Zucker KA.
2. Che F, Nguyen B, Cohen A, Nguyen NT. Prevalence of hiatal hernia Successful laparoscopic repair of paraesophageal hernia. Arch Surg.
in the morbidly obese. Surg Obes Relat Dis. 2013;9(6):920-924. 1995;130:590-596.
3. Kissane NC, Rattner DW. Paraesophageal and other complex 29. Gantert WA, Patti MG, Arcerito M, et al. Laparoscopic repair of
diaphragmatic hernias. In: Yeo CJ, ed. Shackelford’s Surgery of the paraesophageal hiatal hernias. J Am Coll Surg. 1998;186:428-433.
Alimentary Tract. 7th ed. Philadelphia: Saunders; 2012:494. 30. Hallissey MT, Ratliff DA, Temple JG. Paraoesophageal hiatus hernia:
4. Baglaj SM, Noblett HR. Paraesophageal hernia in children: familial surgery for all ages. Ann R Coll Surg Engl. 1992;74:23-25.
occurrence and review of the literature. Ped Surg Int. 1999;15:85-87. 31. Carlson MA, Condon RE, Ludwig KA, Schulte WJ. Management of
5. Hashemi M, Sillin LF, Peters JH. Current concepts in the management intrathoracic stomach with polypropylene mesh prosthesis reinforced
of paraesophageal hiatal hernia. J Clin Gastroenterol. 1999;29:8-13. transabdominal hiatus hernia repair. J Am Coll Surg. 1998;187:227-230.
6. Von Diemen V, Trindade EN, Trindade MR. Hiatal hernia and 32. Stylopoulos N, Gazeelle GS, Rattner DW. Paraesophageal hernias:
gastroesophageal reflux: study of collagen in the phrenoesophageal operation or observation? Ann Surg. 2002;236:492-501.
ligament. Surg Endosc. 2016;30(11):5091-5098. 33. Poulose BK, Gosen C, Marks JM, et al. Inpatient mortality analysis
7. Kleitsch WP. Embryology of congenital diaphragmatic hernia. I. of paraesophageal hernia repair in octogenarians. J Gastrointest Surg.
Esophageal hiatus hernia. Arch Surg. 1958;76:868. 2008;122:1888.
8. Melvin WS, Perry KA. Paraesophageal hernia-open repair. In: Fischer 34. Larusson JH, Zingg U, Hahnloser D, Delport K, Seifert B, Oertli D.
JE, ed. Fischer’s Mastery of Surgery. 6th ed. Philadelphia: Lippincott Predictive factors for morbidity and mortality in patients undergoing
Williams & Wilkins; 2012:760. laparoscopic paraesophageal hernia repair: age, ASA score and
9. Hyun JJ, Bak YT. Clinical significance of hiatal hernia. Gut Liver. operation type influence morbidity. World J Surg. 2009;33:980.
2011;5(3):267-277. 35. Casabella F, Sinanan M, Horgan S, Pellegrini CA. Systematic use
10. Postlewaite RW. Surgery of the Esophagus. 2nd ed. Norwalk, CT: of gastric fundoplication in laparoscopic repair of paraesophageal
Appleton Century-Crofts; 1986. hernias. Am J Surg. 1996;171:485.
CHAPTER
ABSTRACT
Laparoscopic repair of paraesophageal hiatal hernia has
become the standard of care in many centers across the
world including ours at the University of Pittsburgh Medical
Center. In our practice, we were strongly influenced by
the open surgical principles established by Griff Pearson.
However, few centers if any, have been able to come close
to the outstanding outcomes he produced. Dr. Pearson’s
low recurrence rate of 2% with up to ten years of follow-up
and longer has not been duplicated. In our center, we
now have long term follow-up of our laparoscopic results
showing an operative recurrence rate of 3% to 4%, and
another 10% have small medically manageable recurrences.
Some recent studies have had alarming recurrence rates
that may approach 50%. While many laparoscopic surgeons
claim to be able to do these operations, it is important
to continue to scrutinize our results and determine if we
can continue to improve upon our outcomes.
KEYWORDS
Hiatal hernia, gastroesophageal reflux disease, para-
esophageal hernia, GERD
Laparoscopic Paraesophageal Hernia Repair: Technique, Outcomes, and Management of Complications CHAPTER 26 285
Type I Type II
spectrum of associated symptoms because less than 5% the previous 5 to 10 years and substantial changes to their
of patients are truly asymptomatic when questioned diet to avoid hard and sometimes even soft solids.
thoroughly.4 Repair is generally recommended for all When these hernias progress to requiring semiurgent,
symptomatic patients. The management of a truly asymp- nonelective repair in our series, as well as by other sur-
tomatic hernia remains a topic of debate. The incidence geons, they are associated with a significantly increased risk
of a truly asymptomatic PEH is uncommon, and often of perioperative morbidity and mortality. In our series of
patients billed as asymptomatic frequently do suffer from 662 patients who underwent laparoscopic repair of a giant
significant symptoms such as shortness of breath that may PEH, patients admitted electively for laparoscopic repair
not be attributed to the hernia. Often these symptoms had a postoperative mortality rate of 0.5% compared with
have occurred insidiously and patients have learned to 7.5% for patients who underwent urgent repair.12 This can
live with these troublesome limitations and symptoms. In be markedly higher when patients present with gastric
the previous era of primarily open repair, some studies necrosis, massive hemorrhage, or severe aspiration pneu-
estimated that the risk of life-threatening complications monia, albeit the incidence of these more life-threatening
from a PEH was greater than 25% within a relatively short- situations is less common. Thus, when evaluating patients
term follow-up.9 More recently, it has been recognized that who may be minimally symptomatic, it is important to keep
life-threatening events are much less common and some these data in mind. The risk of perioperative mortality
authors have created risk-benefit algorithms to support and/or morbidity with elective and nonelective operation
the notion that life-threatening events are lower than the can be estimated to some degree by the size of the PEH,
risk of undergoing repair.10,11 However, when analyzing the the patient’s functional status, the presence of comorbid
findings of these studies, it is important to make note of conditions, and the patient’s symptom complex. In patients
the definition of minimally symptomatic or asymptomatic with age-adjusted Charleston Comorbidity Index scores of 5
used; in the paper by Stylopoulos, minimal symptoms or less, perioperative morbidity and mortality with elective
were defined as “heartburn that did not affect patient laparoscopic repair is low and increases dramatically when
quality of life.” In our experience, the vast majority of performed urgently. Furthermore, patients with very
patients with radiographic findings of large PEH will have large PEH were much more likely to have obstructive
obstructive symptoms, including dysphagia, postprandial symptoms and to present urgently when compared with
bloating, and chest pain and may not suffer from actual patients with smaller (<75% gastric herniation) PEH.13
heartburn. On occasion, elderly patients in our clinics may Urgent presentations often occur in patients in whom
deny difficulty swallowing but, when questioned further, the presence of the PEH was known much earlier. As
will report significant and unintentional weight loss over such, we recommend elective surgical repair for most
286 SECTION I Esophagus and Hernia
Surgeon
working port
FIGURE 26.2 Surgeon, port placement,
and instrument positions are shown.
The ports are positioned one-third of the
distance from the xiphoid to the
umbilicus.
dissection into the falciform ligament. In patients with a hiatus. Excessive traction or pulling on the stomach should
large protuberant abdomen, dividing the distance from be avoided because this causes unnecessary trauma. To
the xiphoid to umbilicus can be misleading; therefore accomplish sac reduction, the surgeon and assistant grasp
the initial right paramedian port should be placed the hernia sac just inside the hiatus at or near the 12
approximately 2 to 3 inches from the xiphoid process. o’clock position and attempt an atraumatic eversion. By
Insufflation pressures are set at between 12 and 15 mm everting this sac, we attempt to enter the layers of the
Hg, with attention paid to the patient’s hemodynamic state. elongated and attenuated phrenoesophageal ligament. By
In patients with poor cardiopulmonary risk, insufflation careful dissection, the surgeon can then use hemostatic
pressure in the 8- to 10-mm Hg range can be used with energy devices such as the harmonic scalpel (Ethicon,
reasonable visualization if blood pressure problems occur. Cincinnati, Ohio) to open and identify the foamy, areolar
The remainder of the ports are then placed under direct type plane, reminiscent of working in the retroperitoneum
vision. The assistant’s ports are positioned to the left of or perinephric fat. The appearance of this foamy layer is
the midline. The assistant’s left hand holds the camera quite characteristic and critical to the technical success and
through a 10-mm port in the left paramedian line slightly ease of the operation. Once this plane is identified, it is
lower than the position of the initial Hassan port at the important to try to stay within this plane as it is extended
right paramedian line. Two 5-mm ports are then placed up into the mediastinum. While working within this plane,
after full insufflation and just below the costal margins one can visualize the esophagus, the anterior vagus, the
bilaterally. In general, we like to maintain a hand’s breath aorta, and the pleura bilaterally. It is important to note
(9 to 10 cm) between working ports to avoid “scissoring” of that failure to identify this areolar plane can significantly
instruments. Liver retraction can be accomplished through increase the difficulty of the entire operation. Even though
a 5-mm port in the far right lateral subcostal position. these areolar attachments are only minimally vascularized,
we use energy here and minimize blunt dissection to
keep the mediastinum meticulously dry and hemostatic.
REDUCING THE HERNIA SAC Extensive, circumferential mobilization of the esophagus
Following port placement and liver retraction, the operat- is performed high into the mediastinum, laterally to the
ing room table is placed in steep reverse Trendelenburg to pleura and into the posterior mediastinum, and periaortic
facilitate visualization of the hiatus and aid in reducing the area. The dissection can be carried as high as the inferior
herniated abdominal contents. It is important to remember pulmonary veins and beyond if needed.
that the combination of the patient’s dehydrated preopera- Next, we exit the mediastinum and examine the right
tive fasting state and shift in intravascular volume to the crus and divide the gastrohepatic ligament. Occasion-
lower extremities upon steep Trendelenburg positioning ally, a significantly sized accessory hepatic artery may
can lead to a precipitous fall in blood pressure. Thus be encountered, which can be spared with some extra
gradual patient repositioning and initiation of abdominal technical work. However, in most cases, even if sizeable, a
insufflation will allow time for the anesthesiologist to temporary clip can be placed and liver perfusion reassessed
volume load the patient with intravenous fluids and allow in 15 to 20 minutes; if in the judgment of the surgeon,
the patient to respond more favorably to these changes. this vessel should be spared, it can be but with some
Reduction of herniated contents, such as omentum technical challenges. In most cases it can be divided with
and bowel, is performed upon initial assessment of the no significant sequela.
288 SECTION I Esophagus and Hernia
In some patients, there is a favorable angle for a lateral anatomy of the intraabdominal stomach, not just a “pexy.”
suture into the left crus at approximately “3 o’clock.” At We realize that until we have analyzed and published
the completion of the closure, a grasper should be easily our outcomes in this series of patients, the addition of a
introduced through the hiatus, with a small visible space fundoplication should be used in most patients.
surrounding the esophagus circumferentially. This is very At the completion of the operation a nasogastric tube
much an experience and judgment decision: you want can be placed by the anesthesiologist or surgeon under
the space to be minimal because, if too patulous, you risk direct laparoscopic visualization. It is critical that the
herniation of the wrap and/or other abdominal contents, surgeon and anesthesiologist approach this placement
and too tight can produce dysphagia. with care because the obstruction created by the wrap
and the closure can easily cause resistance to passage of
a nasogastric (NG) tube and potential for esophageal
REESTABLISHING THE perforation or at a minimum, suboptimal placement.
ANTIREFLUX BARRIER Although we believe this tube can be removed early on
postoperative day 1, we believe it is essential to enter the
Gastroesophageal reflux is present in only approximately recovery room with an NG tube and an empty stomach
50% of patients at the time of PEH repair, but a number to avoid postoperative retching, vomiting, or hiccupping.
of experts have published results showing the majority of
patients with GPEH have a history of GERD. In addition,
the reduction of the mediastinal sac and dissection of
MESH USE AT THE HIATUS
the esophagus disrupts the phrenoesophageal ligament The use of mesh at the hiatus remains controversial. Early
further and potentially the integrity of the lower esophageal trials with permanent synthetic mesh suggested a reduc-
antireflux barrier contributing to post-op GERD symptoms tion in hernia recurrence rates, but for most esophageal
if a wrap is omitted. There are some recent trials of PEH surgeons the potential complications associated with
repair comparing groups with and without an antireflux synthetic mesh, including erosion and difficult reopera-
fundoplication after the other steps of the hernia repair tions, outweigh the potential benefit. Two randomized
are complete, and the data have shown that a higher trials using biologic absorbable mesh have failed to show
percent of patients undergoing surgical PEH correc- a benefit for the use of this type of mesh. However, it
tion have postoperative reflux, if a fundoplication is not is important to recognize that neither trial aggressively
performed.11,15,16 Surgeon preference and preoperative assessed or treated tension. The fundamentals of hernia
esophageal manometry may help to determine the type surgery dictate that tension is the enemy of any hernia
of fundoplication to be performed: a circumferential repair, and it is logical that this tenet holds true at the
“floppy” fundoplication (two-stitch Nissen over a 54 or 56 hiatus as well. Consequently, future studies need to focus
bougie)17 or a partial fundoplication18,19 are the two most on adequately addressing tension in the form of relaxing
commonly reported. In the past, we routinely performed incisions for crural tension or adding an intentional
the circumferential “floppy” Nissen fundoplication but pneumothorax to create a “floppy diaphragm” to relieve
more recently have moved onto a partial fundoplication or tension during crural repair. In addition, the role of
“near” Nissen to minimize side effects such as dysphagia, Collis gastroplasty for axial esophageal tension should
gas bloat, and flatulence in our generally elderly PEH be further evaluated in controlled trials, and further
population. evaluation of the role of nonpermanent and permanent
We have noted that certain patients, particularly elderly, mesh reinforcement of the crural closure.
frail patients with an upside-down stomach and essentially
100% intrathoracic location, have primarily obstructive
symptoms and minimal heartburn. In this group, we
COMPLICATIONS AND OUTCOMES
are evaluating repair without a fundoplication and are In our center, routine postoperative care includes a barium
putting these data together in an attempt to better clarify esophagram on postoperative day 1 to establish a new
who might be predicted to do well without a wrap. In this anatomic baseline and rule out perforation or leak prior
very specific population, a gastropexy can be considered to initiating oral intake. In addition, given the results of
following careful adherence to the principles of PEH repair, a number of reports of high recurrence rates, the onus
including careful and complete sac dissection, complete is on the surgical team to document their surgical results
stomach mobilization, vagal preservation, and careful immediately postoperatively and then to follow this group
crural closure. Some surgeons have described gastropexy of patients and establish your own recurrence rates. Patients
as a single point of fixation using suture or the placement are discharged on liquid narcotic pain medication for 1
of a gastrostomy tube. We begin the gastropexy near the to 3 days and early are converted to oral liquid Tylenol.
angle of His to the left crus and then follow the cardia Patients are advised to refrain from heavy lifting long
and fundus along the diaphragm, just above the spleen, term and limit lifting to 15 to 20 pounds. In addition,
essentially in a line very near to where the short gastrics used we educate the patient on the avoidance of constipation
to live. We place multiple interrupted horizontal mattress and to watch for and treat early symptoms of gas bloat,
sutures (2 to 0 ethibond). Gastropexy sutures are placed using dietary manipulation and simethicone as needed.
on a diaphragmatic fold just a few millimeters above the Patients follow up in clinic in 2 weeks with a chest x-ray and
spleen, approximately 2 cm apart over a distance of 10 to then annually with a barium esophagram to monitor for
14 cm. By more or less duplicating what used to be the line radiographic recurrence. If any abnormal or concerning
of the short gastrics, we are attempting to recreate normal symptoms are present at any time point, the first step is
290 SECTION I Esophagus and Hernia
an interview with the patient and to review carefully a of mesh reinforcement only if other measures fail; and
barium esophagram, with direct comparison with the one (6) performance of an antireflux procedure or perhaps in
taken the first postoperative day. This close attention to select patients a gastropexy. Laparoscopic repair of PEH
detail facilitates early recognition of relevant symptoms, can provide excellent patient satisfaction and symptom
including dysphagia, and appropriate interventions to resolution when performed by surgeons with extensive
assist with patient comfort and satisfaction with quality of experience in minimally invasive and open esophageal
life. It is important for the surgeon to remain engaged in surgery. In this setting, we have shown that the outcomes,
this process because the patient’s primary care physicians both short and long term, and the reoperation rates with
and even their gastroenterologists may fail to recognize a minimally invasive approach can be comparable with
correctible problems that are related to the PEH repair. the best open series.2,12
Routine dietary changes should include avoiding gassy
foods and slowing down the eating process to avoid excess REFERENCES
gas swallowing, following what we call the “25 chew”
rule (i.e., chewing each bite of food 25 times). We also 1. Kwok H, Marriz Y, Al-Ali S, Windsor JA. Phrenoesophageal ligament
re-visited. Clin Anat. 1999;12:164-170.
recommend four to five small meals per day and avoiding 2. Maziak DE, Todd TR, Pearson FG. Massive hiatus hernia: evaluation
large feast-type meals. and surgical management. J Thorac Cardiovasc Surg. 1998;115(1):53-60,
In the early postoperative period, major postoperative discussion 61–62.
complications include pneumonia, congestive heart failure, 3. Oelschlager BK, Pellegrini CA, Hunter J, et al. Biologic prosthesis
reduces recurrence after laparoscopic paraesophageal hernia repair: a
and pulmonary embolisms can occur in a small subset of multicenter, prospective, randomized trial. Ann Surg. 2006;244:481-490.
patients. Postoperative mortality in the setting of elective 4. Luketich JD, Raja S, Fernando HC, et al. Laparoscopic repair of
repair should be less than 1% but is higher in patients older giant paraesophageal hernia: 100 consecutive cases. Ann Surg.
than 80 years and in patients requiring urgent repair.11,13 2000;232(4):608-618.
In a review of outcomes from more than 650 patients 5. Luketich JD, Maddaus MA. Laparoscopic Collis gastroplasty. In:
Pearson FG, Patterson GA, eds. Pearson’s Thoracic and Esophageal
who underwent laparoscopic giant paraesophageal hiatal Surgery. 3rd ed. Philadelphia: Churchill Livingstone/Elsevier;
hernia repairs, Luketich and colleagues reported major 2008:326-336.
adverse outcomes including pneumonia (4%), pulmonary 6. Mattar SG, Bowers SP, Galloway KD, Hunter JG, Smith CD. Long-term
embolism (3.4%), congestive heart failure (2.6%), need outcome of laparoscopic repair of paraesophageal hernia. Surg
Endosc. 2002;16(5):745-749. [Epub 2002 Feb 8].
for reintubation (2.6%), and postoperative leak (2.5%).12 7. Karmali S, McFadden S, Mitchell P, et al. Primary laparoscopic and
Laparoscopic repair by experienced laparoscopic esopha- open repair of paraesophageal hernias: a comparison of short-term
geal surgeons is associated with a significant decrease in outcomes. Dis Esophagus. 2008;21(1):63-68.
postoperative morbidity as compared with most series of 8. Nason KS, Luketich JD, Qureshi I, et al. Laparoscopic repair of giant
open repair (~25% of patient’s experience complications paraesophageal hernia results in long-term patient satisfaction and a
durable repair. J Gastrointest Surg. 2008;12(12):2066-2075, discussion
after laparoscopic repair as compared with ~60% after open 2075–2077.
repair), although there are no randomized comparative 9. Skinner DB, Belsey RH. Surgical management of esophageal reflux
studies that have been performed.7 The outcomes from and hiatus hernia: long-term results with 1,030 patients. J Thorac
the University of Pittsburgh revealed short postoperative Cardiovasc Surg. 1967;53:33-54.
10. Allen MS, Trastek VF, Deschamps C, Pairolero PC. Intrathoracic
hospitalizations (2 to 3 days) and low 30-day mortality stomach. Presentation and results of operation. J Thorac Cardiovasc
when PEH repair is performed electively. Importantly, Surg. 1993;105(2):253-258, discussion 258–259.
90% of patients reported good to excellent scores on 11. Stylopoulos N, Gazelle GS, Rattner DW. Paraesophageal hernias:
evaluation of their symptomatic outcomes, with only 3.4% operation or observation? Ann Surg. 2002;236(4):492-500, discussion
requiring re-repair for symptom recurrence at long-term 500–501.
12. Luketich JD, Nason KS, Christie NA, et al. Outcomes after a decade
(7 years) follow-up.10,12 We acknowledge that another 10% of laparoscopic giant paraesophageal hernia repair. J Thorac Cardiovasc
or so have small hiatal hernia recurrences, of which the Surg. 2010;139(2):395-404.
majority will be manageable without surgery. However, 13. Ballian N, Luketich JD, Levy RM, et al. A clinical prediction rule
we are not accepting these “small recurrences” as ideal, for perioperative mortality and major morbidity after laparoscopic
giant paraesophageal hernia repair. J Thorac Cardiovasc Surg.
and we are continuing to evaluate our surgical results and 2013;145(3):721-729.
hope to achieve “Pearson-like outcomes” of a less than 14. Frantzides CT, Madan AK, Carlson MA, Stavropoulos GP. A prospec-
2% reoperation rate at long-term follow-up.2 tive, randomized trial of laparoscopic polytetrafluoroethylene (PTFE)
patch repair vs simple cruroplasty for large hiatal hernia. Arch Surg.
2002;137:649-652.
SUMMARY 15. Oelschlager BK, Petersen RP, Brunt LM, et al. Laparoscopic para-
esophageal hernia repair: defining long-term clinical and anatomic
There are several key elements to laparoscopic repair outcomes. J Gastrointest Surg. 2012;16(3):453-459.
of PEH: (1) complete reduction of the hernia sac and 16. Fuller CB, Hagen JA, DeMeester TR, Peters JH, Ritter M, Bremmer
contents; (2) careful preservation of the anterior and CG. The role of fundoplication in the treatment of type II para-
esophageal hernia. J Thorac Cardiovasc Surg. 1996;111:655-661.
posterior vagus nerves; (3) mobilization of the gastro- 17. Davis RE, Awad ZT, Filipi CJ. Technical factors in the creation of a
esophageal fat pad and identification of the GEJ; (4) “floppy” Nissen fundoplication. Am J Surg. 2004;187(6):724-727.
recognition and management of a shortened esophagus 18. O’Reilly MJ, Mullins SG, Saye WB, et al. Laparoscopic posterior
(extensive mediastinal mobilization and performance of partial fundoplication: analysis of 100 consecutive cases. J Laparoendosc
a Collis gastroplasty when necessary); (5) preservation of Surg. 1996;6(3):141-150.
19. el-Sherif AE, Adusumilli PS, Pettiford BL, et al. Laparoscopic clam
crural integrity and closure of the hiatal defect without shell partial fundoplication achieves effective reflux control with
tension with liberal use of an induced pneumothorax reduced postoperative dysphagia and gas bloating. Ann Thorac Surg.
to yield a “floppy” diaphragm during repair; selective use 2007;84(5):1704-1709.
CHAPTER
Open Paraesophageal Hernia Repair
Daniel L. Miller
27
H
iatal hernia (HH) was first recognized more than of patients with PEH that will require surgical care will
400 years ago. In 1610 Ambrose Paré described increase significantly over time, so timing of surgical
a patient with the stomach herniating through correction and what approach and repair is paramount
the esophageal hiatus.1 Bowditch was the first to report to decrease operative morbidity and mortality and to
repair of an HH in 1853, and Akerlund first reported improve short- and long-term outcomes.
paraesophageal herniation in 1926.2,3 In 1945 Harrington
described the first series of patients who underwent HH
repair. 4 In 1951 Allison was the first to attribute the
INDICATION FOR SURGICAL REPAIR
symptoms of gastroesophageal reflux disease (GERD) and Surgical repair is indicated in patients with symptomatic or
acid ingestion to an HH and also described an anatomic complications of PEHs, the timing of which depends upon
repair.5 In 1968 Hill and Tobias were the first to clearly the acuity of presentation and significance of symptoms.
understand the anatomy (gastroesophageal junction [GEJ] Emergency repair is required in patients with acute gastric
and stomach) and clinical implications of paraesophageal volvulus (Fig. 27.3), uncontrolled gastrointestinal bleeding,
hernias (PEHs).6 obstruction, strangulation, perforation, or irreversible
HHs were first classified into three types by the Swedish respiratory comprise secondary to the PEH. As expected
radiologist, Ake Akerlund.2 The current classification a patient with a PEH that presents as an emergency and
scheme defines four types of hiatal or PEHs. The four requires surgical correction is associated with higher
types are classified based on the location of the GEJ mortality rates. 9 Elective repair is recommended in
and the fundus of the stomach in relationship to the patients with PEH who experience chronic symptoms
diagrammatic crura (hiatus). Type I (sliding) hernia is that are increasing in frequency and severity, such as
when the GEJ migrates into the chest only, type II (true) GERD refractory to medical therapy, dysphagia, early
is when the GEJ remains in the abdomen and the fundus satiety, postprandial chest or abdominal pain, postprandial
of the stomach herniates into the chest, type III (mixed) shortness of breath, aspiration, chronic anemia (Cameron
is the combination of type I and II with herniation of both erosions), or vomiting. Surgical repair of these patients
the GEJ and fundus into the chest, and type IV (complex) electively is associated with improved symptoms and
is when other abdominal viscera (colon [Fig. 27.1], small better quality of life (QoL).10 Prophylactic PEH repair
bowel, spleen, pancreas, or omentum) migrate into the (PEHR) in asymptomatic patients is controversial. There
chest with the GEJ and/or stomach. Type I, or sliding is no consensus, but traditionally most surgeons feel that
hernias, are the most common type and account for the very old or debilitated patients should not undergo
approximately 95% of all HHs, with the remaining three surgery, whereas younger and healthier patients with life
types (PEHs) making up the remaining 5% of HHs; 90% expectancy of 5 to 10 years should consider surgery to
of the PEHs are the type III mixed PEHs (Fig. 27.2), and prevent both risk of acute gastric volvulus, especially if
the rarest (<5%) are the type II true PEHs. greater than 50% of stomach in the chest, and potentially
An HH is characterized by enlargement of the opening progressive symptoms. In a recent study, the mortality rate
between the diaphragmatic crura, which allows the stomach from elective repair was estimated to be 1.4%, while the
and other abdominal viscera to elevate into the chest. The probability of developing acute symptoms that would neces-
cause of enlargement of the hiatus is related to increased sitate emergency surgery was 1.1%. Allen and colleagues
intraabdominal pressure creating a transdiaphragmatic followed 23 patients with large PEHs who refused surgery
pressure gradient between the thoracic and abdominal and preferred medical management, for a median of 78
cavities at the GEJ. This pressure gradient results in weak- months (range, 12 to 268 months).11 In four patients,
ness of the phrenoesophageal membrane and widening of progressive symptoms developed, and one patient died
the esophageal hiatus. Conditions that cause an increase from aspiration. They concluded that patients with an
in intraabdominal pressure include obesity, pregnancy, intrathoracic upside-down stomach who have obstructive
chronic constipation, chronic obstructive pulmonary symptoms at initial presentation should undergo repair
disease (COPD) with chronic coughing, and strenuous and that elective operation is safe and effective. However,
jobs with significant amount of lifting. Aging is also a gastric strangulation is extremely rare. The lifetime risk of
significant risk factor for development of PEH. PEH mainly developing acute symptoms requiring emergency surgery
affects older adults, with the median age of presentation decreases exponentially with age older than 65 years.
between 65 and 75 years of age.7 The US population, older
than 65 years of age, increased from 13% to 17% between
2000 and 2015 and is predicted to increase to 24% by
PREOPERATIVE EVALUATION
2060.8 With this increase in the elderly population and Detailed history, physical exam, and endoscopic and
the rise of obesity within the United States, the incidence radiographic evaluation are warranted in patients with
291
Open Paraesophageal Hernia Repair CHAPTER 27 291.e1
ABSTRACT
The incidence of patients developing a hiatal hernia
continue to increase in the United States. With the increase
in the elderly population and the rise of obesity, the
incidence of patients with a paraesophageal hernia (PEH)
that will require surgical care will increase significantly over
time, so timing of surgical correction and what approach
and repair is paramount to decrease operative morbidity
and mortality and to improve short-term and long-term
outcomes. Surgical repair is indicated in patients with
symptomatic or complications of PEH, the timing of which
depends upon the acuity of presentation and significance
of symptoms. Patient evaluation should include a detailed
history (review old operative reports) and physical examina-
tion, and endoscopic and radiographic evaluation are
warranted in patients with PEHs to maximize the best
surgical treatment, approach, and type of procedure. PEHs
can be repaired transabdominally or transthoracically. Even
though laparoscopic approach has become the standard
approach for elective and emergent PEH repair (PEHR),
an open transabdominal or transthoracic approach should
also be in a surgeon’s surgical armamentarium to offer to
patients with complex type IV PEHs who have failed mul-
tiple abdominal PEHR, prior mesh use, or with significant
obesity. Open procedures have similar associated morbidity
and mortality as laparoscopic PEHR. Conversion to an
open approach, transabdominal for general surgeons or
transthoracic for thoracic surgeons, is warranted to improve
outcomes in patients with a complex or recurrent PEH.
KEYWORDS
Paraesophageal hernia
GERD (gastroesophageal reflux disease)
hernia sac
hiatus
fundoplication
laparoscopy
laparotomy
thoracotomy
292 SECTION I Esophagus and Hernia
Hashemi and colleagues demonstrate a 42% recurrence a chest tube is placed laterally above the diaphragm at
rate for the laparoscopic approach as opposed to 15% completion of the repair prior to closure. The peritoneal
for the open approach.13 Open transabdominal approach covering of the crus on the abdominal side is preserved
is used in patients who have had a limited number of when dividing the gastrohepatic omentum from the right
upper abdominal procedures in the past, and reserve the crus of the diaphragm to improve stability of the hiatus
transthoracic approach for patients who have failed previ- tightening. A Penrose drain is used and placed around
ous transabdominal procedures, a history of abdominal the esophagus at the GEJ to elevate the esophagus and
wall mesh, history of abdominal abscess, infection, and stomach to improve dissection of the posterior hiatus.
contamination, and significant elevated body mass index Short gastric vessels are divided with an energy device
(BMI) (>40). The three operative approaches have not (bipolar) to allow complete mobilization of the stomach
been compared with one another in randomized trials, into a normal configuration and for facilitation of the
and the optimal operative approach for PEHR remains planned fundoplication.
controversial and varies most depending on surgeon
training and experience.14 Esophageal Mobilization
In an analysis of approximately 40,000 patients from Distal mobilization of the esophagus and GEJ into the
1999 through 2008 from the Nationwide Inpatient Sample abdomen with sufficient length (4 to 5 cm) of intra-
(NIS) database, 74%, 17%, and 9% were performed open abdominal esophagus is essential for a tension-free PEHR
transabdominally, transthoracically, and laparoscopically, and reduction of recurrence. Energy-assisted intrathoracic
respectively.15 Currently, laparoscopic PEHR has surpassed dissection is warranted to prevent injury to associated
open transabdominal repair as the most commonly per- anatomic structures and more importantly to minimize
formed procedure for PEHs. In the NIS study, transthoracic the injury to the vagus nerves. This dissection is usually
approach was associated with the longest hospital stay (7.8 carried up to the level of the aortic arch to allow for
days), the greatest need for mechanical ventilation (5.6%), a tension-free esophagus. If adequate intraabdominal
and the greatest risk of having pulmonary embolism.15 esophagus cannot be achieved, a lengthening procedure
Laparoscopic approach was associated with the shortest is required; a true shortened esophagus is rare, usually
hospital stay (4.5 days) and the lowest risk of requiring less than 5% of patients. Chronicity of the PEH may
mechanical ventilation (2.3%). In a second study using lead to a higher incidence of a shortened esophagus.
the NIS database, published in 2017, 63,800 patients were Lengthening of the esophagus is performed by a Collis
analyzed from 2000 through 2013 to assess the effect of gastroplasty.17 The Collis procedure creates a gastric tube
minimally invasive PEH surgery (MIS PEH) on patient by vertically stapling the proximal stomach from the
outcomes.16 Abdominal approach was used in 94.2% of angle of His, parallel to a large bougie, 48 to 51 French,
patients (67.1% laparoscopically and 32.9% open) and positioned along the lesser curvature of the stomach (Fig.
5.8% via the thoracic approach (24.5% thoracoscopi- 27.4). The neoesophagus is an elongated gastric tube,
cally and 75.5% open). Patients undergoing MIS PEH thus creating an extension of the esophagus that allows
experienced shorter hospital stay and decreased overall for the new esophagogastric junction to be greater than
cost. Long-term outcome data are not known in these NIS 4 cm in the abdomen. The Collis procedure was originally
studies. Other uncontrolled studies suggest that morbidity performed via a left thoracotomy, but more recently it has
and mortality rates appear lower for laparoscopic PEHR been performed transabdominally via open approach or
compared with the other approaches. Although the risk laparoscopically.18 The open or laparoscopic technique
of radiographic recurrence is higher with laparoscopic is called a wedge Collis gastroplasty because a wedge
approach, reoperation rates are similar. of fundus is resected to allow for vertical placement of
the endoscopic stapler parallel to the lesser curvature,
TRANSABDOMINAL REPAIR thus creating an elongated intraabdominal esophagus
An open or laparoscopic PEHR involves the same sequence (Fig. 27.5).
of steps and principles of repair. An open transabdominal
incision is usually performed via an upper abdominal Closure of Hiatus
incision from the xiphoid to just above the umbilicus. At After complete mobilization of the esophagus, the crura
times to facilitate further hiatal exposure the upper portion of the diaphragm are closed posteriorly to the esophagus.
of the incision is extended to the left of the xiphoid. An The closure of the hiatal defect is one of the most crucial
upper hand retractor is preferred, which is connected to steps in repair of a PEH. The repair must be tension free.
the bed bilaterally and is used instead of circumferential The repair can be performed primarily, with a patch
incisional retractor to allow elevation of the foregut for only, or a combination of primary repair and patch
maximal exposure of the hiatal anatomy. Sequence of reinforcement. Primary closure is usually performed
open surgical steps will now be described in detail. with nonabsorbable sutures, usually 3 to 5, depending on
the size of hiatal defect, in an interrupted fashion; some
Dissection of Hiatus surgeons prefer pledgeted horizontal mattress sutures
To prevent reherniation after PEHR, complete dissection because the cura have no fascial layer. If the crural fibers
and removal of the hernia sac from the mediastinum is are disrupted during the dissection or the primary repair
mandatory. This dissection should be performed meticu- is under tension, the crural closure can be reinforced with
lously to avoid injury to mediastinal pleural, pericardium, biologic mesh, such as porcine dermal matrix or bovine
aorta, and vagal nerves. If a pneumothorax occurs, it is pericardium. In addition, new bioresorbable materials are
not a hemodynamic issue during an open approach and being evaluated and may prove useful at the hiatus to help
294 SECTION I Esophagus and Hernia
A B C
FIGURE 27.4 Sequence of surgical maneuvers for creation of Collis-Nissen gastroplasty for a shortened esophagus. (A) Placement of
60-mm linear staple parallel to an esophageal bougie (51 French) along the lesser curvature of the stomach. (B) Successful firing of the
linear stapler with creation of the neoesophagus. The fundus of the stomach is being wrapped around the neoesophagus posteriorly.
(C) Completion of the Nissen fundoplication below the diaphragm with primary closure of hiatus.
120 mm
120 mm
120 mm
FIGURE 27.6 Recurrent type III paraesophageal hernia (PEH)—status post-mesh reinforcement. Recurrent type III PEH after mesh
reinforcement. (A) Axial computed tomography (CT) scan tomogram of patient showing a recurrent PEH after undergoing laparoscopic
PEH repair (PEHR) with synthetic mesh reinforcement. (B) Coronal CT scan tomogram of patient showing a recurrent PEH after
undergoing laparoscopic PEHR with synthetic mesh reinforcement. (C) Coronal CT scan tomogram of patient showing metal surgical
screws used to secure mesh during hiatus reinforcement.
we perform an anterior partial (Dor) fundoplication to symptoms at 3 and 12 months and significantly lower rates
minimize possible worsening dysphagia postoperatively. By of postoperative esophagitis, 17% versus 53%.22 Other
adding a fundoplication you will reduce the incidence of postoperative complications such as dysphagia and gas
postoperative GERD symptoms by restoring competency bloating did not differ between the two groups.
to the lower esophageal sphincter (LES). We routinely
secure the fundoplication, partial or full, to the anterior Postoperative Management
portion of the hiatus to complete the closure of the defect Patients are admitted to our surgical floor after PEHR.
and hopefully reduce the possibility of recurrence. The Intraoperative orogastric tubes are removed at the comple-
total fundoplication usually measures 2.0 to 2.5 cm in tion of the PEHR in the operating room (OR). Patients
width to help prevent postoperative gas bloat syndrome are given antiemetics for the first 24 hours postoperatively
and is anchored to the esophagogastric junction after to reduce the risk of postoperative nausea and vomiting,
removal of the esophageal fat pad to prevent slippage of which can disrupt the PEHR and cause early recurrence.23
the fundoplication. We have the patient ambulate the night of surgery, as well
In a randomized study, 40 patients who underwent as start chewing gum 3 times a day for 20-minute intervals.
either a cardiophrenicopexy (suture the cardia of the We do not routinely perform a barium swallow the morning
stomach to the diaphragm) or total fundoplication at the after surgery. We allow the patients to start sips of clear
time of PEHR were analyzed. Patients who underwent liquids 24 hours after surgery. The patient’s chest and
fundoplication had significantly fewer postoperative GERD abdomen with respect to reherniation and gastric motility
296 SECTION I Esophagus and Hernia
A B
FIGURE 27.7 Type IV paraesophageal hernia (PEH) in patient with situs inversus. Large type IV PEH containing stomach, colon, and small
bowel in patient with situs inversus. (A) Coronal computed tomography (CT) scan tomogram of patient showing a large type IV PEH
containing stomach, colon, and small bowel in patient with situs inversus. (B) Axial CT scan tomogram of patient showing a large type IV
PEH containing stomach, colon, and small bowel in patient with situs inversus.
Fat pad
Hernia excision
Fat
pad
4. Initial row
of sutures tied,
creating 240
fundoplication 6. Stomach passed into
abdomen and second
row of sutures tied
5. Second row of
sutures placed in
similar fashion and
passed through
diaphragm
Hiatal
closure
FIGURE 27.8 Belsey Mark IV repair. (Netter illustration used with permission of Elsevier, Inc. All rights reserved.)
298 SECTION I Esophagus and Hernia
sutures is secured below the diaphragm and tied to the and were older than 80 years of age, operative mortality
anterior portion of the hiatus. The Belsey repair is an was 8.2%.28 Emergent repair, which was performed in
excellent long-term functional fundoplication to prevent 43% of those patients, was associated with a much higher
GERD symptoms, recurrence, and dysphagia.26 mortality rate compared with the elective repair patients,
15.7% versus 2.4%, respectively, and was the only predictor
Hiatus Repair of mortality in a multivariate analysis. In the two largest
The hiatal closure is performed posteriorly with interrupted series of open transthoracic PEHR, 94 and 240 patients,
nonabsorbable sutures in an interrupted figure-of-eight the postoperative morbidity and mortality rates were 1.7%
configuration, usually 4 to 5 sutures. The hiatal sutures and 2.1%, respectively.29,30
are placed first but not tied until the fundoplication Older patients have an increased incidence of PEH and
transdiaphragmatic sutures are reduced and tied. Sutures can be denied surgical evaluation due to the perception
can be removed if the closure is too tight and could cause of increased complications and mortality. El Lakis and col-
dysphagia. If a myotomy is performed at the time of an leagues examined the influence of age and comorbidities
open transthoracic PEHR for preoperative esophageal on early complications and other short-term outcomes
dysmotlity, then the center sutures from each row are of PEHR.31 The study was a retrospective review of 524
eliminated with continued excellent GERD symptom patients who underwent PEHR for PEHs with more than
prevention. Reinforcement material is not needed in an 50% of stomach involved and were stratified by age (<70,
open transthoracic repair, which significantly reduces the 70 to 79, ≥80 years of age) to determine outcomes. Patients
intraoperative cost. The chest is drained with a single 80 years of age or older had higher American Society
28-French chest tube that can cross the midline and drain of Anesthesiologists (ASA) class, more comorbidities,
the right chest if the pleura was entered during dissection larger PEHs, and higher incidence of type IV PEHs and
of the hernia sac. acute presentation. Patients 80 years of age or older had
more postoperative complications but not more major
Postoperative Management complications; hospital stay was 1 day longer. There was
Patients are again admitted to our surgical floor after no difference in recurrence of their PEH based on age
PEHR. Intraoperative oral gastric tubes are removed of PEHR. After adjustment for comorbidities and other
at the completion of the PEHR in the OR. Patients are factors, age 80 years or older was not a significant factor
given antiemetics for the first 24 hours postoperatively to in predicting severe complications, readmissions within
reduce the risk of postoperative nausea and vomiting. The 30 days, or early recurrence. Older patients with giant
patient ambulates the night of surgery, as well as starts PEHs should be given the opportunity to be evaluated
chewing gum 3 times a day for 20-minute intervals. We by experienced surgeons for PEHR. Advanced age is no
allow the patients to start sips of clear liquids 24 hours longer a contraindication for PEHR.
after surgery. The patient’s chest and abdomen with
respect to reherniation and gastric motility is assessed Efficacy
by daily CXRs. We do not use epidural catheters in our Regardless of approach, PEHRs have been shown to
open transthoracic PEHR patients because of issues with significantly alleviate preoperative symptoms related to
medication-induced hypotension in the elderly and Foley- the PEH. A prospective study of 111 patients found lapa-
related issues in elderly men. Postoperative pain manage- roscopic PEHR to be associated with improved symptoms
ment after an open transthoracic PEHR is controlled by and better QoL at 1 and 3 years after surgery. 32 The
intraoperative parasympathetic injections (T2 to T11), as overall QoL scores at 3 years were 50% higher than the
well as full-thickness latissimus and serratus muscle blocks baseline scores. In a study of 72 patients who underwent
with 300 mL of diluted Exparel. Patients are discharged an open transabdominal PEHR, symptoms were improved
from the hospital usually 4 to 6 days after surgery once compared with baseline.14 The postoperative Short Form-36
bowel function returns and the patient is tolerating a (SF-36) scores were higher than the general population
low-residue diet. Aggressive pulmonary hygiene is carried in six of eight categories and higher than age-matched
out in these patients, including incentive spirometry, population in eight of eight categories. In two large open
flutter valve, and early ambulation. After switching to transthoracic PEHR series with follow-up ranging from
the regional anesthesia blocks with Exparel, respiratory 42 to 92 months, 80% to 86% of patients were symptom
complications have decreased significantly. free and reported satisfactory results with the surgery.29,30
Recurrence
PATIENT OUTCOMES Interpretation of studies reporting recurrence rates are
Morbidity and Mortality limited by different outcome measures used. The rate of
Laparoscopic PEHR is associated with low overall morbid- radiographic recurrence is higher than that of clinical
ity (<3%) and mortality (<2.0%).7,27 The mortality and recurrence. Most patients with a radiographic recurrence
morbidity rates are higher in patients who are 70 years after PEHR are asymptomatic, whereas patients with
of age or older, who underwent an emergent procedure, clinical recurrence can usually have symptoms controlled
and those patients with multiple comorbid conditions. As with medication. Only a small percentage of patients
expected, open transabdominal and transthoracic PEHR will require re-PEHR for complications or intractable
are associated with higher morbidity and mortality rates, symptoms. Laparoscopic PEHRs have a higher radiologic
but not as high as expected. In a retrospective study of 1005 recurrence rate than the two open approaches, but most
patients who underwent an open transabdominal PEHR of these recurrences are clinically silent and do not require
Open Paraesophageal Hernia Repair CHAPTER 27 299
25. Belsey R. Mark IV repair of hiatal hernia by the transthoracic 30. Mazaiak DE, Todd TR, Pearson FG. Massive hiatus hernia; evaluation
approach. World J Surg. 1977;1:475-481. and surgical management. J Thorac Cardiovasc Surg. 1998;115:53-62.
26. Allen MS. Open repair of hiatus hernia: thoracic approach. Chest 31. El Lakis MA, Kaplan SJ, Hubka M, Mohiuddin K, Low DE. The
Surg Clin N Am. 1998;8:431-440. importance of age on short-term outcomes associated with repair of
27. Larusson HJ, Zingg U, Hahnloser D, et al. Predictive factors for giant paraesophageal hernias. Ann Thorac Surg. 2017;103:1700-1709.
morbidity and mortality in patients undergoing laparoscopic para- 32. Lidor AO, Steele KE, Stem M, et al. Long-term quality of life and risk
esophageal hernia repair: age, ASA score and operation type influence factors for recurrence after laparoscopic repair of paraesophageal
morbidity. World J Surg. 2009;33:980-985. hernia. JAMA Surg. 2015;150:424-431.
28. Poulose BK, Gosen C, Marks JM, et al. Inpatient mortality analysis 33. Carlson MA, Condon RE, Ludwig KA, Schulte WJ. Management of
of paraesophageal hernia repair in octogenarians. J Gastrointest Surg. intrathoracic stomach with polypropylene mesh prosthesis reinforced
2008;12:1888-1892. transabdominal hiatus repair. J Am Coll Surg. 1998;187:227-230.
29. Patel HJ, Tan BB, Ye J, et al. A 25-year experience with open primary 34. Ferri LE, Feldman LS, Stanbridge D, et al. Should laparoscopic
transthoracic repair of paraesophageal hernia repair. J Thorac paraesophageal hernia repair be abandoned in favor of the open
Cardiovasc Surg. 2004;127:843-849. approach? Surg Endosc. 2005;19:4-8.
CHAPTER
Diaphragmatic Relaxing Incisions for Crural Tension
During Hiatal Hernia Repair 28
Marc A. Ward
| Steven R. DeMeester
L
aparoscopic repair of a large hiatal hernia with a accommodate without tearing the stitches out. In fact, the
widened hiatus is challenging and objective hernia large amount of shrinkage that occurs with PTFE mesh
recurrence rates are high. In a recent randomized may be beneficial in restoring a more natural contour
trial, the recurrence rate exceeded 50% at 5 years after to the diaphragm over time after a relaxing incision.
laparoscopic paraesophageal hernia.1 Tension on the Absorbable or biologic mesh should be avoided, since it
repair of any hernia contributes to an increased risk for will not permanently repair the defect.
recurrence. This was first recognized for inguinal hernias Both right- and left-sided relaxing incisions are rela-
and led to the incorporation of relaxing incisions and tively straightforward laparoscopic procedures, provided
tension-free repairs. Subsequently component separation that important landmarks are recognized. In our study,
has been adopted to release tension during ventral hernia none of our patients suffered complications or paralyzed
repair.2–4 Logically this concept can be extended to the diaphragms related to the relaxing incisions.8 Furthermore,
hiatus for patients with large crural defects where achieving it is important to recognize that a wide communication
primary approximation of the crural pillars would be between the abdomen and thorax during a laparoscopic
difficult or impossible without significant tension. Similar procedure with 15 mm Hg carbon dioxide capnoperito-
to the tension-free inguinal hernia repair, bridging the neum is well tolerated. Small pleural holes can lead to
crural defect with mesh has been considered. However, a ball-valve effect and tension pneumothorax, but large
bridging with synthetic mesh has been associated with openings are not associated with such problems. In fact,
mesh erosion into the esophagus and is ill-advised (Fig. large openings of the pleura are standard when we repair
28.1).5 Bridging with a biologic mesh is only a temporary large hiatal defects, even without a relaxing incision. This
solution since the bridged biologic or absorbable material allows the mediastinal space to drain into the right, left, or
will go away, setting the stage for hernia recurrence. Park both pleural cavities once the hernia has been repaired.
et al. proposed using the mobilized falciform ligament to After the pleura is opened or a relaxing incision has been
bridge the hiatal defect, but the long-term efficacy of this created, I do not routinely leave drains in the mediastinum
approach is unknown.6 In our opinion a better alternative or chest unless a lung injury with air leak was created or
is a relaxing incision in either the right or left diaphragm, suspected during the dissection.
and occasionally in both.
The concept of a relaxing incision is to create a defect
adjacent to the area of interest that is less critical (muscle
IMPORTANT INITIAL MANEUVERS
or fascia) to allow the principal tissues to come together. The first step to allow crural movement and approximation
In the case of the hiatus, relaxing incisions are ideal since is to ensure that the posterior sac has been well dissected
there is available diaphragm on either side. In addition, off of the left crus. When the sac remains attached to the
there is little (if any) downside to making a relaxing incision left crus, it can restrict movement and reapproximation of
in the diaphragm. Because there is a pressure differential the crura. The second step is to create a left pneumothorax.
between the thorax (negative pressure) and the abdomen This will equilibrate the capnoperitoneum on both sides
(positive pressure), all relaxing incisions must be closed. of the diaphragm, allowing the left diaphragm to become
Even small diaphragmatic defects can lead to herniation floppy and facilitating its movement toward the right crus.
of abdominal contents into the chest.7 We routinely close Just this maneuver alone is enough to reduce measured
these defects with polytetrafluoroethylene (PTFE), since tension during crural approximation by 35.8%, as shown
there is a large experience with this mesh for chest wall in an elegant study by Bradley and colleagues, where
and diaphragm reconstruction. Unlike other types of a tensometer was used intraoperatively to evaluate the
synthetic mesh, the lung does not typically fuse to PTFE effect of tension-reducing techniques.9 If there is still
mesh, so future thoracic surgical procedures, if needed, excessive tension when trying to approximate the crura
will not be made more complex. While there is concern despite these maneuvers, a relaxing incision is the next
about the use of PTFE mesh in a clean-contaminated recommended step.
case, six of our patients in our published series had a
concomitant wedge fundectomy Collis gastroplasty, and
no patient developed an infection of the PTFE mesh used TECHNIQUE FOR RIGHT-SIDED
in the reconstruction.8 Unlike ventral hernia repairs with RELAXING INCISION
mesh, the use of simple interrupted sutures to sew the
PTFE mesh to the edges of the defect with no overlap The right-sided relaxing incision is straightforward and
works well on the diaphragm. This is likely possible because adds only 15 to 30 minutes to the overall procedure. It is
as the mesh contracts, the diaphragm has enough give to the preferred release, but the right crus must be at least a
301
Diaphragmatic Relaxing Incisions for Crural Tension During Hiatal Hernia Repair CHAPTER 28 301.e1
ABSTRACT
It has been shown that the risk of recurrence following any
hernia repair decreases significantly when completed under
tension. To reduce tension at the site of repair, component
separation and relaxing incisions have been adopted
for ventral and inguinal hernias, respectively, which has
led to a reduction in hernia recurrence following these
operations. Logically, this concept can also be applied to
large hiatal hernias, in which tension-free approximation
of the crural pillars would be difficult without additional
interventions. Diaphragmatic relaxing incisions adjacent
to the crura allow the primary hiatal defect the ability to
come together without tension. They can be done on either
the right or the left side of the diaphragm, depending on
the patient’s anatomy. Once sufficient release has been
achieved to allow closure of the hiatus, the diaphragmatic
defect can be reconstructed. The use of synthetic mesh at
the hiatus is not recommended given its association with
mesh erosion into the esophagus. However, the defects
following diaphragmatic relaxing incisions are away from
the esophagus; thus use of synthetic mesh to close these
defects is advised, since even small diaphragmatic defects
can result in herniation of abdominal contents into the
chest. The techniques for right- and left-sided relaxing
incisions reduce tension at the hiatus and should be in
the skill set of surgeons who repair paraesophageal or
large sliding hiatal hernias.
KEYWORDS
Paraesophageal hernia; diaphragmatic relaxing incisions;
PTFE mesh; tension-free repair; hiatal hernia
302 SECTION I Esophagus and Hernia
FIGURE 28.1 The use of synthetic mesh near the diaphragmatic hiatus can result in erosion of the mesh into the esophagus, as seen in
this figure.
IVC H/P
DRI
RC
LC
LC
PN
DISCLOSURES
Dr. DeMeester is a consultant for Gore and Davol/Bard, and
has received a research grant from Davol/Bard.
REFERENCES
FIGURE 28.5 A left relaxing incision is being closed with 2 mm
1. Oelschlager B, Pellegrini C, Hunter J, et al. Biologic prosthesis to
PTFE mesh.
prevent recurrence after laparoscopic paraesophageal hernia repair:
long-term follow-up from a multicenter, prospective, randomized
trial. J Am Coll Surg. 2011;213:461-468.
2. Read RC. The contributions of Usher and others to the elimination
left in the pleural space or mediastinum, and no patient of tension from groin herniorrhapy, 2004. Hernia. 2004;9:208-211.
developed a pleural effusion prior to discharge. On follow- 3. Read RC. Herniology: past, present and future. Hernia. 2009;13:577-580.
up chest x-ray, three patients had developed a pleural 4. Halvorson EG. On the origins of components separation. Plast Reconstr
effusion and two patients required pigtail drainage. An Surg. 2009;124:1545.
asymptomatic mild elevation of the left hemidiaphragm 5. Stadlhuber R, Sherif A, Mittal S, et al. Mesh complications after
prosthetic reinforcement of hiatal closure: a 28-case series. Surg
was seen in one patient. At short-term follow-up, no patient Endosc. 2009;23:1219-1226.
had required reoperation, and no patient developed 6. Park A, Hoogerboord C, Sutton E. Use of the falciform ligament flap
a diaphragmatic hernia through the repaired relaxing for closure of the esophageal hiatus in giant paraesophageal hernia.
incision or evidence of PTFE mesh infection. Objective J Gastrointest Surg. 2012;16(7):1417-1421.
7. Crespin OM, Yates RB, Martin AV, Pellegrini CA, Oelschlager BK.
follow-up in 11 of the 15 patients showed that the repair The use of crural relaxing incisions with biologic mesh reinforcement
was intact in all but one patient who had an asymptomatic, during laparoscopic repair of complex hiatal hernias. Surg Endosc.
trivial (<2 cm) recurrent hernia at 4.5 months.8 2016;30(6):2179-2185.
8. Greene CL, DeMeester SR, Zehetner J, Worrell SG, Oh DS, Hagen JA.
Diaphragmatic relaxing incisions during laparoscopic paraesophageal
CONCLUSION hernia repair. Surg Endosc. 2013;27(12):4532-4538.
9. Bradley DD, Louie BE, Farivar AS, Wilshire CL, Baik PU, Aye RW.
A crural relaxing incision will allow crural repair with Assessment and reduction of diaphragmatic tension during hiatal
reduced tension in patients with a wide hiatus. Since hernia repair. Surg Endosc. 2015;29(4):796-804.
CHAPTER
Collis Gastroplasty for a Foreshortened Esophagus
Stephanie G. Worrell
| Joshua A. Boys
| Steven R. DeMeester
29
N
ormally, several centimeters of the distal esophagus loss of elasticity in the longitudinal esophageal muscle
and the gastroesophageal junction (GEJ) lie below related to chronic loss of intraabdominal fixation of the
the hiatus within the abdomen. When the GEJ, the GEJ. Although any of these histories should increase the
fundus of the stomach, or both migrate into the chest suspicion that a patient may have a short esophagus, none
above the hiatus, a hiatal hernia is present. Intrinsic to are definitive. Objective studies in patients with GERD or
the repair of a hiatal hernia is the need to bring the GEJ, PEH are useful to define the size, type, and reducibility
stomach, and distal esophagus back into the abdomen. of any hiatal hernia, presence of a stricture or erosive
However, since 1950 it has been known that in some esophagitis, esophageal function, and the presence and
patients this can be challenging, particularly those with severity of increased esophageal exposure to refluxed
severe gastroesophageal reflux disease (GERD) or a large gastric juice. A foreshortened esophagus can effectively
hiatal hernia. In these patients esophageal shortening can be ruled out when a hiatal hernia fully reduces on barium
lead to loss of intraabdominal esophageal length and put esophagram, but in any nonreducing hiatal hernia a
tension on the repair of a hiatal hernia. Dr. John. Leigh short esophagus may be present. Therefore, although
Collis described a technique in 1957 to address acquired objective studies can rule out a short esophagus, none can
esophageal shortening.1 His technique, now referred accurately identify its presence. Instead, a foreshortened
to as a Collis gastroplasty, creates an extension to the esophagus can be confirmed only by the intraoperative
esophagus from the high lesser curvature of the stomach. inability to reduce the GEJ below the hiatus by 2 to 3 cm
His gastroplasty was done as a transthoracic procedure. after mediastinal esophageal mobilization and posterior
Subsequently, several techniques have been described to crural closure.
create a similar gastroplasty using a laparoscopic approach.
However, more than 50 years after Dr. Collis described
his procedure for lengthening the esophagus there is
MANAGEMENT OF THE SHORT ESOPHAGUS
still controversy about the existence and prevalence of a Failure to obtain an adequate length of intraabdominal
foreshortened esophagus. Furthermore, the laparoscopic esophagus during hiatal hernia repair has been proposed
management of a short esophagus is challenging, and as as a leading cause for reherniation and breakdown of
a result there is a tendency by many surgeons to ignore the repair.3 Mediastinal mobilization and posterior crural
esophageal length and proceed with a standard repair. closure, particularly in a kyphotic patient, are routinely
Importantly, tension is the enemy of any hernia repair, and used to add esophageal length. To accomplish a fundo-
long-term successful outcomes with hiatal hernia repairs, plication without tension there should be 2 to 3 cm of
as for all other abdominal hernias, require addressing intraabdominal esophagus below the hiatal closure. The
tension when encountered. amount of intraabdominal esophagus during laparoscopic
surgery is deceptive because the pneumoperitoneum
artificially elevates the diaphragm and gives the appearance
IDENTIFYING THE SHORT ESOPHAGUS of more esophageal length than what is actually present.
Patients at risk for acquired esophageal shortening include With deflation of the pneumoperitoneum, the diaphragm
those with advanced GERD with esophagitis, stricture, descends and some of the apparent esophageal length is
long-segment Barrett esophagus, a history of sarcoidosis, lost. If standard methods for esophageal mobilization are
caustic ingestion, or scleroderma and those with a large insufficient to provide 2 to 3 cm of abdominal esophagus,
sliding or paraesophageal hernia (PEH).2,3 In some reports esophageal lengthening is recommended.
patients with a PEH have the highest frequency of a short There are several methods to accomplish a Collis
esophagus.4 The presence of a foreshortened esophagus gastroplasty during a laparoscopic procedure, including
in patients with severe GERD is understandable because advancing a linear stapler through a port in the thorax and
exposure to refluxed gastric juice causes mucosal injury using a circular stapler to make a hole in the stomach and
and can lead to transmural inflammation, fibrosis, and then completing the gastroplasty with a linear stapler.5,6 Our
collagen contraction. An esophageal stricture is strongly preferred approach was described by Terry and colleagues
associated with a shortened esophagus and the need for and is the wedge fundectomy Collis gastroplasty (WFCG)
a gastroplasty. The presence of both a large hiatal hernia technique.7 The WFCG was created with a 52-French
(>5 cm) and an esophageal stricture further increases bougie in place using a 45-mm Endo GIA blue load stapler.
the risk of a shortened esophagus.2 In addition, a history The goal was to excise as small a wedge of fundus as
of a previous failed antireflux procedure with recurrent possible. Given the limitations of articulating Endo GIA
hiatal hernia should raise suspicion that the length of the staplers, we found that to excise only a small portion of the
esophagus is short. The etiology of esophageal shortening fundus it was necessary to create a starfish-shaped piece
in patients with a PEH is unclear but may be related to of the proximal fundus by successively cutting through
305
Collis Gastroplasty for a Foreshortened Esophagus CHAPTER 29 305.e1
ABSTRACT
Over 50 years after Dr. John Leigh Collis described
his procedure for lengthening the esophagus there is
still controversy about the existence and prevalence of
a foreshortened esophagus. Further, the laparoscopic
management of a short esophagus is challenging, and as
a result there is a tendency by many surgeons to ignore
esophageal length and proceed with a standard repair.
Importantly, tension is the enemy of any hernia repair, and
long-term successful outcomes with hiatal hernia repairs,
as for all other abdominal hernias, require addressing
tension when encountered. This chapter will address
the role for, techniques to perform, and outcomes with
a Collis gastroplasty for the foreshortened esophagus.
KEYWORDS
Collis gastroplasty, paraesophageal hernia, short esophagus
306 SECTION I Esophagus and Hernia
the inferior staple line with each successive staple load neoesophagus as possible without inducing excessive
until a mark approximately 3 cm below the angle of tension on the repair.5 It is also possible that the degree
His was reached. The staple line was not reinforced but of shortening in our patients was less than that in the
was buried by the fundoplication. A partial Toupet or series by Jobe et al., because in patients with a very short
complete Nissen fundoplication was added to the WFCG esophagus, the Collis gastroplasty can extend above the
in all patients. Importantly the fundoplication was kept as hiatus. In that circumstance it is not possible to position the
high on the gastroplasty as possible, preferably at the top fundoplication at the top of the gastroplasty. Importantly,
near the GEJ. The importance of this is the fact that the esophagitis in these patients is often asymptomatic. Con-
gastroplasty is made from stomach, and acid production sequently, we recommend that at least one postoperative
by the gastroplasty above the fundoplication can lead to endoscopy be done after a Collis gastroplasty to evaluate
erosive esophagitis in some patients, particularly if there for esophagitis. If esophagitis is found in the setting of
are several centimeters of gastroplasty above the fundoplica- an intact fundoplication, treatment with a proton pump
tion. It is also important to recognize that the gastroplasty inhibitor is recommended to prevent stricture formation
tube is aperistaltic. Therefore bolus transport through the or other complications related to ongoing mucosal injury.
gastroplasty relies on the motility of the distal esophagus A transthoracic Collis gastroplasty has been associ-
above the gastroplasty. Consequently, we are more liberal ated with complications not typically seen with standard
with the use of a partial fundoplication in patients who antireflux surgery, including staple line leaks, abscesses,
have a WFCG added for a shortened esophagus. and fistulas.10 We are always careful to ensure adequate
perfusion of the Collis segment and would avoid a Collis
OUTCOME WITH A COLLIS GASTROPLASTY gastroplasty if there was any compromise of the lesser curve
blood supply due to interruption of the left gastric artery.
Before the introduction of laparoscopic surgery, most In our series of laparoscopic WFCGs we did not have any
antireflux procedures were performed in patients with of these complications. We routinely cover the Collis staple
severe GERD, often with impaired esophageal body func- line with the fundoplication to minimize the risk of a leak
tion. A Collis gastroplasty in these patients frequently led or fistula. Furthermore, the wedge fundectomy technique
to protracted postoperative dysphagia. In a series reported may lead to a wider and more robust portion of fundus
from our center in 1998 a transthoracic Collis gastroplasty that lessens the tension that was sometimes present with
in the presence of preoperative dysphagia was significantly a fundoplication after a traditional transthoracic Collis
associated with a poor postoperative outcome. Many of gastroplasty.
these patients had strictures and severe reflux disease.8 The key issue of course with a Collis gastroplasty is
The availability of potent acid-suppressing medications whether it reduces hernia recurrence rates. We recently
has led to a reduction in the acid-related complications reviewed our experience in 83 patients who had primary
of reflux disease including strictures. Furthermore, the laparoscopic PEH repair (manuscript submitted for publica-
number of patients presenting for elective repair of a PEH tion). In 46 patients (55%), we identified a short esophagus,
in the era of laparoscopic surgery is increasing. In these and these patients were given a WFCG. The remainder
patients a Collis gastroplasty seems to be better tolerated. had a fundoplication alone. At a median follow-up of 9
In contrast to our earlier series, a recent evaluation of months there was objective evidence of a 2-cm or greater
our laparoscopic Collis gastroplasties showed that severe recurrent hernia in two (5.4%) of the fundoplication-alone
reflux disease was less common.9 The Collis gastroplasty group compared with one (2.2%) in the WFCG group
was done in 72% of patients, either for a PEH or during (P = .583). Two of the three recurrent hernias were small
reoperation for a failed fundoplication. Dysphagia was (2 to 3 cm). The single large recurrent hernia developed
a common preoperative symptom; however, it resolved in a patient who had a fundoplication alone and required
in the majority (71%) postoperatively. Importantly, new- reoperation for recurrent symptoms. Based on these data,
onset dysphagia occurred in only two patients (5.5%) one could conclude that a Collis gastroplasty does not
and resolved after one endoscopic dilatation in both alter the frequency of hernia recurrence. However, an
patients. Dysphagia that was present preoperatively and alternative conclusion is that without a Collis gastroplasty,
persisted was typically mild and did not significantly impact patients with a short esophagus would have had a higher
the patient’s diet or lifestyle. The relief of dysphagia in recurrence rate. If true, then the finding of a similar
most patients was likely related to repair of the large recurrent hernia rate in patients deemed to have a short
hiatal hernia and healing of esophagitis. However, we esophagus who had a WFCG compared with those with
also attributed the low rate of new-onset dysphagia to our no esophageal shortening would suggest that addressing
“tailored approach” for a fundoplication, using a Toupet a short esophagus is warranted and improves outcomes.
rather than a Nissen in patients with manometric evidence The expected objective hernia recurrence rate after
of ineffective esophageal motility. laparoscopic PEH repair is known. The randomized trial
A second potential issue with a Collis gastroplasty is acid by Oelschlager and colleagues reported a greater than
production by the neoesophagus above the fundoplica- 50% hernia recurrence rate, and the use of biologic mesh
tion. In our recent series we found that the prevalence did not reduce the rate at 5-year follow-up.11 Recognizing
of esophagitis after laparoscopic Collis gastroplasty was this high failure rate, which we also reported in 2000, we
much lower (11%) than reported by others. It is not have modified our approach.12 It is likely that the high
clear why our prevalence was much less than the 36% recurrence rate is related to the inherent weakness of the
rate reported by Jobe et al., but it may in part be related crural tissue and to unaddressed tension on the repair.
to our efforts to keep the fundoplication as high on the Tension on the repair of any hernia is a harbinger for
Collis Gastroplasty for a Foreshortened Esophagus CHAPTER 29 307
failure. Consequently, we now address lateral tension on 2. Gastal OL, Hagen JA, Peters JH, et al. Short esophagus: analysis of
the crural closure with a diaphragm-relaxing incision predictors and clinical implications. Arch Surg. 1999;134(6):633-636,
discussion 637-638.
and axial tension from a short esophagus with WFCG. 3. Horvath KD, Swanstrom LL, Jobe BA. The short esophagus: patho-
Furthermore, we routinely reinforce the primary crural physiology, incidence, presentation, and treatment in the era of
closure with biologic or absorbable mesh. Using this laparoscopic antireflux surgery. Ann Surg. 2000;232(5):630-640.
approach we have excellent short-term outcomes with a 4. Herbella FAM, Del Grande JC, Colleoni R. Short esophagus: literature
incidence. Dis Esophagus. 2002;15(2):125-131.
very low objective hernia recurrence rate.13 5. Jobe BA, Horvath KD, Swanstrom LL. Postoperative function following
laparoscopic Collis gastroplasty for shortened esophagus. Arch Surg.
CONCLUSION 1998;133(8):867-874.
6. Luketich JD, Grondin SC, Pearson FG. Minimally invasive approaches
Patients found to have a short esophagus during laparo- to acquired shortening of the esophagus: laparoscopic Collis-Nissen
gastroplasty. Semin Thorac Cardiovasc Surg. 2000;12(3):173-178.
scopic hiatal hernia repair are likely at increased risk for 7. Terry ML, Vernon A, Hunter JG. Stapled-wedge Collis gastroplasty
breakdown of the repair and a recurrent hiatal hernia. for the shortened esophagus. Am J Surg. 2004;188(2):195-199.
The first steps to gain esophageal length are mediastinal 8. Ritter MP, Peters JH, DeMeester TR, et al. Treatment of advanced
esophageal mobilization and posterior crural closure. If gastroesophageal reflux disease with Collis gastroplasty and Belsey
partial fundoplication. Arch Surg. 1998;133(5):523-528, discussion
these steps are inadequate a Collis gastroplasty should be 528-529.
added. The wedge fundectomy technique allows esophageal 9. Zehetner J, DeMeester SR, Ayazi S, Kilday P, Alicuben ET, DeMeester
lengthening laparoscopically and is associated with a low TR. Laparoscopic wedge fundectomy for Collis gastroplasty creation
rate of complications. Clear-cut evidence that a laparo- in patients with a foreshortened esophagus. Ann Surg. 2014;260:
scopic Collis gastroplasty reduces hernia recurrence rates 1030-1033.
10. Patel HJ, Tan BB, Yee J, Orringer MB, Iannettoni MD. A 25-year
is lacking; however, tension on the repair of any hernia is experience with open primary transthoracic repair of paraesophageal
associated with an increased failure rate. Consequently, hiatal hernia. J Thorac Cardiovasc Surg. 2004;127(3):843-849.
a Collis gastroplasty in the setting of a foreshortened 11. Oelschlager BK, Petersen RP, Brunt LM, et al. Laparoscopic para-
esophagus is likely to prove beneficial in the long term esophageal hernia repair: defining long-term clinical and anatomic
outcomes. J Gastrointest Surg. 2012;16(3):453-459.
and should be part of the armamentarium of modern 12. Hashemi M, Peters JH, DeMeester TR, et al. Laparoscopic repair of
laparoscopic esophageal surgeons. large type III hiatal hernia: objective followup reveals high recurrence
rate. J Am Coll Surg. 2000;190(5):553-560, discussion 560-561.
REFERENCES 13. Alicuben E, Worrell SG, DeMeester SR. Impact of crural relaxing
incisions, Collis gastroplasty and non-cross-linked human dermal
1. Collis JL. An operation for haitus hernia with short esophagus. mesh crural reinforcement on early hiatal hernia recurrence rates.
J Thorac Surg. 1957;34:768-778. J Am Coll Surg. 2014;219(5):988-992.
CHAPTER
C
rural repair and reduction of hiatal hernia (HH) are hiatus, even for larger HHs. The goal of this chapter is to
paramount to the success of an antireflux surgery. briefly review the advantages and disadvantages of using
Case series from the early laparoscopic era (i.e., mesh at the hiatus.
the 1990s) reported an unacceptably high recurrence
rate of HHs. In addition, some patients were found to be HIATAL HERNIA: RECURRENCE
naturally predisposed to hernia formation, and recurring
hernias were thought to be due in part to inherent defects AND OUTCOMES
in healing. Simultaneously, the routine use of synthetic
mesh in inguinal and ventral hernia repair was gaining RECURRENCE RATES WITH PRIMARY CLOSURE IN
widespread acceptance and was found to be both safe THE EARLY LAPAROSCOPIC ERA
and effective. A 2004 review described a high recurrence rate of eight
These findings set the stage for some practitioners to HH case series published up to that point.8 The studies
advocate for the use of mesh at the hiatus. Soon thereafter, that were evaluated incorporated systematic radiographic
several case series described experience with various follow-up for laparoscopic PEH repair, and objective
types of synthetic mesh materials and configurations of follow-up was available for 277 of 460 patients. The mean
mesh at the hiatus, and these studies were strengthened overall reported recurrence rate was 27% (range, 7% to
by significantly lower recurrence rates in their series. 43%). In 2000 Hashemi et al.9 reported their experience
However, these early reports were followed by isolated with 54 patients who underwent surgery for repair of
case reports describing complications related to mesh use, large PEH at the University of Southern California, a
including catastrophic cases that required esophagogastric large and well-respected esophageal center. In the subset
resections.1 These disastrous complications were thought of patients who underwent laparoscopic repair of large
to result from the synthetic nature of the mesh materials PEH (n = 27), objective follow-up was available for 21
being used, so bioprosthetic materials were suggested as patients (78%). These authors found HH recurrence
attractive alternatives to synthetic mesh. A randomized in 9 of 21 patients (43%), with 8 of 21 patients (38%)
prospective trial described outcomes in patients with also reporting recurrent symptoms.9 This report led to
and without use of biologic mesh and found a significant widespread uncertainty about the safety and efficacy of
decline in recurrence at 6 months’ follow-up when mesh laparoscopy for PEH repair. However, the authors’ high rate
reinforcement was used.2 A large series of 28 patients of recurrent symptoms is unusual because most patients
undergoing reoperation with either synthetic or biologic who undergo laparoscopic PEH repair report good to
mesh at the hiatus was published that demonstrated that excellent symptom control, which may not correlate with
both types of mesh were associated with significant mesh- radiographic recurrence.2,9
related complications.3 Since then, others have reported
that reoperation after the use of mesh is significantly RECURRENCE RATES WITH PRIMARY CLOSURE IN
more challenging than reoperation in patients in whom THE MODERN LAPAROSCOPIC ERA
mesh was not used and that the patients with mesh are Andujar et al.10 reported their systematic radiographic
at greater risk of needing an esophagogastric resection.4,5 follow-up for 166 patients who underwent laparoscopic
Perhaps the benchmark for HH repair, with or without repair of large PEH. At a mean follow-up of 15 months,
mesh, is the sizable open transthoracic series by Maziak they stated a recurrence rate of 25% (5% PEH and 20%
et al.6 They achieved a mean follow-up of 94 months in sliding HH). Nine years later, Gibson et al.11 described
90 of 94 patients in their cohort who underwent surgical the outcomes of their single-center experience of HH
repair of a large paraesophageal hernia (PEH) over a repair in 100 consecutive patients. They reported a very
36-year period. They reported excellent results in 72 of low recurrence rate of 9 of 100 patients, 7 of whom had
90 patients (80%), with anatomic recurrence in only two only a small (<2 cm) recurrent HH. The same group
patients, both of whom underwent reoperation. Table 30.1 later reported medium-term follow-up, describing a total
summarizes the findings of studies on open PEH repair recurrence rate of 25% at mean follow-up of 24 months
with objective follow-up. (5% PEH and 20% small HH).12 They also reported excel-
Since the publication of these early laparoscopic lent continued gastroesophageal reflux disease–related
reports, advancements in minimally invasive surgery quality-of-life scores.
(e.g., improved visualization, better instrumentation, Nason et al.13 reported their long-term follow-up of 187
and greater experience) have helped surgeons achieve patients who underwent laparoscopic repair of a giant
more extensive mediastinal dissection, better crus closure, PEH from 1997 to 2003. Over their median follow-up of
and use of Collis gastroplasty. All of these are associated 77 months, they found a radiographic recurrence rate
with improved outcomes with primary crus repair of the of 15%. Mittal et al.14 later reported 5-year follow-up
308
Mesh at the Hiatus CHAPTER 30 308.e1
ABSTRACT
Crus closure failure with recurrent hiatal hernia is the
proverbial Achilles heel of antireflux surgery, occurring in
more than two-thirds of patients undergoing reoperative
antireflux surgery. Success with use of synthetic mesh
for ventral and inguinal hernias led to its use in hiatal
hernia repair procedures. This was supported by improved
short-term outcomes. Biologic prostheses were advocated to
diminish the complications associated with synthetic mesh.
Mesh-associated complications are often catastrophic and
frequently require esophagogastric resection. In addition,
long-term results do not support the initial improved
outcomes with mesh use and have cast a cloud over their
routine use. In this chapter we discuss the various aspects
of mesh use for hiatus reinforcement.
KEYWORDS
Antireflux surgery; complications; fundoplication; hiatal
hernia; mesh; paraesophageal hernia
Mesh at the Hiatus CHAPTER 30 309
findings in their 73 patients who underwent surgical Different mesh configurations and materials were used in
intervention for intrathoracic stomach. They reported different series. A recent 1-year contrast study described
5%, 11%, and 17% radiographic failure rates at 1, 3, and follow-up after laparoscopic PEH repair and reported a
5 years, respectively, after surgery. Table 30.2 summarizes 27% recurrence of HH larger than 2 cm after biologic
outcomes after laparoscopic PEH repair. mesh reinforcement of HH.21 Lee et al.22 reported an
even higher recurrence rate of 40% at 5-year follow-up
RECURRENCE RATES WITH MESH-REINFORCED with routine use of AlloDerm (LifeCell Corporation,
CRUS CLOSURE Branchburg, New Jersey) for all cruroplasty (not just
An early review of the literature summarized outcomes of PEH). Table 30.3 summarizes recurrence rates and other
432 patients in 22 studies who underwent mesh-reinforced follow-up data from studies reporting mesh-reinforced
HH repair.8 Recurrence rates ranged from 0% to 24%, but closure.
most of the studies in this review reported zero recurrences.
MESH CONFIGURATIONS AND MATERIALS
Early on, synthetic meshes (e.g., polypropylene and
TABLE 30.1 Open Hiatal Hernia Repair polytetrafluoroethylene) were used for hiatus reinforce-
ment as a direct extension of use in ventral and inguinal
Author, Year Study Design Findings hernias. Multiple configurations of varying mesh sizes
Maziak et al., 94 patients Mean follow-up: 94 and shapes were described and are too simply numerous
19986 with massive months. 93% of patients to list individually. The mesh was generally placed either
incarcerated had good or excellent as a bridge to cover a gap in the crural defect (either
PEH. outcome; only 2% had anterior or posterior) or as an overlay over the primary
symptomatic recurrence. crus closure. Some practitioners advocated circumferential
Low and Unger, 72 patients Mean follow-up: 29.8 placement of mesh with a keyhole-shaped opening to
20057 with large months. 18% of patients accommodate the esophagus.28 Others advised creation of
PEH. had recurrent hernia. No relaxing incisions in the diaphragm to allow for primary
patient required revision crus closure and recommended using mesh to bridge
surgery. the created defect.29 Fig. 30.1 shows the various mesh
PEH, Paraesophageal hernia. placement configurations described previously.
FIGURE 30.1 Various mesh placement configurations for repair of a hiatal hernia. (Used with permission from Norton Thoracic Institute,
Phoenix, Arizona.)
Due to the safety concerns described earlier, biologic mesh has been all but abandoned, as has circumferential
mesh was the preferred material for reinforcement. Many mesh placement.
types of available biologic mesh have been reported and
recommended for use, including porcine submucosa
(Surgisis, Cook Medical, Bloomington, Indiana), bovine
MESH-RELATED COMPLICATIONS
pericardium (Varitas, Baxter International, Deerfield, Reports of mesh-related complications were initially
Illinois), human acellular dermis (AlloDerm, LifeCell isolated in the literature, but they raised safety concerns
Corporation, Branchburg, New Jersey), and porcine dermal nonetheless. A series of 28 patients from several different
collagen (Permacol, Medtronic, Dublin, Ireland). More institutions included 17 patients with mesh erosions and 1
recently, synthetic bioabsorbable meshes have gained case of mesh extraction via the esophagus.3 A significant
acceptance; the most widely used of these is Bio-A (Gore number of patients in these series required esophagogastric
Medical, Flagstaff, Arizona), which comes as a prefashioned resection. Parker et al.5 subsequently reported a series of 78
rectangular mesh that can be tailored for posterior hiatal patients from a single institution who required reoperation
reinforcement. In the present era, the use of nonabsorbable for antireflux surgery. They compared 10 patients who had
Mesh at the Hiatus CHAPTER 30 311
prior mesh at the hiatus with the 68 patients who did not a mean follow-up of 76 months, including three patients
and concluded that the incidence of HH recurrence did (3.2%) who required removal of the mesh. The overall
not differ between the groups, which suggested that mesh rate of recurrent HH was 9% in this series during the
does not provide complete protection from recurrence. same follow-up.
However, they also reported a significantly higher need
(30%) for esophagogastric resection in patients who had
prior mesh at the hiatus. The authors cautioned against MESH VERSUS NO MESH
the liberal use of mesh, advising surgeons to weigh the
higher recurrence rate with no mesh against the need COMPARATIVE STUDIES
for esophagogastric resection associated with mesh in the Ten prospective and retrospective comparative studies
event that reoperation was needed. have been carried out that examine primary crus suture
Nandipati et al.4 reported the largest single-center closure versus synthetic mesh reinforcement. Tam et al.30
experience with reoperation for patients in whom mesh summarized these as part of a systematic review, but the
had previously been used at the hiatus. Their series of 26 review was hindered by differences in surgical technique,
patients included three mesh erosions, and they found that mesh type, duration of follow-up, variation in diagnostic
70% of patients had recurrent HH, indicating that mesh tests, and definition of recurrence. The overall recurrence
does not eliminate the risk of developing a recurrent HH. rate of primary suture closure was 63 of 312 (20%); with
They also reported that 8 of 26 patients (31%) required mesh reinforcement, that rate was 32 of 293 (11%) (9 of 10
esophagogastric resection as a corrective procedure and studies included adequate data). The rate of reoperation
there was high associated perioperative morbidity, although in the same cohort for primary suture closure patients was
no postoperative mortalities was reported. 16 of 200 (8%) and 14 of 214 (6.5%) for patients who
Priego et al.27 recently described long-term follow-up received mesh reinforcement (6 of 10 studies included
of 93 patients who had mesh placed at the hiatus at adequate data). Of all reported complications in the entire
the time of primary surgery. They reported 4.3% 30-day cohort, six were mesh-related. Two of these required
mortality, including 1% mesh-specific mortality. In addi- esophageal resection.30 Table 30.4 summarizes outcomes
tion, reoperation was required in five patients (5.4%) at in comparative, nonrandomized studies.
3. Stadlhuber RJ, Sherif AE, Mittal SK, et al. Mesh complications after matrix in large paraesophageal hiatal hernia. J Gastrointest Surg.
prosthetic reinforcement of hiatal closure: a 28-case series. Surg 2008;12(5):811-815.
Endosc. 2009;23(6):1219-1226. 23. Frantzides CT, Carlson MA, Loizides S, et al. Hiatal hernia repair with
4. Nandipati K, Bye M, Yamamoto SR, Pallati P, Lee T, Mittal SK. mesh: a survey of SAGES members. Surg Endosc. 2010;24(5):1017-1024.
Reoperative intervention in patients with mesh at the hiatus is 24. Alicuben ET, Worrell SG, DeMeester SR. Resorbable biosynthetic
associated with high incidence of esophageal resection—a single- mesh for crural reinforcement during hiatal hernia repair. Am Surg.
center experience. J Gastrointest Surg. 2013;17(12):2039-2044. 2014;80(10):1030-1033.
5. Parker M, Bowers SP, Bray JM, et al. Hiatal mesh is associated with 25. Ward KC, Costello KP, Baalman S, et al. Effect of acellular human
major resection at revisional operation. Surg Endosc. 2010;24(12): dermis buttress on laparoscopic hiatal hernia repair. Surg Endosc.
3095-3101. 2015;29(8):2291-2297.
6. Maziak DE, Todd TR, Pearson FG. Massive hiatus hernia: evaluation 26. Chang CG, Thackeray L. Laparoscopic hiatal hernia repair in 221
and surgical management. J Thorac Cardiovasc Surg. 1998;115(1):53-60; patients: outcomes and experience. JSLS. 2016;20(1).
discussion 61–62. 27. Priego P, Perez de Oteyza J, Galindo J, et al. Long-term results and
7. Low DE, Unger T. Open repair of paraesophageal hernia: reassess- complications related to Crurasoft® mesh repair for paraesophageal
ment of subjective and objective outcomes. Ann Thorac Surg. 2005; hiatal hernias. Hernia. 2017;21:291-298.
80(1):287-294. 28. Frantzides CT, Madan AK, Carlson MA, Stavropoulos GP. A prospec-
8. Targarona EM, Bendahan G, Balague C, Garriga J, Trias M. Mesh in tive, randomized trial of laparoscopic polytetrafluoroethylene (PTFE)
the hiatus: a controversial issue. Arch Surg. 2004;139(12):1286-1296; patch repair vs simple cruroplasty for large hiatal hernia. Arch Surg.
discussion 1296. 2002;137(6):649-652.
9. Hashemi M, Peters JH, DeMeester TR, et al. Laparoscopic repair of 29. DeMeester SR. Laparoscopic paraesophageal hernia repair: critical
large type III hiatal hernia: objective followup reveals high recurrence steps and adjunct techniques to minimize recurrence. Surg Laparosc
rate. J Am Coll Surg. 2000;190(5):553-560; discussion 560–561. Endosc Percutan Tech. 2013;23(5):429-435.
10. Andujar JJ, Papasavas PK, Birdas T, et al. Laparoscopic repair of 30. Tam V, Winger DG, Nason KS. A systematic review and meta-analysis
large paraesophageal hernia is associated with a low incidence of of mesh vs suture cruroplasty in laparoscopic large hiatal hernia
recurrence and reoperation. Surg Endosc. 2004;18(3):444-447. repair. Am J Surg. 2016;211(1):226-238.
11. Gibson SC, Wong SC, Dixon AC, Falk GL. Laparoscopic repair of 31. Schmidt E, Shaligram A, Reynoso JF, Kothari V, Oleynikov D. Hiatal
giant hiatus hernia: prosthesis is not required for successful outcome. hernia repair with biologic mesh reinforcement reduces recurrence
Surg Endosc. 2013;27(2):618-623. rate in small hiatal hernias. Dis Esophagus. 2014;27(1):13-17.
12. Furtado RV, Vivian SJ, van der Wall H, Falk GL. Medium-term 32. Asti E, Lovece A, Bonavina L, et al. Laparoscopic management
durability of giant hiatus hernia repair without mesh. Ann R Coll of large hiatus hernia: five-year cohort study and comparison
Surg Engl. 2016;98(7):450-455. of mesh-augmented versus standard crura repair. Surg Endosc.
13. Nason KS, Luketich JD, Qureshi I, et al. Laparoscopic repair of giant 2016;30(12):5404-5409.
paraesophageal hernia results in long-term patient satisfaction and a 33. Granderath FA, Schweiger UM, Kamolz T, Asche KU, Pointner R.
durable repair. J Gastrointest Surg. 2008;12(12):2066-2075; discussion Laparoscopic Nissen fundoplication with prosthetic hiatal closure
2075-2077. reduces postoperative intrathoracic wrap herniation: preliminary
14. Mittal SK, Bikhchandani J, Gurney O, Yano F, Lee T. Outcomes results of a prospective randomized functional and clinical study.
after repair of the intrathoracic stomach: objective follow-up of up Arch Surg. 2005;140(1):40-48.
to 5 years. Surg Endosc. 2011;25(2):556-566. 34. Oelschlager BK, Pellegrini CA, Hunter JG, et al. Biologic prosthesis to
15. Mattar SG, Bowers SP, Galloway KD, Hunter JG, Smith CD. Long-term prevent recurrence after laparoscopic paraesophageal hernia repair:
outcome of laparoscopic repair of paraesophageal hernia. Surg long-term follow-up from a multicenter, prospective, randomized
Endosc. 2002;16(5):745-749. trial. J Am Coll Surg. 2011;213(4):461-468.
16. Ferri LE, Feldman LS, Stanbridge D, Mayrand S, Stein L, Fried GM. 35. Watson DI, Thompson SK, Devitt PG, et al. Laparoscopic repair
Should laparoscopic paraesophageal hernia repair be abandoned of very large hiatus hernia with sutures versus absorbable mesh
in favor of the open approach? Surg Endosc. 2005;19(1):4-8. versus nonabsorbable mesh: a randomized controlled trial. Ann
17. Rathore MA, Andrabi SI, Bhatti MI, Najfi SM, McMurray A. Meta- Surg. 2015;261(2):282-289.
analysis of recurrence after laparoscopic repair of paraesophageal 36. Koetje JH, Irvine T, Thompson SK, et al. Quality of life following
hernia. JSLS. 2007;11(4):456-460. repair of large hiatal hernia is improved but not influenced by use
18. White BC, Jeansonne LO, Morgenthal CB, et al. Do recurrences of mesh: results from a randomized controlled trial. World J Surg.
after paraesophageal hernia repair matter?: ten-year follow-up after 2015;39(6):1465-1473.
laparoscopic repair. Surg Endosc. 2008;22(4):1107-1111. 37. Memon MA, Memon B, Yunus RM, Khan S. Suture cruroplasty
19. Luketich JD, Nason KS, Christie NA, et al. Outcomes after a decade versus prosthetic hiatal herniorrhaphy for large hiatal hernia: a
of laparoscopic giant paraesophageal hernia repair. J Thorac Cardiovasc meta-analysis and systematic review of randomized controlled trials.
Surg. 2010;139(2):395-404, 404.e1. Ann Surg. 2016;263(2):258-266.
20. Le Page PA, Furtado R, Hayward M, et al. Durability of giant hiatus hernia 38. Antoniou SA, Muller-Stich BP, Antoniou GA, et al. Laparoscopic
repair in 455 patients over 20 years. Ann R Coll Surg Engl. 2015;97(3): augmentation of the diaphragmatic hiatus with biologic mesh versus
188-193. suture repair: a systematic review and meta-analysis. Langenbecks Arch
21. Lidor AO, Steele KE, Stem M, Fleming RM, Schweitzer MA, Marohn Surg. 2015;400(5):577-583.
MR. Long-term quality of life and risk factors for recurrence 39. Huddy JR, Markar SR, Ni MZ, et al. Laparoscopic repair of hiatus
after laparoscopic repair of paraesophageal hernia. JAMA Surg. hernia: does mesh type influence outcome? A meta-analysis and
2015;150(5):424-431. European survey study. Surg Endosc. 2016;30(12):5209-5221.
22. Lee YK, James E, Bochkarev V, Vitamvas M, Oleynikov D. Long-term
outcome of cruroplasty reinforcement with human acellular dermal
PART SIX
Barrett Esophagus
CHAPTER
E
sophageal adenocarcinoma (EAC) is a highly lethal that it is centered on the question of which subtypes
disease associated with a survival of less than 20% of metaplastic epithelium arising in the esophagus or
at 5 years.1 The American Cancer Society estimates esophagogastric junction (EGJ) are at risk for neoplastic
that 16,940 new esophageal cancer cases will be diagnosed progression. The definition of BE is best determined by
in the United States in 2017, the majority from EAC, with its malignant potential because what patients carrying the
15,690 cancer-related deaths.2 The incidence of EAC has diagnosis ultimately want to know is whether they are at
been climbing for more than 40 years at a rate greater than risk of developing esophageal cancer.
any other malignancy and with a greater than sevenfold
increase in the United States between 1975 and 2006.3
The poor prognosis relates to the common presentation of CURRENT DEFINITIONS OF
patients with advanced malignancy at the time symptoms BARRETT ESOPHAGUS
manifest, as well as the lack of effective systemic therapies.
Early diagnosis and curative treatment of locoregional The definition of BE according to the current American
disease are essential for improving overall survival, factors Gastroenterological Association (AGA) medical position
emphasizing the importance of appropriate screening statement on the management of BE is “the condition
and follow-up of at-risk populations prior to symptom in which any extent of metaplastic columnar epithelium
development. that predisposes to cancer development replaces the
Barrett esophagus (BE) is the only known precursor stratified squamous epithelium that normally lines the
to EAC and is a strong risk factor, imparting a 30-fold to distal esophagus.”10 The manuscript goes on to state,
125-fold increased risk over that of the general population.4 “Intestinal metaplasia is required for the diagnosis of
Ample evidence supports the value of BE surveillance in Barrett’s esophagus because intestinal metaplasia is the
detecting EAC at an earlier stage, requiring less aggressive only type of esophageal columnar epithelium that clearly
treatment and with a better prognosis, than tumors first predisposes to malignancy.”10 Therefore the diagnosis of
presenting at the time symptoms arise.5,6 The develop- BE requires a combination of endoscopic and histologic
ment of BE is due to metaplasia of the esophageal lining findings, in that the columnar metaplasia must involve the
from squamous to columnar epithelium resulting from tubular esophagus on endoscopy and must contain goblet
the effects of refluxed gastric contents, including acid, cells, which define the presence of intestinal metaplasia
bile, and pancreatic enzymes, in susceptible individuals. (IM), on biopsy specimens.
Depending on how it is defined and the diligence with Agreement does not exist worldwide on the requirement
which it is detected, BE is found in approximately 10% for goblet cells to diagnose BE.11 The British Society
to 15% of patients with symptomatic gastroesophageal of Gastroenterology mandates only histologic proof of
reflux disease (GERD) undergoing endoscopic biopsies.7 columnar mucosa on biopsies taken from the tubular
The pathogenesis of EAC places BE as the important link esophagus; goblet cells do not need to be documented.12
between GERD, the most common malady affecting the In Japan the presence of any endoscopically detectable
foregut, and EAC, the cancer most rapidly increasing in columnar lining of the distal esophagus is adequate to
incidence in Western societies.8 prove the diagnosis of BE; biopsy confirmation is not
Despite being common and playing a key role in EAC mandated.13 The main controversies underlying these
development, BE has been shrouded in controversy since differing definitions are the potential for development of
it was first described 50 years ago.9 Even today, debate EAC in the setting of a columnar-lined esophagus (CLE)
continues over the correct definition relative to the location with or without the presence of goblet cells, as well as in
and type of columnar metaplasia required to establish IM developing distal to the tubular esophagus, so-called
the diagnosis. The debate is by no means esoteric, given cardia intestinal metaplasia (CIM).14
314
Controversies in the Definition of Barrett Esophagus CHAPTER 31 314.e1
ABSTRACT
Esophageal adenocarcinoma (EAC) is a highly lethal
disease associated with a poor prognosis, which relates
to the common presentation of patients with advanced
malignancy at the time symptoms manifest. Early diagnosis
is essential for improving overall survival, a fact that empha-
sizes the importance of appropriate screening and follow-up
of at-risk populations prior to symptom development.
Barrett esophagus (BE) is the only known precursor to
EAC and is a strong risk factor. The pathogenesis of EAC
places BE as the important link between gastroesophageal
reflux disease (GERD), the most common malady affecting
the foregut, and EAC, the cancer most rapidly increasing
in incidence in Western societies. Despite being common
and playing a key role in EAC development, BE has been
shrouded in controversy since it was first described 50
years ago. Even today, debate continues over the correct
definition relative to the location and type of columnar
metaplasia required to establish the diagnosis. The defini-
tion of BE is best determined by its malignant potential
because what patients carrying the diagnosis ultimately
want to know is whether they are at risk of developing
esophageal cancer.
KEYWORDS
Barrett esophagus
esophageal cancer
esophageal adenocarcinoma
columnar-lined esophagus
intestinal metaplasia
cardia intestinal metaplasia
goblet cells
nongoblet esophageal metaplasia
Controversies in the Definition of Barrett Esophagus CHAPTER 31 315
Stomach Stomach
A B
FIGURE 31.2 (A) Barrett’s original conception (1950) of the etiology of a columnar-lined tubular structure in the chest subjacent to the
esophagus, what he believed to be an intrathoracic stomach. (B) Allison and Johnstone’s interpretation (1953) of the etiology of a
columnar-lined tubular structure in the chest, later termed by Barrett (1957) to be columnar-lined esophagus.
316 SECTION I Esophagus and Hernia
FIGURE 31.4 Cardiac (junctional) epithelium with mucous cells. Squamous mucosa
(Photomicrographs courtesy Wei Xu, MD.)
Esophagus Intestinal (specialized)
mucosa in gastric cardia
Cardiac (junctional)
mucosa in gastric
cardia
Oxyntocardiac
Stomach (fundic) mucosa in
gastric cardia
cardiac epithelium usually less than 5 cm in length or proximal stomach. The term gastric cardia has been
and occasionally absent. 14 Patients with cardiac and used to denote the region of the stomach just distal to
oxyntocardiac epithelium were more likely to have an the GEJ, although its boundaries are poorly defined and
abnormal 24-hour ambulatory esophageal pH study and its definition is controversial. The theory that metaplastic
a mechanically defective lower esophageal sphincter on cardiac epithelium has a derivation similar to metaplastic
manometry, both suggesting the presence of GERD, than esophageal epithelium means that the cardia has its origins
patients lacking those epithelial types. in the esophagus, not the stomach, contrary to common
An autopsy study reported in 2000 assessed the histology terminology and conceptualization. Cancers of the cardia,
of the GEJ in adult subjects who had no evidence of according to this theory, are best classified as esophageal
GERD throughout their lives.28 Cardiac epithelium was not gastric,35 consistent with the most recent staging of the
absent in more than half of specimens, and the length of American Joint Committee on Cancer, seventh edition.36
cardiac and oxyntocardiac epithelia ranged from 0.4 to Remarkable in this theory is the premise that, in a complete
8.05 mm, with a squamo-oxyntic gap of less than 5 mm in reversal of the original contention of Norman Barrett
the majority. Half of subjects were found to have a direct that the tubularized CLE he observed was stomach, what
transition from squamous to gastric oxyntic mucosa in at has been called proximal stomach (the “gastric cardia”)
least part of the circumference of the SCJ. These findings lined by metaplastic columnar epithelium is, in fact,
were validated by a later independent study.29 An additional esophagus.
autopsy study in children found cardiac epithelium in all Several types of evidence support the esophageal origin
subjects, but limited to a length between 1 and 4 mm, of cardia adenocarcinoma. Cancers arising in the cardia
with a median of 1.8 mm, far less than the 50-mm length are associated with symptomatic GERD, albeit to a lesser
previously thought to be normal.30 extent than EAC.8 The increasing incidence of cardia
Based on these findings, a direct transition from adenocarcinoma has paralleled the rise in EAC over the
esophageal squamous mucosa to oxyntic gastric mucosa past four decades, although it has deviated from the trend
does exist in some individuals, without a buffer zone of in incidence of cancer arising in the distal stomach. 37
cardiac epithelium. When cardiac epithelium is present In addition, the vast majority of cases of both EAC and
in normal individuals, its length is short, generally limited cardia adenocarcinoma occur in association with IM.38,39
to a few millimeters or less. Additional studies have dem- The presence of dysplastic epithelium arising in cardiac
onstrated that the presence of cardiac epithelium distal (junctional) or fundic mucosa is uncommon in the absence
to the endoscopic GEJ, as defined by the rugal folds, is of coexisting IM.22
associated with GERD, with the length of the segment being
a sensitive indicator of GERD severity.31,32 Finally, cardiac SCREENING AND SURVEILLANCE BIOPSY
mucosa residing distal to the SCJ generally demonstrates PROTOCOLS FOR BARRETT ESOPHAGUS
chronic inflammation in the lamina propria, consistent with Given that IM arising in columnar epithelium below the
GERD, and likely not related to Helicobacter pylori infestation proximal limit of the gastric folds (i.e., CIM) has the same
of the stomach, the other potential cause.33,34 Mucosal pathogenesis as IM arising from the tubular esophagus
inflammation is the major stimulus for the development (i.e., BE), both sites of IM would be expected to have
of IM, both in the esophagus and stomach. the same neoplastic potential. As IM is a GERD-related
If the premise is accepted that cardiac epithelium at the phenomenon, it represents a continuum of disease starting
GEJ is pathologic, related to GERD-induced metaplasia of at the most distal esophagus, where reflux is most severe,
the esophageal mucosa rather than a normal lining of the and progressing cephalad; CIM and BE are not distinct
proximal stomach, an accurate histologic determination clinical entities. The differentiation of CIM from SSBE can
of the GEJ becomes possible. In this model the GEJ is be quite difficult and arbitrary, given that the conventional
defined by the proximal border of gastric oxyntic mucosa boundary is the top of the gastric rugal folds, a landmark
as assessed on biopsies. In a normal subject the GEJ is that can be indiscrete. If the pathogenesis and risk of
located where esophageal squamous mucosa transitions malignant progression is the same for the two entities,
to gastric oxyntic mucosa. In patients with GERD the then distinguishing them is not necessary.
GEJ is located where metaplastic columnar epithelium Although studies have shown an increased risk of EAC
residing within the squamocolumnar gap transitions to with longer segments of IM, a length threshold relative to
gastric oxyntic mucosa. Because this latter transition is risk has not been established.40–44 In a prospective study,
not discernible at the time of endoscopy, the true GEJ in Sikkema et al. found that for every additional centimeter
such circumstances can only be determined by extensive, in BE length the risk of developing high-grade dysplasia
meticulous, and accurately recorded biopsies. (HGD) or EAC increased by 11% over 4 years.45 The
preponderance of evidence, as well as common sense,
suggests that the rate of development of EAC increases
with a larger at-risk surface area of metaplastic epithelium;
IMPLICATIONS OF THE LOCATION OF THE arbitrary cutoffs between CIM, SSBE, and LSBE relative
GASTROESOPHAGEAL JUNCTION to that risk do not seem justified.
Some authors have argued that CIM and SSBE are, in
DEFINITION OF THE CARDIA fact, distinct clinical entities with differing etiologies and
The location of the GEJ has several implications rela- risks of neoplastic progression.46 As evidence, proponents
tive to the diagnosis, classification, and management of of this belief cite differing cytokeratin 7 and 20 immunore-
metaplasia or neoplasia arising in the distal esophagus activity between CIM and BE noted in one study, a finding
Controversies in the Definition of Barrett Esophagus CHAPTER 31 319
that has not been verified in subsequent reports.47–49 The 2. Sampling error, which depends upon the thoroughness
preponderance of literature would suggest that there are of biopsies, the length of CLE, and goblet cell density;
no other immunohistochemical biomarkers either specific and
for mucin or intestinalization (such as CDX2, DAS-1, Hep 3. Goblet cell dynamics.
Par 1, Villin, or MUC2) that can differentiate metaplasia Pseudogoblet cells are mucin-containing columnar cells
occurring in the tubular esophagus from the cardia.49–51 that are difficult to differentiate from true goblet cells. In
In addition, IM arising in the proximal stomach possesses contrast to goblet cells, which typically form as single cells
immunohistochemical features similar to BE and not to in a random distribution, pseudogoblet cells tend to occur
IM arising in the distal stomach.52 in rows within the superficial epithelium. Pseudogoblet
The conceptualization of a continuum of metaplasia cells also lack the triangular nucleus characteristic of true
arising within the esophagus, rather than the discrete goblet cells.58
entities of CIM, SSBE, and LSBE, has implications regard- The number of goblet cells may be small in patients
ing the need to biopsy both above and below the top of with CIM or SSBE. Chandrasoma et al. found IM in 56%
the rugal folds at the time of endoscopy for BE. Most of patients with a 1-cm length of CLE, progressing to 100%
studies of BE screening, surveillance, and progression with a CLE length greater than 5 cm.59 The likelihood of
have assessed biopsies taken from the tubular esophagus detecting goblet cells has been shown to correlate directly
alone, not the gastric cardia. As a result, the cancer risk with the number of biopsies performed at endoscopy as
in patients with CIM is poorly defined.53–56 well as the length of CLE.26,60 Harrison and co-authors
In the AGA Chicago Workshop Recommendations on found a progressive increase in the detection of IM with
the Diagnosis and Management of BE from 2004, state- a higher number of biopsies, reaching 100% when more
ment 9 is: “The normal appearing and normally located than 16 biopsies were taken.61 In addition, goblet cell
squamocolumnar junction should not be biopsied.”22 density has been shown in some studies to vary with the
The subsequent AGA medical position statement on the position along the length of CLE, being highest near the
management of BE from 2011 is quiet on the matter.10 SCJ and lower in the more distal portions.24,62 However,
In addition, the AGA recommendations from 2004 are other reports have shown a more random distribution.59,63
to perform systematic biopsies of both SSBE and LSBE, Current guidelines recommend four quadrant biopsies
the consensus opinion being that they are not distinct every 1 to 2 cm along CLE, with special attention given to
clinical entities.22 Given the continuum of IM length, the regions of mucosal nodularity or irregularity.10,64,65 If an
omission of biopsies from the proximal columnar mucosa adequate number of biopsies is not taken along the entire
just distal to the upper limit of the gastric rugal folds does length of CLE, and if the biopsies are not appropriately
not appear justified, despite the dearth of data delineating dispersed with particular attention paid to the region of
neoplastic risk of IM at this location. Although the risk the SCJ, the presence of goblet cells may be missed. Finally,
of progression is likely lower than for longer segments of goblet cells may vary over time, with therapy, and with
IM based on surface area considerations alone, a discrete progression of disease.60,66 Repeat endoscopy with biopsies
recommendation differing from that for SSBE does not may be necessary to establish the diagnosis of IM.60,66 Each
seem appropriate. of these factors contributes to the unreliability of using
the identification of goblet cells as the pathognomonic
criterion for the presence of IM.
RISK OF NEOPLASTIC PROGRESSION
IN SUBTYPES OF ESOPHAGEAL STEPS IN THE DEVELOPMENT OF
INTESTINAL METAPLASIA
COLUMNAR METAPLASIA Chronic esophageal inflammation and ulceration resulting
from the reflux of gastric contents, including both acid
LIMITATIONS OF USING GOBLET CELLS and bile, drive metaplasia within the esophageal lining.
TO DETERMINE THE PRESENCE OF The conversion of esophageal squamous epithelium to IM
INTESTINAL METAPLASIA is thought to be a stepwise process. Inflamed esophageal
Most cases of adenocarcinoma of the esophagus or GEJ squamous epithelium is first replaced by a multilayer,
arise in a background of IM as defined by the presence of transitional epithelium followed by a single-layer, nonin-
goblet cells, which normally reside in the intestines.6,38,54–56 testinal columnar epithelium; a specialized intestinal-type
Therefore detecting goblet cells in screening or surveil- metaplastic epithelium is the final step.67 According to this
lance biopsies for BE is critical for determining the risk model, the previously described phenotypes of columnar
of subsequent neoplastic progression and the need for metaplasia, including cardiac, oxyntocardiac, and intestinal
follow-up. The dictum “no goblets—no Barrett’s” has types, represent points along a continuum rather than
arisen to underscore their fundamental importance.57 distinct entities. The first step in intestinalization appears
That being said, the highly differentiated goblet cell is to be mediated by upregulation of the Sonic hedgehog
unlikely the precursor to EAC because cancers typically (SHH)–bone morphogenetic protein 4 (BMP-4) signal-
arise from more poorly differentiated cell lines. The goblet ing path, leading to phosphorylation of SMAD proteins
cell merely serves as a marker for the malignant potential (pSMAD), and is modulated by a number of antagonists
of the surrounding metaplastic epithelium. and downstream factors. SHH-BMP-4/pSMAD signaling
A number of factors can affect the detection of goblet is responsible for the induction of genes responsible for
cells, including: the nonintestinal type of metaplasia. The next step is the
1. Differentiation of goblet cells from pseudogoblet cells; induction of genes responsible for intestinalization and
320 SECTION I Esophagus and Hernia
is mediated by the interaction of pSMAD with CDX2, an no difference in the rate of progression to EAC (4.5% and
intestine-specific homeobox gene critical to intestinal 3.6%, respectively). However, the same questions can be
epithelial functions. In the final stages of intestinal dif- raised about their findings relative to errors in sampling
ferentiation, Wnt and Notch signaling also are key.67 for the presence of goblet cells.75
Metaplastic columnar epithelium can reveal biochemical Other studies have contradicted the claim that adeno-
or molecular evidence of intestinal differentiation prior carcinomas of the esophagus and EGJ can arise from
to the development of goblet cells. Studies have shown nongoblet columnar metaplasia. A large population-
that metaplastic columnar mucosa without goblet cells based analysis of 8522 patients with CLE with or without
may express proteins or transcription factors specific to IM was undertaken by Bhat et al. using the Northern
intestinalization, such as CDX2, DAS-1, and Villin, or Ireland BE Registry.76 At a mean follow-up of 7 years,
intestine-specific mucopolysaccharides, such as MUC2, the risk of cancer progression was 0.07% per year in
early in the process of differentiation, although may patients without IM compared with 0.38% per year in
be expressed further once goblet cells appear.48,49,68–70 patients with IM at initial biopsy (P < .001). In a study
Metaplastic esophageal columnar epithelium not contain- of 214 patients by Chandrasoma et al. using a rigorous
ing goblet cells on histology may still be “biologically biopsy protocol of CLE and the GEJ, dysplasia or EAC
intestinalized,”58 with studies finding similar DNA content developed only in patients with proven IM; the risk of
and chromosomal abnormalities to epithelium containing malignant progression was thought either nonexistent or
goblet cells.71,72 Thus goblet cell formation likely represents extremely low for patients without goblet cells.61 Similarly,
the end of a chain of genetic and signaling events ultimately Westerhoff et al. found EAC or dysplasia to arise only in
culminating in histologic intestinalization of metaplastic patients with documented goblet cells.77 Elimination of
cardiac epithelium. the requirement for identification of goblet cells would
have increased the diagnosis of BE by 147% in their study
population without identifying any additional patients who
RISK OF NEOPLASTIC PROGRESSION IN subsequently developed dysplasia or neoplasia. Finally, in
COLUMNAR-LINED ESOPHAGUS WITH OR a study of 45 patients with BE, biopsies of IM revealed
WITHOUT GOBLET CELLS a higher frequency of cancer-associated mutations than
Given the genetic and molecular abnormalities identifiable those obtained from nongoblet metaplastic mucosa.78
in nongoblet esophageal metaplasia, questions exist as to At present, whether nongoblet cell metaplasia of
whether such an epithelium is at risk for progression to the esophagus is a premalignant condition remains
EAC and whether the risk is similar to progression of IM controversial. What is known is that if the definition
in the presence of goblet cells. of BE were expanded to include all patients with CLE
A study of 141 patients who underwent endoscopic without documented IM, the number of patients in need
resection (ER) of small (<2 cm diameter) EAC found that of surveillance would escalate greatly, adding significant
71% had cardiac or fundic epithelium, not IM, adjacent cost without proof of increasing cancer detection and
to the cancer.73 In addition, IM was not observed in any saving lives. More data clearly are necessary regarding
areas of the ER specimens in 57% of cases. Of note, areas the natural history of non-goblet esophageal metaplasia,
of CLE outside of the ER specimens were not biopsied including the cancer risk in both long and short segments,
in this study, so the true incidence of IM in their patient other risk factors for malignant progression, potential
population is not known. lives saved with surveillance, and the cost-effectiveness of
Two large retrospective studies from the United various management strategies, before recommendations
Kingdom found the rates of development of dysplasia can be made to change current practice.
or EAC to be similar in patients with IM compared with
those with CLE without IM.74,75 Gatenby et al. analyzed
3568 biopsies of nondysplastic CLE from 1751 patients
CONCLUSION
with and without IM and found no difference in the rate The definition of BE has evolved over the past 60 years
of development of dysplasia or EAC between the two since CLE was first recognized as an understanding of its
groups.74 The mean length of CLE in the two cohorts pathogenesis has emerged. At the core of the definition
was 5.75 cm and 4.93 cm, respectively. Because the mean of BE is the malignant potential of metaplastic esophageal
number of biopsies was only 2.04 per patient and the columnar epithelium, whether or not it contains goblet
biopsies were randomly distributed with no targeting of the cells and whether or not it is located in the tubular
SCJ, a strong argument can be made that IM was missed esophagus. The risk of developing esophageal cancer
in many patients due to sampling error. This argument is, after all, what concerns patients diagnosed with BE,
is bolstered by the authors’ findings that the rate of IM their families, and their care providers and determines
detection increased with the number of biopsies taken (a the need for surveillance.
24% increase noted for each additional biopsy), and the In the United States and most of the world the presence
fact that 90.8% of the patients initially not found to have of IM must be documented to establish the diagnosis of
IM developed it on subsequent biopsies over the course BE, given the malignant potential of metaplastic columnar
of the next 10 years. Sampling error likely was a major epithelium containing goblet cells. Although the presence
factor negating their conclusion that cancer risk was not of IM is not a requirement for the diagnosis of BE in Great
affected by the presence or absence of IM. Britain and Japan, a paucity of data exists delineating
Kelty et al. analyzed 712 patients with CLE with or the malignant risk associated with nongoblet columnar
without IM over a mean follow-up of 12 years and found epithelium. The available literature in support of the
Controversies in the Definition of Barrett Esophagus CHAPTER 31 321
precancerous nature of CLE not demonstrating IM must 6. Peters JH, Clark GWB, Ireland AP, Chandrasoma P, Smyrk TC,
be interpreted with caution in light of the potential for DeMeester TR. Outcome of adenocarcinoma arising in Barrett’s
esophagus in endoscopically surveyed and nonsurveyed patients. J
significant sampling error, leading to an underappreciation Thorac Cardiovasc Surg. 1994;108:813-821.
of the presence of goblet cells in patients assessed in 7. Dent J, El-Serag HB, Wallander ME, Johansson S. Epidemiology
those studies. of gastro-oesophageal reflux disease: a systematic review. Gut.
Although metaplasia found within what has been termed 2005;54(5):710-717.
8. Lagergren J, Bergstrom R, Lindgren A, Nyren O. Symptomatic gas-
the gastric cardia has not received as much attention as troesophageal reflux as a risk factor for esophageal adenocarcinoma.
IM arising in the tubular esophagus, an extensive body N Engl J Med. 1999;340(11):825-831.
of investigation into the histology of the GEJ supports an 9. Barrett NR. The lower esophagus lined by columnar epithelium.
esophageal origin to the metaplastic mucosa arising just Surgery. 1957;41:881-894.
beyond the proximal limit of the gastric folds. Because 10. Spechler SJ, Sharma P, Souza RF, Inadomi JM, Shaheen NJ.
American Gastroenterological Association medical position state-
CIM shares a common pathogenesis with BE, both entities ment on the management of Barrett’s esophagus. Gastroenterology.
should be handled as a spectrum of metaplastic changes 2011;140:1084-1091.
spanning the GEJ and extending proximally into the 11. Riddell RH, Odze RD. Definition of Barrett’s esophagus: time
esophagus. Only with further study of cohorts of patients for a rethink—is intestinal metaplasia dead? Am J Gastroenterol.
2009;104:2588-2594.
undergoing an aggressive biopsy protocol, including 12. Fitzgerald RC, di Pietro M, Raganuth K, et al. British Society of
adequate sampling of the mucosa just beyond the SCJ, Gastroenterology guidelines on the diagnosis and management of
will the significance of IM in the “gastricized” esophagus Barrett’s oesophagus. Gut. 2014;63:7-42.
be clarified. 13. Ogiya K, Kawano T, Ito E, et al. Lower esophageal palisade vessels and
Until the malignant potential of nongoblet IM arising the definition of Barrett’s esophagus. Dis Esophagus. 2008;21:645-649.
14. Oberg S, Peters JH, DeMeester TR, et al. Inflammation and special-
in CLE is clarified with additional longitudinal studies on ized intestinal metaplasia of cardiac mucosa is a manifestation of
adequate numbers of patients undergoing a thorough gastroesophageal reflux disease. Ann Surg. 1997;226:522-532.
biopsy protocol, the diagnosis of BE is best reserved for 15. Barrett NR. Chronic peptic ulcer of the oesophagus and ‘oesophagitis’.
cases of documented IM. Expanding the definition of Br J Surg. 1950;38:175-182.
16. Allison PR, Johnstone AS. The oesophagus lined with gastric mucous
BE to include all types of columnar metaplasia would membrane. Thorax. 1953;8:87-101.
greatly increase the pool of patients in need of long-term 17. Barrett NR. The lower esophagus lined by columnar epithelium.
surveillance, with an associated increase in costs and Surgery. 1957;41:881-894.
without proven benefit. The impact of the diagnosis of 18. Haggitt RC, Dean PJ. Adenocarcinoma in Barrett’s epithelium.
BE on the individual, both in terms of the psychologic In: Spechler SJ, Goyal RK, eds. Barrett’s Esophagus: Pathophysiology,
Diagnosis and Management. New York: Elsevier Science Publishing;
costs of a precancerous diagnosis and the monetary costs 1985:153-166.
of serial endoscopic assessments, is not trivial. In addition, 19. Hayward J. The lower end of the eoesophagus. Thorax. 1961;16:36-41.
a diagnosis of BE can escalate the price of life insurance, 20. Bremner CG, Hamilton DG, DeMeester TR, et al. Barrett’s esophagus:
with a reported increase of 118% for an otherwise healthy, controversial aspects. In: DeMeester TR, Skinner DB, eds. Esophageal
Disorders: Pathophysiology and Therapy. New York: Raven Press; 1985:233.
nonsmoking male.79 21. McClave SA, Boyce HW Jr, Gottfried MR. Early diagnosis of columnar
Much remains to be learned about BE, including ways lined esophagus: a new endoscopic diagnostic criterion. Gastrointest
to improve the diagnosis, stratify risk, and predict the Endosc. 1987;33:413-416.
response to therapy. Molecular profiling has been studied, 22. Sharma P, McQuaid K, Dent J, et al. A critical review of the diagnosis
particularly in nongoblet CLE, although it needs to be and management of Barrett’s esophagus: the AGA Chicago Workshop.
Gastroenterology. 2004;127:310-330.
validated for cancer risk independent of the presence of 23. Ayazi S, Tamhankar A, DeMeester SA, et al. The impact of gastric
IM. The current position of the AGA is that molecular distention on the lower esophageal sphincter and its exposure to
biomarkers should not be used for risk stratification.10 acid gastric juice. Ann Surg. 2010;252:57-62.
Similarly, serum biomarkers have been studied, although 24. Chandrasoma PT, Makarewicz K, Wickramasinghe K, Ma YL,
DeMeester TR. A proposal for a new validated histologic definition
the available data do not support their utility at present.80 of the gastroesophageal junction. Hum Pathol. 2006;37:40-47.
For now, the presence of IM, as determined by the 25. Paull A, Trier JS, Dalton MD, Camp RC, Loeb P, Goyal RK. The histo-
histologic confirmation of goblet cells, remains the key logic spectrum of Barrett’s esophagus. N Engl J Med. 1976;295:476-480.
to establishing the diagnosis of BE. Until additional data 26. Chandrasoma PT, Wijetunge S, DeMeester SR, Hagen JA, DeMeester
dictate otherwise, those of us providing care to the patient TR. The histologic squamo-oxyntic gap: an accurate and reproduc-
ible marker of gastroesophageal reflux disease. Am J Surg Pathol.
with BE should not be ready to “throw down the goblet”! 2010;34:1574-1581.
27. Chandrasoma PT, Der R, Ma Y, Peters J, DeMeester TR. Histologic
classification of patients based on mapping biopsies of the gastro-
REFERENCES esophageal junction. Am J Surg Pathol. 2003;27:929-936.
28. Chandrasoma PT, Der R, Ma Y, Dalton P, Taira M. Histology of
1. Pohl H, Sirovich B, Welch HG. Esophageal adenocarcinoma inci- the gastroesophageal junction: an autopsy study. Am J Surg Pathol.
dence: are we reaching the peak? Cancer Epidemiol Biomarkers Prev. 2000;24:402-409.
2010;19(6):1468-1470. 29. Sarbia M, Donner A, Gabbert HE. Histopathology of the gastro-
2. https://www.cancer.org/cancer/esophagus-cancer/about/key- esophageal junction. A study on 36 operation specimens. Am J Surg
statistics.html. Accessed 5 December 2017. Pathol. 2002;26:1207-1212.
3. Pera M, Manterola C, Vidal O, Grande L. Epidemiology of esophageal 30. Park YS, Park HJ, Kang GH, Kim CJ, Chi JG. Histology of gastro-
adenocarcinoma. J Surg Oncol. 2005;92:151-159. esophageal junction in fetal and pediatric autopsy. Arch Pathol Lab
4. Cameron AJ, Ott BJ, Payne WS. The incidence of adenocarci- Med. 2003;127:451-455.
noma in columnar-lined (Barrett’s) esophagus. N Engl J Med. 31. Kilgore SP, Ormsby AH, Gramlich TL, et al. The gastric cardia: fact
1985;313(14):857-859. or fiction? Am J Gastroenterol. 2000;95:921-924.
5. Corley DA, Levin TR, Habel LA, Weiss NS, Buffler PA. Surveillance 32. Glickman JN, Fox V, Antonioli DA, Wang HH, Odze RD. Morphology
and survival in Barrett’s adenocarcinomas: a population-based study. of the cardia and significance of carditis in pediatric patients. Am
Gastroenterology. 2002;122:633-640. J Surg Pathol. 2002;26:1032-1039.
322 SECTION I Esophagus and Hernia
33. Der R, Tsao-Wei DD, DeMeester TR, et al. Carditis: a manifestation 57. Batts KP. Barrett’s esophagus—more steps forward. Hum Pathol.
of gastroesophageal reflux disease. Am J Surg Pathol. 2001;25:245-252. 2001;32:357-359.
34. Spechler SJ, Wang HH, Chen YY, Zeroogian JM, Antonioli DA, Goyal 58. Naini BV, Chak A, Ali MA, Odze RD. Barrett’s oesophagus diagnostic
RK. GERD vs. H. pylori infections as potential causes of inflammation criteria: endoscopy and histology. Best Pract Res Clin Gastroenterol.
in thegastric cardia. Gastroenterology. 1997;112:A297. 2015;29:77-96.
35. Chandrasoma PT, Makarewicz K, Wickramasinghe K, Ma YL, 59. Chandrasoma P, Wijetunge S, DeMeester S, Ma Y, et al. Columnar-
DeMeester TR. Adenoarcinomas of the distal esophagus and “gastric lined esophagus without intestinal metaplasia has no proven risk
cardia” are predominantly esophageal adenocarcinomas. Am J Surg of adenocarcinoma. Am J Surg Pathol. 2012;36:1-7.
Pathol. 2007;31:569-575. 60. Oberg S, Johansson J, Wenner J, et al. Endoscopic surveillance of
36. Stephen B Edge, American Joint Committee on Cancer. AJCC Cancer columnar lined esophagus: frequency of intestinal metaplasia detec-
Staging Manual. 7th ed. New York: Springer; 2010:103-115. tion and impact of antireflux surgery. Ann Surg. 2001;234:619-626.
37. Devesa SS, Blot WJ, Fraumeni JF. Changing patterns in the incidence 61. Harrison R, Perry I, Haddadin W, et al. Detection of intestinal
of esophageal and gastric carcinoma in the United States. Cancer. metaplasia in Barrett’s esophagus: an observational comparator
1998;83:2049-2053. study suggests the need for a minimum of eight biopsies. Am J
38. Haggitt RC, Tryzelaar J, Ellis FH, Colcher H. Adenocarcinoma Gastroenterol. 2007;102:1154-1161.
complicating columnar epithelium-lined (Barrett’s) esophagus. Am 62. Chandrasoma PT, Der R, Dalton P, et al. Distribution and significance
J Clin Pathol. 1978;70:1-5. of epithelial types in columnar-lined esophagus. Am J Surg Pathol.
39. Smith RRL, Hamilton SR, Boitnott JK, Rogers EL. The spectrum 2001;25:1188-1193.
of carcinoma arising in Barrett’s esophagus. Am J Surg Pathol. 63. Thompson JJ, Zinsser KR, Enterline HT. Barrett’s metaplasia and
1984;8:563-573. adenocarcinoma of the esophagus and gastroesophageal junction.
40. Rudolph RE, Vaughan TL, Storer BE, et al. Effect of segment Hum Pathol. 1983;14:42-61.
length on risk for neoplastic progression in patients with Barrett’s 64. Wang KK, Sampliner RE. Updated guidelines 2008 for the diagnosis,
esophagus. Ann Intern Med. 2000;132:612-620. surveillance and therapy of Barrett’s esophagus. Am J Gastroenterol.
41. Weston AP, Badr AS, Hassanein RS. Prospective multivariate analysis of 2008;103:788-797.
clinical, endoscopic, and histologic factors predictive of the develop- 65. ASGE Standards of Practice Committee, Evans JA, Early DS, Fukami
ment of Barrett’s multifocal high-grade dysplasia or adenocarcinoma. N, et al. The role of endoscopy in Barrett’s esophagus and other
Am J Gastroenterol. 1999;94:3413-3419. premalignant conditions of the esophagus. Gastrointest Endosc.
42. Menke-Pluymers MB, Hop WC, Dees J, van Blankenstein M, Tilanus 2012;76:788-797.
HW. Risk factors for the development of an adenocarcinoma in 66. Jones TF, Sharma P, Daaboul B, et al. Yield of intestinal metaplasia
columnar-lined (Barrett) esophagus. The Rotterdam Esophageal in patients with suspected short-segment Barrett’s esophagus (SSBE)
Tumor Study Group. Cancer. 1993;72:1155-1158. on repeat endoscopy. Dig Dis Sci. 2002;47:2108-2111.
43. Wani S, Falk G, Hall M, et al. Patients with nondysplastic Barrett’s 67. Krishnadath KK, Wang KK. Molecular pathogenesis of Barrett
esophagus have low risks for developing dysplasia or esophageal esophagus. Gastroenterol Clin North Am. 2015;44:233-247.
adenocarcinoma. Clin Gastroenterol. 2011;9:2207. 68. Hahn HP, Blount PL, Ayub K, et al. Intestinal differentiation in
44. Desai TK, Krishnan K, Samala N, et al. The incidence of esophageal metaplastic, nongoblet columnar epithelium in the esophagus. Am
adenocarcinoma in non-dysplastic Barrett’s esophagus: a meta-analysis. J Surg Pathol. 2009;33:1006-1015.
Gut. 2012;61:970-976. 69. Groisman GM, Amar M, Meir A. Expression of the intestinal marker
45. Sikkema M, Looman CW, Steyergerg EW, et al. Predictors for Cdx2 in the columnar-lined esophagus with and without intestinal
neoplastic progression in patients with Barrett’s esophagus: a (Barrett’s) metaplasia. Mod Pathol. 2004;17:1282-1288.
prospective cohort study. Am J Gastroenterol. 2011;106:1231-1238. 70. Chaves P, Cruz C, Dias Pereira A, et al. Gastric and intestinal dif-
46. Sharma P, Weston AP, Morales T, Topalovski M, Mayo MS, Sampliner ferentiation in Barrett’s metaplasia and associated adenocarcinoma.
RE. Relative risk of dysplasia for patients with intestinal metaplasia Dis Esophagus. 2005;18(6):383-387.
in the distal oesophagus and in the gastric cardia. Gut. 2000;46:9-13. 71. Liu W, Hahn H, Odze RD, Goyal RK. Metaplastic esophageal
47. Glickman JN, Wang H, Das KM, et al. Phenotype of Barrett’s esophagus columnar epithelium without goblet cells shows DNA content
and intestinal metaplasia of the distal esophagus and gastroesophageal abnormalities similar to goblet-containing epithelium. Am J Gastro-
junction. An immunohistochemical study of cytokeratins 7 and 20, enterol. 2009;104:816-824.
DAS-1 and 45M1. Am J Surg Pathol. 2001;25:87-94. 72. Chaves P, Crespo M, Ribeiro C, et al. Chromosomal analysis of
48. Mohammed IA, Streutker CJ, Riddell RH. Utilization of cytokeratins Barrett’s cells: demonstration of instability and detection of the
7 and 20 does not differentiate between Barrett’s esophagus and metaplastic lineage involved. Mod Pathol. 2007;20:788-796.
gastric cardiac intestinal metaplasia. Mod Pathol. 2002;15:611-616. 73. Takubo K, Aida J, Naomoto Y, et al. Cardiac rather than intestinal-type
49. DeMeester SR, Wickramasinghe KS, Lord RV, et al. Cytokeratin and background in endoscopic resection specimens of minute Barrett
DAS-1 immunostaining reveal similarities among cardiac mucosa, adenocarcinoma. Hum Pathol. 2009;40:65-74.
CIM and Barrett’s esophagus. Am J Gastroenterol. 2002;97:2514-2523. 74. Gatenby PA, Ramus JR, Caygill CP, Shepherd NA, Watson A. Relevance
50. Phillips RW, Frierson HF Jr, Moskaluk CA. Cdx2 as a marker of of the detection of intestinal metaplasia in non-dysplastic columnar-
epithelial intestinal differentiation in the esophagus. Am J Surg lined oesophagus. Scand J Gastroenterol. 2008;43:524-530.
Pathol. 2003;27:1442-1447. 75. Kelty CJ, Gough MD, Van Wyk Q, Stephenson TJ, Ackroyd R. Barrett’s
51. Chu PG, Jiang Z, Weiss LM. Hepatocyte antigen as a marker of oesophagus: intestinal metaplasia is not essential for cancer risk.
intestinal metaplasia. Am J Surg Pathol. 2003;27:952-959. Scand J Gastroenterol. 2007;42:1271-1274.
52. White NM, Gabril M, Ejeckam G, et al. Barrett’s esophagus and cardiac 76. Bhat S, Coleman HG, Yousef F, et al. Risk of malignant progression
intestinal metaplasia: two conditions within the same spectrum. Can in Barrett’s esophagus patients: results from a large population-based
J Gastroenterol. 2008;22:369-375. study. J Natl Cancer Inst. 2011;103(13):1049-1057. (Erratum: J Natl
53. Spechler SJ, Zeroogian JM, Antonioli DA, Wang HH, Goyal RK. Cancer Inst. 2013;105(8):581.)
Prevalence of metaplasia at the gastro-oesophageal junction. Lancet. 77. Westerhoff M, Hovan L, Lee C, Hart J. Effects of dropping the
1994;344:1533-1536. requirement for goblet cells from the diagnosis of Barrett’s esophagus.
54. Hirota WK, Loughney TM, Lazas DJ, Maydonovitch CL, Rholl V, Clin Gastroenterol Hepatol. 2012;10:1232-1236.
Wong RKH. Specialized intestinal metaplasia, dysplasia and cancer 78. Bandla S, Peters JH, Ruff D, et al. Comparison of cancer-associated
of the esophagus and esophagogastric junction: prevalence and genetic abnormalities in columnar-lined esophagus tissues with and
clinical data. Gastroenterology. 1999;116:277-285. without goblet cells. Ann Surg. 2014;260:72-80.
55. Ruol A, Parenti A, Zaninotto G, et al. Intestinal metaplasia is the 79. Shaheen NJ, Dulai GS, Ascher B, et al. Effect of a new diagnosis
probable common precursor of adenocarcinoma in Barrett esophagus of Barrett’s esophagus on insurance status. Am J Gastroenterol.
and adenocarcinoma of the gastric cardia. Cancer. 2000;88:2520-2528. 2005;100:577-580.
56. Cameron AJ, Souto EO, Smyrk TC. Small adenocarcinomas of the 80. Bansal A, Fitzgerald RC. Biomarkers in Barrett’s esophagus: role in
esophagogastric junction: association with intestinal metaplasia and diagnosis, risk stratification, and prediction of response to therapy.
dysplasia. Am J Gastroenterol. 2002;97:1375-1380. Gastroenterol Clin North Am. 2015;44:373-390.
CHAPTER
Epidemiology of Barrett Esophagus and Risk Factors
for Progression 32
Oliver M. Fisher
| Reginald V.N. Lord
PREVALENCE AND INCIDENCE OF BARRETT Although puzzlingly much lower than the incidence
ESOPHAGUS rate increase for EAC, studies from several countries have
reported an increase in the incidence of BE.7–11 This seems
Barrett esophagus (BE) is the disease in which the normal to be a true increase as only some of this rise in incidence
squamous lining of the distal esophagus is replaced by a can be attributed to increased use of endoscopy.7,8
metaplastic columnar cell epithelium (termed intestinal
metaplasia [IM]) in response to chronic severe gastro- RISK FACTORS FOR BARRETT ESOPHAGUS
esophageal reflux disease (GERD). Its clinical importance is
as the precursor lesion and major risk factor for esophageal FACTORS ASSOCIATED WITH INCREASED RISK OF
adenocarcinoma (EAC). The epidemiology of EAC has
been investigated thoroughly because of this cancer’s high
BARRETT ESOPHAGUS
lethality and recent extraordinary increase in incidence Sex and Age
rates. Less extensive and less consistent data are available Most epidemiologic data show a 2 : 1 male preponderance
for the epidemiology of BE, which is a difficult disease to among diagnosed cases of BE.5,12 BE patients are also
study in populations, because the majority of individuals typically slightly older than non-BE GERD patients at
with BE are undiagnosed as they have not undergone between 50 and 65 years at the time of diagnosis.5,13,14
upper gastrointestinal endoscopy. Access to endoscopy BE is rare in pediatric populations, being confirmed
is influenced by demographic and other factors such as by histologic presence of IM in approximately 0.12%
the availability of public health care. BE epidemiology of patients less than 20 years of age referred for upper
studies may thus suffer from population selection bias; endoscopy for any indication.15 There are no reports of
for example, the possibility that diagnosed BE cases IM containing BE in a child under the age of 5 years,16
have a different profile to undiagnosed cases or other and within children/adolescent study cohorts, the risk of
confounding factors. BE is increased after the age of 12 years,15 supporting a
Considering these difficulties, it is not surprising that, timeline of some years for BE development.
as shown in Table 32.1, the reported estimates of BE
prevalence vary widely. Studies designed to reduce the Gastroesophageal Reflux
risk of bias include an autopsy study1 and a study that Chronic GERD is the main cause of BE,17 with the risk and
screened for BE in colonoscopy patients with and without length of Barrett disease correlating with the amount and
heartburn2; and the estimated prevalence of BE in patients duration of reflux exposure in the distal esophagus.18–20
without reflux symptoms in these studies is 0.4% to 6%. The exact mechanism by which GERD triggers Barrett
A widely cited estimate of BE prevalence is from an formation remains elusive, but some have postulated
endoscopic screening study of 1000 unselected residents that the refluxate causes an erosive episode during which
who underwent upper gastrointestinal endoscopy in two there is a denuding of the normal squamous epithelium,
communities in northern Sweden.3 Endoscopic findings allowing it to be subsequently repopulated by columnar
of possible BE were present in 10.3% of individuals, with cells. Where these cells are derived from remains a matter
BE (IM) confirmed by histopathology in 1.6% of cases. of ongoing scientific debate.21,22
Thus a standard estimate is that Barrett disease affects Clinical studies have confirmed that severe GERD
approximately 1% to 2% of Western populations. is present in patients with BE and that these patients
Some data indicate that BE affects white non-Hispanic have more esophageal dysmotility and lower distal
Westerners more than Asian or Hispanic people,4,5 but esophageal sphincter pressures compared with patients
a recent meta-analysis including 51 studies with over with erosive or nonerosive esophagitis.18,23,24 Equally, other
greater than 450,000 patients from Asia suggests that studies have shown that BE patients have long exposure
the histologically proven pooled-prevalence of BE is to gastric content with very low pH levels (pH < 2.0 to
1.3% (95% confidence interval [CI], 0.7 to 2.2%; 28 3.0)25 and a high frequency of hiatal hernia (76% in BE
studies, and 298,850 subjects) and thus comparable to vs. 36% in GERD patients).23,24,26,27 In addition to gastric
Western estimates.6 Allowing for the limitations of pooled acid, duodenal juices in the refluxate are thought to be
analyses, such as varying study time periods, BE definitions, important contributors to the formation and propagation
and study populations (the meta-analysis only included of BE.28 The combined exposure of esophageal cells to
one population-based study with 1029 participants), this bile acids and low pH results in DNA damage and oxida-
suggests that BE is not uncommon in Asian (particularly tive stress,29 which in turn propagates BE formation and
Eastern Asian) countries. progression. Similarly, in vitro treatment of esophageal
323
Epidemiology of Barrett Esophagus and Risk Factors for Progression CHAPTER 32 323.e1
ABSTRACT
A standard estimate is that Barrett esophagus (BE) is
present in 1% to 2% of adults in Western populations,
and possibly also in Asian populations, but prevalence
is difficult to estimate as most cases are undiagnosed in
the community. The incidence of BE is increasing but
at a lower rate than for esophageal adenocarcinoma
(EAC). Older age, male sex, presence of a hiatus hernia,
central adiposity, tobacco smoking, and a family history
of gastroesophageal reflux disease, BE, or EAC are risk
factors for BE. Patient height, gastric Helicobacter pylori
infection, and nonsteroidal antiinflammatory drug use are
associated with a decreased risk of BE. Population studies
have in general not found that either acid suppression
medical therapy or antireflux surgery protect against BE
progression. The annual risk of EAC in patients with
BE is often cited as approximately 0.5% per year. The
risk may be lower than this for patients with BE without
dysplasia, and dysplasia remains the major risk factor for
progression. Longer BE segment length and most of the
risk factors for BE development are also risk factors for BE
progression to EAC. The molecular aberrations involved
in the BE to EAC process are exceedingly complex, but
a simple test, p53 protein immunostaining, is reported
to help predict risk of BE progression.
KEYWORDS
Barrett esophagus; epidemiology; risk factors; esophageal
adenocarcinoma; esophageal neoplasms; incidence;
prevalence; antireflux surgery
TABLE 32.1 Summary of Studies Estimating the Prevalence of Barrett Esophagus in Different Populations
324
BE BE
BE Prevalence Prevalence
Prevalence in GERD in Non-GERD Potential Source
Author Year N Study Design Study Population Ethnicity Estimate Patients Patients of Bias Comments
Winters 1987 97 Prospective, Patients with ≥1×/week — 12.40% 12.40% — Selection bias BE specialized epithelium
et al. observational symptoms of GERD was histologically
study confirmed in 50% of
the patients determined
as having BE. Columnar
gastric and junctional
epithelium was also
regarded as “BE” if it
SECTION I Esophagus and Hernia
was confirmed to be in
the tubular esophagus
and/or ≥5 cm proximal
of the gastroesophageal
junction.
Mann et al. 1989 180 Prospective, GERD patients with or “Vast majority 11.00% 11.00% — Selection bias —
cohort study without reflux Caucasian”
esophagitis
Cameron 1990 959 Prospective, Population-based — 0.34% — — — While retrospective in
et al. observational study + prospective nature, this paper is
study search of Mayo regarded as one of the
including Clinic autopsy first articles to suggest
autopsy data material that most Barrett
esophagus patients are
unrecognized.
Clark 1997 241 Prospective, GERD patients — 29.00% 29.00% — Selection bias Biased population as
et al.228 observational undergoing upper patients have GERD
study endoscopy symptoms and 68% of
study population males.
Voutilainen 2000 1128 Prospective, Patients referred for — 1.00% 4.40% — Selection bias BE prevalence estimate is
et al.229 observational upper endoscopy only applicable to
study from primary care patients presenting with
setting with GERD/ upper GI symptoms.
dyspepsia and other
upper-GI symptoms
warranting
endoscopic
investigation
Gerson 2002 110 Prospective, “Asymptomatic” 73% Caucasian, 25.00% — — Selection bias Only 53% of participants
et al.230 observational individuals 14% African had no GERD
study presenting for American, 10% symptoms; biased
screening Hispanic, 4% population as 92%
sigmoidoscopy Pacific Islander/ males with mean age of
Asian >60 years.
Rex et al. 2003 961 Prospective, Patients undergoing 78% Caucasian, 6.80% 8.30% 5.60% Selection bias Biased population as
observational colonoscopy 20.3% black, presenting for
study 1.6% Latin- colonoscopy, 78%
American/Asian white and
predominantly male
(~60%).
Malfertheiner 2005 6215 Prospective, Patients undergoing — 8.39% 8.39% Selection bias In patients who had
et al. observational upper endoscopy for erosive reflux disease
study GERD symptoms prevalence of Barrett
was up to 14%,
whereas in patients with
NERD BE prevalence
was 2.3%. This study
population is also
biased toward GERD
patients.
Ronkainen 2005 1000 Prospective, Random population — 1.60% 2.30% 1.20% — Regarded as a robust
et al. multicenter sample invited to assessment of the true
cohort study participate in prevalence of BE in a
screening upper general population.
endoscopy Problems exist with
regard to
generalizability as only
two northern Swedish
communities were
studied.
Westhoff 2005 378 Prospective, GERD patients 86% white 13.20% 13.20% — Selection bias 86% white and 95%
et al. cohort study presenting for male, with median age
first-time upper of 56 years, all of whom
endoscopy had GERD.
Veldhuyzen 2006 1040 Prospective, Dyspepsia patients 95% Caucasian, 2.40% 2.40% — Selection bias Analyses patients who
van cohort study recruited from 2% black, 1% were promptly referred
Zanten primary care setting Asian, 1% for endoscopy from a
et al.231 Aboriginal/Metis primary care setting;
selection bias exists for
true BE population
prevalence estimates,
as these are
“symptomatic” patients.
Continued
Epidemiology of Barrett Esophagus and Risk Factors for Progression CHAPTER 32
325
326
TABLE 32.1 Summary of Studies Estimating the Prevalence of Barrett Esophagus in Different Populations—cont’d
BE BE
BE Prevalence Prevalence
Prevalence in GERD in Non-GERD Potential Source
Author Year N Study Design Study Population Ethnicity Estimate Patients Patients of Bias Comments
Corley et al. 2008 4205 Observational Evaluation of BE Study population 0.13% — — Potential This study documents
study diagnosis among all included diagnostic that, in a fairly unbiased
patients with a Caucasians, bias as population, the
SECTION I Esophagus and Hernia
estimated.
BE, Barrett esophagus; EAC, esophageal adenocarcinoma; GERD, gastroesophageal reflux disease; GI, gastrointestinal; SEER, Surveillance, Epidemiology and End Results Reporting database.
328 SECTION I Esophagus and Hernia
cells with both bile and acid leads to the expression of studies found an increased risk of BE if leptin levels
intestinal and/or columnar cell markers.30–32 were elevated, but the association was stronger in men in
While GERD can affect up to 20% of Western popula- one study9 and stronger in women in the other study.59
tions,33 only 5% to 10% of patients with GERD develop Another study did not show an association with leptin,
BE.34–39 A higher frequency (e.g., ≥weekly) of GERD symp- but it did show a significant inverse association of elevated
toms is associated with a greatly (10 times) increased risk of adiponectin levels and BE risk compared with colon
BE in population control studies.40,41 A weaker association screening controls (adjusted OR, 0.42; 95% CI, 0.22 to
between symptoms and BE is found in endoscopy studies 0.80).65 Thus while recent studies on the interaction of
of patients with typical symptoms of GERD.13,40,42 It thus central obesity and EAC have elucidated the molecular
remains unclear if BE patients have considerably more pathways by which adipocytokines contribute to EAC
frequent or only slightly more frequent symptoms of reflux formation and propagation, more data from larger studies
compared with non-BE GERD patients.3,38 Frequency are required to further clarify the role for leptin and other
and number of years of reflux symptoms (i.e., chronic- obesity-related cytokines in BE development.
ity of reflux) are better predictors of BE.43,44 Patients
with BE may report that their GERD symptoms have Smoking and Alcohol Consumption
improved in recent years, which is postulated to be related Most, but not all, population-based studies have docu-
to BE development and perhaps reduced esophageal mented an approximately twofold increase in risk of
sensitivity. BE in patients who have ever smoked.40–42,66,67 Contrary
to some earlier findings,68 recent data suggest that a
Obesity greater number of pack-years smoked may be associated
There is a strong reproducible association between obesity with a greater risk of BE,67 but this effect may plateau
and the risk for EAC, indicating that patients with a body at approximately 20 pack-years.69 Furthermore, a recent
mass index (BMI) ≥ 30 kg/m2 bear a 2 to 3 times higher population-based study showed that smoking and GERD
risk of developing this malignancy,45–49 but the associations may exert a synergistic effect on disease development and
with BE have been less consistent. A recent meta-analysis50 progression.49 There are no convincing data supporting an
concluded that there was no significant association with increased risk of BE due to alcohol consumption. On the
BMI and BE when comparing BE and GERD patients. contrary, an inverse association of wine consumption and
However, when using the pooled estimates from three risk of BE has been reported70,71 and a pooled analysis of
studies comparing BE patients with general population Barrett cases compared with controls found a moderate
controls, a significant association between BMI and risk reduction of risk with consumption of wine (OR, 0.71;
for BE could be determined (pooled odds ratio [OR], 1.02 95% CI, 0.60 to 1.00), although without a dose-response
per kg/m2; 95% confidence interval [CI], 1.01 to 1.04; I2 relationship.72
= 0%). Other data from population-based studies suggest
that there is no more than a 50% increase in risk for BE Family History and Genetic Predisposition
with BMIs 30 kg/m2 or greater.40,42,45,51–53 There is most probably a heritable or familial aspect
Central visceral adiposity may be a more important for BE and EAC development in some cases.73,74 This is
risk factor for BE formation than BMI itself.51,52,54 A case- supported by the following observations: concordance
control study found that visceral adipose tissue was 1.5 of cases in both mono- and dizygous twins,75–77 increased
times higher in patients with BE compared with controls54 disease risk in patients with a positive family history,78–81
and population-based studies have identified a significant and the identification through genome-wide association
association between BE and either increasing waist cir- studies (GWAS) of single-nucleotide polymorphisms
cumference51 or waist-to-hip ratio.52 Interestingly, when (SNPs) in genes that render individuals susceptible to
controlling for waist-to-hip ratio, the association between the development of BE and EAC.82–84 SNPs affecting at
BMI and BE has been shown to be almost completely least seven genes (MHC locus, FOXF1, CRCT1, BARX1,
attenuated, suggesting that the obesity effect is mediated FOXP1, TBX5, and GDF7), some of which are involved
through visceral adiposity.52 in esophageal development and inflammatory response
A mechanistic explanation for the relationship with processes, are thought to be significantly associated with
central adiposity includes the unproven postulates that the risk of BE and EAC development.82–84
increased abdominal visceral fat elevates intraabdominal The prevalence estimates of familial Barrett cases vary,
pressure55 or intragastric pressure.56 More likely is that the with one series reporting confirmed family cases of BE in
pressure gradient across the antireflux barrier is increased, 6% of subjects,79 whereas another study found a familial
resulting in hiatus hernia and GERD, and the metabolic prevalence rate of up to 24%.80 The aforementioned
and endocrine activity of visceral adipose tissue in those twin studies have suggested a heritability of up to 30% to
with central obesity (who interestingly are more likely to 40%.75–77 One study estimated 35% (standard error [SE],
be male)9,57–59 may also be important. Central obesity alters 6%) of overall BE risk variance to be explained by common
expression levels of obesity-related and proinflammatory germline genetic variants,85 whereas another recent report
cytokines such as leptin, adiponectin, tumor necrosis provided a more conservative estimate of 9.99% (SE,
factor (TNF)-alpha, interleukin (IL)-6, and insulin-like 1.2%).84 Similarly, another recent study estimated 7%
growth factor.60–63 (95% CI, 3% to 11%) of phenotypic variance for GERD
As leptin is upregulated in obesity and increases pro- to be explained by inheritable SNPs.86 More importantly,
liferation of EAC cells in vitro,64 the association between however, this study also showed that there is approximately
serum leptin levels and BE has been examined.9,59,65 Two 77% (SE, 24%) genetic correlation between GERD and
Epidemiology of Barrett Esophagus and Risk Factors for Progression CHAPTER 32 329
BE and approximately 88% (SE, 25%) between GERD and stomach, colon, and rectum.94,95 Combined observational
EAC, thus providing first evidence for a polygenic basis data suggest a reduction of 44% to 58% in esophageal
for GERD and supporting a polygenic overlap for GERD cancer mortality rates in aspirin users.95 NSAIDs, includ-
with both BE and EAC.86 Taken together, these data also ing aspirin, inhibit the enzymes cyclooxygenase (COX)
suggest that much of the genetic basis for EAC may lie 1 and 2. Elevated COX-2 expression, as found in BE
in the development of BE, rather than in the progression and EAC96–99 can be induced by inflammation, growth
of BE to EAC. It is thus important to assess a complete factors, mitogens, and other cytokines. COX-2 inhibition
family history for patients with BE or EAC. can restore apoptosis100,101 and inhibit cell growth and
proliferation101,102 and neoangiogenesis.103 The potential
FACTORS ASSOCIATED WITH DECREASED RISK OF for COX-2 inhibition through aspirin or other NSAID
administration to prevent BE progression is supported by
BARRETT ESOPHAGUS cohort and case-control studies,104 but further evidence is
Patient Height needed, and the risk/benefit ratio is unclear.105 Whether
A study including 999 cases of EAC, 2061 cases of BE, and this should be routine therapy in patients with BE will
2168 population controls found that height is inversely hopefully become evident with the results of a large (5000
associated with the risk of BE (OR, 0.69; 95% CI, 0.62 to patients) randomized-controlled aspirin and proton pump
0.77 for every 10-cm increase in height) and EAC (OR, inhibitor (PPI) trial in the United Kingdom (The Aspirin
0.70; 95% CI, 0.62 to 0.79 for each 10 cm increase in Esomeprazole Chemoprevention Trial, AspECT; UKCRN
height). This effect was irrespective of gender and the ID 1339, anticipated trial end date October 2018).106,107
estimates were similar across all strata of age, patient Patients who take statin medications for treatment of
education levels, BMI, weight, and whether or not the hypercholesterolemia or as preventive therapy for coronary
patients had GERD symptoms or if they smoked.87 heart disease may have a reduced risk of developing both
BE and EAC.108–110 A meta-analysis suggests that statin
Helicobacter pylori Infection use is significantly associated with a reduced risk of BE
The historic decline of Helicobacter pylori infection preva- compared with controls with a pooled OR of 0.63 (95%
lence in developed countries is temporally aligned with CI, 0.51 to 0.77; 1090 Barrett cases vs. 2085 controls).109
an increased incidence in GERD complications such as A preventive role against BE progression is suggested by
BE and EAC. One explanation for this has been that a recent nested case-control study with more than 1000
H. pylori infections reduce intragastric acidity through US veterans, which found that BE patients who develop
the generation of ammonia or by causing severe corpus EAC are less likely to have used statins compared with BE
gastritis with concomitant destruction of gastric parietal patients who never develop cancer (40.2% vs. 54.0%; P
cells, thus reducing acid production and protecting against < .01).111 With further data, the routine use of NSAIDs
GERD complications.88 While infection with H. pylori or statins for chemoprevention in Barrett disease may be
(particularly CagA+ strains) has consistently been shown recommended.112
to reduce risks of EAC,89 the evidence for this effect has
been less consistently shown for BE.90 A first meta-analysis Acid Suppressive Medical Therapy and
of 12 case-control studies published in 2009 showed that Antireflux Surgery
overall there was no significant difference in H. pylori The main treatment options for GERD are medical acid
infection rates between BE and controls (42.9% vs. 43.9%; suppression therapy and antireflux surgery by fundoplica-
OR, 0.74; 95% CI, 0.40 to 1.37), although in a subgroup tion. Whether medical or surgical therapies protect patients
analysis comparing BE patients with endoscopically with GERD against the development of BE or against
normal controls, H. pylori infection was significantly less BE progression to EAC remains highly contentious.113 A
frequent in BE patients (23.1% vs. 42.7%; OR, 0.50; 95% Cochrane review that pooled data from two randomized
CI, 0.27 to 0.93). An explanation for these discrepan- controlled trials (RCTs) failed to show any major protective
cies may be that many of the included studies provided effect of PPI therapy on BE progression,114 and it is thus
heterogeneous effect estimates, as they were prone to reasonable to state that PPIs are indicated for symptom
selection and information biases. A later meta-analysis91 relief but that they are not BE chemopreventive agents.112
included only four high-quality studies,2,3,92,93 finding an With regard to antireflux surgery, a thorough prospec-
overall protective effect of H. pylori infection on BE risk tive trial from Sweden in particular115 and other surgical
compared with population-based controls (OR, 0.46; 95% series suggest that fundoplication promotes BE regres-
CI, 0.35 to 0.60). Importantly, two of these studies92,93 also sion and prevents dysplasia formation.116–118 Two RCTs
showed that this risk was reduced even when controlling comparing surgery versus medical therapy concluded that
for reflux symptoms, hence suggesting that the protective both provided good symptom control, but slightly better
effect of H. pylori infection cannot be explained simply outcomes were observed in the surgical group.119,120 Four
by reduced gastric acid production. The exact mechanism meta-analyses have examined whether medical therapy or
by which H. pylori infection exerts a protective mechanism surgery is better at preventing BE development and/or
on BE formation thus remains uncertain. progression to EAC.113,121–123 Two of these studies found
no protective effect of antireflux surgery for the develop-
Nonsteroidal Antiinflammatory Drugs and Statins ment of EAC compared with medical therapy,121,122 and a
There is compelling evidence that nonsteroidal anti- third study concluded that because of an observed high
inflammatory drugs (NSAIDs), including aspirin, may postoperative tumor progression rate, antireflux surgery
prevent a variety of cancers, particularly of the esophagus, does not prevent EAC formation in BE patients.123 These
330 SECTION I Esophagus and Hernia
three meta-analyses had methodologic limitations, as each recognition of short-segment BE (SSBE, BE < 3 cm in
type of therapy was pooled into one group irrespective of length) as true Barrett disease,130 including only patients
whether the original studies were comparative. Thus the with long-segment BE (LSBE) is likely to increase EAC
most recent meta-analysis, published in 2016 by Maret- risk estimates.129,131
Ouda,113 aimed to have more rigorous inclusion criteria A widely used estimate of the annual incidence of EAC
by only including comparative studies that contained both in patients with BE was approximately 0.5%, or 1 in 200
treatment arms. This systematic review included 10 studies patients per year,128,132 although it has been noted that even
comparing the risk of EAC after antireflux surgery with this low frequency of EAC may not be observed in clinical
nonoperated GERD patients, 7 studies including patients practice.133 Both Wani et al.134 and Sikkema et al.135 also
with BE and 2 studies comparing the EAC risk following estimated an annual incidence of EAC of approximately
surgery to the background population. The study found 0.6% in pooled analyses but included some patients with
that the risk of EAC was not significantly reduced in a baseline diagnosis of low-grade dysplasia (LGD). When
operated patients compared with medically treated GERD patients with early incident EACs and a baseline diagnosis
patients irrespective of if they had BE or not. However, of high-grade dysplasia (HGD) were excluded, as in the
if the studies were restricted to those performed after study by Yousef et al.,136 an annual EAC incidence rate of
2000, then a significant reduction in pooled incidence 0.41% was estimated. Subsequently, another meta-analysis
rate ratios (IRR) could be found in surgically treated BE by Desai et al. aimed to determine the risk of BE progres-
patients compared with medically treated patients (IRR, sion in patients with a baseline diagnosis of nondysplastic
0.26; 95% CI, 0.09 to 0.79). The authors also found that BE (NDBE).129 This study combined data from 57 studies
the risk for EAC does not revert to that of the back- with 11,434 histologically confirmed cases of NDBE and
ground population for surgically treated patients (IRR, a total of 58,547 patient-years of follow-up. With a total
10.78; 95% CI, 8.48 to 13.71), indicating that patients of 186 incident cases of EAC during follow-up, a pooled
undergoing antireflux surgery need ongoing endoscopic incidence rate of 0.33% (95% CI, 0.28 to 0.38%) was
surveillance. In summary, there is a discrepancy between estimated. This estimate remained the same when only
single-center surgery studies finding BE regression after the highest quality studies were included in a post hoc
fundoplication, whereas this is not shown by multicenter sensitivity analysis (0.33%; 95% CI, 0.26 to 0.40%). This
and population data studies that have so far failed to risk of EAC dropped further in a subgroup analysis of 16
show that this occurs more frequently than expected by studies focusing on patients with nondysplastic SSBE (967
chance alone. patients and 4456 patient-years of follow-up), in whom
an annual EAC incidence rate of 0.19% (95% CI, 0.08 to
Nutrition 0.34%) was estimated. Importantly, this study also found
High intakes of fruits, vegetables, fibers, and even meat that patients with NDBE were more than 10 times more
may be inversely associated with the risk of developing likely to die of another cause than to develop EAC.
BE, whereas high intake of trans-fats may be linked to an Four large-scale population-based studies that include
increased risk of BE.124 Another study in the same cohort more than 30,000 study subjects have been published,
found dietary antioxidants, vitamin C, beta-carotene, and essentially confirming this low incidence rate with annual
vitamin E are also associated with a reduced risk of BE, cancer risks ranging from 0.12% to 0.43% in patients
but the use of antioxidant supplements did not influence with BE.137–140 Thus as illustrated in Fig. 32.1, there has
BE risk.125 A recent population-based study from Ireland been a continuous downward trend in estimates of EAC
suggests that high dietary intake of magnesium significantly incidence rates in patients with BE over recent decades;
reduces the risk of reflux esophagitis and BE (adjusted it is currently considered that the risk of developing EAC
OR, 0.31; 95% CI, 0.11 to 0.87 and adjusted OR, 0.29; in nondysplastic BE is low at approximately 1 case per
95% CI, 0.12 to 0.71, respectively) and that this effect is 300 to 500 patients per year and that BE patients’ risk of
most prominent in the setting of a low calcium:magnesium mortality is increased due to other causes.
intake ratio.126 As a result of these newer data, the effectiveness of
routine surveillance endoscopy among patients with
histologically confirmed non- and never dysplastic BE is
RISK FACTORS FOR THE NEOPLASTIC increasingly being questioned,141–144 particularly in light of
other data showing the inadequacies of current surveillance
PROGRESSION OF BARRETT ESOPHAGUS protocols: up to 60% of incidence EAC cases are diagnosed
within 1 year of the diagnosis of BE, indicating that they
OVERALL RISK OF NEOPLASTIC are probably missed at index endoscopy.138 Current clinical
BARRETT PROGRESSION practice (the Seattle protocol) includes four-quadrant
Estimates of the risk of cancer development in patients biopsies every 1 to 2 cm.145 But even such rigorous biopsy
with BE range from approximately 0% to 3% per patient- strategies typically sample less than 5% of the Barrett
year.127,128 The reasons for higher risk estimates have epithelium, thus rendering sampling error unavoidable.146
been discussed widely127–129 and include factors such as Furthermore, up to 90% of endoscopists do not adhere
publication bias and not accounting for the presence to this laborious surveillance protocol.147 These factors
of baseline dysplasia, which is an intermediate stage of demonstrate the importance of identifying patients at
EAC development.129 Many of the observational studies higher risk of BE progression through determining clinical
included in earlier meta-analyses of BE progression risk and molecular risk factors of progression, as discussed in
were from the 1980s, which is prior to the widespread the subsequent section.
Epidemiology of Barrett Esophagus and Risk Factors for Progression CHAPTER 32 331
1.00 Study size Tobacco use is a risk factor for EAC. 49 The risk is
Small increased for both current (OR 2.3; 95% CI, 1.5 to 3.5)
Intermediate and former smokers (OR 1.5; 95% CI, 1.1 to 2.1) and
Large the risk of EAC increases with cumulative exposure.49
Mean cancer incidence rate
0.75
Smoking has also been found to increase the risk of
progression to HGD and EAC in patients with BE (hazard
ratio [HR], 2.03; 95% CI, 1.2 to 3.17).158 Importantly,
0.50 this risk of progression persists across all strata of
smoking intensity, and current smokers display the
highest risk of malignant BE progression.158 Alcohol
consumption, on the other hand, has not consistently
0.25
shown an association with increased risk of EAC and/or
BE progression.47,49,71
0.71 0.64 0.24
0.00 Recurrent Gastroesophageal Reflux Disease
Large-scale population-based studies suggest that
<1990 1990–2000 >2000 approximately 40% of EAC cases arise in patients with
frequently recurring symptoms of GERD, noticed as
FIGURE 32.1 Changes in estimates of the annual risk of heartburn and/or regurgitation of gastric acid/content
esophageal adenocarcinoma formation in patients with Barrett
at least once a week.159–161 The EAC risk also increases
esophagus over time. Study sizes have been classified as small
with longer symptom duration: patients with 20 years
(blue), intermediate (red), or large (green) depending on the
number of patient follow-up years. Small studies had less than
or longer of reflux symptoms have an almost 3 times
500, intermediate 500 to 1000, and large greater than 1000 higher risk of EAC compared with those with less than
patient-years of follow-up. The figure shows how with increasing 10 years’ duration.162 Moreover, patients who have reflux
study cohort size/follow-up the annual cancer incidence rate has symptoms have an approximately 6 times higher risk of
decreased. EAC compared with those without symptoms. 161 The
clinical relevance of this association, especially for EAC
screening, is limited by the lack of typical GERD symptoms
CLINICAL RISK FACTORS FOR THE PROGRESSION in many patients with EAC, the prevalence of GERD
symptoms in the community (~20% of the population
OF BARRETT ESOPHAGUS report at least weekly symptoms of GERD),163 and the low
Age and Sex annual incidence rates for EAC of 2.7 per 100,000 in the
Men are at increased risk of developing EAC compared general population and 6.0 per 100,000 in white males.164
with women.148 A recent international registry analysis It is nevertheless clear that severe, long-standing and
including over 100,000 cases of EAC showed that only frequent GERD are associated with an increased risk of
22.3% of cases occurred in women, and male-to-female EAC. Consequently, and despite uncertainty with regard to
ratios of up 6 : 1 were reported.149 However, there is some effectiveness, current management guidelines recommend
paucity of data regarding the association of male sex and BE patients with severe and uncontrolled symptoms of
the risk of BE progression to cancer. One of the most reflux require acid suppressive therapy to help prevent
recent population-based studies indicates that men with progression.112,152,153
BE are 2 to 4 times more likely to develop EAC compared
with females.137,150 Barrett Segment Length
The risk of EAC increases with age,149 typically peaking The risk of neoplastic progression increases with the
around the sixth to seventh decade of life.46,137 The dura- length of Barrett epithelium.129,131,151 This was elegantly
tion of patients’ diagnosis of BE is also a risk factor for demonstrated in the study from Germany by Pohl et al.131
malignant progression, with patients having BE for 10 or in which the annual cancer progression rates of patients
more years displaying a greater than 2 times higher risk with long-segment (≥3 cm), short-segment (≥1 to <3 cm),
of HGD/EAC development compared with those with a and ultra-short segment (<1 cm) BE were 0.22%, 0.03%,
diagnosis less than 10 years.151 Consequently, while screen- and 0.01%, respectively. These data imply that to detect
ing strategies have not been formally adopted, it is widely one case of EAC, 450 patients with LSBE would require
accepted that (Caucasian) men over the age of 50 years annual surveillance endoscopy, whereas in SSBE and ultra-
with BE are at increased risk of developing EAC.112,145,152,153 short BE, the corresponding number of patients increases to
3440 and 12,364, respectively. As a result of these differences
Obesity and Smoking in cancer incidence rates, recent guidelines suggest that
EAC is a cancer with one of the strongest associations patients with greater than 3 cm of Barrett segment length
with obesity,154,155 whether measured as increased BMI and no dysplasia should undergo screening endoscopy with
or abdominal obesity.155,156 Whether increased BMI also quadrantic biopsies every 2 to 3 years, whereas in patients
translates to a higher risk of malignant progression in with SSBE, the screening interval can be extended to every
patients with BE is not clear, but some data suggest 3 to 5 years.153 If a nodule, ulcer, or stricture is present,
that, particularly in males, increased waist-to-hip ratio however, there is a high risk of malignancy irrespective of
and waist circumference may confer a higher risk of BE Barrett segment length and reassessment without delay
progression.157 is needed.165
332 SECTION I Esophagus and Hernia
supported by studies showing that increased clonal diversity 4. Abrams JA, Fields S, Lightdale CJ, Neugut AI. Racial and ethnic
as measured by the number of unique clones, Shannon disparities in the prevalence of Barrett’s esophagus among
patients who undergo upper endoscopy. Clin Gastroenterol Hepatol.
diversity index, and genetic divergence of clones, increases 2008;6(1):30-34.
the risk of EAC even when DNA content abnormalities 5. Corley DA, Kubo A, Levin TR, et al. Race, ethnicity, sex and
and 17p LOH are taken into account.205 A recent study temporal differences in Barrett’s oesophagus diagnosis: a large
on clonal diversity in Barrett suggests that there is strong community-based study, 1994–2006. Gut. 2009;58(2):182-188.
6. Shiota S, Singh S, Anshasi A, El-Serag HB. Prevalence of Barrett’s
clonal selection in NDBE disease over time, implying that esophagus in Asian countries: a systematic review and meta-analysis.
the malignant potential of these lesions may be largely Clin Gastroenterol Hepatol. 2015;13(11):1907-1918.
predetermined.206 7. Prach AT, MacDonald TA, Hopwood DA, Johnston DA. Increasing
Finally, a variety of complex molecular regulatory incidence of Barrett’s oesophagus: education, enthusiasm, or
pathways have been suggested to be disrupted during the epidemiology? Lancet. 1997;350(9082):933.
8. Conio M, Cameron AJ, Romero Y, et al. Secular trends in the
development of EAC,178 as evidenced by altered transcrip- epidemiology and outcome of Barrett’s oesophagus in Olmsted
tomic,207–213 methylomic214–218 and proteomic profiles.219–222 County, Minnesota. Gut. 2001;48(3):304-309.
These types of studies are helping elucidate the alterations 9. Kendall BJ, Macdonald GA, Hayward NK, et al. Leptin and the
in regulatory pathways, including disturbances in critical risk of Barrett’s oesophagus. Gut. 2008;57(4):448-454.
10. van Soest EM, Dieleman JP, Siersema PD, Sturkenboom MCJM,
cycle and proliferation processes,223 that are involved in Kuipers EJ. Increasing incidence of Barrett’s oesophagus in the
BE progression. Whole-transcriptome RNA-sequencing has general population. Gut. 2005;54(8):1062-1066.
shown that aside from protein coding genes interacting 11. Coleman HG, Bhat S, Murray LJ, McManus D, Gavin AT, Johnston
in complex gene regulatory networks, a plethora of noncod- BT. Increasing incidence of Barrett’s oesophagus: a population-based
ing genomic elements such as long noncoding RNAs, and study. Eur J Epidemiol. 2011;26(9):739-745.
12. Cook MB, Wild CP, Forman D. A systematic review and meta-analysis
repeat elements are also differentially expressed in EAC of the sex ratio for Barrett’s esophagus, erosive reflux disease, and
compared with less advanced BE stages.213 nonerosive reflux disease. Am J Epidemiol. 2005;162(11):1050-1061.
While encouraging new data for biomarker panels 13. Edelstein ZR, Bronner MP, Rosen SN, Vaughan TL. Risk factors
exist,207,217,224,225 these biomarkers have not progressed past for Barrett’s esophagus among patients with gastroesophageal
reflux disease: a community clinic-based case–control study. Am J
stages 2 or 3 of the National Institutes of Health (NIH)/ Gastroenterol. 2009;104(4):834-842.
National Cancer Institute (NCI) Early Detection Research 14. Eloubeidi MA, Provenzale D. Clinical and demographic predictors of Bar-
Network (EDRN; https://edrn.nci.nih.gov/) clinical rett’s esophagus among patients with gastroesophageal reflux disease:
biomarker development stages,226 indicating that they are a multivariable analysis in veterans. J Clin Gastroenterol. 2001;33(4):
largely preclinical, lacking both external and prospective 306-309.
15. El-Serag HB, Gilger MA, Shub MD, Richardson P, Bancroft J.
validation.169,179 The largest body of evidence concerns the The prevalence of suspected Barrett’s esophagus in children and
potential clinical utility of p53 immunohistochemistry to adolescents: a multicenter endoscopic study. Gastrointest Endosc.
assess protein expression (as reviewed in reference 112). 2006;64(5):671-675.
Numerous studies have proposed that the addition of 16. Sherman PM, Hassall E, Fagundes-Neto U, et al. A global, evidence-
p53 immunohistochemistry may improve the diagnosis based consensus on the definition of gastroesophageal reflux disease
in the pediatric population. Am J Gastroenterol. 2009;104(5):1278-1295.
of dysplasia and improve patient stratification, as either 17. Falk GW. Barrett’s esophagus. Gastroenterology. 2002;122(6):1569-1591.
overexpression or loss of p53 protein expression seems 18. Iascone C, Demeester TR, Little AG, Skinner DB. Barrett’s esophagus.
to be an accurate predictor of the risk of BE progres- Functional assessment, proposed pathogenesis, and surgical therapy.
sion.112,227 The British Society for Gastroenterology is the Arch Surg. 1983;118(5):543-549.
first professional society to recommend the clinical use 19. Taylor JB, Rubenstein JH. Meta-analyses of the effect of symptoms
of gastroesophageal reflux on the risk of Barrett’s esophagus. Am
of p53 immunostaining to help guide management of J Gastroenterol. 2010;105(8):1729-1730; quiz 1738.
patients with BE.112 20. Fass R, Hell RW, Garewal HS, et al. Correlation of oesophageal acid
More objective means of disease evaluation through exposure with Barrett’s oesophagus length. Gut. 2001;48(3):310-313.
biomarkers remains the most viable approach to overcome 21. Phillips WA, Lord RV, Nancarrow DJ, et al. Barrett’s esophagus.
current management limitations as well as biases. This will J Gastroenterol Hepatol. 2011;26:639-648.
22. Souza RF, Krishnan K, Spechler SJ. Acid, bile, and CDX: the ABCs
ultimately lead to those patients who are assessed as being of making Barrett’s metaplasia. Am J Physiol Gastrointest Liver Physiol.
at low risk of progression based on their molecular and 2008;295(2):G211-G218.
clinical disease characteristics being spared laborious and 23. Öberg S, DeMeester TR, Peters JH, et al. The extent of Barrett’s
unnecessary surveillance, while those at highest risk of esophagus depends on the status of the lower esophageal sphincter
and the degree of esophageal acid exposure. J Thorac Cardiovasc
cancer development can be guided toward more aggressive Surg. 1999;117(3):572-580.
eradication treatments. This will reduce the individual 24. Lord RVN, DeMeester SR, Peters JH, et al. Hiatal hernia, lower
and societal burden of EAC. esophageal sphincter incompetence, and effectiveness of Nissen
fundoplication in the spectrum of gastroesophageal reflux disease.
J Gastrointest Surg. 2008;13(4):602-610.
REFERENCES 25. Stein HJ, Hoeft S, Demeester TR. Functional foregut abnormalities
in Barrett’s esophagus. J Thorac Cardiovasc Surg. 1993;105(1):107-111.
1. Cameron AJ, Zinsmeister AR, Ballard DJ, Carney JA. Preva- 26. Avidan B, Sonnenberg A, Schnell TG, Sontag SJ. Hiatal hernia
lence of columnar-lined (Barrett’s) esophagus. Comparison of and acid reflux frequency predict presence and length of Barrett’s
population-based clinical and autopsy findings. Gastroenterology. esophagus. Dig Dis Sci. 2002;47(2):256-264.
1990;99(4):918-922. 27. Cameron AJ. Barrett’s esophagus: prevalence and size of hiatal
2. Rex DK, Cummings OW, Shaw M, et al. Screening for Barrett’s hernia. Am J Gastroenterol. 1999;94(8):2054-2059.
esophagus in colonoscopy patients with and without heartburn. 28. Vaezi MF, Richter JE. Role of acid and duodenogastroesopha-
Gastroenterology. 2003;125(6):1670-1677. geal reflux in gastroesophageal reflux disease. Gastroenterology.
3. Ronkainen J, Aro P, Storskrubb T, et al. Prevalence of Barrett’s 1996;111(5):1192-1199.
esophagus in the general population: an endoscopic study. Gas- 29. Dvorak K, Payne CM, Chavarria M, et al. Bile acids in combination
troenterology. 2005;129(6):1825-1831. with low pH induce oxidative stress and oxidative DNA damage:
334 SECTION I Esophagus and Hernia
relevance to the pathogenesis of Barrett’s oesophagus. Gut. 53. Jacobson BC, Chan AT, Giovannucci EL, Fuchs CS. Body mass index
2007;56(6):763-771. and Barrett’s oesophagus in women. Gut. 2009;58(11):1460-1466.
30. Marchetti M, Caliot E, Pringault E. Chronic acid exposure 54. El-Serag HB, Kvapil P, Hacken-Bitar J, Kramer JR. Abdominal
leads to activation of the cdx2 intestinal homeobox gene in a obesity and the risk of Barrett’s esophagus. Am J Gastroenterol.
long-term culture of mouse esophageal keratinocytes. J Cell Sci. 2005;100(10):2151-2156.
2003;116(8):1429-1436. 55. Wilson LJ, Ma W, Hirschowitz BI. Association of obesity with hiatal
31. Bajpai M, Liu J, Geng X, Souza RF, Amenta PS, Das KM. Repeated hernia and esophagitis. Am J Gastroenterol. 1999;94(10):2840-2844.
exposure to acid and bile selectively induces colonic phenotype 56. Berstad A, Weberg R, Larsen IF, Hoel B, Hauer-Jensen M. Relation-
expression in a heterogeneous Barrett’s epithelial cell line. Lab ship of hiatus hernia to reflux oesophagitis: a prospective study of
Invest. 2008;88(6):643-651. coincidence, using endoscopy. Scand J Gastroenterol. 2009;21(1):55-58.
32. Fitzgerald RC, Omary MB, Triadafilopoulos G. Dynamic effects 57. Donohoe CL, Doyle SL, McGarrigle S, et al. Role of the insulin-
of acid on Barrett’s esophagus. An ex vivo proliferation and dif- like growth factor 1 axis and visceral adiposity in oesophageal
ferentiation model. J Clin Invest. 1996;98(9):2120. adenocarcinoma. Br J Surg. 2012;99(3):387-396.
33. Locke GR, Talley NJ, Fett SL, Zinsmeister AR, Melton LJ. Prevalence 58. Howard JM, Pidgeon GP, Reynolds JV. Leptin and gastro-intestinal
and clinical spectrum of gastroesophageal reflux: a population- malignancies. Obes Rev. 2010;11(12):863-874.
based study in Olmsted County, Minnesota. Gastroenterology. 1997; 59. Thompson OM, Beresford SAA, Kirk EA, Bronner MP, Vaughan TL.
112(5):1448-1456. Serum leptin and adiponectin levels and risk of Barrett’s esophagus
34. Winters C, Spurling TJ, Chobanian SJ, et al. Barrett’s esophagus. A and intestinal metaplasia of the gastroesophageal junction. Obesity
prevalent, occult complication of gastroesophageal reflux disease. (Silver Spring). 2010;18(11):2204-2211.
Gastroenterology. 1987;92(1):118-124. 60. Housa D, Housová J, Vernerová Z, Haluzík M. Adipocytokines and
35. Mann NS, Tsai MF, Nair PK. Barrett’s esophagus in patients with symp- cancer. Physiol Res. 2006;55(3):233-244.
tomatic reflux esophagitis. Am J Gastroenterol. 1989;84(12):1494-1496. 61. Kelesidis I, Kelesidis T, Mantzoros CS. Adiponectin and cancer: a
36. Malfertheiner P, Lind T, Willich S, et al. Prognostic influence of systematic review. Br J Cancer. 2006;94(9):1221-1225.
Barrett’s oesophagus and Helicobacter pylori infection on healing of 62. Garofalo C, Surmacz E. Leptin and cancer. J Cell Physiol. 2006;207(1):
erosive gastro-oesophageal reflux disease (GORD) and symptom 12-22.
resolution in non-erosive GORD: report from the ProGORD study. 63. Doyle SL, Donohoe CL, Finn SP, et al. IGF-1 and its receptor in
Gut. 2005;54(6):746-751. esophageal cancer: association with adenocarcinoma and visceral
37. Westhoff B, Brotze S, Weston A, et al. The frequency of Barrett’s obesity. Am J Gastroenterol. 2011;107(2):196-204.
esophagus in high-risk patients with chronic GERD. Gastrointest 64. Ogunwobi O, Mutungi G, Beales ILP. Leptin stimulates proliferation
Endosc. 2005;61(2):226-231. and inhibits apoptosis in Barrett’s esophageal adenocarcinoma
38. Zagari RM, Fuccio L, Wallander M-A, et al. Gastro-oesophageal cells by cyclooxygenase-2-dependent, prostaglandin-E2-mediated
reflux symptoms, oesophagitis and Barrett’s oesophagus in transactivation of the epidermal growth factor receptor and c-Jun NH
the general population: the Loiano-Monghidoro study. Gut. 2-terminal kinase activation. Endocrinology. 2006;147(9):4505-4516.
2008;57(10):1354-1359. 65. Greer KB, Falk GW, Bednarchik B, Li L, Chak A. Associations
39. Fan X, Snyder N. Prevalence of Barrett’s esophagus in patients of serum adiponectin and leptin with Barrett’s esophagus. Clin
with or without GERD symptoms: role of race, age, and gender. Gastroenterol Hepatol. 2015;13(13):2265-2272.
Dig Dis Sci. 2009;54(3):572-577. 66. Kubo A, Levin TR, Block G, et al. Cigarette smoking and the risk
40. Johansson J, Hakansson H-O, Mellblom L, et al. Risk factors of Barrett’s esophagus. Cancer Causes Control. 2008;20(3):303-311.
for Barrett’s oesophagus: a population-based approach. Scand J 67. Andrici J, Cox MR, Eslick GD. Cigarette smoking and the risk
Gastroenterol. 2007;42(2):148-156. of Barrett’s esophagus: a systematic review and meta-analysis.
41. Anderson LA, Watson RGP, Murphy SJ, et al. Risk factors for Barrett’s J Gastroenterol Hepatol. 2013;28(8):1258-1273.
oesophagus and oesophageal adenocarcinoma: results from the 68. Phillips WA, Lord RV, Nancarrow DJ, Watson DI, Whiteman DC.
FINBAR study. World J Gastroenterol. 2007;13(10):1585-1594. Barrett’s esophagus. J Gastroenterol Hepatol. 2011;26(4):639-648.
42. Smith KJ, O’Brien SM, Green AC, Webb PM, Whiteman DC. 69. Cook MB, Shaheen NJ, Anderson LA, et al. Cigarette smoking
Current and past smoking significantly increase risk for Barrett’s increases risk of Barrett’s esophagus: an analysis of the Barrett’s
esophagus. Clin Gastroenterol Hepatol. 2009;7(8):840-848. and esophageal adenocarcinoma consortium. Gastroenterology.
43. Conio M, Filiberti R, Blanchi S, et al. Risk factors for Barrett’s 2012;142(4):744-753.
esophagus: a case-control study. Int J Cancer. 2002;97(2):225-229. 70. Kubo A, Levin TR, Block G, et al. Alcohol types and sociode-
44. Eisen GM, Sandler RS, Murray S, Gottfried M. The relationship mographic characteristics as risk factors for Barrett’s esophagus.
between gastroesophageal reflux disease and its complications with Gastroenterology. 2009;136(3):806-815.
Barrett’s esophagus. Am J Gastroenterol. 1997;92(1):27-31. 71. Anderson LA, Cantwell MM, Watson RGP, et al. The association
45. Wild CP, Hardie LJ. Reflux, Barrett’s oesophagus and adenocarci- between alcohol and reflux esophagitis, Barrett’s esophagus, and
noma: burning questions. Nat Rev Cancer. 2003;3(9):676-684. esophageal adenocarcinoma. Gastroenterology. 2009;136(3):799-805.
46. Brown LM, Devesa SS, Chow W-H. Incidence of adenocarcinoma 72. Thrift AP, Cook MB, Vaughan TL, et al. Alcohol and the risk of
of the esophagus among white Americans by sex, stage, and age. Barrett’s esophagus: a pooled analysis from the International
J Natl Cancer Inst. 2008;100(16):1184-1187. BEACON Consortium. Am J Gastroenterol. 2014;109(10):1586-1594.
47. Vaughan TL, Davis S, Kristal A, Thomas DB. Obesity, alcohol, and 73. di Pietro M, Fitzgerald RC. Barrett’s oesophagus: an ideal model
tobacco as risk factors for cancers of the esophagus and gastric to study cancer genetics. Hum Genet. 2009;126(2):233-246.
cardia: adenocarcinoma versus squamous cell carcinoma. Cancer 74. Sun X, Elston R, Barnholtz-Sloan J, et al. A segregation analysis
Epidemiol Biomarkers Prev. 1995;4(2):85-92. of Barrett’s esophagus and associated adenocarcinomas. Cancer
48. Lagergren J, Bergström R, Nyren O. Association between body Epidemiol Biomarkers Prev. 2010;19(3):666-674.
mass and adenocarcinoma of the esophagus and gastric cardia. 75. Cameron AJ, Lagergren J, Henriksson C, Nyren O, Locke GR III,
Ann Intern Med. 1999;130(11):883-890. Pedersen NL. Gastroesophageal reflux disease in monozygotic and
49. Whiteman DC, Sadeghi S, Pandeya N, et al. Combined effects of dizygotic twins. Gastroenterology. 2002;122(1):55-59.
obesity, acid reflux and smoking on the risk of adenocarcinomas 76. Mohammed I, Cherkas LF, Riley SA, Spector TT, Trudgill NJ.
of the oesophagus. Gut. 2008;57(2):173-180. Genetic influences in gastro-oesophageal reflux disease: a twin
50. Cook MB, Greenwood DC, Hardie LJ, Wild CP, Forman D. A study. Gut. 2003;52(8):1085-1089.
systematic review and meta-analysis of the risk of increasing adiposity 77. Lembo A, Zaman M, Jones M, Talley NJ. Influence of genetics on
on Barrett’s esophagus. Am J Gastroenterol. 2008;103(2):292-300. irritable bowel syndrome, gastro-oesophageal reflux and dyspepsia:
51. Corley DA, Kubo A, Levin TR, et al. Abdominal obesity and body a twin study. Aliment Pharmacol Ther. 2007;25(11):1343-1350.
mass index as risk factors for Barrett’s esophagus. Gastroenterology. 78. Groves C, Jankowski J, Barker F, Holdstock G. A family history
2007;133(1):34-41. of Barrett’s oesophagus: another risk factor? Scand J Gastroenterol.
52. Edelstein ZR, Farrow DC, Bronner MP, Rosen SN, Vaughan TL. 2005;40(9):1127-1128.
Central adiposity and risk of Barrett’s esophagus. Gastroenterology. 79. Chak A, Ochs-Balcom H, Falk G, et al. Familiality in Barrett’s
2007;133(2):403-411. esophagus, adenocarcinoma of the esophagus, and adenocarcinoma
Epidemiology of Barrett Esophagus and Risk Factors for Progression CHAPTER 32 335
of the gastroesophageal junction. Cancer Epidemiol Biomarkers Prev. apoptosis in human esophageal adenocarcinoma cells. Cancer Res.
2006;15(9):1668-1673. 2000;60(20):5767-5772.
80. Chak A, Lee T, Kinnard MF, et al. Familial aggregation of Bar- 102. Buttar NS, Wang KK, Anderson MA, et al. The effect of selective
rett’s oesophagus, oesophageal adenocarcinoma, and oesopha- cyclooxygenase-2 inhibition in Barrett’s esophagus epithelium: an
gogastric junctional adenocarcinoma in Caucasian adults. Gut. in vitro study. J Natl Cancer Inst. 2002;94(6):422-429.
2002;51(3):323-328. 103. Moran EM. Epidemiological and clinical aspects of nonsteroidal
81. Munitiz V, Parrilla P, Ortiz A, Martinez-de-Haro LF, Yelamos J, anti-inflammatory drugs and cancer risks. J Environ Pathol Toxicol
Molina J. High risk of malignancy in familial Barrett’s esophagus. Oncol. 2002;21(2):193-201.
J Clin Gastroenterol. 2008;42(7):806-809. 104. Wang F, Lv ZS, Fu YK. Nonsteroidal anti-inflammatory drugs and
82. Levine DM, Ek WE, Zhang R, et al. A genome-wide association study esophageal inflammation—Barrett’s esophagus—adenocarcinoma
identifies new susceptibility loci for esophageal adenocarcinoma sequence: a meta-analysis. Dis Esophagus. 2011;24(5):318-324.
and Barrett’s esophagus. Nat Genet. 2013;45(12):1487-1493. 105. Moayyedi P, Jankowski JA. Does long term aspirin prevent cancer?
83. Su Z, Gay LJ, Strange A, Palles C, et al. Common variants at the Br Med J. 2010;341:c7326.
MHC locus and at chromosome 16q24.1 predispose to Barrett’s 106. Deleted in review.
esophagus. Nat Genet. 2012;44(10):1131-1136. 107. Jankowski J, Barr H. Improving surveillance for Barrett’s oesophagus:
84. Palles C, Chegwidden L, Li X, et al. Polymorphisms near TBX5 and AspECT and BOSS trials provide an evidence base. Br Med J.
GDF7 are associated with increased risk for Barrett’s esophagus. 2006;332(7556):1512.
Gastroenterology. 2015;148(2):367-378. 108. Thrift AP. Esophageal adenocarcinoma: the influence of medications
85. Ek WE, Levine DM, D’Amato M, et al. Germline genetic contributions used to treat comorbidities on cancer prognosis. Clin Gastroenterol
to risk for esophageal adenocarcinoma, Barrett’s esophagus, and Hepatol. 2015;13(13):2225-2232.
gastroesophageal reflux. J Natl Cancer Inst. 2013;105(22):1711-1718. 109. Beales IL, Dearman L, Vardi I, Loke Y. Reduced risk of Barrett’s
86. Gharahkhani P, Tung J, Hinds D, et al. Chronic gastroesopha- esophagus in statin users: case–control study and meta-analysis.
geal reflux disease shares genetic background with esophageal Dig Dis Sci. 2015;61(1):238-246.
adenocarcinoma and Barrett’s esophagus. Hum Mol Genet. 110. Alexandre L, Clark AB, Cheong E, Lewis MPN, Hart AR. Systematic
2015;25(4):ddv512-ddv835. review: potential preventive effects of statins against oesophageal
87. Thrift AP, Risch HA, Onstad L, et al. Risk of esophageal adenocar- adenocarcinoma. Aliment Pharmacol Ther. 2012;36(4):301-311.
cinoma decreases with height, based on consortium analysis and 111. Nguyen T, Duan Z, Naik AD, Kramer JR, El-Serag HB. Statin use
confirmed by Mendelian randomization. Clin Gastroenterol Hepatol. reduces risk of esophageal adenocarcinoma in US veterans with
2014;12(10):1667-1676, e1. Barrett’s esophagus: a nested case-control study. Gastroenterology.
88. Labenz J, Malfertheiner P. Helicobacter pylori in gastro-oesophageal 2015;149(6):1392-1398.
reflux disease: causal agent, independent or protective factor? Gut. 112. Fitzgerald RC, di Pietro M, Ragunath K, et al. British Society of
1997;41(3):277-280. Gastroenterology guidelines on the diagnosis and management of
89. Islami F, Kamangar F. Helicobacter pylori and esophageal cancer risk: Barrett’s oesophagus. Gut. 2014;63:7-42.
a meta-analysis. Cancer Prev Res (Phila). 2008;1(5):329-338. 113. Maret-Ouda J, Konings P, Lagergren J, Brusselaers N. Antire-
90. Wang C, Yuan Y, Hunt RH. Helicobacter pylori infection and Barrett’s flux surgery and risk of esophageal adenocarcinoma. Ann Surg.
esophagus: a systematic review and meta-analysis. Am J Gastroenterol. 2016;263(2):251-257.
2009;104(2):492-500. 114. Rees JR, Lao-Sirieix P, Wong A, Fitzgerald RC. Treatment for Barrett’s
91. Fischbach LA, Nordenstedt H, Kramer JR, et al. The association oesophagus. Cochrane Database Syst Rev. 2010;(1):CD004060.
between Barrett’s esophagus and Helicobacter pylori infection: a 115. Öberg S, Wenner J, Johansson J, Walther B, Willén R. Barrett
meta-analysis. Helicobacter. 2012;17(3):163-175. esophagus: risk factors for progression to dysplasia and adenocar-
92. Anderson LA, Murphy SJ, Johnston BT, et al. Relationship between cinoma. Ann Surg. 2005;242(1):49-54.
Helicobacter pylori infection and gastric atrophy and the stages of 116. O’Riordan JM, Byrne PJ, Ravi N, Keeling PWN, Reynolds JV.
the oesophageal inflammation, metaplasia, adenocarcinoma Long-term clinical and pathologic response of Barrett’s esophagus
sequence: results from the FINBAR case-control study. Gut. 2008; after antireflux surgery. Am J Surg. 2004;188(1):27-33.
57(6):734-739. 117. Hofstetter WL, Peters JH, Demeester TR, et al. Long-term outcome
93. Corley DA, Kubo A, Levin TR, et al. Helicobacter pylori infection of antireflux surgery in patients with Barrett’s esophagus. Ann Surg.
and the risk of Barrett’s oesophagus: a community-based study. 2001;234(4):532-538; discussion 538-539.
Gut. 2008;57(6):727-733. 118. Simonka Z, Paszt A, Abrahám S, et al. The effects of laparoscopic
94. Langman MJS, Cheng KK, Gilman EA, Lancashire RJ. Effect Nissen fundoplication on Barrett’s esophagus: long-term results.
of anti-inflammatory drugs on overall risk of common cancer: Scand J Gastroenterol. 2012;47(1):13-21.
case-control study in general practice research database. Br Med 119. Anvari M, Allen C, Marshall J, et al. A randomized controlled trial
J. 2000;320(7250):1642-1646. of laparoscopic Nissen fundoplication versus proton pump
95. Cuzick J, Thorat MA, Bosetti C, et al. Estimates of benefits and inhibitors for the treatment of patients with chronic gastro-
harms of prophylactic use of aspirin in the general population. esophageal reflux disease (GERD): 3-year outcomes. Surg Endosc.
Ann Oncol. 2014;26(1):47-57. 2011;25(8):2547-2554.
96. Botelho NK, Schneiders FI, Lord SJ, et al. Gene expression 120. Mehta RS, Song M, Bezawada N, et al. A prospective study of
alterations in formalin-fixed, paraffin-embedded Barrett esopha- macrophage inhibitory cytokine-1 (MIC-1/GDF15) and risk of
gus and esophageal adenocarcinoma tissues. Cancer Biol Ther. colorectal cancer. J Natl Cancer Inst. 2014;106(4):dju016.
2010;10(2):172-179. 121. Corey KE, Schmitz SM, Shaheen NJ. Does a surgical antireflux
97. Zimmermann KC, Sarbia M, Weber AA, Borchard F, Gabbert HE, procedure decrease the incidence of esophageal adenocarci-
Schrör K. Cyclooxygenase-2 expression in human esophageal noma in Barrett’s esophagus? A meta-analysis. Am J Gastroenterol.
carcinoma. Cancer Res. 1999;59(1):198-204. 2003;98(11):2390-2394.
98. Wilson KT, Fu S, Ramanujam KS, Meltzer SJ. Increased expres- 122. Chang EY, Morris CD, Seltman AK, et al. The effect of antireflux
sion of inducible nitric oxide synthase and cyclooxygenase-2 in surgery on esophageal carcinogenesis in patients with Barrett
Barrett’s esophagus and associated adenocarcinomas. Cancer Res. esophagus. Ann Surg. 2007;246(1):11-21.
1998;58(14):2929-2934. 123. Csendes A. Results of antireflux surgery in patients with Barrett’s
99. Song S, Guha S, Liu K, Buttar NS, Bresalier RS. COX-2 induction by esophagus. Eur Surg. 2008;40(4):154-164.
unconjugated bile acids involves reactive oxygen species-mediated 124. Kubo A, Block G, Quesenberry CP Jr, Buffler P, Corley DA. Effects
signalling pathways in Barrett’s oesophagus and oesophageal of dietary fiber, fats, and meat intakes on the risk of Barrett’s
adenocarcinoma. Gut. 2007;56(11):1512-1521. esophagus. Nutr Cancer. 2009;61(5):607-616.
100. Li M, Lotan R, Levin B, Tahara E, Lippman SM, Xu XC. Aspirin 125. Kubo A, Levin TR, Block G, et al. Dietary antioxidants, fruits, and
induction of apoptosis in esophageal cancer: a potential for che- vegetables and the risk of Barrett’s esophagus. Am J Gastroenterol.
moprevention. Cancer Epidemiol Biomarkers Prev. 2000;9(6):545-549. 2008;103(7):1614-1623.
101. Souza RF, Shewmake K, Beer DG, Cryer B, Spechler SJ. Selective 126. Dai Q, Cantwell MM, Murray LJ, et al. Dietary magnesium,
inhibition of cyclooxygenase-2 suppresses growth and induces calcium:magnesium ratio and risk of reflux oesophagitis, Barrett’s
336 SECTION I Esophagus and Hernia
oesophagus and oesophageal adenocarcinoma: a population-based 151. Sikkema M, Looman CWN, Steyerberg EW, et al. Predictors for
case-control study. Br J Nutr. 2016;115(2):342-350. neoplastic progression in patients with Barrett’s esophagus: a
127. Spechler SJ, Souza RF. Barrett’s esophagus. N Engl J Med. prospective cohort study. Am J Gastroenterol. 2011;106(7):1231-1238.
2014;371(9):836-845. 152. American Gastroenterological Association, Spechler SJ, Sharma P,
128. Shaheen NJ, Crosby MA, Bozymski EM, Sandler RS. Is there publica- Souza RF, Inadomi JM, Shaheen NJ. American Gastroenterological
tion bias in the reporting of cancer risk in Barrett’s esophagus? Association medical position statement on the management of
Gastroenterology. 2000;119(2):333-338. Barrett’s esophagus. Gastroenterology. 2011;140:1084-1091.
129. Desai TK, Krishnan K, Samala N, et al. The incidence of oesophageal 153. Whiteman DC, Appleyard M, Bahin FF, et al. Australian clinical
adenocarcinoma in non-dysplastic Barrett’s oesophagus: a meta- practice guidelines for the diagnosis and management of Barrett’s
analysis. Gut. 2012;61(7):970-976. esophagus and early esophageal adenocarcinoma. J Gastroenterol
130. Spechler SJ, Zeroogian JM, Antonioli DA, Wang HH, Goyal RK. Hepatol. 2015;30(5):804-820.
Prevalence of metaplasia at the gastro-oesophageal junction. Lancet. 154. Thrift AP, Shaheen NJ, Gammon MD, et al. Obesity and risk of
1994;344(8936):1533-1536. esophageal adenocarcinoma and Barrett’s esophagus: a Mendelian
131. Pohl H, Pech O, Arash H, et al. Length of Barrett’s oesophagus randomization study. J Natl Cancer Inst. 2014;106(11):dju252-2.
and cancer risk: implications from a large sample of patients with 155. Corley DA, Kubo A, Zhao W. Abdominal obesity and the risk of
early oesophageal adenocarcinoma. Gut. 2016;65(2):196-201. esophageal and gastric cardia carcinomas. Cancer Epidemiol Biomarkers
132. Wang KK, Sampliner RE. Updated guidelines 2008 for the diagnosis, Prev. 2008;17(2):352-358.
surveillance and therapy of Barrett’s esophagus. Am J Gastroenterol. 156. Hoyo C, Cook MB, Kamangar F, et al. Body mass index in relation
2008;103(3):788-797. to oesophageal and oesophagogastric junction adenocarcinomas:
133. Spechler SJ. Barrett’s esophagus: an overrated cancer risk factor. a pooled analysis from the International BEACON Consortium.
Gastroenterology. 2000;119(2):587-589. Int J Epidemiol. 2012;41(6):1706-1718.
134. Wani S, Puli SR, Shaheen NJ, et al. Esophageal adenocarcinoma in 157. Hardikar S, Onstad L, Blount PL, Odze RD, Reid BJ, Vaughan TL.
Barrett’s esophagus after endoscopic ablative therapy: a meta-analysis The role of tobacco, alcohol, and obesity in neoplastic progression
and systematic review. Am J Gastroenterol. 2009;104(2):502-513. to esophageal adenocarcinoma: a prospective study of Barrett’s
135. Sikkema M, de Jonge PJF, Steyerberg EW, Kuipers EJ. Risk of esophagus. PLoS One. 2013;8(1):e52192.
esophageal adenocarcinoma and mortality in patients with Barrett’s 158. Coleman HG, Bhat S, Johnston BT, McManus D, Gavin AT, Murray
esophagus: a systematic review and meta-analysis. Clin Gastroenterol LJ. Tobacco smoking increases the risk of high-grade dysplasia and
Hepatol. 2010;8(3):235-244; quiz e32. cancer among patients with Barrett’s esophagus. Gastroenterology.
136. Yousef F, Cardwell C, Cantwell MM, Galway K, Johnston BT, Murray 2012;142(2):233-240.
L. The incidence of esophageal cancer and high-grade dysplasia 159. Lagergren J, Bergstrom R, Lindgren A, Nyren O. Symptomatic
in Barrett’s esophagus: a systematic review and meta-analysis. Am gastroesophageal reflux as a risk factor for esophageal adenocar-
J Epidemiol. 2008;168(3):237-249. cinoma. N Engl J Med. 1999;340(11):825-831.
137. Bhat S, Coleman HG, Yousef F, et al. Risk of malignant progression 160. Farrow DC, Vaughan TL, Sweeney C, et al. Gastroesophageal reflux
in Barrett’s esophagus patients: results from a large population-based disease, use of H2 receptor antagonists, and risk of esophageal and
study. J Natl Cancer Inst. 2011;103(13):1049-1057. gastric cancer. Cancer Causes Control. 2000;11(3):231-238.
138. Hvid-Jensen F, Pedersen L, Drewes AM, Sørensen HT, Funch-Jensen 161. Rubenstein JH, Taylor JB. Meta-analysis: the association of oesopha-
P. Incidence of adenocarcinoma among patients with Barrett’s geal adenocarcinoma with symptoms of gastro-oesophageal reflux.
esophagus. N Engl J Med. 2011;365(15):1375-1383. Aliment Pharmacol Ther. 2010;32(10):1222-1227.
139. de Jonge PJF, van Blankenstein M, Looman CWN, Casparie MK, 162. Cook MB, Corley DA, Murray LJ, et al. Gastroesophageal reflux in
Meijer GA, Kuipers EJ. Risk of malignant progression in patients relation to adenocarcinomas of the esophagus: a pooled analysis
with Barrett’s oesophagus: a Dutch nationwide cohort study. Gut. from the Barrett’s and esophageal adenocarcinoma consortium
2010;59(8):1030-1036. (BEACON). PLoS One. 2014;9(7):e103508.
140. Wani S, Falk G, Hall M, et al. Patients with nondysplastic Barrett’s 163. Dent J, El-Serag HB, Wallander M-A, Johansson S. Epidemiology
esophagus have low risks for developing dysplasia or esophageal of gastro-oesophageal reflux disease: a systematic review. Gut.
adenocarcinoma. Clin Gastroenterol Hepatol. 2011;9(3):220-227; 2005;54(5):710-717.
quiz e26. 164. Vaughan TL, Fitzgerald RC. Precision prevention of oesophageal
141. Wood NJ. Barrett esophagus: need for ongoing surveillance called adenocarcinoma. Nat Rev Gastroenterol Hepatol. 2015;12(4):243-248.
into question for patients with non-dysplastic Barrett esophagus. 165. Greene CL, Worrell SG, Attwood SE, et al. Emerging concepts for the
Nat Rev Gastroenterol Hepatol. 2011;8(12):657. endoscopic management of superficial esophageal adenocarcinoma.
142. Kahrilas PJ. The problems with surveillance of Barrett’s esophagus. J Gastrointest Surg. 2016;20(4):851-860.
N Engl J Med. 2011;365(15):1437-1438. 166. Montgomery E, Bronner MP, Goldblum JR, et al. Reproducibility
143. di Pietro M, O’Donovan M, Fitzgerald RC. Where is the truth when of the diagnosis of dysplasia in Barrett esophagus: a reaffirmation.
it comes to cancer risk in Barrett’s esophagus? Gastroenterology. Hum Pathol. 2001;32(4):368-378.
2012;142(5):1245-1247. 167. Reid BJ, Haggitt RC, Rubin CE, et al. Observer variation in
144. Bisschops R. Reflux and Barrett’s disease. Can we stop surveillance the diagnosis of dysplasia in Barrett’s esophagus. Hum Pathol.
after 2011? Endoscopy. 2012;44(04):362-365. 1988;19(2):166-178.
145. Spechler SJ, Sharma P, Souza RF, Inadomi JM, Shaheen NJ. American 168. Kerkhof M, van Dekken H, Steyerberg EW, et al. Grading of
Gastroenterological Association technical review on the manage- dysplasia in Barrett’s oesophagus: substantial interobserver variation
ment of Barrett’s esophagus. Gastroenterology. 2011;140(3):e18-e52; between general and gastrointestinal pathologists. Histopathology.
quiz e13. 2007;50(7):920-927.
146. Harrison R, Perry I, Haddadin W, et al. Detection of intestinal 169. Zeki S, Fitzgerald RC. The use of molecular markers in predicting
metaplasia in Barrett’s esophagus: an observational comparator dysplasia and guiding treatment. Best Pract Res Clin Gastroenterol.
study suggests the need for a minimum of eight biopsies. Am J 2015;29(1):113-124.
Gastroenterol. 2007;102(6):1154-1161. 170. Duits LC, Phoa KN, Curvers WL, et al. Barrett’s oesophagus patients
147. Das D, Ishaq S, Harrison R, et al. Management of Barrett’s esophagus with low-grade dysplasia can be accurately risk-stratified after histo-
in the UK: overtreated and underbiopsied but improved by the logical review by an expert pathology panel. Gut. 2015;64(5):700-706.
introduction of a national randomized trial. Am J Gastroenterol. 171. Dulai GS, Shekelle PG, Jensen DM, et al. Dysplasia and risk of
2008;103(5):1079-1089. further neoplastic progression in a regional Veterans Administration
148. Xie S-H, Lagergren J. The male predominance in esophageal Barrett’s cohort. Am J Gastroenterol. 2005;100(4):775-783.
adenocarcinoma. Clin Gastroenterol Hepatol. 2016;14(3):338-347, e1. 172. Wani S, Falk GW, Post J, et al. Risk factors for progression of low-
149. Edgren G, Adami HO, Weiderpass E, Nyren O. A global assessment of the grade dysplasia in patients with Barrett’s esophagus. Gastroenterology.
oesophageal adenocarcinoma epidemic. Gut. 2013;62(10):1406-1414. 2011;141(4):1179-1186, 1186.e1.
150. Bhat SK, McManus DT, Coleman HG, et al. Oesophageal 173. Curvers WL, ten Kate FJ, Krishnadath KK, et al. Low-grade dysplasia
adenocarcinoma and prior diagnosis of Barrett’s oesophagus: a in Barrett’s esophagus: overdiagnosed and underestimated. Am J
population-based study. Gut. 2015;64(1):20-25. Gastroenterol. 2010;105(7):1523-1530.
Epidemiology of Barrett Esophagus and Risk Factors for Progression CHAPTER 32 337
174. Phoa KN, van Vilsteren FG, Weusten BL, et al. Radiofrequency 197. Li X, Galipeau PC, Sanchez CA, et al. Single nucleotide
ablation vs endoscopic surveillance for patients with Barrett polymorphism-based genome-wide chromosome copy change, loss
esophagus and low-grade dysplasia: a randomized clinical trial. of heterozygosity, and aneuploidy in Barrett’s esophagus neoplastic
JAMA. 2014;311(12):1209-1217. progression. Cancer Prev Res (Phila). 2008;1(6):413-423.
175. Downs-Kelly E, Mendelin JE, Bennett AE, et al. Poor interobserver 198. Ouatu Lascar R, Fitzgerald RC, Triadafilopoulos G. Differentiation
agreement in the distinction of high-grade dysplasia and adeno- and proliferation in Barrett’s esophagus and the effects of acid
carcinoma in pretreatment Barrett’s esophagus biopsies. Am J suppression. Gastroenterology. 1999;117(2):327-335.
Gastroenterol. 2008;103(9):2333-2340. 199. Sirieix PS, O’Donovan M, Brown J, Save V, Coleman N, Fitzgerald
176. Rastogi A, Puli S, El-Serag HB, Bansal A, Wani S, Sharma P. Inci- RC. Surface expression of minichromosome maintenance proteins
dence of esophageal adenocarcinoma in patients with Barrett’s provides a novel method for detecting patients at risk for devel-
esophagus and high-grade dysplasia: a meta-analysis. Gastrointest oping adenocarcinoma in Barrett’s esophagus. Clin Cancer Res.
Endosc. 2008;67(3):394-398. 2003;9(7):2560-2566.
177. Shaheen NJ, Sharma P, Overholt BF, et al. Radiofrequency 200. Murray L, Sedo A, Scott M, et al. TP53 and progression from
ablation in Barrett’s esophagus with dysplasia. N Engl J Med. Barrett’s metaplasia to oesophageal adenocarcinoma in a UK
2009;360(22):2277-2288. population cohort. Gut. 2006;55(10):1390-1397.
178. Clemons NJ, Phillips WA, Lord RV. Signaling pathways in the 201. Bani-Hani K, Martin IG, Hardie LJ, et al. Prospective study of Cyclin
molecular pathogenesis of adenocarcinomas of the esophagus and D1 overexpression in Barrett’s esophagus: association with increased
gastroesophageal junction. Cancer Biol Ther. 2013;14(9):782-795. risk of adenocarcinoma. J Natl Cancer Inst. 2000;92(16):1316-1321.
179. Ong CA, Lao-Sirieix P, Fitzgerald RC. Biomarkers in Barrett’s 202. Maley CC, Galipeau PC, Li X, Sanchez CA, Paulson TG, Reid BJ.
esophagus and esophageal adenocarcinoma: predictors of progres- Selectively advantageous mutations and hitchhikers in neoplasms.
sion and prognosis. World J Gastroenterol. 2010;16(45):5669-5681. Cancer Res. 2004;64(10):3414-3427.
180. Brabender J, Marjoram P, Salonga D, et al. A multigene expression 203. Barrett MT, Sanchez CA, Prevo LJ, et al. Evolution of neoplastic
panel for the molecular diagnosis of Barrett’s esophagus and Barrett’s cell lineages in Barrett oesophagus. Nat Genet. 1999;22(1):106-109.
adenocarcinoma of the esophagus. Oncogene. 2004;23(27):4780-4788. 204. Maley CC, Galipeau PC, Li X, et al. The combination of genetic
181. Brabender J, Marjoram P, Lord RVN, et al. The molecular signature instability and clonal expansion predicts progression to esophageal
of normal squamous esophageal epithelium identifies the presence adenocarcinoma. Cancer Res. 2004;64(20):7629-7633.
of a field effect and can discriminate between patients with Barrett’s 205. Maley CC, Galipeau PC, Finley JC, et al. Genetic clonal diversity
esophagus and patients with Barrett’s-associated adenocarcinoma. predicts progression to esophageal adenocarcinoma. Nat Genet.
Cancer Epidemiol Biomarkers Prev. 2005;14(9):2113-2117. 2006;38(4):468-473.
182. Lord RVN, Salonga D, Danenberg KD, et al. Telomerase reverse 206. Timmer MR, Martinez P, Lau L, et al. 139 Longitudinal single
transcriptase expression is increased early in the Barrett’s meta- cell clonal analysis reveals evolutionary stasis and re-determined
plasia, dysplasia, adenocarcinoma sequence. J Gastrointest Surg. malignant potential in non-dysplastic Barrett’s esophagus. Gastro-
2000;4(2):135-142. enterology. 2016;150(4):S34.
183. Reid BJ, Paulson TG, Li X. Genetic insights in Barrett’s esophagus and 207. Varghese S, Newton R, Ross-Innes CS, et al. Analysis of dysplasia
esophageal adenocarcinoma. Gastroenterology. 2015;149(5):1142-1143. in patients with Barrett’s esophagus based on expression pattern
184. Fisher OM, Lord SJ, Falkenback D, Clemons NJ, Eslick GD, Lord of 90 genes. Gastroenterology. 2015;149(6):1511-1518.
RV. The prognostic value of TP53 mutations in oesophageal 208. van Baal JWPM, Milana F, Rygiel AM, et al. A comparative analysis by
adenocarcinoma: a systematic review and meta-analysis. Gut. SAGE of gene expression profiles of esophageal adenocarcinoma and
2017;66(3):399-410. esophageal squamous cell carcinoma. Cell Oncol. 2008;30(1):63-75.
185. Stachler MD, Taylor-Weiner A, Peng S, et al. Paired exome 209. van Baal JWPM, Milano F, Rygiel AM, et al. A comparative analy-
analysis of Barrett’s esophagus and adenocarcinoma. Nat Genet. sis by SAGE of gene expression profiles of Barrett’s esophagus,
2015;47(9):1047-1055. normal squamous esophagus, and gastric cardia. Gastroenterology.
186. Carter SL, Cibulskis K, Helman E, et al. Absolute quantifica- 2005;129(4):1274-1281.
tion of somatic DNA alterations in human cancer. Nat Biotechnol. 210. Helm J, Enkemann SA, Coppola D, Barthel JS, Kelley ST, Yeatman
2012;30(5):413-421. TJ. Dedifferentiation precedes invasion in the progression from
187. Stephens PJ, Greenman CD, Fu B, et al. Massive genomic rear- Barrett’s metaplasia to esophageal adenocarcinoma. Clin Cancer
rangement acquired in a single catastrophic event during cancer Res. 2005;11(7):2478-2485.
development. Cell. 2011;144(1):27-40. 211. Barrett MT, Yeung KY, Ruzzo WL, et al. Transcriptional analyses
188. Nones K, Waddell N, Wayte N, et al. Genomic catastrophies frequently of Barrett’s metaplasia and normal upper GI mucosae. Neoplasia.
arise in esophageal adenocarcinoma and drive tumorigenesis. Nat 2002;4(2):121-128.
Commun. 2014;5:5224. 212. Nancarrow DJ, Clouston AD, Smithers BM, et al. Whole genome
189. Dulak AM, Stojanov P, Peng S, et al. Exome and whole-genome expression array profiling highlights differences in mucosal defense
sequencing of esophageal adenocarcinoma identifies recurrent driver genes in Barrett’s esophagus and esophageal adenocarcinoma.
events and mutational complexity. Nat Genet. 2013;45(5):478-486. PLoS One. 2011;6(7):e22513.
190. Weaver JM, Ross-Innes CS, Shannon N, et al. Ordering of mutations 213. Fisher OM, Maag JL, Levert-Mignon A, et al. Tu1135 whole tran-
in preinvasive disease stages of esophageal carcinogenesis. Nat scriptome sequencing reveals previously unrecognized alterations in
Genet. 2014;46(8):837-843. Barrett’s esophagus and esophageal adenocarcinoma. Gastroenterology.
191. Reid BJ, Li X, Galipeau PC, Vaughan TL. Barrett’s oesophagus 2016;150(4):S854-S865, e1; quiz e16-e17.
and oesophageal adenocarcinoma: time for a new synthesis. Nat 214. Krause L, Nones K, Loffler KA, et al. Identification of the CIMP-like
Rev Cancer. 2010;10(2):87-101. subtype and aberrant methylation of members of the chromosomal
192. Wijnhoven BPL, Tilanus HW, Dinjens WNM. Molecular biology of segregation and spindle assembly pathways in esophageal adeno-
Barrett’s adenocarcinoma. Ann Surg. 2001;233(3):322-337. carcinoma. Carcinogenesis. 2016;37(4):356-365.
193. Jenkins GJS, Doak SH, Parry JM, D’Souza FR, Griffiths AP, Baxter 215. Smith E, De Young NJ, Pavey SJ, et al. Similarity of aberrant DNA
JN. Genetic pathways involved in the progression of Barrett’s methylation in Barrett’s esophagus and esophageal adenocarcinoma.
metaplasia to adenocarcinoma. Br J Surg. 2002;89(7):824-837. Mol Cancer. 2008;7(1):1.
194. Galipeau PC, Li X, Blount PL, et al. NSAIDs modulate CDKN2A, 216. Wang JS, Guo M, Montgomery EA, et al. DNA promoter hyper-
TP53, and DNA content risk for progression to esophageal adeno- methylation of p16 and APC predicts neoplastic progression in
carcinoma. PLoS Med. 2007;4(2):e67. Barrett’s esophagus. Am J Gastroenterol. 2009;104(9):2153-2160.
195. Nancarrow DJ, Handoko HY, Smithers BM, et al. Genome-wide 217. Jin Z, Cheng Y, Gu W, et al. A multicenter, double-blinded validation
copy number analysis in esophageal adenocarcinoma using study of methylation biomarkers for progression prediction in
high-density single-nucleotide polymorphism arrays. Cancer Res. Barrett’s esophagus. Cancer Res. 2009;69(10):4112-4115.
2008;68(11):4163-4172. 218. Rakyan VK, Down TA, Thorne NP, et al. An integrated resource
196. Lai LA, Paulson TG, Li X, et al. Increasing genomic instability for genome-wide identification and analysis of human tissue-
during premalignant neoplastic progression revealed through high specific differentially methylated regions (tDMRs). Genome Res.
resolution array-CGH. Genes Chromosomes Cancer. 2007;46(6):532-542. 2008;18(9):1518-1529.
338 SECTION I Esophagus and Hernia
219. Shah AK, Lê Cao K-A, Choi E, et al. Glyco-centric lectin magnetic 226. Pepe MS, Etzioni R, Feng Z, et al. Phases of biomarker develop-
bead array (LeMBA)—proteomics dataset of human serum samples ment for early detection of cancer. J Natl Cancer Inst. 2001;93(14):
from healthy, Barrett’s esophagus and esophageal adenocarcinoma 1054-1061.
individuals. Data Brief. 2016;7:1058-1062. 227. Kastelein F, Biermann K, Steyerberg EW, et al. Aberrant p53
220. Kraly JR, Jones MR, Gomez DG, et al. Reproducible two-dimensional protein expression is associated with an increased risk of neo-
capillary electrophoresis analysis of Barrett’s esophagus tissues. plastic progression in patients with Barrett’s oesophagus. Gut.
Anal Chem. 2006;78(17):5977-5986. 2013;62(12):1676-1683.
221. Peng D, Sheta EA, Powell SM, et al. Alterations in Barrett’s -related 228. Clark GW, et al. Short-segment Barrett’s esophagus: a prevalent
adenocarcinomas: a proteomic approach. Int J Cancer. 2007;122(6): complication of gastroesophageal reflux disease with malignant
1303-1310. potential. J Gastrointest Surg. 1997;1:113-122.
222. Zhao J, Chang AC, Li C, et al. Comparative proteomics analysis 229. Voutilainen M, et al. Gastroesophageal reflux disease: prevalence,
of Barrett metaplasia and esophageal adenocarcinoma using two-
dimensional liquid mass mapping. Mol Cell Proteomics. 2007;6(6): clinical, endoscopic and histopathological findings in 1,128 consecu-
987-999. tive patients referred for endoscopy due to dyspeptic and reflux
223. Chao DL, Sanchez CA, Galipeau PC, et al. Cell proliferation, symptoms. Digestion. 2000;61:6-13.
cell cycle abnormalities, and cancer outcome in patients with 230. Gerson LB, et al. Prevalence of Barrett’s esophagus in asymptomatic
Barrett’s esophagus: a long-term prospective study. Clin Cancer Res. individuals. Gastroenterology. 2002;123:461-467.
2008;14(21):6988-6995. 231. Veldhuyzen van Zanten SJ, et al. The prevalence of Barrett’s
224. Bird Lieberman EL, Dunn JM, Coleman HG, et al. Population-based oesophagus in a cohort of 1040 Canadian primary care patients
study reveals new risk-stratification biomarker panel for Barrett’s with uninvestigated dyspepsia undergoing prompt endoscopy.
esophagus. Gastroenterology. 2012;143(4):927-935, e3. Aliment Pharmacol Ther. 2006;23:595-599.
225. Critchley-Thorne RJ, Duits LC, Prichard JW, et al. A tissue systems 232. Hayeck TJ, et al. The prevalence of Barrett’s esophagus in the US:
pathology assay for high-risk Barrett’s esophagus. Cancer Epidemiol estimates from a simulation model confirmed by SEER data. Dis
Biomarkers Prev. 2016;25(6):958-968. Esophagus. 2010;23:451-457.
CHAPTER
Medical and Surgical Therapy for Gastroesophageal
Reflux Disease and Barrett Esophagus 33
Mark R. Wendling
| Brant K. Oelschlager
T
he British surgeon Norman Barrett is famously Western world, where it is the most prominent, the highest
credited for his early description of the lower incidence is in middle-aged white males. It is diagnosed
esophagus lined by columnar epithelium. However, by recognition of columnar mucosa on endoscopy and
he himself did not claim to be the first to describe the confirmed histologically from tissue obtained from biopsy
condition that would later bear his name. His original during the upper endoscopy. The mean age at time of
article in 1950 details numerous previous reports that diagnosis is in the 6th decade of life4 and demonstrates
likely represented this pathology.1 Throughout Barrett’s a male-to-female predominance of approximately 2 : 1.5
impressive career, he proposed many theories that later Obesity is a risk factor for gastroesophageal reflux disease
proved to be correct, but in this instance, he concluded (GERD) due to its effects on the body’s normal antireflux
that the columnar epithelium he observed was gastric mechanisms and is consequently a likely risk factor for
tissue in the setting of a congenitally short esophagus.2 Barrett esophagus. Body mass index (BMI) exists as the
It was not until 3 years later when Allison and Johnstone most common quantifier of obesity; however, many have
published their own findings that this was corrected.3 They suggested that an increased abdominal circumference
demonstrated not only that the columnar-lined viscus was or waist-to-hip ratio may carry more significance, as it is
indeed the esophagus, but they also proposed that the a better measure of abdominal adiposity.6,7 Other inde-
ulcer associated with this condition should be referred pendent risk factors that have been suggested include
to as “Barrett’s ulcer.” smoking8 and presence of a hiatal hernia.9 Conversely, the
If the initial description was difficult to come by, then presence of Helicobacter pylori has demonstrated an inverse
it may have been foreshadowing of things to come. The association with risk of Barrett esophagus for reasons
controversies regarding the definition of Barrett esophagus that are not entirely clear.10 The majority of available
and the epidemiologic descriptions of the disease are so data suggest that the prevalence of Barrett esophagus is
extensive that they have been granted their own chapters in significantly lower in African Americans compared with
this edition. In contrast, something that has been correctly non-Hispanic whites by several times.11–13 Although it is not
postulated from the beginning is that Barrett esophagus entirely understood whether this imbalance exists due to
is a reaction of the esophageal mucosa to chronic injury differences in risk factor profiles or genetic susceptibility,
from refluxate. The tools available in the physician’s a genetic predisposition seems likely. Variances at specific
armamentarium for the treatment of Barrett esophagus genetic loci have been implicated in increasing the genetic
are diverse. These include medication for prophylaxis and susceptibility to developing Barrett esophagus.14 Also,
treatment, antireflux operations, ablation therapies, and the vast majority of patients are diagnosed on their first
surgical resection in select cases. The indications for each endoscopy.15 This suggests that reflux acts as a trigger in
continue to evolve with time. The holy grail of treatment the right genetic environment for the development of
remains prevention of progression along the pathway from Barrett esophagus.
metaplasia to neoplasia, and whether the natural history Barrett esophagus has been categorized by its extent
of Barrett esophagus can be altered remains a topic of either as long (>3 cm) or short (<3 cm) segments. A third,
debate. Current goals of therapy are directed at relief of more controversial, description is intestinal metaplasia at
associated reflux symptoms and healing of esophagitis to the gastroesophageal junction. This likely represents either
prevent complications of nondysplastic Barrett esophagus very short segment Barrett or intestinal metaplasia of the
and removal or obliteration of the tissue when it progresses gastric cardia, and the malignant potential is less clear.
on the continuum toward cancer. Short-segment Barrett represents the large majority of cases,
while long segments likely represent more severe reflux
both in terms of acid exposure and its proximal extent.16
CLINICAL FEATURES OF There has been theoretical concerns that long-segment
BARRETT ESOPHAGUS disease may be associated with a higher incidence of both
dysplasia and adenocarcinoma due to disease burdens on
The mucosa of the esophagus consists of stratified the cellular level, and there is some retrospective data to
squamous epithelium. Barrett esophagus is the meta- support this.17,18 However, the available evidence has not
plastic replacement of this squamous epithelium with been sufficient to justify incorporating length consistently
intestinal columnar epithelium, which predisposes to across current surveillance guidelines. Instead, the presence
the development of adenocarcinoma. Barrett esophagus and degree of dysplasia has provided the marker of risk
has been described across all demographics, but in the and has driven the clinical management. Histologically
339
Medical and Surgical Therapy for Gastroesophageal Reflux Disease and Barrett Esophagus CHAPTER 33 339.e1
KEYWORDS
Barrett esophagus; treatment of Barrett esophagus; GERD;
gastroesophageal reflux disease; chemoprophylaxis; acid
suppression; antireflux surgery; endoscopic mucosal
resection; ablation of Barrett esopahgus; low-grade
dysplasia; high-grade dysplasia; esophagectomy; esophageal
cancer
340 SECTION I Esophagus and Hernia
Probability of mortality
sion therapy should be maximized and repeat biopsies
obtained after a brief period to allow healing. Regardless 0.6
of the classification, there remains a subjective element
to histologic grading, which has been demonstrated by
relatively low interobserver agreement within specimens,
0.4
particularly for LGD. Consequently, a second opinion from
an expert gastrointestinal pathologist is recommended to
confirm a diagnosis of LGD or HGD.
BE unknown, no surveillance
0.2
BE known, no surveillance
SURVEILLANCE BE known, inadequate surveillance
BE known, adequate surveillance
The development of esophageal adenocarcinoma repre-
Censoring (all groups)
sents the most feared complication of Barrett esophagus, 0.0
with mortality increasing with the stage at diagnosis. The
goal of early detection forms the rationale of surveil- 0 2 4 6 8 10 12 14
lance. This efficacy of surveillance has been questioned Follow-up after EAC diagnosis (years)
recently, but poor efficacy is likely related to an excessively
long interval between surveillance endoscopies and an FIGURE 33.1 Mortality related to a prior Barrett esophagus (BE)
inadequate number of biopsies to adequately evaluate the diagnosis and surveillance participation. Kaplan–Meier survival
Barrett segment.19,20 Surveillance for nondysplastic Barrett analysis (+log-rank test) for mortality in esophageal
esophagus should include the use of high-definition adenocarcinoma (EAC) patients, with and without a prior BE
endoscopy with systematic four-quadrant biopsies every diagnosis, with or without participating in an (adequate)
2 cm with separate endoscopic mucosal resection (EMR) surveillance program. BE unknown vs. BE known, no surveillance:
P = .006; BE unknown vs. inadequate surveillance: P = .007,
of mucosal abnormalities at an interval of 1 to 2 years.
BE unknown vs. adequate surveillance: P < .001; adequate vs.
If there is a history of LGD, biopsies should be obtained
inadequate surveillance: P < .001. Adequate surveillance vs. BE
every 1 cm with an interval of 6 to 12 months, although known, no surveillance: P < .001. (From Verbeek RE, Leenders M,
increasingly patients with dysplasia are undergoing ablation Ten Kate FJ, et al. Surveillance of Barrett’s esophagus and
rather than continued surveillance. There is evidence mortality from esophageal adenocarcinoma: a population-based
that surveillance results in the detection of esophageal cohort study. Am J Gastroenterol. 2014;109:1215–1222.)
adenocarcinoma at an earlier stage. For instance, in a
retrospective review of 224 patients, Grant et al.21 found
that those who underwent surveillance had significantly
lower stage tumors at the time of diagnosis. One would TREATMENT OF NONDYSPLASTIC
then expect this to translate to a benefit in mortality
but concern exists that this may represent lead-time bias BARRETT ESOPHAGUS
instead of a true survival benefit. Disputing this is data from
multiple centers showing that with endoscopic resection MEDICAL TREATMENT
of early adenocarcinoma patients are cured of their See Box 33.1.
disease and have survival similar to the general popula-
tion. Corley et al.22 did not demonstrate an association Chemoprophylaxis
between surveillance and a decreased risk of death in a In recent years, there has been much attention paid to
case-control study of 70 patients with esophageal adeno- the subject of chemoprophylaxis against the development
carcinoma diagnosed greater than 6 months after Barrett of esophageal adenocarcinoma. Statins originally were
esophagus, although these results have been criticized shown to have some efficacy in in vitro cell lines. This is
for high overall mortality rates, not using endoluminal thought to be secondary to induction of apoptosis and
therapy in early disease and not differentiating between cyclooxygenase (COX)-2 inhibition.25,26 There have since
adequate surveillance and any surveillance.23 In contrast, been several observational studies published investigating
shortly thereafter, Verbeek et al.24 demonstrated decreased the translational significance of these findings. Although
esophageal adenocarcinoma mortality at 2 and 5 years for these studies have had conflicting results, they are also
those adhering to a surveillance program (hazard ratio heterogeneous and likely underpowered. The largest
[HR], 0.79) (Fig. 33.1). Logically, if early-stage cancer case-control study to investigate the effect of statins in the
is curative in most patients, it makes sense that surveil- setting of Barrett esophagus (311 cases, 856 controls) found
lance at an appropriate interval that allows detection approximately 35% lower odds of developing esophageal
of progression to HGD or early adenocarcinoma will adenocarcinoma in those on statin therapy.27 Two meta-
be beneficial. analyses have attempted addressing the role of statins,
Medical and Surgical Therapy for Gastroesophageal Reflux Disease and Barrett Esophagus CHAPTER 33 341
TABLE 33.1 Clinical Features of Patients With Barrett TABLE 33.2 Comparison of Pre- and Postoperative
Esophagus and Gastroesophageal Reflux Disease and Symptoms in 106 Patients With Barrett Esophagus Who
Esophagogastroduodenoscopy Controls Underwent Laparoscopic Antireflux Surgery
Barrett GERD Controls EGD Controls Preoperative No
(n = 79) (n = 94) (n = 84) Incidence: Resolution: Improvement: Improvement:
Duration of symptoms 16.4 11.8 13 Symptom n (%) n (%) n (%) n (%)
(year)* Heartburn 98 (92%) 69 (70%) 25 (26%) 4 (4%)
Mean age at onset* 35.3 43.7 42.7 Regurgitation 69 (65%) 52 (75%) 6 (9%) 11 (16%)
Esophagitis† 51 (65%) 33 (35%) 24 (29%) Dysphagia 33 (31%) 21 (64%) 6 (18%) 6 (18%)
Esophageal ulcer† 17 (22%) 7 (7%) 6 (7%) Cough 31 (29%) 22 (71%) 2 (6%) 7 (23%)
Esophageal stricture† 21 (27%) 7 (7%) 5 (6%) Chest pain 30 (28%) 20 (67%) 6 (20%) 4 (13%)
Hiatal hernia† 60 (76%) 41 (44%) 31 (37%) Hoarseness 25 (24%) 21 (84%) 1 (4%) 3 (12%)
Severe GERD‡ 67 (85%) 55 (59%) 53 (63%)
From Oelschlager BK, Barreca M, Chang L, Oleynikov D, Pellegrini CA.
EGD, Esophagogastroduodenoscopy; GERD, gastroesophageal reflux Clinical and pathologic response of Barrett’s esophagus to laparoscopic
disease. antireflux surgery. Ann Surg. 2003;238:458–464.
*P < .05 for the Barrett esophagus group versus either control group
(Kruskal-Wallis test).
†
Odds ratios for esophagitis, esophageal ulcer, esophageal stricture, and
hiatal hernia greater than 3 cm for the Barrett esophagus group versus
not significant). There was no difference found in the level
either control group. of symptom control at 3-year follow-up. Also, there was no
‡
Severe GERD was defined as heartburn so painful that it awoke the patient difference found when comparing symptomatic outcomes
or prevented sleeping. or dysphagia rates between those with Barrett esophagus
EGD, Esophagogastroduodenoscopy; GERD, gastroesophageal reflux and those without. Objectively, pH monitoring performed
disease.
Modified from Eisen GM, Sandler RS, Murray S, Gottfried M. The relation-
at 6 months post treatment demonstrated significantly
ship between gastroesophageal reflux disease and its complications with lower percent acid exposure times for those randomized
Barrett’s esophagus. Am J Gastroenterol. 1997;92:27. to surgical therapy compared with PPIs (13.2% to 0.4%
and 7.4% to 4.9%, respectively; P = .002).
Hofstetter et al.47 also demonstrated excellent results in
are currently without a satisfying answer. Nevertheless, 97 patients with Barrett esophagus who underwent antire-
return of a competent mechanical antireflux barrier that flux surgery. At median 5-year follow-up, 79% of patients
prevents all types of reflux would have advantages over had complete resolution of all reflux symptoms with a
medical therapy by addressing both the acid and nonacid 97% patient satisfaction rate reported. The vast majority
components. It is important to remember, however, that of pH studies normalized as well (81%). This is consistent
the anatomic findings that correlate with severe GERD and with a randomized prospective trial comparing medical
Barrett esophagus such as a large hiatal hernia, decreased therapy with fundoplication prior to the popularization
motility, stricture, and a shortened esophagus also make of the laparoscopic approach. Fifty-eight patients with
obtaining a successful surgical result more challenging. Barrett esophagus were randomized to the surgical arm
One must identify these factors prior to an operation. with median 5-year follow-up. Satisfactory symptomatic
Furthermore, evidence indicates that patients with Barrett outcomes were achieved in 91% of patients. The total %
and a failed fundoplication are at increased risk for disease time of pH < 4 decreased from 19.0% preoperatively to just
progression. Thus it is imperative to select patients with 0.6% postoperatively with positive studies found in 15%
Barrett esophagus carefully for antireflux surgery and of patients. Bilitec monitoring to detect duodenogastric
perform the surgery well to minimize the potential for reflux was also used with 92% of these studies being normal
breakdown of the repair. in the surgical arm versus just 25% in those receiving
medical therapy.48
Subjective and Objective Outcomes We have examined the outcomes of those with Barrett
In general, patients with Barrett esophagus have good symp- esophagus who received laparoscopic antireflux surgery
tomatic and functional outcomes after antireflux surgery in our own experience. We demonstrated 86% of patients
(Table 33.2 and Fig. 33.2), although it has been suggested reported improvement in symptoms of heartburn and
that results are inferior to those with uncomplicated GERD. regurgitation and 10/10 median patient satisfactions at
The LOTUS trial,45 published in 2011, represents the first 8-year follow-up. Also, we demonstrated a mean decrease in
multicenter, randomized trial comparing laparoscopic DeMeester score from 54 to 9.49 We have also demonstrated
antireflux surgery with PPIs. Importantly, only patients with that approximately three-quarters of our patients have
good symptomatic control on PPIs were eligible for the normal objective pH data following antireflux surgery.50
study. Both therapies were found to be effective at 5 years. This likely underrepresents the success as it is relatively
A subset analysis of patients with Barrett esophagus from difficult to convince the asymptomatic patient to undergo
this study was published by Attwood46 and has provided objective testing compared with those who are experiencing
insight into the subject. Sixty patients were randomized recurrent gastroesophageal reflux. It seems obvious that
to either standardized laparoscopic antireflux surgery nothing inherent to the metaplastic epithelium itself would
or dose-adjusted esomeprazole. Again, both treatment result in inferior results to surgical attempts of reflux
modalities were found to be effective, with only four control. Most likely, the relationship between Barrett
treatment failures between the groups (1 surgery, 3 PPI, esophagus and surgery represents somewhat of a catch-22
Medical and Surgical Therapy for Gastroesophageal Reflux Disease and Barrett Esophagus CHAPTER 33 343
90
80
70
60
% Time pH < 4
50
40
30
20
10
0
Preop Postop
FIGURE 33.2 Twenty-four-hour distal esophageal pH results before and after Nissen fundoplication in 21 patients with Barrett esophagus
studied preoperatively and postoperatively. Postop, Postoperative; Preop, preoperative. (From Hofstetter WA, Peters JH, DeMeester TR,
et al. Long term outcome of antireflux surgery in patients with Barrett’s esophagus. Ann Surg. 2001;234:532.)
where the same anatomic factors that lead to severe gastro- regression for those who underwent antireflux surgery
esophageal reflux, Barrett esophagus, and need for surgical compared with the medically managed group (40 vs. 16%;
intervention also result in more clinically challenging P = .04). Both these investigations demonstrated signifi-
patients and technically challenging operations, making cantly higher rates of regression in those with short-segment
the pursuit of quality outcomes more difficult. compared with long-segment disease.
The other factor that seems to play a role in the rate
Impact of Antireflux Surgery on the of regression is, perhaps unsurprisingly, actual success in
Metaplasia-Dysplasia-Neoplasia Continuum reconstructing a competent gastric cardia. A cohort of 75
Since GERD is an essential component to the development patients with median follow-up of 8.9 years demonstrated
of Barrett, it is reasonable to assume that effective treatment a rate of regression of 31%. The rate of progression in
of GERD would have a positive effect on the natural history this population was 8%. Patients with progression were
of the disease. Although heavily debated, some also hold much more likely to have a failed fundoplication based
the belief that once Barrett metaplasia has occurred, its on endoscopic evaluation than those who did not progress
natural history cannot be altered. Theoretically, on the (67% vs. 16%; P = .0129).54 In our own experience, we
continuum from metaplasia to neoplasia, there must be a described a 55% regression rate in 90 consecutive patients
point of no return at which the development of esophageal at median 40-month follow-up. Of those who underwent
adenocarcinoma is inevitable. The uncertain point of this functional testing, 89% were found to have normalized
biologic Rubicon coupled with the relatively low incidence pH studies.50
of the disease makes determining the effect of surgical Rates of regression have not been universally reproduc-
therapy on the prevention of esophageal adenocarcinoma ible. Randomized prospective data by Parrilla et al.48 did
extremely difficult and thus far elusive. not find complete regression in any of the 101 patients
In 1980 Brand et al.51 first described regression in 4 of 10 receiving either medical or surgical therapy despite high
patients with Barrett esophagus. Since then, there has been rates of acid normalization on functional studies. There
an abundance of surgical literature from observational was significantly less de novo dysplasia found in the surgical
studies demonstrating varying rates of regression after group (2% vs. 20%). There are several possible explana-
antireflux surgery. In a cohort study of 91 symptomatic tions for the wide variability in findings between studies.
patients with Barrett esophagus, Gurski et al.52 demon- It has been proposed that incorporation of metaplastic
strated a regression rate of 36.4% after antireflux surgery. epithelium in a fundoplication makes identification and
This was compared with a regression rate of 7.1% for 14 biopsy more difficult in short-segment disease resulting
patients receiving medical therapy. In the surgical group, in frequent sampling error, although most endoscopists
the regression was composed of 17 patients with LGD who dismiss any increased difficulty with biopsies after a
reverted to nondysplastic Barrett and 11 patients with fundoplication.
Barrett who regressed to nonmetaplastic epithelium. Eight The likelihood of regression has been shown repeatedly
patients progressed (five and three to LGD and HGD, to be related to both disease length and acid normaliza-
respectively). Zaninotto et al.53 compared a cohort of 45 tion. However, the clinical relevance of regression is
postsurgical patients with 44 with medical management. unknown, as it is possible that Barrett esophagus that is
There was a statistically significant increase in the rate of susceptible to regression may never have progressed at
344 SECTION I Esophagus and Hernia
all, and any length of Barrett is still at risk for cancer. The scope of this chapter; however, level 1 evidence exists
more pressing question is related to progression and the for a reduced risk of adenocarcinoma with PDT in the
role of antireflux surgery in prevention of esophageal setting of HGD and RFA in both LGD and HGD. 57–59
adenocarcinoma. Endoscopic ablation in the setting of nondysplastic Barrett
A flawed study from Sweden used a large database esophagus is not currently recommended, as there is no
to compare the risk of esophageal adenocarcinoma in convincing long-term data from large studies to suggest
patients who underwent antireflux surgery relative to the that ablation of nondysplastic Barrett esophagus decreases
background population. This large-scale study comprised the risk of malignancy. There are two likely reasons for
14,102 patients and 120,514 person-years at risk. A 12-fold this. First, complete ablation without recurrence has not
increase in the risk of developing cancer was found in the been demonstrated by any modality. In a multicenter
antireflux surgery cohort. The authors concluded that trial of patients with nondysplastic Barrett esophagus who
antireflux surgery could not be considered to prevent underwent treatment with high-powered APC in conjunc-
esophageal adenocarcinoma.55 This may be overstating tion with esomeprazole, complete remission was obtained
the evidence. The data support the claim that antireflux in only 77% of patients. This was coupled with 9.8%
surgery does not return the risk of adenocarcinoma to major complication rate comprising bleeding, stenosis, or
that of the baseline population, but that is an unrealistic perforation.60 Second, ongoing acid exposure may sabotage
bar for most interventions in medicine. The study does the effectiveness of ablation. Ferraris et al.61 demonstrated
not address the risk reduction afforded to patients who that even with complete remission of Barrett esophagus,
had antireflux surgery compared with a control group of recurrence was common. Ninety-four patients underwent
patients with reflux disease treated with acid suppression endoscopic ablation with APC and had reflux control with
medications. In a follow-up analysis of this study popula- either medical or surgical therapy. At a mean follow-up
tion, Lofdahl et al.56 identified 55 cases where patients of 36 months, recurrence of intestinal metaplasia was
developed esophageal adenocarcinoma more than 5 identified in 18% of patients. The risk of recurrence was
years after antireflux surgery. They were compared with found to be significantly lower in those who underwent
240 controls matched to age, sex, and calendar year of surgical fundoplication (OR, 0.30). In addition to concerns
antireflux surgery. They found that those who developed about incomplete ablation, ongoing acid exposure, and
esophageal adenocarcinoma were 3 times more likely to recurrence of intestinal metaplasia, there is concern
have recurrent pathologic reflux after their operation than regarding buried Barrett glands under the neosquamous
the control population. This led the authors to conclude epithelium. The risk of these glands is unknown; however,
that perhaps recurrent GERD was responsible for the lack they may also confer an ongoing risk of progression to
of protective effect of antireflux surgery. This is supported cancer.62
by randomized prospective data that found no difference in
progression to carcinoma between medically and surgically Low-Grade Dysplasia
treated patients (5% vs. 3%) at 5 years’ follow-up. In the LGD in the setting of Barrett esophagus represents a diag-
latter group, all progression was associated with clinical nostic entity where the ideal treatment is largely unknown.
and objective recurrence of GERD.48 Logically, if recurrent Large variance has been shown in the interobserver reli-
GERD and a failed fundoplication is a risk factor for ability of the diagnosis. This has led to wide-ranging results
disease progression, a functioning fundoplication must in the literature for treatment and outcome. Certainly,
reduce or stabilize the risk. inflammation from ongoing reflux can lead to histologic
changes that mimic dysplasia, and these resolve with control
of reflux, particularly with antireflux surgery. However,
TREATMENT OF DYSPLASIA AND true dysplasia likely does not regress.
The historically believed rate of progression of Barrett
EARLY CANCER esophagus to adenocarcinoma of 1% per year has more
recently thought to be an overestimation. In fact, recently,
ENDOSCOPIC THERAPY FOR DYSPLASTIC a cohort of patients with LGD was followed for over
BARRETT ESOPHAGUS 950 patient-years. The rate of progression to HGD was
The efficacy of medical and surgical therapy for Barrett found to be 1.6% per year, and progression to esophageal
esophagus as it relates to progression to carcinoma con- adenocarcinoma was 0.44% per year.63 In terms of merely
tinues to be described and debated. In the meantime, making the diagnosis, a Dutch study of 293 patients where
several endoscopic ablative modalities proved beneficial Barrett esophagus was designated with LGD found that
to alter the natural history of Barrett esophagus. In theory, when reviewed by an expert panel, only 27% of the initial
if the metaplastic cells could be eradicated completely in diagnoses were confirmed. The remainder of the cases were
a safe and reliable manner, the risk of progression could downstaged to either nondysplastic Barrett or indefinite
be completely removed. This would negate the need for for dysplasia. Interestingly, the rate of progression was
further surveillance and make many current issues of no 9.1% per year in the confirmed group and only 0.6%
consequence. There are several methods of eradication per year for those with downstaged nondysplastic Barrett
of Barrett esophagus including radiofrequency ablation esophagus.64 Importantly, this 0.6% rate is still higher than
(RFA), photodynamic therapy (PDT), cryoablation, many reports for true nondysplastic Barrett esophagus,
argon plasma coagulation (APC), and EMR (Fig. 33.3), suggesting that even if not confirmed by a second patholo-
each with their own advantages and proponents. The gist, a patient found to have LGD is someone who needs
differentiation between ablative therapies is beyond the to be watched closely.
Medical and Surgical Therapy for Gastroesophageal Reflux Disease and Barrett Esophagus CHAPTER 33 345
A B
FIGURE 33.3 Endoscopic photographs of a mucosal resection. (A) Biopsy shows 1-cm lesion at 9 o’clock position, which is high-grade
dysplasia and intramucosal adenocarcinoma. (B) Pseudopolyp after endoscopic cap/band application. (C) Completed resection. The
lesion was completely excised.
Since there is risk associated with ablative therapy confirmed.64 Conflicting these data is evidence from a
and the risk of progression seems variable within the large Norwegian study showing a significant increase in the
patient population of LGD, emphasis has been placed rate of progression of LGD compared with nondysplastic
on identifying those who are at increased risk within this Barrett even when unconfirmed by a second pathologist.
population. There is some consensus that a patient with In the setting of persistent, multifocal, or long-segment
Barrett esophagus and LGD on a single biopsy should disease, a detailed discussion regarding the risks of abla-
undergo continued surveillance in the absence of high- tion (bleeding, stricture, perforation, continued need
risk features such as mucosal irregularity, multifocal, or for surveillance, progression despite ablation) should
long-segment disease. The timing of the second endoscopy be weighed against potentially decreasing the risk of
has been recommended at a wide interval from 8 weeks progression to carcinoma. These risks were demonstrated
to 12 months depending on the guidelines of several by Phoa et al.58 with 136 patients randomized at a 1 : 1 ratio
societies. LGD present on a second endoscopy is classified to either ablative therapy or continued surveillance for 3
as persistent dysplasia. A recent multicenter, randomize years. Ablation reduced the risk of progression to HGD or
trial found that the number of endoscopies with LGD was adenocarcinoma by 25.0% and the risk of progression to
an independent predictor of progression in their multi- adenocarcinoma by 7.4%. Complete eradication occurred
variate analysis.58 When LGD is not found on subsequent in 92.6% for dysplasia among patients in the ablation group
examination, this is characterized as regressive disease. and 88.2% for intestinal metaplasia. In addition, 19.1%
It is recognized that this may represent true regression, of patients receiving ablation experienced adverse events
sampling error, or interobserver variability in the histologic related to their treatment, with eight patients developing
examination. The significant effect that this determination an esophageal stricture. This study was terminated early
has on treatment decisions makes the need for evaluation due to the superiority of ablation in reducing risk of
by two expert pathologists of paramount importance. A adenocarcinoma. The available evidence suggests that at
large retrospective cohort of the Netherlands’ nationwide a diagnosis of LGD, patients should undergo aggressive
registry found that the risk of progression to HGD or management of their reflux disease, ideally with antireflux
esophageal adenocarcinoma was higher for those who surgery given multiple studies showing regression of
had a diagnosis of LGD confirmed by expert patholo- LGD postfundoplication. Repeat endoscopy and biopsies
gist compared with those where the diagnosis was not are recommended at 3 to 6 months, and if dysplasia
346 SECTION I Esophagus and Hernia
persists, ablation with a low-risk modality such as RFA is continue without change. Mucosal irregularities that
recommended to reduce the risk of progression to HGD demonstrate either LGD or HGD on EMR should undergo
or adenocarcinoma in suitable candidates. ablation of the remainder of the intestinal metaplasia.
An intramucosal (T1a) lesion may be treated with endo-
High-Grade Dysplasia and scopic ablative therapy to the surrounding Barrett, as
Intramucosal Adenocarcinoma the risk of lymph node metastasis is low. The presence of
Although esophagectomy was once the standard of care poorly differentiated histology, lymphovascular invasion,
for those diagnosed with HGD or early carcinoma, EMR or invasion into the submucosa (T1b) should prompt
has recently gained favor, as it has been shown to be discussion of further treatment plan in the setting of a
safe, effective, and less invasive than traditional resec- multidisciplinary oncology group, considering the patient’s
tion.65 EMR allows for better, more definitive histologic functional status, specific risk tolerances, and goals of
examination of the specimen for diagnosis or staging and care. In most circumstances, submucosal (T1b) invasion
is curative in nearly all patients with disease limited to represents a significant risk for lymph node metastases
the mucosa. Patients with Barrett esophagus and either and is an indication for esophagectomy and lymph node
HGD or superficial esophageal adenocarcinoma should dissection.
undergo an in-depth evaluation prior to treatment with Endoscopic techniques have demonstrated safety and
endoscopic therapies. Again, the histopathology evaluation efficacy in the treatment of early esophageal adenocar-
by two independent pathologists with expertise in this area cinoma. Pech et al.65 evaluated a cohort of 114 patients
is emphasized to decrease the rate of false-positive results treated with either transthoracic esophagectomy or EMR in
and subsequent risk of unnecessary therapy. A thorough conjunction with APC for T1 lesions. The major complica-
endoscopic evaluation at centers with high proficiency tion rate was found to be 32% in the surgery group, and
should be undertaken, using high-resolution endoscopes 0% in the endoscopic group. There was one mortality
and a meticulous systematic approach to biopsy and EMR in the surgery group, which was not tumor related, and
of concerning areas. High-resolution modalities such as none in the endoscopic group at approximately 4 years’
narrow band imaging (NBI) have shown promise for follow-up. The recurrence rate after endoscopic therapy
identifying mucosal abnormalities. However, the vascular was 6.6%, of which all were managed endoscopically.
and mucosal patterns seen with NBI have currently not EMR of the known lesion is not adequate therapy as up
been standardized or validated, making utilization outside to one-third of patients with complete resection develop
of high-volume centers limited. If referral to a high-volume metachronous lesions in the residual Barrett mucosa.
center is not possible, we recommend using the Seattle Ablation or resection of the remainder of the intestinal
protocol66 in which four-quadrant biopsies are taken at metaplasia is needed to decrease this risk.
1-cm increments, beginning at the proximal gastric folds Complete resection of T1b lesions is possible with
to the most proximal extent of intestinal metaplasia. EMR or endoscopic submucosal dissection (ESD). The
It is imperative that all mucosal abnormalities receive major factor limiting this as a curative therapy is that
separate targeted biopsies, as they are especially high risk the risk of lymph node metastasis becomes significant as
of harboring pathology of interest. The authors of this the tumor invades into the submucosa.68 Sepesi et al.69
system concluded that 50% of cancers detected using this demonstrated lymph node metastasis in 21% to 50% of
approach would have been missed by standard protocol patients depending on the level of submucosa involved
techniques. (sm1-3) in 29 patients. Increasing rates of lymph node
For patients with nonnodular Barrett esophagus and involvement have not been dependably shown based on
HGD, ablative therapy is preferable to both intense surveil- the depth of the submucosa involved, and neither positron
lance and surgery due to the risk of progression and emission tomography nor endoscopic ultrasound has
favorable morbidity profile.65 Pradad et al.67 performed a proven sensitive enough in these situations to warrant
retrospective review of treatment of HGD in 129 patients routine use. The first and most critical step in staging
who received treatment with a combination of EMR and is endoscopic resection of the lesion with pathologic
PDT. This was compared with 70 patients who underwent evaluation of the depth of invasion and risk factors for
esophagectomy. Mean follow-up was 5 years. None of node metastases including size greater than 2 cm, poor
the patients in either cohort died of esophageal cancer, differentiation, and lymphovascular invasion. There is
although one patient did die because of a postoperative some evidence that low-risk submucosal lesions may be
complication. In a multicenter, sham-controlled trial of treated endoscopically, particularly in patients at increased
127 patients by Shaheen et al.,57 complete eradication risk for esophagectomy.
was obtained in 81% of patients with HGD. The rate of
progression to esophageal adenocarcinoma was decreased RESECTION FOR DYSPLASTIC
from 19% to 2.4% (P = .04). No perforations or deaths BARRETT ESOPHAGUS
were associated with the treatment. As mentioned earlier, until recently, esophagectomy was
In instances of mucosal irregularity due to nodularity, the standard of care for the treatment of Barrett esophagus
ulceration, or irregular mucosal contour, EMR can be both with HGD. Reported rates of esophageal adenocarcinoma
diagnostic and therapeutic. The pathologic findings from found in the final pathology around 40% made any other
the EMR specimen should guide the next steps in therapy. approach prohibitively risky in the surgical candidate.70
If EMR is used in the setting of nondysplastic Barrett Two major changes have occurred that have shifted the
esophagus for mucosal irregularities and no dysplasia is use of formal resection to a second-line treatment. First,
found after complete resection, then surveillance may endoscopic therapy has emerged with a favorable risk
Medical and Surgical Therapy for Gastroesophageal Reflux Disease and Barrett Esophagus CHAPTER 33 347
and efficacy profile as detailed earlier. Second, the rate 6. Corley DA, Kubo A, Levin TR, et al. Abdominal obesity and body
of esophageal adenocarcinoma seems be below previous mass index as risk factors for Barrett’s esophagus. Gastroenterology.
2007;133:34-41, quiz 311.
estimations (as biopsy protocols and EMR make under- 7. Kramer JR, Fischbach LA, Richardson P, et al. Waist-to-hip ratio, but
staging less likely). Recent studies have demonstrated not body mass index, is associated with an increased risk of Barrett’s
values between 6% and 13%.71–73 However, this does not esophagus in white men. Clin Gastroenterol Hepatol. 2013;11:373-381,
mean that there is no longer a role for surgical resection e1.
8. Rubenstein JH, Morgenstern H, Appelman H, et al. Prediction of
in the treatment of dysplastic Barrett or intramucosal Barrett’s esophagus among men. Am J Gastroenterol. 2013;108:353-362.
adenocarcinoma. Although the rate of occult invasive 9. Andrici J, Tio M, Cox MR, Eslick GD. Hiatal hernia and the risk of
esophageal cancer is much lower than 40%, it is not 0%, Barrett’s esophagus. J Gastroenterol Hepatol. 2013;28:415-431.
and factors such as multifocal or nodular disease have 10. Fischbach LA, Graham DY, Kramer JR, et al. Association between
been associated with increased risk73,74; treatment with Helicobacter pylori and Barrett’s esophagus: a case-control study. Am
J Gastroenterol. 2014;109:357-368.
esophagectomy in this group was recently found to have 11. Wang A, Mattek NC, Holub JL, Lieberman DA, Eisen GM. Preva-
increased utility and cost effectiveness compared with lence of complicated gastroesophageal reflux disease and Barrett’s
EMR-RFA.75 Patients who have a relatively low operative esophagus among racial groups in a multi-center consortium. Dig
risk estimation based on a lack of comorbid conditions in Dis Sci. 2009;54:964-971.
12. Corley DA, Kubo A, Levin TR, et al. Race, ethnicity, sex and temporal
the setting of high-risk tumor features should be carefully differences in Barrett’s oesophagus diagnosis: a large community-
counseled regarding the treatment options and potential based study, 1994–2006. Gut. 2009;58:182-188.
outcomes. 13. Nguyen TH, Thrift AP, Ramsey D, et al. Risk factors for Barrett’s
Finally, esophagectomy should be considered in those esophagus compared between African Americans and non-Hispanic
who fail endoscopic therapy. A multicenter sham-controlled Whites. Am J Gastroenterol. 2014;109:1870-1880.
14. Su Z, Gay LJ, Strange A, et al. Common variants at the MHC locus
trial found that after four sessions (12 months’ follow-up), and at chromosome 16q24.1 predispose to Barrett’s esophagus. Nat
HGD persisted in 19% of patients. 57 With continued Genet. 2012;44:1131-1136.
treatment this number was decreased to 7% at 2 years.76 15. Rodriguez S, Mattek N, Lieberman D, Fennerty B, Eisen G. Barrett’s
The study protocol excluded those with nodular Barrett or esophagus on repeat endoscopy: should we look more than once?
Am J Gastroenterol. 2008;103:1892-1897.
disease segments greater than 8 cm, which likely improved 16. Loughney T, Maydonovitch CL, Wong RK. Esophageal manometry
their results. Pech et al. reported a failure rate of 3.7% and ambulatory 24-hour pH monitoring in patients with short and
of endoscopic therapy in a population of patients with long segment Barrett’s esophagus. Am J Gastroenterol. 1998;93:916-919.
HGD65 and 1000 consecutive patients with T1a lesions.77 17. Rudolph RE, Vaughan TL, Storer BE, et al. Effect of segment length
The definition of failed endoscopic therapy remains on risk for neoplastic progression in patients with Barrett esophagus.
Ann Intern Med. 2000;132:612-620.
largely subjective, although failure to make progress in 18. Hirota WK, Loughney TM, Lazas DJ, Maydonovitch CL, Rholl V,
eradication of disease is commonly used. Hunt et al.78 Wong RK. Specialized intestinal metaplasia, dysplasia, and cancer
studied 15 patients who underwent esophagectomy after of the esophagus and esophagogastric junction: prevalence and
endoscopic therapy. They found that those with invasive clinical data. Gastroenterology. 1999;116:277-285.
esophageal cancer on final pathology underwent more 19. Shaheen NJ, Falk GW, Iyer PG, Gerson LB. ACG clinical guideline:
procedures to attempt eradication on average (6.5 vs. diagnosis and management of Barrett’s esophagus. Am J Gastroenterol.
2016;111:30-50.
3). Following esophagectomy at specialized high-volume 20. Fitzgerald RC, di Pietro M, Ragunath K, et al. British Society of
centers, the 5-year survival rates approach 90%.79 Gastroenterology guidelines on the diagnosis and management of
Barrett’s oesophagus. Gut. 2014;63:7-42.
21. Grant KS, DeMeester SR, Kreger V, et al. Effect of Barrett’s esophagus
CONCLUSION surveillance on esophageal preservation, tumor stage, and sur-
vival with esophageal adenocarcinoma. J Thorac Cardiovasc Surg.
Much has been learned in the 66 years since Norman 2013;146:31-37.
Barrett’s early description of the condition that bears his 22. Corley DA, Mehtani K, Quesenberry C, Zhao W, de Boer J, Weiss
name. There have been several paradigm shifts in the way NS. Impact of endoscopic surveillance on mortality from Barrett’s
that we think about and treat the disease. This trend is esophagus-associated esophageal adenocarcinomas. Gastroenterology.
sure to continue and our current practices will someday 2013;145:312-319.e1.
23. Demeester SR, Demeester TR. Ineffective surveillance does not
be historic, as interventions become increasingly effica- improve survival in patients with Barrett’s who progress to adeno-
cious and less invasive. As for now, treatment of Barrett carcinoma. Gastroenterology. 2014;146:588.
esophagus remains steeped in controversy, complex, and 24. Verbeek RE, Leenders M, Ten Kate FJ, et al. Surveillance of Bar-
highly challenging across its spectrum and highlights the rett’s esophagus and mortality from esophageal adenocarcinoma: a
importance of a multispecialty and modality approach to population-based cohort study. Am J Gastroenterol. 2014;109:1215-1222.
25. Konturek PC, Burnat G, Hahn EG. Inhibition of Barret’s adeno-
these patients. carcinoma cell growth by simvastatin: involvement of COX-2 and
apoptosis-related proteins. J Physiol Pharmacol. 2007;58(suppl 3):
REFERENCES 141-148.
26. Ogunwobi OO, Beales IL. Statins inhibit proliferation and induce
1. Barrett NR. Chronic peptic ulcer of the esophagus and “oesophagitis”. apoptosis in Barrett’s esophageal adenocarcinoma cells. Am J
Br J Surg. 1950;38:175-182. Gastroenterol. 2008;103:825-837.
2. Lord RV. Norman Barrett, “doyen of esophageal surgery”. Ann Surg. 27. Nguyen T, Duan Z, Naik AD, Kramer JR, El-Serag HB. Statin use
1999;229:428-439. reduces risk of esophageal adenocarcinoma in US veterans with
3. Allison PR, Johnstone AS. The oesophagus lined with gastric mucous Barrett’s esophagus: a nested case-control study. Gastroenterology.
membrane. Thorax. 1953;8:87-101. 2015;149:1392-1398.
4. Spechler SJ. Barrett’s esophagus. Semin Gastrointest Dis. 1996;7:51-60. 28. Singh S, Singh AG, Singh PP, Murad MH, Iyer PG. Statins are associ-
5. Cook MB, Wild CP, Forman D. A systematic review and meta-analysis ated with reduced risk of esophageal cancer, particularly in patients
of the sex ratio for Barrett’s esophagus, erosive reflux disease, and with Barrett’s esophagus: a systematic review and meta-analysis. Clin
nonerosive reflux disease. Am J Epidemiol. 2005;162:1050-1061. Gastroenterol Hepatol. 2013;11:620-629.
348 SECTION I Esophagus and Hernia
29. Beales IL, Hensley A, Loke Y. Reduced esophageal cancer incidence 52. Gurski RR, Peters JH, Hagen JA, et al. Barrett’s esophagus can and
in statin users, particularly with cyclo-oxygenase inhibition. World J does regress after antireflux surgery: a study of prevalence and
Gastrointest Pharmacol Ther. 2013;4:69-79. predictive features. J Am Coll Surg. 2003;196:706-712, discussion
30. Buttar NS, Wang KK, Anderson MA, et al. The effect of selective 712-713.
cyclooxygenase-2 inhibition in Barrett’s esophagus epithelium: an 53. Zaninotto G, Parente P, Salvador R, et al. Long-term follow-up of
in vitro study. J Natl Cancer Inst. 2002;94:422-429. Barrett’s epithelium: medical versus antireflux surgical therapy.
31. Heath EI, Canto MI, Piantadosi S, et al. Secondary chemoprevention J Gastrointest Surg. 2012;16:7-14, discussion 14-15.
of Barrett’s esophagus with celecoxib: results of a randomized trial. 54. Zehetner J, DeMeester SR, Ayazi S, et al. Long-term follow-up after
J Natl Cancer Inst. 2007;99:545-557. anti-reflux surgery in patients with Barrett’s esophagus. J Gastrointest
32. Kastelein F, Spaander MC, Steyerberg EW, et al. Proton pump Surg. 2010;14:1483-1491.
inhibitors reduce the risk of neoplastic progression in patients with 55. Lagergren J, Ye W, Lagergren P, Lu Y. The risk of esophageal adenocar-
Barrett’s esophagus. Clin Gastroenterol Hepatol. 2013;11:382-388. cinoma after antireflux surgery. Gastroenterology. 2010;138:1297-1301.
33. El-Serag HB, Aguirre TV, Davis S, Kuebeler M, Bhattacharyya A, 56. Lofdahl HE, Lu Y, Lagergren P, Lagergren J. Risk factors for esopha-
Sampliner RE. Proton pump inhibitors are associated with reduced geal adenocarcinoma after antireflux surgery. Ann Surg. 2013;257:
incidence of dysplasia in Barrett’s esophagus. Am J Gastroenterol. 579-582.
2004;99:1877-1883. 57. Shaheen NJ, Sharma P, Overholt BF, et al. Radiofrequency ablation in
34. Nguyen DM, El-Serag HB, Henderson L, Stein D, Bhattacharyya A, Barrett’s esophagus with dysplasia. N Engl J Med. 2009;360:2277-2288.
Sampliner RE. Medication usage and the risk of neoplasia in patients 58. Phoa KN, van Vilsteren FG, Weusten BL, et al. Radiofrequency
with Barrett’s esophagus. Clin Gastroenterol Hepatol. 2009;7:1299-1304. ablation vs endoscopic surveillance for patients with Barrett
35. Gerson LB, Boparai V, Ullah N, Triadafilopoulos G. Oesophageal esophagus and low-grade dysplasia: a randomized clinical trial.
and gastric pH profiles in patients with gastro-oesophageal reflux JAMA. 2014;311:1209-1217.
disease and Barrett’s oesophagus treated with proton pump inhibitors. 59. Overholt BF, Lightdale CJ, Wang KK, et al. Photodynamic therapy
Aliment Pharmacol Ther. 2004;20:637-643. with porfimer sodium for ablation of high-grade dysplasia in Barrett’s
36. Yachimski P, Maqbool S, Bhat YM, Richter JE, Falk GW, Vaezi MF. esophagus: international, partially blinded, randomized phase III
Control of acid and duodenogastroesophageal reflux (DGER) trial. Gastrointest Endosc. 2005;62:488-498.
in patients with Barrett’s esophagus. Am J Gastroenterol. 2015;110: 60. Manner H, May A, Miehlke S, et al. Ablation of nonneoplastic
1143-1148. Barrett’s mucosa using argon plasma coagulation with concomitant
37. Spechler SJ, Sharma P, Traxler B, Levine D, Falk GW. Gastric and esomeprazole therapy (APBANEX): a prospective multicenter
esophageal pH in patients with Barrett’s esophagus treated with evaluation. Am J Gastroenterol. 2006;101:1762-1769.
three esomeprazole dosages: a randomized, double-blind, crossover 61. Ferraris R, Fracchia M, Foti M, et al. Barrett’s oesophagus: long-term
trial. Am J Gastroenterol. 2006;101:1964-1971. follow-up after complete ablation with argon plasma coagulation
38. Thrift AP, Kramer JR, Qureshi Z, Richardson PA, El-Serag HB. Age and the factors that determine its recurrence. Aliment Pharmacol
at onset of GERD symptoms predicts risk of Barrett’s esophagus. Ther. 2007;25:835-840.
Am J Gastroenterol. 2013;108:915-922. 62. Gray NA, Odze RD, Spechler SJ. Buried metaplasia after endoscopic
39. Eisen GM, Sandler RS, Murray S, Gottfried M. The relationship ablation of Barrett’s esophagus: a systematic review. Am J Gastroenterol.
between gastroesophageal reflux disease and its complications with 2011;106:1899-1908, quiz 1909.
Barrett’s esophagus. Am J Gastroenterol. 1997;92:27-31. 63. Wani S, Falk GW, Post J, et al. Risk factors for progression of low-
40. Kauer WK, Peters JH, DeMeester TR, Ireland AP, Bremner CG, Hagen grade dysplasia in patients with Barrett’s esophagus. Gastroenterology.
JA. Mixed reflux of gastric and duodenal juices is more harmful to 2011;141:1179-1186, 1186.e1.
the esophagus than gastric juice alone. The need for surgical therapy 64. Duits LC, Phoa KN, Curvers WL, et al. Barrett’s oesophagus patients
re-emphasized. Ann Surg. 1995;222:525-531, discussion 31-33. with low-grade dysplasia can be accurately risk-stratified after histologi-
41. Sun D, Wang X, Gai Z, Song X, Jia X, Tian H. Bile acids but not cal review by an expert pathology panel. Gut. 2015;64:700-706.
acidic acids induce Barrett’s esophagus. Int J Clin Exp Pathol. 2015;8: 65. Pech O, Bollschweiler E, Manner H, Leers J, Ell C, Holscher AH.
1384-1392. Comparison between endoscopic and surgical resection of mucosal
42. Vaezi MF, Richter JE. Synergism of acid and duodenogastroesopha- esophageal adenocarcinoma in Barrett’s esophagus at two high-
geal reflux in complicated Barrett’s esophagus. Surgery. 1995;117: volume centers. Ann Surg. 2011;254:67-72.
699-704. 66. Reid BJ, Blount PL, Feng Z, Levine DS. Optimizing endoscopic
43. Fein M, Ireland AP, Ritter MP, et al. Duodenogastric reflux potentiates biopsy detection of early cancers in Barrett’s high-grade dysplasia.
the injurious effects of gastroesophageal reflux. J Gastrointest Surg. Am J Gastroenterol. 2000;95:3089-3096.
1997;1:27-32, discussion 33. 67. Prasad GA, Wang KK, Buttar NS, et al. Long-term survival following
44. Stein HJ, Feussner H, Kauer W, DeMeester TR, Siewert JR. Alkaline endoscopic and surgical treatment of high-grade dysplasia in Barrett’s
gastroesophageal reflux: assessment by ambulatory esophageal esophagus. Gastroenterology. 2007;132:1226-1233.
aspiration and pH monitoring. Am J Surg. 1994;167:163-168. 68. Badreddine RJ, Prasad GA, Lewis JT, et al. Depth of submucosal
45. Galmiche JP, Hatlebakk J, Attwood S, et al. Laparoscopic antireflux invasion does not predict lymph node metastasis and survival of
surgery vs esomeprazole treatment for chronic GERD: the LOTUS patients with esophageal carcinoma. Clin Gastroenterol Hepatol.
randomized clinical trial. JAMA. 2011;305:1969-1977. 2010;8:248-253.
46. Attwood SE, Lundell L, Hatlebakk JG, et al. Medical or surgical 69. Sepesi B, Watson TJ, Zhou D, et al. Are endoscopic therapies appropri-
management of GERD patients with Barrett’s esophagus: the LOTUS ate for superficial submucosal esophageal adenocarcinoma? An
trial 3-year experience. J Gastrointest Surg. 2008;12:1646-1654, [discus- analysis of esophagectomy specimens. J Am Coll Surg. 2010;210:418-427.
sion 54-55]. 70. Edwards MJ, Gable DR, Lentsch AB, Richardson JD. The rationale
47. Hofstetter WL, Peters JH, DeMeester TR, et al. Long-term outcome for esophagectomy as the optimal therapy for Barrett’s esophagus
of antireflux surgery in patients with Barrett’s esophagus. Ann Surg. with high-grade dysplasia. Ann Surg. 1996;223:585-589, discussion
2001;234:532-538, discussion 538-539. 589-591.
48. Parrilla P, Martinez de Haro LF, Ortiz A, et al. Long-term results 71. Nasr JY, Schoen RE. Prevalence of adenocarcinoma at esophagectomy
of a randomized prospective study comparing medical and surgical for Barrett’s esophagus with high grade dysplasia. J Gastrointest Oncol.
treatment of Barrett’s esophagus. Ann Surg. 2003;237:291-298. 2011;2:34-38.
49. Morrow E, Bushyhead D, Wassenaar E, et al. The impact of laparo- 72. Wang VS, Hornick JL, Sepulveda JA, Mauer R, Poneros JM. Low
scopic anti-reflux surgery in patients with Barrett’s esophagus. Surg prevalence of submucosal invasive carcinoma at esophagectomy for
Endosc. 2014;28:3279-3284. high-grade dysplasia or intramucosal adenocarcinoma in Barrett’s
50. Oelschlager BK, Barreca M, Chang L, Oleynikov D, Pellegrini esophagus: a 20-year experience. Gastrointest Endosc. 2009;69:777-783.
CA. Clinical and pathologic response of Barrett’s esophagus to 73. Konda VJ, Ross AS, Ferguson MK, et al. Is the risk of concomitant
laparoscopic antireflux surgery. Ann Surg. 2003;238:458-464, discussion invasive esophageal cancer in high-grade dysplasia in Barrett’s
464-466. esophagus overestimated? Clin Gastroenterol Hepatol. 2008;6:159-164.
51. Brand DL, Ylvisaker JT, Gelfand M, Pope CE 2nd. Regression of 74. Buttar NS, Wang KK, Sebo TJ, et al. Extent of high-grade dysplasia
columnar esophageal (Barrett’s) epithelium after anti-reflux surgery. in Barrett’s esophagus correlates with risk of adenocarcinoma.
N Engl J Med. 1980;302:844-848. Gastroenterology. 2001;120:1630-1639.
Medical and Surgical Therapy for Gastroesophageal Reflux Disease and Barrett Esophagus CHAPTER 33 349
75. Hu Y, Puri V, Shami VM, Stukenborg GJ, Kozower BD. Comparative 78. Hunt BM, Louie BE, Dunst CM, et al. Esophagectomy for failed
effectiveness of esophagectomy versus endoscopic treatment for endoscopic therapy in patients with high-grade dysplasia or intramu-
esophageal high-grade dysplasia. Ann Surg. 2016;263:719-726. cosal carcinoma. Dis Esophagus. 2014;27:362-367.
76. Shaheen NJ, Overholt BF, Sampliner RE, et al. Durability of radiofre- 79. Peters JH, Clark GW, Ireland AP, Chandrasoma P, Smyrk TC,
quency ablation in Barrett’s esophagus with dysplasia. Gastroenterology. DeMeester TR. Outcome of adenocarcinoma arising in Barrett’s
2011;141:460-468. esophagus in endoscopically surveyed and nonsurveyed patients.
77. Pech O, May A, Manner H, et al. Long-term efficacy and safety of J Thorac Cardiovasc Surg. 1994;108:813-821, discussion 21-22.
endoscopic resection for patients with mucosal adenocarcinoma of
the esophagus. Gastroenterology. 2014;146:652-660.e1.
CHAPTER
C
ompared with the general population, patients BE and when dysplastic changes are reported, that this is
with Barrett esophagus (BE) have a higher risk a true reflection of a potential neoplastic process. Chapter
of developing esophageal adenocarcinoma (EAC) 31 has addressed the issues related to the definition of
with the risk increasing in those patients who develop Barrett intestinal metaplasia that, when present, will lead to
dysplasia—low-grade dysplasia (LGD) or high-grade endoscopic surveillance. If there is a diagnosis of dysplasia
dysplasia (HGD).1 In the past, patients with HGD or at endoscopy, the management implications change for the
proven intramucosal adenocarcinoma (IMC) will have been patient and the clinician. However, pathologic consensus
considered for an esophagectomy, if they were medically of the histologic diagnosis can be a problem, and the dys-
suitable. This major procedure clearly removes the pathol- plastic changes may not be present on subsequent biopsies,
ogy, as well as the whole Barrett segment, but at a cost of particularly LGD. The diagnosis of HGD, confirmed by a
morbidity and mortality. Now the evidence supports the second pathologist with gastrointestinal (GI) expertise,
use of endoscopic therapies with endoscopic resection of carries a clear risk for progression to adenocarcinoma such
localized lesions such as early mucosal adenocarcinoma that all these patients should be considered for definitive
or segments of HGD, aiming to completely remove the treatment of the neoplastic lesion as well as ablation of the
neoplastic segment. However, the residual untreated BE whole BE segment. The alternative is an esophagectomy.5–7
is left in situ with a significant risk of recurrent neoplasia, The presence of visible lesions (nodules or ulceration)
being reported in up to one-third of patients.2,3 This has led in a segment of HGD carries significant implications. If
to a focus on therapies aimed at eradicating this high-risk nodules are present, there is a 2.6 times potential for
mucosa allowing regrowth of squamous epithelium. With progression to EAC,8 and if ulceration is present, the
the incidence of BE and adenocarcinoma increasing in risk of the presence of EAC in a high-grade dysplastic
Western countries along with improved outcomes and segment has been reported to be 80%, compared with
greater experience of Barrett mucosal ablation, there has 52% if there was no ulceration.9 Endoscopic resection of
been a trend to expand ablation to patients who have not these abnormalities, if possible, offers better pathologic
had HGD or IMC but may have LGD or even consider staging and complete resection and should be performed
ablation in patient with nondysplastic BE (NDBE). Thus before attempts at BE ablation. Patients with a diagnosis
for every patient the role and value of these treatments of IMC that has been endoscopically completely removed
must be weighed against the impact of the treatment on will need ablation of the residual BE segment.
the individual from a cancer prevention perspective, the Patients with confirmed LGD in BE will require more
impact on quality of life both after treatment and in the regular endoscopic surveillance, and there is evidence to
long term, and the cost implications to the community. support considering this group for complete BE ablation.
The optimal ablative therapy would completely eliminate However, the diagnosis of LGD needs to be secure. There
the mucosa to the submucosa, in a single session, with has been a meta-analysis of the outcomes from endoscopic
very few side effects, and offer the patient a lifetime therapies for LGD, which assessed 37 studies comprising
guarantee of no recurrence following complete squamous 521 patients. The multiple techniques of ablation used
re-epithelialization of the treated segment.4 We are not provided complete eradication (CE) of dysplasia in 88.9%
there yet. The ablation techniques that have been used and intestinal metaplasia (IM) in 67.8%, with a pooled
include endoscopic resection of a focal lesion or of the incidence of progression to cancer of 3.9 (95% CI, 1.27 to
complete BE segment and mucosal ablation techniques 9.1) per 1000 patient-years. The authors concluded that
such as radiofrequency ablation (RFA), photodynamic there was likely to be histologic overdiagnosis of LGD.
therapy (PDT), argon plasma coagulation (APC), laser RFA was the most safe and effective option in this group,
therapy, and cryotherapy. Presently the most commonly but ablation did not eliminate the risk of progression to
applied techniques are endoscopic resection of a focal HGD or EAC.10
abnormality with RFA of the residual dysplastic or non- The potential for “overdiagnosis” of LGD and the
dysplastic Barrett esophagus (NDBE). The indications consequences have been highlighted by a number of
for Barrett ablation, along with the techniques and their studies. In a group of patients diagnosed with LGD by
outcomes, will be described in this chapter. community pathologists, where subsequent review by expert
GI pathologists occurred, the diagnosis was downgraded
in 85% of patients to NDBE or indefinite dysplasia.11 In
WHICH PATIENTS WITH BARRETT the “downgraded group” the risk of progression to HGD
SHOULD HAVE ABLATION? or EAC was 0.49% per patient per year, compared with
the group confirmed to have LGD where the progression
For a clinician to consider performing a significant inter- was 13.4%.11 In a Dutch study, the percentage of diagnoses
vention on a patient, one must be secure the diagnosis is downgraded from LGD to indeterminate or NDBE was
350
Ablation for Patients With Barrett or Dysplasia CHAPTER 34 350.e1
ABSTRACT
The evidence now supports endoscopic therapy as the gold
standard in patients with high-grade dysplasia in Barrett
esophagus. The combination of endoscopic mucosal
resection of visible abnormalities and associated ablation
of the residual intestinal metaplasia with radiofrequency
ablation (RFA) offers low morbidity with the most efficient
and durable short- and long-term outcomes. Other abla-
tive techniques include photodynamic therapy, argon
plasma coagulation, and cryotherapy, but these may have
higher morbidity and higher recurrence rates of intestinal
metaplasia and higher rates of buried Barrett compared
with RFA. The use of mucosal ablation for low-grade
dysplasia can be considered under strict pathologic and
clinical guidelines but should not be used in patients
with nondysplastic intestinal metaplasia. All patients who
have had endoscopic therapy for Barrett dysplasia require
long-term acid suppression therapy and careful endoscopic
follow-up.
KEYWORDS
Barrett esophagus, low-grade dysplasia, high-grade
dysplasia, Barrett ablation, endoscopic mucosal resection,
sequential radical endoscopic resection, radiofrequency
ablation, photodynamic therapy, argon plasma coagulation,
cryotherapy, subsquamous Barrett
Ablation for Patients With Barrett or Dysplasia CHAPTER 34 351
73%, where expert pathologists reviewed the histology of sampling, performed at the surveillance endoscopy. The
293 referred patients. At a median follow-up of 39 months, patient perspective was highlighted in a study assessing
21 of the 75 (27%) confirmed LGD patients progressed patient preference in the management of NDBE where the
to HGD/EAC. The rate of progression was measured at options, hypothetically, were put for either chemopreven-
9.1% per patient-year compared with the pathologically tion with aspirin and 3 to 5 yearly surveillance endoscopies
reviewed NDBE group who had a rate of progression of or endoscopic ablation. The patients preferred ablation.21
0.6 per patient-year.12 For ablation to be a realistic option in asymptomatic
The multicenter European SURF trial randomized 136 patients with NDBE, the treatment must be safe, be effec-
patients with LGD to receive RFA ablation to the BE or tive, have durable long-term results, and be cost effective.
surveillance, with both groups receiving proton pump The safest, most effective treatment is RFA. In a cohort
inhibitor (PPI) therapy. There was CE of the dysplasia in study of patients with NDBE who had RFA and regular
93% of the RFA group and 28% of the controls and CE of follow-up with treatment of residual or recurrent IM, the
the IM in 88% of the RFA group compared with zero in complete regression rate was 70% at 1 year22 and 92%
the controls. At 3-year follow-up, the rate of progression at 5 years.23 This group of patients required multiple
to HGD was 1.5% after RFA compared with 26.5% in the endoscopies and the need for more intense surveillance
surveillance group, and the rate of progression to EAC was than recommended for nontreated NDBE. The impact of
1.5% compared with 8.8%, respectively. They reported a this approach on the patients’ longevity and quality of life
neoplasia progression rate of 8% per year.13 The rate of along with the cost effectiveness is yet to be clearly defined.
progression of LGD confirmed by an expert pathologist was With respect to cost effectiveness of BE ablation, in 2004,
further assessed in a cohort study of 170 patients, with 45 a study examined the management of HGD comparing
undergoing RFA ablation and 125 surveillance with median endoscopic surveillance, endoablation (using PDT), and
follow-up of more than 2 years. The rate of progression esophagectomy. Endoablation with PDT was shown to be
to HGD or EAC in the RFA group was 0.77% compared the most effective strategy.24 A recent literature review
with 6.6% in the surveillance group. The measured risk concluded that endoscopic therapy for dysplastic BE,
of progression after RFA had a hazard ratio of 0.08 (95% using PDT or RFA, was cost effective compared with
CI, 0.01 to 0.61). In the surveillance group, following a esophagectomy.25 One single-institution cohort study
multivariate analysis, independent factors associated with reported the cost of PDT to be 5 times that of RFA. 26
the risk of progression were nodularity in the BE and Using a Markov model, assessing patients with HGD,
multifocal dysplasia.14 RFA with continued surveillance is more cost effective
For LGD to be considered confirmed, guidelines recom- than endoscopic surveillance and esophagectomy when
mend a second pathologist with GI expertise review the a cancer develops.27
biopsies, and for a second endoscopy and biopsy after 6 For patients with LGD, if confirmed by two pathologists
months to reassess the BE and confirm the continued and the diagnosis remains stable on repeated endoscopies
presence of the LGD.15–18 In this group of patients with when assessing quality-adjusted life years (QALYs), RFA is
confirmed LGD, the data for a higher risk for progression more cost effective than continued regular surveillance
to HGD and IMC are clear.13 Using this definition for LGD, and performing the RFA when HGD develops.27 Although
guidelines in the United States recommend endoscopic the American Society of Gastroenterology (ASGE) has
ablation to be appropriate for patients with confirmed recommended ablation is an “option” for NDBE,16 the
LGD.18 In the UK the published guidelines in 201415 were evidence suggests this is not cost effective, due to the low
updated in 2016 when reviewed evidence supported the rate of progression to adenocarcinoma in this group of
role of RFA to ablate confirmed LGD.19 The factors that patients.25,27
have been associated with higher rates of progression of At this time, for NDBE there is no evidence that supports
LGD to HGD/EAC include male gender, NDBE present the routine use of ablative therapies. In the future, for
for more than 10 years, length of BE (>3 cm), persistent ablation to be considered, there will need to be a number
esophagitis, multifocal dysplasia, and the presence of of factors better defined. They may include selection of
nodules in the BE mucosa.5,20 In a consensus statement patients with longer life expectancy than average BE
from the BOBCAT group, it was agreed that there was patients; identifying patients with higher risk for progres-
“moderate quality evidence” to support the ablation of sion than the average patient without dysplasia, most likely
the high-risk LGD group. The criteria for considering through a panel of biomarkers; identifying which patients
ablating this group were consensus of the diagnosis by will do well with RFA and which group has a higher risk of
two expert pathologists, persistence of the LGD over time, failing; and finally creating a situation whereby minimal
multifocal dysplasia, and longer BE segments.17 surveillance is required, as well as a reduced need to
“touch up” the ablated segments.28
300 patients being considered for endoscopic resection. FIGURE 34.1 Appearance post radiofrequency ablation.
The selection criteria were visible lesion, multiple lesions,
and lesions larger than 15 mm or poorly lifting with diameter to allow selection of the correct size ablation
submucosal injection.40 En bloc resection of the relevant balloon catheter.42 The ablation catheter is positioned, and
lesion occurred in 90%. Residual IM required further then there is a double application of the energy applied
treatment (median 2) in 62%, using EMR, ESD, APC, or (Fig. 34.1). RFA offers the most reliable destruction of the
RFA. At a median of 20 months, the CE rate of neoplasia mucosa to the level of the submucosa. The efficacy of this
was 92% and IM 73%. There were two patients who had technique at the time of the first application is reduced
a delayed hemorrhage and three with perforation, all if there is a large hiatus hernia, longer segments of BE,
treated endoscopically. The stricture rate was 60% with esophagitis, and a narrow esophagus.42
regular dilations required; for some patients this was long The pivotal AIM trial was an RCT, which compared
term.40 A study of 30 patients reported IMC and HGD RFA with a sham endoscopy, with both groups receiving
complete resection rates of 97%, with en bloc removal of high-dose PPI therapy, and with recruitment on a 2 : 1
the abnormal lesion in 90%. The R0 rate was 39% due to ratio, respectively. The 127 patients were stratified for
lateral margin involvement with dysplasia.41 the grade of dysplasia and the length of IM, and had
ESD is more time-consuming, has more serious com- intensive postprocedural surveillance. In the patients
plications, and requires an extra level of expertise. There with HGD, at 12 months, on an intention-to-treat (ITT)
is no clear evidence for improved rates of CE of dysplasia analysis, following a mean of 3.5 treatments per patient,
or IM or reduced rate of progression to EAC, compared there was CE of the dysplasia in 81% compared with 19%
with SRER or focal EMR followed by BE ablation notably in the controls. The segment of intestinal metaplasia was
using RFA. It would seem that selective use of ESD in eradicated in 74% in the RFA group, compared with zero
specialist units will remain the standard unless there are in the control group. In this high-risk population, one
RCTs showing a benefit over EMR in the future. The patient (2%) in the RFA group progressed to EAC and
outcomes are summarized in Table 34.2. four (19%) in the sham group.6 This trial also recruited
patients with LGD. At 12 months, in the RFA group CE of
MUCOSAL ABLATION the dysplasia occurred in 91% of the patients compared
The most frequently used mucosal ablation technique is with 23% in the control patients. The number needed
RFA, because this procedure produces a more predictable to treat (NNT) to avoid one persistent dysplasia was 1.5
degree of mucosal eradication and has a low side-effect patients.6 At 3 years, the RFA group had CE of all dysplasia
profile. in 94% of patients and IM in 91%.43
Patients may require a number of treatments to achieve
RADIOFREQUENCY ABLATION complete BE ablation. A study in the United States reported
Ablation occurs through the transfer of heat using radio- CE after one treatment in 29%, two treatments 35%,
frequency energy applied to the mucosa of the esophagus. and for the rest it took up to 10 treatments. CE was
The energy is delivered with an RFA delivery HALO device achieved more quickly in younger patients and those with
(BARRX Medical, Sunnydale, California) 10 to 12 J/cm2, short-segment BE. This group reported a recurrence rate
40 W/cm2. The devices used to transfer the energy are a of IM of 20% at 1 year and 33% at 2 years. Others have
HALO 360, 10 cm balloon used for circumferential energy also reported up to 55% of patients will require repeat
transfer, or HALO 90 and 60 devices to allow treatment ablations after the first 12 months to achieve ablation
of Barrett islands. The energy delivered causes water levels above 90%.43
vaporization, protein coagulation, and tissue necrosis. Larger cohort studies have confirmed the efficacy of
The HALO 360 involves a sizing balloon that generates RFA. Most of these studies included patients with neoplastic
pressure volume data and determines the esophageal lesions endoscopically removed with subsequent RFA
354 SECTION I Esophagus and Hernia
TABLE 34.4 Outcomes from Photodynamic Therapy for TABLE 34.5 Outcomes from Argon Plasma Coagulation
Barrett Esophagus for Barrett Esophagus
Outcomes % of Patients Outcomes % of Patients
EFFICACY EFFICACY
Complete eradication of dysplasia 50%–80% Complete eradication of dysplasia 67%
Complete eradication of IM 13%–52% Complete eradication of IM 38%–99%
DURABILITY DURABILITY
5 years complete eradication of dysplasia 48% 1 year complete eradication of IM 84%
8 years complete eradication of IM 66%
COMPLICATIONS
16 years complete eradication of IM 50%
Stricture 9%–36%
Photosensitivity 6%–43% COMPLICATIONS
Buried Barrett 14%–20% Stricture 9%–11%
Buried Barrett 21%–24%
IM, Intestinal metaplasia.
IM, Intestinal metaplasia.
profile that includes a high rate of stricture formation at 1 year of 87% and at 7 years 74%, compared with the
and photosensitivity. APC is relatively easy to apply, but medically treated patients who had an IM ablation rate of
durability of the CE rate is disappointing, with recurrence 79% at 1 year and 54% at 7 years. The numbers in the trial
of the IM occurring in long-term studies. The results were small and the difference between methods of acid
from APC can be dependent upon the operator, as well control was not significant. There was also no difference in
as factors such as the energy settings, APC mode, and the the rate of dysplasia in each group.70 Presently antireflux
distance of the catheter from the mucosa and time of surgery is not considered to be an antineoplastic measure
energy transmission. For longer segments of Barrett the in patients with BE, so that the indications for surgery
procedure can be very time-consuming. The complication in patients who have had BE ablation is the same as for
profile is not high, but strictures do occur. The data from a population of patients with GERD who have not had
CRYO studies have not suggested equivalent efficacy or the procedure.18
durability when compared with RFA, and the numbers
relate to small cohort studies. CRYO or APC may be
reasonable alternative therapies if a patient had recurrence
following previous RFA ablative therapy.
SURVEILLANCE
It will typically take a number of episodes of treatments
to achieve CE or near CE of the BE segment, no matter
ROLE OF ACID CONTROL BEFORE AND which ablative technique is being used. There are varying
AFTER ABLATION TREATMENT rates of recurrence of dysplasia and IM with the different
ablative techniques, with RFA ablation achieving the more
Following the mucosal injury, most studies recommend a durable results. One study reported a 9% per year rate of
minimum medical therapy of a double-dose PPI follow- recurrence of IM post apparent complete BE eradication
ing ablation. Once the injury has healed, patients are by RFA ablation. Often further treatment episodes will
maintained with a normal dose of PPI. One of the lower be required after the first year.43 The mean time to IM
rates of recurrence of IM (6%) following RFA has been recurrence in patients with NDBE has been reported to
reported from a study in the Netherlands where patients be around 2 years.52 A study of 166 patients, assessing the
received high-dose PPI (twice daily esomeprazole 40 mg), long-term control following treatment of neoplastic BE
ranitidine (300 mg at night), and sucralfate (5 mL [200 mg/ with multiple modalities including endoscopic resection,
mL] 4 times a day for 2 weeks) and then a normal dose of PDT, and RFA, or a combination of these treatments,
maintenance PPI.46 The role of maintenance therapy is not reported CE of the neoplastic BE in 95% and IM 83%.
clear. In the general population there is a paucity of data to At follow-up of 33 months (range 18 to 58 months), the
support routine use of regular acid suppression in a patient recurrent IM rate was 35% and dysplasia 9%. Retreat-
with BE. PPI therapy is typically prescribed for symptom ment achieved remission in 90%.89 It is clear that for
control of gastroesophageal reflux disease (GERD). A patients treated with Barrett neoplasia, long-term careful
systematic review of cohort studies reported that the use surveillance will be necessary.26
of PPIs in patients with NDBE produced a decreased risk Presently the data used to inform the surveillance
of progression to neoplastic BE compared with no acid periods following BE ablation is based on cohort studies
suppression.84 There is evidence that incomplete control with the longer term follow-up details, and recommen-
of GERD may be associated with increased recurrence dations have been reported in recent guideline docu-
rates of IM following successful ablation,85,86 although ments.18,45,46 For patients who have had HGD and/or IMC
there is no evidence for the degree of control and the resected with subsequent BE ablation, it is recommended
relationship to the rate of recurrence. The presence of they should have 3-monthly reviews in the first year,
uncontrolled acid reflux before RFA ablation predisposes 6-monthly reviews in the second year, and then annually. For
patients to a higher rate of incomplete response after RFA.87 patients who have had ablation for LGD, it is recommended
Thus at this time, despite the level of evidence not being they have 6-monthly reviews in the first year and then
high, patients should have their GERD controlled with annually. The tubular esophagus and the gastroesophageal
a PPI prior to ablative therapy, and it is recommended junction should be assessed visually, and four-quadrant
that patients are maintained on a regular PPI following biopsies at 1 cm levels over the previous IM segment
ablation therapy. and thecardia should be taken. Recurrent metaplasia/
The role of antireflux surgery in the form of a fundo- dysplasia should be treated according to the histology of
plication to control GERD long term has been assessed the lesion.18
through analysis of data in the US RFA registry. There Whether the surveillance periods can be extended will
was no difference in the CE of dysplastic epithelium remain a source of research with assessment of longer-term
or intestinal metaplasia among 301 patients who had a data. The AIM II trial assessed the durability of RFA
fundoplication, compared with those who had not had in patients with NDBE, 2 to 6 cm in length, at 5 years
antireflux surgery.88 Another group assessed the role of posttreatment, in 50 out of the 70 patients recruited.
APC and acid control compared with surveillance and acid The CE rate was 92%, and those who had focal IM had
control. They performed two RCTs where the difference this successfully eradicated with focal RFA. In this group
in each trial was the form of long-term acid control, of patients there were no patients with dysplasia or
one trial with regular PPI and the other group having buried Barrett.23 It may be that once dysplasia is cleared,
antireflux surgery. The group who had antireflux surgery the times between surveillance endoscopies can be
(a fundoplication) had a rate of at least 95% IM ablation extended.
Ablation for Patients With Barrett or Dysplasia CHAPTER 34 359
22. Sharma VK, Wang KK, Overholt BF, et al. Balloon-based, circumferen- patients with neoplastic Barrett’s esophagus. Clin Gastroenterol Hepatol.
tial, endoscopic radiofrequency ablation of Barrett’s esophagus: 1-year 2013;11(6):636-642.
follow-up of 100 patients. Gastrointest Endosc. 2007;65(2):185-195. 45. Haidry RJ, Dunn JM, Butt MA, et al. Radiofrequency ablation and
23. Fleischer DE, Overholt BF, Sharma VK, et al. Endoscopic radiofre- endoscopic mucosal resection for dysplastic Barrett’s esophagus and
quency ablation for Barrett’s esophagus: 5-year outcomes from a early esophageal adenocarcinoma: outcomes of the UK National
prospective multicenter trial. Endoscopy. 2010;42(10):781-789. Halo RFA Registry. Gastroenterology. 2013;145(1):87-95.
24. Shaheen NJ, Inadomi JM, Overholt BF, Sharma P. What is the best 46. Phoa KN, Pouw RE, Van Vilsteren FGI, et al. Remission of Barrett’s
management strategy for high grade dysplasia in Barrett’s oesophagus? esophagus with early neoplasia 5 years after radiofrequency ablation
A cost effectiveness analysis. Gut. 2004;53(12):1736-1744. with endoscopic resection: a Netherlands cohort study. Gastroenterology.
25. Gerson LB. Cost-analyses studies in Barrett’s esophagus: what is 2013;145(1):96-104.
their utility? Gastroenterol Clin N Am. 2015;44(2):425-438. 47. van Vilsteren FGI, Herrero LA, Pouw RE, et al. Predictive factors
26. Ertan A, Zaheer I, Correa AM, Thosani N, Blackmon SH. for initial treatment response after circumferential radiofrequency
Photodynamic therapy vs radiofrequency ablation for Barrett’s ablation for Barrett’s esophagus with early neoplasia: a prospective
dysplasia: efficacy, safety and cost-comparison. World J Gastroenterol. multicenter study. Endoscopy. 2013;45(7):516-525.
2013;19(41):7106-7113. 48. Korst RJ, Santana JS, Rutledge JR, et al. Patterns of recurrence and
27. Hur C, Choi SE, Rubenstein JH, et al. The cost-effectiveness of persistent intestinal metaplasia after successful radiofrequency ablation
radiofrequency ablation for Barrett’s esophagus. Gastroenterology. of Barrett’s esophagus. J Thorac Cardiovasc Surg. 2013;145:1529-1534.
2012;142(5):S73. 49. Orman ES, Kim HP, Bulsiewicz WJ, et al. Intestinal metaplasia recurs
28. Bergman JJGHM, Corley DA. Barrett’s esophagus: who should infrequently in patients successfully treated for Barrett’s esophagus
receive ablation and how can we get the best results? Gastroenterology. with radiofrequency ablation. Am J Gastroenterol. 2013;108(2):187-195.
2012;143(3):524-526. 50. Dulai PS, Pohl H, Levenick JM, Gordon SR, MacKenzie TA, Rothstein
29. Peters FR, Brakenhoff KPM, Curvers WL, et al. Histologic evaluation RI. Radiofrequency ablation for long- and ultralong-segment Barrett’s
of resection specimens obtained at 293 endoscopic resections in esophagus: a comparative long-term follow-up study. Gastrointest
Barrett’s esophagus. Gastrointest Endosc. 2008;67(4):604-609. Endosc. 2013;77(4):534-541.
30. Mino-Kenudson M, Hull MJ, Brown I, et al. EMR for Barrett’s 51. Gupta M, Iyer PG, Lutzke L, et al. Recurrence of esophageal intestinal
esophagus-related superficial neoplasms offers better diagnostic metaplasia after endoscopic mucosal resection and radiofrequency
reproducibility than mucosal biopsy. Gastrointest Endosc. 2007;66(4): ablation of Barrett’s esophagus: results from a US multicenter
660-666. consortium. Gastroenterology. 2013;145(1):79-86.
31. May A, Gossner L, Behrens A, et al. A prospective randomized trial 52. Pasricha S, Bulsiewicz WJ, Hathorn KE, et al. Durability and predictors
of two different endoscopic resection techniques for early stage of successful radiofrequency ablation for Barrett’s esophagus. Clin
cancer of the esophagus. Gastrointest Endosc. 2003;58(2):167-175. Gastroenterol Hepatol. 2014;12(11):1840-1847.
32. Peters FR, Kara MA, Curvers WL, et al. Multiband mucosectomy for 53. Morales TG, Camargo E, Bhattacharyya A, Sampliner RE. Long-
endoscopic resection of Barrett’s esophagus: feasibility study with term follow-up of intestinal metaplasia of the gastric cardia. Am J
matched historical controls. Eur J Gastroenterol Hepatol. 2007;19(4): Gastroenterol. 2000;95(7):1677-1680.
311-315. 54. Wani S, Puli SR, Shaheen NJ, et al. Esophageal adenocarcinoma in
33. Pouw RE, Seewald S, Gondrie JJ, et al. Stepwise radical endoscopic Barrett’s esophagus after endoscopic ablative therapy: a meta-analysis
resection for eradication of Barrett’s oesophagus with early neoplasia and systematic review. Am J Gastroenterol. 2009;104(2):502-513.
in a cohort of 169 patients. Gut. 2010;59(9):1169-1177. 55. Haidry R, Lovat L, Sharma P. Radiofrequency ablation for Bar-
34. Lopes CV, Hela M, Pesenti C, et al. Circumferential endoscopic rett’s dysplasia: past, present and the future? Curr Gastroenterol Rep.
resection of Barrett’s esophagus with high-grade dysplasia or early 2015;17(3):13.
adenocarcinoma. Surg Endosc. 2007;21(5):820-824. 56. Dunn JM, Mackenzie GD, Banks MR, et al. A randomised controlled
35. Larghi A, Lightdale CJ, Ross AS, et al. Long-term follow-up of com- trial of ALA vs. Photofrin photodynamic therapy for high-grade dyspla-
plete Barrett’s eradication endoscopic mucosal resection (CBE-EMR) sia arising in Barrett’s oesophagus. Lasers Med Sci. 2013;28(3):707-715.
for the treatment of high grade dysplasia and intramucosal carcinoma. 57. Kelty CJ, Marcus SL, Ackroyd R. Photodynamic therapy for Barrett’s
Endoscopy. 2007;39(12):1086-1091. esophagus: a review. Dis Esophagus. 2002;15(2):137-144.
36. Peters FP, Kara MA, Rosmolen WD, et al. Stepwise radical endo- 58. Zoepf T, Alsenbesy M, Jakobs R, Apel D, Rosenbaum A, Riemann
scopic resection is effective for complete removal of Barrett’s JF. Photodynamic therapy (PDT) versus argon plasma coagulation
esophagus with early neoplasia: a prospective study. Am J Gastroenterol. (APC) for ablative therapy of Barrett’s esophagus. Gastrointest Endosc.
2006;101(7):1449-1457. 2003;57(5):Ab139.
37. van Vilsteren FGI, Pouw RE, Seewald S, et al. Stepwise radical 59. Ackroyd R, Brown NJ, Davis MF, et al. Photodynamic therapy
endoscopic resection versus radiofrequency ablation for Barrett’s for dysplastic Barrett’s oesophagus: a prospective, double blind,
oesophagus with high-grade dysplasia or early cancer: a multicentre randomised, placebo controlled trial. Gut. 2000;47(5):612-617.
randomised trial. Gut. 2011;60(6):765-773. 60. Overholt BF, Wang KK, Burdick JS, et al. Five-year efficacy and safety
38. Chadwick G, Groene O, Markar SR, Hoare J, Cromwell D, Hanna GB. of photodynamic therapy with photofrin in Barrett’s high-grade
Systematic review comparing radiofrequency ablation and complete dysplasia. Gastrointest Endosc. 2007;66(3):460-468.
endoscopic resection in treating dysplastic Barrett’s esophagus: 61. Gray J, Fullarton GM. Long term efficacy of photodynamic therapy
a critical assessment of histologic outcomes and adverse events.
(PDT) as an ablative therapy of high grade dysplasia in Barrett’s
Gastrointest Endosc. 2014;79(5):718-731.
39. Cao Y, Liao C, Tan A, Gao Y, Mo Z, Gao F. Meta-analysis of endoscopic oesophagus. Photodiagn Photodyn. 2013;10(4):561-565.
submucosal dissection versus endoscopic mucosal resection for 62. Fayter D, Corbett M, Heirs M, Fox D, Eastwood A. A systematic
tumors of the gastrointestinal tract. Endoscopy. 2009;41(9):751-757. review of photodynamic therapy in the treatment of pre-cancerous
40. Chevaux JB, Piessevaux H, Jouret-Mourin A, Yeung R, Danse E, skin conditions, Barrett’s oesophagus and cancers of the biliary tract,
Deprez PH. Clinical outcome in patients treated with endoscopic brain, head and neck, lung, oesophagus and skin. Health Technol
submucosal dissection for superficial Barrett’s neoplasia. Endoscopy. Asses. 2010;14(37):1-7.
2015;47(2):103-112. 63. Kelty CJ, Ackroyd R, Brown NJ, Stephenson TJ, Stoddard CJ, Reed
41. Neuhaus H, Terheggen G, Rutz EM, Vieth M, Schumacher B. MWR. Endoscopic ablation of Barrett’s oesophagus: a randomized-
Endoscopic submucosal dissection plus radiofrequency ablation of controlled trial of photodynamic therapy vs. argon plasma coagula-
neoplastic Barrett’s esophagus. Endoscopy. 2012;44(12):1105-1113. tion. Aliment Pharm Therap. 2004;20(11-12):1289-1296.
42. Subramanian CR, Triadafilopoulos G. Endoscopic treatments for 64. Attwood SE, Lewis CJ, Caplin S, Hemming K, Armstrong G. Argon
dysplastic Barrett’s esophagus: resection, ablation, what else? World beam plasma coagulation as therapy for high-grade dysplasia in
J Surg. 2015;39(3):597-605. Barrett’s esophagus. Clin Gastroenterol Hepatol. 2003;1(4):258-263.
43. Shaheen NJ, Overholt BF, Sampliner RE, et al. Durability of radiofre- 65. Dumot JA, Greenwald BD. Argon plasma coagulation, bipolar
quency ablation in Barrett’s esophagus with dysplasia. Gastroenterology. cautery, and cryotherapy: ABC’s of ablative techniques. Endoscopy.
2011;141(2):460-468. 2008;40(12):1026-1032.
44. Bulsiewicz WJ, Kim HP, Dellon ES, et al. Safety and efficacy of 66. Ragunath K, Krasner N, Raman VS, Haqqani MT, Phillips CJ, Cheung
endoscopic mucosal therapy with radiofrequency ablation for I. Endoscopic ablation of dysplastic Barrett’s oesophagus comparing
Ablation for Patients With Barrett or Dysplasia CHAPTER 34 361
argon plasma coagulation and photodynamic therapy: a randomized vs. endoscopic surveillance of patients with Barrett’s esophagus after
prospective trial assessing efficacy and cost-effectiveness. Scand J antireflux surgery. Gastrointest Endosc. 2004;59(1):1-7.
Gastroenterol. 2005;40(7):750-758. 81. Gray NA, Odze RD, Spechler SJ. Buried metaplasia after endoscopic
67. Pereira-Lima JC, Busnello JV, Saul C, et al. High power setting argon ablation of Barrett’s esophagus: a systematic review. Am J Gastroenterol.
plasma coagulation for the eradication of Barrett’s esophagus. Am 2011;106(11):1899-1908.
J Gastroenterol. 2000;95(7):1661-1668. 82. Bronner MP, Overholt BF, Taylor SL, et al. Squamous overgrowth is
68. Madisch A, Miehlke S, Bayerdoerffer E, et al. Long-term follow-up not a safety concern for photodynamic therapy for Barrett’s esophagus
after complete ablation of Barrett’s esophagus with argon plasma with high-grade dysplasia. Gastroenterology. 2009;136(1):56-64.
coagulation. World J Gastroenterol. 2005;11(8):1182-1186. 83. Zhou C, Tsai TH, Lee HC, et al. Characterization of buried glands
69. Milashka M, Calomme A, Van Laethem JL, et al. Sixteen-year follow-up before and after radiofrequency ablation by using 3-dimensional optical
of Barrett’s esophagus, endoscopically treated with argon plasma coherence tomography (with videos). Gastrointest Endosc. 2012;76(1):
coagulation. United Eur Gastroenterol. 2014;2(5):367-373. 32-40.
70. Sie C, Bright T, Schoeman M, et al. Argon plasma coagulation ablation 84. Singh S, Garg SK, Singh PP, Iyer PG, El-Serag HB. Acid-suppressive
versus endoscopic surveillance of Barrett’s esophagus: late outcomes medications and risk of oesophageal adenocarcinoma in patients
from two randomized trials. Endoscopy. 2013;45(11):859-865. with Barrett’s oesophagus: a systematic review and meta-analysis.
71. Hage M, Siersema PD, van Dekken H, et al. 5-Aminolevulinic acid Gut. 2014;63(8):1229-1237.
photodynamic therapy versus argon plasma coagulation for abla- 85. Kahaleh M, Van Laethem JL, Nagy N, Cremer M, Deviere J. Long-term
tion of Barrett’s oesophagus: a randomised trial. Gut. 2004;53(6): follow-up and factors predictive of recurrence in Barrett’s esophagus
785-790. treated by argon plasma coagulation and acid suppression. Endoscopy.
72. Shaheen NJ, Greenwald BD, Peery AF, et al. Safety and efficacy of 2002;34(12):950-955.
endoscopic spray cryotherapy for Barrett’s esophagus with high-grade 86. Ferraris R, Fracchia M, Foti M, et al. Barrett’s oesophagus: long-term
dysplasia. Gastrointest Endosc. 2010;71(4):680-685. follow-up after complete ablation with argon plasma coagulation
73. Gosain S, Mercer K, Twaddell WS, Uradomo L, Greenwald BD. and the factors that determine its recurrence. Aliment Pharm Therap.
Liquid nitrogen spray cryotherapy in Barrett’s esophagus with 2007;25(7):835-840.
high-grade dysplasia: long-term results. Gastrointest Endosc. 2013;78(2): 87. Krishnan K, Pandolfino JE, Kahrilas PJ, Keefer L, Boris L, Komanduri
260-265. S. Increased risk for persistent intestinal metaplasia in patients
74. Pasricha PJ, Hill S, Wadwa KS, et al. Endoscopic cryotherapy: with Barrett’s esophagus and uncontrolled reflux exposure before
experimental results and first clinical use. Gastrointest Endosc. radiofrequency ablation. Gastroenterology. 2012;143(3):576-581.
1999;49(5):627-631. 88. Shaheen NJ, Kim HP, Bulsiewicz WJ, et al. Prior fundoplication does
75. Greenwald BD, Dumot JA, Horwhat JD, Lightdale CJ, Abrams not improve safety or efficacy outcomes of radiofrequency ablation:
JA. Safety, tolerability, and efficacy of endoscopic low-pressure results from the US RFA Registry. J Gastrointest Surg. 2013;17(1):21-28.
liquid nitrogen spray cryotherapy in the esophagus. Dis Esophagus. 89. Guarner-Argente C, Buoncristiano T, Furth EE, Falk GW, Ginsberg
2010;23(1):13-19. GG. Long-term outcomes of patients with Barrett’s esophagus and
76. Anders M, Lucks Y, El-Masry MA, et al. Subsquamous extension high-grade dysplasia or early cancer treated with endoluminal
of intestinal metaplasia is detected in 98% of cases of neoplastic therapies with intention to complete eradication. Gastrointest Endosc.
Barrett’s esophagus. Clin Gastroenterol Hepatol. 2014;12(3):405-410. 2013;77(2):190-199.
77. Chennat J, Ross AS, Konda VJA, et al. Advanced pathology under 90. Zeki S, Fitzgerald RC. The use of molecular markers in predicting
squamous epithelium on initial EMR specimens in patients with dysplasia and guiding treatment. Best Pract Res Clin Gastroenterol. 2015;
Barrett’s esophagus and high-grade dysplasia or intramucosal carci- 29(1):113-124.
noma: implications for surveillance and endotherapy management. 91. Galipeau PC, Li XH, Blount PL, et al. NSAIDs modulate CDKN2A,
Gastrointest Endosc. 2009;70(3):417-421. TP53, and DNA content risk for progression to esophageal adeno-
78. Prasad GA, Wang KK, Halling KC, et al. Utility of biomarkers in carcinoma. PLoS Med. 2007;4(2):342-354.
prediction of response to ablative therapy in Barrett’s esophagus. 92. Das A, Singh V, Fleischer DE, Sharma VK. A comparison of endoscopic
Gastroenterology. 2008;135(2):370-379. treatment and surgery in early esophageal cancer: an analysis of
79. Hornick JL, Mino-Kenudson M, Lauwers GY, Liu WT, Goyal R, Odze Surveillance, Epidemiology and End Results data. Am J Gastroenterol.
RD. Buried Barrett’s epithelium following photodynamic therapy 2008;103(6):1340-1345.
shows reduced crypt proliferation and absence of DNA content 93. Corley DA, Mehtani K, Quesenberry C, Zhao W, de Boer J, Weiss
abnormalities. Am J Gastroenterol. 2008;103(1):38-47. NS. Impact of endoscopic surveillance on mortality from Barrett’s
80. Ackroyd R, Tam W, Schoeman M, Devitt PG, Watson DI. Prospective esophagus-associated esophageal adenocarcinomas. Gastroenterology.
randomized controlled trial of argon plasma coagulation ablation 2013;145(2):312-319, e311.
PART SEVEN
Esophageal Cancer
CHAPTER
E
sophageal cancer accounts for 1% of new cancer Worldwide, the incidence of esophageal cancer varies
diagnoses in the United States annually and 2.6% by more than 21-fold, with SCC being the predominant
of cancer-related deaths.1 An estimated 0.5% of subtype.6 The highest-risk area, referred to as the esophageal
men and women will be diagnosed with esophageal cancer belt, extends from the Middle East to northeast China,
cancer at some point during their lifetime. The two most where the incidence of SCC is more than 100 cases per
common subtypes of primary esophageal cancer include 100,000 people annually.5,6,11,12 High rates of esophageal
adenocarcinoma (EAC) and squamous cell carcinoma SCC are also seen in Southern and Eastern Africa, whereas
(SCC). These differ tremendously in their natural history, the lowest rates are found in Western Africa.6
epidemiologic pattern, and risk factors. SCC arises from
the native squamous epithelium of the esophagus. Chronic MORTALITY AND PROGNOSIS OF PATIENTS WITH
inflammation due to environmental exposures causes ESOPHAGEAL CANCER
progression to dysplasia and eventually malignant change.2 Esophageal cancer is the 11th leading cause of cancer-
EAC typically arises in areas of the esophagus where related death in the United States.1 It accounted for an
the squamous epithelium is replaced by columnar-lined estimated 15,690 deaths in 2016.4 Over the past decade, the
metaplastic epithelium (Barrett esophagus), usually due death rate due to esophageal cancer has been declining
to the presence of gastroesophageal reflux.3 an average of 0.8% per year (see Fig. 35.1). Across all
This chapter reviews the epidemiologic pattern, risk races, the death rate is approximately 5 times higher in
factors, and clinical manifestations of esophageal cancer men than women. The overall relative 5-year survival rate
and its histologic subtypes. has increased from 4% in the 1970s to 18.7% currently.1
Improved prognosis is seen in patients with localized
disease. With complete surgical resection, the relative 5-year
EPIDEMIOLOGY survival rate is approximately 90% for pTis tumors, 75%
for pT1, 45% for pT2, 30% for pT3, and 10% to 15% for
INCIDENCE pT4 disease.13 Unfortunately, more than 30% of patients
In 2016, an estimated 16,910 people will be diagnosed with have metastatic disease at the time of presentation, with
esophageal cancer in the United States alone.4 Worldwide, a relative 5-year survival of 4.5% (Table 35.1).1,4
over 450,000 people are diagnosed annually. Over the past
few decades, there has been a major shift in the incidence AGE, SEX, AND RACE DISTRIBUTION
of esophageal cancer worldwide with trends differing by Esophageal cancer is more common with increased age.
histologic subtype.5,6 The majority of new cases are diagnosed in people aged 65
In the United States, the overall incidence of esophageal to 74 years with a median age at diagnosis of 67.1 Overall,
cancer has been falling an average of 1.4% per year over it has a male preponderance (7 : 1) with a high of 11 : 1
the past decade (Fig. 35.1). The most recent Surveillance in those aged 50 to 54 and a low of 4 : 1 in those aged 75
Epidemiology, and End Results (SEER) data estimate 4.3 to 79. This predilection for males is seen irrespective of
new cases annually per 100,000 men and women. Across all race/ethnicity.14
races, EAC is the most common histologic subtype in the Across races, substantial differences are seen in the
United States and Europe. Since the 1970s, the incidence incidence of esophageal cancer and the distribution
of EAC has increased at a rate greater than any other of the histologic subtypes. The age-adjusted incidence
malignancy in the United States.7 The absolute incidence rate of esophageal cancer is highest in white and black
of EAC has increased from 0.4 case per 100,000 in 1975 to men (7.9 and 7.2 per 100,000 people, respectively) and
2.58 cases per 100,000 in 2009.8 During this same period, lowest in men of Asian/Pacific Islander descent (3.4 per
the incidence of SCC has steadily decreased.7,9,10 100,000). Over the past three decades, the incidence of
362
Epidemiology, Risk Factors, and Clinical Manifestations of Esophageal Cancer CHAPTER 35 362.e1
ABSTRACT
Worldwide, over 450,000 people are diagnosed with
esophageal cancer annually. In the United States alone,
esophageal cancer accounts for 1% of new cancer diagnoses
and 2.6% of cancer related deaths annually. The two
most common subtypes of primary esophageal cancer
include adenocarcinoma and squamous cell carcinoma.
These differ tremendously in their natural history, epi-
demiologic pattern, and risk factors. Over the past few
decades, there has been a major shift in the incidence
of esophageal cancer worldwide with trends differing by
histologic subtype. These differences may be accounted
for by changes in the prevalence of known risk factors,
such as gastroesophageal reflux and obesity. This chapter
reviews the epidemiologic pattern, risk factors, and clinical
manifestations of esophageal cancer and its histologic
subtypes.
KEYWORDS
Esophageal cancer; esophageal adenocarcinoma; esopha-
geal squamous cell carcinoma; epidemiology
Epidemiology, Risk Factors, and Clinical Manifestations of Esophageal Cancer CHAPTER 35 363
6 30
White males
5 25 Black males
Rate per 100,000 people
3
15
2
10
1 Incidence
Mortality 5
0
1975 1980 1985 1990 1995 2000 2005 2010 0
Year 1975 1980 1985 1990 1995 2000 2005 2010 2015
Year
FIGURE 35.1 Trend in incidence and mortality from esophageal
cancer (1975–2013, all ages, all races, both sexes). (Data from FIGURE 35.2 Trend in age-adjusted incidence rate of esophageal
Howlader N, Noone AA, Krapcho M, et al. SEER Cancer Statistics cancer by race and sex (1975–2013). (Data from Howlader N,
Review, 1975–2013. http://seer.cancer.gov/csr/1975_2013/.) Noone AA, Krapcho M, et al. SEER Cancer Statistics Review,
1975–2013. http://seer.cancer.gov/csr/1975_2013/.)
TABLE 35.1 Stage Distribution at Diagnosis and tumors are defined as those located between the distal
Respective 5-Year Relative Survival
5 cm of the esophagus and the proximal 5 cm of the gastric
Stage 5-year Relative cardia. Siewert and Stein subclassified these tumors into
Stage at Diagnosis Distribution (%) Survival (%) three types based on their location with relation to the
Localized (confined to 20 41.3 Z line: type I (esophageal), type II (cardiac), and type
primary site) III (subcardiac).17 The 7th edition of the American Joint
Regional (spread to 31 22.8 Commission on Cancer (AJCC) Cancer Staging Manual
regional lymph nodes) grouped EGJ tumors with esophageal cancer for purposes
Distant (cancer has 38 4.5 of staging and treatment.18
metastasized)
Unknown (unstaged) 11 12.4
RISK FACTORS FOR ESOPHAGEAL CANCER
Data based on SEER 18, 2006–2012, both sexes, all races.
Data from Howlader N, Noone AA, Krapcho M, et al. SEER Cancer Sta- There are several risk factors that are associated with the
tistics Review, 1975–2013. <http://seer.cancer.gov/csr/1975_2013/>. development of EAC and SCC. The risk factors and their
effects are summarized in Table 35.2.
ACHALASIA
ANATOMIC DISTRIBUTION OF Achalasia has been associated with an increased risk of
ESOPHAGEAL CANCER esophageal SCC, presumably due to chronic mucosal
irritation caused by nitrosamines released from bacteria in
Cancer of the cervical esophagus is rare. SCC is evenly dis- stagnant food debris.25 The incidence rate of malignancy in
tributed within the middle and lower thoracic esophagus, these patients varies significantly throughout the literature,
whereas 75% of all EAC is located in the distal esophagus.16 with a 10- to 50-fold increase in relative risk compared
Tumors of the esophagogastric junction (EGJ) represent with the general population.26–28 Leeuwenburgh et al.
an entity that has historically been difficult to classify. EGJ found an incidence of 0.34% per year of follow-up with
364 SECTION I Esophagus and Hernia
without dysplasia, the annual risk of esophageal cancer hypotheses.7 They noted a shift in the anatomic distribution
is 0.25%, compared with 6% for those with high-grade of esophageal cancer with an increased incidence only
dysplasia.57 Risk factors for progression to cancer include in the distal third of the esophagus where EAC is usually
chronic GERD, presence of hiatal hernia, advanced age, found. This would argue that reclassification of SCC does
male sex, white race, tobacco use, and obesity. The risk not account for the increased incidence of EAC. Similarly,
of cancer also increases as the segment length of Barrett it cannot be explained by reclassification of tumors of the
increases, with long-segment Barrett (≥3 cm) having a gastric cardia because the incidence of tumors below the
transition rate of 0.22% per year compared with 0.01% gastroesophageal junction has also increased over the past
in ultra-short segment Barrett (<1 cm).58 three decades. Finally, they noted that over the past 30
years, the mortality from EAC has increased over sevenfold
PROTON PUMP INHIBITORS with only a minor change in the proportion of patients
The increased incidence of EAC has correlated with the diagnosed with localized disease, excluding improved
increased use of proton pump inhibitors (PPIs) since diagnosis as a cause for the rise in incidence. Because of
their discovery in the 1980s. Studies have shown that up these data, they concluded that the increased incidence
to half of patients with EAC report PPI use.54 While PPIs of EAC represented a true increase in disease burden.
decrease the acidity of the refluxate into the esophagus, The increase in esophageal cancer may be explained
they do not eliminate reflux. In fact, studies have shown by changes in the prevalence of its known risk factors.
that reflux of bile salts also contributes to the development The most notable change has been in the increased
of Barrett esophagus.59,60 prevalence of obesity, especially in the United States.
As reviewed earlier in this chapter, several studies have
LOWER-SPHINCTER RELAXING MEDICATIONS identified obesity and central adiposity as risk factors
Medications that lower the resting pressure of the lower for EAC, possibly due to hormonal effects or due to the
esophageal sphincter (LES) are thought to increase the risk increased prevalence of GERD among obese individuals.
of esophageal cancer by promoting increased gastroesopha- Over the past three decades, the increased incidence of
geal reflux. A case-control study by Lagergren et al. found EAC has been greatest in the Netherlands, United States,
a positive association between past use of sphincter-relaxing and Spain, in decreasing order. However, Kroep et al.
medications (nitroglycerin, anticholinergics, β-adrenergic identified that the prevalence of obesity has not followed
agonists, aminophyllines, and benzodiazepines) and the the same trend, signifying that there must be other key
risk of EAC.61 They also demonstrated a higher prevalence drivers in the rise of EAC incidence.68 Parallels have also
of reflux symptoms among users of LES-relaxing agents. been noted with the rising prevalence of GERD, PPI use,
After adjusting for reflux symptoms, the positive association and H. pylori eradication, particularly in the Western world.
almost disappeared suggesting that promotion of reflux
is the link between the use of sphincter-relaxing drugs
and EAC. Ranka et al. demonstrated a similar positive
CLINICAL MANIFESTATIONS
association with cancer risk and the use of bronchodilators, The symptoms of esophageal cancer vary by stage. Early-
theophylline, and calcium channel blockers.62 stage tumors are typically asymptomatic, which is why
over 50% of patients present with regionally advanced
HELICOBACTER PYLORI or metastatic disease.4 The most common presenting
Helicobacter pylori infection is linked to several gastrointestinal symptoms include dysphagia (74%), weight loss (57%),
malignancies including gastric cancer.63 Conversely, studies and odynophagia (17%).69,70 Other symptoms such as
have shown that colonization of the stomach with H. pylori, cough, dyspnea, hoarseness, and pain (abdominal, back,
particularly cytotoxin-associated gene A (CagA)-positive retrosternal) may indicate more extensive disease.71 In
strains, may protect against the development of EAC.64 A patients with EAC, up to two-thirds of patients have a
recent meta-analysis also demonstrated a negative associa- history of reflux symptoms.54
tion with SCC in East Asian populations but no effect in For early-stage disease, the physical examination is
Western populations.65 It is hypothesized that H. pylori usually unremarkable. Patients with metastatic disease may
induces gastric atrophy and decreased acid production, have hepatomegaly, pleural effusion, or lymphadenopathy,
hence decreasing acid reflux into the distal esophagus.66 particularly in the left supraclavicular fossa (Virchow
Another suggested mechanism is through decreased gastric node).71
secretion of the hormone ghrelin, which subsequently Routine endoscopic screening of the general popula-
decreases appetite, leading to lower rates of obesity, a tion is generally not recommended, as the incidence
known risk factor for EAC.67 of esophageal cancer is low. The American College of
Gastroenterology 2016 guidelines recommend endoscopic
screening for Barrett esophagus in men with chronic or
WHY IS THE EPIDEMIOLOGY CHANGING? frequent symptomatic GERD with two or more risk factors
Several theories have been proposed to explain the chang- for EAC (i.e., age >50, Caucasian race, central obesity,
ing epidemiology of esophageal cancer worldwide. The rise current or past tobacco use, family history of Barrett or
in EAC has been the most dramatic change. Some have esophageal cancer in a first-degree relative).72 In women,
raised the question of whether the rise in the incidence of they recommend considering screening on a case-by-case
EAC is actually a result of reclassification of related cancers basis as determined by the presence of multiple risk
or improved diagnosis due to better-quality endoscopy. factors listed above. In patients with Barrett esophagus,
Pohl and Welch used the SEER database to assess these surveillance is recommended every 3 to 5 years in the
366 SECTION I Esophagus and Hernia
absence of dysplasia, with more frequent intervals and/ 27. Sandler RS, Nyren O, Ekbom A, et al. The risk of esophageal cancer
or therapy in the setting of dysplasia. in patients with achalasia. A population-based study. JAMA. 1995;274:
1359-1362.
28. Dunaway PM, Wong RK. Risk and surveillance intervals for squa-
REFERENCES mous cell carcinoma in achalasia. Gastrointest Endosc Clin N Am.
2001;11:425-434, ix.
1. Howlader N, Noone AA, Krapcho M, et al. SEER Cancer Statistics 29. Leeuwenburgh I, Scholten P, Alderliesten J, et al. Long-term esopha-
Review, 1975–2013. http://seer.cancer.gov/csr/1975_2013/. geal cancer risk in patients with primary achalasia: a prospective
2. Dawsey SM, Lewin KJ, Wang GQ, et al. Squamous esophageal study. Am J Gastroenterol. 2010;105:2144-2149.
histology and subsequent risk of squamous cell carcinoma of the 30. Brucher BL, Stein HJ, Bartels H, et al. Achalasia and esophageal
esophagus. A prospective follow-up study from Linxian, China. cancer: incidence, prevalence, and prognosis. World J Surg. 2001;25:
Cancer. 1994;74:1686-1692. 745-749.
3. Conteduca V, Sansonno D, Ingravallo G, et al. Barrett’s esophagus 31. Riboli E, Norat T. Epidemiologic evidence of the protective effect of
and esophageal cancer: an overview. Int J Oncol. 2012;41:414-424. fruit and vegetables on cancer risk. Am J Clin Nutr. 2003;78:559S-569S.
4. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J 32. Freedman ND, Park Y, Subar AF, et al. Fruit and vegetable intake
Clin. 2016;66:7-30. and esophageal cancer in a large prospective cohort study. Int J
5. Pennathur A, Gibson MK, Jobe BA, et al. Oesophageal carcinoma. Cancer. 2007;121:2753-2760.
Lancet. 2013;381:400-412. 33. Yamaji T, Inoue M, Sasazuki S, et al. Fruit and vegetable consumption
6. Torre LA, Bray F, Siegel RL, et al. Global cancer statistics, 2012. CA and squamous cell carcinoma of the esophagus in Japan: the JPHC
Cancer J Clin. 2015;65:87-108. study. Int J Cancer. 2008;123:1935-1940.
7. Pohl H, Welch HG. The role of overdiagnosis and reclassification 34. Jemal A, Center MM, DeSantis C, et al. Global patterns of cancer
in the marked increase of esophageal adenocarcinoma incidence. incidence and mortality rates and trends. Cancer Epidemiol Biomarkers
J Natl Cancer Inst. 2005;97:142-146. Prev. 2010;19:1893-1907.
8. Hur C, Miller M, Kong CY, et al. Trends in esophageal adenocarci- 35. Tran GD, Sun XD, Abnet CC, et al. Prospective study of risk factors
noma incidence and mortality. Cancer. 2013;119:1149-1158. for esophageal and gastric cancers in the Linxian general population
9. Trivers KF, Sabatino SA, Stewart SL. Trends in esophageal cancer inci- trial cohort in China. Int J Cancer. 2005;113:456-463.
dence by histology, United States, 1998–2003. Int J Cancer. 2008;123: 36. Islami F, Kamangar F, Nasrollahzadeh D, et al. Oesophageal cancer
1422-1428. in Golestan Province, a high-incidence area in northern Iran—a
10. Giri S, Pathak R, Aryal MR, et al. Incidence trend of esophageal squa- review. Eur J Cancer. 2009;45:3156-3165.
mous cell carcinoma: an analysis of Surveillance Epidemiology, and 37. Islami F, Boffetta P, Ren JS, et al. High-temperature beverages and
End Results (SEER) database. Cancer Causes Control. 2015;26:159-161. foods and esophageal cancer risk—a systematic review. Int J Cancer.
11. Torre LA, Siegel RL, Ward EM, et al. Global cancer incidence and 2009;125:491-524.
mortality rates and trends---an update. Cancer Epidemiol Biomarkers 38. Lukanich JM. Section 1. Epidemiological review. Semin Thorac
Prev. 2016;25:16-27. Cardiovasc Surg. 2003;15:158-166.
12. Kamangar F, Dores GM, Anderson WF. Patterns of cancer incidence, 39. Dong LM, Kristal AR, Peters U, et al. Dietary supplement use and
mortality, and prevalence across five continents: defining priorities risk of neoplastic progression in esophageal adenocarcinoma: a
to reduce cancer disparities in different geographic regions of the prospective study. Nutr Cancer. 2008;60:39-48.
world. J Clin Oncol. 2006;24:2137-2150. 40. Thompson CL, Khiani V, Chak A, et al. Carbohydrate consumption
13. Rice TW, Rusch VW, Apperson-Hansen C, et al. Worldwide esophageal and esophageal cancer:an ecological assessment. Am J Gastroenterol.
cancer collaboration. Dis Esophagus. 2009;22:1-8. 2008;103:555-561.
14. Mathieu LN, Kanarek NF, Tsai HL, et al. Age and sex differences 41. Zimmermann KC, Sarbia M, Weber AA, et al. Cyclooxygenase-2 expres-
in the incidence of esophageal adenocarcinoma: results from the sion in human esophageal carcinoma. Cancer Res. 1999;59:198-204.
Surveillance, Epidemiology, and End Results (SEER) Registry 42. Trifan OC, Hla T. Cyclooxygenase-2 modulates cellular growth and
(1973–2008). Dis Esophagus. 2014;27:757-763. promotes tumorigenesis. J Cell Mol Med. 2003;7:207-222.
15. Cook MB, Chow WH, Devesa SS. Oesophageal cancer incidence in 43. Liao LM, Vaughan TL, Corley DA, et al. Nonsteroidal anti-
the United States by race, sex, and histologic type, 1977–2005. Br J inflammatory drug use reduces risk of adenocarcinomas of the
Cancer. 2009;101:855-859. esophagus and esophagogastric junction in a pooled analysis.
16. Zhang Y. Epidemiology of esophageal cancer. World J Gastroenterol. 2013; Gastroenterology. 2012;142:442-452. e445, quiz e422–443.
19:5598-5606. 44. Vaughan TL, Dong LM, Blount PL, et al. Non-steroidal anti-
17. Siewert JR, Stein HJ. Classification of adenocarcinoma of the inflammatory drugs and risk of neoplastic progression in Barrett’s
oesophagogastric junction. Br J Surg. 1998;85:1457-1459. oesophagus: a prospective study. Lancet Oncol. 2005;6:945-952.
18. Rice TW, Blackstone EH, Rusch VW. 7th edition of the AJCC cancer 45. Anderson LA, Johnston BT, Watson RG, et al. Nonsteroidal anti-
staging manual: esophagus and esophagogastric junction. Ann Surg inflammatory drugs and the esophageal inflammation-metaplasia-
Oncol. 2010;17:1721-1724. adenocarcinoma sequence. Cancer Res. 2006;66:4975-4982.
19. Prabhu A, Obi KO, Rubenstein JH. The synergistic effects of alcohol 46. Calle EE, Rodriguez C, Walker-Thurmond K, et al. Overweight,
and tobacco consumption on the risk of esophageal squamous cell obesity, and mortality from cancer in a prospectively studied cohort
carcinoma: a meta-analysis. Am J Gastroenterol. 2014;109:822-827. of U.S. adults. N Engl J Med. 2003;348:1625-1638.
20. Pandeya N, Williams GM, Sadhegi S, et al. Associations of duration, 47. Corley DA, Kubo A, Zhao W. Abdominal obesity and the risk of
intensity, and quantity of smoking with adenocarcinoma and squamous esophageal and gastric cardia carcinomas. Cancer Epidemiol Biomarkers
cell carcinoma of the esophagus. Am J Epidemiol. 2008;168:105-114. Prev. 2008;17:352-358.
21. Kamangar F, Chow WH, Abnet CC, et al. Environmental causes of 48. Steffen A, Huerta JM, Weiderpass E, et al. General and abdominal
esophageal cancer. Gastroenterol Clin North Am. 2009;38:27-57, vii. obesity and risk of esophageal and gastric adenocarcinoma in the
22. Corrao G, Bagnardi V, Zambon A, et al. A meta-analysis of alcohol European Prospective Investigation into Cancer and Nutrition. Int
consumption and the risk of 15 diseases. Prev Med. 2004;38:613- J Cancer. 2015;137:646-657.
619. 49. Thrift AP, Shaheen NJ, Gammon MD, et al. Obesity and risk of
23. Bagnardi V, Blangiardo M, La Vecchia C, et al. A meta-analysis of esophageal adenocarcinoma and Barrett’s esophagus: a Mendelian
alcohol drinking and cancer risk. Br J Cancer. 2001;85:1700-1705. randomization study. J Natl Cancer Inst. 2014;106.
24. Freedman ND, Abnet CC, Leitzmann MF, et al. A prospective study 50. Turati F, Tramacere I, La Vecchia C, et al. A meta-analysis of body
of tobacco, alcohol, and the risk of esophageal and gastric cancer mass index and esophageal and gastric cardia adenocarcinoma. Ann
subtypes. Am J Epidemiol. 2007;165:1424-1433. Oncol. 2013;24:609-617.
25. Eckardt AJ, Eckardt VF. Current clinical approach to achalasia. World 51. Long E, Beales IL. The role of obesity in oesophageal cancer
J Gastroenterol. 2009;15:3969-3975. development. Therap Adv Gastroenterol. 2014;7:247-268.
26. Zendehdel K, Nyren O, Edberg A, et al. Risk of esophageal adeno- 52. El-Serag HB, Sweet S, Winchester CC, et al. Update on the epidemiol-
carcinoma in achalasia patients, a retrospective cohort study in ogy of gastro-oesophageal reflux disease: a systematic review. Gut.
Sweden. Am J Gastroenterol. 2011;106:57-61. 2014;63:871-880.
Epidemiology, Risk Factors, and Clinical Manifestations of Esophageal Cancer CHAPTER 35 367
53. Lagergren J, Bergstrom R, Lindgren A, et al. Symptomatic gastro- 63. Parsonnet J, Friedman GD, Vandersteen DP, et al. Helicobacter pylori
esophageal reflux as a risk factor for esophageal adenocarcinoma. infection and the risk of gastric carcinoma. N Engl J Med. 1991;325:
N Engl J Med. 1999;340:825-831. 1127-1131.
54. Lada MJ, Nieman DR, Han M, et al. Gastroesophageal reflux disease, 64. Islami F, Kamangar F. Helicobacter pylori and esophageal cancer risk:
proton-pump inhibitor use and Barrett’s esophagus in esophageal a meta-analysis. Cancer Prev Res (Phila). 2008;1:329-338.
adenocarcinoma: trends revisited. Surgery. 2013;154:856-864, discus- 65. Xie FJ, Zhang YP, Zheng QQ, et al. Helicobacter pylori infection and
sion 864–856. esophageal cancer risk: an updated meta-analysis. World J Gastroenterol.
55. Shaheen N, Ransohoff DF. Gastroesophageal reflux, Barrett esophagus, 2013;19:6098-6107.
and esophageal cancer: scientific review. JAMA. 2002;287:1972-1981. 66. Chow WH, Blaser MJ, Blot WJ, et al. An inverse relation between
56. Kahrilas PJ, Shaheen NJ, Vaezi MF, et al. American Gastroenterological cagA+ strains of Helicobacter pylori infection and risk of esophageal
Association medical position statement on the management of and gastric cardia adenocarcinoma. Cancer Res. 1998;58:588-590.
gastroesophageal reflux disease. Gastroenterology. 2008;135:1383-1391. 67. Whiteman DC, Sadeghi S, Pandeya N, et al. Combined effects of
1391 e1381–1385. obesity, acid reflux and smoking on the risk of adenocarcinomas
57. Spechler SJ. Barrett esophagus and risk of esophageal cancer: a of the oesophagus. Gut. 2008;57:173-180.
clinical review. JAMA. 2013;310:627-636. 68. Kroep S, Lansdorp-Vogelaar I, Rubenstein JH, et al. Comparing trends
58. Pohl H, Pech O, Arash H, et al. Length of Barrett’s oesophagus and in esophageal adenocarcinoma incidence and lifestyle factors between
cancer risk: implications from a large sample of patients with early the United States, Spain, and the Netherlands. Am J Gastroenterol.
oesophageal adenocarcinoma. Gut. 2016;65:196-201. 2014;109:336-343, quiz 335, 344.
59. Nasr AO, Dillon MF, Conlon S, et al. Acid suppression increases 69. Daly JM, Fry WA, Little AG, et al. Esophageal cancer: results of an
rates of Barrett’s esophagus and esophageal injury in the presence American College of Surgeons Patient Care Evaluation Study. J Am
of duodenal reflux. Surgery. 2012;151:382-390. Coll Surg. 2000;190:562-572, discussion 572–563.
60. Alsalahi O, Dobrian AD. Proton pump inhibitors: the culprit for 70. Fein R, Kelsen DP, Geller N, et al. Adenocarcinoma of the esophagus
Barrett’s esophagus? Front Oncol. 2014;4:373. and gastroesophageal junction. Prognostic factors and results of
61. Lagergren J, Bergstrom R, Adami HO, et al. Association between therapy. Cancer. 1985;56:2512-2518.
medications that relax the lower esophageal sphincter and risk for 71. Enzinger PC, Mayer RJ. Esophageal cancer. N Engl J Med. 2003;349:
esophageal adenocarcinoma. Ann Intern Med. 2000;133:165-175. 2241-2252.
62. Ranka S, Gee JM, Johnson IT, et al. Non-steroidal anti-inflammatory 72. Shaheen NJ, Falk GW, Iyer PG, et al. ACG clinical guideline: diagnosis
drugs, lower oesophageal sphincter-relaxing drugs and oesophageal and management of Barrett’s esophagus. Am J Gastroenterol. 2016;111:
cancer. A case-control study. Digestion. 2006;74:109-115. 30-50.
CHAPTER
ABSTRACT
The most common two malignant tumors involving the
esophagus and the esophagogastric junction are adeno-
carcinoma, which is more common in Western countries,
and squamous cell carcinoma, which is more commonly
seen in Asia. Diagnosis and identification of tumor cell
type is done with endoscopic biopsies. Accurate staging
of esophageal cancer is important as it impacts prognosis
and facilitates staged directed treatment. Staging is done
according to the TNM classification with the current
staging version being published by the AJCC (UICC)
7th Edition. Staging is done on the basis of depth of
penetration in the wall of the esophagus or stomach (T
stage), the number of lymph nodes involved (N stage) and
evidence for metastatic spread to other organs (M stage).
Anatomic location of the cancer in the esophagus, R status
(Resection margins) and G status (Tumor differentiation)
can also play an important role in assigning tumor stage
and deciding on treatment approach. Staging that is
assigned following investigation is typified as “clinical
stage”; staging following endoscopic or surgical resection
is classified as “pathologic stage.”
The most important investigations to provide direct
information regarding accurate clinical stage include upper
endoscopy, computed tomography scans, positron emission
tomography scans, and endoscopic ultrasounds. In selected
patients, endoscopic mucosal resections, endobronchial
ultrasound, MRI scans, and thoracoscopic and laparoscopic
procedures can also play a role. The results of tumor
staging are best reviewed in a multidisciplinary tumor board
where all the information can be assessed and appropriate
treatment can be assigned in a multidisciplinary fashion.
KEYWORDS
Esophageal cancer, staging, diagnosis
Esophageal Cancer Diagnosis and Staging CHAPTER 36 369
UES 15 cm
Cervical esophagus
Sternal notch 20 cm
Upper thoracic
Azygos vein 25 cm
Middle thoracic
Inferior pulmonary vein 30 cm
Lower thoracic
EGJ 40 cm
FIGURE 36.1 Description of the anatomic landmarks of the esophagus. EGJ, Esophagogastric junction; UES, upper esophageal sphincter.
radiotherapy than adenocarcinoma, but the long-term The nodal classification (N) is one of the most con-
outcome of therapy appears to be similar. Adenocarcinoma troversial aspects in TNM staging. Different N stages are
may be associated with a better long-term prognosis after defined as:
resection than SCC.14 Another histologic subtype that • N0: No positive node
has been reported is the primary mixed adenosquamous • N1: 1 to 2 nodes
carcinoma. It is a rare kind of malignancy characterized • N2: 3 to 6 nodes
by mixed glandular and squamous differentiation as well • N3: 7 or more nodes
as a propensity for aggressive clinical behavior.15 There is no consensus on the ideal number of nodes
that must be resected for optimal staging. Data suggest
that the number of lymph nodes recovered—rather than
HISTOLOGIC GRADE their location—is an independent predictor of survival
The tumor grade reflects the histologic aggressiveness of the after esophagectomy. In the Surveillance, Epidemiology,
cancer. It can be an indicator of how quickly a tumor is likely and End Results (SEER) database, when 12 lymph nodes
to grow and spread. Cancers that are “well-differentiated” were examined, significant reduction in mortality was
tend to grow and spread at a slower rate than tumors that noted compared with no lymph node evaluation. Moreover,
are “undifferentiated” or “poorly differentiated.” Tumor patients who had 30 or more lymph nodes examined
grade is defined by numbers: 1, 2, 3, or 4 or by X depending had significantly lower mortality than any other groups.17
on the amount of abnormality, shown as16: This may be secondary to enhanced N staging, or to a
• GX: Grade cannot be assessed (undetermined grade) therapeutic effect of lymphadenectomy. In addition, the
• G1: Well differentiated (low grade) number of involved lymph nodes can be used to predict
• G2: Moderately differentiated (intermediate grade) the likelihood of systemic disease. 18 There is general
• G3: Poorly differentiated (high grade) agreement that a two-field lymph node dissection should
• G4: Undifferentiated (high grade) be done in an invasive cancer. For lymph node mapping,
a lymph node map that extends the nomenclature and
numbering system used for the staging of non–small cell
TUMOR, NODE, METASTASIS lung cancer can be used (Fig. 36.2).
CLASSIFICATION AND UPDATES IN Distant metastasis is simply designated as:
THE SEVENTH EDITION • M0: no distant metastases
• M1: distant metastases.
The TNM staging system for all solid tumors was devised
by Pierre Denoix between 1943 and 1952. Currently, it is UPDATES IN THE SEVENTH EDITION
maintained and developed by the American Joint Com- The seventh edition of TNM staging is the most updated
mittee on Cancer (AJCC) and the Union for International version published in 2010. This update involved the analysis
Cancer Control (UICC). The TNM staging system is of data on 4627 patients treated with esophagectomy without
updated every 6 to 8 years to include advances in our induction or adjuvant therapy.19 Changes as compared with
understanding of cancer prognostication. The TNM the AJCC sixth edition are reviewed in Table 36.1.
system classifies and groups cancers primarily by the Two major revisions were made for the T stage: Tis,
extent of local tumor invasion into the esophageal wall or high-grade dysplasia, now includes all noninvasive
and advanced invasion into adjacent structures (T), the neoplastic epithelium, which was previously termed
status of regional draining lymph nodes (N), and the carcinoma-in-situ. T1–T3 stages remained the same as in
presence or absence of distant metastases (M). The T the sixth edition. T4 lesions have been subcategorized into
stage is assessed by evaluation of the depth of invasion T4a, resectable cancers infiltrating the pleura, pericardium,
into the four distinct layers involving the esophageal wall or diaphragm; and T4b, unresectable cancers infiltrating
and adventitia according to the following nomenclature: structures, such as the aorta, vertebral body, or trachea-
• TX: Primary tumor cannot be assessed bronchi and carotid vessels.
• T0: No evidence of primary tumor In the seventh TNM edition, major modifications were
• Tis: High-grade dysplasia observed for the N stage. The sixth edition defined regional
• T1: Tumor invading mucosal lamina propria, muscularis nodes (N1) as those in the periesophageal, mediastinal,
mucosae, or submucosa and perigastric areas, but cervical and celiac nodes were
• T1a: Tumor invading into the lamina propria or regarded as “distant” metastases and designated M1a and
muscularis mucosae M1b. Stage M1b included visceral organ metastases.
• T1b: Tumor invading submucosa In the seventh edition, a regional node was redefined
• T2: Tumor invading muscularis propria to include any paraesophageal node extending from the
• T3: Tumor invading adventitia thoracic inlet to celiac axis. The subclassifications of M1a
• T4: Tumor invading adjacent structures and M1b have been eliminated, as has MX. In addition,
• T4a: Resectable tumor invading pleura, pericardium, the seventh edition accounted for the nodal burden by
or diaphragm classifying the number of involved lymph nodes into
• T4b: Unresectable tumor invading other adjacent categories: N1, 1 to 2; N2, 3 to 6; N3, 7 or more.
structures, such as aorta, vertebral body, or trachea. The seventh edition also noted the following changes:
T stage is important in the prognostication and is a different staging system for adenocarcinoma and SCC
crucial to determining suitability for surgical resection (Table 36.2); the precise definition of the three types of GEJ
and establishing a treatment plan. tumors based on location, including all three exclusively
Esophageal Cancer Diagnosis and Staging CHAPTER 36 371
FIGURE 36.2 Lymph node stations suggested by the American Joint Committee on Cancer manual. 1, Supraclavicular; 2R, right upper
paratracheal nodes; 2L, left upper paratracheal nodes; 3P, posterior mediastinal nodes; 4R, right lower paratracheal nodes; 4L, left lower
paratracheal nodes; 5, aortopulmonary nodes; 6, anterior mediastinal nodes; 7, subcarinal nodes; 8M, middle paraesophageal nodes; 8L,
lower paraesophageal nodes; 9, pulmonary ligament nodes; 10R, right tracheobronchial nodes; 10L, left tracheobronchial nodes; 15,
diaphragmatic nodes; 16, paracardial nodes; 17, left gastric nodes; 18, common hepatic nodes; 19, splenic nodes; 20, celiac nodes.
TABLE 36.2 American Joint Committee on Cancer Seventh Edition Stage Groupings
0
II
III
nodes secondary to the interference from the primary Porta hepatis lymph node
tumor. The sensitivity of PET is poor in the identification
of lymph node involvement ranging from 38% to 82%.
This is especially true for nodes in the middle and lower
mediastinum, where most primary tumors are found. In
one study, the sensitivity of PET for detecting cervical,
upper thoracic, and abdominal nodes was 78%, 82%, and
60%, respectively, but it was only 38% and 0%, respectively,
for the mid- and lower mediastinum.31 PET scan specificity
is much better than its sensitivity for the N staging with
a high pooled specificity ranging from 76% to 95% as
compared with 77% to 89% in CT scans.32,33
The main utility of PET scanning is in its ability to
identify the presence of distant metastasis as compared
with contrast CT alone. A study recently demonstrated
a change in the plan of management from curative into
palliative in 47% of cases seen to be negative on CT, but
were positive for distant metastasis on PET.34 In addition,
Luketich et al. reported 69% sensitivity, 93.4% specific-
FIGURE 36.6 Enlarged porta hepatis lymph node on cross-
ity, and 84% accuracy in detecting metastases with PET
sectional computed tomography scan.
compared with 46.1% sensitivity, 73.8% specificity, and
63% accuracy with CT.35 A meta-analysis of 12 publications
on the use of PET in esophageal cancer showed that sensitivity and specificity of 25% and 94%, respectively.38 A
the pooled sensitivity and specificity for the detection of CT performed with intravenous and oral contrast provides
locoregional metastases were 0.51 and 0.84, respectively. anatomic details that are critical before considering surgical
The PPV and negative predictive value (NPV) were 0.60 treatment. Obliteration of the fat plane between the
and 0.46, respectively. For distant metastases, the pooled esophagus and the aorta, trachea and bronchi, and the
sensitivity and specificity were 0.67 and 0.97, respectively. pericardium is suggestive of invasion, but the paucity of
The corresponding PPV and NPV were 0.92 and 0.83. When fat often makes this assessment unreliable. Thickening or
two studies that had particularly low sensitivities for the indentation of the normally flat membranous trachea and
detection of distant metastases were excluded (probably left main bronchus also is suggestive of invasion, but it
because they included more early tumors), the pooled should always be confirmed by bronchoscopic examination.
sensitivity improved to 0.72 and the specificity to 0.95.36 When the area of contact between the esophagus and the
This study highlights that the accuracy of PET in detecting aorta extends beyond 90 degrees of the circumference, an
locoregional nodes is only moderate. EUS fine-needle 80% accuracy of infiltration is reported. However, a CT
aspiration (FNA) is more accurate in this regard. However, scan cannot reliably distinguish the various T stages; T1
PET is clearly more accurate for identifying distant nodal and T2 lesions generally show an esophageal wall thickness
and visceral metastases. between 5 and 15 mm, and T3 lesions have a thickness
The diagnostic utility of PET in early-stage disease (Tis, greater than 15 mm, but this is far from accurate.
T1) may be low because the incidence of lymph node The sensitivity of detecting mediastinal and abdominal
metastases increases with the increasing T stage. Cost nodal involvement is suboptimal with CT because size
studies to support PET in early-stage disease are unlikely. alone is used as a diagnostic criterion. Intrathoracic
PET should routinely be performed in patients in whom and abdominal nodes greater than 1 cm are considered
standard staging methods (CT and EUS) demonstrate enlarged, and supraclavicular nodes with a short axis
regional invasive cancer with no distant metastatic disease greater than 0.5 cm and retrocrural nodes greater than
as it has been shown to be an independent predictor of 0.6 cm are pathologic.39 However, normal-sized lymph
overall survival in patients with nonmetastatic esophageal nodes may contain metastatic deposits, and enlargement
cancer. PET/CT improves the detection of metastatic of lymph nodes may be due to reactive and inflammatory
disease and, thus, can often result in a change of the processes. Sensitivity and specificity of detecting lymph
management strategy.37 Overall management strategies node involvement is 50% and 83%, respectively.32
were reported to be changed because of PET findings in The main value of CT lies in its ability to detect
3% to 20% of patients in a recent meta-analysis.36 distant systemic disease, such as hepatic, adrenal, and
lung metastases. With adenocarcinoma of the GEJ and
CT SCAN AND MAGNETIC RESONANCE IMAGING gastric cardia, peritoneal metastases are more likely than
CT is typically performed as the first radiologic test in with squamous cell cancer of the tubular esophagus. CT
the staging evaluation of an endoscopically diagnosed scanning is inferior to laparoscopy in detecting peritoneal
esophageal cancer. An esophageal wall thickness of greater metastases. Solitary lung metastases are rare in patients with
than 5 mm on a CT scan is generally regarded as abnormal esophageal carcinoma and, thus, when seen on CT, are
and warrants further investigation. A CT scan can display more likely to be primary lung cancer or benign nodules
the lesion and surrounding structures, regional organ and additional assessment including PET is required.
invasion, and lymph node metastasis (Fig. 36.6). Its role Magnetic resonance imaging (MRI) presents the advan-
is particularly important to exclude T4 disease with a tage of direct multiplanar imaging capabilities, which may
376 SECTION I Esophagus and Hernia
be of particular use in assessing tracheobronchial, aortic, stenotic esophagus, maneuvers such as dilatation of the
and pericardial invasion. Conventional MRI correctly lumen can be selectively considered, or different instru-
assessed the T stage in 60% of patients40 versus 81% ments, such as small-caliber ultrasound catheter or a
when using high-resolution T2-weighted MRI.41 A study wire-guided echoendoscope without fiberoptics, might
performed using high-resolution T2-weighted MRI with be used. However, there are currently no comparative
faster sequences and cardiorespiratory motion gating in studies to determine their accuracy. In view of these
combination with diffusion-weighted imaging demonstrated conflicting results, the only method that is currently
detection rates of 33% for T1, 58% for T2, 96% for T3, capable of accurately determining the depth of invasion
and 100% for T4 carcinomas.42 Experience with MRI has of a small visible lesion is EMR (see later).51
shown limitations similar to those of CT, especially with EUS improves the accuracy of N staging by identifying
respect to the low detection rate of mediastinal lymph suspicious nodal characteristics and by providing cytologic
nodes.43 confirmation of nodal metastatic disease aspirated by FNA
biopsy. Optimal criteria for identifying malignant lymph
ENDOSCOPIC ULTRASOUND AND ENDOSCOPIC nodes based upon EUS criteria and for helping to select
ULTRASONOGRAPHY/FINE-NEEDLE ASPIRATION patients for whom EUS/FNA is required continue to
Several modalities have been used to assess the depth of evolve. The modified EUS criteria (four standard criteria
invasion of cancer into the tubular esophagus. Dedicated plus EUS-identified celiac lymph nodes, >5 lymph nodes,
EUS operating at frequencies of 7.5 and 12 MHz are able or EUS T3/4 tumor) were more accurate than standard
to identify the wall as a five-layered structure (Fig. 36.7) criteria (hypoechoic, smooth border, round, or width
with relatively accurate sensitivities ranging from 81% to >5 to 10 mm) at identifying malignant lymph nodes.52
92%44 and provide information on the presence of abnor- Compared with the gold standard final pathologic evalua-
mal or enlarged lymph nodes. However, low-frequency tion of esophageal resected specimen along with dissected
endoscopes cannot visualize the muscularis mucosae. lymph nodes, sensitivity, specificity, and accuracy of EUS/
Therefore authorities consider the deeper the tumor, the FNA for locoregional lymph nodes are all over 85%53,54;
higher the sensitivity of EUS.45,46 There are conflicting data in particular, sensitivity is highest for cervical and upper
on the ability to accurately discriminate invasion of the thoracic paraesophageal, infracarinal, left paratracheal, and
mucosa and submucosa (i.e., the T1a versus T1b stages). recurrent laryngeal nodes. A prospective study comparing
A comparative prospective blinded trial concluded that the performance characteristics of CT, EUS, and EUS/
neither standard probes nor newer high-resolution 20-MHz FNA for preoperative lymph node staging of esophageal
probes are able to accurately distinguish intramucosal carcinoma in 125 patients demonstrated that EUS/FNA
from submucosal tumor invasion.47 In contrast, Thosani was more sensitive than CT (83% vs. 29%) and more
et al. demonstrated a sensitivity and a specificity for T1a accurate than CT (87% vs. 51%) or EUS (87% vs. 74%)
tumors of 85% and 87%, respectively, and 86% for both for nodal staging.55
sensitivity and specificity for T1b tumors.48 The accuracy EUS accuracy is operator-dependent and interobserver
of EUS in determining T4 stage was found to be 86%.49 reliability was found to be influenced by experience and
Overstaging may occur as peritumoral edematous tumor stage.56 Agreement among experienced endosonog-
changes may be mistaken for a tumor; on the other raphers for both T and N stage was good, except for T2
hand, understaging can result when tumor penetration tumors in which agreement was poor. There is a tendency
is below the resolution of sonography.50 Other technical to overstage T2 cancers by expert endosonographers in
obstacles might preclude complete EUS examination, 8% to 14% of cases due to peritumoral inflammation.57
such as stenotic esophageal tumors. To negotiate the However, FNA nodal aspiration should be considered
in all cases of celiac, porta hepatis, cervical, and upper
thoracic adenopathy. It should only be performed when
nodes are accessible and when the primary tumor is not
in the pathway of the aspiration needle.
ENDOSCOPIC RESECTION
Any malignant, visible lesion cannot be assumed to be
limited to the mucosa; in fact, very small cancers may
penetrate into the submucosa. EMR is a technique by
which the mucosa, and in most patients part or all of
the submucosa, are resected down to the muscularis
propria for definitive histologic diagnosis. It was first
described in 1992 by Dr. Inoue from Japan.58 Different
techniques have been described, but the more popular
method involves the use of a cap that fits over the end
of a standard endoscope. The aim is to excise the speci-
men in one piece; however, piecemeal excision remains
acceptable, but raises the potential for incomplete resec-
tion and makes pathologic evaluation of the resection
margins more complex. This technique has evolved and
FIGURE 36.7 Esophageal wall layers identified by endoscopic
revolutionized the staging and treatment of superficial
ultrasound.
Esophageal Cancer Diagnosis and Staging CHAPTER 36 377
esophageal adenocarcinoma—particularly the 20% of sensitivity and specificity of tracheal involvement were 92%
all esophageal adenocarcinoma in the United States and 83%, respectively, compared with 59% and 56%, and
presenting at early stage (T1) with disease confined to the 75% and 73% in CT and MRI, respectively.66 In addition,
mucosa or submucosa.59 It permits accurate assessment of other recent studies have demonstrated that endobronchial
depth of infiltration, presence of lymphovascular invasion, ultrasound (EBUS) has a greater accuracy in evaluating
and degree of differentiation, thus providing a relevant tracheobronchial invasion by esophageal neoplasia when
estimation of the risk for local lymph node metastasis. compared with conventional bronchoscopy, CT, and EUS.67
However, histopathologic interpretation of esophageal A study by Liberman et al. highlighted the utility
reflux (ER) specimens may not be straightforward. There of EBUS combined with EUS in sampling peritumoral
is a high rate of discordance (up to 48%) between patholo- esophageal lymph nodes as EUS/FNA run a high risk of
gists’ assessment for the depth of tumor invasion in ER false-positive cytopathology secondary to contamination
specimens, particularly for lesions called T1b.60 of the sample by piercing the primary tumor. Adding
The appropriate treatment modality is best selected in EBUS to EUS for staging esophageal cancer in this study
a multidisciplinary fashion with EMR providing the most added valuable information for patients with upper and
accurate staging. Several studies have demonstrated that ER mid-esophageal tumors owing to the inability to assess
is safe and effective for complete resection with excellent local lymph node invasion in the peritracheal area and
comparable long-term disease control, lower rate of com- subcarinal region. A total of 12% of patients were upstaged
plications but higher rate of recurrence when compared with the addition of EBUS to EUS in this study, and that
with surgery for high-grade dysplasia and intramucosal has an important impact on patient prognostication and
adenocarcinoma (T1aN0). In these lesions the risk of treatment planning. However, this technique has not yet
lymph node involvement or hematogenous dissemination seen wide application.68
is estimated to be less than 2%,61 justifying a nonsurgical
approach.62,63 For lesions penetrating into the submucosa, LAPAROSCOPY AND THORACOSCOPY
more definitive treatment, such as esophagectomy with To improve preoperative staging, diagnostic laparoscopy
lymphadenectomy, is advocated. or thoracoscopy is selectively performed. Small-volume
intraperitoneal metastases can be difficult to diagnose
BRONCHOSCOPY AND noninvasively by either CT or PET scans. Laparoscopic or
ENDOBRONCHIAL ULTRASOUND thoracoscopic assessment can help in identifying occult
Flexible bronchoscopy is performed to assess tumor intraperitoneal or intrathoracic distant metastases as well
invasion into the tracheobronchial tree and to evaluate as sampling regional lymph nodes in some situations, such
mediastinal lymph nodes, especially in situations where as in cancers at the GEJ or in the distal esophagus.69,70
the presence of an obstructing esophageal tumor prevents This approach might limit aggressive treatment to patients
the progression of an echoendoscope to perform EUS or with locally advanced or metastatic disease.
when the tumor is located in the mid- and upper portions Laparoscopic staging includes visual inspection of
of the esophagus. the peritoneal cavity and surface of the liver, as well as
Signs of involvement include a widened carina, external the potential for laparoscopic ultrasound examination
compression, tumor infiltration, and fistulization. The of the liver, collection of peritoneal fluid for cytologic
last two signs contraindicate resection.64 The gross mac- examination, and biopsy of suspicious lesions.
roscopic bronchoscopic appearance may not be accurate; In a study of 53 patients who underwent preoperative
in fact, a bulge of the posterior tracheobronchial wall evaluation including minimally invasive staging, a change
or minimal mucosal changes (erythema, edema) do not in the stage originally assigned by CT scan and EUS
confirm neoplastic invasion of the wall, and biopsy plus occurred in 32.1% of patients with adenocarcinoma of
brush cytology is recommended. In one study involving the esophagus.71 The multiinstitutional study from the
patients with supracarinal cancer, endoluminal tumor Cancer and Leukemia Group B (CALGB 9380) reported
mass, protrusion of the posterior tracheal wall, and signs on combined thoracoscopic and laparoscopic staging in
of mucosal invasion were visible in 5.9%, 28.6%, and 4.1% 113 patients; the strategy was feasible in 73% of patients.
of bronchoscopic examinations, respectively. However, in Thoracoscopy and laparoscopy identified nodes or meta-
only 8.6% of 220 bronchoscopic examinations, cancer static disease missed by CT in 50% of patients, by MRI in
invasion was proved by biopsy or cytology. Bronchoscopy 40%, and by EUS in 30%. Although no deaths or major
excluded 18.1% of otherwise potentially operable patients complications occurred, it did involve general anesthesia,
from surgery because of airway invasion, with an overall one-lung anesthesia, a median operating time of 210
accuracy of 93.3%.65 minutes, and a hospital stay of 3 days.72
Endobronchial ultrasound has been investigated as Currently diagnostic thoracoscopic assessment is not
a staging tool. The diagnosis of tracheobronchial inva- used except in highly selective cases where other staging
sion was based on an interruption in the most external studies suggest metastatic disease. Many consider diagnostic
hyperechoic layer of the tracheobronchus (corresponding laparoscopy in patients with adenocarcinoma with extensive
to its adventitia). In one study of 26 patients determined gastric involvement, or in patients with suspicious CT and
to be invasion-free by bronchoscopic ultrasound, only 2 PET findings.
had invasion. This compares with 7 of 22 patients who EUS/FNA is an effective approach for sampling
were found to have invasion not suggested by CT alone. mediastinal nodes. However, laparoscopy has its role,
In another study evaluating esophageal or thyroid cancer especially for adenocarcinoma of the lower esopha-
patients with suspicion of tracheobronchial invasion, the gus or GEJ tumors since the chance of metastases in
378 SECTION I Esophagus and Hernia
the abdomen is considerably greater compared with of approximately 50% when correlated with the percent-
SCC of the esophagus. Laparoscopy can be of use in age of residual viable cells in the surgical specimen and
diagnosing abdominal metastases, such as peritoneal survival.80
secondaries or identifying unsuspected cirrhosis, which A potentially more accurate diagnostic modality that
is a relative contraindication to surgical resection. Its has the potential to predict tumor response early in the
value is minimal for more proximally located tumors.73 course of neoadjuvant treatment is FDG-PET scanning,
Following CT and EUS, laparoscopy may upstage nodal which has been shown to be moderately sensitive in
status in 0% to 21% of patients and downstage it in esophageal cancer. The evaluation of FDG-PET for early
4% to 19%. Together with other findings, change in monitoring of nonsurgical therapy response is described
management can occur in up to 20% of patients. 74 in several small studies and in one large phase II trial,
Even though thoracoscopy and laparoscopy can identify with promising results.80–84
some additional patients with advanced disease, the The MUNICON study recruited 110 patients who
application of these approaches should involve our underwent neoadjuvant chemotherapy, of whom 54 had
highly selected patients.75 metabolic response after 2 weeks of induction therapy.
The increased accuracy of PET/CT may decrease the Nonresponders underwent immediate surgery while
application of other invasive staging methods, especially responders had surgery after a full course of treatment.
with regard to distant disease. It does seem that with better Survival in nonresponders (median, 26 months) was
PET, CT, and EUS, staging laparoscopy is less indicated. inferior to responders, but did not appear to be worse
Laparoscopic assessment is indicated in cases where liver than a historical cohort who received chemotherapy
metastases or peritoneal metastases are suspected and and surgery, despite incomplete chemotherapy.85 This
confirmation is required. study suggests the potential feasibility of a PET-guided
Proper staging of esophageal cancer patients is criti- treatment algorithm.
cal in view of the wide available variation in treatment In a study on patients who were assessed before and
approaches. Fig. 36.8 highlights all steps for the work-up of after chemoradiation, Brücher et al. demonstrated
patients with esophageal cancer. Following the completion a sensitivity in detecting a response of 100%, with a
of staging and physiologic assessment, patients should be corresponding specificity of 55%. The PPV and NPV
considered for presentation at a multidisciplinary tumor were 72% and 100%, respectively. 81 Flamen et al.
board whenever it is feasible. also demonstrated the value of PET for evaluation of
response. Serial PET had a predictive accuracy of 78%
for a major response, with a sensitivity of 71% and a
THERAPY MONITORING specificity of 82%.82
Neoadjuvant chemoradiotherapy followed by surgery has False negatives (overestimation of response) is due
demonstrated a favorable long-term outcome compared to residual micrometastatic cancer foci falling below the
with surgery alone.76 However, in a proportion of patients, detection threshold. False positives (underestimation of
the objective anticipated response will be insufficient, or response) usually occur at the primary tumor site where
the disease will progress in spite of neoadjuvant therapy. inflammatory reactions may increase FDG uptake. FDG-
These findings can result in changing the management PET scanning cannot demonstrate a complete pathologic
strategy as chemoradiotherapy side effects could lead to response to neoadjuvant chemoradiation, although clinical
increased perioperative morbidity. In addition, prolonged trials examining this issue are ongoing.
but ineffective preoperative treatment will inevitably delay
appropriate surgical therapy. In addition, the ability to
identify nonresponders will increase the ability to tailor
SURVEILLANCE STRATEGY
therapy. Surveillance, following either surgical or medical manage-
Conventional imaging techniques for monitoring ment of esophageal cancer, is typically done to implement
nonsurgical therapy, such as CT and EUS, are based salvage therapy in cases of locoregional failure. However,
on morphologic imaging. General restrictions of these the number of salvageable cases after recurrences is small.
methods include difficulty in distinguishing a viable tumor Sudo et al. demonstrated that only 2% of 518 esophageal
from necrotic or fibrotic tissue and delay between cell adenocarcinoma patients who underwent trimodality
kill and tumor shrinkage.77 Nevertheless, CT scanning is therapy were salvageable when they recurred at a median
still widely applied for this purpose, partly because of its of 29.3 months of follow-up, which raised the question of
wide availability. In a study involving the use of CT scans, the benefit of the surveillance/salvage strategies.86
CT after chemoradiotherapy accurately staged the T Following nonsurgical management, it is difficult to
classification in only 42%; it overstaged 36% of patients, evaluate the lesion as distinguishing between fibrosis,
and understaged 20%. CT had a sensitivity of 65%, a inflammation, and true histologic response is not easy.
specificity of 33%, a PPV of 58%, and an NPV of 41% in The posttreatment surveillance strategy suggested by
evaluating the pathologic tumor response.78 the consensus-based guidelines from the National Com-
EUS is typically not recommended for response assess- prehensive Cancer Network include the following steps:
ment because of its limited accuracy and poor reproduc- • History and physical examination every 3 to 6 months
ibility. Moreover, EUS is invasive and not always feasible, for 1 to 3 years, then every 6 months for years 4 and
especially in case of postradiation esophagitis or severe 5, then annually
stricturing.79 A study assessing the prognostic value of • Complete blood count and chemistry profile, as clinically
endoscopic biopsy and EUS demonstrated an accuracy indicated
Esophageal Cancer Diagnosis and Staging CHAPTER 36 379
STAGING/INVESTIGATION ALGORITHM
Presentation with
dysphagia/weight loss
EGD-US ± FNA
Multidisciplinary tumor
board
FIGURE 36.8 Scheme of stepwise approach in diagnosing suspected cancer in the esophagus. Ca, Cancer; CT, computed tomography;
EGD, esophagogastroduodenoscopy; FDG, 2-[18F]-fluoro-2-deoxyglucose; FNA, fine-needle aspiration; F/U, follow-up; HGD, high-grade
dysplasia; PET, positron emission tomography; Rx, treatment; US, ultrasound.
380 SECTION I Esophagus and Hernia
• Radiologic imaging and upper GI endoscopy, as clinically 21. Levine MS, Chu P, Furth EE, Rubesin SE, Laufer I, Herlinger H.
indicated Carcinoma of the esophagus and esophagogastric junction: sensitivity
of radiographic diagnosis. AJR Am J Roentgenol. 1997;168:1423.
• Dilation for anastomotic stenosis 22. Graham DY, Schwartz JT, Cain GD, Gyorkey F. Prospective evalua-
• Nutritional counseling. tion of biopsy number in the diagnosis of esophageal and gastric
carcinoma. Gastroenterology. 1982;82:228-231.
23. Bloomfeld RS, Bridgers DI 3rd, Pineau BC. Sensitivity of upper endos-
ACKNOWLEDGMENT copy in diagnosing esophageal cancer. Dysphagia. 2005;20:278-282.
24. Jacobson BC, Hirota W, Baron TH, et al. The role of endoscopy
We would like to acknowledge Dr. Law for the use of parts of his in the assessment and treatment of esophageal cancer. Gastrointest
chapter published in the seventh edition. Endosc. 2003;57:817-822.
25. Zargar SA, Khurro MS, Jan GM, Mahajan R, Shah P. Prospective
comparison of the value of brushings before and after biopsy in the
REFERENCES endoscopic diagnosis of gastroesophageal malignancy. Acta Cytol.
1991;35:549-552.
1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA Cancer 26. Faigel DO, Deveney C, Phillips D, Fennerty MB. Biopsy negative
J Clin. 2013;63:11-30. malignant esophageal stricture: diagnosis by endoscopic ultrasound.
2. Baquet CR, Commiskey P, Mack K, Meltzer S, Mishra SI. Esophageal Am J Gastroenterol. 1998;93:2257-2260.
cancer epidemiology in blacks and whites: racial and gender dispari- 27. Heresbach D, Leray E, d’Halluin PN, et al. Diagnostic accuracy of
ties in incidence, mortality, survival rates and histology. J Natl Med esophageal capsule endoscopy versus conventional upper digestive
Assoc. 2005;97:1471-1478. endoscopy for suspected esophageal squamous cell carcinoma.
3. Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer Endoscopy. 2010;42:93-97.
J Clin. 2005;55:10-30. 28. Hofstetter W, Swisher SG, Correa AM, et al. Treatment outcomes
4. Zhang Y. Epidemiology of esophageal cancer. World J Gastroenterol. of resected esophageal cancer. Ann Surg. 2002;236:376-385.
2013;19:5598-5606. 29. Rosenbaum S, Stergar H, Antoch G, Veit P, Bockisch A, Kühl H.
5. Hur C, Miller M, Kong CY, et al. Trends in esophageal adenocarci- Staging and follow-up of gastrointestinal tumors with PET/CT.
noma incidence and mortality. Cancer. 2013;119:1149-1158. Abdom Imaging. 2006;31:25-35.
6. Li JY. Epidemiology of esophageal cancer in China. Natl Cancer Inst 30. Stahl A, Ott K, Weber WA, et al. FDG PET imaging of locally advanced
Monogr. 1982;62:113-120. gastric carcinomas: correlation with endoscopic and histopathological
7. MacDonald WC, MacDonald JB. Adenocarcinoma of the esophagus findings. Eur J Nucl Med Mol Imaging. 2003;30:288-295.
and/or gastric cardia. Cancer. 1987;60:1094. 31. Kato H, Kuwano H, Nakajima M, et al. Comparison between positron
8. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer emission tomography and computed tomography in the use of the
J Clin. 2012;62:10-29. assessment of esophageal carcinoma. Cancer. 2002;94:921.
9. Soetikno R, Kaltenbach T, Yeh R, Gotoda T. Endoscopic mucosal 32. van Vliet EP, Heijenbrok-Kal MH, Hunink MG, Kuipers EJ, Siersema
resection for early cancers of the upper gastrointestinal tract. J Clin PD. Staging investigations for oesophageal cancer: a meta-analysis.
Oncol. 2005;23:4490-4498. Br J Cancer. 2008;98:547-557.
10. Sjoquist KM, Burmeister BH, Smithers BM, et al. Survival after 33. Rankin SC, Taylor H, Cook GJ, Mason R. Computed tomography
neoadjuvant chemotherapy or chemoradiotherapy for resectable and positron emission tomography in the pre-operative staging of
oesophageal carcinoma: an updated meta-analysis. Lancet Oncol. oesophageal carcinoma. Clin Radiol. 1998;53:659-665.
2011;12:681. 34. Hocazade C, Özdemir N, Yazici O, et al. Concordance of positron
11. Buas MF, Vaughan TL. Epidemiology and risk factors for gastro- emission tomography and computed tomography in patients with
esophageal junction tumors: understanding the rising incidence locally advanced gastric and esophageal cancer. Ann Nucl Med.
of this disease. Semin Radiat Oncol. 2013;23:3-9. 2015;29:621-626.
12. Meltzer CC, Luketich JD, Friedman D, et al. Whole-body FDG 35. Luketich JD, Friedman DM, Weigel TL, et al. Evaluation of distant
positron emission tomographic imaging for staging esophageal metastases in esophageal cancer: 100 consecutive positron emission
cancer: comparison with computed tomography. Clin Nucl Med. tomography scans. Ann Thorac Surg. 1999;68:1133.
2000;25:882. 36. van Westreenen HL, Westerterp M, Bossuyt PM, et al. Systematic
13. Siewert JR, Katja O. Are squamous and adenocarcinomas of the review of the staging performance of 18F-fluorodeoxyglucose positron
esophagus the same disease? Semin Radiat Oncol. 2007;17:38-44. emission tomography in esophageal cancer. J Clin Oncol. 2004;22:3805.
14. Siewert JR, Stein HJ, Feith M, Bruecher BL, Bartels H, Fink U. 37. Flamen P, Lerut T, Haustermans K, Van Cutsem E, Mortelmans
Histologic tumor type is an independent prognostic parameter L. Position of positron emission tomography and other imaging
in esophageal cancer: lessons from more than 1,000 consecutive diagnostic modalities in esophageal cancer. Q J Nucl Med Mol Imaging.
resections at a single center in the Western world. Ann Surg. 2001;234: 2004;48:96-108.
360-367. 38. Romagnuolo J, Scott J, Hawes RH, et al. Helical CT versus EUS with
15. Zhang HD, Chen CG, Gao YY, et al. Primary esophageal adenosqua- fine needle aspiration for celiac nodal assessment in patients with
mous carcinoma: a retrospective analysis of 24 cases. Dis Esophagus. esophageal cancer. Gastrointest Endosc. 2002;55:648.
2014;27:783-789. 39. van Overhagen H, Becker C. Diagnosis and staging of carcinoma
16. American Joint Committee on Cancer. AJCC Cancer Staging Manual. of the esophagus and gastroesophageal junction, and detection of
7th ed. New York: Springer; 2010. postoperative recurrence, by computed tomography. In: Myers M,
17. Groth SS, Virnig BA, Whitson BA, et al. Determination of the ed. Neoplasms of the Digestive Tract: Imaging, Staging and Management.
minimum number of lymph nodes to examine to maximize survival Philadelphia: Lippincott-Raven; 1998:31.
in patients with esophageal carcinoma: data from the Surveillance, 40. Wu LF, Wang BZ, Feng JL, et al. Preoperative TN staging of esophageal
Epidemiology and End Results database. J Thorac Cardiovasc Surg. 2010; cancer: comparison of miniprobe ultrasonography, spiral CT and
139:612-620. MRI. World J Gastroenterol. 2003;9:219-224.
18. Peyre CG, Hagen JA, DeMeester SR, et al. Predicting systemic 41. Riddell AM, Allum WH, Thompson JN, Wotherspoon AC, Richardson
disease in patients with esophageal cancer after esophagectomy: a C, Brown G. The appearances of oesophageal carcinoma demon-
multinational study on the significance of the number of involved strated on high-resolution, T2-weighted MRI, with histopathological
lymph nodes. Ann Surg. 2008;248:979-985. correlation. Eur Radiol. 2007;17:391-399.
19. Rice TW, Rusch VW, Ishwaran H, Blackstone EH; Worldwide 42. Sakurada A, Takahara T, Kwee TC, et al. Diagnostic performance
Esophageal Cancer Collaboration. Cancer of the esophagus and of diffusion-weighted magnetic resonance imaging in esophageal
esophagogastric junction: data-driven staging for the seventh edition cancer. Eur Radiol. 2009;19:1461-1469.
of the American Joint Committee on Cancer/International Union 43. Lehr L, Rupp N, Siewert JR. Assessment of resectability of esophageal
Against Cancer Cancer Staging Manuals. Cancer. 2010;116:3763- cancer by computed tomography and magnetic resonance imaging.
3773. Surgery. 1998;103:344.
20. Enzinger PC, Mayer RJ. Esophageal cancer. N Engl J Med. 2003;349:2241- 44. Hasegawa N, Niwa Y, Arisawa T, Hase S, Goto H, Hayakawa T.
2252. Preoperative staging of superficial esophageal carcinoma: comparison
Esophageal Cancer Diagnosis and Staging CHAPTER 36 381
of an ultrasound probe and standard endoscopic ultrasonography. invasion by supracarinal esophageal cancer: a prospective study in
Gastrointest Endosc. 1996;44:388. 166 patients. Chest. 2001;119:1652.
45. Puli SR, Reddy JB, Bechtold ML, Antillon D, Ibdah JA, Antillon 66. Nishimura Y, Osugi H, Inoue K, Takada N, Takamura M, Kinosita H.
MR. Staging accuracy of esophageal cancer by endoscopic ultra- Bronchoscopic ultrasonography in the diagnosis of tracheobronchial
sound: a meta-analysis and systematic review. World J Gastroenterol. invasion of esophageal cancer. J Ultrasound Med. 2002;21:49-58.
2008;14:1479-1490. 67. Wakamatsu T, Tsushima K, Yasuo M, et al. Usefulness of preoperative
46. Young PE, Gentry AB, Acosta RD, Greenwald BD, Riddle M. Endo- endobronchial ultrasound for airway invasion around the trachea:
scopic ultrasound does not accurately stage early adenocarcinoma esophageal cancer and thyroid cancer. Respiration. 2006;73:651-657.
or high-grade dysplasia of the esophagus. Clin Gastroenterol Hepatol. 68. Liberman M, Hanna N, Duranceau A, Thiffault V, Ferraro P. Endo-
2010;8:1037. bronchial ultrasonography added to endoscopic ultrasonography
47. May A, Gunter E, Roth F, et al. Accuracy of staging in early esopha- improves staging in esophageal cancer. Ann Thorac Surg. 2013;96:232.
geal cancer using high resolution endoscopy and high resolution 69. Kaushik N, Khalid A, Brody D, Luketich J, McGrath K. Endoscopic
endosonography: a comparative, prospective, and blinded trial. Gut. ultrasound compared with laparoscopy for staging esophageal cancer.
2004;53:624-640. Ann Thorac Surg. 2007;83:2000.
48. Thosani N, Singh H, Kapadia A, et al. Diagnostic accuracy of EUS 70. Bryan RT, Cruickshank NR, Needham SJ, et al. Fielding laparoscopic
in differentiating mucosal versus submucosal invasion of superficial peritoneal lavage in staging gastric and oesophageal cancer. Eur J
esophageal cancers: a systematic review and meta-analysis. Gastrointest Surg Oncol. 2001;27:291.
Endosc. 2012;75:242. 71. Luketich JD, Meehan M, Nguyen NT, et al. Minimally invasive
49. Rösch T. Endosonographic staging of esophageal cancer: a review surgical staging for esophageal cancer. Surg Endosc. 2000;14:700.
of literature results. Gastrointest Endosc Clin N Am. 1995;5(3):537. 72. Krasna MJ, Reed CE, Nedzwiecki D, et al. CALGB 9380: a prospec-
50. Saunders HS, Wolfman NT, Ott DJ. Esophageal cancer radiologic tive trial of the feasibility of thoracoscopy/laparoscopy in staging
staging. Radiol Clin North Am. 1997;35:281-294. esophageal cancer. Ann Thorac Surg. 2001;71:1073.
51. Pouw RE, Heldoorn N, Alvarez Herrero L, et al. Do we still need 73. Stein HJ, Kraemer SJ, Feussner H, Fink U, Siewert JR. Clinical value
EUS in the workup of patients with early esophageal neoplasia? A of diagnostic laparoscopy with laparoscopic ultrasound in patients
retrospective analysis of 131 cases. Gastrointest Endosc. 2011;73:662. with cancer of the esophagus or cardia. J Gastrointest Surg. 1997;1:167.
52. Vazquez-Sequeiros E, Levy MJ, Clain JE, et al. Routine vs. selec- 74. Yoon HH, Lowe VJ, Cassivi SD, Romero Y. The role of FDG-PET and
tive EUS-guided FNA approach for preoperative nodal staging of staging laparoscopy in the management of patients with cancer of
esophageal carcinoma. Gastrointest Endosc. 2006;63:204-211. the esophagus or gastroesophageal junction. Gastroenterol Clin North
53. Eloubeidi MA, Wallace MB, Reed CE, et al. The utility of EUS Am. 2009;38:105.
and EUS-guided fine needle aspiration in detecting celiac lymph 75. Wallace MB, Nietert PJ, Earle C, et al. An analysis of multiple staging
node metastasis in patients with esophageal cancer: a single-center management strategies for carcinoma of the esophagus: computed
experience. Gastrointest Endosc. 2001;54:714. tomography, endoscopic ultrasound, positron emission tomography,
54. Giovannini M, Seitz JF, Monges G, Perrier H, Rabbia I. Fine-needle and thoracoscopy/laparoscopy. Ann Thorac Surg. 2002;74:1026.
aspiration cytology guided by endoscopic ultrasonography: results 76. Gebski V, Burmeister B, Smithers BM, et al. Survival benefits from
in 141 patients. Endoscopy. 1995;27:171. neoadjuvant chemoradiotherapy or chemotherapy in oesophageal
55. Vazquez-Sequeiros E, Wiersema MJ, Clain JE, et al. Impact of lymph carcinoma: a meta-analysis. Lancet Oncol. 2007;8:226-234.
node staging on therapy of esophageal carcinoma. Gastroenterology. 77. Westerterp M, van Westreenen HL, Reitsma JB, et al. Esophageal
2003;125:1626. cancer: CT, endoscopic US, and FDG PET for assessment of response
56. Fockens P, Van den Brande JH, van Dullemen HM, van Lanschot JJ, to neoadjuvant therapy—systematic review. Radiology. 2005;236:841-
Tytgat GN. Endosonographic T-staging of esophageal carcinoma: a 851.
learning curve. Gastrointest Endosc. 1996;44:58. 78. Jones DR, Parker LAJ, Detterbeck FC, Egan TM. Inadequacy of
57. Souquet JC, Napoleon B, Pujol B, Valette PJ, Chollet R, Lambert computed tomography in assessing patients with esophageal carci-
R. Endosonography-guided treatment of esophageal carcinoma. noma after induction chemoradiotherapy. Cancer. 1999;85:1026.
Endoscopy. 1992;1:324. 79. Sloof GW. Response monitoring of neoadjuvant therapy using CT,
58. Inoue H, Takeshita K, Hori H, Muraoka Y, Yoneshima H, Endo M. EUS, and FDG-PET. Best Pract Res Clin Gastroenterol. 2006;20:941-957.
Endoscopic mucosal resection with a cap-fitted panendoscope for 80. Schneider PM, Metzger R, Schaefer H, et al. Response evaluation by
esophagus, stomach, and colon mucosal lesions. Gastrointest Endosc. endoscopy, rebiopsy, and endoscopic ultrasound does not accurately
1993;39:58-62. predict histopathologic regression after neoadjuvant chemoradiation
59. Das A, Singh V, Fleischer DE, Sharma VK. A comparison of endoscopic for esophageal cancer. Ann Surg. 2008;248:902.
treatment and surgery in early esophageal cancer: an analysis of 81. Brücher BL, Weber W, Bauer M, et al. Neoadjuvant therapy of
Surveillance, Epidemiology and End Results data. Am J Gastroenterol. esophageal squamous cell carcinoma: response evaluation by positron
2008;103:1340-1345. emission tomography. Ann Surg. 2001;233:300.
60. Worrell SG, Boys JA, Chandrasoma P, et al. Inter-observer variability 82. Flamen P, Van Cutsem E, Lerut A, et al. Positron emission tomography
in the interpretation of endoscopic mucosal resection specimens for assessment of the response to induction radiochemotherapy in
of esophageal adenocarcinoma: interpretation of ER specimens. locally advanced oesophageal cancer. Ann Oncol. 2002;13:361.
J Gastrointest Surg. 2016;20:140-145. 83. Wieder HA, Brucher BL, Zimmermann F, et al. Time course of tumor
61. Manner H, Pech O, Heldmann Y, et al. Efficacy, safety, and long-term metabolic activity during chemoradiotherapy of esophageal squa-
results of endoscopic treatment for early stage adenocarcinoma of mous cell carcinoma and response to treatment. J Clin Oncol. 2004;
the esophagus with low-risk sm1 invasion. Clin Gastroenterol Hepatol. 22:900-908.
2013;11:630-635. 84. Cerfolio RJ, Bryant AS, Ohja B, Bartolucci AA, Eloubeidi MA. The
62. Fernando HC, Luketich JD, Buenaventura PO, Perry Y, Christie NA. accuracy of endoscopic ultrasonography with fine-needle aspiration,
Outcomes of minimally invasive esophagectomy (MIE) for high-grade integrated positron emission tomography with computed tomography,
dysplasia of the esophagus. Eur J Cardiothorac Surg. 2002;22:1. and computed tomography in restaging patients with esophageal
63. Schmidt HM, Mohiuddin K, Bodnar AM, et al. Multidisciplinary treat- cancer after neoadjuvant chemoradiotherapy. J Thorac Cardiovasc
ment of T1a adenocarcinoma in Barrett’s esophagus: contemporary Surg. 2005;129:1232-1241.
comparison of endoscopic and surgical treatment in physiologically 85. Lordick F, Ott K, Krause BJ, et al. PET to assess early metabolic
fit patients. Surg Endosc. 2016;30:3391-3401. response and to guide treatment of adenocarcinoma of the oesopha-
64. Cheung HC, Siu KF, Wong J. A comparison of flexible and rigid gogastric junction: the MUNICON phase II trial. Lancet Oncol. 2007;
endoscopy in evaluating esophageal cancer patients for surgery. 8:797-805.
World J Surg. 1988;12:117. 86. Sudo K, Taketa T, Correa AM, et al. Locoregional failure rate after
65. Riedel M, Stein HJ, Mounyam L, Lembeck R, Siewert JR. Extensive preoperative chemoradiation of esophageal adenocarcinoma and
sampling improves preoperative bronchoscopic assessment of airway the outcomes of salvage strategies. J Clin Oncol. 2013;31:4306-4310.
CHAPTER
T
he detection and treatment of high-grade dysplasia to EAC.3–6 On the other hand, many patients presenting
and early esophageal cancer overlap to the point that with a new diagnosis of EAC do not have BE and have
they are a continuum. In some situations they can never had symptoms of reflux. Perhaps that is because
be difficult to separate and exist simultaneously. Because “silent” reflux is present in this subgroup of patients
existing studies often combine the descriptions of the two, or there are other variables driving the progression to
for the purposes of this chapter we will consider both as neoplasia that are not fully understood. We have a little
superficial neoplasia. better understanding about disease progression when
Esophagectomy, long considered the only curative BE is present. Through an accumulation of incompletely
option for esophageal cancer, is no longer standard therapy understood genetic alterations, disease can progress from
for high-grade dysplasia (HGD) and early esophageal nondysplastic intestinal metaplasia to low-grade dysplasia
cancer limited to the mucosa. Recognition of disease (LGD), HGD, and invasive EAC.7 However, an orderly
biology in combination with bridging technologic gaps progression through the metaplasia-dysplasia-cancer
have brought about a change in the treatment paradigm sequence may not be recognized or even present in every
that is both disruptive to previous therapies and represents patient with a Barrett-associated cancer. It is possible that
a clear advance in therapy. Organ-sparing procedures episodic observation of the esophagus misses those events
including endoscopic resection (ER) in combination in real time, if they have occurred. Another explanation
with various forms of ablation have been shown to be the may be that the disease has the ability to skip intermediate
preferable option for the treatment of early esophageal stages of progression once a critical genetic mutation
neoplasia in the vast majority of cases.1–4 or mutational load is reached. Nonetheless, screening
In the United States the incidence of esophageal endoscopies are one of the most important elements
adenocarcinoma (EAC) has been increasing, whereas in discovering early, curable disease that can be treated
squamous cell cancers of the esophagus are decreasing.5 without removal of the esophagus. Unfortunately, it is the
In fact, the rise in incidence of EAC is inconsistent with rare patient with early disease who manifests observable
most other cancers, which are either stable or declining. symptoms that would set into motion the need for an
The reasons for this are not completely clear, but what endoscopy. It is far more common that symptoms are
is known is that this is an insidious disease. Symptoms the harbinger of locally advanced or advanced disease.
occur late in the disease process, leaving most cases of Therefore devising a cost-effective screening strategy that
EAC diagnosed at an advanced stage, a point where the will detect early disease is challenging. Given the low yield
disease is lethal in the majority of patients. Therefore of endoscopy/cancer diagnoses, current recommenda-
an early diagnosis constitutes our best hope of impact- tions by expert consensus panels indicate that there is a
ing survival. Emphasis on public awareness regarding lack of sufficient evidence to support routine screening
the risk factors for EAC has perhaps stimulated a trend for Barrett-related disease. According to the American
toward esophageal surveillance, despite a lack of clarity Gastroenterological Association report “well-established
within the gastroenterology society guidelines. People risk factors for Barrett’s esophagus include advanced age,
with long-standing reflux who are not realizing benefits male sex, white ethnicity, GERD, hiatal hernia, elevated
from medical therapy or those who are simply undergoing body mass index, and a predominantly intraabdominal
routine colon screening and describe a long history of distribution of body fat. [However], despite the consider-
reflux to their gastroenterologists are often undergoing able published data available on risk factors for Barrett’s
screening. This informal screening method has been one esophagus, few attempts have been made to apply this
of the only mechanisms for us to routinely observe and information systematically in the design of guidelines
treat early stage esophageal disease. on who to screen for the condition.”1 The vexing issue
is that we are unable to focus screening efforts upon a
sufficiently high-risk population to justify the expense.
SCREENING AND SURVEILLANCE Notwithstanding the lack of concrete guidelines, most
Screening for squamous cell carcinoma (SCC) of the experts would recommend that any patient who has had
esophagus should be considered for patient populations long-standing reflux, has a family history of esophageal
with known risk factors, such as tobacco and alcohol neoplasia, continues to be symptomatic with GERD despite
use, Plummer-Vinson syndrome, tylosis palmaris et plan- therapy, or has unexplained anemia should consider
taris, history of head and neck cancer, caustic injury, endoscopic screening of the esophagus. Likewise, longi-
and achalasia. Increased body mass index (BMI), hiatal tudinal studies have shown that initial presentation with
hernia, and gastroesophageal reflux disease (GERD) are all erosive esophagitis (Los Angeles [LA] Classification of
considered to be contributors to the formation of intestinal GERD grade C/D) and regular proton pump inhibitor
metaplasia (Barrett esophagus [BE]), a precursor lesion (PPI) use were independently associated with progression
382
Endoscopic Management of High-Grade Dysplasia and Superficial Esophageal Carcinoma CHAPTER 37 382.e1
ABSTRACT
Esophagectomy is no longer considered standard therapy
for high-grade dysplasia and early esophageal cancer
limited to the mucosa. Organ-sparing procedures including
endoscopic resection in combination with various forms
of ablation have been shown to be the preferable option
for the treatment of early esophageal neoplasia in most
cases. This chapter will discuss the data that have led to
this paradigm shift in treatment, along with specifying ideal
candidates for endoscopic therapies, current treatment
modalities, clinical and safety outcomes, and specific
management recommendations.
KEYWORDS
Dysplasia
Early stage esophageal carcinoma
Radiofrequency ablation
Cryotherapy
Endoscopic resection
Esophagectomy
Intramucosal cancer
Endoscopic treatment
Mucosal resection
Esophageal adenocarcinoma
Esophageal squamous cell carcinoma
Endoscopic Management of High-Grade Dysplasia and Superficial Esophageal Carcinoma CHAPTER 37 383
TABLE 37.1 Guidelines for Patients With Barrett ENDOSCOPIC DIAGNOSIS AND STAGING
Esophagus With and Without Dysplasia A detailed inspection of the Barrett segment with high-
Diagnosis Recommendations definition white light endoscopy with or without magnifica-
Nondysplastic EGD every 3–5 years
tion is essential for the detection of mucosal abnormalities,
BE Four-quadrant biopsies every 2 cm nodularity, and cancer. Studies reveal that the longer the
Focused biopsy on areas of abnormality time spent inspecting the Barrett segment, the higher the
Low-grade Confirm results with expert pathologist likelihood of detecting suspicious lesions. In a study of
dysplasia Follow-up EGD with biopsies in 6 months 112 patients undergoing surveillance, the authors reported
Consider endoscopic resection and/or ablation that endoscopists who averaged greater than 1 minute of
Surveillance after therapy is necessary inspection time per centimeter of Barrett detected more
High-grade Confirm results with expert pathologist suspicious lesions than those who spent less time.16
dysplasia Refer for endoscopic resection and/or ablation Topographic detection of suspicious lesions is probably
Consider esophagectomy in refractory cases also aided by various techniques for mucosal enhancement.
Surveillance after endoscopic therapy is A number of visual aids such as chromoendoscopy, narrow
necessary band imaging (NBI; Olympus, Center Valley, Pennsylvania),
BE, Barrett esophagus; EGD, esophagogastroduodenoscopy. autofluorescence imaging, confocal laser endomicroscopy,
and volumetric laser endomicroscopy have become available
for detailed visualization and characterization of mucosal
to BE, warranting screening in these populations.8 After and cellular architecture. Although the literature has not
BE has been discovered, endoscopic efforts convert from clearly defined the benefit of these advanced imaging
screening to surveillance, the purpose of which is to techniques, we advocate using mucosal enhancements that
detect the presence of esophageal lesions that are at are either electronic or vital staining as a routine practice
high risk of progressing to EAC. Although the benefit to facilitate the detection of HGD and early neoplasia.
of surveillance has also not been conclusively proved, These modalities can be very helpful in targeting biopsies.17
several studies have shown that patients who contracted Upper endoscopy with ultrasound plays an important
EAC while under surveillance presented at a lower stage role in the diagnosis and staging of esophageal cancer. In
of disease and had improved survival when compared with cases of intermediate to locally advanced EAC, endoscopic
those not undergoing surveillance.9,10 In fact, surveillance ultrasound (EUS) is the principal method used to estimate
guidelines have been developed under the assumption the depth and regional nodal status of the disease. In
that the practice will reduce deaths. contrast, EUS maintains a limited role for the evaluation
Surveillance guidelines vary according to the degree of patients with HGD and early EAC. The accuracy of EUS
of dysplasia detected in the esophageal biopsies (Table staging in these disease categories is modest at best.18 In
37.1). Mucosal irregularities identified on endoscopy need a study describing the accuracy of endoscopy compared
particular attention and should be biopsied with multiple with EUS, the sensitivity of EUS staging for mucosal tumors
bites taken to increase accuracy. An expert pathologist was 90%, whereas for submucosal tumors it was 46%,
should confirm the detection of dysplasia or cancer.11 which was not significantly different from the sensitivity
Although older literature advocated for surveillance, of high-resolution endoscopy in experienced hands.19 A
patients with confirmed HGD should now be referred for systematic review compared EUS with endoscopic mucosal
treatment.11–14 As discussed in the introduction, bridging resection (EMR) or surgical pathology for early (T1-T2)
a technology gap that enables efficient and safe therapy tumors. In that study EUS predicted the depth of the
has changed the recommendations for treatment of target lesion with 67% accuracy (12 studies, n = 132).
HGD. Without intervention, the progression risk from flat Because some patients had multiple lesions, analysis on
HGD to EAC is high, ranging from 6% to 19% per year. an individual patient basis reduced the accuracy to only
Patients with visible lesions, such as a nodular esophagus, 56%.20 Given the available data, most experts would not
have a much higher risk of progression (40% to 70%).13 recommend EUS in patients with a flat Barrett segment
There is also a risk of concomitant adenocarcinoma in and HGD detected by biopsy. There are several factors
patients diagnosed with HGD. Historical data have given that may contribute to the suboptimal accuracy of EUS
us somewhat of a perspective on this risk. In one study, in early EAC, including wall thickening due to chronic
patients who underwent esophagectomy for HGD actually inflammation, presence of a duplicated muscularis mucosa
harbored invasive cancer in approximately 50% of the often seen around the junction, anatomic changes at the
resected specimens.13 But to be equitable, other more level of the gastroesophageal junction (GEJ)/cardia, and
recent studies using better endoscopy techniques and endoscopist’s experience.20 Despite these drawbacks, we
superior equipment suggest that the rate of undiagnosed continue to advocate for EUS in select cases in which there
cancer may be as low 11% in patients without visible is a possibility of detecting malignant lymphadenopathy.
lesions.15 Fine-needle aspiration (FNA) is especially important to
Surveillance as a treatment strategy for HGD should verify the presence of disease in any suspicious lymph
be reserved for those who are unable or unwilling to nodes. In a study of 25 patients undergoing EUS evaluation
undergo therapy. In these cases endoscopy should be (13 with intramucosal adenocarcinoma), 7 patients were
scheduled every 3 months, with random four-quadrant found to have suspicious lymphadenopathy. FNA confirmed
biopsies every 1 cm with focused biopsies on any mucosal malignancy in five of these seven patients. EUS identified
irregularity. five patients (20%) who were unsuitable candidates for
384 SECTION I Esophagus and Hernia
endoscopic therapy.21 Studies like this one highlight the intrinsic properties of the esophagus and the energy
importance of a careful, individual approach to patients setting used for the RFA. This will typically ablate through
with early esophageal cancer. the epithelium and into the lamina propria layer of the
EMR is the preferred diagnostic and therapeutic tool esophageal mucosa. Multiple endoscopic treatments may be
for patients with nodular Barrett or early adenocarcinoma. required to attain complete eradication of dysplasia (CE-D)
EMR is capable of removing small- to moderate-sized and/or metaplasia, with the goal being total resolution
mucosal and superficial submucosal lesions. It is superior to of metaplasia. Treatments are usually performed every 2
EUS for the assessment of depth of invasion by the simple to 3 months until therapeutic goals are met, after which
fact that depth is determined by histologic examination surveillance is continued.
under the microscope. In practice, studies that compared Several studies have demonstrated the efficacy, safety,
EMR with esophagogastroduodenoscopy [EGD]-EUS and durability of RFA to treat dysplastic Barrett.14,24–26
report that the pathologic assessment was changed by In the AIM trial, a randomized multicenter study, 127
EMR in 30% to 48% of cases.22,23 A prospective study of 75 patients with dysplastic BE were assigned to receive either
patients with biopsy-proven HGD or early cancer reports RFA or sham procedure.14 Eighty-four patients were
that the pathology from an EMR changed the tumor randomized to the RFA treatment group (42 patients
grading or staging in 48% of patients (downgrading in with HGD and 42 with LGD). After an average of 3.5
28% and upgrading in 20%).22 Another study of 293 EMR treatments per patient, CE-D occurred in 81% of HGD
procedures performed on focal lesions indicated that 30% patients assigned to the ablation group as compared with
of patients changed therapy as a result of ER compared with 19% of those assigned to the control group (P < .001).
clinical assessment.23 These are very significant findings; Overall, 77.4% of patients in the RFA group had complete
imagine redirecting the recommended therapy of an eradication of intestinal metaplasia (CE-IM) compared with
individual patient from combined multimodality therapy 2.3% of those in the control group (P < .001). Progression
consisting of chemoradiation followed by esophagectomy from HGD to cancer occurred in 4 of 21 patients in the
to an organ-sparing, curative EMR. control group and in 1 of 42 patients in the RFA treated
In summary, optimal evaluation is fundamental to the group (P = .045). Longer follow-up at 2 years indicated
management of dysplastic and early-stage mucosal and that among subjects with initial HGD, there was CE-D in
selected submucosal lesions. ER is an important diagnostic 93% and CE-IM in 89%. At 3 years, CE-D was reported
and potentially therapeutic tool for patients with early in 98% of patients and CE-IM in 91%. The annual rate
neoplastic disease. of progression to EAC among those treated with RFA was
0.55% per patient per year.24
The safety profile of RFA has also been described. The
ENDOSCOPIC TREATMENT OF HIGH-GRADE most common complications reported include: chest
DYSPLASIA/INTRAMUCOSAL CANCER pain lasting less than 1 week, strictures requiring dilation
(6% to 8%), and gastrointestinal hemorrhage (1%).14,27
ABLATION Although operator experience has been shown to impact
• Endoscopic therapy currently consists of methods that the success of RFA and the number of sessions needed to
either ablate (destroy) or resect tissue. achieve CE-IM, it does not appear to impact the risk of com-
• Advantages of ablation therapies include ability to apply plications including stricture formation, gastrointestinal
over large surface areas. hemorrhage, perforation, or hospitalization.
• Repeat applications are possible and typically required Several factors have been identified as predictors of a
to eradicate long segments of metaplasia/dysplasia. poor response after RFA. In a large, multicenter study of
• Ablation therapies do not provide a pathologic specimen 278 patients, poor initial response was defined less than
and are not as effective in treating elevated or ulcerated 50% regression of the BE 3 months after circumferential
lesions. balloon-based RFA. Thirty-six patients (13%) were deemed
• Ablative therapies include photodynamic therapy, to have a poor initial response. Predictive factors for a
thermal laser, argon plasma coagulation (APC), mul- poor response included the presence of active reflux
tipolar electrocoagulation, radiofrequency ablation esophagitis, regeneration of Barrett at the ER scar, relative
(RFA), and cryoablation. This chapter will focus on esophageal narrowing prior to RFA, and number of years
describing RFA and cryoablation, given the evidence with dysplastic changes before RFA.28
of effectiveness, ease of application, and low risk of Following RFA, endoscopic surveillance is indicated to
adverse effects. monitor for recurrence, which can occur with an incidence
at 1 year ranging from 5% to 25%.29,30 Currently, there are
RADIOFREQUENCY ABLATION no consensus recommendations regarding surveillance
interval in postablation patients. Some experts recommend
RFA destroys tissue through a heat-energy system applied surveillance endoscopy every 3 months for the first year,
to the esophageal mucosa. Mucosal ablation is performed every 6 months for the second year, and then annually.30,31
under endoscopic guidance followed by immediate
débridement of the ablated area (Fig. 37.1). The ablation CRYOTHERAPY
treatment is then immediately repeated in the same area, Cryotherapy is an ablative technique that causes tissue
so that there is full treatment of the Barrett segment all destruction by application of liquid nitrogen or carbon
within one endoscopy session. The depth of ablation is dioxide gas (Fig. 37.2). Small areas can be treated (2 to
typically between 500 and 1000 µm, depending on the 3 cm) while covering approximately one-third or one-half
Endoscopic Management of High-Grade Dysplasia and Superficial Esophageal Carcinoma CHAPTER 37 385
A B
C D
FIGURE 37.1 (A) Long-segment circumferential Barrett esophagus with dysplasia/intramucosal cancer. (B) Endoscopic view with narrow
band imaging (Olympus). (C) Treatment with the HALO 90 (Medtronic) device. (D) Endoscopic appearance postradiofrequency ablation
treatment, new Z-line. (E) Retroflex view of neosquamous lining.
of the luminal circumference with each application. 87% had CE of all dysplasia.33 The most common adverse
Multiple areas can be treated in one endoscopic session. events reported included strictures in 3% that responded
On average, three to four endoscopies are needed to to endoscopic dilation, and chest pain in 2%, managed
completely ablate a long segment of disease, and the on an outpatient basis. With regard to cryoablation for
procedures can be performed approximately every 6 to early cancer, a multicenter retrospective study of liquid
8 weeks.31,32 nitrogen for esophageal cancer was published in 2010.
There are no randomized controlled studies assessing CE of cT1a tumors occurred in 18 of 24 (75%) patients.
the efficacy of cryotherapy in the treatment of dysplastic BE. For cT1b (submucosal) tumors, CE was seen in four of six
In a multicenter retrospective study of liquid nitrogen in patients (60%) with a mean follow-up of 11.8 months.33
patients with BE-HGD, 97% of patients had CE of HGD and Caveats must be made that these cancers were clinically
386 SECTION I Esophagus and Hernia
A B
C D
FIGURE 37.2 (A and B) Endoscopic appearance of nodular dysplastic Barrett esophagus with intramucosal cancer. (C) Cryotherapy
application. (D) Appearance immediately after cryotherapy.
staged with the inherent deficiencies of EUS staging, are often adequately managed by endoscopic ablation
and because ablation was chosen over ER no tissue was therapies alone, but there are times when a strategy of
provided to pathology for analysis. Overall, it seems that resection may be more definitive, cheaper, and effective.
cryotherapy has a bit better pain profile and may be more Short segments of flat HGD are amenable to complete ER
durable in applications where the mucosa is not flat. As with a low risk of complication.44 This is especially true
example, a single-center retrospective cohort study of 121 when the area is not circumferential (circumferential
patients with Barrett with dysplasia or intramucosal cancer, resections will cause serious stricture issues).45
16 patients who failed RFA14 or had recurrent dysplasia2 Invasive lesions amenable to EMR must be superfi-
underwent cryotherapy as salvage. After cryotherapy, 75% cial (T1) and small enough to be completely resected
of patients had CE-D and 31% had CE-IM.34 endoscopically. Patients with the lowest risk for regional
or systemic disease will have lesions that are less than
ENDOSCOPIC RESECTION 2 cm in maximal dimension. Well-differentiated mucosal
ER is currently the leading therapy for nodular BE and adenocarcinomas are associated with very low rates of
suspected early esophageal tumors.35–43 This procedure lymph node metastases (<3%) and can be managed
originally gained acceptance as a treatment for polyps and with endoscopic therapies, whereas tumors invading the
early-staged cancers within the colon. It later translated submucosa may have a substantial risk of lymph node
into treatment for esophageal squamous cell carcinoma metastases (in excess of 20%) and should be referred
within Asian medical centers. Now it is used throughout for esophagectomy.46
the world after groups in Germany, the Netherlands, and Step by step ER (Fig. 37.3):
North America have shown that this modality is a preferable 1. Procedure setup and setting. Patients are advised to remain
treatment option over surgery for HGD and early EAC.37–40 nothing by mouth (NPO) overnight and preferably stay
In patients diagnosed with HGD, the main goal of ER on a liquid diet the evening before. Performing this
is to resect visible lesions within the esophagus. Solitary procedure with food in the stomach is inconvenient and
nodules, areas of nodularity, and superficial ulcers are all perhaps incurs a bit more risk. The appropriate setting
high-risk areas for concurrent, unrecognized carcinoma is in a monitored suite with full anesthesia capability.
or progression to invasive disease. Areas of flat HGD We perform the procedure with monitored sedation in
Endoscopic Management of High-Grade Dysplasia and Superficial Esophageal Carcinoma CHAPTER 37 387
A B
FIGURE 37.3 (A) Endoscopic appearance of an early adenocarcinoma. (B) Cautery marks the margins of resection. (C) Immediate
postresection appearance.
most cases, but it is not absolutely imperative that the 4. Choose a method of resection. Multiple methods of EMR
patient is under a general anesthetic. Avoiding cough, or endoscopic submucosal dissection (ESD) are cur-
hiccoughs, and excess patient movement does help at rently used. EMR with a cap and snare technique or
the most critical point of the case when the tissue is mucosal banding followed by a cauterized snare will
acquired into a cap or band. resect into the submucosal layer of the esophagus,
2. Identification of the lesion. High-definition equipment and the procedure is facile after a relatively short
is mandatory. The endoscopy equipment is critical learning curve. Smaller lesions and selected lesions that
to the outcome of therapy. Techniques that improve are up to 2 cm can be removed entirely en bloc with
topographic visualization are encouraged because early suction cap techniques. Larger lesions resected with
lesions around the GEJ/cardia can be easily overlooked. an EMR-cap typically require a “piecemeal” approach
Endoscopists must be experienced at detecting early in which the lesion is removed by multiple applica-
esophageal/gastroesophageal pathology. tions of the cap and snare to include overlapping
3. Outline the area of resection. The area to be removed should areas, ultimately resulting in a complete resection.
be marked prior to introducing a cap for therapy. We are In cases where a lesion is tough to lift or is a little
frequently surprised at how an obvious lesion becomes larger, we will often use a larger oblique cap as these
difficult to see once a scope with a cap partially obstructs are capable of handling larger tissue areas. Another
our view in the esophagus. Chromoendoscopy of some alternative is ESD. This more aggressive approach
sort can facilitate marking the lesion with adequate has the potential to achieve an en bloc resection for
margins. The endoscopist may also choose to place larger or deeper lesions; however, ESD skills are more
marks (cautery, ink, or clips) on the tissue to be resected difficult to master and there are more frequent reports
that will orient the tissue margins for the clinician and of complications, such as perforation or bleeding. For
pathologist. This is probably only relevant in situations this reason it is performed only in selected centers
in which there is a single resection because piecemeal where specialized training has been performed and
resections render peripheral margins interpretable in there is adequate system support to increase the acuity
most situations. of care.
388 SECTION I Esophagus and Hernia
Retrieve all of the tissue samples and grossly examine 2% to 3% for a T1a lesion without LVI to 60%+ for T1b
the specimen to confirm complete capture and adequate lesions with LVI.38,40 Size of tumor and differentiation are
margins. Take more tissue around the margins with other factors that have been shown to be independent
additional ER if necessary to achieve a complete resec- prognostic variables in some studies, in which lesions
tion. Pinning the tissue to cork will help the pathologist less than 2 cm and well to moderately differentiated
to orient the deepest margin; omitting this step allows are less likely to harbor concurrent adenopathy.38,40
the tissue to invert its shape into a little mushroom According to some reports it is possible to identify
button that may lead to misinterpretation of the deep patients with early submucosal disease at low risk for
margin. nodal disease that may be adequately treated by ER
5. Postresection management. Patients are discharged home alone.49 Given that there is still some question about
on the same day unless there are comorbidities or interobserver variability regarding submucosal invasion
complications that direct otherwise. We typically allow and the presence of LVI even among expert pathologists,
full liquid diet on the first day, then liberalize the diet selecting appropriate patients who can avoid resection
when symptoms allow, starting on a soft diet for the should be done only at centers of excellence and in
first few days after the procedure. Patients and caregiv- the presence of a multidisciplinary group.
ers need to be educated on expected outcomes and 7. Follow-up. It has been our practice to perform endoscopy
potential signs of complications. Expected short-term every 2 to 3 months while in the process of ablating
outcomes will include mild to moderate chest discomfort residual metaplasia/dysplasia, followed by increasing
and odynophagia for a few days after resection so we intervals between procedures depending on individual
may provide a prescription compound that includes findings. The need for cross-sectional imaging is debated
an ionic binder to help heal the iatrogenic ulcer and in small T1a lesions given the low risk of regional or
a local anesthetic to alleviate symptoms. Narcotics are distant metastasis, and it is not at all indicated for
not usually required; in our experience, GEJ ER is more patients with dysplasia only. Patients at higher risk for
easily tolerated than a high to mid-esophageal lesion. regional and distant disease such as those with deeper,
Mild to moderate dysphagia can be a frequent complaint or larger lesions, or those patients with LVI who have
associated with healing but most often subsides after opted for endoscopic treatment alone based on risk for
6 to 8 weeks. Most patients are already on an acid esophageal resection should undergo imaging every 4
reduction medication, so we recommend continuing to 6 months and consider EUS to screen for regional
with a high-dose PPI for at least 2 weeks after the ER, disease.
then resuming the previous dose that the patient was
taking. If not on a PPI previously, we will typically keep
a patient on them at least until the next endoscopic ENDOSCOPIC RESECTION, MUCOSAL
evaluation. Patients with lower EAC with or without ABLATION, OR BOTH?
Barrett are typically maintained on PPIs indefinitely,
but the evidence for this practice after eradication of After initial endoscopic therapy directed at nodular
IM is lacking. areas, any remaining areas of metaplasia/dysplasia should
Complications specific to EMR include perforation, undergo ablation with the goal of total eradication. Past
an event that is seen in less than 0.1% of cases using reports indicated that untreated segments increased the
most cap techniques. This is in contrast to reports of risk of metachronous lesions to as high as 30%.43,44 Combin-
perforation upward of 40% for aggressive ESD proce- ing EMR with ablation to eradicate residual intestinal
dures. Fortunately almost all perforations associated metaplasia is both safe and efficacious. Pouw et al. reported
with ER can be managed nonoperatively. Bleeding their experience with 23 patients who underwent EMR for
risk is approximately 2% for EMR. 47 Stricture may visible lesions; 16 patients had early cancer and 7 patients
occur, and this depends on previous pathology in the had BE-HGD. In this study, RFA was performed at least 6
individual patient and the extent of circumferential weeks after the EMR. CE-neoplasia was achieved in 100%
resection. Removing more than 50% circumference of patients (median follow-up 22 months) and CE-IM in
significantly increases the risk of stricture.48 Complete 88% of patients.42
circumferential resections can be performed, but we Another retrospective study reported on the treatment
typically like to stage procedures that will require circum- of 65 patients with EMR and RFA for nodular disease and
ferential resections into several episodes. Single-session an additional 104 patients treated with RFA alone for flat
circumferential resection is associated with healing to segments of Barrett disease. There were no significant
high-grade stenosis; therefore stenting at the time of differences in eradication of dysplasia and intramucosal
resection should be considered.45,48 carcinoma between the two groups. The complication
6. Individualizing therapy based on interpretation of pathol- rates were also similar, including strictures that occurred
ogy. Lesions that are treated successfully with EMR in 4.6% of the EMR-before-RFA patient group and in 7.7%
are most often limited to the mucosa. Submucosal of patients in the RFA only group.43 This study suggests
invasion increases the risk of lymph node invasion that similar results may be achieved when ER is reserved
and/or cancer related events of recurrence or death. for dysplasia-associated lesions that are visible while RFA
Lymphovascular invasion (LVI) is the most important alone may suffice for flat pathology.
prognostic determinant of outcome for resected early- Of course there is the possibility that one could resect
stage cancer.38 In current surgical literature the risk of the entire area of pathology in the esophagus using ER
nodal involvement has been estimated to increase from techniques. This may be favorable in situations in which
Endoscopic Management of High-Grade Dysplasia and Superficial Esophageal Carcinoma CHAPTER 37 389
there is limited, noncircumferential disease. In contrast, 7. Goldblum JR. Barrett’s esophagus and Barrett’s-related dysplasia.
combined EMR with ablation may be the preferred Mod Pathol. 2003;16:316-324.
8. Malfertheiner P, Nocon M, Vieth M, et al. Evolution of gastro-
approach over stepwise radical ER for the treatment of oesophageal reflux disease over 5 years under routine medical
Barrett associated with early cancer. In a Dutch multicenter care—the ProGERD study. Aliment Pharmacol Ther. 2012;35(1):154-164.
study, patients with a BE segment less than or equal to 9. Corley DA, Levin HR, Habel LA, Weiss NS, Buffler PA. Surveillance
5 cm containing HGD/early cancer were randomized and survival in Barrett’s adenocarcinoma: a population based study.
Gastroenterology. 2002;122:633-640.
to stepwise radical endoscopic resection (SRER) or ER 10. Fountoulakis A, Zafirellis K, Donlan K, et al. Effect of surveillance
followed by RFA. Although both groups achieved excellent of Barrett’s oesophagus on clinical outcome of oesophageal cancer.
(>90%) and comparable rates of CE-neoplasia and CE-IM, Br J Surg. 2004;91:997-1003.
those in the SRER group developed a significantly higher 11. Fernando HC, Murthy SC, Hofstetter W, et al. The Society of Thoracic
number of strictures requiring endoscopic dilation (88% Surgeons practice guideline series: guidelines for the management
of Barrett’s esophagus with high-grade dysplasia. Ann Thorac Surg.
in the SRER group compared with 14% in the EMR-RFA 2009;87:1993-2002.
group; P < .001).42 12. Phoa KN, van Vilsteren FG, Weusten BL, et al. Radiofrequency
Endoscopic treatment compares favorably with esopha- ablation vs endoscopic surveillance for patients with Barrett
gectomy as well. A retrospective comparison of endoscopic esophagus and low-grade dysplasia: a randomized clinical trial.
JAMA. 2014;311(12):1209-1217.
therapy performed in 40 patients (22 with HGD and 18 13. Rastogi A, Puli S, El-Serag HB, Bansal A, Wani S, Sharma P. Inci-
with intramucosal cancer [IMC]) with esophagectomy in dence of esophageal adenocarcinoma in patients with Barrett’s
61 patients (13 with HGD and 48 with IMC) reported that esophagus and high-grade dysplasia: a meta-analysis. Gastrointest
there was no difference in survival between the two groups Endosc. 2008;67:394-398.
(94% at 3 years). However, compared with esophagectomy, 14. Shaheen NJ, Sharma P, Overholt BF, et al. Radiofrequency abla-
tion in Barrett’s esophagus with dysplasia. N Engl J Med. 2009;360:
endoscopic therapy was associated with significantly lower 2277-2288.
morbidity (39% vs. 0%; P < .0001).46 One should note that 15. Konda VJ, Ross AS, Ferguson MK, et al. Is the risk of concomitant
endoscopic therapy was not finite in this study, consisting invasive esophageal cancer in high-grade dysplasia in Barrett’s
of 102 ERs and 79 ablations in 40 patients. Nonetheless, esophagus overestimated? Clin Gastroenterol Hepatol. 2008;6:159-164.
16. Gupta N, Gaddam S, Wani SB, Bansal A, Rastogi A, Sharma P. Longer
cost comparisons still seem to favor endoscopic treatment inspection time is associated with increase detection of high-grade
over esophageal resection in most situations.50 dysplasia and esophageal adenocarcinoma in Barrett’s esophagus.
In summary, EMR followed by ablation is an effective Gastrointest Endosc. 2012;76:531-538.
treatment modality for early cancer arising in the setting 17. Panossian AM, Raimondo M, Wolfsen HC. State of the art in the
of BE. However, these techniques are best performed at endoscopic imaging and ablation of Barrett’s esophagus. Dig Liver
Dis. 2011;43:365-373.
high-volume referral centers by experienced endoscopists. 18. Thosani N, Singh H, Kapadia A, et al. Diagnostic accuracy of EUS
The outcomes may not apply to general practices. in differentiating mucosal versus submucosal invasion of superficial
esophageal cancers: a systematic review and meta-analysis. Gastrointest
Endosc. 2012;75(2):242-253.
CONCLUSION 19. May A, Günter E, Roth F, et al. Accuracy of staging in early oesopha-
geal cancer using high resolution endoscopy and high resolution
Endoscopic therapies have been shown to be effective endosonography: a comparative, prospective, and blinded trial. Gut.
and safe in the treatment of patients with BE-HGD and 2004;53:634-640.
early esophageal cancer while preserving the esophagus. 20. Young PE, Gentry AB, Acosta RD, Greenwald BD, Riddle M. Endo-
Attempts at total eradication of all dysplasia/metaplasia scopic ultrasound does not accurately stage early adenocarcinoma
or high-grade dysplasia of the esophagus. Clin Gastroenterol Hepatol.
followed by close surveillance is recommended after 2010;8:1037-1041.
endoscopic therapy because of the risk of recurrence. 21. Shami VM, Villaverde A, Stearns L, et al. Clinical impact of con-
Given the complexities in the evaluation and management ventional endosonography and endoscopic ultrasound-guided
of these patients, they are best managed by a multidisci- fine needle aspiration in the assessment of patients with Barrett’s
esophagus and high-grade dysplasia or intramucosal carcinoma
plinary team in a tertiary referral center with expertise who have been referred for endoscopic ablation therapy. Endoscopy.
in esophageal diseases. 2006;38:157-161.
22. Moss A, Bourke MJ, Hourigan LF, et al. Endoscopic resection for
Barrett’s high-grade dysplasia and early esophageal adenocarcinoma:
REFERENCES an essential staging procedure with long-term therapeutic benefit.
Am J Gastroenterol. 2010;105:1276-1283.
1. Spechler SJ, Sharma P, Souza RF, et al. American Gastroenterological 23. Peters FP, Brakenhoff KP, Curvers WL, et al. Histologic evaluation
Association Medical Position Statement on the management of of resection specimens obtained at 293 endoscopic resections in
Barrett’s esophagus. Gastroenterology. 2011;140:1084-1091. Barrett’s esophagus. Gastrointest Endosc. 2008;67:604-609.
2. Sampliner RE. Practice guidelines on the diagnosis, surveillance and 24. Shaheen NJ, Overholt BF, Sampliner RE, et al. Durability of radiofre-
therapy of Barrett’s esophagus. Am J Gastroenterol. 1998;93:1028-1031. quency ablation in Barrett’s esophagus with dysplasia. Gastroenterology.
3. Jung KW, Talley NJ, Romero Y, et al. Epidemiology and natural 2011;141:460-468.
history of intestinal metaplasia of the gastroesophageal junction 25. Ganz RA, Overholt BF, Sharma VK, et al. Circumferential ablation
and Barrett’s esophagus: a population-based study. Am J Gastroenterol. of Barrett’s esophagus that contains high-grade dysplasia: a U.S.
2011;106:1447-1455. multicenter registry. Gastrointest Endosc. 2008;68:35-40.
4. Hvid-Jensen F, Pedersen L, Drewes AM, Sørensen HT, Funch-Jensen 26. Fleischer DE, Overholt BF, Sharma VK, et al. Endoscopic radiofre-
P. Incidence of adenocarcinoma among patients with Barrett’s quency ablation for Barrett’s esophagus: 5-year outcomes from a
esophagus. N Engl J Med. 2011;365:1375-1383. prospective multicenter trial. Endoscopy. 2010;42:781-789.
5. Hur C, Miller M, Kong CY, et al. Trends in esophageal adenocarci- 27. Bulsiewicz WJ, Kim HP, Dellon ES, et al. Safety and efficacy of
noma incidence and mortality. Cancer. 2013;119(6):1149-1158. endoscopic mucosal therapy with radiofrequency ablation for
6. Yates M, Cheong E, Luben R, et al. Body mass index, smoking, patients with neoplastic Barrett’s esophagus. Clin Gastroenterol Hepatol.
and alcohol and risks of Barrett’s esophagus and esophageal 2013;11(6):636-642.
adenocarcinoma: a UK prospective cohort study. Dig Dis Sci. 28. Van Vilsteren FG, Alvarez Herrero L, Pouw RE, et al. Predictive factors
2014;59(7):1552-1559. for initial treatment response after circumferential radiofrequency
390 SECTION I Esophagus and Hernia
ablation for Barrett’s esophagus with early neoplasia: a prospective 40. Alvarez Herrero L, Pouw RE, van Vilsteren FG, et al. Risk of lymph
multicenter study. Endoscopy. 2013;45:516-525. node metastasis associated with deeper invasion by early adenocar-
29. Orman ES, Kim HP, Bulsiewicz WJ, et al. Intestinal metaplasia recurs cinoma of the esophagus and cardia: study based on endoscopic
infrequently in patients successfully treated for Barrett’s esophagus resection specimens. Endoscopy. 2010;42(12):1030-1036.
with radiofrequency ablation. Am J Gastroenterol. 2013;108(2):187-195, 41. May A, Gossner L, Pech O. Local endoscopic therapy for intraepi-
[quiz 196]. thelial high-grade neoplasia and early adenocarcinoma in Barrett’s
30. Titi M, Overhiser A, Ulusarac O, et al. Development of subsquamous esophagus: acute-phase and intermediate results of a new treatment
high-grade dysplasia and adenocarcinoma after successful radiofre- approach. Eur J Gastroenterol Hepatol. 2002;14:1085-1091.
quency ablation of Barrett’s esophagus. Gastroenterology. 2012;143: 42. Pouw RE, Wirths K, Eisendrath P, et al. Efficacy of radiofrequency
564-566. ablation combined with endoscopic resection for Barrett’s esophagus
31. Greenwald BD, Dumot JA. Cryotherapy for Barrett’s esophagus and with early neoplasia. Clin Gastroenterol Hepatol. 2010;8:23-29.
esophageal cancer. Curr Opin Gastroenterol. 2011;27:363-367. 43. Kim HP, Bulsiewicz WJ, Cotton CC, et al. Focal endoscopic mucosal
32. Shaheen NJ, Greenwald BD, Peery AF, et al. Safety and efficacy of resection before radiofrequency ablation is equally effective and safe
endoscopic spray cryotherapy for Barrett’s esophagus with high-grade compared with radiofrequency ablation alone for the eradication
dysplasia. Gastrointest Endosc. 2010;71:680-685. of Barrett’s esophagus with advanced neoplasia. Gastrointest Endosc.
33. Greenwald BD, Dumot JA, Abrams JA, et al. Endoscopic spray 2012;76:733-739.
cryotherapy for esophageal cancer: safety and efficacy. Gastrointest 44. Van Vilsteren FG, Pouw RE, Seewald S, et al. Stepwise radical
Endosc. 2010;71:686-693. endoscopic resection versus radiofrequency ablation for Barrett’s
34. Sengupta N, Ketwaroo GA, Bak DM, et al. Salvage cryotherapy oesophagus with high-grade dysplasia or early cancer: a multicenter
after failed radiofrequency ablation for Barrett’s esophagus-related randomized trial. Gut. 2011;60:765-773.
dysplasia is safe and effective. Gastrointest Endosc. 2015;82:443-448. 45. Conio M, Fisher DA, Blanchi S, Ruggeri C, Filiberti R, Siersema
35. Pech O, Bollschweiler E, Manner H, Leers J, Ell C, Holscher AH. PD. One-step circumferential endoscopic mucosal cap resection of
Comparison between endoscopic and surgical resection of mucosal Barrett’s esophagus with early neoplasia. Clin Res Hepatol Gastroenterol.
esophageal adenocarcinoma in Barrett’s esophagus at two high- 2014;38(1):81-91.
volume centers. Ann Surg. 2011;254(1):67-72. 46. Zehetner J, DeMeester SR, Hagen JA, et al. Endoscopic resection and
36. Pech O, Behrens A, May A, et al. Long-term results and risk factor ablation versus esophagectomy for high-grade dysplasia and intramu-
analysis for recurrence after curative endoscopic therapy in 349 cosal adenocarcinoma. J Thorac Cardiovasc Surg. 2011;141:39-47.
patients with high-grade intraepithelial neoplasia and mucosal 47. Pech O, May A, Manner H, et al. Long-term efficacy and safety of
adenocarcinoma in Barrett’s oesophagus. Gut. 2008;57(9):1200-1206. endoscopic resection for patients with mucosal adenocarcinoma of
37. Manner H, Pech O, Heldmann Y, et al. Efficacy, safety, and long-term the esophagus. Gastroenterology. 2014;146(3):652-660.
results of endoscopic treatment for early-stage adenocarcinoma of 48. Hanaoka N, Ishihara R, Uedo N, et al. Refractory strictures despite
the esophagus with low-risk sm1 invasion. Clin Gastroenterol Hepatol. steroid injection after esophageal endoscopic resection. Endosc Int
2013;11(6):630-635. [quiz e45]. Open. 2016;4(3):E354-E359.
38. Lee L, Ronellenfitsch U, Hofstetter WL, et al. Predicting lymph 49. Manner H, May A, Pech O, et al. Early Barrett’s carcinoma with “low-
node metastases in early esophageal adenocarcinoma using a simple risk” submucosal invasion: long-term results of endoscopic resection
scoring system. J Am Coll Surg. 2013;217(2):191-199. with a curative intent. Am J Gastroenterol. 2008;103(10):2589-2597.
39. Pouw RE, van Vilsteren FG, Peters FP, et al. Randomized trial 50. Pohl H, Sonnenberg A, Strobel S, Eckardt A, Rösch T. Endoscopic
on endoscopic resection-cap versus multiband mucosectomy for versus surgical therapy for early cancer in Barrett’s esophagus: a
piecemeal endoscopic resection of early Barrett’s neoplasia. Gastrointest decision analysis. Gastrointest Endosc. 2009;70(4):623-631.
Endosc. 2011;74(1):35-43.
CHAPTER
Multimodality Therapy in the Management of Locally
Advanced Esophageal Cancer 38
Jonathan Cools-Lartigue
| Lorenzo Ferri
E
sophageal cancer (EC) remains a devastating malig- theoretic advantages have been put forward favoring
nancy with a low rate of cure. Results in patients both approaches.
with early-stage disease (stage I to II) remain more Given the observation that a majority of patients present
promising, with long-term survival rates between 60% and with locally advanced disease, preoperative chemotherapy
90%.1–4 Unfortunately the majority of patients present could serve to downstage seemingly unresectable lesions,
with locally advanced or advanced disease.1–5 This is rendering them amenable to complete oncologic (R0)
associated with high rates of systemic recurrence and has resection. Furthermore, in the preoperative setting, optimal
demonstrated poor outcomes when treated with surgery drug delivery may be achieved given an intact blood
alone. Thus while surgery continues to play a central role supply. Finally, the administration of chemotherapy in
in curative intent therapy for EC, a significant number the preoperative setting provides a unique opportunity
of patients present with a burden of disease precluding to observe the clinical efficacy of the drug regimen in
upfront surgery.1 Accordingly, the modern approach to question. Assessment of tumor response has important
the majority of patients with locally advanced disease is prognostic significance and by identifying patients who do
the implementation of multimodality treatment strategies.1 not respond, alternate treatment strategies can be selected
Complementary therapeutic modalities include chemo- in lieu of an initial ineffective and morbid approach.6
therapy and radiation therapy, which have demonstrated Proponents for the postoperative approach highlight the
efficacy with respect to improved local and distant control.1 possibility of overtreatment based on inaccurate staging.
However, considerable variability in the application of Pathologic analysis of surgical specimens provides ultimate
multimodality regimens is observed in the literature to staging and can thus direct treatment to patients with risk
date.1 This includes the application of neoadjuvant or factors for recurrence who stand to benefit the most.7,8
adjuvant chemotherapy alone or in conjunction with Finally, controversy does exist regarding the optimum
radiation therapy, as well as the exact drug regimens and chemotherapeutic regimen. The majority of randomized
radiation doses employed.1 Furthermore, considerable controlled trials (RCTs) to date have employed doublet
variability in the surgical approach to esophagectomy is platinum–based therapy, but triplet regimens have been
evident in the literature. In keeping with these observa- employed and may demonstrate improved efficacy with
tions, controversy regarding the optimal regimen persists. regard to tumor regression and survival.
What remains clear is that a meticulous approach to this Based on these postulates, a number of important
vulnerable patient population with respect to staging, questions can be formulated. First, does perioperative che-
surgical technique, and adjuvant local and systemic thera- motherapy improve survival in patients with esophageal cancer
pies is necessary to provide the best possible outcome.1 and, if so, does this depend on tumor histology? Second, if so, is
Herein, the literature providing the rationale for modern chemotherapy more effective when administered in the preoperative
multimodality approaches is reviewed, and the advantages or postoperative setting? Third, do triplet regimens confer an
and disadvantages of each are laid out. advantage over doublets?
ABSTRACT
The management of locally advanced esophageal cancer
remains complex and necessitates a multimodality approach
in order to achieve favorable survival outcomes. Given
the evidence to date, several generalizations regarding
effective local and systemic treatment strategies can be
drawn. With respect to local control, excellent surgery
remains paramount as evidenced by improved outcomes
associated with R0 resection. En bloc esophagectomy has
been shown to provide excellent results, and its impact with
respect to survival in large randomized trials is currently
under elucidation in the NeXT, TOP GEAR, and Neo
AEGIS trials. The inability to achieve R0 resection in a
significant proportion of patients mandates additional
local therapy. For patients with squamous cell carcinoma,
neoadjuvant chemoradiotherapy (CRT) provides excellent
results and represents standard therapy. With respect to
patients with esophageal adenocarcinoma, the benefit
of neoadjuvant CRT is less pronounced but remains an
acceptable standard. Excellent results have been reported
with docetaxel-based triplets, which appear to be borne out
in contemporary meta-analysis. Accordingly, neoadjuvant
chemotherapy alone remains an acceptable standard
in this patient population. In the present chapter, the
literature to date is extensively reviewed, and the up-to-date
standardized treatment strategies are outlined.
KEYWORDS
Esophageal adenocarcinoma, Squamous cell carcinoma,
chemoradiation, chemotherapy, en bloc esophagectomy
392 SECTION I Esophagus and Hernia
analysis demonstrates improved survival outcomes in in the previous studies, median and 2-year survival times
patients who demonstrate a response to the preoperative were improved (chemotherapy vs. surgery, 42.2 months
regimen.10,11,13,14,19 vs. 13.8 months; P = .008 and 59% vs. 33%, respectively).
Roth et al. demonstrated improved survival in squamous Ancona et al. similarly randomized patients with SCC to two
cell carcinoma (SCC) patients who demonstrated a major cycles of preoperative cisplatin and 5FU or curative intent
(47%) or complete (5%) response to vinblastine-, cisplatin-, surgery alone.19 Combined complete and major response
and bleomycin-based chemotherapy (median survival, 20 rates of resected tumors in the neoadjuvant arm were 40%,
months vs. 6 months; P = .008). Patients who responded to with a complete response rate of 12.8%. No difference in
preoperative therapy also fared better than patients who median survival was noted on an intention to treat basis
received surgery alone.13 Schlag similarly demonstrated (24 and 25 months for surgery alone vs. neoadjuvant
significant increases in survival for patients with SCC chemotherapy, respectively).19 In the 40% of patients
who demonstrated a response (minor 12%, major 32%, who demonstrated a major response to chemotherapy,
complete 6%) to 3 cycles of cisplatin and 5-fluorouracil a significant improvement with respect to median (53
(5FU)-based chemotherapy.14 Schlag et al. randomized months) and 3-year (74%) and 5-year (60%) survival was
SCC patients to receive 2 cycles of cisplatin and 5FU-based observed compared with patients undergoing surgery
chemotherapy in the preoperative setting or to curative alone (28 months, 46%, 26%; P = .01) and nonresponders
intent surgery alone. The authors demonstrated reduced to chemotherapy (19 months, 38%, 19%; P < .05). In
recurrence rates, predominantly as a result of improved keeping with this theme, the survival benefit was most
locoregional control in patients who received chemo- pronounced in the 12.8% of patients who demonstrated
therapy compared with surgery alone. This finding likely a complete response, leading the authors to conclude
relates to higher R0 resection rates in chemotherapy-treated that pathologic response is a significant determinant of
patients (67% vs. 35%; P = .003). This did not translate long-term outcome in addition to R0 resection.19
into an overall survival (OS) advantage (median and 2-year Kelsen et al. randomized patients with both SCC and
survival in chemotherapy vs. surgery alone, 16.8 months esophageal adenocarcinoma (EAC) in roughly equal
vs. 13 months, and 44% vs. 31%, respectively; P = .17).14 proportions to three cycles of preoperative cisplatin and
However, in patients who demonstrated a response, as 5FU or curative intent surgery alone.10 Complete response
Multimodality Therapy in the Management of Locally Advanced Esophageal Cancer CHAPTER 38 393
rate in patients receiving chemotherapy was 2.5%, with a and 5FU versus surgery alone.15 Patients randomized to
major objective response observed in 19%.10 In keeping preoperative therapy demonstrated improved outcome
with previous studies, survival was only improved in patients with respect to R0 resection rate (87% vs. 74%; P = .04)
who demonstrated a major response (response vs. no and OS (5-year survival chemotherapy vs. surgery alone
response, hazard ratio [HR], 2.83; 95% confidence interval 38% vs. 24%; P < .05).15 In the study by Cunningham et al.,
[CI], 1.84 to 4.35; P < .001).10 patients with gastric and lower esophageal/GE junction
Collectively, these data suggest that effective chemo- tumors (25%) were randomized to receive perioperative
therapy, as measured by an objective regression response, chemotherapy with 5FU, cisplatin, and epirubicin. In
is protective and confers a significant survival advantage, patients receiving perioperative therapy, an improvement
in addition to R0 resection in patients with both SCC in R0 (CS vs. surgery alone, 79.3% vs. 70.3%; P = .03)
and EAC compared with surgery alone. Furthermore, resection rate and a tendency for smaller tumors and less
effective chemotherapy may improve R0 resection rates advanced nodal disease was noted.9 In keeping with these
and thus contribute to improved survival through improved findings, improved OS was noted in the chemotherapy
locoregional control. arm compared with surgery alone (HR, 0.75; 95% CI, 0.6
More recent studies, including the MRC/OE2, Boonstra, to 0.93; P = .009; 5-year survival, 36.3% vs. 23%).9
Ychou, and Cunningham (MAGIC) trials, were positive on Collectively, the data demonstrate that when given in the
an intent to treat basis, supporting the use of neoadjuvant perioperative setting, chemotherapy can effectively reduce
chemotherapy over surgery alone in patients with locally tumor burden, facilitate curative resection, and impart a
advanced esophageal cancer.9,12,15,17,18 In these studies, significant survival benefit in patients with locally advanced
histologies were commonly mixed. Therapeutic regimens SCC or EAC. With respect to whether efficacy differs
included cisplatin and 5FU-based doublets, with the excep- according to tumor histology, overall the data suggest
tion of the Cunningham (MAGIC) trial, which administered that chemotherapy provides a benefit in both SCC and
triplet therapy encompassing an anthracycline in addition EAC. The MRC/OE2 study specifically assessed treatment
to cisplatin and 5FU. All studies administered chemotherapy effects according to histology and found no evidence of
in the preoperative and postoperative periods, with the a difference in the two, with a significant reduction in
exception of the MRC study, where chemotherapy was both T and N stage overall.12,18 Furthermore, the benefits
administered in the preoperative setting only. of chemotherapy with respect to survival did not differ
In the MRC study and its subsequent follow-up (OE2), according to histology, with absolute 5-year survival rates
preoperative cisplatin and 5FU was associated with a in chemotherapy versus surgery alone of 22.6% versus
significant reduction in primary tumor size and regional 17.6% in EAC patients and 25.5% versus 17% in SCC
lymph node positivity, compared with specimens from patients, respectively.12,18 Thus effective chemotherapy may
untreated patients.12,18 This was associated with improved be considered in esophageal cancer patients in addition
R0 resection rates in patients subject to neoadjuvant to surgery, regardless of histology.
chemotherapy (60% vs. 54%, P < .001). Similarly, OS was
improved in patients subject to neoadjuvant treatment DO TRIPLET REGIMENS CONFER AN
(median and 2-year survival, neoadjuvant chemotherapy ADVANTAGE OVER DOUBLETS?
followed by surgery versus surgery alone: 16.8 months The studies mentioned thus far have predominantly
vs. 13.3 months and 43% vs. 34%, respectively; HR, 0.79; employed chemotherapy regimens composed of a
95% CI, 0.67 to 0.93; P = .004).12,18 In long-term follow-up doublet.7,8,10–12,14–19 These are typically composed of a
(OE2), patients subject to chemotherapy demonstrated platinum agent and 5FU. The rationale for this regimen
reduced T and N stage compared with patients receiving stems from the observation of significant response rates
surgery alone following pathologic analysis. Overall 5-year in conjunction with a good toxicity profile. The MAGIC
survival was 23% in neoadjuvant group versus 17.1% in trial suggests that treatment with a chemotherapy triplet
patients randomized to surgery alone (P < .001).12,18 via the addition of an anthracycline to the cisplatin and
Boonstra et al. conducted a clinical trial comparing peri- 5FU doublet is also highly effective.9 Proponents of triplet
operative chemotherapy using a regimen of cisplatin and therapy highlight the potential to improve response
etoposide in patients with SCC.17 Following chemotherapy, rates with acceptable toxicity.5,20–26 A recent meta-analysis
a partial response rate of 40% and complete response examined 21 randomized studies comparing a doublet
rate of 7% were observed. No difference in R0 resection to triplet regimen in patients with distal esophageal,
(71% vs. 57%) rate or lymph node positivity between esophagogastric, and gastric adenocarcinoma.20 Overall,
the two treatment arms was noted; however, significantly outcomes in 3475 patients were assessed. The authors
more patients in the surgery alone arm demonstrated demonstrated improved OS (HR, 0.9; 95% CI, 0.83 to
unresectable tumors or underwent an R2 resection. 17 0.97) and progression-free survival (PFS) (HR, 0.8; 95%
Accordingly, a significant survival advantage in patients CI, 0.69 to 0.93), favoring taxane-based triplet therapies.
receiving chemotherapy was noted (median, 2- and 5-year Fluoropyrimidine-based triplets were associated with an
survival CS vs. surgery alone, 16 months vs. 12 months, objective response rate (ORR) compared with doublets
42% vs. 30%, and 26% vs. 17%, respectively; P = .03).17 alone (RR, 1.25; 95% CI, 1.09 to 1.44). This benefit,
The trials by Ychou and Cunningham et al. included however, was associated with increased incidence of grade
patients with adenocarcinoma of the esophagus and 3 to 4 thrombocytopenia, infection, and mucositis by a
stomach.9,15 Ychou et al. randomized patients, of whom factor of approximately 2.20
75% harbored lower esophageal or gastroesophageal With respect to SCC, doublet therapies composed of
(GE) junction tumors to receive perioperative cisplatin cisplatin and 5FU have demonstrated effectiveness, as
394 SECTION I Esophagus and Hernia
demonstrated by the studies noted previously.7,8,10–12,14–19 investigating postoperative cisplatin and 5FU in patients
Similarly, triplet therapies comprising an additional taxane with SCC versus surgery alone. In patients who were
have also demonstrated efficacy.20–26 However, to date no ultimately found to harbor node positive disease, adjuvant
extensive randomized data exist specifically comparing chemotherapy was associated with a significant improve-
doublet to triplet regimens in patients with squamous ment in 5-year disease-free survival (DFS) compared with
histology. surgery alone (52% vs. 38%; P = .041). Conversely, there
Recently there has been much interest in taxane-based was a nonsignificant trend toward worse DFS outcomes in
triplets in the neoadjuvant setting, with excellent oncologic node negative patients receiving adjuvant chemotherapy
results from phase II studies employing docetaxel in both compared with surgery alone (70% vs. 76%; P = .433).7
major histologic subtypes.20–26 We have previously shown These seemingly contradictory results led to a follow-up
excellent tolerance and response (pathologic complete study comparing the same regimen given either pre- or
response [pCR] of 10%) in a phase II trial in patients postoperatively. In this study, the authors directly compared
with locally advanced (cT3 and/or N1) adenocarcinoma preoperative versus postoperative chemotherapy in 330
of the esophagus, GE junction, and stomach using a patients with locally advanced esophageal SCC. Patients
regimen of docetaxel, cisplatin, and 5FU (DCF).21 Long- received two cycles of cisplatin and 5FU, either preceding
term results demonstrated an impressive 5-year survival or following curative intent surgery.16 OS was significantly
of over 55%, despite a high burden of residual disease improved in patients who received preoperative com-
(median positive pathologic lymph nodes = 5).5 These pared with postoperative therapy, with 5-year survival of
results have been replicated in several German series, 55% versus 43% (P = .04).16 Overall response rates to
including a recent study employing a docetaxel-based chemotherapy were 38%, which translated into fewer
regimen, but replacing cisplatin with oxaliplatin (FLOT: preoperative-treated patients being found to harbor T4 or
5-Flourouracil, Leucovorin, Oxaliplatin, and Taxotere N+ tumors. Furthermore, significantly more patients who
[Docetaxel]) for esophagogastric adenocarcinoma with received neoadjuvant chemotherapy underwent curative
impressive pathologic response rates—20% pCR and (R0) resection (96% vs. 91%; P = .04). Patients in the
another 20%, with near pCR (<10% residual tumor).23 adjuvant arm experienced increased toxicity and reduced
Another German study employing a similar regimen completion (75% vs. 85%; P = .04) of therapy compared
demonstrated a pCR rate of 15% and a median OS of just with patients in the neoadjuvant arm. Overall, the results
over 4 years.24 These strong data set the foundation for a of the studies to date support the use of preoperative
large German multicenter phase three trial investigating chemotherapy with respect to compliance, assessment of
FLOT versus the regimen established by the MAGIC tumor response, and survival.16
trial (ClinicalTrials.gov identifier, NCT01216644), ECF
(epirubicin, cisplatin, and 5-fluorouracil regimen) in
patients with locally advanced esophagogastric adeno- NEOADJUVANT CHEMORADIATION
carcinoma. This study has completed accrual. Several VERSUS SURGERY ALONE
Japanese studies have supported the use of taxane-based
triplets, most notably docetaxel, in esophageal SCC.22,25,26 The studies looking at chemotherapy administration in
Indeed, impressive results from these studies have led to the perioperative setting demonstrated that patients with a
inclusion of DCF as one of three arms in the ongoing marked tumor response to therapy experienced improved
large Japanese multiinstitutional NExT (JCOG1109) trial survival outcomes.10,16,17,19 This benefit was observed when
investigating neoadjuvant chemotherapy (cisplatin/5FU tumor response was noted both within the primary tumor
or docetaxel/cisplatin/5FU) versus cisplatin/5FU with and regional lymph nodes.10,16,17,19 In an attempt to improve
concurrent radiotherapy.27 This study is well underway, local response and R0 resection rates, the addition of radia-
with preliminary results due in 2017. tion to chemotherapy in the preoperative setting has been
attempted and extensively studied. A useful framework
for assessing the added benefit of chemoradiation to
IS CHEMOTHERAPY MORE EFFECTIVE surgery alone is to look at its efficacy with respect to local/
WHEN ADMINISTERED IN THE PRE- OR regional control and any additional control of systemic
POSTOPERATIVE SETTING? recurrence. Furthermore, the toxicity of various treatment
Current data are conflicted regarding the use of postopera- regimens needs to be determined. Stated otherwise, does
tive chemotherapy alone in the adjuvant setting only.7,8 neoadjuvant chemoradiation improve local/regional control
The 1997 study by Ando et al. randomized patients with compared with surgery alone in patients with esophageal SCC
esophageal SCC to receive either curative intent surgery and EAC? Does it provide improved systemic control compared
alone or adjuvant chemotherapy with cisplatin and vin- with surgery alone? Can any benefits be achieved with acceptable
desine, delivered in two cycles.8 Patients predominantly toxicity? Along these lines, the bulk of evidence to date
harbored locally advanced disease; however, in this study has helped establish neoadjuvant chemoradiation as a
the exact rate of curative R0 resection was not clearly standard of care in the management of EC. Randomized
indicated. No survival benefit was observed with the addi- trials comparing chemoradiation (CRT) to surgery alone
tion of postoperative chemotherapy, although the efficacy are outlined in Table 38.2.28–40
of the regimen and uncertainty regarding the extent of Until the publication of the CROSS trial in 2008, the
resection diminish the generalizability of the study.8 bulk of the studies examining neoadjuvant CRT failed
In 2003, Ando and colleagues published a Japanese to demonstrate a significant survival benefit over surgery
Clinical Oncology Group multiinstitutional phase III trial alone with the exception of the Walsh and Tepper trials.
Multimodality Therapy in the Management of Locally Advanced Esophageal Cancer CHAPTER 38 395
However, while grossly negative, the additional trials were low, particularly in patients receiving surgery alone.33
listed provide important information concerning the As in previous studies, pCR was higher in SCC at 27%
effectiveness of various neoadjuvant CRT regimens with versus 9% in EAC patients, which was lower than response
respect to both local and distant control. rates observed in contemporary studies. No appreciable
difference in survival was noted between groups.33
DOES NEOADJUVANT CHEMORADIATION The CROSS trial is the largest positive trial performed
IMPROVE LOCAL/REGIONAL CONTROL COMPARED to date comparing neoadjuvant CRT to surgery alone and
WITH SURGERY ALONE IN PATIENTS WITH has established a standard therapy for both esophageal
ESOPHAGEAL SCC AND EAC? DOES IT PROVIDE SCC and adenocarcinoma in much of the West.35,36 The
authors employed a slightly different chemotherapeutic
IMPROVED SYSTEMIC CONTROL COMPARED WITH approach composed of a weekly regimen of relatively
SURGERY ALONE? low-dose taxane and carboplatin in conjunction with
With respect to local control, the evidence to date sug- concurrent radiation at a dose of 41.4 Gy. With respect to
gests that neoadjuvant CRT is beneficial.28,31–33,35–37,39,40 local control, in the adenocarcinoma arm a pCR rate of
In particular, the benefit is greater in SCC compared 25% was again noted compared with 49% in patients with
with adenocarcinoma. For example, the Nygaard study, SCC.35,36 In keeping with previous studies, R0 resection
conducted exclusively in patients with SCC, demonstrated was achieved in a greater proportion of patients receiving
a trend toward improved R0 resection rates with neo- neoadjuvant CRT compared with surgery alone (92% vs.
adjuvant CRT.28 This translated into improved survival 69%).35,36 In addition, significantly more patients undergo-
compared with patients who received no radiation at ing surgery alone were found to harbor metastatic lymph
all.28 In the Le Prise study, also exclusively conducted nodes compared with patients receiving neoadjuvant CRT
in SCC patients, a significant downstaging effect can be (75% vs. 31%), despite comparable preoperative clinical
inferred based on the finding of significantly more T3 staging.35,36 Finally, on long-term follow-up, a significant
and T4 tumors in the non-CRT group.29 However, no reduction in locoregional recurrence was noted in patients
difference in R0 resection and a very low pCR rate were with either adenocarcinoma or SCC following neoadjuvant
observed, which may have contributed to the negative CRT compared with surgery alone (22% vs. 38%; HR,
nature of the study overall. Similarly, a relatively low dose 0.45; 95% CI, 0.3 to 0.66; P < .001), thus supporting the
of chemotherapy and radiation (20 Gy) were used.29 The locoregional benefit of neoadjuvant CRT.35,36
Walsh study, which was a positive study in patients with Bass and colleagues demonstrated analogous findings
adenocarcinoma, demonstrated a pCR rate of 25% and a in 211 esophageal cancer (SCC and EAC) patients with
significant downstaging effect, with 42% of patients treated respect to pCR and improved locoregional control fol-
with CRT found to harbor positive lymph nodes at the lowing neoadjuvant CRT.38 For patients with EAC pCR
time of surgery, compared with 82% of patients in the rates were lower than for patients with SCC, again at 25%
surgery alone arm (P < .001).30 The Bosset study, again a versus 31%.38 Downstaging of mediastinal disease following
negative study in patients with SCC, demonstrated a 26% neoadjuvant CRT was inferred based on a 64% rate of
pCR and significantly fewer lymph node metastases at the lymph node positivity in patients treated with surgery
time of surgery (25% vs. 57%) in patients who received alone versus 29% in patients undergoing neoadjuvant
CRT compared with surgery alone.31 The authors also CRT. Again, this effect was more pronounced in SCC
demonstrated a significant increase in R0 resection rate patients, with 85% being node negative following CRT
following neoadjuvant CRT (81% vs. 69%).31 versus 58% in adenocarcinoma patients.38
The study by Urba et al. yielded similarly results and was The two final studies performed to date by Lv et al.
a powerful indicator of improved local/regional control and Cao et al. included patients with SCC exclusively
in patients treated with neoadjuvant CRT.32 Furthermore, and similarly showed comparable results with respect to
it suggested differential efficacy in patients harboring SCC oncologic resection, with improved R0 resection rates in
versus adenocarcinoma, with the former deriving a greater patients following neoadjuvant CRT, along with an elevated
benefit.32 Patients with mixed histology were enrolled. pCR pCR (22.3%), albeit lower than what was observed in the
rates differed between SCC and adenocarcinoma patients, CROSS trial.39,40 The preoperative regimen was 5FU and
with the latter exhibiting a 25% response rate consistent cisplatin-based, which may account for this disparity.
with previous studies. SCC patients exhibited an improved Taken together, the bulk of evidence to date dem-
response at 38%. R0 resection rate was improved in the onstrates improved local control following neoadjuvant
CRT group (96% vs. 90%), although surgical quality was CRT compared with surgery alone, as evidenced by a
excellent in both arms. Overall no survival benefit was significant clinical and pathologic response rate, a sig-
observed as a result of CRT. However, in patients who nificant reduction in lymph node disease burden within
exhibited a pCR, a survival benefit could be appreciated the mediastinum, and consequently, an improved R0
(median, 1-, 3-year survival pCR vs. no pCR 49.7 months, resection rate. However, this benefit is more pronounced in
86%, 64% vs. 12 months, 52%, 19% respectively; P = .01). patients with SCC compared with EAC. Another interesting
Importantly this benefit appeared to be driven by improved finding is that for chemoradiotherapy (CRT) in EAC,
local control with no difference in systemic recurrence the pCR rate hovers around a consistent 20% to 25%
rates between the two groups.32 The trial by Burmeister irrespective of the chemotherapy regimen and amount
et al. also showed improved R0 resection rates in patients of radiotherapy. However, in patients who do respond to
randomized to neoadjuvant CRT compared with surgery neoadjuvant therapy, this improved local control translates
alone (80% vs. 59%), although R0 resection rates overall into improved survival.
Multimodality Therapy in the Management of Locally Advanced Esophageal Cancer CHAPTER 38 397
with respect to distant recurrence (22.5% vs. 28.9%; P marker of the efficacy of the regimen and an opportunity to
= .31) or OS (median OS 31.8 months).41 Furthermore, modify treatment in the face of failure of a given regimen.
in-hospital mortality was significantly higher in patients Second, effective regimens aimed at controlling systemic
randomized to CRT (11.1% vs. 3.4%; P = .049).41 Taken disease seem logical, given that the bulk of mortality in
together, the data demonstrate an oncologic benefit for EC patients stems from distant, rather than local, disease.
neoadjuvant CRT in patients with locally advanced EC (T3 Third, omitting radiation therapy prior to surgery may
and/or N+ stage III and above), which is not maintained minimize treatment-related morbidity and allow it to be
in patients with earlier stage disease, owing to increased reserved for recurrent disease.41 Finally, the results of
treatment-related complications. radiation therapy in EAC, the predominant histology in the
West, are somewhat disappointing, hovering consistently
in the vicinity of 25% regardless of the CRT regimen in
CHEMOTHERAPY VERSUS CHEMORADIATION question.30,31,35–37
BEFORE SURGERY In addition to the potential benefits of chemotherapy,
no study to date has specifically mandated en bloc
In patients with locally advanced esophageal cancer, the esophagectomy and involved field lymphadenectomy in
omission of perioperative therapy represents substandard surgical patients. This may contribute to the improved
care by modern standards.1 However, both chemotherapy local control observed in patients receiving CRT. Such a
and chemoradiation therapy have demonstrated efficacy in scenario has been observed in patients with gastric cancer,
the neoadjuvant setting, and the superiority of one strategy as demonstrated by the Macdonald study, wherein rates
over another has not been clearly demonstrated.1 Advan- of D2 dissection were low, necessitating additional local
tages for both modalities have been put forward—namely therapy in the form of radiation for adequate disease
improved R0 resection rates and locoregional control in control.42 Results from several single-institution series in
regimens employing radiation, compared with a focus on which en bloc esophagectomy was routinely performed
systemic control in regimens employing chemotherapy reveal a local-regional recurrence rate of 5% to 10%,21,43,44
alone. Several oncologic surrogate metrics are frequently significantly lower than the consistent 25% to 30% rate
employed when comparing studies, most notably pCR seen with a standard esophagectomy.11,35,36,45,46 Furthermore,
and complete resection (R0) rate. However, it must be in a randomized trial of transhiatal versus transthoracic
stressed that irrespective of these two pathologic results, esophagectomy (with modified en bloc), local recurrence
the ultimate goal of treatment for patients with esophageal and DFS were significantly improved in the transthoracic
cancer is improving long-term outcomes, measured by OS study arm.47 Thus it is possible that adequate local control
and DFS. With this in mind, we can look at the available can be achieved in at least a subset of patients with locally
modalities of treatment for esophageal cancer in two broad advanced disease with high-quality surgery, as represented
terms—systemic therapy (chemotherapy and targeted by an en bloc esophagectomy, potentially obviating the
agents) versus local therapy (radiotherapy and surgery). need for additional local therapy in the form of radio-
Theoretically, the goal of these two approaches is different; therapy. Accordingly, it is possible that in patients in
systemic therapies should address distant disease with whom an en bloc esophagectomy is performed, the only
its effectiveness measured by OS and the rate of distant additional therapy required is systemic treatment, given
metastatic failure, and local therapies should address the low rate of local-regional recurrence in these patients.
the primary tumor and regional lymph node disease, The literature directly comparing neoadjuvant CRT and
the effectiveness of which is measured by the metrics of chemotherapy are limited. Nonetheless, several randomized
complete resection (R0) and local-regional recurrence. trials on this topic have been performed and are demon-
strated in Table 38.3. The published studies to date suffer
THE CASE FOR NEOADJUVANT CHEMORADIATION from low accrual and the fact that en bloc esophagectomy
In early studies looking at the perioperative use of chemo- was not part of the treatment plan. Regardless, no trial to
therapy or radiation therapy, patients who demonstrate a date has demonstrated a clearly superior modality, and
pathologic response to therapy and in particular a complete both represent currently acceptable standards.48–50
response tend to derive the greatest survival benefit. To Stahl et al. randomized patients with distal one-third
date, the highest rates of pCR have been achieved with and esophagogastric junction (EGJ) adenocarcinoma to
CRT regimens.31,35,37,40 From the standpoint of locoregional receive preoperative chemotherapy with cisplatin and
control, CRT is associated with improved R0 resection 5FU, or preoperative concurrent CRT with the same
rates and significant mediastinal lymph node sterilization, regimen in addition to 30 Gy of radiation.48 In total, 126
which adds to the local benefit provided by surgery.31,35,37,40 patients were enrolled, with 64 patients randomized to
Furthermore, a benefit with respect to distant control chemotherapy and surgery and 62 randomized to CRT,
has been observed with CRT.13,30,31,34,35,37,40 This has been with the trial being terminated early due to poor accrual.
attributed to the systemic effect of chemotherapy on any Patients receiving CRT demonstrated improved pCR
micrometastatic disease present, as well as on the improved (15.6% vs. 3%; P = .03) and a node-negative surgical
rates of lymph node clearance.13,30,31,34,35,37,40 specimen (36.7% vs. 64.4%; P = .01). R0 resection rates
were comparable in the two groups (69.5% chemotherapy
THE CASE FOR NEOADJUVANT CHEMOTHERAPY [CT] and 72% CRT). In-hospital mortality was increased
Administration of chemotherapy in the preoperative setting in patients receiving neoadjuvant CRT compared with
provides a number of advantages. First, response to the CT at 10.2% vs. 3.8%, but this difference did not reach
regimen can be observed prior to surgery, providing a statistical significance, likely due to low power (P = .26).
Multimodality Therapy in the Management of Locally Advanced Esophageal Cancer CHAPTER 38 399
Median and 3-year survival were improved in the CRT 35 Gy of radiation. Overall, 75 patients were randomized,
arm (see Table 38.3), but this did not reach significance. with 36 receiving chemotherapy and 39 receiving CRT
Subgroup analysis revealed that OS was improved in prior to surgery. As in the study by Stahl, the study was
patients who underwent R0 resection and had negative terminated early due to poor accrual. R0 resection rate
lymph nodes following surgery, irrespective of which was comparable, at 80.5% and 84.6% in patients receiving
group they were randomized to.48 However, because the chemotherapy or CRT, respectively. The pCR rate was 13%
rate of pathologically negative nodes was higher in the in patients who received CRT versus 0% in those who
CRT group, the authors posit that CRT could provide a received chemotherapy alone. This difference was not
significant survival advantage in addition to R0 resection, associated with reduced locoregional, distant recurrence,
as evidenced by the trend toward improved survival. or survival benefit.49
The study by Burmeister was similar in that patients The most recent study from Sweden by Klevebro et al.
with esophageal or EGJ adenocarcinoma were random- randomized 181 patients to neoadjuvant chemotherapy
ized to undergo neoadjuvant chemotherapy alone or (nCT), comprising three cycles of cisplatin and 5FU or neo-
concurrent chemoradiation.49 Patients received a 5FU/ adjuvant chemoradiation (nCRT) therapy with concomitant
cisplatin-based regimen either alone or in conjunction with 40 Gy.50 Surgery included a two-field lymphadenectomy in
400 SECTION I Esophagus and Hernia
all cases. A total of 91 and 90 patients were allocated to a modified MAGIC protocol versus CRT, according to
each arm, respectively. The study predominantly included the CROSS protocol. An Australian trial (TOP-GEAR)
patients with EAC; however, nearly one-third had squamous is randomizing patients with esophageal and gastric
histology. As in the prior two studies, pCR was increased adenocarcinoma to chemotherapy (ECF), then surgery
in patients receiving nCRT compared with nCT (28% vs. versus ECF, followed by CRT then surgery, with a series of
9%, P = .002). Similarly, pCR was associated with negative adjuvant cycles of ECF. The latest trial to investigate this
regional nodes on final pathology in 90% of patients. topic is a German initiative (ESOPEC) randomizing EAC
Accordingly, 35% of nCT treated patients were found to patients to the docetaxel-based FLOT or CROSS (taxol/
be node negative, compared with 65% of nCRT patients, carboplatin with concurrent radiotherapy) regimens in
despite a 63% rate of node positivity on initial staging the neoadjuvant setting. Results from these trials are
in both groups. However, this did not translate into a expected over the next 5 years and hope to provide some
survival benefit on an intention-to-treat or per protocol clarity to this topic.
analysis. Subgroup analysis revealed improved outcomes
following nCRT in patients with SCC that did not reach
statistical significance compared with nCT. No trend was
DEFINITIVE CHEMORADIATION
observed in adenocarcinoma patients. Furthermore, a An examination of the data following combined CRT for
trend toward increased severe postoperative complications esophageal cancer reveals that a significant number of
was appreciated.50 patients experience a pCR as well as mediastinal downstag-
Several criticisms with respect to each of these trials can ing and nodal clearance. This finding begs the question as
be put forward. First, all were relatively underpowered to to whether patients who demonstrate a complete response
detect a clear difference in treatments. Second, none of require surgery at all. Instead, should it be reserved for
the regimens employed in either treatment arm (nCT or patients who fail to respond adequately? A number of
nCRT) reflect those employed in the landmark studies to studies have addressed this question specifically and are
date for each modality.9,35 Proponents of chemotherapy outlined in Table 38.4.52–57 Overall, the main drawback
alone highlight the positive results in GEJ adenocarcinoma is a prohibitive rate of local failure, even in seemingly
patients following the MAGIC trial, which employed triplet complete responders in the range of 40% to 60%.52–57
therapy with epirubicin, cisplatin, and 5FU in the pre- and Thus the current standard in many centers is to reserve
postoperative setting, as well as the promising results of definitive chemoradiation for poor surgical candidates,
regimens incorporating docetaxel (DCF or FLOT).9,35 a strategy that is best suited for SCC histology, given the
Similarly, the CROSS trial employed carboplatin, paclitaxel, poorer response rate of adenocarcinoma to radiotherapy.
and concurrent 41.4 Gy.9,35 No randomized trial to date The initial study demonstrating a survival benefit for
has directly compared these regimens. Finally, none of combined CF-based CRT in EC patients was conducted
these studies have required en bloc esophagectomy with by Herskovic et al.52,53 The authors randomized patients
involved field lymphadenectomy in their treatment pro- with predominantly squamous histology to undergo CRT
tocols, which may explain the findings noted in the Stahl with 50 Gy, compared with radiation alone with 64 Gy.
study, wherein node negativity correlated with improved The authors showed improved outcomes with CRT versus
survival, despite an equal R0 resection rate with both radiation alone with respect to local and distant recurrence,
treatment modalities. However, there are retrospective as well as a significant survival advantage (5 year 26% vs.
data on this topic that attempt to address the influence 0% radiation alone). Despite a high local failure rate
of en bloc esophagectomy on the choice of neoadjuvant approaching 50%, the 5-year survival observed was in line
therapy. Spicer and colleagues combined data on locally with standard surgery–based treatment at that time, thus
advanced (cT3N+) EAC patients from three centers, all suggesting a role for definitive chemoradiation.52,53 In
performing routine en bloc esophagectomy, two of which an attempt to improve upon these results, Minsky et al.,
prefer neoadjuvant chemotherapy (cisplatin and 5FU performed a similarly structured trial, this time random-
[CF] or DCF) and one CRT.51 No difference in any of the izing patients to receive combined CF-based CRT with
pathologic surrogates of oncologic efficacy (R0 or lymph either 50.4 or 64.8 Gy of concurrent radiation.54 Squamous
node retrieval) differed, although CRT was associated histology was present in 95% of included patients, and
with an increased pCR rate, as would be expected. With as in the previous study, there was a high local failure
more than 100 patients in each study group, there was rate regardless of radiation dose in the vicinity of 50% to
no difference in either DFS or OS between patients who 60%. Survival rates, however, were similarly comparable.54
received chemotherapy or CRT.51 Given the relatively favorable survival outcomes noted
Several ongoing trials will help resolve this ongoing in these studies and the observation that adequate local
question. The Japanese JCOG1109 (NExT trial) is a three- control is achieved in nearly half the patients subject
arm phase III trial comparing neoadjuvant chemotherapy to definitive chemoradiation without exposure to the
with radiation therapy in patients with locally advanced morbidity associated with esophagectomy, a prospective
esophageal SCC undergoing en bloc esophagectomy—the comparison between chemoradiation with and without
only trial to do so.27 The chemotherapy regimens include surgery was undertaken by Stahl et al.55 Induction che-
preoperative docetaxel/cisplatin/5FU and cisplatin/5FU motherapy with cisplatin, etoposide, and 5FU in three
versus preoperative cisplatin/5FU-based CRT. Similarly, cycles was followed by concurrent etoposide, cisplatin, and
the ongoing Irish ICORG 10 to 14 trial (Neo-AEGIS) 65 Gy in the CRT alone group. In the surgery arm, this
will directly compare the treatment outcomes in patients regimen was given with 40 Gy of radiation, followed by
with locally advanced esophageal adenocarcinoma using transthoracic esophagectomy. Following surgery, pCR and
Multimodality Therapy in the Management of Locally Advanced Esophageal Cancer CHAPTER 38 401
Local Failure
Study Year Patients Regimen Histology (CRT vs. XRT) pCR R0 Survival (CRT vs. XRT)
Herskovic/ 1992 129 5FU, cisplatin +50 Gy vs. SCC/EAC PD 27%, LR NA NA Median 12.5 mo vs.
Cooper et al. 64 Gy 16% vs. PD 8.9 mo
(RTOG 85-01) 40%, LR 24% 5 yr 26% vs. 0%*
Minsky et al. 2002 236 5FU, cisplatin + 50.4 Gy SCC/EAC 50% vs. 55% NA NA Median 13 mo vs.
(RTOG 94-05/ vs. 5FU, cisplatin + 18.1 mo
INT 0123) 64.8 Gy 2 yr 31% vs. 40%
Stahl et al. 2005 172 Induction 5FU, cisplatin, SCC 2 yr 59% vs. 35% 82% Median 14.9 mo vs.
etoposide + concurrent 36%* 16.4 mo
cisplatin, etoposide + 3 yr 24.4% vs.
65 Gy vs. 5FU, 31.3%
cisplatin, etoposide +
concurrent cisplatin,
etoposide + 40 Gy + Sx
Bedenne et al. 2007 259 Induction 5FU, cisplatin + SCC/EAC 43% vs. 33.6%* 23% 75% Median 19.3 mo vs.
45 Gy 17.7 mo
Concurrent 5FU, cisplatin 2 yr 39.8% vs.
+ 15–20 Gy vs. 33.6%
induction + Sx
Conroy et al. 2014 267 FOLFOX + 50 Gy vs. 5FU, SCC/EAC PD 45% vs. NA NA Median 20.2 mo vs.
(PRODIGE5/ cisplatin + 50 Gy 46% 17.5 mo
ACCORD17) 3 yr 19.9% vs.
26.9%
*Denotes statistical significance.
CRT, Chemoradiotherapy; EAC, esophageal adenocarcinoma; FOLFOX, folinic acid, 5FU, and oxaliplatin; 5FU, 5-fluorouracil; LR, local recurrence;
pCR, pathologic complete response; PD, persistent disease; SCC, squamous cell carcinoma; Sx, surgery; XRT, radiation therapy.
node negativity was appreciated in 35% of patients with mortality was 1% in the nonsurgical arm and 9% in the
nodal sterilization alone in an additional 33%. Survival surgical arm (P = .002). No difference in survival was
analysis revealed a significant reduction in cancer-related appreciated between treatment groups, with median
mortality in patients who underwent surgery. However, and 2-year survival between the CRT and surgical arms
this was associated with a greater incidence of treatment- of 19.3 months versus 17.7 months and 39.8% versus
related death. In-hospital mortality was 12.8% in patients 33.6%, respectively. Although there was no difference
who received surgery, versus 3.5% in patients who did in the frequency of metastasis between the two groups,
not (P = .03). Survival at 2 and 3 years was equivalent local failures were more common in patients randomized
in both groups (39.9%, 95% CI, 29.4 to 50.4, CRT + Sx; to CRT alone (HR no surgery vs. surgery, 1.63; 95% CI,
35.4%, 95% CI, 25.2 to 45.6, and 31.3% vs. 24.4%). In 1.04 to 2.55; P = .03). Collectively these data indicate that
keeping with the data presented thus far, local control chemoradiation provides comparable survival outcomes to
was improved in patients subject to surgery. Freedom multimodality therapy in appropriately selected patients.
from progression in CRT alone was 40.7% versus 64.3% Although surgery is associated with improved local control,
in CRT + Sx (P ≥ .003). Multivariate analysis revealed that it comes at the cost of increased treatment-associated
the most important prognostic factor, however, was tumor mortality.56
response to therapy. Those patients in whom a response Taken together, these results suggest that in patients
was noted demonstrated survival approaching 50% at 5 with locally advanced SCC, definitive CRT and CRT +
years, irrespective of treatment arm. In nonresponders, surgery offer equivalent survival results, with a reduction
however, R0 resection improved survival, increasing survival in treatment-associated morbidity and mortality compared
from 17.9% at 3 years to 32%.55 with the addition of surgery. Although local control was
Additional high-quality evidence supports these find- improved in patients subject to surgery, subgroup analysis
ings.56 Bedenne et al. randomized patients to surgery or suggests that surgery is rescuing patients who do not
definitive chemoradiation therapy following induction demonstrate an adequate response to CRT, thus providing
therapy with a CF-based regimen. Patients who dem- local control because of CRT failure. This is evidenced
onstrated a response were subsequently randomized to by the survival rates observed in patients who responded
surgery or additional chemoradiation, for a total dose of to treatment versus those who did not.
45 to 66 Gy. Patients who did not respond proceeded to Given the collective results thus far, the majority of
surgery. Patients predominantly harbored SCC (90%). R0 centers employ definitive CRT for patients who are unfit
resection was achieved in 75% of patients, and pCR was for surgery, and numerous studies are underway in search
noted in 23% of surgical specimens. Treatment-related of the optimal regimen.
402 SECTION I Esophagus and Hernia
Staging
SCC+EAC EAC
Local therapy alone Neoadjuvant chemotherapy
Up-front surgery preferred 5FU/platinum/taxane-based triplets (MAGIC)
over XRT associated with 5FU/cisplatin-based doublets
increased morbidity Neoadjuvant chemoradiotherapy
Platinum/taxane doublet + 41.4 Gy (CROSS)
SCC
CRT regimen, as opposed to neoadjuvant CT alone for adenocarcinoma: an FNCLCC and FFCD multicenter phase III
appropriately selected patients with EAC. In both groups, trial. J Clin Oncol. 2011;29:1715-1721.
16. Ando N, Kato H, Igaki H, et al. A randomized trial comparing
surgery is aimed at achieving an R0 resection, which is postoperative adjuvant chemotherapy with cisplatin and 5-fluorouracil
effectively achieved with en bloc esophagectomy. Thus versus preoperative chemotherapy for localized advanced squamous
the optimal treatment of esophageal cancer necessitates cell carcinoma of the thoracic esophagus (JCOG9907). Ann Surg
a multidisciplinary approach tailored to the individual Oncol. 2012;19:68-74.
17. Boonstra JJ, Kok TC, Wijnhoven BP, et al. Chemotherapy followed
patient, taking into account their tumor location, histology, by surgery versus surgery alone in patients with resectable esopha-
and performance status. In this manner, the selection of geal squamous cell carcinoma: long-term results of a randomized
a regimen most likely to improve their survival and limit controlled trial. BMC Cancer. 2011;11:181.
their treatment-related morbidity becomes increasingly 18. Allum WH, Stenning SP, Bancewicz J, Clark PI, Langley RE. Long-term
possible. The spectrum of available regimens is likely results of a randomized trial of surgery with or without preoperative
chemotherapy in esophageal cancer. J Clin Oncol. 2009;27:5062-5067.
to increase in the future, and a number of randomized 19. Ancona E, Ruol A, Santi S, et al. Only pathologic complete response
studies are underway that may answer some of the ongoing to neoadjuvant chemotherapy improves significantly the long term
questions in the treatment of this devastating disease. survival of patients with resectable esophageal squamous cell carci-
noma: final report of a randomized, controlled trial of preoperative
chemotherapy versus surgery alone. Cancer. 2001;91:2165-2174.
REFERENCES 20. Mohammad NH, ter Veer E, Ngai L, Mali R, van Oijen MGH, van
Laarhoven HWM. Optimal first-line chemotherapeutic treatment
1. Cools-Lartigue J, Spicer J, Ferri LE. Current status of management in patients with locally advanced or metastatic esophagogastric
of malignant disease: current management of esophageal cancer. carcinoma: triplet versus doublet chemotherapy: a systematic literature
J Gastrointest Surg. 2015;19:964-972. review and meta-analysis. Cancer Metastasis Rev. 2015;34:429-441.
2. Amin RN, Parikh SJ, Gangireddy VG, Kanneganti P, Talla S, Daram S. 21. Ferri LE, Ades S, Alcindor T, et al. Perioperative docetaxel, cisplatin,
Early esophageal cancer specific survival is unaffected by anatomical and 5-fluorouracil (DCF) for locally advanced esophageal and
location of tumor: a population-based study. Can J Gastroenterol gastric adenocarcinoma: a multicenter phase II trial. Ann Oncol.
Hepatol. 2016;2016:613-640. 2012;23:1512-1517.
3. Tapias LF, Mathisen DJ, Wright CD, et al. Outcomes with open and 22. Watanabe M, Nagai Y, Kinoshita K, et al. Induction chemotherapy
minimally invasive Ivor Lewis esophagectomy after neoadjuvant with docetaxel/cisplatin/5-fluorouracil for patients with node-positive
therapy. Ann Thorac Surg. 2016;101:1097-1103. esophageal cancer. Digestion. 2011;83:146-152.
4. Defoe SG, Pennathur A, Flickinger JC, et al. Retrospective review 23. Schulz C, Kullmann F, Kunzmann V, et al. NeoFLOT: multicenter
of patients with locally advanced esophageal cancer treated at the phase II study of perioperative chemotherapy in resectable adenocar-
University of Pittsburgh. Am J Clin Oncol. 2011;34:587-592. cinoma of the gastresophageal junction or gastric adenocarcinoma—
5. Sudarshan M, Alcindor T, Ades S, et al. Survival and recurrence very good response predominantly in patients with intestinal type
patterns after neoadjuvant docetaxel, cisplatin, and 5-fluorouracil tumors. Int J Cancer. 2015;137:678-685.
(DCF) for locally advanced esophagogastric adenocarcinoma. Ann 24. Lorenzen S, Thuss-Patience P, Al-Batran SE, et al. Impact of pathologic
Surg Oncol. 2015;22:324-330. complete response on disease-free survival in patients with esopha-
6. Ku GY, Kriplani A, Janjigian YY, et al. Change in chemotherapy gogastric adenocarcinoma receiving preoperative docetaxel-based
during concurrent radiation followed by surgery after a suboptimal chemotherapy. Ann Oncol. 2013;24:2068-2073.
positron emission tomography response to induction chemotherapy 25. Hara H, Tahara M, Daiko H, et al. Phase II feasibility study of preop-
improves outcomes for locally advanced esophageal adenocarcinoma. erative chemotherapy with docetaxel, cisplatin, and fluorouracil for
Cancer. 2016;122:2083-2090. esophageal squamous cell carcinoma. Cancer Sci. 2013;104:1455-1460.
7. Ando N, Iizuka T, Ide H, et al. Surgery plus chemotherapy compared 26. Watanabe M, Baba Y, Yoshida N, et al. Outcomes of preoperative
with surgery alone for localized squamous cell carcinoma of the chemotherapy with docetaxel, cisplatin, and 5-fluorouracil followed by
thoracic esophagus: a Japan Clinical Oncology Group Study— esophagectomy in patients with resectable node-positive esophageal
JCOG9204. J Clin Oncol. 2003;21:4592-4596. cancer. Ann Surg Oncol. 2014;21:2838-2844.
8. Ando N, Iizuka T, Kakegawa T, et al. A randomized trial of surgery 27. Nakamura K, Kato K, Igaki H, et al. Three-arm phase III trial
with and without chemotherapy for localized squamous carcinoma comparing cisplatin plus 5-FU (CF) versus docetaxel, cisplatin plus
of the thoracic esophagus: the Japan Clinical Oncology Group Study. 5-FU (DCF) versus radiotherapy with CF (CF-RT) as preoperative
J Thorac Cardiovasc Surg. 1997;114:205-209. therapy for locally advanced esophageal cancer (JCOG1109, NExT
9. Cunningham D, Allum WH, Stenning SP, et al. Perioperative study). Jpn J Clin Oncol. 2013;43:752-755.
chemotherapy versus surgery alone for resectable gastresophageal 28. Nygaard K, Hagen S, Hansen HS, et al. Pre-operative radiotherapy
cancer. N Engl J Med. 2006;355:11-20. prolongs survival in operable esophageal carcinoma: a randomized,
10. Kelsen DP, Winter KA, Gunderson LL, et al. Long-term results of multicenter study of pre-operative radiotherapy and chemotherapy.
RTOG trial 8911 (USA Intergroup 113): a random assignment trial The second Scandinavian trial in esophageal cancer. World J Surg.
comparison of chemotherapy followed by surgery compared with 1992;16:1104-1109, discussion 1110.
surgery alone for esophageal cancer. J Clin Oncol. 2007;25:3719-3725. 29. Le Prise E, Etienne PL, Meunier B, et al. A randomized study
11. Law S, Fok M, Chow S, Chu KM, Wong J. Preoperative chemotherapy of chemotherapy, radiation therapy, and surgery versus surgery
versus surgical therapy alone for squamous cell carcinoma of the for localized squamous cell carcinoma of the esophagus. Cancer.
esophagus: a prospective randomized trial. J Thorac Cardiovasc Surg. 1994;73:1779-1784.
1997;114:210-217. 30. Walsh TN, Noonan N, Hollywood D, Kelly A, Keeling N, Hennessy
12. Medical Research Council Esophageal Cancer Working Group. Surgical TP. A comparison of multimodal therapy and surgery for esophageal
resection with or without preoperative chemotherapy in esophageal adenocarcinoma. N Engl J Med. 1996;335:462-467.
cancer: a randomised controlled trial. Lancet. 2002;359:1727-1733. 31. Bosset JF, Gignoux M, Triboulet JP, et al. Chemoradiotherapy
13. Roth JA, Pass HI, Flanagan MM, Graeber GM, Rosenberg JC, Steinberg followed by surgery compared with surgery alone in squamous-cell
S. Randomized clinical trial of preoperative and postoperative cancer of the esophagus. N Engl J Med. 1997;337:161-167.
adjuvant chemotherapy with cisplatin, vindesine, and bleomycin for 32. Urba SG, Orringer MB, Turrisi A, Iannettoni M, Forastiere A,
carcinoma of the esophagus. J Thorac Cardiovasc Surg. 1988;96:242-248. Strawderman M. Randomized trial of preoperative chemoradia-
14. Schlag PM. Randomized trial of preoperative chemotherapy for tion versus surgery alone in patients with locoregional esophageal
squamous cell cancer of the esophagus. The Chirurgische Arbe- carcinoma. J Clin Oncol. 2001;19:305-313.
itsgemeinschaft Fuer Onkologie der Deutschen Gesellschaft Fuer 33. Burmeister BH, Smithers BM, Gebski V, et al. Surgery alone versus
Chirurgie Study Group. Arch Surg. 1992;127:1446-1450. chemoradiotherapy followed by surgery for resectable cancer of the
15. Ychou M, Boige V, Pignon JP, et al. Perioperative chemotherapy esophagus: a randomised controlled phase III trial. Lancet Oncol.
compared with surgery alone for resectable gastresophageal 2005;6:659-668.
404 SECTION I Esophagus and Hernia
34. Tepper J, Krasna MJ, Niedzwiecki D, et al. Phase III trial of trimodal- 47. Omloo JM, Lagarde SM, Hulscher JB, et al. Extended transthoracic
ity therapy with cisplatin, fluorouracil, radiotherapy, and surgery resection compared with limited transhiatal resection for adenocarci-
compared with surgery alone for esophageal cancer: CALGB 9781. noma of the mid/distal esophagus: five-year survival of a randomized
J Clin Oncol. 2008;26:1086-1092. clinical trial. Ann Surg. 2007;246:992-1000, discussion 1000–1001.
35. van Heijl M, van Lanschot JJ, Koppert LB, et al. Neoadjuvant 48. Stahl M, Walz MK, Stuschke M, et al. Phase III comparison of
chemoradiation followed by surgery versus surgery alone for patients preoperative chemotherapy compared with chemoradiotherapy in
with adenocarcinoma or squamous cell carcinoma of the esophagus patients with locally advanced adenocarcinoma of the esophagogastric
(CROSS). BMC Surg. 2008;8:21. junction. J Clin Oncol. 2009;27:851-856.
36. Shapiro J, van Lanschot JJ, Hulshof MC, et al. Neoadjuvant chemo- 49. Burmeister BH, Thomas JM, Burmeister EA, et al. Is concurrent
radiotherapy plus surgery versus surgery alone for esophageal or radiation therapy required in patients receiving preoperative che-
junctional cancer (CROSS): long-term results of a randomised motherapy for adenocarcinoma of the esophagus? A randomised
controlled trial. Lancet Oncol. 2015;16:1090-1098. phase II trial. Eur J Cancer. 2011;47:354-360.
37. van Hagen P, Hulshof MC, van Lanschot JJ, et al. Preoperative 50. Klevebro F, Alexandersson von Dobeln G, Wang N, et al. A random-
chemoradiotherapy for esophageal or junctional cancer. N Engl J ized clinical trial of neoadjuvant chemotherapy versus neoadjuvant
Med. 2012;366:2074-2084. chemoradiotherapy for cancer of the esophagus or gastro-esophageal
38. Bass GA, Furlong H, O’Sullivan KE, Hennessy TP, Walsh TN. junction. Ann Oncol. 2016;27:660-667.
Chemoradiotherapy, with adjuvant surgery for local control, confers 51. Spicer JD, Stiles BM, Sudarshan M, et al. Preoperative chemoradiation
a durable survival advantage in adenocarcinoma and squamous cell therapy versus chemotherapy in patients undergoing modified en
carcinoma of the esophagus. Eur J Cancer. 2014;50:1065-1075. bloc esophagectomy for locally advanced esophageal adenocarcinoma:
39. Cao XF, He XT, Ji L, Xiao J, Lv J. Effects of neoadjuvant radiochemo- is radiotherapy beneficial? Ann Thorac Surg. 2016;101:1262-1269,
therapy on pathological staging and prognosis for locally advanced discussion 1969–1270.
esophageal squamous cell carcinoma. Dis Esophagus. 2009;22:477-481. 52. Herskovic A, Martz K, al-Sarraf M, et al. Combined chemotherapy
40. Lv J, Cao XF, Zhu B, Ji L, Tao L, Wang DD. Long-term efficacy of and radiotherapy compared with radiotherapy alone in patients
perioperative chemoradiotherapy on esophageal squamous cell with cancer of the esophagus. N Engl J Med. 1992;326:1593-1598.
carcinoma. World J Gastroenterol. 2010;16:1649-1654. 53. Cooper JS, Guo MD, Herskovic A, et al. Chemoradiotherapy of locally
41. Mariette C, Dahan L, Mornex F, et al. Surgery alone versus chemora- advanced esophageal cancer: long-term follow-up of a prospective
diotherapy followed by surgery for stage I and II esophageal cancer: randomized trial (RTOG 85-01). Radiation Therapy Oncology
final analysis of randomized controlled phase III trial FFCD 9901. Group. JAMA. 1999;281:1623-1627.
J Clin Oncol. 2014;32:2416-2422. 54. Minsky BD, Pajak TF, Ginsberg RJ, et al. INT 0123 (Radiation Therapy
42. Macdonald JS, Smalley SR, Benedetti J, et al. Chemoradiotherapy Oncology Group 94-05) phase III trial of combined-modality therapy
after surgery compared with surgery alone for adenocarcinoma of the for esophageal cancer: high-dose versus standard-dose radiation
stomach or gastroesophageal junction. N Engl J Med. 2001;345:725-730. therapy. J Clin Oncol. 2002;20:1167-1174.
43. Altorki N, Skinner D. Should en bloc esophagectomy be the standard 55. Stahl M, Stuschke M, Lehmann N, et al. Chemoradiation with and
of care for esophageal carcinoma? Ann Surg. 2001;234:581-587. without surgery in patients with locally advanced squamous cell
44. Hagen JA, DeMeester SR, Peters JH, Chandrasoma P, DeMeester TR. carcinoma of the esophagus. J Clin Oncol. 2005;23:2310-2317.
Curative resection for esophageal adenocarcinoma: analysis of 100 56. Bedenne L, Michel P, Bouche O, et al. Chemoradiation followed by
en bloc esophagectomies. Ann Surg. 2001;234:520-530, discussion surgery compared with chemoradiation alone in squamous cancer
530–521. of the esophagus: FFCD 9102. J Clin Oncol. 2007;25:1160-1168.
45. Dresner SM, Griffin SM. Pattern of recurrence following radical 57. Conroy T, Galais MP, Raoul JL, et al. Definitive chemoradiotherapy
oesophagectomy with two-field lymphadenectomy. Br J Surg. with FOLFOX versus fluorouracil and cisplatin in patients with
2000;87:1426-1433. esophageal cancer (PRODIGE5/ACCORD17): final results of a
46. Hulscher JB, van Sandick JW, Tijssen JG, Obertop H, van Lanschot randomised, phase 2/3 trial. Lancet Oncol. 2014;15:305-314.
JJ. The recurrence pattern of esophageal carcinoma after transhiatal
resection. J Am Coll Surg. 2000;191:143-148.
CHAPTER
Surgical Approaches to Remove
the Esophagus: Open 39A
B.J. Noordman
| S.M. Lagarde
| B.P.L. Wijnhoven
|
ABSTRACT
Surgical resection remains the cornerstone of therapy for
patients with resectable cancer of the esophagus in the
absence of systemic metastases. Surgery, most of the time
combined with neoadjuvant therapy in current practice,
offers the highest likelihood of cure for patients with
locoregional disease. To obtain the best results, the manage-
ment of esophageal cancer should be individualized and
based on a combination of factors including the physiologic
status of the patient, tumor type and location, and stage
of disease. In this chapter, we describe the different open
surgical approaches to remove the esophagus in patients
with esophageal cancer. Although minimally invasive
techniques are increasingly applied, the benefits of fully
minimally invasive esophagectomy have not yet been proven
unequivocally, and an open or hybrid esophagectomy
remains the standard procedure to remove the esopha-
gus in many leading high-volume centers worldwide. At
present, the only strong available evidence comes from
preliminary results of the French randomized MIRO trial
comparing hybrid transthoracic esophagectomy (TTE,
laparoscopic gastric mobilization , and open thoracotomy)
with fully open TTE. These results suggest that hybrid
TTE significantly reduces postoperative complications
compared with open TTE (odds ratio for postoperative
morbidity, 0.31; 95% confidence interval, 0.18 to 0.55;
P = .0001; percentage of pulmonary complications: 17.7%
vs. 30.1% vs. P = .037). The open thoracic part of hybrid
TTE is similar to that of fully open TTE (this chapter),
whereas the laparoscopic (abdominal) part is described
in detail in another chapter.
KEYWORDS
Esophagectomy; transhiatal; transthoracic; salvage surgery;
lymphadenectomy; postoperative complications; neoad-
juvant therapy
406 SECTION I Esophagus and Hernia
TABLE 39A.1 Relationship Between Tumor Depth (T-stage) and Lymph Node Status (N-stage) for Esophageal
Adenocarcinoma
Prevalence of Node Number of Involved Number With 1–4 Number With >4
Tumor Depth Metastases (%)* Nodes (Median [IQR])† Involved Nodes (%)‡ Involved Nodes (%)§
Intramucosal (T1A) 1/16 (6) 2 (n/a) 1/16 (6) 0/16 (0)
Submucosal (T1B) 5/16 (31) 1 (n/a) 4/16 (25) 1/16 (6)
Intramuscular (T2) 10/13 (77) 2 (1–4) 9/13 (69) 1/13 (8)
Transmural (T3) 47/55 (85) 5 (3–13.5) 22/55 (40) 25/55 (45)
*χ2 = 42.0, P < .0001 (chi-square test for trend).
χ = 11.02, P = .0116 (Kruskal-Wallis; includes only patients with involved nodes).
† 2
Modified from Hagen JA, DeMeester SR, Peters JH, Chandrasoma P, DeMeester TR. Curative resection for esophageal adenocarcinoma: analysis of
100 en bloc esophagectomies. Ann Surg. 2001;234:520-530.
has penetrated the submucosal layer, up to one-half of pleura lying on both sides of the azygos arch is incised,
patients will have nodal metastases.20 More than 80% of and the arch is ligated or closed with a stapling device and
patients with invasion of the muscularis propria will have subsequently transected. The pleura cranial to the azygos
at least one involved lymph node.21 In the presence of arch is incised and saved to create a pedicled “flap” to
transmural invasion, nodal involvement will be present cover the subsequent intrathoracic anastomosis. The right
in more than 85%, and the median number of involved paratracheal nodes are removed in between the trachea,
nodes and the proportion of patients with more than superior vena cava, and the azygos arch. The right vagal
four involved nodes increases (Table 39A.1).22 Extended nerve and the bronchial artery are divided. The vagal
lymphadenectomy as performed during TTE increases the nerve should not be divided with use of electrocautery to
chance of removal of all tumor-positive lymph nodes and prevent injury to the right recurrent nerve. The pleura
theoretically improves regional tumor control and perhaps overlying the lateral aspect of the vertebral bodies is incised
even long-term survival. However, high-quality clinical from the level of the azygos arch to the diaphragm, and
evidence on the optimal extent of lymphadenectomy is the intercostal veins are divided between ligatures or
absent, especially in the present era of neoadjuvant treat- clips where they enter the azygos vein. A dissection plane
ment. Consequently, individual opinions and institutional is then created following each intact intercostal artery
preferences currently dominate the choice of surgical to reach the adventitial plane of the aorta. Dissection
technique and extent of lymphadenectomy. continues across the anterior surface of the aorta until
the left mediastinal pleura is reached. Direct branches of
TECHNIQUE OF OPEN EN BLOC the thoracic aorta to the esophagus should be carefully
TRANSTHORACIC ESOPHAGECTOMY ligated before dividing. One or two communicating veins
En bloc TTE is performed through a right thoracotomy to the hemiazygos need to be ligated as they pass behind
and a midline laparotomy. The proximal anastomosis is the aorta. The mediastinal tissue posteriorly between
performed either through an extra incision made at the the azygos vein and the aorta just above the diaphragm
left side of the neck or in the chest (see “Anastomosis”). includes the thoracic duct, which should be identified and
When a cervical anastomosis is performed, the procedure transected at this stage. A heavy nonresorbable ligature
starts with a thoracotomy followed by the abdominal part should be placed caudally to prevent the development of
of the operation, whereas in case of an intrathoracic a chylothorax. The dissection can be ended at the level
anastomosis the laparotomy is performed prior to the of both crura of the diaphragm.
thoracic phase. The anterior portion of the dissection is performed
The thoracic dissection includes removal of the azygos along the previously incised inferior pulmonary ligament.
vein with its associated nodes, the thoracic duct, and the Hereby, the posterior aspect of the pericardium is freed by
paratracheal, subcarinal, paraesophageal, and parahiatal blunt and sharp dissection. The pericardium should only be
nodes in continuity with the resected esophagus. Nodes removed when the tumor is adherent. Once the left medi-
in the aortopulmonary window are removed separately. astinal pleura is reached, the plane can be connected with
The block of tissue removed is bounded laterally on each the previous dissection over the aorta. Sometimes the left
side by the excised mediastinal pleura, anteriorly by the pleura is incised. The thoracic esophagus is then encircled
pericardium and membranous part of the trachea, and with a Penrose drain for traction. The anterior dissection
posteriorly by the aorta and vertebral bodies. is then continued cephalad along the pericardium until
During the thoracic phase the patient is placed in the subcarinal nodes are encountered. Careful dissection
the left lateral decubitus position with a posterolateral along the right main bronchus up to the carina and then
thoracotomy performed entering the chest through the distally along the left main bronchus allows for removal
fifth or sixth intercostal space. The inferior pulmonary of the entire subcarinal node basin in continuity with the
ligament is divided to the level of the inferior pulmonary resected esophagus. At this point, the anterior dissection is
vein. The pleura overlying the right main bronchus is also transitioned to the wall of the esophagus by dividing
divided, taking into account its membranous part. The the left vagal nerve where it crosses the left main bronchus.
Surgical Approaches to Remove the Esophagus: Open CHAPTER 39A 407
The esophagus is separated from the membranous part (consequently a narrower tube). When an intrathoracic
of the trachea. In case of an intrathoracic anastomosis, anastomosis is performed, more of the right gastric vessels
the esophagus is divided above the level of the azygos can be preserved; consequently, a wider tube can be
arch. In case of a cervical anastomosis, the dissection created. Finally, the staple line is oversewn.
is continued toward the root of the neck. The lymph
nodes in the aortopulmonary window can be dissected Technique of Transhiatal Esophagectomy
after identification of the left vagal nerve. The left vagal The operation begins with an abdominal lymph node
nerve is divided between ligatures at the level of the left dissection and gastric mobilization (see “Technique of
main bronchus. The proximal side is carefully moved Open En Bloc Transthoracic Esophagectomy”). Next,
upward with use of the same ligature, thus preventing the tendinous part of the esophageal hiatus is incised
damage to the left recurrent nerve when dissecting the anteriorly or the muscular part is incised circumferentially
aortopulmonary window nodes. The proximal thoracic duct after division of the diaphragmatic vein with ligatures. This
is also ligated and cut at the level of the fourth vertebral ensures removal of any potentially involved parahiatal
body where it crosses from right to left. nodes, but it also enlarges the hiatal opening that facilitates
The abdominal portion of the operation begins with a the lower mediastinal dissection. Placement of appropriate
midline laparotomy and inspection of the peritoneal cavity retractors through the widened esophageal hiatus allows for
and liver. Normally, segments two and three of the liver en bloc dissection of all the fatty tissue and lymph nodes
are mobilized by incising the left triangular ligament with surrounding the lower thoracic esophagus under visual
electrocautery. The flaccid part of the lesser omentum is control as far as possible. Under normal circumstances,
identified and incised in the direction of the right crus. The this can be done up to the level of the inferior pulmonary
right gastric artery is identified, and the lesser omentum veins. To not damage the thoracic duct, care should be
is further mobilized. Then the gastrocolic omentum is taken not to dissect at the right side of the thoracic aorta.
divided, carefully preserving the gastroepiploic arcade. Subsequently, the gastric tube is created and the cervical
This dissection should begin distally at the level of the esophagus is exposed (see “Cervical Anastomosis”). The
pylorus, continuing proximally to include division of the upper thoracic esophagus is delivered into the cervical
short gastric vessels. The short gastric vessels should be wound and it is divided in the neck. A large-bore vein
divided as close as possible to the spleen to preserve as stripper is inserted through the cervical esophagus and
many collateral vessels to the fundus as possible. In this brought to the gastric remnant. After a long tape is tied
fashion, an omental wrap around the future anastomosis to the distal part of the transected esophagus, it is bluntly
can also be created. stripped from the neck toward the abdomen, while the
All of the lymph node–bearing tissue overlying the adhesions between the esophagus and surrounding
proximal border of the hepatic artery and portal vein is structures are manually freed via the widened hiatus. In
removed. This dissection is continued proximally along the lower mediastinum, the vagal nerve trunks that are
the hepatic artery to its origin from the celiac axis. The separated from the esophagus by this maneuver can be
retroperitoneal tissue above the pancreas overlying the divided below the carina with use of scissors. The right
right crus of the diaphragm is dissected medially and lateral attachments are mobilized by a similar maneuver
superiorly to remain attached to the esophagectomy passing the right hand anterior to the esophagus and using
specimen. Attention is then turned to the greater curvature the thumb and index finger to bluntly dissect the right
of the stomach where the gastrocolic omentum is divided. lateral attachments. The tape tied follows the inverting
The gastric fundus is rotated to the right to continue the esophagus from the neck to the abdomen. The esophagus
dissection in the retroperitoneum, removing all of the is everted again and the resection specimen is sent for
node-bearing tissue above the splenic artery and overlying pathologic examination. The tape is now sutured onto
the left crus of the diaphragm. The musculature of the the top of the gastric tube (which has been created at an
diaphragmatic hiatus is then incised (in case of a bulky earlier stage; see previous text). The gastric tube can be
tumor) to meet the incision made in the diaphragm wrapped in a bowel bag or laparoscopic camera bag to
during the thoracic dissection. Often the diaphragmatic facilitate atraumatic passage and can be brought up to the
vein needs to be ligated. Retracting the stomach anteriorly, neck by pulling gently on the tape and pushing the gastric
ample exposure of the celiac axis can be achieved to allow tube into the mediastinum. Care should be taken to avoid
for ligation of the coronary vein (=left gastric vein). After rotation of the gastric tube. A cervical anastomosis can
this, the upper abdominal lymph adenectomy around the subsequently be performed (see “Cervical Anastomosis”).
celiac trunk can be completed. The left gastric artery is
divided at its origin. A Kocher maneuver can be performed RECONSTRUCTION
if needed to allow additional mobility of the stomach. In the far majority of patients undergoing resection for
Reconstruction is preferably performed by creation of esophageal cancer, reconstruction is performed using a
a gastric tube after resection of the gastric cardia. The gastric conduit, where only a single anastomosis is required.
gastric tube is created using a linear stapling device. The The major disadvantages of using the stomach include
staple line should begin on the upper fundus at least 5 cm the almost complete lack of peristaltic activity and the
from the distal limit of the tumor and should continue tendency for persistent reflux into the remaining cervical
to a point along the lesser curvature corresponding to esophagus that is directly connected to the acid-secreting
the fourth or fifth branch of the right gastric artery in stomach. In long-term survivors, this ongoing reflux can
the case of a cervical anastomosis, where more length result in the development of interstitial metaplasia (Barrett)
can be achieved by staying closer to the greater curve in the cervical remnant.23 The need to preserve length may
408 SECTION I Esophagus and Hernia
also result in more limited margins, especially for large sternocleidomastoid muscle. This incision should extend
or very distal tumors that can result in local recurrence. from the sternal notch to a point halfway to the ear lobe.
As a result, when there is extensive involvement of the The omohyoid, sternohyoid, and sternothyroid muscles are
stomach and the esophagus, the use of an antiperistaltic divided laterally, and the jugular vein and carotid sheath
or isoperistaltic left colon interposition is preferred. Also, are lateralized. The middle thyroid vein and inferior
in cases where creation of a (sufficiently oxygenated) thyroid artery are ligated. Dissection is then continued
gastric tube is technically not possible (e.g., history of posteriorly to the esophagus, down to the dissection plane
gastric surgery or aberrant blood supply of the stomach), with the prevertebral fascia, into the thoracic inlet where
reconstruction is performed using a colonic interposition. the dissection plane performed during the thoracotomy is
During TTE, the surgeon can choose between an anas- reached. A dissection plane is then created between the
tomosis at the cervical level or in the chest. In contrast, a esophagus and the trachea. The esophagus is encircled
THE always requires an anastomosis in the neck. Despite with a Penrose drain and the upper thoracic esophagus is
the increased rate of recurrent laryngeal nerve damage, delivered into the neck. The esophagus is divided at the
leakage, and possible stricture formation, some surgeons level of the thoracic inlet and the specimen is removed
prefer a cervical anastomosis during TTE, because of a via the abdomen after tying a tape to the esophagus. The
longer proximal tumor-free margin and a theoretically cervical remnant should not be too long, thus prevent-
reduced morbidity in case of an anastomotic leak.24,25 ing that the anastomosis will ultimately retract into the
The latter is founded on the assumption that a leakage upper chest with a possibly increased risk of intrathoracic
of a cervical anastomosis is more likely to be confined to manifestation in case of leakage.
the neck instead of leaking into the pleural cavity and With use of the tape, which is tied to top of the gastric
mediastinum. However, a meta-analysis on this topic did not tube, the gastric pull-up can be completed. The previously
show differences in pulmonary complications (OR, 0.86; created gastric tube can be wrapped in a plastic bag to
95% CI, 0.13 to 5.59; P = .87) and tumor recurrence (OR, facilitate atraumatic passage to the neck. Care should be
2.01; 95% CI, 0.68 to 5.91; P = .21), which suggests that a taken to avoid excessive tension on the stomach or its
cervical anastomosis after TTE does not decrease the risk gastroepiploic arcade during this maneuver and to avoid
of thoracic complications compared with an intrathoracic twisting of the stomach. The anastomosis is performed
anastomosis.26 Interestingly, two large retrospective studies between the remaining cervical esophagus and the gastric
showed that the risk of intrathoracic manifestations due tube. We prefer to perform an end-to-end anastomosis
to leakage of a cervical anastomosis is significantly less with single-layer running suture. Several nonabsorbable
in patients after THE than in patients who underwent sutures should be placed to normalize the size of the
TTE. This is probably explained by the difference in hiatus to prevent visceral herniation into the thorax. A
mediastinal dissection and pleural resection. After THE, nasogastric decompression tube is then carefully passed
the bilaterally intact parietal pleura may confine infections, as well as a nasojejunal feeding tube. Alternatively, one
which prevents extension to the pleural cavity and mediasti- can choose for a percutaneous jejunal feeding tube.
num.27,28 Notably, these studies were performed before the
introduction of neoadjuvant therapy. Studies comparing INTRATHORACIC ANASTOMOSIS
cervical with intrathoracic anastomoses in patients who In the case of an intrathoracic anastomosis, the proximal
underwent neoadjuvant therapy are lacking. The CROSS part of the esophagus is divided just above the arch of
trial comparing neoadjuvant chemoradiotherapy plus the azygos vein. With care to prevent rotation, the cardia
surgery with surgery alone, in which most anastomoses together with the gastric tube is delivered through the
were performed at cervical level, showed no significant hiatus into the thoracic cavity, and the surgical specimen
difference in leakage rate.29 Nevertheless, preoperative (i.e., esophagus and cardia) is removed. After placement
radiotherapy likely affects anastomotic healing, especially of four to eight sutures around the esophagus (PDS 3.0), a
if the fundus (i.e., the future tip of the gastric tube) was purse-string Prolene 1.0 suture is placed, and after careful
located within the radiation field. Theoretically, the gastric inflation of a 30-mL balloon of a catheter, the diameter
tube can be shorter in case of an intrathoracic anastomosis, of the circular stapler is estimated and the anvil is placed.
with potentially improved oxygenation of the tip and thus Subsequently, the gastrotomy is made at the tip of the
enhanced anastomotic healing. On the contrary, radiation gastric tube, the circular stapling device is introduced, and
damage on the intrathoracic esophageal remnant might an end-to-side anastomosis is created using a 25 mm or
hamper intrathoracic anastomotic healing. This topic is 29 mm circular stapling device. The gastrotomy is closed
currently subject of investigation in an ongoing Dutch with a linear stapler and the linear staple line is oversewn.
randomized trial comparing cervical with intrathoracic A nasogastric tube is passed into the distal stomach. After
anastomosis after neoadjuvant chemoradiotherapy (ICAN completion of the anastomosis, omental tissue is wrapped
trial, Dutch Trial Registry number: NTR4333). around the anastomosis (omentoplasty).
colon are mobilized completely. The left segment of We routinely perform a catheter jejunostomy to provide
the colon to be interposed derives its arterial supply early postoperative enteral feeding and to avoid the need
from the ascending branch of the left colic artery and for parenteral nutrition in the event of postoperative
usually corresponds to the segment extending from the complications such as an anastomotic leak. The jeju-
midtransverse colon to the proximal descending colon. nostomy catheter is removed when the patient is able
This segment is mobilized by dissecting the middle colic to maintain body weight by oral feedings, usually 3 to 4
artery back to its origin from the superior mesenteric weeks postoperatively.
artery where it arises as a single trunk in most patients.
After the middle colic artery and vein have temporarily
been occluded to ensure adequate collateral flow through
COMPLICATIONS
the marginal artery, these vessels are ligated and divided. Despite recent improvements in perioperative manage-
The apex of the arc portended by the vascular pedicle ment, postoperative morbidity and mortality following
is then marked with a suture and the distance from this esophagectomy for cancer remain significant. These are
point to the neck is measured with an umbilical tape. This large, technically demanding operations that are often
tape is used to measure proximally from the first marking performed on patients with compromised cardiopulmo-
stitch to determine the point of transection of the proximal nary function. Nutritional disturbances are also common
colon. The divided colon is then passed through the because of the combined effects of the cancer itself and
bed of the resected esophagus wrapped in a bowel bag, the obstructing mass in the esophagus.
and a single-layer monofilament running anastomosis is Recent audits suggest a hospital mortality rate varying
performed to the remaining cervical esophagus. Traction from 3.5% to 9% in the West.30,31 Complication rates
is gently applied to the colon from within the abdomen varying from 17% to 74% are reported in both open
to eliminate redundancy, and the colon is secured to the and minimally invasive esophagectomy series.32,33 This
left crus of the diaphragm with a nonabsorbable suture. wide range of complication rates can be explained by
The colon is then divided with a linear stapler 5 to 10 cm the variations in definitions of complications and the
below the point where it enters the abdomen. Care should absence of standardization of time periods defining post-
be exercised not to leave too long of an intraabdominal operative deaths.34,35 Accurate comparison of outcomes
segment of colon, as this will result in food retention. The between centers to improve the quality of care requires
mesentery should be divided immediately adjacent to the consistency in definitions and data collection. Therefore
wall of the colon to avoid injury to the vascular pedicle. an international system for defining and recording post-
A single-layered anastomosis is then performed between operative complications associated with esophagectomy
the distally divided colon and the Roux-en-Y jejunal loop, has been developed.36
and colon continuity is restored by a colocolostomy. Complications occurring in a randomized trial compar-
Alternatively, the left colon can be used in antiperistaltic ing open TTE with open THE for esophageal adenocar-
position, which is based on a vascular pedicle of the middle cinoma are summarized in Table 39A.2.37 Pulmonary
colic artery and vein. In this way, the interposed segment complications including pneumonia (defined as isolation
can be longer by making use not only of the descending of a pathogen from a sputum culture and an infiltrate on
colon, but also (part of) the sigmoid colon. chest x-ray) and atelectasis (defined as lobar collapse on
Finally, the right colon can be used including the chest x-ray) are among the most common complications,
ileocecal valve in an isoperistaltic position and again occurring in 57% and 27% of patients who underwent
based on the middle colic vessels. The advantage of this TTE or THE, respectively. These complications can be
technique is that the ileocecal valve will act as an antifreflux minimized by early ambulation and careful attention
mechanism at the proximal anastomosis. to adequate pain control. Prevention of aspiration can
TABLE 39A.2 Postoperative Complications Occurring in 220 Primary Resections for Esophageal Adenocarcinoma in a
Randomized Trial Comparing Transthoracic Esophagectomy and Transhiatal Esophagectomy
Complication Transthoracic Esophagectomy (%) Transhiatal Esophagectomy (%) P Value
Pulmonary complications* 65 (57) 29 (27) <.001
Cardiac complications 30 (26) 17 (16) .10
Anastomotic leakage† 18 (16) 15 (14) .85
Subclinical 8 (7) 9 (8)
Clinical 10 (9) 6 (6)
Vocal-cord paralysis‡ 24 (21) 14 (13) .15
Chylous leakage 11 (10) 2 (2) .02
Wound infection 11 (10) 8 (8) .53
*Pulmonary complications include pneumonia (indicated by isolation of a pathogen from a sputum culture and an infiltrate on chest x-ray) and atelec-
tasis (indicated by lobar collapse on chest x-ray).
†
The definition for subclinical anastomotic leakage was anastomotic leakage seen only on contrast radiography, and clinical anastomotic leakage was
defined as anastomotic leakage resulting in a cervical salivary fistula (all patients had cervical anastomoses).
‡
In most cases, vocal-cord paralysis was temporary.
Data from Hulscher JB, van Sandick JW, de Boer AG, et al. Extended transthoracic resection compared with limited transhiatal resection for adenocar-
cinoma of the esophagus. N Engl J Med. 2002;347:1662–1669.
410 SECTION I Esophagus and Hernia
be achieved by keeping the patient constantly in the more frequently after THE than after TTE. Moreover,
semi-upright position and by meticulous attention to lymph node yield was higher after TTE (mean difference
maintaining a functioning nasogastric tube. When neces- of eight nodes; 95% CI, 1 to 14; P = .02). The results of
sary, a mini-tracheostomy can provide invaluable assistance this meta-analysis should be interpreted with caution,
in clearing retained secretions. because both randomized and nonrandomized studies
Cardiac complications occur in approximately 26% were included. This probably introduced a selection bias
and 16% of TTE and THE patients, with the development in favor of the THE group, because patients with more
of atrial fibrillation accounting for the majority of these advanced tumors probably have been treated preferentially
complications. The shift of body fluids and the extensive via the chest.44 On the other hand, more frail patients
mediastinal dissection that causes a systemic inflammatory may have been offered a THE because of no need for a
response likely play a role in the pathogenesis. Although thoracotomy. Finally, the enhanced short-term recovery
these are generally self-limiting, they do require cardiac after THE could not be confirmed in a large (more than
monitoring and treatment, which can prolong the intensive 17,000 patients), multicenter observational study that
care unit stay. Atrial fibrillation can also, for example, compared TTE with THE; no differences were found in
be caused by anastomotic dehiscence with secondary morbidity and mortality. However, a preference for THE
mediastinitis or by mechanical irritation by a chest tube. in patients with poor performance status probably resulted
For these underlying causes, specific measures are needed. in selection bias in favor of patients who underwent TTE.45
Anastomotic complications occur in 10% to 30% of Proponents of the transhiatal approach explain dif-
patients depending on the definition and the type of ferences in survival by stage that have been consistently
reconstruction performed.38 Most of these leaks can be reported as being due to stage migration. This occurs
managed with local drainage and antibiotic administration when positive nodes in the extended part of the dissec-
if the vascular supply to the reconstruction is adequate. We tion increase pN stage in patients with a more favorable
recommend early endoscopy in any patient who is known or prognosis compared with patients with the same number
suspected to have a substantial leak to exclude potentially of positive nodes after a limited dissection during THE.
life-threatening conduit ischemia, which can be present To address this issue, Altorki et al. have reported outcome
in as many as 14% of patients with an anastomotic leak.39 following en bloc TTE and transhiatal resections performed
in patients with T3N-positive (stage III) disease.46 In this
group of patients, the effect of stage migration was sup-
RESULTS posed to be limited, because all had locally advanced
Long-term survival following esophagectomy depends tumors with lymph node involvement. They reported
on several factors including age, gender, weight loss, 4-year survival of 35% after en bloc resection, which was
histologic subtype, depth of tumor invasion, radicality significantly better than the 11% survival observed after
of the resection, and the number of involved lymph THE. Ultimately, this debate can only be resolved by the
nodes.29,40,41 The impact of surgical approach on long-term completion of a large randomized controlled trial. To date,
survival remains the subject of debate. only one such large trial (HIVEX) has been reported by
In a retrospective analysis from nine high-volume Hulscher et al.37 This trial randomized 220 patients with
centers on 2303 patients (60% adenocarcinoma [AC], AC of the mid-to-distal esophagus or the gastric cardia
40% squamous cell carcinoma [SCC]) who underwent substantially involving the esophagus between THE and
R0 resections, it was shown that a high total number of TTE. By avoiding a thoracotomy, artificial ventilation
resected nodes is an independent prognostic factor of time (1 day after THE vs. 2 days after TTE; P < .001) and
(favorable) survival after primary surgery. The optimal hospital stay (15 days after THE vs. 19 days after TTE; P
threshold for survival benefit was removal of 23 nodes, < .001) were shorter, and pulmonary complications were
and the operation most likely to achieve this number was reported less frequently (27% after THE vs. 57% after
found to be an en bloc transthoracic resection.42 These TTE; P < .001) after THE than after TTE. Nevertheless,
findings are arguments in favor of TTE over THE. In in-hospital mortality was comparable between both groups
contrast, a nonrandomized study by two British high-volume (2% after THE and 4% after TTE; P = .45). Interestingly,
centers showed similar long-term survival after THE and the more extended TTE was not associated with a higher
TTE for patients with SCC (12%) or AC (88%), while percentage of tumor-free resection margins (72% after
hospital stay was significantly shorter after THE.43 This THE vs. 71% after TTE), whereas the median number of
advantage in short-term recovery after THE over TTE resected lymph nodes was 2 times higher after TTE than
without substantially jeopardizing oncologic outcome after THE (median 31 vs. 16; P < .001). This high lymph
was confirmed in a recent meta-analysis of 52 studies node yield did not translate into a significantly better
that included 3389 TTE patients and 2516 THE patients 5-year overall survival (34% after THE and 36% after TTE;
(48% SCC, 52% AC). In addition to the significantly P = .71).47 However, in a subsequent subgroup analysis of
shorter hospital stay (4 days less in patients who underwent patients with a truly esophageal (Siewert type 1) cancer,
THE; 95% CI, 1 to 7; P < .01), THE was associated with and more specifically in patients with a limited number
shorter operation time (85 minutes shorter; 95% CI, 40 (1–8) of positive lymph nodes, an improved long-term
to 129; P < .001), less pulmonary complications (17.3% survival was found after TTE (23% after THE vs. 64%
vs. 21.4%; OR, 1.37; 95% CI, 1.05 to 1.79; P = .02), and TTE; P = .02). Given the post hoc design of this analysis,
lower postoperative mortality (7.2% vs. 10.6%; OR, 1.48; the effect of stage migration on improved survival of TTE
95% CI, 1.20 to 1.83; P < .001). On the other hand, patients cannot be excluded, because more lymph nodes
anastomotic leaks and recurrent nerve palsies occurred were resected after TTE. Furthermore, the relevance of
Surgical Approaches to Remove the Esophagus: Open CHAPTER 39A 411
these results is unclear for patients with SCC (only patients surgery alone suffices in this subgroup of patients. This
with AC were included). The conclusion of the HIVEX is supported by the high rate of radical resections (92%)
trial was that in patients with advanced, truly esophageal in the surgery-alone arm of the French trial. However, the
cancer (Siewert type 1), TTE is the preferred technique generalizability of the FFCD 9901 trial is questionable due
(especially in case of a limited number of positive nodes), to the low case volume of most participating centers, the
while THE suffices in patients with a tumor located at the high toxicity of the nCRT regimen with less sophisticated
esophagogastric junction (EGJ) (Siewert type 2) and in radiation techniques compared with the CROSS trial, and
patients with a poor performance status (especially in case a remarkably high postoperative mortality rate (11.1%).
of pulmonary comorbidities), without clinically suspected Therefore we caution to conclude that patients with
nodes at or above the carina.47 early-stage esophageal cancer should not undergo nCRT.
We believe that in the absence of high-quality evidence
on the specific effect of nCRT on early-stage tumors,
ROLE OF NEOADJUVANT THERAPY the results from the CROSS trial (which also included
Increasingly, the management of esophageal cancer stage II cancers) should be leading.53
has focused nowadays on multimodality therapy, with The CROSS trial and the FFCD 9901 trial included both
neoadjuvant chemotherapy or chemoradiotherapy being AC and SCC. Although nCRT also significantly improves
administered to nearly all patients with locally advanced survival in patients with AC, the maximum benefit of
disease in many centers. The concept of neoadjuvant nCRT is observed in SCC, which is known to be more
therapy in esophageal cancer was spurred by a general radiosensitive than AC.29,49 Three small underpowered
disappointment in the results of primary resections, which randomized trials comprising 119, 75, and 131 patients,
resulted in survival of 35% or less at 5 years.37 respectively, with esophageal AC did not show significant
Many studies have been performed to test the additional differences in survival between nCRT followed by surgery
value of preoperative neoadjuvant therapy to surgical and nCT followed by surgery. Nevertheless, higher rates
resection. A meta-analysis showed that both neoadjuvant of pCR, R0, and ypN0 were found in the nCRT groups,
chemotherapy and neoadjuvant chemoradiotherapy and two of these three trials showed a (nonsignificant)
improve long-term survival.48 Furthermore, this meta- benefit in favor of nCRT.54–57 The optimal neoadjuvant
analysis showed a (nonsignificant) benefit of neoadjuvant treatment for esophageal AC remains undetermined and
chemoradiotherapy (nCRT) over neoadjuvant chemo- is currently investigated in the randomized Neo-AEGIS
therapy (nCT) by comparison of the treatment arms (perioperative MAGIC chemotherapy vs. preoperative
of several trials (hazard ratio [HR] for overall mortality CROSS chemoradiotherapy, in adenocarcinoma of the
for nCRT vs. nCT, 0.88; 95% CI, 0.76 to 1.01; P = .07). esophagus and esophagogastric junction) trial, which is
Unfortunately, direct comparisons are limited, especially likely to be reported in 2021.58
for patients with AC. nCRT has a significant downstaging effect on both
Since the publication of this meta-analysis, the multi- the primary tumor and the regional lymph nodes. In the
center randomized CROSS trial was completed, compar- nCRT-arm of the CROSS trial, a substantial number of
ing nCRT plus surgery with surgery alone in patients patients (29% overall, 49% SCC, 23% AC) did not have
with esophageal or junctional cancer (both SCC and any vital tumor left in the resection specimen. This observa-
AC).29,49 The applied regimen (carboplatin and paclitaxel tion led to the imperative to reconsider the necessity of
with 41.4 Gy concurrent radiotherapy) had low toxicity standard esophagectomy in all patients who undergo
compared with earlier trials that mostly used cisplatin nCRT. Therefore the feasibility of an active surveillance
and fluorouracil. Median survival doubled from 24% in strategy in patients with a clinically complete response
the surgery-alone group to 49% in the nCRT group (HR, (cCR) after nCRT is currently being explored. In this
0.68; 95% CI, 0.53 to 0.88; P = .003), with a 5-year survival so-called SANO (Surgery As Needed in Oesophageal
advantage of 14% (33% vs. 47%). The superior survival cancer patients) approach, surgical resection would be
in the surgery-alone arm of the CROSS trial compared offered only to patients in whom residual disease is highly
with that in earlier randomized trials, indicates that the suspected or proven after nCRT. Before SANO can be
survival benefit can be attributed to improved survival in tested in a prospective clinical trial, we aim to determine
the multimodality arm, and is not due to poor survival in the accuracy of clinical detection of residual disease after
the surgery-alone arm.50,51 Based on these results, nCRT nCRT in the present preSANO trial. 59 Furthermore,
according to the CROSS regimen plus surgery is now the French phase II/III randomized ESOSTRATE trial
considered standard of care in many countries. comparing standard surgery with surgery on demand in
The favorable results of the CROSS trial were not case of recurrence in patients with a cCR after nCRT is
confirmed in a recently completed French randomized currently being initiated (ClinicalTrials.gov identifier:
trial (Fédération Francophone de Cancérologie Digestive NCT02551458).60
[FFCD] 9901 trial) comparing nCRT plus surgery with As outlined previously, the randomized HIVEX trial
surgery alone in stage I and II esophageal cancer patients. comparing THE with TTE for subcarinal AC only included
The applied neoadjuvant regimen consisted of cisplatin patients with primary surgery. In that trial, TTE did not
and fluorouracil with 45 Gy concurrent radiotherapy. improve the rate of tumor-free margins (72% after THE
No differences in 3-year overall survival rate and radical vs. 71% after TTE) but roughly doubled the number of
resection rate were found between both treatment arms.52 resected nodes (median ± standard deviation = 16 ± 9
Based on the FFCD 9901 trial, the standard use of after THE vs. 31 ± 14 after TTE; P < .001). As discussed
nCRT for early-stage tumors can be debated. Possibly, previously, a retrospective international study has shown
412 SECTION I Esophagus and Hernia
that after primary surgery, the number of resected nodes is nCRT has familiarized surgeons with surgical resection
correlated with a favorable long-term survival.42 However, in an irradiated surgical field.
it has been reported that chemoradiotherapy reduces Results of salvage surgery after failed dCRT were analyzed
lymph node yield from within the radiotherapy field.61–63 in a nonrandomized phase II trial.68 Forty-three patients
Importantly, in the patients after primary surgery from were treated with induction CT (5-fluorouracil [5-FU],
the CROSS trial, the total number of resected nodes and cisplatin and paclitaxel) followed by CRT (5-FU and cis-
the number of resected positive nodes were positively platin with concurrent 50.4 Gy). CT scans of the chest and
correlated. However, this positive association completely abdomen, positron emission tomography (PET, optional
disappeared in patients who underwent nCRT. Further- but encouraged), esophagogastroscopy with biopsies,
more, after surgery alone, the total number of removed and endoscopic ultrasound (EUS) were performed after
nodes was positively correlated with overall survival (HR completion of CRT and serially thereafter. Twenty patients
per 10 additionally resected nodes, 0.76; P = .007), which underwent salvage esophagectomy because of residual
corresponds with the earlier retrospective international or recurrent disease without signs of distant metastases.
study.42,64 Interestingly, this positive correlation between One-year overall survival was 71% (95% CI, 54% to 82%).
the number of resected nodes and survival was absent after Nevertheless, a subsequent phase III trial was not initiated,
nCRT (HR, 1.00; P = .98). The randomized design of the because the intended predefined minimal 1-year survival
CROSS trial renders differences between both treatment rate of 77.5% was not achieved. This predefined 1-year
groups unlikely as an explanation for the (disappearance survival rate was deducted from the RTOG database,
of the) association in this post hoc analysis. These results consisting mainly of SCC patients. The proportion of ACs
question the necessity of maximization of surgical lymph in this trial was 73%. Moreover, three CRT-related deaths
node dissection after nCRT, both for prognostication and were reported. Theoretically, elimination or mitigation
for therapeutic purposes. of induction CT from the regimen might have reduced
The same phenomenon was identified in a large retro- treatment-related toxicity and increased the chance of
spective comparison of 307 patients who underwent nCRT achieving the target 1-year survival rate of 77.5%.68
according to CROSS plus surgery and 301 patients who Furthermore, a more recent retrospective propensity
underwent nCT according to MAGIC followed by surgery. matched analysis compared patients undergoing salvage
In the nCRT group, the association between lymph node esophagectomy (n = 308) with patients who underwent
harvest and survival was absent. However, in the nCT neoadjuvant chemoradiotherapy followed by planned
group, extent of lymphadenectomy seemed to be positively esophagectomy (n = 540). In-hospital mortality was
correlated with progression-free survival. Again, these comparable (but high) in both groups (8.4% vs. 9.3%).
data question the necessity for maximization of surgical Differences in postoperative complications were found for
lymph node retrieval specifically after nCRT. However, anastomotic leak (17.2% vs. 10.7%; P = .007) and wound
extended lymphadenectomy seems of importance in infection (18.5% vs. 12.3%; P = .026), which were both
patients who undergo nCT followed by surgery (or surgery more frequent in patients who underwent salvage surgery.
alone).65 At 3-year follow-up, groups had comparable overall (43.3%
These indirect arguments need confirmation in a vs. 40.1%; P = .542) and disease-free survival rates (39.2%
randomized trial comparing TTE with extended lymph- vs. 32.8%; P = .232), suggesting that salvage surgery can
adenectomy and THE with limited lymphadenectomy offer acceptable short- and long-term results in a selected
in patients with (Siewert type 1) esophageal cancer who group of patients.69
undergo nCRT. We believe that such trial should focus
on truly esophageal cancer, and not on junctional cancer,
because it already has been shown that THE suffices in
SUMMARY
junctional cancer if the patients undergo primary surgery, Changes in the diagnosis, evaluation, and pre- and
let alone in patients with junctional cancer who have been postoperative treatment of cancer of the esophagus and
treated with preoperative nCRT. esophagogastric junction have resulted in improved
prognosis for patients with this uncommon, but deadly,
disease. A tailored approach to the management of these
SALVAGE SURGERY patients can now result in an overall 5-year survival of
Definitive CRT (dCRT) is frequently applied in patients about 50%, which is a dramatic improvement compared
with SCC of the proximal part of the esophagus (i.e., with the dismal results reported in the (recent) past.
above the carina) and in patients not fit for surgery. Nevertheless, the optimal surgical approach remains
Although organ preservation is a considerable advantage unclear. The widely applied use of multimodality treatment
in the nonoperative strategy of dCRT, this approach is (especially nCRT) questions the necessity of maximization
associated with high rates (up to 51%) of recurrence or of surgical lymph node retrieval, and the introduction
persistence of locoregional disease.66 In these patients, of MIE might further decrease postoperative morbidity
salvage esophagectomy is an option after failed dCRT with with reduction of especially pulmonary complications.
curative intent. This selective surgery is more demanding However, the lack of high-quality evidence on these topics
than primary esophagectomy. Thanks to centralization of has led to persistence of substantial differences in the
care with improvement in patient selection, in surgical treatment approach between individual institutions. These
technique and in perioperative management, periopera- differences underline the ongoing need for well-designed
tive morbidity and mortality nowadays have substantially clinical trials on specific topics in the field of esophageal
decreased.67 Furthermore, the increased application of cancer surgery.
Surgical Approaches to Remove the Esophagus: Open CHAPTER 39A 413
45. Connors RC, Reuben BC, Neumayer LA, Bull DA. Comparing update of the POET phase III study. J Clin Oncol. 2016;34(suppl;
outcomes after transthoracic and transhiatal esophagectomy: a abstr 4031).
5-year prospective cohort of 17,395 patients. J Am Coll Surg. 2007;205: 58. Keegan N, Keane F, Cuffe S. Neo-AEGIS: a randomized clinical
735-740. trial of neoadjuvant and adjuvant chemotherapy (modified MAGIC
46. Altorki NK, Girardi L, Skinner DB. En bloc esophagectomy improves regimen) versus neoadjuvant chemoradiation (CROSS protocol) in
survival for stage III esophageal cancer. J Thorac Cardiovasc Surg. adenocarcinoma of the esophagus and esophagogastric junction.
1997;114:948-955, [discussion 955-956]. J Clin Oncol. 2014;32(suppl; abstr TPS4145).
47. Omloo JM, Lagarde SM, Hulscher JB, et al. Extended transthoracic 59. Noordman BJ, Shapiro J, Spaander MC, et al. Accuracy of detecting
resection compared with limited transhiatal resection for adenocarci- residual disease after cross neoadjuvant chemoradiotherapy for
noma of the mid/distal esophagus: five-year survival of a randomized esophageal cancer (preSANO Trial): rationale and protocol. JMIR
clinical trial. Ann Surg. 2007;246:992-1000, [discussion 1000-1001]. Res Protoc. 2015;4:e79.
48. Sjoquist KM, Burmeister BH, Smithers BM, et al. Survival after 60. Putora PM, Bedenne L, Budach W, et al. Oesophageal cancer: explor-
neoadjuvant chemotherapy or chemoradiotherapy for resectable ing controversies overview of experts’ opinions of Austria, Germany,
oesophageal carcinoma: an updated meta-analysis. Lancet Oncol. France, Netherlands and Switzerland. Radiat Oncol. 2015;10:116.
2011;12:681-692. 61. Robb WB, Dahan L, Mornex F, et al. Impact of neoadjuvant chemora-
49. Shapiro J, van Lanschot JJ, Hulshof MC, et al. Neoadjuvant chemo- diation on lymph node status in esophageal cancer: post hoc analysis
radiotherapy plus surgery versus surgery alone for oesophageal of a randomized controlled trial. Ann Surg. 2015;261:902-908.
or junctional cancer (CROSS): long-term results of a randomised 62. Taflampas P, Christodoulakis M, Gourtsoyianni S, Leventi K, Melissas
controlled trial. Lancet Oncol. 2015;16:1090-1098. J, Tsiftsis DD. The effect of preoperative chemoradiotherapy on
50. Walsh TN, Noonan N, Hollywood D, Kelly A, Keeling N, Hennessy lymph node harvest after total mesorectal excision for rectal cancer.
TP. A comparison of multimodal therapy and surgery for esophageal Dis Colon Rectum. 2009;52:1470-1474.
adenocarcinoma. N Engl J Med. 1996;335:462-467. 63. Lykke J, Roikjaer O, Jess P. Danish Colorectal Cancer Group. Tumour
51. Tepper J, Krasna MJ, Niedzwiecki D, et al. Phase III trial of trimodal- stage and preoperative chemoradiotherapy influence the lymph
ity therapy with cisplatin, fluorouracil, radiotherapy, and surgery node yield in stages I–III rectal cancer: results from a prospective
compared with surgery alone for esophageal cancer: CALGB 9781. nationwide cohort study. Colorectal Dis. 2014;16:O144-O149.
J Clin Oncol. 2008;26:1086-1092. 64. Talsma AK, Shapiro J, Looman CW, et al. Lymph node retrieval
52. Mariette C, Dahan L, Mornex F, et al. Surgery alone versus chemora- during esophagectomy with and without neoadjuvant chemoradio-
diotherapy followed by surgery for stage I and II esophageal cancer: therapy: prognostic and therapeutic impact on survival. Ann Surg.
final analysis of randomized controlled phase III trial FFCD 9901. 2014;260:786-793.
J Clin Oncol. 2014;32:2416-2422. 65. Markar SR, Noordman BJ, Mackenzie H, et al. Multimodality treat-
53. Shapiro J, van Lanschot JJ, Hulshof MC, van der Gaast A. Effectiveness ment for esophageal adenocaricnoma: multi-center propensity-score
of neoadjuvant chemoradiotherapy for early-stage esophageal cancer matched study. Ann Oncol. 2017;28:519-527.
(letter to the editor). J Clin Oncol. 2015;33:288-289. 66. Cooper JS, Guo MD, Herskovic A, et al. Chemoradiotherapy of locally
54. Stahl M, Walz MK, Stuschke M, et al. Phase III comparison of advanced esophageal cancer: long-term follow-up of a prospective
preoperative chemotherapy compared with chemoradiotherapy in randomized trial (RTOG 85-01). Radiation Therapy Oncology
patients with locally advanced adenocarcinoma of the esophagogastric Group. JAMA. 1999;281:1623-1627.
junction. J Clin Oncol. 2009;27:851-856. 67. Markar SR, Schmidt H, Kunz S, Bodnar A, Hubka M, Low DE.
55. Burmeister BH, Thomas JM, Burmeister EA, et al. Is concurrent Evolution of standardized clinical pathways: refining multidisciplinary
radiation therapy required in patients receiving preoperative che- care and process to improve outcomes of the surgical treatment of
motherapy for adenocarcinoma of the oesophagus? A randomised esophageal cancer. J Gastrointest Surg. 2014;18:1238-1246.
phase II trial. Eur J Cancer. 2011;47:354-360. 68. Swisher SG, Winter KA, Komaki RU, et al. A Phase II study of
56. Klevebro F, Alexandersson von Dobeln G, Wang N, et al. A a paclitaxel-based chemoradiation regimen with selective surgi-
randomized clinical trial of neoadjuvant chemotherapy versus cal salvage for resectable locoregionally advanced esophageal
neoadjuvant chemoradiotherapy for cancer of the oesophagus or cancer: initial reporting of RTOG 0246. Int J Radiat Oncol Biol Phys.
gastro-oesophageal junction. Ann Oncol. 2016;27(4):660-667. 2012;82:1967-1972.
57. Stahl M, Riera-Knorrenschild J, Stuschke M, et al. Preoperative 69. Markar S, Gronnier C, Duhamel A, et al. Salvage surgery after
chemoradiotherapy and the long-term run in curative treatment chemoradiotherapy in the management of esophageal cancer: is it
of locally advanced oesophagogastric junction adenocarcinoma: a viable therapeutic option? J Clin Oncol. 2015;33:3866-3873.
CHAPTER
Surgical Approaches to Remove the Esophagus:
Minimally Invasive 39B
Arianna Barbetta
| Daniela Molena
ABSTRACT
Even if technically demanding and requiring a significant
learning curve, the feasibility and safety of minimally
invasive esophagectomy have been proven. Moreover,
several benefits have been reported with minimally invasive
surgery over the open approach, such as postoperative
pain reduction, faster recovery, and decrease of cardiopul-
monary complications, blood loss, and length of stay. The
short and long-term oncologic outcomes after minimally
invasive esophagectomy are similar to those observed with
open approaches. Minimally invasive esophagectomy is
a valid alternative approach to open esophagectomy for
both esophageal benign disease and cancer.
KEYWORDS
Minimally invasive esophagectomy
esophageal cancer
esophagectomy
MI-Ivor Lewis esophagectomy
lymphadenectomy
intrathoracic anastomosis
surgical outcomes
pulmonary complication
416 SECTION I Esophagus and Hernia
Hepatic artery
Azygos vein
FIGURE 39B.3 The esophagus is completely mobilized to the The EEA (Covidien; Minneapolis, Minnesota) stapler
thoracic inlet and divided with an endoscopic stapler. The is introduced through an opening made in the proximal
esophageal stump is completely freed up from the trachea and gastric conduit. The conduit is pulled over the stapler to
the other mediastinal structures. allow the piston to come out next to the greater curvature
of the stomach in the area chosen for the anastomoses.
The anvil and the stapler are engaged, and the stapler is
midesophagus or higher, the lymphadenectomy is extended fired to complete the anastomosis. We most often use the
to the paratracheal nodal stations. A cervical anastomosis largest Orvil (25 mm), and due to the esophageal thickness
is also considered in these cases. we prefer using the green stapler (4.8-mm staples). Using
The gastric conduit is then pulled up into the chest a linear stapler, the excess gastric tip is resected, leaving
and completely divided from the esophagus with a linear at least 1 cm of tissue between this staple line and the
stapler. It is extremely important that the gastric tube anastomosis. To protect the anastomosis posteriorly and
has a proper orientation to avoid spiraling or twisting of avoid fistulization into the airways in case of leakage, the
the conduit. The inferomedial port site is extended up greater fat pad is positioned between the conduit and
to approximately 3 cm and a wound protector is placed the airways. Finally, to avoid abdominal organ herniation
to allow the retrieval of the specimen and the introduc- into the chest, the gastric conduit is bound to the right
tion of a circular stapler in the chest. The specimen is diaphragm by an interrupted suture.
removed into a bag through the wound protector and Transthoracic end-to-side anastomosis using a circular stapler:
sent for pathologic analysis of the gastric margin. The With this technique, a 25- or 28-mm anvil is placed into
operation is completed with an intrathoracic anastomosis the esophageal stump from the operating field, and the
(Fig. 39B.4). stump is tightly closed around the anvil with a purse string.
Several different intrathoracic anastomoses have been The anastomosis is then completed in the same fashion
performed with MIE; these include both handsewn and as explained earlier.
stapled techniques. Transthoracic circular, transoral Intrathoracic side-to-side anastomosis using a linear stapler:
circular, and side-to-side linear stapled are among the The proximal stump of the esophagus and the gastric
most common mechanical anastomoses. conduit are aligned with the esophagus coming over the
Transoral circular stapler: This is our preferred approach. stomach. The esophageal stump can be left completely
The Orvil is inserted through the patient’s mouth into the open or closed with a stapler. The stapler is introduced
esophageal stump. The orogastric tube connected with through a small gastrostomy placed next to the greater
the anvil is passed through a small opening next to the curve, approximately 5 or 6 cm away from the top of the
staple line and removed through one of the trocars. After gastric conduit. If the esophagus was divided with a stapler,
transposing the conduit into the posterior mediastinum, a small esophagostomy is performed in the middle of the
we use fluorescence imaging and indocyanine green esophageal stump to introduce the stapler. The side-to-side
to evaluate the microvascular perfusion to target our anastomosis is then completed with an Endo GIA stapler.
anastomoses. Correct assessment of the gastric conduit The common enterotomy can be closed with a linear
length is also necessary to avoid either an excessive tension stapler (a longer anastomosis is required in this case) or
on the anastomosis or twisting and folding of the distal with interrupted stitches or running suture.
stomach above the diaphragm and impairment in gastric A stapled anastomosis is often preferred with the Ivor
emptying.16 Lewis approach and has been shown to be associated with
418 SECTION I Esophagus and Hernia
lower incidence of stricture (especially with the linear to the proximal end. After the specimen is retrieved from
stapling).17 The circular anastomosis is usually quicker to the abdomen, the gastric conduit is pulled up into the
perform and does not require a long esophageal stump.18 neck through the posterior mediastinum and anastomosed
Although with distal esophageal cancer the proximal to the cervical esophagus. The anastomoses can be done
margin of resection is often not a problem to obtain, we with handsewn or stapled techniques (usually linear
prefer to place the anastomosis above the azygos vein so stapler).
that there is not much redundant stomach within the The minimally invasive approach as compared
abdomen, which can predispose to severe gastroesophageal with open THE offers a superior visualization of the
reflux. mediastinum. The magnified visualization by the lapa-
For patients with a proximal thoracic tumor located roscopic camera associated with a less “blind” dissection
near the airways or with extensive Barrett esophagus or may account for decreased blood loss and transfusion
if there is a concern with obtaining a proximal margin rates.22,23 Decreased rates of respiratory complication,
free of tumor, a modified McKeown esophagectomy or a shorter hospital and intensive care unit stay, faster
three-hole esophagectomy with a cervical anastomosis return of bowel function, and lower costs have also been
may represent a better option. reported.23,24
24. Bhayani NH, Gupta A, Dunst CM, et al. Esophagectomy with tho- 37. Yibulayin W, Abulizi S, Lv H, et al. Minimally invasive oesophagectomy
racic incisions carry increased pulmonary morbidity. JAMA Surg. versus open esophagectomy for resectable esophageal cancer: a
2013;148(8):733-738. meta-analysis. World J Surg Oncol. 2016;14:304.
25. Suzuki Y, Urashima M, Ishibashi Y, et al. Hand-assisted laparoscopic 38. Sihag S, Kosinski S, Gaissert HA, et al. Minimally invasive versus
and thoracoscopic surgery (HALTS) in radical esophagectomy with open esophagectomy for esophageal cancer: a comparison of early
three-field lymphadenectomy for thoracic esophageal cancer. Eur J surgical outcomes from the Society of Thoracic Surgeons national
Surg Oncol. 2005;31:1166-1174. database. Ann Thorac Surg. 2016;101:1281-1289.
26. Huang L, Onaitis M. Minimally invasive and robotic Ivor Lewis 39. Dantoc MM, Cox MR, Eslick GD. Does minimally invasive esopha-
esophagectomy. J Thorac Dis. 2014;6(S3):S314-S321. gectomy (MIE) provide for comparable oncologic outcomes to
27. Tapias LF, Mathisen DJ, Wright CD, et al. Outcomes with open and open techniques? A systematic review. J Gastrointest Surg. 2012;16(3):
minimally invasive Ivor Lewis esophagectomy after neoadjuvant 486-494.
therapy. Ann Thorac Surg. 2016;101(3):1097-1103. 40. Dolan JP, Kaur T, Diggs BS, et al. Impact of comorbidity on
28. Maas KW, Cuesta MA, van Berge Henegouwen MI, et al. Quality outcomes and overall survival after open and minimally invasive
of life and late complication after minimally invasive compared to esophagectomy for locally advanced esophageal cancer. Surg Endosc.
open esophagectomy: results of a randomized trial. World J Surg. 2013;27(11):4094-4103.
2015;39(8):1986-1993. 41. Palazzo F, Rosato E, Chaudhary A, et al. Minimally invasive
29. Luketich JD, Pennathur A, Franchetti Y, et al. Minimally invasive esophagectomy provides significant advantage compared with open
esophagectomy: results of a prospective phase II multicenter trial—the or hybrid esophagectomy for patient with cancer of the esophagus
Eastern Cooperative Oncology Group (E2202) study. Ann Surg. 2015; and gastroesophageal junction. J Am Coll Surg. 2015;4:672-679.
261(4):702-707. 42. Shen Y, Zhang Y, Tan L, et al. Extensive mediastinal lymphadenectomy
30. Biere SS, van Berge Henegouwen MI, Maas KW, et al. Minimally during minimally invasive esophagectomy: optimal results from a
invasive versus open oesophagectomy for patients with oesophageal single center. J Gastrointest Surg. 2012;16:715-721.
cancer: a multicenter, open-label, randomised controlled trial. Lancet. 43. Sihag S, Wright CD, Wain JC, et al. Comparison of perioperative
2012;379:1887-1892. outcomes following open versus minimally invasive Ivor Lewis
31. Lv L, Hu W, Ren Y, et al. Minimally invasive esophagectomy versus oesophagectomy at a single, high-volume centre. Eur J Cardiothorac
open esophagectomy for esophageal cancer: a meta-analysis. Onco Surg. 2012;42:430-437.
Targets Ther. 2016;9:6751-6762. 44. Sudarshan M, Ferri L. A critical review of minimally invasive esopha-
32. Gurusamy KS, Pallari E, Midya S, et al. Laparoscopic versus open gectomy. Surg Laparosc Endosc Percutan Tech. 2012;22(4):310-318.
transhiatal oesophagectomy for oesophageal cancer. Cochrane Database 45. Sundaram A, Geronimo JC, Willer BL, et al. Survival and quality
Syst Rev. 2016;(3):CD011390. of life after minimally invasive esophagectomy: a single surgeon
33. Wang H, Yaxing S, Mingxiang F, et al. Outcomes, quality of life, experience. Surg Endosc. 2012;26(1):168-176.
and survival after esophagectomy for squamous cell carcinoma: 46. Briez N, Piessen G, Bonnetain F, et al. Open versus laparoscopically-
a propensity score-match comparison of operative approaches. J assisted oesophagectomy for cancer: a multicenter randomized
Thorac Cardiovasc Surg. 2015;149:1006-1015. controlled phase III trial—the MIRO trial. BMC Cancer. 2011;11:
34. Luketich J, Alvero-Rivera M, Buenaventura PO, et al. Minimally 310.
invasive esophagectomy. Outcomes in 222 patients. Ann Surg. 47. Avery KN, Metcalfe C, Berrisford R, et al. The feasibility of a random-
2003;238:486-495. ized controlled trial of esophagectomy for esophageal cancer—the
35. Mao T, Fang W, Gu Z, et al. Comparison of perioperative outcomes ROMIO (Randomized Oesophagectomy: Minimally Invasive or
between open and minimally invasive esophagectomy for esophageal Open) study: protocol for a randomized control trial. Trials. 2014;15:
cancer. Thorac Cancer. 2015;6:303-306. 2000.
36. Maas KW, Biere SS, Van Hoogstraten IM, et al. Immunological 48. Straatman J, van der Wielen N, Cuesta MA, et al. Minimally invasive
changes after minimally invasive or conventional esophageal resection versus open esophagectomy resection. Three-year follow-up of the
for cancer. A randomized trial. World J Surg. 2014;38(1):131-137. previously reported randomized controlled trial: the TIME trial.
doi:10.1007/s00268-013-2233-0. Ann Surg. 2017;266(2):232-236.
CHAPTER
Surgical Approaches to Remove the Esophagus:
Vagal-Sparing 39C
Steven R. DeMeester
T
o date, no therapy has been proven superior to and in patients with esophageal cancer limited to the
esophagectomy for the cure of patients with early- mucosa. Importantly, a vagal-sparing procedure is only
stage esophageal cancer. The primary goal of surgery applicable to patients with intramucosal tumors and no
is complete (R0) resection of the tumor to maximize the evidence of lymphadenopathy since preserving the vagus
opportunity for cure and minimize the incidence of local nerves precludes the ability to perform an adequate
recurrence. However, with early-stage disease and a high lymphadenectomy along the left gastric artery and in the
likelihood of long-term survival, there is an increasing periesophageal mediastinal tissues. Therefore a biopsy
focus on postesophagectomy quality of life, especially since showing cancer in an area of nodularity or ulceration
there are endoscopic alternatives for these early lesions. requires initial endoscopic resection to confirm that the
This has prompted us to scale back the extent of the tumor is limited to the mucosa.3 Submucosal tumors
resection and preserve the vagal nerves to try to provide have a significant risk for lymph node metastases and
the benefits of complete resection, while minimizing tumor invasion into this layer is a contraindication for
some of the morbidity associated with esophagectomy in a vagal-sparing approach. Relative contraindications to
appropriate candidates. a vagal-sparing esophagectomy include the presence of
an esophageal stricture, history of caustic injury to the
esophagus, or prior antireflux or esophageal surgery
WHY A VAGAL-SPARING ESOPHAGECTOMY? (repair of perforation or congenital trachea-esophageal
An esophagectomy is a major operation associated with fistula) since in these circumstances mediastinal scaring
significant perioperative and long-term physiologic may prohibit safe stripping of the esophagus or may lead
alterations. During the procedure, the dissection, to vagal disruption even if the stripping is accomplished
typically involving the mediastinum and the abdomen, safely. Further, diabetes or evidence of impaired gastric
leads to extensive third spacing and volume shifts in emptying should be considered a relative contraindication
the perioperative period. These volume shifts frequently for a vagal-sparing procedure using a colon interposition
produce hemodynamic alterations and in some patients to the intact stomach. Lastly, prior gastric surgery such as
cardiopulmonary compromise. Later, the gastrointestinal a pyloroplasty may preclude an advantage to preserving
alterations associated with esophagectomy and reconstruc- the vagal nerves, although even in this setting avoidance
tion often include dumping, diarrhea, early satiety, and of postvagotomy diarrhea may be a sufficient reason to
gastroesophageal reflux symptoms. A laparoscopic vagal- spare the vagus nerves, if possible.
sparing esophagectomy minimizes the dissection associated
with an esophagectomy, since the esophagus is stripped
out of the mediastinum without formal dissection. In
SURGICAL APPROACH
addition, many of the gastrointestinal alterations associated A vagal-sparing esophagectomy with gastric pull-up has
with an esophagectomy are secondary to division of the some similarities to a transhiatal operation, except the
vagus nerves, and vagal preservation minimizes dumping, esophagus is stripped out of the mediastinum and no
diarrhea, and depending on the type of reconstruction mediastinal or transhiatal dissection is done, and the
early satiety and reflux symptoms compared with other left gastric arterial trunk is preserved, along with vagal
types of esophagectomy and reconstruction.1,2 Lastly, the innervation to the pylorus. The operation commences
technique for a vagal-sparing esophagectomy with gastric in the abdomen, and with a minimum of dissection, the
pull-up allows preservation of the left gastric arterial hiatus is opened and the anterior and posterior vagal
trunk and branches to the pylorus. This improves the trunks are encircled with a vessel loop. The vagus nerves
perfusion of the proximal portion of the graft and may are retracted gently toward the patient’s right, and the
reduce anastomotic leaks and stenosis. gastroesophageal fat pad is dissected, beginning on the
left of the esophagus and stomach, such that it allows
the anterior vagus nerve to be brought well over to the
INDICATIONS FOR A VAGAL-SPARING right of the esophagus. Failure to do this step will lead in
ESOPHAGECTOMY most cases to inadvertent injury of the anterior vagus nerve
during the subsequent steps of the procedure. Once the
A vagal-sparing procedure should be considered in any anterior vagus is safely over to the right of the esophagus,
patient with a benign process such as end-stage achalasia a highly selective vagotomy is performed starting just above
or gastroesophageal reflux disease, in patients with high- the crow’s foot near the antrum of the stomach. This is
grade dysplasia in squamous mucosa or Barrett esophagus, necessary if the stomach is to be used as the esophageal
421
Surgical Approaches to Remove the Esophagus: Vagal-Sparing CHAPTER 39C 421.e1
ABSTRACT
To date no therapy has been proven superior to esophagec-
tomy for the cure of patients with early-stage esophageal
cancer. The primary goal of surgery is complete (R0) resec-
tion of the tumor in order to maximize the opportunity
for cure and minimize the incidence of local recurrence.
However, with early-stage disease and a high likelihood
of long-term survival, there is an increasing focus on
postesophagectomy quality of life, especially since there
are endoscopic alternatives for these early lesions. This has
prompted us to scale back the extent of the resection and
preserve the vagal nerves to try to provide the benefits of
complete resection while minimizing some of the morbidity
associated with esophagectomy in appropriate candidates.
KEYWORDS
Esophageal cancer, Barrett esophagus, esophagectomy,
vagus nerves
422 SECTION I Esophagus and Hernia
replacement, and is beneficial with a colon interposition mediastinum after the esophagus has been removed. The
to reduce gastric acidity and the potential for ulceration esophagus comes out inverted with the mucosa external
in the colon graft. The highly selective vagotomy precisely to the muscular wall similar to taking off a sock inside-out.
follows the lesser curve of the stomach up to the point In general, bleeding is minimal, and very little force is
where the distal esophagus is reached and the vagal nerve required to pull the esophagus out. Resistance should
trunks are completely separated from the esophagus. raise concern, and excessive resistance should prompt
This dissection is facilitated by sequential grasping of the conversion to a transhiatal procedure.
stomach with Babcock clamps along the lesser curve, and Importantly, in patients with high-grade dysplasia or
by using an advanced energy source in this very vascular intramucosal cancer, all layers of the esophagus are stripped
area. Avoidance of a hematoma or bleeding during this out so as not to inadvertently leave any dysplastic mucosa
dissection is critical to prevent unintended injury to the or tumor behind. However, in patients with achalasia, only
distal vagal branches. the mucosa need be stripped out. This is accomplished in
At this point, the gastroesophageal junction should a similar fashion, except a cervical esophageal myotomy
be completely exposed and the lesser curve above the is made and the mucosa alone is encircled with a heavy
crow’s foot skeletonized. If the stomach is to be used tie, leaving the remaining muscular wall of the esophagus
for esophageal replacement, then the greater curve is intact. After carefully securing the mucosal tie around the
mobilized in the same fashion as for a standard gastric vein stripper, the mucosa can be stripped out from below,
pull-up. However, if the colon is to be used then there leaving the muscular wall of the esophagus in place. This
is no need to mobilize the greater curve completely. works nicely through a gastrotomy high on the anterior
Instead, the omentum is detached from the transverse fundus of the stomach. The mucosa is stapled off just
colon and a window created near the left crus by dividing distal to the squamocolumnar junction, and the anterior
the most proximal one or two short gastric and posterior gastrotomy is closed.
pancreaticogastric vessels. This creates a passage from the The next step is to dilate the mediastinal tract to prevent
lesser sac to the hiatus for the colon graft. The colon is constriction of the graft. I sequentially dilate the tract using
mobilized in standard fashion based on the ascending a 90 cc balloon Foley catheter progressively filled with
branch of the left colic artery whenever possible.4 The saline and pulled up through the mediastinum. Typically
necessary length of colon is marked out by measuring two to three passes are made to ensure an adequate tract
the distance from the tip of the left ear to the xiphoid is created. This is particularly important in patients who
anteriorly with an umbilical tape, and then marking a have a normal-caliber esophagus at the time of stripping.
similar distance on the colon starting from the point The graft can then be brought up through the posterior
where the left colic vessels tether the graft and going mediastinal tract.
proximally. The colon can then be divided and placed When a gastric pull-up is being used, the stomach is
in the pelvis for later use. tubularized in standard fashion, leaving the crow’s foot
Next, attention is directed to the left neck. The esopha- intact. The gastric tube is then pulled up through the
gus is exposed, and after placing a Penrose drain around posterior mediastinum and an esophagogastric anastomosis
the esophagus to facilitate traction, blunt dissection is is constructed in standard fashion. The vascular supply
accomplished with a finger to free the upper mediastinal of the gastric tube is typically excellent, since the left
portion of the esophagus, a nasogastric tube is inserted, gastric artery has been preserved and only the branches
and the esophagus is irrigated with a dilute povidone-iodine to the skeletonized lesser curve region were divided. This
(Betadine) solution to reduce mediastinal contamination usually leaves several branches intact to the antrum, and in
during the subsequent stripping procedure. The nasogastric combination with preserved right gastric and gastroepiploic
tube is then removed. Next, a gastrotomy is made near arteries, this leads to excellent graft perfusion in most
the gastroesophageal junction, or alternatively the cardia patients. After completing the cervical anastomosis, the
is divided with a stapler and a small portion of the staple graft is gently pulled into the abdomen to eliminate
line is opened to provide access to the esophageal lumen. redundancy and sutured to the crura to prevent herniation
A standard vein stripper is then passed retrograde up of abdominal organs into the mediastinum. At this point,
the esophagus and brought out the anterior wall of the the operation is complete with the exception of passing
cervical esophagus, as distally as possible. The esophagus a nasogastric tube and placing a feeding jejunostomy
is ligated distal to the exit site of the vein stripper in the tube. Since the antral innervation has been preserved,
neck using a heavy suture, and the cervical esophagus no pyloroplasty should be performed.
is divided at the site where the vein stripper comes out. When a gastric pull-up is planned, the vagal-sparing
The divided distal end of the esophagus is then suture procedure is readily adapted to a fully laparoscopic
ligated and tied securely. The use of several Endoloops approach. The gastric mobilization as well as the highly
can facilitate secure ligation of the esophagus around selective vagotomy are straightforward laparoscopic
the vein stripper head. This is a critical step since, if procedures. I have found that the use of a 4-cm incision
the ligatures slip, the vein stripper will merely pull out, in the midline with placement of a hand port facilitates
leaving the partially stripped esophagus somewhere in stripping the esophagus out (via the hand port) and
the mediastinum. After changing the vein stripper to subsequent dilatation of the mediastinal tract. The graft
the large head, the esophagus is inverted on itself by is pulled up attached to a chest tube, and the cervical
pulling the vein stripper from below. It is useful to leave esophagogastric anastomosis is accomplished in standard
a long umbilical tape tied to the distal end of the cervical fashion. Similar to an open procedure, the gastric tube
esophagus to provide access to the tract in the posterior should be sutured to the left crus to prevent torsion of
Surgical Approaches to Remove the Esophagus: Vagal-Sparing CHAPTER 39C 423
the graft or herniation of abdominal organs into the A stapled cologastric anastomosis is then done to the
posterior mediastinum. proximal posterior fundus, using a 75-mm GIA stapler,
When a colon graft is used with preservation of the and a nasogastric tube is guided into the stomach. Next,
intact, innervated stomach, there are several important the colocolostomy is accomplished in standard fashion,
technical considerations. First, only the cardia immediately with care taken to avoid traction on the left colic vessels
below the gastroesophageal junction is excised, leaving the or the marginal artery supplying the graft. Typically this
remaining stomach in place. A highly selective vagotomy requires that the right colon be brought up into the left
is performed along the lesser curvature to reduce acid upper quadrant. Lastly, the mesenteric defects are closed
secretion and provide protection from the development and a feeding jejunostomy is placed.
of cologastric anastomotic ulcers. There is no need to
do an extensive mobilization of the greater curvature.
Instead, only the proximalmost one to two short gastric
CONCLUSION AND SUMMARY
vessels, along with the posterior pancreaticoduodenal A vagal-sparing esophagectomy is the ideal operation for
vessels, are divided, so that there is an approximately high-grade dysplasia or cancer confined to the mucosa,
10 cm window created near the left crus of the diaphragm. and in patients with end-stage benign disease. It is nearly
Importantly, the colon graft is passed up posterior to the always a “one and done” therapy that eliminates the
stomach through this window, into the hiatus, and then diseased mucosa and the need for further interventions,
up through the posterior mediastinum. In patients with and offers reduced side effects and better functional
achalasia, where only the mucosa was stripped out through outcome compared with other types of esophagectomy.
an anterior gastrotomy, the entire muscular tube of the A vagal-sparing procedure is only an option with patients
native esophagus remains intact, and a sufficient-sized at low risk for lymph node metastases, since no formal
hole must be cut into the residual muscular tube along lymphadenectomy is performed. Therefore any nodules
the left lateral aspect near the hiatus to allow the colon or lesions in a patient with Barrett or squamous cell
graft to be pulled up inside. If all layers of the esophagus cancer must first undergo endoscopic resection to confirm
have been stripped out, as in patients with high-grade if they are malignant and to pathologically determine
dysplasia or intramucosal cancer, then that issue does not the depth of invasion. Long-term good outcome can be
exist since the muscular wall of the esophagus is gone and achieved using either the tubularized stomach with pyloric
only the mediastinal tract is present at the hiatus. The innervation maintained or a colon interposition to the
esophago-colo anastomosis is done either with a stapled intact, innervated stomach for esophageal replacement.
or hand-sewn technique in an end-to-end fashion. If the
muscular tube of the esophagus has been preserved, it
can be pulled up like a sheath to cover the anastomosis REFERENCES
proximally. The colon is then pulled firmly down into 1. Banki F, Mason RJ, DeMeester SR, et al. Vagal-sparing esophagectomy: a
the abdomen to eliminate any redundancy and sutured more physiologic alternative. Ann Surg. 2002;236(3):324-335; discussion
335–336.
to the left crus of the diaphragm to prevent twisting of 2. Peyre CG, DeMeester SR, Rizzetto C, et al. Vagal-sparing esopha-
the graft or herniation of abdominal contents into the gectomy: the ideal operation for intramucosal adenocarcinoma and
mediastinum. In particular, sutures should be placed Barrett’s with high-grade dysplasia. Ann Surg. 2007;246:665-674.
between the colon graft and the posterior aspect of the 3. Greene CL, Worrell SG, Attwood SE, et al. Emerging concepts for the
hiatus near crural decussation since herniation can occur endoscopic management of superficial esophageal adenocarcinoma.
J Gastrointest Surg. 2016;20(4):851-860.
underneath the colon graft if these sutures are omitted. 4. Maish MS, DeMeester SR. Indications and technique of colon and
The colon is divided approximately 10 to 15 cm distal jejunal interpositions for esophageal disease. Surg Clin North Am.
to the hiatus, taking care not to injure the vascular arcade. 2005;85(3):505-514.
CHAPTER
T
he use of minimally invasive esophagectomy (MIE) mega-esophagus, refractory stricture, intractable reflux
has increased over the past several years. The term resistant to surgical interventions, and multiple failed
minimally invasive can refer to performing either or hiatal hernia operations.
both the thoracic and abdominal phases of the operation
with either laparoscopic or robotic assistance. Transhiatal
esophagectomy is another form of MIE that avoids a chest
EQUIPMENT
incision. Recent studies have demonstrated that MIE has The Da Vinci Surgical System is currently the only US
benefits with decreases in blood loss, chest tube duration, Food and Drug Administration (FDA)-approved robotic
length of stay, and respiratory complications versus open system for surgery. The surgeon sits at a console some
esophagectomy1–4 and maybe even reduces cost.5 Melvin distance from the patient, who is positioned on an
et al. were the first to report robotic esophagectomy in operating table close to the robotic unit with its four
2002.6 Since then, the use of robotic technology for either/ robotic arms. The robotic arms incorporate remote center
both the abdominal and thoracic phases of the operation, technology, in which a fixed point in space is defined,
whether a transhiatal, Ivor Lewis, or modified McKeown and about it the surgical arms move so as to minimize
approach is taken, has become increasingly common. stress on the thoracic or abdominal wall during manipula-
tions. The small proprietary Endowrist instruments attached
INDICATIONS to the arms are capable of a wide range of high-precision
movements. These are controlled by the surgeon’s hand
Most candidates for esophagectomy are also candidates for movements via “master” instruments at the console. The
attempted MIE and therefore also candidates for robotic “master” instruments sense the surgeon’s hand movements
esophagectomy. There are few specific contraindications and translate them electronically into scaled-down micro-
for the use of robotic technology. The need to perform movements to manipulate the small surgical instruments.
an en bloc resection of aorta or intrathoracic trachea or Hand tremor is filtered out by a 6-Hz motion filter. The
carina along with the esophagectomy, which has been surgeon observes the operating field through console
safely applied to selected patients, would generally be binoculars. The image comes from a maneuverable high-
considered a contraindication to robotic esophagectomy.7,8 definition stereoscopic camera (endoscope) attached to
Prior thoracic or abdominal surgery can make a robotic one of the robot arms. The console also has foot pedals
approach more challenging due to the presence of that allow the surgeon to engage and disengage different
adhesions, but lysis of adhesions can be performed to instrument arms, reposition the console “master” controls
permit its use. Comorbidities or poor functional status without the instruments themselves moving, and activate
that would otherwise make patients suboptimal candidates electric cautery. A second optional console allows tandem
for esophagectomy generally would also apply to offering surgery and training. Da Vinci currently offers both the
robotic esophagectomy, although robotic esophagectomy Xi and Si systems. The Xi system is newer and features an
may permit surgeons to offer esophagectomy to somewhat overhead beam that permits rotation of the instrument
older and sicker patients by decreasing the perioperative arms, allowing for greater flexibility in terms of direction
complication rate (especially respiratory complications).9 of approach of the robot to the patient. Compared with
However, caution in patient selection should be applied, as the Si, the Xi also has thinner instrument arms, longer
the physiologic effects of complications such as anastomotic instruments themselves, and the option to switch the
leak and chylothorax remain significant regardless of camera to any arm/port.
whether they occur in a robotic or open esophagectomy.
Early-stage (T1a and early T1b) esophageal cancers can
be managed with endoscopic mucosal resection (EMR).
PREOPERATIVE EVALUATION
Generally, if a lesion is not amenable to EMR or is T1b A thorough history and physical should be performed,
or deeper on final pathologic analysis, esophagectomy focusing on key points such as Barrett esophagus, gas-
may be considered. If EMR for early-stage esophageal troesophageal reflux disease, motility disorders such as
cancer is performed in the context of Barrett esophagus, achalasia, prior surgeries, functional status, and cardiac
radiofrequency ablation (RFA) to promote regression and respiratory comorbidities. Smoking cessation should
of Barrett should also be considered. Patients with per- be encouraged and alcohol use should be noted to screen
sistent high-grade dysplasia following attempted RFA for cirrhosis and warn of possible withdrawal issues in the
are also candidates for esophagectomy. Benign indica- perioperative period. Patients undergoing esophagec-
tions for esophagectomy include end-stage achalasia or tomy for neoplasm should receive a whole-body positron
424
Surgical Approaches to Remove the Esophagus: Robotic CHAPTER 39D 424.e1
ABSTRACT
Robotic esophagectomy offers similar benefits to the
patient as minimally invasive esophagectomy (IE) through
thoracoscopic and/or laparoscopic techniques. We con-
sider any patient that is a candidate for esophagectomy
to be a candidate for robotic esophagectomy as well.
Robotic techniques can be applied to Ivor Lewis, modified
McKeown, or transhiatal esophagectomies. We describe
our technique for the abdominal and thoracic phases
(performed robotically), and cervical phase, of these
operations. Advantages of using robotics for esophagectomy
include the ability of the surgeon to control the camera
and retraction rather than relying on an assistant, better
surgeon ergonomics for what can be long and complex
cases, the simplicity of using the platform for visualization
of real-time perfusion with indocyanine green (ICG) dye,
and greater ease of performing an anastomosis in the
chest. Operative times, perioperative results, morbidity,
and mortality have been comparable to non-robotic MIE.
KEYWORDS
Esophagectomy; esophageal cancer; robotic surgery;
minimally invasive surgery
Surgical Approaches to Remove the Esophagus: Robotic CHAPTER 39D 425
emission tomography (PET)-computed tomography (CT) to induction therapy; however, this did not translate to
scan to evaluate for possible metastatic disease, unless this survival.13 Chiu et al., though, found that delayed surgery
is obvious from chest/abdominal CT scans alone. If fairly (defined as >8 weeks after chemoradiation) was associated
convincing combined radiologic and clinical evidence with decreased 5-year survival for patients with squamous cell
exists for metastatic disease (e.g., weight loss, widespread carcinoma that demonstrated a complete clinical response.14
adenopathy or liver/lung nodules), biopsy confirmation
of metastatic disease may not be necessary. However,
single-site M1 disease should likely be confirmed with tissue
CHOICE OF OPERATION
diagnosis. Location of tumor, synchronous lesions, and The abdominal and/or thoracic phase of the esophagec-
presence/extent of Barrett esophagus should be noted tomy can be performed with robotic assistance. Choice
on the preoperative endoscopy. Tumors extending into of type of esophagectomy (Ivor Lewis, McKeown, or
the proximal stomach may require a partial gastrectomy transhiatal) can be surgeon-dependent, with some prefer-
and different reconstructive approach; tumors in the ring a neck anastomosis due to the decreased incidence of
mid-esophagus should generally be approached via a mediastinal leaks, and others preferring a chest anastomosis
McKeown type operation rather than Ivor Lewis. An due to the risk of recurrent laryngeal nerve injury. Loca-
adequate margin may be difficult to achieve for tumors tion of the tumor may dictate this decision; for instance,
in the proximal one-third of the esophagus; these patients a midthoracic tumor is best suited for resection of the
are better suited for definitive chemoradiation, although entire intrathoracic esophagus with a neck anastomosis.
in some centers, laryngoesophagectomy may be an option.
Some investigators have suggested that patients with
preoperative dysphagia may not need an endoscopic TECHNICAL DETAILS
ultrasound (EUS) given that 90% of them had T3 to T4
disease, a finding that has been corroborated by others.10,11 ABDOMINAL PHASE
However, although the presence of symptoms such as The operation starts in the abdomen for Ivor Lewis or
dysphagia is a very specific finding for the presence of a transhiatal esophagectomies. Port placement is shown
T3 or greater lesion, the absence of symptoms does not in Fig. 39D.1. The camera port is located 18 cm inferior
necessarily indicate that the patient does not have a T3 or from the xiphoid process and is generally placed first.
greater lesion. Given that being T3 or deeper and/or the A 30-degree down camera or a 0-degree lens is used.
presence of N1 or greater disease dictates the performance Inspection of the abdomen is performed for liver and
of induction chemoradiation at our institution, we also peritoneal metastases prior to placing the other ports.
consider EUS a critical part of the preoperative evaluation. A single left robotic arm and two right robotic arms are
Performance of induction chemoradiation for T2N0 used. These ports should be placed no more than 2 to
lesions is variable. We prefer preoperative therapy since 3 cm superior to the camera port to avoid problems with
a significant percentage will have nodal disease, although the angle of the instruments when dividing the greater
there is a controversy regarding patients who are older omentum off the greater curvature of the stomach toward
than 75 years. Brain imaging is performed if the patient the pylorus. The robotic arms should be around 9 cm
has neurologic symptoms or headaches that are concerning apart from each other if an Si system is used (8 cm if an
for intracranial metastases. Bronchoscopy is done if the Xi system is used). If there is not enough room on the
patient has an esophageal cancer of the proximal or middle left side of the abdomen to place the ports straight across,
esophagus to rule out airway invasion. Patients who remain the robotic arm closer to the camera can be staggered
candidates for esophagectomy after the aforementioned slightly in front of the other one. If using the Si system,
testing generally also receive pulmonary function testing the second right robotic arm can be a 5 mm and the other
and stress testing. No specific diagnostic procedures are robotic arms 8 mm. Stapling of the conduit is performed
performed for robotic esophagectomy per se. via the assistant port. If using the Xi system and robotic
After the completion of induction chemoradiation, stapling is desired, the left robotic arm should be a 12-mm
restaging PET-CT should be performed. Patients who port; the rest of the robotic ports are 8-mm ports. A
develop progression of disease or metastases are offered 5-mm port for the liver retractor is placed as close to the
palliative management strategies. Patients who have per- costal margin and laterally as possible (just over the right
sistent disease or show a response (complete or partial colon). A 12-mm assistant port is placed in the patient’s
resolution of fluorodeoxyglucose (FDG) avidity of the right lower quadrant and triangulated behind the left
lesion on PET-CT scan) are scheduled for esophagectomy robotic arm port and camera port. Insufflation should
from 8 to 12 weeks after the conclusion of chemoradiation, be delivered via this port during the case.
once they have recovered reasonably well from the side The patient is placed in steep reverse Trendelenburg
effects of induction therapy. Data on the optimal interval position and the liver retractor is positioned under the left
between completion of chemoradiation and surgery lateral lobe of the liver to expose the esophageal hiatus.
are mixed. Kim et al. showed no difference in terms of We use a Snowden Pencer articulating pretzel retractor
perioperative risk, pathologic response, or overall survival (Becton Dickinson; Franklin Lakes, New Jersey) for this
between patients who were resected more than 8 weeks purpose. If using an Si system, the head of the bed is
after chemoradiation versus those resected less than 8 turned so that the robot can approach it from over the
weeks after.12 Lee et al. demonstrated that prolonging the head. If using an Xi system, the bed does not have to be
interval after chemoradiation for esophageal adenocarci- turned. The robot is carefully driven in, making sure that
noma increased the pathologic complete response rate its arms do not collide with the patient’s head and upper
426 SECTION I Esophagus and Hernia
Right robotic
Liver
arm #1
retractor
18 cm Right robotic
Assistant arm #2
port 9c m 9c
m 9c m >3 cm
Camera port
body. The robotic arms are docked to the ports, and the left gastric artery above the replaced left hepatic artery.
robotic phase begins. We then perform a circumferential dissection around the
We generally use the following instruments during the esophagus at the hiatus, carrying it into the mediastinum
dissection: left robotic arm—Cadiere forceps; right robotic for a few centimeters but trying to avoid excessive trauma
arm—Vessel Sealer; second right robotic arm—thoracic or widening of the hiatus or entering the left pleura, to
grasper (Si system), tip-up fenestrated grasper (Xi system). avoid increasing the risk of paraconduit herniation. If
Division of the greater omentum from the greater curvature a transhiatal esophagectomy is planned, however, this
of the stomach starts by entering the lesser sac between mediastinal dissection should be performed as high as
the stomach and the left side of the transverse colon. possible. The use of the fenestrated bipolar forceps in the
Care is taken to identify the gastroepiploic vessels and right hand can be helpful for atraumatically dissecting the
assiduously avoid them. The greater omentum is divided underside of the esophagus. During this, the second right
from the patient’s left toward the right until the pylorus robotic arm is used to retract the esophagus up and to
is reached. The surgeon then switches direction and goes the patient’s left (screen right). A 1-inch-thick or greater
from the entry point into the lesser sac toward the spleen Penrose drain is placed around the esophagus and the
and fundus while paying extra attention to the short gastric ends are either tied or stapled together. The left gastric
vessels. During this process, the second robotic arm is used pedicle is identified and the surrounding fat is dissected
to hold the greater omentum/colon in one direction; the off of the vein and artery. Depending on the patient, this
assistant can grasp the stomach and retract in the other vessel can be approached either from the lesser curvature
direction. An omental flap should be preserved during side or from underneath the stomach as it is lifted (greater
this dissection to be wrapped around the anastomosis curvature side). Test clamp of the pedicle is then performed
and protect the airway. Once the short gastric vessels are prior to division of the left gastric artery and vein with the
divided, the surgeon then works on the left side of the stapler. The vessels may be stapled individually or together.
esophageal hiatus, working from the top of the hiatus down Next, botulinum toxin injection with 100 units in 4 mL
underneath the esophagus so that the area beneath the of saline at the pylorus is performed. Alternatively, a gastric
esophagus is as clear as possible to facilitate encircling the emptying procedure such as pyloromyotomy or pyloroplasty
esophagogastric junction later. Attachments of the stomach may be done at the discretion of the surgeon, although
to the retroperitoneum should also be divided at this point. this increases the operative time without a significant
Next, the lesser sac is entered through the lesser improvement in results.16 The pylorus should be able to
omentum. An accessory or replaced left hepatic artery reach the esophageal hiatus with little tension; if tension
originating from the left gastric artery can be located in exists or it cannot reach, further mobilization of the pylorus
this area, as up to 12% of patients may have this variation.15 from the greater omentum and a Kocher maneuver should
Our practice is to test clamp any significant vessel in this be performed. Care should be taken to avoid injuring
location and perform a visual check for liver perfusion structures in the portal triad during mobilization. At this
prior to dividing it. We have not had this occur, but if liver point, the surgeon needs to confirm that the nasogastric
appears ischemic with the test clamp, the operation could tube has been withdrawn to 20 cm or so. A starting point
be aborted, or the surgeon could attempt to divide the on the lesser curvature of the stomach is selected, and
Surgical Approaches to Remove the Esophagus: Robotic CHAPTER 39D 427
the perigastric fat going from the edge of the stomach the right robotic arm port (8 mm in Xi system, 12 mm in
to the opening in the lesser omentum is divided with the Si system), which is in line with the anterior superior iliac
Vessel Sealer. The gastric conduit is created with a stapler spine. The left robotic arm port (12 mm in Xi system if
(4 mm staple height, 45 to 60 mm length) using right completely robotic linear stapling of anastomosis is desired,
robotic arm #2 (placed on the fundus) and the assistant 8 mm if not desired or in Si system) is placed 9 cm (10 cm
(grasping or retracting the antrum) to stretch the stomach in Si system) away from the camera port in line with the
out. The specimen is not completely divided from the anterior iliac spine. The second left robotic arm port
conduit so that the conduit may be pulled up into the (8 mm in Xi system, 5 mm in Si system) is placed posterior
chest or neck with the abdomen. A suture is placed at to the midaxillary line just above the diaphragm. The
the distal part of the staple line near the pylorus so that assistant port (12 mm) is placed in a position triangulated
the end of the staple line can be easily seen from the behind the left robotic arm and camera port just above
chest. If a neck anastomosis is being performed, part of the diaphragm. We generally use the following instruments
the lesser curve of the specimen may be resected at this during the dissection: left robotic arm—Cadiere forceps,
point to help facilitate its passage through the thoracic right robotic arm—thoracic dissector, second right robotic
inlet. The Penrose drain and specimen are pushed up arm—thoracic grasper (Si system), and tip-up fenestrated
into the mediastinum. grasper (Xi system). The lung is retracted anteriorly with
Finally, a jejunostomy tube is placed. If the anatomy the second left robotic arm, and the inferior pulmonary
is suitable (e.g., exposure to the proximal jejunum is in ligament is divided. Lymph nodes from stations 8 and
front of the camera rather than directly below or behind 9 are resected. The esophagus is dissected off of the
it), this can be done robotically. In general, however, pericardium until the carina is exposed. Lymph nodes
we have found it simpler and quicker to perform this from station 7 are resected so that the right mainstem
laparoscopically after undocking the robot. Three 2-0 bronchus, carina, and left mainstem are clearly visible.
absorbable sutures are placed in the proximal jejunum Thermal injury to the airway is carefully avoided. The
in a triangulated fashion and the ends are exteriorized. dissection is carried toward the azygos vein, which is then
The jejunostomy tube is placed with a Seldinger technique isolated and divided with a vascular load staple fired as
over a wire after dilating the tract and tying the sutures. posteriorly as possible. The dissection of the esophagus is
then carried proximally and into the thoracic inlet, now
THORACIC PHASE staying closer to the esophagus to avoid a tracheal injury. If
The single-lumen endotracheal tube may be exchanged performing a cervical anastomosis, this dissection is carried
for a double-lumen endotracheal tube during closure of as high as possible. Next, the esophagus is dissected off
the skin incisions to expedite transition to the thoracic of the aorta posteriorly, taking care to avoid the thoracic
phase. The patient is positioned in a nearly prone posi- duct, which runs especially close to the esophagus near
tion. It is important to keep the right arm/shoulder low the azygos vein. This is carried from the thoracic inlet
during positioning so it will not get in the way of the toward the diaphragm until the dissection plane from
right robotic arm. Patient positioning and port placement the abdominal phase is reached and the Penrose drain
are demonstrated in Fig. 39D.2. The right robotic arm is grasped. The Penrose drain can then be retracted with
port (8 mm; 12 mm if completely robotic linear stapling the second left robotic arm both anteriorly and laterally
technique is desired) is placed in the axilla. A long trocar to help facilitate the remainder of the dissection. We try
can be used to help avoid collisions with the patient. The not to enter the left pleural space unless this needs to be
camera port is placed about 9 cm (10 cm in Si system) from resected for gross disease.
If a McKeown esophagectomy is to be performed, the
Penrose is pushed up into the thoracic inlet to be retrieved
through the neck. Much of the esophagus in the distal
3 neck can be dissected via the chest with a right robotic
approach; we believe that this is an added advantage of
a robotic approach and may significantly decrease the
incidence of recurrent laryngeal nerve injury. The chest
2 C 1 tube is placed (additional tube placed in left pleural space
if entered), insufflation gas turned off, ports removed,
port sites are checked for bleeding, the lung reinflated,
A and incisions closed. Then the abdominal and cervical
phases of the operation are begun.
If an Ivor Lewis esophagectomy is to be performed, the
specimen and conduit are brought up into the thorax,
mostly with the assistant using an atraumatic grasper such
as empty ring forceps, making sure that the conduit is not
twisted (staple line directed laterally). The previously placed
suture at the distal staple line in the lesser curve should be
FIGURE 39D.2 Patient positioning and port placement for thoracic visible and brought just into the chest above the hiatus. The
phase of robotic esophagectomy. 1, Right robotic arm; 2, left esophagus is transected at the desired length with robotic
robotic arm; 3, second left robotic arm; A, assistant port; shears and the proximal and distal margins are sent for
C, camera port. pathologic examination. The specimen is divided from the
428 SECTION I Esophagus and Hernia
gastric conduit with the stapler (4-mm staple height) and deployment of the tip of the stapler into the aorta. Mul-
together with the Penrose removed. The gastric conduit is tiple attempts at extending the tip through the conduit
oriented so that the anastomosis will be located posteriorly. wall should also be avoided. The tip of the stapler is
The conduit is tacked to pleura and/or transected vagus nerve linked to the anvil, which can be facilitated with the use
to keep it in place during the anastomosis, which orients it of the laparoscopic anvil grasper.
and prevents tension. The anastomosis can be completely • The stapler is fired. The rims of tissue excised by the
hand-sewn, completely stapled (linear or circular stapler), stapler should be examined; if they are not complete,
or a combination of the two (linear stapler “posterior” wall the corresponding area of the anastomosis should be
and hand-sewn “anterior” wall). The optimal approach to checked and closed with sutures.
performing the anastomosis has not been described. We have The anastomosis should be inspected and any question-
performed all of these and will describe each technique: able areas should have sutures placed to close them.
Endoscopy can be performed routinely if so desired and
Hand-Sewn Anastomosis the integrity of the anastomosis checked by insufflating it
• The gastrotomy is made on the posterior wall of the while it is submerged under saline or water. The omental
conduit for an “end-to-side” anastomosis, at least 2 cm flap is wrapped around the anastomosis, protecting the
proximal to the tip of the conduit and away from the airway from it, and sutured in place. The gastric conduit
staple line. is secured to the diaphragm at the hiatus. A chest tube
• A row of 3-0 silk sutures is placed for the outer layer is placed and this phase of the operation concluded as
(“posterior”). described before.
• An inner layer of 3-0 absorbable sutures is placed for
the inner layer (“posterior”). CERVICAL PHASE
• The “anterior” wall for the anastomosis is closed with The cervical phase of the operation is similar to that
interrupted 3-0 absorbable sutures. performed during open operations. During the thoracic
• An outer “anterior” layer of 3-0 silk sutures is placed. phase of a McKeown esophagectomy, we perform extensive
periesophageal dissection into the thoracic inlet. We find it
Combination Stapled and Hand-Sewn Anastomosis helpful to place a Penrose drain around the esophagus and
• After the gastrotomy is performed, the esophagus and either staple or tie the ends together and push the drain
conduit are lined up and a stapler is fired to create a into the superior mediastinum so that the esophagus can
20 to 30 mm common wall for the “posterior” wall for be more easily encircled during the neck dissection. An
the anastomosis. The stapler can be deployed either incision anterior and parallel to the left sternocleidomastoid
by the assistant or in the left robotic arm (Xi system). is made. The platysma is divided. The omohyoid muscle is
• The “anterior” wall is closed in two layers as described encountered and divided. The carotid sheath containing
earlier. the common carotid artery and the internal jugular vein
is gently retracted laterally. The trachea is gently retracted
Completely Stapled Anastomosis (Linear Stapler) medially, taking care to avoid the use of metal retractors
• The “posterior” wall of the anastomosis is created as that could injure the recurrent laryngeal nerve in the
described previously. tracheoesophageal groove. The esophagus is followed
• Five 3-0 silk interrupted sutures are placed and tied to inferiorly until the intrathoracic dissection plane is reached
approximate the muscle and mucosa of the “anterior” and the Penrose drain is identified.The Penrose is used to
wall. pull the esophagus into the incision. The specimen and
• These sutures are held up and the “anterior” wall of conduit are pulled gently up through the chest. This can be
the anastomosis is created with the linear stapler. The done with the camera in the abdomen to make sure that the
stapler generally needs to be deployed in the location of conduit does not twist during the process and to facilitate
the right robotic arm port (either robotically, in which passage of the lesser curvature of the specimen or omental
case the surgeon needs to upsize to a 12-mm port, or fat attached to conduit through the hiatus. The esophagus
after undocking that port and having the assistant place is then divided sharply, and the specimen is stapled off of
a stapler through it). the gastric conduit, taking care to prevent the conduit from
retracting back into the mediastinum with a nontraumatic
Completely Stapled Anastomosis (Circular Stapler) clamp. A gastrotomy is made in the conduit on the posterior
• A purse string with 3-0 nonabsorbable monofilament wall, and a 3.5 to 4.5 mm tall stapler is fired to create the
suture is placed in the esophagus, making sure to posterior wall of the anastomosis. Interrupted silk sutures
incorporate the mucosal layer. are placed and tied anteriorly to approximate the mucosa
• The anvil is placed in the esophagus and the purse and muscle, and a stapler is used to create the anterior wall
string is tied. of the anastomosis, making sure to incorporate mucosa of
• An additional purse-string suture is placed if there is both the esophagus and stomach along the entire edge.
any gap around the anvil. If there is concern about the integrity of the anastomosis
• A gastrotomy is created at the tip of the conduit. Retrac- and enough redundancy exists, a buttressing layer over the
tion sutures can be placed to help. staple line can be created by Lembert-type interrupted 3-0
• The stapler is positioned through the gastrotomy with the silk sutures. We do not routinely test the anastomosis for
end directed toward the posterior aspect of the conduit. leaks intraoperatively or leave a nasogastric tube. We do
The tip of the stapler is extended, going through the not typically leave a drain in place. The incision is then
wall of the conduit. Caution should be exercised to avoid closed in layers with absorbable suture.
Surgical Approaches to Remove the Esophagus: Robotic CHAPTER 39D 429
Lymph Estimated
Nodes Blood Loss Operative Time Overall Major
Name, Year # Pts Dissected Operative Approach (mL) (min) Leak Rate Morbidity Mortality
Cerfolio, 85 22 Ivor Lewis (lap/robot 35 361 4.3% 36.4% 3.5% 30-day
201617 abd, robot chest) 11% 90-day
Hernandez, 52 20 Ivor Lewis (robot abd/ NR 442 3.8% 26.9% 0% (“hospital”)
201318 chest)
De la 50 18.5 Ivor Lewis (robot abd/ NR 445 4% 28% 0% (“hospital”)
Fuente, chest)
201319
Sarkaria, 21 20 Ivor Lewis (n = 17) 300 556 14% (grade 24% (grade III 4.8%
201320 and McKeown (n = II or or greater) (“postoperative”)
4), robot abd/chest greater)
Dunn, 40 20 Transhiatal (robot 100 311 25% NR 2.5% 30-day
201321 mediastinal
dissection)
Weksler, 11 19 McKeown (robot abd/ 200 445 9.1% 36.4% 0% (“hospital”)
201222 chest)
Park, 114 44 McKeown (lap/robot 209 420 12.3% NR 2.5% 90-day
201623 abd, robot chest) (grade II
or greater)
Boone, 47 29 Ivor Lewis (lap abd, 625 450 21% NR 6.4%
200924 robot chest) (“postoperative”)
Kernstine, 14 18 Ivor Lewis (lap/robot 275 11.2 hours 14% 29% 0% 30-day
200725 abd, robot chest) (total room 1 patient (7.1%)
time) died at 72 days
Galvani, 18 14 Transhiatal (robot 54 267 33% NR 0% 30-day
200826 abd)
Hodari, 54 16 Ivor Lewis (lap abd, 74.4 362 6.8% NR 2% 30-day
201527 robot chest)
Coker, 23 15 Transhiatal (robot 100 231 9% NR 4% 30-day
201428 abd)
Harrison, 43 12 Transhiatal (28), NR 309 23% 41.8% 4.7%
201529 McKeown (7), Ivor (“postoperative”)
Lewis (5); robot abd
Luketich, 1033 21 McKeown (n = 481), NR NR 5% NR 1.68% 30-day
201230 Ivor Lewis (n = 530) (requiring 2.8% 30-day or
Lap abd/VATS chest surgery) hospital
abd, Abdominal; lap, laparoscopic; NR, not recorded; VATS, video-assisted thoracoscopic surgery.
430 SECTION I Esophagus and Hernia
optics, improved dexterity, favorable ergonomics, and carcinoma: does delayed surgery impact outcome? Ann Surg Oncol.
the ability to control the retraction and camera without 2013;20:4245-4251.
15. Hiatt JR, Gabbay J, Busuttil RW. Surgical anatomy of the hepatic
an assistant.33,34 Disadvantages of the robotic platform arteries in 1000 cases. Ann Surg. 1994;220:50-52.
include cost, and complexity in terms of developing robotic 16. Cerfolio RJ, Bryant AS, Canon CL, Dhawan R, Eloubeidi MA. Is
skills, personnel issues, room layout, and robot docking, botulinum toxin injection of the pylorus during Ivor Lewis esopha-
although these may be surmounted with a formal training gogastrectomy the optimal drainage strategy? J Thorac Cardiovasc
Surg. 2009;137:565-572.
paradigm.35 Long-term oncologic outcomes specific to 17. Cerfolio RJ, Wei B, Hawn MT, Minnich DJ. Robotic esophagectomy
robotic esophagectomy are not yet well described, although for cancer: early results and lessons learned. Semin Thorac Cardiovasc
it would be expected that they would be comparable to Surg. 2016;28(1):160-169.
those for nonrobotic MIE. 18. Hernandez JM, Dimou F, Weber J, et al. Defining the learning
curve for robotic-assisted esophagogastrectomy. J Gastrointest Surg.
2013;17:1346-1351.
CONCLUSION 19. De la Fuente SG, Weber J, Hoffe SE, Shridhar R, Karl R, Meredith KL.
Initial experience from a large referral center with robotic-assisted
Robotic esophagectomy can be done safely with comparable Ivor Lewis esophagogasrectomy for oncologic purposes. Surg Enodsc.
intraoperative parameters, morbidity, and outcomes to 2013;27:3339-3347.
20. Sarkaria IS, Rizk NP, Finley DJ, et al. Combined thoracoscopic and
nonrobotic MIE, while offering certain more subjective laparoscopic robotic-assisted minimally invasive esophagectomy using
advantages to the surgeon. a four-arm platform: experience, technique and cautions during early
procedure development. Eur J Cardiothorac Surg. 2013;43:e107-e115.
21. Dunn DH, Johnson EM, Morphew JA, Dilworth HP, Krueger JL,
REFERENCES Banerji N. Robot-assisted transhiatal esophagectomy: a 3-year single-
center experience. Dis Esophagus. 2013;26:159-166.
1. Xie MR, Liu CQ, Guo MF, Mei XY, Sun XH, Xu MQ. Short-term 22. Weksler B, Sharm P, Moudgill N, Chojnacki KA, Rosato EL.
outcomes of minimally invasive Ivor-Lewis esophagectomy for Robot-assisted minimally invasive esopahgectomy is equivalent to
esophageal cancer. Ann Thorac Surg. 2014;97:1721-1727. thoracoscopic minimally invasive esophagectomy. Dis Esophagus.
2. Biere SS, van Berge Henegouwen MI, Maas KW, et al. Minimally 2012;25:403-409.
invasive versus open oesophagectomy for patients with oesophageal 23. Park SY, Kim DJ, Yu WS, Jung HS. Robot-assisted thoracoscopic
cancer: a multicenter, open-label, randomized controlled trial. Lancet. esophagectomy with extensive mediastinal lymphadenectomy: experi-
2012;379:1887-1892. ence with 114 consecutive patients with intrathoracic esophageal
3. Weksler B, Sharma P, Moudgill N, Chojnacki KA, Rosato EL. cancer. Dis Esophagus. 2016;29(4):326-332.
Robot-assisted minimally invasive esophagectomy is equivalent to 24. Boone J, Schipper ME, Moojen WA, Borel Rinkes IH, Cromheecke
thoracoscopic minimally invasive esophagectomy. Dis Esophagus. GJ, van Hillegersberg R. Robot-assisted thoracoscopic oesophagectomy
2012;25:403-409. for cancer. Br J Surg. 2009;96:878-886.
4. Clark J, Sodergren MH, Purkayastha S, et al. The role of robotic 25. Kernstine KH, DeArmond DT, Shamoun DM, Campos JH. The
assisted laparoscopy for oesophagogastric oncological resection; an first series of completely robotic esophagectomies with three-field
appraisal of the literature. Dis Esophagus. 2011;24:240-250. lymphadenectomy: initial experience. Surg Endosc. 2007;21:2285-2292.
5. Lee S, Sudarshan M, Li C, et al. Cost-effectiveness of minimally 26. Galvani CA, Gorodner MV, Moser F, et al. Robotically assisted lapa-
invasive versus open esophagectomy for esophageal cancer. Ann roscopic transhiatal esophagectomy. Surg Endosc. 2008;22(1):188-195.
Surg Oncol. 2013;20:3732-3739. 27. Hodari A, Park KU, Lace B, Tsiouris A, Hammoud Z. Robot-assisted
6. Melvin WS, Needleman BJ, Krause KR, et al. Computer-enhanced minimally invasive Ivor Lewis esophagectomy with real-time perfusion
robotic telesurgery: initial experience in foregut surgery. Surg Endosc. assessment. Ann Thorac Surg. 2015;100(3):947-952.
2002;16:1790-1792. 28. Coker AM, Barajas-Gamboa JS, Cheverie J, et al. Outcomes of
7. Cong Z, Diao Q, Yi J, et al. Esophagectomy combined with aortic robotic-assisted transhiatal esophagectomy for esophageal cancer
segment replacement for esophageal cancer invading the aorta. after neoadjuvant chemoradiation. J Laparoendoscopic Adv Surg Tech
Ann Thorac Surg. 2014;97:460-466. A. 2014;24:89-94.
8. Van Raemdonck D, Van Cutsem E, Menten J, et al. Induction therapy 29. Harrison LE, Yiengpruksawan A, Patel J, Itskovich A, Lee B, Korst
for clinical T4 oesophageal carcinoma; a plea for continued surgical R. Robotic gastrectomy and esophagogastrectomy: a single center
exploration. Eur J Cardiothorac Surg. 1997;11:828-837. experience of 105 cases. J Surg Oncol. 2015;112:888-893.
9. Li J, Shen Y, Tan L, et al. Is minimally invasive esophagectomy 30. Luketich JD, Pennathur A, Awais O, et al. Outcomes after minimally
beneficial to elderly patients with esophageal cancer? Surg Endosc. invasive esophagectomy: review of over 1000 patients. Ann Surg.
2015;29:925-930. 2012;256:95-103.
10. Ripley RT, Sarkaria IS, Grosser R, et al. Pretreatment dysphagia in 31. Van der Sluis PC, Ruurda JP, van der Horst S, et al. Robot-assisted
esophageal cancer patients may eliminate the need for staging by minimally invasive thoraco-laparoscopic esophagectomy versus open
endoscopic ultrasonography. Ann Thorac Surg. 2016;101:226-230. transthoracic esophagectomy for resectable esophageal cancer, a
11. Fang TC, Oh YS, Szabo A, Khan A, Dua KS. Utility of dysphagia randomized controlled trial (ROBOT trial). Trials. 2012;13:230.
grade in predicting endoscopic ultrasound T-stage of non-metastatic 32. Sarkaria IS, Bains MS, Finley DJ, et al. Intraoperative near-infrared
esophageal cancer. Dis Esophagus. 2016;29(6):642-648. fluorescence imaging as an adjunct to robotic-assisted minimally
12. Kim JY, Correa AM, Vaporciyan AA, et al. Does the timing of invasive esophagectomy. Innovations (Phila). 2014;9:391-393.
esophagectomy after chemoradiation affect outcome? Ann Thorac 33. Wei B, D’Amico TA. Thoracoscopic versus robotic approaches:
Surg. 2012;93:207-212. advantages and disadvantages. Thorac Surg Clin. 2014;24:177-188.
13. Lee A, Wong AT, Schwartz D, Weiner JP, Osborn VW, Schreiber D. 34. Ruurda JP, van der Sluis PC, van der Horst S, van Hillegersberg R.
Is there a benefit to prolonging the interval between neoadjuvant Robot-assisted minimally invasive esophagectomy for esophageal
chemoradiation and esophagectomy in esophageal cancer? Ann cancer: a systematic review. J Surg Oncol. 2015;112:257-265.
Thorac Surg. 2016;102(2):433-438. 35. Cerfolio RJ, Bryant AS, Minnich DJ. Starting a robotic program in
14. Chiu C, Chao Y, Chang H, et al. Interval between neoadjuvant general thoracic surgery: why, how, and lessons learned. Ann Thorac
chemoradiotherapy and surgery for esophageal squamous cell Surg. 2011;91:1729-1737.
CHAPTER
Extent of Lymphadenectomy for Esophageal Cancer
Alexander W. Phillips
| S. Michael Griffin
40
T
he extent of lymphadenectomy as part of an esopha-
gectomy for cancer remains a controversial issue. LYMPH NODE TIERS
The aggressive nature of the disease often means Extent of lymph node dissection is commonly divided
that both local nodal and distant metastases exist at the into three fields: the upper abdomen, the mediastinum,
time of presentation. As such, locally advanced disease and the neck (Fig. 40.2). Three-field dissection therefore
in which potential cure is intended is frequently treated relates to removal of nodal tissue from each of these
with neoadjuvant modalities. The debate on degree of areas. A lack of clarity exists on the exact definition of
lymphadenectomy hinges largely on the belief that a radical a two-field lymph node resection. The lack of clarity is
dissection provides improved locoregional control and because of differences in prevalence of squamous cell
thus improved survival. However, it is also worth noting cancer in Japan and the East and in those patients from
that extended lymphadenectomy also provides improved Western countries where adenocarcinoma has become
staging, which can allow better patient counseling and the prevalent cause.
may influence the use of adjuvant treatment as further Where SCC of the esophagus is the common pathology,
studies are performed into its role. a two-field dissection is usually described as removal of
nodal tissue from the upper abdomen (around the celiac
artery) and in the inferior and superior mediastinum
LYMPHATIC DRAINAGE OF THE ESOPHAGUS and along both recurrent laryngeal nerves. To contrast
AND PATTERNS OF SPREAD this, in countries where adenocarcinoma has become the
common variant, a two-field dissection is usually regarded
Knowledge of the lymphatic drainage of the esophagus is as removal of tissue from the upper abdomen and inferior
a key component for having a rationale for lymphadenec- mediastinum. This usually extends only to the level of the
tomy. The esophagus traverses three body compartments, carina and reflects the usual anatomic location of these
and lymph flow can occur in a wide pattern of spread. tumors in the lower esophagus or at the esophagogastric
The embryologic origin of the esophagus is from the junction.
branchial arches and pharyngeal pouches from above, Although some surgeons do not perform a formal
and the splanchnic mesoderm below. These join during lymphadenectomy irrespective of the surgical approach,
early embryologic development but remain demarcated the following describes what is commonly accepted to be
at the level of the tracheal bifurcation leading to bilateral the groups of nodes resected with each “field” of dissection.
lymphatic drainage (Fig. 40.1).
In early esophageal cancer the suggestion is that lymph
node spread, when it occurs, follows these anatomic
ABDOMINAL LYMPH NODE DISSECTION
pathways, implying that the tumor location is key to This is commonly regarded as the first “field” of lymph-
determining which nodes are likely to be involved. Thus adenectomy and includes the following abdominal lymph
nodal involvement for tumors above the tracheal bifurca- node stations: the superior gastric group, celiac trunk
tion is preferentially to those in the upper mediastinum nodes, and common hepatic nodes (see Fig. 40.1 and
and neck, whereas those below this point will metastasize Table 40.1).
toward the celiac axis. Tumors located at the bifurcation The superior gastric nodes include those that are
may metastasize in either direction. Skipping of lymph paracardial, which are frequently involved and are related
node stations in these early tumors is rare.1 to the most superior branch of the left gastric artery, and
Lymph node involvement appears to be more common so both left and right paracardial should be considered
with squamous cell carcinoma (SCC) compared with a single group. The lesser curve nodes are involved with
adenocarcinoma when the tumor has invaded into the tumor spread to the celiac trunk and are also considered
muscularis mucosae (T1a-M3), approaching up to 12%,2,3 part of the superior gastric group, as are the nodes found
compared with only 1.3% in adenocarcinoma.4–6 along the length of the left gastric artery.
There is also an extensive submucosal lymphatic network The celiac trunk nodes include those around the celiac
that allows longitudinal communication between the axis at the root of the left gastric artery, common hepatic
proximal and distal drainage systems. In tumors that are artery, and splenic artery, and they should be removed
more advanced and have potentially led to a blockage of en bloc with the primary lesion. Disease beyond this
one of the primary pathways, this submucosal system allows drainage point represents metastatic disease. Similarly,
eccentric lymph node involvement. This is an important disease involvement beyond the hepatic nodes equates
consideration because most patients are diagnosed at a to metastatic disease.
more advanced state and T3 disease is associated with an The involvement of nodes in both paracardial regions
up to 85% chance of lymph node involvement.7,8 and lesser curve are the most commonly involved with
431
Extent of Lymphadenectomy for Esophageal Cancer CHAPTER 40 431.e1
ABSTRACT
The extent of lymphadenectomy as part of an esophagec-
tomy for cancer remains a controversial issue. The aggres-
sive nature of the disease often means that both local nodal
and distant metastases exist at the time of presentation. As
such, locally advanced disease in which potential cure is
intended is frequently treated with neoadjuvant modalities.
The debate on degree of lymphadenectomy hinges largely
on the belief that a radical dissection provides improved
locoregional control and thus improved survival. However,
it is also worth noting that extended lymphadenectomy
also provides improved staging, which can allow better
patient counseling and may influence the use of adjuvant
treatment as further studies are performed into its role.
KEYWORDS
Esophagectomy
esophageal cancer
lymphadenectomy
staging esophageal cancer
adenocarcinoma
squamous cell carcinoma
432 SECTION I Esophagus and Hernia
Cervical field
(three field)
Thoracic field
(two field)
Abdominal field
(one field)
these suggestions, the incidence of local recurrence for belief that it confers improved locoregional control, and
either subtype is between 25% and 50%, indicating that hence cure rates, as well as improving the overall staging
a meticulous lymphadenectomy may have a profound of the disease.17
impact on providing locoregional control. It has been established that complete resection (R0)
is an important indicator of prognosis and that those
patients with microscopic (R1) or macroscopic (R2)
IMPLICATIONS OF LYMPHADENECTOMY residual disease have a significantly bleaker prognosis.
There is still a great deal of debate regarding the extent of Approximately 25% to 50% of patients may develop
lymphadenectomy that should be performed and whether locoregional recurrence after a two-field lymphadenec-
a more radical resection confers any survival advantage. tomy. Anderegg et al. indicated that the location of nodal
The argument for an extensive lymphadenectomy lies in a involvement is an independent predictor of survival in
434 SECTION I Esophagus and Hernia
11.2%
TABLE 40.1 Nodes According to Anatomic Region
t
Diaphragmatic nodes
ne
Superior gastric lymph nodes 12.3%
Paracardial nodes
Lesser curve nodes
8.2%
Left gastric artery nodes
k.
Celiac trunk nodes
Common hepatic nodes
oo
From Akiyama H. Surgery for Cancer of the Esophagus. Baltimore: Lip-
pincott Williams & Wilkins; 1990.
yb
patients with adenocarcinoma of the distal esophagus and
esophagogastric junction. In their series of 479 patients
who underwent transthoracic esophagectomy and two-field
er
involvement was 35 months compared with 16 months cell carcinoma from 100 patients. (From Stein HJ, Theisen J,
for those with nodes at the celiac trunk and 15 months Siewert J-R. Surgical resection for esophageal cancer: role of
for those with involvement in the proximal field.18 In an extended lymphadenectomy. In: Fielding JWL, Hallissey MT, eds.
su
earlier study, Hulscher demonstrated a trend toward greater Upper Gastrointestinal Surgery. London: Springer; 2005:322.)
survival at 5 years of patients undergoing transthoracic
esophagectomy with extended en bloc lymphadenectomy not suggest that a higher yield of lymph nodes equated
://
when compared with a transhiatal approach in patients to better overall survival.22 It is worth noting that in both
with lower esophageal and junctional adenocarcinomas.19 these studies the lymph node yields were generally low
tp
The risk of involvement of lymph nodes in early cancers with the upper border of the third quartile only 15 nodes
has been previously established. Griffin et al. demonstrated in van der Schaaf’s study, and 20 nodes in Lagergren’s
that 12% of patients with adenocarcinoma invading the study. Both papers demonstrated that, as expected, a
ht
submucosa may have lymph node involvement, although greater number of metastatic nodes is associated with
none of the patients with tumor confined to the mucosa decreased survival but also that a higher ratio of positive
were lymph node positive.20 Although this highlights that to negative nodes was associated with increased mortality.
local endoscopic resection is not appropriate in patients Indeed the aspect of lymph node ratio has been evalu-
with submucosal invasion, it also reiterates the importance ated a number of times. Several studies have demonstrated
of carrying out a meticulous lymphadenectomy even when that lymph node ratio is a prognostic factor; however,
the cancer is deemed as at an early stage. this is often coupled with the caveat that if insufficient
A number of studies have attempted to determine the nodes are obtained, this ratio ceases to be useful as a
impact of the extent of lymphadenectomy by using the prognostic tool due to the fact that there has been, in
number of nodes retrieved as a surrogate for the extent essence, incomplete sampling.
of dissection. Van der Schaaf et al. reviewed the outcomes Chen et al. evaluated the impact of lymph node ratio
from a Swedish nationwide study of 1044 resections over for esophageal squamous cell cancers from more than
a 23-year period, by comparing outcomes according to 2000 patient in China and concluded that this was a better
nodal yield quartiles.21 They concluded that more extensive predictor of prognosis than using the “N” category. Their
clearance did not equate to improved survival. Similarly suggestion was to use an “Nr” ratios category of 0, 0%
Lagergren et al. looked at patients from a single center to 10%, 10% to 20%, and greater than 20% rather than
again using lymph node yield as a surrogate for extent of N0-N3 with predicted survivals of 61%, 41%, 33%, and 23%,
lymphadenectomy. Again, the results from this study did respectively.23 Similar findings were made by Bhamidipati
Extent of Lymphadenectomy for Esophageal Cancer CHAPTER 40 435
t
12.8% or worse.31 Patients who are considered for neoadjuvant
ne
treatment are likely to have more locally advanced disease,
10.2%
and this would suggest they will benefit from a more
extensive lymphadenectomy.
k.
Ideally it would be possible to tailor the lymphadenec-
tomy according to disease stage. However, this is reliant
oo
on accurate assessment of which nodes are involved, and
this currently remains difficult. Sentinel node techniques
have been attempted, but the occurrence of skip metastases
yb
and unpredictability of nodal involvement makes this
currently unreliable.21
Peyre et al. looked at whether systemic disease could
be predicted by the number of lymph nodes involved.32
er
patients with adenocarcinoma. (Modified from Stein HJ, Theisen the probability of systemic disease exceeds 50% when
J, Siewert J-R. Surgical resection for esophageal cancer: role of more than three nodes are involved and approaches 100%
extended lymphadenectomy. In: Fielding JWL, Hallissey MT, eds. when greater than eight nodes are involved.
su
Upper Gastrointestinal Surgery. London: Springer; 2005:323.) A further study by the same group sought to evaluate
the impact on the number of lymph nodes removed
et al. in a cohort of US patients who predominantly had on outcomes in esophageal cancer. This study looked
://
adenocarcinoma. They also concluded that lymph node at 2303 esophageal cancer patients, again who did not
ratio was an independent predictor of prognosis in their receive neoadjuvant or adjuvant treatment. Their findings
tp
cohort of 347 patients undergoing surgery and was more demonstrated that the number of nodes recovered was an
useful than “N” category.24 independent predictor of survival after esophagectomy and
The idea that an extensive lymphadenectomy leads suggested an optimum threshold of 23 nodes. Furthermore,
ht
to better locoregional control is founded in a belief that they suggested that this optimum finding of 23 lymph
leaving positive lymph nodes behind equates to leaving nodes indicated that it was not only the number of nodes
disease behind. High lymph node yields do not necessarily that affected survival but also their location.33
equate to an extensive lymphadenectomy, and it is the Omloo et al. found that an extended resection is
location from which the nodes come that is of higher beneficial in patients with fewer than eight lymph node
importance. Most recently Phillips et al. attempted to metastases.30 Thus, given the difficulty in identifying such
evaluate the impact of the extent of lymphadenectomy on patients, a complete two-field lymphadenectomy should
outcomes from patients with adenocarcinoma undergo- be carried out in all patients where nodal involvement is
ing transthoracic esophagectomy following neoadjuvant a concern. Ideally, based on an individual’s risk factors,
treatment. This study looked at the location of positive it would be possible to tailor surgical approach34 to try
nodes after resection, and evaluated how many extra and optimize outcomes and minimize risks. The ability to
recurrences would have occurred had nodes in respec- tailor the lymphadenectomy so that metastatic nodes are
tive fields been left behind. This study suggested a 23% removed, although technically possible, is fraught with a
decrease in survival had no formal lymphadenectomy high false-negative rate.35
been carried out, but interestingly the impact of leaving
abdominal nodes behind in this cohort of patients with THREE-FIELD LYMPHADENECTOMY
lower esophageal and junctional adenocarcinomas was Debate around the benefit of a three-field lymphad-
predicted as minimal.25 enectomy is long-standing. In Japan its practice has
436 SECTION I Esophagus and Hernia
been widely used for many years, but in the West this
approach is still performed relatively infrequently. This
MORBIDITY OF LYMPHADENECTOMY
may be in part due to concerns regarding morbidity and Extended lymphadenectomy has been associated by some
mortality when including a cervical lymphadenectomy with increased morbidity. Transhiatal esophagectomy
and also due to extrapolation of the already discussed has been shown to have fewer complications than a
argument that lymph node involvement equates to systemic transthoracic esophagectomy with extended en bloc lymph-
disease and lymphadenectomy does not confer a survival adenectomy.19 Respiratory problems are the main cause of
advantage. major problems after esophagectomy, due to pneumonia
Lerut et al. looked at the long-term outcomes of and acute respiratory distress syndrome (ARDS). The
three-field lymphadenectomy in a cohort of patients extensive dissection places the tracheobronchial tree
who received only unimodality treatment (prior to the at risk of direct injury and may also occur in a delayed
integration of neoadjuvant treatment for locally advanced fashion if cautery has been used inadvertently in close
disease). Their findings are of interest, with a 0% survival proximity. Similarly risk to the recurrent laryngeal nerve
for adenocarcinomas of the gastroesophageal junction is increased with a more extensive lymph node dissection,
and 12% survival for distal-third adenocarcinomas where as is risk to the thoracic duct and hence the possibility of
cervical nodes were involved. This rose to 28% 5-year a significant chyle leak.
survival in middle-third squamous cell cancers. The It has been postulated that increased respiratory
implication was that cervical nodes should not necessarily morbidity may be related to one-lung ventilation, which
t
be regarded as metastatic. There were other interesting increases capillary permeability and leads to pulmonary
ne
findings from this study; 25% of patients had cervical lymph edema; however, mediastinal lymphadenectomy also
node involvement, and in three-quarters of these patients impairs lymphatic flow, which can compound this edema.
this had not been detected on the initial staging investiga- A recent study from Sweden sought to establish the
k.
tions. (It is worth noting that endoscopic ultrasound was impact of lymphadenectomy on quality of life on esopha-
used routinely as part of staging, but positron emission geal cancer patients. This study did not find any association
oo
tomography was introduced only in the final year of this with the number of lymph nodes removed and the quality
cohort.) A further interesting finding was that although of life either at 6 months or 5 years post surgery.40
the 5-year survival for distal-third adenocarcinomas with However, despite these concerns regarding morbidity,
yb
positive nodes was 12%, 4-year survival was 36%, perhaps it is inconclusive that a significant difference exists in
indicating this surgery has a role in palliating disease. 8 experienced hands.
There was no significant apparent increase in complica-
tions, and recurrent laryngeal nerve damage was low
er
at 2.6%.
SUMMARY
Although Lerut et al. evaluated a mixed cohort of The American Joint Committee on Cancer revised the
rg
adenocarcinomas and SCCs, Japanese studies involving seventh edition of the TNM staging system for esophageal
patients with SCC indicate almost a third of patients will cancer in 2010. The revised system evaluated the N category
have cervical node involvement, explaining why three-field by looking at the number of positive lymph nodes. This
su
lymphadenectomy is widely used in Japan. has continued on into the eighth edition of the TNM
A number of surgeons have attempted to determine staging system, which has been published and should be
if anatomic location of the tumor should influence the used from 2018.41 However, neither system specifies the
://
need for cervical lymphadenectomy. Chen et al. reviewed number of nodes that constitutes an adequate yield to allow
1715 patients with SCC and found that there was increased accurate staging, and it is well known that an inadequate
tp
cervical node involvement the more proximal the tumor yield may lead to understaging and thus stage migration.
occurred, with 44% of proximal-third tumors having It is felt by some that a comprehensive lymphadenectomy
cervical node involvement, but strikingly 23% of lower-third is associated with increased morbidity due to the increased
ht
tumors also had cervical node involvement.36 This may surgical trauma. However, a number of studies have refuted
indicate the importance of three-field lymphadenectomy this claim and found little difference in morbidity when
for squamous cell cancers. Indeed, a number of system- a more extensive lymphadenectomy is carried out.
atic reviews, of predominantly squamous populations, Although it is true that an extensive lymphadenectomy
have demonstrated superior survival with three-field will lead to more accurate staging, advocates of a minimal
lymphadenectomy.37–39 dissection will point to the fact that there is little evidence
There is a paucity of data for three-field lymphadenec- for postoperative adjuvant treatment on long-term survival.
tomy in patients with adenocarcinoma, particularly in However, such information may be valuable in counseling
the era of neoadjuvant treatments, and the poor overall a patient regarding prognosis, and future studies may
survival demonstrated by Lerut indicates that routine reveal that postoperative chemotherapy does confer a
cervical dissection may not be merited. Each patient with survival advantage.
adenocarcinoma presenting with extensive mediastinal Although there has been a variation in findings of
nodal disease should be considered for a three-field the impact of lymphadenectomy, it would be intuitive to
dissection. The decision will be dictated by extent of suppose that leaving positive nodes behind is likely to
disease, likelihood of recurrence, and overall fitness. The hasten recurrence. However, the effect of neoadjuvant treat-
protocol in our department is to consider each patient in ments may in some patients sterilize any remaining disease.
the multidisciplinary setting and give the options to the It is impossible to predict this impact and so we would
patient and their family. unequivocally advocate an en bloc lymphadenectomy.
Extent of Lymphadenectomy for Esophageal Cancer CHAPTER 40 437
REFERENCES 20. Griffin SM, Burt AD, Jennings NA. Lymph node metastasis in early
esophageal adenocarcinoma. Ann Surg. 2011;254(5):731-737.
1. Stein HJ, Feith M, Bruecher BLDM, Naehrig J, Sarbia M, Siewert JR. 21. van der Schaaf M, Johar A, Wijnhoven B, Lagergren P, Lagergren
Early esophageal cancer: pattern of lymphatic spread and prognostic J. Extent of lymph node removal during esophageal cancer surgery
factors for long-term survival after surgical resection. Ann Surg. and survival. JNCI J Natl Cancer Inst. 2015;107(5):djv043.
2005;242(4):566-573, discussion 573–575. 22. Lagergren J, Mattsson F, Zylstra J, et al. Extent of lymphadenec-
2. Endo M, Yoshino K, Kawano T, Nagai K, Inoue H. Clinicopathologic tomy and prognosis after esophageal cancer surgery. JAMA Surg.
analysis of lymph node metastasis in surgically resected superficial 2016;151(1):32-39.
cancer of the thoracic esophagus. Dis Esophagus Off J Int Soc Dis 23. Chen J-W, Xie J-D, Ling Y-H, et al. The prognostic effect of perineural
Esophagus. 2000;13(2):125-129. invasion in esophageal squamous cell carcinoma. BMC Cancer.
3. Kodama M, Kakegawa T. Treatment of superficial cancer of the 2014;14:313.
esophagus: a summary of responses to a questionnaire on super- 24. Bhamidipati CM, Stukenborg GJ, Thomas CJ, Lau CL, Kozower BD,
ficial cancer of the esophagus in Japan. Surgery. 1998;123(4):432- Jones DR. Pathologic lymph node ratio is a predictor of survival in
439. esophageal cancer. Ann Thorac Surg. 2012;94(5):1643-1651.
4. Alvarez Herrero L, Pouw R, van Vilsteren F, et al. Risk of lymph node 25. Phillips AW, Lagarde SM, Navidi M, Disep B, Griffin SM. Impact
metastasis associated with deeper invasion by early adenocarcinoma of extent of lymphadenectomy on survival, post neoadjuvant
of the esophagus and cardia: study based on endoscopic resection chemotherapy and transthoracic esophagectomy. Ann Surg.
specimens. Endoscopy. 2010;42(12):1030-1036. 2017;265(4):750-756.
5. Ancona E, Rampado S, Cassaro M, et al. Prediction of lymph 26. Hosch S, Kraus J, Scheunemann P, et al. Malignant potential and
node status in superficial esophageal carcinoma. Ann Surg Oncol. cytogenetic characteristics of occult disseminated tumor cells in
2008;15(11):3278-3288. esophageal cancer. Cancer Res. 2000;60(24):6836-6840.
t
6. Leers JM, DeMeester SR, Oezcelik A, et al. The prevalence of lymph 27. Izbicki JR, Hosch SB, Pichlmeier U, et al. Prognostic value of
ne
node metastases in patients with T1 esophageal adenocarcinoma. immunohistochemically identifiable tumor cells in lymph nodes of
Ann Surg. 2011;253(2):271-278. patients with completely resected esophageal cancer. N Engl J Med.
7. Hagen JA, DeMeester SR, Peters JH, Chandrasoma P, DeMeester 1997;337(17):1188-1194.
TR. Curative resection for esophageal adenocarcinoma: analysis 28. Glickman JN, Torres C, Wang HH, et al. The prognostic significance
k.
of 100 en bloc esophagectomies. Ann Surg. 2001;234(4):520-530, of lymph node micrometastasis in patients with esophageal carcinoma.
discussion 530–531. Cancer. 1999;85(4):769-778.
8. Lerut T, Nafteux P, Moons J, et al. Three-field lymphadenectomy 29. Vazquez-Sequeiros E, Wang L, Burgart L, et al. Occult lymph
oo
for carcinoma of the esophagus and gastroesophageal junction node metastases as a predictor of tumor relapse in patients
in 174 R0 resections: impact on staging, disease-free survival, and with node-negative esophageal carcinoma. Gastroenterology.
outcome: a plea for adaptation of TNM classification in upper-half 2002;122(7):1815-1821.
esophageal carcinoma. Ann Surg. 2004;240(6):962-972, discussion 30. Omloo JMT, Lagarde SM, Hulscher JBF, et al. Extended transthoracic
yb
972–974. resection compared with limited transhiatal resection for adenocarci-
9. Akiyama H. Surgery for Cancer of the Esophagus. Baltimore: Lippincott noma of the mid/distal esophagus. Ann Surg. 2007;246(6):992-1001.
Williams & Wilkins; 1990. 31. Rizk NP, Ishwaran H, Rice TW, et al. Optimum lymphadenectomy
10. Fujita H, Sueyoshi S, Tanaka T, Shirouzu K. Three-field dissection for esophageal cancer. Ann Surg. 2010;251(1):46-50.
er
for squamous cell carcinoma in the thoracic esophagus. Ann Thorac 32. Peyre CG, Hagen JA, DeMeester SR, et al. Predicting systemic disease
Cardiovasc Surg. 2002;8(6):328-335. in patients with esophageal cancer after esophagectomy. Ann Surg.
11. Takahashi H, Arimura Y, Masao H, et al. Endoscopic submucosal 2008;248(6):979-985.
rg
dissection is superior to conventional endoscopic resection as a 33. Peyre CG, Hagen JA, DeMeester SR, et al. The number of lymph nodes
curative treatment for early squamous cell carcinoma of the esophagus removed predicts survival in esophageal cancer: an international
(with video). Gastrointest Endosc. 2010;72(2):255-264, 264.e1–2. study on the impact of extent of surgical resection. Ann Surg.
su
13. Bollschweiler E, Baldus SE, Schröder W, et al. High rate of lymph-node 2015;39(10):2492-2499.
metastasis in submucosal esophageal squamous-cell carcinomas and 35. Grotenhuis BA, Wijnhoven BPL, van Marion R, et al. The sentinel
adenocarcinomas. Endoscopy. 2006;38(2):149-156. node concept in adenocarcinomas of the distal esophagus and gas-
tp
14. Siewert JR, Stein HJ, Feith M, Bruecher BL, Bartels H, Fink U. troesophageal junction. J Thorac Cardiovasc Surg. 2009;138(3):608-612.
Histologic tumor type is an independent prognostic parameter 36. Chen J, Wu S, Zheng X, et al. Cervical lymph node metastasis
in esophageal cancer: lessons from more than 1,000 consecutive classified as regional nodal staging in thoracic esophageal squamous
ht
resections at a single center in the Western world. Ann Surg. 2001; cell carcinoma after radical esophagectomy and three-field lymph
234(3):360-367, discussion 368–369. node dissection. BMC Surg. 2014;14(1):110.
15. Law SY, Fok M, Wong J. Pattern of recurrence after oesophageal resec- 37. Ma G-W, Situ D-R, Ma Q-L, et al. Three-field vs two-field lymph node
tion for cancer: clinical implications. Br J Surg. 1996;83(1):107-111. dissection for esophageal cancer: a meta-analysis. World J Gastroenterol.
16. Wayman J, Bennett MK, Raimes SA, Griffin SM. The pattern of 2014;20(47):18022-18030.
recurrence of adenocarcinoma of the oesophago-gastric junction. 38. Ye T, Sun Y, Zhang Y, Zhang Y, Chen H. Three-field or two-field
Br J Cancer. 2002;86(8):1223-1229. resection for thoracic esophageal cancer: a meta-analysis. Ann Thorac
17. Griffin SM, Raimes SA, Shenfine J, eds. Oesophagogastric Surgery: A Surg. 2013;96(6):1933-1941.
Companion to Specialist Surgical Practice. 5th ed. Edinburgh: Saunders 39. Shang Q-X, Chen L-Q, Hu W-P, Deng H-Y, Yuan Y, Cai J. Three-field
Ltd; 2013. lymph node dissection in treating the esophageal cancer. J Thorac
18. Anderegg MCJ, Lagarde SM, Jagadesham VP, et al. Prognostic Dis. 2016;8(10):E1136-E1149.
significance of the location of lymph node metastases in patients 40. Schandl A, Johar A, Lagergren J, Lagergren P. Lymphadenec-
with adenocarcinoma of the distal esophagus or gastroesophageal tomy and health-related quality of life after oesophageal cancer
junction. Ann Surg. 2016;264(5):847-853. surgery: a nationwide, population-based cohort study. BMJ Open.
19. Hulscher JBF, van Sandick JW, de Boer AGEM, et al. Extended 2016;6(8):e012624.
transthoracic resection compared with limited transhiatal resection 41. Rice TW, Ishwaran H, Ferguson MK, Blackstone EH, Goldstraw P.
for adenocarcinoma of the esophagus. N Engl J Med. 2002;347(21): Cancer of the esophagus and esophagogastric junction: an eighth
1662-1669. edition staging primer. J Thorac Oncol. 2017;12(1):36-42.
CHAPTER
Willy Coosemans
| Toni Lerut
t
operation as a transthoracic esophagectomy more safely.
ne
the esophagus2
Over the following decades pioneers, such as Denk,3
1913—W. Denk: cadaver and experimental animal studies
Ohsawa,4 Grey Turner,5 Adam and Phemister,6 Sweet,21
on the transhiatal resection of the esophagus3
Ivor Lewis,7 McKeown,8 Belsey,22 and Orringer,9 further
1933—T. Ohsawa: first report on transthoracic esophageal
k.
developed and refined the surgical techniques as we use
resection and esophagogastrostomy4
them today. However, postoperative mortality remained
1933—G. Turner: first transhiatal resection5
oo
high well into the 1970s. Better insights in medical oper-
1938—W. Adams and D. Phemister: first single stage
ability and better perioperative management hallmarked
transthoracic resection and reconstruction in the United
the 1980s and 1990s where operative mortality was brought
States6
down to below 5%,23 and is now approximately 1% to 2%
yb
1946—I. Lewis: esophageal resection and esophagogas-
in many centers of experience.
trostomy via a right thoracotomy and laparotomy7
Better selection in terms of oncologic operability
1976—K. McKeown: description of a three-hole
through the introduction of the computed tomography
er
esophagectomy8
scan, the positron emission tomography scan, and endo-
1978—M. Orringer: popularizes transhiatal esophagectomy
scopic ultrasound have resulted in a sharp decrease of futile
in the Western hemisphere9
rg
ABSTRACT
Esophagectomy followed by reconstruction is considered
one of the most challenging interventions on the alimen-
tary tract. Today most esophagectomies are performed for
cancer of the esophagus and gastroesophageal junction.
Other indications are decompensated achalasic mega-
eesophagus, sequelae of caustic burns, after multiple redo
surgeries for reflux. The stomach, shaped into a narrow
gastric tube, is the most commonly used conduit for
reconstruction due to its favorable length, reliable vascular
supply, the need for only one single anastomosis, and, in
general, good to excellent sustainable quality of deglutition
and life. However, dependent on each individual patient’s
particularities, reconstruction may require to choose
another conduit. Colon and jejunum—sometimes to be
used as an isolated loop with free vascular anastomosis or
as composite grafts—are the available alternatives. Tubular
skin flaps in an extremely rare situation may become the
t
ne
last resort option. Combined with a multitude of different
access routes, including the recent minimally invasive
techniques, as well as different levels of anastomosis, it is
clear that there are myriad options available when planning
k.
an esophagectomy and reconstruction. A tailored approach
for each individual patient guided by an experienced
oo
surgical team that is familiar with all conduits available
and able to adapt to every situation in order to offer the
patient the best possible type of reconstruction is the key
yb
to success. This chapter provides an in-depth description
of the techniques and results of the whole spectrum of
options for esophageal replacement.
er
KEYWORDS
esophagus
rg
esophagectomy
reconstruction
su
interposition
stomach
colon
://
jejunum
tp
ht
Options for Esophageal Replacement CHAPTER 41 439
t
intramural deposits higher up in the esophagus may
ne
Common Celiac Left gastric Splenic
require the surgeon to switch the level of the anastomosis hepatic trunk artery artery
from the chest to the neck. artery
Finally, a critical determinant is the surgeon’s expertise,
k.
and that of the whole team involved in the pre- and FIGURE 41.1 View on the upper abdominal compartment with
postoperative management. vascularization of the stomach.
oo
In other words, esophagectomy followed by reconstruc-
tion is one of the most complex and difficult operations at the level of the gastroepiploic arcade. In some instances,
on the alimentary tract, which requires an experienced the right gastroepiploic artery may end about half way up
yb
surgical team familiar with all available conduits, and who the greater curvature connecting to the left gastroepiploic
are able to adapt to every situation in order to offer the artery only through delicate arterioles in the omentum,
patient the best possible type of reconstruction. which need to be respected during mobilization (Fig. 41.2).
Historically, the first attempts for reconstruction were Moreover, the pioneering work by Liebermann-Meffert
er
tried by Billroth12 as early as 1879. Skin was the first material et al.,26 using corrosion casts on cadaver specimens, clearly
used for reconstruction,13 but the stomach, colon, and shows that the submucosal plexus is thinning out near the
rg
jejunum—in order of their frequency of use—became top of the fundus, and that there is a watershed zone with
the three classic substitutes over time. They are mostly a clear decrease of the intertwining connections between
used as a single pedicled transposition, but can be used the submucosal microcirculation of the left side (lesser
su
as free vascularized grafts (in particular the jejunum) or curvature) and the right side (greater curvature).
occasionally as composite replacements. Studies have shown that after full mobilization of the
Combining conduit choice with a multitude of different greater curvature of the stomach and ligation of the left
://
access routes, including the recent minimally invasive and right gastric artery, the oxygen tension at the top of the
techniques, and different levels of anastomosis, it is clear fundus decreases substantially, and, after transposing the
tp
that there are a myriad of options available when plan- conduit up to the neck, it falls to approximately 50%.27,28
ning an esophagectomy and reconstruction for cancer. A Efforts have been made to prevent this effect by preopera-
tailored approach for each individual patient guided by tive conditioning of the vascular supply; that is, ligating
ht
an experienced surgeon is the key to success. or embolizing the left gastric artery a few days before the
planned esophagectomy and reconstruction.29–32 However,
the results have been equivocal and one prospective,
STOMACH randomized trial failed to show any advantage. Therefore,
Today, gastroplasty is by far the most preferred conduit it is of paramount importance not to traumatize the top
for replacing the esophagus in over 95% of cases. It is end of the stomach in order to avoid anastomotic leak
indeed the quickest and “simplest” organ surgery for or eventually fundic necrosis.30
reconstruction. It has a robust arterial and venous supply
and submucosal plexus. After mobilization, the stomach TECHNIQUE OF GASTROPLASTY WITH THE
will receive its blood supply from the right gastroepiploic CREATION OF A GASTRIC TUBE
artery with the venous blood drained via the right gas- In general, most surgeons prefer to start with a laparotomy/
troepiploic vein (Fig. 41.1). There is only the need for laparoscopy in order to first inspect the abdominal cavity
a single anastomosis. As the stomach is a very flexible to exclude the presence of unforeseen metastasis before
organ, it can easily reach the neck for a cervical or even mobilization of the stomach. Alternatively, one may prefer
hypopharyngeal anastomosis. Its main disadvantage is to start with a thoracotomy/video-assisted thoracoscopic
the potential of reflux and related aspiration problems. surgery (VATS) in order to assess resectability, particularly
Regarding the “robustness” of the blood supply, the in the case of a questionable T3-4 tumor. Mobilization
surgeon must be aware of some anatomic variations of the stomach is started by opening the gastrohepatic
440 SECTION I Esophagus and Hernia
t
ne
A B
k.
FIGURE 41.2 Favorable (A) and unfavorable (B) pattern of the right gastroepiploic arcade. (From Siewert R, Hölscher A. 3 Eingriffe beim
Ösophaguskarzinom. Jejunuminterposition. In: Siewert, ed. Breitner Chirurgische Operationslehre. Band IV: Ösophagus, Magen und
oo
Duodenum. Vienna: Urban & Schwarzenberg; 1989:33. Fig. 3-36a,b.) yb
ligament so that the hiatus and the right pillar of the vessels here come through the omental intertwining small
right crus come into view. In some patients, there will vessels. At this level, it is important to stay away from the
be a separate left hepatic artery present. If this artery greater curvature and dissect more on the periphery of
is small, a couple of millimeters in diameter, it can be the omentum so these connections can be preserved.
er
ligated without causing harm to the liver. A larger size Coming at the level of the short gastric vessels connecting
left hepatic artery needs to be dissected carefully down to the spleen and the proximal greater curvature, it is again
rg
its origin, usually at the left gastric artery, and both need preferable to stay as close to the spleen as possible when
to be preserved in order to avoid major, possibly lethal, dividing the vessels. Again this allows preservation of the
hepatic necrosis. small delicate connections within the submucosal plexus.
su
This is followed by the mobilization of the greater Preserving as much as possible of the omentum at that
curvature by dividing, distally to the gastroepiploic vascular level also allows it to be used to protect the anastomosis
arcade, the omentum and its feeding vascular branches, later on. Obviously, great care is taken not to damage the
://
as well as the attachments to the transverse colon in this spleen. Today, with the help of ultrasonic devices, the risk
area. Usually, the easier point to start the dissection of the of damage requiring eventually iatrogenic splenectomy
tp
greater curvature is about halfway up the greater curvature. has substantially decreased and, in fact, should no longer
In some cases, due to previous episodes of pancreatitis occur. Once the short gastric vessels are divided, the
for example, the retrogastric area will be obliterated gastrodiaphragmatic area is dissected so that, after retrac-
ht
making the dissection more difficult. The utmost care tion of the fundus to the right, the left pillar of the right
has to be taken not to damage the gastroepiploic vessels. crus will come into view, allowing further dissection of
Especially in laparoscopic mobilization, care should be the entire retrogastric area covering the hiatus.
taken never to grab the gastroepiploic vessels nor the The hiatus can now be opened by incising the phreno-
greater curvature of the stomach with the laparoscopic esophageal ligament (if not already done from above
forceps because of the earlier described very important during the thoracic portion of the case). In case of a
but fragile microcirculation. In laparoscopic mobilization, gastroesophageal junction (GEJ) tumor, a rim of the
it is very helpful to pull on the lesser curvature, as it will diaphragmatic hiatus is excised and left in continuity with
be stapled out afterward. the esophagus in order to ensure a complete R0 resection.
The dissection is further continued upward along Now the inferior mediastinum becomes widely opened
the greater curvature and the gastroepiploic arcade is and accessible so the distal esophagus can be dissected
divided at the level of the inferior pole of the spleen at out and a tape passed around it.
the point where the left gastroepiploic artery ends in the Next comes the ligation of the left gastric artery. This
splenic artery. is done after the dissection of lymph nodes around the
In a number of patients, the right gastroepiploic artery left gastric artery and the celiac axis extending over the
interrupts somewhere midway or two-thirds of the way common hepatic and the splenic artery.
down on the greater curvature (Fig. 41.2). The connections Both the left gastric artery and accompanying vein are
with the left gastroepiploic artery or the short gastric ligated and divided separately. The further dissection is
Options for Esophageal Replacement CHAPTER 41 441
t
lower half and in GEJ cancers.33
ne
The creation of a gastric tube has been enormously
facilitated by the introduction of the staplers. If the access
route is a left thoracoabdominal one, the stapling starts
k.
from above, at the top of the fundus. A linear cutting
stapler will be placed at the top end of the fundus about
oo
5 cm laterally from the GEJ (Fig. 41.3). The staplers are
then placed in a vertical direction. Several staplers will
be necessary. The end result is a long gastric tube with a
yb
width of approximately 4 to 5 cm. If the access is through
a laparoscopic approach, the stapling starts from the
lesser curvature at the level of the crow’s foot, about 4 cm
proximal to the pylorus.
er
esophagus, lesser curvature, and gastric tube will be pulled the fundus. ([A] Reprinted with permission from the Journal of the
up via the cervical or thoracic incision and exteriorized. American College of Surgeons, formerly Surgery Gynecology &
tp
Handsewn Anastomosis
On the outer sides of the esophagus and gastric tube,
two sutures of nonresorbable monofilament 3-0 are
placed between the muscularis of the esophagus and
the seromuscular layer of the stomach. Separate sutures
t
or continuous running suture will complete the posterior
ne
outer layer of the anastomosis. Care is taken to avoid any
tension between the two conduits (Fig. 41.5).
Then with the electrocautery an incision of approxi-
k.
mately 2 cm (maximum 3 cm) is made, widely opening the
lumen of the stomach. This incision is made at a minimum
oo
distance of 1 cm away from the posterior outer layer.
The same maneuver is done on the esophageal wall
comprising the whole width of the esophagus and again
yb
a minimum of 1 cm away from the outer layer. The slimy
mucous content of the gastric tube and esophagus are care-
fully removed with suction and swabs, avoiding any spillage
in the operative field, and both lumina are disinfected
er
A B
FIGURE 41.5 Handsewn anastomosis: (A) posterior mucosal layer; (B) posterior inner layer starting from corner to corner in a running
fashion.
t
ne
the deepest point. The gastric tube is incised at the base
of the pentagon (see Fig. 41.7). Next, monofilament 3-0
k.
resorbable sutures are placed in the corners outwards.
Then, two monofilament 4-0 resorbable sutures are placed
oo
in the middle of the incision, bringing the base of the
E pentagon together and aligning the gastric and esophageal
TUB walls. In between these sutures, a 45-mm linear stapler
yb
is fired over a distance of ~35 mm (Fig. 41.8). This will
create a V-shaped back wall, allowing a wider passage
(Fig. 41.9). This technique prevents narrowing down
during the cicatrization of the anastomosis, resulting in
er
FIGURE 41.6 Trimming of the tip of the gastric tube by a linear improving the patient’s QOL.36,37 The base of the back
stapler. E, Esophagus; TUB, gastric tube. wall, lateral to the stapler line, is completed with separate
4-0 monofilament sutures. Then, the nasogastric tube can
su
redundant proximal part. Otherwise, a blind sac would be pushed through the anastomosis, down in the gastric
form, which can act as a pseudodiverticulum, possibly conduit, to decompress the stomach after surgery. The
impairing the passage of food down the tube. This resec- front wall of the anastomosis is closed with a continuous
://
tion is done using a stapler, taking care that the line of two-layer suture with the earlier placed monofilament
transection is about 2 cm away from the anastomosis line resorbable suture for the inner layer and the nonresorbable
tp
to avoid ischemia with the risk of necrosis of the gastric suture for the outer layer, as shown in Fig. 41.6. The tip
wall in between (Fig. 41.6). The staple line is oversewn of the gastric tube is resected by using a linear stapler as
via a running nonabsorbable 3-0 monofilament suture. described. After careful hemostasis, a small Redon-type
ht
Usually, some omentum is available to wrap around the drain is placed in the neck to prevent the accumulation
suture line as a protection against potential leaking. It of blood, and the incision is closed. This semimechanical
is of paramount importance to avoid any traumatization anastomosis results in improved dysphagia scores for solids
of the tissues by clamps, forceps, or other instruments. and semi-solids, and significantly reduces the need for
As the anastomosis is now finished, the esophagus-gastric dilations, in particular repeat dilatations.37
tube complex is gently pushed back into the thoracic Whether performing a handsewn or semimechanical
inlet and, after leaving behind a Redon-type drain, the anastomosis, before closing the cervicotomy, a mini-
cervicotomy is closed in layers. tracheostomy tube may be inserted through the crico-
thyroid ligament under tracheoscopic guidance from
Semimechanical Anastomosis the anesthesiologist. A mini-tracheostomy facilitates
If the length of the conduit is sufficient (>5 cm overlap), a the aspiration of secretions, which can be of particular
semimechanical Orringer or modified Collard end-to-side assistance to patients who have insufficient strength to
anastomosis is preferred (Fig. 41.7).35 To avoid leakage, cough them up. Then, the platysma muscle is closed,
from traumatizing the tissues the tip of the gastric conduit followed by the skin suture.
will be resected with another linear stapler as described. It
starts by placing five separate 3-0 nonresorbable stitches, in Intrathoracic Stapled Anastomosis
the form of a pentagon, between the muscular esophageal In recent years, as a result of the increased interest in
wall and the gastric serosal layer, with the tip being at minimally invasive esophagectomy (MIE), there has
444 SECTION I Esophagus and Hernia
×
× ×
× ×
S N
A B
t
ne
C D E
k.
oo
yb
FIGURE 41.7 (Left) The modified Collard anastomosis: creation of an esophagogastrostomy on a gastric tube. (Right) (A) Serosal stitches
in pentagonic form. (B) Placement of corner stitches. (C) Linear stapler insertion. (D) Status after firing the stapler resulting in a V-shaped
posterior wall. (E) Manual running closure of the front wall. (Left from Collard JM, et al. Terminalized semimechanical side-to-side suture
er
technique for cervical esophagogastrostomy. Ann Thorac Surg. 1998;65:814–817. Right from Ferguson M, ed. Esophagus. In: Thoracic
Surgery Atlas. Philadelphia: Saunders; 2007:222.)
rg
su
E
://
E
tp
ht
TUB
TUB
Anvil in distal
Esophagus
esophagus
t
ne
C © Heart Lung and Esophageal Surgery Institute
University of Pittsburgh Medical Centre
© Heart Lung and Esophageal Surgery Institute
B University of Pittsburgh Medical Centre
k.
oo
© Heart Lung and Esophageal Surgery Institute
A University of Pittsburgh Medical Centre
FIGURE 41.10 Intrathoracic EEA anastomosis during minimally invasive Ivor Lewis esophagectomy. (A) The EEA anvil is positioned in the
yb
proximal esophagus and secured with a purse-string suture. (B) A gastrotomy is made along the staple line near the tip of the gastric
conduit, through which the EEA stapler is delivered. (C) After the anastomosis is completed, the tip of the gastric conduit is excised to
complete the reconstruction. (From Schuchert MJ, Luketich JD, Landreneau RJ. Management of esophageal cancer. Curr Probl Surg.
er
2010;47:845–946.)
of the transected esophageal wall after making sure the who popularized the technique in the 1980s. Claimed
rg
frozen section is free of tumor. This purse string is inserted advantages are that it preserves part of the stomach and
loosely using a monofilament nonabsorbable 3-0 suture related gastric function and that it has the ability to reach
(Fig. 41.10A). Second, the detached anvil is introduced. the pharynx. In this respect, it also has been advocated
su
This can be somewhat tricky requiring the help of one or for benign diseases. For malignant diseases it is rarely
two Babcock-type clamps to exert counter traction. The used nowadays.
purse string is now securely tied around the central rod The blood supply is based on the left gastroepiploic
://
of the anvil. Often after this maneuver, there may still be artery, which arises from the splenic artery and from the
some tissue protruding that requires the placement of a short gastric vessels, and requires a careful dissection in
tp
second purse string. the hilum of the spleen. In essence, the liberation of the
In both variants, a gastrotomy is then performed to greater curvature is as described earlier. Typically, the
introduce the head of the circular stapler with the tip of right gastroepiploic artery is divided about 4 cm proximal
ht
the shaft fully retracted. The head is carefully positioned to the pylorus (Fig. 41.11). The stomach is now divided
well away from the vertical staple line and ensures that the starting at the greater curvature with a linear cutting
gastric conduit is not twisted. By turning the screw system stapler placed vertically, the subsequent staplers being
of the gun, the central rod with pin is pushed through the placed in parallel with the greater curvature at a distance
gastric wall. Then the anvil and shaft are clicked together of approximately 3 to 4 cm away from the border and up
with the help of clamps (Fig. 41.10B). Further turning on to about two-thirds the length of the greater curvature. If
the screw mechanism on the stapler device approximates a greater length is needed, a modification can be done
the head and anvil tightly. The stapler is then firmly fired by incorporating the pylorus into the tube (Fig. 41.12).
and removed and the two “doughnuts” are inspected for The branches from the right gastroepiploic artery to the
completeness. The esophageal doughnut is submitted for pylorus are preserved and the artery itself is divided close
final pathologic examination of the margin. A nasogastric to its origin from the gastroduodenal artery. To make the
tube is passed through the anastomosis and the gastric conduit, the first stapler is placed at the lesser curvature
remnant is resected using a linear stapler (Fig. 41.10C). at the level of the antrum and then further worked out
in parallel with the greater curvature. The duodenum
REVERSED GASTRIC TUBE is transected with a linear cutting stapler just below the
The principle of the reversed gastric tube was described pylorus. Bringing up a Roux-en-Y jejunal limb to the distal
by Beck and Carrel14 as early as in 1905, and later by stomach in order to avoid biliary reflux restores continuity
Jianu,39 but it is mainly the Rumanian surgeon, Gavriliu,40 with the duodenum.
446 SECTION I Esophagus and Hernia
A B C
t
ne
k.
oo
yb
er
D
rg
FIGURE 41.11 Graphic depiction of the technique to create a reversed gastric tube (Gavriliu) for esophageal replacement. (A) Inspection of
the blood supply to the stomach and preservation of the gastroepiploic artery for creating the tube. (B) The use of a stapler to create the
su
tube along the greater curvature of the stomach. (C) The completed reversed tube is brought up to the chest for the esophageal
anastomosis. (D) Perioperative view. ([A, B, C] From Chandler NM, Colombani PM. Ashcroft’s Pediatric Surgery. Philadelphia: Elsevier;
2014:351–364.)
://
Short gastric
arteries
t
Left gastro-
ne
FIGURE 41.12 Extended reversed gastric tube. (Illustration epiploic
artery
reprinted with permission from Steichen FM, Wolsch RA, eds.
Mechanical Sutures in Operations on the Esophagus &
Gastroesophageal Junction. Woodbury, CT: Cine-Med; 2005.)
k.
oo
profile over time. This is due to autonomous recovery of
acid secretion from the parietal cells.
yb
Consequently, a substantial number of patients will
suffer from reflux symptoms after esophagectomy and
gastroplasty. The main symptoms are heartburn, regurgita-
tion, dysphagia, vomiting, and aspiration pneumonia.
er
and regurgitation at 3 months, increasing to 21% after 1 FIGURE 41.13 (A) Preparation of the nonreversed gastric tube. The
year. There was a significant difference in the incidence circular stapler creates the defect required for insertion of the
tp
of reflux symptoms, esophagitis, and anastomotic stenosis linear stapler. (B) Omental wrap of the nonreversed gastric tube
when comparing infraaortic intrathoracic anastomosis before cervical transposition. (From Fell SC, Ximenes-Netto M.
with cervical anastomosis. At 3 months, reflux symptoms Gastric tubes: reversed and nonreversed. In: Patterson G, et al.
ht
were present in 5% of patients with a cervical anastomosis, Pearson’s Thoracic and Esophageal Surgery. 3rd ed. Philadelphia:
Churchill Livingstone; 2008:656–662.)
and in 30% of patients with an intrathoracic anastomosis
(P = .097). At 1 year, these figures were 4% and 50%,
respectively (P = .001).
Esophagitis at 3 months was present in 6% after cervical pump inhibitors) has resulted in a major decrease of
anastomosis and 43% after intrathoracic anastomosis persistent disabling reflux-related problems.
(P = .02). At 1 year, these figures were 8% and 53%,
respectively (P = .001). Gastric Emptying–Related Symptoms:
Anastomotic stenosis was present at 3 months in 12% Pyloric Drainage, Yes or No?
after cervical anastomosis and 30% after intrathoracic Vagal denervation results in chronic dysmotility of the
anastomosis (6% and 17%, respectively [not significant], gastric remnant and an outlet dysfunction of the pylorus,
at 1 year). The latter figures suggest that, contrary to which may cause delayed emptying. This may induce a wide
general opinion, cervical anastomosis does not necessarily spectrum of symptoms: early satiety, postprandial fullness,
result in a higher incidence of stenosis when compared heartburn, high dysphagia, aspiration, and pneumonia.
with intrathoracic anastomosis. Gutschow et al.42 reported This spectrum of symptoms affects up to half of the patients
that 38% of patients had reflux in the remnant esophagus and is truly disabling in approximately 5% to 10%.
at 3 years or more after esophagectomy. Fortunately, the The addition of a gastric drainage procedure, which can
introduction of potent antiacid medication (i.e., proton be a pyloroplasty, pyloromyotomy, digitoclasy, or botulinum
448 SECTION I Esophagus and Hernia
t
time. From this trial, the authors concluded that routine .01), 52% (P = .001), and 49% (P = .01), respectively. At 6
ne
pyloroplasty is not necessary after esophageal cancer months, the differences between the two groups became
resection and reconstruction when using the stomach smaller (for a solid bolus, 92% vs. 89%; for a full meal, 88%
as a substitute. vs. 73%), but the percentage of asymptomatic patients in
k.
Chattopadhyay et al.47 randomized 24 patients who the pyloroplasty group remained significantly higher than
underwent esophagectomy for esophageal cancer into that in the no pyloroplasty group (86% vs. 53%; P = .01).
oo
a pyloroplasty and a no pyloroplasty group. Pre- and Gastric emptying was further assessed by studying the 50%
postoperative gastric emptying was evaluated using a emptying time of an indium 113--labeled semi-solid meal
technetium 99m--labeled liquid meal. All surviving patients at 6 months. There was significantly slower emptying in
yb
were followed up for between 6 months and 4 years. the no pyloroplasty group compared with the pyloroplasty
There was no difference in clinical outcome (i.e., fullness, group (24.3 and 6.6 minutes, respectively; P = .01).
regurgitation, vomiting, heartburn) between the two These studies seem to indicate a trend favoring pyloric
groups. Postoperative gastric emptying was significantly drainage for both the early and late outcome of gastric
er
delayed in both postoperative groups compared with the emptying, food intake, and related nutritional status.
preoperative results. Although the delay in gastric emptying Nevertheless, some patients, although asymptomatic, may
rg
was significantly less (P = .001) in the pyloroplasty group, have delayed gastric emptying, whereas others are symp-
it was still significantly prolonged when compared with tomatic while having normal gastric emptying, indicating
preoperative values (P = .001). The authors concluded that there is a great variation in the individual pattern of
su
that postoperatively, patients may have symptoms as a gastric emptying, and thus reflecting individual differences
result of this delayed emptying, but pyloroplasty fails to in gastric tube activity (Table 41.1).
effectively prevent them, thus questioning the need for A meta-analysis by Urschel et al.50 of all existing ran-
://
randomizing 20 patients in the pyloroplasty and 20 patients and obstructive foregut symptoms, but favoring no drainage
in the no pyloroplasty group following esophagectomy for to prevent bile reflux-related complications.
carcinoma and retrosternal gastric pull-up of the whole Finally, a recent meta-analysis by Akkerman et al.51
ht
stomach with cervical esophagogastric anastomosis. In concluded that the results regarding the benefit of pyloric
the pyloroplasty group, aspiration, food vomiting, and drainage procedures on gastric function remain contradic-
heartburn were absent. One patient complained of early tory, but do not support its routine use. More recently,
satiety. In the no pyloroplasty group, aspiration occurred the botulinum toxin (Botox) injection directly into the
in four patients, with three fatal outcomes, food vomiting sphincter has been advocated as a promising method to
in one, early satiety in two, and heartburn in two. The prevent or relieve obstructive symptoms. Cerfolio et al.52
results of this trial suggest that pyloroplasty should be compared pyloroplasty, pyloromyotomy, no drainage, and
performed on the retrosternal stomach in order to prevent preoperative Botox injection into the pylorus. At day four,
the potentially lethal effects of gastric stasis. gastric delay as measured by a timed barium swallow was
The largest randomized study was performed by Fok 96%, 93%, 96%, and 59%; P = .001. Hospital length of stay
et al.49 with a meticulous analysis of eating abilities and (P = .015) and operative times (P = .037) were shorter in
gastric emptying function in the early postoperative the Botox group. Follow-up (mean, 40 months) showed
period as well as during long-term follow-up. In this study, symptoms of biliary reflux to be lowest in the Botox group
the whole stomach had been used for reconstruction (P = .024). They concluded that injection of the pylorus
in all patients. In the early postoperative period, the with Botox at the time of esophagogastrectomy is safe, and
daily nasogastric aspirate was not significantly different. decreases operative time when compared with pyloroplasty
Subjective eating abilities at 2 weeks were as follows: for or pyloromyotomy. In addition, it can improve early gastric
the pyloroplasty group the ability to consume a solid bolus emptying, decrease respiratory complications, shorten
Options for Esophageal Replacement CHAPTER 41 449
t
was associated with 5% delayed emptying, 14% pneumonia, improve gastric emptying in normal subjects and in patients
ne
and 10% reflux. For transhiatal esophagectomy, these with diabetic gastroparesis or postvagotomy gastroparesis.
figures were 4%, 5%, and 14%, respectively. In a randomized, clinical trial, Burt et al.59 showed that
the percentage of gastric retention, as measured by a
k.
Gastric Emptying–Related Symptoms: Gastric Tube technetium 99m--labeled solid meal, at 90 minutes was
Versus Whole Stomach? 37% in patients receiving erythromycin versus 88% in
oo
Bemelman et al.55 studied the impact of the size of the patients receiving a placebo (P < .001). These studies
gastric substitute on delayed postoperative emptying. were performed in the immediate postoperative period.
Delayed gastric emptying was seen in 38% of patients in Further studies are needed to investigate the impact of
yb
whom the whole stomach was used; 14% of patients in prolonged administration of erythromycin.
whom substitution with distal two-thirds stomach was per- In patients with persistent symptoms of gastric outlet
formed; and 3% of patients in whom a tubulized stomach obstruction, balloon dilatation has become a valuable
was used. The addition of a pyloroplasty in each of the option. Lanuti et al.60 reported a successful outcome in
er
three groups did not affect the incidence of delayed gastric 36 out of 38 of the patients after pylorus dilatation for
emptying. The authors suggest that the small gastric tube refractory gastric outlet symptoms. A total of 42% had
rg
leads to a rapid increase of intraluminal gastric pressure no previous pyloric drainage procedure, while 52% did.
when the stomach is filled, facilitating gastric emptying Finally, in cases of persistent disabling symptoms despite
by the effect of the law of Laplace. Consequently, more such treatment, rescue pyloroplasty is a valuable last resort
su
gastric stasis occurs when the whole stomach is used for option as reported by Datta et al.61 In this report, 9 out
esophageal replacement. Similar results were reported of 13 patients were successfully treated.
by Barbera et al.56
://
Collard et al.,57 on the other hand, found a better Intestinal Metaplasia and Gastric
recovery of gastric motility, as reflected on manometric Drainage Procedures
tp
tracings, when using the whole stomach versus a gastric The combination of biliary and acid reflux is commonly
tube. The motility index progressively increased with believed to play a central role in the pathogenesis of Barrett
time in both groups of patients, but motor recovery was metaplasia in patients suffering from gastroesophageal
ht
better in whole-stomach patients than in those receiv- reflux disease. The ablation of the lower esophageal
ing a gastric tube. Even after 3 years, motor recovery sphincter mechanism at the time of esophagectomy and
remained significantly higher in whole-stomach patients. the vagotomy-induced pyloric dysfunction with possible
These differences might be explained by the fact that related enterogastric biliary reflux, are of increasing
resection of the lesser curvature partly destroys both the concern in relation to the risk of developing Barrett
organizer and effector command ganglia in the myenteric metaplasia, especially in long-term survivors. As early as
plexus. Long-term alimentary comfort was suggested to 1992, da Rocha et al.62 reported that 8.3% of 48 patients
be significantly better with an interposed whole stomach developed Barrett metaplasia in the esophageal remnant in
than after gastric tube reconstruction. the long term, after subtotal esophagectomy. Franchimont
The results of the studies by Bemelman et al. and et al.63 found a 13.5% incidence of newly developed Barrett
Collard et al. are clearly in conflict, and further elucidation esophagus in the cervical stump after esophagectomy
of strictly mechanical effects versus intrinsic functional for cancer with a median time to diagnosis of 489 days
effects is necessary. From a systematic literature review, (range 43 to 1172 days). Proton pump inhibition early
Akkerman et al. 51 found a superiority of the gastric after surgery did not appear to influence the development
tube compared with the whole stomach in most of the of Barrett esophagus.
studies. Pyloric drainage is not significantly associated In a series of 39 patients, Oberg et al.64 noticed a 47%
with the risk of developing delayed gastric emptying after prevalence of metaplastic columnar mucosa within the
esophagectomy. cervical esophagus. Intestinal-type metaplasia was found
450 SECTION I Esophagus and Hernia
in three patients, who all showed an abnormal exposure the viscosity of the intraluminal contents, or to drugs such
to both acid and bilirubin on 24-hour monitoring. as the α-glucosidase inhibitor acarbose, which decreases
Other studies suggested that gastric drainage procedures the rapid absorption of glucose, or native somatostatin
favor enterogastric bile reflux. or the somatostatin analogue octreotide, which alters gut
Wang et al.65 compared patients with and without transit and inhibits the release of vasoactive mediators
pyloroplasty. Those with pyloroplasty were found to have into the bloodstream.72
a higher incidence of bile regurgitation (55.5% vs. 8.6%,
respectively). Thirty-three of these patients underwent a Quality of Life After Surgery
technetium 99m--hydroxyiminodiacetic acid (HIDA) test Advances in proper selection, surgical techniques, including
showing a high incidence of enterogastric bile reflux (60%), MIE, as well as the multimodality strategies, have resulted
whereas no enterogastric bile reflux could be demonstrated in significantly better outcomes for esophageal carcinoma
in patients who had not had a gastric drainage procedure. patients over the last two decades. Currently, 5-year survival
Gutschow et al.66 found a prevalence of esophagitis figures close to 50% are no longer an exception. However,
of 26.5% in patients who received a gastric drainage significant morbidity as a consequence of postoperative
procedure versus 9.5% in patients who did not. In the complications can affect up to half of all patients who have
group of patients receiving a gastric drainage procedure, surgery.73 This morbidity can be related to a multitude of
6.7% developed Barrett metaplasia. It is currently unknown factors, but is mainly due to anastomotic complications
whether these patients have the same risk of developing (e.g., leak or stenosis), pulmonary complications as a direct
t
adenocarcinoma as is seen in the classic reflux-induced consequence of the surgery itself (e.g., blood loss, length
ne
Barrett’s population. But these data are suggesting that it is of surgery with prolonged intraoperative atelectasis), or
better not to perform any type of drainage procedure. In as the consequence of reflux-related chronic aspiration,
cases of persistent disabling symptoms, balloon dilatation or due to undesired functional side effects as described
k.
appears to be a viable option. earlier (e.g., gastric outlet obstruction). It takes up to 1
year for patients to recover their baseline health-related
oo
Dumping and Diarrhea quality of life (HRQL), and this baseline level is never
After esophagectomy followed by gastroplasty, many patients achieved if disease recurs within 2 years.74 As a result, a
complain of diarrhea and dumping (-like) symptoms, with paradigm shift is causing more focus on QOL.
yb
a reported incidence of between 10% and 50%. McLarty Obviously, it is of paramount importance for surgeons
et al.67 suggested that symptoms are early postprandial to make relentless efforts to minimize complications
abdominal and vasomotor symptoms resulting from osmotic through meticulous surgical techniques or by develop-
fluid shifts and the release of vasoactive neurotransmitters, ing newer less traumatic techniques. In this respect,
er
and late symptoms are secondary to reactive hypoglycemia. the advent of minimally invasive surgery seems to be
Diarrhea, abdominal cramps, nausea, dizziness, postpran- promising, as shown in a randomized, controlled trial by
rg
dial sweating, and hypotension are the main complaints. Biere et al.75 Several studies have shown that MIE has a
These dumping symptoms are thought to be provoked beneficial effect on postoperative outcome and HRQL. A
by the accelerated gastric emptying. Hölscher et al.68 study by Nafteux et al.76 compared the outcome between
su
demonstrated a markedly accelerated emptying of semi- open esophagectomy (OE) and MIE for early cancer.
solid food in comparison to the stomach of a control The blood loss was less (P = .01) and the duration of
group, but complete emptying of the intrathoracic stomach operation longer (P = .001) in MIE. Hospital mortality
://
was very much delayed compared with transit through a (P = .66) and postoperative complications (P = .34) were
normal esophagus. Banki et al.69 compared the results comparable. However, respiratory complications (P =
tp
of a vagal nerve sparing esophagectomy plus coloplasty .008) and intensive care unit admission (P = .02) were
versus a control group and a group of patients after higher in OE. Gastrointestinal complications (P = .005),
standard esophagectomy with gastric pull-up. Patients with that is, gastroparesis (P = .004) were more frequent in
ht
a vagal sparing esophagectomy had a complete absence MIE. At 3 months, postoperative fatigue, pain (general),
of postoperative diarrhea and a low (7%) incidence of and gastrointestinal pain were less in MIE (P = .09, P =
dumping. In the gastric pull-up group, the incidence of .05, and P = .01, respectively). In the following months,
diarrhea was 50% and that of dumping 10%. Some authors these differences faded out, and at 1 year there was no
argue that a gastric drainage procedure increases the longer a difference. However, the values did stay below
incidence of dumping. In fact, 10% to 30% of patients the preoperative baseline, indicating long-term persistence
undergoing pyloroplasty will develop dumping syndrome, of the impact of the intervention on HRQL.
1% to 5% being refractory to conservative management.70 Reducing postoperative complications and related
Sinha et al.71 compared the results of subtotal esophagec- morbidity also translates into earlier discharge, which may
tomy and cervical esophagogastrostomy with and without have a beneficial effect on postoperative overall HRQL.
pyloroplasty. Evidence of dumping on a provocation test This was clearly shown in another study by Nafteux et al.77
was noted in 18% of the pyloroplasty group but in none of Indeed, patients with early discharge (length of stay [LOS]
the nonpyloroplasty group. An effective relief of dumping <10 days) indicated, at 3 and 12 months postoperatively,
symptoms can be achieved with dietary modifications to significantly better HRQL scores in the functional scales
minimize the ingestion of simple carbohydrates and to (physical, emotional, social, and role functioning) and in
exclude fluid intake during the ingestion of the solid the symptoms scales (fatigue, nausea, dyspnea, appetite
portion of a meal. More severely affected patients may loss, and dry mouth) when compared to those patients
respond to agents such as pectin and guar, which increase with LOS greater than 10 days. Return to the level of the
Options for Esophageal Replacement CHAPTER 41 451
reference population scores was achieved at 1 year in the entire esophagus up to the pharynx. The blood supply
LOS 10 days or less group for almost all the scales but from the left colic artery is robust, and the presence of
not in the LOS greater than 10 days’ group. a marginal artery, the artery of Drummond, close to the
A well-known downside of MIE is its steep learning colon, permits a linear interposition procedure without
curve. It is generally accepted that a minimum of 25 MIEs redundancy or mechanical kinking (Fig. 41.14A). If the
is required to overcome the learning curve. This may be dif- stomach is retained, it offers the potential of reduced
ficult to obtain given the fact that most centers are dealing delayed gastric emptying and reduced reflux. However, it is
with a rather limited number of annual esophagectomies. a complex procedure requiring at least three anastomoses
Therefore the experience of the surgeon and the entire and thus longer operative time and higher rates of necrosis,
team involved is of paramount importance to optimize morbidity, and mortality have been reported.
not only the oncologic but also the functional outcome of Its use is contraindicated in the presence of an aortic
esophagectomy. This is particularly true in an era where aneurysm or extensive atheromatosis disease involving the
other nonsurgical approaches (endoluminal resections for superior mesenteric artery, previous colon surgery, severe
early cancer, definitive radiochemotherapy for advanced colonic inflammatory bowel disease (IBD), or tumors.
squamous cell carcinoma) are increasingly challenging Scattered diverticulosis may be a relative contraindication.
the outcomes of surgery as the cornerstone when aiming In the past, colonic interposition was preferred over
at therapy with curative options for esophageal cancer. gastric pull-up in young patients with early cancer and
thus a long life expectancy. However, since early cancers
t
(T1aN0M0) are now mostly treated by endoluminal resec-
COLON
ne
tion, and due to the increasing experience with gastric
Vuillet17 using the left colon on a cadaver, and Kelling18 pull-up and the related excellent-to-very-good functional
using transverse colon in a patient with esophageal cancer, results obtained, the colon nowadays is rarely used as the
k.
in 1911 independently described the clinical principles for first option. A colon interposition is used as a second option
the use of the colon as an esophageal substitute. In the substitute when the stomach cannot be used or when the
oo
1950s and 1960s, the use of the colon as a substitute was stomach has to be resected for oncological reasons, such
further popularized by Orsoni,78 Reboud,79 Waterston,80 as adenocarcinoma involving the distal esophagus as well
Belsey,81 Lortat-Jacob,82 and others. as the majority of the lesser curvature .
yb
The main advantage of using a colon is its versatility. The blood supply from the colon comes from both the
The length is freely available for replacement of the superior and the inferior mesenteric artery. Classically,
er
Middle colic Superior Marginal Left colic Arc of Riolan Marginal artery
artery mesenteric artery artery artery of Drummond
rg
su
Right
colic
artery
://
tp
Ileocolic
artery
ht
FIGURE 41.14 Vascularization of the colon. (A) Classic pattern. (B) Arc of Riolan. ([A] from Albright JB, Beaty J. Colorectal Surgery.
Philadelphia: Elsevier; 2013:403–425. [B] from Gordon PH, Nivatvongs S, eds. Principles and Practice of Surgery for the Colon, Rectum
and Anus. 2nd ed. St. Louis: Quality Medical Publishing; 1999:27.)
452 SECTION I Esophagus and Hernia
t
arteries proximal to the roots of the superior mesenteric the classical preparation group (P = .03). The addition
ne
artery, whereas the marginal of Drummond exists well of oral antibiotics may further reduce the risk of infec-
distal to the roots of the mesenteric artery. The presence tion, but this is controversial. Just before the start of the
of an arc of Riolan usually precludes the use of a long- surgery and for the subsequent 48 hours, broad spectrum
k.
segment colon interposition. Among the many technical antibiotics, including anaerobe covering, are administered
options, there are two main types of grafts traditionally intravenously (Table 41.2).
oo
used for esophageal reconstruction (Fig. 41.15). The right
colon graft is created using the middle colic vessels as a Technique of Left Colon Interposition
pedicle; this usually involves dividing the right colic and The preferred technique is a left colonic interposition.
yb
ileocolic vessels. A segment from the terminal ileum to The arterial flow comes from the left vessels, and is inter-
the ascending colon is interposed in an isoperistaltical posed in an isoperistaltic way.84 This technique allows for
fashion. A left colon graft is created based off the ascending adequate length to make an anastomosis either in the
branch of the left colic artery and the inferior mesenteric thorax or in the neck with an excellent vascular supply
er
vein as a pedicle. The middle colic vessels are divided, if no compromising anatomic variations are present.
preserving the communication between the right and left The left colon is almost completely mobilized from its
rg
branches of these vessels. A segment from the transverse peritoneal attachments by incising the white line of Toldt
colon to the splenic flexure is used for interposition in down to where the inferior mesenteric artery is identified.
an isoperistaltic fashion. In a long-segment interposition, The transverse colon is detached from the omentum and
su
the hepatic flexure is also included. This may require the the splenic flexure is mobilized as well. On the right side,
division of a right colic artery. if a long-segment interposition is planned, the hepatic
://
tp
ht
A B C
FIGURE 41.15 Modalities for colon interposition. (A) Right colon pedicled on middle colic artery. (B) Left colon pedicled on the ascending
branch of left colic artery. (C) Ileocolic graft pedicled on the right colic artery. (From Watanabe M, Mine S, Nishida K, Kurogochi T,
Okamura A, Imamura Y. Reconstruction after esophagectomy for esophageal cancer patients with a history of gastrectomy. Gen Thorac
and Cardiovasc Surg. 2016;64:457–463.)
Options for Esophageal Replacement CHAPTER 41 453
t
ne
FIGURE 41.16 Transillumination. (From Popovici ZI. Atlas of
flexure, the right colon, and the cecum are mobilized Advanced Operative Surgery. Philadelphia: Elsevier;
by incising the peritoneal fold. Once the colon is fully 2013:113–125.)
k.
mobilized, the arterial vascularization is identified. In a
slim patient, the vessels can be easily seen in the mesentery
oo
of the colon. In obese patients, this may be more difficult
as the mesenteric fat obscures the view. This can be solved
using transillumination (Fig. 41.16). A thorough inspection
yb
of all the vessels is performed with special attention to the
continuity of the marginal arteries and possible anatomic
variations of the colic arteries. The ascending branch of
the left colic artery is identified. Usually, this is a robust
er
Ligated midcolic
artery, and pulsations can easily be palpated. In a left vessels
colic transposition, this will be the artery on which the
rg
Marginal artery
measured. The real length needed is now measured over
the artery and marginal arteries—not by measuring the
tp
Ascending branch
needed, they are also placed at the base of the right colic of left colic artery
artery and on the marginal artery at the site of the proximal
part of the colon selected for transection according to Inferior mesenteric
the measurement. The blood to the future conduit now vein
comes only from the ascending branch of the left colic
artery. Usually, pulsations are visible and palpable over the
entire length of the isolated segment. This may require
some time as the colon may have gone into spasm as a FIGURE 41.17 The measurement of the coloplasty is based on the
result of the dissection. The topical application of lidocaine length of vascular arcades rather than on the length of the colon
or papaverine may relieve the spasm. If pulsations are itself. (From DeMeester TR. Esophageal replacement with colon
not evident, removing the clamps and reapplying them interposition. Oper Techn Cardiac Thorac Surg. 1997;2:73–86.)
after some 20 to 30 minutes may be needed to solve the
problem. In the absence of visible/palpable pulsations,
Doppler flow measurement may be used. marginal artery at the point of proximal transection of
The middle colic artery can now be divided and ligated. the colon is clamped, but only at the side of the colon
To do so, both the artery and the vein are dissected out remaining in situ. After the transection of the marginal
as close as possible to their origin. This is a most delicate artery, the flow to the nonclamped side can be immediately
maneuver and should be done using fine clamps. The evaluated.
454 SECTION I Esophagus and Hernia
The proximal part of the colon is now transected and carefully cleaned with an iso-Betadine swab. On the
allowing the entire mobilized loop to be moved upward. gastric side, the anastomosis is placed close to the greater
The transection is performed by using a linear cutting curvature one-third of the length down from the fundus
stapler. The colon is passed behind the stomach through to the pylorus on the posterior side of the stomach. An
an avascular opening in the lesser gastrohepatic omentum 8- to 10-cm segment of the graft is retained underneath the
up to the level of the hiatus, usually by attaching it to the diaphragm in the high pressure zone, creating an antireflux
distal esophagus before exteriorizing the esophagus via device similar to the principles governing a classic antireflux
the cervical incision. Great care is taken not to axially procedure (Fig. 41.21). After finishing the anastomosis, the
twist the colonic interposition. fundus will fall as a flap valve over the intraabdominal part
The colon interposition, by now in spasm, is gently of the colon acting as the additional, second component of
stretched out so the precise length needed to make the an effective antireflux barrier against reflux colitis. This is
anastomosis with the stomach can be determined. This an essential feature of the procedure. The anastomosis is
maneuver guarantees a straight position of the conduit made with an inner layer of 3-0 absorbable and an outer
in the chest, minimizing the risk for redundancy whilst layer of 3-0 nonabsorbable suture material, with both layers
avoiding too much tension that would compromise the using a running suture.
vascularization at the top end. At this point, the left The colocolic anastomosis is now fashioned between
colon will be transected using a linear cutting stapler. the right and left colon, which are easily brought together
The marginal artery is left intact, but its small branches given the extensive mobilization. The opening in the
t
to the colon are dissected and divided over a distance mesenterium of colon can be closed to avoid herniation
ne
of about 1 cm on both sides of the transection line each and possible strangulation of the small intestines, or can
Figs. 41.18 and 41.19. be left widely open.
The first anastomosis is the cologastric anastomosis (Fig. The cervical anastomosis is fashioned in exactly the same
k.
41.20). This is an end-to-side anastomosis. The linear staple way as described before when using the stomach. However,
line at the top end is removed and the colon is opened the preference is here to use handsewn anastomosis as
oo
yb
2 cm
er
rg
su
://
B
tp
ht
A C
FIGURE 41.18 (A) Preparation of a left colon loop pedicled on the ascending branch of the left colic artery. (B) The distal marginal artery is
not interrupted; the branches to the colon are divided. (C) Perioperative view. ([B] from DeMeester TR. Esophageal replacement with
colon interposition. Oper Techn Cardiac Thorac Surg. 1997;2:73–86.)
Options for Esophageal Replacement CHAPTER 41 455
MC
AC
AC
A B
FIGURE 41.19 Perioperative view: (A) Short segment. (B) Long sement. AC, Ascending branch of left colic artery; MC, ligated middle colic
artery.
t
ne
k.
oo
yb
er
rg
su
://
tp
FIGURE 41.20 Cologastric anastomosis on the posterior wall of the stomach close to the greater curvature one-third of the length down
from the top of the fundus to the pylorus. (Illustration based on La Chirurgia dell’esofago. Stipa – Belsey. Atlante: trattamento del
carcinome dell’esofago e del cardias. Esofago-colo-gastroplastica. Piccin Editore Padova; 1980:463. Fig 14A.)
ht
it allows for better adaptation for possible incongruence vascularization at the future top end of the marginal
between cervical esophagus and the colon (Fig. 41.22). artery. In the scenario of a long segment, the cecum and
At the time of the cervical anastomosis, a nasogastric last ileal loop may be needed as part of the interposition.
tube is pushed down through the anastomosis into the In this situation, the right ileocolic artery also has to be
colon and through the cologastric anastomosis in order divided. The further steps of the intervention are similar
to decompress the stomach. A temporary placement of to those described for the left colon.
a gastrostomy may be useful as well.
Variations
Technique of Right Colon Interposition Most authors prefer isoperistaltic colon interposition
The right colon is used as a first choice by some surgeons, assuming that the colon as substitute retains its capacity to
or in case the vascularization of the left colon is compro- episodically propel the solid bolus in an aboral direction.
mised, such as after previous sigmoid resection.85 The graft Therefore, placing the conduit in an antiperistaltic fashion
will be based on the middle colic artery. The principles is believed to increase the risk for aspiration.84 However,
of dissection and measuring the length needed are the in some situations where the surgeon, due to variations
same as for the left colon interposition. The right colic in the vascular anatomy, is forced to pedicle the conduit
artery is identified and along this artery the mesenterium on the right colic artery, an antiperistaltic interposition
is incised. Test clamping is performed to assess sufficient of the conduit may be the only option.
456 SECTION I Esophagus and Hernia
Other variations are often related to technical, mostly “V.” In such a situation, the solution is to place a DeBakey
vascular, difficulties discovered at the time of surgery. vascular clamp on the mesenteric artery followed by the
Occasionally the middle colic artery may split immediately resection of the middle colic artery with a patch suturing
after its origin from the superior mesenteric artery and the mesenteric artery and the patch with fine Prolene
without an overarching marginal artery in between the 6-0 (Fig. 41.23).
When impaired venous drainage is suspected, a “super-
drainage” can be performed in order to avoid congestion
of the transplant. The marginal vein is anastomosed to
the anterior or exterior jugular vein, or the internal
thoracic vein, using microsurgical techniques.85,86 The
so-called “supercharged colon interposition” consists of
an arterial anastomosis to avoid ischemic necrosis of the
graft, usually between the stump of the right or middle
colic artery and the superior thyroid artery or internal
mammary artery.87
1/3 Outcomes
Colon interposition is a very complex intervention, and
t
the reported postoperative mortality after esophagectomy
ne
for cancer ranges between 10% and 20% being somewhat
higher than the postoperative mortality after gastric
pull-up.88,89 One of the specific causes of mortality or
k.
severe morbidity is graft necrosis, which, according to
the literature, is seen in approximately 5% of the cases.
oo
Common causes are damaging the vessels by clamps or
ligatures at the time of surgery, leading to thrombosis
2/3
and subsequent necrosis, rotation of the vascular pedicle
yb
at various time points during the surgery—in particular
when bringing up the conduit into the neck—tearing off
vessels during the same maneuver, too much narrowing of
the hiatus strangulating the vascular pedicle, and failure
er
the intraabdominal part of the coloplasty. (Based on La Chirurgia by staged reconstruction when the patient has completely
dell’esofago. Stipa – Belsey. Atlante: trattamento del carcinome recovered.
dell’esofago e del cardias. Esofago-colo-gastroplastica. Piccin Anastomotic leaks after short-segment coloplasty with
su
Editore Padova; 1980:461. Fig 12.) intrathoracic anastomosis are rare, but after long-segment
://
tp
ht
t
ne
k.
oo
A B
yb
FIGURE 41.24 (A) Protruding colon at the cervical anastomosis
level. (B) Redundancy of the colon interposition after retrosternal
placement.
er
FIGURE 41.23 Middle colic artery with V split close to the superior 38 years. Seventeen patients (25%) developed redundancy,
mesenteric artery: resection with patch. (Based on DeMeester TR. which was noticed at three levels: supraortic, supradia-
tp
Esophageal replacement with colon interposition. Oper Techn phragmatic, and infradiaphragmatic. All patients were
Cardiac Thorac Surg. 1997;2:73–86.) symptomatic, requiring revisional surgery in 15 cases
to solve the problem. de Delva et al.91 described their
ht
interposition with anastomosis in the neck, leaks are experience with revisional surgery for late complications.
observed in about 10%. They usually can be treated They treated 12 patients for redundancy, performing
conservatively and require rarely revisional surgery.88,89 A what they called a “box car” resection. This is a segmental
particular problem at the cervical level of the coloplasty is resection of the redundant part, preserving the marginal
bulging of its supraclavicular portion. Due to air swallowing artery followed by reanastomosis.
during speech or while eating, the thin walled colon bulges Another late complication is the occurrence of fibrosis
out. Beside the unaesthetic aspect, the bulging may, in the at the top end of the graft. This is believed to be the
long term, cause dysphagia so the patient has to manually result of a venous ischemia caused by congestion of the
push down the food bolus. This would eventually require venous drainage. Such strictures are very difficult to
revisional surgery by excising the protruded part (Fig. dilate. de Delva et al.91 described stricturoplasty, which
41.24A). An important complication, in particular in later is a longitudinal incision through the stricture that will
follow-up, is redundancy of the interposed colon (Fig. be closed horizontally to solve the problem.
41.24B). This is seen more after retrosternal interposition, Reflux related colitis may occur. The best treatment
when using right-sided access for the esophagectomy is prevention using an appropriate surgical technique to
and, to a lesser extent, after left-sided or transhiatal minimize the risk of reflux as described earlier. Reflux-
esophagectomy. The redundancy will increase in size over related colitis usually responds well to proton pump
time and eventually will result in a mechanical kinking. inhibitors. Occasionally, a reflux ulcer may cause a colo-
The symptoms are dysphagia and aspiration due to stasis gastric stenosis requiring surgical revision.
458 SECTION I Esophagus and Hernia
More problematic may be regurgitation and aspiration this is its rather segmental vascular configuration, making
due to reflux and/or stasis in the conduit. If it does not it more difficult to prepare a long segment compared
respond to conservative therapy, revisional surgery may to the stomach and colon, putting it at a higher risk
be required to remediate it. Within this context, some for ischemia and necrosis.96 Jejunum can be used as an
authors, especially when dealing with benign disorders interposition, a Roux-en-Y loop, or as a free vascular graft.
in children or young adults, advocate to add a partial It is most commonly used in esophageal cancer surgery
type of antireflux procedure,92 claiming a significant to reconstruct continuity in patients who had previous
improvement in reflux control without increasing stasis gastrectomy or in patients in whom a total gastrectomy
and dysphagia in late follow-up. is needed for oncologic reasons, such as GEJ tumors
The occurrence of intrinsic pathology in the interposed extending into the stomach.
segment, such as IBD or colon cancer in late follow-up In general, a relatively short segment Roux-en-Y will
has been described on an anecdotal basis, and is treated suffice to allow an intrathoracic anastomosis at the level
according to the related standards of care. of the inferior pulmonary vein or somewhat higher up
at the level of the aortic arch. The longer the segment
needed, the more jejunal arteries that must be divided
JEJUNOPLASTY (JEJUNAL INTERPOSITION) in order to obtain an adequate length.
César Roux is credited for having performed the first The blood supply to the jejunum comes from the superior
jejunoplasty in 1907 in a 12-year-old child suffering from a mesenteric artery (Fig. 41.25A). The branches are displayed
t
severe caustic injury. It was a presternal esophago-jejuno- in a segmental pattern. However, anatomic variations fre-
ne
gastrostomy. That patient died at the age of 53.16 quently occur causing technical difficulties when preparing a
Longmire93 was the first to describe a long-segment jejunal loop. In an unfavorable setting, there is a ladder-type
jejunal interposition with microvascular augmentation. distribution with the formation of secondary arches (Fig.
k.
The complexity of the operation and a lack of appropriate 41.25B). In this setting, there is a tendency of the jejunum
microsurgical instrumentarium precluded widespread to retain its curved redundant appearance.
oo
use in spite of these early reports, demonstrating the
technical feasibility of the augmented blood supply to the Technique of Roux-en-Y Jejunoplasty
long-segment pedicled jejunal interposition. The utility The ligament of Treitz is identified and the vascularization
yb
of a small bowel conduit for esophageal reconstruction is inspected. In a slim patient this will be quite easy, but
was confirmed by Allison et al.94 who, in 1957, reported in more obese patients with a fatty mesenterium a transil-
that most patients had normal nutritive intake and work lumination is necessary. The first jejunal artery is identified
capacity at the 3-year follow-up. In 1945, Thompson95 and preserved to guarantee adequate vascularization of
er
performed a presternal jejunoplasty as the first step to treat the first 15 cm departing from the ligament of Treitz.
a mid-one-third squamous cell cancer, with the resection Then the length that will be required has to be assessed
rg
being performed as the second step. after which the proximal two to three segmental arteries
Jejunal interposition is the third most commonly used of this segment are visualized, with transillumination if
modality for esophageal reconstruction. The reason for needed. The arteries are carefully dissected out from the
su
://
tp
ht
A B
Secondary Primary
arcade arcade
FIGURE 41.25 Segmental vascularization of the jejunum arising from the superior artery. (A) Favorable; (B) Unfavorable “ladder type”
pattern. PA, Primary arcade; SA, secondary arcade. (Based on Siewert R, Hölscher A. 3 Eingriffe beim Ösophaguskarzinom.
Jejunuminterposition. In: Siewert R, ed. Breitner Chirurgische Operationslehre. Band IV: Ösophagus, Magen und Duodenum. Urban &
Schwarzenberg; 1989:47. Fig 3-60.)
Options for Esophageal Replacement CHAPTER 41 459
t
ne
A B
k.
FIGURE 41.26 Preparation of a jejunal loop. (A) Transillumination. (B) Clamping the segmental arteries to be divided. ([A] from Maier A,
Pinter H, Tomaselli F, et al. Retrosternal pedicled jejunum interposition: an alternative for reconstruction after total esophago-gastrectomy.
Eur J Cardiothorac Surg. 2002;22:661–665; [B] from Siewert R, Hölscher A. 3 Eingriffe beim Ösophaguskarzinom. Jejunuminterposition.
oo
In: Siewert R, ed. Breitner Chirurgische Operationslehre. Band IV: Ösophagus, Magen und Duodenum. Urban & Schwarzenberg;
1989:48. Fig 3-62.)
yb
mesenterium. With atraumatic bulldog vascular clamps, For the latter, a 25 or 28 mm diameter device is used.
the adequacy of the vascularization is assessed (Fig. 41.26). The anvil is introduced in the transected esophagus and
The arteries and veins are now individually ligated and a purse string is placed around its central rod as described
er
divided. With two or three arteries divided, sufficient earlier (Fig. 41.28). After inspecting the completeness of
length of the conduit is obtained to bring the conduit up the anastomosis, the jejunal stump is closed using a linear
rg
to the level of the inferior pulmonary vein or even higher cutting stapler close to the circular anastomosis in order
up to the level of the aortic arch. If a longer segment is to prevent the formation of a pseudodiverticulum in the
needed, an additional artery can be divided, although the long run. The anastomosis can be performed in the same
su
more arteries that are interrupted the higher the risk for way via a VATS approach.
ischemia at the top end of the conduit (Fig. 41.27A and B). Another alternative to be used in an open setting
Some additional maneuvers can be done to increase is to perform a semimechanical anastomosis. Two stay
://
the length. One is to divide and ligate one or two primary sutures are placed between the top end of the jejunum
arcades—the vascular continuity being guaranteed through and the esophagus approximately 4 cm cranial of its cut
tp
the secondary arcades close to the jejunum. This maneu- end. Then a small incision is made in the jejunum on its
ver also can be used to avoid too much redundancy of antimesenteric side. The two halves of a linear cutting
the conduit resulting from the segmental aspect of the device are introduced in the esophagus and jejunum,
ht
vascularization. After preparation of the loop, the jejunum respectively, and fired. This will be the posterior, mechani-
is transected about 15 cm from the ligament of Treitz cal, wall of the anastomosis. The remaining open part of
using a cutting linear stapler, and the jejunal loop is the esophageal circumference and the jejunal incision
brought up into the chest through the mesentery of the (i.e., the anterior wall) is closed with two layers of running
transverse colon. For a short segment this is easy, but if sutures (Fig. 41.29).101 This type of anastomosis of course
the loop has to be longer this can be more difficult due requires at least 4 cm of esophagus available proximal from
to the fact that the jejunum has the tendency to retain its cut end—the latter requiring another 5 cm proximal
its curved structure. This may cause problems in the later from the upper extent of the tumor.
follow-up (e.g., stasis leading to distention, dysphagia, and The jejuno-jejunal end-to-side anastomosis finally
regurgitation with aspiration). Segmental resection(s) of concludes the reconstruction. The proximal 15 cm are
a redundant jejunum preserving the vascular integrity has anastomosed to the vertical limb of the jejunum at a
been described.97 In case of compromised vascularization distance of at least 70 cm from the top end in order to
at the top end of the jejunum, a supercharged arterial maximally prevent biliary reflux.
anastomosis on the internal mammary artery or inferior
thyroid artery may rescue the situation (Fig. 41.27C).98–100 Technique of Jejunal Interposition
The esophago-jejunal anastomosis can be performed An alternative to the Roux-en-Y limb is the interposition
manually as described earlier with two layers of either of a jejunal loop known as the Merendino operation (Fig.
interrupted or running suture, or by using a circular stapler. 41.30).102 This can be an option in the cases where the colon
460 SECTION I Esophagus and Hernia
Supercharge
t
ne
A B C
k.
oo
FIGURE 41.27 Esophageal reconstruction using a pedicled jejunum. (A) Roux-en-Y reconstruction anastomosis between the middle
esophagus and pedicled jejunum. (B) Intrathoracic anastomosis with upper esophagus and elongated pedicled jejunum. (C) Cervical
anastomosis using a supercharged jejunal pedicle. (From Watanabe M, Mine S, Nishida K, Kurogochi T, Okamura A, Imamura Y.
yb
Reconstruction after esophagectomy for esophageal cancer patients with a history of gastrectomy. Gen Thorac Cardiovasc Surg.
2016;64:457–463.)
er
is unavailable, and when preservation of the gastric reservoir after reconstruction. At the time of follow-up, 95% (20/21)
rg
function is preferred. Some authors have advocated it as of the patients were tolerating a regular diet, and 76.2%
the first choice in cases of high grade dysplasia (HGD) (16/21) did not require any supplemental alimentation.
of early T1aN0M0.103 However, as these conditions are Ninety-five percent (20/21) of the patients were free from
su
now increasingly treated by endoscopic resections, this reflux symptoms, and 80.9% (17/21) had no dumping
operation is rarely indicated in cancer surgery. symptoms. Similar results are reported by Stephens et al.107
The principles of preparation of the conduit are exactly
://
the same as described for the Roux-en-Y loop preparation, Free Vascular Grafts
and the different anastomoses: esophago-jejunal, gastro- The introduction and popularization of microsurgical
tp
jejunal, and jejuno-jejunal are identical. technology in the 1970s boosted the interest in using
free vascularized intestinal grafts, which, until then, had
Functional Results been performed only on an anecdotal basis, but were
ht
The results of jejunal Roux-en-Y and interposition are described by Alexis Carrel in dogs in 1906.15 In theory,
generally good. Postoperative mortality is around 3% to stomach based on the right gastroepiploic artery,108 colon
5%, and the leak rate around 5% to 10%.104 Dysphagia, based on the middle colic artery, and jejunum based on
regurgitation, diarrhea, and reflux-related heartburn are the most robust segmental artery, can be used.
rare. However, some patients with Roux-en-Y reconstruc- In practice, jejunum is the preferred option. It is
tion, even if adhering to a 70-cm distance from proximal claimed that jejunum, because of its preserved peristaltic
to distal anastomosis, will suffer from persistent debilitating mechanism, has the better functional outcome. Free
biliary reflux. This may be very difficult to treat and vascularized bowel segments also can be used as onlay
may eventually require revisional surgery replacing the patches to cover localized large defects in the esophagus
anastomosis further down on the jejunum. that cannot be closed primarily.109 Free vascularized colon
Given its technical difficulty, long-segment jejunoplasty or jejunum grafts are mostly used in patients who need a
is infrequently performed, but from the available data in laryngopharyngectomy for oncologic reasons, as a primary
literature the results seem to be similar to those obtained indication or as a rescue option.110,111 The intervention
after short-segment jejunoplasty.105,106 In a study by Ascioti is performed by two teams: a visceral surgery team and
et al.,99 a long segment supercharged pedicled jejunoplasty a plastic surgery team; the latter is specifically for the
was attempted in 26 patients. There was no postopera- microvascular anastomoses.
tive mortality. Functional results were available in 95.4% The intervention starts in the abdomen with the prepara-
(21/22) of the patients who survived at least 6 months tion of the jejunal loop. A suitable segment of jejunum is
Options for Esophageal Replacement CHAPTER 41 461
SKIN
Historically, skin tubes were among the first attempts to
restore continuity,13 but since the introduction of the
technique of microvascular anastomosis, skin tubes are
t
no longer used with the exception of myocutaneous
ne
flaps to cover large defects in the neck or radial forearm
flaps to close defects in the esophageal wall or to bridge
short gaps. Skin or myocutaneous flaps are other options,
k.
particularly in pharyngeal–cervical esophageal reconstruc-
tions. Bulky myocutaneous flaps from the pectoralis major
oo
or deltoideopectoral muscle are used if soft tissue is needed
to cover cervical defects, especially after extensive neck
dissection and/or radiation therapy. Forearm flaps are
yb
thinner and based on the radial artery and vein, and are
used as free vascular grafts. They are most useful as an
onlay patch for lateral pharyngeal defects and salivary
fistulas. Also, they can be rolled into a tube to replace
er
FIGURE 41.28 Stapled esophagojejunal anastomosis. (From available, such as a frozen abdomen.112,113
Siewert R, Hölscher A. 3 Eingriffe beim Ösophaguskarzinom. Like free bowel transfers, these are highly specialized
plastic and reconstructive techniques primarily used by
su
is to retrieve a segment of jejunum that has a sufficiently originated by Theodor von Billroth.12 In 1879, he resected
large diameter to match with the base of the pharynx. an upper esophageal carcinoma together with the larynx
The jejunum is transected proximally and distally with a and the thyroid gland. The wound was “encouraged to
linear cutting stapler. Then the mesentery is divided on close” with the hope of reepithelialization of the neck
both sides of the feeding artery and draining vein. After channel. Bougies were passed to maintain the lumen.
removing the specimen, the blood vessels are flushed with Unfortunately the patient died of mediastinitis when a
a cold heparin-containing solution to prevent thrombosis bougie was passed through the mediastinum.
after which they are occluded with small nontraumatic Very recently, the interventional endoscopy group from
microsurgery bulldogs. The loop is brought up to the neck Marseille reported their experience with the so-called
and the anastomosis between the pharynx and the jejunum “rendezvous” technique for stenosis or partial necrosis of
is performed in two layers. This first anastomosis stabilizes an interposition, where there is a mediastinal segment of
the jejunum protecting against inadvertent movements/ more than 5 cm without reconstruction. In the described
displacements during and after the microsurgery. The technique they use a scope from the pharynx down to
loop is placed in an isoperistaltic fashion. Then the plastic the mediastinum and another one going up from the
surgery team comes in to perform the vascular microanas- stomach, trying to reach each other. When necessary,
tomosis, making the connection with the inferior thyroid a stent is placed, or several dilatations are performed,
artery and the external jugular vein with 10-0 monofilament in order to create a tube of controlled fibrotic tissue to
suture material (Fig. 41.32). If the length of the vascular bridge the defect.114,115
462 SECTION I Esophagus and Hernia
t
ne
A B
k.
FIGURE 41.29 Semimechanical anastomosis. (A) The linear stapler introduced into a jejunal opening and into the severed esophageal
stump creates the posterior wall. (B) The anterior wall is completed by manual suture. (From Peracchia A. Total gastrectomy and
oo
Roux-en-Y reconstruction. In: Patterson G, et al. Pearson’s Thoracic and Esophageal Surgery. 3rd ed. Philadelphia: Churchill Livingstone;
2008:613-619.) yb
Vagus nerve
er
Esophagus
rg
Diaphragm
su
Jejunal segment
Cardia
://
Vascular
pedicle to
jejunal
tp
segment
ht
Jejunostomy
CONDUIT PLACEMENT
Ligament
The route used to position the esophageal substitute is
of Treitz often determined by the extent of visceral resection, the
location of the anastomosis, and the incisions and exposure
FIGURE 41.30 Merendino procedure: interposition of a jejunum required for the resection. The classic interventions for
loop. (From Merendino KA, Dillard DH. The concept of sphincter cancer of the esophagus are the left thoracoabdominal
substitution by an interposed jejunal segment for anatomic and approach with intrathoracic anastomosis as popularized
physiologic abnormalities at the esophagogastric junction. With by Sweet, and with a cervical anastomosis as advocated
special reference to reflux esophagitis, cardiospasm and by Belsey; right thoracotomy and laparotomy with
esophageal varices. Ann Surg. 1955;September:488. Fig 1.) intrathoracic anastomosis by Ivor Lewis; three-hole right
thoracotomy, laparotomy, and cervicotomy by McKeown;
Options for Esophageal Replacement CHAPTER 41 463
Vertebral
column
Clavicle
Sternum
Dome of 3
diaphragm
t
ne
k.
FIGURE 41.32 Free vascular jejunal graft after implantation. (From
oo
Lee HS, et al. Free jejunal graft for esophageal reconstruction
using end-to-side vascular anastomosis and extended pharyngo-
jejunostomy. Ann Thorac Surg. 2012;93:1850–1854.)
yb
TABLE 41.3 Most Commonly Used Available Routes,
Advantages and Disadvantages
er
MEDIASTINAL mediastinum is
scarred (e.g.,
su
previous surgery)
Transhiatal No opening of the Limitation of lymph
chest node and lateral
://
section plane
clearance
Transthoracic: Direct visualization Displacement lung,
tp
higher tendency toward redundancy because of the larger of patients with esophageal cancer. Dis Esophagus. 2015 Oct 15;
retrosternal space. Some authors systematically perform doi:10.1111/dote.12405; [Epub ahead of print].
26. Liebermann-Meffert DM, Meier R, Siewert JR. Vascular anatomy
a resection of the overlying half of the manubrium and of the gastric tube used for esophageal reconstruction. Ann Thorac
the attached head of the clavicle. Surg. 1992;54:1110-1115.
The lateral transpleural route, anterior to or behind 27. Cooper GJ, Sherry KM, Thorpe JA. Changes in gastric tissue
the lung hilum, and the antethoracic route, are very oxygenation during mobilisation for oesophageal replacement.
Eur J Cardiothorac Surg. 1995;9(3):158-160.
rarely used. 28. Jacobi CA, Zieren HU, Müller JM, et al. Anastomotic tissue oxygen
Finally, a subcutaneous route is available in the extremely tension during esophagectomy in patients with esophageal carci-
rare case that any of the previously mentioned routes are noma. Eur Surg Res. 1996;28(1):26-31.
not available. 29. Bludau M, Hölscher AH, Vallböhmer D, et al. Ischemic conditioning
of the gastric conduit prior to esophagectomy improves mucosal
oxygen saturation. Ann Thorac Surg. 2010;90(4):1121-1126.
REFERENCES 30. Farran L, Miro M, Alba E, et al. Preoperative gastric conditioning
in cervical gastroplasty. Dis Esophagus. 2011;24(4):205-210.
1. Czerny V. Neue operationen: resection des oesophagus. Zentr Chir. 31. Schröder W, Hölscher AH, Bludau M, et al. Ivor-Lewis esophagec-
1877;4:433-434. tomy with and without laparoscopic conditioning of the gastric
2. Torek F. The first successful case of resection of the thoracic portion conduit. World J Surg. 2010;34(4):738-743.
of the esophagus for carcinoma. Surg Gynecol Obstet. 1913;16:614- 32. Veeramootoo D, Shore AC, Wajed SA. Randomized controlled
617. trial of laparoscopic gastric ischemic conditioning prior to
3. Denk W. Zur Radikaloperation des Oesophaguskarzinoms (vorläufige minimally invasive esophagectomy, the LOGIC trial. Surg Endosc.
t
Mittelung). Zentralbl Chir. 1913;40:1065-1068. 2012;26(7):1822-1829.
ne
4. Ohsawa T. Esophageal surgery (in Japanese). Nippon Geka Gakkai 33. Akiyama H, Tsurumaru M, Udagawa H, et al. Radical lymph
Zasshi. 1933;34:1318-1590. node dissection for cancer of the thoracic esophagus. Ann Surg.
5. Turner GG. Excision of the thoracic oesophagus for carcinoma with 1994;220(3):364-372.
reconstruction of an extrathoracic gullet. Lancet. 1933;ii:315-316. 34. Silberhumer G, Györi G, Burghuber C, et al. The value of protect-
k.
6. Adams W, Phemister DB. Carcinoma of the lower thoracic esophagus. ing the longitudinal staple line with invaginating sutures during
Report of a successful resection and esophago-gastrostomy. J Thorac esophageal reconstruction by gastric tube pull-up. Dig Surg.
Surg. 1938;7:621-627. 2009;26(4):337-341.
oo
7. Lewis I. The surgical treatment of carcinoma of the oesophagus 35. Collard JM, Romagnoli R, Goncette L, et al. Terminalized semi-
with special reference to a new operation for growths of the middle mechanical side-to-side suture technique for cervical esophagogas-
third. Br J Surg. 1946;34(133):18-31. trostomy. Ann Thorac Surg. 1998;65(3):814-817.
yb
8. McKeown KC. Total three-stage esophagectomy for cancer of the 36. Deng XF, Liu QX, Zhou D, et al. Hand-sewn vs linearly stapled
esophagus. Br J Surg. 1976;63(4):259-262. esophagogastric anastomosis for esophageal cancer: a meta-analysis.
9. Orringer MB. Esophagectomy without thoracotomy. J Thorac World J Gastroenterol. 2015;21(15):4757-4764.
Cardiovasc Surg. 1978;76(5):643-654. 37. Nafteux P. The continuous quest for quality improvement after
er
10. Cushieri A. Endoscopic oesophagectomy through a right thoraco- esophagectomy for cancer. thesis. Chapter 8: Impact of change in
scopic approach. J R Coll Surg Edinb. 1992;37(1):7-11. anastomotic type on complications and health-related quality of life
11. Luketich JD, Alvelo-Rivera M, Buenaventura PO, et al. Minimally after esophagectomy, gastric tubulation and cervical anastomosis for
rg
invasive esophagectomy: outcomes in 222 patients. Ann Surg. cancer: semimechanical and hand-sewn. 201;157-181Ed. Peeters,
2003;238(4):486-494. Leuven.
12. Billroth CAT. Totalextirpation des Ganzenoesophagus vom Pharynx 38. Barbour AP, Rizk NP, Gonen M, et al. Adenocarcinoma of the
su
bis zum sternum: ein Totalextirpation des Ganzenlarynx mit des gastroesophageal junction: influence of esophageal resection margin
ganzen Schilddruse. Verhandl Deut Ges Chir. 1879;8:7-9. and operative approach on outcome. Ann Surg. 2007;246(1):1-8.
13. von Mikulicz J. Ein fall von Resektion des Carcinomatosen 39. Jianu A. Gastrostomie und Oesophagoplastik. Dtsch Z Chir.
Oesophagus mit plastischen Ersatz des exeidirten Stuckes. Prag 1912;118:383.
://
Med Wochenschr. 1886;11:93-94. 40. Gavriliu D. Replacement of the esophagus by a reverse gastric tube.
14. Beck C, Carrel A. Demonstration of specimens illustrating a method Curr Probl Surg. 1975;12(10):36-64.
of formation of a prethoracic esophagus. Ill Med J. 1905;7:463-464. 41. Ximenes Netto M, Silva RO, Vieira LF, et al. The reversed gastric
tp
15. Carrel A. The surgery of blood vessels. Bull Johns Hopkins. tube revisited: a useful replacement for benign disease. South Am
1907;190:18-27. J Thorac Surg. 1998;5:22.
16. Roux CK. L’oesophago-jejuno gastromose: nouvelle operation 42. Gutschow C, Collard JM, Romagnoli R, et al. Denervated stomach
ht
pour retreeissement infranchissable de l’oesophage. Sem Med. as an esophageal substitute recovers intraluminal acidity with time.
1907;27:37-40. Ann Surg. 2001;233:509-514.
17. Vuillet H. De l’oesophagoplastie et des diverses modifications. 43. Shibuya S, Fukudo S, Shineha R, et al. High incidence of reflux
Semin Med. 1911;31:529. esophagitis observed by routine endoscopic examination after
18. Kelling G. Oesophagoplastik mit hilfe des quercolon. Zentralbl Chir. gastric pull-up esophagectomy. World J Surg. 2003;27:580-583.
1911;38:1209-1212. 44. De Leyn P, Coosemans W, Lerut T. Early and late functional results in
19. Rowbotham S. Laryngeal intubation in anaesthetics. Br Med J. patients with intrathoracic gastric replacement after oesophagectomy
1923;1(3261):1090-1091. for carcinoma. Eur J Cardiothorac Surg. 1992;6:79-85.
20. Magill W. Endotracheal anesthesia. Proc R Soc Med. 1928;22(2):83-88. 45. Manjari R, Padhy AK, Chattopadhyay TK. Emptying of the intratho-
21. Sweet RH. Carcinoma of the esophagus and cardiac end of the racic stomach using three different pylorus drainage procedures:
stomach: immediate and late results of treatment by resection of results of a comparative study. Surg Today. 1996;26(8):581-585.
primary esophagogastric anastomosis. JAMA. 1947;135:485. 46. Huang G, Zhang DC, Zhang DW. A comparative study of resection
22. Belsey RH. Surgical exposure of the esophagus. In: Skinner DBJ, of carcinoma of the esophagus with and without pyloroplasty. In:
Belsey RH, eds. Management of Esophageal Disorders. Philadelphia: Demeester T, Skinner D, eds. Esophageal Disorders: Pathophysiology
WB Saunders; 1988:192-201. and Therapy. New York: Raven Press; 1985:383-388.
23. Jamieson GG, Mathew G, Ludemann R, et al. Postoperative mortality 47. Chattopadhyay T, Gupta S, Padhy A, et al. Is pyloroplasty necessary
following oesophagectomy and problems in reporting its rate. Br following intrathoracic transposition of stomach? Results of a
J Surg. 2004;91(8):943-947. prospective clinical study. Aust N Z J Surg. 1991;61:366-369.
24. Lerut T, Moons J, Coosemans W, et al. Multidisciplinary treatment of 48. Mannell A, McKnight A, Esser J. Role of pyloroplasty in the
advanced cancer of the esophagus and gastroesophageal junction: a retrosternal stomach: result of prospective, randomized controlled
European center’s approach. Surg Oncol Clin N Am. 2008;17:485-502. trial. Br J Surg. 1990;77:57-59.
25. Straatman J, Joosten PJ, Terwee CB, et al. Systematic review of 49. Fok M, Cheng S, Wong J. Pyloroplasty versus no drainage in gastric
patient-reported outcome measures in the surgical treatment replacement of the esophagus. Am J Surg. 1991;162:447-452.
Options for Esophageal Replacement CHAPTER 41 465
50. Urschel JD, Blewett CJ, Young JE, et al. Pyloric drainage (pyloro- cancer: a multicentre, open-label, randomised controlled trial.
plasty) or no drainage in gastric reconstruction after esophagec- Lancet. 2012;379:1887-1892.
tomy: a meta-analysis of randomized controlled trials. Dig Surg. 76. Nafteux P, Durnez J, Moons J, et al. Assessing the relationships
2002;19:160-164. between health-related quality of life and postoperative length of
51. Akkerman RD, Haverkamp L, van Hillegersberg R, et al. Surgical hospital stay after oesophagectomy for cancer of the oesophagus
techniques to prevent delayed gastric emptying after esophagectomy and the gastro-oesophageal junction. Eur J Cardiothorac Surg.
with gastric interposition: a systematic review. Ann Thorac Surg. 2013;44(3):525-533.
2014;98(4):1512-1519. 77. Nafteux P, Moons J, Coosemans W, et al. Minimally invasive
52. Cerfolio RJ, Bryant AS, Canon CL, et al. Is botulinum toxin oesophagectomy: a valuable alternative to open oesophagectomy
injection of the pylorus during Ivor Lewis [corrected] esophago- for the treatment of early oesophageal and gastro-oesophageal
gastrectomy the optimal drainage strategy? J Thorac Cardiovasc Surg. junction carcinoma. Eur J Cardiothorac Surg. 2011;40(6):1455-1463.
2009;137(3):565-572. 78. Orsoni P, Lemaire M. Technique des oesophagoplasties par le
53. Eldaif SM, Lee R, Adams KN, et al. Intrapyloric botulinum injection colon transverse et descendant. J Chir (Paris). 1951;67:491-505.
increases postoperative esophagectomy complications. Ann Thorac 79. Reboud E, Picaud R, Rouzaud R, et al. Left colon transplants in
Surg. 2014;97(6):1959-1964. esophageal surgery. Mem Acad Chir (Paris). 1968;94(10):324-333.
54. Finley R, Lamy A, Clifton J, et al. Gastrointestinal function following 80. Waterston D. Colonic replacement of esophagus. Surg Clin North
esophagectomy for malignancy. Am J Surg. 1995;169:471-475. Am. 1964;44:1441-1447.
55. Bemelman WA, Taat CW, Slors JFM, et al. Delayed postoperative 81. Belsey R. Reconstruction of the esophagus with left colon. J Thorac
emptying after esophageal resection is dependent on the size of Cardiovasc Surg. 1965;49:33-53.
the gastric substitute. J Am Coll Surg. 1995;180:461-464. 82. Lortat-Jacob JL, Giuli R. Esophageal replacement. Prog Surg.
56. Barbera L, Kemen M, Wegener M, et al. Effect of site and width 1973;12:77-95.
of stomach tube after esophageal resection on gastric emptying. 83. Leal AJ, Tannuri AC, Tannuri U. Mechanical bowel preparation for
t
Zentralbl Chir. 1994;119:240-244. esophagocoloplasty in children: is it really necessary? Dis Esophagus.
ne
57. Collard JM, Tinton N, Malaise J, et al. Esophageal replacement: 2013;26(5):475-478.
gastric tube or whole stomach? Ann Thorac Surg. 1995;60:261-266. 84. Thomas PA, Gilardoni A, Trousse D, et al. Colon interposition
58. Swanson EW, Swanson SJ, Swanson RS. Endoscopic pyloric balloon for oesophageal replacement. Multimed Man Cardiothorac Surg.
dilatation obviates the need for pyloroplasty at esophagectomy. 2009;2009(603).
k.
Surg Endosc. 2012;26(7):2023-2028. 85. Hamai Y, Hihara J, Emi M, et al. Esophageal reconstruction using
59. Burt M, Scott A, Williard W, et al. Erythromycin stimulates gastric the terminal ileum and right colon in esophageal cancer surgery.
emptying after esophagectomy with gastric replacement: a random-
oo
Surg Today. 2012;42:342-350.
ized clinical trial. J Thor Cardiovasc Surg. 1996;11:649-654. 86. Saeki H, Morita M, Harada N, et al. Esophageal replacement
60. Lanuti M, DeDelva P, Morse CR, et al. Management of delayed by colon interposition with microvascular surgery for patients
gastric emptying after esophagectomy with endoscopic balloon with thoracic esophageal cancer: the utility of superdrainage. Dis
yb
dilatation of the pylorus. Ann Thorac Surg. 2011;91(4):1019-1024. Esophagus. 2013;26:50-56.
61. Datta J, Williams NN, Conway RG, et al. Rescue pyloroplasty for 87. Shirakawa Y, Naomoto Y, Noma K, et al. Colonic interposition and
refractory delayed gastric emptying following esophagectomy. supercharge for esophageal reconstruction. Langenbecks Arch Surg.
Surgery. 2014;156(2):290-297. 2006;391:19-23.
er
62. da Rocha JR, Cecconello I, Zilberstein B, et al. Barrett esophagus 88. Pompeo E, Nofroni I, Van Raemdonck D, et al. Esophagocoloplasty
in the esophageal stump after subtotalesophagectomy with cervical for congenital, benign and malignant diseases. Surgical and long-
esophagogastroplasty. Rev Hosp Clin Fac Med Sao Paulo. 1992;47:69-70. term functional results. Eur J Cardiothorac Surg. 1996;10(7):561-567.
rg
63. Franchimont D, Covas A, Brasseur C, et al. Newly developed Barrett’s 89. Gust L, Ouattara M, Coosemans W, et al. European perspective
esophagus after subtotalesophagectomy. Endoscopy. 2003;35:850-853. in thoracic surgery-eso-coloplasty: when and how? J Thorac Dis.
64. Oberg S, Johansson J, Wenner J, et al. Metaplastic columnar 2016;8(suppl 4):S387-S398.
su
mucosa in the cervical esophagus after esophagectomy. Ann Surg. 90. Jeyasingham K, Lerut T, Belsey RH. Functional and mechanical
2002;235:338-345. sequelae of colon interposition for benign oesophageal disease.
65. Wang L, Huang MH, Huang BS, et al. Gastric substitution for Eur J Cardiothorac Surg. 1999;15(3):327-331. [discussion 331–332].
resectable carcinoma of the esophagus; an analysis of 368 cases. 91. de Delva PE, Morse CR, Austen WG Jr, et al. Surgical management of
://
Ann Thorac Surg. 1992;53:289-294. failed colon interposition. Eur J Cardiothorac Surg. 2008;34(2):432-437.
66. Gutschow C, Collard J, Romagnoli R, et al. Bile exposure of the 92. Vasseur Maurer S, Estremadoyro V, Reinberg O. Evaluation of
denervated stomach as an esophageal substitute. Ann Thorac Surg. an antireflux procedure for colonic interposition in pediatric
tp
life. Ann Thorac Surg. 1997;63:1568-1572. 94. Allison PR, Wooler GH, Gunning AJ. Esophagojejunogastrostomy.
68. Hölscher A, Voit H, Buttermann G, et al. Function of the J Thorac Surg. 1957;33:738-748.
intrathoracic stomach as esophageal replacement. World J Surg. 95. Thompson VC. Oesophagectomy for carcinoma: restoration of
1988;12:835-844. function by an ante-thoracic jejunal graft. Proc R Soc Med. 1946;39:
69. Banki F, Mason RJ, DeMeester SR, et al. Vagal-sparing esophagec- 310.
tomy: a more physiologic alternative. Ann Surg. 2002;236:324-335. 96. Gaur P, Blackmon SH. Jejunal graft conduits after esophagectomy.
70. Humphrey C, Johnston D, Walker B, et al. Incidence of dumping J Thorac Dis. 2014;6(suppl 3):S333-S340.
after truncal and selective vagotomy with pyloroplasty and 97. Wright C, Cuschieri A. Jejunal interposition for benign esophageal
highly selective vagotomy without drainage procedure. Br Med J. disease. Technical considerations and long-term results. Ann Surg.
1972;3:785-788. 1987;205(1):54-60.
71. Sinha S, Padhy A, Chattopadhyay T. Dumping syndrome in the 98. Watanabe W, Mine S, Nishida K, et al. Reconstruction after
intra-thoracic stomach. Trop Gastroenterol. 1997;8:131-133. esophagectomy for esophageal cancer patients with a history of
72. Hasler W. Dumping syndrome. Curr Treat Options Gastroenterol. gastrectomy. Gen Thorac Cardiovasc Surg. 2016;64(8):457-463.
2002;5:139-145. 99. Ascioti AJ, Hofstetter WL, Miller MJ, et al. Long-segment, super-
73. Hulscher JB, van Sandick JW, de Boer AG, et al. Extended trans- charged, pedicled jejunal flap for total esophageal reconstruction.
thoracic resection comhapared with limited transhiatal resection J Thorac Cardiovasc Surg. 2005;130:1391-1398.
for adenocarcinoma of the esophagus. N Engl J Med. 2002;347: 100. Blackmon SH, Correa AM, Skoracki R, et al. Supercharged pedicled
1662-1669. jejunal interposition for esophageal replacement: a 10-year experi-
74. Lagergren P, Avery KN, Hughes R, et al. Health-related quality of ence. Ann Thorac Surg. 2012;94:1104-1111. [discussion 1111-1113].
life among patients cured by surgery for esophageal cancer. Cancer. 101. Peracchia A, Rosati R. Total gastrectomy and Roux-en-Y reconstruc-
2007;110:686-693. tion. In: Patterson AG, Cooper JD, Deslauriers J, Lerut A, Luketich
75. Biere SS, van Berge-Henegouwen MI, Maas KW, et al. Minimally JD, Rice T, eds. Pearson’s Thoracic and Esophageal Surgery. Philadelphia:
invasive versus open oesophagectomy for patients with oesophageal Churchill Livingstone; 2008:613-619 [Chapter 57].
466 SECTION I Esophagus and Hernia
102. Merendino KA, Dillard DH. The concept of sphincter substitution 109. Jurkiewicz MJ, Paletta CEL. Free jejunal graft. In: Current Therapy
by an interposed jejunal segment for anatomic and physiologic in Cardiothoracic Surgery. Philadelphia: B.C. Decker; 1989:206.
abnormalities at the esophagogastric junction; with special reference 110. Jurkiewicz MJ. Vascularized intestinal graft for reconstruction of the
to reflux esophagitis, cardiospasm and esophageal varices. Ann cervical esophagus and pharynx. Plast Reconstr Surg. 1965;36:509-517.
Surg. 1955;142(3):486-506. 111. Hester TR, McConnel FM, Nahal F, et al. Reconstruction of cervical
103. Stein H, Feith M, Siewert JR. Distal esophageal resection and esophagus, hypopharynx and oral cavity using free jejunal transfer.
jejunum interposition for early Barrett carcinoma. Zentralbl Chir. Am J Surg. 1980;140(4):487-491.
2001;126(suppl 1):9-13. [German]. 112. Chen H, Tang Y. Microsurgical reconstruction of the esophagus.
104. Blackmon SH, Correa AM, Skoracki R, et al. Supercharged pedicled Semin Surg Oncol. 2000;19:235.
jejunal interposition for esophageal replacement: a 10-year experi- 113. Nakatsuka T, Harii K, Asato H, et al. Comparative evaluation in
ence. Ann Thorac Surg. 2012;94(4):1104-1111. pharyngo-oesophageal reconstruction: radial forearm flap compared
105. Maier H, Pinter H, Tomaselli F, et al. Retrosternal pedicled jejunum with jejunal flap: a 10 year experience. Scand J Plast Reconstr Surg
interposition: an alternative for reconstruction after total esophago- Hand Surg. 1997;32:307.
gastrectomy. Eur J Cardiothorac Surg. 2002;22(5):661-665. 114. Gonzalez JM, Vanbiervliet G, Gasmi M, et al. Efficacy of the endo-
106. Ring WS, Varco RL, L’Heureux PR, et al. Esophageal replacement scopic rendez-vous technique for the reconstruction of complete
with jejunum in children: an 18 to 33 year follow-up. J Thorac esophageal disruptions. Endoscopy. 2016;48:179-183.
Cardiovasc Surg. 1982;83:918. 115. Perbtani Y, Suarez AL, Wagh MS. Emerging techniques and
107. Stephens EH, Gaur P, Hotze KO, et al. Super-charged pedicled efficacy of endoscopic esophageal reconstruction and lumen
jejunal interposition performance compares favorably with a gastric restoration for complete esophageal obstruction. Endosc Int Open.
conduit after esophagectomy. Ann Thorac Surg. 2015;100(2):407-413. 2016;4(2):E136-E142.
108. Yamagishi M, Ikeda N, Yonemoto T. An isoperistaltic gastric tube. New
method of esophageal replacement. Arch Surg. 1970;100(6):689-692.
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
CHAPTER
Palliative Therapy for Esophageal Cancer
Dennis Wells
| Virginia R. Litle
42
T
he American Cancer Society estimates that 16,940 carried out in conjunction with other modalities such
new cases of esophageal cancer will be diagnosed as stenting or ablation to provide longer relief for the
in 2016, and in the same year there will be 15,690 patient. The Savary bougie dilators produce both radial and
deaths from esophageal cancer.1 Unfortunately, more longitudinal forces to the esophagus (Fig. 42.1), whereas
than half of patients present with advanced disease with a balloon dilator produces only a radial force (Fig. 42.2).
symptoms of dysphagia, weight loss, and less often bleed- The balloon dilator is better for a longer stricture, such
ing.2 Palliative interventions are needed to optimize quality as a malignant stricture, and the bougie is better for the
of life by reducing hospital admissions for aspiration and more focal anastomotic or postfundoplication stricture.
bleeding, as well as allow them to enjoy eating. The goals The balloon dilation can be done under direct vision
t
for treating malignant dysphagia are ease of treatment and with a through-the-scope dilator or over a guidewire and
ne
short hospital stay. For patients who may be candidates with fluoroscopic guidance to confirm safe placement
for palliative chemotherapy or chemoradiation therapy, of the balloon across the tumor. When the esophageal
the only palliative procedural intervention may be a lumen is very narrow, expandable metal stent placement
k.
need for a percutaneous gastrostomy tube for caloric without initial dilation may lead to stent infolding and an
supplementation. initial dilation may be indicated. Similarly, after expand-
oo
Depending on the degree of dysphagia and the per- able metal stent placement, careful balloon dilation may
formance status of the patient, a significant number of facilitate immediate expansion. However, overzealous
patients may need endoscopic interventions to improve dilation may lead to a less snug fit of the stent and early
yb
oral intake and nutritional status as the first step in their migration. Dilation is often used before and after ablative
treatment algorithm. Patients who cannot swallow their therapy including cryoablation or laser therapy to allow
saliva need an urgent endoscopic intervention to reduce passage of the endoscope and laser fiber. Complications
their risk of aspiration, a potentially fatal event. after esophageal dilation range from pain and fever to
er
Most endoscopic therapies provide initial improvement perforation. Although risk of perforation is uncommon
in dysphagia but vary in their durability. After improve- in experienced hands, the endoscopist should have a
rg
ment in oral nutrition, performance may improve and low threshold for further diagnostic studies, such as an
other therapies may be tolerated. In this chapter, we intraoperative esophagram with gastroview or postoperative
review all the commonly used methods for endoscopic barium esophagography, to rule this out, in particular in
su
palliation of malignant dysphagia. In general, these the patient with excessive pain or pneumomediastinum,
methods deliver rapid relief and allow intake of a soft to pneumoperitoneum, or a pneumothorax after dilation.
regular diet with some modifications depending on the If there is no leak, the patient can be started on clear
://
individual. Dilation and endoscopic stents are considered liquids for 24 hours, followed by a soft diet as tolerated. In
the mainstay of palliative options. As the durability of the some cases, if the perforation is minor and contrast agent
tp
relief of dysphagia is variable, a multimodality approach is shows no or minimal extravasation, with good drainage
often required. Laser therapy with thermal ablation and into the esophageal lumen, patients can be conservatively
nonthermal photodynamic therapy have been options managed and treated with antibiotics and nothing by
ht
for many years; however, cryoablation has now joined the mouth for a short period of observation. In some cases,
ranks. The palliative options for bleeding or obstructing immediate deployment of a covered, expandable metal
esophageal cancer are summarized in Table 42.1. On the stent will seal more substantial leaks. It is imperative that
rare occasion, in a very good performance with incurable an experienced thoracic surgeon evaluate these patients
disease, esophagectomy may be indicated for palliation because, even in the era of covered expandable stents,
when other interventions fail. In our practice, advances in surgical intervention may be indicated for perforations.
endoscopic palliative modalities have made this unusual.
STENTS
ESOPHAGEAL DILATION With the advent of the self-expanding metal stent (SEMS),
Malignant esophageal dysphagia can be relieved imme- now more than 20 years ago, endoscopic palliation is now
diately to some degree with dilation with a bougie or easier and associated with fewer complications than with
a balloon dilator. If this is the only therapy, recurrent the historical open traction and push-through techniques.3
dysphagia will occur in almost all patients within 1 to 2 A SEMS can be placed via upper endoscopy with fluoro-
weeks. This temporary intervention may allow improve- scopic guidance and does not require general anesthesia.
ment of oral intake, particularly fluids, for a short period The primary risks are aspiration at time of placement, so
of time while other therapies, including chemotherapy general anesthesia may reduce the risk of periprocedural
and radiation, are initiated. Dilation is more commonly aspiration. Communication with the anesthesiologist is
467
Palliative Therapy for Esophageal Cancer CHAPTER 42 467.e1
ABSTRACT
The American Cancer Society estimates that 16,940 new
cases of esophageal cancer will be diagnosed in 2016, and
more than half of the patients will present with malignant
dysphagia. Patients who cannot swallow their saliva need
an urgent endoscopic intervention to reduce their risk of
aspiration, a potentially fatal event. Palliative interventions
include esophageal dilation with balloon or bougie dilator,
stenting with self-expanding metal stents, and ablative
therapies such as photodynamic therapy and cryoabla-
tion. Chemotherapy, brachytherapy, and external beam
radiation also provide palliation and may be offered to
those not needing an urgent intervention. This chapter
summarizes the current benefits and indications for the
different modalities and introduces the newer topic of
cryoablation.
KEYWORDS
t
ne
Malignant dysphagia, endoscopic stenting, photodynamic
therapy, cryoablation
k.
oo
yb
er
rg
su
://
tp
ht
468 SECTION I Esophagus and Hernia
t
Weight
ne
Modality Bleeding Obstruction Loss
Balloon dilation N Y N
Self-expanding metal stent N Y N
k.
LASER
Thermal laser (Nd:YAG) Y Y N
oo
Photodynamic therapy Y Y N
Cryoablation Y Y N
Radiation therapy N Y N
Chemotherapy N Y N
yb
FIGURE 42.1 The Savary dilators (up to 60 French in size) are
Percutaneous gastrostomy tube N N Y
placed over a wire to dilate benign and malignant esophageal
Nd:YAG, Neodymium:yttrium-aluminum-garnet. strictures with fluorosopic guidance.
er
rg
su
://
tp
ht
A B
FIGURE 42.2 The balloon dilators (A) can be placed through the endoscope working channel and then inflated under direct vision or with
fluoroscopic guidance with a handheld pump (B). (Images used with permission from Boston Scientific Corp.)
Palliative Therapy for Esophageal Cancer CHAPTER 42 469
Table 42.2 summarizes the more commonly used After adequate seating in the esophagus, the intrinsic
SEMS for malignant obstruction. There are several radial force of the most expandable metal stents leads to
types of expandable metal stents, and all share many continued expansion to its maximal diameter, depending
features but have minor design modifications, which on the severity of the obstruction. Thus, if a stent appears
t
may offer advantages of one over the other. The Evolu- patent but not open completely on initial placement, the
ne
tion Controlled-Release Stent or Z-stent (Wilson Cook, clinician can repeat the endoscopy the next day or obtain a
Winston-Salem, North Carolina) is composed of stainless barium esophagram to see if the stent has expanded fully.
steel in either a mesh or zigzag design. It is released with Poor initial stent deployment, especially with significant
k.
a sheath and pusher rod mechanism, which is meant to infolding, may indicate an improper stent diameter or
optimize accurate placement. The Ultraflex and WallFlex length. Immediate intervention, such as gentle dilation, or
oo
SEMS (Boston Scientific, Watertown, Massachusetts) are even stent removal may be necessary if an obvious failure
knitted stents deployed with the removal of a string. After is observed. Complications from stent placement can occur
deployment, typical stent diameter ranges from 18 to early and late. The early complications include aspiration
23 mm and stent length is 10 to 15 cm.6
yb
pneumonia during the procedure, esophageal perforation,
The high radial force of SEMS may result in postopera- malpositioned stent, airway compression and compromise
tive pain, which is usually mild and transient. In unusual with mid to upper esophageal stents, persistent obstruc-
cases it may persist and require removal. A SEMS is available tion, and pain. Late complications include intractable
er
covered or uncovered. The covered stents reduce tumor reflux, stent obstruction, or migration. The risk of delayed
ingrowth except at the ends, which are uncovered to reduce perforation or erosion into adjacent structures is rare but
rg
stent migration. When tumor ingrowth of the stent occurs has been described.9 Delayed removal or repositioning
or overgrowth at the end of the stent occurs, additional of expandable esophageal stents is generally avoided but
stents can be applied, or neodymium:yttrium-aluminum- in experienced hands has been done. The obvious risk
su
garnet (Nd:YAG) laser or photodynamic therapy (PDT) is esophageal trauma with the potential for esophageal
laser can be used to ablate the tumor. Thermal laser perforation. In summary, expandable metal stents provide
may be limited in these situations because the laser can rapid improvement in malignant dysphagia, allowing
://
damage the stent nitinol itself. Some of the newer designs patients a short or no hospital stay and improved quality
of expandable metal stents (Alveolus) offer full-length of life. The need for repeat intervention is common in
tp
coverage with options for more stable esophageal wall patients who live longer than a few months. In these
contact that may minimize migration yet allow full-length cases a careful investigation of the causes for stent failure
coverage. should be made to determine if additional dilation, stent
ht
Self-expanding plastic stents (SEPSs), including Polyflex placement, or laser therapy may be indicated to provide
(Boston Scientific), have more commonly been used to continued relief of obstruction.
treat benign esophageal lesions because these stents are Recent advancements in stents also include the develop-
easier to retrieve given less tissue reaction in comparison ment of drug-eluting stents,10 radioactive stents,11,12 as
to metal stents. Other benefits of plastic stents include well as biodegradable stents.13 The role of these types of
lower cost, although they are prone to stent migration. stents for palliation of esophageal malignancy remains
Malignant esophageal disease and palliation has typically to be determined.
been addressed with SEMS. Conio et al. randomized
patients to receive either the Polyflex or the Ultraflex stent
and reported no difference in palliation of dysphagia but a NEODYMIUM:YTTRIUM-ALUMINUM-GARNET
significantly higher rate of complication, particularly stent LASER TREATMENT
migration,7 in the SEPS group. In addition, a meta-analysis
conducted by Yakoub et al. in London also reported metal Thermal laser therapy for esophageal cancer was initially
stents performed superiorly to plastic stent for palliative described in 1982 and involved use of argon or Nd:YAG
management of malignant dysphagia.8 lasers. The Nd:YAG laser has since been proven to be
In our practice, we use a combination of endoscopic and more effective than the argon laser but suffers from some
fluoroscopic guidance to measure the length of esophageal technical limitations and a perforation rate as high as 7%
obstruction before deploying the expandable metal stents. to 10%. PDT was approved by the US Food and Drug
470 SECTION I Esophagus and Hernia
Administration in 1996 for the treatment of malignant tumor necrosis after PDT is limited to 5 mm. This provides
dysphagia and has a low perforation rate (1%) and is more a safety factor in minimizing risk of esophageal perforation
durable than Nd:YAG laser ablation. Thermal laser remains but also can limit its effectiveness for large bulky tumors,
a useful tool for ablating the unusual bleeding esophageal especially when significant extrinsic compression is present.
tumor acutely. The Nd:YAG laser uses a wavelength of Full-thickness perforation can occur, but in the largest
1064 nm. For both bleeding and obstructing esophageal series to date, esophageal perforation occurred in only 5
lesions, 50 to 90 W is used with a pulse duration of 0.3 of 215 patients (1.5%).19 Balloon dilation was performed
to 1.0 second. The greatest risk with thermal laser use is before PDT to allow passage of the endoscope through the
esophageal perforation, which can occur directly from obstructing tumor. It was unclear whether the mechanical
the laser or with concurrent dilation. The laser treat- dilation of the esophagus, the use of PDT, or a combination
ment often begins with esophageal dilation either with a of the two factors contributed to this complication. In
pneumatic balloon or Savary dilator with a guidewire and the same series of 215 patients, esophageal stricture after
fluoroscopic guidance. The general concept is to work PDT occurred in only 1.5% of patients; however, in more
off the luminal surface of the tumor to minimize the contemporary series of PDT for palliation and treatment,
risk of perforation. The tumor is then circumferentially the stricture rate approaches 20%.20 This lower rate of
ablated with the endoscopically placed Nd:YAG laser stricture after PDT for palliation may be explained by the
in a point-by-point method. Depending on the length fact that less normal esophagus is exposed to the laser
of the endoluminal tumor, Nd:YAG can be more time light when a bulky tumor is present. The combination
t
consuming than PDT as the nonthermal laser diffuser of radiation therapy, chemotherapy, and PDT typically
ne
probe allows administration of light over a greater surface increases the risk of stricture formation. Whether a true
area. Tumors with a significant component of extraluminal PDT-induced stricture or tumor progression results in
mass may fail laser therapy due to extrinsic compression luminal narrowing is sometimes difficult to ascertain in
k.
and may be better palliated with an expandable metal the palliative setting. Ideal candidates for palliative PDT
stent. One study comparing thermal laser to esophageal have locally advanced esophageal carcinoma with primarily
oo
stenting found similar relief in dysphagia and survival, endoluminal disease and minimal stricture or extrinsic
but the dysphagia relief lasted significantly longer in the compression. The overall advantages of PDT for treating
laser-treated group in whom the patients had significant locally advanced esophageal cancer include improve-
gastric involvement.14 The main complication of Nd:YAG
yb
ment in malignant dysphagia within days of treatment,
laser therapy is esophageal perforation, which occurred minimal pain, and, in some cases of gastroesophageal
in 7% of 118 patients in a randomized multicenter trial junction tumors, less reflux; and in cases of high cervical
of patients undergoing PDT versus Nd:YAG therapy for esophageal obstruction, there is less concern over tracheal
er
palliation of esophageal cancer. Esophageal dilation had compression resulting from stent expansion of the cervical
accompanied laser therapy in half of the perforation cases. esophagus. The main disadvantages include the skin
rg
Fever, nausea, and postoperative respiratory insufficiency photosensitivity in patients with a limited life expectancy,
are all potential perioperative morbidities of the laser the costs of specialized equipment and the photosensitizing
treatment.15 Fistulas and strictures occur as late complica- agent, and limitations in efficacy when significant, bulky,
su
tions in 10% of patients.16 Thermal laser therapy may be extrinsic compression is present. The more common side
valuable in cases where a stent cannot be placed easily, effects are chest pain, nausea, odynophagia, and vomiting.
such as in the cervical esophagus where proximal airway Mediastinitis and tracheoesophageal fistulas are rare.21
://
antireflux valve. However, in an early randomized study Medical, Baltimore, Maryland) system has been available
of 30 patients, the valve did not prevent gastroesophageal for several years, and despite the paucity of literature for
reflux.17 More recently 38 patients were enrolled at nine palliative options, it has been reported for ablative purposes
centers to covered SEMS with antireflux valve or SEMS for more than 10,000 treatments.22 TruFreeze involves
without a valve.18 The conclusion was gastroesophageal spraying liquid nitrogen directly on the bulky tumor. A
reflux disease (GERD) was improved but there were more gastric decompression tube is required to reduce the risk
obstructions of the SEMS in the valve group. of injury to the enteric viscera from the rapidly expanding
nitrogen gas. If a decompression tube cannot be placed
across a bulky endoluminal obstruction, then cryospray is
PHOTODYNAMIC THERAPY not an initial palliative option. As the cryogen is released
Photodynamic therapy is a nonthermal laser process into the esophagus, a transmural freeze occurs. Cardiac
using selective endoscopic delivery of light with a specific arrhythmias may occur from the freeze affecting the heart.
wavelength to activate a photosensitizing agent that results A preoperative cardiac assessment including a recent
in tumor ablation and restores endoluminal patency. electrocardiogram is recommended in the preoperative
Currently PDT is primarily performed with porfimer setting. These risks and side effects should be discussed
sodium (Photofrin; Concordia Laboratories Inc., Oakville, with the patient prior to surgery.
Ontario), a hematoporphyrin derivative with a light activa- For the truFreeze cryotherapy technique, an orogastric
tion wavelength of 630 nm. The depth of penetration and decompression tube is placed, the cryospray catheter is
Palliative Therapy for Esophageal Cancer CHAPTER 42 471
A B
t
ne
FIGURE 42.3 (A and B) TruFreeze liquid nitrogen is delivered through the endoscope to ablate the endoluminal tumor under direct vision.
primed, and the decompression tube is placed on continu- and stricture formation, the latter of which can be treated
k.
ous suction. Expose the upper abdomen and apply gentle successfully with dilation. In a German retrospective study
pressure consistently throughout the procedure. Typically of 139 patients with incurable esophageal cancer treated
oo
the clinician applies the cryospray (−196°C) under direct with EBRT alone, EBRT with BT, or with BT alone, EBRT ±
vision for 10 to 20 seconds and allows complete thawing BT provided the longest dysphagia-free survival, at greater
of tissue in between freezes (~2 to 3 minutes) (Fig. 42.3). than 90% at 6 months.24 Overall symptom relief occurred
yb
Three separate freezes are administered, then the catheter in 72% of patients for up to 5 months. Complications of
is defrosted with the decompression tube on suction the EBRT with BT in a series of 148 patients included
still. Cryotherapy complications are uncommon but can fistula (5% of cases), stricture (27%), and bleeding (4%),25
include perforation, chest pain, and arrhythmias. The other and the combination of EBRT with BT is not considered
er
technology is the carbon dioxide cryospray (Polar Wand; GI safe by all radiation oncologists.26
Supply, Camp Hill, Pennsylvania) and is also administered
rg
catheter to allow a flow of CO2 at 6 to 8 L/min. We await alone requires 4 to 6 weeks for improvement in swallow-
future reports of the palliative success with the Polar Wand. ing.27 The addition of chemotherapy to a 3-week course of
external beam radiation (35 to 40 Gy) results in significant
://
radiation is another modality to locally ablate obstructing dysphagia scoring system. The minor complication rate was
esophageal cancer. It has been used as a salvage therapy low and included radiation pneumonitis and infections.
after external beam radiation therapy (EBRT), as well as There was a 6% treatment-related mortality rate. The
resulting in greater than 50% relief in dysphagia.23 The median time to improvement was 6 weeks after the start
benefits are a good tumor response but with stricture of chemoradiation therapy. Part of the delay to clinical
rates of 38% with high-dose rate BT (3 weekly fractions of improvement is undoubtedly related to the significant
500 cGy).23 Patients typically return for several treatment incidence of esophagitis secondary to the treatment.
fractions over 2 to 3 weeks to obtain the full treatment dose. Thus, although a tumor may be shrinking, esophagitis
In addition, BT requires specialized staff and expensive may contribute to the delay to satisfactory oral intake.
delivery equipment with limited availability. For each In a phase III trial of palliative epirubicin, oxaliplatin,
treatment the patient undergoes esophageal dilation and and capecitabine (EOX) versus docetaxel, cisplatin,
placement of the after loading catheter. Afterward they are 5-fluorouracil, and leucovorin (DCF) for esophageal and
transported to the BT delivery room, where the catheter is gastric adenocarcinomas, DCF provided a longer median
loaded and approximately 500 cGy is administered, which survival by 2.4 months and with fewer side effects than
penetrates to a depth of 0.5 to 1 cm. Contraindications EOX, with primary complications being nausea, pain, and
to the procedure include the presence of an esophageal thromboembolic events up to 13.8%.29
fistula because this is a known potential complication of Chemotherapy alone as palliation is suboptimal because
the procedure. Other complications include perforation of the delay in relief. Clinicians reported in an observational
472 SECTION I Esophagus and Hernia
study in the United Kingdom that only 53% of patients 11. Tian D, Wen H, Fu M. Comparative study of self-expanding metal
completed their course of palliative chemotherapy and only stent and intraluminal radioactive stent for inoperable esophageal
squamous cell carcinoma. World J Surg Oncol. 2016;14(1):18.
9% of those older than 75 years completed it.30 Patients 12. Liu N, Liu S, Xiang C, et al. Radioactive self-expanding stents give
who cannot tolerate chemoradiation therapy should be superior palliation in patients with unresectable cancer of the
offered a palliative stent. Chemoradiation therapy can esophagus but should be used with caution if they have had prior
take longer than a month to see a benefit but may be radiotherapy. Ann Thorac Surg. 2014;98(2):521-526.
13. Krokidis M, Burke C, Spiliopoulos S, et al. The use of biodegradable
a good choice for cervical esophageal lesions. For distal stents in malignant oesophageal strictures for the treatment of
esophageal cancers, SEMS can produce significant reflux dysphagia before neoadjuvant treatment or radical radiotherapy: a
when placed across the gastroesophageal junction, and feasibility study. Cardiovasc Intervent Radiol. 2013;36(4):1047-1054.
ablative therapies may be better endoscopic options. 14. Loizou LA, Grigg D, Atkinson M, Robertson C, Bown SG. A prospec-
tive comparison of laser therapy and intubation in endoscopic
palliation for malignant dysphagia. Gastroenterology. 1991;100(5 Pt
COMMENTS 1):1303-1310.
15. Lightdale CJ, Heier SK, Marcon NE, et al. Photodynamic therapy
This chapter reviews the palliative approaches to the with porfimer sodium versus thermal ablation therapy with Nd:YAG
patient with dysphagia who has inoperable disease. The laser for palliation of esophageal cancer: a multicenter randomized
trial. Gastrointest Endosc. 1995;42(6):507-512.
recent addition to the armamentarium is cryoablation. 16. Heier SK, Rothman KA, Heier LM, Rosenthal WS. Photodynamic
There is no single palliative therapy that works well for all therapy for obstructing esophageal cancer: light dosimetry and
patients, and, indeed, not all patients with mild degrees of randomized comparison with Nd:YAG laser therapy. Gastroenterology.
t
dysphagia need palliation. It is critical that we recognize 1995;109(1):63-72.
ne
that the ability to restore oral nutrition is a valuable aspect 17. Homs MY, Wahab PJ, Kuipers EJ, et al. Esophageal stents with
antireflux valve for tumors of the distal esophagus and gastric cardia:
of allowing the patient to regain an important component a randomized trial. Gastrointest Endosc. 2004;60(5):695-702.
of his or her quality of life, which is far superior and 18. Coron E, David G, Lecleire S, et al. Antireflux versus conven-
k.
more cost effective than feeding tubes or parenteral tional self-expanding metallic stents (SEMS) for distal esophageal
nutrition. In a retrospective cohort study of 736 patients cancer: results of a multicenter randomized trial. Endosc Int Open.
2016;4(6):E730-E736.
oo
in The Netherlands, investigators found an amalgam of 19. Litle VR, Luketich JD, Christie NA, et al. Photodynamic therapy as
approaches to palliate patients with inoperable esophageal palliation for esophageal cancer: experience in 215 patients. Ann
cancer, with best supportive care begin the most common Thorac Surg. 2003;76(5):1687-1692, discussion 1692-1693.
approach.31 Ideally, the clinician dealing with esophageal
yb
20. Brantingham CR, Beekman BE, Te Groen DM. Peripheral venous
cancer patients should be well versed in all endoscopic incompetence and the urban terrain. The relationship of venous
incompetence and flat surfaces, and a system for artificially varying
interventions and the palliative aspects and limitations the terrain. J Am Podiatry Assoc. 1967;57(12):547-554.
of chemotherapy and radiation therapy; however, tools
er
2005;92(3):246-256. 24. Welsch J, Kup PG, Nieder C, et al. Survival and symptom relief after pal-
3. Cusumano A, Ruol A, Segalin A, et al. Push-through intubation: liative radiotherapy for esophageal cancer. J Cancer. 2016;7(2):125-130.
effective palliation in 409 patients with cancer of the esophagus 25. Laskar SG, Lewis S, Agarwal JP, Mishra S, Mehta S, Patil P. Com-
tp
and cardia. Ann Thorac Surg. 1992;53(6):1010-1014. bined brachytherapy and external beam radiation: an effective
4. Knyrim K, Wagner HJ, Bethge N, Keymling M, Vakil N. A controlled approach for palliation in esophageal cancer. J Contemp Brachytherapy.
trial of an expansile metal stent for palliation of esophageal obstruc- 2015;7(6):453-461.
ht
tion due to inoperable cancer. N Engl J Med. 1993;329(18):1302-1307. 26. Tai P, Yu E. Esophageal cancer management controversies: radiation
5. Ell C, May A, Hahn EG. Gianturco-Z stents in the palliative treatment oncology point of view. World J Gastrointest Oncol. 2014;6(8):263-274.
of malignant esophageal obstruction and esophagotracheal fistulas. 27. Siersema PD, Dees J, van Blankenstein M. Palliation of malignant
Endoscopy. 1995;27(7):495-500. dysphagia from oesophageal cancer. Rotterdam Oesophageal Tumor
6. Hindy P, Hong J, Lam-Tsai Y, Gress F. A comprehensive review of Study Group. Scand J Gastroenterol Suppl. 1998;225:75-84.
esophageal stents. Gastroenterol Hepatol (N Y). 2012;8(8):526-534. 28. Harvey JA, Bessell JR, Beller E, et al. Chemoradiation therapy is
7. Conio M, Repici A, Battaglia G, et al. A randomized prospective effective for the palliative treatment of malignant dysphagia. Dis
comparison of self-expandable plastic stents and partially covered Esophagus. 2004;17(3):260-265.
self-expandable metal stents in the palliation of malignant esophageal 29. Ochenduszko S, Puskulluoglu M, Konopka K, et al. Comparison of
dysphagia. Am J Gastroenterol. 2007;102(12):2667-2677. efficacy and safety of first-line palliative chemotherapy with EOX
8. Yakoub D, Fahmy R, Athanasiou T, et al. Evidence-based choice and mDCF regimens in patients with locally advanced inoperable or
of esophageal stent for the palliative management of malignant metastatic HER2-negative gastric or gastroesophageal junction adeno-
dysphagia. World J Surg. 2008;32(9):1996-2009. carcinoma: a randomized phase 3 trial. Med Oncol. 2015;32(10):242.
9. Christie NA, Buenaventura PO, Fernando HC, et al. Results of 30. Groene O, Crosby T, Hardwick RH, Riley S, Greenaway K, Cromwell
expandable metal stents for malignant esophageal obstruction in D. A population-based observational study on the factors associated
100 patients: short-term and long-term follow-up. Ann Thorac Surg. with the completion of palliative chemotherapy among patients with
2001;71(6):1797-1801, discussion 1801-1802. oesophagogastric cancer. BMJ Open. 2015;5(3):e006724.
10. Shaikh M, Choudhury NR, Knott R, Garg S. Engineering stent based 31. Opstelten JL, de Wijkerslooth LR, Leenders M, et al. Variation in
delivery system for esophageal cancer using docetaxel. Mol Pharm. palliative care of esophageal cancer in clinical practice: factors
2015;12(7):2305-2317. associated with treatment decisions. Dis Esophagus. 2017;30(2):1-7.
CHAPTER
Anastomotic Complications After Esophagectomy:
Frequency, Prevention, and Management 43
Tamar B. Nobel
| Jessica G.Y. Luc
| Daniela Molena
S
ince Dobromysslow described the first esophageal severe anastomotic leak (35.8 vs. 54.8 months, and 34
resection with successful anastomotic reconstruc- vs. 47.9 months, respectively) and 35% increased risk of
tion in 1901, the outcomes of esophagectomy have recurrence.5,6 These grave consequences make prevention,
significantly improved and this surgery is now the mainstay when possible, and early identification of leaks of utmost
of treatment for esophageal cancer.1 Data from the Society importance to minimize the impact of such events.
t
of Thoracic Surgeons (STS) National Database reported
ne
a major complication rate of 33.1% and mortality rate of CLASSIFICATION AND INCIDENCE
3.1%.2 However, anastomotic complications continue to
pose a technical challenge for surgeons and negatively The incidence of anastomotic leak after esophagectomy
k.
impact patient postoperative recovery. Familiarity with such ranges from 0% to 35%.7 Analyses of STS database results
complications and the associated principles of management reported a 12% incidence of leaks requiring medical
oo
may better guide perioperative patient care and thus or surgical intervention.2 Variability in the definition of
mitigate the long-term sequelae of such events. In this anastomotic leak has led to wide discrepancies in reported
chapter, we discuss the etiology, diagnosis, and treatment rates. Bruce et al. determined that fewer than 40% of
yb
of anastomotic complications after esophagectomy. articles pertaining to anastomotic leak after esophagectomy
It is important to understand the techniques used included the criteria used to establish the diagnosis, and
for esophageal resection in order to better address another systematic review noted definitions included in
diversions from the anticipated postoperative course. only 28.3% of studies with 22 different descriptions used.7,8
er
Currently, the most commonly used surgical approaches Possible explanations for differences in reported out-
to esophagectomy include the Ivor Lewis esophagectomy comes in the literature include variable diagnostic tools
rg
(ILE); transhiatal esophagectomy (THE); McKeown, or used to identify a leak, time frame used for evaluation
tri-incisional esophagectomy (TIE); and thoracoabdomi- (e.g., 30 vs. 90 day), and anastomotic location.5 There
nal esophagectomy (TAE). The procedure type dictates have been multiple groups that have sought to reach
su
the anastomotic site; TAE and ILE use an intrathoracic a consensus on a definition for anastomotic leak. The
anastomosis, whereas TIE and THE use a cervical anasto- International Multispecialty Anastomotic Leak Global
motic location. When reconstructing the esophagus, the Improvement Exchange group (IMAGINE) defined
://
stomach is most commonly used as a conduit. Alternatives gastrointestinal postoperative anastomotic leak as “a defect
include pedicled colonic or small bowel, and rarely, small of the integrity in a surgical join between two hollow
tp
bowel free graft. Finally, there are multiple methods for viscera with communication between the intraluminal and
performing the anastomosis. The double layer handsewn extraluminal compartments.”9 In 2015 the Esophagectomy
technique was first described in 1942 by Churchill and Complications Consensus Group (ECCG), a group of 21
ht
Sweet and has since been modified; it now includes several high-volume esophageal surgeons from 14 countries, met
variations including continuous versus interrupted suture, to create a standardized system for defining and recording
single- or double- layer sutures, and different suture types. complications after esophagectomy (Table 43.1).10
More recent developments have led to increased use of Another commonly referenced classification system was
mechanical stapling devices, which include both circular proposed by Lerut et al. (Table 43.2).11 In this definition,
and linear techniques.3,4 These differences in methodologic clinically occult leaks diagnosed on routine postoperative
approaches may impact the likelihood and presentation imaging comprise the most minor end of the spectrum
of anastomotic complications following esophagectomy. while conduit necrosis is the most severe presentation.
Leakage at the anastomotic site is the Achilles heel The type of leak has significant clinical implications, with
of every esophagectomy. It is associated with a threefold grade I/II being relatively low risk and grade III and IV
increased risk of postoperative death and up to 60% having mortality rates near 60% and 90%, respectively.4
mortality. Furthermore, such events lead to increased
length of hospital stay, delayed initiation of oral feeding,
and increased risk of reoperation. Anastomotic leak may
RISK FACTORS
also influence oncologic outcome; in a multicenter study Failed anastomotic healing is affected by multiple inter-
of 2994 postoperative patients with esophageal cancer, related factors. An understanding of these predictive
Markar et al. demonstrated a significant reduction in variables is important in the preoperative setting because
both overall and disease-free survival in patients with modifiable factors may be better optimized, and fixed
473
Anastomotic Complications After Esophagectomy: Frequency, Prevention, and Management CHAPTER 43 473.e1
ABSTRACT
This chapter examines the etiology, diagnostic techniques,
and treatment approaches of anastomotic complications
after esophagectomy. An understanding of these much
feared events may help mitigate the impact of the associ-
ated negative clinical sequelae. Anastomotic leak has a wide
variation in presentation, from an incidental radiologic
finding to a catastrophic septic event. Patients should
be adequately resuscitated, and further management is
dependent upon the location of the leak and clinical
presentation. Mild cases may be observed or managed
with local drainage. The most severe cases are conduit
necrosis. In suspected cases, direct visualization of the
conduit is necessary, and subsequent resection must be
performed if the diagnosis is confirmed. Anastomotic leak
is a frequent cause of anastomotic stricture; however, later
presentation must raise suspicion for tumor recurrence.
In benign cases, endoscopic dilation is often sufficient.
t
ne
Leaks are also a frequent cause of conduit airway fistulas.
Such events are rare but are life threatening and must
rapidly be identified. In such cases, a trial of conservative
treatment may be pursued with subsequent endoscopic
k.
therapy and, ultimately, surgical resection if this is not
successful.
oo
KEYWORDS
anastomotic leak
yb
esophagectomy
conduit necrosis
anastomotic stricture
er
t
• Identified endoscopically and not associated with free
ne
Lerut T, Coosemans W, Decker G, et al. Anastomotic complications after
anastomotic or conduit leak esophagectomy. Dig Surg. 2002;19:92-98.
• Treatment—surgical therapy not involving esophageal
diversion
k.
Type III: Conduit necrosis extensive esophagectomy often present with malnutrition. Poor
• Treatment—treated with conduit resection with diversion nutritional status, commonly defined as hypoalbuminemia
oo
CHYLE LEAK or weight loss, has previously been demonstrated to be
Type I: Treatment – enteric dietary modifications associated with increased risk of anastomotic leak. 12,13
Type II: Treatment – total parenteral nutrition Analysis of the STS database found that heart failure,
yb
Type III: Treatment – interventional or surgical therapy* hypertension, renal insufficiency, and procedure type
Severity Level were predictive of anastomotic leak.14 Use of neoadjuvant
< 1 liter output/day therapy has been associated with increased risk of leak
> 1 liter output/day in some series, although this appears to be dependent
er
For example, a chyle leak initially producing 1200 ml/day and on radiation dose given. A review of 1939 patients dem-
successfully treated by stopping enteric feeds and initiating onstrated no increase in complications in patients who
rg
Dietary modification allowed gastrointestinal system, the esophagus does not have a
Type II: Injury requiring elective surgical procedure, for serosal layer. This, in combination with the longitudi-
tp
example, thyroplasty or medialization procedure nal orientation of the muscle fibers of the esophagus,
Type III: Injury requiring acute surgical intervention (due to contributes to increased fragility and decreased suture
aspiration or respiratory issues), for example, thyroplasty or security.4,13,17 Attempts to create additional protection using
ht
medialization procedure a free peritoneal patch did not impact the rate of leak
Severity Level but did increase the rate of stricture.18 Furthermore, the
Unilateral esophagus is primarily located within the thorax, which
Bilateral may create a unique challenge for esophageal anastomoses
For example, a unilateral vocal cord injury requiring elective
when compared with other types of gastrointestinal leaks.
medialization procedure. Final Type IIA
The negative intrathoracic pressure may draw gastric
*Does not include elective insertion of additional surgical or interventional fluid across anastomotic lines and result in leakage. Such
chest drains. leaks into the pleural space may affect the dynamics
GI, Gastrointestinal; TPN, total perenteral nutrition.
From Low DE, Alderson D, Cecconello I, et al. International consensus of
of intrathoracic pressure and result in respiratory and
standardization of data collection for complications associated with hemodynamic instability.19
esophagectomy: esophagectomy Complications Consensus Group (ECCG). The ability of an anastomosis to heal successfully is
Ann Surg 2015;262:286-294. directly associated with maintenance of adequate tissue
perfusion to the conduit. In the course of gastric mobi-
characteristics may be used for risk stratification and lization and resection of the lesser curvature, the major
patient selection in attempts to minimize poor outcomes.12 vessels are ligated; this leaves approximately 60% of the
Esophagectomy is most often used to treat esophageal gastric tube to be supplied by the right gastroepiploic
cancer and also has a role in management of motility disor- vessels and the remaining proximal portion is supplied by
ders; therefore it is not surprising that patients undergoing small collateral vessels. As such, the anastomosis, which is
Anastomotic Complications After Esophagectomy: Frequency, Prevention, and Management CHAPTER 43 475
usually made at the fundus, is within the most ischemic Laparoscopic surgery has become the standard approach
part. Caution must be exercised to minimize trauma to many gastrointestinal operations because it has been
to the collateral vessels during dissection.4 Tension on demonstrated to be safe with better postoperative results.
the anastomosis can lead to impaired blood supply and Similarly, there has been increased utilization of minimally
result in ischemia. Some authors have suggested that the invasive approaches to esophagectomy. A 2016 Cochrane
utilization of the whole stomach as opposed to a gastric Review evaluating open versus laparoscopic THE observed
tube may better preserve blood supply by not disturbing fewer overall complications and decreased length of
collateral circulation. Arguments against use of the whole stay.28 Biere et al. performed a randomized control trial
stomach claim that use of a gastric tube yields less gastric comparing open versus minimally invasive transthoracic
distention and decreased surface area for acid secretion, esophagectomy and demonstrated significant postoperative
both of which can be detrimental to the anastomosis benefits, including decreased incidence of pulmonary
and also lead to poor quality of life. A narrow gastric infections, decreased length of stay, decreased opera-
conduit with longer length can better reach the cervical tive blood loos, and improved quality of life at 6 weeks
anastomosis site without tension but must be balanced postoperatively and postoperative pain scores. In addition,
against excessive narrowing leading to impaired blood no differences were demonstrated in number of lymph
supply.11,12,19 Controversy remains as to whether other nodes harvested and rate of R0 resection.29
types of conduit, most commonly colonic, have a similar
incidence of leak; however, colonic interposition does PREVENTION OF LEAK
t
require three anastomoses as opposed to one when using
ne
a gastric conduit.4 Preoperative patient optimization with correction of malnu-
Cervical anastomosis has been associated with a higher trition, optimization of medical conditions, and decreasing
incidence of leakage than thoracic, with rates ranging steroid use have been demonstrated to help decrease risk
k.
from 2% to 26%.20 This may possibly be explained by of leak.5 Procedures performed at low-volume centers
the longer distance required for blood supply to travel have been associated with increased risk of postoperative
oo
to the anastomosis site in the neck.14 In comparison, and long-term mortality, as well as increased risk of any
thoracic anastomoses are associated with leak rates of 0% to (and severe) leaks, which highlights the importance of
9.3%.11,14,20 There was previously suggestion in the literature careful surgical technique in minimizing the incidence of
yb
that thoracic leaks are associated with higher postopera- leak. Intraoperatively, there have been attempts to better
tive mortality, but more recent data do not support this characterize perfusion to the anastomosis because there
finding when considering early (30-day) outcomes.14,20,21 is no standardized therapy. Multiple approaches exist,
The higher leak rate associated with cervical anastomosis including assessment of graft color, temperature, and
er
is more accepted because it is generally contained and checking Doppler signals. Doppler is limited to assessment
thus associated with lower risk. Thoracic anastomotic of macrocirculation. Fluorescence imaging is a promis-
rg
breakdown is often associated with mediastinal soilage ing approach to assessment of the microcirculation and
and may have a more severe clinical course.19 macrocirculation of the gastric conduit. Weaker perfusion,
The type of anastomotic technique selected most as assessed by intraoperative laser-assisted fluorescent dye
su
often depends upon the surgeon’s preference. Stapled angiography, has been demonstrated to be correlated to
anastomotic technique has been used for approximately leak and may be a promising approach.30
25 years and began to increase in popularity due to easier Following stomach mobilization, a 50% drop in
://
reconstruction, especially in areas where exposure and gastric fundus oxygen tension has been measured, and
access are limited. Heitmiller et al. described an 0.8% leak further study has described a direct correlation between
tp
rate after two-layer handsewn anastomosis.13,22 Blackmon intraoperative gastric fundus oximetry and resultant
et al. published a propensity-matched analysis compar- anastomotic success. Ischemic preconditioning has been
ing outcomes between side-to-side stapled anastomosis, described as an attempt to preoperatively redistribute
ht
end-to-end circular stapled anastomosis, and handsewn, gastric blood supply. Embolization of the left and right
with no significant difference in leak rate noted.23 Mul- gastric and splenic arteries was performed 2 to 3 weeks
tiple randomized control trials have compared various before esophagectomy, thus leaving the stomach dependent
techniques in attempt to identify the safest approach on the right gastroepiploic artery. Results in animal models
but have failed to identify a significant difference in leak have demonstrated decreased anastomotic leak using these
rate.24-26 Reconstruction can be performed either via an techniques, but similar results have not been demonstrated
anterior retrosternal route or a posterior mediastinal route in humans. An additional technical consideration includes
when performing a cervical anastomosis. Urschel et al. division of the interclavicular ligament to alleviate conduit
performed a meta-analysis that failed to demonstrate a venous obstruction.4,5
significant difference in leak rate; however, the posterior Postoperatively, gastric distention may contribute
route is alleged to have the advantages of shorter distance to anastomotic tension and inhibit venous drainage.
to the anastomosis, resulting in decreased tension, avoid- Minimization of distention with use of nasogastric tube
ance of foregut angulation, less cardiopulmonary morbidity, and promotility agents may help to alleviate this. Pyloro-
soft tissue coverage of potentially devascularized airways, myotomy or pyloroplasty at time of original procedure may
and preservation of the skeletal structures of the thoracic also reduce incidence of this complication. In addition,
inlet. Anterior reconstruction is primarily justified due to use of pharmacologic agents has been previously studied.
the minimization of tumor recurrence by avoiding the Administration of prostaglandin E1 within an hour of
posterior tumor bed.27 gastric tube creation has been shown to increase blood
476 SECTION I Esophagus and Hernia
flow to the tissue but has failed to show a clinical benefit The management of intrathoracic leaks depends upon
to date.4 the degree of sepsis. Patients with contained collections
may be treated with percutaneous drainage. Patients with
hemodynamic stability in the setting of sepsis should be
DIAGNOSIS OF LEAK taken to the operating room for washout and drainage.
Diagnosis of anastomotic leaks is dependent upon the size Conservative operative treatment, with débridement and
and location of the leak. In cases of conduit necrosis, patients refashioning of the anastomosis, has been described with
often present within 48 to 72 hours with catastrophic sepsis. 25% risk of recurrent leak.5,11,33
Clinical suspicion is necessary to correctly diagnose more Recently, there has been increased interest in stenting
subtle leaks. The most obvious signs include bilious output as a mechanism of closure of anastomotic defects with
in surgical drains. Cervical leaks usually present within 5 to possible adjunctive percutaneous drainage of associated
10 days with associated fever, drainage, and wound erythema. fluid collections. In cases in which less than 30% of cir-
Intrathoracic leaks may be less apparent. Unexplained cumference of the anastomosis has broken down, stent
low-grade fever, tachycardia, or leukocytosis should prompt placement may be appropriate. Covered self-expanding
suspicion.4,5,13 A proposed risk assessment score found that stents are preferred to facilitate future removal. Stent
C-reactive protein, white blood cell count, and albumin migration is a common problem, which may be minimized
levels could be used to predict major complications with with the use of longer stents or with adjuncts, includ-
sensitivity of 89% and specificity of 63%.31 ing clipping or suturing. Other risks associated include
t
Many practices perform routine contrast esophagram inadequate coverage resulting in persistent leak, plugging,
ne
studies postoperatively. Usually water-soluble contrast is and erosion into surrounding structures. Freeman et al.
used initially followed by barium. Grade I leaks are usually reported outcomes of 17 patients treated with stents
clinically silent and detected on such routine imaging. after intrathoracic anastomotic leak and demonstrated
k.
Computed tomography (CT) scan with oral contrast is 94% success in healing with 18% stent migration. The
more sensitive than contrast swallow study, and it is our majority (82%) of patients were able to resume a diet 72
oo
preferred approach when suspicious for leak. Endoscopy hours after stenting, and stents were removed at a mean
is a useful adjunctive tool to confirm leak and guide of 17 days.5,12,34 Endoluminal vacuum therapy has emerged
potential intervention. Despite concern regarding an as a new alternative approach to anastomotic leak after
yb
invasive procedure in the setting of a fresh anastomosis, esophagectomy; the sponge is placed endoscopically into
systematic endoscopy following esophagectomy has been an intraluminal or intracavitary position and the vacuum
demonstrated to be safe.5,13 then closes the esophageal defect and drains any associated
collection via an intranasal route.35
er
MANAGEMENT
CONDUIT NECROSIS
rg
optimization of respiratory function, broad-spectrum subclinical ischemia to frank necrosis.36 The sequelae
antibiotics, and nutrition, preferably enteral downstream of conduit ischemia may be quite serious. Patients with
from the site of leakage. Furthermore, collections near ischemia may ultimately develop a stricture or, more
://
the site of the anastomosis should be drained.5,32 In the concerning, an anastomotic leak requiring reoperation.
absence of diffuse conduit necrosis, further treatment Conduit necrosis is the most severe grade of anastomotic
tp
is dependent upon the location of leak and severity of leak, as described previously. Mortality after conduit
clinical presentation. necrosis may exceed 90%, highlighting the importance
Cervical leaks that are small and well contained may be of rapid identification and treatment of this problem,
ht
managed conservatively with nothing by mouth. Antibiotics and when possible, prevention.37
may often not be necessary in the absence of sepsis. Multiple risk factors for conduit necrosis have been
Repeat contrast study or endoscopy may be performed identified, including improper technique in the creation or
to document healing. Larger, contained cervical leaks manipulation of the gastric tube, radiation, low periopera-
with associated wound erythema or fluctuance require tive cardiac output, postoperative hypotension, previous
drainage. This may often be accomplished by opening upper abdominal surgery, malnutrition, peptic ulcer
up the surgical wound. In cases in which a small defect disease, twisting of the stomach conduit as it passes into the
is observed, direct suture closure or stenting may be mediastinum, and a tight hiatal opening. Radiation creates
performed to accelerate healing. Median time to closure a fibrotic reaction that decreases the microvascular blood
is approximately 2 to 3 weeks. If patients fail to respond supply of the conduit. By using caution while mobilizing
promptly to management with the above approaches, the conduit and taking caution in maintaining adequate
inadequate drainage or conduit necrosis should be con- patient perfusion in the perioperative period, these factors
sidered and further investigation should be undertaken. may be minimized.36,38
Surgical management in the setting of uncontrolled sepsis Patients with conduit necrosis most commonly present
includes débridement and assessment of the viability of with severe sepsis within the first week postoperatively. They
the conduit, with possible resection in setting of necrosis. may initially have an unexplained tachycardia or leukocytosis
Subsequently, there is a high risk of stricture formation with rapid clinical decompensation. The most important
following cervical anastomotic leaks.4,5,11 principle for management of such patients is to have a
Anastomotic Complications After Esophagectomy: Frequency, Prevention, and Management CHAPTER 43 477
t
diagnosis, there are multiple diagnostic options available. of severe anastomotic failure, secondary to ischemia, are
ne
Traditionally, an upper gastrointestinal contrast study may more likely to be resistant to serial dilation. In addition to
be used to identify presence of an anastomotic leak and endoscopic dilation, it is important to administer proton-
may possibly demonstrate cobblestoning of the mucosa. pump inhibitor treatment to prevent recurrence.11
k.
CT with oral contrast may yield additional information
regarding associated effusions, although a normal exam CONDUIT AIRWAY FISTULA
oo
does not exclude the possibility of ischemia and fluid
and air within the mediastinum may be normal in the The esophagogastric anastomosis lies close to the lung
postoperative state. Endoscopy should be the next diagnostic parenchyma and membranous airway, which increases
modality used.37,38 Although there was historical concern
yb
the risk of possible fistulization between the airway and
regarding anastomotic disruption with this study, Page et al. the conduit. Such events, although rare with reported
demonstrated that endoscopy within 1 week of operation incidence of 0.04% to 0.3%, may be life-threatening and
was safe and effective in detecting anastomotic leakage.39 must be rapidly identified.38
er
A proposed endoscopic grading system identifies various Multiple patient characteristics and perioperative
degrees of ischemia with the goal of guiding management complications may precipitate this event. A strong cor-
rg
(Table 43.3).40 In the setting of a small leak, nonoperative relation has been made between neoadjuvant therapy and
management with possible stenting will allow the patient risk of fistula formation, likely due to secondary tissue
to recover. Patients with a completely necrotic conduit injury and ischemia. Most often, conduit airway fistulas
su
are most often septic and brought to the operating room occur in the setting of anastomotic leak secondary to
for emergent exploration. A video-assisted thorascopic leakage of enteric contents resulting in tissue necrosis
investigation or thoracotomy in the case of an intratho- and erosion. Intraoperative tracheobronchial injury is
://
racic anastomosis or reopening of the neck incision must a rare event with reported incidence of 1.8% during
be performed to visualize the conduit. If this confirms THE and 0.8% during transthoracic esophagectomy. It
tp
necrosis, then the conduit must be resected and the patient may serve as another potential source of airway conduit
should be diverted with an end esophagostomy, venting fistulization postoperatively due to the opening in the
gastrostomy, and feeding jejunostomy. Care should be airway. 44 Additional etiologies may include ischemia
ht
taken to maintain the longest possible length of remaining due to extensive dissection around the trachea, stent
esophagus to facilitate future reconstruction; however, erosion after management of leak, after endoscopic
patients should be resuscitated and sepsis should resolve dilation of anastomotic stricture, or cuff-induced tra-
prior to any such attempts. Options for reconstruction cheal necrosis in the setting of prolonged endotracheal
include colonic interposition or jejunal transfer.37,38 intubation.38,45
Patients may initially present with mild symptoms, such
as cough with oral intake, or more serious symptoms,
ANASTOMOTIC STRICTURE such as recurrent pneumonia. In the most extreme cases,
Anastomotic strictures often arise in the setting of previous patients may present with mediastinitis. A high index of
anastomotic leak. Reported incidence of stricture after suspicion will enable more prompt identification, which is
esophagectomy is 10% to 40%; however, as strictures of utmost importance as aspiration of gastric secretions into
are often diagnosed in the setting of patient-reported the lungs can lead to a severe pneumonitis and respiratory
dysphagia, it is possible that the true incidence is much compromise. Oral contrast radiologic studies are usually
higher. Strictures that present early postoperatively are most the first diagnostic tool but may not visualize very small
often benign. Later presentation should raise suspicion fistulas. Endoscopic evaluation is the preferred diagnostic
for tumor recurrence.4,5 modality to localize the fistula although mucosal folding
Anastomotic technique has been correlated with inci- within the conduit may obscure small openings. As such,
dence of stricture and is associated with the resultant size of bronchoscopic evaluation at same time as endoscopic
478 SECTION I Esophagus and Hernia
assessment is the best way to characterize the size and 13. Mitchell JD. Anastomotic leak after esophagectomy. Thorac Surg
location of the fistula.12,38,45 Clin. 2006;16:1-9.
14. Kassis ES, Kosinski AS, Ross P Jr, et al. Predictors of anastomotic leak
The management of fistulas should be based upon the after esophagectomy: an analysis of the Society of Thoracic Surgeons
size and site of the fistula in addition to the severity of the General Thoracic Database. Ann Thorac Surg. 2013;96:1919-1926.
symptoms. In patients with benign clinical presentation, 15. Escofet X, Manjunath A, Twine C, et al. Prevalence and outcome
attempts at conservative treatment (nothing by mouth, of esophagogastric anastomotic leak after esophagectomy in a UK
regional cancer network. Dis Esophagus. 2010;23:112-116.
antibiotics) may be considered; however, failure to heal 16. Mungo B, Molena D, Stem M, et al. Does neoadjuvant therapy for
within 4 to 6 weeks should prompt further intervention. esophageal cancer increase postoperative morbidity or mortality?
Endoscopic approaches have been described including Dis Esophagus. 2015;28:644-651.
attempts at closure using fibrin glue, hemostastic clips, 17. Akiyama H. Esophageal anastomosis. Arch Surg. 1973;107:512-514.
and mesh plugs.38,45 Stent placement may be considered 18. Van Oosterom FJ, Van Lanschot JJ, Oosting J, et al. A free peritoneal
patch does not affect the leakage rate but increases stricture formation
as a temporizing measure in severe cases to control con- of a cervical esophago-gastrostomy. Dig Surg. 1999;16:379-384.
tamination. Ultimately, a surgical repair may be required. 19. Chen KN. Managing complications I: leaks, strictures, emptying,
The preferred approach is to repair the anastomosis and reflux, chylothorax. J Thorac Dis. 2014;6:S355-S363.
close the airway defect with interposition of vascularized 20. Biere SS, Maas KW, Cuesta MA, et al. Cervical or thoracic anastomosis
after esophagectomy for cancer: a systematic review and meta-analysis.
soft tissue interposed between the suture lines to prevent Dig Surg. 2011;28:29-35.
recurrent fistulization. In severe cases in which the conduit 21. Chang AC, Ji H, Birkmeyer NJ, et al. Outcomes after transhiatal
must be excised, the preferred approach is with resection and transthoracic esophagectomy for cancer. Ann Thorac Surg.
t
and esophagostomy with delayed reconstruction after the 2008;85:424-429.
ne
patient is clinically stabilized from sepsis. When performing 22. Heitmiller RF, Fischer A, Liddicoat JR. Cervical esophagogastric
anastomosis: results following esophagectomy for carcinoma. Dis
reconstruction, colonic interposition in the substernal Esophagus. 1999;12:264-269.
space can be used to avoid reoperation in the area of 23. Blackmon SH, Correa AM, Wynn B, et al. Propensity-matched analysis
k.
inflammation. In addition, jejunal interposition can be of three techniques for intrathoracic esophagogastric anastomosis.
used.12,38,45,46 Ann Thorac Surg. 2007;83:1805-1813.
24. Beitler AL, Urschel JD. Comparison of stapled and hand-sewn
oo
esophagogastric anastomoses. Am J Surg. 1998;175:337-340.
SUMMARY 25. Law S, Fok M, Chu KM, et al. Comparison of hand-sewn and stapled
esophagogastric anastomosis after esophageal resection for cancer: a
yb
Anastomotic complications after esophagectomy can be prospective randomized controlled trial. Ann Surg. 1997;226:169-173.
rapidly life-threatening. Awareness of the types of complica- 26. Soluja SS, Ray S, Pal S, et al. Randomized trial comparing side-to-
side stapled and hand-sewn esophagogastric anastomosis in neck.
tions that can occur and their clinical presentations is J Gastrointest Surg. 2012;16:1287-1295.
critical for early recognition and diagnosis.
er
27. Urschel JD, Urschel DM, Miller JD, et al. A meta-analysis of random-
ized controlled trials of route of construction after esophagectomy
for cancer. Am J Surg. 2001;182:470-475.
REFERENCES
rg
brustabschitte. Zentralbl Chir. 1901;28:1. 29. Biere SS, van Berge Henegouwen MI, Maas KW, et al. Minimally
2. The Society of Thoracic Surgeons General Thoracic Surgery Database invasive versus open oesophagectomy for patients with oesophageal
Task Force. The Society of Thoracic Surgeons composite score for cancer: a multicentre, open-label, randomised controlled trial. Lancet.
evaluating esophagectomy for esophageal cancer. Ann Thorac Surg. 2012;379:1887-1892.
://
4. Cassivi SD. Leaks, strictures and necrosis: a review of anastomotic 31. Noble F, Curtis N, Harris S, et al. Risk assessment using a novel
complications following esophagectomy. Semin Thorac Cardiovasc score to predict anastomotic leak and major complications after
Surg. 2004;16:124-132. oesophageal resection. J Gastrointest Surg. 2012;16:1083-1095.
ht
5. Messager M, Warlaumont M, Renaud F, et al. Recent improvements 32. Martin LW, Hofstetter W, Swisher SG, et al. Management of
in management of esophageal anastomotic leak after surgery for inrathoracic leaks following esophagectomy. Adv Surg. 2006;40:173-
cancer. Eur J Surg Oncol. 2017;43:258-269. 190.
6. Markar S, Gronnier C, Duhamel A, et al. The impact of severe anasto- 33. Whooley BP, Law S, Alexandrou A, et al. Critical appraisal of the
motic leak on long-term survival and cancer recurrence after surgical significance of intrathoracic anastomotic leakage after esophagectomy
resection for esophageal malignancy. Ann Surg. 2015;262:972-980. for cancer. Am J Surg. 2001;181:198-203.
7. Blencowe NS, Strong S, McNair AGK, et al. Reporting of short-term 34. Freeman RK, Vyverberg A, Ascioti AJ. Esophageal stent placement
clinical outcomes after esophagectomy: a systematic review. Ann for the treatment of acute intrathoracic anastomotic leak after
Surg. 2012;255:658-666. esophagectomy. Ann Thorac Surg. 2011;92:204-208.
8. Bruce J, Krukowski ZH, Al-Khairy G, et al. Systematic review of the 35. Kuehn F, Loske G, Schiffmann L, et al. Endoscopic vacuum therapy
definition and measurement of anastomotic leak after gastrointestinal for various defects of the upper gastrointestinal tract. Surg Endosc.
surgery. Br J Surg. 2001;88:1157-1168. 2017;doi:10.1007/s00464-016-5404-x; [Epub ahead of print].
9. Chadi SA, Fingergut A, Berho M, et al. Emerging trends in the 36. Wormuth JK, Heitmiller RF. Esophageal conduit necrosis. Thorac
etiology, prevention and treatment of gastrointestinal anastomotic Surg Clin. 2006;16:11-22.
leakage. J Gastrointest Surg. 2016;20:2035-2051. 37. Dickinson KJ, Blackmon SH. Management of conduit necrosis
10. Low DE, Alderson D, Cecconello I, et al. International consensus of following esophagectomy. Thorac Surg Clin. 2015;25:461-470.
standardization of data collection for complications associated with 38. Meyerson SL, Mehta CK. Managing complications II: conduit failure
esophagectomy: Esophagectomy Complications Consensus Group and conduit airway fistulas. J Thorac Dis. 2014;6:S364-S371.
(ECCG). Ann Surg. 2015;262:286-294. 39. Page RD, Asmat A, McShane J, et al. Routine endoscopy to detect
11. Lerut T, Coosemans W, Decker G, et al. Anastomotic complications anastomotic leakage after esophagectomy. Ann Thorac Surg.
after esophagectomy. Dig Surg. 2002;19:92-98. 2013;95:292-298.
12. Jones CE, Watson TJ. Anastomotic leakage following esophagectomy. 40. Oezcelik A, Banki F, Ayazi S, et al. Detection of gastric conduit ischemia
Thorac Surg Clin. 2015;25:449-459. or anastomotic breakdown after cervical esophagogastrostomy: the
Anastomotic Complications After Esophagectomy: Frequency, Prevention, and Management CHAPTER 43 479
use of computed tomography scan versus early endoscopy. Surg 44. Hulscher JB, ter Hofstede E, Kloek J, et al. Injury to the major
Endosc. 2010;24:1948-1951. airways during subtotal esophagectomy: incidence, management
41. Orringer MB, Marshall B, Stirling MC. Transhiatal esophagec- and sequelae. J Thorac Cardiovasc Surg. 2000;120:1093-1096.
tomy for benign and malignant disease. J Thorac Cardiovasc Surg. 45. Buskens CJ, Hulscher JBF, Fockens P, et al. Benign trachea-neo-
1993;105:265-276. esophageal fistulas after subtotal esophagectomy. Ann Thorac Surg.
42. Tretino P, Pompeo E, Nofroni I, et al. Predictive value of early 2001;72:221-224.
postoperative esophagoscopy for occurrence of benign stenosis 46. Boyd M, Rubio E. The utility of stenting in the treatment of airway
after cervical esophagogastrostomy. Endoscopy. 1997;29:840-844. gastric fistula after esophagectomy for esophageal cancer. J Bronchology
43. Briel JW, Tamhankar AP, Hagen JA, et al. Prevalence and risk factors Interv Pulmonol. 2012;19:232-236.
for ischemia, leak and stricture of esophageal anastomosis: gastric
pull-up versus colon interposition. J Am Coll Surg. 2004;198:536-541.
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
PART EIGHT
Miscellaneous Esophageal
Conditions
CHAPTER
t
ne
T
he esophagus has been dubbed the organ of determine its etiology. An endoscopist must be prepared
symptoms not signs. Inflammation of this dynamic to take appropriate biopsies at the time of the procedure.
k.
conduit will elicit manifestations that can severely Radiologic studies may also help to shed light on the
impact a patient’s life. These symptoms are universal cause and to rule out complications, such as stricture
oo
regardless of the cause of inflammation. They may include formation or abnormal anatomy. The specific esophageal
dysphagia, odynophagia, heartburn (pyrosis?), chest pain, study is double-contrast barium swallow. This study should
nausea, vomiting, hematemesis, anorexia, and weight loss. be avoided in patients with severe dysphagia who are at
yb
So-called atypical respiratory symptoms may also occur, risk for aspiration. Otherwise it provides a “roadmap”
such as cough, bronchospasm, and aspiration. Due to the prior to endoscopy. Other tests including chest computed
high prevalence of reflux disease, the vast majority of these tomography (CT), manometry, pH/impedance, and
patients who have very similar symptoms will be labeled nuclear scintigraphy are obtained only on an individual
er
as having gastroesophageal reflux disease (GERD) and basis to confirm a specific suspicion.
will receive antireflux medication that will fail to resolve Laboratory tests can be helpful to rule out immunosup-
rg
cause a fibrotic reaction leading to benign strictures. Deep initially treat the complications of this disease, such as
ulceration may occasionally cause severe bleeding or, severe bleeding, malnutrition, or esophageal perforation.
tp
more rarely, even perforation and mediastinitis. In addi- Esophagitis frequently causes clinical symptoms, includ-
tion, malignancy has been associated with long-standing ing dysphagia, odynophagia, and regurgitation, which
esophagitis, especially with reflux when associated with compel patients to seek evaluation and treatment. The
ht
ABSTRACT
Esophagitis frequently causes clinical symptoms, including
dysphagia, odynophagia, and regurgitation, that compel
patients to seek evaluation and treatment. The most
common cause of these symptoms is esophageal inflam-
mation related to reflux; however, causes of nonreflux
esophagitis are an increasingly important diagnostic
consideration. Nonreflux esophagitis remains relatively rare
in clinical practice, but its incidence has seen a dramatic
increase over the last two decades. In particular, there has
been a rapid expansion in the incidence and prevalence of
eosinophilic esophagitis. Other causes include infections
esophagitis (fungal, viral, and tuberculous), medication-
induced esophagitis, radiation esophagitis, and acute
esophageal necrosis. This chapter outlines these causes
and examines the clinical presentation, epidemiology,
diagnostic work-up, and management of each distinct
cause.
t
ne
KEYWORDS
Nonreflux esophagitis; eosinophilic esophagitis; pill
esophagitis
k.
oo
yb
er
rg
su
://
tp
ht
Nonreflux Esophagitis CHAPTER 44 481
Eosinophilic esophagitis
Infectious esophagitis • Candida
• Cytomegalovirus
• Herpes
• Tuberculosis
Radiation-induced esophagitis
Pill esophagitis
Acute esophageal necrosis
EOSINOPHILIC ESOPHAGITIS
EoE is a chronic inflammatory condition, characterized by
symptomatic esophageal dysfunction with intraepithelial
t
eosinophilic infiltration on pathologic examination.1 The
ne
presence of esophageal eosinophilia was initially described
in the presence of GERD. 2 Individual case reports of
symptomatic eosinophilia in the absence of GERD began
k.
to emerge in the 1970s3 and 1980s,4,5 but it was not until FIGURE 44.1 Food impaction in eosinophilic esophagitis with
1993 that Attwood et al. described the current entity.6 characteristic concentric rings. (From Gonsalves N, Policarpio-
oo
Over the last two decades, there has been tremendous Nicolas M, Zhang Q, Rao MS, Hirano I. Histopathologic variability
growth in the observed incidence of this disease and a and endoscopic correlates in adults with eosinophilic esophagitis.
parallel growth in our understanding.3 Gastrointest Endosc. 2006;64:313–319.)
yb
CLINICAL PRESENTATION
EoE can present at any age, but the clinical presentation treatment cessation.14 With years of persistent inflamma-
differs between adults and children.2 In adults, the most tion, the disease seems to progress toward a fibrostenotic
er
common symptom of EoE is dysphagia, particularly to phenotype; according to one large case series, the risk of
solid foods. In one series, dysphagia was the presenting fibrostenotic complications doubles with each decade.15
rg
complaint in 83% of patients.7 The dysphagia is usually The disease is more common in Caucasians and has a male
intermittent and rarely accompanied by odynophagia. In predominance, which is consistently reported at 3 to 4 : 1
severe circumstances, persistent dysphagia and odynopha- across all ages.16,17 There is also a strong association with
su
gia can lead to malnutrition.8 Other symptoms include atopic diseases, including a correlation with environmental
heartburn (30% to 60% of patients) and noncardiac chest and food allergies.18
pain.2 Of particular clinical importance is the frequency There is an increasingly large body of evidence showing
://
with which EoE causes food impaction requiring acute that EoE is rapidly increasing in incidence and in preva-
intervention (Fig. 44.1). In one series, food impaction was lence, and that this increase cannot be explained simply
tp
the presenting complaint in 42% of patients with EoE,9 by increased awareness. Current estimates indicate that
and more than half of adult patients with impaction may the prevalence is likely between 40 and 90 cases per
have esophageal eosinophilia.10 When interviewing patients 100,000 persons in the United States.16 This prevalence is
ht
complaining of dysphagia, it is important to elicit a history consistent with other Western countries, namely Australia,
of impaction events leading to retching or regurgitation, Switzerland, Spain, and Canada.14 In patients undergoing
and EoE must be considered in all patients presenting endoscopy for dysphagia, the prevalence of EoE is between
with impaction.2 12% and 22%.11,19
In children, the most common presenting complaints In terms of incidence, multiple studies suggest
are heartburn and abdominal pain, with vomiting or that the incidence of EoE has been rapidly increasing
decreased appetite.11 One large series demonstrated that since its discovery 20 years ago. In a case series from
38% of children with EoE report heartburn as their primary Hamilton County, Ohio, the incidence increased from
complaint upon presentation and 31% report abdominal 9 per 100,000 to 12.8 per 100,000 people over a 3-year
pain or dyspepsia.12 Other symptoms include growth period.20 This finding is consistent with another report
failure and, rarely, hematemesis.13 Infants and toddlers from Olmsted County, Minnesota, showing a dramatic
present with difficulty feeding, described by caregivers as increase in incidence over the 15-year period from 1990
“gagging” or “choking.” In contrast to adults, dysphagia until 2005.21 A population-based, prospective study from
is uncommon in children until adolescence.12 the Swiss EoE study group provides more evidence for a
marked increase in incidence. Like the Minnesota group,
EPIDEMIOLOGY the Swiss study compares the observed incidence with the
In children and adults, EoE is a chronic disease that rate of upper endoscopy.22 In both studies, the rise in EoE
does not resolve spontaneously and often recurs after incidence in excess of increases in the upper endoscopy
482 SECTION I Esophagus and Hernia
Persistent esophageal
symptoms
Endoscopy
w/ biopsy
Esophageal eosinophilia
on biopsy
No Yes
Eosinophilic Reflux?
esophagitis
1st line Tx
t
ne
• Swallowed topical PPI-responsive GERD
corticosteroids esophageal
• Dietary elimination eosinophilia
k.
Tx
2nd line Tx Tx
oo
• Systemic corticosteroids PPI • Lifestyle modification
• Mast cell stabilizers • Antireflux meds
• Leukotriene receptor antagonists • Antireflux surgery
yb
• Immunotherapies
er
FIGURE 44.2 Diagnostic and therapeutic algorithm for eosinophilic esophagitis. GERD, Gastroesophageal reflux disease; PPI, proton pump
inhibitor; Tx, therapy.
rg
rate suggests that there is a real increase in disease rather interpreted in isolation.”24 Thus, the diagnostic criteria
than just an increase in awareness and detection. include a combination of clinical and pathologic findings;
su
The marked increase in incidence of EoE parallels no single finding is pathognomonic (Fig. 44.2).
increases in other allergic disease, including asthma, aller- First, the patient must have symptoms of esopha-
gic rhinitis, atopic dermatitis, and various food allergies.14 geal dysfunction. Second, pathologic examination of
://
One potential explanation is the “hygiene hypothesis,” esophageal biopsy specimens must demonstrate eosinophil-
which is a frequently cited and much-discussed theory predominant inflammation with a characteristic peak value
tp
initially based on epidemiologic studies. It points to the of at least 15 eosinophils per high-power field (HPF);
decreased infection burden in industrialized nations with previous guidelines have used various thresholds for
an associated increase in atopic diseases and proposes a diagnosis, ranging from 15 to 30 eosinophils/HPF.1 Third,
ht
causative link. There are animal models that support the secondary causes of esophageal eosinophilia should be
“hygiene hypothesis” for certain specific autoimmune excluded. Fourth, the mucosal eosinophilia is confined to
diseases, but therapeutic approaches to allergic disease the esophagus and persists after a proton pump inhibitor
based on the reintroduction of specific infections have (PPI) trial. Patients who meet other criteria for EoE but
shown mixed results.23 The “hygiene hypothesis” holds a have histologic and symptomatic resolution with PPI
plausible explanation for the rising incidence of EoE, but therapy may fall into the diagnostic criteria for a separate
requires further study. Overall, the rapid rise of EoE is an entity called PPI-responsive esophageal eosinophilia (REE),
area of much debate and research. Current thinking is which is considered distinct from both EoE and GERD.
that the increase is likely multifactorial, and hypotheses Finally, a response to EoE-specific treatment, such as
include a variety of immunologic-, environmental-, and dietary elimination or topical corticosteroids, supports
microbiome-related topics.14 the diagnosis of EoE.23
To fulfill the ACG histologic criteria for diagnosis, an
DIAGNOSTIC WORK-UP endoscopic evaluation with biopsy is required. Current
According to the recently published guideline from the recommendations call for two to four biopsy specimens
American College of Gastroenterology (ACG), EoE is from both the proximal and distal portions of the esopha-
defined as, “a clinicopathologic disorder diagnosed by gus.23 The physician performing the initial endoscopic
clinicians taking into consideration both clinical and evaluation should also obtain biopsies of the antrum and/
pathologic information without either of these parameters or duodenum in patients with small intestine symptoms
Nonreflux Esophagitis CHAPTER 44 483
or endoscopic abnormalities to rule out other causes of promotes esophageal tissue remodeling and is essential for
esophageal eosinophilia, such as Crohn disease or rare the fibrostenotic changes observed in chronic EoE.29 This
infectious causes.25 To evaluate for GERD in patients with work in the disease’s biochemical pathogenesis suggests a
esophageal eosinophilia, pH monitoring has historically role for targeted anti-IL-13 or anti-IL-5, which are currently
been a useful diagnostic test.24 However, a more recent under development.28
cohort study of patients with esophageal eosinophilia Outside of the observed association with other atopic
showed a high incidence (71%) of pathologic reflux, conditions, there do not appear to be direct systemic
and many patients without reflux had improvement in consequences of the esophageal inflammation of EoE.
symptoms and pathologic response to PPI therapy. This There is some evidence of associated changes in serum
group found that pH testing was not a useful predictor of laboratory values. To date, several studies in adults and
patient response to PPI therapy.26 Therefore, pH testing children have demonstrated that approximately 40% to
may be useful in select patients, but it is not required in 50% of patients with EoE have an increased number of
the work-up of EoE.24 circulating eosinophils.24 This systemic eosinophilia has
The ACG diagnostic criteria are devoid of radiographic led to research into potential biomarkers to gauge the
or endoscopic findings; however, there are several charac- response to treatment. Currently, there is insufficient
teristics of EoE that may be apparent and are increasingly evidence to support the routine monitoring of peripheral
important clinically. Endoscopic findings of EoE include eosinophil count, total immunoglobulin E levels, or any
the development of multiple concentric rings (see Fig. other surrogate biomarker.24
t
44.1), “trachealization” or “feline esophagus,” narrow-
ne
ing, linear furrows, white exudates, and edema.1 These ALLERGIC TRIGGERS
endoscopic findings may be present in 90% to 95% of The environmental triggers that initiate the immunologic
cases.2 Furthermore, the endoscopic appearance of EoE cascades outlined previously have also received a good
k.
is under examination as a clinically relevant therapeutic deal of attention since the first descriptions of EoE in
outcome measure, and an endoscopic grading system to the 1990s. The success of dietary therapies targeting the
oo
classify disease severity has been introduced.23,24 identification and removal of dietary antigens suggests an
essential role for food-borne allergens in the pathogenesis
PATHOPHYSIOLOGY of EoE.13 In contrast, there is much debate about the role
yb
The progression of chronic inflammation to fibrostenotic of aeroallergens.
complications helps demonstrate the effect of prolonged An early case report of a 21-year-old female with EoE
eosinophilic infiltration on esophageal function. 15 showed symptomatic and histologic exacerbations during
Manometry studies in patients with biopsy-confirmed EoE pollen season with remission during the winter months; in
er
show a potential association with nonspecific dysmotility, subsequent years, several other small case series suggested
particularly of the lower esophagus, and lower esophageal a potential role of aeroallergens.31 More recently, work
rg
sphincter dysfunction.27,28 In severe cases, fibrostenotic from a national US pathology database demonstrated
pathology including ring and stricture formation, or clear geographic variability in the prevalence of EoE with
narrowing, are a source of morbidity. These fibrostenotic a strong correlation between colder climates and increased
su
complications are associated with increased symptom disease prevalence in the United States.32 Because climate
duration and underscores the importance of managing is a major determinant of local flora, the authors of this
the characteristic inflammation of EoE.13 study hypothesize that airborne antigens could trigger EoE
://
At a biochemical level, the pathogenesis of EoE is driven in these climates. However, a recent meta-analysis found no
by myriad genetic, environmental, and immunologic seasonal variation in EoE incidence, which speaks against
tp
factors.2 Early descriptions of the disease observed an the importance of aeroallergens as triggers.33 To date, there
association with atopic conditions that has been supported is no consensus on the causative role of aeroallegens in
by more recent work, earning the disease its early name, EoE and, outside the clear role of food-borne allergens,
ht
esophageal asthma.1 Put succinctly, EoE appears to be a host there remains much debate about the types of antigens
response to environmental allergens. A T helper (Th)2- that initiate the immunologic cascade of EoE.
cell-mediated immune response involving interleukin
(IL)-13,29 and to a lesser extent IL-4 and IL-5,30 stimulates MANAGEMENT
the allergic response and recruits eosinophils into the Medical management of EoE is challenging due to the
esophagus.13 disease’s chronic nature and its propensity to relapse
Analysis of RNA specimens of patients with biopsy- upon cessation of therapy. Therapeutic options are some-
confirmed EoE demonstrates that IL-13-induced pathways what limited and can be broadly grouped into medical
are major drivers of inflammation. Under the influence of therapy and dietary elimination. There are a number of
an IL-13-induced keratinocyte transcriptome that includes targeted immunotherapies currently under investigation,
eotaxin-3 (a potent recruiter of eosinophils), the esopha- but at present the classes of medication available are
geal tissue experiences an influx of eosinophils, mast cells, corticosteroids, leukotriene inhibitors, mast cell stabilizers,
and lymphocytes. Subsequent epithelial cell hyperplasia, and PPIs. Endoscopic dilation is an option for patients
elongation of papillae, and lamina propria remodeling with strictures. (See Fig. 44.2.)
correlate with the fibrostenotic changes observed grossly.28 The challenge of effectively managing EoE is com-
Histopathologic analysis of biopsy specimens taken from pounded by the lack of a consensus regarding therapeutic
patients with EoE, and esophageal tissue from a mouse endpoints. The most recent ACG guidelines stipulate,
model of EoE, suggests that IL-5-mediated eosinophilia “while complete resolution of symptoms and pathology
484 SECTION I Esophagus and Hernia
is an ideal endpoint, acceptance of a range of reductions 8- or 12-week course. The most common adverse effect
in symptoms and histology is a more realistic and practi- is oral candidiasis, which may occur in up to 20% of
cal goal of clinical practice.”24 The recommendation is patients.2 Consensus guidelines state that the type and
graded as conditional, and the strength of evidence is duration of steroid therapy depends upon the specific
graded as low. The guideline reflects a paucity of data clinical circumstance. More research is required to guide
regarding the degree of reduction in eosinophil density optimum dosing, duration, and potential consequences
required to protect against esophageal injury, and treat- of prolonged use.25
ment endpoints in the literature have shown considerable Systemic steroids are only recommended for cases
variation. Furthermore, symptom-based endpoints have refractory to topical steroids or where a rapid improvement
proven difficult to quantify.24 In spite of these difficulties, in symptoms is needed.24 There is one clinical trial that
some medicines have proven to be effective in obtaining demonstrates the efficacy of oral steroids in children. In
symptomatic relief and histologic improvement. comparison to topical fluticasone, oral prednisone led
As part of the diagnostic work-up for EoE, other causes to a more robust histologic resolution, but it was at the
for esophageal eosinophilia must be excluded and a PPI expense of an increased number of adverse events.37 In
trial must be initiated. Two clinical scenarios exist where general, the side-effect profile of systemic steroid therapy
patients with symptomatic esophageal eosinophilia will makes it a less desirable option.
respond to PPI therapy: patients with reflux causing There is little current evidence to support the use
esophagitis with eosinophilia, and patients without reflux, of other medical therapies for EoE. Some data suggest
t
who for somewhat unclear reasons have an eosinophilia significant symptomatic improvement with the leukotriene
ne
that responds to PPI therapy.24 The latter group falls receptor antagonist montelukast,8 but there is no evi-
under the diagnosis of PPI-REE. To exclude underlying dence that it induces a histologic response.25 There is a
reflux or PPI-REE and thus confirm the diagnosis of EoE, theoretic role for the mast cell stabilizer cromolyn, but
k.
all patients who meet the diagnostic criteria should be no evidence of therapeutic benefit in clinical practice.24
given a 2-month course of PPI therapy.2 The mechanism Current research is ongoing to evaluate the highly selective
oo
of PPI efficacy in PPI-REE remains unclear, but some anti-IL-5 antibodies (mepolizumab and reslizumab), and
in vitro evidence suggests that PPIs may have intrinsic there also may be a role for a monoclonal antibody against
eosinophil-reducing effects. This is due to direct inhibition IL-13, but there is not yet enough evidence to support
of cytokine-stimulated increase in eotaxin-3 mRNA34 and the use of any biologic agent.2
yb
effect is independent of acid reduction effects.
The frequency with which esophageal eosinophilia DIETARY THERAPY
responds to PPI therapy has led some authors to speculate Dietary therapy consists of strategies to remove the
er
on the role of acid-reducing procedures as a treatment of food-borne allergens that trigger EoE.3 There are three
EoE, particularly in children. Early case reports of patients strategies currently accepted as effective treatment for
rg
who underwent fundoplication for refractory reflux EoE in both children and adults.24 The first is conversion
symptoms and were later found to have EoE showed no to elemental or amino-acid-based formula. This method
benefit. Theoretic concern over increasing the incidence completely eliminates all food allergens, but there are
su
of food impaction makes fundoplication potentially more practical limitations in terms of cost and alterations in
risky, but there may be a role for “surgical” management quality of life. The second is targeted elimination of foods
of EoE. Some others posit that more research into the guided by allergy testing. The third is empiric removal
://
role of fundoplication is warranted.35 of the six most common triggers of EoE: soy, egg, milk,
Once a PPI trial has confirmed the diagnosis of EoE, wheat, nuts, and seafood. Multiple retrospective studies and
tp
first-line therapy is swallowed topical steroids, either several meta-analyses have demonstrated the superiority
budesonide or fluticasone, for an initial 8-week trial.24 of the elemental diet, although it is also the most labor
Several randomized controlled trials in children showed intensive and disruptive.2 Targeted elimination using results
ht
significant symptomatic relief and histologic improvement from specific allergy testing have proven to be generally
with topical steroids, with the most robust responses to ineffective, without evidence of widespread clinical or
viscous formulations.36,37 histologic remission.13
In adults, multiple trials have demonstrated histologic In the pediatric population, there is strong evidence
remission with topical corticosteroid therapy.2 A trial of for the effectiveness of dietary approaches. Some trials
swallowed aerosolized fluticasone against placebo in 42 demonstrate efficacy in adults; however, in general, dietary
patients showed a complete histologic response in 62% therapy is less well studied in the adult population.13 The
of patients compared with 0% in the placebo group.38 treatment generally lasts for 4 to 6 weeks, with stepwise
Another randomized, double-blind, placebo-controlled reintroduction of foods after achieving remission.24 Patients
trial examined the efficacy of a budesonide suspension who opt for elemental diets have the longest reintroduction
in 36 patients over 14 years of age. This group found a process. Food groups are generally reintroduced one at
significant reduction in eosinophils from 68.2 to 5.5/HPF a time to facilitate identification of individual triggers.2
in the treatment arm, compared with an insignificant Current recommendations call for clinical assessment
reduction in the placebo group. Furthermore, patients and endoscopy with biopsy to monitor inflammatory
had significantly improved dysphagia scores and many response whenever foods are being withdrawn or reintro-
patients had resolution of endoscopic findings.39 duced to the diet.24 Clearly dietary elimination and allergen
In general, topical corticosteroids are well tolerated. identification is incredibly time and labor intensive. It is
There is no evidence of adrenal axis suppression after an also quite costly. A multidisciplinary approach involving the
Nonreflux Esophagitis CHAPTER 44 485
ENDOSCOPIC DILATION
In patients with severe fibrostenotic complications of EoE,
including focal stricture and narrow-caliber esophagus,
endoscopic dilation is an effective treatment. Current
recommendations call for dilation in symptomatic patients
with strictures that have persisted in spite of medical or
dietary therapy, or in patients with “severely symptomatic
esophageal stenosis.”24
Esophageal dilation was one of the original therapies
for symptomatic EoE, and dilation produces long-lasting
symptom relief.40 Dilation was initially associated with high
complication rates; in particular, rates of perforation in
some early series were as high as 8%.13 A more recent
meta-analysis from 2011 demonstrated symptomatic relief
in 92% of patients for up to 1 to 2 years after dilation
t
with a very low rate of true perforation (<0.1%).41 It is
ne
important to note that the symptomatic improvement
following dilation has no effect on the eosinophil-induced FIGURE 44.3 White exudates characteristic of Candida esophagitis.
(From Rosolowski M, Kierzkiewicz M. Etiology, diagnosis and
inflammation.13
k.
treatment of infectious esophagitis. Prz Gastroenterol.
Current guidelines advocate for small increases in
2013;8:333–337.)
esophageal diameter over multiple sessions, but do not
oo
make recommendations on dilation technique.24 Regard-
less of the specific technique, whether wire-guided or
non-wire-guided bougie or through-the-scope balloons, pseudomembranes causing esophageal stenosis. The find-
ings can be graded on a scale of one to four.8 Endoscopic
yb
patients must be counseled on the expected postopera-
tive pain associated with dilation13 and the potential for evaluation has a sensitivity of 100% and a specificity of
complication.25 83%.8 Biopsy, which is not required for diagnosis, shows
yeast and pseudohyphae invading mucosal cells.42
er
Fungal, viral, bacterial, and parasitic causes of infectious evaluation is appropriate and symptoms should resolve in
esophagitis have all been described.42 On the whole, infec- 3 to 5 days. If there is no timely resolution of symptoms,
tious esophagitis is rare and tends to occur in patients with endoscopy with biopsy is needed due to the high incidence
su
antibiotic use.43 The prognosis is generally tied to the if the disease is severe enough to limit intake by mouth.
severity of comorbid conditions.8 Regardless of the causative The current recommendation is for oral fluconazole with
tp
agent, infectious esophagitis usually presents with acute a loading dose of 400 mg followed by 200 to 400 mg daily.
onset of symptoms such as dysphagia or odynophagia.42 In refractory cases, other azoles may be used; echinocan-
dins or amphotericin B are also acceptable alternatives.
ht
t
acceptable alternative but has been more extensively FIGURE 44.4 Herpetic ulcers in herpes simplex virus esophagitis.
ne
studied in CMV retinitis.50 For ganciclovir-resistant disease, (From Gurvits GE, Shapsis A, Lau N, Gualtieri N, Robilotti JG.
other antiviral therapies include cidofovir and foscarnet. Acute esophageal necrosis: a rare syndrome. J Gastroenterol.
CMV immunoglobulin, in combination with antiviral 2007;42:29–38.)
k.
medications or as a monotherapy, is not currently indicated
due to heterogeneity among studies evaluating its use.48 individual cases suggest that the condition occurs most
oo
often in immunocompromised individuals.1 TE can occur
HERPES ESOPHAGITIS as a result of direct extension of tuberculosis from contigu-
Herpes esophagitis, caused by the herpes simplex virus ous mediastinal structures. Symptoms of dysphagia or
yb
(HSV), is typically seen in immunocompromised patients. odynophagia may result, and ulceration or fistula formation
It usually occurs as an opportunistic infection secondary to may be evident on endoscopy. The diagnosis should be
viral reactivation in patients with underlying HIV infection, considered in patients with esophageal fistulas or sinus
malignancy, immunosuppressive therapy, or severe illness.51 tracts on endoscopy; reports describe the development
er
It occurs less frequently than CMV esophagitis and the of bronchoesophageal and tracheoesophageal fistulas.8
two viral infections may coexist.8 There are case reports The differential diagnosis of esophageal fistulas must
rg
of HSV esophagitis in immunocompetent hosts, but these include Crohn disease and esophageal carcinoma; work-up
are exceedingly rare; in these patients the disease appears should include bronchoscopy, endoscopy with biopsy,
to be self-limited.51 Within the last few years there have and CT scan. Histologic features may include caseat-
su
been several published reports of acute herpes esophagitis ing and noncaseating granulomas in combination with
in immunocompetent patients with EoE; many of these chronic inflammation and scarring. Acid-fast bacilli may
cases were identified in the absence of steroid therapy. be recognized on Ziehl–Neelsen staining.1
://
The coincidence of these two rare conditions raises the Most cases of fistulas require surgical resection with
possibility of a causal relationship. antituberculous treatment, although recent reports dem-
tp
Patients generally present with symptoms of odynopha- onstrate that endoscopic intervention or nonoperative
gia, dysphagia, retrosternal chest pain, or fever, and there management may be effective in select patients.53,54 One
is often a recent history of upper respiratory infection or case series, albeit with just two patients, demonstrated effi-
ht
orolabial lesions.1,51 On endoscopy, lesions are composed cacy of fistula resection with pedicled pleural flaps securing
of small vesicles, 1 to 3 mm in diameter, which slough to suture lines.55 Another case series of six Chinese patients
leave well-circumscribed ulcers with discrete edges (Fig. with dysphagia caused by TE demonstrated the efficacy of
44.4).8 The lesions are most commonly found in the surgical management. In this series, five patients underwent
mid- to lower esophagus.51 Biopsies of herpes esophagitis video-assisted thoracoscopic surgery (VATS) with excision
is similar to herpesvirus infection at other sites; cytopathic or enucleation and drainage of large mediastinal lymph
effects include enlarged and multinucleated cells with nodes, and one patient with an esophagopleural fistula
marginated chromatin and nuclear molding. Intranuclear underwent thoracotomy with débridement and fistula
inclusions may be large, eosinophilic and glassy, or powdery repair coverage with a diaphragmatic muscle flap. These
and homogeneous. Optimal histologic diagnosis requires patients had uneventful postoperative courses, with relief
sampling of ulcer edges, because the virus infects squamous from dysphagia at the time of discharge.52
cells of intact epithelium.1 Pharmacologic management is
with antiviral medicines, and acyclovir is frequently used.51 DRUG-INDUCED ESOPHAGITIS
TUBERCULOUS ESOPHAGITIS Drug-induced esophagitis, or “pill esophagitis,” occurs
Tuberculous esophagitis (TE) is rare, even in countries when direct esophageal injury occurs as a result of an
where tuberculosis is relatively common.52 Reports of ingested medicine.1 Since the original description in
Nonreflux Esophagitis CHAPTER 44 487
1970, over 100 different medicines have been reported may be managed with dietary modifications and local
to cause drug-induced esophagitis.56 The most commonly anesthetics; supportive care with calorie supplementation
implicated medicines include bisphosphonates, nonste- or intravenous hydration may be required if adequate
roidal antiinflammatory drugs (NSAIDs), tetracyclines, intake by mouth becomes difficult. Grade 3 esophagitis
vitamin C, and potassium chloride tablets.8 Patients usually generally requires inpatient management and some
present with retrosternal pain or heartburn, less commonly patients require enteric access with a gastrostomy tube
odynophagia or dysphagia. The pathogenesis is linked to to ensure adequate hydration and nutrition.
systemic and local actions of the ingested medicine, and The risk of stricture formation correlates with the grade
the injury is most likely to occur at sites of anatomic or of esophagitis. In patients receiving treatment for SCLC,
pathologic narrowing.57 24% of patients with grade 3 esophagitis developed an
On endoscopy, drug-induced esophagitis most com- esophageal stricture compared with just 2% of patients
monly causes ulcers (82% of patients) with occasional with grade 1 or 2 disease.62 In both adults and children,
bleeding (24% of patients); erosions are less common esophageal strictures are generally amenable to endoscopic
(18% of patients) and strictures are rare (3%).58 Histologic dilation.65
features are generally nonspecific; they range from tiny, Both anterograde and retrograde approaches have been
punctate erosions to large circumferential ulcers with shown to be safe and effective; regardless of the technique,
granulation tissue and fibrinopurulent exudate.1 Specific patients usually require multiple dilations. 64 Dilation
medicines tend to cause consistent patterns of injury; can be successful even in cases of complete stricture.64
t
for example, the polarizable crystalline foreign material Surgical intervention is generally not required; cases of
ne
characteristic of alendronate causes a histiocytic giant-cell iatrogenic esophageal perforation following attempted
reaction.1 dilation may be managed conservatively. The strongest
In the majority of cases, pill esophagitis is self-limiting predictor for failure of endoscopic dilation appears to
k.
and resolves without complications. The key to prompt be time to onset of esophageal stricture.65
resolution is correct and timely diagnosis with removal
oo
of the causative agent whenever possible.57 Symptoms
generally improve in 7 to 10 days. Rarely, cases may be
ACUTE ESOPHAGEAL NECROSIS
complicated by significant bleeding, which may require AEN is a rare but potentially lethal disease; it likely
endoscopic intervention and local epinephrine injection.58
yb
represents a final common pathway for a number of
Endoscopic dilation may be required in cases of stricture conditions that can lead to severe esophageal injury. 66
formation; in rare circumstances these can take weeks The pathogenesis is believed to be multifactorial and
to improve without dilation. Surgery is reserved for the an ischemic phenomenon is likely the major driver of
er
management of complications that may develop during the esophageal lesions. Hypoperfusion due to shock,
treatment.57 atherosclerosis, thromboembolic disease, and cardiac
rg
Radiation esophagitis is extremely common in patients are known associations with alcoholism, cocaine abuse,
who receive radiation therapy to the thorax, head, or malnutrition, malignancy, and general debilitation.67
neck region.1 It occurs in adults and children.59,60 Some Patients with AEN are generally quite ill, with myriad
://
authors refer to esophagitis as an “inevitable” result of comorbid conditions. The most common presenting
radiotherapy and concurrent chemotherapy for thoracic symptoms are hematemesis and/or melena (71% to 90%
tp
cancer61; the associated dysphagia, odynophagia, and pain of cases).66,67 On endoscopy, the disease is characterized
represent the major dose-limiting toxicity of treatment.62 by a circumferential black appearance, giving the condi-
Given the frequency of symptomatic esophagitis in patients tion its alternate name of “black esophagus” (Fig. 44.5).
ht
receiving therapeutic radiation, the National Cancer The mortality is high, but depends upon the comorbid
Institute has published a common terminology criterion conditions, and reports range from 15% to 36%.66
for adverse events that grade dysphagia and esophagitis The treatment is generally supportive, and is directed
on a scale from 1 to 5.63 at treating underlying medical conditions and maximizing
Radiotherapy causes congestion, edema, and erosion tissue perfusion. This includes aggressive resuscitation,
of the esophageal mucosa. This damage likely occurs due optimization of acid suppression, and treatment with
to obliterative endarteritis and microvascular damage antibiotics if sepsis is present. Surgical intervention is
that renders the esophageal mucosa ischemic, leading warranted in cases where AEN progresses to perfora-
to fibrosis.64 There appears to be a sensitization effect of tion, or in the presence of mediastinitis or a mediastinal
chemotherapy. Symptoms of dysphagia, odynophagia, and abscess. If these conditions arise, surgery should be
substernal pain are extremely common; the incidence of pursued expediently. Patients who survive are at risk for
esophagitis is as high as 40% in head/neck cancer patients, esophageal strictures, which may occur as early as 1 week
and the incidence of grade 1 to 2 acute esophagitis was after the initial diagnosis. These strictures may require
81% in patients undergoing concurrent radiation and serial endoscopy with dilation.67 In our experience, the
chemotherapy for small-cell lung cancer (SCLC). resulting strictures are frequently long segment, and
The management of radiation esophagitis depends endoscopic dilation is less likely to be effective. Surgical
upon the grade. Low-grade esophagitis (grade 1 to 2) management is often necessary.
488 SECTION I Esophagus and Hernia
16. Dellon ES, Jensen ET, Martin CF, Shaheen NJ, Kappelman MD.
The prevalence of eosinophilic esophagitis in the United States.
Clin Gastroenterol Hepatol. 2014;12(4):589-596.
17. Francios JP, Tam V, Liacouras CH, Spergel JM. A case-control
study of sociodemographic and geographic characteristics of 335
children with eosinophilic esophagitis. Clin Gastroenterol Hepatol.
2009;7(4):415-419.
18. Roy-Ghanta S, Larosa DF, Katzka DA. Atopic characteristics of adult
patients with eosinophilic esophagitis. Clin Gastroenterol Hepatol.
2008;6(5):531-535.
19. Ricker J, McNear S, Cassidy T, et al. Routine screening for eosino-
philic esophagitis in patients presenting with dysphagia. Therap Adv
Gastroenterol. 2011;4(1):27-35.
20. Noel RJ, Putnam PE, Rothenberg ME. Eosinophilic esophagitis. N
Engl J Med. 2004;351(9):940-941.
21. Prasad GA, Alexander GA, Schleck CD, et al. Epidemiology of
eosinophilic esophagitis over 3 decades in Olmsted County, Min-
nesota. Clin Gastroenterol Hepatol. 2009;7(10):1055-1061.
22. Hruz P, Straumann A, Bussmann C, et al. Escalating incidence
of eosinophilic esophagitis: a 20-year prospective, population-
based study in Olten County, Switzerland. J Allergy Clin Immunol.
2011;128(6):1349-1350.
t
23. Okada H, Kuhn C, Feillet H, Back JF. The ‘Hygiene Hypothesis’
ne
for autoimmune and allergic disease: an update. Clin Exp Immunol.
2010;160:1-9.
FIGURE 44.5 Characteristic “black esophagus” in acute 24. Dellon ES, Gonsalves N, Hirano I, Furuta GT, Liacouras CA, Katzka
esophageal necrosis. (From Shafa S, Sharma N, Keshishian J, DA. ACG clinical guideline: evidence based approach to the diagnosis
k.
Dellon ES. The black esophagus: a rare but deadly disease. ACG and management of esophageal eosinophilia and eosinophilic
Case Rep J. 2016;3:88–91.) esophagitis (EoE). Am J Gastroenterol. 2013;108:679-692.
25. Liacouras CA, Furuta GT, Hirano I, et al. Eosinophilic esophagitis:
oo
updated consensus recommendations for children and adults.
J Allergy Clin Immunol. 2011;128(1):3-20.
REFERENCES 26. Molina-Infante J, Ferrando-Lamana L, Ripoll C, et al. Esophageal
yb
eosinophilic infiltration responds to proton pump inhibition in
1. Almashat SJ, Duan L, Goldsmith JD. Non-reflux esophagitis: a most adults. Clin Gastroenterol Hepatol. 2011;9(2):110-117.
review of inflammatory diseases of the esophagus exclusive of reflux 27. Lucendo AJ, Pascual-Turrion JM, Navarro M, et al. Endoscopic,
esophagitis. Semin Diagn Pathol. 2014;31:89-99. bioptic, and manometric findings in eosinophilic esophagitis before
er
2. Kavitt RT, Hirano I, Vaezi MF. Diagnosis and treatment of eosino- and after steroid therapy: a case series. Endoscopy. 2007;39:765-771.
philic esophagitis in adults. Am J Med. 2016;129:924-934. http:// 28. Lucendo AJ, Castillo P, Martin-Chavarri S, et al. Manometric find-
dx.doi.org/10.1016/j.amjmed.2016.04.024. ings in adult eosinophilic oesophagitis: a study of 12 cases. Eur J
rg
3. Attwood S, Furuta GT. Eosinophilic esophagitis: historical perspective Gastroenterol Hepatol. 2007;19(5):417-424.
on an evolving disease. Gastroenterol Clin North Am. 2015;43(2):185-199. 29. Blanchard C, Mingler MK, Vicario M, et al. IL-13 involvement in
4. Picus D, Frank PH. Eosinophilic esophagitis. AJR Am J Roentgenol. eosinophilic esophagitis: transcriptome analysis and reversibility
su
an emerging problem with unique esophageal features. Gastrointest 32. Hurrell JM, Genta RM, Dellon ES. Prevalence of esophageal eosino-
Endosc. 2004;59(3):335-361. philia varies by climate zone in the United States. Am J Gastroenterol.
8. Attwood SEA, Lamb CA. Eosinophilic oesophagitis and other non- 2012;107(5):698-706.
ht
reflux inflammatory conditions of the oesophagus: diagnostic imaging 33. Lucendo AJ, Arias A, Redondo-Gonzalez O, Gonzalez-Cervera J.
and management. Best Pract Res Clin Gastroenterol. 2008;22(4):639-660. Seasonal distribution of initial diagnosis and clinical recrudescence
9. Remedios M, Campbell C, Jones DM, Kerlin P. Eosinophilic of eosinophilic esophagitis: a systematic review and meta-analysis.
esophagitis in adults: clinical, endoscopic, histologic findings, and Allergy. 2015;70:1640-1650.
response to treatment with fluticasone propionate. Gastrointest Endosc. 34. Cheng E, Zhang X, Huo X, et al. Omeprazole blocks eotaxin-3 expres-
2006;63:3-12. sion by oesophageal squamous cells from patients with eosinophilic
10. Desai TK, Stecevic V, Chang CH, Goldstein NS, Badizadegan K, oesophagitis and GORD. Gut. 2013;62:824-832.
Furuta GT. Association of eosinophilic inflammation with esophageal 35. Rea F, Caldara T, Tambucci R, et al. Eosinophilic esophagitis: is it
food impaction in adults. Gastrointest Endosc. 2005;61:795-801. also a surgical disease? J Pediatr Surg. 2013;48:304-308.
11. Prasad GA, Talley NJ, Romero Y, et al. Prevalence and predictive 36. Aceves SS, Bastian JF, Newbury RO, Dohil R. Oral viscous budesonide:
factors of eosinophilic esophagitis in patients presenting with a potential new therapy for eosinophilic esophagitis in children. Am
dysphagia: a prospective study. Am J Gastroenterol. 2007;102:2627-2632. J Gastroenterol. 2007;102:2271-2279.
12. Kapel RC, Miller JK, Torres C, Aksoy S, Lash R, Katzka DA. Eosino- 37. Schaefer ET, Fitzgerald JF, Molleston JP, et al. Comparison of oral
philic esophagitis: a prevalent disease in the United States that affects prednisone and topical fluticasone in the treatment of eosinophilic
all age groups. Gastroenterology. 2008;134:1316-1321. esophagitis: a randomized trial in children. Clin Gastroenterol Hepatol.
13. Dellon ES, Liacouras CA. Advances in clinical management of 2008;6(2):165-173.
eosinophilic esophagitis. Gastroenterology. 2014;147(6):1238-1254. 38. Alexander JA, Jung KW, Arora AS, et al. Swallowed fluticasone
14. Dellon ES. Epidemiology of eosinophilic esophagitis. Gastroenterol improves histologic but not symptomatic response of adults with
Clin North Am. 2014;43(2):201-218. eosinophilic esophagitis. Clin Gastroenterol Hepatol. 2012;10(7):742-749.
15. Dellon ES, Kim HP, Sperry SLW, Rybnicek DA, Woosley JT, Shaheen 39. Staumann A, Conus S, Degen L, et al. Budesonide is effective in
NJ. A phenotypic analysis shows eosinophilic esophagitis is a progres- adolescent and adult patients with active eosinophilic esophagitis.
sive fibrostenotic disease. Gastrointest Endosc. 2014;79(4):577-585. Gastroenterology. 2010;139:1526-1537.
Nonreflux Esophagitis CHAPTER 44 489
40. Schoepfer AM, Gonsalves N, Bussmann C, et al. Esophageal dilation 54. Catano J, Cardeno J. Perforated tuberculosis lymphadenitis. Am J
in eosinophilic esophagitis: effectiveness, safety, and impact on the Trop Med Hyg. 2013;88(6):1009-1010.
underlying inflammation. Am J Gastroenterol. 2010;105:1062-1070. 55. Ramo OJ, Salo JA, Isolauri J, Luostarinen M, Mattila SP. Tuberculous
41. Bohn ME, Richter JE. Review article: oesophageal dilation in fistula of the esophagus. Ann Thorac Surg. 1996;62:1030-1032.
adults with eosinophilic oesophagitis. Aliment Pharmacol Ther. 56. Abid S, Mumtaz K, Jafri W, et al. Pill-induced esophageal injury: endo-
2011;33:748-757. scopic features and clinical outcomes. Endoscopy. 2005;37(8):740-744.
42. Rosolowski M, Kierzkiewicz M. Etiology, diagnosis and treatment of 57. Zografos GN, Georgiadou D, Thomas D, Kaltsas G, Digalakis M.
infectious esophagitis. Prz Gastroenterol. 2013;8(6):333-337. Drug-induced esophagitis. Dis Esophagus. 2009;22:633-637.
43. Underwood JA, Williams JW, Keat RF. Clinical findings and risk factors 58. Kim SH, Jeong JB, Kim JW, et al. Clinical and endoscopic
for Candida esophagitis in outpatients. Dis Esophagus. 2006;16(2):66-69. characteristics of drug-induced esophagitis. World J Gastroenterol.
44. Mimidis K, Papadopoulos V, Margaritis V, et al. Predisposing factors 2014;20(31):10994-10999.
and clinical symptoms in HIV-negative patients with Candida oesopha- 59. Lar DR, Foroutan HR, Su WT, Wolden SL, Boulad F, La Quaglia MP.
gitis: are they always present? Int J Clin Pract. 2005;59(2):210-213. The management of treatment-related esophageal complication in
45. Samonis G, Skordilis P, Maraki S, et al. Oropharyngeal candidiasis children and adolescents with cancer. J Pediatr Surg. 2006;41:495-499.
as a marker for esophageal candidiasis in patients with cancer. Clin 60. Werner-Wasik M. Treatment-related esophagitis. Semin Oncol.
Infect Dis. 1998;27(2):283-286. 2005;32(3):S60-S66.
46. Kanda N, Yasuba H, Takahashi T, et al. Prevalence of esophageal 61. Yu Y, Guan H, Done Y, Xing L, Li X. Advances in dosimetry and
candidiasis among patients treated with inhaled fluticasone propio- biological predictors of radiation-induced esophagitis. Onco Targets
nate. Am J Gastroenterol. 2003;98(10):2146-2148. Ther. 2016;9:597-603.
47. Pappas PG, Kauffman CA, Andes D, et al. Clinical practice guidelines 62. Grant JD, Shirvani SM, Tang C, et al. Incidence and predictors
for the management of candidiasis: 2009 update by the Infectious of severe acute esophagitis and subsequent esophageal stricture
Diseases Society of America. Clin Infect Dis. 2009;48:503-535. in patients treated with accelerated hyperfractionated chemora-
t
48. Baroco AL, Oldfield EC. Gastrointestinal cytomegalovirus disease diation for limited-stage small cell lung cancer. Pract Radiat Oncol.
ne
in the immunocompromised patient. Curr Gastroenterol Rep. 2015;5:e383-e391.
2008;10:409-416. 63. Cancer Therapy Evaluation Program, Common Terminology Criteria
49. Ozaki T, Yamashita H, Kaneko S, et al. Cytomegalovirus disease of for Adverse Events, Version 3.0, DCTD, NCI, NIH, DHHS; 2003.
the upper gastrointestinal tract in patients with rheumatic disease: a http://ctep.cancer.gov. Accessed 9 August 2006.
k.
case series and literature review. Clin Rheumatol. 2013;32:1683-1690. 64. Francis DO, Hall E, Dang JH, Vlacich GR, Netterville JL, Vaezi
50. Martin D, Sierra-Madero J, Walmsley S, et al. A controlled trial of MF. Outcomes of serial dilation for high grade radiation-related
valganciclovir as induction therapy for cytomegalovirus retinitis. N esophageal strictures in head and neck cancer patients. Laryngoscopy.
oo
Engl J Med. 2002;346:1119-1126. 2015;125(4):856-862.
51. Canalejo E, Duran FG, Cabello N, Martinez JG. Herpes esophagitis 65. Tuna Y, Kocak E, Dincer D, Koklu S. Factors affecting the success
in healthy adults and adolescents: report of 3 cases and review of of endoscopic bougie dilation of radiation-induced esophageal
yb
the literature. Medicine (Baltimore). 2010;89:204-210. stricture. Dig Dis Sci. 2012;57:424-428.
52. Ni B, Lu X, Gong Q, et al. Surgical outcome of esophageal tubercu- 66. Shafa S, Sharma N, Keshishian J, Dellon ES. The black esophagus:
losis secondary to mediastinal lymphadenitis in adults: experience a rare but deadly disease. ACG Case Rep J. 2016;3(2):88-91.
from a single center in China. J Thorac Dis. 2013;5(4):498-505. 67. Gurvits GE, Shapsis A, Lau N, Gualtieri N, Robilotti JG. Acute
er
53. Manca S, Fois AG, Santoru L, et al. Unusual clinical presentation esophageal necrosis: a rare syndrome. J Gastroenterol. 2007;42:29-38.
of thoracic tuberculosis: the need for a better knowledge of illness.
Am J Case Rep. 2015;16:240-244.
rg
su
://
tp
ht
CHAPTER
E
sophageal duplication cyst is one of two types of CLINICAL PRESENTATION
foregut duplication cysts. The other type is the
bronchogenic duplication cyst, and they are both The presentation of an esophageal duplication cyst
classified together due to their common embryologic depends on its size, location, and effect on surrounding
origin. Duplication cysts are a rare entity, with most structures. It is also related to the age of the patient.
literature consisting of case reports or small case series. For instance, infants and children often present with
It is estimated that foregut duplication cysts make up respiratory distress, cough, or recurrent pneumonias.8,9
20% of all gastrointestinal duplication cysts. Esophageal Although the majority of adults are asymptomatic, present-
t
duplication cysts are the second most common benign ing symptoms include progressive dysphagia to solids and
ne
posterior mediastinal lesion in children after bronchogenic liquids, epigastric or abdominal pain, and retrosternal chest
cyst.1 A review of almost 50,000 autopsies revealed an discomfort.8,10-13 Physical examination is often noncontribu-
incidence of esophageal duplication cyst of 1 in 8200, tory. There are isolated reports of acute, severe abdominal
k.
with 60% occurring in the lower third esophagus, 17% pain in patients who present with cyst perforation or
in middle third, and 23% in upper third.2 There is a hemorrhage.3,8
oo
male predominance in a ratio of 2 : 1, and duplication There are also rare cases of patients presenting weight
cyst has been associated with congenital abnormalities loss and lymphadenopathy due to malignancy arising from
such as small intestinal duplication, esophageal atresia, within the cyst.14-16
and spinal abnormalities.3
yb
Malignant transformation within an esophageal duplica-
tion cyst is an extremely rare event, with only a handful of
reported cases.14-17 The age at presentation ranged widely
PATHOPHYSIOLOGY from 18 to 60 years old, with no gender predominance and
er
During the fifth to eighth week of fetal life, as the variable size from 3 to 10 cm. The clinical presentation
esophagus elongates, its epithelium grows and the cells of these malignancies varied from an incidental finding
rg
obliterate the lumen of the esophagus. The esophagus to dysphagia, fever, and pain.
produces secretions, which form into vacuoles. If the The differential diagnosis for a submucosal esophageal
vacuoles persist because of failure of proper alignment lesion includes other nonepithelial tumors such as leio-
su
and coalescence, duplication cysts may be formed. The cyst myoma, gastrointestinal stromal tumor (GIST), sarcoma,
often occurs on the right because of the elongation of the lymphoma, lipoma, and other posterior mediastinal masses.
viscera and dextrorotation of the stomach.2 Esophageal For intraabdominal esophageal duplication cyst the dif-
://
duplication cysts are more commonly of the cystic form ferential diagnosis should include pancreatic pseudocyst,
(80%), which has no communication with the lumen of dermoid cyst, cystadenoma, and cystadenocarcinoma.5
tp
ABSTRACT
Esophageal duplication cysts are rare entities with most
literature comprising case reports or small case series.
There is a spectrum of clinical presentation. Children
tend to present with symptoms while adults tend to be
asymptomatic. Work-up includes a CT or MRI. Esopha-
gogastroduodenoscopy and endoscopic ultrasound are
also useful. Esophageal duplication cysts have a small
chance of malignant transformation. We recommend
surgical management, which we will review. However, we
acknowledge that each case should be considered on an
individual basis.
KEYWORDS
Esophageal duplication cyst, diagnosis, management,
endoscopic ultrasound, malignant transformation
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
Esophageal Duplication Cyst CHAPTER 45 491
In rare cases, duplication cysts have been diagnosed may need to be decompressed to facilitate dissection
using technetium 99m to detect ectopic gastric mucosa from the esophagus and surrounding structures and
in the adult patient, or ultrasound in the prenatal (third intercostal extirpation. The vagus nerves should be
trimester) period.20 The use of prenatal ultrasound to identified and preserved. The operation is similar to
detect esophageal duplication cyst has not been commonly enucleation of a submucosal tumor of the esophagus.
reported, and other infants with esophageal duplication There can be significant adhesions to adjacent structures
cysts have had normal prenatal ultrasounds.21 and, more importantly, to the underlying mucosa of the
esophagus.
ENDOSCOPY Patients should be informed of the risk and potential
Esophagogastroduodenoscopy reveals a submucosal lesion. implications of esophageal injury as a result of cyst resec-
Depending on the size of the lesion, they can cause nar- tion. This significant complication has been previously
rowing of the esophageal lumen as well as signs of extrinsic reported and is typically managed with immediate primary
compression.22 In cases of malignant transformation an repair with or without soft tissue buttressing (e.g., inter-
associated esophageal stricture may be present.15 Since the costal muscle, pedicled pericardial fat pad–Brewer’s patch,
mid-1990s there has been increasing use of endoscopic omentum, etc.). In some instances, successful management
ultrasound (EUS). EUS has been useful to determine of a perforation at the time of resection has required
if the lesion is intramural or extramural and cystic or esophagectomy.5,26 If possible, the esophageal muscularis
solid.4 On EUS, there is usually a three- to five-layer cyst propria should be reapproximated to prevent the forma-
t
wall with anechoic or hypoechoic, homogeneous internal tion of a pseudodiverticulum, which may require further
ne
contents. The lesion has regular margins and arises from surgical correction.7,13 Before reapproximation of the
the submucosal layer or extrinsic to the gut wall.23 However, muscularis, the mucosal integrity at the repair site should
in a case series of 19 mediastinal cysts, the EUS find- always be assessed. This is preferably accomplished with
k.
ings were not as consistent.24 EUS with a 12-MHz probe the flexible esophagoscope, which allows for both direct
described the presence of both hypoechoic and anechoic mucosal examination and insufflation of the esophageal
oo
cysts with and without the presence of cell wall layers on lumen to test for leakage while the posterior mediastinum
EUS. Three of the patients without a visible cystic wall is submerged under irrigation fluid.
on EUS underwent surgical resection that histologically Thoracoscopic resection may be associated with a
shorter hospital stay.7 Incomplete resection can lead to
yb
confirmed the diagnosis of esophageal duplication cyst. The
others without cell wall layers were managed conservatively recurrence.11 Resection of cysts near the gastroesophageal
and classified as nonspecific simple cysts. The role of junction may result in incompetence of the lower esopha-
EUS–fine-needle aspiration (FNA) is controversial because geal sphincter and the development or exacerbation of
er
of the risk of infection. In the aforementioned case series, gastroesophageal reflux symptoms. This can usually be
four patients developed infections following endoscopic managed medically.13
rg
drainage, one of which was life-threatening. FNA should For the rare cases of malignant transformation, and
be avoided when the lesion is clearly cystic in nature (i.e., locally advanced disease, patients should be treated on
anechoic). However, if the lesion is hypoechoic, FNA can a case-by-case basis. In our review, patients with locally
su
be considered after appropriate prophylactic antibiotics advanced disease received adjuvant chemotherapy and
have been administered.4,24 radiation.16 If they were not considered surgical candidates
because of morbidity of resection or distant metastasis,
://
ment of esophageal duplication cyst. Fine-needle aspiration worsening cough associated with chest pain. She had a past
under EUS guidance represents the least invasive treatment medical history of asthma. A chest x-ray showed a posterior
option for these patients. Although major surgery may be mediastinal mass (Fig. 45.1). The chest CT scan revealed
avoided, there should be a discussion about the potential an 8.5 × 4.5 cm homogeneous, well-circumscribed posterior
for recurrence, and of the fact that this treatment modality mediastinal mass (Figs. 45.2 to 45.5). An esophagogastro-
cannot prevent potential complications, such as bleeding, duodenoscopy and EUS confirmed extrinsic compression
ulceration, perforation, or malignant transformation, of the esophagus from 30 to 36 cm from the level of
especially in young individuals.18,22,25 the incisors. The mass was extrinsic to the wall of the
esophagus on EUS. The patient underwent thoracoscopic
OPERATIVE enucleation of the esophageal duplication cyst (Fig. 45.6).
Surgical resection is often recommended because of The muscularis of the esophagus was repaired with a
the risks of ulceration, hemorrhage, perforation, and monofilament suture (Figs. 45.7 and 45.8). Intraoperative
the small risk of malignant transformation associated endoscopy confirmed the absence of any mucosal injury.
with esophageal duplication cyst. As opposed to simple The final pathology report was consistent with the pre-
observation, surgery provides a definitive diagnosis and operative diagnosis. Follow-up barium swallow at 1 and 6
decreases the frequency of follow-up investigations. years revealed a normal-appearing esophagus (Fig. 45.9).
Complete surgical excision can be safely conducted via She has developed no postoperative complications at
open or thoracoscopic approach.7,10,12,13 The cystic lesion 6 years.
492 SECTION I Esophagus and Hernia
Duplication cyst
t
ne
200 mm
Duplication cyst
k.
oo
FIGURE 45.1 Frontal and lateral radiographic projections of the chest showing a large round retrocardiac mass with smooth interface with
the adjacent lung. Differential includes esophageal solid mass and hiatal hernia. The normal position of the gastric fundus is below the
diaphragm. The lack of air fluid level makes hiatal hernia less likely on the differential.
yb
er
rg
su
://
tp
Left atrium
ht
Aorta
Duplication
cyst
Duplication cyst
Gastric fundus
FIGURE 45.2 Maximum intensity projection (MIP) 12-mm slab images in coronal and sagittal plane showing the esophageal duplication
cyst in relation to the posterior mediastinal structures (esophagus, left atrium, aorta, thoracic spine) and the diaphragm, and gastric
fundus. Images obtained in the pulmonary circulatory vascular phase after injection of intravenous contrast on Toshiba Aquilion One
(Toshiba Canada, Markham, ON, Canada) with axial volume acquisition at 0.625 mm and postprocessing reconstruction on Terarecon
Aquarius Intuition software (Terarecon, Foster City, California).
Esophageal Duplication Cyst CHAPTER 45 493
Duplication cyst
Duplication cyst
t
Esophagus
ne
k.
FIGURE 45.3 Volume-rendered computed tomography images with half volume coronal and oblique sagittal cut views showing the
esophageal duplication cyst in relation to the posterior mediastinal structures (esophagus, left atrium, aorta, thoracic spine) and the
oo
diaphragm, and gastric fundus. Lower density structures (lungs, fat) projected as a color mask for emphasis. Images obtained in the
pulmonary circulatory vascular phase after injection of intravenous contrast on Toshiba Aquilion One (Toshiba Canada, Markham, Ontario,
Canada) with axial volume acquisition at 0.625 mm and postprocessing reconstruction on Terarecon Aquarius Intuition software
(Terarecon, Foster City, California).
yb
er
SUMMARY
rg
Esophagus
elderly patient with a small asymptomatic cyst, a period
of observation with serial imaging to ensure stability may
be a reasonable approach after detailed discussion of the
potential complications related to esophageal duplication
120 mm
cysts.
Of all treatment options reviewed herein, we favor
surgical excision because it is the best modality to obtain
a definitive diagnosis and because it probably has the best
FIGURE 45.4 Axial 5-mm computed tomography image showing likelihood of preventing long-term complications and
the esophageal duplication cyst in relation to the posterior recurrence. Esophageal duplication cyst is a relatively
mediastinal structures (esophagus, left atrium, aorta, thoracic rare condition and, as a result, the scientific evidence
spine). Images obtained in the pulmonary circulatory vascular to support our recommendation is and will likely
phase after injection of intravenous contrast on Toshiba Aquilion remain limited. In that light, we acknowledge that it is
One (Toshiba Canada, Markham, Ontario, Canada) with axial appropriate to individualize management decisions on a
volume acquisition at 0.625 mm and postprocessing case-by-case basis.
reconstruction on Terarecon Aquarius Intuition software
(Terarecon, Foster City, California).
494 SECTION I Esophagus and Hernia
Esophagus
Esophagus
Aorta
Aorta
120 mm 120 mm
t
ne
Duplication cyst
k.
Esophagus
oo
Aorta
120 mm
yb
FIGURE 45.5 Axial T2 turbo spin echo (TSE) 5-mm, axial T1 TSE 5-mm, and axial T1 three-dimensional volumetric interpolated breath-
er
hold examination fat-saturated (3D VIBE FS) postcontrast images showing the cyst in the same plane as the accompanying axial
computed tomography scan. High signal intensity (SI) on T2 and low SI on T1 are consistent with water or near-water cystic contents.
rg
On postcontrast image there is a thin enhancing wall with no solid nodular components. Images obtained on a Siemens Symphony 1.5T
magnet (Siemens Healthcare, Erlangen, Germany).
su
Right lung
://
Muscularis propria
Right inferior
ht
pulmonary vein
Mucosa
Muscularis edges
REFERENCES
Azygous vein 1. Watanobe I, Ito Y, Akimoto E, et al. Laparoscopic resection of an
intra-abdominal esophageal duplication cyst: a case report and
Bronchus literature review. Case Rep Surg. 2015;2015:940768.
intermedius 2. Arbona JL, Fazzi JG, Mayoral J. Congenital esophageal cysts: case
report and review of literature. Am J Gastroenterol. 1984;79(3):177-182.
3. Neo EL, Watson DI, Bessell JR. Acute ruptured esophageal duplication
cyst. Dis Esophagus. 2004;17(1):109-111.
4. Faigel D, Burke A, Ginsberg G, Stotland BR, Kadish SL, Kochman ML.
The role of endoscopic ultrasound in the evaluation and management
of foregut duplications. Gastrointest Endosc. 1997;45(1):99-103.
5. Martin N, Kim J, Verma S, et al. Intra-abdominal esophageal duplica-
tion cysts: a review. J Gastrointest Surg. 2007;11:773-777.
6. Uppal P, Kaur J, Agarwala S, Gupta AK, Safaya R, Kabra SK. Com-
municating oesophageal duplication cyst with heterotopic pancreatic
tissue—an unusual cause of recurrent pneumonia in an infant. Acta
Vagus nerve Paediatr. 2010;99:1432-1433.
7. Cioffi U, Bonavina L, De Simone M, et al. Presentation and surgical
management of bronchogenic and esophageal duplication cysts in
adults. Chest. 1998;113(6):1492-1496.
t
8. Lee HS, Jeon HJ, Song CW, et al. Esophageal duplication cyst
FIGURE 45.8 The mediastinal pleural is reapproximated over the
ne
complicated with intramural hematoma—case report. J Korean Med
enucleation. Care was taken to preserve both vagus nerves. Sci. 1994;9(2):188-196.
9. Rangasami R, Chandrasekharan A, Archana L, Santhosh J. Case
report: antenatal MRI diagnosis of esophageal duplication cyst.
k.
Indian J Radiol Imaging. 2009;19(1):75-77.
10. Achildi O, Grewal H. Congenital anomalies of the esophagus.
Otolaryngol Clin North Am. 2007;40(1):219-244.
oo
11. Al-Sadoon H, Wiseman N, Chernick V. Recurrent thoracic duplication
Standing cyst with associated mediastinal gas. Can Respir J. 1998;5(2):149-151.
Prone 12. Chaudhary V, Rana SS, Sharma V, et al. Esophageal duplication cyst
in an adult masquerading as submucosal tumor. Endosc Ultrasound.
yb
2013;2(3):165-167.
13. Herbella FA, Tedesco P, Muthusamy R, Patti MG. Thoracoscopic resec-
tion of esophageal duplication cysts. Dis Esophagus. 2006;19(2):132-134.
14. Lee MY, Jensen E, Kwak S, Larson RA. Metastatic adenocarcinoma
er
Gastric fundus Gastric fundus 20. Jeung MY, Gasser B, Gangi A, et al. Imaging of cystic masses of the
mediastinum. Radiographics. 2002;22 Spec:S79-S93.
21. Wootton-Gorges SL, Eckel GM, Poulos ND, Kappler S, Milstein JM.
ht
S
ubmucosal tumors (SMTs) of the esophagus and in diameter. However, concerns for patient compliance,
gastroesophageal junction (GEJ) are a group of rare, the cost of repeat surveillance, the potential for delayed
benign lesions. This chapter will review the various diagnosis of malignancy, and patient anxiety may lead
types of esophageal and GEJ SMTs. We intend to focus physicians to recommend resection of small esophageal
on advancements in the imaging modalities, medical SMTs.4
treatment, and surgical options for these tumors, including Treatment options for esophageal SMTs have evolved
descriptions of evolving surgical techniques. in recent years. Targeted molecular therapy has greatly
t
Our understanding of the prevalence of esophageal improved outcomes for gastrointestinal stromal tumors
ne
SMTs comes largely from autopsy studies from the mid-20th (GIST), and recommendations for optimizing medical
century, which identified an overall prevalence of about therapy are evolving. Observation and surveillance of small
0.5% for benign esophageal tumors and cysts.1,2 Only and asymptomatic esophageal SMTs may be possible now
k.
18% of all esophageal tumors were benign; the remaining with better diagnostic techniques in cases of well defined,
82% were malignant.1 Benign esophageal and GEJ SMTs small, asymptomatic, benign esophageal lesions with low
oo
are found in people of all ages from infancy to elderly, potential for malignancy.2 Historically, surgical resection
with the majority presenting in the fourth through sixth was recommended for esophageal or GEJ SMTs, for both
decades, and with male predominance. treatment and definitive diagnosis. Open laparotomy or
yb
The development of symptoms related to esophageal thoracotomy for surgical enucleation or esophagectomy
SMTs is largely correlated to the size of the tumor. Present- was recommended. Minimally invasive techniques including
ing symptoms may include dysphagia and regurgitation due laparoscopy, video-assisted thoracoscopic surgery (VATS),
to intraluminal obstruction, pain, pulmonary symptoms and robotic surgical techniques have gained acceptance
er
due to extramural mass effect on the airways, or bleeding.3 with improved outcomes in postoperative pain, pulmonary
Physical examination is usually unrevealing unless the complications, and length of stay. More recently, purely
rg
tumor is quite large; therefore, discovery of these tumors endoscopic techniques for resection of select esophageal
is often incidental. About 50% of esophageal SMTs are SMTs have been shown to be feasible and safe when
asymptomatic, and identified on endoscopy or imaging performed by experienced endoscopists.
su
a clearer depiction of a small lesion, but findings may Surgical resection is the mainstay of treatment for
be subtle, suggesting esophageal wall thickening only. leiomyoma. Indications for surgical resection include
tp
A contrast esophagram usually shows a smooth-walled symptomatic lesions, inability to rule out malignancy
indentation on the column of contrast.2 The differential or distinguish from GIST, atypical imaging findings,
diagnosis for SMTs is broad, including benign solid tumor, overlying mucosal erosion or dysplastic changes, regional
ht
cyst, vascular anomaly, or cancer. As advancements in lymphadenopathy, large tumors (size recommendation is
imaging technology have developed, confidence in variable), and tumor growth during surveillance.5
determining the benign nature of these lesions without Traditionally SMTs, particularly leiomyomas, were
tissue for diagnosis has grown. However, in some cases, resected by either enucleation via right or left thoracotomy,
multiple imaging modalities may still not definitively rule or laparotomy for GEJ tumors. With open access to the
out malignancy. Standard endoscopic forceps biopsy or esophagus or GEJ, the surgeon can visualize and palpate
endoscopic ultrasound (EUS)-guided fine-needle aspiration the tumor to precisely localize it. The overlying pleura is
(FNA) may be inadequate or indeterminate due to the incised, and the longitudinal and circular muscle split.
submucosal location of the tumor, and whether biopsy The tumor is dissected from the surrounding muscle,
is advisable. For large or symptomatic SMTs, surgical submucosa, and enucleated from the underlying mucosa.
resection is generally recommended. However, surveillance Any mucosal injuries (incidence reported at 7% in open
versus surgical resection may be considered for small, cases) are closed primarily, and the overlying muscle
asymptomatic tumors. layers and pleura are generally closed (Fig. 46.1).2,6 The
Due to the overall rare incidence of SMTs, evidence- safety and efficacy of the open approach has been well
based recommendations are mostly limited to findings documented.5,7,8 Mortality of thoracotomy with enucle-
from a few small series. The exception to this is leiomyoma, ation is less than 1.3%, and around 90% of patients were
which is relatively frequent. Observation has been generally reported to be symptom free after 5 years.5 Esophagectomy
recommended for asymptomatic lesions less than 3 cm may be indicated for tumors greater than 8 cm or diffuse
496
Submucosal Tumors of the Esophagus and Gastroesophageal Junction CHAPTER 46 496.e1
ABSTRACT
Tumors originating in the submucosal space of the
esophagus and proximal stomach comprise a hetero-
geneous group of benign neoplasms, half of which are
asymptomatic. Removal is indicated when these tumors
generate symptoms, including dysphagia or globus or
they present with concerning manifestations, such as
ulceration/hemorrhage, large size or increased growth
during surveillance. Treatment includes medical, endo-
scopic, and surgical approaches with organ-sparing return
to baseline function in many cases.
KEYWORDS
Submucosal tumors
gastrointestinal stromal tumor (GIST)
esophageal
gastric
endoscopy
t
ne
enucleation
resection
k.
oo
yb
er
rg
su
://
tp
ht
Submucosal Tumors of the Esophagus and Gastroesophageal Junction CHAPTER 46 497
t
for the identification of small tumors and testing mucosal
ne
integrity after enucleation (Fig. 46.2).15 A few cases of
robotic enucleation for esophageal leiomyoma have been
reported. Surgeons who have used this technique suggest
k.
that the benefits include improved dexterity, precision,
and visibility, allowing minimized esophageal mobilization
oo
and avoidance of mucosal injury. 6 Up-front cost, the
requirement of a bedside surgeon, and docking time are
B concerns that must be individually weighed against the
potential benefits of this approach.6,12
yb
Minimally invasive tumor localization can be improved
FIGURE 46.1 (A) The esophageal muscular fibers are split and the with intraoperative endoscopy for specific localization
leiomyoma is bluntly extracted from the esophageal wall. (B) Once of small lesions, and intraluminal balloons or bougies
er
removed, the muscular defect is reapproximated. to help optimize the position of the lesion for surgical
resection.5,15,16 Laparoscopic transgastric or intragastric
rg
A B
FIGURE 46.2 Tumor localization was performed under esophagoscopy, and myotomy was performed at the tumor level. (A) Tumor
localization by transillumination. (B) Myotomy at the tumor level. (From Jeon HW, Choi MG, Lim CH, Park JK, Sung SW. Intraoperative
esophagoscopy provides accuracy and safety in video-assisted thoracoscopic enucleation of benign esophageal submucosal tumors. Dis
Esophagus. 2015;28:438.)
498 SECTION I Esophagus and Hernia
A B
t
ne
k.
oo
yb
er
C
rg
FIGURE 46.3 (A) Intraluminal submucosal lesion. (B) Partially enucleated lesion. (C) Intracorporeal closure of the mucosal and submucosal
su
defect. (From Mino JS, Guerron AD, Monteiro R, et al. Long-term outcomes of combined endoscopic/laparoscopic intragastric
enucleation of presumed gastric stromal tumors. Surg Endosc. 2016;30:1749.)
://
Capnogastrum is achieved with or without the aid of a Later, the technique was used for Barrett esophagus with
coda balloon to occlude the gastric outlet. This approach high-grade dysplasia or intramucosal adenocarcinoma.
tp
may help to avoid excess resection and minimize the The endoscopic submucosal dissection (ESD) technique
disruption of the angle of His at the GEJ, and has been is commonly used for tunneling in peroral endoscopic
successful for tumors as large as 7 cm.18-20 Enucleation may myotomy (POEM), a natural orifice transluminal endo-
ht
be possible with combined endoscopic and laparoscopic scopic surgery (NOTES) procedure for the treatment of
instruments, but full-thickness resection is performed achalasia. Endoscopic resection has grown in popularity
if enucleation with clear margins is not possible. The and indications have expanded as more surgeons and
specimen can be removed via a specimen bag through a gastroenterologists gain experience with ESD and resection
trocar site or transorally with the endoscope via a grasper techniques. The particular approach depends upon the
or snare. The gastrotomy incisions are primarily closed characteristics of the tumor and the experience of the
with suture, but large defects from resection may be closed endoscopist. A disadvantage of endoscopic resection is
with an endoscopic stapler. If resection of a portion of the the learning curve, though it is reported to be relatively
cardia is required and will leave a significant deformity of short for experienced endoscopists.21
the GEJ, fundoplication should be considered to prevent
postoperative reflux. The development of subsequent ENDOSCOPIC MUCOSAL RESECTION
worsened or new onset gastroesophageal reflux disease Endoscopic mucosal resection (EMR; Fig. 46.4) may be
(GERD) after resection of GEJ lesions or enucleation of used for small, superficial tumors (<2 cm, superficial to
midesophageal lesions has been reported.8,16 submucosa). EMR separates the full thickness of mucosa
Endoscopic resection has been used more frequently from the underlying muscle layers. With this technique,
in the last several years. Originally endoscopic resection a lifting solution is injected into the submucosal layer
techniques were used for the curative resection of well- beneath the tumor to create a pseudopolyp, which can
differentiated squamous cell carcinoma of the esophagus. then be banded and snared to complete the resection.22,23
A B
t
ne
k.
oo
yb
er
rg
su
C D
://
tp
ht
E F
FIGURE 46.4 Endoscopic mucosal resection of a submucosal tumor (SMT) using a ligation device in the esophagus. (A) An SMT is
observed in the lower esophagus. (B) Saline solution with a small amount of epinephrine and indigo carmine dye is injected beneath the
lesion to elevate it. (C) The lesion is then aspirated into the ligation device, followed by deployment of the elastic band. (D) Snare
resection is performed using a blended electrosurgical current. (E) The lesion is completely removed. (F) Inner surface of the resected
specimen. (From Kahng DH, Kim GH, Park DY, et al. Endoscopic resection of granular cell tumors in the gastrointestinal tract: a single
center experience. Surg Endosc. 2013;27:3230.)
500 SECTION I Esophagus and Hernia
t
A simple cautery snare can be used for a well-lifted small
ne
lesion, but usually other techniques may be required to TUNNEL TECHNIQUE
avoid piecemeal resection for an adherent tumor. For submucosal tunneling (Figs. 46.6 and 46.7), a lift
Cap-assisted EMR can be used with a single-channel is established similar to that described in the previous
k.
endoscope with the cap in place. The targeted lesion sections, except starting 2 to 3 cm proximal to the proximal
then can be fully suctioned into the cap prior to snaring edge of the lesion. A 1 to 2 cm mucosotomy is made and
oo
it. For more precise resection, a variceal band can be a submucosal tunnel created to approach the lesion—as
applied to ligate and position the lesion as a pseudopolyp described in the POEM technique first described by Inoue
so that the snare can be placed under direct vision. As an in humans.26 Continue in this plane to lift the mucosa off
yb
alternative to the cap-assisted technique, an endoscopic the lesion; then it can be dissected off of and out of the
grasper can be used to lift the lesion for snaring, but this muscularis. This may require full muscle thickness incision,
requires a dual-channel endoscope. The resection bed is exposing the mediastinal connective tissue, but the risk
then checked for bleeding or perforation. The specimen of infection or leak is minimized by careful preservation
er
can be sucked into the cap and retrieved with the scope, and clip closure of the mucosa. Once it is completely
or with grasping forceps or a retrieval net. freed, the lesion is extracted with forceps or a snare.
rg
frequently in Asia, endoscopists worldwide have begun technique is suggested to have faster recovery and healing
to accept and use this technique. ESD is indicated for time over standard ESD.24 In this case, even if the resection
larger, deeper tumors. ESD variations include endoscopic was full thickness through the circular and longitudinal
://
submucosal tunneling or, in select cases, even full-thickness muscle fibers, the closed mucosa should prevent leak or
resection. ESD is indicated for tumors larger than 2 cm development of mediastinitis or peritonitis, similar to
tp
or involving layers deep into the mucosa, which cannot full-thickness POEM.
be captured en bloc by the limited dimensions of the Intraoperative concerns including capnothorax,
EMR cap. Feasible size ranges from 2 to 8 cm depending capnomediastinum, or capnoperitoneum are usually
ht
on author, though greater than 5 cm usually requires conservatively managed as the CO 2 (which should be
piecemeal resection or at least dividing the lesion to allow used exclusively as opposed to air insufflation) is quick
extraction.24,25 ESD involves a customized incision through to absorb. Occasionally needle or Veress needle decom-
the normal mucosa and dissection beneath and around pression or a percutaneous drain may be needed if the
the lesion through the submucosal plane. The dissection patient is symptomatic or if ventilator pressures become
is performed with one of several varieties of endoscopic supraphysiologic.
electrosurgical knives. Delayed perforation, leak, or fistula are dreaded postop-
erative concerns.4,24 A follow-up swallow study can help rule
ENDOSCOPIC SUBMUCOSAL out leak before feeding. A liquid diet is allowed for several
DISSECTION TECHNIQUE days to allow healing before advancing to solid meals.
The line of the planned incision is demarcated with cautery, Bleeding or persistence of a painful ulcer is of concern,
marking every few millimeters along the circumference particularly if the mucosal defect is not closed. Proton
of the lesion as the intended margin of resection, a few pump inhibitor (PPI) or mucosal protecting agents may be
millimeters away from the lesion to allow for shrinkage required for short-term use while ulcers heal, and follow-
of the tissue while still preserving the pathologic margin up endoscopy should be performed to ensure healing.
(5 mm for malignant lesion to avoid a coagulation effect Stricture formation may occur, particularly for large lesions
on the tumor). The margin of the specimen may be requiring extensive resection. For any endoscopic resection
similarly marked for subsequent pathologic orientation. A technique, local recurrence due to inadequate resection is
Submucosal Tumors of the Esophagus and Gastroesophageal Junction CHAPTER 46 501
A B
t
ne
k.
oo
yb
er
rg
su
C D
://
tp
ht
E F
FIGURE 46.5 Endoscopic submucosal dissection of a submucosal tumor (SMT) in the stomach. (A) An SMT is observed in the upper body of
the stomach. (B) Marking dots are made around the lesion and saline with epinephrine and indigo carmine is then injected into the
submucosa beneath the lesion. (C) A complete circumferential incision is made using an insulated-tip (IT) knife. (D) Submucosal dissection is
made using an IT knife. (E) The lesion is completely removed. (F) Inner surface of the resected specimen. (From Kahng DH, Kim GH, Park
DY, et al. Endoscopic resection of granular cell tumors in the gastrointestinal tract: a single center experience. Surg Endosc. 2013;27:3230.)
502 SECTION I Esophagus and Hernia
A B
t
ne
C D
k.
FIGURE 46.6 (A) Submucosal tunneling: a submucosal tunnel 5 cm above the tumor is created by endoscopic mucosectomy. Epinephrine
oo
or similar solution is injected into submucosa to separate the superficial mucosa from muscularis propria and create a submucosal cavity.
(B) Tumor separation and resection: the submucosal tumor is separated from surrounding tissue and dissected from muscularis propria
and mucosa. (C) Removal of the submucosal tumor: the submucosal tumor is totally extracted and carefully removed with an insulated-
yb
tip knife through the submucosal tunnel. (D) Closure of the mucosal entry orifice: after complete hemostasis in the submucosal tunnel,
the mucosal entry orifice is tightly closed with hemostatic tips. (From Chen WS, Zheng XL, Jin L, Pan XJ, Ye MF. Novel diagnosis and
treatment of esophageal granular cell tumor: report of 14 cases and review of the literature. Ann Thorac Surg. 2014;97:298.)
er
of primary concern. Follow-up interval endoscopy (in some successfully with repeat endoscopic resection. At the
rg
protocols even before discharge) and EUS at 3 months 12-month follow-up there were no instances of recurrence.4
and 12 months is generally recommended, though this Declining morbidity and mortality have been observed
may vary depending on tumor pathology. over time with all surgical techniques.2 SMTs overall have
su
Endoscopic resections appeal to patients and clinicians excellent prognosis; most require limited resection and
because they allow the least invasive treatment for SMTs. most patients fully recover without persistent symptoms.
Outcomes are good, with few serious complications,
://
improved postoperative pain, and length of stay compared TUMORS BY TYPE: LEIOMYOMA
to other surgical techniques. In the largest study to date A leiomyoma is the most common benign esophageal or
tp
of ESD for SMT by He et al, 224 cases were reviewed.4 GEJ tumor of mesenchymal origin, sharing a common
The SMTs were localized 41% esophageal and 6% cardia cell origin with GIST and schwannoma.1,2 Sixty to seventy
and were resected with en bloc ESD. These were mostly percent of benign esophageal tumors are leiomyomas.
ht
small tumors, mean diameter 13.6 mm, 39% originating Twelve percent of all gastrointestinal leiomyomas originate
from muscularis mucosae, 51% from muscularis propria, in the esophagus.27 Leiomyomas of the esophagus are rare,
and 10% from submucosa. Submucosal tunnel endoscopic representing only 0.4% to 1% of all esophageal tumors;
resection was performed for lesions greater than 2 cm, esophageal cancer is 50-fold more common.2,13,28 The ratio
involving muscularis propria, or at the GEJ. The majority of leiomyoma of the esophagus in men and women is
of tumors in the esophagus were leiomyomas, and the 2 : 1. These tumors have been reported in both adults and
majority of those at the cardia were GIST. Of these patients children. Though rare in children younger than 12 years of
5% of attempts at ESD failed; 1.8% required conversion age, 90% of cases in children have diffuse leiomyomatosis
to surgery for bleeding or perforation. Overall, 6.25% of affecting extensive portions of the esophagus, and may
cases included perforation, and all but one were managed be related to other syndromes.2,29,30
conservatively. There were no cases of mediastinitis or The most common location for leiomyomas is in the
fistula, and no development of diverticulum at the resection distal two-thirds of the esophagus. A large study reported
site during the course of follow-up. Procedure time varied the esophageal location as upper third in 8.5%, middle
from 10 to 180 minutes with a mean of 47 minutes. Length third in 38.2%, lower third in 46.5%, and at the GEJ
of stay mean was 4.9 days, with a range of 1 to 31 days. in 6.8%. 31 Most commonly, leiomyomas originate in
There was a 1.3% incidence of incomplete resection. All intramural layers of the esophagus, but variations may
of these cases of incomplete resection were identified on develop extraluminally, extending into the mediastinum,
the first surveillance EUS postoperatively and were treated or intraluminally with polypoid features.2 Leiomyomas
Submucosal Tumors of the Esophagus and Gastroesophageal Junction CHAPTER 46 503
A B C
t
ne
k.
Submucosal
tunnel
oo
yb
Entrance
of tunnel
D E F
er
rg
su
://
tp
ht
G H I
FIGURE 46.7 Submucosal tunnel endoscopic resection for a submucosal tumor (SMT) of the esophagogastric junction (EGJ). (A) SMT at
the EGJ. (B) Endoscopic ultrasound (EUS) showing a lesion originating from the muscularis propria (MP) layer (arrow). (C) Submucosal
injection for marking tumor location preoperatively to prevent mistaking the target tissue in the tunnel cavity. (D) A 2-cm longitudinal
mucosal incision was made approximately 5 cm proximal to the SMT (arrow). (E) The submucosal tunnel is established (arrow). (F)
Separating the tumor from the MP layer using the hybrid knife. (G) The mucosal entry incision is sealed with several clips. (H) Irregularly-
shaped, completely resected specimen (maximum diameter 30 mm). (I) Macroscopic findings of the resected specimen revealed a
leiomyoma (H&E, ×20). (From Wang XY, Xu MD, Yao LQ, et al. Submucosal tunneling endoscopic resection for submucosal tumors of
the esophagogastric junction originating from the muscularis propria layer: a feasibility study (with videos). Surg Endosc. 2014;28:1973.)
may originate from muscularis mucosae, from muscularis a few millimeters to up to 29 cm. Leiomyomas over 1000 g
propria, or from submucosa of the esophagus.32 are considered “giant.”2 Half of all leiomyomas are less
Grossly, a leiomyoma is a firm, rubbery, encapsulated than 5 cm in diameter, 85% are less than 10 cm. Ninety-
lesion with intact overlying mucosa. It is usually uniform seven percent of esophageal leiomyomas are solitary, but
or it can be asymmetrically shaped. The size ranges from leiomyomatosis involving the entire esophageal smooth
504 SECTION I Esophagus and Hernia
t
for CD117 and CD34, differentiating it from GIST.28,34,35 SMA, smooth muscle actin.
ne
Leiomyoma has been reported in association with other Data from Mansour KA, Hatcher CR, Haun CL. Benign tumors of the
foregut disorders, including achalasia, other esophageal esophagus: experience with 20 cases. South Med J. 1977;70:461; Sweet
RH, Soutter L, Valenzuela CT. Muscle wall tumors of the esophagus. J
motility disorders, hiatal hernia, and epiphrenic diverticu-
k.
Thorac Surg. 1954;27:13, discussion 35; Nemir P Jr, Wallace HW, Fal-
lum. Resection of the tumor has been shown to correct lahnejad M. Diagnosis and surgical management of benign diseases of
dysmotility in some cases.36 Leiomyomatosis has been the esophagus. Curr Probl Surg. 1976;13:1; Reed CE. Benign tumors of
oo
rarely identified in association with some genetic disorders. the esophagus. Chest Surg Clin N Am. 1994;4:769; Herrera JL. Benign
and metastatic tumors of the esophagus. Gastroenterol Clin North Am.
Li-Fraumeni syndrome, caused by a p53 mutation, is 1991;20:775; Avezzano EA, Fleischer DE, Merida MA, et al. Giant fibro-
associated with sarcoma, breast cancer, brain cancer, and vascular polyps of the esophagus. Am J Gastroenterol. 1990;85:299; Went
rarely esophageal leiomyomatosis.30 Alport syndrome, an
yb
PT, Dirnhofer S, Bundi M, et al. Prevalence of KIT expression in human
X-linked recessive syndrome with glomerular nephropathy, tumors. J Clin Oncol. 2004;22:4514; Miettinen M, Sarlomo-Rikala M, Sobin
sensorineural deafness, and ocular abnormalities, has LH, et al. Esophageal stromal tumors: a clinicopathologic, immunohisto-
chemical, and molecular genetic study of 17 cases and comparison with
been reported in association with diffuse leiomyomatosis,
er
A B
t
ne
FIGURE 46.9 Barium swallow (A) and computed tomogram (B) reveal a large heterogeneous esophageal mass (arrow) measuring ∼3.3 cm.
(From Kernstine KH, Andersen ES, Falabella A, Ramirez NA, Anderson CA, Beblawi I. Robotic fourth-arm enucleation of an esophageal
leiomyoma and review of literature. Innovations [Phila]. 2009;4:355.)
k.
oo
yb
er
rg
su
A
://
can help predict histologic diagnosis.4 An ultrasound of alpha. Of mesenchymal SMTs, GIST is less common than
leiomyoma usually identifies a hypoechoic, homogeneous, leiomyoma, but more common than schwannoma, and
well-demarcated lesion without locoregional lymphade- has greater malignant potential than either of these. GIST
nopathy (Fig. 46.11).27 Large, irregular, heterogeneous, usually presents in older patients with the median age at
and locoregional lymphadenopathy are more suspicious presentation in the sixth decade, and rarely in patients
for malignant tumor.2,41 Endoscopic forceps or EUS-guided under age 30.46 The incidence of GIST is slightly higher
FNA biopsy of suspected leiomyoma is controversial. in males than in females; the reported incidence from
Traditionally, in the surgical literature, biopsy was con- Surveillance, Epidemiology, and End Results (SEER)
demned as a source of adhesions to mucosa, adding to data was 54% in men and 46% in women between 1992
the difficulty of operative resection and increasing the risk and 2000.47 The incidence of GIST of all sites based
of mucosal injury, bleeding, leak, fistula, or mediastinitis.8 on population studies is estimated to be about 10 cases
However, others report no technical disadvantage after per million, with only 1% of these being esophageal in
biopsy.6,12,13,16,28,42,43 Proponents of biopsy suggest that origin.46-48 Most esophageal GIST develops in the lower
EUS-guided needle biopsy is safe and is indicated in third, which is consistent with the increased distribution
lesions with a diameter greater than 2 cm, growth during of interstitial cells of Cajal.49
observation, or PET avidity to help guide management.14,27 The most common location for GIST is in the stomach,
Immunohistochemistry is diagnostic, but adequate tissue but they can develop anywhere in the gastrointestinal (GI)
may be difficult to obtain, with 50% to 80% of EUS-FNA tract. Esophageal GIST symptoms are similar to leiomyoma,
t
samples being inadequate or pathologically inconclusive. and may include dysphagia, globus, pain, or bleeding.
ne
Without biopsy, the accuracy of EUS predicting histologic Grossly, GIST is a firm, solid lesion. Histologically, GIST
diagnosis is about 80%, and the accuracy in predicting layer is a highly cellular lesion; the majority are spindle cell
of origin of the tumor is 74.6%.4 EUS with FNA biopsy type, fewer are epithelioid type, and rarely, a combination
k.
sensitivity is 95%, specificity 100%, and diagnostic yield of the two cellular patterns may be included.46 Molecular
85%, and it increases with three or more needle passes genetic markers that can help to differentiate GIST from
oo
to improve diagnostic yield.27,44 leiomyoma or schwanomma include positive KIT CD117
In the case of a tumor with clinical and imaging char- and CD34. Few GIST tumors (about 5%) are KIT nega-
acteristics consistent with a very small, benign leiomyoma, tive, in which case DOG1, a calcium-dependent chloride
yb
observation alone may be sufficient. Although there is not channel protein may be tested and seems to be consistently
a definitive consensus, some experts in favor of surveil- expressed in GIST regardless of the type of mutation.46
lance suggest it is acceptable in lesions less than 2 cm Endoscopy is typically the first modality to identify
in diameter; others suggest a cutoff of 4 to 5 cm.21 Size, esophageal GIST. GIST greater than 2 cm should prompt
er
location, symptoms, and features of the tumor should be a staging work-up including CT of the abdomen/pelvis
carefully considered when proceeding with observation with contrast and chest CT.50 PET may be useful to clarify
rg
to determine the frequency of repeat imaging. Patients ambiguous imaging findings or evaluate metastatic disease
should be monitored clinically for the development of in consideration of resection, though some leiomyoma
symptoms, with some protocols suggesting endoscopy may be FDG avid as well.46 EUS features may include
su
and EUS or CT every 6 months to 2 years.10 At least one increased echogenicity of GIST compared to leiomyoma,
case of overlying squamous cell carcinoma and another but this also has not been shown to be consistently reliable
of high-grade dysplasia associated with leiomyoma have for diagnosis. EUS-guided FNA is generally included in
://
been reported.2,42 The underlying mechanism is thought the diagnostic algorithm when GIST is suspected, and
to be increased stress and exposure of overlying mucosa is recommended in current National Comprehensive
tp
Leiomyosarcoma, a malignant esophageal mesenchymal passes to improve diagnostic yield.27,44 In cases suspicious
tumor, has a worse prognosis than other SMTs. Survival for esophageal GIST that are readily resectable, some
in the Chinese population at 1, 3, and 5 years is 60.2%, experts have recommended avoiding preoperative FNA
42.8%, and 32.1%, respectively. Prognosis is better than to avoid potential tumor rupture and dissemination,
other forms of esophageal cancer due to slow growth and for concerns of complicating resection margins with
and relatively late presentation of metastatic disease. scar tissue.8,46 However, biopsy results may change the
The treatment of leiomyosarcoma is primarily by surgical management approach. Resection is recommended even
resection. Though this tumor is not typically radiosensitve, for small GIST less than 2 cm with high-risk features,
radiotherapy may be attempted in cases of unresectable while GIST less than 2 cm with low-risk features may be
disease.45 considered for surveillance. The decision to obtain FNA
biopsy should be individualized based on suspected tumor
GASTROINTESTINAL STROMAL TUMOR type and extent of disease.50 For unresectable or metastatic
GIST is a mesenchymal tumor of the esophagus with tumors, tissue diagnosis is required to guide chemotherapy.
specific molecular genetic features that differs from We favor EUS with FNA biopsy of SMTs concerning for
leiomyoma and schwannoma. GIST is the most common GIST to obtain precise measurement of the tumor size,
soft tissue sarcoma of the gastrointestinal tract, and devel- mitotic activity in aspirated cells, and endoscopic features
ops as the result of mutations leading to the abnormal such as mucosal ulceration or erosion that may affect the
expression of c-KIT or platelet-derived growth factor surgical strategy.
Submucosal Tumors of the Esophagus and Gastroesophageal Junction CHAPTER 46 507
t
should be handled carefully, and the specimen should be with a postoperative diet. The testing of GIST for KIT and
ne
extracted in a bag to avoid seeding peritoneum or port PDGFRA is strongly recommended. If these are missing,
sites. Extended anatomic resection is rarely indicated then further staining and mutation analysis, such as BRAF
as intramural extension is usually limited; therefore and SDH should be considered, as these tumor types are
k.
organ-sparing surgery in the case of the esophagus is less likely to respond to imatinib. For GIST that does not
recommended, and esophagectomy should be avoided respond or becomes resistant to imatinib, alternative
oo
when possible. Lymphadenectomy is generally not required therapies include sunitinib, and subsequently regorafenib
unless there are pathologically positive nodes. If multiple kinase inhibitors. Adjuvant imatinib is recommended for
organs are involved then extensive resection should not be at least 3 years for high-risk GIST.50
yb
first-line therapy. Re-resection of microscopically positive Previously, the majority of GIST were considered to
resection margins is generally not indicated.50 Resection be benign, but given the propensity for recurrence and
of distant metastases is not recommended except for metastases, more recently it has been accepted that
palliation.52 any GIST has potential for malignant behavior. “The
er
Even in complete resection, about 50% of patients with term benign has in general been replaced semantically
malignant GIST at any site will develop recurrence or with very low risk. Several risk classification systems have
rg
metastasis with median recurrence of about 2 years and been developed; the National Institutes of Health (NIH)
5-year survival of 50%.50,53 Although surgical resection is Fletcher criteria (Table 46.2) are simple and widely
still the primary treatment for GIST without metastatic used. These criteria include tumor size (>5 cm) and
su
disease, the standard of care has changed to include mitotic rate (>5/HPF), and modified versions include
targeted drug therapy since the introduction of tyrosine other aggressive features. The modified NIH or Joensuu
kinase inhibitors because complete resection alone is criteria also incorporate tumor rupture and nongastric
://
commonly not curative. Evaluation of the risk of relapse primary tumor site as higher risk features that affect
is essential in determining the future prognosis of patients prognosis. Esophageal GIST is rare, but it does seem
tp
and tailoring multimodal treatment regimens, as well as to have a worse prognosis than GIST in other areas of
a surgical approach.54 Consideration for adjuvant therapy the GI tract. Feng et al. compared esophageal to gastric
is based on risk stratification for recurrence. GIST in a case match series that matched tumor size,
ht
Imatinib is a tyrosine kinase inhibitor that was originally mitotic rate, and adjuvant imatinib therapy. For the 135
used to treat leukemia. It was first found to be useful for esophageal GIST in this series, disease-free survival and
metastatic GIST in 2000. In phase II and III trials, imatinib disease-specific survival were around 65%, which was
has been shown to be safe and effective for GIST, and significantly lower compared to gastric GIST. Of note,
is now used as adjuvant therapy to improve recurrence- most of the esophageal cases were for high-risk GIST
free survival for localized GIST, based on significant (57% were >5 cm large, and/or high mitotic rate) though
improvement in recurrence-free survival in the Z9001 only 28% received imatinib. Additionally, the type of
trial of adjuvant imatinib after complete surgical resection surgical resection was not specified, and death related
of GIST.50,55 Imatinib is also indicated for GIST that is to esophagus-specific recurrence or distant metastases
definitively unresectable, recurrent, or metastatic, prior to were not differentiated.49
consideration of resection or re-resection. In patients with The optimal surgical approach for GIST has not been
unresectable and/or metastatic GIST, imatinib has good definitively determined, and remains somewhat contro-
overall response rates and progression-free survival, with versial. In general, GIST in any location requires only a
an objective response in 50% of patients.50 Neoadjuvant margin negative resection as per the most recent consensus
imatinib may be used to improve the resectability of GIST guidelines from the NCCN.50 Specifically, esophagus-sparing
and to allow for a less invasive resection strategy. Imatinib surgery is recommended for GEJ GIST, where the goal
is well tolerated and may be continued until just before is complete resection with minimal morbidity. However,
surgery and resumed quickly in the adjuvant setting along historically, there is a trend for poorer overall prognosis
508 SECTION I Esophagus and Hernia
for esophageal GIST compared to GIST of other organs. cells of Cajal, and can present anywhere in the GI tract.
Theoretically, this may be due to the lack of serosalization Approximately 60% develop in the stomach, but only
of this organ, or the complexity of resection compared about 5% are esophageal. Grossly, schwannoma is a firm,
to intraabdominal serosalized organs. Some experts have irregular, rubbery, tan lesion.63 Histologically, schwannoma
suggested that esophagectomy for GIST is likely more is moderately cellular with spindle cells, a rim of lymphoid
oncologically sound, based on SEER data and on small cells, and, in some cases, melanin pigmentation. S-100
series citing difficulty with enucleation due to large size, and GFAP proteins are positive on immunohistochemi-
friability, and adherence of the GIST (all were high-risk cal staining, while c-KIT, CD34, and SMA are negative
GIST: >5 cm and >5 mitotic figures).14,54 The data cited (Fig. 46.12).
are limited, as esophageal GIST is a very rare tumor, with The presentation and work-up of schwannoma is similar
heterogeneous risk of malignancy. SEER data reviewed in to leiomyoma and GIST, depending on the size and loca-
2005 reported an alarming 14% 5-year survival rate after tion of the tumor. Dysphagia is the most typical presenting
diagnosis if it is an esophageal GIST, with better outcomes symptom. Endoscopy will show a submucosal protrusion
noted for “partial or total organ removal,” though this with intact mucosa, and EUS shows hypoechoic tumors in
included limited details of the surgical procedures and did the submucosa or muscularis propria.62 Though typically
not document the risk classification for esophageal GIST. benign, schwannomas may be FDG avid on PET, similar
Also, only 45% of esophageal GIST was localized at the time to leiomyoma or GIST.64-66 Schwannoma tends to develop
of diagnosis, probably representing a group of higher-risk more proximally in the esophagus; therefore, tumors
t
GIST. Furthermore, the SEER data were for cases treated of larger size may present with tracheal compression or
ne
prior to 2002, before imatinib therapy was available and paresthesia in addition to esophageal obstruction.65,67-69
resection alone was the primary treatment.47 These results Malignant potential is dependent upon tumor size and may
are unlikely comparable to modern multimodal therapy. rarely be associated with metastases to lymph nodes, with
k.
Locoregional recurrence of GIST is a known risk despite only a handful of malignant cases reported.70-72 Chemo-
esophagectomy for high-risk tumors.43 therapy and radiation are ineffective for schwannoma; the
oo
More recent studies, since the onset of endoscopic mainstay of treatment is resection, with similar guidelines as
enucleation of SMTs, include a small number of esophageal leiomyoma. Surgical options include endoscopic, minimally
GIST. The results, although limited to short-term data in invasive, or open enucleation versus esophagectomy for
very large or invasive tumors.16,24,73,74 Postoperative outcome
yb
a few cases, are promising for a low rate of recurrence
or metastasis. Of note, in general, these endoscopically is dependent on the aggressiveness of the tumor, and
resectable GISTs were of smaller size; therefore, they had negative margins of resection. Only one case of recurrence
a lower risk of malignancy than those in prior reports. has been reported after resection.65
er
though data are limited in number of cases and short-term Granular cell tumor (GCT) is an SMT that can be found
follow-up.4,24,32,56,57 Two cases of endoscopic resections in any organ system and is rare in the esophagus. It is most
that developed distant metastases were high-risk GIST frequently found in the submucosa of the tongue (40%),
su
primary tumors.32 Results also have been favorable in skin (30%), and breast (15%), and less commonly the GI
many reports of open or minimally invasive enucleation tract (5%), of which only one-third are esophageal.75-79 Most
for GIST, especially for low-risk tumors.16,17,43,58-62 esophageal GCTs are isolated, but may be multiple, and in
://
It seems that small tumors with low-risk features are 5% to 14% of cases may affect multiple organ systems.75
amenable to enucleation by endoscopic, minimally invasive, Only about 2% to 3% of GCTs are malignant. A gender
tp
or open surgery with probable low risk of recurrence or preponderance has not been reliably identified, and the
metastatic disease. Esophagectomy may be considered average age of presentation is in the fourth decade.75,80
for larger tumors with high mitotic rate, for a more thor- Grossly, a GCT appears as a submucosal bulge into
ht
ough oncologic resection. However, in small series, even the esophageal lumen and classically has a pale yellow
complete resection with esophagectomy did not prevent coloration “molar-shaped” polypoid lesion, though this is
locoregional recurrence. The morbidity of esophageal not always present (Fig. 46.13).79 Microscopically, GCT cells
resection is a significant concern. There is no definitive have small nuclei with abundant granular cytoplasm and
answer to the best surgical approach for esophageal GIST, overlying mucosa with pseudoepitheliomatous hyperplasia.
so a multidisciplinary evaluation of any high-risk GIST GCT cells have electron microscopy features similar to
is advisable, and close surveillance is necessary after any schwannoma, and stain for neural proteins S-100 and
type of resection. We are unlikely to definitively answer neuron-specific enolase.81 GCTs considered to be malignant
this question due to the rarity and heterogeneity of this may demonstrate aggressive histologic findings, such as
disease. This should be a case-by-case decision with the increased cellularity and atypia; or aggressive clinical
patient, surgeon, and oncologist. findings, such as large size (>4 cm), rapid growth, or
recurrence.79 Both local invasion and metastasis have
SCHWANNOMA been reported for malignant GCTs.2
Schwannoma is the least common of the esophageal The size of GCT correlates with the presentation of
mesenchymal SMTs. The majority of these tumors present symptoms.80 Only 50% are symptomatic, with presentation
between the fourth and sixth decade of life, and there similar to other SMTs. The work-up is similar as well,
is no apparent gender predominance. Like the other with endoscopy as previously described. EUS can help
mesenchymal SMTs, schwannoma arises from interstitial to determine the layer of origin, and typically identifies
Submucosal Tumors of the Esophagus and Gastroesophageal Junction CHAPTER 46 509
A
FIGURE 46.13 Endoscopic view of granular cell tumor of the
t
ne
esophagus. A yellow submucosal tumor is noted with normal
overlying mucosa. (From An S, Jang J, Min K, et al. Granular cell
tumor of the gastrointestinal tract: histologic and
immunohistochemical analysis of 98 cases. Hum Pathol.
k.
2015;46:815.)
oo
a hyperechoic solid mass surrounded by hypoechoic
submucosa without continuity to the muscularis propria.82
yb
The ideal treatment for GCTs is somewhat controversial.
Some authors advocate the resection of all tumors due
to their unknown malignant potential; others suggest
criteria for the resection of tumors that are relatively
er
HEMANGIOMA
tp
A B
FIGURE 46.14 (A) Endoscopic image of esophageal submucosal lesion. (B) Endoscopic ultrasound image demonstrating no blood flow
t
through the lesion. (From Chedgy FJ, Bhattacharyya R, Bhandari P. Endoscopic submucosal dissection for symptomatic esophageal
ne
cavernous hemangioma. Gastrointest Endosc. 2015;81:998.)
k.
and simple, or large and multinodular, and may appear
anywhere along the length of the esophagus, though
oo
most are in the middle and distal thirds.92 EUS with color
Doppler may help to identify a vascularized lesion with
continuity to major vascular structures, but may also show
no flow.93,94 Microscopically, hemangioma has a benign
yb
proliferation of cavernous vascular spaces, with fibrous
septations. CT or magnetic resonance imaging (MRI)
with intravenous contrast may be particularly helpful to
er
t
ne
A B
k.
oo
yb
er
*
rg
due to normal overlying mucosa. One case of fibrovascular food or foreign material, or may suggest a leiomyoma or
polyp harboring a well-differentiated liposarcoma has GIST (Fig. 46.16).102,105 MRI may show a hyperintense mass
been reported.104 on T1-weighted imaging, with decreased signal intensity
Barium esophagram of a fibrovascular polyp shows a on T2-weighted imaging.106
filling defect that moves within the esophageal lumen. Resection is recommended for all fibrovascular polyps
A fibrovascular polyp may be missed endoscopically or due to the potential for airway compromise. This has
on esophagram, due to the proximal origin and normal traditionally been performed with open surgery via a
overlying mucosa in up to one-third of cases.100 Fibro- transcervical approach (Fig. 46.17) or thoracotomy in
vascular polyps are echo-dense on EUS. CT will show a which the esophagus is opened opposite the origin
heterogeneous mass, which may be confused with retained of the polyp; the polyp and stalk is excised and the
512 SECTION I Esophagus and Hernia
REFERENCES
1. Plachta A. Benign tumors of the esophagus. Review of literature
and report of 99 cases. Am J Gastroenterol. 1962;38:639-652.
2. Brock H. Benign tumors and cysts of the esophagus. In: Yeo CJ, ed.
Shackelford’s Surgery of the Alimentary Tract. Philadelphia: Saunders;
2012.
3. Choong CK, Meyers BF. Benign esophageal tumors: introduction,
incidence, classification, and clinical features. Sem Thorac Cardiovasc
Surg. 2003;15(1):3-8.
4. He G, Wang J, Chen B, et al. Feasibility of endoscopic submucosal
dissection for upper gastrointestinal submucosal tumors treatment
and value of endoscopic ultrasonography in pre-operation assess
and post-operation follow-up: a prospective study of 224 cases in
a single medical center. Surg Endosc. 2016;30(10):4206-4213.
5. Lee LS, Singhal S, Brinster CJ, et al. Current management of
esophageal leiomyoma. J Am Coll Surg. 2004;198(1):136-146.
6. Kernstine KH, Andersen ES, Falabella A, Ramirez NA, Anderson
CA, Beblawi I. Robotic fourth-arm enucleation of an esopha-
geal leiomyoma and review of literature. Innovations (Phila).
2009;4(6):354-357.
FIGURE 46.17 Delivery of giant fibrovascular polyp through the
t
7. Asteriou C, Konstantinou D, Lalountas M, et al. Nine years experi-
ne
cervical esophagotomy. (From Peltz M, Estrera AS. Resection of a ence in surgical approach of leiomyomatosis of esophagus. World
giant esophageal fibrovascular polyp. Ann Thorac Surg. J Surg Oncol. 2009;7:102.
2010;90:1018.) 8. Bonavina L, Segalin A, Rosati R, Pavanello M, Peracchia A.
Surgical therapy of esophageal leiomyoma. J Am Coll Surg.
k.
1995;181(3):257-262.
9. Rendina EA, Venuta F, Pescarmona EO, et al. Leiomyoma of the
mucosal defect and esophagotomy closed.107 Esophagec- esophagus. Scand J Thorac Cardiovasc Surg. 1990;24(1):79-82.
oo
tomy for fibrovascular polyp has only been required in 10. Samphire J, Nafteux P, Luketich J. Minimally invasive techniques
a handful of cases. Kanaan and DeMeester reported a for resection of benign esophageal tumors. Sem Thorac Cardiovasc
case of a particularly large polyp, with features initially Surg. 2003;15(1):35-43.
11. Jiang G, Zhao H, Yang F, et al. Thoracoscopic enucleation of
yb
suggestive of GIST or leiomyosarcoma, requiring tran- esophageal leiomyoma: a retrospective study on 40 cases. Dis
shiatal esophagectomy.105 In other cases, endoscopic Esophagus. 2009;22(3):279-283.
removal has been successful, though some authors raise 12. Elli E, Espat NJ, Berger R, Jacobsen G, Knoblock L, Horgan S.
concerns for the potential hemorrhage from the vascular
er
Robotic-assisted thoracoscopic resection of esophageal leiomyoma.
stalk, or the potential for incomplete resection and a Surg Endosc. 2004;18(4):713-716.
13. Haddad J, Bouazza F, Barake H, Liberale G, Flamen P, Nakadi
propensity for recurrence. Transorally, the fibrovascu- IE. Surgical strategy in abnormally increased Fluorine-18
rg
lar polyp may be resected with open or endoscopic fluorodeoxyglucose uptake in an asymptomatic lower esophageal
instruments, including polypectomy snare and cautery, submucosal tumor—Report of a case. Int J Surg Case Rep. 2014;5(9):589-
or with energy devices, such as ultrasonic shears.108,109 593.
su
14. Blum MG, Bilimoria KY, Wayne JD, de Hoyos AL, Talamonti MS,
Regardless of approach, principles of surgical resection Adley B. Surgical considerations for the management and resection
include complete removal of the base of the polyp to of esophageal gastrointestinal stromal tumors. Ann Thorac Surg.
avoid the risk of local recurrence. EUS may be helpful
://
2007;84(5):1717-1723.
intraoperatively to ensure complete endoscopic resection. 15. Jeon HW, Choi MG, Lim CH, Park JK, Sung SW. Intraoperative
No treatment deaths have been reported for either open esophagoscopy provides accuracy and safety in video-assisted
tp
24. Liu BR, Song JT, Kong LJ, Pei FH, Wang XH, Du YJ. Tunneling 48. Lott S, Schmieder M, Mayer B, et al. Gastrointestinal stromal tumors
endoscopic muscularis dissection for subepithelial tumors originating of the esophagus: evaluation of a pooled case series regarding
from the muscularis propria of the esophagus and gastric cardia. clinicopathological features and clinical outcome. Am J Cancer Res.
Surg Endosc. 2013;27(11):4354-4359. 2015;5(1):333-343.
25. Park YS, Park SW, Kim TI, et al. Endoscopic enucleation of upper-GI 49. Feng F, Tian Y, Liu Z, et al. Clinicopathologic features and clinical
submucosal tumors by using an insulated-tip electrosurgical knife. outcomes of esophageal gastrointestinal stromal tumor: evaluation
Gastrointest Endosc. 2004;59(3):409-415. of a pooled case series. Medicine (Baltimore). 2016;95(2):e2446.
26. Inoue H, Ikeda H, Hosoya T, et al. Submucosal endoscopic tumor 50. NCCN. Soft Tissue Sarcoma Version 2.2017; 2017. https://
resection for subepithelial tumors in the esophagus and cardia. www.nccn.org/professionals/physician_gls/PDF/sarcoma.pdf.
Endoscopy. 2012;44(3):225-230. 51. Lok KH, Lai L, Yiu HL, Szeto ML, Leung SK. Endosonographic
27. Baysal B, Masri OA, Eloubeidi MA, Senturk H. The role of EUS surveillance of small gastrointestinal tumors originating from
and EUS-guided FNA in the management of subepithelial lesions of muscularis propria. J Gastrointest Liver Dis. 2009;18(2):177-180.
the esophagus: a large, single-center experience. Endosc Ultrasound. 52. Nunobe S, Sano T, Shimada K, Sakamoto Y, Kosuge T. Surgery
2015;doi:10.4103/2303-9027.155772. including liver resection for metastatic gastrointestinal stromal
28. Dendy M, Johnson K, Boffa DJ. Spectrum of FDG uptake in large tumors or gastrointestinal leiomyosarcomas. Jpn J Clin Oncol.
(>10 cm) esophageal leiomyomas. J Thorac Dis. 2015;7(12):E648- 2005;35(6):338-341.
E651. 53. Rossi CR, Mocellin S, Mencarelli R, et al. Gastrointestinal stromal
29. Burgos R, Muniz E, Rosa ER, Olivares CJ, Romaguera J. Compre- tumors: from a surgical to a molecular approach. Int J Cancer.
hensive management of diffuse leiomyomatosis in a patient with 2003;107(2):171-176.
Alport syndrome. P R Health Sci J. 2013;32(4):200-202. 54. Pisters PW, Patel SR. Gastrointestinal stromal tumors: current
30. Kazarin O, Vlodavsky E, Guralnik L, Kremer R, Lachter J, Bar-Sela G. management. J Surg Oncol. 2010;102(5):530-538.
Association between esophageal leiomyomatosis and p53 mutation. 55. DeMatteo RP, Ballman KV, Antonescu CR, et al. Adjuvant imatinib
t
Ann Thorac Surg. 2013;95(4):1429-1431. mesylate after resection of localised, primary gastrointestinal stromal
ne
31. Hatch GF 3rd, Wertheimer-Hatch L, Hatch KF, et al. Tumors of tumour: a randomised, double-blind, placebo-controlled trial.
the esophagus. World J Surg. 2000;24(4):401-411. Lancet. 2009;373(9669):1097-1104.
32. Fei BY, Yang JM, Zhao ZS. Differential clinical and pathological 56. Guo J, Liu Z, Sun S, Liu X, Wang S, Ge N. Ligation-assisted endo-
characteristics of esophageal stromal tumors and leiomyomata. Dis scopic enucleation for treatment of esophageal subepithelial lesions
k.
Esophagus. 2014;27(1):30-35. originating from the muscularis propria: a preliminary study. Dis
33. Seremetis MG, Lyons WS, deGuzman VC, Peabody JW Jr. Leio- Esophagus. 2015;28(4):312-317.
myomata of the esophagus. An analysis of 838 cases. Cancer. 57. Huang ZG, Zhang XS, Huang SL, Yuan XG. Endoscopy dissection
oo
1976;38(5):2166-2177. of small stromal tumors emerged from the muscularis propria in the
34. Depypere L, Coosemans W, Nafteux P. Fluorine-18-fluorodeoxyglu- upper gastrointestinal tract: preliminary study. World J Gastrointest
cose uptake in a benign oesophageal leiomyoma: a potential pitfall Endosc. 2012;4(12):565-570.
yb
in diagnosis. Interact Cardiovasc Thorac Surg. 2012;14(2):234-236. 58. Isaka T, Kanzaki M, Onuki T. Long-term survival after thoracoscopic
35. Sousa RG, Figueiredo PC, Pinto-Marques P, et al. An unusual cause enucleation of a gastrointestinal stromal tumor arising from the
of pseudoachalasia: the Alport syndrome-diffuse leiomyomatosis esophagus. J Surg Case Rep. 2015;2015(2):doi:10.1093/jscr/rju155.
association. Eur J Gastroenterol Hepatol. 2013;25(11):1352-1357. 59. Coccolini F, Catena F, Ansaloni L, Lazzareschi D, Pinna AD.
er
36. Amer KM, Payne HR, Jeyasingham K. The relevance of abnormal Esophagogastric junction gastrointestinal stromal tumor: resection
motility patterns in intra-mural oesophageal leiomyomata. Eur J vs enucleation. World J Gastroenterol. 2010;16(35):4374-4376.
Cardio Thoracic Surg. 1996;10(8):634-640. 60. Jiang P, Jiao Z, Han B, et al. Clinical characteristics and surgical
rg
37. Sweet RH, Soutter L, Valenzuela CT. Muscle wall tumors of the treatment of oesophageal gastrointestinal stromal tumours. Eur J
esophagus. J Thorac Surg. 1954;27(1):13-31, discussion 31-35. Cardio Thorac Surg. 2010;38(2):223-227.
38. Storey CF, Adams WC Jr. Leiomyoma of the esophagus; a report 61. Lee HJ, Park SI, Kim DK, Kim YH. Surgical resection of
su
of four cases and review of the surgical literature. Am J Surg. 1956; esophageal gastrointestinal stromal tumors. Ann Thorac Surg.
91(1):3-23. 2009;87(5):1569-1571.
39. Shimada Y, Okumura T, Nagata T, et al. Successful enucleation of 62. Chen X, Li Y, Liu X, et al. A report of three cases of surgical removal
a fluorine-18-fluorodeoxyglucose positron emission tomography of esophageal schwannomas. J Thorac Dis. 2016;8(5):E353-E357.
://
positive esophageal leiomyoma in the prone position using sponge 63. Liu D, Yang Y, Qi YU, Wu K, Zhao S. Schwannoma of the esophagus:
spacer and intra-esophageal balloon compression. Gen Thorac a case report. Oncol Lett. 2015;10(5):3161-3162.
Cardiovasc Surg. 2012;60(8):542-545. 64. Matsuki A, Kosugi S, Kanda T, et al. Schwannoma of the esophagus:
tp
40. Winant AJ, Gollub MJ, Shia J, Antonescu C, Bains MS, Levine MS. a case exhibiting high 18F-fluorodeoxyglucose uptake in positron
Imaging and clinicopathologic features of esophageal gastrointestinal emission tomography imaging. Dis Esophagus. 2009;22(4):E6-E10.
stromal tumors. AJR Am J Roentgenol. 2014;203(2):306-314. 65. Kassis ES, Bansal S, Perrino C, et al. Giant asymptomatic primary
ht
41. Todaro P, Crino SF, Ieni A, Pallio S, Consolo P, Tuccari G. Intraparietal esophageal schwannoma. Ann Thorac Surg. 2012;93(4):e81-e83.
esophageal leiomyomas diagnosed by endoscopic ultrasound-guided 66. Makino T, Yamasaki M, Takeno A, et al. Thoracoscopic enucleation of
fine-needle aspiration cytology: Cytological and immunocytochemical esophageal schwannoma exhibiting (18)F-fluorodeoxyglucose uptake
features in two cases. Oncol Lett. 2014;8(1):123-126. on positron emission tomography. Dis Esophagus. 2013;26(3):331-
42. Ahn SY, Jeon SW. Endoscopic resection of co-existing severe 332.
dysplasia and a small esophageal leiomyoma. World J Gastroenterol. 67. Yoon HY, Kim CB, Lee YH, Kim HG. An obstructing large schwan-
2013;19(1):137-140. noma in the esophagus. J Gastrointest Surg. 2008;12(4):761-763.
43. Robb WB, Bruyere E, Amielh D, et al. Esophageal gastrointestinal 68. Chen HC, Huang HJ, Wu CY, Lin TS, Fang HY. Esophageal
stromal tumor: is tumoral enucleation a viable therapeutic option? schwannoma with tracheal compression. Thorac Cardiovasc Surg.
Ann Surg. 2015;261(1):117-124. 2006;54(8):555-558.
44. Kim GH, Park DY, Kim S, et al. Is it possible to differentiate gastric 69. Tomono A, Nakamura T, Otowa Y, et al. A case of benign esophageal
GISTs from gastric leiomyomas by EUS? World J Gastroenterol. schwannoma causing life-threatening tracheal obstruction. Ann
2009;15(27):3376-3381. Thorac Surg. 2015;21(3):289-292.
45. Ma S, Bu W, Wang L, et al. Radiotherapy treatment of large 70. Murase K, Hino A, Ozeki Y, Karagiri Y, Onitsuka A, Sugie S. Malig-
esophageal leiomyosarcoma: a case report. Oncology Lett. 2015;9(5): nant schwannoma of the esophagus with lymph node metastasis:
2422-2424. literature review of schwannoma of the esophagus. J Gastroenterol.
46. Demetri GD, von Mehren M, Antonescu CR, et al. NCCN Task Force 2001;36(11):772-777.
report: update on the management of patients with gastrointestinal 71. Wang S, Zheng J, Ruan Z, Huang H, Yang Z, Zheng J. Long-term
stromal tumors. J Natl Compr Canc Netw. 2010;8(suppl 2):S1-S41, survival in a rare case of malignant esophageal schwannoma cured
quiz S42-S44. by surgical excision. Ann Thorac Surg. 2011;92(1):357-358.
47. Tran T, Davila JA, El-Serag HB. The epidemiology of malignant 72. Mishra B, Madhusudhan KS, Kilambi R, Das P, Pal S, Srivastava
gastrointestinal stromal tumors: an analysis of 1,458 cases from DN. Malignant schwannoma of the esophagus: a rare case report.
1992 to 2000. Am J Gastroenterol. 2005;100(1):162-168. Korean J Thorac Cardiovasc Surg. 2016;49(1):63-66.
514 SECTION I Esophagus and Hernia
73. Kitada M, Matsuda Y, Hayashi S, Ishibashi K, Oikawa K, Miyokawa 92. Govoni AF. Hemangiomas of the esophagus. Gastrointest Radiol.
N. Esophageal schwannoma: a case report. World J Surg Oncol. 1982;7(2):113-117.
2013;11:253. 93. Chedgy FJ, Bhattacharyya R, Bhandari P. Endoscopic submucosal
74. Naus PJ, Tio FO, Gross GW. Esophageal schwannoma: first report of dissection for symptomatic esophageal cavernous hemangioma.
successful management by endoscopic removal. Gastrointest Endosc. Gastrointest Endosc. 2015;81(4):998.
2001;54(4):520-522. 94. Ha C, Regan J, Cetindag IB, Ali A, Mellinger JD. Benign esophageal
75. Giacobbe A, Facciorusso D, Conoscitore P, Spirito F, Squillante tumors. Surg Clin N Am. 2015;95(3):491-514.
MM, Bisceglia M. Granular cell tumor of the esophagus. Am J 95. Ramo OJ, Salo JA, Bardini R, Nemlander AT, Farkkila M, Mattila
Gastroenterol. 1988;83(12):1398-1400. SP. Treatment of a submucosal hemangioma of the esophagus
76. Sarma DP, Rodriguez FH Jr, Deiparine EM, Weilbaecher TG. using simultaneous video-assisted thoracoscopy and esophagoscopy:
Symptomatic granular cell tumor of the esophagus. J Surg Oncol. description of a new minimally invasive technique. Endoscopy.
1986;33(4):246-249. 1997;29(5):S27-S28.
77. Subramanyam K, Shannon CR, Patterson M, Davis M, Gourley WK. 96. Kim AW, Korst RJ, Port JL, Altorki NK, Lee PC. Giant cavernous
Granular cell myoblastoma of the esophagus. J Clin Gastroenterol. hemangioma of the distal esophagus treated with esophagectomy.
1984;6(2):113-118. J Thorac Cardiovasc Surg. 2007;133(6):1665-1667.
78. Kahng DH, Kim GH, Park DY, et al. Endoscopic resection of 97. Kobara H, Mori H, Masaki T. Successful en bloc resection of an
granular cell tumors in the gastrointestinal tract: a single center esophageal hemangioma by endoscopic submucosal dissection.
experience. Surg Endosc. 2013;27(9):3228-3236. Endoscopy. 2012;44(suppl 2 UCTN):E134-E135.
79. An S, Jang J, Min K, et al. Granular cell tumor of the gastrointestinal 98. Shigemitsu K, Naomoto Y, Yamatsuji T, et al. Esophageal heman-
tract: histologic and immunohistochemical analysis of 98 cases. gioma successfully treated by fulguration using potassium titanyl
Hum Pathol. 2015;46(6):813-819. phosphate/yttrium aluminum garnet (KTP/YAG) laser: a case
80. Coutinho DS, Soga J, Yoshikawa T, et al. Granular cell tumors of report. Dis Esophagus. 2000;13(2):161-164.
t
the esophagus: a report of two cases and review of the literature. 99. Aoki T, Okagawa K, Uemura Y, et al. Successful treatment of an
ne
Am J Gastroenterol. 1985;80(10):758-762. esophageal hemangioma by endoscopic injection sclerotherapy:
81. Reed CE. Benign tumors of the esophagus. Chest Surg Clin N Am. report of a case. Surg Today. 1997;27(5):450-452.
1994;4(4):769-783. 100. Caceres M, Steeb G, Wilks SM, Garrett HE Jr. Large pedunculated
82. Tada S, Iida M, Yao T, Miyagahara T, Hasuda S, Fujishima M. polyps originating in the esophagus and hypopharynx. Ann Thorac
k.
Granular cell tumor of the esophagus: endoscopic ultrasono- Surg. 2006;81(1):393-396.
graphic demonstration and endoscopic removal. Am J Gastroenterol. 101. Peltz M, Estrera AS. Resection of a giant esophageal fibrovascular
1990;85(11):1507-1511. polyp. Ann Thorac Surg. 2010;90(3):1017-1019.
oo
83. Mineo TC, Biancari F, Francioni F, Trentino P, Casciani CU. Conserva- 102. Levine MS, Buck JL, Pantongrag-Brown L, Buetow PC, Hallman
tive approach to granular cell tumour of the oesophagus. Three JR, Sobin LH. Fibrovascular polyps of the esophagus: clinical,
case reports. Scand J Thorac Cardiovasc Surg. 1995;29(3):141-144. radiographic, and pathologic findings in 16 patients. AJR Am J
yb
84. Catalano F, Kind R, Rodella L, et al. Endoscopic treatment of Roentgenol. 1996;166(4):781-787.
esophageal granular cell tumors. Endoscopy. 2002;34(7):582-584. 103. Images of the month: Giant fibrovascular polyp of the esophagus.
85. Yasuda I, Tomita E, Nagura K, Nishigaki Y, Yamada O, Kachi H. Am J Gastroenterol. 2012;107(10):1473.
Endoscopic removal of granular cell tumors. Gastrointest Endosc. 104. Beylergil V, Simmons MZ, Ulaner G, Jurcic J, Hibshoosh H, Carras-
er
1995;41(2):163-167. quillo JA. FDG PET/CT findings in a rare case of giant fibrovascular
86. Esaki M, Aoyagi K, Hizawa K, et al. Multiple granular cell tumors polyp of the esophagus harboring atypical lipomatous tumor/
of the esophagus removed endoscopically: a case report. Gastrointest well-differentiated liposarcoma. Clin Nucl Med. 2014;39(3):288-291.
rg
Endosc. 1998;48(5):536-539. 105. Kanaan S, DeMeester TR. Fibrovascular polyp of the esophagus
87. Tsai SJ, Lin CC, Chang CW, et al. Benign esophageal lesions: requiring esophagectomy. Dis Esophagus. 2007;20(5):453-454.
endoscopic and pathologic features. World J Gastroenterol. 106. Goto A, Suzuki M, Iizuka K, et al. Regurgitation of a mass into
su
89. Riemenschneider HW, Klassen KP. Cavernous esophageal heman- 108. Chauhan S, Draganov P. Endoscopic removal of two giant fibrovas-
gioma. Ann Thorac Surg. 1968;6(6):552-556. cular polyps of the esophagus using the “two channel, two devices
90. Rodrigues-Pinto E, Pereira P, Macedo G. Bluish discoloration technique. Gastrointest Endosc. 2011;73(5):1036-1037.
tp
of the esophagus: cavernous hemangioma of the pharynx and 109. Lobo N, Hall A, Weir J, Mace A. Endoscopic resection of a giant
larynx with esophageal involvement. Endoscopy. 2015;47(suppl 1 fibrovascular polyp of the oesophagus with the assistance of ultra-
UCTN):E213-E214. sonic shears. BMJ Case Rep. 2016;2016:doi:10.1136/bcr-2015-214158.
ht
91. Cantero D, Yoshida T, Ito T, Suzumi M, Tada M, Okita K. Esophageal 110. Pallabazzer G, Santi S, Biagio S, D’Imporzano S. Difficult
hemangioma: endoscopic diagnosis and treatment. Endoscopy. polypectomy-giant hypopharyngeal polyp: case report and literature
1994;26(2):250-253. review. World J Gastroenterol. 2013;19(35):5936-5939.
CHAPTER
Caustic Esophageal Injury
Daniel French
| Sudhir Sundaresan
47
A
caustic esophageal ingestion involves damage to the sodium hydroxide.10 Animal studies have shown variable
wall of the esophagus, secondary to direct contact degrees of injury with 10% sodium hydroxide exposed for
with an acid or base, through a well-described 60 seconds.11 The total alkalinity or acidity of a substance
inflammatory response. Knowledge of the pathophysiology also appears to determine the depth of the injury.12
of caustic injuries guides complex management decisions
through the multiple phases of these potentially life- SUBSTANCES
threatening injuries. The substances consumed are often household cleaners
or hair products. In North America, alkalizing agents
are more commonly used for cleaning, whereas in India
EPIDEMIOLOGY
t
strong acids are readily available and used for cleaning.13,14
ne
There are two etiologies of caustic ingestion: accidental Commercially available drain cleaners (e.g., Liquid Plumr
and intentional. The age of the patient is predictive of the [Clorox, Oakland, California] and Drano [S.C. Johnson &
etiology. Accidental ingestions occur in pediatric patients Son, Inc., Racine, Wisconsin]) and oven cleaners contain
k.
younger than 5 years, whereas intentional ingestions occur varying concentrations of sodium hydroxide. Dishwasher
in adults and adolescents as an act of self-harm. detergents contain phosphates. Hair relaxers contain
oo
The 2013 Annual Report of the American Associa- sodium or calcium hydroxide or ammonium hydroxide.
tion of Poison Control Centers’ National Poison Data Lye is a generic term used to describe high concentrations
System found the majority of caustic ingestions occurred of sodium hydroxide or potassium hydroxide used for
in children younger than 5 years,1 thus the majority of
yb
cleaning. Bleach is an alkali made of various chemical
the epidemiologic data is published in the pediatric compounds and thus the pH varies by brand. Mister
literature. Retrospective studies consistently report low Plumber and Lysol toilet cleaners are acids. The majority
social economic status, low parental education, residence of swimming pool cleaners are acids.
er
children ingest an alkaline substance, most commonly a The pathophysiology differs for ingestion of acid (coagula-
household cleaning agent. Often the substance was not tion necrosis) and alkali (liquefaction necrosis).
stored in a labeled container. Manufacturers in developing There is a perception that acid injury is less severe.
su
countries have poor compliance with the use of safety Acids cause instant burning when ingested, resulting in a
caps2-5; conversely, this injury appears to be declining in lower quantity consumed compared with alkali substances.
developed countries.5 Esophageal mucosal exposure to acid also results in for-
://
The pathophysiology of caustic injuries is different for more significant injury to the stomach compared with the
ingestion of acidic and alkali substances. In addition to esophagus because of gastric retention of the substance
pH, the severity of the injury depends on the following due to pyloric spasm. However, a study focusing only on
factors: viscosity, concentration, amount ingested, contact acidic ingestions reported a significantly higher grade of
time, and comorbidities. Most studies report the substances injury to the esophagus compared with the stomach.13
consumed; however, the volume is not reported because The pathophysiology resulting from an alkali ingestion
this information cannot be gathered reliably from the is more complex than for acid ingestion. Animal studies in
patient. One study estimated the minimum consumption at the 1950s and 1960s demonstrated that corrosive substances
50 to 200 mL and attempted to predict the grade of injury damage the esophagus through ischemia, thrombosis, and
based on volume and concentration of alkali substance inflammation. Johnson studied 85 dogs after consumption
consumed, acknowledging this inherent difficulty.6 Not of 10% sodium hydroxide and described three phases of
surprisingly, more severe esophageal injuries occur in adults injury based on the gross and histologic examination of
attempting suicide compared with pediatric accidental the esophagus from the sacrificed dogs.11,15 Haller et al.
ingestions, because larger quantities of the corrosive exposed cats to 10% sodium hydroxide for 1 minute,
substance are ingested.7 Substances with a pH < 2 or pH > documenting similar findings to Johnson.11 These phases
12 are known to cause significant damage.8 Full-thickness of injury are described in Table 47.1.
injury to the esophagus occurs with 10 seconds of contact The first phase, the acute necrotic phase, occurs during
with 22.5% sodium hydroxide9 or 1 second with 30% the first 72 hours of ingestion, featuring four findings:
515
Caustic Esophageal Injury CHAPTER 47 515.e1
ABSTRACT
Caustic injuries most commonly present as accidental
ingestions in children or intentional ingestions in ado-
lescents or adults. Pathophysiologic models are described
for both alkali and acidic injuries. However, the clinical
management is similar for all substances. Initial assessment
involves clinical assessment, laboratory studies, endoscopic
examination, and possibly imaging studies to grade the
severity of the injury and ultimately guide management.
Minor injuries are typically treated with a trial of oral
intake, moderate injuries require close monitoring in
a hospital setting, and severe injuries typically require
surgical intervention. Long-term complications include
recurrent stricturing and development of malignancy.
KEYWORDS
caustic injury, corrosive injury, esophagus
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
516 SECTION I Esophagus and Hernia
t
blood vessels of the injury that deposition of dense connective tissue
ne
(5) Infiltration of fibroblasts leads to esophageal narrowing due to annular fibrosis
(6) Early collagenous (stricture).11,15
producing connective The depth of injury is dependent on the pH of the
k.
tissue
substance and the duration of mucosal exposure. These
Cicatrization and 3 weeks to (1) Ongoing formation of
factors ultimately determine the severity of the injury in
stricturing 3 months collagenous connective
oo
tissue
all three phases, the patient’s associated clinical condition,
(2) Submucosa and
and risk of subsequent stricture formation. Therefore the
muscularis replaced with degree of injury is an important predictor of long-term
yb
dense fibrosis outcome, and attempts to improve outcomes have focused
(3) Decreasing inflammatory on interventions based on understanding the underlying
response pathophysiology. Despite the studies cited here describing
separate pathophysiology for alkali and acid injuries, the
er
Data from Haller JA Jr, Andrews HG, White JJ, Tamer MA, Cleveland WW.
Pathophysiology and management of acute corrosive burns of the
esophagus: results of treatment in 285 children. J Pediatr Surg.
1971;6(5):578–584; and Johnson EE. A study of corrosive esophagitis.
CLINICAL FEATURES
su
(1) death of cells through coagulation of proteins, (2) an laboratory studies, imaging, and endoscopic exams will
intense inflammatory response, (3) thrombosis of vessels, be used to make management decisions for a patient with
tp
and (4) infiltration of the esophagus wall and underlying a caustic injury.
tissues with hemorrhage and bacteria. Johnson’s studies On history, it is important to characterize the substance
showed that the submucosal layer and occasionally the ingested, including the name of the substance, texture
ht
muscularis were liquefied through a process called liquefac- (solid or liquid), quantity ingested, and the timing of the
tion necrosis.11,15 Complete liquefaction of the esophagus ingestion relative to time of presentation. In accidental
and trachea and part of the lungs has been described ingestions the time passed since ingestion and quantity
with ingestion of a large quantity of alkaline substances.16 consumed are usually less than in intentional ingestions
Johnson described an intense but variable inflammatory in the adult population. Patients must be questioned
response.15 Interestingly, four decades later using a mouse regarding dysphagia, odynophagia, refusal to drink,
model and intravital microscopy, Osman et al. reported chest pain, vomiting, and epigastric pain. Dysphagia,
minimal inflammatory cells in the early phases of the odynophagia, refusal to drink in pediatric patients, and
injury.17 However, the same study did confirm thrombosis chest pain may represent esophageal injury. Vomiting is
of the microcirculation.17 Thrombosis inhibits blood concerning for recurrent exposure of the esophagus to
flow contributing to necrosis in the compromised tissue the caustic substance and risk of aspiration. Epigastric
and delays arrival of inflammatory cells. This concept is pain is concerning for gastric injury or perforation.
supported by a later study of esophageal injury in rats that On physical exam, special attention to the patient’s vital
showed increased free radicals after 24 hours and persisting signs is essential to allow for early aggressive interventions
for 72 hours.18 The fourth finding results from the first: that can be lifesaving. The mouth should be inspected
with necrosis of the protective layers of the esophagus, for signs of mucosal injury. Drooling, hoarseness, and
the muscularis propria is exposed to the contents of the stridor must also be noted. The presence of oral mucosal
esophagus, allowing bacterial invasion into the esophageal injury and drooling has been reported to increase the
Caustic Esophageal Injury CHAPTER 47 517
t
tivariate retrospective analysis of 210 patients identified a
ne
probability of significant esophageal injury.19 Hoarseness white blood cell count greater than 20,000 on presentation
and stridor can indicate injury to the larynx and airways. as an independent predictor of death after caustic inges-
The abdomen should be examined because tenderness tion.23 A retrospective study of 129 patients focusing on
k.
can represent a gastric injury or even perforation. acid-base status determined that patients with a serum pH
It has been recognized that correlating signs and symp- less than 7.22 or base excess less than 12 would require
oo
toms of patients presenting with caustic ingestion can be operative intervention.24-26 Despite the results of these
difficult to correlate with severity of injury.20 A detailed studies, it is highly unlikely a single laboratory value can
retrospective review of pediatric patients concluded no fully characterize the severity of the injury.
yb
single or group of signs or symptoms can be used to
determine the severity of the esophageal injury. However, IMAGING
they did report that all patients with grade 2 or 3 injury on Chest radiographs are helpful in the initial and ongoing
endoscopic exam (Table 47.2) were symptomatic.21 Another assessment of patients with a caustic injury, paying par-
er
smaller retrospective review found that all patients with ticular attention for mediastinal air and pleural effusions
endoscopically proven injury were symptomatic, whereas all that could suggest perforation. The role of a contrast
rg
asymptomatic patients did not have endoscopic evidence esophagram to assess for esophageal perforation has
of injury.22 A more recent study assessed prospectively a been reviewed.26 There is ongoing debate between the
series of 148 patients (majority adults) to create a model use of barium and water-soluble contrast: barium is felt
su
to predict severity of esophageal injury. Drooling, buccal to decrease pneumonitis in the setting of aspiration but
mucosal burn, and elevated white blood cell count were irritate the peritoneal and pleural cavity if a perforation
useful parameters to predict significant esophageal injury.19 is present, whereas water-soluble contrast has the opposite
://
Given these contradicting results, no single finding should effects.27 Ultimately, with the availability of high-resolution
be used to rule out a significant injury and all patients endoscopy and cross-sectional imaging, esophagrams are
tp
should undergo endoscopy for objective assessment. not likely required in the initial assessment, and instead can
The three clinical phases may be described as follows: be used very selectively in the later course to characterize
acute, intermediate, and chronic. In the acute phase strictures or assess for delayed perforations.
ht
the priority and purpose of the clinical assessment is to CT scans are used increasingly to assess these patients
guide resuscitation, assess for necrosis or perforation of and for prognostic purposes. Fig. 47.1 shows images from
the esophagus or stomach, and determine disposition. In a male who ingested toilet cleaner sustaining a serious
the intermediate phase, repeated assessment is done to caustic injury requiring urgent gastrectomy, subtotal
assess for necrosis or delayed perforation. The chronic esophagectomy, cervical esophagostomy, and feeding
phase requires attention to three common consequences tube with delayed reconstruction with a colonic interposi-
of caustic injuries, which include stricture formation, tion graft. The grading system shown in Table 47.3 was
dysmotility, and development of cancer. developed based on a retrospective review of 49 patients
who presented with a caustic injury. CT imaging done
within 72 hours of the injury was used to assign a grade to
INITIAL INVESTIGATIONS the esophageal wall edema and inflammation of adjacent
AND MANAGEMENT tissues. The grade system was designed to predict the
subsequent development of a stricture as documented
A patient with a caustic injury should be managed as a on esophagram completed at a later date.28 Another
critically ill patient, necessitating stabilization of the patient study looked at the role of CT in guiding management
prior to completing all investigations. The priorities of of patients with grade 3b injuries on endoscopy. Seventy-
airway, breathing, and circulation must be followed. A two patients who had a CT scan were compared with 125
patient presenting with respiratory distress may require patients who did not. This study excluded patients who
518 SECTION I Esophagus and Hernia
A B
t
ne
k.
oo
yb
er
C D
rg
FIGURE 47.1 Computed tomography (CT) images of a male patient presenting with a severe caustic injury after ingestion of toilet cleaner.
Images of the proximal (A) and distal (B) esophagus show inflammation of the esophageal wall with lack of enhancement. Abdominal
images show inflammation and lack of enhancement in the proximal stomach (C) and pneumatosis and free air anterior to the liver
su
suggesting gastric necrosis with perforation (D). The patient underwent a gastrectomy and subtotal esophagectomy with a cervical
esophagostomy. He was eventually reconstructed with a colon interposition graft.
://
TABLE 47.3 Grading System for Computed Tomography rate of esophagectomy in patients undergoing a CT scan
Findings without an increase in mortality.29 A systematic review of the
literature noted a lack of randomized controlled data but
ht
Grade CT Findings
Grade I No definite swelling of esophagus wall (<3 mm
concluded that CT cannot replace endoscopy.30 Essentially,
within normal limits) the literature supports CT imaging as a useful adjunct to
Grade II Edematous wall thickening (>3 mm) without endoscopy in guiding management in hemodynamically
periesophageal soft tissue infiltration stable patients.
Grade III Edematous wall thickening with periesophageal
soft tissue infiltration plus well-demarcated ENDOSCOPY
tissue interface Direct visualization of the esophageal mucosa has been
Grade IV Edematous wall thickening with periesophageal considered the ideal means of evaluating a patient with
soft tissue infiltration plus blurring of tissue a caustic injury. The grading system shown in Table 47.2
interface or localized fluid collection around was developed based on 88 patients having a total of
esophagus or descending aorta 381 endoscopy procedures.6 Fig. 47.2 shows endoscopic
CT, Computed tomography. images several hours after a male patient ingested sodium
From Ryu HH, Jeung KW, Lee BK, et al. Caustic injury: can CT grading hydroxide. The injury was grade 2b, and he was ultimately
system enable prediction of esophageal stricture? Clin Toxicol (Phila.). managed without surgery. In contrast, the images in Fig.
2010;48(2):137–142. 47.3 were acquired 8 hours after a female ingested an
unidentified quantity of Drano. These injuries are grade 3b,
and the patient required emergent esophagogastrectomy.
The final pathology revealed transmural necrosis with early
Caustic Esophageal Injury CHAPTER 47 519
A B
t
ne
k.
oo
yb
er
rg
C
su
FIGURE 47.2 Endoscopic images of a male patient several hours after ingestion of a sodium hydroxide. Examination of the esophagus (A),
://
body (B) and antrum (C) of the stomach show diffuse ulcerations of the esophagus and stomach consistent with a grade 2a caustic
injury. The patient was managed without surgical intervention.
tp
neutrophil infiltration. Despite the established utility of authors also emphasized that endoscopy allowed for
endoscopy, there are two persistent areas of debate: (1) early prognosis, early feeding, and earlier discharge in
ht
the need to perform routine endoscopy on all patients the setting of normal findings.32 In summary, larger and
presenting with a caustic ingestion; and (2) the role of current studies support the need to perform endoscopy
advancing the endoscope through an area of injury to in all patients to grade the esophageal injury.
fully visualize the entire esophagus. The second debate involves the advancement of the
The first debate involves mainly pediatric patients endoscope through any area of injury. Although some
because in adults there is a general consensus that all authors argue that this increases the risk of perforation,
patients require endoscopic assessment of their esophagus. others argue that a complete assessment of the esophagus
As noted, in the pediatric population most ingestions are is required to fully grade the injury and make treatment
accidental involving consumption of a small quantity of decisions. A recent report suggests that the opinion against
the caustic agent. A retrospective review of 28 pediatric full endoscopic assessment relates to the previous use
patients reported that all four asymptomatic patients in of rigid endoscopy.33 However, rigid esophagoscopy is
the series had no findings at the time of endoscopy.22 rarely used currently and slimmer flexible endoscopes
However, a larger study that reviewed 206 children with are now widely available. Thus, although there is a lack
caustic injury found that 22 of 57 patients with normal of clinical data to settle this debate, it is reasonable to
clinical findings had abnormal endoscopic findings graded predict a lower iatrogenic perforation rate with flexible
between 1 and 2b.31 Another larger study found poor endoscopy, and a complete endoscopic examination should
sensitivity and specificity of clinical findings, concluding be performed on all patients suspected to have a caustic
endoscopy is mandatory in all pediatric patients. These ingestion.
520 SECTION I Esophagus and Hernia
A B
t
ne
k.
oo
yb
er
rg
C
su
FIGURE 47.3 Endoscopic images of a female patient 8 hours after ingestion of Drano. Examination of the stomach body (A), fundus (B),
://
and pylorus (C) show diffuse necrosis consistent with a grade 3b injury. The duodenal mucosa distal to the pylorus was not injured. The
patient required emergent esophagogastrectomy where the final pathology revealed transmural necrosis with early neutrophil infiltration.
tp
Endoscopic ultrasound (EUS) is a relatively new tech- and observed for 1 to 2 days; grade 2b and 3a injuries
nology used for prognostication for caustic esophageal should be kept nil per os (NPO) and monitored in an
ht
injuries. A retrospective review of 18 adult patients found intensive care unit, with follow-up endoscopy in 5 to 7
that EUS can safely visualize the entire esophageal wall days to ensure improvement/healing before starting oral
and demonstrate the depth of the injury. However, there intake; and grade 3b or 4 injuries will require urgent
was no added benefit to EUS when used in conjunction operative intervention.26,27,36-38 Although there is no high-
with endoscopy.34 A smaller review of 11 adult patients level evidence to guide management of patients with
reported that EUS can demonstrate injuries that penetrate grade 2b and higher injuries, there are data from multiple
the muscularis predicting the formation of strictures and case series to guide medical and surgical treatments for
concluded that EUS is useful for prognosticating.35 In individual patients.
summary, EUS appears to be safe but has an undefined
role in management of caustic injuries. MEDICAL MANAGEMENT
The goal of medical management has been to prevent
acute perforation and chronic strictures. Evidence is
TREATMENT IN ACUTE PHASE lacking to clearly recommend interventions to achieve
After a secure airway is confirmed and intravenous these goals, but there is evidence supporting avoidance
resuscitation has begun, the investigations noted earlier of some common medical interventions.
should be promptly initiated to grade the injury and Agents to induce emesis are contraindicated because
guide management of the patient. In very general terms: emesis reexposes the esophagus to the caustic substance
grade 1 and 2a injuries can be given a trial of oral intake and a potential second insult. Ingestion of neutralizing
Caustic Esophageal Injury CHAPTER 47 521
agents has not been proven to benefit the patient but experience and preference. In grade 3a injuries a feeding
can lead to an exothermic reaction, further injuring jejunostomy should be considered to provide long-term
the esophagus. Ingestion of milk and activated charcoal enteral nutrition. Survival has been reported after upper
also have no proven benefit but can inhibit endoscopic abdominal exentration including esophogastrectomy,
examination of the esophagus.26 total pancreatectomy, and duodenectomy.43 Resection of
Antibiotics were originally used in conjunction with the colon obviously limits later reconstructive options.
steroids. Data from prospective randomized controlled trials If tissue viability is uncertain, a scheduled “second look”
are lacking; however, many authors use broad-spectrum should be conducted in 12 to 24 hours.
antibiotics in the management of caustic injuries.24,37
However, other authors have published similar rates of
mortality and morbidity without antibiotics,36 and antibiot-
TREATMENT IN INTERMEDIATE PHASE
ics have not been shown to prevent strictures.39 Because It is patients with grade 2 to 3a injury who will require
the pathophysiology involves bacterial infiltration with management in the intermediate phase. The majority of
higher-grade injuries, antibiotics can be justified and grade 1 injuries will not require any significant manage-
should be administered. ment, and grade 3b and 4 injuries will be recovering
Although often used, the utility of proton pump inhibi- from removal of the esophagus and stomach. The two
tors (PPIs) has not yet been clearly defined in prospective areas of focus of care include nutritional support and
randomized controlled trials.24,27 A prospective study of 13 stricture prophylaxis.
t
patients with grades 1 to 3 injury receiving an infusion of
ne
omeprazole followed by omeprazole intravenously twice a NUTRITIONAL SUPPORT
day revealed significant healing on repeat endoscopy at 72 A review of 315 patients managed with a nutritional
hours compared with a retrospective review of patients not protocol based on endoscopic grade revealed a 93%
k.
receiving a PPI.40 However, other authors have published success rate with nonoperative management. Grade 0
similar rates of mortality and morbidity without acid and 1 injuries were allowed a normal diet and discharged
oo
suppression.36 There is no basis for recommendation of home after 1 day of observation in hospital. Grade 2 to
routine use of PPIs in this setting. 3A injuries were placed on total parenteral nutrition
Early administration of steroids has been a point of (TPN) and had repeat endoscopy at 1 week. If the injury
yb
controversy for many years. Animal studies from the was improved or healed the patients were given a trial
1950s to 1960s suggested a benefit to steroids in reduc- of oral intake; otherwise they were supported with TPN
ing the acute inflammatory response in the esophagus with repeat endoscopy after 3 weeks. Patients with grade
and associated lung injuries. 15 This inhibition of the 3a injury who were expected to have a prolonged course
er
acute inflammatory response has also been proposed to without oral intake had a feeding jejunostomy placed.
decrease the incidence of stricture in the later phases of All patients with grade 3b or 4 injuries were taken to the
rg
the injury. Unfortunately, this pharmacologic model has operating room for exploration.38 A subsequent literature
not been proven in two large meta-analyses. A systematic review recommended a similar approach.27
pooled analysis of 50 years of data showed no benefit from Placement of a nasogastric (NG) tube theoretically
su
steroids in grade 2 injuries but did show an increased allows a route for enteral nutrition and a guide for future
risk of perforation and infection, leading the authors to stricture dilation.26,36,37 Risks include perforation during
recommend against the use of steroids for grade 1 to 3 insertion, erosion, and exacerbating reflux secondary to
://
injuries.41 A meta-analysis of 12 studies comparing patients stenting open natural sphincters. In one study, 53 patients
with grades 2 and 3 injuries who received and did not had an NG tube safety placed endoscopically over a wire
tp
receive steroids for longer than 8 days reported higher into the duodenum and fed through it for 8 weeks. When
rates of strictures in patients who received steroids. The compared with 43 patients managed with jejunostomy
available evidence shows no benefit from steroids, and feeds for 8 weeks, there was no difference in stricture rate
ht
in fact a possible harm, to patients with a caustic injury.42 or morbidity.36 There is no evidence to support routine
use of NG tubes in caustic injuries.
SURGICAL MANAGEMENT
Patients with grades 3b or 4 injuries on esophagoscopy, STRICTURE PROPHYLAXIS
usually adults committing deliberate ingestion, require Strictures predominantly affect grade 2 and 3 injuries.41 As
an operation.38 Although surgery is not contraindicated noted, there is not strong evidence to show that antibiotics
in children, all efforts should be made to preserve their or PPI in the acute phase of the injury prevent strictures.
esophagus because of the challenges and long-term However, there is evidence to show steroids do not prevent
complications associated with restoring continuity of the strictures and may actually be harmful to the patient.
gastrointestinal tract in a growing patient.20 In addition to these systemic therapies, there has been
The principles of damage control surgery apply in considerable effort to identify compounds that can be
caustic injuries: careful assessment of the esophagus and applied topically or injected into the esophageal wall to
stomach, removal of necrotic/nonviable tissue, preservation prevent stricture formation, progression, and recurrence.
of hemostasis, and deferring definitive reconstruction. If Several of these are summarized later; most have remained
the esophagus is nonviable, this will require esophagectomy experimental. Therefore management of strictures involves
with cervical esophagostomy. There is debate regarding mostly mechanical dilation or stenting.
the role of laparotomy versus laparoscopy to examine the Triamcinolone, a corticosteroid, has been injected
stomach. Ultimately, this choice depends on the surgeon’s into strictures to prevent progression.44 Mitomycin C, a
522 SECTION I Esophagus and Hernia
chemotherapy agent known to disrupt DNA, has been placement. The authors concluded stents should be used
used in trials of topical application and injection.45,46 In for caustic injuries instead of serial dilation. In this series,
a prospective trial, mitomycin C was shown to reduce one tracheoesophageal fistula is reported resulting from
esophageal strictures in 18 pediatric patients with caustic stent placement compared with six major complications
injuries without systemic adverse events. 47 However, from serial dilations.62 Another study described antegrade
concerns regarding systemic absorption of this agent insertion of stent for 2 to 3 months with success in 80%
and the possibility of long-term secondary malignancy of patients.63 A review of the literature quotes the efficacy
persist. The chemotherapy agent 5-fluorouracil (5-FU, of stenting at 50% to 72% with approximately a 10%
an antimetabolite with antiproliferative effect through migration rate. In summary, stenting is a reasonable
interference with pyrimidine metabolism) has been applied option to avoid serial dilations in patients with strictures;
intraluminally in rats with caustic injuries and been shown however, timing and patient selection must be carefully
to reduce esophageal strictures.48 individualized.
Multiple antioxidant agents, the majority with antifi-
brotic and antiinflammatory properties, have been tried in
animal models, but no human trials exist.27 Pirfenidone,
TREATMENT IN CHRONIC PHASE
palifermin, tamoxifen, olmesartan, tenoxicam, and The objective in the chronic phase of the injury is to restore
glucagon-like peptide 2 (GLP-2) have been shown in rats functional gastrointestinal continuity. The patients who
to decrease esophageal strictures after caustic injury.49-54 underwent emergent resection require a reconstructive
t
There has also been an attempt to use mesenchymal stem operation, whereas those who did not often experience
ne
cells to prevent strictures in rats following caustic injury.55 severe stricture formation or dysmotility.
With lack of a medical means to prevent stricture forma- Strictures are reported to occur in 6.3% to 13.8% and
tion, the predominant management has been dilatation 23.1% to 71% of grade 2 and 3 injuries, respectively.42
k.
of the strictures after they become evident. The role of Because strictures start forming in the intermediate
early dilation to prevent clinically significant strictures phase and continue to remodel in the chronic phase,
oo
has been debated for two reasons. 56 Firstly, stricture the management of strictures crosses the two phases. As
dilatation requires physically disrupting annular fibrosis, described previously, endoscopic dilation is the first line
which will heal with further fibrosis; increased dilation of therapy for strictures and some series report improved
yb
is thus potentially associated with increased stricture outcomes with early stent placement. Complications of
rate. Secondly, it is not recommended to dilate between stricture formation necessitating surgical intervention
7 and 21 days after the injury when necrotic tissue is include obliteration of the lumen, nondilatable stricture
sloughing and new collagen is forming, increasing the (i.e., failure to achieve satisfactory patency despite multiple
er
risk of esophageal perforation.24,57,58 Savary dilators are dilations), patient refusal to undergo serial dilatations, or
safer than balloon dilatation because the operator can perforation during dilatation.
rg
feel resistance to guide the dilation, thereby decreasing There are two considerations in surgical management of
the perforation risk.27,56 The overall rate of perforation patients in the chronic phase: (1) resection versus bypass
has been reported to be between 5% and 32% depending of strictures, and (2) optimal conduits for reconstruction.
su
on operator expertise.58 For some patients the surgical options are dictated by the
A retrospective review of 125 pediatric patients found extent of resection in the earlier phases of the injury,
early dilatation prevented the subsequent development of or the location or pattern of the injury. In patients with
://
recalcitrant strictures.57 Placement of a gastrostomy tube intentional ingestions, establishment of psychiatric stability
has been reported to facilitate retrograde dilatations.58 is vital prior to undertaking major reconstructive surgery.
tp
on-demand dilatations after completing this regimen.59 of mucoceles and development of cancer in the injured
Intraluminal stents have been advocated to avoid serial organ. Esophageal carcinoma has been reported to occur
dilations; however, there is an absence of prospective and in more than 30% of cases in both injured and noninjured
controlled data to provide clear evidence that stenting portions of the esophagus, which has been used to justify
is superior to dilation. The majority of data come from performing total esophagectomy.64 Arguments against
the pediatric population in which a variety of stents esophageal resection include the risks of operating in a
have been used at different time intervals relative to surgical field that has sustained a significant inflammatory
the caustic injury.56 Polypropylene stents covered with response (e.g., injury to the airway or thoracic duct) and
silicone were used in 10 pediatric patients with caustic the long latent period until development of cancer (15
strictures who failed 6 to 9 months of dilation. Stents were to 40 years).65 Ultimately, there is a lack of high-level
retained for 20 to 225 days, yielding a 50% cure rate.60 In evidence and therefore a decision must be made on an
another study, customized polytetrafluoroethylene (PTFE) individual basis considering the patient’s life-expectance,
stents were placed for 9 to 14 months in 11 patients comorbidities, ability to withstand a major procedure, and
who previously failed 3 months of dilation. Eight of the associated potential complications.
11 patients were eventually able to resume a normal
diet.61 A case series describing management of benign RECONSTRUCTION
strictures (70% were secondary to caustic injuries) treated Isolated injury to the esophagus allows for reconstruction
with silicon stents report fewer dilations with early stent with a gastric conduit or a colonic interposition graft. A
Caustic Esophageal Injury CHAPTER 47 523
gastric conduit is commonly used for reconstruction in repair.22 The incidence of GERD has been reported to be
oncologic resection and requires only a single anastomosis. 63.5% in pediatric patients after a caustic injury. The length
Colon interposition involves greater operative time and of stricture has been associated with an increased rate of
complexity, with three anastomoses. A review of 28 patients GERD. It is possible that GERD promotes stricturing in
treated on average 5 months after a caustic injury by addition to the underlying insult from the caustic injury.
transhiatial esophagectomy with cervical esophagogas- Prolonged and severe GERD, especially from a young age,
trostomy reported good overall survival, but almost half may increase the risk of Barrett esophagus and eventual
of the patients developed strictures at the anastomosis adenocarcinoma.74
requiring an average of three dilations.66 Other authors
implicate the increased incidence of reflux and aspiration
associated with gastric conduits as the basis to recommend
SUMMARY
a colon graft instead—highlighting the colon as a straight The majority of caustic injuries occur as accidental inges-
and superior functioning conduit.67 tions in children, whereas the minority is intentional
Significant gastric injury mandates that a colon ingestions in adolescents and adults. The pathophysiology
interposition graft be used. A review of 32 patients who is well described through animal models. Endoscopy
had antesternal colonic interposition graft for persistent is required to grade injuries and guide management.
strictures or to restore gastrointestinal conduit after Clinical assessment, blood work and CT imaging are
esophagogastrectomy report good functional outcomes.65 useful adjuncts to endoscopy. In general, grade 1 and
t
If the stomach and colon are both unavailable as 2a injuries are managed with a trial of oral intake, grade
ne
conduits, more esoteric options must be considered. 2b and 3a injuries require close monitoring in hospital,
Jejunal interposition grafts are described but technically and grade 3b and 4 injuries require emergent surgical
challenging, requiring microvascular anastomosis. Reverse intervention. In nonoperative patients, there is evidence to
k.
gastric tubes are also described but known to be technically show steroids are not beneficial and controversies persist
difficult.68 regarding other medical therapies. Ultimately, mechanical
oo
Severe strictures involving the pharynx and esophagus dilation is often required for management of strictures.
may require reconstruction with a colonic interposition Long-term complications of caustic injuries include an
graft or free jejunal graft. Pharyngocoloplasty does allow increased risk of malignancy, strictures, dysmotility, and
yb
for restoration of gastrointestinal continuity with good GERD. Caustic injuries remain a surgical disease often
survival outcomes, but patients do report significantly requiring careful decision making regarding resection
lower quality of life compared with patients who can while planning for complex reconstructions to restore
undergo an esophagocoloplasty.69 Free jejunal grafts have continuity of the gastrointestinal tract.
er
nolaryngol. 2006;70(10):1791-1798.
4. Riffat F, Cheng A. Pediatric caustic ingestion: 50 consecutive cases
and a review of the literature. Dis Esophagus. 2009;22(1):89-94.
LONG-TERM CONSIDERATIONS 5. Contini S, Swarray-Deen A, Scarpignato C. Oesophageal corrosive
injuries in children: a forgotten social and health challenge in
The most serious long-term complication after a caustic developing countries. Bull World Health Organ. 2009;87:950.
injury is the development of malignancy; benign 6. Zargar SA, Kochhar R, Mehta S, Mehta SK. The role of fiber-
complications such as esophageal stricture, dysmotility, optic endoscopy in the management of corrosive ingestion and
modified endoscopic classification of burns. Gastrointest Endosc.
gastroesophageal reflux disorder (GERD), and recurrent 1991;37(2):165-169.
aspiration are far more common. 7. Poley JW, Steyerberg EW, Kuipers EJ, et al. Ingestion of acid and
The incidence of adenocarcinoma and squamous cell alkaline agents: outcome and prognostic value of early upper
carcinoma is 2% to 8% after esophageal caustic injury, endoscopy. Gastrointest Endosc. 2004;60(3):372-377.
8. Arevalo-Silva C, Eliashar R, Wohlgelernter J, Elidan J, Gross M.
with a latency of 15 to 40 years after the injury.73 However, Ingestion of caustic substances: a 15-year experience. Laryngoscope.
some series report a rate as high as 31.3%.64 Standardized 2006;116(8):1422-1426.
surveillance regimens are not established but should be 9. Krey H. On the treatment of corrosive lesions in the oesophagus;
considered on an individual basis as part of the patient’s an experimental study. Acta Otolaryngol Suppl. 1952;102:1-49.
long-term care. 10. Kirsh MM, Ritter F. Caustic ingestion and subsequent damage
to the oropharyngeal and digestive passages. Ann Thorac Surg.
Tracheoesophageal fistula has been described after 1976;21(1):74-82.
caustic injury to the esophagus and will require resection 11. Haller JA Jr, Andrews HG, White JJ, Tamer MA, Cleveland WW.
of the fistula and possibly the esophagus along with airway Pathophysiology and management of acute corrosive burns of
524 SECTION I Esophagus and Hernia
the esophagus: results of treatment in 285 children. J Pediatr Surg. 37. Kay M, Wyllie R. Caustic ingestions in children. Curr Opin Pediatr.
1971;6(5):578-584. 2009;21(5):651-654.
12. Hoffman RS, Howland MA, Kamerow HN, Goldfrank LR. Comparison 38. Cabral C, Chirica M, de Chaisemartin C, et al. Caustic injuries of
of titratable acid/alkaline reserve and pH in potentially caustic the upper digestive tract: a population observational study. Surg
household products. J Toxicol Clin Toxicol. 1989;27(4-5):241-246. Endosc. 2012;26(1):214-221.
13. Zargar SA, Kochhar R, Nagi B, Mehta S, Mehta SK. Ingestion of 39. Salzman M, O’Malley RN. Updates on the evaluation and management
corrosive acids. Spectrum of injury to upper gastrointestinal tract of caustic exposures. Emerg Med Clin North Am. 2007;25(2):459-476.
and natural history. Gastroenterology. 1989;97(3):702-707. 40. Cakal B, Akbal E, Koklu S, et al. Acute therapy with intravenous
14. Lahoti D, Broor SL. Corrosive injury to the upper gastrointestinal omeprazole on caustic esophageal injury: a prospective case series.
tract. Indian J Gastroenterol. 1993;12(4):135-141. Dis Esophagus. 2013;26(1):22-26.
15. Johnson EE. A study of corrosive esophagitis. Laryngoscope. 41. Fulton JA, Hoffman RS. Steroids in second degree caustic burns of
1963;73:1651-1696. the esophagus: a systematic pooled analysis of fifty years of human
16. Emoto Y, Yoshizawa K, Shikata N, Tsubura A, Nagasaki Y. Autopsy data: 1956–2006. Clin Toxicol (Phila). 2007;45(4):402-408.
results of a case of ingestion of sodium hydroxide solution. J Toxicol 42. Pelclova D, Navratil T. Do corticosteroids prevent oesophageal
Pathol. 2016;29(1):45-47. stricture after corrosive ingestion? Toxicol Rev. 2005;24(2):125-129.
17. Osman M, Russell J, Shukla D, Moghadamfalahi M, Granger 43. Guarino S, Shobayo F, Qureshi YA, Daley F, Alaraimi B, Patel B.
DN. Responses of the murine esophageal microcirculation to Upper abdominal exenteration: a life saving procedure following
acute exposure to alkali, acid, or hypochlorite. J Pediatr Surg. caustic ingestion. Dig Liver Dis. 2014;46(4):386-387.
2008;43(9):1672-1678. 44. Kochhar R, Ray JD, Sriram PV, Kumar S, Singh K. Intralesional
18. Gunel E, Caglayan F, Caglayan O, Akillioglu I. Reactive oxygen radical steroids augment the effects of endoscopic dilation in corrosive
levels in caustic esophageal burns. J Pediatr Surg. 1999;34(3):405-407. esophageal strictures. Gastrointest Endosc. 1999;49(4):509-513.
19. Havanond C, Havanond P. Initial signs and symptoms as prognostic 45. Uhlen S, Fayoux P, Vachin F, et al. An alternative conservative
t
indicators of severe gastrointestinal tract injury due to corrosive treatment for refractory esophageal stricture in children? Endoscopy.
ne
ingestion. J Emerg Med. 2007;33(4):349-353. 2006;38(4):404-407.
20. Gaudreault P, Parent M, McGuigan MA, Chicoine L, Lovejoy FH Jr. 46. Berger M, Ure B, Lacher M. Mitomycin C in the therapy of
Predictability of esophageal injury from signs and symptoms: a study recurrent esophageal strictures: hype or hope? Eur J Pediatr Surg.
of caustic ingestion in 378 children. Pediatrics. 1983;71(5):767-770. 2012;22(2):109-116.
k.
21. Gorman RL, Khin-Maung-Gyi MT, Klein-Schwartz W, et al. Initial 47. El-Asmar KM, Hassan MA, Abdelkader HM, Hamza AF. Topical
symptoms as predictors of esophageal injury in alkaline corrosive mitomycin C can effectively alleviate dysphagia in children
ingestions. Am J Emerg Med. 1992;10(3):189-194. with long-segment caustic esophageal strictures. Dis Esophagus.
oo
22. Gupta SK, Croffie JM, Fitzgerald JF. Is esophagogastroduodenos- 2015;28(5):422-427.
copy necessary in all caustic ingestions? J Pediatr Gastroenterol Nutr. 48. Duman L, Buyukyavuz BI, Altuntas I, et al. The efficacy of single-dose
2001;32(1):50-53. 5-fluorouracil therapy in experimental caustic esophageal burn. J
yb
23. Rigo GP, Camellini L, Azzolini F, et al. What is the utility of selected Pediatr Surg. 2011;46(10):1893-1897.
clinical and endoscopic parameters in predicting the risk of death 49. Orozco-Perez J, Aguirre-Jauregui O, Salazar-Montes AM, Sobrevilla-
after caustic ingestion? Endoscopy. 2002;34(4):304-310. Navarro AA, Lucano-Landeros MS, Armendariz-Borunda J. Pir-
24. Cheng HT, Cheng CL, Lin CH, et al. Caustic ingestion in adults: fenidone prevents rat esophageal stricture formation. J Surg Res.
er
ingestion injuries. Surg Today. 2003;33(7):483-485. burns. Exp Ther Med. 2014;8(4):1087-1091.
26. Ramasamy K, Gumaste VV. Corrosive ingestion in adults. J Clin 51. Elmas O, Cevik M, Demir T, Ketani MA. Effect of oral tamoxifen
Gastroenterol. 2003;37(2):119-124. on the healing of corrosive oesophageal burns in an experimental
su
27. Contini S, Scarpignato C. Caustic injury of the upper gastro- rat model. Interact Cardiovasc Thorac Surg. 2014;19(3):351-356.
intestinal tract: a comprehensive review. World J Gastroenterol. 52. Dereli M, Krazinski BE, Ayvaz S, et al. A novel approach for preventing
2013;19(25):3918-3930. esophageal stricture formation: olmesartan prevented apoptosis.
28. Ryu HH, Jeung KW, Lee BK, et al. Caustic injury: can CT grading Folia Histochem Cytobiol. 2014;52(1):29-35.
://
system enable prediction of esophageal stricture? Clin Toxicol (Phila). 53. Tekin M, Topaloglu N, Kucuk A, Deniz M, Yildirim S, Erdem H.
2010;48(2):137-142. Protective effect of glucagon-like peptide-2 in experimental corrosive
29. Chirica M, Resche-Rigon M, Pariente B, et al. Computed tomog- esophagitis. Dis Esophagus. 2015;28(3):258-261.
tp
raphy evaluation of high-grade esophageal necrosis after corrosive 54. Erbas M, Kiraz HA, Kucuk A, et al. Effects of tenoxicam in experimen-
ingestion to avoid unnecessary esophagectomy. Surg Endosc. tal corrosive esophagitis model. Dis Esophagus. 2015;28(3):253-257.
2015;29(6):1452-1461. 55. Kantarcioglu M, Caliskan B, Demirci H, et al. The efficacy of mes-
ht
30. Bonnici KS, Wood DM, Dargan PI. Should computerised tomography enchymal stem cell transplantation in caustic esophagus injury: an
replace endoscopy in the evaluation of symptomatic ingestion of experimental study. Stem Cells Int. 2014;2014:939674.
corrosive substances? Clin Toxicol (Phila). 2014;52(9):911-925. 56. Bonavina L, Chirica M, Skrobic O, et al. Foregut caustic injuries:
31. Temiz A, Oguzkurt P, Ezer SS, Ince E, Hicsonmez A. Predictability results of the World Society of Emergency Surgery consensus confer-
of outcome of caustic ingestion by esophagogastroduodenoscopy ence. World J Emerg Surg. 2015;10:44. eCollection 2015.
in children. World J Gastroenterol. 2012;18(10):1098-1103. 57. Tiryaki T, Livanelioglu Z, Atayurt H. Early bougienage for relief of
32. Boskovic A, Stankovic I. Predictability of gastroesophageal caustic stricture formation following caustic esophageal burns. Pediatr Surg
injury from clinical findings: is endoscopy mandatory in children? Int. 2005;21(2):78-80.
Eur J Gastroenterol Hepatol. 2014;26(5):499-503. 58. Contini S, Garatti M, Swarray-Deen A, Depetris N, Cecchini S,
33. Bosnali O, Moralioglu S, Celayir A, Pektas OZ. Is rigid endoscopy Scarpignato C. Corrosive oesophageal strictures in children: outcomes
necessary with childhood corrosive ingestion? A retrospective after timely or delayed dilatation. Dig Liver Dis. 2009;41(4):263-268.
comparative analysis of 458 cases. Dis Esophagus. 2016;30(3):1-7. 59. Nijhawan S, Udawat HP, Nagar P. Aggressive bougie dilatation
34. Chiu HM, Lin JT, Huang SP, Chen CH, Yang CS, Wang HP. Predic- and intralesional steroids is effective in refractory benign esopha-
tion of bleeding and stricture formation after corrosive ingestion geal strictures secondary to corrosive ingestion. Dis Esophagus.
by EUS concurrent with upper endoscopy. Gastrointest Endosc. 2016;29(8):1027-1031.
2004;60(5):827-833. 60. Broto J, Asensio M, Vernet JM. Results of a new technique in the
35. Kamijo Y, Kondo I, Kokuto M, Kataoka Y, Soma K. Miniprobe treatment of severe esophageal stenosis in children: poliflex stents.
ultrasonography for determining prognosis in corrosive esophagitis. J Pediatr Gastroenterol Nutr. 2003;37(2):203-206.
Am J Gastroenterol. 2004;99(5):851-854. 61. Atabek C, Surer I, Demirbag S, Caliskan B, Ozturk H, Cetinkursun
36. Kochhar R, Poornachandra KS, Puri P, et al. Comparative evaluation S. Increasing tendency in caustic esophageal burns and long-term
of nasoenteral feeding and jejunostomy feeding in acute corrosive polytetrafluorethylene stenting in severe cases: 10 years experience. J
injury: a retrospective analysis. Gastrointest Endosc. 2009;70(5):874-880. Pediatr Surg. 2007;42(4):636-640.
Caustic Esophageal Injury CHAPTER 47 525
62. Foschia F, De Angelis P, Torroni F, et al. Custom dynamic stent for after total laryngopharyngo-oesophagectomy. Eur J Cardiothorac Surg.
esophageal strictures in children. J Pediatr Surg. 2011;46(5):848-853. 2013;44(2):258-262.
63. Wang RW, Zhou JH, Jiang YG, et al. Prevention of stricture with 69. Chirica M, Brette MD, Faron M, et al. Upper digestive tract reconstruc-
intraluminal stenting through laparotomy after corrosive esophageal tion for caustic injuries. Ann Surg. 2015;261(5):894-901.
burns. Eur J Cardiothorac Surg. 2006;30(2):207-211. 70. Kim SH, Kim HK, Kim K, Shim YM. Outcome of free jejunal transfer
64. Okonta KE, Tettey M, Abubakar U. In patients with corrosive using the end-to-side arterial anastomosis technique as a pharyngo-
oesophageal stricture for surgery, is oesophagectomy rather than oesophageal substitute: a 15-year experience. Eur J Cardiothorac Surg.
bypass necessary to reduce the risk of oesophageal malignancy? 2013;44(3):520-524.
Interact Cardiovasc Thorac Surg. 2012;15(4):713-715. 71. Sa YJ, Kim YD, Kim CK, Park JK, Moon SW. Recurrent cervical
65. Gvalani AK, Deolekar S, Gandhi J, Dalvi A. Antesternal colonic esophageal stenosis after colon conduit failure: use of myocutaneous
interposition for corrosive esophageal stricture. Indian J Surg. flap. World J Gastroenterol. 2013;19(2):307-310.
2014;76(1):56-60. 72. Imaizumi A, Liem AA, Yang CF, Chen W, Chen SH, Chen HC.
66. Harlak A, Yigit T, Coskun K, et al. Surgical treatment of caustic Long-term outcomes of simultaneous skin and bowel flaps for
esophageal strictures in adults. Int J Surg. 2013;11(2):164-168. esophageal reconstruction. Ann Plast Surg. 2015;75(2):180-185.
67. Boukerrouche A. Left colonic graft in esophageal reconstruc- 73. Millar AJ, Cox SG. Caustic injury of the oesophagus. Pediatr Surg
tion for caustic stricture: mortality and morbidity. Dis Esophagus. Int. 2015;31(2):111-121.
2013;26(8):788-793. 74. Iskit SH, Ozcelik Z, Alkan M, Turker S, Zorludemir U. Factors
68. Huang PM, Chen CN, Yang TL, Ko JY, Lee JM, Cheng NC. affecting the prevalence of gastro-oesophageal reflux in child-
Supercharged reversed gastric tube technique: a microvascular hood corrosive oesophageal strictures. Balkan Med J. 2014;31(2):
anastomosis procedure for pharyngo-oesophageal reconstruction 137-142.
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
CHAPTER
P
erforation of the esophagus is a potentially serious of esophageal perforation include foreign body ingestion,
and life-threatening medical emergency. Given blunt or penetrating trauma, iatrogenic operative injury,
the diverse etiologies and wide variety in clinical tumor, and tumor necrosis following cancer-related therapy,
presentations of the perforated esophagus, the managing such as radiation or chemotherapy.
physician must possess a thorough understanding of the
principles behind its treatment, as well as have access
to an array of therapeutic tools, to provide an optimal
PRESENTATION
outcome. Surgical therapy has occupied a prominent Esophageal perforations are quite variable in their mani-
place in the management armamentarium of esophageal festations and presentations depending upon multiple
t
perforation since Norman Barrett reported the first case factors, including:
ne
of successful primary repair in 1946.1 Over the subsequent • Etiology
decades, surgical techniques have been developed and • Size
refined, and improvements in antibiotics, critical care, • Location
k.
radiologic imaging, and percutaneous interventions have • Associated esophageal pathology
evolved. More recently, nonoperative and endoscopic • Time interval since perforation
oo
approaches have been introduced as alternatives to surgery • Extent of neck, mediastinal, pleural or abdominal
in appropriately selected patients, further improving the contamination
ability to treat this condition.2 Despite these advances, • Patient comorbidities
yb
the morbidity and mortality following perforation of the The most common presenting symptom is chest pain,
esophagus remain high, especially in cases of diagnostic although complaints of odynophagia, neck or abdominal
delay, underscoring the importance of prompt recognition pain, dyspnea, crepitus, fevers, and chills may occur in
and appropriate treatment of this malady. Therefore addition or in isolation. Symptoms may be mild, particularly
er
considerable clinical judgment is required on the part if the perforation is contained or of recent onset. Within
of the managing physician because treatment decisions the first 8 to 24 hours following injury, frank sepsis with
rg
may have a major impact on outcomes. tachycardia, hypotension, altered mental status and respira-
tory failure may develop, particularly if the contamination
is extensive. Given the potentially serious consequences of
ETIOLOGY
su
1724. He described the demise of Barron van Wassenaer, recent onset, especially following vomiting or a history of
the Grand Admiral of the Dutch fleet, after an episode of recent esophageal instrumentation.
tp
self-induced vomiting following a large meal. Spontaneous During the initial evaluation, a thorough history focused
esophageal perforation, which has come to be known as on preexisting dysphagia, heartburn, or regurgitation,
Boerhaave syndrome, is due to an abrupt increase in the as well as prior esophageal surgery or known esophageal
ht
intraluminal pressure of the esophagus during forceful disorders, is important because concomitant esophageal
emesis leading to full-thickness rupture of the esophageal pathology may have contributed to the perforation or
wall. The perforation is generally toward the left and may may influence the treatment paradigm.
involve the thoracic or abdominal portion of the esophagus.
Fortunately, Boerhaave syndrome remains relatively
uncommon, accounting for only 15% of esophageal
DIAGNOSIS
perforations.3 With the frequent utilization of flexible After a detailed clinical history and physical examination
endoscopy and its adjuncts for esophageal diagnostics and have been completed, the most useful and expeditious
therapeutics, the most common cause of esophageal per- initial screening examination for evaluation of the
foration currently is iatrogenic injury following esophageal patient with a suspected esophageal perforation is a plain
instrumentation, accounting for nearly 60% of all cases. upright chest radiograph. This quick, readily available,
Given the elective nature of most such interventions and and inexpensive study may reveal a pleural effusion,
the fact that the perforations are often both immediately pneumothorax, pneumoperitoneum, subcutaneous or
recognized and occur with minimal initial contamination mediastinal emphysema, or mediastinal widening suggestive
of surrounding tissues due to an empty stomach, the of perforation. However, of importance, is the fact that
outcomes may be better than those following spontaneous a normal chest x-ray does not rule out the possibility
perforation, where diagnosis may be delayed and extensive of a leak, especially in the early time period after the
contamination may occur. Other contemporary etiologies suspected event, because it may be contained or inadequate
526
Etiology and Management of Esophageal Perforation CHAPTER 48 526.e1
ABSTRACT
Perforation of the esophagus is a potentially serious and
life-threatening medical emergency. Given the diverse
etiologies and wide variety in clinical presentations of the
perforated esophagus, the managing physician must possess
a thorough understanding of the principles behind its
treatment, as well as have access to an array of therapeutic
tools, to provide an optimal outcome. Surgical therapy
has occupied a prominent place in the management
armamentarium of esophageal perforation since Norman
Barrett reported the first case of successful primary repair
in 1946. Over the subsequent decades, surgical techniques
have been developed and refined, and improvements in
antibiotics, critical care, radiologic imaging, and percutane-
ous interventions have evolved. More recently, nonoperative
and endoscopic approaches have been introduced as
alternatives to surgery in appropriately selected patients,
further improving the ability to treat this condition. Despite
t
ne
these advances, the morbidity and mortality following
perforation of the esophagus remain high, especially in
cases of diagnostic delay, underscoring the importance
of prompt recognition and appropriate treatment of
k.
this malady. Therefore considerable clinical judgment is
required on the part of the managing physician because
oo
treatment decisions may have a major impact on outcomes.
KEYWORDS
yb
Esophagus
esophageal perforation
Boerhaave syndrome
er
esophageal stents
endoscopic clips
su
t
ne
FIGURE 48.2 Computed tomography of the chest revealing a distal
esophageal perforation with associated pleural effusion. Note the
leakage of orally administered contrast.
k.
oo
abdomen, or neck, as well as the size of associated fluid
collections. A negative study with water-soluble contrast
should be followed by one using thin barium to increase
yb
FIGURE 48.1 Lateral view of a Gastrografin esophagram the sensitivity of the examination. Films also should be
demonstrating an esophageal perforation (arrow) with taken with the patient in both the left lateral and right
extravasation of contrast into the mediastinum. (From Soose RJ, lateral decubitus positions because proper positioning
also may increase the sensitivity of the study in detecting
er
of 10% to 38%.4,5
Computed tomography (CT) has proven to be an
time may have elapsed to allow for the development of a extremely useful diagnostic modality for assessing esopha-
su
recognizable radiographic abnormality. geal perforations and guiding their management (Fig.
The most frequently used radiologic test specifically 48.2). Findings on CT suggestive of perforation include
to assess for esophageal perforation has been the con- pneumomediastinum, pneumoperitoneum, subcutaneous
://
trast esophagram (Fig. 48.1). The study traditionally is emphysema, mediastinal fluid or inflammation, pleural
commenced with a water-soluble contrast agent, such as effusion, or abdominal abscess. Importantly, CT provides
tp
diatrizoate meglumine and diatrizoate sodium solution accurate information regarding the location and size of
(Gastrografin), due to a concern for exacerbating medi- any extraesophageal fluid collections in need of operative
astinal, pleural, or abdominal contamination if barium intervention or percutaneous drainage.5,6
ht
were used and leaked. However, contrast esophagography The finding of frank leakage of oral contrast is not
requires an alert and cooperative patient who is able sensitive and should not be the sole criterion to prove
to swallow without aspirating. In the intubated patient, or disprove the diagnosis.
contrast can be administered via a nasoesophageal tube. Flexible fiberoptic upper endoscopy serves a critical
A risk of water-soluble contrast is the potential to induce role in the assessment of a wide array of esophageal
a severe chemical pneumonitis if aspirated into the lung pathology, including perforations (Fig. 48.3). Not only
parenchyma. Therefore judgment must be exercised in does endoscopic assessment facilitate the determina-
using this study for the elderly patient or others considered tion of the location and size of a mucosal injury, it also
at high risk for aspiration, because a resultant pulmonary allows identification of concomitant mucosal ischemia or
injury could be a catastrophic insult added to the morbidity ulceration, as well as more chronic or subacute pathology
of the perforation. such as a stricture, diverticulum, or malignancy. Some
The exam typically provides a reliable assessment of perforations are subtle and do not demonstrate an
esophageal anatomy, the location and size of a perforation, obvious mucosal tear but rather only an ecchymotic or
and the presence of significant coexisting esophageal slightly disrupted mucosa that flutters with insufflation.
pathology, such as an esophageal stricture, malignancy, Although concern may exist about the safety of flexible
diverticulum, or motility disorder. In addition, a determina- endoscopy in the setting of acute perforation, given the
tion can be made as to whether the perforation is contained need for insufflation and the risk of inducing a tension
or freely leaking into the mediastinum, pleural spaces, pneumothorax or exacerbating pneumoperitoneum,
528 SECTION I Esophagus and Hernia
t
FIGURE 48.3 Flexible upper endoscopy revealing a full-thickness fluid collections, and provision of nutritional support. Any
ne
perforation of the esophagus after dilation of a stricture. management strategy, whether nonoperative, endoscopic,
or operative, must include these essential components
of therapy.
k.
the examination typically can be completed safely in After an esophageal perforation is considered in the
experienced hands and with appropriate attention to differential diagnosis, prior to a definitive diagnosis being
oo
detail. Insufflation should be kept to a minimum, and reached, initial interventions should include avoidance of
consideration should be given to placement of a chest food or liquids by mouth, administration of intravenous
tube prior to the procedure if concern exists about the fluids, and initiation of broad-spectrum antibiotics. An
yb
consequences of a pneumothorax. extensive body of literature has shown a significant decrease
No single diagnostic study is absolutely reliable in the in sepsis-related mortality with the expeditious administra-
evaluation of esophageal perforation. Esophagography tion of appropriate antibiotic therapy.7,8 Leaked enteric
with water-soluble contrast agents is limited by its false contents cause a chemical burn in the surrounding tissues
er
negative rate and risk of aspiration pneumonitis, as well as of the neck, mediastinum, pleural spaces, or peritoneal
by its relative insensitivity in assessing mucosal pathology. cavity and may lead to sequestration of large amounts of
rg
In addition, leaked barium risks exacerbating mediastinal, fluid, further exacerbating the hypotension that results
pleural, or abdominal sepsis. CT lacks sensitivity in locating from sepsis. Antibiotics should be directed toward enteric
the exact site of a perforation and detecting concomitant organisms, including gram-positive, gram-negative, and
su
esophageal pathology. Endoscopy is invasive, requires anaerobic bacteria, as well as fungi. Antifungal therapy is
sedation, risks exacerbating pneumothorax or pneumo- particularly relevant in individuals with a recent history of
peritoneum, and does not facilitate a determination of using proton pump inhibitors because they are known to
://
the extent of extraesophageal contamination. increase the risk of fungal colonization in the stomach.9
As a result of the limitations of each of these diagnostic Closed tube thoracostomy should be considered early to
tp
modalities, considerable judgment is required on the part drain large pleural effusions or treat a pneumothorax,
of the managing physician to determine the sequence and even while preparations are made for definitive diagnosis
optimal utilization of studies, taking into consideration and intervention.
ht
can be treated with intravenous antibiotics and nothing Advances in engineering and manufacturing led to the
by mouth with close observation in a monitored setting. development of self-expanding metallic stents (SEMSs)
The duration of therapy is dictated by the patient’s clinical in the 1990s. These stents are constructed of steel alloys,
course. Follow-up endoscopy or radiographic imaging is such as nitinol (nickel and titanium) or elgiloy (cobalt,
useful in determining the resolution of the perforation nickel, and chromium), and are woven, knitted, or cut
and the timing of resumption of an oral diet. Of course, a into a mesh. They are introduced into the esophagus over
clinical decline with the development of sepsis mandates a guidewire in a compressed state using a delivery system.
an expeditious investigation and strong consideration of Upon deployment, they are designed to self-expand up
more invasive therapeutic measures. to a predetermined diameter. Advantages of such stents
compared with prior versions include the ability to use
ENDOSCOPIC MANAGEMENT flexible upper endoscopy for placement and the smaller
Advances in endoscopic technologies have led to novel esophageal luminal diameter requirements for insertion.
techniques for management of esophageal perforations and Subsequent versions of SEMSs incorporated a partial
their sequelae. Currently available endoscopic modalities covering of polyurethane, silicone, or other polymers to
used in the management of perforations include endolumi- prevent tumor ingrowth along most of the stent length.
nal suturing (OverStitch; Apollo Endosurgery, Inc., Austin, However, the proximal and distal 1.5 cm were left bare
Texas), through-the-scope (TTS) clips, over-the-scope to facilitate better adherence to the mucosa and allow
clips (OTSC System; Ovesco Inc., Tübingen, Germany), ingrowth of granulation tissue as a means to achieve stent
t
endoscopic vacuum therapy (EVT), and the placement of fixation. This construct helps to prevent migration but
ne
covered esophageal stents. These technologies have been also makes extraction at a later date difficult or impos-
used alone or in combination as part of an endoscopic sible without risking significant injury to the esophagus
treatment paradigm, and as either primary therapy or and surrounding structures. Due to their potential for
k.
salvage after failed surgical repair of an esophageal rupture. erosion when left in place for extended periods of time,
The most extensive experience to date has been with partially covered stents are indicated only for cases of
oo
stenting. end-stage malignancy when life expectancy is short or in
The first report of successful esophageal stent placement the preoperative setting for resectable cancer with the
was by Symonds in 1887 using a prosthesis made of ivory expectation that the esophagus and stent subsequently
and silver.12 Multiple varieties of rigid, hollow prostheses
yb
will be removed.
placed with the assistance of transoral delivery systems With advancements in materials, the next iteration in
were introduced much later by Mousseau et al., Celestin, stent technology was the introduction of solid, collaps-
and Atkinson et al., among others.13-15 Because of their ible, self-expanding plastic stents (SEPSs) coated with
er
large diameter, bulk, and stiff construct, all of these stents silicone. These devices can be inserted in a compressed
were difficult and hazardous to place and extract. Rigid state with minimal esophageal dilation, deployed within
rg
stents were used traditionally for palliation of dysphagia the esophageal lumen, and used to occlude esophageal
in patients with locoregionally advanced, obstructing perforations or fistulas. Because these stents are solid,
esophageal neoplasms and a short life expectancy. Versions ingrowth of tumor or granulation tissue is not a concern,
su
with a soft outer sponge encasing the main shaft also although the decreased fixation increases the rate of stent
were created for occlusion of esophageal perforations migration.16 However, they possess the advantage over
and fistulas (Fig. 48.4). partially covered SEMSs of being removable at a later date.
://
t
design and sizing. Historically, primary repair was the most frequently used
ne
therapy of the perforated esophagus; it remains the gold
standard against which other forms of treatment should be
significant patient discomfort, or those spanning the gas- measured. The traditional dogma was that repair should
k.
troesophageal junction that are impossible to occlude due be undertaken only within the first 24 hours following
to the bulbous nature of the gastric cardia. A perforation a perforation, given concerns about tissue friability and
oo
occurring in a dilated esophagus, such as may be found poor healing in cases in which inflammation and infection
in the setting of achalasia, may also prove impossible to were of a longer duration. Data from the 1980s and
occlude with a stent. Even in a normal esophagus, satisfac- early 1990s suggested that leak rates following primary
yb
tory occlusion of a leak may not be achieved, particularly closure were higher when operation was delayed beyond
if the stent is malpositioned or is of an inappropriate size 24 hours compared with earlier repair.19 Over time, and
or length. with increasing experience, primary repair was advocated,
Complications associated with esophageal stents include in cases of treatment delayed beyond 24 hours, to avoid
er
ongoing or recurrent esophageal leakage, migration, the morbidity associated with esophageal resection or
luminal obstruction, erosion, pain, bleeding, and inextract- exclusion and the need for a second, major reconstructive
rg
ability at a later date (Fig. 48.5). In addition, for stents procedure at a later date.20,21
crossing the esophagogastric junction and compromising Primary surgical repair is best performed in two layers,
the lower esophageal high-pressure zone, the induction the first consisting of the mucosa and submucosa and
su
or exacerbation of gastroesophageal reflux is a significant the second of the overlying circular and longitudinal
concern. The endoscopist also must be alert to the possibil- muscle layers of the esophagus. In addition, buttressing
ity of isolating an extraluminal fluid collection, preventing of the suture lines with vascularized tissue such as pleura,
://
it from draining back into the esophagus. The timing pericardial fat, intercostal muscle, omentum, or gastric
of stent removal is a matter of judgment, balancing the fundus is frequently added (Fig. 48.6). For perforations of
tp
likelihood of healing of the perforation against potential the distal intrathoracic esophagus, a left thoracotomy via
complications, such as erosion or obstruction, which may the seventh intercostal space is generally used. For perfora-
arise the longer the stent is left in place. tions of the proximal to mid intrathoracic esophagus, a
ht
Persistent or recurrent esophageal leaks after stenting right thoracotomy via the fourth to sixth intercostal space
can occur for a multitude of reasons, including malposi- is preferable. Of course, perforations of the cervical or
tioning, incorrect size leading to inadequate proximal or intraabdominal esophagus require incisions in the neck
distal stent apposition to the esophageal mucosa, tearing or abdomen, respectively.
or degradation of the stent covering, leakage between A key principle underlying successful esophageal repair
two separate stents, and stent migration.18 A number of is that the deeper mucosal defect commonly extends
techniques have been advocated to prevent “stent leaks” beyond the more superficial muscular one. A myotomy
due to migration, including bridling and endoscopic created proximal and distal to the injury is generally neces-
clipping or suturing, although success rates have varied sary to identify the mucosal edges and facilitate complete
and reports proving efficacy are lacking. closure (Fig. 48.7). The mucosa, once adequately exposed,
Over the past few years, EVT has been advocated as should be débrided back to healthy, noninflamed tissue
a promising modality for managing esophageal perfora- and reapproximated with absorbable or nonabsorbable
tions and contiguous fluid collections. Similar to vacuum suture. The muscular layer can be closed more superficially
therapy for superficial wound infections, EVT involves with interrupted sutures. Care must be taken to prevent
application of negative pressure in the region of the luminal narrowing in the process of closure; use of a
esophageal defect, administered via an electronically transoral bougie is a consideration to ensure an adequate
controlled vacuum device attached to a nasoesophageal esophageal diameter. The repair can be buttressed with
tube-mounted polyurethane sponge. No commercially adjacent viable tissue, as described.
Etiology and Management of Esophageal Perforation CHAPTER 48 531
Pleural flap
t
ne
k.
oo
yb
er
C
rg
FIGURE 48.6 Pleural flap coverage of a large esophageal defect. (A) After mobilization of the esophagus, a pleural flap is raised. (B) The
flap is wrapped around the esophagus, covering the perforation with or without primary repair of the esophageal defect. (C) The flap is
su
sutured to itself. Sutures are placed proximally and distally at the margins of the flap, as well as to the perforation itself (if not primarily
closed), to tack the pleura firmly to the esophageal musculature. (From Grillo HC, Wilkins FW. Esophageal repair following late diagnosis
of intrathoracic perforation. Ann Thorac Cardiovasc Surg. 1975;20:337, by permission of the Society of Thoracic Surgeons.)
://
tp
At the time of repair, washout with decortication, of the perforated esophagus, as well as irrigation and
débridement, and drainage of the contaminated spaces drainage of the infected pleural and mediastinal spaces
is also important. A drain should be placed near, but and decortication of the ipsilateral lung. Thus thoracotomy
ht
not directly abutting, the esophageal suture line in case or thoracoscopy incisions generally are made on the side
a recurrent leak develops in the early postoperative of a pleural effusion. During the abdominal phase of the
period. In addition, a feeding tube should be placed to operation, a gastric or jejunal feeding tube should be
facilitate nutritional support while the esophagus heals. placed for postoperative nutritional support. A transhiatal
The tube may be placed into the stomach or jejunum, resection may be the best choice in some cases, avoid-
either operatively (via laparotomy or laparoscopy) or using ing the morbidity of chest incisions and the need for
percutaneous techniques under endoscopic guidance, single-lung ventilation when a transthoracic operation is
depending upon surgeon preference, available resources, undertaken.
and the clinical circumstances. If sepsis has arisen or is imminent because of a delay in
diagnosis or the magnitude of extraluminal contamination,
Esophagectomy foregut reconstruction should be delayed due to concern
If clinical judgment determines that esophageal repair of conduit ischemia from sepsis-related hypoperfusion
or stenting is not feasible, or the esophagus is deemed and resultant anastomotic breakdown. In this situation
unsalvageable as in cases of end-stage achalasia, esophageal a cervicothoracic end esophagostomy should be created
resection may be the best option in selected cases of to allow drainage of oral secretions. Important surgical
esophageal perforation. A transthoracic approach, either details include a left neck approach to facilitate dissection
open or performed with minimally invasive techniques, of the cervical esophagus, protection of the recurrent
is commonly used because it allows for both the removal laryngeal nerves, and preservation of as much proximal
532 SECTION I Esophagus and Hernia
Extent of
mucosal tear
t
submucosal plane to separate the mucosa
ne
from the muscularis propria, improving
exposure of the limits of the mucosal defect.
(From Whyte RI, Iannettoni MD, Orringer MB.
k.
Intrathoracic esophageal perforation: the merit
of primary repair. J Thorac Cardiovasc Surg.
oo
1995;109:140.)
yb
esophagus as possible to aid in future reconstruction. A occur, resulting in ongoing soilage if the perforation has
long length of esophageal remnant may permit place- not healed. Esophageal diversion also runs the risk of
ment of the esophagostomy on the chest wall, allowing a pooled secretions within the esophageal lumen, leading
er
more secure application of an external drainage bag and to retching or nausea, and rarely should be necessary.
increased patient comfort compared to a stoma placed
rg
tip of the esophagostomy is not uncommon, leading to as well. Cervical perforations are generally well tolerated
the need for dilation or surgical revision in a significant because fascial investments surrounding the esophagus
tp
proportion of cases (unpublished data). in the neck promote containment. Many pharyngeal or
cervical esophageal leaks resolve with antibiotics alone. If
Esophageal Diversion the clinical condition deteriorates, a neck abscess develops,
ht
The most definitive method to divert the flow of gastro- or fluid tracks into the mediastinum, surgical management
intestinal contents from the mediastinum is esophagectomy should be pursued, including wide drainage through a
with end esophagostomy. However, some surgeons have cervical incision. In most operative situations, drainage is
advocated proximal and distal esophageal diversion sufficient and attempts at direct visualization and repair
without esophagectomy to facilitate foregut reconstruc- of the perforation may prove difficult, unsuccessful, and
tion after recovery. Proximal diversion is most commonly unnecessary.
accomplished by end cervical esophagostomy,22 although Because there are typically only a few centimeters of
placement of an esophageal T-tube23 and side cervical intraabdominal esophagus, perforations into the peritoneal
esophagostomy24 have been described as alternatives. The cavity are less common than those occurring into the
latter option may be technically challenging, especially thorax. However, abdominal perforations can occur,
in the setting of obesity or when the patient’s neck is particularly with endoscopic interventions aimed at the
otherwise thick. Distal diversion can be achieved by surgical lower esophageal sphincter, such as pneumatic dilation
division at the gastroesophageal junction using a linear or peroral endoscopic myotomy (POEM) for achalasia,
cutting stapler or, alternatively, using a noncutting stapler balloon or rigid dilation for benign strictures or tumors,
without division of tissues. This latter technique for distal or a variety of endoscopic mucosal resective or ablative
diversion has been touted to make future reconstruction technologies. An important anatomic consideration in
easier, although it should be used with caution. Early patients with sliding or paraesophageal hernias is that the
recanalization of the stapled, undivided esophagus can hernia sac is bounded cranially by the phrenoesophageal
Etiology and Management of Esophageal Perforation CHAPTER 48 533
ligament, rendering the herniated distal esophagus an hours after a perforation often cited as the critical window.
intraabdominal structure. For all of these reasons, a distal Mortality rates in the setting of treatment occurring after
esophageal perforation may need to be addressed by 24 hours are quoted to be double of those recognized
laparotomy or laparoscopy. and managed earlier.2,25,26 Similarly, the mortality following
Preoperative CT imaging may assist in the decision spontaneous perforations (36%) is almost double that
as to whether a distal perforation is best approached from iatrogenic injuries (19%), likely representing the
through the chest or through the abdomen. When repair more rapid diagnosis in the latter cohort and the fact
is performed below the diaphragm, the stomach can be that such perforations typically occur in the setting of
used as a buttress by fashioning a partial or complete an empty stomach.2
fundoplication to cover the repair site suture line. Although Nonoperative approaches in carefully selected patients
placing a fundoplication above the hiatus in the chest have shown superior outcomes to cohorts undergoing
has been described, creating an iatrogenic hiatal hernia, surgery, consistent with the less severe nature of the
common sense dictates that such repairs may lead to perforations in the former and the morbidity associated
the problems inherent in any herniated fundic wrap with operative intervention. Altorjay et al. reported on a
including the potential for ischemia, pain, obstruction, series of 20 patients treated with an initial nonoperative
or gastroesophageal reflux. paradigm.11 Sixteen patients (80%) were managed success-
fully; four (20%) required operative intervention, after
PREEXISTING ESOPHAGEAL PATHOLOGY which two patients (10%) died. The morbidity rate was
t
Esophageal perforation may occur due to, or in the face 20%. The authors commented that the mortalities resulted
ne
of, preexisting esophageal disease; both entities must from an incorrect decision to manage in a nonoperative
be considered when devising a treatment plan. The fashion; the patients were negatively impacted by the delays
concomitant pathology may influence the applicability in operative intervention, underscoring the importance of
k.
of primary repair or esophageal stenting and whether appropriate patient selection. More recently, Keeling et al.
esophageal resection is advisable. reported on a series of 25 patients treated nonoperatively
oo
A classic example is the patient with achalasia perforated using strict criteria similar to those described in Table
at time of pneumatic dilation. Repair of the perforation 48.1.27 Their treatment algorithm resulted in a mortality of
alone will not ameliorate the dysphagia and regurgita- 8% and morbidity of 48%, with the two deaths occurring
yb
tion associated with the underlying disease. Of equal in patients with metastatic esophageal cancer who refused
importance, the unrelieved distal esophageal obstruc- operative intervention.
tion imposed by a poorly relaxing or hypertensive lower Analyses of the surgical treatment of esophageal perfora-
esophageal sphincter may increase the risk of postopera- tion have shown widely varied results. In a meta-analysis
er
tive leak from the repair site. In this circumstance the of 572 patients published in 2004 the mortality rates
best approach is to close the perforation in two layers, ranged from 0% to 80% among the case series reviewed.2
rg
perform a distal esophageal (modified Heller) myotomy For patients undergoing primary repair (n = 322) the
on the contralateral side of the esophagus away from average mortality was 12%. After esophagectomy (n = 129),
the perforation, and create a partial fundoplication that mortality was 17%, whereas after esophageal exclusion (n
su
serves as both an antireflux valve and a buttress of the = 34), mortality was 24%. Surgical drainage alone (n =
mucosal repair. In patients with end-stage achalasia or a 88) was associated with a 36% mortality rate.
recalcitrant esophageal stricture, repair alone may lead Since the introduction of esophageal stents, enthusi-
://
to ongoing dysphagia or regurgitation; such individuals asm for their application in the treatment of esophageal
may benefit from esophagectomy. perforation has been increasing. Results from cohorts
tp
Patients with esophageal malignancy should be consid- treated endoscopically compare favorably to surgical repair.
ered for definitive surgical resection of their cancer at the A meta-analysis published in 2011 of 267 patients treated
time a perforation occurs, assuming clinical stability, lack with esophageal stenting found a success rate of 85%.28
ht
of metastatic disease, and both a sufficient performance Fifty-nine percent of patients required concurrent drainage
status and a reasonable life expectancy. Alternatively, a of extraesophageal fluid collections. Mortality for stented
stent may be placed for temporary occlusion of the leak patients was 13%, similar to patients undergoing surgical
with the option of esophagectomy at a later date when repair. Thirty-four percent of patients had a stent-related
the patient is more stable and a thorough preoperative complication, including stent migration in 29%, bleeding
evaluation has been completed. However, the presence of in 2%, and tissue overgrowth in 5%. Surgical intervention
tumor will lead to nonhealing of a perforation, making for incomplete occlusion or stent-related complications
subsequent stent extraction unfeasible without esophageal was necessary in 13% of patients. Patients at increased
resection. risk for failure of endoscopic management include those
with a perforation of the cervical esophagus or at the
gastroesophageal junction, and those with esophageal
OUTCOMES injuries longer than 6 cm.29
The results following therapy of esophageal perforation The role of endoscopic clipping as therapy for esopha-
historically were poor with significant morbidity and geal perforation using either TTS clips or OTSCs was
mortality. A meta-analysis of 726 patients treated for a assessed in a literature review.30 The analysis included 127
perforated esophagus between 1990 and 2003 revealed an patients from 38 articles and concluded that TTS clips
overall mortality rate of 18%.2 Delays in treatment clearly are effective in the treatment of early perforations (<24
are associated with a worse outcome, with the first 24 hours old) with limited contamination and measuring
534 SECTION I Esophagus and Hernia
less than 10 mm in length, whereas lesions up to 20 mm 2. Brinster CJ, Singhal S, Lee L, et al. Evolving options in the manage-
can be treated with OTS clipping. ment of esophageal perforation. Ann Thorac Surg. 2004;77:1475-1483.
3. Sepesi B, Raymond DP, Peters JH. Esophageal perforation: surgi-
Recent literature reviews of EVT for various types of cal, endoscopic and medical management strategies. Curr Opin
upper gastrointestinal defects assessed data on more than Gastroenterol. 2010;26:379-383.
200 patients, with success rates ranging from 70% to 4. Swanson JO, Levine MS, Redfern RO, Rubesin SE. Usefulness of
100%.31,32 Although the results of both endoscopic clipping high-density barium for detection of leaks after esophagogastrectomy,
total gastrectomy, and total laryngectomy. AJR Am J Roentgenol.
and EVT appear encouraging, they are limited to small 2003;181:415-420.
case series and retrospective cohort studies from highly 5. Fadoo F, Ruiz DE, Dawn SK, et al. Helical CT esophagography for
selected patient populations with limited extraluminal the evaluation of suspected esophageal perforation or rupture. AJR
contamination. Lacking a control population undergoing Am J Roentgenol. 2004;182:1177-1179.
expectant observation, the studies do not demonstrate the 6. White CS, Templeton PA, Attar S. Esophageal perforation: CT
findings. AJR Am J Roentgenol. 1993;160:767-770.
added value of these endoscopic interventions. 7. Ferrer R, Martin-Loeches I, Phillips G, et al. Empiric antibiotic
treatment reduces mortality in severe sepsis and septic shock from the
CONCLUSION first hour: results from a guideline-based performance improvement
program. Crit Care Med. 2014;42:1749-1755.
8. Dellinger RP, Levy MM, Rhodes A, et al. Surviving sepsis campaign:
Esophageal perforation remains a challenging clinical international guidelines for management of severe sepsis and septic
problem and a potentially life-threatening emergency. shock. Crit Care Med. 2012;2013(41):580-637.
Considerable skill, judgment, and creativity are necessary 9. Elsayed H, Shaker H, Whittle I, Hussein S. The impact of systemic
t
on the part of the managing physician in devising a treat- fungal infection in patients with perforated oesophagus. Ann R Coll
ne
ment strategy that effectively addresses the perforation Surg Engl. 2012;94:579-584.
10. Cameron JL, Kieffer RF, Hendrix TR, et al. Selective nonoperative
and its sequelae while minimizing morbidity. management of contained intrathoracic esophageal disruptions.
The tools residing within the surgeon’s armamentarium Ann Thorac Surg. 1979;27:404-408.
k.
continue to evolve, although the principles underlying 11. Altorjay A, Kiss J, Voros A, Bohak A. Nonoperative management of
therapy remain constant and guide decision making. The esophageal perforations. Is it justified? Ann Surg. 1997;225:415-421.
12. Symonds CJ. The treatment of malignant stricture of the oesophagus
oo
basic tenets of closing, occluding, exteriorizing, or diverting by tubage or permanent catheterism. Br Med J. 1887;1:870-873.
proximal and distal to the esophageal defect, draining 13. Mousseau M, LeForestier J, Barbin J. Role of permanent intubation
associated extraluminal fluid collections, alleviating distal in palliative treatment of esophageal cancer. Arch Mal Appar Dig Mal
yb
esophageal obstruction, treating infection with antibiotics, Nutr. 1956;45:208-214.
and providing supportive care including nutrition are 14. Celestin LR. Permanent intubation in inoperable cancer of
the oesophagus and cardia: a new tube. Ann R Coll Surg Engl.
critical whether an operative, endoscopic, or noninvasive 1959;25:165-170.
approach is pursued. A treatment plan must not only
er
29. Freeman RK, Ascioti AJ, Giannini T, Mahidhara RJ. Analysis of 31. Kuehn F, Loske G, Schiffmann L, Gock M, Klar E. Endoscopic
unsuccessful esophageal stent placements for esophageal perforation, vacuum therapy for various defects of the upper gastrointestinal
fistula, or anastomotic leak. Ann Thorac Surg. 2012;94:959-964, tract. Surg Endosc. 2017;doi:10.1007/s00464-016-5404-x; [Epub ahead
discussion 964–955. of print].
30. Lázár G, Paszt A. Mán E. Role of endoscopic clipping in the treatment 32. Newton NJ, Sharrock A, Rickard R, Mughal M. Systematic review of
of oesophageal perforations. World J Gastrointest Endosc. 2016;8:13- the use of endo-luminal topical negative pressure in oesophageal
22. leaks and perforations. Dis Esophagus. 2017;30:1-5.
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
CHAPTER
T
he incidence of esophageal perforations is on the results from a sudden increase in intraesophageal pressure.
rise. Iatrogenic causes remain the most common He described his findings in a 1724 pamphlet detailing
and continue to increase in an era of frequent use postmortem observations of Baron de Wassenaer, the
of endoscopy for diagnostic and therapeutic procedures. Grand Admiral of Holland, who suffered a fatal esophageal
Despite many advances in care, the mortality rate for an rupture as a result of self-induced vomiting in an attempt
esophageal perforation remains high, with some series to relieve discomfort following unrestrained consumption
citing 12% to 50%. of food.3 Although most widely perceived to be associated
The esophagus passes through the neck, chest, and with vomiting, spontaneous perforations may also be
abdomen, so surgeons managing perforations must be seen in situations of forceful valsalvae—weight lifting,
t
experienced with the unique anatomic considerations for child-birth, and defecation.
ne
approaching a perforation in any of these levels/locations. Esophageal disruption may occur in the setting of
Many factors must be considered when managing these penetrating or, less frequently, blunt trauma. Gunshot
patients, including acuity of presentation, contamination, wounds account for 75% of penetrating injury followed
k.
size of leak, cause of leak, and comorbid conditions. Those by stab wounds and other mechanisms.4 The majority of
caring for esophageal perforation must be experienced injuries involve the cervical esophagus. Unfortunately
oo
in endoscopic procedures, esophageal resection, and these injuries continue to carry a very high mortality
complex esophageal reconstruction. Many of the current with a large series published in 2013 by Patel et al.
techniques involve a hybrid approach of stenting with a citing 44%.5
yb
muscle buttress. Blunt trauma resulting in esophageal perforation
is exceedingly rare with only about 100 cases reported
in the literature. The mechanism of injury is debated
ANATOMIC CONSIDERATIONS and may be caused by anterior-posterior compression of
er
The esophagus is a long muscular tube that begins at the the esophagus between the sternum and spine, severe
pharynx and ends at the gastroesophageal junction past hyperextension, increased intraluminal pressure, and
rg
the crura of the diaphragm, thus passing through three chest compression with perforation occurring in a manner
anatomic fields. The esophagus lacks a serosal layer, making similar to Boerhaave’s, or possibly ischemic injury from
it more susceptible to leak and less forgiving with surgical rapid deceleration with delayed perforation.6
su
repair. The inner circular and outer longitudinal muscular Ingestion of caustic materials may result in severe
layers are often weakened by perforation and do not hold esophageal injury and in some cases perforation. The
sutures well. Perforations are often underestimated as the extent of injury is dependent on a variety of factors includ-
://
infection spreads through the submucosal plane and is ing type of material, amount consumed, and length of time
covered by muscular tissue. it is in contact with tissues. Caustic materials are generally
tp
The cause of perforation often dictates the location. categorized as acidic or alkali. Acids are generally poor to
Important anatomic landmarks to appreciate, as they are taste and irritating, resulting in smaller volume ingestion.
common sites of perforation, are described in Table 49.1. They cause a coagulative necrosis, form an eschar, and
ht
ABSTRACT
The incidence of esophageal perforations is on the rise,
and iatrogenic causes remain the most common and
continue to increase in an era of frequent use of endoscopy
for diagnostic and therapeutic procedures. Despite many
advances in care, the mortality rate for an esophageal
perforation remains high, with some series citing 20%. The
esophagus passes through the neck, chest and abdomen,
so surgeons managing perforations must be experienced
with the unique anatomic considerations for approach-
ing a perforation in any of these levels/locations. Many
factors must be considered when managing these patients,
including acuity of presentation, contamination, size of
leak, cause of leak, and comorbid conditions. Those
caring for esophageal perforation must be experienced
in endoscopic procedures, esophageal resection, and
complex esophageal reconstruction.
t
KEYWORDS
ne
Esophagus, perforation, esophagectomy, esophageal stent,
tissue regeneration
k.
oo
yb
er
rg
su
://
tp
ht
Management of Esophageal Perforations and Leaks CHAPTER 49 537
t
Unique problems specific to this location include complete vascular rings, a right-sided aortic arch, and
ne
the classic dysphagia lusoria (arteria lusoria), which is impairment in swallowing due to compression
from an aberrant right subclavian artery.
Gastroesophageal The lower esophageal sphincter is another location where iatrogenic injury often occurs; however, the
k.
junction most common etiology of perforation here remains Boerhaave’s. This unfortunately can result in
contamination of the mediastinum, bilateral pleural spaces, and the abdomen because of its location.
oo
NPO, Nil per os.
yb
in inflammation or perforation and leak may also result immunocompromised patients may fail to have a classic
in stricture formation. presentation and often warrant a more aggressive imaging
approach with subtle signs such as a mere tachycardia,
PATIENT PRESENTATION denoting infection.
er
taneous esophageal rupture of a middle-aged man who and leaks as timing of intervention correlates with outcome.
consumes excessive food and alcohol, has active vomiting The work-up of any patient with suspected esophageal
and retching, and develops chest pain and subcutaneous perforation begins with a detailed history and physical
://
emphysema. The Anderson triad includes subcutaneous examination. Particular attention should be given to a
emphysema, rapid respirations, and abdominal rigidity. history of instrumentation, trauma, food consumption,
tp
Patients presenting with a high esophageal perforation— occupation, recent activity, and signs and symptoms of
in the neck or piriform sinus—may describe neck pain, malignancy such as weight loss or dysphagia. In patients
a change in their voice (generally a more nasal sound presenting with hemodynamic instability, this should be
ht
secondary to inflammation of the vocal cords), dysphagia, addressed immediately with placement of large bore IVs,
hemoptysis, or crepitus (a crunching sound or sensation fluid administration, and aggressive monitoring.
when pushing on the skin secondary to the accumulation If esophageal perforation is suspected, an anteroposterior,
of subcutaneous emphysema; Fig. 49.1). lateral upright chest, and abdominal radiographs should
Intrathoracic perforations often present with symptoms be obtained promptly. Subcutaneous emphysema, pneu-
such as chest or back pain, dysphagia, dyspnea, bleeding, momediastinum, new effusion, pneumothorax, and pleural
vomiting, or signs and symptoms of sepsis. Signs of an thickening are indicators of perforation. A plain radiograph
intrathoracic perforation include pleural effusion, pneu- may prove diagnostic in 80% of patients with suspected
mopericardium, pneumomediastinum, or pneumothorax. iatrogenic perforations. Radiographs are not only diagnostic
Intraabdominal perforations are more likely to present but also assist in localizing the defect; midesophageal
with abdominal pain and distention. Signs of an abdominal perforations manifest with right-sided efffusion, whereas
perforation of the esophagus include pneumoperitoneum, distal esophageal injuries show left-sided effusion.
or free fluid detected by either exam or imaging (Fig. 49.2). The gold standard diagnostic study remains a contrast
With an uncontained perforation, polymicrobial infec- swallow study with the treating surgeon present. The
tion with bacteria such as Staphylococcus, Pseudomonas, esophagram is performed fluoroscopically with the patient
Streptococcus, and Bacteroides typically occurs within the positioned obliquely in standing or semierect during
first 12 hours. Patients begin to develop tachycardia, swallowing of contrast. This facilitates the identification
fluid sequestration, fever, and a leukocytosis. In addition, of subtle leaks (Fig. 49.3). The false-negative rate of
538 SECTION I Esophagus and Hernia
t
Traumatic Penetrating or blunt trauma to neck or torso Strong association with neck hyperextension
ne
Cancer Erosion of an esophageal tumor Gas near or abutting the tumor on imaging
Extension of surrounding tumor through esophageal wall
Paraesophageal Incarceration with necrosis of the distal esophagus Evidence of left pleural effusion or abdominal fluid on
k.
hernia imaging studies
Foreign body Ingestion of a substance (i.e., chicken bone) that Upper esophageal impaction at the sphincter
oo
becomes lodged
Esophageal webs
Eosinophilic esophagitis
Esophagitis Inflammation and erosion of ulceration Immunocompromised patient
yb
Peptic ulcers
Zollinger–Ellison syndrome
Barrett ulcer
er
Potassium
Quinidine
su
NSAIDs
Sustained-release formulations
CMV, Cytomegalovirus; EMR, endoscopic mucosal resection; NSAIDs, nonsteroidal antiinflammatory drugs; POEM, per oral endoscopic myotomy.
://
tp
false-negative rate—extravasating in only 50% to 80% of setting of perforation must be approached cautiously by
cases of esophageal perforation—and carries a risk of an expert endoscopist, but skilled thoracic surgeons are
severe pneumonitis when aspirated. Barium has a higher often more comfortable performing thoracic irrigation
diagnostic accuracy; however, it persists in the space after in the setting of a chronic esophageal perforation and
imaging and may complicate further imaging, making a can often immediately follow with appropriate surgical
closed perforation appear present after it has sealed.11 intervention while under the same anesthesia. Patient
Patients who are unable to swallow or who are intubated should be intubated and under general anesthesia in
may alternatively have a computed tomography (CT) the operating room. In this setting, pleural effusions may
scan performed (Fig. 49.4). CT is also helpful in cases of also be sampled to determine if they are transudative or
suspected perforation that are unable to be identified on exudative effusions that could require additional treatment.
upper gastrointestinal (GI) evaluation, and a nasogastric Endoscopy is being used increasingly as an effective method
tube (NGT) should be placed under fluoroscopy to allow for the treatment of some patients with perforation.12
for contrast injection to better identify and delineate the Esophageal strictures, similarly, are worked up with
leak. It is important to ensure that endotracheal tube cuff an esophagram and endoscopy to denote the location
is inflated to prevent aspiration. CT scan is useful not and extent of esophageal involvement. Endoscopy is an
only in identifying the site of leak, but also in delineating important adjunct to the diagnosis to rule out malignancy
associated abscess of fluid collections that may require as the cause of esophageal narrowing appreciated on a
drainage. contrast study.
Management of Esophageal Perforations and Leaks CHAPTER 49 539
t
ne
A B
k.
FIGURE 49.1 A patient presenting with a piriform sinus perforation manifesting with severe subcutaneous emphysema. Image (A) exhibits
oo
extensive subcutaneous emphysema at the level of the neck and (B) demonstrates the emphysema to have dissected down to the level
of the mid-thorax.
yb
One of the initial steps in patient management is deter-
mining whether or not the esophagus is salvageable. The
traditional management of esophageal leaks and fistulas
er
t
ne
A
k.
oo
yb
er
rg
su
://
B
tp
FIGURE 49.3 Patient with chronic distal esophageal stricture underwent attempted dilation resulting in contained perforation. The
perforation was identified on esophagram (A). Endoscopy visualized the defect (B).
ht
Surgical Approach
Cervical Perforations. The preferred surgical approach for
cervical perforations is through a left neck, oblique incision
along the anterior border of the sternocleidomastoid
muscle. Small, contained perforations require only wide
drainage and GI rest. Larger perforations may be repaired
or require diversion. Despite our recommendation of
placing a diversion below the level of the clavicle on the
anterior chest wall, cervical perforations may have to be
diverted onto the neck (Fig. 49.7).
Thoracic Perforations. Thoracic perforations must be
divided into two categories based on proximal or distal
thoracic esophagus. Perforations in the upper two-thirds
of the thoracic esophagus can be approached surgically
through a right posterolateral thoracotomy. We prefer a
muscle-sparing thoracotomy in the fourth or fifth intercos-
tal space, and an intercostal muscle can be harvested upon
entry to provide a muscular flap for the repair (Fig. 49.8).
t
Ideally, a diverted ostomy should be brought out below
ne
the clavicle when the esophagus cannot be repaired. This
type of anterior chest wall diversion provides a longer
length of esophagus for future reconstruction and allows
k.
easier management of the ostomy device. Another benefit
FIGURE 49.5 Esophageal perforation localized endoscopically. of a distal diversion is patient satisfaction, as patients find
oo
Removable stent was placed to cover the defect. these devices are easily hidden under clothing and thereby
less socially isolating.
Perforations of the lower third of the esophagus are
yb
approached through a left posterolateral thoracotomy
through the sixth or seventh intercostal space. Similarly,
er
rg
Radiographic evaluation
://
*
Resuscitation Condition
Antibiotics worsens
Monitoring Resectable <48 h >48 h
carcinoma
Severe stricture
Caustic injury
Reinforced primary Resection/diversion
repair and wide with immediate
drainage of infected or delayed
Resection spaces reconstruction
*Endoscopic stenting may be used as an adjunct to surgery and/or the FIGURE 49.6 Management of
primary treatment in select patients esophageal perforations. CT,
Computed tomography.
542 SECTION I Esophagus and Hernia
Mouth Esophagus
Esophagostomy
Trachea
0 Vicryl
suture
tying off
esophagus
Diaphragm
t
ne
Stapled
end of
k.
Blue sutures esophagus
oo
yb
er
rg
su
://
FIGURE 49.7 Esophageal diversion is required in some cases of long-segment esophageal injury or those involving more than 50% of the
tp
circumference. When possible, the esophagostomy should be brought out onto the chest below the clavicle. This allows for greater ease
of care for the patient. (Illustration courtesy Mike DeLaFlor.)
ht
intercostal muscle flaps should be harvested as the chest space augments healing. Chest tube size should be tailored
is entered and sparing both the latissimus and serratus to intraoperative findings—tubes smaller than 32 French
each time the chest is entered allows the surgeon to plan can obstruct easily when frankly purulent material is found
for future complication management. at the time of surgery.
Video-assisted thoracoscopic surgery (VATS) should be Jejunostomy and gastrostomy tubes should be consid-
reserved for early presentation if adequate débridement ered to facilitate gastric drainage and enteral feeding in
can be achieved.15 The disadvantage of a VATS approach any esophageal perforation case but is necessary for the
is the lack of the ability to harvest an intercostal muscle diverted patient or in whom prolonged nil per os status
flap upon entry for buttressing. is anticipated. The tubes should be placed in such a way
Abdominal Perforations. Abdominal perforations can that it does not compromise conduit preparation for
be approached laparoscopically or through a midline reconstruction.
abdominal incision. Again, here the principal goal is Diligent monitoring of tubes and drains as well as
débridement and repair versus drainage. checking of labs and radiographs is essential. Do not
Adjunctive Considerations. Pulmonary decortication at remove tubes early. We prefer to bridal our NGT to avoid
the time of surgery is important to facilitate pulmonary inadvertent removal. Counseling patients, families, and
reexpansion for respiratory stability. Furthermore, it is nursing staff about the complexity of replacing inadver-
believed that expansion of the lung to minimize pleural tently removed tubes is time well spent.
Management of Esophageal Perforations and Leaks CHAPTER 49 543
A B
t
ne
k.
oo
yb
er
C
rg
FIGURE 49.8 A patient who presented with a thoracic esophageal perforation just proximal to the gastroesophageal junction (A). Surgical
su
approach was left thoracotomy, débridement of the esophagus (B) with primary repair, and buttressing with an intercostal muscle
flap (C).
://
Broad-spectrum antibiotic therapy should be continued been published, the results have been promising in some
until the sensitivities of offending agents are confirmed. circumstances with low procedural morbidity, low overall
tp
Microbes commonly responsible for infections related to patient mortality, and high success in reestablishment of
esophageal perforations include Staphylococcus, Pseudomonas, intestinal continuity.16,17
Streptococcus, and Bacteroides, and adequate coverage for each
ht
Shutter
A B
t
ne
k.
oo
yb
er
C
rg
su
://
tp
ht
FIGURE 49.9 A patient who presented with an intrathoracic esophageal perforation. The perforation was visible on (A) esophagram and (B)
endoscopically. Patient presented early with limited contamination. (C) A stent was deployed with good coverage and (D) no residual leak
was noted on repeat radiographic study.
Boston Scientific, Merit Medical Endoteck, EndoChoice, Patient Selection for Endoscopic Management
and Taewoong Medical Company. Similar to any type of surgery, patient selection is of
The original SEMS were uncovered stents (Fig. 49.9). paramount importance. Identifying patients who are
These stents were generally left in situ and had signifi- suitable for nonoperative management has not been
cant tumor and tissue granulation and ingrowth. Once well defined. However, Cameron 24 in 1979 proposed
significant tissue ingrowth had occurred, these stents could key considerations, and Altorjay25 expanded on these
not easily be removed. Covered stents were developed two decades later. They included early diagnosis of an
to prevent the tissue ingrowth. The metal framework intramural perforation, transmural perforation within
of the stent is covered in polytetrafluoroethylene (most the neck or mediastinum with free drainage back into
commonly) through the length of the entire stent for fully the esophagus on esophagram, the absence of benign or
covered and the middle portion, leaving the proximal malignant obstructive esophageal disease, and minimal
and distal ends uncovered for partially covered stents.23 symptomatology without evidence of sepsis. Another
Management of Esophageal Perforations and Leaks CHAPTER 49 545
commonly used indication for nonoperative therapy has the group emphasized the importance of patient selection.
been cases in which patients were deemed unsuitable for Contraindications to esophageal stenting for anastomotic
surgery secondary to significant comorbid conditions. leak included nonviable conduit; leaks that were within
Endoluminal stenting has been highlighted to have 2 cm of the cricopharyngeus muscle to avoid significant
the following benefits: less procedural morbidity than patient discomfort; severe angulation of the conduit; and
surgery and rapid closure of the perforation, which quickly esophagojejunal conduits, as this resulted in obstruction
eliminates ongoing soilage of the mediastinum and pleura and erosion.
and allows for earlier initiation of oral nutrition. Leak after laparoscopic bariatric surgery occurs in up
Our indications for stenting continue to expand. We to 7% of cases in some series. This severe complication
use stenting most commonly in iatrogenic injuries that is difficult to manage and can result in significant patient
are discovered immediately with minimal contamination. morbidity and cost. Covered SEMS have been shown to
Newer approaches include endoluminal suturing using be effective in acute leaks when combined with drainage
the Apollo device or clipping with the resolution device or of associated collections. Simon et al.28 in their series of
Ovesco clips. We find acute intentional incisions, similar nine patients with leaks after sleeve gastrectomy showed a
to those created during per oral endoscopic myotomy 78% resolution of leak. Similarly, Eubanks et al.29 showed
(POEM), that are easily closed with the resolution clips. In a resolution of 84% in their 19 patients who were stented
addition, larger perforations or full-thickness perforations after gastric bypass. Importantly, symptomatic improvement
often require a larger clipping device such as the Ovesco. was also seen in 90% of patients with oral feeding started
t
When clipping is not feasible, intraluminal suturing can immediately in 79%. Chronic leaks have a significantly
ne
always augment healing when the tissue is intact enough reduced success rate in healing the chronic leak; however,
to hold the suture. Unfortunately, many esophageal they minimize ongoing sepsis and allow patients to con-
perforations result in damage to the mucosa such that it tinue enteral nutrition.30
k.
becomes too friable to suture. Regardless of the indication, it is essential to couple
stenting with adequate drainage of any associated infected
oo
Esophageal Stenting for Perforations and Leaks space.31 While once considered an indication for open
Essential components of nonoperative management for surgery, mediastinal contamination is no longer a contra-
esophageal disruption remain the same as with an open indication to endoscopic management, and drainage of
yb
strategy: drainage if required, prevention of ongoing small collections may be accomplished endoscopically with
contamination, and nutritional support while healing. In gentle aspiration and irrigation. In addition, double-pigtail
some cases, only endoscopic treatment may be required, stents may be placed through the esophageal wall into an
and in others, it serves as an adjunct to surgical treatment abscess cavity temporarily to provide ongoing drainage
er
as noted earlier. for a few days. Larger abscesses and free pleural effusions
Stents have been shown themselves to be highly suc- often require the placement of additional surgical drains.
rg
cessful in a few major arenas: iatrogenic perforations with In these cases, however, the utilization of a stent may help
early identification, esophageal anastomotic leaks, and to avoid esophageal resection or diversion.
gastric staple line leak after weight loss surgery. After stent deployment, it is important to assess for
su
Iatrogenic perforations with early identification are adequacy of coverage with an on-table contrast fluoroscopic
ideally suited for covered stent placement. The stent examination. If there is no evidence of ongoing leak,
effectively occludes the leak preventing mediastinal con- the stent is sutured into place to prevent migration. If
://
tamination, allows for early institution of oral, enteral inadequate seal is perceived with ongoing leak, the stent
nutrition, and allows for accelerated patient recovery is repositioned. Importantly, the contrast should also be
tp
compared with open surgery. In their series of 17 patients, seen to flow distally through the end of the stent. Distal
Freeman et al. demonstrated that with thoughtful patient obstruction leads to stasis, impairs proper healing, and may
selection, stents could be placed safely with a low mor- result in additional injury to the gastrointestinal tissues.
ht
bidity, mortality, and high rate of success. See Fig. 49.9 In some cases when no gastrostomy tube is present, we
for an example of endoscopic stenting for intrathoracic leave an NGT for ongoing decompression of the esophagus
esophageal perforation. and stomach. The NGT is left to gravity drainage as active
More reports are emerging describing endoscopically suction in the region of the stent can cause collapse and
placed stents for the treatment of anastomotic leaks. Most predispose to migration.
published series are small, but they cite good outcomes.13,26
Stenting after esophagectomy is challenging, as the diam-
eter of the esophagus is quite a bit smaller than the gastric
POSTSTENTING CONSIDERATIONS
conduit. This allows for more reflux up around the stent Once a stent is placed, judicious ongoing management and
to the level of the anastomosis and makes the stent more surveillance is required. If a patient develops evidence of
prone to migration. D’Cunha et al.27 reported one of the sepsis, the stent should be investigated for migration or
larger series in 2011. Their subgroup analysis of patients failure to adequately seal the leak. In some cases, the stent
with anastomotic leak included 22 patients, and they had can be repositioned with more effective coverage, while
a 60% success rate with healing. It should be noted that in other cases, surgical intervention must be pursued.32
the time to healing averaged 40 days and ranged from 22 Our protocol for management of patients after
to 120 days, and the mortality rate was 14%. The average esophageal stenting is to obtain an esophagram within
time to per os (PO) intake was 6 days, and some patients 24 hours. If no leak is present, the patient is initiated on
were able to begin oral intake immediately. Importantly, a room-temperature liquid diet that is advanced to soft
546 SECTION I Esophagus and Hernia
solids (essentially, anything you would feed a 1-year-old theory of “stent-guided regeneration and reepithelializa-
baby—nothing too hot or cold and nothing that does tion.” They could show that by bridging gap and providing
not dissolve into small pieces within a few minutes to a scaffold, full reepithelialization can be accomplished.
prevent stent impaction). They receive special counseling The exciting future for esophageal surgery lies in the
regarding diet from our nutritionist as well as a handout field of regenerative medicine. In 2012 Bradylak et al.
to follow at home. They return within 2 weeks for repeat published their experience, creating extracellular matrices,
endoscopy, evaluation for healing, and ongoing leak. In repopulating with autologous cells, and implanting into
some cases, the esophagus is fully healed, and the stent animal models. They observed that these biologic scaf-
can be removed; in others, the stent is repositioned or folds provide a framework for structural and functional
exchanged to change the pressure points and resecured. regeneration of tissues.
In both cases, an esophagram is performed within another In 2016, Dua et al.44 published their experience with
24 hours to confirm the absence of leak. Once a stent in vivo regeneration for a 5-cm esophageal defect. They
has been removed, the soft diet is continued for a total placed a stent as the framework to maintain the shape of
of 2 to 6 weeks. the esophagus and wrapped it with a dermal matrix. Endo-
scopic monitoring over time revealed normal squamous
mucosa ingrowth and endoscopic ultrasound showed the
COMPLICATIONS OF STENTING neoesophagus to have five layers, and peristalsis was seen
Early complications of stenting occur immediately or on manometry. The neoesophagus demonstrated both
t
within 2 to 4 weeks of stent placement. These include structural and functional regeneration. Phase 1 and 2
ne
patient discomfort, bleeding, acid reflux, perforation, clinical trials are needed, but this new technology could
stent leak, and most commonly stent migration.33–37 Stent revolutionize all aspects of esophageal surgery.
migration has been cited as the predominant problem in
k.
a number of series. New techniques and technology have
been developed to mitigate this issue, including the use
CONCLUSION
oo
of partially covered stents, which allows some ingrowth The management of esophageal perforation is constantly
of tissue proximally and distally to stabilize the stent, and evolving. With the increased use of endoscopy for both
suturing the stents into place.13 diagnostic and therapeutic interventions, it is a problem
yb
Stent leak is a feared complication of stent placement that should not expect a cure. Despite more minimally
and has been reported in 10% to 40% of patients after invasive approaches coming into vogue in the last 10 years,
stent placement. Stephens et al.38 developed a classification the basic principles of management remain unchanged:
system for the leaks (types 1 to 5) with recommenda- early identification, resuscitation, débridement, repair/
er
tions on management of each type. Importantly, prompt diversion, drainage, and enteric access for feeding. A
recognition of a leak is essential to achieve good outcome. concern our group shares is the dissemination of endo-
rg
Knowledge of which factors may predispose to a stent luminal techniques used by those not surgically trained.
leak led to a more timely identification and intervention This disconnected management strategy may often result in
for ongoing leaks. stenting without drainage of an adjacent abscess, placing a
su
Delayed complications may occur weeks to months stent directly next to the aorta or adjacent drains, placing
after stent placement. These include stent migration, stents without a pexy resulting in migration, and excessive
tumor ingrowth resulting in difficult extraction (Yoon), experimentation. The future of esophageal perforation
://
and perhaps the most feared complication, injury to may hold repair with the use of adjunctive techniques
surrounding structures. Case reports and institutional such as tissue regeneration, deploying matrices embedded
tp
series detail esophageal fistulization and tumor erosion.39–42 with growth factors, and more advanced endoluminal
Some of these complications may be avoided by earlier techniques.
extraction. Studies have shown that extraction is not
ht
9. Cohen MS, Kaufman A, DiMarino AJ, Cohen S. Eosinophilic 29. Eubanks S, Edwards CA, Fearing NM, et al. Use of endoscopic stents
esophagitis presenting as spontaneous esophageal rupture. Clin to treat anastomotic complications after bariatric surgery. J Am Coll
Gastroenterol Hepatol. 2007;5:24. Surg. 2008;206(5):935-938.
10. Ferguson DD. Evaluation and management of benign esophageal 30. Puig CA, Waked TM, Baron TH, et al. The role of endoscopic stents
strictures. Dis Esophagus. 2005;18(6):359-364. in the management of chronic anastomotic and staple line leaks
11. Madan R, Blair RJ, Chick JF. Complex iatrogenic esophageal injuries: and chronic strictures after bariatric surgery. Surg Obes Relat Dis.
an imaging spectrum. AJR Am J Roentgenol. 2015;204(2):W116-W125. 2014;10(4):613-617.
12. Freeman RK, Van Woerkom JM, Ascioti AJ. Esophageal stent place- 31. Abbas G, Schuchert MJ, Pettiford BL, et al. Contemporaneous
ment for the treatment of iatrogenic intrathoracic esophageal management of esophageal perforation. Surgery. 2009;146:749,
perforation. Ann Thorac Surg. 2007;83(6):2003-2007. discussion 755.
13. Blackmon SH, Santora R, Schwarz P, Barroso A, Dunkin BJ. Utility 32. Dickinson KJ, Blackmon SH. Endoscopic techniques for the
of removable esophageal covered self-expanding metal stents for management of esophageal perforation. Thorac Cardivasc Surg.
leak and fistula management. Ann Thor Surg. 2010;89(3):931-936. 2015;20(3):251-278.
14. Martin LW, Hofstetter W, Swisher SG, Roth JA. Management of intra- 33. Eubanks S, Edwards CA, Fearing NM. Use of endoscopic stents to
thoracic leaks following esophagectomy. Adv Surg. 2006;40:173-190. treat anastomotic complications after bariatric surgery. J Am Coll
15. Nguyen NT, Hinojosa MW, Fayad C, Wilson SE. Minimally invasive Surg. 2008;206:935-938.
management of intrathoracic leaks after esophagectomy. Surg Innov. 34. Ott C, Ratiu N, Endlicher E, et al. Self-expanding stents in esophageal
2007;14(2):96-101. disease: various indications, complications and outcomes. Surg Endosc.
16. Leers JM, Vivaldi C, Schafer H, et al. Endoscopic therapy for 2007;21:889-896.
esophageal perforation or anastomotic leak with a self-expandable 35. Sabharwal T, Hamady MS, Chui S, Atkinson S, Mason R, Adam A.
metallic stent. Surg Endosc. 2009;23:2258. A randomised prospective comparison of the Flamingo Wallstent
17. Kiev J, Amendola M, Bouhaidar D, Sandhu BS, Zhao X, Maher J. and Ultraflex stent for palliation of dysphagia associated with lower
t
A management algorithm for esophageal perforation. Am J Surg. third oesophageal carcinoma. Gut. 2003;52:922-926.
ne
2007;194:103. 36. Wang MQ, Sze DY, Wang ZP, Gao ZQ, Want YA, Dake MD. Delayed
18. Adler DG, Pleskow EK. Closure of a benign tracheoesophageal fistula complications after esophageal stent placement for treatment of
by using a coated, self-expanding plastic stent in a patient with a malignant esophageal obstruction and esophagorespiratory fistulas.
history of esophageal atresia. Gastrointest Endosc. 2005;61:765-768. J Vasc Interv Radiol. 2001;12:465-474.
k.
19. Zhou JH, Jiang YG, Wang RW, et al. Management of corrosive esopha- 37. Radecke K, Gerfen G, Trichel U. Impact of self-expanding plastic
geal burns in 149 cases. J Thorac Cardiovasc Surg. 2005;130:449-456. esophageal stent on various esophageal stenosis, fistulas and leak-
20. Kauer WKH, Stein HJ, Dittler HJ, Siewert JR. Stent plantation as a ages: a single-center experience in 39 patients. Gastrointest Endosc.
oo
treatment option in patients with thoracic anastomotic leaks after 2005;61:812-818.
esophagectomy. Surg Endosc. 2002;22:50-53. 38. Stephens EH, Correa AM, Kim MP, Gauer P, Blackmon SH. Clas-
21. Fischer A, Thomusch O, Benz S, von Dobschuetz E, Baier P, Hopt sification of esophageal stent leaks: leak presentation, complication,
yb
UT. Nonoperative treatment of 15 benign esophageal perfora- and management. Ann Thorac Surg. 2014;98(1):297-304.
tions with self-expandable covered metal stents. Ann Thorac Surg. 39. Tomaselli F, Maier A, Pinter H, Smolle-Juttner F. Management
2006;81:467-472. of iatrogenous esophageal perforation. Thorac Cardiovasc Surg.
22. Mumtaz H, Barone GW, Ketel BL, Ozdemir A. Successful management 2002;50:168-173.
er
of a nonmalignant esophageal perforation with a coated stent. Ann 40. Kennedy C, Steger A. Fatal hemorrhage in stented esophageal
Thorac Surg. 2002;74:1233-1235. carcinoma: tumor necrosis of the aorta. Cardiovasc Intervent Radiol.
23. Hindy P, Hong J, Lam-Tsai Y, Gress F. A comprehensive review of 2001;24:443-444.
rg
esophageal stents. Gastroenterol Hepatol. 2012;8(8):526-534. 41. Yoon CJ, Shin JH, Song HY, Lim JO, Yoon HK, Sung KB. Removal
24. Cameron JL, Kieffer RF, Hendrix TR, Mehigan DG, Baker RR. of retrievable esophageal and gastrointestinal stents: experience
Selective nonoperative management of contained intrathoracic with 113 patients. Am J Roentgenol. 2004;183:1437-1444.
su
esophageal disruptions. Ann Thorac Surg. 1979;27:404. 42. Homan N, Nofts MR, Klingenberg–Noftz RD, Ludwig D. Delayed
25. Altorjay A, Kiss J, Voros A, Bohák A. Nonoperative management of complications after placement of self-expanding stents in malignant
esophageal perforations: is it justified? Ann Surg. 1997;225:415. esophageal obstruction: treatment strategies and survival rate. Dig
26. Langer FB, Wenzl E, Prager G, et al. Management of postoperative Dis Sci. 2008;53(2):334-340.
://
esophageal leaks with the Polyflex self-expanding covered plastic 43. Amrani L, Menard C, Berdah S, et al. From iatrogenic digestive
stent. Ann Thorac Surg. 2005;70:398-404. perforation to complete anastomotic disunion: endoscopic stenting as
27. D’Cunha J, Rueth NM, Groth SS, Maddaus MA, Andrade RS. a new concept of “stent-guided regeneration and re-epithelialization.
tp
Esophageal stents for anastomotic leaks and perforations. J Thorac Gatrointest Endosc. 2009;69(7):1282-1287.
Cardiovasc Surg. 2011;142:39-46. 44. Dua KS, Hoga WJ, Aadam AA, Gasparri M. In-vivo oesophageal
28. Simon F, Siciliano I, Gillet A, Castel B, Coffin B, Msika S. Gastric leak regeneration in a human being by use of a non-biological scaffold
ht
after laparoscopic sleeve gastrectomy: early covered self-expandable and extracellular matrix. Lancet. 2016;388(10039):55-61. [Epub 08
stent reduces healing time. Obes Surg. 2013;23:687-692. April 2016].
PART NINE
Hernia
CHAPTER
t
ne
A
hernia is described as a protrusion of an organ or of minimizing the risk for subsequent reoperations by
tissue from its normal cavity. This protrusion may employing the best evidence-based approach for the
k.
extend outside the abdominal wall or between first hernia repair.2 Multiple comorbidities have been
body cavities. Hernias vary in presentation including identified to increase the risk of infection following hernia
oo
congenital, umbilical and epigastric hernias, inguinal, repair. A higher infection occurrence then augments
traumatic flank hernias, and incisional hernias to name the risk of recurrence. Comorbidities that increase rates
a few. In addition, hernia etiology may differ based upon of postoperative infections include smoking, diabetes,
yb
type. A congenital hernia, present at birth, is the result of chronic corticosteroid use, immunosuppression, coronary
defective development of the abdominal wall; alternatively, artery disease, chronic obstructive pulmonary disease, low
hernias may be acquired later in life as the result of preoperative serum albumin levels, prolonged operative
injury to the abdominal wall via trauma or surgery. An times, and use of absorbable synthetic mesh.3
er
acquired hernia may be attributed to overexertion, weight Many classification systems have also been developed
lifting, jumping from a high distance, or violent coughing to categorize hernias to better understand who is at risk.
rg
episodes, although underlying connective tissue disorders Global classifications for wound status include the
may also be a contributing factor. In recent years, the role Centers for Disease Control and Prevention (CDC) wound
of connective tissue disease such as Marfan syndrome, classification. This system categorized all surgeries into
su
Ehlers-Danlos syndrome, and osteogenesis imperfecta one of four groups: clean, clean-contaminated, contami-
have shown a predisposition for hernia development. nated, and dirty. Each CDC class was assigned a risk of
Similarly, syndromes such as polycystic kidney disease, postoperative wound infections by the CDC in 1985. 4
://
known for an abnormal extracellular matrix production, Clean wounds are known uninfected wounds without
have been demonstrated to be associated with up to a entry into any visceral tracts that carry a low risk of 1%
tp
43% incidence of hernias. It is believed that abnormalities to 5%. Clean-contaminated wounds are those in which
in collagen metabolism contribute to hernia formation the respiratory, alimentary, genital, or urinary tracts are
and high recurrence rates in these populations. Multiple entered under controlled conditions, increasing infection
ht
genomics projects are ongoing, looking at candidate genes risk to 3% to 11%. Contaminated wounds include open,
responsible for the production of type I and III collagens fresh, accidental wounds, those with major breaks in
as well as matrix metalloproteinases (MMPs). sterile technique, gross spillage from the gastrointestinal
The goal of ventral hernia repair generally includes tract, and presence of nonpurulent inflammation. The
closure of the midline without excess tension. Many risk risk of infection in a contaminated wound is 10% to 17%.
factors influence the longevity of that repair including Finally, dirty or infected wounds include those with old
patient factors (increased intraabdominal pressure, dimin- traumatic wounds, retained devitalized tissue, and those
ished tissue integrity) and technical factors (infection, with an existing clinical infection or perforated viscus.3
lateral mesh detachment, missed hernia). It is estimated Their risk is expectedly elevated to greater than 27%
that 75% of all recurrences are due to infection and for postoperative surgical site infection (SSI). The CDC
inadequate repair material fixation and/or overlap.1 Each updated the estimated SSI rates again in 1985 and 1991
type of hernia has specific hernia recurrence rates. Midline following changes in antibiotic coverage and technique
laparotomies for nonhernia surgery carry a 25% risk of with 2.1% for clean, 3.3% for clean-contaminated, 10%
developing an incisional hernia. Five-year reoperation rates to 17% for contaminated, and over 27% for dirty.4 CDC
for incisional hernia repairs have been reported at 24% classification focuses only on the characteristics of the
following the first reoperation, 35% after the second, 39% wound at the time of repair.
after the third, with the 7-year rate after three operations The Ventral Hernia Working Group (VHWG) clas-
nearing 50%. These data underscore the importance sification system developed a grading scale to predict
548
Basic Concepts and Factors Associated With Ventral Hernia Recurrence CHAPTER 50 548.e1
ABSTRACT
Ventral hernias occur commonly and are associated with
not infrequent complications and significant recurrence
rates often resulting in future procedures. Comprehensive
strategies to improve outcomes include preoperative
optimization, evidence-based perioperative care, and use
of appropriate surgical techniques. Tobacco abuse, obesity,
and poorly controlled diabetes have been identified as
significant risk factors for poor outcomes in ventral hernia
repair and require optimization. The use of immune
modulating diet supplements perioperatively may be
considered in complex ventral hernia repairs to further
outcomes. Recognition of patients with either a history
of drug resistant infections or carrier status is necessary
to appropriately tailor antibiotic therapy. Appropriate
surgical techniques and biomaterial selection for ventral
hernia repair further minimizes complications and recur-
rence rates. The culmination of optimal preoperative
t
ne
care, surgical technique, and postoperative management
effectively improves hernia outcomes.
k.
oo
yb
er
rg
su
://
tp
ht
Basic Concepts and Factors Associated With Ventral Hernia Recurrence CHAPTER 50 549
t
in the area 3 cm above
ing the review of 299 open ventral hernia repairs with
ne
and below the umbilicus
determination of SSI rates for each grade. Following this Iliac L3 Between a horizontal line
analysis, the modified system was developed representing 3 cm below the umbilicus
three grades. VHWG-M grade 1 includes low risk hernias and the inguinal region
k.
with low risk of complications and no prior wound infec- Lumbar L4 Lateral and dorsal to the
tions with a reported surgical site occurrence (SSO) rate anterior axillary line
oo
of 14%. VHWG-M grade 2 hernias represent patients with
significant comorbidities. VHWG-M grade 2 patients experi-
ence postoperative SSO rates of 27%. VHWG-M grade 3 and lateral primary hernias as well as incisional hernias.
yb
includes hernias with all degrees of bacterial contamination The classification system allows for more specific discussion
including clean-contaminated, contaminated, and dirty and comparison of hernias with a common nomenclature.
procedures with SSO rates of 46%.5 Midline hernias are designated from the xiphoid process
The Ventral Hernia Risk Score (VHRS) was developed to the pubic bone and medial to the lateral margin of the
er
using a single-center data within a Veterans Affairs popula- rectus sheath on both sides. Lateral hernias occur from
tion to stratify SSI risk based upon wound classification, costal margin to inguinal region and from the lateral margin
rg
comorbid conditions, and technique. The VHRS assigns of the rectus sheaths to the lumbar region. Hernias are
points ascribed to each of five clinical attributes: con- demarcated by size: W1, 1 to 4 cm; W2, 4 to 10 cm; and W3,
comitant hernia repair (2 points), creation of skin flaps greater than or equal to 10 cm as well as recurrent nature.
su
(2 points), American Society of Anesthesiologists (ASA) Table 50.1, adapted from the European Hernia Society
score 3 or greater (2 points), body mass index (BMI) 40 definitions, depicts the types of hernias that can occur on
or more kg/m2 (3 points), and incision class 4 or dirty (7 the abdominal wall.8 A standardized classification for hernias
://
points). There are five VHRS subgroups with progressive should be used to allow for effective description of outcome
associated risk for wound infection based upon the range measures as pertaining to hernia groups, recognizing that
tp
of points accumulated: group I (0 points), group II (2 to not all hernia repairs are associated with comparable results.
3 points), group III (4 points), group IV (5 to 10 points),
and group V (11 to 16 points). The risk of SSI development
ht
than 160 mg/dL. HbA1c was shown to be a stronger hernia patients, active smokers experience more frequent
predictor of adverse events compared with preoperative hernia recurrences and postoperative infections than
diabetes mellitus status or perioperative glucose.11 Other comparable nonsmoking patients.18 Smoking cessation
studies have identified a similar association between HbA1c efforts rank highly among the most challenging initiatives
and surgical outcomes. Although the optimal preopera- for patients and physicians. Smoking recidivism rates are
tive HbA1c has not been established, efforts should be not insignificant following attempts at smoking cessation.
made to optimize glycemic control and HbA1c prior to However, considering the increased morbidity and cost
consideration of any elective hernia repair to enhance associated with hernia complications and recurrences,
postoperative outcomes. attempts at smoking cessation prior to elective ventral
In 2001 Latham et al. reported a 7.9% SSI rate in hernia repair should be attempted. Although extenuat-
patients with a HbA1c greater than 8, a twofold risk com- ing circumstances may compel patients and surgeons to
pared with patients with an HbA1c less than 8.12 Endara proceed with elective repair in patients using cigarettes,
et al. found that HbA1c greater than 6.5% was associated the risks should be carefully considered preoperatively, as
with increased rates of dehiscence after surgical wound each hernia recurrence carries a greater risk of recurrence
closure similar to the finding by Goodenough et al. in than the prior repair. Patients with multiple recurrent
which 6.5% was found to be the threshold HbA1c level at hernias are generally not suitable for elective repair while
which complication rates increase.11,13 Underwood et al. smoking, but first-time repairs in smokers should be
treated diabetic patients with HbA1c greater than 8% similarly discouraged to avoid the creation of recurrent-
t
preoperatively, resulting in considerable blood glucose hernia patients.
ne
level improvements on the day of surgery.14 The American Alcohol abuse is associated with an increased risk of
Diabetes Association recommends that outpatient manage- bleeding, wound, and cardiopulmonary complications.
ment of diabetes should ideally include a combination Alcohol abuse is categorized as ingestion of five or more
k.
of target HbA1c less than 7%, preprandial blood glucose drinks (60 g of ethanol) a day. Abstinence from alcohol for
level of 90 to 130 mg/dL, and a peak postprandial blood 1 month preoperatively reduces postoperative morbidity
oo
glucose level of less than 180 mg/dL.15 While no finite with reduced responses to surgical stress, improved cardiac
recommendation exists, most experts agree that attempts and immune dysfunction.19
to obtain a HbA1c less than 8 should be made prior to
WEIGHT OPTIMIZATION
yb
elective hernia repair with a goal as close to 6.5% as
feasible. BMI is considered a significant predictor for surgical site
In Goodenough et al. publication, one-third of patients occurrence. Morbidly obese patients are at a higher risk for
without a history of diabetes were found to have HbA1c the development of abdominal wall defects and progression
er
greater than 6.5% on screening. These patients were of the size of the defects due to increased intraabdominal
more likely to have a BMI greater than 30, have additional pressure and poor wound healing potential.20 The relation-
rg
comorbidities including coronary artery disease, and ship between obesity and surgical complications including
have chronic pulmonary obstructive disease, as well as SSIs has been reported in colorectal surgery, and more
categorized as non-Caucasian.11 Based on these findings, recently, the association has been specifically related to
su
we recommend obtaining HbA1c levels on all patients ventral hernia repair.21 BMI is considered a significant
with BMI greater than 30 kg/m2 presenting for elective predictor for SSI when analyzed as a continuous variable,
ventral hernia repair. thus demonstrating that SSI risk increases with increas-
://
Smoking and alcohol use have been shown to negatively intervention.6 While the ideal BMI for elective hernia repair
impact postoperative outcomes. Current smokers have an is often debated, BMI should be evaluated in the context
increased risk of pulmonary and wound complications of individual patient and hernia characteristics when
ht
following operation. The effects of nicotine on a cellular determining the best strategy for patient management.
level include vasoconstriction and tissue level hypoxia Other factors such as risk for incarceration, crescendo
correlating with tissue nicotine levels, increased platelet symptomatology, and rapidity of hernia progression may
aggregation, and reduced fibroblast migration. Carbon influence decision making. For example, patients with
monoxide levels also reduce oxygen delivery to the tissues, small defects with large volumes of incarcerated bowel
leaving smokers who consume greater than 20 cigarettes may be at risk for significant intestinal loss in the event
hypoxic most of the day.16 The additive effects of tissue of strangulation. In such circumstances, elective repair
hypoxia, reduced fibroblast proliferation, reduced collagen may be appropriate in high BMI patients despite the
1 to 3 ratios and reduced overall collagen deposition in increased perioperative risks.
smokers leads to increased wound infections and reduced Accordingly, patients amenable to a laparoscopic
tissue strength. Smoking one cigarette decreases cutaneous approach might be considered for an elective operation
and subcutaneous blood flow by 38.1%.17 Smoking cessa- at higher BMI levels due to the reduced likelihood of
tion of 4 weeks preoperatively has been shown to reduce postoperative infections in the morbidly obese compared
wound infection rates from 12% in 1 pack per day smokers with open hernia repairs. While conversion to an open
to 1%, values comparable with patients who have never procedure is always possible, conversion rates are low
smoked. In that investigation, there was no difference in many large laparoscopic ventral hernia repair series.
between transdermal nicotine patch and placebo patches As the incidence of obesity in the adult population
used for smoking cessation techniques. Specifically, in increases, surgeons are evaluating the outcomes of this
Basic Concepts and Factors Associated With Ventral Hernia Recurrence CHAPTER 50 551
t
BMI less than 40 kg/m2.23 Other studies have found the Experimental studies have shown nutritional supplements
ne
laparoscopic approach to offer a safe alternative for obese containing l-arginine, omega-3 polyunsaturated fatty acids,
patients but with apparent increased risk of recurrence and nucleotides boost immune responsiveness after surgery
of hernia for this subset of patients.24–26 or trauma. l-Arginine is a semiessential amino acid and
k.
Weight loss surgery is an effective method of weight a precursor of nitric oxide, which is the most important
loss, and ventral hernia repair can be safely combined endothelial vasodilator. L-arginine has been shown to
oo
with weight loss procedures such as Roux-en-Y gastric improve wound healing, restore postoperative depressed
bypass and sleeve gastrectomy or performed in stages.27–29 macrophage function and lymphocyte responsiveness,
A large national database study showed that patients who and augment resistance to infections. Intake of additional
yb
underwent ventral hernia repair in conjunction with omega-3 polyunsaturated fatty acids alters cell membrane
either laparoscopic Roux-en-Y gastric bypass or sleeve phospholipid content and prostaglandin synthesis that
gastrectomy had increased incidence of SSI but not overall is theorized as an important factor in suppression of
morbidity in the 30-day postoperative time period, again the generalized inflammatory response and subsequent
er
providing evidence of the seeming appropriateness of the immunosuppression and capillary leakage after major
combination of these procedures.30 However, long-term surgery.33 Elevated omega-3 levels also inhibit the metabo-
rg
outcome studies specifically addressing hernia recurrence lism of arginine. An oral immune-enhancing nutritional
rates in these populations are lacking. supplement taken for 5 days preoperatively has been
As a patient’s BMI approaches 30 kg/m2, the risk of shown to result in increased preoperative serum arginine
su
infection and recurrence reduce; however, an optimal concentration and decreased number of postoperative
cutoff BMI for ventral hernia repair in the obese population infections with preserved renal function.34 Similarly, low
is still debatable. In general, a BMI of less than 40 kg/m2 albumin is also independently associated with major
://
may be considered safe for repair supported by the work complications that can be improved with preoperative
of Tsereteli et al.23 nutrition augmentation. As a component of our enhanced
tp
among patients presenting with incisional hernias as this is the impact of this dietary regimen in ventral hernia repair
a significant risk factor for hernia formation. In a landmark has not been independently studied, the numerous studies
study in the 1980s, a history of previous wound infection demonstrating the benefits in other patient populations
predicted a greater risk for subsequent wound infection cannot be disregarded.33,34
following incisional hernia repair.31 Houck et al. noted that Traditional surgical and anesthesia dogma has required
41% of patients with prior wound infections developed patients to abstain from oral intake for at least 8 hours
an infection following subsequent repairs compared with prior to an elective operation. Despite years of practice,
an infection rate of 12% in patients without previous no scientific evidence exists to support the basis for this
infections (P < .05), thus emerging the concept that once fast, and in a Cochrane review, there was no evidence
an abdominal wall is infected, the risk for subsequent to suggest a shortened fluid fast results in an increased
infections remains elevated. 31 Subsequent infections risk of aspiration, regurgitation, or morbidity compared
increase the risk of hernia recurrence by 80% with a with the standard fasting policy.35 Patients following a
relative risk of 4.3 compared with noninfected repairs.32 standard fast present with depleted glycogen stores in
Although the presence of prior infections represents a the liver, which increases the demand for amino acids
nonmodifiable risk factor for future complications, this production resulting in protein catabolism following
increased risk associated with prior infections should surgical stressors rather than tissue repair. Recent protocols
heighten the awareness of other modifiable risk factors have transitioned to providing a clear fluid that contains
to facilitate optimal outcomes. a relatively high concentration of complex carbohydrates
552 SECTION I Esophagus and Hernia
2 to 3 hours before induction of anesthesia allowing open operations both contribute to stress hyperglycemia
patients to undergo surgery in a metabolically fed state with by promoting a proinflammatory response.11 Emergent
adequate glycogen stores. As little as 400 mL of a 12.5% conditions preclude the opportunity to perform patient
drink of mainly maltodextrins (e.g., Gatorade) reduces optimization for modifiable risk factors. However, awareness
preoperative thirst, hunger, anxiety and postoperative of the inherent risk factors will allow for postoperative
insulin resistance, resulting in less postoperative losses of treatment of these modifiable conditions.
nitrogen and protein and better-maintained lean body mass
and muscle strength.36 Our enhanced recovery protocol
includes administration of 32 ounces of a carbohydrate-
HERNIA PREVENTION
rich, clear fluid 3 to 4 hours preoperatively as a method Incisional hernia formation occurs in a significant number
of ensuring euvolemia and adequate glycogen stores as of patients undergoing abdominal surgery and is the most
elective hernia repair commences. common complication of a laparotomy. The rate of hernia
occurrence after laparotomy approaches 25% by 3 years
with no gold standard for prevention.42 Two prospective
METHICILLIN-RESISTANT STAPHYLOCOCCUS studies evaluating incisional hernia rates have been con-
AUREUS PROPHYLAXIS ducted. While physical examination is the most common
method for detecting incisional hernias, ultrasonography
Ventral hernia patients with a history of methicillin-resistant may be used as an adjunct to detect clinically occult
t
Staphylococcus aureus (MRSA) at any prior location (blood, hernias. Early detection of occult hernias may facilitate
ne
urine, sputum, wound) experience increased wound identification of patients at risk for hernia progression.43
infection rates compared with those patients without a While neither ultrasonography nor examination has 100%
history for MRSA.37 In an effort to decolonize known specificity, incisional hernia rates are likely underreported.44
k.
MRSA carriers or reduce the risk of colony-forming units While operation rates are often considered a surrogate
(CFU) of the skin flora with unknown MRSA status, the for incisional hernia rates, minimally symptomatic hernias
oo
use of mupirocin intranasal ointment twice daily combined are often managed nonoperatively and further contribute
with chlorhexidine showers for 5 days preoperatively is to the underestimation of hernia incidence.
considered effective with reports of a 44% reduction in Hernia prophylaxis represents an opportunity to prevent
SSI.38,39 Although decontamination protocols have not
yb
the initial development of hernia, thus minimizing the
been evaluated independently in hernia patients, MRSA subsequent morbidity associated with complications and
decolonization has been investigated as a component of recurrences. The optimal technique for laparotomy closure
a hernia-enhanced recovery protocol.40 Identification of has been exhaustively investigated. Prior studies have
er
patients with a history of MRSA infection often repre- evaluated suture choice, suture technique, stitch length,
sents the greatest barrier to implementation. As a result, and mesh reinforcement at the time of laparotomy to
rg
mupirocin and chlorhexidine showers may be prescribed reduce the risk of incisional hernia formation.
preoperatively to all patients undergoing ventral hernia
repair due to the low cost, infrequent side effects, and SUTURE SELECTION
su
bleeding. The maintenance of a patient’s temperature, of hernia formation associated with rapidly absorbable
rather than restoration following temperature decreases, suture materials. Long-lasting absorbable sutures have the
is paramount. Normothermia may be accomplished with benefits of a permanent suture without suture site pain
a suitable warming device and warmed intravenous fluids and development of wound sinuses and have become the
should be used routinely to keep body temperature greater standard over a permanent suture.45
than 36°C. Monitoring is essential to titrate warming
devices and to avoid hyperpyrexia.41 Warming devices such
as forced-air heating blankets and warmed intravenous
STITCH LENGTH
fluids should be routinely used to keep body temperature Equally important to suture material is stitch length and
greater than 36°C. suture size. Initial work evaluating at 4 : 1 suture--wound
length ratio dates back to 1976 with Jenkins reporting
suture-to-wound length ratio greater than 4 : 1 reducing
EMERGENCY OPERATIONS laparotomy failure (e.g., abdominal burst wound).46 This
Emergency operations are known risk factors for postopera- type of closure is based on the Pythagorean theorem in
tive complications and hernia recurrence. Patients present- which 1 cm of travel along an abdominal closure will
ing with surgical emergencies may not have the opportunity require approximately 4 cm of suture material. However,
to be optimized through risk factor optimization and this only accounts for the length and width of suture
present with high levels of stress cortisol. Emergency and required and does not account for the the thickness of the
Basic Concepts and Factors Associated With Ventral Hernia Recurrence CHAPTER 50 553
abdominal wall. Competing ideals debate the ideal suture- loop No. 1 PDS versus running closure with an onlay
to-wound length ratio and whether it should incorporate of lightweight polypropylene (3 cm of overlap). Hernia
the thickness of tissue and be as high as 6 : 1.47 occurrence was determined by computed tomography (CT)
In a prospective study of 450 patients undergoing imaging and physical exam at 12 months with 1.5% hernia
laparotomy, the impact of suture-to-wound length ratios rate in the mesh group versus 35.9% rate in the nonmesh
was compared. Hernia rates for closures with less than group.52 The technique of mesh prophylaxis and 4 : 1
4 : 1 suture-to-wound length ratio was 24% compared with closure was applied to patients undergoing emergent and
9% when the ratio was greater than 4 : 1. In a subsequent elective colorectal surgery in an RCT published in Annals
study, Israelsson et al. performed a single-site prospective of Surgery in 2015. Fifty-four patients were closed with 4 : 1
trial comparing midline laparotomy closure with 2-0 technique alone compared with 53 patients reinforced with
polydiaxanone (PDS) running suture with 5 to 8 mm tissue onlay polypropylene (PP) mesh, 2.5 cm overlap laterally.
bites and 5 to 8 mm advances (short stitch) compared with Suture only had a 31.5% rate of hernia development
a No. 1 PDS with 1 cm bites and 1 cm advances (large compared with 11.3% with mesh (P = .011), no difference
stitch). The small stitch group experienced a reduced in mesh in seroma, SSI, evisceration, or mortality.53 The
infection rate of 5.1% compared with 9.6% infection rate PRIMA trial is a double-blinded RCT ongoing to compare
in the large stitch group. Furthermore, the small stitch primary suture closure, glued onlay mesh augmentation
hernia rate was reduced to one-third of that of the large (OMA), and sublay mesh augmentation (SMA). Initial
stitch group, 4.7% compared with 17.1% for short stitch 30-day results, published in 2015, showed no difference in
t
and long stitch, respectively.48 SSI, hematoma, reintervention, or readmission; however,
ne
In a multicenter randomized controlled trial in Europe there was an increased OR of seroma formation using
(STITCH trial), 609 patients were randomly assigned to onlay low-weight PP mesh augmentation (OMA vs. primary
undergo a large stitch or small stitch laparotomy closure, suture [PS]: OR, 4.3; P = .004; OMA vs. SMA: OR, 2.09;
k.
and were followed for a year with either physical examina- P = .003).54 Mesh augmentation for hernia prevention, even
tion or ultrasound. In these matched cohorts based upon in high-risk patient populations, has been proven safe and
oo
age, gender, BMI, comorbid conditions, type of surgery effective; however, it has not become mainstream due to
including colorectal, upper GI, gynecologic, or vascular, lack of acceptance or reimbursement.
the large stitch patients had fewer numbers of stitches
yb
(25 vs. 45) and shorter overall suture length used (95 vs.
110 cm) for equivalent laparotomy lengths. The ratio of
MANAGEMENT OF A HERNIA
suture to length of wound was longer in the small stitch (5 Once a hernia develops, there are multiple components
vs. 4.3), and the operative duration required to perform involved in the optimal repair to avoid recurrence. Flum
er
the short stich closure was 4 minutes longer in the small et al. reported that the 5-year reoperative rate following
stitch cohort. Postoperative complications were no different, incisional hernia repair was 23.8% after the first reopera-
rg
morbidity (ileus, cardiac, pneumonia), SSIs, rate of burst tion, 35.3% after the second, and 38.7% after the third.55
abdomen (1%), and hospital length of stay. Hernia rates Many mechanisms of recurrence have been described in
were greater in the large stitch versus short stitch groups the literature including technical failures such as knot
su
(21% vs. 12%). While the STITCH study demonstrated a tying, suture pullout, inadequate mesh overall, lateral
reduction in hernia rates with the short stitch technique, no detachment of mesh, or inadequate choice of mesh or
reduction in SSI rates were seen.49 While this study clearly fixation as well as occurrence of infection, trauma, or
://
demonstrates the benefits of technique upon incisional increased intraabdominal pressure.1 The goal of any hernia
hernia rates, the mean BMI of patients enrolled in the operation is approximation of the midline abdominal
tp
STITCH trial is 24 kg/m2, and the technique has not been wall without tension. This goal is affected by the tissue
demonstrated to be effective in obese populations. integrity. Determining the degree of tension at the time
of repair has been challenging. The use of tensiometers
ht
No. 1 Prolene running suture closure or sublay PP mesh integration, reduce wound complications, and have tissue
with 2 to 4 cm overlap and a midline closure if possible. coverage to minimize exposure to superficial SSIs as
The goal follow-up was 36 months; however, the average well as intraperitoneal contents. An onlay repair secures
was only 26 months. In primary incisional hernias, the mesh to the anterior fascia, typically involving dissection
recurrence rate using suture alone was 43% compared of devascularizing skin flaps and primary closure of the
with 24% in the group that used mesh (P < .02). This fascia below the mesh.
trend continued in patients undergoing repair of first Inlay repair places the mesh within the hernia defect,
recurrence, 58% suture alone versus 20% mesh. Pooled securing it circumferentially to the edges of the fascia
data often quoted reports 46% rate of hernia recurrence without mesh-tissue integration. Sublay repair refers to
in suture alone and 23% rate following mesh repair (P < retrorectus, commonly referred to as Rives-Stoppa, or pre-
.005) with more patients in the suture-only arm developing peritoneal mesh placement, with the mesh location beneath
hernia recurrences early around 12 months compared the rectus complex and primary closure of fascia over the
with mesh repairs.32 The recurrence rate on this cohort mesh. This position allows for mesh-tissue integration to
of patients was updated by Burger et al. in 2004, with a load-bearing tissues from both the posterior rectus sheath
10-year cumulative rate of recurrence of 63% for suture and the anterior myofascial complex while protecting the
repair and 32% for mesh (P < .001).56 Using mesh repair mesh from superficial exposure and minimizing skin flap
of hernias less than 2 cm in size was debated until 2013. creation. The underlay repair denotes an intraperitoneal
Christoffersen et al. reported results of a prospective position with mesh secured to the anterior abdominal
t
cohort study using the Danish Ventral Hernia registry wall, placed open, or laparoscopically. Advantages include
ne
comparing elective open mesh versus suture-only repairs mesh protection from superficial SSI and lack of skin flaps;
of small umbilical and epigastric hernias. A total of 4786 however, there is an increased risk of deep/organ space
patients were included and followed up to 4 years with infections and complications secondary to adjacent viscera.
k.
a reoperation rate for recurrence of 2.2% in the mesh The incidence of hernia recurrence due to mesh loca-
group and 5.6% in the suture-alone group. Subset analysis tion is best described in Awad et al. in 2005, in which
oo
in suture-alone group showed similar rates between non-, 119 papers were reviewed and the data were pooled to
slow-, and fast-absorbable sutures. Rates of reoperation determine the cause of each hernia recurrence.1 Criticisms
were also similar in regard to mesh position (inlay/plug of this paper include unreported length of follow-up, lower
yb
4.9%, sublay 2.5%, onlay 2.2%, and intraperitoneal 0.7%). than previously reported hernia recurrence rates, and a
The rates of reoperation cannot be correlated directly to heterogeneous patient population. Regardless, they were
recurrence rates; however, this paper does support the able to report a difference in hernia recurrence based on
use of mesh reinforcement of hernias 2 cm or greater.57 mesh location that is consistent with the understanding
er
Recommendations regarding the extent of overlap are greatest risk of recurrence is associated with an inlay
often defined according to the size of the defect and repair (12.7%). This is consistent with the mainstream
vary from 2 to 5 cm, with a 5-cm mesh overlap in every understanding that a larger mesh-to-fascia interface
su
direction from the defect generally accepted as an ideal.58–60 allows for improved strength and longevity compared
Sufficient overlap depends on the location of the mesh, with an inlay, where the interface is completely dependent
location of the defect, and limitations associated with upon the suture strength approximating mesh to fascia.
://
adjacent osseous structures that can limit fixation. An Onlay, sublay, and intraperitoneal repairs were similar in
important component of mesh selection is defect sizing. recurrence rates (5%, 4.4%, and 3.6%, respectively).1 A
tp
Ventral hernia defect area will increase with a rise in meta-analysis regarding mesh placement in open ventral
intraabdominal pressure (IAP); therefore a defect size hernia repairs was published in 2016. Twenty-one studies
is in constant flux during daily activities. For example, reported on recurrence rates predominantly detected based
ht
resting IAP is 2 to 4 mm Hg and is increased with activity on clinical exam during a 5- to 60-month follow-up. The
such as jumping (IAP of 170 mm Hg), coughing (IAP of pooled recurrence rate was 16.5% for onlay, 30.2% for
100 mm Hg), Valsalva maneuver (IAP of 40 mm Hg), and inlay, 7.0% for sublay, and 14.7% for underlay. The rate
standing (IAP of 20 mm Hg).61 of SSI based on mesh location was also reported based on
Multiple models have been developed in attempts to 10 studies and had 16.9% for onlay, 31.3% for inlay, 3.7%
answer the force required to displace the mesh-fascial for sublay, and 16.7% for underlay.63 The recurrence rates
interface. A recent dynamic in vivo study laparoscopically in this meta-analysis were higher than those published by
measured midline hernia defect sizes in the vertical and Awad et al.; however, the elongated duration of follow-up
horizontal position at an IAP of 8 and 15 mm Hg to likely accounted for additional recurrence detection. Across
determine the defect area. Results confirm that increasing all studies, mesh placed in an open sublay position (e.g.,
IAP increases defect area by an average of 25% when retrorectus, preperitoneal) is associated with the lowest
varying pressure between 8 and 15 mm Hg. When using a risk of recurrence.
large mesh overlap of 5 cm or greater, precise measurement
of the defect area becomes less important.62 BRIDGE VERSUS BUTTRESS
The goal of any open ventral hernia repair is approxima-
MESH LOCATION tion of the midline discussed in the literature since the
Multiple positions for mesh placement have been described, 1920s. Hernia repairs performed laparoscopically include
and the ideal mesh placement should allow for mesh-tissue primary fascial closure (PFC) of the defect via transfascial
Basic Concepts and Factors Associated With Ventral Hernia Recurrence CHAPTER 50 555
t
demonstrated that the time required to close the fascia configurations (Table 50.2). While favorable characteristics
ne
during laparoscopic hernia repairs did not significantly for the ideal mesh have been elucidated, no single mesh
increase the total operative time, thus maintaining opera- is considered the gold standard for all hernia repairs.
tive costs.64 Differing results have been published regarding Accordingly, many mesh materials are marketed for use
k.
postoperative pain, with the most recent literature noting in specific conditions based upon patient factors or choice
no difference.67 of surgical technique.
oo
In the setting of open midline abdominal wall recon- Mesh materials can generally be classified based
struction, Booth et al. noted increased hernia recurrence upon one of three categories: synthetic, biologic, or
with bridge repair (55.6% vs. 7.7%; P < .001) with an bioabsorbable. Synthetic mesh materials include those
yb
interval to recurrence 9 times quicker in bridged cohort. fabricated primarily from polypropylene, polyester, or
The bridged repair group experienced a sixfold higher polytetrafluoroethylene. Both polypropylene and polyes-
overall complication rate than the mesh-reinforced group ter mesh materials are woven meshes with interstices to
(P < .001).68 Hernias larger than 15 cm repaired in a facilitate integration into the host tissues. These materials
er
bridging fashion was an independent risk factor for recur- are generally the most cost-effective material for hernia
rence. Criticisms include a high frequency of biologic repair, although there may be some limitations to their
rg
mesh, majority using porcine acellular dermal matrix over use.70 Both polypropylene and polyester mesh materials
synthetic, which had no recurrences in a bridge position. are not suitable for placement within the peritoneal
Booth et al. literature is used to support the statement that cavity adjacent to the intestines without the addition
su
a biologic mesh should not be used in a bridge position of an adhesion barrier. These synthetic meshes may be
due to a high rate of recurrence. used as an onlay or in the retrorectus or preperitoneal
space to avoid contact between the mesh and the viscera.
://
undisputed following the results of a landmark study by minimizes visceral adhesion and a surface with greater
Luijendijk et al. in 1999 in which the rate of recurrence porosity to enhance tissue integration. Numerous studies
associated with primary ventral and incisional hernias have compared the characteristics of these materials from
was twofold greater in those repaired without the use of the standpoint of strength, contracture, and adhesion
a prosthetic mesh.32 In a subsequent publication of this formation.71–73 Polytetrafluoroethylene meshes may be
trial, the 10-year recurrence rates for suture repair were advantageous in certain situations due to their dual-sided
63%, while the recurrence rates associated with mesh fabrication process, and the strength of integration is less
repair were 32%.56 These studies demonstrated for the than that of polypropylene.71,74
first time the significant benefits associated with the use Synthetic mesh materials have historically been con-
of a prosthetic mesh when repairing ventral hernias. sidered contraindicated in the presence of any degree
However, these studies also highlighted the significant rate of contamination due to concerns for mesh infection.
of recurrence associated with ventral hernia repair over In a study of over 30,000 ventral hernia repairs, patients
time. Other authors have demonstrated similar trends in undergoing ventral hernia repair in clean-contaminated or
ventral hernia recurrence rates with increasing incidence contaminated surgical fields were found to have increased
of hernia recurrence and hernia reoperation rates over risks for superficial SSI (2.53 vs. 3.84 OR), deep incisional
prolonged periods of time.55,69 SSI (3.09 vs. 5.33 OR) in clean-contaminated and contami-
Prosthetic mesh choices have evolved significantly nated wounds, respectively, compared with CDC class 1
since the time of the initial work by Luijendijk et al. with wounds.75 This study affirmed the beliefs of many surgeons
556 SECTION I Esophagus and Hernia
that permanent synthetic meshes should be avoided in surgical settings to support their use.85 However, a meta-
contaminated surgical fields. Although enterotomy or analysis from 2014 evaluating 1229 hernia repair patients
unplanned bowel resection at the time of hernia repair among eight studies demonstrated no difference in hernia
is relatively rare, the placement of a synthetic mesh at recurrence rates between biologic and synthetic mesh
the time of these event is associated with a threefold hernia repairs with statistically fewer wound infections in
increased incidence of complications, a fourfold increase the biologic hernia repair group.86 In a study of 761 ventral
in 30-day reoperations, and a 10-fold increased incidence hernia repairs performed with suture, synthetic, or biologic
of enterocutaneous fistula formation in a series of 1124 mesh, there was no difference in SSI rates or hernia
patients undergoing elective hernia repair.76 In fact, the recurrence rates between groups.87 Although this study
presence of an enterotomy is considered to be a significant was retrospective, the groups were similar in wound class,
predictor for the need for mesh explantation.77 hernia size, comorbid conditions, use of fascial releases,
The role of synthetic mesh in the setting of contami- and acuity of hernia presentation. These studies suggest
nated surgical fields has more recently been debated. In a that biologic mesh outcomes may rival that of synthetic
series of 100 ventral hernia repairs in clean-contaminated mesh, but further studies directly comparing outcomes
or contaminated surgical fields, the incidence of mesh are needed prior to drawing definitive conclusions.
removal 10 months following initial operation was 4% The greatest barrier to adoption of the use of biologic
with one patient developing an enterocutaneous fistula.78 mesh materials for hernia repair is cost. In a value-based
In this study, lightweight polypropylene mesh was placed health care environment, the use of cost-efficient materials
t
in the retrorectus or preperitoneal space in all but one is paramount. The costs of biologic meshes are significantly
ne
repair. This same group has more recently reported their greater than the cost of synthetic mesh.70 In the United
outcomes with lightweight synthetic mesh hernia repairs States, health care costs are generally reimbursed through
in CDC class 1 wounds and noted an increased incidence a diagnosis-related group (DRG) payment. The choice
k.
of hernia recurrences with the use of lightweight meshes of mesh material, biologic or synthetic, is unlikely to
(22.9%) versus midweight polypropylene mesh (10.6%).79 impact the facility reimbursement, and accordingly, the
oo
Further investigation into the role of synthetic meshes in added costs associated with more expensive materials will
contaminated fields will need to be conducted to appreciate directly impact a hospital’s profit margin. Reynolds et al.
the benefits, risks, and costs. reported average financial losses of $8370 with the use of
yb
Biologic mesh emerged as an alternative to synthetic biologic mesh during ventral hernia repair, while synthetic
mesh hernia repair in the early 21st century as a class of mesh repairs were associated with a positive contribution
materials derived from either human or animal sources. margin of $3110, and an overall profit of $60 inclusive of
These materials provided an alternative to the use of indirect costs.70 Because of the cost differences between
er
synthetic mesh in complex hernia repair, although the meshes with direct impact upon a hospital’s finances,
cost associated with these materials was significantly greater it is incumbent upon the surgeon to select appropriate
rg
than synthetic meshes.70 Biologic mesh materials may be materials based upon individual patient characteristics.
derived from numerous tissue sources including dermis, Biosynthetic meshes have emerged as a third class of
small intestine submucosa, urinary bladder, pericardium, mesh materials. This group of materials includes fabricated
su
and liver, to name a few. Each of these materials undergoes meshes derived from polylactic acid, polyglycolic acid,
unique processing to remove cellular elements, leaving trimethylene carbonate, silk, and poly-4-hydroxybutyrate.
an intact extracellular matrix.80 This class of materials has emerged as a more cost-effective
://
Biologic grafts are typically used in surgical wounds alternative to biologic mesh. These materials provide a
with some level of contamination as a strategy to minimize scaffold to reinforce the abdominal wall while ultimately
tp
the risk for mesh infection. However, similar to synthetic resorbing. Preclinical evidence with biosynthetic meshes
meshes, biologic grafts are not cleared by the US Food have demonstrated predictable degradation and hernia
and Drug Administration for use in contaminated settings repair strength in excess of the native abdominal wall
ht
and many of these materials include warnings against use strength.88 A prospective observational trial of 107 patients
in the setting of infection within the instructions for use.81 undergoing ventral hernia repair with biosynthetic mesh
Nevertheless, biologic grafts have been used with demonstrated a clinical recurrence rate of 17.3% at 2
increasing frequency in contaminated surgical fields as years.89 Although several trials evaluating biosynthetic
an alternative to sutured repairs, synthetic mesh repairs, meshes are ongoing, there is inadequate evidence at this
or the use of tissue flaps. Early experiences with biologic time to support widespread adoption.
meshes demonstrated poor outcomes when these grafts Table 50.3 includes classification of prosthetic materials.
were used as a bridging material to span a hernia defect
with recurrence rates approaching 100%.82 However, when
biologic meshes are used in conjunction to a component
POSTOPERATIVE MANAGEMENT
separation in contaminated settings, reasonable outcomes Following effective preoperative and intraoperative deci-
can be expected with recurrence rates up to 31%.83 Not sion making, critical postoperative management of risk
dissimilar from outcomes with synthetic meshes, biologic factors and limiting infections are key to reducing hernia
mesh hernia repair outcomes are impacted by mesh complications.
location, with lower recurrence rates demonstrated with
retrorectus mesh positioning.84 DIABETES MELLITUS CONTROL
There is little evidence directly comparing outcomes The current guidelines from surgical societies recom-
with synthetic and biologic mesh materials in contaminated mend intraoperative glucose levels less than 180 mg/
Basic Concepts and Factors Associated With Ventral Hernia Recurrence CHAPTER 50 557
t
ePTFE + lightweight polypropylene
ne
Coated prosthesis a. Lightweight or midweight Reduced chronic pain and recurrence, no need to
b. Absorbable or nonabsorbable coating close peritoneum in laparoscopic inguinal hernia
Materials for coating: repair, coating decreases the adherence of protein
k.
1. Omega-3 fatty acid coagulum and inhibits partially the initiation of the
2. Absorbable complex carbohydrate- inflammatory cascade
oat beta glucan
oo
3. Carboxymethylcellulose-sodium
hyaluronate-polyethylene glycol
4. Oxidized regenerated cellulose
yb
5. Titanium
Biologic prosthesis Types: Human dermis, porcine dermis, Used in contaminated fields, acutely incarcerated
porcine small intestine submucosa, fetal groin hernias associated with tissue necrosis and/or
er
dL while guidelines from nonsurgical societies recom- infection rate (1.27% no antibiotics, 2.53% antibiotic;
mend that glucose levels be no less than 140 mg/dL P = .364). Critics of this study include low overall infection
in noncritically ill hospitalized patients. 90 In addition, occurrences.93 Ríos et al. in 2001 published results from
treatment of perioperative short-term hyperglycemia can 216 patients undergoing incisional hernia reconstruction
improve outcomes. Kwon and colleagues reported that using polypropylene mesh who met criteria for a clean
patients who experienced perioperative hyperglycemia operation. The surgical wound infection rate was 13.6%
were at increased risk of infection, reoperation, and for antibiotic prophylaxis arm compared with 26.3%
death, but hyperglycemic patients who received insulin in those without preoperative antibiotic prophylaxis,
were at no greater risk than those with normal glucose which was statistically significant following multivariate
levels.91,92 analysis.94 A retrospective study published in 2013 evaluat-
ing deep and superficial infection rates following hernia
ANTIBIOTIC CHEMOPROPHYLAXIS repair treated with either two doses, 2 days, or 4 days of
Antibiotic prophylaxis in clean surgery with implanta- antibiotic chemoprophylaxis with no significant difference
tion of prosthetic material is widely accepted; however, a in infection rates between groups during a 4- to 6-week
recent prospective randomized control study evaluated follow-up.95 Standard CDC-based recommendations for
the administration of single-dose cefazolin for prosthetic antibiotic chemoprophylaxis are recommended. Antibiotic
hernia repairs. This study group was compared with no prophylaxis duration for prior mesh infections is not
antibiotic prophylaxis with no significant difference in available in the literature.
558 SECTION I Esophagus and Hernia
DRAINS CONCLUSION
Often prophylactic routine use of subcutaneous drains The goal of a hernia operation is to optimize the current
reportedly decreases wound complications such as infec- repair by preventing complications and recurrence. This is
tion, seroma, or hematoma. However, the indication for done by identifying the potential risk factors preoperatively,
the use of drains in ventral hernia repair is not studied. with appropriate optimizing of nutrition, blood glucose,
Although closed-suction surgical drains are used to remove weight, smoking, and contamination prior to operative
fluids (seroma, lymph, blood, abscess) accumulating repair. Surgery should include avoiding visceral injury,
in the surgical field, the occurrence of postoperative closing midline to the extent possible, using component
seroma, hematoma, and abscess is independent of drain separation, and reinforcing with prosthetic material
presence. Drains also have several disadvantages includ- appropriate for the situation. Perioperatively, expedit-
ing patient discomfort, limitations of mobilization, and ing ambulation, nutrition, and mitigating risk factors
increasing inpatient and home nursing care. There are no for infection and complications are important steps to
RCTs comparing drains versus no drains after incisional augment future hernia recurrence prevention.
hernia repair.96 The use of subcutaneous drains following
laparotomy closure has not shown a benefit.8 In settings
where drains are used per surgeon preference, prolonged REFERENCES
drain duration for a goal of 40 mL/day effluent was
t
1. Awad ZT, Puri V, LeBlanc K, et al. Mechanisms of ventral hernia
linked to BMI of 35 or more and operations longer than
ne
recurrence after mesh repair and a new proposed classification.
210 minutes. There was no relationship between the J Am Coll Surg. 2005;201(1):132-140.
incidences of seroma and drain duration, but there was 2. Breuing K, Bulter CE, Ferzoco S, et al. The Ventral Hernia Working
a direct linear relationship between wound complications Group. Incisional ventral hernias: review of the literature and
k.
and drain duration while adjusting for obesity.97 recommendations regarding the grading and technique of repair.
Surgery. 2010;148(3):544-558.
The use of retrorectus drains following Rives-Stoppa 3. Hart D, Postlethwait RW, Brown IW Jr, Smith WW, Johnson PA.
oo
or transverse abdominis muscle release has not been Postoperative wound infections: a further report on ultraviolet
published in the literature to date; however, postoperative irradiation with comments on the recent (1964) national research
seromas and hematomas in this location can be more council cooperative study report. Ann Surg. 1968;167(5):728-
743.
yb
difficult to access percutaneously and anecdotally supports 4. Garner JS. CDC guideline for prevention of surgical wound infec-
the placement of postoperative drains. The effect of tions, 1985. Supersedes guideline for prevention of surgical wound
drain placement in this location needs to be researched infections published in 1982. (Originally published in November
further.
er
1985). Revised. Infect Control. 1986;7(3):193-200.
5. Kanters AE, Krpata DM, Biatnik JA, Novitsky YM, Rosen MJ. Modified
ACTIVITY RESTRICTIONS hernia grading scale to stratify surgical site occurrence after open
ventral hernia repairs. J Am Coll Surg. 2012;215(6):787-793.
rg
No prospective studies were found on the restriction of 6. Berger RL, Li LT, Hicks SC, Davila JA, Kao LS, Liang MK. Develop-
physical activity after abdominal incisions. In questionnaires ment and validation of a risk-stratification score for surgical site
sent to surgeons in both 1998 and again in 2014, there occurrence and surgical site infection after open ventral hernia
su
2015;19(1):1-24.
an adverse impact on the return to normal activity and 9. Tebby J, Lecky F, Edwards A, et al. Outcomes of polytrauma patients
delay the return to work. with diabetes mellitus. BMC Med. 2014;12:111.
ht
17. Sorensen LT, Jorgensen S, Petersen LJ, et al. Acute effects of nicotine pilot study and preliminary outcomes. Plast Reconstr Surg. 2004;134(4
and smoking on blood flow, tissue oxygen, and aerobe metabolism suppl 2):151S-159S.
of the skin and subcutis. J Surg Res. 2009;152(2):224-230. 41. Kurz A, Sessler D I, Lenhardt R. Perioperative normothermia to
18. Sorensen LT, Hemmingsen UB, Kirkeby LT, Kallehave F, Jørgensen reduce the incidence of surgical-wound infection and shorten
LN. Smoking is a risk factor for incisional hernia. Arch Surg. hospitalization. N Engl J Med. 1996;334(19):1209-1216.
2005;140(2):119-123. 42. Fink C, Baumann P, Wente MN, et al. Incisional hernia rate 3 years
19. Tonnesen H, Rosenberg J, Nielsen HJ, et al. Effect of preoperative after midline laparotomy. Br J Surg. 2014;101(2):51-54.
abstinence on poor postoperative outcome in alcohol misusers: 43. Diener MK, Voss S, Jensen K, Buchler MW, Seiler CM. Elective
randomized controlled trial. BMJ. 1999;318(7194):1311-1316. midline laparotomy closure: the inline systematic review and
20. Sugerman HJ, Kellum JM Jr, Reines HD, et al. Greater risk of meta-analysis. Ann Surg. 2010;251(5):843-856.
incisional hernia with morbidly obese than steroid-dependent 44. Young J, Gilbert AI, Graham MF. The use of ultrasound in the
patients and low recurrence with prefascial polypropylene mesh. diagnosis of abdominal wall hernias. Hernia. 2007;11(4):347-351.
Am J Surg. 1996;171(1):80-84. 45. van ‘t Riet M, Steyerberg EW, Nellensteyn J, Bonjer HJ, Bonjer
21. Wick EC, Hirose K, Shore AD, et al. Surgical site infections and HJ, Jeekel J. Meta-analysis of techniques for closure of midline
cost in obese patients undergoing colorectal surgery. Arch Surg. abdominal incisions. Br J Surg. 2002;89(11):1350-1356.
2011;146(9):1068-1072. 46. Jenkins TP. The burst abdominal wound: a mechanical approach.
22. Eid MG, Mattar SG, Hamad G, et al. Repair of ventral hernias in Br J Surg. 1976;63(11):873-876.
morbidly obese patients undergoing laparoscopic gastric bypass 47. Varshney S, Manek P, Johnson CD. Six-fold suture: wound length
should not be deferred. Surg Endosc. 2004;18(2):207-210. ration for abdominal closure. Ann R Coll Surg Engl. 1999;81(5):333-336.
23. Tsereteli Z, Pryor BA, Heniford BT, Park A, Voeller G, Ramshaw 48. Israelsson LA, Jonsson T. Suture length to wound length
BJ. Laparoscopic ventral hernia repair (LVHR) in morbidly obese ratio and healing of midline laparotomy incisions. Br J Surg.
patients. Hernia. 2007;12(3):233-238. 1993;80(10):1284-1286.
t
24. Froylich D, Segal M, Weinstein A, Hatib K, Shiloni E, Hazzan D. 49. Deerenberg EB, Harlaar JJ, Steyerberg EW, et al. Small bites versus
ne
Laparoscopic versus open ventral hernia repair in obese patients: large bites for closure of abdominal midline incisions (STITCH):
a long-term follow-up. Surg Endosc. 2015;30(2):670-675. a double-blind, multicentre, randomized controlled trial. Lancet.
25. Lee J, Mabardy A, Kermani R, Lopez M, Pecquex N, McCluney A. 2015;386(10000):1254-1260.
Laparoscopic vs open ventral hernia repair in the era of obesity. 50. Gutierrez de la Pena C, Medina Achirica C, Dominguez-Adame E,
k.
JAMA Surg. 2013;148(8):723-726. Medina Diez J. Primary closure of laparotomies with high risk of
26. Regner JL, Mrdutt MM, Munoz-Maldonado Y. Tailoring surgical incisional hernia using prosthetic material: analysis of usefulness.
approach for elective ventral hernia repair based on obesity and
oo
Hernia. 2003;7(3):134-136.
National Surgical Quality Improvement Program outcomes. Am J 51. El-Khadrawy OH, Moussa G, Mansour O, Hashish MS. Prophylactic
Surg. 2015;210(6):1024-1030. prosthetic reinforcement of midline abdominal incisions in high-risk
27. Newcomb WL, Polhill JL, Chen AY, et al. Staged hernia repair patients. Hernia. 2009;13(3):264-274.
yb
preceded by gastric bypass for the treatment of morbidly obese 52. Caro-Tarrago A, Olona Casas C, Jimenez Salido A, Duque Guilera
patients with complex ventral hernias. Hernia. 2008;12(5):465-469. E, Moreno Fernandez F, Vicente Guillen V. Prevention of incisional
28. Praveen Raj P, Senthilnathan P, Kumaravel R, et al. Concomitant hernia in midline laparotomy with an onlay mesh: a randomized
laparoscopic ventral hernia mesh repair and bariatric surgery: clinical trial. World J Surg. 2014;38(9):2223-2230.
er
a retrospective study from a tertiary care center. Obes Surg. 53. Garcia-Urena MA, Lopez-Monchus J, et al. Randomized controlled
2012;22(5):685-689. trial of the use of a large-pore polypropylene mesh to prevent inci-
29. Raziel A, Sakran N, Szold A, Goitein D. Concomitant bariatric and sional hernia in colorectal surgery. Ann Surg. 2015;261(5):376-381.
rg
ventral/incisional hernia surgery in morbidly obese patients. Surg 54. Timmermans L, Eker HH, Steyerberg EW, et al. Short-term results
Endosc. 2012;28(4):1209-1212. of a randomized controlled trial comparing primary suture with
30. Spaniolas K, Kasten K, Mozer A, et al. Synchronous ventral primary glued mesh augmentation to prevent incisional hernia.
su
hernia repair in patients undergoing bariatric surgery. Obes Surg. Ann Surg. 2015;261(2):276-281.
2015;25(10):1-5. 55. Flum DR, Horvath K, Koepsell T. Have outcomes of incisional
31. Houck JP, Rypins EB, Sarfeh IJ, et al. Repair of incisional hernia. hernia repair improved with time? A population-based analysis.
Surg Gynecol Obstet. 1989;169(5):397-399. Ann Surg. 2003;237(1):129-135.
://
32. Luijendijk RW, Hop WC, van den Tol MP, et al. A comparison of 56. Burger JW, Luijendijk RW, Hop WC, Halm JA, Verdaasdonk
suture repair with mesh repair for incisional hernia. N Engl J Med. EG, Jeekel J. Long-term follow-up of a randomized controlled
2000;343(6):392-398. trial of suture versus mesh repair of incisional hernia. Ann Surg.
tp
preoperative oral immune-enhancing nutritional supplement on elective repair in small umbilical and epigastric hernias. A Nation-
patients at high risk of infection after cardiac surgery: a randomized wide Register Study. World J Surg. 2013;37(11):2548-2552.
placebo-controlled trial. Lancet. 2001;358(9283):696-701. 58. Conze J, Klinge U, Schumpelick V. Incisional hernia. Chirurg.
35. Brady M, Kinn S, Stuar P. Preoperative fasting for adults to 2005;76(9):897-909.
prevent perioperative complications. Cochrane Database Syst Rev. 59. Binnebosel M, Rosch R, Junge K, et al. Biomechanical analyses
2013;(4):CD004423. of overlap and mesh dislocation in an incisional hernia model in
36. Noblett SE, Watson DS, Huong H, Davison B, Hainsworth PJ, Horgan vitro. Surgery. 2007;142(3):365-371.
AF. Pre-operative oral carbohydrate loading in colorectal surgery: 60. Lambrecht J. Overlap-coefficient for the relationship between mesh
a randomized controlled trial. Colorectal Dis. 2006;8(7):563-569. size and defect size in laparoscopic ventral hernia surgery. Hernia.
37. Ousley J, Baucom RB, Poulose BK, et al. Previous methicillin-resistant 2011;15(4):473-474.
Staphylococcus aureus infection independent of body site increases 61. Cobb WS, Burns JM, Kercher KW, Matthews BD, Norton J, Todd
odds of surgical site infection after ventral hernia repair. J Am Coll Heniford B. Normal intraabdominal pressure in healthy adults. J
Surg. 2015;221(2):470-477. Surg Res. 2005;129(2):231-235.
38. Buehlmann M, Frei RM, Fenner LM, Dangel M, Fluckiger U, Widmer 62. Quandell H, O’Dwyer PJ. Relationship between ventral hernia defect
AF. Highly effective regimen for decolonization of methicillin- area and intra-abdominal pressure: dynamic in vivo measurement.
resistant Staphylococcus aureus carriers. Infect Control Hosp Epidemiol. Surg Endosc. 2016;30(4):1480-1484.
2008;29(6):510-516. 63. Holihan JL, Bondre I, Askenasy EP, et al. Ventral Hernia Outcomes
39. Bode LG, Kluytmans JA, Wertheim HF, et al. Preventing surgical-site Collaborative (VHOC) Writing Group. Sublay versus underlay in
infections in nasal carriers of Staphylococcus aureus. N Engl J Med. open ventral hernia repair. J Surg Res. 2016;202:26-32.
2010;362(1):9-17. 64. Zeichen MS, Lujan HJ, Mata WN, et al. Closure versus non-closure
40. Fayezizadeh M, Rosen MJ, Novitsky YW, Novitsky YW. Enhanced of hernia defect during laparoscopic ventral hernia repair with
recovery after surgery pathway for abdominal wall reconstruction: mesh. Hernia. 2013;17(5):589-596.
560 SECTION I Esophagus and Hernia
65. Nguyen DH, Nguyen MT, Askenasy EP, Kao LS, Liang MK. Primary bio-prosthetic mesh in the repair of contaminated ventral hernias.
fascial closure with laparoscopic ventral hernia repair: systematic Hernia. 2013;17(1):31-35.
review. World J Surg. 2014;38:3097-3104. 85. Huerta S, Varshney A, Patel PM, Mayo HG, Livingston E. Biological
66. Wennergren JE, Askenasy EP, Greenberg JA, et al. Laparo- mesh implants for abdominal hernia repair: US Food and Drug
scopic ventral hernia repair with primary fascial closure versus Administration approval process and systematic review of its efficacy.
bridged repair: a risk-adjusted comparative study. Surg Endosc. JAMA Surg. 2016;151(4):374-381.
2015;30(8):3231-3238. 86. Darahzereshki A, Goldfarb M, Zehetner J, et al. Biologic versus
67. Clapp ML, Hicks SC, Awad SS, Liang MK. Trans-cutaneous closure nonbiologic mesh in ventral hernia repair: a systematic review and
of central defects (TCCD) in laparoscopic ventral hernia repairs meta-analysis. World J Surg. 2014;38(1):40-50.
(LVHR). World J Surg. 2013;37(1):42-51. 87. Bondre IL, Holihan JL, Askenasy EP, et al. Suture, synthetic,
68. Booth JH, Garvey PB, Baumann DP, et al. Primary fascial closure or biologic in contaiminated ventral hernia repair. J Surg Res.
with mesh reinforcement is superior to bridged mesh repair in 2016;200(2):488-494.
abdominal wall reconstruction. J Am Coll Surg. 2013;217(6):999-1009. 88. Deeken CR, Matthews BD. Characterization of the mechanical
69. Helgstrand F, Rosenberg J, Kehlet H, Jorgensen LN, Bisgaard T. strength, resorption properties, and histologic characteristics of a
Nationwide prospective study of outcomes after elective incisional fully absorbable material (poly-4-hydroxybutyrate-PHASIX Mesh)
hernia repair. J Am Coll Surg. 2013;2016(2):217-228. in a porcine model of hernia repair. ISRN Surg. 2013;28:238067.
70. Reynolds D, Davenport DL, Korosec RL, Roth JS. Financial implica- 89. Open Complex Ventral Incisional Hernia Repair Using Biosynthetic
tions of ventral hernia repair: a hospital cost analysis. J Gastrointest Material for Midline Fascial Closure Reinforcement (COBRA).
Surg. 2013;17(1):159-166. ClinicalTrials.gov NCT01325792. Accessed 2015.
71. Iannitti DA, Hope WW, Tsikitis V. Strength of tissue attachment 90. Qaseem A, Humphrey LL, Chou R, Snow V, Shekelle P, Clinical
to composite and ePTFE grafts after ventral hernia repair. JSLS. Guidelines Committee of the American College of Physicians. Use of
2007;11(4):415-421. intensive insulin therapy for the management of glycemic control in
t
72. Burger JW, Halm JA, Wijsmuller AR, ten Raa S, Jeekel J. Evaluation hospitalized patients: a clinical practice guideline from the American
ne
of new prosthetic meshes for ventral hernia repair. Surg Endosc. College of Physicians. Ann Intern Med. 2011;154(4):260-267.
2006;20(8):1320-1325. 91. Kwon S, Thompson R, Dellinger P, Farrohki E, Flum D. Importance
73. Young RM, Gustafson R, Dinsmore RC. Sepramesh vs. Dualmesh of perioperative glycemic control in general surgery: a report from
for abdominal wall hernia repairs in a rabbit model. Curr Surg. the surgical care and outcomes assessment program. Ann Surg.
k.
2004;61(1):77-79. 2013;257(1):8-14.
74. Levy S, Plymale MA, Miller MT, Davenport DL, Roth JS. Laparoscopic 92. Buchleitner AM, Martinez-Alonso M, Hernandez M, Solà I, Mauricio
parastomal hernia repair: no different than a laparoscopic ventral D. Perioperative glycemic control for diabetic patients undergoing
oo
hernia repair? Surg Endosc. 2016;30(4):1542-1546. surgery. Cochrane Database Syst Rev. 2012;(9):CD007315.
75. Choi JJ, Palaniappa NC, Dallas KB, Rudich TB, Colon MJ, Divino 93. Mehrabi Bahar M, Jabbari Nooghabi A, Jabbari Nooghabi M,
CM. Use of mesh during ventral hernia repair in clean-contaminated Jangjoo A. The role of prophylactic cefazolin in the prevention of
yb
and contaminated cases: outcomes of 33,832 cases. Ann Surg. infection after various types of abdominal wall hernia repair with
2012;255(1):176-180. mesh. Asian J Surg. 2015;38(3):139-144.
76. Gray SH, Vick CC, Graham LA, Finan KR, Neumayer LA, Hawn 94. Ríos A, Rodríguez JM, Munitiz V, Alcaraz P, Pérez Flores D, Parrilla P.
MT. Risk of complications from enterotomy or unplanned bowel Antibiotic prophylaxis in incisional hernia repair using a prosthesis.
er
resection during elective hernia repair. Arch Surg. 2008;143(6):582- Hernia. 2001;5(3):148-152.
586. 95. Ioannidis O, Paraskevas G, Varnalidis I, et al. Hernia mesh repair of
77. Hawn MT, Gray SH, Snyder CW, Graham LA, Finan KR, Vick CC. anterior abdominal wall and antibiotic chemoprophylaxis: multiple
rg
Predictors of mesh explantation after incisional hernia repair. Am doses of antibiotics failed to prevent or reduce wound infection.
J Surg. 2011;202(1):28-33. Chirurgia. 2013;108(6):835-839.
78. Carbonell AM, Criss CN, Cobb WS, Novitsky YW, Rosen MJ. Out- 96. Gurusamy KS, Allen VB. Wound drains after incisional hernia
su
comes of synthetic mesh in contaminated ventral hernia repairs. repair. Cochrane Database System Rev. 2013;(12):CD005570.
J Am Coll Surg. 2013;217(6):991-998. 97. Plymale MA, Harris JW, Roth JS, Smith N, Levy S, Scott Roth J.
79. Cobb WS, Warren JA, Ewing JA, Burnikel A, Merchant M, Carbonell Abdominal wall reconstruction: the uncertainty of the impact of
AM. Open retromuscular mesh repair of complex incisional drain duration upon outcomes. Am Surg. 2016;82(3):207-211.
://
hernia: predictors of wound events and recurrence. J Am Coll 98. Kehlet H, Callesen T. Recommendations for convalescence
Surg. 2015;220(4):606-613. after hernia surgery. A questionnaire study. Ugeskr Laeger.
80. Bachman S, Ramshaw B. Prosthetic material in ventral hernia 1998;160(7):1008-1009.
tp
repair: how do I choose? Surg Clin North Am. 2008;88(1):101-112. 99. Pommergaard HC, Burcharth J, Danielsen A, et al. No consensus
81. Primus FE, Harris HW. A critical review of biologic mesh use in on restrictions on physical activity to prevent incisional hernias
ventral hernia repairs under contaminated conditions. Hernia. after surgery. Hernia. 2013;18(4):495-500.
ht
t
completely resolve the respiratory problems and should be buds form.
ne
performed urgently in the neonatal period. Subsequent 2. Pseudoglandular Phase: defined lobar structures are
investigations have shown that although the diaphragmatic present by week 6. From week 7 to 16, airways dif-
defect requires surgical repair, success depends more on ferentiate into bronchioles and terminal bronchioles.
k.
the preoperative and postoperative management of the 3. Canalicular Phase: during week 16 to 24, crude air sacs
associated physiological derangements. and pneumocytes develop.
oo
In the last few decades, there have been numerous 4. Saccular Phase: until term, continued remodeling of
advances in prenatal diagnosis, neonatal care (ventilatory alveolar airspaces occurs, along with production of
management, surfactant, inhaled nitric oxide [iNO], high- surfactant.
yb
frequency oscillatory ventilation [HFOV], extracorporeal 5. Alveolar Phase: shortly after birth, alveoli begin to
membrane oxygenation [ECMO]), and anesthetic and appear, with maturation and multiplication taking
surgical techniques. Despite this, the mortality remains place until the age of 8.
as high as 20% to 30%. The CDH Study Group is an
er
to answer questions and monitor outcomes. Since 1995, and due to the small capacity of the abdomen, it extends
112 centers have participated, collecting data on more into the umbilical cord. At 10 weeks, the abdomen is large
than 8000 children and has generated 35 manuscripts.2 enough to accommodate the elongating loops of bowel,
su
These investigations, along with others, have advanced and they return while undergoing a counterclockwise
our knowledge and ability to treat this once uniformly rotation around the base of the mesentery.3,4 If closure
fatal neonatal disease. of the pleuroperitoneal canal has not occurred by the
://
KEYWORDS
CDH, diaphragmatic hernia, congenital hernia, EC
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
562 SECTION I Esophagus and Hernia
CLASSIFICATION
t
ne
GENERAL Esophageal Aortic Posterolateral defect:
hiatus hiatus Bochdalek hernia
The overall incidence of CDH is 1 : 2000 to 1 : 5000 live
k.
births.9,10 Left-sided defects are more common (80% to
FIGURE 51.1 Location of diaphragmatic defect.
85%), followed by right (15%) or bilateral (<5%) defects.
oo
In general, posterolateral left-sided hernias are more likely
to be symptomatic and identified in the neonatal period,
whereas anterior defects may go undetected.
they may present with incarceration or strangulation.14
yb
BOCHDALEK (POSTEROLATERAL) HERNIA Most often, there is low associated morbidity if they are
The classic description of the congenital defect was by repaired promptly. Surgical repair is most often per-
Bochdalek, for whom the defect was named. Typically, a formed via a transabdominal (open vs. laparoscopic)
er
the thoracic cavity. The size of the defect is variable and in the pentalogy of Cantrell.15 The syndrome involves a
may range from the classic small defect to complete ventral midline developmental defect resulting in anterior
diaphragmatic agenesis (Fig. 51.1). Most commonly, small diaphragmatic hernia, epigastric omphalocele, sternal
su
bowel, spleen, stomach, liver, and/or colon may be in the cleft, intracardiac defect (ventricular septal defect most
chest. A nonmuscularized membrane forms a hernia sac commonly), and pericardial defect (usually results in
in 10% to 20%.11 ectopia cordis). Pentalogy is associated with an extremely
://
and associated anomalous hepatic venous drainage is present symptomatically within the first 6 hours of life.16,17
common. There is a high incidence of preterm complica- Overall, cardiac anomalies are most common, affecting
tions, associated comorbidities, and frequent need for between 25% and 40%. They most commonly involve the
ECMO.12 Bilateral congenital defects are rare (<5%) cardiac outflow tract, including ventricular septal defects,
and are nearly always fatal due to arrest of bilateral lung tetralogy of Fallot, transposition, and coarctation.18 Limb
development.13 anomalies are seen in 30% and include limb shortening
and costovertebral defects.19 The most common airway
MORGAGNI (ANTERIOR) HERNIA anomalies are congenital tracheal stenosis and bron-
Morgagni hernias account for less than 2% of diaphrag- chial stenosis on the ipsilateral side.20 Other reported
matic defects. They are typically located anteromedially defects are esophageal atresia, omphalocele, and cleft
on either side at the junction of the septum transversum palate.16
and the anterior thoracic wall. They are most often Approximately 10% will have chromosomal anomalies
asymptomatic, as the amount of pulmonary compression is such as trisomy (21, 18, 13) or other syndromes (Frey,
minimal compared to posterolateral defects. In addition, Beckwith-Wiedemann, and Fryns). When CDH is associ-
patients typically have transverse colon herniation into ated with an abnormal karyotype, the outcome is poor.21
the anterior mediastinum covered by a peritoneal sac. Nearly all stillborn infants with CDH have associated
Most are discovered incidentally as an anterior chest lethal anomalies, with neural tube defects and cardiac
mass or suspected pneumonia in an older child. Rarely, anomalies being most common.
Congenital Diaphragmatic Hernia CHAPTER 51 563
t
prenatal ultrasound, as intermittent herniation of viscera
ne
into the thoracic cavity may occur.22 In addition, CDH may
be confused with congenital cystic lung malformations.
When CDH is identified prenatally, it should prompt
k.
a search for associated anomalies, especially those affect-
ing the cardiac and nervous systems. Fetal karyotyping
oo
should be offered with either chorionic villi sampling or
amniocentesis, whichever is appropriate for gestational age FIGURE 51.2 Fetal magnetic resonance imaging (coronal view)
at diagnosis. The parents should be encouraged to deliver demonstrates presence of bowel in the left chest consistent with
yb
in an appropriate tertiary care center with capabilities of congenital diaphragmatic hernia.
advance respiratory care strategies and urgent pediatric
surgical availability.
er
diagnosed CDH is the lung-to-head ratio (LHR).26 Ultra- been used to characterize complex fetal anomalies (Fig.
sound is used to measure the fetal lung area relative to 51.2). Ultrafast MRI uses shorter acquisition times to
the head circumference, therefore providing an indirect eliminate issues with fetal movement. An additional benefit
su
assessment of the contralateral lung volume and estimating is reduced intraobserver variability of fetal lung volume
the likelihood of pulmonary hypoplasia. In general, an LHR measurements commonly seen with ultrasound techniques.
less than 1.0 is associated with low survival, while a ratio It has proven superior to “liver-up” versus “liver-down,” as
://
of more than 1.4 have improved outcomes.23,26 Original it allows the exact quantification of the amount of liver
cutoffs for LHR were between gestational age 24 to 26 herniated into the chest.31 It appears from systematic
tp
weeks. Due to a more rapid growth of fetal lung compared reviews that measurement of fetal lung volumes, liver
to the head circumference that occurs during gestation, position, and side of defect identified with MRI correlates
the observed-to-expected (O/E) LHR was introduced, as with neonatal survival.32
ht
[H]
t
ne
k.
A B
oo
FIGURE 51.3 Chest radiograph demonstrating left diaphragmatic hernia immediately postnatally. Although the chest is opacified early (A),
continued introduction of air results in eventual bowel distention (B).
yb
intrathoracic viscera. Mediastinal compression may shift gas exchange. Reduction of hernia contents into the
the trachea to the contralateral side or produce tension abdomen rarely results in reexpansion of the lungs, as
physiology with obstructed venous return. Pallor, cyanosis, they are hypoplastic rather than atelectatic.34 Due to these
er
sternal retractions, and grunting all signify increased work findings, the focus has now shifted to delayed repair after
of breathing and impending deterioration. medical stabilization. CDH is now considered a physiologic
rg
A chest radiograph should be done soon after birth, emergency instead of a surgical one.
which typically demonstrates loops of bowel located in the The resuscitation of a neonate with CDH starts with
chest. The nasogastric tube location is useful for identifying prompt endotracheal intubation, as ventilation by mask
su
the stomach position within the chest or abdomen. Lower or Ambu bag may cause significant bowel distention and
chest opacification may be seen with initial radiographs, worsening ventilation.
as the bowel has not yet accumulated with air (Fig. 51.3). A nasogastric tube should be inserted to allow gastric
://
In addition to the chest x-ray, additional imaging and intestinal decompression. Typically, umbilical venous
studies include a transthoracic echocardiogram and head and arterial catheters are placed along with oxygen satura-
tp
ultrasound. The echocardiogram is performed to rule tion probes in the pre- and postductal locations to allow for
out associated anomalies and assess ventricular function/ shunt estimation. Excessive stimuli can easily exacerbate
pulmonary hypotension. This is best performed on day pulmonary pressures and lead to increased shunt flow/
ht
of life 2 to not overestimate the degree of pulmonary desaturations. For this reason, infants are kept sedated
hypertension. Complex imaging such as a chest ultra- in beds/incubators with radiant warmers and external
sound, computerized tomography (CT), MRI, or upper stimulation is limited. Muscle paralysis should only be used
gastrointestinal contrast study are not necessary to confirm in extreme circumstances due to ventilatory consequences
presence of a diaphragmatic hernia. Right-sided defects and added morbidity.
may warrant MRI to rule out hepatopulmonary fusion With advances in postnatal resuscitation and care, the
and to assess the vascular anatomy of the mediastinum survival rate has improved significantly. One of the most
and liver.33 beneficial changes seems to be the idea of less aggressive
ventilatory strategies. Previously, aggressive hyperventila-
tion and induced alkalosis (PaCO2 as low as 20 mm Hg)
TREATMENT AND OUTCOMES were used with the thought that this would minimize
pulmonary hypertension and the associated shunting. To
MEDICAL OPTIMIZATION achieve these extremely low PaCO2 levels, very high peak
CDH was previously felt to constitute a surgical emergency, inspiratory pressures were used, which are potentially
and infants typically underwent surgery in the first few damaging to the lungs and may ultimately worsen chronic
hours of life. It was later understood that the majority of lung disease. In 1985, Wung and colleagues proposed
patients showed a deterioration in respiratory mechanics permissive hypercapnia as a way to limit barotrauma and
after a brief postoperative “honeymoon period” of adequate still achieve sufficient tissue oxygenation.35 This concept
Congenital Diaphragmatic Hernia CHAPTER 51 565
has been widely applied, and hypercapnia (PaCO2 60 to requires systemic heparin anticoagulation to prevent
65 mm Hg) is tolerated (permissive hypercapnia) as long thrombosis of the extracorporeal circuit and oxygenator.
as the pH is greater than 7.2.36,37 Hemorrhagic complications may be seen in up to 60% and
include bleeding at the cannulation site, surgical site, head,
VENTILATORY ADJUNCTS chest, and gastrointestinal tract.48 Due to hemorrhagic
Failure of conventional ventilation is suggested by worsen- complications, surgical repair of CDH is typically delayed
ing tachypnea (evidenced by retractions, paradoxical chest until ECMO decannulation.
movement), inadequate oxygenation (preductal oxygen
saturation <85%), severe hypercarbia (PaCO2 >65 mm TIMING OF REPAIR
Hg), or worsening pulmonary hypertension (widened Current management involves early stabilization and
gap between pre- and postductal oxygen saturations). If delayed repair, although the exact timing of surgery is
customary measures cannot achieve adequate oxygenation not known.49,50 The period of preoperative stabilization
and ventilation, then nonconventional ventilatory modes varies between institutions, but most allow ventilatory
may be used. HFOV is one of the more popular modes weaning and serial cardiac echocardiograms to evaluate
due to its gentle ventilation properties and limitation of improvement in pulmonary hypertension. For the stable
barotrauma.38,39 infant without pulmonary hypertension, repair can be
Persistent pulmonary hypertension is the major factor safely done after a 48-hour period of stabilization and
increasing pulmonary resistance and causing right-to-left adjustment to postnatal life.
t
shunting with hypoxemia. In theory, vasodilator therapy Repair on ECMO poses a significant challenge due to
ne
should lead to improvement, but it has shown to be systemic heparinization and difficulty in transportation
unsuccessful due to inadequate pulmonary vasodilation to the operative suite. Outcomes appear to be improved
and increased shunting from systemic hypotension. Nitric for those who undergo surgical repair following ECMO,
k.
oxide, also known as endothelium-derived relaxing factor, with significantly increased survival, lower rates of surgical
directly stimulates cyclic GMP in vascular smooth muscle, bleeding, and decreased total duration of ECMO therapy
oo
causing relaxation. iNO diffuses across the alveoli to the compared to those repaired on pump.51
vascular smooth muscle, allowing for selective pulmonary
vasodilation. iNO has been shown to induce improvement SUMMARY
yb
in oxygenation when used at low doses (10 to 20 parts In summary, there are a few major principles of initial
per million) and is now used as an adjunct to ventilatory preoperative CDH management.35-48,52
strategies.40 Unfortunately, clinical studies have been 1. Minimizing the onset and impact of pulmonary
mixed; however, improved survival or decreased need for hypertension.
er
ECMO has been consistently demonstrated.41,42 2. Gentle ventilation with permissive hypercapnia
minimizes iatrogenic injury. Unconventional modes
EXTRACORPOREAL OXYGENATION
rg
which the persistent fetal circulation can resolve. The traditional approach is through a subcostal incision
The most common parameter used to determine the on the side of the defect. Once the abdomen is entered,
need for ECMO is the oxygenation index (OI), derived the bowel is reduced from the chest with gentle downward
ht
from the formula: OI = (FiO2 × Mean Airway Pressure) traction. Typically, the spleen and liver, if present, are the
/ PaO2. The indication for instituting ECMO is variable, last organs to be reduced. Great care should be exercised
but it is typically when OI is greater than 25 to 40.44 Other during mobilization, as the spleen and liver may develop
indications are consistent preductal oxygen saturations subcapsular hematomas and life-threatening hemorrhage
less than 85%, persistent metabolic acidosis, or hypoten- with traumatic reduction. If a hernia sac is present, it should
sion refractory to pressors. ECMO is contraindicated in be excised to minimize the risk of recurrence. Although
prematurity (gestational age <34 weeks), head ultrasound the anterior rim of diaphragm is usually prominent, the
demonstrating severe intraventricular hemorrhage, irrevers- posterior rim is typically diminished and obscured in
ible cardiac disease, or with another lethal congenital the retroperitoneal tissue. The posterior diaphragmatic
anomaly.45 tissue must then be mobilized from the retroperitoneum,
ECMO can be accomplished via either a venoarte- revealing the size of the diaphragmatic defect. If adequate,
rial (VA) or venovenous (VV) technique with cannulas a primary repair with interrupted nonabsorbable suture
placed in the carotid artery (VA) and internal jugular material is preferred (Fig. 51.4).
vein (VA/VV). Traditionally, VA bypass has been used If the size of the defect precludes primary closure,
due to the underlying pulmonary hypertension. The use of prosthetic mesh is indicated (Fig. 51.5). The most
efficacy and survival with VV bypass has been found to commonly used mesh is Gore-Tex, although more recent
be comparable to VA, with lower short-term neurologic biologic and absorbable mesh have also been used with no
sequelae due to avoidance of carotid ligation.46,47 ECMO difference in recurrence or postoperative complications.53
566 SECTION I Esophagus and Hernia
t
ne
Before After
k.
oo
yb
er
rg
su
://
During hernia.
ht
Overall, use of prosthetic mesh is associated with a higher thoracoscopic repair in an infant was reported in 1995.56
risk of recurrence, especially with a large initial defect Initial opposition for thoracoscopic repair was due to
or complete diaphragmatic agenesis.54 With extremely the concern that high end-tidal CO2 requiring increas-
large defects and loss of abdominal wall domain, return ingly high inspiratory pressures would worsen pulmonary
of viscera into the abdomen may result in inappropriately hypertension.57 Potential advantages are improved cosmesis,
high intraabdominal pressures. A Silastic sheet can be used improved surgical field visualization, and avoidance of
between fascial edges as a temporizing measure, with slow thoracotomy-associated musculoskeletal deformities.
closure over the ensuing days-weeks. Eventual abdominal The operation is most often performed with a standard
closure leads to a ventral hernia that can be dealt with endotracheal tube. Due to the associated pulmonary
outside the neonatal period. Additional techniques for hypoplasia, there is usually adequate room after the bowel
repair of defects include internal oblique rotation flap is reduced into the abdomen. Contralateral mainstem
or split abdominal wall muscle flap.55 intubation may also be used, if tolerated. The infant is
then positioned transversely at the end of the bed in lateral
THORACOSCOPIC APPROACH decubitus positioning (Fig. 51.6). The Veress needle is
Minimally invasive surgical techniques for repair of inserted in the 5th intercostal space at the lower edge of
CDH are increasingly being used. The first successful the scapula (mid-posterior axillary line), and low pressure
Congenital Diaphragmatic Hernia CHAPTER 51 567
Diaphragm Abdominal wall CO2 insufflation (3 to 5 mm Hg) is used. Typically, three
ports are used and include a 5-mm port in the site of the
Veress insertion (4-mm camera), 3-mm port in the left
anterolateral chest wall (bowel grasper), and 5/3-mm
convertible port in the right posterolateral chest wall
(bowel grasper, needle driver). The viscera are gently
reduced with blunt graspers into the abdominal cavity. The
posterolateral edge of the diaphragm is unfurled from the
retroperitoneal tissue revealing the extent of the hernia,
and needles can be introduced through the 3- to 5-mm
port or through the chest wall. The defect is then closed
with interrupted nonabsorbable sutures (2-0 Ethibond)
beginning superior-medially and ending inferior-laterally
(Fig. 51.7). Sutures are placed approximately 1 cm apart,
and knots may be tied either intra- or extracorporeally.
Some surgeons claim added benefit to “roughing up”
the edge of the smooth diaphragmatic membrane using
cautery.58 The most difficult technical portion involves the
t
Posterolateral defect: Bochdalek hernia lateral corner, where sutures must be placed around the
ne
Repair with mesh rib and tied extracorporeally in a subcutaneous pocket.
Initial reports in 2003 suggested that although appropri-
FIGURE 51.5 Open surgical repair with patch closure in large ate for Morgagni defects or those diagnosed outside the
k.
defects nonamenable to primary repair. neonatal period, the minimally invasive surgery (MIS)
approach for a newborn could not be recommended
oo
yb
er
rg
su
://
Left hand
ht
FIGURE 51.6 Operative positioning for thoracoscopic repair includes placing the patient transversely on the operative table with the
surgeon standing at the neonate’s head and monitors at the foot of the bed.
568 SECTION I Esophagus and Hernia
t
ne
A B
FIGURE 51.7 Thoracoscopic view of the chest before reduction (A) and with interrupted nonabsorbable sutures closing the diaphragmatic
k.
defect (B).
oo
due to the high failure rate and frequent rise in PCO2 Diaphragm Abdominal wall
levels.59 Other reports also corroborated a high recur-
rence rate between 20% and 40%,60,61 and more recent
yb
literature includes recurrence rates of 2% to 8%.62,63 A
recent meta-analysis including all nine published studies
to date (507 patients) supports thoracoscopic repair due
er
MORGAGNI
su
Bochdalek hernia
Primary repair
REFERENCES 27. Jani J, Nicolaides KH, Keller RL, et al. Observed to expected lung
area to head circumference ratio in the prediction of survival in
1. Gross RE. Congenital hernia of the diaphragm. Am J Dis Child. fetuses with isolated diaphragmatic hernia. Ultrasound Obstet Gynecol.
1946;71:579-592. 2007;30(1):67-71.
2. Harting MT, Lally KP. The congenital diaphragmatic hernia study 28. Hedrick HL, Danzer E, Merchant A, et al. Liver position and lung-to-
group registry update. Semin Fetal Neonatal Med. 2014;19(6):370-375. head ratio for prediction of extracorporeal membrane oxygenation
3. Moore KL, Persaud TVN, Torchia MG, eds. The Developing Human: and survival in isolated left congenital diaphragmatic hernia. Am J
Clinically Oriented Embryology. 10th ed. Philadelphia: Saunders; 2016. Obstet Gynecol. 2007;197(4):422 e1-422 e424.
4. Moore KL, Persaud TVN, Torchia MG, eds. Before We Are Born: Essentials 29. Albanese CT, Lopoo J, Goldstein RB, et al. Fetal liver position and
of Embryology and Birth Defects. 9th ed. Philadelphia: Saunders; 2016. perinatal outcome for congenital diaphragmatic hernia. Prenat
5. Davies G, Reid L. Growth of the alveoli and pulmonary arteries in Diagn. 1998;18(11):1138-1142.
childhood. Thorax. 1970;25(6):669-681. 30. Mullassery D, Ba’ath ME, Jesudason EC, Losty PD. Value of liver
6. Rottier R, Tibboel D. Fetal lung and diaphragm development in herniation in prediction of outcome in fetal congenital diaphragmatic
congenital diaphragmatic hernia. Semin Perinatol. 2005;29(2):86-93. hernia: a systematic review and meta-analysis. Ultrasound Obstet Gynecol.
7. Rennie J. Rennie & Roberton’s Textbook of Neonatology. 5th ed. Elsevier; 2010;35(5):609-614.
2012. 31. Lazar DA, Ruano R, Cass DL, et al. Defining “liver-up”: does
8. Yamataka T, Puri P. Pulmonary artery structural changes in pulmonary the volume of liver herniation predict outcome for fetuses with
hypertension complicating congenital diaphragmatic hernia. J Pediatr isolated left-sided congenital diaphragmatic hernia? J Pediatr Surg.
Surg. 1997;32(3):387-390. 2012;47(6):1058-1062.
9. Langham MR Jr, Kays DW, Ledbetter DJ, Frentzen B, Sanford LL, 32. Mayer S, Klaritsch P, Petersen S, et al. The correlation between lung
Richards DS. Congenital diaphragmatic hernia. Epidemiology and volume and liver herniation measurements by fetal MRI in isolated
outcome. Clin Perinatol. 1996;23(4):671-688. congenital diaphragmatic hernia: a systematic review and meta-analysis
t
10. Dott MM, Wong LY, Rasmussen SA. Population-based study of of observational studies. Prenat Diagn. 2011;31(11):1086-1096.
ne
congenital diaphragmatic hernia: risk factors and survival in Met- 33. Katz S, Kidron D, Litmanovitz I, Erez I, Dolfin Z. Fibrous fusion
ropolitan Atlanta, 1968-1999. Birth Defects Res A Clin Mol Teratol. between the liver and the lung: an unusual complication of right
2003;67(4):261-267. congenital diaphragmatic hernia. J Pediatr Surg. 1998;33(5):
11. Zamora IJ, Cass DL, Lee TC, et al. The presence of a hernia sac in 766-767.
k.
congenital diaphragmatic hernia is associated with better fetal lung 34. Sakai H, Tamura M, Hosokawa Y, Bryan AC, Barker GA, Bohn DJ.
growth and outcomes. J Pediatr Surg. 2013;48(6):1165-1171. Effect of surgical repair on respiratory mechanics in congenital
12. Hedrick HL, Crombleholme TM, Flake AW, et al. Right congenital diaphragmatic hernia. J Pediatr. 1987;111(3):432-438.
oo
diaphragmatic hernia: Prenatal assessment and outcome. J Pediatr 35. Wung JT, James LS, Kilchevsky E, James E. Management of infants
Surg. 2004;39(3):319-323, discussion 319–323. with severe respiratory failure and persistence of the fetal circulation,
13. Neville HL, Jaksic T, Wilson JM. Bilateral congenital diaphragmatic without hyperventilation. Pediatrics. 1985;76(4):488-494.
yb
hernia. J Pediatr Surg. 2003;38(3):522-524. 36. Kays DW, Langham MR Jr, Ledbetter DJ, Talbert JL. Detrimental
14. Kimmelstiel FM, Holgersen LO, Hilfer C. Retrosternal (Morgagni) effects of standard medical therapy in congenital diaphragmatic
hernia with small bowel obstruction secondary to a Richter’s incarcera- hernia. Ann Surg. 1999;230(3):340-348, discussion 348–351.
tion. J Pediatr Surg. 1987;22(11):998-1000. 37. Boloker J, Bateman DA, Wung JT, Stolar CJ. Congenital diaphrag-
er
15. Cantrell JR, Haller JA, Ravitch MM. A syndrome of congenital defects matic hernia in 120 infants treated consecutively with permissive
involving the abdominal wall, sternum, diaphragm, pericardium, hypercapnea/spontaneous respiration/elective repair. J Pediatr Surg.
and heart. Surg Gynecol Obstet. 1958;107(5):602-614. 2002;37(3):357-366.
rg
16. Fauza DO, Wilson JM. Congenital diaphragmatic hernia and associated 38. Reyes C, Chang LK, Waffarn F, Mir H, Warden MJ, Sills J. Delayed
anomalies: their incidence, identification, and impact on prognosis. repair of congenital diaphragmatic hernia with early high-frequency
J Pediatr Surg. 1994;29(8):1113-1117. oscillatory ventilation during preoperative stabilization. J Pediatr
su
17. Losty PD, Vanamo K, Rintala RJ, Donahoe PK, Schnitzer JJ, Lloyd Surg. 1998;33(7):1010-1014, discussion 1014–1016.
DA. Congenital diaphragmatic hernia—does the side of the defect 39. Cacciari A, Ruggeri G, Mordenti M, et al. High-frequency oscillatory
influence the incidence of associated malformations? J Pediatr Surg. ventilation versus conventional mechanical ventilation in congenital
1998;33(3):507-510. diaphragmatic hernia. Eur J Pediatr Surg. 2001;11(1):3-7.
://
18. Cohen MS, Rychik J, Bush DM, et al. Influence of congenital heart 40. Kinsella JP, Neish SR, Shaffer E, Abman SH. Low-dose inhalation
disease on survival in children with congenital diaphragmatic hernia. nitric oxide in persistent pulmonary hypertension of the newborn.
J Pediatr. 2002;141(1):25-30. Lancet. 1992;340(8823):819-820.
tp
19. Laudy JA, Van Gucht M, Van Dooren MF, Wladimiroff JW, Tibboel 41. Finer NN, Barrington KJ. Nitric oxide for respiratory failure in infants
D. Congenital diaphragmatic hernia: an evaluation of the prognostic born at or near term. Cochrane Database Syst Rev. 2001;(4):CD000399.
value of the lung-to-head ratio and other prenatal parameters. Prenat 42. Finer NN, Sun JW, Rich W, Knodel E, Barrington KJ. Random-
ht
Diagn. 2003;23(8):634-639. ized, prospective study of low-dose versus high-dose inhaled nitric
20. Nose K, Kamata S, Sawai T, et al. Airway anomalies in patients with oxide in the neonate with hypoxic respiratory failure. Pediatrics.
congenital diaphragmatic hernia. J Pediatr Surg. 2000;35(11):1562-1565. 2001;108(4):949-955.
21. Benjamin DR, Juul S, Siebert JR. Congenital posterolateral 43. Bartlett RH, Gazzaniga AB, Jefferies MR, Huxtable RF, Haiduc
diaphragmatic hernia: associated malformations. J Pediatr Surg. NJ, Fong SW. Extracorporeal membrane oxygenation (ECMO)
1988;23(10):899-903. cardiopulmonary support in infancy. Trans Am Soc Artif Intern Organs.
22. Adzick NS, Harrison MR, Glick PL, Nakayama DK, Manning FA, 1976;22:80-93.
deLorimier AA. Diaphragmatic hernia in the fetus: prenatal diagnosis 44. Does extracorporeal membrane oxygenation improve survival in
and outcome in 94 cases. J Pediatr Surg. 1985;20(4):357-361. neonates with congenital diaphragmatic hernia? The Congenital
23. Lipshutz GS, Albanese CT, Feldstein VA, et al. Prospective analysis Diaphragmatic Hernia Study Group. J Pediatr Surg. 1999;34(5):720-724,
of lung-to-head ratio predicts survival for patients with prena- discussion 724–725.
tally diagnosed congenital diaphragmatic hernia. J Pediatr Surg. 45. Annich G. ECMO: Extracorporeal Cardiopulmonary Support in Critical
1997;32(11):1634-1636. Care. 4th ed. Ann Arbor, MI: ELSO; 2012.
24. Wilson JM, Fauza DO, Lund DP, Benacerraf BR, Hendren WH. 46. Dimmitt RA, Moss RL, Rhine WD, Benitz WE, Henry MC, Vanmeurs
Antenatal diagnosis of isolated congenital diaphragmatic hernia is KP. Venoarterial versus venovenous extracorporeal membrane
not an indicator of outcome. J Pediatr Surg. 1994;29(6):815-819. oxygenation in congenital diaphragmatic hernia: the Extracorporeal
25. Benachi A, Cordier AG, Cannie M, Jani J. Advances in prenatal Life Support Organization Registry, 1990-1999. J Pediatr Surg.
diagnosis of congenital diaphragmatic hernia. Semin Fetal Neonatal 2001;36(8):1199-1204.
Med. 2014;19(6):331-337. 47. Guner YS, Khemani RG, Qureshi FG, et al. Outcome analysis
26. Metkus AP, Filly RA, Stringer MD, Harrison MR, Adzick NS. Sono- of neonates with congenital diaphragmatic hernia treated with
graphic predictors of survival in fetal diaphragmatic hernia. J Pediatr venovenous vs venoarterial extracorporeal membrane oxygenation.
Surg. 1996;31(1):148-151, discussion 151–152. J Pediatr Surg. 2009;44(9):1691-1701.
570 SECTION I Esophagus and Hernia
48. Vazquez WD, Cheu HW. Hemorrhagic complications and repair of 60. Cho SD, Krishnaswami S, Mckee JC, Zallen G, Silen ML, Bliss DW.
congenital diaphragmatic hernias: does timing of the repair make a Analysis of 29 consecutive thoracoscopic repairs of congenital
difference? Data from the Extracorporeal Life Support Organization. diaphragmatic hernia in neonates compared to historical controls.
J Pediatr Surg. 1994;29(8):1002-1005, discussion 1005–1006. J Pediatr Surg. 2009;44(1):80-86, discussion 86.
49. Langer JC, Filler RM, Bohn DJ, et al. Timing of surgery for congenital 61. Gander JW, Fisher JC, Gross ER, et al. Early recurrence of congenital
diaphragmatic hernia: is emergency operation necessary? J Pediatr diaphragmatic hernia is higher after thoracoscopic than open repair:
Surg. 1988;23(8):731-734. a single institutional study. J Pediatr Surg. 2011;46(7):1303-1308.
50. West KW, Bengston K, Rescorla FJ, Engle WA, Grosfeld JL. Delayed 62. Liem NT. Thoracoscopic approach in management of congenital
surgical repair and ECMO improves survival in congenital diaphrag- diaphragmatic hernia. Pediatr Surg Int. 2013;29(10):1061-1064.
matic hernia. Ann Surg. 1992;216(4):454-460, discussion 460–462. 63. Huang JS, Lau CT, Wong WY, Tao Q, Wong KK, Tam PK. Thora-
51. Partridge EA, Peranteau WH, Rintoul NE, et al. Timing of repair coscopic repair of congenital diaphragmatic hernia: two centres’
of congenital diaphragmatic hernia in patients supported by experience with 60 patients. Pediatr Surg Int. 2015;31(2):191-195.
extracorporeal membrane oxygenation (ECMO). J Pediatr Surg. 64. Zhu Y, Wu Y, Pu Q, Ma L, Liao H, Liu L. Minimally invasive surgery
2015;50(2):260-262. for congenital diaphragmatic hernia: a meta-analysis. Hernia.
52. Wilson JM, Lund DP, Lillehei CW, Vacanti JP. Congenital diaphrag- 2016;20(2):297-302.
matic hernia—a tale of two cities: the Boston experience. J Pediatr 65. Harrison MR, Adzick NS, Longaker MT, et al. Successful repair
Surg. 1997;32(3):401-405. in utero of a fetal diaphragmatic hernia after removal of herni-
53. Romao RL, Nasr A, Chiu PP, Langer JC. What is the best prosthetic ated viscera from the left thorax. N Engl J Med. 1990;322(22):
material for patch repair of congenital diaphragmatic hernia? 1582-1584.
Comparison and meta-analysis of porcine small intestinal submucosa 66. Wilson JM, DiFiore JW, Peters CA. Experimental fetal tracheal
and polytetrafluoroethylene. J Pediatr Surg. 2012;47(8):1496-1500. ligation prevents the pulmonary hypoplasia associated with fetal
54. Moss RL, Chen CM, Harrison MR. Prosthetic patch durability in nephrectomy: possible application for congenital diaphragmatic
t
congenital diaphragmatic hernia: a long-term follow-up study. J hernia. J Pediatr Surg. 1993;28(11):1433-1439, discussion 1439–1440.
ne
Pediatr Surg. 2001;36(1):152-154. 67. Hedrick MH, Estes JM, Sullivan KM, et al. Plug the lung until it
55. Scaife ER, Johnson DG, Meyers RL, Johnson SM, Matlak ME. The split grows (PLUG): a new method to treat congenital diaphragmatic
abdominal wall muscle flap—a simple, mesh-free approach to repair hernia in utero. J Pediatr Surg. 1994;29(5):612-617.
large diaphragmatic hernia. J Pediatr Surg. 2003;38(12):1748-1751. 68. Chiba T, Albanese CT, Farmer DL, et al. Balloon tracheal occlusion
k.
56. Silen ML, Canvasser DA, Kurkchubasche AG, Andrus CH, Naunheim for congenital diaphragmatic hernia: experimental studies. J Pediatr
KS. Video-assisted thoracic surgical repair of a foramen of Bochdalek Surg. 2000;35(11):1566-1570.
hernia. Ann Thorac Surg. 1995;60(2):448-450. 69. Harrison MR, Keller RL, Hawgood SB, et al. A randomized trial
oo
57. Bliss D, Matar M, Krishnaswami S. Should intraoperative hypercapnea of fetal endoscopic tracheal occlusion for severe fetal congenital
or hypercarbia raise concern in neonates undergoing thoracoscopic diaphragmatic hernia. N Engl J Med. 2003;349(20):1916-1924.
repair of diaphragmatic hernia of Bochdalek? J Laparoendosc Adv 70. Al-Maary J, Eastwood MP, Russo FM, Deprest JA, Keijzer R. Fetal
yb
Surg Tech A. 2009;19(suppl 1):S55-S58. tracheal occlusion for severe pulmonary hypoplasia in isolated
58. Davenport M, Rothenberg SS, Crabbe DC, Wulkan ML. The great congenital diaphragmatic hernia: a systematic review and meta-analysis
debate: open or thoracoscopic repair for oesophageal atresia or of survival. Ann Surg. 2016;264:929-933.
diaphragmatic hernia. J Pediatr Surg. 2015;50(2):240-246. 71. Dekoninck P, Gratacos E, Van Mieghem T, et al. Results of fetal
er
59. Arca MJ, Barnhart DC, Lelli JL Jr, et al. Early experience with endoscopic tracheal occlusion for congenital diaphragmatic hernia
minimally invasive repair of congenital diaphragmatic hernias: and the set up of the randomized controlled TOTAL trial. Early
results and lessons learned. J Pediatr Surg. 2003;38(11):1563-1568. Hum Dev. 2011;87(9):619-624.
rg
su
://
tp
ht
CHAPTER
Ventral Hernia and Abdominal Release Procedures
Heidi J. Miller
| Yuri W. Novitsky
52
V
entral herniation presents a set of common, yet hernia sac of peritoneum that contains visceral organs.
diverse and complex problems in the surgical Other bulges that may appear similarly, but are not true
world. It is a surgical disease with wide variation in hernias, are diastasis recti and eventration. Diastasis recti
management, variable outcomes, and high volumes. More is the thinning and broadening of the linea alba that leads
than 2 million laparotomies are completed in the United to a bulge at the midline and is usually asymptomatic. An
States every year, and it is estimated that up to 28% of eventration is a bulge resulting from lack of muscle tone
these will develop into ventral incisional hernias. Adding in the abdominal wall due to trauma, denervation, and
an additional 20% or more of primary congenital and surgical or congenital absence of the muscle. Neither
acquired hernias,1 this leads to an astounding incidence of diastasis recti nor eventration has a fascial defect and
t
ventral hernias in the United States alone. In 2006 more there is no hernia sac in either scenario.
ne
than 365,000 hernias were repaired in both inpatient and Ventral hernia can be further defined by location and
outpatient settings. The cumulative incidence of ventral origin. An incisional hernia occurs at any previous surgical
hernias is increasing each year by an estimated 3%, which site of the anterior abdominal wall. Traumatic hernias
k.
correlates with reported recurrence rates as high as 43%, occur due to injury of the fascia and musculature of
even after mesh repair.2 This leads to a calculation of the abdominal wall, and can be in any location. Lateral
oo
nearly 500,000 ventral hernia repairs in the United States abdominal wall hernias, also known as flank hernias,
during 2015. The enormity of the ventral hernia problem are often caused by blunt trauma with disruption of the
corresponds with a high cost of care, with an estimated attachments of the lateral muscles of the abdominal wall.
yb
$3.2 billion spent in 2006 for ventral hernia repairs alone. Subxiphoid hernias are located just inferior to the xiphoid
Within this spending there is great variability in cost per at the midline. Epigastric hernias can be spontaneous or
patient or hernia because complex ventral hernias can incisional at the midline between the xiphoid and the
lead to much higher costs, longer lengths of stay, and umbilicus. Umbilical hernias are located at the umbilicus
er
increased mortality rates in a small portion of patients and can be congenital or acquired spontaneously. Hypo-
compared with the majority.1 gastric hernias at the midline, inferior to the umbilicus are
rg
This disparity likely mirrors the complexity and rare spontaneous occurrences. Suprapubic and parailiac
variability that ventral herniation presents to surgeons, hernias occur adjacent to the bony structures of the pelvis.
characteristics that lead to a lack of consensus in the Finally, spigelian hernias are spontaneous hernias that
su
literature and among hernia surgeons on the ideal repair occur along the semilunar line, typically at its junction
technique. Traditional repair results have been fairly poor with the arcuate line of Douglas.
and the field of herniorrhaphy has finally been recognized The European Hernia Society (EHS) has introduced
://
by the surgical community as an important subspecialty.3 It a ventral and incisional hernia classification system in an
continues to be a rapidly changing field with innovations attempt to create a common language for the evaluation
tp
in mesh technology and surgical technique. Modern and treatment of ventral hernias.4 Primary and incisional
approaches include laparoscopy, open suture repair, hernias are separated in this classification system (Tables
mesh repair, component separation, and abdominal 52.1 and 52.2). A primary ventral hernia is divided by
ht
wall reconstruction. In the search for the ideal ventral location between midline hernias (epigastric and umbilical)
hernia repair, the surgeon must consider cost-savings; and lateral hernias (spigelian and lumbar). The size of
risk adjustment by patient comorbidities; complexity of primary ventral hernias is categorized into small (<2 cm),
the hernia, such as recurrences and nonhealing wounds; medium (2 to 4 cm), and large (>4 cm). Any weakness or
improved mesh incorporation; and decreasing the risk protrusion at the site of a surgical scar defines an incisional
of recurrence. Because there is no perfect repair for hernia for the EHS, and because these hernias are more
all ventral hernias, it is paramount for a surgeon to be variable, the classification system is slightly more involved.
familiar with a wide array of techniques and have a defined First, incisional hernias are divided into medial or lateral,
algorithm for evaluating and managing patients with ventral with the lateral edge of the rectus abdominis being the
hernias. dividing line. Midline hernias are placed into one of five
vertical zones (M1 to M5) ranging from subxiphoid to
suprapubic. Lateral hernias are divided into four zones
DEFINITIONS (L1 to L4) with subcostal, flank, and iliac stacked medial
Hernia is derived from Latin meaning “rupture” or to the anterior axillary line and lumbar (L4) hernias
“protruding viscous” and a ventral hernia is a protrusion arising anywhere dorsolateral to this line. Incisional hernias
of viscera, usually intestine, through the layers of the are divided into recurrent or not, and both length and
anterior abdominal wall. A true hernia has a defect in width are taken into consideration. Hernias with multiple
the fascia of the abdominal wall and the formation of a defects are measured at the point of greatest distance in
571
Ventral Hernia and Abdominal Release Procedures CHAPTER 52 571.e1
ABSTRACT
Ventral herniation presents a set of common, yet diverse
and complex problems in the surgical world. It is a surgi-
cal disease with wide variation in management, variable
outcomes, and high volumes. In 2006 more than 365,000
hernias were repaired in both inpatient and outpatient
settings and an expected 500,000 ventral hernia repairs
in the United States during 2015. The enormity of the
ventral hernia problem corresponds with a high cost of
care, with an estimated US$3.2 billion in 2006 for ventral
hernia repairs alone; there is great variability in cost per
patient or hernia because complex ventral hernias can
lead to much higher costs, longer lengths of stay, and
mortality rates in a small portion of patients.1 This disparity
likely mirrors the complexity and variability that ventral
herniation presents to surgeons, characteristics that lead
to a lack of consensus in the literature and among hernia
surgeons on the ideal repair technique. Traditional repair
t
ne
results have been fairly poor and the field of herniorrhaphy
has been recognized by the surgical community as an
important subspecialty.3 It is a rapidly changing field with
innovations in mesh technology and surgical technique.
k.
Modern approaches include laparoscopy, open suture
repair, mesh repair, component separation, and abdominal
oo
wall reconstruction. In the search for the ideal ventral
hernia repair, the surgeon must consider cost savings,
risk adjustment by patient comorbidities, complexity of
yb
the hernia, such as recurrences and nonhealing wounds,
improved mesh incorporation, and decreasing the risk
of recurrence. There is no perfect repair for all ventral
er
KEYWORDS
su
t
Epigastric M2 and protects the spine from hyperextension.
ne
Umbilical M3 Laterally, the abdominal wall is constructed of three
Infraumbilical M4 layered flat muscles. From superficial to deep these are
Suprapubic M5 the external oblique (EO), internal oblique (IO), and
k.
Lateral Subcostal L1 transversus abdominis muscles. The transversalis fascia
Flank L2 is the deepest layer of fascia in the abdominal wall and
oo
Iliac L3 separates the transversus abdominis muscle from the
Lumbar L4 peritoneum. The EO muscle runs inferior-medially,
Recurrent? Yes No originating at the lower costal margin and inserting at
yb
the linea alba, iliac crest, and pubic tubercle to form the
Length: cm Width: cm
inguinal ligament. The IO muscle runs perpendicular
Width W1 W2 W3 to the EO with its origination at the lateral half of the
cm <4 cm ≥4–10 cm ≥10 cm inguinal ligament, anterior iliac spine, and thoracolumbar
er
Modified from Muysoms FE, Miserez M, Berrevoet F, et al. Classification fascia and inserting into the lower ribs and linea alba.
of primary and incisional abdominal wall hernias. Hernia. 2009;13:407–414. Finally, the transversus abdominis muscle runs horizontally
rg
su
://
Anterior layer,
Rectus sheath External
tp
oblique muscle
External (cut away)
oblique muscle
ht
Internal
Posterior layer, oblique muscle
Rectus sheath
Inferior Rectus
epigastric abdominis
vessels muscle
Peritoneum
from the iliac crest, lateral inguinal ligament, and costal are present from birth and include complex entities such
cartilage to insert into the linea alba and joins the IO to as omphalocele and gastroschisis, or more straightforward
form the conjoint tendon. defects such as primary umbilical or epigastric hernias.
Each muscle body is surrounded by its relevant fascia More than 80% of primary congenital umbilical hernias
and these laminal layers join to form aponeurotic connec- will close spontaneously before the age of 5 and will not
tions for each flat muscle. Medially the three aponeuroses require repair. However, congenital epigastric hernias may
form the linea semilunaris, which lies at the lateral edge be symptomatic with incarcerated preperitoneal fat and
of the rectus abdominis muscle. The rectus abdominis require surgical intervention.5 Acquired ventral hernias
is a vertically oriented muscle originating at the pubic are of the spontaneous or incisional variety. Spontaneous
symphysis and inserting into the fifth to seventh costal hernias most often occur at weaknesses of the abdominal
cartilages. In the upper third of the abdomen, the EO wall, along the midline, or at the arcuate line or spigelian
aponeurosis and anterior lamina of the IO aponeurosis fascia. However, trauma to the abdominal wall may also
fuse to form the anterior rectus sheath, while the poste- lead to herniation in other locations. Incisional hernias
rior lamina of the IO overlies the transversus abdominis are defined as any ventral herniation that is located at a
muscle body. The transversus abdominis muscle extends previous surgical site or incision, including trocar sites.
medially in the upper abdomen, just deep to the rectus Spontaneous ventral hernias are diagnosed in adulthood
muscle, and the posterior lamina eventually joins the and are usually the effect of increased abdominal pressure
transversus abdominis aponeurosis to form the posterior related to obesity, pregnancy, ascites, or other factors.
t
rectus sheath. In the middle third of the abdomen, the Increased abdominal pressure leads to enlargement of the
ne
transversus abdominis muscle ends more laterally and hernia defect as well as increased likelihood of incarcera-
the posterior rectus sheath is formed by the aponeurosis tion. Fig. 52.2 shows the anatomic location of acquired
of the transversus abdominis and the posterior lamina of and congenital hernias.
k.
the IO aponeurosis. In the lower third of the abdomen, Epigastric hernias occur at the midline above the
below the arcuate line, the IO and transversus abdominis umbilicus where perforating neurovascular bundles travel
oo
aponeurosis fuse with the EO aponeurosis as part of the through the fascial layers that interlace to create the linea
anterior rectus fascia, leaving only peritoneum deep to alba. These defects are usually quite small but often have
the rectus abdominis. an incarcerated mushroom of preperitoneal fat that can
yb
Medial to the body of the rectus abdominis muscle, be quite symptomatic for the patient.
all of the flat muscle aponeuroses fuse to create the linea
alba and the midline of the anterior abdominal wall. The
linea alba is widest at the xiphoid, where the rectus muscles
er
Umbilical hernias occur through the base of the A meta-analysis of prophylactic mesh placement during
umbilicus or in the surrounding tissue. The umbilical laparotomy in high-risk patients also found a significant
ring is formed as the flat discus of fetal cells begins its decrease in incisional hernia development with prophylac-
three-dimensional fold and is surrounded by the amniotic tic mesh placement, longer operating room (OR) times,
cavity. It eventually fuses with the linea alba and rectus and shorter hospital stays.13 Therefore, prophylactic use
complexes, and the umbilical cord is created. Here the of mesh during closure of laparotomy in high-risk patients
linea alba is penetrated by the umbilical veins and arter- can be considered to reduce the risk of incisional hernia,
ies and after birth a cicatrix is formed as these vessels but no strong recommendations can be made for the
degenerate into the falciform and umbilical ligaments. It general population.
is thought that this is the weakest part of the abdominal Laparoscopic port site hernias are a growing problem
wall; either the cicatrix itself or the tissues surrounding within the incisional hernia category given the increasing
it are weak, allowing for vulnerability and the formation use of laparoscopy and robotics, and the exploration of
of umbilical or periumbilical hernias.5 Other patient single-site minimally invasive approaches. The incidence
factors may add to the stress of the abdominal wall and of port site hernias has been shown to be between 0% and
contribute to the formation of ventral hernias, such as 5.2%. Ninety-six percent of port site hernias occur at 10-mm
disorders of collagen formation, sleep apnea, steroid and larger port sites, and 86% occur at the umbilicus.14
use, and smoking or chronic lung diseases. Whatever the However, more recent studies have shown rates reaching
etiology, umbilical hernias are a common problem and 22% to 39%15,16 after ventral hernia repair with large mesh
t
approximately 200,000 repairs are done in the United and bariatric surgeries. It is generally believed that any port
ne
States every year. larger than 5 mm in diameter should be closed to prevent
Incisional hernias occur in up to 40% of patients after port site hernias. However, it remains controversial, and we
midline laparotomy. An innate problem with suture closure believe that dilating, noncutting trocars less than 12 mm
k.
of laparotomy is that tension is required to approximate do not require closure.17,18 Single-incision laparoscopic
the rectus abdominis muscles and counter tension of the and robotic surgeries have been found to have slightly
oo
lateral abdominal wall musculature. Such tension may higher incidence of port site hernias of 3%, although this
contribute to overtightening of sutures and ischemia to has improved over time and more recent studies have
the midline tissues.6 shown no difference.19 Port site hernias may be difficult to
yb
There are three types of incisional hernias: acute diagnose because they may occur early, late, intrafascially,
wound dehiscence and evisceration due to sutures tearing or as Richter-type hernias. Wound- and patient-related
through tissue, subacute with early gapping of the tissues factors may also play a role in the formation of port site
approximated under tension, and chronic remodeling of hernias: infection, delayed healing, steroids, collagen
er
the scar tissue causing “Swiss-cheese” or “cheese-cutting” disorders, obesity, and increased abdominal pressure are
hernias, formed as sutures cut through the weakened all known to predispose to herniation.
rg
with all midline layers en bloc.7 Two recent randomized complaints of a bulge in their abdominal wall. The bulge
controlled trials have found that small stitches, incorporat- may be more noticeable after physical activity, Valsalva
tp
ing about 5 mm of midline tissue and taken about 5 mm maneuvers, coughing, or any other activities that increase
apart reduces the risk of infection, wound dehiscence, and intraabdominal pressure. Some patients will present with
incisional hernia development.8,9 This is a change from complaints of pain or discomfort along with the bulge,
ht
historical surgical dogma where 1 cm was felt to be the although about 25% of patients will be asymptomatic.20
proper amount of tissue and distance between stitches. Pain associated with ventral hernias may be related to
Proper “small bite” technique should be confirmed at the incarceration and at times relieved with rest, lying down,
time of closure by measuring the suture-to-wound-length or reduction of the hernia contents. The most concerning
ratio to be at least 4 to 1.10 Although small bites may take presentations are incarcerated ventral hernias with signs
longer to complete than the traditional larger bites, the of bowel obstruction or strangulation, which is more
added operating room time accounts only for a small common with small defects. Other complaints that may
increased cost and the cost savings with the reduced accompany a ventral hernia are cosmetic, gastrointestinal
incidence of ventral hernia and complications is much or urinary symptoms, generalized pain, back pain, and
greater.11 There has also been some recent interest in respiratory difficulties.20
exploring the use of prophylactic mesh at closure of The mere presence of a ventral hernia is often con-
primary laparotomy for incisional hernia prevention. sidered a de facto indication for a surgical intervention.
A randomized controlled trial of mesh versus suture Other than incarceration and strangulation, generalized
closure after abdominal aortic aneurysm repair found a pain is the most common reasons surgeons choose to
fourfold reduction in hernia development with the use of intervene on ventral hernias.20 However, many ventral
prophylactic mesh. This study also showed a longer time hernia patients have complex medical histories and the
to hernia development for those with mesh placement, repairs are fraught with potential complications. Morbid-
and no increase in complications or mesh infections.12 ity rates after ventral hernia repair can reach 60% and
Ventral Hernia and Abdominal Release Procedures CHAPTER 52 575
mortality rates can be as high as 5.3%.21 Despite high weight loss goals. In a recent retrospective review at our
complication and morbidity rates, surgery is often felt institution of over 800 ventral hernia repairs, we found that
to be the primary treatment option for ventral hernias, a body mass index (BMI) greater than 45 increased the risk
largely because the natural history of ventral/incisional of postoperative wound infection.24 As a result, we have now
hernias is unstudied. A recent series of watchful waiting for set a BMI greater than 45 as an upper limit to qualify for an
ventral hernias showed no change in pain, functionality, elective open ventral hernia repair. Pulmonary and cardiac
or quality of life after 2 years of watchful waiting for problems must be addressed, and patients should be evalu-
minimally symptomatic ventral hernias larger than 9 cm.21 ated and treated for sleep apnea if they present with risk
The risk of incarceration or strangulation was less than factors. Diabetes control should be optimized with a goal
5% and another 8% of patients requested repair for for an HbA1c below 7.5. Any age-appropriate screening,
symptoms within the 2 years.21 This study is limited by a particularly a screening colonoscopy, is mandatory. Finally,
small sample size, but the low rates of complications and preoperative counseling is important to set the patient
progression of disease mirrors the results of a randomized up with appropriate expectations for the repair, results,
controlled trial for minimally symptomatic inguinal hernias postoperative care, and recovery. All possible outcomes
that has shown that watchful waiting can be safe.22 Also, should be discussed, including potential wound or mesh
although patients with symptomatic ventral hernias report infections and recurrence. It is important to understand
improvement in pain and quality of life, patients with the patient’s goals and their view of what outcomes may
limited symptomatology report similar pain levels before or may not be unacceptable.
t
and after repair.20,23 However, follow-up data have shown
ne
a high rate of disease progression and although watchful PRINCIPLES OF VENTRAL HERNIA REPAIR
waiting may be relatively safe, we believe most patients
should be optimized and undergo repair in the earlier Tissue-based repairs of ventral hernias have been developed
k.
stages of the disease. in an attempt to decrease recurrence rates; the Mayo repair
used a “vest-over-pants” technique and mattress sutures to
PATIENT EVALUATION
oo
imbricate fascial edges and approximate healthy tissue.
Initial intake begins with a thorough history and physical, However, long-term follow-up found similar recurrence
which should identify comorbidities and risk factors for rates to fascial suture approximation. In the 1990s, a
yb
complications. The physical exam should delineate previous randomized controlled trial showed a 50% decrease in
surgical scars, skin issues, the presence of stomas, fistulas, ventral hernia recurrence with the use of mesh compared
exposed mesh, or sinuses and hernia qualities such as to suture repair.25 The decreased rate of recurrence was
location and defect size. As an adjunct to the physical even more significant in smaller hernias, although with an
er
exam, computed tomography (CT) of the abdomen and associated increase in complications. Surgical site infections
pelvis can help with operative planning. We routinely and the presence of abdominal aortic aneurysms increased
rg
obtain a noncontrast CT for moderate and large ventral the risk of hernia development. Patients randomized to
defects. Oral contrast may be useful in the presence of mesh repair suffered more complications, but had lower
gastrointestinal or obstructive symptoms, and intravenous levels of postoperative pain.25
su
contrast may be helpful to identify soft tissue infections The introduction of prosthetics has been a game
or delineate vasculature. Finally, a thorough review of changer in the repair of ventral hernias, and today the
the patient’s operative history is necessary to identify any use of mesh reinforcement is recommended in the majority
://
potential difficulties with the planned repair and to discover of ventral hernia repairs. Prosthetics used in hernia repair
the type and location of any previous mesh placement. are in constant evolution and no “ideal” mesh has been
tp
Based on this evaluation, we then decide on the type of discovered. The qualities of the ideal mesh include good
repair most appropriate for the patient. We repair only tissue incorporation, limited foreign body reaction, and
very small (<2 cm) primary ventral hernias with a suture sufficient strength to withstand the forces of the abdominal
ht
repair. Appropriate patients for laparoscopic hernia repair wall along with good flexibility and compliance.26,27
include recurrent ventral hernias with small defects and Prosthetics are first divided by material type and can be
primary hernias that are larger than 2 cm but less than 10 synthetic, biologic, or biosynthetic. Synthetic meshes are
to 12 cm in width. We usually make an exception for women usually made of polypropylene, polyester, or polytetrafluo-
of childbearing age and avoid the use of laparoscopy roethylene (PTFE). Biologic meshes are either cadaveric
and/or permanent meshes. The defect size must also be allograft or xenograft tissue grafts that are processed to
considered in relation to the size of a patient, because reduce host reaction and improve tissue integration. Finally,
a 10-cm defect in a large male patient is quite different the newest category of biosynthetic meshes are made
than a 10-cm defect in a petite female patient. For larger of slowly absorbable biodegradable synthetic polymers.
or more complex hernias we use component separation Synthetic and biosynthetic meshes can be further divided
strategies for abdominal wall reconstruction. into monofilament or multifilament construction, micro- or
macroporous, and heavy, midweight, or lightweight types.
PREPARATION FOR SURGERY Finally, to reduce visceral adhesion and fistula forma-
To prepare the patient for surgery, we require smoking tion, there are covered meshes made for intraabdominal
cessation of at least 4 weeks and will test for nicotine placement. It is important for the surgeon to be educated
metabolites at preanesthesia testing. Weight loss counseling on the types of meshes available for hernia repair, the
should also be provided for obese patients, with referral benefits and limitations of each mesh type, and to have an
to bariatric surgery, if necessary, and agreement on set algorithm for choosing mesh in individual hernia repairs.
576 SECTION I Esophagus and Hernia
We prefer to use a macroporous, midweight monofilament developed a four-grade scale of ventral hernias: Grade 1
polypropylene mesh for most extraperitoneal repairs. includes healthy patients with no comorbidities or history
However, heavyweight polypropylene meshes are used of wound infections or contamination; grade 2 captures
in cases where more significant support of the repair is comorbid patients without infection or contamination;
needed. Our use of biologic meshes is limited to actively grade 3 includes higher-risk patients with potential for
infected fields and in cases of staged repair with a planned contamination with the presence of a stoma, history of
hernia. The role of slowly absorbable biosynthetics is wound infection, or opening of the intestinal tract; and
currently evolving. grade 4 captures patients with active infections, fistulas,
The most appropriate positioning of mesh within the or contamination.29 This grading scale was evaluated
abdominal wall is as big a question as which type of mesh and modified 2 years later by Kanters et al., who found
to use. Neither of these questions has been answered in that increased grade correlated with increased risk of
the literature, although 75% to 80% of hernia repairs recurrence as well as surgical site occurrence (SSO). 30
involve mesh to reduce the risk of recurrence. Mesh can However, they found significant difference between grades
be placed as an onlay, sublay, underlay, or interposition 2 and 3 with regard to SSO and proposed a modification
(Fig. 52.3). Onlay mesh is placed above the anterior of the grading scale to include only three grades. grade
rectus sheath, whereas sublay mesh is positioned within 1 remains the same low-risk group, grade 2 is mostly
the layers of the abdominal wall, typically retromuscular the same class of comorbid patients, with the added
or in the preperitoneal plane. An underlay mesh is placed inclusion of a history of wound infection, and grade 3 is
t
within the abdominal cavity underneath the peritoneum. the contaminated group.30 One of two major limitations
ne
Recurrence rates are lowest with sublay or retromuscular of the modified grading system by Kanters et al. is that it
mesh placement, followed by underlay mesh placement.28 segregated the groups based on often clinically irrelevant
Interposition mesh placement has the highest recurrence SSOs and not SSIs. Furthermore, there were not a sufficient
k.
rate, reported to be as high as 80%.28 Interposition mesh number of patients with dirty (CDC wound class 4) wounds
placement is fraught with complications and also holds the resulting in those patients being grouped together with
oo
highest SSI rate, whereas sublay mesh placement has the patients with contamination and active infection in the
lowest SSI rate and lowest rate of mesh excision.28 However, grade 3 category.
there is significant discrepancy in the literature regarding The VHWG also provided guidance in the repair of
yb
outcomes related to mesh location, which accentuates the ventral hernias with the first step including the evaluation
need for clinical decision making during ventral hernia of the hernia grade. The choice of repair technique
repair. Mesh location, like mesh type, should be decided and type of prosthetic used is then decided. For grade
based upon patient characteristics, anticipated technique, 3 and 4 hernias, the VHWG does not recommend the
er
and the characteristics of the hernia in question. use of synthetic mesh and states that biologics can be
To help with clinical decision making in the repair considered. However, Carbonell and colleagues have found
rg
of ventral hernias, as well as the ability to study and that synthetic mesh can be used in clean-contaminated
compare outcomes, the Ventral Hernia Working Group and contaminated cases with favorable outcomes.31 More
(VHWG) proposed a ventral hernia grading scale and recently Majumder and colleagues showed that the use of
su
recommendations for repair. 29 In 2010, the VHWG biologic mesh in clean-contaminated and contaminated
ventral hernia repairs was associated with a markedly
increased risk of surgical site events (SSEs), SSIs, and
://
continues to decline.
t
laparoscopic to open ventral hernia repair and found for ventral hernia.43,44 eTEP for ventral hernias includes
ne
no difference in recurrence rates and similar quality of extraperitoneal balloon dissection in the subcutaneous,
life after 6 months.33,37 The short-term improvements retromuscular, or preperitoneal planes, and allows for
after laparoscopy include less postoperative pain, quicker defect closure, component separation if needed, and
k.
rehabilitation and return to work, decreased wound infec- wide prosthetic reinforcement in a sublay position. Yet
tion rates, and better cosmetic outcomes. The frequently another approach is the endoscopic-assisted transhernia
oo
discussed downsides of laparoscopic ventral hernia repair mini-open sublay repair (MILOS) developed in Germany.
are increased risk of incidental enterotomy during lysis of MILOS achieves wide dissection using endoscopic or direct
adhesions, increased seroma and hematoma development, visualization with a lighted trocar through small incisions.
yb
and increased development of intraabdominal adhesions The dissection is taken through the hernia sac and into
due to an intraperitoneal prosthetic; potentially longer the extraperitoneal plane allowing for closure of the
operative times, and persistent bulging at the site of defect and wide mesh overlap.45,46 Furthermore, Belyansky
defects with bridged techniques.33,35,37 In addition to these and colleagues applied minimally invasive techniques for
er
potential complications, traditional LVHR technique laparoscopic abdominal wall reconstruction with posterior
leaves mesh intraperitoneally, in close proximity to bowel, component separation via TAR and extraperitoneal mesh
rg
thus requiring the use of a covered mesh, which is more reinforcement with very encouraging early results.47 More
expensive and more likely to harbor infection due to the recently, they used eTEP principles to perform retrorectus
barrier coating. The overall impact of mesh placed in or posterior component separation, largely avoiding
su
direct contact with bowel is unclear. intraperitoneal dissection. Although the refinement of
Even coated mesh confers increased risk for intra- the technique is ongoing, this approach, coined eTAR, may
abdominal adhesions, with some studies showing one-third evolve as one of the preferred, yet technically demanding,
://
of patients having significant adhesions at reoperation, the techniques for abdominal wall reconstructions.
potential for harboring infection, and increased cost.38,39 With the advent of robotics in many surgical specialties,
tp
Additionally, standard methods of fixation necessitate an robotic-assisted ventral hernia repair has recently gained
expensive laparoscopic tacking device, and both permanent interest because it may confer the benefits of a minimally
and absorbable tacks are correlated with increased post invasive approach while also allowing for a shorter learning
ht
operative and chronic pain.40 Finally, intraperitoneal mesh curve. Surgeons are able to use a technique similar to an
is often anchored with transfascial sutures to stretch the open approach (uncoated mesh in the extraperitoneal
mesh and limit shrinkage and hernia recurrence. Trans- space, midline restoration, avoidance of transfascial sutures
fascial sutures have been shown to cause ischemia of the and tacks) through minimally invasive access. Detractors
abdominal wall and are correlated with increased postop- of the robotic approach cite heightened cost, although
erative and prolonged pain and potentially incite recurrent no comparative cost data on this subject currently exist.
hernias.41 Moreover, it is important to point out that robotic repairs
Despite its myriad disadvantages, the LVHR technique allow the surgeon to avoid expensive, and at times painful,
has evolved very little since it initially gained accep- tack fixation, and significantly more expensive composite
tance and popularity in the early 1990s. More recently, meshes with antiadhesive coating, likely offsetting the
however, it has been suggested that defect closure prior costs of robotic instruments. To date, there has only been
to mesh placement could alleviate some of the shortfalls one retrospective study comparing laparoscopic ventral
of the traditional LVHR. Potential benefits of defect hernia repair to robotic ventral hernia repair, with both
closure include reduced wound morbidity by reducing approaches using an intraperitoneal mesh (IPOM).48
potential space for seroma and hematoma formation, This study showed longer operative times for robotic
lower rates of recurrence, improved abdominal wall cases and lower rates of complications and recurrences.
functionality by reapproximation of linea alba, and better The major technical differences were defect closure and
cosmesis.42 circumferential suturing of the mesh in robotic cases.48
578 SECTION I Esophagus and Hernia
Several authors have also reported on feasibility of robotic of the hernia are reduced using two atraumatic graspers,
ventral hernia repair and preperitoneal inguinal hernia and the hernia sac is typically left intact. We then measure
repair concomitant with robotic prostatectomy.49,50 the defect intracorporeally using transabdominal spinal
As surgeons become more comfortable with laparoscopy needles placed at each edge.
and mesh technologies evolve, there is potential for lapa- We routinely close fascial defects during laparoscopic
roscopic or robotic ventral hernia repair to grow into a hernia repairs. This is a widely debated topic among
place of superiority. As of now, it is one of many options for hernia surgeons and thus far the literature has shown
incisional hernia repair in the surgeon’s armamentarium a trend toward improved outcomes with primary fascial
and should be deployed under an appropriate algorithm. closure during laparoscopic hernia repair, citing lower
recurrence rates, lower rates of seroma formation, and
OPERATIVE TECHNIQUE—LAPAROSCOPIC VENTRAL improved patient satisfaction.51,52 However, the data are
HERNIA REPAIR sparse and a randomized controlled trial comparing
The patient is positioned supine on the operating table bridged laparoscopic repair to defect closure with synthetic
and the arms are tucked. Preoperative antibiotics and mesh is ongoing. We use a laparoscopic “shoelacing”
venothromboembolism (VTE) prophylaxis are given prior technique of figure-of-eight permanent sutures to close our
to incision. Orogastric tubes are placed, as we reserve defects.42 Briefly, a vertical line is drawn down the middle
nasogastric patients for those with an extensive adhesiolysis of the defect, 3-cm intervals are marked (beginning at the
or bowel congestion from incarceration. A urinary catheter upper edge of the defect), and stab wounds are created.
t
is placed in all patients; in those with infraumbilical exten- The figure-of-eight transfascial sutures are placed using
ne
sions of the hernia defect, we place a three-way catheter a suture passer with monofilament permanent sutures
to allow for instillation of saline to facilitate intraoperative (Fig. 52.4A). Each suture incorporates 1 to 2 cm of fascial
bladder identification. The abdomen is sterilized and edge and the sac is left in situ. After all sutures are placed,
k.
draped using an iodophor-impregnated drape (Ioban; 3M, the pneumoperitoneum is released and the sutures are
St. Paul, Minnesota) as an extra layer of protection against tied from outside in, leaving buried subcutaneous knots.
oo
mesh contamination. We typically enter the abdomen The abdomen is reinsufflated and the defect closure is
using an optical trocar in the subcostal region; however confirmed.
access should be individualized according to the surgeon’s The mesh is introduced through a 12-mm port placed
yb
comfort. Additional 5-mm ports are placed laterally on near the midline and/or near the closed defect, in a
the side of entry and another two 5-mm ports are placed location where it will be subsequently covered with mesh.
on the contralateral side to aid in the placement and We use a covered mesh in an underlay position and often
securing of the mesh. Adhesiolysis is carried out sharply, have the luxury of a mesh deployment device. If this is
er
minimizing the use of electrocautery or energy devices unavailable, four monofilament sutures should be placed
to prevent burn injuries to underlying bowel. Contents in four quadrants of the mesh prior to introduction into
rg
su
://
tp
ht
A B
FIGURE 52.4 Laparoscopic ventral hernia repair (LVHR) technique with shoelacing technique defect closure. (A) Interrupted figure-of-eight
sutures passed transfacially with a suture passer along the length of the defect. (B) LVHR repair with intraperitoneal mesh after defect
closure showing circumferential tacks, lateral fixation sutures, and transfascial buttressing sutures just lateral to closure. (Based on
Orenstein SB, Novitsky YW. Laparoscopic ventral hernia repair with defect closure. In: Novitsky YW, ed. Hernia Surgery: Current
Principles. New York: Springer; 2016:235.)
Ventral Hernia and Abdominal Release Procedures CHAPTER 52 579
the abdominal cavity. The mesh is unrolled and transfascial the cicatricial tissue closing the umbilical ring. In children
sutures are pulled through, keeping the mesh on stretch, younger than 2 years old, most umbilical hernias close
with a suture passer. We recommend starting fixation by spontaneously; however, in adults these hernias tend to
pulling the upper or lower stitch first, followed by the enlarge with time.
lateral ones. The edge of the mesh is secured with metal or Repair of an umbilical hernia, as described by William
absorbable tacks at 1-cm intervals. We then place transfacial Mayo, using a vertical fascial overlap technique was dis-
stitches on either side of the closed defect (within 2 cm cussed earlier. This operation (or simple fascial closure) is
of the midline) to fix the mesh, take tension off our still performed frequently today by many surgeons. These
“primary” defect closure, and redistribute the tension on repairs are effective and may be the preferred technique
the mesh (see Fig. 52.4B). The ports are removed under for small umbilical hernias with no tension after fascial
direct vision, pneumoperitoneum is released, and the approximation, but larger hernias have been shown to
incisions are closed with subcutaneous sutures. have a recurrence rate of up to 28%.55
The most catastrophic potential complication of lapa- The introduction of mesh prosthetics has appropri-
roscopic ventral hernia repair is small bowel injury during ately had an impact on umbilical hernia repair. These
adhesiolysis, especially if they are missed.53 Enterotomy has tension-free repairs, which have been popularized for
been reported in an average of 1.7% to 3.3% of patients other ventral hernias, may have a role in umbilical hernia
in recent series of laparoscopic ventral hernia repairs.54 repair. The largest randomized controlled trial looking at
If an enterotomy occurs, the mortality rate is reported to mesh in umbilical hernia repair was completed in 2001 by
t
be 1.7% if it is recognized and repaired. However, if the Arroyo et al. This randomized controlled trial compared
ne
enterotomy is missed, the mortality rate increases to 7.7%.54 primary suture repair and mesh repair in 200 patients
Management of a recognized intraoperative enterotomy with umbilical hernias.56 The two patient groups were
varies according to the type and extent of the injured comparable with regard to age, sex, hernia defect size,
k.
intestine and the type of mesh available. Small lacera- and American Society of Anesthesiologists (ASA) class.
tions in the small intestine or bladder without significant Operative times and complications were not statistically
oo
contamination may not be an absolute contraindication to different. The mean follow-up was 64 months. The major
mesh placement either laparoscopically or by open means. difference was the recurrence rate of 11% in the suture
In the event of fecal spillage, the bowel should be repaired repair group versus 1% in the mesh repair group (P =
yb
and the adhesiolysis completed. A delayed hernia repair .0015). A review of all randomized controlled trials and
is generally warranted if a prosthetic is required. At times, observational studies found no difference in complication
the patient may be placed on antibiotics and returned to rates associated with mesh use, and an odds ratio of 0.09
the operating room in 3 or 4 days for definitive repair. The in favor of mesh in randomized controlled trials and
er
safer option, however, is to perform a primary repair of 0.40 in observational studies support the use of mesh in
the hernia defect or repair with a biologic mesh, but the reducing umbilical hernia recurrence.57
rg
long-term durability of these repairs is poor. We believe The ideal technique for placement of a prosthetic during
that placement of an intraperitoneal synthetic mesh in the umbilical hernia repair remains debatable. Laparoscopic
presence of significant contamination is contraindicated. techniques have recently been proposed for umbilical
su
Another alternative that should be strongly considered is hernias as well. The technical aspects are essentially the
a conversion to laparotomy. This would include careful same as those applied to other ventral hernia defects. The
inspection of the entire bowel for other unrecognized laparoscopic approach takes longer to perform, tends to
://
injuries, and repair or resection of the involved segment have fewer complications, and has no recurrences reported
followed by primary closure or extraperitoneal repair in a small retrospective series. Criticism of the laparoscopic
tp
SPECIAL CONSIDERATIONS around but not through the umbilicus has the potential
to avoid the wound-related complications associated with
UMBILICAL HERNIAS an incision directly over the mesh.
Umbilical hernias are relatively common in the adult There are many effective methods to repair umbilical
population and are another example of a spontaneous hernias. Each patient must be evaluated individually, and
ventral hernia. More than 166,000 umbilical hernia repairs one method of repair may not apply to all cases. Small
are performed annually in the United States, making it primary umbilical defects in low-risk patients can probably
the second most prevalent abdominal wall hernia after be repaired with sutures alone and achieve acceptable
inguinal hernia.1 These umbilical hernias can be the results. As the defect size increases, particularly in obese
result of a recurrence or persistent congenital umbilical patients or manual laborers, a mesh prosthetic should be
hernia. In 90% of patients, it is an acquired defect that is a considered. Whether the repair is better performed via an
direct result of chronically increased abdominal pressure. open or laparoscopic approach is controversial because
Numerous factors have been linked to increased abdominal prospective data are not available. Improvements in mesh
pressure including multiparous status, obesity, and cirrhosis prosthetics may continue to guide the ideal approach.
with ascites.4 Umbilical hernias tend to be more common
in females and often develop in the fourth to fifth decade SPIGELIAN HERNIAS
of life. The fascial ring that constitutes the neck of the Adriaan van der Spiegel, a Belgian anatomist, was the
hernia can be dense and is formed by gradual yielding of first to describe the semilunar line as a concave region
580 SECTION I Esophagus and Hernia
at the lateral border of the rectus muscle formed by the patient factors may also play a role. Radical prostatectomy
aponeurosis of the IO. More than 100 years later, in 1764, is the most common operative procedure that leads to the
Klinkosh identified the “hernia of the spigelian line” as development of a suprapubic defect. Similar defects are
a distinct entity. also seen after operations involving the uterus, urinary
Although spigelian hernias are rare (accounting for bladder, and sigmoid colon.
0.1% to 2% of all abdominal wall hernias), its diagnostic Suprapubic hernias may manifest as vague lower
incidence has been rising because of improved imaging abdominal discomfort, urinary symptoms, or a palpable
technology and incidental identification during laparos- mass. The diagnosis of a suprapubic hernia may be missed
copy. Spigelian hernias usually occur in the sixth and because of the similarity of features with more common
seventh decades and affect both sexes and sides equally. inguinal hernias. However, a thorough physical examina-
Most are acquired, and nearly 50% of patients with tion will demonstrate close proximity of the mass, defect,
spigelian hernias have a history of previous laparotomy or both to the pubis and not the external inguinal ring.
or laparoscopy.58 Other factors that have been implicated Although suprapubic hernias may be a source of significant
in contributing to the development of these hernias are abdominal pain, bowel incarceration requiring emergency
alterations in compliance of the abdominal wall as a repair is extremely rare.
result of morbid obesity, multiple pregnancies, prostatic Primary repair of traumatic suprapubic hernias may
enlargement, chronic pulmonary disease, and rapid weight be a viable alternative if the herniorrhaphy is undertaken
loss in obese patients.58 without delay. With time, the rectus muscle retracts and can
t
A spigelian hernia is a challenge to diagnose and lead to significant tension if primary repair is performed.
ne
requires a high index of suspicion. Pain is the most Thus, a mesh repair is preferred for most traumatic and
common initial complaint. The fascial defect is masked incisional suprapubic hernia repairs. Several approaches
by the intact overlying EO aponeurosis, thus complicating to mesh placement for suprapubic hernias have been
k.
physical examination.59 In addition, a palpable mass (when described. The open preperitoneal approach provides
present) may mimic an abdominal wall lipoma or desmoid excellent delineation of the bladder and pubis and allows
oo
tumor. Although abdominal imaging may be helpful, the for appropriate inferior fixation of the mesh in contrast
findings of unusual abdominal complaints in the proper to an onlay style of repair. The laparoscopic approach to
anatomic location should alert one to the possibility of a suprapubic herniorrhaphy can also allow for a solid repair,
yb
spigelian hernia. More than half of all spigelian hernias provided that mobilization of the bladder is performed.
are diagnosed intraoperatively.58 This can be facilitated by using a three-way urinary
Given the small neck of these hernias, 20% to 30% catheter. The bladder is instilled with 300 mL of saline
require emergency intervention.58–60 Thus, even incidental and can be clearly visualized for adequate mobilization
er
spigelian hernias should be repaired electively to avoid to expose the entire pubis, Cooper ligament, and the
incarceration. Surgical management of these hernias iliac vessels. This is imperative to prevent inadequate
rg
has typically been accomplished via a transverse incision overlap of the mesh and early recurrence. Regardless of
and primary repair. Primary repairs have been associated the approach (open or laparoscopic), the dissection can
with a low, but real recurrence rate of about 4%.59 As be challenging because of the close proximity of these
su
expected, mesh repairs have been successfully applied to hernias to bony, vascular, and nerve structures, and to the
treat spigelian hernias. Few or no recurrences at long-term bladder.
follow-up have been reported by investigators.59 More
://
although long-term recurrence outcomes are awaited.60 Many complex and often multiply recurrent incisional
Evidence-based surgical recommendations are limited hernias are not amenable to basic repairs or traditional
by the rarity of this condition, and a recommendation laparoscopic techniques. Those patients frequently need
ht
regarding suture- or mesh-based repair (either open or more advanced reconstructive procedures to address their
laparoscopic) is not clear at present for the treatment of defects in order to provide a durable and functional repair.
spigelian hernias. These patients require abdominal release or component
separation procedures to provide for excessive-tension-free
SUPRAPUBIC HERNIAS closure and reduce the risk of recurrence. Component
The abdominal oblique aponeurosis, rectus abdominis separations include a variety of techniques wherein the
musculature, and rectus sheath insert on the symphysis layers of the abdominal wall are strategically divided and
pubis. Suprapubic hernias result from disruption of these separated for the purpose of medialization of the rectus
musculotendinous elements of the lower abdominal wall muscles and restoration of the linea alba. In other words,
and usually occur after blunt abdominal trauma or pelvic the redundancy of layers of anterior and lateral abdominal
surgery. The origin of traumatic suprapubic hernias is often walls allows for sacrifice of one or several of its components
through a ruptured rectus muscle at or near its insertion to provide for myofasciocutaneous medialization aimed
to the pubic bone. In contrast, incisional suprapubic at restoring near-normal anatomy and physiology to the
hernias develop as a result of apical pubic osteotomy or entire abdominal wall.
iatrogenic detachment of the rectus muscle from its pubic
insertion to improve visualization during pelvic surgery. ANTERIOR COMPONENT SEPARATION
Inadequate tissue purchase inferiorly during closure may In 1990, Ramirez refined the technique of EO release that
result in hernia formation, although infection and other had been described in the early 20th century.61 About a
Ventral Hernia and Abdominal Release Procedures CHAPTER 52 581
decade after its initial introduction, anterior component fascia using electrocautery. These flaps are raised from the
separation (ACS) became the most commonly employed costal margin to the inguinal ligament and laterally to the
reconstructive technique. Given the wide dissection midclavicular or anterior axillary line. The EO aponeurosis
required to reach the EO for division and separation, it can be clearly identified by the directionality of its fibers
also allows for wide mesh overlap. However, ACS is also and is incised 1 to 2 cm lateral to the semilunar line in
fraught with a high potential for wound complications a vertical fashion from the costal margin to just above
due to creation (and devascularization) of large skin the inguinal ligament. It is crucial to properly identify
flaps raised during dissection. In fact, wound morbidity and protect the semilunar line. If this incision is made
of ACS has been reported to be 26% to 63%.62 However, lateral enough to expose the muscle fibers of the EO,
refinements that are described later, such as endoscopic then the muscle fibers are divided using electrocautery
ACS and perforator-sparing ACS, have successfully reduced and right angle dissector to avoid injury to the underlying
the wound complication rates to 2% to 26%.62 IO aponeurosis (Fig. 52.5A). The muscle fibers should
be divided along the length of the incision and can be
Operative Technique extended above the costal margin to aid in the closure of
The patient is positioned supine on the operating table subxiphoid or epigastric defects. Once the EO complex
with arms out, and an orogastric tube and urinary catheter is divided, the EO muscle should be “separated” laterally
are placed. The abdomen is prepared with hair removal from the underlying IO aponeurosis using blunt dissection
and sterilized widely. A laparotomy incision is made, and along the avascular plane. Complete separation of the
t
as needed can include teardrop or elliptical incisions to EO is critical to remove its contribution to the tension
ne
excise attenuated skin and scar tissue. A transverse incision of the lateral abdominal wall, thus facilitating subsequent
can also be used and is often deployed when simultane- medialization of the rectus abdominis complex. If this
ous panniculectomy or abdominoplasty is planned. Safe release does not allow for sufficient medial advancement,
k.
surgical access to the abdomen is imperative to avoid a posterior rectus sheath release can also be performed.
bowel injury and contaminating the field. In some cases, This release is accomplished by incising the posterior
oo
it is possible to remain extraperitoneal to avoid a hostile rectus sheath along the muscle body of the rectus to
abdomen and the added time and morbidity of a long allow the rectus to be freed from its fascial encasement
adhesiolysis. Once the abdomen is entered, a safe and and facilitate its medialization. Both maneuvers were
described by Ramirez et al.61
yb
complete adhesiolysis between the viscera and abdominal
wall should be performed. Once adequate release has been performed, mesh
Next, the anterior fascia of the rectus abdominis muscle placement can be completed. The choice may depend
is identified and subcutaneous flaps are created. The on wound class, comorbidities, and other factors. Mesh
er
skin and soft tissue are released from the anterior rectus positioning is also a choice the surgeon must make at
rg
su
://
tp
ht
FIGURE 52.5 Anterior component separation (ACS) technique. (A) Skin flaps are raised laterally from midline, and bilateral external oblique
fascia and muscle are cut to expose internal oblique muscle fibers to allow for midline closure of healthy anterior rectus sheath to
re-create the linea alba. (B) Wide onlay mesh overlap after ACS with mesh fixed to the cut edges of the EO and re-created linea alba.
(Based on Silverman R. Open component separation. In: Rosen M, ed. Atlas of Abdominal Wall Reconstruction. Philadelphia: Elsevier;
2012:131–138.)
582 SECTION I Esophagus and Hernia
this point. The mesh can be placed as an intraperitoneal closure of the soft tissue with absorbable sutures and skin
underlay, as a sublay in the posterior rectus space, or as staples.
an onlay above the anterior rectus fascia (see Fig. 52.5B).
Intraperitoneal mesh must be secured with circumferential PERIUMBILICAL PERFORATOR-SPARING ANTERIOR
transfascial sutures along the edge of the mesh, closing the COMPONENT SEPARATION
space to abdominal contents. It must also be secured with The skin of the central abdominal wall gets its blood
some tension across the midline to allow for approximation supply from the deep inferior and superior epigastric
of the fascia ventral to the taut mesh. To accomplish vessels. The deep epigastric vessels divide into a vast
that, we measure half the width of the mesh from the network of musculocutaneous perforating branches and
midline closure to place our lateral sutures. Onlay mesh are concentrated in the periumbilical region to provide
is placed after re-creation of the linea alba and is secured the majority of vascularity to the central abdominal wall.
to the cut edges of the EO aponeurosis. Fibrin glue has Therefore, to combat the significant wound morbidity
also been used as an adjunct in mesh fixation for onlay caused by the creation of the necessary large undermined
meshes.63 Large-bore closed-suction drains should be left subcutaneous flaps, Dumanian was first to develop a
ventral to the mesh, regardless of its position within the perforator-sparing component separation technique.
abdominal wall. The goal is to maintain pulsatile blood flow to the reap-
Midline fascial closure should be completed with absorb- proximated tissue of the hernia repair and to release lateral
able monofilament suture either in a running fashion tension at the midline.64,65 Wound complications have been
t
or using interrupted figure-of-eight sutures to achieve found to be reduced by 50% to 90% with techniques that
ne
reconstruction of the linea alba and approximation of the allow for preservation of the periumbilical perforators of
rectus abdominis muscles. Soft tissue closure after ACS is the abdominal wall.64,65
of great importance because subcutaneous flaps can be a Perforator-sparing ACS can be completed via lateral
k.
source of major wound and mesh morbidity. Dead space incisions or via tunneling from the midline at the inferior
between subcutaneous tissue and fascia or mesh must be and superior portions of the laparotomy incision (Fig.
oo
dealt with to avoid seroma and hematoma formation. This 52.6). If a lateral incision is used, a 6-cm incision should be
can be achieved with closed-suction drains or suturing made transversely at the costal margin and subcutaneous
the soft tissue back down to the fascia with progressive tissues divided until the EO fascia is identified. The EO
yb
tension suture technique. Any old scars, as well as ischemic fascia is incised and the EO muscle is bluntly separated
or devascularized, attenuated and redundant skin and from the underlying IO fascia. A tunnel in this plane is
soft tissue must be excised. We then suggest a layered developed bluntly with finger dissection and the aid of a
er
Vertical
rg
subcutaneous
tunnels
su
://
tp
Release of
ht
external
oblique
aponeurosis
Horizontal
subcutaneous
access tunnels
FIGURE 52.6 Perforator-sparing anterior component separation with lateral subcutaneous tunnel dissection. (Based on Dumanian GA.
Open anterior component separation with perforator preservation. In: Novitsky YW, ed. Hernia Surgery: Current Principles. New York:
Springer; 2016:150–158.)
Ventral Hernia and Abdominal Release Procedures CHAPTER 52 583
t
ENDOSCOPIC SEPARATION OF COMPONENTS significant overlap placed under the muscular abdominal
ne
In an effort to minimalize the morbidity of the ACS, a wall can be explained by Pascal’s principles of hydrostatics.
minimally invasive, endoscopic component separation Because the intraabdominal cavity functions as a cylinder,
technique (ECST) was described by Lowe et al. in 2000.66 the pressure is distributed uniformly to all aspects of the
k.
This was created to address the extensive lateral dissection system. Consequently, the same forces that are attempting
required for EO release, thus creating large subcutaneous to push the mesh through open hernia defects are also
oo
flaps and requiring ligation of perforating vessels and holding the mesh in place against the intact abdominal
leading to wound infection rates of 25% to 57%. ECST wall. In this manner, the prosthetic is held in place by
provides the benefit of medialization of the abdominal intraabdominal pressure.
yb
wall allowing for tension-free closure during hernia repair,
while limiting subcutaneous dissection and preserving deep Operative Technique
periumbilical perforators.67 It also allows for dissection and A full laparotomy incision is made and adhesiolysis is
release within a clean plane, avoiding ostomy sites and carried out as needed. The posterior rectus release begins
er
infected prosthetics or midline processes.68 ECST must be with the incision of the posterior rectus sheath at its most
accompanied by midline closure and mesh reinforcement; medial border, confirming the location of the rectus
rg
options include open laparotomy at the midline with onlay muscle ventral to the initial incision. This incision is taken
mesh placement after raising subcutaneous flaps, open along the full length of the rectus muscle, cranially to
laparotomy with retrorectus dissection and sublay mesh the xiphoid and caudally to the arcuate line and into the
su
placement, or laparoscopic ventral hernia repair with defect space of Retzius. The posterior sheath is dissected free
closure and intraperitoneal underlay mesh placement. from the rectus abdominus muscle medially to laterally
The patient is placed in the supine position with arms using blunt dissection and electrocautery. This is a mostly
://
tucked. A 2- to 4-cm transverse incision is made near the avascular plane until the lateral edge of the rectus muscle
costal margin at the tip of the 11th rib. The EO muscle is reached and deep perforating vessels are encountered.
tp
fibers are identified and separated from the underlying IO As the lateral edge is reached, the epigastric vessels can
fascia. A round balloon dissector is introduced between be visualized and should remain with the muscle body.
the EO muscle and IO fascia and pushed blindly toward The dissection is complete when the semilunar line has
ht
the inguinal ligament medial to the anterior superior iliac been reached. It is important to identify neurovascular
spine. The balloon dissector is then inflated sequentially bundles as they perforate the posterior rectus sheath just
from distal to proximal, and then aimed toward the costal medial to the linea semilunaris. Superiorly this dissection
margin and inflated in the superior space. Three ports leads into the subxiphoid space and the insertions of
are placed and the EO insertion is divided sharply and posterior sheath at the xiphoid can be incised to allow
with cautery along with the subcutaneous tissue and fascia access into this plane across the midline. Inferiorly, the
ventral to the dissection plane. Ventral hernia repair is then dissection leads into the space of Retzius; blunt dissection
completed in the fashion chosen by the operating surgeon. is performed to skeletonize the pubic symphysis and
Advantages of ECST over open ACS include lower rates bilateral Cooper ligaments.
of wound necrosis, dehiscence, and infections, with wound After bilateral retrorectus dissections are performed,
complication rates reduced by almost 50%.69 However, posterior sheaths are reapproximated using running
open ACS has shown a lower recurrence rate and lower absorbable sutures. A mesh of choice is measured to
rate of intraabdominal abscess formation.67 Limitations fit the retrorectus space and is fixed with transfascial
of ECST include the possibility of inadvertent division sutures at the lateral edges, ensuring sufficient overlap
of the linea semilunaris and creation of postoperative of the repaired hernia defect and physiologic tension
flank hernias with limited options for reinforcement of of the mesh. Large-bore closed-suction drains may be
the area of release, limited options for mesh placement, placed ventral to the mesh if preferred. Finally, the linea
and limited utility for nonmidline defects. alba is re-created ventral to the mesh by approximating
584 SECTION I Esophagus and Hernia
anterior rectus fascia at the midline using slowly absorbable must make a transition into the preperitoneal space and
monofilament suture. Skin and soft tissue are closed in divide the arcuate line laterally at its junction with the
a layered fashion. semilunar line.
Lateral and retroperitoneal dissections are then under-
TRANSVERSUS ABDOMINIS RELEASE taken with the aid of a Kittner dissector using proper
We first performed the transversus abdominis release traction and counter-traction to separate the TAM from the
(TAR) procedure in 2006 with initial presentation of our underlying transversalis fascia. Once this plane is created,
early experience at the 2009 World Hernia Conference. it can be continued laterally using blunt dissection until
Although it was originally met with skepticism, after the retroperitoneum is reached and the psoas muscle is
subsequent publication of the technical description and identified, if necessary. Dissection following the lateral
outcomes in 2012, the TAR is rapidly becoming one of edge of the psoas muscle caudally allows for identification
the fastest growing approaches to major abdominal wall of the entire myopectineal orifice and allows access to the
reconstructions. The TAR procedure, a type of posterior space of Retzius and Cooper ligaments. In women, round
component separation, is essentially an extension of ligaments are divided. In men, spermatic cord should be
the posterior rectus release described earlier. The TAR identified and protected.
approach allows for significant medial advancement of The superior dissection is now undertaken, and depend-
the posterior rectus sheath, creates a large retromuscular ing on the location of the hernia, the cranial extent of the
pocket of vascularized tissue for placement of large meshes, dissection may be in the epigastrium or the subxiphoid
t
and the reinforcement of the entire visceral sac. Moreover, space. In the epigastrium, the posterior sheath must be
ne
it allows for the medialization of the rectus muscles and disconnected from its insertion into the linea alba bilater-
effective reconstruction of the linea alba.74 The TAR ally to allow for sufficient mesh overlap cranially from the
technique has been shown to be durable and reliable in a hernia defect. This is done by incising the insertion of the
k.
number of series with recurrence rates between 3.7% and posterior sheath just lateral to the linea alba bilaterally,
5%.75,76 It has also been found to be safe and effective for allowing intact linea alba to lift away and disconnecting
oo
repairing recurrent ventral incisional hernias following either side of the posterior sheath. This should be done
ACS. Finally, TAR allows access to the space above the for at least 5 cm to allow for sufficient mesh overlap and
xiphoid and costal margins, aiding in repair of complex prevent superior recurrences. For defects that already reach
yb
subcostal and subxiphoid hernias. into the epigastrium, the subxiphoid and substernal spaces
are entered to allow for sufficient dissection and mesh
Operative Technique placement. To enter these spaces, the posterior sheath
The patient is positioned supine on the operating table insertions to the xiphoid are divided. The most cranial
er
with both arms out. A urinary catheter and orogastric transversus abdominis muscle fibers must be divided and
tube are placed preoperatively, and the patient is prepped the pretransversalis plane will lead into the retrosternal
rg
from nipples to thighs and laterally to the edge of the bed. space. One must be careful not to divide the fibers of the
The abdomen is draped using an iodoform-impregnated anterior diaphragm because they interdigitate with the
drape (Ioban; 3M, St. Paul, Minnesota). First, a generous transversus abdominis at this level. Once this dissection
su
midline incision is made and can be completed in linear, into the pretransversalis plane is completed bilaterally,
elliptical, or teardrop fashion for excision of old scar the two planes can be connected by dissecting above the
and excess soft tissue. Although some surgeons have subxiphoid fat pad. This dissection may be continued
://
advocated a completely extraperitoneal dissection,77 we into the retrosternal space by separating transversalis
prefer a complete adhesiolysis to remove all visceral adhe- fascia from the diaphragm until the central tendon of
tp
sions to the anterior abdominal wall. This reduces the the diaphragm is exposed.
risk of injury to underlying organs and allows for lateral The next step is to reconnect medialized posterior
retromuscular dissection and release. We reserve interloop layers and to re-create a continuous visceral sac with
ht
adhesiolysis for patients who had obstructive symptoms running absorbable sutures. Any small holes in the
in the preoperative period. posterior layer may be closed with figure-of-eight or
We then proceed with the posterior rectus sheath interrupted absorbable sutures, and any larger gaps in
release and retrorectus dissection as described earlier, the posterior layer can be filled with an interposition of
continuing this dissection laterally to the linea semilunaris polyglactin (Vicryl Mesh; Ethicon, Somerville, New Jersey)
and carefully identifying and preserving perforating or biologic mesh. Once the posterior sheath is closed, the
neurovascular bundles (Fig. 52.7). Exposure and division abdominal contents are now excluded from the operative
of the transversus abdominis muscle (TAM) is begun by space. Generous irrigation of the retromuscular space is
the incision of the posterior sheath 1 cm medial to the completed. We use antibiotic-laden irrigation (cephazolin
perforating vessels, as cephalad as possible. This incision 3 g, gentamicin 240 mg, Bacitracin 50,000 units per 3 L
exposes the muscular aspect of the TAM in the superior of normal saline) with a power irrigator in contaminated
portion of the abdomen and the aponeurotic aspect in or clean-contaminated cases.
the inferior portion. The TAM fibers are then divided with A mesh is then measured to fill the entire retromuscular
electrocautery, using a right-angled dissector to isolate and space. We generally favor a midweight macroporous poly-
lift the fibers from the underlying transversalis fascia and/ propylene mesh in most of our cases, although a heavier
or peritoneum. This division is completed cranial-caudally weight polypropylene may be used for unusual hernias
until there is complete transection of the TAM and its such as flank hernias or in cases where the linea alba is
aponeurosis. At the level of the arcuate line, the surgeon either not restored or restored under significant tension.
Ventral Hernia and Abdominal Release Procedures CHAPTER 52 585
A B
t
ne
k.
oo
yb
er
rg
C D
su
FIGURE 52.7 Transversus abdominis release (TAR) technique. (A) Incision of the transversus abdominis muscle after posterior rectus
sheath release and incision. (B) Medialization of posterior sheath with TAR and lateral dissection. (C) Closure of the posterior sheath after
bilateral TARs and exclusion of abdominal contents. (D) Reinforcement of visceral sac with large retromuscular mesh sublay and
://
re-creation of the linea alba. (Based on Novitsky YW. Posterior component separation via transversus abdominis muscle release: the TAR
procedure. In: Novitsky YW. Hernia Surgery: Current Principles. New York: Springer; 2016:117–136.)
tp
Fixation of the mesh is done with long-lasting absorbable dissection if the anatomy is unfamiliar. Similarly, there
sutures to bilateral Cooper ligaments and transfascial is a possibility of injury and destabilization of the lateral
ht
sutures at the xiphoid. Lateral transfascial fixation can be abdominal wall and resultant bulging/hernias, if the linea
done with a straight suture passer or a Reverdin needle, semilunaris or the IOs are injured during the ACS. Fur-
although recently we have found this to be superfluous, thermore, there is a significant risk for soft tissue necrosis
because the large mesh has limited mobility within the after raising soft tissue flaps to allow for EO release, which
retromuscular space. Large closed-suction drains are placed can be quite morbid, with possible exposure and infection
ventral to the mesh and brought out laterally through of the underlying mesh. Component separations should
the abdominal wall. Finally, the anterior fascia is closed be used for select patients only and indications for the
with running or interrupted figure-of-eight sutures, and procedure should be part of an algorithm for treatment
the skin and soft tissue are closed in layers after excision of complex abdominal wall hernias.
of any excess tissue.
Clean field
Synthetic mesh
<2 cm <6 cm
t
ne
>6 cm >15 cm Underlay
k.
Treat as Open TAR Mesh excision
incisional open TAR
oo
FIGURE 52.8 Hernia repair algorithm for clean fields. LVHR, Laparoscopic ventral hernia repair; MIS, minimally invasive surgery; RS,
Rives-Stoppa; TAR, transversus abdominis release.
yb
than 6 cm will receive an LVHR with defect closure and
Contaminated field wide mesh overlap. We deploy the laparoscopic shoelacing
er
There is limited use of minimally invasive hernia repair in alvimopan, and a scheduled early diet advancement.
contaminated fields; however we may consider a minimally We have found our outcomes greatly improved with the
invasive surgery (MIS) retromuscular repair for clean- use of the pathway; as expected, patients have a quicker
contaminated cases with concomitant procedures. We time to bowel function and advancement to regular diet.
generally approach contaminated fields with an open TAR Furthermore, we have seen a shorter hospital length of
and most often will use synthetic mesh. Biosynthetic or stay, decreased to an average of 4 days from 6.1, and a
biologic meshes are used for patients with latent or active lower 90-day readmission rate.78 We also have integrated
methicillin-resistant Staphylococcus aureus (MRSA) infections. a nurse practitioner into the care of our abdominal wall
Parastomal hernia repairs are described elsewhere in this reconstruction patients, allowing for better preoperative
book, but are included here within our algorithm. For education, easier access to clinic appointments, and earlier
small parastomal hernias, we use a laparoscopic mesh intervention for potential problems.
repair and prefer the Sugarbaker method. For large or
recurrent parastomal hernias, or those with ventral hernia
components, we use synthetic mesh in the retromuscular
CONCLUSION
space using TAR with ostomy reinforcement with mesh. Ventral hernia is a complex disease process that requires
Finally, for dirty fields we will generally approach with the general surgeon to have a wide armamentarium of
planned staging and close primarily or reinforce with a repair techniques. An understanding of abdominal wall
biologic underlay. Grossly infected meshes are excised anatomy and physiology is key for restoration of abdominal
t
and primary repair is reinforced with an underlay biologic wall function. Prevention of incisional hernias is still
ne
mesh to delay a hernia recurrence. This strategy allows us under evaluation, but proper closure of laparotomies
to leave the retromuscular space intact for future definitive and thoughtful use of prophylactic mesh may reduce the
reconstruction via TAR. incidence of this costly and morbid disease. Utilization
k.
of advanced reconstructive techniques requires careful
patient selection, preoperative optimization and intricate
ENHANCED RECOVERY PATHWAYS AFTER
oo
knowledge of abdominal wall anatomy and technical
ABDOMINAL WALL RECONSTRUCTION nuances. Surgical mishaps result in significant patient
morbidity, need for reoperations, and significant nega-
yb
Since 2013, our institution has implemented a multimodal tive impact on quality of life. Techniques for repair and
enhanced recovery after surgery protocol, also known as technologies of prosthetics will continue to evolve as we
the “pathway,” to optimize abdominal wall reconstruction search for the ideal repair with restoration of normal
patients and outcomes. Our pathway begins at the initial abdominal function and durable long-term results for
er
evaluation and includes preoperative optimization that patients with routine and complex hernias.
requires patient participation and buy-in. Weight loss
rg
ventral hernia repair: making the case for hernia research. Hernia.
in mind, we recently evaluated our data and identified a 2012;16:179-183.
BMI of 45 as an upper cutoff for elective repairs, given 3. Flum DR, Horvath K, Koepsell T. Have outcomes of incisional hernia
tp
12. Bevis PM, Windhaber RA, Lear PA, Poskitt KR, Earnshaw JJ, Mitch- 37. Basile F, Biondi A, Donati M. Surgical approach to abdominal wall
ell DC. Randomized clinical trial of mesh versus sutured wound defects: history and new trends. Int J Surg. 2013;11:S20-S23.
closure after open abdominal aortic aneurysm surgery. Br J Surg. 38. Franklin ME, Dorman JP, Glass JL, Balli JE, Gonzalez JJ. Laparo-
2010;97:1497-1502. scopic ventral and incisional hernia repair. Surg Laparosc Endosc.
13. Bhangu A, Fitzgerald JE, Singh P, Battersby N, Marriott P, Pinkney T. 1998;8:294-299.
Systematic review and meta-analysis of prophylactic mesh placement 39. Brown CN, Finch JG. Which mesh for hernia repair? Ann R Coll
for prevention of incisional hernia following midline laparotomy. Surg Engl. 2010;92:272-278.
Hernia. 2013;17:445-455. 40. Colavita PD, Tsirline VB, Belyansky I, et al. Prospective, long-term
14. Helgstrand F, Rosenberg J, Bisgaard T. Trocar site hernia after laparo- comparison of quality of life in laparoscopic versus open ventral
scopic surgery: a qualitative systematic review. Hernia. 2011;15:113-121. hernia repair. Ann Surg. 2012;256:714-723.
15. Boldó E, Perez de Lucia G, Aracil JP, et al. Trocar site hernia after 41. Prasad P, Tantia O, Patle NM, Khanna S, Sen B. Laparoscopic ventral
laparoscopic ventral hernia repair. Surg Endosc. 2007;21:798-800. hernia repair: a comparative study of transabdominal preperitoneal
16. Scozzari G, Zanini M, Cravero F, Passera R, Rebecchi F, Morino M. versus intraperitoneal onlay mesh repair. J Laparoendosc Adv Surg
High incidence of trocar site hernia after laparoscopic or robotic Tech A. 2011;21:477-483.
Roux-en-Y gastric bypass. Surg Endosc. 2014;28:2890-2898. 42. Orenstein SB, Dumeer JL, Monteagudo J, Poi MJ, Novitsky YW.
17. Johnson WH, Fecher AM, McMahon RL, Grant JP, Pryor AD. VersaStep Outcomes of laparoscopic ventral hernia repair with routine
trocar hernia rate in unclosed fascial defects in bariatric patients. defect closure using ‘shoelacing’ technique. Surg Endosc. 2011;25:
Surg Endosc. 2006;20:1584-1586. 1452-1457.
18. Chiong E, Hegarty PK, Davis JW, Kamat AM, Pisters LL, Matin 43. Daes J. The enhanced view–totally extraperitoneal technique for
SF. Port-site hernias occurring after the use of bladeless radially repair of inguinal hernia. Surg Endosc. 2012;26:1187-1189.
expanding trocars. Urology. 2010;75:574-580. 44. Daes J. Endoscopic subcutaneous approach to component separation.
19. Agaba EA, Rainville H, Ikedilo O, Vemulapali P. Incidence of port-site J Am Coll Surg. 2014;218:e1-e4.
t
incisional hernia after single-incision laparoscopic surgery. JSLS. 45. Köckerling F, Botsinis MD, Rohde C, Reinpold W. Endoscopic-
ne
2014;18:204-210. assisted linea alba reconstruction plus mesh augmentation for
20. Evans KK, Chim H, Patel KM, Salgado CJ, Mardini S. Survey on ventral treatment of umbilical and/or epigastric hernias and rectus abdominis
hernias: surgeon indications, contraindications, and management diastasis—early results. Front Surg. 2016;3:27.
of large ventral hernias. Am Surg. 2012;78:388-397. 46. Reinpold W. Neue Techniken in der Narben- und Bauchwand-
k.
21. Bellows CF, Robinson C, Fitzgibbons RJ, Webber LS, Berger DH. hernienchirurgie. Chirurgische Allgemeine. 2013;14:331-337.
Watchful waiting for ventral hernias: a longitudinal study. Am Surg. 47. Belyansky I, Zahiri HR, Park A. Laparoscopic transversus abdominis
2014;80:245-252. release, a novel minimally invasive approach to complex abdominal
oo
22. Fitzgibbons RJ, Giobbie-Hurder A, Gibbs JO, et al. Inguinal hernia wall reconstruction. Surg Innov. 2016;23:134-141.
in minimally symptomatic men. JAMA. 2006;295:285-293. 48. Gonzalez AM, Rabaza JR, Seetharamaiah R, Donkor C, Romero R,
23. Lauscher JC, Loh JC, Grone J, Buhr HJ, Ritz J-P, Rieck S. Oligo- Kosanovic R, et al. Laparoscopic vs robotic ventral hernia repair: a
yb
symptomatic vs. symptomatic incisional hernias—who benefits from single group experience. http://www.sages.org/wp-content/uploads/
open repair? Langenbecks Arch Surg. 2011;396:179-185. posters/2013/45846.jpg 2013 Accessed 6 January 16.
24. Wen Y, Fayezizadeh M, Majumder A, Miller H, Novitsky Y. Analysis of 49. Tran H. Robotic single-port hernia surgery. JSLS. 2011;15:309-314.
ventral hernia repair in the obese: defining the goals of preoperative 50. Tayar C, Karoui M, Cherqui D, Fagniez PL. Robot-assisted laparoscopic
er
weight loss; 2016. In: 17th Annual Hernia Repair. mesh repair of incisional hernias with exclusive intracorporeal
25. Burger JW, Luijendijk RW, Hop WC, Halm JA, Verdaasdonk EG, Jeekel suturing: a pilot study. Surg Endosc. 2007;21:1786-1789.
J. Long-term follow-up of a randomized controlled trial of suture 51. Nguyen DH, Nguyen MT, Askenasy EP, Kao LS, Liang MK. Primary
rg
versus mesh repair of incisional hernia. Ann Surg. 2004;240:578-583, fascial closure with laparoscopic ventral hernia repair: systematic
discussion 583–585. review. World J Surg. 2014;38:3097-3104.
26. Brown CN, Finch JG. Which mesh for hernia repair? Ann R Coll 52. Booth JH, Garvey PB, Baumann DP, et al. Primary fascial closure
su
Surg Engl. 2010;92:272-278. with mesh reinforcement is superior to bridged mesh repair for
27. Bringman S, Conze J, Cuccurullo D, et al. Hernia repair: the search abdominal wall reconstruction. J Am Coll Surg. 2013;217:999-1009.
for ideal meshes. Hernia. 2010;14:81-87. 53. Berger D, Bientzle M, Müller A. Postoperative complications after
28. Albino FP, Patel KM, Nahabedian MY, Sosin M, Attinger CE, Bhanot laparoscopic incisional hernia repair. Incidence and treatment. Surg
://
29. Breuing K, Butler CE, Ferzoco S, et al. Incisional ventral hernias: literature. JSLS. 2007;11:408-414.
review of the literature and recommendations regarding the grading 55. Celdran A, Bazire P, Garcia-Urena M, Marijuán JL. H-hernioplasty:
and technique of repair. Surgery. 2010;148:544-558. a tension-free repair for umbilical hernia. Br J Surg. 1995;82:
ht
30. Kanters AE, Krpata DM, Blatnik JA, Novitsky YM, Rosen MJ. Modified 371-372.
hernia grading scale to stratify surgical site occurrence after open 56. Arroyo A, García P, Pérez F, Andreu J, Candela F, Calpena R.
ventral hernia repairs. J Am Coll Surg. 2012;215:787-793. Randomized clinical trial comparing suture and mesh repair of
31. Carbonell AM, Criss CN, Cobb WS, Novitsky YW, Rosen MJ. Outcomes umbilical hernia in adults. Br J Surg. 2001;88:1321-1323.
of synthetic mesh in contaminated ventral hernia repairs. J Am Coll 57. Aslani N, Brown CJ. Does mesh offer an advantage over tissue
Surg. 2013;217:991-998. in the open repair of umbilical hernias? A systematic review and
32. Majumder A, Winder JS, Wen Y, Pauli EM, Belyansky I, Novitsky YW. meta-analysis. Hernia. 2010;14:455-462.
Comparative analysis of biologic versus synthetic mesh outcomes in 58. Montes IS, Deysine M. Spigelian and other uncommon hernia
contaminated hernia repairs. Surgery. 2016;160:828-838. repairs. Surg Clin North Am. 2003;83:1235-1253.
33. Al Chalabi H, Larkin J, Mehigan B, McCormick P. A systematic 59. Larson DW, Farley DR. Spigelian hernias: repair and outcome for
review of laparoscopic versus open abdominal incisional hernia 81 patients. World J Surg. 2002;26:1277-1281.
repair, with meta-analysis of randomized controlled trials. Int J Surg. 60. Patle NM, Tantia O, Sasmal PK, Khanna S, Sen B. Laparoscopic
2015;20:65-74. repair of spigelian hernia: our experience. J Laparoendosc Adv Surg
34. Schluender S, Conrad J, Divino CM, Gurland B. Robot-assisted Tech A. 2010;20:129-133.
laparoscopic repair of ventral hernia with intracorporeal suturing. 61. Ramirez OM, Ruas E, Dellon AL. ‘Components separation’ method
Surg Endosc. 2003;17:1391-1395. for closure of abdominal-wall defects: an anatomic and clinical study.
35. Misiakos EP, Patapis P, Zavras N, Tzanetis P, Machairas A. Current Plast Reconstr Surg. 1990;86:519-526.
trends in laparoscopic ventral hernia repair. JSLS. 2015;19:e2015.00048. 62. Pauli EM, Rosen MJ. Open ventral hernia repair with component
36. Sauerland S, Walgenbach M, Habermalz B, Seiler CM, Miserez M. separation. Surg Clin North Am. 2013;93:1111-1133.
Laparoscopic versus open surgical techniques for ventral or inci- 63. Shahan CP, Stoikes NF, Webb DL, Voeller GR. Sutureless onlay
sional hernia repair. Cochrane Database Syst Rev. 2011;(3):CD007781, hernia repair: a review of 97 patients. Surg Endosc. 2015;30:3256-3261.
doi:10.1002/14651858.CD007781.pub2. doi:10.1007/s00464-015-4647-2.
Ventral Hernia and Abdominal Release Procedures CHAPTER 52 589
64. Saulis AS, Dumanian GA. Periumbilical rectus abdominis perforator 72. Wantz GE. Incisional hernioplasty with Mersilene. Surg Gynecol Obstet.
preservation significantly reduces superficial wound complica- 1991;172:129-137.
tions in ‘separation of parts’ hernia repairs. Plast Reconstr Surg. 73. Iqbal CW, Pham TH, Joseph A, Mai J, Thompson GB, Sarr MG.
2002;109:2275-2280, discussion 2281–2282. Long-term outcome of 254 complex incisional hernia repairs
65. Butler CE, Campbell KT. Minimally invasive component separation using the modified Rives-Stoppa technique. World J Surg. 2007;31:
with inlay bioprosthetic mesh (MICSIB) for complex abdominal 2398-2404.
wall reconstruction. Plast Reconstr Surg. 2011;128:698-709. 74. Novitsky YW, Elliott HL, Orenstein SB, Rosen MJ. Transversus
66. Lowe JB, Garza JR, Bowman JL, Rohrich RJ, Strodel WE. Endoscopi- abdominis muscle release: a novel approach to posterior component
cally assisted ‘components separation’ for closure of abdominal wall separation during complex abdominal wall reconstruction. Am J
defects. Plast Reconstr Surg. 2000;105:720-729, quiz 730. Surg. 2012;204:709-716.
67. Switzer NJ, Dykstra MA, Gill RS, et al. Endoscopic versus open 75. Novitsky YW, Fayezizadeh M, Majumder A, Neupane R, Elliott HL,
component separation: systematic review and meta-analysis. Surg Orenstein SB. Outcomes of posterior component separation with
Endosc. 2015;29:787-795. transversus abdominis muscle release and synthetic mesh sublay
68. Harth KC, Rosen MJ. Endoscopic versus open component reinforcement. Ann Surg. 2016;264(2):226-232.
separation in complex abdominal wall reconstruction. Am J Surg. 76. Jones CM, Winder JS, Potochny JD, Pauli EM. Posterior component
2010;199:342-347. separation with transversus abdominis release. Plast Reconstr Surg.
69. Ghali S, Turza KC, Baumann DP, Butler CE. Minimally invasive 2016;137:636-646.
component separation results in fewer wound-healing complications 77. Johnson KC, Miller MT, Plymale MA, Levy S, Davenport DL, Roth
than open component separation for large ventral hernia repairs. JS. Abdominal wall reconstruction: a comparison of totally extra-
J Am Coll Surg. 2012;214:981-989. peritoneal and transabdominal preperitoneal approaches. J Am Coll
70. Stoppa RE. The treatment of complicated groin and incisional Surg. 2016;222:159-165.
hernias. World J Surg. 1989;13:545-554. 78. Majumder A, Fayezizadeh M, Neupane R, Elliott HL, Novitsky YW.
t
71. Rives J, Pire JC, Flament JB, Palot JP, Body C. [Treatment of large Benefits of multimodal enhanced recovery pathway in patients under-
ne
eventrations. New therapeutic indications apropos of 322 cases]. going open ventral hernia repair. J Am Coll Surg. 2016;222:1106-1115.
Chirurgie. 1985;111:215-225. doi:10.1016/j.jamcollsurg.2016.02.015.
k.
oo
yb
er
rg
su
://
tp
ht
CHAPTER
Evan Alicuben
I
nguinal hernia repair is a frequently performed crown of death, as injury due to dissection in this area can
operation, and laparoscopic inguinal hernia repair lead to significant hemorrhage. There are several other
has become increasingly prevalent, particularly for the dangerous areas of dissection with laparoscopic hernia
repair of bilateral or recurrent hernias. The first described repair. The “triangle of doom” is located between the vas
laparoscopic inguinal hernia repair was completed in deferens medially and the gonadal vessels laterally. The
1990 by Ger in canines1; the procedure has since evolved external iliac vessels, deep circumflex iliac vein, genital
to include the use of a prosthetic mesh to cover the branch of the genitofemoral nerve, and femoral nerve
t
myopectineal orifice. There are two commonly performed are located within this triangle.
ne
techniques: the transabdominal preperitoneal repair The key nerves in this area are located lateral to the
(TAPP) and the totally extraperitoneal repair (TEP). This deep inguinal ring (Fig. 53A.2). In the laparoscopic
chapter reviews the anatomy, technical considerations, approach, the following nerves with cutaneous innervation
k.
benefits, and possible complications of laparoscopic may be encountered from laterally to medially: the lateral
inguinal hernia repair. femoral cutaneous nerve, the anterior femoral cutaneous
oo
nerve, femoral nerve, femoral branch of the genitofemoral
SURGICAL ANATOMY OF THE REGION nerve, and genital branch of the genitofemoral nerve.
The area inferior to the iliopubic tract and lateral to the
yb
A comprehensive understanding of the anatomy of the gonadal vessels is known as the “triangle of pain,” where
preperitoneal space is critical to the performance of a the lateral femoral cutaneous nerve and femoral branch
safe and effective laparoscopic inguinal hernia repair.2 of the genitofemoral nerve are found. Tacks placed in this
The anterior approach to inguinal hernia repair involves area may injure either of these nerves. Together, the area
er
recognition of the anatomy from a superficial to deep between the vas deferens medially and the iliopubic tract
position; this is in contrast to laparoscopic repair, which superiorly and laterally constitutes “the square of doom,”
rg
requires identification of the critical structures from a where tacks and electrocautery should never be applied
reversed viewpoint. to avoid nerve injury.
The median umbilical ligament covers the urachus and The previous anatomic discussion is required to
su
travels from the umbilicus to the bladder (Fig. 53A.1). understand the landmarks that define the three spaces
The paired medial umbilical ligaments are remnants associated with groin hernias (Fig. 53A.3):
of the fetal umbilical arteries. The inferior epigastric 1. Indirect inguinal hernia—lateral to the inferior epi-
://
vessels originate from the external iliac vessels and have gastric vessels.
a peritoneal covering creating the paired lateral umbilical 2. Direct inguinal hernia—medial to the inferior epigastric
tp
folds. These folds originate medial to the deep inguinal vessels and lateral to the border of the rectus abdominis
ring and travel to the arcuate line, where the inferior muscle within the triangle of Hesselbach.
epigastric vessels enter the rectus sheath. The medial 3. Femoral hernia—below the iliopubic tract, medial to
ht
inguinal fossa is the space located between the medial the external iliac vein, and lateral to Cooper ligament.
umbilical ligament and lateral umbilical fold bilaterally. All three of these spaces should be covered during the
This is the space associated with direct inguinal hernias. laparoscopic approach with an appropriately sized mesh.
The lateral inguinal fossa is the depression lateral to the
lateral umbilical fold. Indirect inguinal hernias develop at
this site. The pectineal (Cooper) ligament is formed from
LAPAROSCOPIC VERSUS OPEN REPAIR
fascia and periosteum and travels along the pectineal line The decision to perform an inguinal hernia repair via a
of the pubic bone. The iliopubic tract is a thickened band laparoscopic versus an anterior approach remains difficult,
of fibers from the transversalis fascia that joins laterally to complex, and nuanced. Tension-free anterior mesh repair
the iliac crest and inserts medially on the pubic tubercle has been the gold standard because of its low recurrence
and pectineal line. rate and lack of need for specialized equipment. Critiques
The major vascular structures in this region are generally of the laparoscopic approach include the higher in-hospital
located medial to the deep inguinal ring. The inferior costs due to more expensive equipment, the possibility
epigastric vessels branch from the external iliac vessels and of intraabdominal organ or vascular injury, and the steep
travel to supply the anterior abdominal wall. A vascular learning curve. Studies have indicated that greater than
connection may be noted in some patients between the 250 operations are required to become experienced in this
obturator and external iliac vessels crossing the superior technically challenging operation.3-5 In fact, a recurrence
pubic ramus. This is known as the “corona mortis,” or rate of greater than 10% has been reported for surgeons
590
Inguinal Hernia Repair: Laparoscopic CHAPTER 53A 590.e1
ABSTRACT
Symptomatic inguinal hernias are a commonly encountered
surgical problem. Repair can be performed with an open or
minimally invasive approach. Laparoscopic inguinal hernia
repair is now widely performed, especially in the setting
of bilateral or recurrent inguinal hernias. This chapter
reviews the anatomy, benefits, and possible complications
of laparoscopic inguinal hernia repair. Laparoscopic
inguinal hernia repair is also compared with the traditional
anterior approach, and the operative techniques of both
the transabdominal preperitoneal repair and the totally
extraperitoneal repair are reviewed. The outcomes of
these two procedures are also examined.
KEYWORDS
Inguinal hernia, femoral hernia, hernia repair, mesh,
laparoscopy, transabdominal preperitoneal repair, totally
extraperitoneal repair
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
Inguinal Hernia Repair: Laparoscopic CHAPTER 53A 591
t
ne
Femoral branch
genitofemoral nerve
k.
Danger zone
oo Femoral nerve
yb
er
rg
su
Genital branch
genitofemoral nerve
://
Lateral femoral
tp
cutaneous nerve
ht
Iliacus muscle
Psoas muscle
who have performed fewer than 250 procedures.3 For this Nevertheless, most large trials have failed to consistently
reason, some have argued that the laparoscopic approach show a difference in recurrence rates between the two
should be reserved for experienced centers. Conversely, approaches.8-11 The biggest exception is a study out of
many trials have shown significantly less early postoperative the Veterans Administration (VA) system from 2004.3
pain resulting in less narcotic use and swifter return to In this study, more than 2000 male patients within the
normal activity.6,7 VA system were randomized to an open or laparoscopic
592 SECTION I Esophagus and Hernia
Area of
direct hernia
Area of
indirect hernia
Internal
inguinal
ring
Pubic
tubercle Edge of
Inferior inguinal
epigastric vein ligament
Iliopubic tract
t
ne
Cooper
ligament
Gonadal
Femoral canal vessels
k.
oo
External iliac
Vas deferens
yb vessels
er
rg
inguinal hernia repair with mesh. With 2-year postoperative additional incisions. In recurrent hernias, especially those
follow-up, the laparoscopic group had a significantly with previous open repair, a posterior approach results
://
higher recurrence rate than the anterior approach in in dissection through native tissue planes. Some studies
the repair of primary hernias (10.1% vs. 4.0%). However, have suggested this may lead to improved recurrence
tp
recurrence rates were similar if the repairs of recurrent compared to redo anterior repair.13 Inguinal hernias in
hernias were compared (10.0% vs. 14.1%). This study females may be better performed laparoscopically given
has steered many surgeons to repair unilateral primary that the posterior placement of mesh allows for coverage
ht
inguinal hernias through an open approach. of the femoral space, therefore addressing the incidence
More recent research has attempted to clarify this of femoral recurrence seen in Lichtenstein repair. The
dilemma. In a meta-analysis of 27 randomized controlled importance of this concept was illustrated in a series of
trials of primary unilateral inguinal hernia repair including hernia repairs in women which showed a 41.6% rate of
7161 patients, O’Reilly et al. found that the laparoscopic femoral hernia found during operations for recurrences.14
approach resulted in an increased risk of recurrence.12
Interestingly, compared with open repair, TEP had
increased rates of recurrence but TAPP had equivalent OPERATIVE TECHNIQUE
rates. Additionally, the laparoscopic approach was associ-
ated with higher perioperative complication risk. This was TRANSABDOMINAL PREPERITONEAL REPAIR
attributed to a higher complication risk with TAPP but The patient is placed in the supine position with both
equivalent risk with TEP. However, laparoscopic repair arms tucked. An indwelling catheter is placed in the
resulted in reduced risk of chronic groin pain and numb- bladder to prevent view obstruction and decrease
ness compared with the anterior approach. the risk of injury to the bladder during dissection of
The laparoscopic repair technique may be better the preperitoneal space. The monitor is placed at the
suited for bilateral or recurrent inguinal hernias and foot of the table. The surgeon stands behind the shoulder
hernias in women. Bilateral hernias may be repaired opposite to the hernia, and the camera assistant stands
through the same set of port sites and do not require on the other side of the patient. Steep Trendelenburg is
Inguinal Hernia Repair: Laparoscopic CHAPTER 53A 593
5-mm
trocar 5-mm
trocar
Bladder
Spermatic
Optional Iliac vessels
additional Vas deferens
vessels
t
grasper
ne
FIGURE 53A.5 Creation of the peritoneal flap. (From Eubanks S.
Hernias. In: Sabiston DC Jr, Lyerly HK, eds. Sabiston Textbook of
k.
Surgery: The Biological Basis of Modern Surgical Practice. 15th
ed. Philadelphia: Saunders; 1997:1226.)
FIGURE 53A.4 Trocar placement for transabdominal preperitoneal
oo
hernia repair.
Inferior epigastric
vessels Musculoaponeurotic arch
of transversus abdominis
yb
required to remove the small bowel from the pelvis and Rectus abdominis
adequately visualize the area to be dissected. Pubis
Three ports are required for this operation. A 10-mm Iliopubic tract
port for the laparoscope is placed at the umbilicus. Two
er
Iliac vessels
to open the peritoneum at the anterior abdominal wall Spermatic
without the 30-degree angle. vessels
tp
trocars are inserted too low, it can be very difficult to Hernias. In: Sabiston DC Jr, Lyerly HK, eds. Sabiston Textbook of
produce an adequately sized peritoneal flap or easily Surgery: The Biological Basis of Modern Surgical Practice. 15th
maneuver the tacking device or the fibrin glue sprayer. ed. Philadelphia: Saunders; 1997:1226.)
On the other hand, if they are placed too high, the small
bowel may get in the way. Therefore optimal placement electrocautery is used with care in this area to avoid nerve
of trocars and a 30-degree laparoscope are essential to injury. The femoral nerve is present under the iliopubic
success. tract at the lateral aspect of the dissection, but this nerve
The peritoneal flap may be incised from lateral to is not commonly visualized during this procedure. In
medial or medial to lateral. If a lateral-to-medial dissection thin patients, the lateral femoral cutaneous nerve and
is chosen, the incision begins medially to the anterior the genitofemoral nerve may be identified.
superior iliac spine (Fig. 53A.5). This then extends medi- In males, the spermatic cord structures are dissected
ally, staying at least 2 cm above the deep inguinal ring and free of the peritoneal flap. This involves separating the
hernia defect, and ends at the medial umbilical ligament. cord structures, including the vas deferens, from the
Blunt dissection is used to enlarge the peritoneal flap peritoneum and the hernia sac. The peritoneum must
and expose the critical landmarks in the preperitoneal be dissected quite inferiorly, as inadequate mobilization
space, including the pubic tubercles, Cooper ligament, can result in folding of the mesh after peritoneal closure
and the iliopubic tract (Fig. 53A.6).This dissection of the and early recurrence. If the view of the operative site is
areolar tissue can be done with minimal hemostasis, and obscured with blood, it can be irrigated and aspirated
594 SECTION I Esophagus and Hernia
Mesh stapled into ring to avoid nerve injury. Nerve injury can lead to severe
Cooper ligament chronic pain due to neuroma formation around the staple
and around medial or tack. Then tacks may be placed laterally and medially;
Optional additional edge first
laterally, it is essential to stay above the iliopubic tract, but
grasper
tacks placed medially are inserted into the rectus muscle
and on Cooper ligament. Usually two staples or tacks are
placed in Cooper ligament and one or two in the rectus
muscle. Staples or tacks are often used for fixation in
laparoscopic inguinal hernia repair due to the risk of
shrinkage or migration of the mesh.
After the mesh is secured, pneumoperitoneum pressure
is reduced to 10 mm Hg. The peritoneal flap is then
positioned over the mesh and closed with tacks. Absorbable
tacks are preferred to prevent subsequent adhesions to
the tacks. Complete mesh coverage is essential to prevent
exposure of the mesh to the underlying small bowel.
This can lead to the creation of adhesions and possible
small bowel obstruction. If possible, tacking is performed
t
in an overlap fashion. The peritoneal flap can also be
ne
FIGURE 53A.7 Placement and fixation of the mesh. (From Eubanks
S. Hernias. In: Sabiston DC Jr, Lyerly HK, eds. Sabiston Textbook closed using a continuous running suture per surgeon
of Surgery: The Biological Basis of Modern Surgical Practice. 15th preference. The fascia is routinely sutured at the umbilical
ed. Philadelphia: Saunders; 1997:1226.) port during closure.
k.
TOTALLY PREPERITONEAL HERNIA REPAIR
oo
or cleaned with gauze placed intraabdominally. Direct The arcuate line of Douglas is a transitional line; above
hernia sacs and small indirect hernia sacs can often be it, the rectus muscle has a defined anterior and posterior
easily reduced during this dissection. sheath composed of the aponeurotic fascia of the internal
yb
In the case of a very large indirect inguinoscrotal hernia, oblique and transversus abdominis muscles. Below the
the distal part of the sac can be divided and left within the arcuate line, all fascial layers of the abdominal muscles are
scrotum if it cannot be reduced. This dissection begins anterior to the rectus muscle, and behind the rectus muscle
with gentle and atraumatic separation of the sac from itself there is only the transversalis fascia. Preperitoneal
er
the spermatic cord structures. As the sac is separated, dissection is performed below the arcuate line, which is
it is divided, but care should always be taken to ensure located approximately midway between the umbilicus
rg
that the vas deferens is not included with the sac. It may and pubis.
be easier to locate the vas deferens before starting the The operation is begun by making a 10-mm incision
division of the hernia sac, but often a gradual division below the umbilicus. Two retractors are used to slide the
su
of the sac will allow for complete separation of the sac incision to the right if the hernia is located on the right
from the cord. Once the peritoneal sac is completely side, or to the left if the hernia is located on that side.
excised, the operation continues as usual. The distal The anterior rectus sheath on the side of the hernia is
://
portion of the divided sac is left open in the inguinal then opened under direct vision, and two stay sutures are
canal and the proximal part is ligated using an Endoloop placed on each edge. The rectus muscle is then bluntly
tp
or clips. This method is employed if the complete sac separated using two retractors so that the posterior fascia
cannot be reduced as significant dissection at the distal can be visualized. At this point, it is essential not to open
sac in larger hernias can cause hematoma formation, the posterior fascia of the rectus muscle but instead to
ht
ischemic orchitis, or atrophy of the testicle. dissect inferiorly toward the symphysis pubis in an oblique
When the hernia sac has been completely reduced and fashion, using either the index finger or a small peanut
dissection of the preperitoneal space is complete, the mesh with an angulation of about 30 degrees. That will lead to
is rolled and introduced through the umbilical port using the preperitoneal space below the arcuate line of Douglas.
a grasper (Fig. 53A.7). Mesh of an appropriate size should The preperitoneal space is then dissected using a
be used, typically a 15- by 10-cm piece will be adequate balloon space maker under direct vision with a 30-degree
for one side. Once inside the abdominal cavity, the mesh laparoscope (Fig. 53A.8). While the balloon is inflated, the
is unrolled and positioned so that it covers the direct, rectus muscle should be seen anteriorly and superiorly,
indirect, and femoral spaces. Fixation is then performed and the preperitoneal fat and peritoneum should be seen
with fibrin glue, tacks, or suture. Some surgeons feel that posteriorly. It is preferred that the inferior epigastric vessels
complications of fixation, and associated vascular or nerve stay anteriorly with the rectus muscle, as otherwise they
entrapment can be avoided if a significantly large piece may hamper dissection and possibly require ligation. The
of mesh is placed. Our technique of choice currently is to balloon should stay in place for approximately 15 seconds
use fibrin glue. The fibrin glue is sprayed over the mesh to tamponade any bleeding. The space-maker balloon is
in a thin layer, especially over Cooper ligament and the then removed and the Hasson port is introduced.
lateral aspect of the mesh. However, if one chooses to Two 5-mm ports are placed at the midline between the
use tacks, the mesh fixation can begin at the midportion, umbilicus and the symphysis pubis to allow for operating on
“three fingers” above the superior limit of the internal both sides. The space created using this technique is small
Inguinal Hernia Repair: Laparoscopic CHAPTER 53A 595
Peritoneum
Pubis
Bladder
A B
t
ne
Expanded preperitoneal space
maintained with insufflation
Balloon inflated in
k.
preperitoneal space
oo
Pubis
yb
er
Peritoneum
Bladder
rg
C D
FIGURE 53A.8 Balloon dissection of the preperitoneal space. (A) A balloon dissector is introduced into the preperitoneal space. (B) The
su
balloon dissector is advanced to the pubis in the preperitoneal space. (C) The balloon is inflated under direct vision with the laparoscope.
(D) The balloon is removed and the preperitoneal space is insufflated with carbon dioxide. (From Shadduck PP, Schwartz LB, Eubanks
WS. Laparoscopic inguinal herniorrhaphy. In: Pappas TN, Schwartz LB, Eubanks WS, eds. Atlas of Laparoscopic Surgery. Philadelphia:
://
and limits the movement of the laparoscopic instruments. used to separate the sac from the cord structures and vas
Care should be taken to avoid perforating the peritoneum, deferens. Small hernia sacs are usually easily reduced. If
ht
since an opening in the peritoneum will allow CO2 to the hernia sac is large and dissection difficult, it may be
escape into the abdominal cavity, which will subsequently prudent to amputate the sac, leave the distal part of the
compress the space and reduce it further. If this occurs, a sac open, and close the proximal opening either with
Veress needle can be inserted into the abdomen to allow clips or an Endoloop. The mesh is then placed into the
for the exit of CO2. The perforation can be then be closed preperitoneal space and positioned as mentioned in
using an Endoloop or a 5-mm clip applier. the previous section. Once the mesh is in place, fibrin
After the ports are in place, Cooper ligament is identi- sealant or tacks can be used to fix the mesh in position.
fied medially and the iliac vein’s position is noted as well. The fibrin sealant can be applied as an aerosol spray,
These two structures are key anatomic landmarks in this which is attached to one of the open trocars to prevent
procedure, as they aid in defining the inferior aspect of excessive pressure inside the abdomen. This fibrin glue
the dissection. It is possible to injure the iliac vein, and can be placed above or below the lightweight, large-pore
there have been reports of ligation of the iliac vein when mesh. It is important to use a lightweight, large-pore mesh
it was mistaken for the hernia sac. Once the iliac vein is to avoid or minimize the risk of mesh contraction and a
identified after careful dissection, the next step is locating foreign body reaction. The operation ends by desufflating
the inferior epigastric vessels. This will help delineate the the preperitoneal space and gradually removing the ports.
internal ring and Hesselbach triangle. The vas deferens A grasper is used to hold the mesh in place as the CO2 is
is identified by following the internal ring medially and evacuated. After port removal, the anterior fascia is closed
toward the iliac vein. Gentle, blunt dissection is then with the previously placed stay sutures.
596 SECTION I Esophagus and Hernia
t
shown an advantage of one approach over the other.5,15-20 with a cannula while preparations are made for suture
ne
The largest published series is from the Herniamed ligation using transfascial sutures placed with fascial closure
registry, which is a multicenter Internet-based database devices.
with participating surgeons in Germany, Austria, and
k.
Switzerland. In their published series of 17,587 patients Bowel Injury
who underwent laparoscopic repair of a primary unilateral Injury to bowel most commonly occurs during entry into
oo
inguinal hernia, the TEP and TAPP techniques had similar the abdomen with the trocar or Veress needle and is
intraoperative complication rates but the TEP method had estimated to occur in 0.13% of cases.28 Most frequently,
significantly fewer postoperative complications.21 Subgroup the small bowel is injured and this can be detected intra-
yb
analysis indicated that these findings could be attributed operatively in two-thirds of cases. Missed bowel injuries
to a difference in bleeding and seroma formation. Risk usually present with peritonitis and sepsis approximately
factors for the development of complications included one week postoperatively. An experienced laparoscopic
increased defect size and the presence of a scrotal hernia. surgeon may repair the injury via intracorporeal sutur-
er
Of note, these characteristics were more commonly seen ing or stapling, taking care to avoid narrowing of the
in the TAPP group, and this may explain the difference in intraluminal diameter.
rg
patients, the TEP approach was associated with a signifi- trocar or during dissection. Small injuries can be managed
cantly increased rate of intraoperative complications (TEP with urinary drainage, but larger defects require repair.
6.3% vs. TAPP 2.8%) and a longer operative time (TEP If such an injury is recognized, laparoscopic repair is an
://
80.3 minutes vs. TAPP 73.0 minutes), but the rates of option if the experience of the surgeon is sufficient. If
postoperative complication were similar.22 Overall these unrecognized, injury may present later with hematuria,
tp
trials, Wei et al. reported no significant difference in The risk of developing urinary retention is significantly
operative time, cost, complications, or time to return to higher in laparoscopic repair compared with the
usual activities.23 Given the failure to show benefit in the anterior approach, given the use of general anesthesia
TEP group, this study suggested that the TAPP approach compared with the possibility of local anesthesia for
may be the preferred strategy, especially for nonexpert open repairs. The actual reported incidence varies
laparoscopic surgeons. widely up to 22%, and is higher in older patients with
history of prostatic symptoms.29-31 In one study, postopera-
COMPLICATIONS tive narcotic use of 6.5 mg or greater of morphine or
The incidence of complications following laparoscopic morphine equivalent was an independent risk factor
inguinal hernia repair has gradually declined and should for the development of urinary retention. 31 Narcotic
continue to decline as the procedure becomes more avoidance with the usage of antiinflammatory medica-
prevalent. The reported rates of complications vary widely tions, acetaminophen, and regional nerve blocks may
and reach as high as 25%, depending on the definition help to decrease the risk of retention. The amount of
of morbidity put forth by the studies.7,24 intraoperative fluids given, the presence of a urinary
catheter during surgery, and operative time were not
Vascular Injuries associated with increased risk. Intermittent catheterization
Significant vascular injury during the procedure is most or a temporary indwelling catheter usually resolves the
likely to occur while gaining access to the peritoneal cavity. symptoms.
Inguinal Hernia Repair: Laparoscopic CHAPTER 53A 597
t
in inguinal hernia repair. Ann Surg. 2005;242:344.
artery. Duplex ultrasound should be performed to evaluate
ne
5. Wake BL, McCormack K, Fraser C, Vale L, Perez J, Grant AM.
for testicular torsion. Patients are managed expectantly Transabdominal pre-peritoneal (TAPP) vs totally extraperitoneal
with antiinflammatory agents for symptoms. Testicular (TEP) laparoscopic techniques for inguinal hernia repair. Cochrane
necrosis may ensue and eventually result in testicular Database Syst Rev. 2005;(1):CD004703.
k.
atrophy; however, this happens in a minority of cases.24 6. Grant AM. Laparoscopic versus open groin hernia repair: meta-analysis
of randomized trials based on individual patient data. Hernia.
Some believe that laparoscopic repair may result in less 2002;6:2.
oo
testicular ischemia due to a higher dissection on the 7. Memon MA, Cooper NJ, Memon B, Memon MI, Abrams KR.
spermatic cord. However, postoperative color Doppler Meta-analysis of randomized clinical trials comparing open and
ultrasonography has shown no measurable difference in laparoscopic inguinal hernia repair. Br J Surg. 2003;90:1479.
8. Pikoulis E, Tsigris C, Diamantis T, et al. Laparoscopic preperitoneal
blood flow.32
yb
mesh repair or tension-free mesh plug technique? A prospective study
of 471 patients with 543 inguinal hernias. Eur J Surg. 2002;168:587.
Groin Pain 9. Eklund A, Rudberg C, Smedberg S, et al. Short-term results of a
The etiology and management of postinguinal groin
er
randomized clinical trial comparing Lichtenstein open repair with
pain following hernia repair remains poorly understood totally extraperitoneal laparoscopic inguinal hernia repair. Br J Surg.
2006;93:1060.
and reflects the complex nature of this problem. The 10. Eklund AS, Montgomery AK, Rasmussen C, Sandbue RP, Bergkvist
rg
incidence has varied greatly but has been reported to be LA, Rudberg CR. Low recurrence rate after laparoscopic (TEP) and
as high as 53%, with 1% requiring referral to specialty open (Lichtenstein) inguinal hernia repair. Ann Surg. 2009;249:33.
pain clinic.33,34 Laparoscopic repair causes significantly 11. Langeveld HR, van’t Riet M, Weidema WF, et al. Total extraperitoneal
su
20. Sharma D, Yadav K, Hazrah P, Borgharia S, Lal R, Thomas S. 28. Van der Voort M, Heijnsdijk EAM, Gouma DJ. Bowel injury as a
Prospective randomized trial comparing laparoscopic transabdominal complication of laparoscopy. Br J Surg. 2004;91:1253.
preperitoneal (TAPP) and laparoscopic totally extraperitoneal (TEP) 29. Koch CA, Grinberg GG, Farley DR. Incidence and risk factors
approach for bilateral inguinal hernias. Int J Surg. 2015;22:110. for urinary retention after endoscopic hernia repair. Am J Surg.
21. Kockerling F, Bittner R, Jacob DA, et al. TEP versus TAPP: comparison 2006;191:381.
of the perioperative outcome in 17,587 patients with a primary 30. Dulucq JL, Wintringer P, Mahajna A. Laparoscopic totally extraperi-
unilateral inguinal hernia. Surg Endosc. 2015;29:3750. toneal inguinal hernia repair: lessons learned from 3,100 hernia
22. Gass M, Scheiwiller A, Sykora M, Metzger J. TAPP or TEP for repairs over 15 years. Surg Endosc. 2009;23:482.
recurrent inguinal hernia? Population-based analysis of prospective 31. Patel JA, Kaufman AS, Howard RS, Rodriguez CJ, Jessie EM. Risk
data on 1309 patients undergoing endoscopic repair for recurrent factors for urinary retention after laparoscopic inguinal hernia
inguinal hernia. World J Surg. 2016;40(10):2348-2352. repairs. Surg Endosc. 2015;29:3140.
23. Wei FX, Zhang YC, Han W, Zhang YL, Shao Y, Ni R. Transabdominal 32. Gurbulak EK, Gurbulak B, Algun IE, et al. Effects of totally extra-
preperitoneal (TAPP) versus totally extraperitoneal (TEP) for peritoneal (TEP) and Lichtenstein hernia repair on testicular blood
laparoscopic hernia repair: a meta-analysis. Surg Laparosc Endosc flow and volume. Surgery. 2015;158:1297.
Percutan Tech. 2015;25:375. 33. Poobalan AS, Bruce J, Cairns W, et al. A review of chronic pain after
24. Bittner R, Schmedt CG, Schwarz J, Kraft K, Leibl BJ. Laparoscopic inguinal herniorrhaphy. Clin J Pain. 2003;19:48.
transperitoneal procedure for routine repair of groin hernia. Br J 34. Gitelis ME, Patel L, Deasis F, et al. Laparoscopic totally extraperitoneal
Surg. 2002;89:1062. groin hernia repair and quality of life at 2-year follow-up. J Am Coll
25. Schafer M, Lauper M, Krahenbuhl L. A nation’s experience of Surg. 2016;223:153.
bleeding complications during laparoscopy. Am J Surg. 2000;180:73. 35. Singh AN, Bansal VK, Misra MC, et al. Testicular functions, chronic
26. Azevedo JLMC, Azevedo OC, Miyahira SA, et al. Injuries caused groin pain, and quality of life after laparoscopic and open mesh
by Veress needle insertion for creation of pneumoperitoneum: a repair of inguinal hernia: a prospective randomized controlled trial.
t
systematic literature review. Surg Endosc. 2009;23:1428. Surg Endosc. 2012;26:1304.
ne
27. Tamme C, Scheidbach H, Hampe C, Schneider C, Kockerling F. 36. Barazanchi AWH, Fagan PVB, Smith BB, Hill AG. Routine neurectomy
Totally extraperitoneal endoscopic inguinal hernia repair (TEP). of inguinal nerves during open onlay mesh hernia repair. Ann Surg.
Surg Endosc. 2003;17:190. 2016;264:64.
k.
oo
yb
er
rg
su
://
tp
ht
CHAPTER
Inguinal Hernia Repair: Open
Kamran Samakar
| Kulmeet K. Sandhu
| Namir Katkhouda
53B
I
nguinal hernia repair is the most common general running between the internal (deep) inguinal and external
surgical procedure in the United States with approxi- (superficial) inguinal rings. In men, the inguinal canal
mately 800,000 performed annually. Throughout its contains the spermatic cord and, in women, it contains
long history, many techniques have been proposed for the the round ligament. During open inguinal hernia repair,
repair of inguinal hernias. Modern-day repair of inguinal the iliohypogastric, ilioinguinal, and genital branches
hernias is based on the tenets of minimizing tension and of the genitofemoral nerves are encountered (Fig. 53B.3).
the use of mesh to provide a lasting repair. In this chapter, The ilioinguinal and iliohypogastric nerves can be identi-
we review the most common techniques for the surgical fied as they pass between the external and internal oblique
repair of inguinal hernias, relevant anatomy, and the muscles. The genital branch of the genitofemoral nerve
t
postoperative complications of herniorrhaphy. is generally found outside the area of dissection behind
ne
the cord structures.
BACKGROUND Inguinal hernias are classified anatomically as direct or
indirect. Indirect inguinal hernias are the most common
k.
A hernia is defined as an abnormal protrusion or bulge type of hernia in both sexes. Indirect hernias are a protru-
of an organ or tissue through the surrounding walls. The sion of the hernia sac at the internal ring, lateral to the
oo
prevalence of inguinal hernias is estimated at 5% to 10% inferior epigastric vessels. Indirect inguinal hernias are the
of the population of the United States. Inguinal hernia result of a patent processus vaginalis. In contrast, the sac
repair is the most common general surgery procedure in of a direct inguinal hernia protrudes medial to the inferior
yb
the United States, with approximately 800,000 performed epigastric vessels, within Hesselbach triangle. Hesselbach
annually. triangle is formed by the inguinal ligament (Poupart
Inguinal hernias have a variety of clinical manifestations, ligament) inferiorly, the inferior epigastric vessels laterally,
ranging from a painless groin bulge to pain without a and the rectus abdominis muscle medially. Direct hernias
er
bulge. Common symptoms include pain described as dull are a result of weakness in the floor of the inguinal canal.
discomfort or pinching in the groin, which may or may
rg
lying flat are all associated with inguinal hernias. The most remains the most common approach to primary unilateral
common finding on physical examination in adults is a hernias.5 The exact choice of repair may vary depending
groin bulge. In most cases, diagnosis of an inguinal hernia on the use of mesh as well as operative technique. Based
://
can be made by history and physical examination alone.1 on multiple large systematic reviews, various hernia society
When the diagnosis is not apparent, ultrasonography of guidelines generally advocate the use of mesh in a tension-
tp
the groin should be considered as the initial diagnostic free technique for hernia repair.6-8 Hernia recurrence
modality.2 Other imaging modalities include computed rates using mesh typically range from 1% to 5% and are
tomography (CT) and magnetic resonance imaging (MRI), estimated to be significantly lower than nonmesh repairs.9
ht
both of which can aid in the evaluation of an inguinal The ideal mesh used for inguinal hernia repair should be
hernia. lightweight, macroporous, and inexpensive.10 Still, the use
Consideration of hernia repair depends on the patient’s of mesh in contaminated fields and complicated hernia
symptoms and the potential for incarceration or stran- repairs remains controversial with evidence to suggest
gulation. Common practice is to offer elective repair that it may be done safely in certain circumstances.11-16
of symptomatic inguinal hernias for patients who are
physically fit for surgery. Nonoperative management of CONVENTIONAL ANTERIOR OPEN APPROACH
asymptomatic or minimally symptomatic inguinal hernias Open anterior inguinal hernia repairs generally follow the
has been shown to be a safe and acceptable approach.3,4 same initial steps: skin incision along the lines of Langer,
Femoral hernias should be repaired at the time of diagnosis deepening of the incision through Camper and Scarpa
due to the increased risk of strangulation. fascia to the external oblique aponeurosis, and incision
of the external oblique through the external ring. Once
ANATOMY the external oblique aponeurosis has been incised, the
The inguinal canal is formed by the aponeurosis of the superior flap is created by bluntly sweeping off the internal
external oblique muscle anteriorly, the transversalis fascia, oblique muscle. The ilioinguinal and iliohypogastric
and the transversus abdominis muscles posteriorly (Figs. nerves are identified and preserved. Selective use of
53B.1 and 53B.2). The canal is approximately 4 cm in neurectomy is advocated in cases of inadvertent trauma or
length and is located cephalad to the inguinal ligament presumed injury due to mesh entrapment.17-19 Inferiorly,
599
Inguinal Hernia Repair: Open CHAPTER 53B 599.e1
ABSTRACT
Inguinal hernia repair is the most common general surgical
procedure in the United States, with roughly 800,000
performed annually. Throughout its long history, many
techniques have been proposed for repair of inguinal
hernias. Modern-day repair of inguinal hernias is based
on the tenets of minimizing tension and the use of mesh
to provide a lasting repair. In this chapter, we review the
most common techniques for surgical repair of inguinal
hernias, relevant anatomy, and postoperative complications
of herniorrhaphy.
KEYWORDS
open inguinal hernia repair, Lichtenstein repair, Bassini
repair, Shouldice repair, McVay repair, femoral hernia
repair, inguinal anatomy, inguinodynia, tension-free
mesh repair, complications of hernia repair
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
600 SECTION I Esophagus and Hernia
Internal oblique
Transversalis
fascia
Transversus
Spermatic
abdominis
cord
Deep circumflex
Conjoint tendon iliac artery,
Inguinal ascending branch
ligament, Inferior epigastric
reflected part artery and venae
Dorsal nerve comitantes
of penis Femoral canal
Dorsal artery Femoral vein
of penis
Femoral artery
Deep dorsal
vein of penis
t
ne
FIGURE 53B.1 Anterior view of the muscles of the groin Spermatic Genitofemoral
area (parts of the external and internal oblique have cord nerve,
been removed). (From Standring S, Ellis H, Healy JC, femoral branch
Superficial
et al., eds. Gray’s Anatomy: The Anatomical Basis of
k.
dorsal vein
Clinical Practice. 40th ed. Philadelphia: Churchill of penis
Livingstone; 2008; Fig. 61.15.)
oo
yb
Internal mammary artery
er
Rectus muscle
Transversus abdominis
muscle
Skin
rg
Internal oblique
Rectus abdominis muscle Sup. epigastric artery
muscle
External oblique
muscle
su
Transversus abdominis
Linea alba
://
Deep epigastric
vessels Peritoneum Internal oblique muscle
tp
Transversalis
Transversus abdominis fascia
muscle External oblique muscle
ht
Transversalis fascia
the cord structures are separated from the inferior flap Penrose drain. Once the Penrose is placed for retraction,
of the external oblique aponeurosis and blunt dissection dissection of the cord is performed in order to identify
is carried onto the pubic tubercle. Using both index an indirect hernia sac. The indirect hernia sac is then
fingers, the surgeon creates a window behind the cord dissected free from the cord structures up to the level of
structures at the pubic tubercle to allow for passage of a the internal ring. The sac can either be high-ligated with
Inguinal Hernia Repair: Open CHAPTER 53B 601
AP IE
CL
RP PB
DC
UA
IV IPA
IP
VD IA
B GB
FB
A B
FN
GN LC
IM
CI IS
U
IL
A B PM
FIGURE 53B.3 (A) Preperitoneal view of the right groin depicting the so-called triangle of doom (A) and the triangle of pain or the electrical
t
hazard zone (B). (B) Cadaveric preparation that shows the structures included within these triangles that could be damaged during a
ne
preperitoneal herniorrhaphy. AP, Anterior pubic branch and iliopubic vein; B, bladder (reflected posteriorly); CI, common iliac artery;
CL, Cooper ligament; DC, deep circumflex iliac vessels; FB, femoral branch of the genitofemoral nerve; FN, femoral nerve; GB, genital
branch of the genitofemoral nerve; GN, genitofemoral nerve; IA, external iliac artery; IE, inferior epigastric vessels; IL, ilioinguinal nerve;
k.
IM, musculus iliacus; IP, iliopubic tract; IPA, iliopectineal arch; IS, internal spermatic vessels; IV, external iliac vein; LC, lateral femoral
cutaneous nerve; PB, anastomotic pubic branch; PM, musculus psoas major; RP, retropubic vein; U, ureter; UA, umbilical artery; VD, vas
deferens. (From Greene FL, Ponsky JL. Endoscopic Surgery. Philadelphia: Saunders; 1994:365.)
oo
yb
division and suture closure or it can simply be inverted fascia. Care should be taken not to entrap the ilioinguinal,
and reduced into the preperitoneal space.20 If a direct iliohypogastric, or genital branches of the genitofemoral
hernia is present, a purse-string suture can be placed in nerves when placing sutures. The main limitation of this
the transversalis fascia at the base of the hernia to allow technique is that it does not address femoral hernias.
er
performed after incising the cremasteric muscle longitu- plug may be used, and the plug is inserted into the hernia
dinally to fully mobilize the sac. Similarly, direct hernias defect and secured to the edges with suture (Fig. 53B.5).
are circumferentially dissected and reduced back into The plug can similarly be positioned within the indirect
://
the preperitoneal space. A large mesh prosthesis is then hernia and secured in place along the edges as well (Fig.
tailored to the shape and size of the patient’s anatomy to 53B.6). In this way the plug can act as a preperitoneal
tp
facilitate overlap of 2 cm onto the pubic tubercle, 4 cm underlay mesh. The patch is then positioned in a fashion
above Hesselbach triangle, and 5 to 6 cm lateral to the similar to the Lichtenstein technique along the inguinal
internal ring. The mesh is sutured to the pubic tubercle space. Limitations of this technique include the possibility
ht
on either side and then secured in a continuous fashion of meshoma and pain requiring mesh explantation, mesh
along the shelving edge of the inguinal ligament inferiorly migration, and erosion of the mesh into adjacent organs/
until it is at least 1 cm lateral to the insertion of the internal structures.21-23
oblique muscle into Poupart ligament. Similarly, the mesh
is secured superiorly to the rectus sheath and subsequently PREPERITONEAL REPAIR
to the internal oblique aponeurosis with interrupted Preperitoneal mesh placement plays a central role in many
sutures. Two tails are created in the mesh by incising it hernia repair techniques, including those described by
from the lateral edge to create a slit that encircles the Nyhus-Condon, Wantz, Read, Rives, Stoppa, and Kugel.24
spermatic cord and reconstructs the internal ring. The The key component of the preperitoneal repair is place-
mesh tails encircling the cord are anchored in a fashion ment of a large mesh in the preperitoneal space between
that overlaps the superior and inferior tails in a manner the transversalis fascia and the peritoneum. The preperi-
that creates a new internal ring fitting snugly around toneal space is accessed anteriorly through the inguinal
the spermatic cord. This is accomplished by suturing the floor to allow for placement of the mesh. Subsequently,
tails together and tucking the ends of the tails under the the mesh is secured with interrupted sutures and serves
external oblique aponeurosis. Creation of this shutter valve as reinforcement of the transversalis fascia. The external
at the internal ring is a critical step for preventing indirect oblique is then closed in a continuous fashion. A limitation
hernia recurrence. The superior and inferior tails can of this approach is the need for blunt dissection in the
then be secured to the underlying internal oblique and preperitoneal space, which can lead to structural injury,
602 SECTION I Esophagus and Hernia
t
ne
k.
oo
A B
FIGURE 53B.4 Lichtenstein repair. (A) One of the sutures demonstrates the approximation of the inferior edge of the prosthesis to the
yb
inguinal ligament. The second suture will include the inferior surface of the superior tail and the inferior surface of the inferior tail just
lateral to the internal ring as well as the inguinal ligament to create a “shutter” valve. (B) The shutter valve has been completed and the
superior and medial surfaces have been sutured to the underlying internal oblique muscle and anterior rectus sheath, respectively. This
er
older illustration shows continuous suture on the superomedial border of the prosthesis, but interrupted sutures are now preferred by
most surgeons to minimize the incidence of nerve entrapment. (From Kurzer M, Belsham PA, Kark AE. The Lichtenstein repair for groin
hernias. Surg Clin North Am. 2003;83:1110.)
rg
su
://
tp
ht
hematoma formation, and potential scar tissue formation, The McVay repair is similar to the Bassini repair except
thus limiting future surgical dissection. for the use of Cooper ligament instead of the inguinal
ligament for the medial portion of the repair. The con-
BILAYER MESH REPAIR joined tendon is sutured to Cooper ligament from the
Bilayer mesh repair combines components of the pubic tubercle and extends along the ligament until as
Lichtenstein repair and the preperitoneal repair. The far as the edge of the femoral sheath. The final stitch
repair is performed by bluntly dissecting a pocket in to the Cooper ligament is known as the transition stitch
the preperitoneal space for placement of the mesh deep and includes the inguinal ligament and may include the
to the transversalis fascia with a superficial layer placed in medial aspect of the femoral sheath as well. This repair is
front of the transversalis fascia. Prefabricated mesh systems commonly used to address femoral hernias by narrowing
are available for this type of repair. Limitations of this the femoral ring, but it can cause considerable tension and
repair are similar to those that apply to all preperitoneal requires a relaxing incision to accomplish. The relaxing
repair techniques. incision is performed by incising the anterior rectus sheath
from the pubic tubercle cephalad for several centimeters
TISSUE (NONMESH) REPAIR along the fusion of the external oblique aponeurosis with
Numerous types of tissue repair are described in the surgi- the sheath.
cal literature. The most commonly used tissue repairs in
modern times are those of Shouldice, Bassini, and McVay. FEMORAL HERNIAS
t
Among the nonmesh repairs, the Shouldice technique is Femoral hernias are rare and typically seen in women. A
ne
preferred because it has the lowest associated recurrence femoral hernia occurs through the femoral canal, which
rate.25 Nonetheless, nonmesh repairs have a reported is bound by the inguinal ligament anteriorly, the pectineal
recurrence as high as 35% and have been shown to be ligament posteriorly, the femoral vein laterally, and the
k.
clearly inferior to tension-free mesh repair through a lacunar ligament medially. Typically a femoral hernia will
number of randomized controlled trials.26-28 produce a bulge below the inguinal ligament; however, it
oo
The Shouldice technique is an anterior approach may also present over the inguinal ligament. Femoral hernia
that involves division of all the layers of the floor of the repair can be performed by a preperitoneal approach,
inguinal canal with reduction of the hernia followed by Cooper ligament repair (McVay), or laparoscopically. The
yb
reconstruction of the inguinal canal. Once the transversalis essential components of a femoral hernia repair include
fascia is split from the internal ring to the pubic crest, dissection and reduction of the hernia sac and closure of
reconstruction of the canal is performed using a four-layer the defect either through approximation of the iliopubic
overlap technique and continuous suture. The repair tract to Cooper ligament or through the use of mesh.
er
starts at the pubic tubercle by approximating the iliopubic Unlike inguinal hernias, all femoral hernias should be
tract to the underside of the lateral edge of the rectus repaired once they are diagnosed due to the higher risk
rg
pubic tubercle, this suture line approximates the medial not limited to surgical site infection, urinary retention,
flap tissue to the shelving edge of the inguinal ligament. orchitis, seroma, hematoma, injury to the vas deferens,
tp
The internal oblique and transversus abdominis are then hydrocele, testicular descent, bowel or bladder injury,
approximated to the shelving edge of the inguinal liga- osteitis pubis, prosthetic complications, and wound
ment. The final suture line is then reversed; it runs laterally complications. Although some of these complications
ht
and secures the lower flap of the external oblique over are related to underlying disease processes, others are
the internal oblique in a similar fashion to the previous directly related to technical aspects of the repair.
suture line. The major limitation of this technique is that
it is technically difficult and hard to reproduce. SURGICAL SITE INFECTION
The Bassini repair strengthens the weakened inguinal Inguinal hernia repairs are generally considered clean
floor by suturing the conjoined tendon to the inguinal operations except for instances of contamination,
ligament from the pubic tubercle to the area of the inter- which may occur inadvertently or during the repair of
nal ring. This repair begins with the standard anterior strangulated hernias. The risk of surgical site infection
approach and subsequently divides the transversalis fascia is estimated to be up to 5% after open repair. The use
along the inguinal canal. Exposure of this space allows for of prophylactic preoperative antibiotics is controversial,
inspection of possible femoral hernias. Once the hernia sac with several studies concluding that there is no benefit
is high-ligated, reconstruction of the floor is performed by to this practice.29-31 Moreover, the use of prosthetic mesh
suturing the three layers of transversalis fascia, transversus does not confer a greater risk of infection or substantiate
abdominis, and internal oblique muscle to the inguinal the need for prophylaxis. Surgical site infections can
ligament. Classic descriptions of this technique include generally be managed with open drainage, local wound
an initial stitch of the three layers to the periosteum of care, and oral antibiotics. Mesh infection may lead to a
the pubic tubercle and the rectus sheath. Laterally, the chronically draining sinus tract and ultimately require
repair extends until closure of the internal ring. mesh explantation.
604 SECTION I Esophagus and Hernia
t
advocate the use of laparoscopic repair for recurrent
ne
1. Rosenberg J, Bisgaard T, Kehlet H, et al. Danish Hernia Database
hernias. Similarly, anterior repair may be best suited for recommendations for the management of inguinal and femoral
recurrent hernias performed in a preperitoneal fashion. hernia in adults. Dan Med Bull. 2011;58:C4243.
Recurrent hernias should almost always be managed with 2. Robinson A, Light D, Kasim A, Nice C. A systematic review and
k.
a mesh repair. meta-analysis of the role of radiology in the diagnosis of occult
inguinal hernia. Surg Endosc. 2013;27:11.
3. Fitzgibbons RJ Jr, Giobbie-Hurder A, Gibbs JO, et al. Watchful
CHRONIC PAIN (INGUINODYNIA)
oo
waiting vs. repair of inguinal hernia in minimally symptomatic men:
Nerve injury and chronic pain are often underrecognized a randomized clinical trial. JAMA. 2006;295:285.
and potentially debilitating complications of inguinal 4. O’Dwyer PJ, Norrie J, Alani A, Walker A, Duffy F, Horgan P. Observa-
tion or operation for patients with an asymptomatic inguinal hernia:
yb
hernia repair. Nerve injury can occur from traction, mesh a randomized clinical trial. Ann Surg. 2006;244:167.
or suture entrapment, electrocautery, and transection. 5. Simons MP, Aufenacker T, Bay-Nielsen M, et al. European Hernia
The nerves most commonly affected by open hernia Society Guidelines on the treatment of inguinal hernia in adult
repair are the ilioinguinal, iliohypogastric, and genital
er
patients. Hernia. 2009;13:343.
branch of the genitofemoral nerves. While chronic pain 6. Society for Surgery of the Alimentary Tract. SSAT patient care
guidelines. Surgical repair of groin hernias. J Gastrointest Surg.
is often associated with nerve injury, it may also result 2007;11:1228.
rg
from hernia recurrence, mesh-related problems, and 7. Scott NW, McCormack K, Graham P, Go PM, Ross SJ, Grant AM.
infection. Chronic pain is defined as pain lasting longer Open mesh versus non-mesh for repair of femoral and inguinal
than 3 months postoperatively and has been reported to hernia. Cochrane Database Syst Rev. 2002;(4):CD002197.
su
10. Earle DB, Mark LA. Prosthetic material in inguinal hernia repair:
The clinical presentation of chronic pain after hernia how do I choose? Surg Clin North Am. 2008;88:179.
repair can be heterogeneous and variable. Careful 11. Ge BJ, Huang Q, Liu LM, Bian HP, Fan YZ. Risk factors for bowel
evaluation of symptoms and physical findings should
ht
19. Reinpold WM, Nehls J, Eggert A. Nerve management and chronic 26. Nordin P, Bartelmess P, Jansson C, Svensson C, Edlund G. Randomized
pain after open inguinal hernia repair: a prospective two phase trial of Lichtenstein versus Shouldice hernia repair general surgical
study. Ann Surg. 2011;254:163. practice. Br J Surg. 2002;89:45-49.
20. Stylianidis G, Haapamaki MM, Sund M, Nilsson E, Nordin P. 27. Danielsson P, Isacson S, Hansen MV. Randomized study of Lichtenstein
Management of the hernia sac in inguinal hernia repair. Br J Surg. compared with Shouldice inguinal hernia repair by surgeons in
2010;97:415. training. Eur J Surg. 1999;165:49-53.
21. Chuback JA, Sing RS, Sills C, Dick LS. Small bowel obstruction 28. McGillicuddy JE. Prospective randomized comparison of the Shouldice
resulting from mesh plug migration after open inguinal hernia and Lichtenstein hernia repair procedures. Arch Surg. 1998;133:974-978.
repair. Surgery. 2000;127:475. 29. Sanchez-Manuel FJ, Seco-Gil JL. Antibiotic prophylaxis for hernia
22. Kingsnorth AN, Hyland ME, Porter CA, Sodergren S. Prospective repair. Cochrane Database Syst Rev. 2003;(2):CD003769.
double-blind randomized study comparing Perfix plug-and-patch 30. Tzovaras G, Delikoukos S, Christodoulides G, et al. The role of
with Lichtenstein patch in inguinal hernia repair: one year quality antibiotic prophylaxis in elective tension-free mesh inguinal hernia
of life results. Hernia. 2000;4:255-258. repair: results of a single-centre prospective randomized trial. Int J
23. Amid PK. Classification of biomaterials and their related complications Clin Pract. 2007;61(2):236-239.
in abdominal wall hernia surgery. Hernia. 1997;1:12-19. 31. Aufenacker TJ, van Geldere D, van Mesdag T, et al. The role of
24. Amid PK. Groin hernia repair: open techniques. World J Surg. 2005;29: antiobiotic prophylaxis in prevention of wound infection after Lich-
1046. tenstein open mesh repair of primary inguinal hernia: a multicenter
25. Amato B, Moja L, Panico S, et al. Shouldice technique versus other double-blind randomized controlled trial. Ann Surg. 2004;240:955-966.
techniques for inguinal hernia repair. Cochrane Database Syst Rev. 32. Towfigh S, Neumayer L. Inguinal hernia. In: Cameron JL, ed. Current
2012;(4):CD001543. Surgical Therapy. 11th ed. Philadelphia: Elsevier Saunders; 2014.
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
CHAPTER
A
side from ventral and inguinal hernias, there lumbar triangle, or Petit triangle, is an upright triangle
are less common hernias of the abdominal wall bordered by the iliac crest inferiorly, the external abdomi-
and pelvis that may come to the attention of a nal oblique muscle anterolaterally, and the latissimus dorsi
general surgeon from time to time. It is important for posteromedially. The floor is formed by the lumbodorsal
the surgeon to have a general knowledge of these defects fascia and transversalis muscle.8 A medially displaced
so that they may be included in the differential diagnosis latissimus dorsi muscle, alterations in the origin of the
and treated appropriately (and sometime expeditiously) external abdominal oblique muscle, and the presence
if needed. These include hernias of the lumbar area in of a Hartmann fissure at the vertex of the triangle may
the lower back, as well as obturator, sciatic, and perineal increase the likelihood of visceral protrusion through the
t
hernias found within the pelvis. In this chapter, we will inferior lumbar triangle (Fig. 54.3).8 Herniation occurs
ne
review the important anatomy, clinical presentation and more commonly in the superior triangle as it has a greater
evaluation, and surgical treatment of each one of these surface area, is a more vulnerable area of weakness (there
rare and unique entities. is only transversalis fascia at its lower margin), and is not
k.
penetrated by neurovascular bundles.7–9
In addition to the anatomic classification, lumbar hernias
LUMBAR HERNIA
oo
can further be categorized as congenital or acquired based
on etiology. Congenital defects of the lumbar region make
ANATOMY AND CLASSIFICATION up 10% to 20% of lumbar hernias.9,10 Although often
yb
Lumbar hernias are exceedingly rare, posterolateral, unilateral when presenting early in life, some patients
abdominal wall defects containing retroperitoneal fat may have bilateral hernias and develop symptoms in
or viscera. Lumbar hernias were first described several late adulthood as a result of progressive posterolateral
hundred years ago. Barbette suggested the entity of a abdominal muscle weakening.11–13 Congenital lumbar
er
lumbar hernia in 1672, but the first published case in hernias are most often associated with lumbocostovertebral
the medical literature was in 1731 by DeGarangeot.1 He syndrome, although it has been reported with vertebral
rg
reported the first incarcerated lumbar hernia on autopsy.2 anomalies, anal atresia, cardiac defects, tracheoesophageal
Twenty years later, Ravaton described the surgical reduction fistulas, renal anomalies, and limb defects (VACTERL
of a strangulated lumbar hernia.3 The French surgeon syndrome), congenital diaphragmatic hernia, and atrial
su
Jean-Louis Petit in 1774 was the first to describe an inferior septal defects, among other congenital malformations.13,14
lumbar hernia through the anatomic boundary now often Acquired lumbar hernias constitute the majority of
referred to as Petit triangle.4 More than two centuries after lumbar defects encountered clinically and can be subdi-
://
its first description, Grynfelt and Lesshaft independently vided into primary or secondary acquired hernias.9,10,15
reported visceral herniation through a superior lumbar Primary acquired hernias occur spontaneously, are often
tp
defect, now commonly known as a Grynfelt-Lesshaft small in size, are confined to the borders of the superior or
hernia.5,6 inferior lumbar triangles, and represent 55% of acquired
Lumbar hernias occur through a parietal wall defect lumbar hernias.15 Risk factors for the development of a
ht
in the lumbar region whose boundaries are the 12th rib primarily acquired lumbar hernia include advanced age,
superiorly, the iliac crest inferiorly, the erector spinae chronic malnutrition or debilitation, obesity, chronic
muscle medially, and the posterior border of the external cough, and previous wound infections or a history of
abdominal oblique laterally (Fig. 54.1). They can be sepsis.16 On the other hand, secondary acquired hernias are
classified anatomically into superior or inferior lumbar often the result of trauma or previous surgical procedures
hernias based on two well-defined areas of weakness. The in the lumbar region.9 These defects may be diffuse,
superior lumbar triangle is an inverted triangle bordered extending beyond the margins of the lumbar triangle.9,10
by the 12th rib superiorly, the internal abdominal oblique It is a rare complication of surgical procedures involving
muscle anterolaterally, and the quadratus lumborum flank incisions. Examples include open partial or com-
muscle posteromedially.7 The latissimus dorsi and the plete nephrectomy, adrenalectomy, and abdominal aortic
aponeurosis of the transversalis muscle form the roof and aneurysm repair. It may also occur at previous bone graft
the floor of the triangle, respectively. Areas of weakness donor sites.9,17 The iliac crest is a common donor site for
include the region immediately below the costal margin autogenous bone grafts given that it is easily accessible and
where the transversalis fascia is not reinforced by the supplies ample amounts of both cancellous and cortical
external abdominal oblique muscle, the area at which the bone.18 Herniation of intraabdominal contents through
12th dorsal intercostal neurovascular bundle penetrates the resulting bone defect is uncommon, occurring only
the fascia, and the area between the ligament of Henle in up to 5% of cases.18–23 Blunt trauma is another rare
and the inferior costal margin (Fig. 54.2).8 The inferior cause of secondary acquired lumbar hernias, with less
606
Lumbar, Pelvic, and Uncommon Hernias CHAPTER 54 606.e1
ABSTRACT
This chapter discusses less common hernias of the abdomi-
nal wall and pelvis, which rarely come to the attention
of a general surgeon. These include hernias of the
lumbar triangles in the lower back, namely the superior
lumbar hernia (or Grynfelt-Lesshaft hernia) and inferior
lumbar hernia (or Petit hernia). Obturator hernias,
characterized by the classic Howship-Romberg sign, will
also be discussed. In addition, the extremely rare sciatic
and perineal hernias will be reviewed.
KEYWORDS
Superior lumbar hernia, Grynfelt-Lesshaft hernia, inferior
lumbar hernia, Petit hernia, obturator hernia, Howship-
Romberg sign, sciatic hernia, perineal hernia
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
Lumbar, Pelvic, and Uncommon Hernias CHAPTER 54 607
t
Anatomic Relations of Lumbar Hernia
ne
k.
Bowel loop entering obturator foramen Obturator Hernia
FIGURE 54.1 Lumbar and obturator hernias. (From Yabara S, Rosenthal R. Lumbar, obturator, sciatic, and perineal hernias. In: Floch MH,
su
12th rib
ht
Aponeuroses of the
oblique and transverse m.
Quadratus lumborum m.
* FIGURE 54.2 Cross-sectional anatomy of the superior
Latissimus dorsi m.
lumbar triangle. The asterisk denotes the area where
Posteroinferior serratus m. a lumbar hernia may occur. (From Aguirre DA, Rivero
Thoracolumbar fascia OM, Martinez J. Normal anatomy of the abdominal
Erector spinae m. wall. In: Sahani DV, ed. Abdominal Imaging.
Philadelphia: Saunders; 2011:1439–1443.)
Aponeuroses of the
oblique and transverse m.
* Iliac crest
FIGURE 54.3 Cross-sectional anatomy of the inferior Latissimus dorsi m.
lumbar triangle. The asterisk denotes the area where Quadratus lumborum m.
a lumbar hernia may occur. (From Aguirre DA, Rivero
OM, Martinez J. Normal anatomy of the abdominal Thoracolumbar fascia
wall. In: Sahani DV, ed. Abdominal Imaging. Erector spinae m.
Philadelphia: Saunders; 2011:1439–1443.)
t
lipoma, rhabdomyoma, sarcoma and other malignant
ne
growths, abscess, hematoma, or a renal mass.15,26 Pain
is variable and can range from mild local discomfort
to severe, diffuse intestinal colic. 15 Depending on
k.
the contents of the hernia, the pain can travel down
the distribution of the sciatic nerve or be referred
oo
to the anterior abdomen, especially when incarceration
or panniculitis is present.15,17,27 Patients may also have
gastrointestinal complaints such as nausea, vomiting, R
and/or bloating. 17,28 When a mass is palpated in the
yb
lumbar region, it is often soft and fluctuates in size.
In addition, the mass may protrude with coughing
or bearing down and even produce bowel sounds on
er
ultrasound, this may be indicated by intraluminal gas, is placed in a lateral decubitus position contralateral to
which projects as an area of focal dense echogenicity the side of the hernia. General or spinal anesthesia may
with acoustic shadowing.28 be used based on the surgeon’s discretion. A generous
lumbar incision is made, and exploration of the hernia
TREATMENT is performed (Fig. 54.5A).26 The edges of the fascial
The natural progression of lumbar hernias is a gradual defect are defined circumferentially. After the hernia
increase in size over time.15,26,27 Despite this, approximately sac is identified and its contents reduced, the sac may
25% of patients will present with incarcerated bowel, be excised or inverted.26 If the defect is small, it may be
and 10% to 18% will demonstrate evidence of strangula- repaired primarily with nonabsorbable sutures (see Fig.
tion.9,15,27,31 Given the risk of associated complications 54.5B–D), although larger defects require a mesh-enforced
and the increased complexity of repairing large hernias, repair with reapproximation of the overlying muscle
surgical intervention upon diagnosis of a lumbar hernia is layers.9,10,26,31–33,36 Nonabsorbable mesh may be secured in an
prudent when the patient’s medical condition permits.7,15,27 extraperitoneal position with 3 to 5 cm of overlap.7,10,12,26,34,36
The repair of a lumbar hernia is challenging, and various In a case series by Cavallaro et al., seven patients with
techniques of repair have been described, including simple both Petit- and Grynfelt-type hernias underwent open
repair, musculofascial flaps, free grafts, and repair using repair with placement of an extraperitoneal mesh.10 They
synthetic mesh.9,16,25,29,32 More recent studies have evalu- reported no recurrences over a median of 25 months.
ated the efficacy of laparoscopic or retroperitoneoscopic Synthetic mesh was secured to the surrounding muscles
t
repair.15,27,33–35 Still, the rarity of lumbar hernias and lack and the 12th rib or iliac crest for superior and inferior
ne
of sufficient data preclude a standardized approach or defects, respectively. Solaini et al. described a separate
optimal timing in the management of this condition.15,27 case of open repair of a superior lumbar hernia with dart
Lumbar hernias are traditionally approached through an mesh (Bard Mesh Dart, a small monofilament knitted
k.
open or anterior technique. In this approach the patient polypropylene) fixed to the 12th rib.29 The mesh was fixed
oo
Transverse External
fascia oblique
yb
Latissimus dorsi
er
Incision
rg
Alternative
incision
su
A
Flap of fascia from
://
Crest of
ilium
B
FIGURE 54.5 The Dowd operation for lumbar hernia. (A) Line
of incision. (B) Turning up a flap of the fascia lata and
aponeurosis of the gluteus maximus and medius muscles
and suturing it to the lumbar fascia and external oblique
Flap of fascia from and latissimus dorsi muscles. (C) The flap sutured.
C latissimus dorsi (D) Closing the remaining gap with a flap of fascia from
the latissimus dorsi. (From Watson LF. Hernia. 3rd ed.
D St. Louis: Mosby; 1948.)
610 SECTION I Esophagus and Hernia
medially to the quadratus lumborum, externally to the sutures and tacks (Fig. 54.7).27 The use of bone anchors
internal oblique, and inferiorly to the serratus posterior. has been described to secure mesh to the iliac crest in
The patient was discharged on the first postoperative day inferior lumbar hernias.34 In a study of seven patients who
and did not demonstrate a recurrence at 11 months of underwent laparoscopic transabdominal incisional lumbar
follow-up. The open or anterior approach may be prefer- herniorrhaphy with polypropylene mesh reinforcement,
able when the lumbar defect is small and well defined there were no complications or recurrences over a mean
with adequate surrounding musculoaponeurotic tissue.10 follow-up of 34 months.27 The laparoscopic transabdominal
Interestingly, simple primary closure of lumbar hernias approach may also be preferable in patients with previous
without mesh has demonstrated acceptable outcomes over lumbar surgery.27
a short follow-up period.7 However, most authors agree Unlike the transabdominal laparoscopic technique, the
that mesh reinforcement is an important component in retroperitoneoscopic (or totally extraperitoneal [TEP])
ensuring the durability of the repair, particularly for larger repair avoids penetration into the abdominal cavity.26,37,38
hernia defects.10,15,28,29,31,36,37 A flank incision is made and a balloon dissector is used
The alternative to open repair is a laparoscopic repair. to create a retroperitoneal plane and adequate operat-
The laparoscopic technique may be performed through ing space.35,37,38 Nonabsorbable mesh is placed in an
a transabdominal or retroperitoneoscopic approach. extraperitoneal position. Habib reported no recurrence
In the transabdominal approach, patient positioning is out to 2 years in a patient who underwent tension-free
similar to that of the open approach. A lumbar roll may retroperitoneoscopic repair of a superior lumbar hernia
t
be placed to increase the distance between the inferior with polypropylene mesh.38
ne
rib margin and the iliac crest.27 Dissection occurs along The management of TLH deserves special mention.
the white line of Toldt. The colon is mobilized and the The timing of the repair (early vs. delayed) and the
borders of the underlying lumbar defect are identified surgical approach to those patients with TLH require
k.
(Fig. 54.6).26 The hernia sac and its contents are reduced. individualization based on the patient’s clinical status,
Mesh may then be secured with transabdominal full- presence of concomitant injuries, mechanism of injury,
oo
thickness bites using a combination of nonabsorbable size of hernia, patient factors, and radiologic findings.
Early repair is recommended in hemodynamically stable
patients without major associated injuries.25 In patients with
yb
extensive tissue loss and contamination, it is preferable
to defer hernia repair until life-threatening injuries such
as visceral injury have been addressed.9 A transperitoneal
approach is recommended by some authors in TLH to
er
containing incarcerated colon (From Varban O. Lumbar hernia patch can be used for the coverage of large defects. It
after breast reconstruction. Int J Surg Case Rep. 2013, 4[10]: can also be used in the face of contamination because it
is biologically inactive leading to decreased inflammation
ht
869–871.)
and adhesion formation, has good tensile strength, and
may be resistant to infection.9
Although the laparoscopic transabdominal and retro-
peritoneoscopic approaches require less tissue dissection
and provide an improved anatomic view in comparison to
the open technique, there is no consensus regarding the
preferred approach to lumbar hernias.27,38 Moreno-Egea
et al. performed the only reported comparative study of
open versus laparoscopic lumbar hernia repair.33 They
demonstrated that a laparoscopic technique provides
advantages such as decreased morbidity and length of stay,
with quicker return to normal activity and no significant
increase in cost.33 A classification system has been proposed
to help guide the management of lumbar hernias, which
classifies them into four categories based on six individual
FIGURE 54.7 Completed laparoscopic mesh repair of a lumbar criteria: size, location, contents within the hernia, the
hernia. (From Varban O. Lumbar hernia after breast presence of muscular atrophy, origin, and a history of
reconstruction. Int J Surg Case Rep. 2013;4[10]:869–871.) recurrence.8 Although this classification system has yet to
Lumbar, Pelvic, and Uncommon Hernias CHAPTER 54 611
be validated, it may provide a basis to aid in the clinical external obturator muscle along the posterior division
management of lumbar hernias. of the obturator nerve. In this variation the sac can be
found situated posterior to the adductor brevis muscle.
Lastly, the least frequent pathway occurs with the entire
OBTURATOR HERNIA sac penetrating and lying in between the internal and
external obturator muscles.
ANATOMY AND CLASSIFICATION
The obturator hernia was first described at the Royal CLINICAL PRESENTATION AND DIAGNOSIS
Academy of Sciences of Paris by Pierre Roland Arnaud Often referred to as “the little old lady’s hernia,” obturator
de Ronsil in 1724.39 More than 100 years later, Obre hernias are usually discovered in thin, elderly females in
performed the first successful repair.40 Currently, these their seventh to eighth decade of life. Obturator hernias
rare defects represent only 0.05% to 1.4% of all hernias are 6 to 9 times more common in women.42,48,49 This is
encountered by clinicians.41–47 An obturator hernia occurs likely secondary to differences in pelvic bone anatomy and
when intraabdominal viscera or extraperitoneal tissue angulation given females have a larger and wider pelvis
project through the obturator canal, an osteofibrous with a horizontally inclined obturator canal.49 Malnutri-
tunnel that courses from the pelvis to the proximal thigh tion and a thin body habitus are also significant risk
and is penetrated by the obturator neurovascular bundle factors.40,50 Emaciation results in a loss of preperitoneal
(Figs. 54.8 and 54.9; see also Fig. 54.1). The foramen is fat and connective tissue normally concealing the obtura-
t
created by the superior pubic ramus superiorly, the body tor canal.40,43 Conditions that increase intraabdominal
ne
and inferior ramus of the pubic bone interiorly, and the pressure and promote laxity of the pelvic floor, such as
ramus and body of the ischium inferiorly.40 Loss of the pregnancy and multiparity, may potentially predispose
protective extraperitoneal areolar and adipose tissue overly- patients to herniation.43 Of note, obturator hernias occur
k.
ing the obturator canal creates various anatomic pathways predominantly in the right pelvis due to the fact that the
through which herniation can occur (Fig. 54.10).42 In sigmoid colon often covers the left obturator foramen.51
oo
the most common pathway, the hernia sac protrudes The hernia sac usually contains small bowel, although
through the external opening of the canal along the other organs, including the colon, appendix, Meckel
anterior division of the obturator nerve, with the sac lying diverticulum, bladder, and adnexa have been reported.51
yb
beneath the pectineus muscle of the thigh. An alternative The natural development of an obturator hernia occurs
pathway is between the upper and middle fasciculi of the in three successive stages.42 The first stage begins with
er
rg
Obturator
FIGURE 54.8 A diagram of normal pelvic floor
su
Iliopsoas
Aponeurosis of Pectineus
adductor brevis
Caps artic.
Adductor
longus
Obturator membrane
Peritoneum
t
ne
Ischium Adductor magnus muscle
k.
FIGURE 54.10 Side view of the anatomy of the obturator canal. A hernia can follow the path of either the anterior or posterior branch of
oo
the obturator nerve. (From Stamatiou D, Skandalakis LJ, Zoras O, et al. Obturator hernia revisited: surgical anatomy, embryology,
diagnosis, and technique of repair. Am Surg. 2011;77:1147–1156.)
yb
the bulging of preperitoneal fat through the obturator
foramen. In the second, or developmental stage, prolonga-
tion of the peritoneum and formation of a true hernia sac
er
t
and further studies are needed to elucidate the role of
ne
(arrow) protruding outside the right obturator foramen. (From Kim
JJ, Jung H, Oh SJ, et al. Laparoscopic transabdominal laparoscopic, extraperitoneal, and inguinal approaches
preperitoneal hernioplasty of bilateral obturator hernia. Surg in the management of an obturator hernia.42
Laparosc Endosc Percutan Tech. 2005;15:106.) After the obstructed segment of bowel is identified,
k.
reduction may be performed by simple gentle traction,
if possible. In some instances, reduction of the herniated
oo
clinical benefit of preoperative CT is controversial. Several contents may be challenging and require incision of the
previously published studies have reported a reduced obturator membrane, pubic osteotomy, or use of the
rate of complications, need for small bowel resection, “water pressure method” when incarcerated intestine is
and mortality with the use of CT imaging.42,51,56 However, present.42,64 In the water pressure method, an 8-French
yb
more recent studies have been unable to demonstrate Nelaton catheter is inserted on the side of the hernia into
an added benefit with preoperative CT.46,53,59 The role of the obturator foramen and water is injected into the hernia
CT may be important as it provides a prompt diagnosis sac to promote intestinal reduction via water pressure.64
er
of an obturator hernia and allows surgical planning, After reduction of the hernia contents is achieved, the
although it should never delay surgical intervention as defect may be repaired using various methods. Small
rg
any delay in treatment will significantly increase mortal- defects can be repaired by reapproximation of the hernia
ity.56,57 Ultimately, CT is an incredibly valuable tool, but sac.40,51 Larger defects require fascial closure (i.e., suturing
the diagnosis of an obturator hernia is best made in the the pectineus muscle to the periosteum of the pubic
su
operating room. The utility of other diagnostic modalities, bone), coverage with local tissue (the round ligament,
such as contrast herniography and ultrasonography, have uterine fundus, ipsilateral ovary, or bladder), or prosthetic
been described but their accuracy and relevance have yet mesh reinforcement.40,42,53,56 Although prosthetic mesh is
://
and presents with an acute intestinal obstruction.40,42 As years were significantly reduced (0% vs. 22%) with the use
a result, surgeons may prefer to conservatively manage of prosthetic mesh.57 Nevertheless, prospective long-term
such patients, given their inherently high surgical risk.56 studies are needed to compare the efficacy and durability
Despite this, urgent surgical intervention is crucial in of the various repair techniques described.
the management of obturator hernias because the rate
of strangulation reaches 50% to 75% and may increase
further with any delay in treatment.42,51 The classic surgical SCIATIC HERNIA
approach is through an exploratory laparotomy, although
in recent years, inguinal and laparoscopic extraperitoneal ANATOMY AND CLASSIFICATION
techniques have been described.40,61–63 Sciatic hernias are the rarest of the pelvic floor hernias.
Exploration by laparotomy allows a complete assessment First described by Papen in 1750, only slightly more than
of intraabdominal viscera and pathology and evaluation 100 cases have been reported in the literature since then.69
of both obturator canals and facilitates bowel resection The sciatic notch is located between the posterior inferior
if required.42 It is performed through a lower midline iliac spine and ischial spine. Three distinctive anatomic
incision and preferred in patients with suspected intestinal spaces are created within the sciatic notch, and visceral
perforation, strangulation, or peritoneal inflammation.40 herniation can occur through any one of these potential
Laparoscopic transabdominal repair for obturator hernias openings. The greater and lesser sciatic foramina are
has demonstrated good outcomes in both elective and formed by the sacrospinous and sacrotuberous ligaments,
614 SECTION I Esophagus and Hernia
Gluteus medius
muscle
Post. sup. iliac spine
Piriformis
Greater sciatic muscle
foramen
Sacrospinous
ligament
Suprapiriform
space
Sacrotuberous
ligament
Infrapiriform
Lesser sciatic space
foramen Tuberosity
t
of ischium
Sacrotuberous
ne
ligament
FIGURE 54.13 Posterolateral view of the pelvis showing the greater Sacrospinous
k.
and lesser sciatic foramina and their ligamentous and osseous ligament
boundaries. (From Watson LF. Hernia. 3rd ed. St. Louis: CV
Mosby; 1948.) Below
oo
sacrospinous
ligament
yb
respectively (Fig. 54.13; see also Fig. 54.8). The greater Posterior femoral cutaneous nerve
sciatic foramen is further subdivided by the piriformis
Sciatic nerve
muscle into the suprapiriform space and the infrapiriform
space (Fig. 54.14). The precise mechanism by which a
er
and gluteus maximus muscle are associated pathologic (1), through the infrapiriform space (2), or below the sacrospinous
findings.70,71 Still, the etiology of sciatic hernias is likely ligament (3). Also shown are the gluteus medius muscle (4),
more complex and involves interplay between different
su
The clinical diagnosis of a sciatic hernia is extremely dif- of the world literature (1900–2008). Am J Surg. 2010;199[1]
ficult because symptoms are usually limited to nonspecific 52–59.)
tp
or pain radiating down the posterior thigh, may be the Differential diagnoses include a gluteal abscess, arterial
presenting symptom in a patient with compression of aneurysm, or a lipoma.74
the sciatic nerve by the hernia sac. Alternatively, sciatic The clinical diagnosis of a sciatic hernia requires high
hernias may present with acute intestinal obstruction or clinical suspicion and often adjunctive diagnostic imaging.
strangulation. An interesting variant described in the The diagnosis of a sciatic hernia can be made using various
literature is a ureteric sciatic hernia.73 In this type of hernia, imaging modalities including barium studies, transgluteal
the presence of a ureteral segment or bladder within the ultrasound, CT, and MRI.69 The preferred diagnostic imaging
hernia sac can lead to obstructive uropathy. Small bowel, is CT (Fig. 54.15) because it allows for a rapid diagnosis,
ovarian tissue, colon, and Meckel diverticulum may also delineates regional anatomy, and may provide informa-
protrude through the sciatic foramen. tion regarding intestinal viability. In the case of a ureteric
There are no pathognomonic findings on physical sciatic hernia, diagnosis is traditionally achieved with an
exam that can facilitate the diagnosis of a sciatic hernia. excretory or retrograde urogram, which will demonstrate
Identification of a bulge or protrusion is exceedingly rare a pathognomonic “curlicue ureter” (Fig. 54.16).75
given the location of the hernia and the fact that the
gluteus maximus muscle is disproportionately large and TREATMENT
overlies the sciatic foramen. Therefore viscera protruding Sciatic hernias are treated surgically. The surgical approach
through the sciatic notch would be difficult to appreciate may be abdominal transperitoneal, abdominal extraperi-
unless fairly large in size. Pelvic or rectal examination toneal, gluteal, or a combination of multiple techniques
Lumbar, Pelvic, and Uncommon Hernias CHAPTER 54 615
t
incarcerated sciatic hernia: ultrasound diagnosis. Br J Radiol.
ne
2002;75:381.)
PERINEAL HERNIAS
k.
ANATOMY AND CLASSIFICATION
Perineal hernias are rare hernias involving the pelvic
oo
floor. The first case of a perineal hernia was documented
by De Garangeot in 1743, and the first surgical repair
was reported by Moscowitz in 1916.77 Perineal hernias
yb
are initially classified as congenital or acquired. Purely
congenital perineal hernias are extremely rare, and only
nine cases have been documented in the literature.77
Acquired perineal hernias are classified as either primary
er
Bulbospongiosus m.
Ischiocavernosus m.
Deep transverse perineal m.
(urogenital diaphragm)
1
Superficial transverse
perineal m.
2
External anal sphincter m.
3
Levator ani m.
Gluteus maximus m.
Coccygeus m.
t
and perineocele. In: Raz S, ed. Female Urology. Philadelphia: sac
ne
Saunders Elsevier; 2008:743–750. Modified from Cali RL, Pitsch
RM, Blatchford GJ, Thorson A, Christensen MA. Rare pelvic floor FIGURE 54.18 Anterior perineal hernia. When the hernia descends
hernias. Dis Colon Rectum. 1992;35:604.) only into the posterior portion of the labium major, it is known as
a pudendal or vaginolabial hernia. (From Watson LF. Hernia. 3rd
k.
ed. St. Louis: Mosby; 1948.)
oo
pubococcygeus) or in a space between the levator ani and
the coccygeus muscle. Most posterior perineal hernias are perineal hernias. If urinary symptoms are present, further
secondary, and the true incidence of primary posterior work-up with cystourethrography and possibly cystoscopy is
yb
perineal hernias is unknown. Posterior hernias can be recommended. Plain radiographs or contrast studies (i.e.,
found in either sex, but they are 3 to 5 times more common barium enema) may show small or large bowel within the
in women.81 Secondary posterior perineal hernias are hernia. With the availability of CT today, most clinicians
incisional hernias through the pelvic floor that typically will obtain a CT to accurately identify the contents within
er
present within the first postoperative year. the hernia, as well as to assess the size of the pelvic floor
defect. Furthermore, pelvic MRI is quickly becoming a
CLINICAL PRESENTATION AND DIAGNOSIS
rg
tinence.82 On physical exam, patients with anterior perineal with reduction and possible resection of the hernia sac
hernias have a palpable mass in the posterior aspect of and primary closure of the defect. It is important to
tp
the labia (Fig. 54.18). This needs to be distinguished from remember that there is not always a clearly defined hernia
other similar conditions, such as a Bartholin gland abscess, sac because extraperitoneal structures can herniate through
labial cysts, lipoma, hematoma, or inguinal hernia. An the pelvic floor defect. In addition, the size of the defect
ht
anterior perineal hernia can usually be reduced into the can be extensive, especially following pelvic exenteration.80
pelvic floor inferiorly or below the pubic ramus, whereas Repair of such a large defect usually necessitates advanced
a true inguinal hernia will reduce by passing over or reconstruction of the pelvic floor using mesh, free fascial
superior to the pubic ramus. grafts, or pedicled muscle flaps.86–88
Patients with a posterior perineal hernia are usually Using a traditional open approach, the perineal hernia
asymptomatic, but when it causes symptoms, they usually defect can be repaired via an abdominal, perineal, or
complain of a perineal mass that produces pain and dis- a combined approach. Laparoscopic repair with mesh
comfort, especially upon sitting. Physical exam will reveal a reinforcement has also been described and is becoming
soft, reducible mass between the anus and ischial tuberosity more common.89,90 To date, there are no controlled trials or
or occasionally ventral to the gluteus maximus muscle.83 studies that have shown any approach to be superior to the
Nausea, vomiting, and other signs of intestinal obstruc- others. The choice of approach is best determined by the
tion are rarely seen in perineal hernias due to the usually size of the defect, the patient’s comorbidities and previous
wide pelvic floor defect, as well as the laxity of the tissues. surgical history, the condition of the pelvic tissues, and the
However, bowel obstruction and perineal skin breakdown experience and comfort level of the operating surgeon.
have been reported in perineal hernias that occur following The abdominal approach via laparotomy allows for
pelvic surgery.84,85 excellent exposure of the pelvic floor defect and hernia sac,
A thorough history and physical exam focusing on the especially for those cases where the defect is large requir-
vaginal, perineal, and rectal exams is important for all ing reconstruction with mesh.86,91 The open abdominal
Lumbar, Pelvic, and Uncommon Hernias CHAPTER 54 617
t
ne
A B
k.
oo
yb
er
rg
su
://
tp
C D
ht
I230 I230
FIGURE 54.19 Coronal T2-weighted magnetic resonance images. (A) Rest image showing a midline rectum (arrow) in a patient with
obstructed defecation. (B) Strain image demonstrating the rectum (arrows) herniating through a defect in the right levator ani complex.
(C) Postoperative rest image after a combined abdominoperineal approach to simple levator hernia repair with no absorbable sutures.
(D) The rectum remains midline after hernia repair, and the patient’s symptoms of obstructed defecation were resolved. (From Kaufman
HS, Buller JL, Thompson JR, et al. Dynamic pelvic MR imaging and cystocolpoproctography alter surgical management of pelvic floor
disorders. Dis Colon Rectum. 2001;44:1575; discussion 1584.)
approach is necessary in cases involving strangulation to larger defects (as seen following pelvic exenteration)
allow for possible resection of the hernia contents. Patients require more complex reconstruction usually with mesh.
should be positioned in a modified lithotomy position The perineal approach is the preferred initial approach,
to allow access to the perineum if needed. In addition, especially for small defects or otherwise straightforward
placing the patient in some degree of Trendelenburg will cases. Advantages to this approach include less morbidity
allow the viscera to fall out of the pelvis and therefore to the patient, as well as the ability to remove any excess or
optimize visualization. In general, after obtaining adequate redundant skin accompanying larger hernias.86 A notable
exposure, the hernia sac is identified, dissected out, and disadvantage is limited exposure, making it difficult to lyse
reduced. Small defects can be repaired primarily, whereas adhesions as well as to evaluate for recurrent cancer or
618 SECTION I Esophagus and Hernia
t
ne
k.
oo
A B C
yb
FIGURE 54.20 Site-specific rectocele repair of a distal defect with separation of the rectovaginal fascia from the perineal body.
(A) Dissection of a discrete defect. (B) Repair of the defect to the perineal body. (C) Onlay dermal graft attached to the levators laterally,
the rectovaginal fascia apically, and the perineal body distally. (From Kohli N, Miklos JR. Dermal graft-augmented rectocele repair.
er
visceral injuries. It may also be difficult to properly fixate of recurrent perineal hernias following previous repair
any mesh used in the reconstruction, and recurrences as with mesh has also been described.96
high as 23% have been reported with this approach.92 Despite the various approaches, recurrence rates follow-
su
For the perineal approach (Fig. 54.20A–C) the patient ing perineal hernia repair remain a significant concern.
is generally positioned similar to that used for an open A recent study looked at 40 patients who underwent
procedure (lithotomy and Trendelenburg position). The various techniques of perineal hernia repair following
://
skin is carefully incised over the hernia sac. For anterior APR. A primary recurrence was noted in 13 patients, with
perineal hernias, this incision is made over or along a second recurrence in 3 patients.97
tp
is then primarily repaired in layers using nonabsorbable Lumbar and pelvic hernias are unique hernias that are
suture with or without mesh reinforcement. A combined seldom seen in a surgeon’s clinical practice. Presenting
or abdominoperineal approach provides the benefits of symptoms are often vague and underappreciated, which
each approach individually but is generally associated makes lumbar and pelvic hernias difficult to diagnose.
with increased morbidity. However, the increased availability of CT and MRI has
Laparoscopic transabdominal repairs of perineal hernias made it easier to make the diagnosis. These imaging
are becoming more common as more experience is gained modalities also allow clinicians to completely evaluate
in the repair of these rare defects. The well-known benefits the size of the anatomic defect and the contents within
of less postoperative pain and quicker recovery have been the hernia sac. As with most hernias throughout the
documented following the repair of these pelvic hernias.93 body, they can be repaired successfully via an open or
Laparoscopic repair almost always uses reinforcement laparoscopic approach, either with or without mesh
with mesh, and repairs with both synthetic and biologic reinforcement.
mesh have been performed successfully.94 Combined
laparoscopic approaches have also been described in the REFERENCES
literature.89,95 The hernia sac is generally mobilized and
1. Jeannel M. La hernie lombaire. Arch Prov Chir Paris. 1903;11:389-418.
reduced laparoscopically, with subsequent primary suture 2. DeGarangeot RJC. Traite des Operations de Chirurgie. 1731;1:369.
repair of the defect performed via an open perineal 3. Cavallaro G, Sadighi A, Paparelli C, et al. Anatomical and surgical
approach. Even successful redo-laparoscopic mesh repair considerations on lumbar hernias. Am Surg. 2009;75(12):1238-1241.
Lumbar, Pelvic, and Uncommon Hernias CHAPTER 54 619
4. Petit JL. Traité des maladies chirurgicales, et des operations qui 35. Grauls A, Lallemand B, Krick M. The retroperitoneoscopic repair
leur convenient. Didot. 1774;2:256. of a lumbar hernia of Petit. Case report and review of literature.
5. Grynfelt J. Quelque mots sur la hernie lombaire. Montpellier Méd. Acta Chir Belg. 2004;104:330-334.
1866;16:323. 36. Carbonell AM, Kercher KW, Sigmon L, et al. A novel technique of
6. Lesshaft P. Die lumbal gegend. Anat Chirurgischer. Hinsicht Arch f lumbar hernia repair using bone anchor fixation. Hernia. 2005;9:22-25.
Anat u Physiol u Wissensch Med Leipzig. 1870;37:264. 37. Arca MJ, Heniford BT, Pokorny R, et al. Laparoscopic repair of
7. Zhou X, Nve JO, Chen G. Lumbar hernia: clinical analysis of 11 lumbar hernias. J Am Coll Surg. 1998;187:147-152.
cases. Hernia. 2004;8:260-263. 38. Habib E. Retroperitoneoscopic tension-free repair of lumbar hernia.
8. Moreno-Egea A, Baena EG, Calle MC, Martínez JA, Albasini JL. Hernia. 2003;7:150-152.
Controversies in the current management of lumbar hernias. Arch 39. Gray SW, Skandalakis JE, Soria RE, et al. Strangulated obturator
Surg. 2007;142:82. hernia. Surgery. 1974;75:20-27.
9. Esposito TJ, Federak I. Traumatic lumbar hernia: case report and 40. Stamatiou D, Skandalakis LJ, Zoras O, et al. Obturator hernia
literature review. J Trauma. 1994;37:123-126. revisited: surgical anatomy, embryology, diagnosis, and technique
10. Cavallaro G, Sadighi A, Miceli M, Burza A, Carbone G, Cavallaro of repair. Am Surg. 2011;77(9):1147-1157.
A. Primary lumbar hernia repair: the open approach. Eur Surg Res. 41. Kulkarni SR, Punamiya AR, Naniwadekar RG, et al. Obturator hernia:
2007;39:88-92. a diagnostic challenge. Int J Surg Case Rep. 2013;4:606-608.
11. Hickey M, Buick RG. Bilateral congenital lumbar hernia. Ir J Med 42. Losanoff JE, Richman BW, Jones JW. Obturator hernia. J Am Coll
Sci. 1982;151:388-389. Surg. 2002;194(5):657-663.
12. Lichtenstein IL. Repair of large diffuse lumbar hernias by an 43. Bjork KJ, Mucha P, Cahill DR. Obturator hernia. Surg Gynecol Obstet.
extraperitoneal binder technique. Am J Surg. 1986;151:501-504. 1988;167:217-222.
13. Vagholkar K, Dastoor K. Congenital lumbar hernia with lumbocos- 44. Lo CY, Lorentz TG, Lau PWK. Obturator hernia presenting as small
tovertebral syndrome: a case report and review of the literature. bowel obstruction. Am J Surg. 1994;167:396-398.
t
Case Rep Pediatr. 2013;2013:532910. http://www.ncbi.nlm.nih.gov/ 45. Skandalakis LJ. Obturator Hernia. 5th ed. New York: Lippincott
ne
pubmed/24159401. Accessed June 2016. Williams and Wilkins; 2002.
14. Harris K, Dorn C, Bloom B. Lumbocostovertebral syndrome with 46. Yokoyama Y, Yamaguchi A, Isogai M, et al. Thirty-six cases of obturator
associated VACTERL anomalad: a neonatal case report. J Perinatol. hernia: does computed tomography contribute to postoperative
2009;29:826-827. outcome? World J Surg. 1999;23:214-216.
k.
15. Heinford BT, Iannitti DA, Gagner M. Laparoscopic inferior and 47. Maharaj D, Maharaj S, Young L, et al. Obturator hernia repair—a
superior lumbar hernia repair. Arch Surg. 1997;132:1141-1144. new technique. Hernia. 2002;6(1):45-47.
16. Sutherland RS, Gerow RR. Hernia after dorsal incision into lumbar 48. Cali RL, Pitsch RM, Blatchford GJ, et al. Rare pelvic floor hernias:
oo
region: a case report and review of pathogenesis and treatment. report of a case and review of the literature. Dis Colon Rectum.
J Urol. 1995;153:382. 1992;35:604-612.
17. Stamatiou D, Skandalakis JE, Skandalakis LE, Mirilas P. Lumbar 49. Temple DF, Miller RE. Incarcerated obturator hernia: two case reports
yb
hernia: surgical anatomy, embryology and technique of repair. Am and review of the literature. J Natl Med Assoc. 1980;72:513-515.
Surg. 2009;75:202. 50. Tateno Y, Adachi K. Sudden knee pain in an underweight, older
18. Danikas D, Theodorou SJ, Stratoulias C, et al. Hernia through an iliac woman: obturator hernia. Lancet. 2014;384:206.
crest bone-graft donor site. Plast Reconstr Surg. 2003;20(131):61-63. 51. Rodriguez-Hermosa JI, Codina-Cazador A, Maroto-Genover A, et al.
er
19. Prabhu R, Kumar N, Shenoy F. Iliac crest bone graft donor site hernia: Obturator hernia: clinical analysis of 16 cases and algorithm for its
not so uncommon. BMJ Case Rep. 2013;http://casereports.bmj.com/ diagnosis and treatment. Hernia. 2008;12:289-297.
content/2013/bcr-2013-010386.long. Accessed June 2016. 52. Yokoyama T, Munakata Y, Ogiwara M, et al. Preoperative diagnosis
rg
20. Oldfield MD. Iliac hernia after bone grafting. Lancet. 1945;245(6357): of strangulated obturator hernia using ultrasonography. Am J Surg.
810-812. 1997;174:76-78.
21. Kaushik R, Attri AK. Incisional hernia from iliac bone grafting 53. Nasir BS, Zendejas B, Ali SM, et al. Obturator hernia: the Mayo
su
sites—a report of two cases. Hernia. 2003;7(4):227-228. Clinic experience. Hernia. 2012;16(3):315-319.
22. Pyrtek LJ, Kelly CC. Management of herniation through large iliac 54. Nazarian S, Narayanan A, Chang S. Diagnosis of an obturator
bone defects. Ann Surg. 1960;152:998-1003. hernia by CT. BMJ Case Rep. 2015;http://casereports.bmj.com/
23. Auleda J, Bianchi A, Tibau R, Rodriguez-Cano O. Hernia through iliac content/2015/bcr-2015-212239.long. Accessed June 2016.
://
crest defects. A report of four cases. Int Orthop. 1995;19(6):367-369. 55. Hodgins N, Cieplucha K, Conneally P, et al. Obturator hernia: a case
24. Burt BM, Afifi HY, Wantz GE, Barie PS. Traumatic lumbar hernia: report and review of the literature. Int J Surg Case Rep. 2013;4:889-892.
report of cases and comprehensive review of the literature. J Trauma. 56. Kammori M, Mafune K, Hirashima T, et al. Forty-three cases of
tp
65. Ng DCK, Tung KLM, Tang CN, Li M. Fifteen-year experience in 81. Poon FW, Lauder JC, Finlay IG. Perineal herniation. Clin Radiol.
managing obturator hernia: from open to laparoscopic approach. 1993;47:49.
Hernia. 2014;18:381-386. 82. Anderson WR. Pudendal hernia. Unusual cause of labial mass. Obstet
66. Yokoyama T, Kobayashi A, Kikuchi T, Hayashi K, Miyagawa S. Gynecol. 1968;32:802.
Transabdominal preperitoneal repair for obturator hernia. World J 83. Cali RL, Pitsch RM, Blatchford GJ, Thorson A, Christensen MA.
Surg. 2011;35(10):2323-2327. Rare pelvic floor hernias. Dis Colon Rectum. 1992;35:604.
67. Karashima R, Kimura M, Taura N, Shimokawa Y, Nishimura T, Baba 84. Ego-Aguirre E, Spratt JS, Butcher HR, Bricker EM. Repair of perineal
H. Total extraperitoneal approach for incarcerated obturator hernia hernia developing subsequent to pelvic exenteration. Ann Surg.
repair. Hernia. 2016;20(3):479-482. 1964;159:66.
68. Shapiro K, Patel S, Choy C, Chaudry G, Khalil S, Ferzli G. Totally extra- 85. Kelly AR. Surgical repair of post-operative perineal hernia. Aust N
peritoneal repair of obturator hernia. Surg Endosc. 2004;18(6):954-956. Z J Surg. 1960;29:243.
69. Losanoff JE, Basson MD, Gruber SA, Weaver DW. Sciatic hernia: 86. Abdul Jabbar AS. Postoperative perineal hernia. Hernia. 2002;6:188.
a comprehensive review of the world literature (1900–2008). Am J 87. Frydman GM, Polglase AL. Perineal approach for polypropylene
Surg. 2010;199:52. mesh repair of perineal hernia. Aust N Z J Surg. 1989;59:895.
70. Gaffney LB, Schanno JF. Sciatic hernia. Am J Surg. 1958;95:974-975. 88. Brotschi E, Noe JM, Silen W. Perineal hernias after proctectomy.
71. Ghahremani GG, Michael AS. Sciatic hernia with incarcerated ileum: Am J Surg. 1985;149:301.
CT and radiographic diagnosis. Gastrointest Radiol. 1991;16:120- 89. Franklin ME, Abrego D, Parra E. Laparoscopic repair of postoperative
122. perineal hernia. Hernia. 2002;6:42.
72. Miklos JR, O’Reilly MJ, Saye WB. Sciatic hernia as a cause of chronic 90. Ghellai AM, Islam S, Stoker ME. Laparoscopic repair of postoperative
pelvic pain in women. Obstet Gynecol. 1998;6:998-1001. perineal hernia. Surg Laparosc Endosc Percutan Tech. 2002;12:119.
73. Rotchild TPE. Ureteral hernia: report of a case of herniation of the 91. Beck DE, Fazio VW, Jagelman DG, et al. Postoperative perineal
ureter into the sciatic foramen. Arch Surg. 1969;98:96-98. hernia. Dis Colon Rectum. 1987;30:21.
t
74. Bernard AC, Lee C, Hoskins J, et al. Sciatic hernia: laparoscopic 92. So JB, Palmer MT, Shellito PC. Postoperative perineal hernia. Dis
ne
transabdominal extraperitoneal repair with plug and patch. Hernia. Colon Rectum. 1997;40:954.
2010;14:97. 93. Sorelli PG, Clark SK, Jenkins JT. Laparoscopic repair of primary
75. Beck WC, Baurys W, Brochu J. Herniation of the ureter into the perineal hernias: the approach of choice in the 21st century. Colorectal
sciatic foramen (‘curlicue ureter’). JAMA. 1952;179:441. Dis. 2012;14(2):e72-e73.
k.
76. Gee J, Munson JL, Smith JJIII. Laparoscopic repair of ureterosciatic 94. Abbas Y, Garner J. Laparoscopic and perineal approaches to perineal
hernia. Urology. 1999;54:730-733. hernia repair. Tech Coloproctol. 2014;18(4):361-364.
77. Stamatiou D, Skandakakis JE, Skandalakis LJ, Mirilas P. Perineal 95. Gomes Portilla A, Cendoya I, Uzquiza E, et al. Giant perineal hernia.
oo
hernia: surgical anatomy, embryology, and technique of repair. Am Laparoscopic mesh repair complemented by a perineal cutaneous
Surg. 2010;76(5):474-479. approach. Hernia. 2010;13:199.
78. Salameh JR. Primary and unusual abdominal wall hernias. Surg Clin 96. Goedhart-de Haan AM, Langenhoff BS, Petersen D, Verheijenet
yb
North Am. 2008;88:45. PM. Laparoscopic repair of perineal hernia after abdomino-
79. Rayhanabad J, Sassani P, Abbas MA. Laparoscopic repair of perineal perineal excision. Hernia. 2015;http://link.springer.com/articl
hernia. JSLS. 2009;13:237. e/10.1007%2Fs10029-015-1449-3. Accessed June 2016.
80. Twiss CB, Rosenblum N. Perineal hernia and perineocele. In: Raz 97. Mjoli M, Sloothaak DA, Buskens CJ, Bemelman WA, Tanis PJ. Perineal
er
S, Rodriguz LV, eds. Female Urology. Vol. 76. Philadelphia: Elsevier; hernia repair after abdominoperineal resection: a pooled analysis.
2008:743-750. Colorectal Dis. 2012;e400-e406.
rg
su
://
tp
ht
CHAPTER
Mesh: Material Science of Hernia Repair
Samuel Wade Ross
| David A. Iannitti
55
M
esh is defined as a network of interlaced material HISTORY
with a lattice-like structure, which in medicine has
become synonymous with use for reinforcement of Hernias have been recognized as medical problems since
hernia repairs. Mesh use has become ubiquitous as inguinal the time of the Egyptians in 1500 bc and have gone
hernia repair (IHR) is one of the most common procedures through a long evolution of treatment.14 The Greeks and
performed in the world.1 Additionally, IHR is the most Romans were the first to realize that inadequate technique
common type of hernia operation performed, with an in surgical abdominal closure was linked to incisional
estimated 700,000 performed annually in the United hernias, and Galen was the first to describe mass closure
t
States alone.1,2 The core principles of IHR remain the techniques in the second century ad. Galen also advocated
ne
same regardless of which technique is performed (excision for use of the paramedian incision to prevent incisional
of the hernia sac tension, reduction of intraabdominal hernia, a technique now proven to reduce hernia rates.15
contents, and tension-free closure); however, the methods In the 18th and 19th centuries, hernias were described
k.
used between various surgical approaches and individual and differentiated in detail by anatomists, but not much
techniques differ significantly. An open approach via headway was made in the operative repair of hernias until
oo
groin incision is still the most common IHR performed after the advancements in anesthesia and antisepsis in the
in the United States,3 and the modified Lichtenstein late 19th century that allowed for modern aseptic and
repair has become the most common open repair type general anesthetic techniques. This allowed for complex
since its introduction in 1984.4–6 In addition, the majority
yb
tissue repair techniques such as the Bassini repair, which
of repairs are now performed with a prosthetic mesh, as is still in use today.16
its use in an open repair has reduced hernia recurrence Modern hernia repair is characterized by the use of
from 7% to 1%.7 Laparoscopic repairs inherently require mesh reinforcement, which was popularized in 1958
er
the use of mesh and have had an increasing utilization in by the work of Usher who first described the use of PP
the past 20 years as laparoscopic education and training mesh.13 Lesser known was that the first artificial mesh
rg
around the globe, with an estimated 350,000 performed the deleterious side effect of toxic sulfur silver buildup.14
in the United States and 300,000 performed in Europe Subsequently, stainless steel or tantalum gauze was used in
per year.10,11 Both IHR and VHR are one of the top five the early 20th century, but these were plagued with high
://
procedures performed by graduating general surgery infection and complication rates. It was not until the dawn
residents consistently each year.12 Mesh reinforcement of plastic science after World War II and the creation of PP,
tp
of hernia defects was popularized by the work of Usher polyester, and polyfluoroethylene (PTFE) and expanded
who first described the use of polypropylene (PP) mesh, polyfluoroethylene (ePTFE) that suitable materials were
and today, mesh use is ubiquitous in VHR.13 However, present for the creation of malleable, pliable, and durable
ht
few surgeons understand the inherent differences in the meshes with relatively low complication rates.14
materials and properties of the mesh they use every day, In 1968, Rives and Stoppa described broad mesh rein-
the evolution of mesh technology and material science, forcement of groin hernias using Dacron grafts17 and then
different scenarios in which a certain mesh may be most applied this technique to large abdominal wall defects with
effective, and most importantly, that mesh choice can mesh in the retrorectus position, which was popularized
effect critical patient outcomes. in the 1980s.18,19 The onlay technique was first described
Therefore the goal of this chapter is to describe the by Chevrel in 1979, which involves mesh being placed
evolution and history of mesh for use for hernia repair, superficial to the anterior rectus fascia following suture
the physical properties on which we evaluate mesh, the repair of the fascial defect.20,21 Another mesh technique,
underlying materials used to construct them, the classifica- intraperitoneal mesh placement, was previously looked
tion schemas used to differentiate mesh, the nonmesh upon unfavorably due to the risk of mesh erosion into
adjuncts being used in combination with and used to viscera and formation of fistulas. However, with creation
fixate mesh, new technologies used to construct mesh of barrier-coated meshes and use of biologic and absorb-
(especially biologic and absorbable synthetic mesh), as able synthetic meshes, intraperitoneal mesh placement
well as the inherent risks and complications that can result is now ubiquitous and the way most laparoscopic VHR
due to mesh implantation. Additionally, we hope to focus is performed.22,23 The first laparoscopic VHR repair was
on evidence-based scenarios and recommendations on described by LeBlanc in 1993,22 and recently, the field of
when and what type of mesh is appropriate. robotic surgery has provided an opportunity to explore
621
Mesh: Material Science of Hernia Repair CHAPTER 55 621.e1
ABSTRACT
The field of hernia repair remains dynamic given the
ongoing evolution of materials for hernia repair as well
as surgical techniques. This chapter reviews the history
and philosophy of hernia repair along with data support-
ing use of reinforcing materials. We review in detail the
design and construction of permanent synthetic meshes,
biologic grafts, and absorbable synthetic meshes used today
in inguinal and ventral hernia repairs. Principles and
techniques for mesh fixation, mesh specific complications,
and the role of antiadhesive in antimicrobial coatings
are discussed.
KEYWORDS
Inguinal hernia
Ventral hernia
Mesh
Hernia technique
t
ne
Hernia materials
Synthetic hernia mesh
Biologic mesh
Absorbable synthetic mesh
k.
oo
yb
er
rg
su
://
tp
ht
622 SECTION I Esophagus and Hernia
advanced minimally invasive techniques; the first robotic drawbacks that make it ideal in certain situations. To guide
VHR was described in 2003.24 the use of mesh selection, the Ventral Hernia Working
Perhaps even more significant than minimally invasive Group published a grading scale for hernias by level of
techniques, the introduction of biologic-derived mesh and hernia and patient complexity and contamination: grade
absorbable synthetic mesh has changed the playing field 1 hernias are clean fields in low-risk patients without
for hernia repair. Interestingly, the first biologic reinforce- significant comorbidities; grade 2 are clean fields but
ments for hernia repair were with frozen cadaveric dermis, with patient comorbidities including diabetes, smoking,
tensor fasciae latae, and dural mater tissues in the 1930s and obesity; grade 3 are potential contamination such
and 1940s, but these had poor results.14 The renaissance as the presence of an ostomy, enterotomy, prior wound
of the biologic mesh came about in the early 2000s when infection; and grade 4 are mesh infection present or a
acellular dermal matrix (AlloDerm; Allergan, Dublin, septic wound.41
Ireland) was created from cadaveric human dermis for In general, synthetic mesh is ideal in lower grade
use in reconstructive and burn procedures. Later it was hernias, and biologic mesh should be used in higher-grade
studied for potential use in VHR in contaminated fields hernias. The use of these meshes to avoid mesh infection of
first in a pig model25 and then for use in stoma site hernias synthetic mesh should be balanced with the much higher
in humans.26 With the use in hernia repair solidified, financial cost associated with biologic mesh, as well as the
the number and type of biologic mesh developed has higher hernia recurrence rate, as these meshes stretch
exploded in the last decade with various tissue sources, over time. Indeed, in lower risk, lower grade hernias,
t
xenografts, allografts and now even absorbable synthetic, synthetic mesh has been found to be more cost effective
ne
and biosynthetic created meshes. Where in previous years than biologic mesh.42 For this reason, mesh companies
there were only a handful of mesh options, now the number have developed new mesh materials that have biochemical
of meshes available gives surgeons a plethora of options in scaffolds that will dissolve over time and allow for tissue
k.
how best to repair hernias tailored to specific patients.27 ingrowth: so-called absorbable synthetic, resorbable, or
biosynthetic mesh (terms vary in literature).
oo
EVIDENCE FOR MESH REPAIR
Mesh reinforcement for hernia repair is a much less SYNTHETIC
yb
debated topic than in prior decades, and it is currently
standard of care in developed nations given the high MATERIAL
rate of hernia recurrence with primary tissue repairs. The first plastic synthetic mesh was originally composed of
IHR with mesh has been studied for longer than VHR, PP, which is a permanent monofilament carbon polymer
er
with the first randomized control trial dating from the that is flexible and biologically inert.43 PP is composed of
1990s, and several meta-analyses showing that the risk hydrophobic polymer with alternating methyl moieties.
rg
of hernia recurrence is reduced 50% to 75% with mesh The monofilament nature allows for large pores, which
reinforcement compared with tissue repair alone.28,29 The facilitates tissue ingrowth and less interstices for bacteria
landmark study on this topic in VHR by Luijendijk et al. to set up biofilms, which are more likely the smaller the
su
in 2000 described a 43% hernia recurrence rate at 3 years pore and in multifilamentous mesh.44 The majority of
with suture repair compared with a 24% rate for mesh commercially available mesh is still made from PP, but
repair,30 which increased to 63% recurrence at 6 years.31 various evolutions of the original monofilament large-pore
://
Numerous studies have verified these results, and a recent mesh have been created. An in situ image of a PP knit mesh
meta-analysis clearly demonstrated that mesh repair results for IHR is displayed in Fig. 55.1. The body uses the PP as
tp
in less recurrence.32 Most surgeons acknowledge that it is a lattice to build scar tissue on, which is the process for
standard of care to perform VHR with mesh.7,31,33 Despite tissue reinforcement, but it is also a drawback if it is placed
this, there have been some proponents of performing in direct contact with the viscera. Adhesion formation can
ht
primary VHR with component separation alone without be severe if placed intraperitoneally, resulting in bowel
the use of a prosthetic mesh to buttress the repair.34–37 obstructions and mesh erosion into bowel.45,46 Therefore,
However, recurrence after VHR with component separation coatings have been added to PP to prevent visceral adhesion
alone can result in inferior outcomes and is reported to formation when it is placed in body cavities.47,48
occur anywhere from 5% to 23% in short-term follow-up After the introduction of PP meshes, PTFE and ePTFE
of 2 years.38–40 meshes were synthesized. PTFE is a sheet of hydrophobic
inert fluoropolymer that the body does not incorporate
into but encapsulates around as a foreign body. One
MESH CATEGORIES reason for this is that the material is highly negatively
Materials science for mesh has come a long way since charged, so water and oils will not adhere to the material.
the time of silver wire weaves, and in the past 15 years, PTFE is generally not currently used since the introduc-
the market for number and types of mesh has multiplied tion of ePTFE. This material is more microporous than
exponentially. The original synthetic plastic mesh repair PTFE but still does not allow for tissue ingrowth when
described by Usher was with a Marlex mesh composed of made as a flat sheet.46 The principal advantage of ePTFE
PP,13 and since that time, hernia mesh has evolved from is that it can be placed directly onto viscera and not
this single synthetic material to three broad classes of form adhesions, as depicted in Fig. 55.2. This must be
materials: permanent synthetic, biologic, and absorbable balanced with the fact that it is highly susceptible to
synthetic. Each category of mesh has its own benefits and bacterial colonization in addition to seroma formation.
Mesh: Material Science of Hernia Repair CHAPTER 55 623
t
FIGURE 55.1 In situ image of a wide-pore lightweight expanded polyfluoroethylene mesh that resists adhesion formation
ne
polypropylene mesh for onlay Lichtenstein-type repair for inguinal and can be placed directly next to viscera. (Copyright 2016 Davol,
hernia repair. (Copyright 2016 Davol, Inc. All rights reserved. Used Inc. All rights reserved. Used with the permission of Davol, Inc.)
with the permission of Davol, Inc.)
k.
FILAMENT, WEAVE, KNIT
oo
Dualmesh
Mesh can differ significantly in how the strands of material,
usually PP, are arranged, from very simple one-to-one
yb
knits to intricate and complicated weaves. The choice
in the configuration can affect bursting pressure, tensile
strength, and elasticity of the mesh.43 In addition, mesh
can have a multifilament, braided, or almost ropelike
er
2ND SE 10KV 800X 33.3µm 02222 tion. Just as in fabric, knit mesh involves creating loops of
material using multiple needles with one thread. Woven
tp
FIGURE 55.2 Scanning electron microscope image of the rough mesh involves a similar process but with multiple threads
side of DualMesh expanded polyfluouroethylene, designed to
at right angles.
increase surface area to facilitate better tissue incorporation.
In general, knit mesh will have larger pore size, elasticity,
ht
(Copyright 2016 Davol, Inc. All rights reserved. Used with the
and will stretch in all directions, called isotropic stretch.53
permission of Davol, Inc.)
However, woven mesh will only stretch in a direction
oblique to the intersection of the thread weaves, called
Most surgeons would recommend that infected ePTFE must anisotropic stretch, which is a little understood but vitally
always be explanted.46,49 To combat the issue of poor tissue important aspect to consider when orienting mesh for
incorporation, different-sided ePTFE mesh has become implantation.54 Mesh placed with the greatest direction
available in DualMesh (Gore Medical, Flagstaff, Arizona), of force on the vector oblique from the weave will stretch
which allows textures of ePTFE with a corduroy side for more than if oriented in the direction of the weave threads.
more surface area for tissue incorporation and a smooth A closeup of a monofilament woven PP mesh is displayed
side for visceral contact.50 An electronic microscope image in Fig. 55.4. These factors of filament number, size of the
of the corduroy side is seen in Fig. 55.3. filament, and knit versus weave are what contribute mostly
Polyester mesh was created after PP and ePTFE, but to mesh weight, tensile strength, and burst strength.
while available, is not generally used in the United States,
since it has similar properties to PP mesh. It is created MESH WEIGHT AND BURST STRENGTH
from polymers of terephthalic acid, which is hydrophilic Synthetic mesh varies by weight with three broad categories:
and can be degraded by hydrolysis.51 Commonly available lightweight, midweight, and heavyweight mesh, which
synthetic mesh categorized by physical properties are have different inherent tensile strengths depending on
displayed in Table 55.1. their weave and material composition.55 The weight of the
624
Barrier Weight
Material Strategy Category Filament Structure Pore Size Mesh Brand Name Additional Comments
Polypropylene None Light Mono Knit Macro Bard Soft For all hernia repair types, Soft is lighter weight, larger pore, Plug
version available
Light Mono Knit Macro Prolene Soft For all hernia repair types, Soft is lighter weight, larger pore
Light Mono Weave Macro Progrip Self-gripping mesh for laparoscopic IHR facilitated by polylactic acid
mesh component, absorbs over 18 months, tacking not required
Light Mono Weave Macro 3D Max Light Contoured mesh for laparoscopic IHR, Light version is lighter weight
Light Mono Knit Macro Prolite, Prolite Ultra For IHR, Ultra is lighter weight, larger pore
ht
Light Mono Weave Macro VitaMesh Blue For all hernia repairs, dyed blue for better visualization
Light Multi Knit Macro Ultrapro, Ultrapro Monocryl as one of the filaments, absorbs over 2 weeks, leaving
tp
SECTION I Esophagus and Hernia
FIGURE 55.4 Micrograph of a lightweight polypropylene woven Open pore mesh design, 35x magnification
t
mesh. (Copyright 2016 Davol, Inc. All rights reserved. Used with
ne
FIGURE 55.5 Micrograph of large-pore polypropylene mesh.
the permission of Davol, Inc.)
(Copyright 2016 Davol, Inc. All rights reserved. Used with the
permission of Davol, Inc.)
k.
mesh actually refers to the mass of the material in a set
area, which is in actuality density. Lightweight mesh is also
oo
considered less than 35 g/m2, and these PP meshes are in Fig. 55.5. The long-term strength of the mesh repair is
usually monofilament, smaller caliber woven meshes with not from the mesh itself but in creating tissue ingrowth and
large pores of up to 4 mm.56 Midweight mesh is between fibrosis, and there is some data to suggest that too small
35 and 60 g/m2 and has larger caliber filaments, is woven,
yb
of a pore size prevents fibrosis from occurring through
and has larger pores. Heavyweight mesh has a density of the mesh. This “bridging fibrosis” is key to sustained
60+ g/m2, and a knitted, monofilament PP mesh such as tissue strength.56 Given the need for balance between
Marlex has a weight of 95 g/m2.55 Heavyweight mesh has bridging fibrosis and early mesh failure in lightweight
er
higher tensile strength in the range of 1200 N compared mesh and decreased porosity and bacterial clearance in
with light (540 N) and midweight mesh (560 N). However, heavyweight mesh, midweight mesh may theoretically be
rg
given the concern for mesh sensation, stiffness, and ability the ideal uncoated synthetic mesh. However, there is no
to clear bacterial infection, in the past decade there randomized clinical data to support midweight mesh as
has been trend away from using heavyweight for lighter superior, and surgeons should be knowledgeable about
su
weight mesh. Recently data have suggested that quality mesh materials so they can choose the most appropriate
of life does not suffer with heavyweight mesh57 and that mesh for each clinical situation.
lightweight mesh can suffer catastrophic central failure
://
the use of midweight and heavyweight mesh, especially was a flat sheet of metal or plastic. But in recent years
for VHR, has had a resurgence. shaped mesh technology has become available to create
mesh designed to fulfill certain roles and contoured to
ht
t
precursors to absorbable coated mesh and are designed
ne
indirect inguinal space. (Copyright 2016 Davol, Inc. All rights
reserved. Used with the permission of Davol, Inc.) to facilitate tissue ingrowth into the PP side and have an
inert ePTFE side to prevent visceral attachments. These are
still on the market today, but they have all the drawbacks
k.
associated with ePTFE mesh and have been associated with
increased propensity to infections (particularly methicillin-
oo
resistant Staphylococcus aureus [MRSA]), and low rates
of mesh salvage after infection.66–68 The high degree
of contraction and stiffening of combined permanent
yb
materials have led to patient discomfort and the mesh
pulling free of the abdominal wall with subsequent hernia
recurrence.59 Difficulty with laparoscopic handling of
composite meshes has also been noted by surgeons.69 The
er
dimensional mesh placed in the preperitoneal inguinal space for coated PP mesh are commercially available, including
laparoscopic transabdominal preperitoneal hernia repair. omega-3 fatty acid coatings (O3FA), 71 hyaluronate-
tp
(Copyright 2016 Davol, Inc. All rights reserved. Used with the
carboxymethylcellulose (HCMC) hydrogel,72 and oxidized
permission of Davol, Inc.)
regenerated cellulose (ORC).73 HCMC is interesting in
that the hydrogel layer is absorbable but bound to the
ht
t
reserved. Used with the permission of Davol, Inc.)
ne
per square centimeter given the cost of procuring and
k.
processing the tissues from cadaveric or animal donors.
Regardless of cross-linkage, these meshes are limited in
oo
their durability as they lose their strength as the body
degrades and incorporates the collagen scaffold of the
mesh. Human studies of biologic mesh use in contaminated
yb
fields has elucidated that these hernias have an exceedingly
high hernia recurrence rate of 31.3% at 5 years78 and 56%
or higher when a biologic bridge repair is performed.79
Most hernia specialists reserve biologic mesh for con-
er
2016 Davol, Inc. All rights reserved. Used with the permission of
it was possible for biologic mesh to become chronically
Davol, Inc.)
infected and require explantation. These meshes should be
tp
Graft Source Tissue Source Cross-Linkage Mesh Brand Name Additional Comments
Human Dermis No AlloDerm Small sizes, requires refrigeration and subsequent
rehydration, highly elastic
Allomax Acellular, sterilized by gamma radiation
FlexHD No refrigeration or rehydration, reduced elasticity
Porcine Dermis Yes Permacol Acellular, sterilized by gamma radiation, no refrigeration
necessary
CollaMend Acellular and lyophilized
No Strattice Acellular, available in large sizes, fenestrated versions
available to facilitate ingrowth
XenMatrix, Acellular, available in large sizes, AB has drug elution of
XenMatrix AB minocycline and rifampin for 7 days
Intestinal submucosa Surgisis Acellular, no refrigeration but does require rehydration
Bovine Dermis No SurgiMend Fetal bovine derived, requires rehydration
Pericardium Veritas Initially used for staple line reinforcement, now for VHR
t
Tutopatch Fenestrated mesh also available, 5-year shelf life
ne
Yes Periguard Can be used in multiple scenarios where patch is needed,
less studied for hernia repair, requires rehydration
VHR, Ventral hernia repair.
k.
oo
granulation tissue ingrowth in open abdomens and chronic
wounds.82 Unfortunately, there is a high fistula rate of
9% to 17%.72,83,84
yb
The first modern mesh in this class was Gore Bio-A,
which is constructed from 67% polyglycolic acid and
33% trimethylene carbonate, and was first marketed for
hernia repair in the early 2010s.85,86 However, the material
er
reversal, enterocutaneous fistula takedown, and/or current hydroxybutyrate knit mesh (brand name Phasix). (Copyright 2016
mesh infection, and at 2 years there was only a 17% Davol, Inc. All rights reserved. Used with the permission of Davol,
recurrence rate.88
ht
Inc.)
A newer material that is truly a biosynthetic product
is poly-4-hydroxybutyrate (P4HB), which is created using
genetically engineered Escherichia coli to produce a protein area in mesh science and may offer more durable options
substrate. This is then refined and polymerized and knitted than biologic mesh. Currently, however, there have been
into a mesh pattern, which has a much longer absorption no head-to-head comparisons of biologic, synthetic, and
period, 12 to 18 months.89 Mechanically, this mesh has biosynthetic mesh. Available absorbable synthetic mesh
properties similar to an uncoated light- to midweight PP categorized by physical properties is displayed in Table 55.3.
mesh. A micrograph of a P4HB mesh is pictured in Fig.
55.11. However, unlike Bio-A or biologic mesh, this mesh
will cause significant adhesion formation; therefore a
MESH FIXATION
hydrogel-coated version has become available, and early The technology for securing mesh to the abdominal wall
animal data show that it functions similar to a coated has not evolved as rapidly as the prostheses themselves,
synthetic PP mesh.90 The only current human data on and most surgeons still use simple suture to secure mesh
this mesh are from reconstructive surgery after deep in open IHR, VHR, and UHR. Laparoscopic IHR brought
inferior epigastric perforator flap creation, which showed about the need for new fixation methods given the small
decreased abdominal bulge when P4HB was used as an space, and the first tacking devices appeared soon after the
onlay.91 There are many other materials coming to market introduction of the technique. The initial design for these,
in this category, and in general, this is the fastest growing which has continued today, was permanent titanium tackers
Mesh: Material Science of Hernia Repair CHAPTER 55 629
t
is highest in uncoated PP mesh placed directly into the
ne
peritoneum adjacent to the viscera and can erode into
the small intestine, colon, and even bladder.97–99 However,
ePTFE and polyester can erode as well and, in severe
k.
uncontrolled fistulas, can even result in necrotizing fasciitis
of the abdominal wall.100 These complications are often
oo
underreported since patients may present to physicians
other than the index surgeon and since presentation
can occur very delayed from the original surgery. These
yb
complications are disabling, both financially and in terms
of patient quality of life. Surgical repair of the erosion
is usually mandated to explant the mesh and resect the
fistula track.
er
tacks that could stimulate adhesion formation leading to smoker, or if there was a coated, ePTFE, or composite
bowel obstructions92 or even erode into other structures,93 mesh, the patients had 100% need for mesh resection.
tp
but later designs have added a plastic cap to prevent this, Patients with a lightweight, PP mesh did best, and some
pictured in Fig. 55.12. Since titanium is permanent, some were able to be managed with antibiotics and percutaneous
have theorized chronic pain after laparoscopic IHR could drain placement.101
ht
antimicrobial properties, and silver has had a resurgence 9. Heniford BT, Matthews BD, Box EA, et al. Optimal teaching
in wound care, dressing, and even endotracheal tubes to environment for laparoscopic ventral herniorrhaphy. Hernia.
2002;6(1):17-20.
prevent pneumonia. Studies have shown that mesh bonded 10. Poulose BK, Shelton J, Phillips S, et al. Epidemiology and cost of
with nanoparticles of silver can decrease bacterial load by ventral hernia repair: making the case for hernia research. Hernia.
greater than 99% when inoculated with Staphylococcus aureus 2012;16(2):179-183.
and E. coli.104 Similarly, gold-palladium nanoparticles bound 11. Sauerland S, Walgenbach M, Habermalz B, Seiler CM, Miserez M.
Laparoscopic versus open surgical techniques for ventral or incisional
to mesh have demonstrated 0% surgical site infection hernia repair. Cochrane Database Syst Rev. 2011;(3):CD007781.
when inoculated with Staphylococcus epidermidis.105 However, 12. Accreditation Council for Graduate Medical Education. Case Log
given the rarity of these materials widespread use of Statistical Reports; 2013. http://www.acgme.org/acgmeweb/tabid/
these technologies may be cost prohibitive. Lysostaphin 274/DataCollectionSystems/ResidentCaseLogSystem/CaseLogs
is an antistaphylococcal protein that has been shown StatisticalReports.aspx Accessed 15 September 2016.
13. Usher FC, Ochsner J, Tuttle LL Jr. Use of marlex mesh in the
to decrease bacterial loads and increase survival in rat repair of incisional hernias. Am Surg. 1958;24(12):969-974.
models when applied to PP mesh and could be a cheaper 14. Sanders DL, Kingsnorth AN. From ancient to contemporary times:
alternative.106,107 Actual antibiotics have been combined a concise history of incisional hernia repair. Hernia. 2012;16(1):1-7.
with drug-eluting particles that are then bound to mesh. 15. Muysoms FE, Antoniou SA, Bury K, et al. European Hernia Society
guidelines on the closure of abdominal wall incisions. Hernia.
In addition, vancomycin,108 ofloxacin,109 and rifampin/ 2015;19(1):1-24.
minocycline110 have been studied in vitro and animals, 16. McClusky DA 3rd, Mirilas P, Zoras O, Skandalakis PN, Skandala-
but no human studies have been performed. kis JE. Groin hernia: anatomical and surgical history. Arch Surg.
t
2006;141(10):1035-1042.
ne
17. Rives J, Stoppa R, Fortesa L, Nicaise H. [Dacron patches and their
SUMMARY AND RECOMMENDATIONS place in surgery of groin hernia. 65 cases collected from a complete
series of 274 hernia operations]. Ann Chir. 1968;22(3):159-171.
Mesh science has come a long way since the days of 18. Stoppa RE. The treatment of complicated groin and incisional
k.
hand weaving together silver threads, and with the inci- hernias. World J Surg. 1989;13(5):545-554.
dence and number of hernia repairs increasing every 19. Rives J, Pire JC, Flament JB, Palot JP, Body C. [Treatment of large
eventrations. New therapeutic indications apropos of 322 cases].
oo
year, the technology and capital investment in mesh Chir Mem Acad Chir. 1985;111(3):215-225.
hernia repair will likely continue to escalate. Synthetic 20. Alexandre JH. Jean-Paul Chevrel (1933–2006). Hernia.
meshes, especially large-pore midweight PP, continue 2007;11(4):293-296.
yb
to offer the most robust repairs when performed in 21. Stoikes N, Webb D, Powell B, Voeller G. Preliminary report of a
clean cases, but the surgeon should use discretion in sutureless onlay technique for incisional hernia repair using fibrin
glue alone for mesh fixation. Am Surg. 2013;79(11):1177-1180.
the choice of mesh for contaminated and dirty fields. 22. LeBlanc KA, Booth WV. Laparoscopic repair of incisional abdominal
While biologic mesh has been well studied over the last
er
2. Masukawa K, Wilson SE. Is postoperative chronic pain syndrome 28. Scott NW, McCormack K, Graham P, Go PM, Ross SJ, Grant AM.
higher with mesh repair of inguinal hernia? Am Surg. 2010;76(10): Open mesh versus non-mesh for repair of femoral and inguinal
1115-1118. hernia. Cochrane Database Syst Rev. 2002;(4):CD002197.
3. Saleh F, Okrainec A, D’Souza N, Kwong J, Jackson TD. Safety of 29. Collaboration EUHT. Repair of groin hernia with synthetic
laparoscopic and open approaches for repair of the unilateral mesh: meta-analysis of randomized controlled trials. Ann Surg.
primary inguinal hernia: an analysis of short-term outcomes. Am 2002;235(3):322-332.
J Surg. 2014;208(2):195-201. 30. Luijendijk RW, Hop WC, van den Tol MP, et al. A comparison of
4. Lichtenstein IL, Shulman AG, Amid PK, Montllor MM. The tension- suture repair with mesh repair for incisional hernia. N Engl J Med.
free hernioplasty. Am J Surg. 1989;157(2):188-193. 2000;343(6):392-398.
5. Amid PK. The Lichtenstein repair in 2002: an overview of causes 31. Burger JW, Luijendijk RW, Hop WC, Halm JA, Verdaasdonk
of recurrence after Lichtenstein tension-free hernioplasty. Hernia. EG, Jeekel J. Long-term follow-up of a randomized controlled
2003;7(1):13-16. trial of suture versus mesh repair of incisional hernia. Ann Surg.
6. Smietanski M, Bigda J, Zaborowski K, Worek M, Sledzinski Z. 2004;240(4):578-583, discussion 583–575.
Three-year follow-up of modified Lichtenstein inguinal hernioplasty 32. den Hartog D, Dur AH, Tuinebreijer WE, Kreis RW. Open surgi-
using lightweight poliglecaprone/polypropylene mesh. Hernia. cal procedures for incisional hernias. Cochrane Database Syst Rev.
2009;13(3):239-242. 2008;(3):CD006438.
7. Vrijland WW, van den Tol MP, Luijendijk RW, et al. Randomized 33. Mathes T, Walgenbach M, Siegel R. Suture versus mesh repair in
clinical trial of non-mesh versus mesh repair of primary inguinal primary and incisional ventral hernias: a systematic review and
hernia. Br J Surg. 2002;89(3):293-297. meta-analysis. World J Surg. 2016;40(4):826-835.
8. Zendejas B, Ramirez T, Jones T, et al. Trends in the utilization of 34. DiCocco JM, Fabian TC, Emmett KP, Magnotti LJ, Goldberg SP, Croce
inguinal hernia repair techniques: a population-based study. Am J MA. Components separation for abdominal wall reconstruction:
Surg. 2012;203(3):313-317, discussion 317. the Memphis modification. Surgery. 2012;151(1):118-125.
Mesh: Material Science of Hernia Repair CHAPTER 55 631
35. Vargo D. Component separation in the management of the difficult 59. Novitsky YW, Harrell AG, Cristiano JA, et al. Comparative evaluation
abdominal wall. Am J Surg. 2004;188(6):633-637. of adhesion formation, strength of ingrowth, and textile properties
36. de Vries Reilingh TS, van Goor H, Charbon JA, et al. Repair of of prosthetic meshes after long-term intra-abdominal implantation
giant midline abdominal wall hernias: “components separation in a rabbit. J Surg Res. 2007;140(1):6-11.
technique” versus prosthetic repair : interim analysis of a randomized 60. Lichtenstein IL, Shore JM. Simplified repair of femoral and recurrent
controlled trial. World J Surg. 2007;31(4):756-763. inguinal hernias by a “plug” technic. Am J Surg. 1974;128(3):439-444.
37. Espinosa-de-los-Monteros A, Dominguez I, Zamora-Valdes D, 61. Magnusson J, Nygren J, Gustafsson UO, Thorell A. UltraPro Hernia
Castillo T, Fernandez-Diaz OF, Luna-Torres HA. Closure of midline System, Prolene Hernia System and Lichtenstein for primary
contaminated and recurrent incisional hernias with components inguinal hernia repair: 3-year outcomes of a prospective randomized
separation technique reinforced with plication of the rectus muscles. controlled trial. Hernia. 2016;20(5):641-648.
Hernia. 2013;17(1):75-79. 62. Bell RC, Price JG. Laparoscopic inguinal hernia repair using an
38. Fischer PE, Fabian TC. Decreasing the reherniation rate using anatomically contoured three-dimensional mesh. Surg Endosc.
a modified components separation technique. World J Surg. 2003;17(11):1784-1788.
2007;31(11):2266, author reply 2267–2268. 63. Leber GE, Garb JL, Alexander AI, Reed WP. Long-term complications
39. Jernigan TW, Fabian TC, Croce MA, et al. Staged management associated with prosthetic repair of incisional hernias. Arch Surg.
of giant abdominal wall defects: acute and long-term results. Ann 1998;133(4):378-382.
Surg. 2003;238(3):349-355, discussion 355–347. 64. Halm JA, de Wall LL, Steyerberg EW, Jeekel J, Lange JF. Intraperi-
40. Ko JH, Wang EC, Salvay DM, Paul BC, Dumanian GA. Abdominal wall toneal polypropylene mesh hernia repair complicates subsequent
reconstruction: lessons learned from 200 “components separation” abdominal surgery. World J Surg. 2007;31(2):423-429, discussion
procedures. Arch Surg. 2009;144(11):1047-1055. 430.
41. Ventral Hernia Working G, Breuing K, Butler CE, et al. Inci- 65. Koehler RH, Begos D, Berger D, et al. Minimal adhesions to ePTFE
sional ventral hernias: review of the literature and recommenda- mesh after laparoscopic ventral incisional hernia repair: reoperative
t
tions regarding the grading and technique of repair. Surgery. findings in 65 cases. JSLS. 2003;7(4):335-340.
ne
2010;148(3):544-558. 66. Cobb WS, Carbonell AM, Kalbaugh CL, Jones Y, Lokey JS. Infection
42. Fischer JP, Basta MN, Mirzabeigi MN, Kovach SJ 3rd. A comparison of risk of open placement of intraperitoneal composite mesh. Am
outcomes and cost in VHWG grade II hernias between Rives-Stoppa Surg. 2009;75(9):762-767, discussion 767–768.
synthetic mesh hernia repair versus underlay biologic mesh repair. 67. Cobb WS, Harris JB, Lokey JS, McGill ES, Klove KL. Incisional
k.
Hernia. 2014;18(6):781-789. herniorrhaphy with intraperitoneal composite mesh: a report of
43. Bilsel Y, Abci I. The search for ideal hernia repair; mesh materials 95 cases. Am Surg. 2003;69(9):784-787.
and types. International J Surg. 2012;10(6):317-321. 68. Greenberg JJ. Can infected composite mesh be salvaged? Hernia.
oo
44. Blatnik JA, Krpata DM, Jacobs MR, Gao Y, Novitsky YW, Rosen MJ. 2010;14(6):589-592.
In vivo analysis of the morphologic characteristics of synthetic mesh 69. Gal I, Balint A, Szabo L. [Results of laparoscopic repair of abdominal
to resist MRSA adherence. J Gastrointest Surg. 2012;16(11):2139-2144. wall hernias using an ePTFE-polypropylene composite mesh].
yb
45. Borrazzo EC, Belmont MF, Boffa D, Fowler DL. Effect of prosthetic Zentralbl Chir. 2004;129(2):92-95.
material on adhesion formation after laparoscopic ventral hernia 70. Deeken CR, Faucher KM, Matthews BD. A review of the composi-
repair in a porcine model. Hernia. 2004;8(2):108-112. tion, characteristics, and effectiveness of barrier mesh prostheses
46. Matthews BD, Pratt BL, Pollinger HS, et al. Assessment of adhesion utilized for laparoscopic ventral hernia repair. Surg Endosc.
er
adhesions with use of hyaluronic acid-carboxymethylcellulose mem- tissue response to a novel omega-3 fatty acid bioabsorbable barrier
brane: a randomized clinical trial. Ann Surg. 2002;235(2):193-199. macroporous mesh after intraperitoneal placement. Surg Innov.
48. van ‘t Riet M, de Vos van Steenwijk PJ, Bonthuis F, et al. Prevention 2009;16(1):46-54.
su
of adhesion to prosthetic mesh: comparison of different barriers 72. van’t Riet M, de Vos van Steenwijk PJ, Bonjer HJ, Steyerberg EW,
using an incisional hernia model. Ann Surg. 2003;237(1):123- Jeekel J. Mesh repair for postoperative wound dehiscence in the
128. presence of infection: is absorbable mesh safer than non-absorbable
49. Petersen S, Henke G, Freitag M, Faulhaber A, Ludwig K. Deep mesh? Hernia. 2007;11(5):409-413.
://
prosthesis infection in incisional hernia repair: predictive factors 73. Harrell AG, Novitsky YW, Peindl RD, et al. Prospective evaluation of
and clinical outcome. Eur J Surg Acta Chir. 2001;167(6):453-457. adhesion formation and shrinkage of intra-abdominal prosthetics
50. Dolce CJ, Keller JE, Stefanidis D, et al. Evaluation of soft tissue in a rabbit model. Am Surg. 2006;72(9):808-813, discussion 813–804.
tp
attachments to a novel intra-abdominal prosthetic in a rabbit 74. Deerenberg EB, Mulder IM, Grotenhuis N, Ditzel M, Jeekel J, Lange
model. Surg Innov. 2012;19(3):295-300. JF. Experimental study on synthetic and biological mesh implantation
51. Novitsky YW, Harrell AG, Hope WW, Kercher KW, Heniford BT. in a contaminated environment. Br J Surg. 2012;99(12):1734-1741.
ht
Meshes in hernia repair. Surg Technol Int. 2007;16:123-127. 75. Kathju S, Nistico L, Melton-Kreft R, Lasko LA, Stoodley P. Direct
52. Harrell AG, Novitsky YW, Kercher KW, et al. In vitro infectability demonstration of bacterial biofilms on prosthetic mesh after ventral
of prosthetic mesh by methicillin-resistant Staphylococcus aureus. herniorrhaphy. Surg Infect (Larchmt). 2015;16(1):45-53.
Hernia. 2006;10(2):120-124. 76. Kolker AR. Multilayer reconstruction of abdominal wall defects with
53. Tomaszewska A. Mechanical behaviour of knit synthetic mesh used acellular dermal allograft (AlloDerm) and component separation.
in hernia surgery. Acta Bbioeng Biomech. 2016;18(1):77-86. Ann Plast Surg. 2005;55(1):36-41, discussion 41–32.
54. Cordero A, Hernandez-Gascon B, Pascual G, Bellon JM, Calvo 77. Deeken CR, Melman L, Jenkins ED, Greco SC, Frisella MM, Matthews
B, Pena E. Biaxial mechanical evaluation of absorbable and BD. Histologic and biomechanical evaluation of crosslinked and
nonabsorbable synthetic surgical meshes used for hernia repair: non-crosslinked biologic meshes in a porcine model of ventral
physiological loads modify anisotropy response. Ann Biomed Eng. incisional hernia repair. J Am Coll Surg. 2011;212(5):880-888.
2016;44(7):2181-2188. 78. Rosen MJ, Krpata DM, Ermlich B, Blatnik JA. A 5-year clinical
55. Cobb WS, Burns JM, Peindl RD, et al. Textile analysis of heavy experience with single-staged repairs of infected and contami-
weight, mid-weight, and light weight polypropylene mesh in a nated abdominal wall defects utilizing biologic mesh. Ann Surg.
porcine ventral hernia model. J Surg Res. 2006;136(1):1-7. 2013;257(6):991-996.
56. Cobb WS, Kercher KW, Heniford BT. The argument for lightweight 79. Booth JH, Garvey PB, Baumann DP, et al. Primary fascial closure
polypropylene mesh in hernia repair. Surg Innov. 2005;12(1):63-69. with mesh reinforcement is superior to bridged mesh repair for
57. Ladurner R, Chiapponi C, Linhuber Q, Mussack T. Long term abdominal wall reconstruction. J Am Coll Surg. 2013;217(6):999-1009.
outcome and quality of life after open incisional hernia repair—light 80. Huntington CR, Cox TC, Blair LJ, et al. Biologic mesh in ventral
versus heavy weight meshes. BMC Surg. 2011;11:25. hernia repair: outcomes, recurrence, and charge analysis. Surgery.
58. Petro CC, Nahabet EH, Criss CN, et al. Central failures of light- 2016;160(6):1517-1527.
weight monofilament polyester mesh causing hernia recurrence: 81. Kim M, Oommen B, Ross SW, et al. The current status of biosynthetic
a cautionary note. Hernia. 2015;19(1):155-159. mesh for ventral hernia repair. Surg Technol Int. 2014;25:114-121.
632 SECTION I Esophagus and Hernia
82. Levasseur JC, Lehn E, Rignier P. [Experimental study and clinical 97. Chand M, On J, Bevan K, Mostafid H, Venkatsubramaniam AK.
use of a new material in severe postoperative evisceration of the Mesh erosion following laparoscopic incisional hernia repair.
abdomen (author’s transl). Chirurgie. 1979;105(7):577-581. Hernia. 2012;16(2):223-226.
83. Fabian TC, Croce MA, Pritchard FE, et al. Planned ventral hernia. 98. Riaz AA, Ismail M, Barsam A, Bunce CJ. Mesh erosion into the
Staged management for acute abdominal wall defects. Ann Surg. bladder: a late complication of incisional hernia repair. A case
1994;219(6):643-650, discussion 651–643. report and review of the literature. Hernia. 2004;8(2):158-159.
84. Greene MA, Mullins RJ, Malangoni MA, Feliciano PD, Richardson JD, 99. Foda M, Carlson MA. Enterocutaneous fistula associated with
Polk HC Jr. Laparotomy wound closure with absorbable polyglycolic ePTFE mesh: case report and review of the literature. Hernia.
acid mesh. Surg Gynecol Obstet. 1993;176(3):213-218. 2009;13(3):323-326.
85. Negro P, Gossetti F, Dassatti MR, Andreuccetti J, D’Amore L. 100. Moussi A, Daldoul S, Bourguiba B, Othmani D, Zaouche A.
Bioabsorbable Gore BIO-A plug and patch hernia repair in young Gas gangrene of the abdominal wall due to late-onset enteric
adults. Hernia. 2012;16(1):121-122. fistula after polyester mesh repair of an incisional hernia. Hernia.
86. Burgess PL, Brockmeyer JR, Johnson EK. Amyand hernia repaired 2012;16(2):215-217.
with Bio-A: a case report and review. J Surg Educ. 2011;68(1):62-66. 101. Augenstein VA, Cox TC, Hlavacek C, et al. Treatment of 161
87. Priego Jimenez P, Salvador Sanchis JL, Angel V, Escrig-Sos J. Consecutive Synthetic Mesh Infections: Can Mesh Be Salvaged?
Short-term results for laparoscopic repair of large paraesophageal 2016. Washington, DC: Americas Hernia Society.
hiatal hernias with Gore Bio A(R) mesh. International J Surg. 102. Perez-Kohler B, Garcia-Moreno F, Brune T, Pascual G, Bellon JM.
2014;12(8):794-797. Preclinical bioassay of a polypropylene mesh for hernia repair
88. Rosen MJ, Bauer JJ, Harmaty M, et al. Multicenter, prospective, lon- pretreated with antibacterial solutions of chlorhexidine and allicin:
gitudinal study of the recurrence, surgical site infection, and quality an in vivo study. PLoS ONE. 2015;10(11):e0142768.
of life after contaminated ventral hernia repair using biosynthetic 103. Yabanoglu H, Arer IM, Caliskan K. The effect of the use of synthetic
absorbable mesh: the COBRA study. Ann Surg. 2017;265:205-211. mesh soaked in antibiotic solution on the rate of graft infec-
t
89. Deeken CR, Matthews BD. Characterization of the mechanical tion in ventral hernias: a prospective randomized study. Int Surg.
ne
strength, resorption properties, and histologic characteristics of a 2015;100(6):1040-1047.
fully absorbable material (Poly-4-hydroxybutyrate-PHASIX Mesh) 104. Kumar V, Jolivalt C, Pulpytel J, Jafari R, Arefi-Khonsari F. Develop-
in a porcine model of hernia repair. ISRN Surg. 2013;2013:238067. ment of silver nanoparticle loaded antibacterial polymer mesh
90. Scott JR, Deeken CR, Martindale RG, Rosen MJ. Evaluation of a fully using plasma polymerization process. J Biomed Mater Res A.
k.
absorbable poly-4-hydroxybutyrate/absorbable barrier composite 2013;101(4):1121-1132.
mesh in a porcine model of ventral hernia repair. Surg Endosc. 105. Saygun O, Agalar C, Aydinuraz K, et al. Gold and gold-palladium
2016;30(9):3691-3701. coated polypropylene grafts in a S. epidermidis wound infection
oo
91. Wormer BA, Clavin NW, Lefaivre JF, et al. Reducing postoperative model. J Surg Res. 2006;131(1):73-79.
abdominal bulge following deep inferior epigastric perforator flap 106. Belyansky I. The addition of lysostaphin dramatically improves
breast reconstruction with onlay monofilament poly-4-hydroxybu- survival, protects porcine biomesh from infection, and improves
yb
tyrate biosynthetic mesh. J Reconstr Microsurg. 2016;33(1):8-18. graft tensile shear strength. J Surg Res. 2011;171(2):409-415.
92. Withers L, Rogers A. A spiral tack as a lead point for volvulus. JSLS. 107. Belyansky I, Tsirline VB, Montero PN, et al. Lysostaphin-coated
2006;10(2):247-249. mesh prevents staphylococcal infection and significantly improves
93. Reynvoet E, Berrevoet F. Pros and cons of tacking in laparoscopic survival in a contaminated surgical field. Am Surg. 2011;77(8):1025-
er
tematic review and meta-analysis. Surg Endosc. 2012;26(5):1269-1278. for soft tissue repair. Surg Innov. 2016;23(5):442-455.
tp
ht
CHAPTER
T
he stomach is a remarkable organ that aids in and the parietal portion of the anterior abdominal wall.
digestion, regulating nutrition, and controlling Posteriorly, the pancreas (neck, body, and tail), left kidney,
appetite. The complex physiologic processes by and adrenal grand border the stomach. The spleen sits
which the stomach exerts its endocrine and nutritional posterolaterally and the transverse colon inferiorly. The
functions have been researched for decades and there is two points of attachment are at the gastroesophageal
still much to be learned. This chapter on the anatomy junction superiorly and the retroperitoneal duodenum.
and physiology of the stomach aims to equip the surgeon Ligamentous attachments also help to further anchor the
with the detailed knowledge of not only the gross anatomy stomach to surrounding organs: gastrophrenic (dia-
and vascular supply of the stomach, but also the physiologic phragm), hepatogastric or lesser omentum (liver), gas-
t
properties behind the complex process of gastric acid trosplenic or gastrolienal (spleen), and the gastrocolic or
ne
secretion and hormonal regulation related to digestion. greater omentum (transverse colon).
The anatomic regions of the stomach can be distin-
EMBRYOLOGIC DEVELOPMENT guished based on surgical landmarks (Fig. 56.3). Beginning
k.
superiorly from the abdominal portion of the esophagus
The stomach arises from the embryonic endoderm and and the gastroesophageal junction, the cardiac portion
oo
comprises a portion of the foregut along with the esopha- of the stomach follows just inferiorly and the fundus of
gus, the first portion of the duodenum, as well as the the stomach is superior and to the left extending above
liver, bile ducts, and pancreas. During the fourth week of the gastroesophageal junction, forming a sharp angle
yb
gestation, the foregut is oriented as a craniocaudal tube with the distal esophagus known as the cardiac notch.
with the primitive stomach and first portion of the duo- The corpus or body of the stomach extends and curves
denum forming the caudal end. The ventral mesogastrium inferiorly as a distensible reservoir and forms a sharp
and the dorsal mesogastrium are attached to the stomach medial border called the lesser curvature to the right and
er
anteriorly and posteriorly and suspend the stomach in a lateral border called the greater curvature on the left.
the peritoneal cavity. The greater and lesser curvatures The gastric antrum of the stomach is not anatomically
rg
of the stomach are formed because the dorsal portion of distinguishable but is estimated to be a region from the
the gastric wall grows at a faster rate than the ventral angular notch along the distal lesser curvature to a point
portion.1 At approximately weeks 7 and 8, as the foregut along an inferior line to the distal greater curvature. The
su
develops, it rotates 90 degrees clockwise on its long gastric antrum empties into the pyloric canal leading to
axis so that the ventral mesogastrium is positioned to the pyloric sphincter, a palpable thickened ring of smooth
the right of the stomach and the dorsal mesogastrium is muscle that empties into the first portion of the
://
and hepatoduodenal ligaments and contains the liver, its outermost serosal layer, which is contiguous with the
which grows rapidly and pushes the stomach to the left lesser and greater omenta anteriorly and the anterior wall
portion of the peritoneal cavity. The dorsal mesogastrium of the lesser sac posteriorly. The muscularis externa of
ht
develops into the greater omentum, comprised of the the stomach wall comprises three layers: the outermost
gastrophrenic, gastrosplenic, and gastrocolic ligaments longitudinal muscle layer, the middle circular muscle
and is where the spleen is located during development. layer, and the innermost oblique muscle layer (Fig. 56.4).
This rotation also positions the left vagal nerve trunk The longitudinal muscle layer of the stomach is concen-
anterior to the stomach and the right vagal nerve trunk trated proximally at the gastroesophageal junction and
in the posterior position. The stomach descends as along the greater and lesser curvatures, and subsequently
cephalad structures grow and is eventually located between spreads unevenly over the corpus until joining more
T10 and L3 in the adult. densely near the pylorus. Deep to the longitudinal muscle
fibers, the circular muscle layer covers the stomach
completely and is contiguous with the lower esophageal
GROSS ANATOMY AND ANATOMIC sphincter muscle proximally and forms a thickened band
RELATIONSHIPS OF THE STOMACH at the pylorus distally. The innermost oblique muscle layer
is blended proximally with the circular muscle layer at
The stomach is a dilated cylindrical J-shaped organ that the collar of Helvetius and splays incompletely over the
rests in the epigastric and left hypochondrial region of anterior and posterior gastric walls. The submucosa, which
the abdomen at the level of the first lumbar vertebra (Fig. is also the strength layer of the gastric wall, is the next
56.2). It is bordered anteriorly by the left hemidiaphragm, layer deep to the muscle layers followed by the muscularis
the left lobe of the liver and a portion of the right lobe, mucosae, and finally the mucosa, which contains the
634
Anatomy and Physiology of the Stomach CHAPTER 56 634.e1
ABSTRACT
The stomach is a remarkable organ that aids in digestion,
regulating nutrition, and controlling appetite. The complex
physiologic processes by which the stomach exerts its
endocrine and nutritional functions have been researched
for decades, and there is still much to be learned. This
chapter on the anatomy and physiology of the stomach
aims to equip the surgeon with the detailed knowledge
of not only the gross anatomy and vascular supply of the
stomach but also the physiologic properties behind the
complex process of gastric acid secretion and hormonal
regulation related to digestion.
KEYWORDS
Gastric anatomy, gastric physiology, stomach embryology,
gastric mucosa, gastric motility, gastric digestion, motilin,
ghrelin
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
Anatomy and Physiology of the Stomach CHAPTER 56 635
6th week
(10 mm)
Greater
Lesser
Stomach curvature
curvature 7th week
8th week (16 mm)
(19 mm)
Duodenum
A B
Ventral
mesogastrium
t
ne
Stomach
Adult
k.
Dorsal
mesogastrium Omental bursa
C D
oo
FIGURE 56.1 Positional changes of the developing stomach. (A and B) Anterior view of the stomach rotating on its longitudinal axis.
(C and D) Transverse section of peritoneal attachments during rotation of the stomach. (E) Shape of the stomach at various prenatal
yb
stages and in the adult. ([A–D] Based on Sadler TW. Langman’s Medical Embryology. 5th ed. Baltimore: Williams & Wilkins; 1985.
In Skandalakis LJ, et al. Stomach. In: Skandalakis JE, et al, eds. Skandalakis’ Surgical Anatomy. New York: McGraw-Hill; 2004
[Chapter 15]. [E] From Lewis FT. The development of the stomach. In: Keibel WP, Mall FP, eds. Manual of Human Embryology.
er
Philadelphia: JB Lippincott; 1912. In Skandalakis LJ, et al. Stomach. In: Skandalakis JE, et al, eds. Skandalakis’ Surgical Anatomy.
New York: McGraw-Hill; 2004 [Chapter 15].)
rg
Celiac trunk
su
Gallbladder
ht
Stomach
(posterior wall)
Spleen
Hepatoduodenal
ligament
Pancreas
Right kidney
Gastrocolic
ligament
Lesser Transverse colon
omental space
t
The stomach receives input from both the sympathetic
ne
lamina propria (LP) composed of connective tissue, blood and parasympathetic nervous systems and also originates
vessels, and mucosal epithelium. The inner surface of the afferent fibers of the enteric nervous system (ENS) that
stomach can be visualized as multiple irregular folds, provide input to the sympathetic (via splanchnic nerves)
k.
termed rugae, which help to increase surface area of the and parasympathetic systems (via the vagus). The ENS is
stomach and flatten out to allow the stomach to expand considered the third branch of the autonomic nervous
oo
and accommodate meals. system (the other two being the sympathetic and para-
sympathetic) and although it is relatively poorly understood,
it is recognized that it contains as many neurons as the
VASCULAR, LYMPHATIC, AND NEURAL spinal cord and can function autonomously.2
yb
SUPPLY TO THE STOMACH Presynaptic efferent parasympathetic neurons originate
in the dorsal motor nucleus and travel in the left and
The vasculature of the stomach contains a well-developed right vagus nerves that enter the abdomen through the
er
network of anastomosing vessels that stem from the celiac esophageal hiatus on the anterior and posterior surfaces
trunk (Fig. 56.5). This rich blood supply makes ischemia of the esophagus, respectively (Fig. 56.6). These fibers
rg
of the stomach rare and can make control of gastric synapse with postsynaptic neurons located between the
hemorrhage a significant challenge. The greater and circular and longitudinal muscle layers—the myenteric
lesser omenta contain the majority of the blood vessels (Auerbach) plexus—and within the submucosal (Meissner)
su
supplying the stomach. The left gastric artery is a direct plexus. Afferent fibers originating in the stomach travel
branch from the celiac trunk and courses along the lesser in the vagus and synapse with cell bodies in the nucleus
curvature to anastomose distally with the right gastric of the solitary tract of the brainstem.3
://
artery, most often a branch of the common hepatic artery. Presynaptic efferent sympathetic fibers travel in the
The gastroduodenal artery also branches from the common sympathetic chain alongside the eighth to tenth thoracic
tp
hepatic (proximal to the right hepatic artery) and it vertebrae and synapse with neurons located in the splanch-
supplies the greater curvature with the right gastroepiploic nic (celiac) ganglia before terminating in the gastric
(right gastro-omental) artery. The left gastroepiploic (left neuronal plexuses. Sympathetic afferents from the stomach
ht
gastro-omental) branches from the splenic artery at the have cell bodies located in the dorsal root ganglia of the
superior and proximal portion of the greater curvature thoracic spinal nerves.3
before anastomosing with the right gastroepiploic artery.
The short gastric arteries supply the fundus and proximal
body of the stomach by branching from the splenic hilum,
unlike the other vessels that course through the greater
MICROSCOPIC ANATOMY AND PHYSIOLOGY
and lesser omenta. OF THE STOMACH
Venous drainage of the stomach parallels the arterial
blood supply with eventual drainage into the portal vein. GASTRIC MUCOSA
The left gastric vein (coronary vein) and right gastric vein The mucosal lining of the stomach is characterized by a
course along the lesser curvature and drain directly into simple columnar epithelium (SCE) that uniformly lines
the portal vein. The greater curvature is drained by the the stomach. Surface mucous cells (SMs) make up the
right gastroepiploic vein into the superior mesenteric gastric pits (GPs), which lead to long, branched, tubular
vein and by the left gastroepiploic vein, which empties glands, giving the gastric mucosa a leafy appearance,
into the splenic vein. The splenic vein also drains termed the gastric foveolae. Each gland has distinct regions
the short gastric veins and the inferior mesenteric vein from the surface down: the gastric pit, isthmus, neck, and
and finally joins the superior mesenteric vein to form base (Fig. 56.7). The stomach can be divided into three
the portal vein. In cases of portal hypertension, portal glandular regions and these glands are composed of
Esophagus Fundus
SCE
Cardia
Longitudinal
layer
Pyloric Circular
orifice Pyloric Lesser layer
Muscularis LP
Duodenum sphincter curvature SSE
Oblique
layer
Body
Greater curvature MM
Pylorus
ECG
Gastric folds
B
A Stomach regions, anterior view
t
ne
Simple columnar
k.
Blood vessel
epithelium Lymph vessel
oo
Lamina
Mucosa
propria
yb
Muscularis
mucosae
Submucosa Artery
er
Oblique Vein
layer Subumucosal
nerve plexus
rg
Circular
Muscularis layer Myenteric
nerve plexus
su
Longitudinal
layer
Serosa
P P P
tp
ht
GG
D MM
FIGURE 56.4 The stomach wall: (A) Anterior view of the stomach regions and the muscle layers. (B) Transitional epithelium between the
esophagus and the stomach. Stratified squamous epithelium (SSE) in the esophagus becomes simple columnar epithelium (SCE) in the
proximal stomach. The lamina propria (LP) underlies the epithelium and the muscularis mucosa (MM) is deep to the LP with esophageal
cardiac glands (ECG) pictured. (C) The simple columnar epithelium of the gastric mucosa contains gastric pits leading to gastric glands
with various cell types. Additional layers of the stomach wall are illustrated. (D) Histologic section of the gastric mucosa illustrating the
relation of the gastric pits (P) leading into the gastric glands (GG) inferiorly bordered by muscularis mucosa (MM). (Modified from Mescher
AL. Digestive tract. In: Mescher AL, ed. Junqueira’s Basic Histology. 14th ed. New York: McGraw-Hill; 2016.)
638 SECTION II Stomach and Small Intestine
Branches
to greater Stomach
omentum
L. gastric a.
R. gastroepiploic a.
Splenic a. and v.
Hepatic a. L. gastroepiploic a.
t
Gastro-
ne
duodenal
a. and v.
Pancreas
k.
Sup. pan-
creatico-
Tr.
oo
duodenal a.
colon
Panc. duct
yb
Jejunum
Desc.
Duodenum
Sup. colon
mesenteric
er
Ileocolic a.
Abd.
aorta Inf. mesen-
su
teric a.
FIGURE 56.5 Vascular supply of the foregut. The stomach is shown reflected cephalad and the pancreatic duct is exposed.
://
tp
various cell types: cardiac glands in the cardia, oxyntic and are heavily concentrated in the neck, the basal
glands in the fundus and body, and antral glands in the chief (zymogenic) cells that secrete pepsinogen, and
pyloric portion of the stomach.4 enterochromaffin-like (ECL) cells that produce
ht
Cardiac glands are chiefly comprised of mucous cells histamine—a powerful stimulus for parietal cell acid
along with a few scattered parietal cells, undifferentiated production—located throughout the gland.
cells in the neck, and a majority of endocrine cells at the The antral mucosa is distinct from fundus/body mucosa
base of the glands. The cardiac glands make up the 10- to in its lack of acid-producing cells and greater proportion
30-mm transition zone between the squamous epithelium of gastrin-secreting G cells. Gastric mucosal cells secrete
of the distal esophagus and the oxyntic glands of the an electrolyte-rich solution that aids in churning, mixing,
fundus, and have a primary function of producing mucus. and lubricating food. Gastric fluid also acts as a vehicle
Although thought to be congenital, the expression of for proteolytic enzymes that are active in the fluid phase.
these glands varies among ethnic populations.5 When the The volume and electrolyte composition of gastric fluid
basal half of these glands expresses more parietal cells, depend on stimuli such as vagal/cholinergic tone and
they are termed oxyntocardiac glands. hormonal/paracrine factors (i.e., gastrin, histamine). In
Oxyntic glands are located in the fundus and body healthy individuals, the basal secretory rate of the stomach
of the stomach and are appropriately named for their is more than 60 mL of fluid hourly, which, in experimental
acid-producing functions, based on the Greek oxynein, studies, can increase to more than double that when
meaning “acid-forming.” The main cell types are the stimulated by histamine. Total average daily fluid produc-
surface epithelial cells, the mucous cells located in the tion is more than 1.5 L. The electrolyte composition of
GPs and in the isthmus and neck, the parietal cells gastric fluid is similarly dependent on external stimuli
that secrete hydrochloric acid (HCl) and intrinsic factor and is summarized in Table 56.1.
Anatomy and Physiology of the Stomach CHAPTER 56 639
Left vagus n.
Rt. vagus n.
TABLE 56.2 Important Gastric Secretory Products
t
Modified from Barrett KE. Gastric secretion. In: Barrett KE, et al., eds.
ne
Gastrointestinal Physiology. 2nd ed. New York: McGraw-Hill; 2014
[chapter 3].
k.
FIGURE 56.6 Diagram of the vagal innervation of the human
stomach. (From Menguy R: Surgery of Peptic Ulcer. Philadelphia: whereas deeper secretory cells turn over at a much slower
oo
Saunders; 1976.) rate.
Parietal Cells
yb
Parietal (oxyntic) cells are located in the neck and deeper
TABLE 56.1 Gastric Fluid Volume and Electrolyte
parts of the GGs and secrete HCl and intrinsic factor.
Composition in 200 Healthy Volunteers This cell has a rounded or pyramidal appearance with a
round nucleus and a highly eosinophilic cytoplasm due
er
H+ (mEq/L) 27.1 ± 13.8 95.0 ± 20.6 verted to a hydrogen ion (H+) and a hydroxide ion (OH−)
Na+ (mEq/L) 48.1 ± 15.7 23.4 ± 6.1 (Fig. 56.8). The H+ is pumped into the gastric lumen in
K+ (mEq/L) 13.4 ± 3.1 15.2 ± 2.2 exchange for K+, which is maintained in the cytosol above
su
Cl− (mEq/L) 98.5 ± 20.1 139.9 ± 16.3 chemical equilibrium by the basolateral Na+/K+ ATPase
From Meeroff JC, Rofrano JA, Meeroff M. Electrolytes of the gastric juice
and the sodium/potassium/chloride cotransporter
(NKCC1). Research in recent years has proposed that
://
Knowing the various secretory products produced by This process is catalyzed by the carbonic anhydrase II
gastric cells is important in understanding the grand enzyme. A chloride ion is simultaneously transported
scheme of the stomach’s complex role in digestion (Table across the basolateral membrane into the parietal cell
56.2).4,6 lumen and across the apical membrane to combine with
H+ to form HCl. When the parietal cell is stimulated by
Mucous Cells acetylcholine (Ach), histamine, or gastrin to secrete gastric
Mucous cells are located on the surface and in the neck acid, important intracellular events take place as the cell
of the gastric glands (GGs). SMs that line the stomach shifts from a resting to a secretory state. In particular,
lumen and the GPs have a columnar shape and secrete intracellular canaliculi and tubulovesicles that house the
an alkaline, highly viscous mucous substance rich in H+/K+ proton pumps fuse together and with the apical
bicarbonate ions that helps to protect the stomach mucosa membrane of the cell. This amplifies the working surface
from abrasive food particles and erosive gastric acid. The area of the cell 5- to 10-fold and the concentration of
mucous cells located deeper in the isthmus and neck of proton pumps to the apical membrane increases as well
the GGs secrete a more acidic mucin substance and have as the parietal cell’s power to produce HCl.4
a rounded nuclei with apical secretory granules. These Proton pump inhibitors (PPIs) block acid secretion by
mucous neck cells (MNs) are the anchored pluripotent directly inhibiting the H+/K+ exchange ATPase at the
stem cells that divide to replace all other cell types in the apical membrane. Inhibitors such as omeprazole are weak
gastric gland. The SMs have a 4- to 7-day turnover rate, bases and become protonated in the highly acidic
640 SECTION II Stomach and Small Intestine
GP SM Surface mucous
GP cell (secretes
alkaline fluid
NM containing mucin)
Gastric
NM P pit
Mucous neck
cell (secretes
acidic fluid
P containing
NM
mucin)
t
pepsinogen
ne
P and
C Gastric
gland gastric lipase)
k.
G cell
C
(enteroendocrine
cells that secrete
oo
gastrin into
MM the blood)
MM
P
yb
A C D
FIGURE 56.7 Gastric glands (GG): (A) The long, coiled GGs penetrate the complete thickness of the mucosa, from the gastric pits (GP) to
er
the muscularis mucosae (MM). (B) In the neck of a gastric gland, below the surface mucous cells (SM) lining the gastric pit, are small
mucous neck cells (NM), scattered individually or clustered among parietal cells (P) and stem cells that give rise to all epithelial cells of
the glands. The numerous parietal cells are large distinctive cells often bulging from the tubules, with central nuclei surrounded by
rg
intensely eosinophilic cytoplasm with unusual ultrastructure. Chief cells (C) begin to appear in the neck region. Around these tubular
glands are various cells and microvasculature in connective tissue. (C) Near the MM, the bases of these glands contain fewer parietal
cells (P) but many more zymogenic chief cells (C). Chief cells are found in clusters, with basal nuclei and basophilic cytoplasm. From
su
their apical ends chief cells secrete pepsinogen, the zymogen precursor for the major protease pepsin. Zymogen granules are often
removed or stain poorly in routine preparations. (Both x200; H&E stain) (D) Diagram showing general morphology and functions of major
gastric gland cells. (Modified from Mescher AL. Digestive tract. In: Mescher AL, ed. Junqueira’s Basic Histology. 14th ed. New York:
://
McGraw-Hill; 2016.)
tp
environment immediately surrounding the parietal cell factor binding. Upon reaching the terminal ileum, the
membrane and subsequently form a covalently linked intrinsic factor–B12 complex is endocytosed by specialized
ht
mercapto complex that inactivates the enzyme.9 By blocking epithelial cells. Patients undergoing proximal gastric
the final step in the acid-secretion pathway, PPIs are able resection or total gastrectomy and those with pernicious
to attenuate acid secretion stimulated by gastrin/histamine anemia require parenteral injections of vitamin B12.
and vagal/cholinergic pathways.
Intrinsic factor is a 45-kDa glycoprotein that is secreted Chief Cells
by parietal cells and is required for cobalamin (vitamin Chief (zymogenic) cells are predominantly located in the
B12) uptake in the terminal ileum. Secretion of intrinsic basal regions of the GGs. These protein-secreting cells
factor is independent of acid secretion and is unaffected have an abundance of rough endoplasmic reticulum and
by PPIs and histamine receptor blockers, although a apical granules containing pepsinogen, an inactive precur-
higher gastric pH may inhibit absorption of food-bound sor protein secreted into the gastric lumen by a compound
vitamin B12. Although intrinsic factor is synthesized and exocytosis. Pepsinogen is a 42-kDa proenzyme that
secreted in the acidic environment of the stomach, it undergoes catalytic cleavage in the acidic environment
binds with cobalamin at an optimum pH of approximately of the stomach and is converted to pepsin, which can
7 and is fairly resistant to breakdown by acid and proteolytic activate additional molecules of pepsinogen and effectively
enzymes in the stomach. Vitamin B12 is initially bound by degrades collagen. At higher pH, proteolytic activity is
haptocorrin (R factor), after which, exposure to the higher diminished because of denaturation, with irreversible loss
pH and proteolytic enzymes of the duodenum dissociates of proteolytic function above pH 7.2. Stimuli for the
the haptocorrin–B12 complex and allows for intrinsic secretion of pepsinogen are similar to that of gastric acid
Anatomy and Physiology of the Stomach CHAPTER 56 641
t
ne
k.
oo
yb
er
FIGURE 56.8 Molecular mechanics of parietal cell acid production. The intracellular events after ligand binding to the parietal cell are
depicted. Gastrin binds to the type B CCK receptor and acetylcholine binds to M3 receptors to stimulate phospholipase C (PLC) through
rg
a G protein–linked mechanism. Activated PLC converts membrane-bound phospholipids into inositol triphosphate (IP3), which stimulates
the release of intracellular calcium from intracellular calcium stores. The increase in intracellular calcium leads to the activation of protein
su
kinases, which activate H+/K+-ATPase. Histamine binds to its H2 receptor to stimulate adenylate cyclase, which also occurs through a G
protein–linked mechanism. Activation of adenylate cyclase leads to an increase in intracellular cyclic adenosine monophosphate (cAMP)
levels, which activates protein kinases. Activated protein kinases stimulate a phosphorylation cascade that results in increased levels of
://
phosphoproteins, which activate the proton pump. Activation of the proton pump leads to extrusion of cytosolic hydrogen in exchange
for extracytoplasmic potassium. In addition, chloride is secreted through a chloride channel located at the luminal side of the membrane.
ATP, Adenosine triphosphate; ATPase, adenosine triphosphatase; Gs, stimulatory guanine nucleotide protein; Gi, inhibitory guanine
tp
and include Ach, gastrin, gastrin-releasing peptide (GRP), cells (open) of the pylorus, which secrete gastrin promoting
cholecystokinin, and nitric oxide. Increasing intracellular gastric acid secretion.
calcium is the primary cellular mechanism by which G Cells. Gastrin is produced by G cells located primarily
exocytosis of pepsinogen occurs. in the antral mucosa. It is a crucial stimulus for gastric
acid secretion. Gastrin undergoes post-translational cleav-
Enteroendocrine Cells age and modification prior to its release into the circula-
Enteroendocrine cells are epithelial cells distributed along tion, resulting in several isoforms of the peptide with
the mucosa of the entire digestive tract that release varying biologic activities and half-lives. Gastrin-17 is the
hormones in a paracrine or endocrine fashion. These predominant active form, although varying amounts of
cells are part of the diffuse neuroendocrine system (DNES) gastrin-34 and gastrin-71 are produced in healthy and
and can be categorized as “closed” if the cellular apex is hypersecretory disease states, such as gastrinoma. Stimuli
surrounded by neighboring cells and “open” if the apical for gastrin release include Ach from vagal efferent fibers,
portion of the cell is open to the gastric gland lumen with luminal peptides and amino acids, alkaline gastric pH,
chemoreceptors to sample luminal contents. The various and calcitonin. Gastrin release is inhibited by somatostatin
enteroendocrine cells of the stomach are the enterochro- and acidic pH.
maffin cells (closed), which secrete serotonin and substance The effect of gastrin on acid secretion is indirect and
P to increase gut motility; the D cells of the pylorus, which involves ECL cells, which express the cholecystokinin-2
secrete somatostatin and act on other DNES cells; and G (CCK-2) receptor (a G protein–coupled transmembrane
642 SECTION II Stomach and Small Intestine
t
secretion through direct inhibition of gastrin secretion
ne
from antral G cells, histamine from ECL cells, and acid
from parietal cells. Somatostatin is produced by D cells,
+ which are present in both oxyntic glands of the body and
k.
fundus and the pyloric glands. D cells possess intercalating
Parietal Cell cytoplasmic processes that are in close contact with target
oo
+ + cells and facilitate the paracrine action of somatostatin
on multiple mediators of gastric acid secretion.
The effect of somatostatin is mediated through soma-
yb
FIGURE 56.9 The central role of the enterochromaffin-like (ECL) cell tostatin receptor 2 (SSTR-2), which is expressed in G
in regulation of acid secretion by the parietal cell is depicted. As cells, parietal cells, and ECL cells of the gastric mucosa.
shown, after ingestion of a meal, vagal fibers are stimulated and There is a complex crosstalk between somatostatin signaling
release acetylcholine (cephalic phase). Acetylcholine binds to M3 and the gastrin-histamine axis. Somatostatin binding
er
receptors located on the ECL cell, parietal cell, and G cell to blunts histamine-stimulated translocation of the H+/K+
cause the release of histamine, hydrochloric acid, and gastrin, exchange ATPase to the apical membrane in parietal cells.
rg
respectively. Acetylcholine also interacts with M3 receptors Somatostatin also prevents gastrin-induced histamine
located on the D cell to inhibit somatostatin release. Food within release in ECL cells by the same G protein–inhibitory
the gastric lumen also stimulates the G cell to release gastrin,
mechanism.
su
shown). The principal stimulus for activation of the D cell is antral gastrin levels and low gastric pH exert a negative feedback
luminal acidification (not shown). effect that increases somatostatin levels and serves to
blunt acid secretion. Finally, there is a baseline level of
ht
t
by increasing vagal tone—and, subsequently, histamine Prostaglandins Increase mucosal blood COX-1- and
ne
release—thereby augmenting gastric acid production. (PGE2, PGI2) flow and reduce acid COX-2-
However, ghrelin levels peak prior to meals and drop production locally. expressing
sharply in the postprandial period, indicating that ghrelin epithelial cells
k.
has a primary role in basal rather than inducible acid Neuropeptides Increase prostaglandin Nonadrenergic,
secretion. Ghrelin is also known to be a stimulant of gastric (Bombesin) production (induce noncholinergic
oo
motility and the ghrelin peptide exhibits marked sequence expression of COX-2) efferent neurons
homology with another gastrointestinal hormone, motilin, COX, Cyclooxygenase.
which is discussed in Gastric Motility, later in this chapter.
yb
Finally, ghrelin has been linked with the stimulation of
appetite and feeding, as well as increased adiposity. prevention of gastric mucosal barrier disruption, stimula-
tion of mucus secretion, and stimulation of nonparietal
PROTECTIVE FACTORS OF THE GASTRIC MUCOSA alkaline secretion.13
er
The acidic pH and proteolytic enzymes present in the Neurohormonal peptides are also involved in gastric
gastric lumen combine to facilitate the breakdown of mucosal protection. There is a neural reflex arc involving
rg
dietary proteins, enhance the absorption of nutrients such autonomic sensory neurons and noncholinergic afferent
as iron and vitamin B12, and minimize the threat posed neurons that secrete bombesin in response to noxious
by pathogenic microbes. This environment also poses a chemical stimuli such as capsaicin. Bombesin, in turn,
su
risk of damage to the gastric mucosa itself. Various protec- mediates an increase in mucosal blood flow both directly
tive factors are necessary to prevent injury to and break- and through stimulation of cyclooxygenase (COX), leading
down of the gastric mucosa (Table 56.3). to increased production of prostaglandins.
://
cells of the mucosa. Mucus consists of mucin glycoproteins, Gastric acid production is maintained at a low basal rate
surface phospholipids, and water. Disulfide bonds crosslink in the fasting state by the tonic inhibition of acid secretion
adjacent mucin molecules and their oligosaccharides by somatostatin from gastric D cells. Somatostatin acts in
ht
provide a viscoelastic structure that expands with hydration. a paracrine manner on G cells in the antrum, along with
The mucin layer is further stabilized by trefoil factors, ECL and parietal cells in the fundus and body of the
which are small peptides that interact with the carbohydrate stomach to suppress gastrin, histamine, and acid secretion.
side chains. The hydrophobicity of the mucous layer can Gastric acid secretion as it relates to a meal occurs in
be attributed to phospholipids, and bicarbonate ions three phases: cephalic, gastric, and the intestinal phase.
secreted at the base of the mucous layer provide an Most gastric acid secretion occurs in the gastric phase.10
additional layer of protection from an injuriously low pH Prior to the ingestion of food, olfactory, gustatory,
by neutralization. cephalic, and visual stimuli begin to increase gastric acid
Gastric mucosal blood flow is critical to mucosal health. production and stimulate gastric motility.14,15 Higher brain
Prostaglandins are important regulators of mucosal blood centers send information to the dorsal vagal complex in
flow and crucial mediators of mucosal health. As an response to the premeal stimuli. Subsequently, vagal
example, nonsteroidal antiinflammatory drugs cause output activates enteric nerves to release GRP and Ach.
altered prostaglandin synthesis and can lead to erosive GRP stimulates the antral G cells to release gastrin, which
gastritis and gastrointestinal bleeding. Prostaglandins are activates parietal and chief cells in an endocrine fashion.
a family of long-chain fatty acid derivatives of arachidonic Ach inhibits somatostatin release from D cells resulting
acid with vasoactive and neurohormonal properties. The in disinhibition of gastrin, histamine, and acid release, as
mechanisms of prostaglandin-induced mucosal protection well as the direct stimulation of antral G cells and parietal
are varied and include enhancement of blood flow, cells.
644 SECTION II Stomach and Small Intestine
Once feeding begins and the meal enters the gastric activation of PLC and increases in intracellular calcium.
lumen, chemical and mechanical factors add to the M2 and M4 receptors located on D cells mediate the
continued vagal stimulation from the cephalic phase to decreased somatostatin release seen with increased vagal
promote continued gastric secretion. Luminal amino acids tone.
and short peptides, released from dietary protein by the
action of pepsin from chief cells, activate receptors on G Neuropeptides
cells to release gastrin. Alcoholic beverages, coffee, and The ENS is increasingly recognized as a complex and
dietary calcium also activate gastrin release.16 The stomach dynamic system. As mentioned earlier, there is a complex
distends as it receives a meal (receptive relaxation) and interaction between the ENS and the parasympathetic
stretch receptors active long and short reflex arcs that and sympathetic autonomic nervous systems. Although
stimulate vagal nerves to release Ach to activate parietal efferent signals are primarily conducted via the cholinergic
cells directly, or release Ach to activate ECL cells to secrete neurotransmitter Ach, preganglionic afferents and post-
histamine and GRP to activate G cells to secrete gastrin ganglionic fibers of the ENS utilize a variety of peptide
for indirect gastric acid secretion. Distention and the neurotransmitters including Ach, GRP, vasoactive intestinal
presence of luminal peptides produce continued gastric polypeptide, pituitary adenylate cyclase-activating poly-
acid secretion and increased motility.17,18 peptide, nitric oxide, substance P, and neuropeptide Y.22
Finally, in the intestinal phase, gastric acid secretion These substances may have direct effects on acid secretion
is returned to its basal level by several mechanisms. as in the case of Ach on parietal cells or indirect effects
t
Decreased sensory stimuli and gastric distention following through modulation of gastrin, histamine, or somatostatin
ne
a meal lead to a tapering of the cephalic and gastric phase release.23 Many have important implications in gastric
responses. The gastrin released during the cephalic and mucosal health and response to, and/or prevention of,
gastric phases exerts negative feedback through D cells mucosal injury.24
k.
in the antrum, which releases somatostatin leading to
inhibition of gastrin release.19 Food entering the duodenum GASTRIC MOTILITY
oo
leaves the gastric mucosa exposed to the full acidifying The stomach has a salient role in initiating important
effect of parietal cell proton production leading to che- events in digestion relating to its powerful secretory func-
moreceptor activation and neural reflex release of calci- tion but also because of its complex motility properties.
yb
tonin gene-related peptide (CGRP) near D cells to release The stomach is a great homogenizer and helps to mechani-
somatostatin. This results in restoration of the tonic cally break down food to small particles making digestion
inhibition of somatostatin upon acid secretion and a easier. It also serves as a reservoir with the ability for
return to basal acid production. The marked increase in controlled release of food distally to the small intestine
er
gastric blood flow during this phase boosts gastric cell depending on the content and consistency of meals for
secretory function. ease of assimilation downstream. Approximately 200 kcal/h
rg
modulators of gastric acid secretion and gastric motility. motor complex (MMC) is the term that describes this
Additionally, parasympathetic afferent input from the complex process.16
stomach provides important information on gastric pH, The proximal portion of the stomach, consisting of
://
secretion, and emptying along with visceral sensory input the cardia, fundus, and proximal body, serve as a reservoir,
relating to nausea and satiety. Highlighting the importance whereas the distal body and antrum mix gastric contents.
tp
of afferent signals from the stomach is the finding that The pylorus acts as a sphincter to control the rate at which
75% to 90% of vagal neurons are afferent fibers.20 Nocicep- ingested material exits the stomach and its relaxation is
tive stimuli travel primarily through the sympathetic key for the proper functioning of the MMC. Smooth
ht
afferents. There does appear to be crossover with vagal muscle in the different regions of the stomach determine
parasympathetic afferents that may produce clinically the type of contractions that take place as part of the
confusing pain syndromes as a result of similar sensory MMC, namely, phasic or tonic contractions. Phasic contrac-
pathways for the heart, esophagus, and stomach. This can tions occur in the distal stomach and are initiated and
lead to the mislabeling of upper gastrointestinal pathology terminated in a matter of seconds. Tonic contractions,
such as gastritis or gastroesophageal reflux as anginal on the other hand, occur proximally in the stomach and
heart pain.21 last for several minutes at a time. Each type of contraction
is important in the stomach carrying out its primary
Acetylcholine function in each region.
Presynaptic vagal efferent neurons synapse with neurons There are three major phases of gastric motility: fasting,
of the myenteric and submucosal plexuses that, in turn, accommodation, and postprandial. Fasting motility occurs
stimulate gastrin release from antral G cells, leading to between meals and is characterized by stereotypical move-
increased acid secretion. Postsynaptic fibers also provide ments that last about 100 minutes. This process is known
direct stimulation to parietal cells. Redundancy in this as the MMC, where the stomach is able to do “housekeep-
pathway makes pharmacologic suppression of acid secretion ing” and ensure that any indigestible materials do not
incomplete when targeting only one pathway, as with H2 remain in the stomach and potentially obstruct the lumen,
receptor blockers. Ach binds to the M3 muscarinic recep- causing pathologic conditions such as bezoars. Fasting
tors located on G cells, parietal, and chief cells, causing motility consists of three phases: phase I, motor quiescence;
Anatomy and Physiology of the Stomach CHAPTER 56 645
t
During feeding, MMCs disappear and the stomach acts operation, although they do not appear to influence
ne
as a reservoir, although not in a passive capacity. Feeding gastrointestinal motility in the physiologic state.32 A reflex
is characterized by adaptive and receptive relaxation of arc involving sensory afferent neurons throughout the
gastric smooth muscle in the body and fundus, allowing gastrointestinal tract, interacting with noncholinergic,
k.
expansion of the stomach to several times its resting nonadrenergic efferent neurons, produces the global
volume with minimal increases in intragastric pressure. hypomotility seen following manipulation of the stomach
oo
Adaptive relaxation involves a local reflex arc with activa- or intestine during surgery. These discoveries have impor-
tion of stretch receptors in the gastric wall in response tant implications for the treatment of postoperative ileus,
to elevation of intragastric pressure. Mechanoreceptors a major cause of prolonged hospitalization, increased
yb
relay impulses via afferent sensory neurons in the myenteric costs, and morbidity following surgery.
plexus that lead to release of nitric oxide and relaxation
of smooth muscle fibers.28 The adaptive relaxation reflex Motilin
remains largely intact following proximal vagotomy and Motilin is a 22-amino-acid peptide synthesized by M cells
er
functions maximally at higher gastric volumes. Alternatively, of the proximal small intestine and is the principal stimu-
receptive relaxation is triggered by passage of food through lant of MMCs. It is a close homologue of another
rg
the gastroesophageal junction and is transmitted by gastrointestinal peptide hormone, ghrelin, which is co-
cholinergic vagal efferent fibers originating in the nucleus secreted along with motilin and exhibits 50% sequence
of the solitary tract. Interruption of vagal innervation to homology.33 Motilin plasma level peaks correspond closely
su
the stomach as in proximal (truncal) vagotomy abolishes to periods of high MMC activity and drop quickly following
receptive relaxation and leads to decreased emptying time initiation of a meal when phase III contractions are
of liquids as a result of increased intragastric pressure abolished. Motilin acts in an endocrine fashion on both
://
during feeding. gastric smooth muscle and nerves of the myenteric plexus
Postprandial gastric motility serves to triturate and mix through binding with the motilin receptor GPR 38, a
tp
stomach contents and empty the meal in a manner that G protein–coupled receptor.34,35 Macrolide antibiotics
maximizes the digestive and absorptive capacity of the such as erythromycin act as motilin receptor agonists
small intestine. Solids and liquids are handled differently resulting in the promotility qualities that are employed
ht
by the stomach as reflected by distinct physiologic mecha- in patients with delayed gastric emptying.
nisms and rates of emptying for each. Liquids tend to
empty rapidly following the initiation of a meal, whereas Ghrelin
solids experience a lag period where emptying is minimal As mentioned earlier, ghrelin is a closely related peptide
for a period of time.29 Human and animal studies have hormone secreted by X/A-like cells of the oxyntic gland
identified several important aspects of differential emptying of the stomach.36 Despite their close relation, ghrelin
of solids and liquids: gastric distention and activation of acts at a separate receptor from motilin and there is no
inhibitory and excitatory reflex arcs modulate intragastric evidence that the two receptors exhibit cross reactivity to
pressure and rate of emptying of liquids28; antral propulsive the two ligands. Ghrelin’s role in regulating gastrointestinal
and triturating contractions regulate the emptying of motility is not completely understood but evolving,
solids29; pyloric and fundal pressure affect emptying of and evidence exists for promotility actions similar to
solids by acting as a peristaltic pump and preventing motilin.
the passage of large particles to the duodenum30; and
the anatomic configuration of the stomach leads to sedi-
mentation of solids in the dependent portion of the
ACKNOWLEDGMENTS
stomach, whereas liquids preferentially flow toward The authors would like to acknowledge Jonathan C. King and
the pylorus leading to rapid emptying of liquids during O. Joe Hines for their contributions to the previous edition’s
the lag phase. chapter.
646 SECTION II Stomach and Small Intestine
REFERENCES 20. Berthoud HR, Neuhuber WL. Functional and chemical anatomy of
the afferent vagal system. Auton Neurosci. 2000;85(1-3):1-17.
1. Skandalakis LJ, Colborn GL, Weidman TA, Kingsnorth AN, Skan- 21. Qin C, Chandler MJ, Miller KE, Foreman RD. Responses and afferent
dalakis JE, Skandalakis PN. Stomach. In: Skandalakis JE, Colburn pathways of C(1)-C(2) spinal neurons to gastric distension in rats.
GL, Weidman TA, et al., eds. Skandalakis’ Surgical Anatomy. New Auton Neurosci. 2003;104(2):128-136.
York: McGraw-Hill; 2004 [chapter 15]. 22. Smith VC, Dhatt N, Buchan AM. The innervation of the human
2. Furness JB, Costa M. Types of nerves in the enteric nervous system. antro-pyloric region: organization and composition. Can J Physiol
Neuroscience. 1980;5(1):1-20. Pharmacol. 2001;79(11):905-918.
3. Scharoun SL, Barone FC, Wayner MJ, Jones SM. Vagal and gastric 23. Schubert ML, Peura DA. Control of gastric acid secretion in health
connections to the central nervous system determined by the transport and disease. Gastroenterology. 2008;134(7):1842-1860.
of horseradish peroxidase. Brain Res Bull. 1984;13(4):573-583. 24. Gyires K. Neuropeptides and gastric mucosal homeostasis. Curr Top
4. Barrett KE. Gastric secretion. In: Gastrointestinal Physiology. 2nd ed. Med Chem. 2004;4(1):63-73.
New York: McGraw-Hill; 2014 [chapter 3]. 25. Szurszewski JH. A migrating electric complex of canine small intestine.
5. Huang Q. Controversies of cardiac glands in the proximal stomach: Am J Physiol. 1969;217(6):1757-1763.
a critical review. J Gastroenterol Hepatol. 2011;26(3):450-455. 26. Dooley CP, Di Lorenzo C, Valenzuela JE. Variability of migrating
6. Mescher AL. Digestive tract. In: Junqueira’s Basic Histology. 14th ed. motor complex in humans. Dig Dis Sci. 1992;37(5):723-728.
New York: McGraw-Hill; 2016. 27. Luiking YC, van der Reijden AC, van Berge Henegouwen GP,
7. Kaufhold MA, Krabbenhoft A, Song P, et al. Localization, trafficking, Akkermans LM. Migrating motor complex cycle duration is deter-
and significance for acid secretion of parietal cell Kir4.1 and KCNQ1 mined by gastric or duodenal origin of phase III. Am J Physiol.
K+ channels. Gastroenterology. 2008;134(4):1058-1069. 1998;275(6 Pt 1):G1246-G1251.
8. Heitzmann D, Warth R. No potassium, no acid: K+ channels and 28. Desai KM, Sessa WC, Vane JR. Involvement of nitric oxide in the
gastric acid secretion. Physiology (Bethesda). 2007;22:335-341. reflex relaxation of the stomach to accommodate food or fluid.
t
9. Tutunji MF, Qaisi AM, El-Eswed BI, Tutunji LF. Reactions of sulfenic Nature. 1991;351(6326):477-479.
ne
acid with 2-mercaptoethanol: a mechanism for the inhibition of 29. Camilleri M, Malagelada JR, Brown ML, Becker G, Zinsmeister AR.
gastric (H+-K+)-adenosine triphosphate by omeprazole. J Pharm Sci. Relation between antral motility and gastric emptying of solids and
2007;96(1):196-208. liquids in humans. Am J Physiol. 1985;249(5 Pt 1):G580-G585.
10. Di Mario F, Goni E. Gastric acid secretion: changes during a century. 30. Haba T, Sarna SK. Regulation of gastroduodenal emptying of solids
k.
Best Pract Res Clin Gastroenterol. 2014;28(6):953-965. by gastropyloroduodenal contractions. Am J Physiol. 1993;264(2 Pt
11. Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K. 1):G261-G271.
31. van Helden DF, Laver DR, Holdsworth J, Imtiaz MS. Generation
oo
Ghrelin is a growth-hormone-releasing acylated peptide from stomach.
Nature. 1999;402(6762):656-660. and propagation of gastric slow waves. Clin Exp Pharmacol Physiol.
12. Wang G, Lee HM, Englander E, Greeley GH Jr. Ghrelin—not just 2010;37(4):516-524.
another stomach hormone. Regul Pept. 2002;105(2):75-81. 32. Plourde V, Wong HC, Walsh JH, Raybould HE, Tache Y. CGRP
yb
13. Miller TA. Protective effects of prostaglandins against gastric mucosal antagonists and capsaicin on celiac ganglia partly prevent postopera-
damage: current knowledge and proposed mechanisms. Am J Physiol. tive gastric ileus. Peptides. 1993;14(6):1225-1229.
1983;245(5 Pt 1):G601-G623. 33. Poitras P, Peeters TL. Motilin. Curr Opin Endocrinol Diabetes Obes.
14. Rogers J, Raimundo AH, Misiewicz JJ. Cephalic phase of colonic 2008;15(1):54-57.
er
pressure response to food. Gut. 1993;34(4):537-543. 34. Takeshita E, Matsuura B, Dong M, Miller LJ, Matsui H, Onji M.
15. Katschinski M. Nutritional implications of cephalic phase gastro- Molecular characterization and distribution of motilin family recep-
intestinal responses. Appetite. 2000;34(2):189-196. tors in the human gastrointestinal tract. J Gastroenterol. 2006;41(3):
rg
16. Barrett KE. Gastric motility. In: Gastrointestinal Physiology. 2nd ed. 223-230.
New York: McGraw-Hill; 2014 [chapter 8]. 35. Xu L, Depoortere I, Tomasetto C, et al. Evidence for the presence
17. Schiller LR, Walsh JH, Feldman M. Distention-induced gastrin of motilin, ghrelin, and the motilin and ghrelin receptor in neurons
su
release: effects of luminal acidification and intravenous atropine. of the myenteric plexus. Regul Pept. 2005;124(1-3):119-125.
Gastroenterology. 1980;78(5 Pt 1):912-917. 36. De Vriese C, Delporte C. Ghrelin: a new peptide regulating growth
18. Yeo CJ, Bastidas JA, Schmieg RE Jr, Zinner MJ. Meal-stimulated hormone release and food intake. Int J Biochem Cell Biol.
absorption of water and electrolytes in canine jejunum. Am J Physiol. 2008;40(8):1420-1424.
://
1990;259(3 Pt 1):G402-G409.
19. Meshkinpour H, Dinoso VP Jr, Lorber SH. Effect of intraduodenal
administration of essential amino acids and sodium oleate on motor
tp
E
ndoscopy is the gold standard for providing visual aggregation is a relative contraindication for therapeutic
access to the endoluminal space of the gastrointestinal procedures.
(GI) tract for diagnostic and therapeutic purposes.
As a diagnostic tool, it allows for direct visualization of
the mucosal surface and permits the identification of EQUIPMENT
abnormalities including mucosal changes, polyps, strictures,
and external compression. As a therapeutic tool, the ENDOSCOPY TOWER
t
included instrument channel allows passage for instruments An endoscopy tower is used to concentrate and organize
ne
that can sample mucosa for pathologic examination, treat the equipment needed to perform endoscopy within a
bleeding, dilate strictures, and access other organ patholo- compact and portable unit. It generally includes a digital
gies through the wall of the GI tract. Upper GI endoscopy video processor, which allows connection of the endoscope
k.
allows for examination of the esophagus, stomach, and electronics to provide signal-to-video and post-processing
proximal duodenum down to the ligament of Treitz. The capabilities. It should also allow capture and saving of
oo
small bowel between the ligament of Treitz and the ileoce- still pictures and video for documentation purposes. A
cal valve is more difficult to access although techniques light source is also important, interfacing with the light-
such as video capsule endoscopy and balloon endoscopy guide cables of the scope to provide illumination that
yb
have expanded the roles of endoscopic evaluation and travels to the tip of the endoscope. Finally, a video monitor
therapy. Endoscopy is a growing domain and a majority provides a display capability that allows the endoscopist
of GI pathology either involves the mucosa directly or and assistant(s) to directly visualize the magnified picture
can be identified indirectly from within the lumen of the produced by the camera at the tip of the scope.
er
INDICATIONS AND CONTRAINDICATIONS production of a water jet for lavage through the auxiliary
water channel. A radiofrequency generator or other source
General indications for endoscopy include persistent of energy is usually also bundled, allowing the delivery
su
upper abdominal symptoms despite appropriate medical of monopolar cautery and bipolar/thermal energy. Finally,
therapy or have associated worrisome symptoms such as an accessory insufflator can be used to allow luminal
vomiting, weight loss, anorexia, dysphagia, and odyno- distention with carbon dioxide.
://
suspicious radiographic abnormalities, or surveillance for The modern endoscope includes an objective lens and a
premalignant lesions or conditions such as familial adeno- charge-coupled device (CCD) camera at the tip. A light-
matous polyposis. Indications for therapy include tissue guide system enables transmission of illumination from
ht
sampling and excision, treatment of acute upper GI the light source. Cables allow for angulation and deflection
bleeding (UGIB), foreign body retrieval, dilation or stent- of the instrument tip. The endoscope also supplies chan-
ing of strictures or leaks, and placement of feeding tubes.1 nels for insufflation and optionally an auxiliary water
Relative contraindications for endoscopy under con- channel. Importantly, there is also an instrument channel
scious sedation include inability to tolerate the procedure through which therapeutic modalities can be delivered
or sedation, inadequate patient cooperation, and suspicion (Fig. 57.1). The size of the instrument channel can vary
of perforated viscus. It is possible to perform endoscopy from scope to scope, with diagnostic scopes having smaller
in these cases under general anesthesia, although the channels and therapeutic scopes with larger or even
risks and benefits must be weighed carefully and that the double channels. Mini-scopes also exist to facilitate passage
potential result of the endoscopy should change future through tight strictures. Moreover, the objective and
management. Endoscopy in patients with a suspicion of instrument channels can be placed at the tip of the instru-
perforation is best done with carbon dioxide insufflation ment (end viewing) or to the side of the instrument (side
and the ability to quickly decompress pneumoperitoneum viewing), which are adapted to visualization of specific
surgically if required. Surgical anastomoses are considered regions of the GI tract.
safe to be evaluated endoscopically. Patients with recent
myocardial infarction, stroke, or pneumonia should be PREPARATION
assessed independently for risk of worsening their existing Prior to endoscopy, an assessment of general medical
comorbidity. Coagulopathy or inhibition of platelet condition, medication allergies, the patient’s ability to
647
Diagnostic and Therapeutic Endoscopy of the Stomach and Small Bowel CHAPTER 57 647.e1
ABSTRACT
Upper gastrointestinal endoscopy allows for the direct
visualization and treatment of pathologies involving the
stomach, small bowel, and surrounding organs. Mucosal
changes, polyps, and tumors can be diagnosed, biopsied,
excised, or sampled. Upper gastrointestinal bleeding
lesions and ulcers can be controlled with thermal or
mechanical therapies. Strictures and external compression
can be dilated using hydrostatic balloons or self-expanding
metal stents. Biliary and pancreatic pathologies can be
identified and biopsied or drained transluminally. The
small bowel beyond the ligament of Treitz can be inves-
tigated using video capsule endoscopy and balloon
endoscopy. Endoscopy has thus become a cornerstone in
providing diagnostic and therapeutic capabilities to the
skilled gastrointestinal surgeon.
KEYWORDS
t
ne
Diagnostic endoscopy, peptic ulcer disease, therapeutic
endoscopy, gastrointestinal bleeding, endoscopic mucosal
resection, endoscopic submucosal dissection, self-expanding
metal stent, balloon dilatation, deep enteroscopy, video
k.
capsule endoscopy
oo
yb
er
rg
su
://
tp
ht
648 SECTION II Stomach and Small Intestine
TECHNIQUE
The patient is placed in the left lateral decubitus position.
A bite block is placed to protect the teeth and to allow
easy gliding of the endoscope with limited impedance.
The endoscope is initially passed along, following the
t
Instrument surface of the tongue and curved to view the epiglottis.
ne
Light-guide channel Passing under the epiglottis allows visualization of the
vocal cords. It is important to inspect the vocal cords to
FIGURE 57.1 Example layout of the distal end of the endoscope. rule out any lesions, polyps, cord paresis, or reflux laryn-
k.
gitis. Passing the scope under and to the side of the
arytenoid cartilages, the endoscope is then advanced back
oo
tolerate the procedure, and prior reactions to sedation toward the midline with gentle pressure and air insufflation.
should be undertaken. Prior airway issues should be It is important not to forcefully advance the endoscope
explored. In case of therapeutic intervention, an up-to-date blindly in this location, especially in older patients. The
yb
coagulation panel and platelet count should be checked presence of a pulsion (Zenker) diverticulum or cervical
to ensure it is within safe ranges. If fluoroscopy is consid- osteophytes may impede the passage of the scope. Asking
ered, pregnancy testing should be considered for women the patient to swallow with a flexed neck position, or
of childbearing age. Informed consent should be taken performing a jaw-thrust can help pass the endoscope
er
detailing the risks and benefits of the procedure. beyond the upper esophageal sphincter.
For routine upper GI endoscopy, the last oral intake The mucosa of a healthy esophagus is glistening white,
rg
of solids should be 6 hours and clear liquids 2 hours prior representing the squamous epithelium. The esophagus
to intervention.2 This helps prevent regurgitation and is usually straight, narrow, devoid of remnant liquid or
aspiration and improves endoscopic visualization. In cases solid particles, and shows active peristalsis. It can, however,
su
of known gastroparesis or obstruction, a longer fasting become tortuous in older patients. In achalasia, the
period or the prior insertion of a nasogastric tube to esophagus may become dilated and even sigmoid shaped.
empty the stomach may be appropriate. In GI bleeding, There can be presence of mid and lower esophageal
://
lavage may be performed through a nasogastric tube both diverticula. Further information about the endoscopy of
as a diagnostic tool and to help clear clots and blood, the esophagus can be found in Chapter 7. The gastro-
tp
Typically, patients are first anesthetized using a topical to the columnar epithelium of the stomach. This transition
solution of local anesthetic spray or gargle, to decrease is identified with a change in color from glistening white
the gag reflex. Removable partial and complete dentures to a pinkish-brown. The body and fundus of the stomach
are removed. Adequate monitoring must be maintained also display rugae, a series of ridges and folds along the
throughout the procedure for safety of conscious sedation. internal surface of the stomach. The stomach can be
It includes the use of pulse oximetry, a blood pressure distended to flatten the rugae at this juncture to allow
cuff, and visual cues of respiratory rate and effort. Certain for complete inspection of its mucosal surface.
patients with knowledge of the procedure and low anxiety Upon entering the stomach, it is important to note
may be able to tolerate an unsedated short diagnostic whether there is any liquid or solid food particles present,
procedure. especially in the fundus situated to the left side of the
For the majority of patients, the procedure is more field of view. A large, distended stomach may signify a
comfortable with conscious sedation. Sedation protocols downstream obstruction or poor emptying. It is wise to
usually include a combination of a short-acting narcotic use the suction channel of the scope to aspirate as much
with a benzodiazepine delivered intravenously (e.g., a of the liquid contents in the fundic pool of the stomach
combination of fentanyl and midazolam), with a dose as possible before proceeding further, especially if there
titrated to previous history of sedation, gender, and age.3 is higher risk of regurgitation and aspiration. Pushing the
For older patients, it is recommended to start with a lower endoscope forward with a slight clockwise rotation places
dose and to supplement sedation as required. Adjunctive it along the greater curve of the stomach. As the endoscope
Diagnostic and Therapeutic Endoscopy of the Stomach and Small Bowel CHAPTER 57 649
t
position (Fig. 57.2). This is accomplished by pulling the
ne
endoscope back, which reduces the gastric loop, and
paradoxically advances the scope to the second and third Long position
parts of the duodenum. Examination of the ampulla can
k.
be difficult with a straight viewing scope because it is
situated on the medial curve of the second part of
oo
the duodenum. This area can sometimes be seen on the
right-hand side as the endoscope is pulled back in the
short position. A side viewing scope allows an en face
yb
position toward the ampullary complex and is the best
method for obtaining visualization and access to ampulla.
Once the duodenum is thoroughly examined, it is
possible to pull back into the stomach and perform ret-
er
t
pharmacotherapy and its eradication should be docu-
ne
mented with repeat testing.6
Vascular abnormalities may also be identified and
treated on upper endoscopy. Gastric varices are similar
k.
to, and may be present with or independently from,
FIGURE 57.3 Diffuse gastritis in the body of the stomach. Testing
esophageal varices. They are usually located in the fundal
oo
for Helicobacter pylori must always be done in this situation. area of the stomach. Presence of both esophageal and
gastric varices usually implies portal hypertension, whereas
isolated gastric varices are often caused by splenic venous
yb
hypertension alone. This is often due to splenic vein
thrombosis associated with pancreatitis or pancreatic
malignancy. Varices appear as serpentine dilated submu-
cosal veins that cross over existing gastric rugae and are
er
t
ne
FIGURE 57.5 Mild gastric antral vascular ectasia.
k.
oo
FIGURE 57.7 Multiple fundic gland polyps seen in the proximal part
of the stomach.
yb
gland polyp, which are usually numerous, sessile polyps
found in the fundus of the stomach (Fig. 57.7). They do
not have a stalk, have the same color as the gastric mucosa,
er
FIGURE 57.6 Full retroflex view of the gastroesophageal junction considered malignant, but they often arise in the context
showing a hiatal hernia. of metaplastic chronic gastritis and may contain metaplasia
or dysplasia.10 Polyps greater than 1 cm and with pedun-
ht
t
resection (EMR) and endoscopic submucosal dissection
ne
(ESD) allow for the excision of larger lesions in a piecemeal
or en bloc fashion at a deeper submucosal plane.
FIGURE 57.8 A large sessile duodenal periampullary polyp. The Standard biopsy forceps are an excellent and easy-to-use
k.
ampulla is visible to the right of the polyp. tool for obtaining tissue. For example, forceps can be
used for obtaining random specimen to rule out celiac
oo
disease in the duodenum or H. pylori organisms in the
stomach. They can also be used to obtain partial samples
from larger lesions or complete samples of very small
yb
lesions. The most common model of biopsy forceps today
is the double-bite model. These forceps feature two closable
metal cups and a central needle spike. The central spike
provides deeper purchase by impaling the tissue to be
er
t
ne
k.
B
oo
yb
er
rg
su
://
tp
lesion, thereby lifting it. Failure of the saline lift is a band. EMR can be considered curative for malignant
worrisome sign of deeper invasion of the lesion. The lifted lesions that are completely excised with negative margins,
lesion can then be excised using a snare in standard are limited to the mucosa (T1a), are well-differentiated,
fashion. The second method is EMR with a cap-and-snare and do not have lymphovascular invasion.12 Risks associated
technique, which allows for the excision down into the with EMR include bleeding, perforation, and stricturing.
submucosa and enables accurate examination of the depth Post-EMR, patients are typically placed on proton pump
of invasion of malignant lesions (Fig. 57.11). The distal inhibitors until reassessment of the EMR site at repeat
end of the endoscope is fitted with a cap that is preloaded endoscopy confirms appropriate healing (Fig. 57.12).
with elastic bands, similar to a variceal bander. The tissue Larger lesions that are beyond the size of the EMR
to be excised is centered and suctioned into the cap until cap-and-band device can be removed by ESD. In this
it fills the cap completely. The band is then fired over the technique, the margins of the tumor are marked with
suctioned tissue, trapping the tissue into a pseudopolyp. cautery. A lift is then performed in the submucosal space
A polypectomy snare is then used to remove the trapped with a starch-based volume expander mixed with blue
lesion at its “stalk,” at or above the level of the elastic dye. An incision along the premarked margin is made
654 SECTION II Stomach and Small Intestine
t
ne
k.
oo
yb
B
er
rg
su
://
tp
ht
C
FIGURE 57.11 Steps of a cap-and-snare endoscopic
mucosal resection. (A) A cap fitted with elastic bands is
advanced onto the lesion. (B) The lesion is suctioned up
into the cap and the elastic band is fired, creating a
pseudopolyp. (C) A polypectomy snare is used to excise
the pseudopolyp at the elastic band.
with a needle knife and a cap-fitted endoscope. The cap however, far more technically challenging, has a higher
is then used as a retraction device to continue dissection risk of perforation, and is time consuming to perform.
along the submucosal plane using the needle knife. This
enables precise dissection in the proper submucosal plane, CONTROL OF GASTROINTESTINAL BLEEDING
direct control over the margins of the lesions, and results UGIB is one of the most common indications for thera-
in a higher en bloc resection specimen rate.13 ESD is, peutic intervention in upper endoscopy. Timing of
Diagnostic and Therapeutic Endoscopy of the Stomach and Small Bowel CHAPTER 57 655
t
along the posterior gastric lesser curve or posterior
ne
duodenal wall.18 Lesions with lower risk for rebleeding
include clean-based ulcer with minor oozing and black
FIGURE 57.12 Previously seen sessile duodenal polyp in Fig. 57.8 or red spots. Lower-risk lesions have a good prognosis
k.
has been excised with a saline-lift endoscopic mucosal resection and do not require endoscopic treatment.
technique.
Multiple tools can be used for endoscopic treatment
oo
and can be categorized as injection, thermal, and mechani-
intervention is determined by the clinical scenario and cal therapy.19 The goal of injection therapy is to provide
patients can be categorized as requiring early endoscopy, hemostasis through a combination of the tamponade
yb
surveillance, or discharge. Patients considered to be at a effect due to the volume of the injection itself and vaso-
higher risk for severe UGIB include those older than 60 constricting or sclerosing effect due to the injected
years, patients with known concomitant liver disease compound. The most commonly used compound for
including varices, patients with witnessed hematemesis or injection of nonvariceal bleeding sources is dilute 1 : 10,000
er
hematochezia, which are signs of ongoing bleeding, or epinephrine, which provides local compression and adds
patients with tachycardia or hypotension on presentation.14 a vasoconstrictive effect. Other compounds include normal
rg
These patient characteristics are associated with higher saline solution, which provides local compression alone,
mortality from a bleeding episode and benefit from early and ethanol or sodium morrhuate, which have sclerosing
endoscopic management. A low threshold should be effect. Thrombin, fibrin, or cyanoacrylate glues may provide
su
maintained for use of early endoscopy because the con- primary tissue sealing. Injection therapy alone has been
sequences of missing a severe UGIB are significantly shown to be suboptimal and insufficient to prevent rebleed-
higher than the morbidity of additional nontherapeutic ing. Injection therapy with dilute epinephrine is best used
://
preparation prior to safe performance of endoscopy. mechanical clipping, or sclerosant or sealant injection.20
Proton pump inhibitors should be started prior to endos- Complications from injection are usually related to the
copy, which reduces the severity of the bleeding and the injected compound itself; epinephrine rarely causing
ht
requirement of therapy at endoscopy, but does not change arrhythmias or hypertension, whereas sclerosants have a
clinical outcomes and mortality.15 Patients with ongoing higher rate of local effects such as tissue necrosis, strictures,
UGIB may require volume repletion and blood product bleeding, and perforation.
transfusion. Careful monitoring in an intensive care unit Thermal options include contact methods and non-
may be required. Coagulation parameters must be obtained contact methods and are based on delivery of energy-
and any anticoagulant should be reversed but should not inducing direct tissue coagulation, contraction, and
delay endoscopy. These patients may be at high risk of desiccation. Contact methods include the delivery of
aspiration and may benefit from airway control and protec- monopolar and bipolar energy and require the use of a
tion using endotracheal intubation. It is possible to perform radiofrequency generator. Monopolar energy completes
lavage of the stomach by instilling warm saline in a large- a circuit through an active electrode and a grounding
bore naso- or orogastric tube as a diagnostic tool and to pad. Multiple instruments such as hot biopsy forceps and
facilitate future endoscopic visualization. An aspirate snares can serve as the active electrode. Bipolar energy
yielding blood confirms UGIB, but a negative sample does includes both electrodes at the tip of the instrument,
not preclude bleeding and can miss up to 20% of bleeding. completing the circuit locally. Bipolar energy can be used
Promotility agents such as intravenous erythromycin can directly by completing the circuit at the level of the tissue
also be used prior to endoscopy to improve clearance of to be treated itself or can be used as part of a heater
the upper GI tract but has not been shown to improve probe, using thermocoupling to generate a constant
clinical outcomes.16 temperature and induce coagulation through direct heat
656 SECTION II Stomach and Small Intestine
transfer. Contact probes are advantageous in allowing for powder achieves hemostasis by acting as a scaffold for
direct pressure tamponade of bleeding using the probe platelets and coagulation factors, promoting thrombus
itself and cause a deeper burn than noncontact methods. formation. Investigational methods for tissue approxima-
This deeper burn does increase the risk, albeit rare, of tion include over-the-scope clips and endoscopic suturing.
peptic ulcer perforations and precipitation of bleeding. For larger lesions, a “bear-claw” type clip includes an
Some contact probes are combined with an injection automatically closing spring mechanism and comes pre-
needle allowing for delivery of combination therapy using loaded on an applicator cap in the open position.23 Once
a single instrument. Noncontact methods include argon the scope is properly positioned, the clip can be fired
plasma coagulation, which uses the flow of argon gas from allowing the two parts of the clip to close onto the desired
the tip of the instrument.21 The argon gas is ignited using tissue. Over-the-scope clips lead to a deeper purchase
an electrode, turning it into argon plasma, which travels compared to standard endoscopic clips and may remain
to the nearest conductive tissue, and is often used for in position much longer. Endoscopic suturing devices can
creating a superficial burn. The superficiality of the burn also provide direct mechanical approximation but are
renders complications rare, but the argon gas itself can unwieldy to use in the context of UGIB.
create overdistention of the GI tract, subcutaneous After initial endoscopic treatment of patients with
emphysema, and pneumoperitoneum. high-risk stigmata, it is advisable to monitor the patient
The most commonly deployed mechanical hemostatic for at least 72 hours and continue proton pump inhibitors
device is the endoscopic clip. Standard through-the-scope to prevent risk of rebleeding.24 Appropriate management
t
endoscopic clips are loaded through the instrument of UGIB with combination therapy is highly effective and
ne
channel and are generally made of a two-pronged metal initial hemostasis is achieved in up to 95% of cases. If
clip attached to a deployment handle that allows opening, appropriate initial management was performed, routine
closing, and firing of the clip. Their handles often allow repeat endoscopy in 24 hours is not necessary. Nonetheless,
k.
rotation of the clip to facilitate positioning. Endoscopic there is a risk for rebleeding in 5% to 10% of all cases.
clips provide direct tissue approximation and superficial Patients with rebleeding should undergo a second endo-
oo
vessel closure and can also be used for closure of perfora- scopic assessment and treatment if appropriate. If endo-
tion (Fig. 57.13). Through-the-scope clips generally dis- scopic hemostasis fails, second-line options include
lodge within a week. Other methods for mechanical percutaneous embolization and surgical treatment, which
yb
hemostasis include detachable loops, which are similar is further described in Chapter 59. In patients with peptic
to loop ligature devices. These are most useful in control- ulcer disease, H. pylori testing should be carried out. A
ling the stalk of a lesion to be snared, thereby preventing negative H. pylori test during an episode of bleeding is
bleeding once the lesion is excised. Endoscopic band considered to have lower sensitivity and further testing
er
ligators were initially created to band esophageal varices should be repeated at a later time. H. pylori infections
and consist of elastic bands loaded onto a cap, but can should be eradicated and eradication should be docu-
rg
also be used to ligate gastric varices and Dieulafoy lesions. mented to reduce the recurrence rate of peptic ulcer
Emerging methods for hemostasis include Hemospray disease. In patients with ulcer complications due to NSAIDs,
(Cook Medical, Winston-Salem, North Carolina), an their cessation must be considered. If NSAIDs cannot be
su
inorganic mineral powder that is sprayed onto the bleeding discontinued, concomitant use of a proton pump inhibitor
lesion using a carbon dioxide cartridge.22 The adherent may decrease risk of recurrent bleeding, although there
is still associated risk.
://
t
ne
k.
FIGURE 57.16 Example of a covered self-expanding metal stent.
The stent flares out at both ends to decrease risk of migration.
FIGURE 57.14 Image through an endoscopic band ligator after
oo
placement of a band. This technique can also be used as first
step in the cap-and-snare technique for endoscopic mucosal
resection. and pressurization of the balloon through the catheter
yb
to specific pressures and sizes. It is not uncommon to
require multiple balloon dilations. It is recommended to
size the initial dilation to the tightness of the stricture
and use successively larger balloon sizes over time to
er
t
loaded onto the guidewire for delivery through the stricture
ne
either through the scope or beside the scope. Dilation
FIGURE 57.17 Picture of an over-the-scope clip placed to help or deployment can then proceed using a combination of
prevent migration of a fully covered stent placed across a fluoroscopic and endoscopic guidance.
k.
strictured gastrojejunostomy. Endoscopic treatment of stricturing lesions is not
without risk. Balloon dilation of tight strictures or larger
oo
(>15 mm) balloon sizes carries a non-negligible risk of
perforation and peritonitis. Tearing of the mucosa and
submucosa at the site of dilation may also induce bleeding.
yb
A stent during its expansion phase also causes abdominal
pain and a potential risk for perforation, especially if the
stent is positioned across a corner and there is pressure
on the edges of the stent. Placement of a duodenal stent
er
TRANSLUMINAL THERAPY
Although endoscopy is best adapted to treat lesions inside
ht
FIGURE 57.18 View of the antrum after deployment of an the lumen of the GI tract, the proximity of other organs
uncovered duodenal stent with prepyloric position of the proximal
to the stomach and duodenum allows pathologies in
end of the stent for palliative treatment of duodenal obstruction by
adjacent organs to be treated transluminally. The most
pancreatic cancer.
common example of this is a percutaneous gastrostomy,
which is described in Chapter 58. Another example is the
be useful in treating fistulas, anastomotic leaks, and perfora- drainage of pancreatic pseudocysts by creating an endo-
tions (Fig. 57.17). They are also easier to reposition and scopic cystogastrostomy.
remove after insertion. Uncovered stents, on the other In general, we wait 6 to 8 weeks after the episode of
hand, allow tissue ingrowth that prevents migration, but pancreatitis to allow for the pseudocyst wall to mature
can ultimately lead to stent occlusion due to ingrowth before drainage of large symptomatic non-resolving
(Fig. 57.18). Both types of stent usually also feature flared pseudocysts.32 This image-guided endoscopic technique
ends to further increase the radial holding pressure and is usually done under endoscopic ultrasound guidance.
decrease risk of migration. SEMS placed across a malignant Assessment of patient tolerance is important because the
stricture will gradually dilate the stricture as the stent procedure can be uncomfortable and it is often better
returns to its original shape. Optimal placement of a performed under general anesthesia. Preprocedure
SEMS is centered on the stricturing lesion as the stent antibiotics are given. If gallstone pancreatitis is suspected,
undergoes a certain amount of foreshortening as it prior clearance of the biliary tract with standard endoscopic
expands. retrograde cholangiopancreatography is recommended.
Diagnostic and Therapeutic Endoscopy of the Stomach and Small Bowel CHAPTER 57 659
t
ne
k.
FIGURE 57.19 A fluoroscopic view after insertion of two self- FIGURE 57.20 A fluoroscopic view after insertion of a double pigtail
oo
expanding metal stents, a biliary and duodenal stent, for palliation stent serving as cystogastrostomy for pseudocyst drainage. A
of pancreatic cancer. There is external compression on the transpapillary stent is also seen draining the same pseudocyst.
duodenum as seen from the narrowing of the duodenal stent.
yb
This malignant stricture will be expanded as the stent regains its
shape over the course of the next 2 days.
er
t
ne
for easier future pancreatic debridement. More investi-
gational transluminal therapies include the potential for
leaving the luminal space altogether by creating a gas-
k.
trostomy and entering the peritoneal cavity. This experi-
mental branch of minimally invasive surgery is termed FIGURE 57.23 Angiodysplasia as seen from a video capsule. This
oo
NOTES and features “scarless” abdominal operations such is a common cause for obscure gastrointestinal bleeding in the
as cholecystectomy and appendectomy, which are done small intestine.
through the wall of the stomach or vagina.
yb
ENDOSCOPY OF THE SMALL BOWEL
The midgut, defined as small intestine between the ampulla
er
a push-and-pull technique. In selected patients, it is possible 5. Marshall B, Warren JR. Unidentified curved bacilli in the stomach
to combine antegrade and retrograde balloon endoscopy of patients with gastritis and peptic ulceration. Lancet. 1984;323
(8390):1311-1315.
to examine the entire small bowel.38 However, it is more 6. Chey WD, Wong BC. American College of Gastroenterology guideline
common to use video capsule endoscopy to identify and on the management of Helicobacter pylori infection. Am J Gastroenterol.
situate the lesion first followed by balloon enteroscopy to 2007;102(8):1808-1825.
perform therapy.39 Therapy can be delivered through the 7. Jabbari M, Cherry R, Lough JO, Daly DS, Kinnear DG, Goresky CA.
Gastric antral vascular ectasia: the watermelon stomach. Gastroenterol-
instrument channel with specialized long instruments. ogy. 1984;87(5):1165-1170.
This enables tissue sampling, hemostasis, injection, stricture 8. Carmack SW, Genta RM, Schuler CM, Saboorian MH. The current
dilation, and foreign body removal in the small intestine. spectrum of gastric polyps: a 1-year national study of over 120,000
Balloon enteroscopy can also be used to access the duo- patients. Am J Gastroenterol. 2009;104(6):1524-1532.
denum and ampulla in patients with surgically altered 9. Genta RM, Schuler CM, Robiou CI, Lash RH. No association between
gastric fundic gland polyps and gastrointestinal neoplasia in a study
anatomy.40 However, balloon enteroscopy requires advanced of over 100,000 patients. Clin Gastroenterol Hepatol. 2009;7(8):849-854.
endoscopic skill, is particularly time consuming, and may 10. Jain R, Chetty R. Gastric hyperplastic polyps: a review. Dig Dis Sci.
be poorly tolerated by patients without general anesthesia. 2009;54(9):1839-1846.
Potential complications include perforation, deep mucosal 11. Hiki N, Yamamoto Y, Fukunaga T, et al. Laparoscopic and endoscopic
cooperative surgery for gastrointestinal stromal tumor dissection.
tears, and acute pancreatitis. Surg Endosc. 2008;22(7):1729-1735.
With the advances in deep enteroscopy, the use of 12. Ono H, Kondo H, Gotoda T, et al. Endoscopic mucosal resection
intraoperative enteroscopy has been reduced. Despite for treatment of early gastric cancer. Gut. 2001;48(2):225-229.
t
balloon endoscopy, total endoscopy is not feasible in all 13. Oka S, Tanaka S, Kaneko I, et al. Advantage of endoscopic submucosal
ne
patients and there often remains a part of the small bowel dissection compared with EMR for early gastric cancer. Gastrointest
Endosc. 2006;64(6):877-883.
that cannot be accessed by deep enteroscopic techniques. 14. Blatchford O, Murray WR, Blatchford M. A risk score to predict
This is especially the case for patients with previous need for treatment for upper-gastrointestinal haemorrhage. Lancet.
k.
abdominal surgery, bowel obstruction, or coagulopathy.41 2000;356(9238):1318-1321.
Intraoperative enteroscopy is associated with a higher rate 15. Sreedharan A, Martin J, Leontiadis GI, et al. Proton pump inhibitor
treatment initiated prior to endoscopic diagnosis in upper gastro-
oo
of complications, including wound infection and intestinal intestinal bleeding. Cochrane Database Syst Rev. 2010;(7):CD005415.
ischemia. As such, it should not be used as a first-line 16. Barkun AN, Bardou M, Martel M, Gralnek IM, Sung JJ. Prokinetics
diagnostic tool to limit negative surgical explorations. in acute upper GI bleeding: a meta-analysis. Gastrointest Endosc.
yb
2010;72(6):1138-1145.
17. Rockall TA, Logan RF, Devlin HB, Northfield TC. Risk assessment
CONCLUSION after acute upper gastrointestinal haemorrhage. Gut. 1996;38(3):
316-321.
Upper endoscopy has become a cornerstone in providing
er
19. Hwang JH, Fisher DA, Ben-Menachem T, et al. The role of endoscopy
therapeutic endoscopy has provided the skilled GI surgeon in the management of acute non-variceal upper GI bleeding. Gas-
with an increasing variety of techniques and novel instru- trointest Endosc. 2012;75(6):1132-1138.
ments. These endoscopic techniques are often improve-
su
22. Sung JJ, Luo D, Wu JC, et al. Early clinical experience of the safety
and effectiveness of Hemospray in achieving hemostasis in patients
ACKNOWLEDGMENT with acute peptic ulcer bleeding. Endoscopy. 2011;43(4):291-295.
ht
31. Maire F, Hammel P, Ponsot P, et al. Long-term outcome of biliary in patients with obscure gastrointestinal bleeding. Am J Gastroenterol.
and duodenal stents in palliative treatment of patients with unresect- 2005;100(11):2407-2418.
able adenocarcinoma of the head of pancreas. Am J Gastroenterol. 37. Khashab MA, Pasha SF, Muthusamy VR, et al. The role of deep
2006;101(4):735-742. enteroscopy in the management of small-bowel disorders. Gastrointest
32. Bergman S, Melvin WS. Operative and nonoperative management Endosc. 2015;82(4):600-607.
of pancreatic pseudocysts. Surg Clin North Am. 2007;87(6):1447-1460, 38. Xin L, Liao Z, Jiang YP, Li ZS. Indications, detectability, positive
[ix]. findings, total enteroscopy, and complications of diagnostic double-
33. Varadarajulu S, Bang JY, Sutton BS, Trevino JM, Christein JD, Wilcox balloon endoscopy: a systematic review of data over the first decade
CM. Equal efficacy of endoscopic and surgical cystogastrostomy for of use. Gastrointest Endosc. 2011;74(3):563-570.
pancreatic pseudocyst drainage in a randomized trial. Gastroenterology. 39. Pasha SF, Leighton JA, Das A, et al. Double-balloon enteroscopy
2013;145(3):583-590.e1. and capsule endoscopy have comparable diagnostic yield in small-
34. Shah RJ, Shah JN, Waxman I, et al. Safety and efficacy of endoscopic bowel disease: a meta-analysis. Clin Gastroenterol Hepatol. 2008;6(6):
ultrasound-guided drainage of pancreatic fluid collections with 671-676.
lumen-apposing covered self-expanding metal stents. Clin Gastroenterol 40. Itoi T, Ishii K, Sofuni A, et al. Single-balloon enteroscopy–assisted
Hepatol. 2015;13(4):747-752. ERCP in patients with Billroth II gastrectomy or Roux-en-Y anasto-
35. Widmer J, Singhal S, Gaidhane M, Kahaleh M. Endoscopic ultrasound- mosis (with video). Am J Gastroenterol. 2010;105(1):93-99.
guided endoluminal drainage of the gallbladder. Dig Endosc. 41. Bonnet S, Douard R, Malamut G, Cellier C, Wind P. Intraoperative
2014;26(4):525-531. enteroscopy in the management of obscure gastrointestinal bleeding.
36. Triester SL, Leighton JA, Leontiadis GI, et al. A meta-analysis of the Dig Liver Dis. 2013;45(4):277-284.
yield of capsule endoscopy compared to other diagnostic modalities
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
CHAPTER
Access and Intubation of the Stomach and
Small Intestine 58
David S. Shapiro
| Stephanie C. Montgomery
I
ntubation of the gastrointestinal (GI) tract occurs in Fig. 58.1). The feasibility of placement, potential dura-
frequently in the course of patient care. Enteral access, tion of use, and route of enteral access are equally
whether via the nasal or percutaneous route, is procured important considerations in determining the optimal
in the majority of instances for decompression or nutrition. intestinal intubation for nutrition. Gastric access for feeding
Intestinal access is indicated for diagnostic and therapeutic may be of little value or even detrimental in patients with
reasons in a variety of disorders, including bowel obstruc- a high risk for aspiration or impaired gastric emptying.
tion, gastric outlet obstruction, gastroesophageal reflux, Decisions should be based on consideration of underlying
t
GI bleeding, and disorders of motility (Table 58.1). Despite medical and comorbid conditions, the anticipated length
ne
the frequent need and indication for gastric and small of time that enteral access will be required, and the setting
intestinal intubation in modern medical and surgical in which it will occur. Some conditions may offer relative
practices, the means of access and the access devices contraindications or completely preclude enteral intuba-
k.
themselves carry innate risks that must be considered. tion. Obstructions of the nasopharynx, esophagus, or
Serious, even potentially fatal, complications may result proximal stomach are absolute contraindications to
oo
from the placement or management of enteral tubes. nasoenteric intubations, and usually contraindicate any
Feasibility, appropriateness, timing, and route of access endoscopically placed tubes. Coagulopathy, ascites, obesity,
must all be considered to determine the proper patient previous abdominal surgery, and gastroesophageal varices
yb
and procedure. are all relative contraindications to enteral tube placement
Providers who are intubating the stomach and small by any method.
intestine should always consider potential complications
of tube placement when examining possible indications.
er
and an indwelling tube may compromise this goal as a descriptions of nasogastric tubes and intestinal intubation
result of common in-place protocols and patient interest. date from the 17th century.3 Modern tubes are known
Two meta-analyses of more than 3000 patients support eponymously for the individual who introduced them
://
decompression in selective postoperative settings only into clinical practice. In 1921, Levin described a single
(extensive adhesiolysis, known gastroparesis, mechanically lumen catheter fenestrated at the distal end for decompres-
tp
ventilated patients, etc.), and these authors suggest that sion with low intermittent suction or gravity drainage, or
other routine postoperative use should be abandoned.1 feeding.4 A modification of the Levin tube is now widely
Any benefit of gastric decompression carries a concomitant used and known as a Salem sump.4a The Salem sump tube
ht
risk of aspiration, sinusitis, and pressure-related skin and has a second lumen that permits air to be withdrawn into
soft tissue ulceration, as described. There are also inherent the stomach, or sump, during suctioning, thereby avoiding
procedural risks, including nasopharyngeal injury, epistaxis, adherence of the tube to the gastric mucosa, and possible
and even pneumothorax. Significantly more pulmonary injury. This tube is used most commonly today for GI
complications occurred in patients with nasogastric tubes decompression in the setting of gastroparesis, mechanical
placed routinely, although there was no difference in bowel obstruction, or functional inertia of the bowel or
wound-related complications when compared with selective ileus.
placement of tubes for vomiting and gastric distention.2 In 1934, Miller and Abbott first introduced a long,
Patients undergoing laparotomy or other surgical balloon-tipped intestinal tube designed to pass into the
procedures may be intolerant of intragastric nutrition or intestine via gentle advancement and peristalsis; subsequent
medication administration in the early postoperative modifications included percutaneous, weighted, multilu-
period. Akin to GI intubation for decompression, the men, and silicone models. Temporary placement of a
benefit of accessing the GI tract for nutritive and/or long tube into the small intestine for decompression was
pharmaceutical reasons must be weighed against the described by White in 1956 and was later popularized by
potential risks. Further, the addition of a surgically created Baker, who devised his own eponymous tube.5–7 Generally,
enteral access must be considered if the patient is antici- long tubes have weighted or balloon-tipped ends and are
pated to be unable to take sufficient calories by mouth intended to pass distally to provide intestinal or gastric
in the longer postoperative period (see the decision tree decompression.
663
Access and Intubation of the Stomach and Small Intestine CHAPTER 58 663.e1
ABSTRACT
Intubation of the gastrointestinal (GI) tract occurs fre-
quently in the course of patient care. Enteral access,
whether via the nasal or percutaneous route and gastric
or intestinal in destination, is procured in the vast majority
of instances for decompression or nutrition. Intestinal
access is indicated for diagnostic and therapeutic reasons
in a variety of disorders, including bowel obstruction,
gastric outlet obstruction, gastroesophageal reflux, GI
bleeding, and disorders of motility. Despite the ubiquitous
need and indication for gastric and small intestine intuba-
tion in modern medical and surgical practices, both the
means of access and the access devices themselves carry
innate risks that must be considered. Serious, even poten-
tially fatal, complications may result from the placement
or management of enteral tubes. Feasibility, appropriate-
ness, timing, and route of access must all be considered
to determine the proper patient and procedure.
t
ne
KEYWORDS
Stomach, small intestine, gastrostomy, enterostomy,
percutaneous endoscopic gastrostomy, intubation, access,
k.
nutrition
oo
yb
er
rg
su
://
tp
ht
664 SECTION II Stomach and Small Intestine
TABLE 58.1 Enteral Access: Common Methods With Indications and Contraindications
t
7th ed. Philadelphia: Saunders; 2013.
ne
Indication Transnasal
Short-term
k.
GASTRIC SUMP
Question
oo
Decision
No Open
Yes GASTROSTOMY
Does patient have
Possible
yb
history of multiple
abdominal Percutaneous
procedures? fluoroscopic or CT-
Patient requires Can patient
er
guided
gastrointestinal Long-term undergo GASTROSTOMY
decompression EGD?
rg
Patient requires
enteral access for
Transnasal
nutrition and/or Anticipated
Short-term GASTRIC
://
Open
JEJUNOSTOMY
Does patient have
Nutrition Insufficient history of multiple
Long-term Percutaneous
per os calorie intake abdominal
procedures? fluoroscopic or CT-
guided
JEJUNOSTOMY
FIGURE 58.1 Decision tree for gastric/small intestine intubation and methodologies for possible techniques and devices. CT, Computed
tomography; EGD, esophagogastroduodenoscopy; per os, by mouth. (Created by David S. Shapiro, MD.)
Nasoenteric tubes designed for feeding are similar to for passage and manipulation. The most widely used of
long tubes and are intended to traverse the pylorus, but these tubes, introduced by Dobbie and Hoffmeister in
unlike decompressive tubes, they are generally of smaller the 1970s, is now recognized as the Dobbhoff tube. The tip
caliber and made of softer materials than standard sump of this tube is slightly larger and heavier than the remainder
or long tubes. These tubes often require a stiffening stylus of the catheter, which may or may not facilitate passage
Access and Intubation of the Stomach and Small Intestine CHAPTER 58 665
with the pylorus.8 Unassisted bedside placement into a tubes demonstrated no advantage.10 Although others
postpyloric position can be simple, and a multitude of have described long tubes as a secondary treatment and
methodologies have been suggested. Promotility agents, have anecdotally noted value in this legacy therapy, most have
patient positioning, insufflation of air, and other methods abandoned it.
have been advocated to assist with advancement into a
postpyloric position, but results are mixed. Endoscopic,
radiologic, magnetic, and electromagnetic methods have
CONTRAINDICATIONS
been described, and will be discussed later. Soft feeding tubes (Dobbhoff, Keofeed, Cortrak and similar
tubes) are smaller bore and useful for gastric or postpyloric
feeding and may be placed at bedside transnasally. Naso-
INDICATIONS enteric access is a simple, useful, and reasonably comfort-
Obstruction and the need for decompression are the most able means of enteral access when desired in patients with
common indications for nasogastric and nasoenteric indications. Contraindications to nasoenteral access include
intubation. Less commonly, intubation of the stomach or nasopharyngeal obstruction, esophageal obstruction or
small intestine is used for diagnostic or therapeutic means, perforation, recent foregut manipulation or surgery, and
including gastric lavage and evacuation of gastric contents craniofacial trauma. While orogastric intubation is the
in the initial management of upper GI bleeding or toxic preferred route for access in the presence of craniofacial
ingestion. Diagnostic uses are numerous and include trauma, it may not be practical in the patient without a
t
aspiration to determine the presence of drugs or toxins, secured airway. Coagulopathy is also an important con-
ne
measurement of gastric secretions, volume of output, or traindication when intubation is placed for nonurgent
pH, and for the procurement of specimens for culture purposes to avoid epistaxis or other bleeding.
of Mycobacterium or Helicobacter pylori. Therapeutic uses
k.
for gastric and enteral intubation are well defined. METHODS OF BEDSIDE INTUBATION
Decompression of air or enteric contents is very common,
oo
and is often used in the setting of ileus, mechanical bowel Nasoenteric intubation is easily done at the bedside, but
obstruction, gastric dilation, perioperative gastric drainage, despite the simplicity of the procedure, it can have several
or reduction of aspiration risk in select patients. The pitfalls leading to complications. Patients vary in their
yb
routine use of postoperative nasogastric decompression level of cooperation, alertness, and cognitive capacity.
after abdominal surgery has fallen out of favor. The evi- Consent should be obtained according to institutional
dence does suggest that selective use in patients with the requirements, and should include an assessment of
indications listed earlier, including chronic nausea and benefits, risks, and experience.
er
vomiting, is associated with more frequent pulmonary For patients who are awake, alert, and cooperative,
complications than routine postoperative tube decompres- Fowler position is helpful with a 90- degree angle preferred.
rg
sion.9 Decompression is integral to the management of A chair may be used, but a stretcher or bed may provide
intestinal obstruction; it relieves any advancement of fluid better patient comfort. The patient’s neck should be
or gas from the stomach and can be used to determine slightly flexed to avoid endotracheal placement. The
su
improvement because the volume of output will decrease patient should be in a quiet room because distractions
if the obstruction improves. can be problematic. The patient should understand the
In terms of decompressive treatment of intestinal reason for the procedure, the steps involved, and be
://
obstruction, nasogastric decompression was often sufficient prepared for the uncomfortable nature of nasogastric
to relieve the obstruction from the influx of air and fluid. intubation. Parenteral anxiolytic and analgesic agents are
tp
In the case of partial intestinal obstruction, decompression usually not necessary, and can complicate appropriate
may effect relief of obstruction within 48 hours. Persistent passage.
obstructions will warrant further diagnostic investiga- Assessment for nasal passage patency is important,
ht
tion and possible operative management. In patients especially in a patient with a history of septum abnormality.
with suspected complete intestinal obstruction, nasogastric An emesis basin and protective barrier (towel, drape)
intubation is important in the preoperative resuscitative may be helpful to the patient. Inhalation through the
period to decompress the stomach and minimize aspira- nose with each nostril sequentially obstructed can help
tion, but surgical management remains the mainstay of decide which passage to use.
therapy. The tube should be warmed by sliding it repeatedly
Intraluminal plication of at-risk bowel following through gloved hands to soften the structure and create
extensive adhesiolysis using a long tube (Baker, Cantor, a slight curve in the tube. Most tubes are marked with
others) has been described and evaluated in the litera- centimeter indicators to identify the length of the indwell-
ture. Although the technique is encouraged by some, it ing section, but some may be unlabeled. The correct
has gradually fallen out of favor because complications distance of insertion should be about 50 cm for intragastric
associated with an enterostomy-placed Baker tube are placement, and usually more than 65 cm for postpyloric
prohibitive. Nasally introduced tubes have been suggested placement.
as having efficacy in decompression of partially obstructed Prior to insertion, a water-based lubricant with or
bowel, but the results are mixed. However, without without local topical anesthetic (2% lidocaine, viscous)
gastric decompression, symptomatic relief from nausea should be applied to the tube and to the nasal passages.
and/or emesis may not be achieved. One randomized Anesthetic sprays (benzocaine, butamben) may also be
control study comparing short and long decompressive used, but will not offer any lubrication.
666 SECTION II Stomach and Small Intestine
The tube is inserted into the nostril with a trajectory only method used to confirm location. Aspiration of
toward the posterior nasopharynx, parallel to the patient’s enteric or gastric material may also offer confidence in
ipsilateral angle of the jaw or pinnae. The tube should some settings, but the tube contents are unclear during
not be inserted in the cephalad direction because this passage and there is no proof of tip location. Radiographic
will result in the tube curling in the nasopharynx or evidence remains the mainstay for tip location confirmation
trauma to the nasal mucosa. Maintaining the tube along and should be performed routinely after intubation.
the floor of the nasal passage may facilitate entry into the Contrast may be required in some settings to determine
posterior pharynx. As the tube reaches the posterior the course of the tube, and thinned solutions of barium
nasopharynx and some mild resistance is met, gentle are likely the least expensive and simplest to use. Devices
pressure will facilitate the tip of the tube turning caudally currently exist that use electromagnetic sensors at the
to descend into the oropharynx. tube tip (Cortrak, CORPAK MedSystems, Buffalo Grove,
The patient may be given a cup of water with a straw Illinois), and have been reviewed (Fig. 58.2). Retrospective
and permitted to sip steadily and swallow water (once the and prospective studies exist demonstrating reliable use
tube is inserted and the first resistance is met), which of an active positioning detection system to follow the
facilitates closing the epiglottis and allowing directed tube as it is passed by the provider. Systems using real-time
passage into the upper esophagus. Patients who experience imaging to provide gastric or small bowel intubation are
severe pain, gagging, anxiety coughing, respiratory distress, useful bedside adjuvants to the provider, and provide a
or significant resistance noted by the inserter should be level of confidence. Resistance, patient symptoms, and
t
permitted respite before subsequent attempts are made. training of personnel should be considered when deciding
ne
Gagging, coughing, or other respiratory distress may be how best to confirm location. In experienced hands,
signs of laryngeal passage and should prompt the provider electromagnetic detection is useful and provides cost
to remove the tube and restart. savings with respect to radiographic imaging, but is not
k.
Once passage into the esophagus is successful, the tube yet the gold standard.11 Placement in the respiratory
should continue to advance until the desired depth is system occurred in up to 3.2% of patients in one study,
oo
reached. Once in the stomach, at about 50 to 55 cm, the and pneumothorax occurred in up to 1.2%, with an
tube’s position should be assessed. Confirmation of tube associated mortality.12–14 These authors advocate a simple
placement should be accomplished prior to using the multimodality method for confirmation, including an
yb
tube for any indication, and many methods have been experienced procedurist, accurate reporting regarding
described. ease of passage, patient symptoms and signs, and the use
Insufflation with auscultation alone is an unreliable of routine electromagnetic detection with selective radio-
technique, and although it may be helpful in the distal graphic imaging or with routine radiographic imaging.
er
tube position for some providers, it should not be the The risk of pneumothorax, intrapleural placement (and
rg
su
R L
://
tp
ht
R L
FIGURE 58.2 The Cortrak2 Enteral Access System provides direct feedback from an electromagnetic tip in an enclosed tube system. The
device is followed in real time and provides a map of tube passage.
Access and Intubation of the Stomach and Small Intestine CHAPTER 58 667
subsequent administration of nutrition), or other complica- gastrocutaneous fistula, and is the most common form of
tions, although low, has dramatic and morbid complications long-term enteral access performed by surgeons. This form
that should be regarded as “never events” in the realm of enteral access is preferred because it minimizes patient
of patient safety. discomfort in the long term, eliminates the nasal passage
irritation associated with transnasal enteral access devices,
and obviates the need for frequent changes because of
ACCESS TO THE STOMACH OR clogging or inadvertent removal of the tube. Multiple
SMALL INTESTINE surgical techniques have been described for the insertion
of gastrostomies including open techniques, endoscopically
Access to the stomach or small intestine may be provided placed tubes, and laparoscopic maneuvers.
by any number of interventions or devices. Access may
be endeavored for either enteral nutritional support or Open Gastrostomy
decompression and other therapeutics. Access for nutrition The open approach for gastrostomy tube placement is
may include multiple methods. Tube access can provide performed most commonly in the technique described
a route for cyclic bolus feedings, which are physiologically by Martin Stamm in 1894. The procedure is performed
more advantageous because they mirror normal daily under operating room conditions and via a small upper
functional eating. This type of bolus nutrition is sometimes midline incision. The greater omentum and transverse
impractical in the inpatient setting, and therefore continu- colon are identified and gently retracted downward to
t
ous enteral support is provided by intragastric or postpyloric identify the stomach. The stomach is grasped and manual
ne
continuous feeding. Although these may be provided by traction is applied upward to identify a relatively avascular
nasoenteric routes as described earlier, the duration of site along the anterior aspect of the greater curvature to
need should be part of the decision tree to select the place the tube. This area ideally should be away from the
k.
nasoenteric route versus nonnasoenteric, or more invasive, pylorus and antrum and without undue tension on the
methods. Decision making should include discussions of stomach. Once the site is chosen, a stab incision is made
oo
feeding technique as well. Cyclic bolus feedings into the in the left upper quadrant of the abdomen away from
stomach result in variations in blood insulin levels, resulting the course of the epigastric vessels and approximately 2
in lipolysis and the preferred anabolic state.15 Gastric to 3 finger breadths below the costal margin. The tract
yb
atony and cholestasis may also be improved by intragastric is made by passage of a blunt clamp through the skin
nutrition. Gastric intubation is also more simply accom- incision and into the peritoneal cavity using firm but
plished than postpyloric or jejunal access because it can controlled pressure with attention to avoid visceral injury.
be performed at the bedside without sedation or special Electrocautery can be useful to assist the surgeon by
er
equipment in most situations. Pharyngeal obstruction, making an opening in the peritoneal layer at the point
foregut surgery, hiatal hernia, esophageal pathology of contact with the clamp. Many different tubes may be
rg
including varices, and other sequelae of portal hyperten- used, but a 22- or 24-French tube with a balloon or
sion, among other concerns, may be contraindications to mushroom tip is commonly selected. The tube is passed
direct transnasal or transoral access. Furthermore, long- through the abdominal wall using the in situ clamp. The
su
term nutritional support, especially in skilled nursing surgeon may create a single or double purse string at the
facilities and centers for rehabilitation, may preclude the chosen site with permanent suture. The purse strings
temporary nature of the nasogastric or orogastric route. should accommodate the chosen tube’s diameter without
://
These patients may require a more appropriate surgical redundant serosa and should be placed before the gas-
or procedure-based method of providing enteral nutrition trostomy is made. These sutures are left untied for tying
tp
(see Fig. 58.1, Decision Tree). Short-term versus long-term after the tube has been passed. A gastrostomy is made in
needs may be arbitrary in some settings, but if enteral the center of the purse-string suture using electrocautery,
nutrition is required for more than 14 days in a patient and the tube is passed through the gastrostomy and
ht
who cannot tolerate sufficient nutrition orally, more advanced several centimeters into the gastric lumen (Fig.
definitive tube-based nutrition is indicated. In addition, 58.3). If the tube has a balloon, it is now inflated and the
anesthetic risk assessments, anatomic limitations, body purse-string sutures are carefully tied to secure the tube.
habitus, prior abdominal surgery, gastroparesis or gastric Careful attention must be used to avoid balloon puncture
obstruction, insufficient absorptive intestinal surface, and during fixation. Four to six stay sutures are placed around
inflammatory bowel disease may be contraindications to the gastrostomy (Fig. 58.4) to affix the anterior wall of
particular methodologies. the stomach to the anterior wall entry site on the parietal
peritoneum (Figs. 58.5 and 58.6). These are usually silk
sutures. The tube is secured to the skin using nylon sutures.
PROCEDURES FOR INTUBATION OF THE The abdominal incision is closed in the customary manner.
STOMACH AND SMALL INTESTINE
Percutaneous Endoscopic Gastrostomy
GASTROSTOMY The percutaneous endoscopic gastrostomy (PEG), an
A gastrostomy is an opening created in the wall of the endoscopically placed gastrostomy tube, has allowed for
stomach that connects to the skin through the abdomi- safe and efficient placement of long-term feeding access
nal wall. The wall of the anterior stomach is apposed without the requirement of prolonged general anesthesia
directly to the parietal peritoneum by sutures and/or by or a laparotomy incision. Few contraindications exist for
the tube itself. This leads to the existence of a planned PEG, but include upper abdominal surgery, patients with
668 SECTION II Stomach and Small Intestine
t
tube is passed into the stomach. (Illustration by Jacob Wood,
ne
MD.) ascites, or other recent abdominal surgical procedures.
The two most frequently used techniques for PEG place-
ment include the “push” and the “pull” methods. Both
k.
approaches may be performed in the operating room, in
the endoscopy suite, or at the bedside with careful monitor-
oo
ing of vital signs and procedural tolerance. Appropriate
anesthesia is usually obtained with conscious sedation and
infiltration of local anesthesia at the surgical site just prior
yb
to incision. The patient’s upper abdominal hair is clipped,
the abdomen is prepared with surgical preparatory solu-
tion, and sterile draping is applied. Both the push and
pull methods use esophagogastroscopy and air insufflation
er
the tied purse-string sutures, ready for gastropexy. (Illustration by lumination settings are usually available on most gastro-
Jacob Wood, MD.) scopic apparatus and should be used to assist in visualization.
tp
t
abdominal wall. (Illustration by Jacob Wood, MD.)
ne
k.
oo
yb
FIGURE 58.10 The completed percutaneous endoscopic
er
MD.)
subsequently removed and a guidewire is passed through modate this easily. In either method, once the tube emerges
in a Seldinger-like technique, into the gastric lumen. After from the skin incision, it is grasped and pulled gently
several centimeters of the wire have been advanced, the upward until the intraluminal button is in contact with
://
snare is loosened slightly to allow the cannula to be pulled the gastric mucosa. During the placement, the intragastric
from the stomach. The snare is retightened to firmly grasp end of the tube or “button” will reach the patient’s mouth
tp
the wire alone (Fig. 58.8). The endoscopist then withdraws prior to being passed into the patient. At this point, we
the scope from the patient and the wire is pulled along advocate the reintroduction of the gastroscope to inspect
with the snare. The assistant must ensure that the wire is the placement of the tube and to verify that it is in proper
ht
easily advanced. Using the most common pull method, position without undue tension (Fig. 58.10). The button
the gastroscope is removed and the gastrostomy tube, should be in close but sufficiently loose contact with the
equipped with a wire loop, is affixed to the guidewire mucosa. These authors advocate spinning the tube on its
exiting the patient’s mouth and the assistant exerts gentle axis; if the mucosa is seen to move with the button, it is
traction on the wire traversing the abdominal wall (Fig. too tight. At this point, the outer bolster is affixed to the
58.9). This gentle pulling moves the gastrostomy tube tube and pushed carefully to rest at skin level to maintain
down the patient’s esophagus and into the stomach and this position. The bolster should not be advanced to the
finally exits the skin incision. In the push method, the point that it applies pressure to the skin and a distance
gastrostomy tube is threaded over the guidewire withdrawn of 2 to 3 mm from the skin is appropriate. The bolster
through the patient’s mouth. The gastroscope is removed may be sutured to the skin, but this step is unnecessary
and the wire is passed through the gastrostomy tube until with contemporary tubes. The use of occlusive dressings
it is completely through the tube. The tube is gently is universally discouraged because it encourages infection.
pushed along the wire until the long, tapered tip passes Routine and uncomplicated insertion of PEGs can be
through the incision. This is somewhat facilitated by slight done immediately following the procedure.
tension being exerted on the abdominal side of the
guidewire. Of note, the tubes are often equipped with a Jejunostomy
stiffened, disposable end that exits first through the skin. The most common operative approach for the open
The junction of the soft tube and the stiffened end usually jejunostomy is the Witzel technique, named for Friedrich
670 SECTION II Stomach and Small Intestine
t
ne
FIGURE 58.11 A single purse string is placed in the small intestine FIGURE 58.12 Prior to the enteropexy, a “tunnel” is created in a
k.
wall, and left untied as the tube is placed into the lumen. Witzel enterostomy, imbricating serosa of the small intestine to
(Illustration by Jacob Wood, MD.) secure the tube, and creating a functional “one-way valve.”
Should the tube become dislodged, the likelihood of leakage of
oo
succus entericus is decreased. (Illustration by Jacob Wood, MD.)
Witzel (1865–1925). During a laparotomy, the surgeon
yb
must first identify the ligament of Treitz and then progress
along the jejunum until a point 15 to 20 cm distally. This
chosen area should be sufficiently mobile to allow for
tension-free apposition to the anterior abdominal wall. A
er
sonography is used to achieve access. This may be facilitated imperative. This process can be time consuming and costly
by the use of effervescent material given by mouth or via and care should be taken to avoid this preventable issue.
nasogastric or nasoenteric means. The tubes placed If the tube has been in place for fewer than 10 to 14 days,
percutaneously are usually of smaller caliber, and are immediate laparotomy and replacement of the tube is
subject to occlusion with medications, nutritional material, indicated. Some providers advocate for an abdominal
or other means. binder or girdle to secure transabdominal tubes. Caution
must be used to avoid pressure, shear injury, or other
phenomena, but this can keep patients without sufficient
MANAGEMENT AND COMPLICATIONS OF motor control from unintentional dislodgement of their
INTESTINAL TUBES access.
t
method that was chosen to place the tube. It is common patients.
ne
practice to begin feeding at a low rate and increase as
tolerated until the goal rate is reached; however some
surgeons advocate beginning at the goal rate if the patient REFERENCES
k.
was tolerating the goal rate preoperatively.
1. Nelson R, Tse B, Edwards S. Systematic review of prophylactic
Reported rates of gastrostomy complications vary from nasogastric decompression after abdominal operations. Br J Surg.
oo
0.4% to 22.5% of patients.17,18 Most of these complications 2005;92(6):673-680.
are minor, but life-threatening complications can and do 2. Cheatham ML, Chapman WC, Key SP, Sawyers JL. A meta-analysis
occur. The surgeon should remain vigilant and treat them of selective versus routine nasogastric decompression after elective
laparotomy. Ann Surg. 1995;221(5):469-476, [discussion 476–468].
yb
assertively. Infection of the gastrostomy site is a common 3. Boyes RJ, Kruse JA. Nasogastric and nasoenteric intubation. Crit
complication and an infection is reported to occur once Care Clin. 1992;8(4):865-878.
for every 2.1 years that a gastrostomy is in situ.19 Signs of 4. Levin A. A new gastroduodenal catheter. J Am Med Assoc. 1921;76:1007.
a gastrostomy site infection include leakage of feeds, 4a. McConnell EA. Ensuring safer stomach suctioning with the Salem
er
erythema of the site, purulent drainage, and induration sump tube. Nursing. 1977;7(9):54-57.
5. Baker JW. A long jejunostomy tube for decompressing intestinal
of the skin. Local wound care usually suffices; however, obstruction. Surg Gynecol Obstet. 1959;109:518-520.
rg
patients sometimes require a local incision-and-drainage 6. Baker JW. Stitchless plication for recurring obstruction of the small
procedure if abscess formation has occurred. Antibiotics bowel. Am J Surg. 1968;116(2):316-324.
may also be indicated. 7. White RR. Prevention of recurrent small bowel obstruction due to
su
and eventually a larger gastrocutaneous fistula. Excessive 9. Wolff BG, Pembeton JH, van Heerden JA, et al. Elective colon and
compression of the tissue between the external bumper rectal surgery without nasogastric decompression. A prospective,
tp
17. Cyrany J, Rejchrt S, Kopacova M, Bures J. Buried bumper syndrome: 20. Rosenberger LH, Newhook T, Schirmer B, Sawyer RG. Late accidental
a complication of percutaneous endoscopic gastrostomy. World J dislodgement of a percutaneous endoscopic gastrostomy tube: an
Gastroenterol. 2016;22(2):618-627. underestimated burden on patients and the health care system.
18. Vanis N, Saray A, Gornjakovic S, Mesihovic R. Percutaneous endo- Surg Endosc. 2011;25(10):3307-3311.
scopic gastrostomy (PEG): retrospective analysis of a 7-year clinical 21. Ao P, Sebastianski M, Selvarajah V, Gramlich L. Comparison of
experience. Acta Inform Med. 2012;20(4):235-237. complication rates, types, and average tube patency between jeju-
19. Clarke E, Pitts N, Latchford A, Lewis S. A large prospective audit nostomy tubes and percutaneous gastrostomy tubes in a regional
of morbidity and mortality associated with feeding gastrostomies in home enteral nutrition support program. Nutr Clin Pract.
the community. Clin Nutr. 2016;doi:10.1016/j.clnu.2016.01.008. 2015;30(3):393-397.
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
CHAPTER
Surgery for Peptic Ulcer Disease
Abubaker Ali*
| Bestoun H. Ahmed
| Michael S. Nussbaum
59
G
astroduodenal peptic ulcer disease (PUD) is a equal to 1 cm. PPI use has reduced the risk of peptic
common problem with significant geographic ulcer hemorrhage.7
variation in prevalence. In Western countries the Current indications for surgical intervention are as
incidence of PUD has steadily declined. Recent population- follows:
based studies have shown a prevalence rate of 4% with 1. Protracted bleeding despite endoscopic therapy. Bleed-
20% of patients having asymptomatic ulcers.1 In developing ing is the most common complication of PUD in the
countries, the prevalence is much higher. In a recent United States, with an incidence of approximately 100
population-based study from China, 17.2% of the popula- per 100,000 population.
tion had endoscopically documented duodenal or gastric 2. Perforation is the second most common with an annual
t
ulcers; however, more than 70% of these patients were incidence of 11 operations per 100,000 population.
ne
asymptomatic. Such variations are likely related to the Perforations are associated with the highest rate of
prevalence of Helicobacter pylori, smoking, and use of mortality.5
ulcerogenic drugs, such as nonsteroidal antiinflammatory 3. Obstruction occurs as a consequence of scarring fol-
k.
drugs (NSAIDs). In a systematic review of the literature lowing healing of prepyloric and/or duodenal ulcers.
covering developed countries, the annual incidence of 4. Intractability despite maximum medical therapy is an
oo
PUD ranged from 0.1% to 0.19% for physician-diagnosed uncommon indication for operation.
PUD and 0.01% to 0.17% when based upon hospitalized 5. Inability to rule out cancer when an ulcer remains
patients.1 In a 1996 Veterans Affairs study the prevalence despite treatment and negative endoscopic biopsies.
of PUD in H. pylori–positive patients was found to be 2%.2
yb
This is of particular importance with gastric ulcers.
In the United States, there has been a decrease in the The goals of surgical procedures are to
prevalence and number of hospitalizations for PUD. 1. Permit ulcer healing
Between 1993 and 2006 the rate of PUD-related admissions 2. Prevent or treat ulcer complications
er
decreased by 30%, with a larger decrease in duodenal 3. Address the underlying ulcer etiology
ulcer–related admissions than gastric ulcers. Such pref- 4. Minimize postoperative digestive consequences
rg
erential decreases in duodenal versus gastric ulcer disease No single procedure satisfies all of these objectives. To
likely relate to testing for H. pylori and introduction of choose the best operation, the surgeon must consider the
more potent and successful therapeutic regimens.3 The characteristics of the ulcer (location, chronicity, type of
su
advent of histamine H2-receptor antagonists (H2 blockers) complication), the likely etiology (acid hypersecretion,
in the 1970s was responsible for an initial 40% decrease drug induced, possible role of H. pylori), the patient (age,
in incidence of ulcer operations. The later development nutrition, comorbid illnesses, condition on presentation),
://
of proton pump inhibitors (PPIs) in the late 1980s led to and the operation (mortality rate, side effects). In some
further acid reduction and faster, more efficient healing respects, all ulcer operations represent a compromise:
tp
of active ulcer disease. The development of PPIs has not The morbidity of ulcer disease is replaced by the morbidity
only influenced elective medical management but also of the operation. Finally, surgeon experience must play
has had an effect in the emergency setting. When com- a role in the choice of operation; nowadays, most surgical
ht
bined, PPIs and endoscopic treatment have further residents complete their training with little experience
decreased the need for emergency operation. with the more complex procedures. Undoubtedly this
PUD complications include bleeding, perforation, and influences their choices for both elective and emergent
gastric outlet obstruction. There has been a significant operations.
downward trend in the incidence of these complications.
Although older studies demonstrated a stable or even an
increase in the number of patients admitted with one of
HISTORY OF SURGICAL TREATMENT OF
these complications, recent studies demonstrated that the PEPTIC ULCER DISEASE
rate of perforation and bleeding has been decreasing in
the United States.4 Complications of PUD vary depending PHYSIOLOGIC DISCOVERY
on the geographic location, with bleeding being the most Initial operations for ulcer disease were based on local
common in the United States, whereas obstruction may control without a good understanding of the physiology
be more common in other locations in the world.5,6 involved. As the physiology of digestion and acid produc-
Risk factors for peptic ulcer complications and their tion was delineated, the operations changed and were
recurrence included NSAID and/or acetylsalicylic acid subsequently shifted toward addressing the current
use, H. pylori infection, and ulcer size greater than or understanding of the cause of PUD.
Benjamin Brodie, an English physiologist and surgeon,
*Supported by a grant from the Foundation for Surgical Fellowships. in 1814, described the vagus nerves and their connection
673
Surgery for Peptic Ulcer Disease CHAPTER 59 673.e1
ABSTRACT
Gastroduodenal peptic ulcer disease (PUD) is a common
problem with significant geographic variation in preva-
lence. In Western countries, the incidence of PUD has
steadily declined and the prevalence is much higher in
developing countries. Such variations are likely related
to the prevalence of Helicobacter pylori, smoking, and the
use of ulcerogenic drugs, such as nonsteroidal antiinflam-
matory drugs. The advent of histamine H 2-receptor
antagonists (H2 blockers) in the 1970s and the development
of proton pump inhibitors (PPIs) in the late 1980s led to
further acid reduction and faster, more efficient healing
of active ulcer disease. The combined use of PPIs and
endoscopic treatment has further decreased the need for
emergency operation. PUD complications include bleeding,
perforation, and gastric outlet obstruction. There has
been a significant downward trend in the incidence of
these complications. Complications of PUD also vary
t
ne
depending on the geographic location. Bleeding is the
most common in the United States, and obstruction may
be more common in other locations in the world. The
goals of surgical procedures are to permit ulcer healing,
k.
prevent or treat ulcer complications, address the underlying
ulcer etiology, and minimize postoperative digestive
oo
consequences.
KEYWORDS
yb
Peptic ulcer disease, gastric ulcer, duodenal ulcer,
antrectomy, pyloroplasty, Billroth, vagotomy, Helicobacter
pylori, nonsteroidal antiinflammatory drugs (NSAIDs),
er
with the production of gastric acid. Then in 1822, William benign gastric ulcer were presented. The data showed a
Beaumont, an army surgeon, cared for Alexis St. Martin, mortality rate of only 2.2% and that other, previously used
a man who sustained a shotgun wound to the abdomen. operations were being performed with much less
Beaumont treated his wound but expected him to die; frequency.10
however, the patient survived and was left with a gastro- In 1921 Andre Latarjet described the anatomy of the
cutaneous fistula. In 1825 Beaumont began to study the vagus nerves and applied that knowledge clinically by
patient’s digestive process by tying food to a string and performing an anatomically complete vagotomy for dys-
inserting it through the fistula into the stomach and pepsia. Subsequently, he observed postoperative issues
observe how it had been digested. He also studied the with inadequate gastric emptying and included a gastro-
gastric fluid from the fistula. In 1833 Beaumont published jejunostomy.12 Despite the improved understanding in
his findings as Experiments and Observations on the Gastric vagal anatomy and physiology, therapeutic vagotomy
Juice, and the Physiology of Digestion.8 remained an obscure treatment option for PUD. Lester
The discovery of the three separate, yet related, phases Dragstedt, a physiologist and surgeon at the University
of gastric secretion and its involvement in the consumption of Chicago and later at the University of Florida, was
and digestion of a meal defined the surgical treatment paramount in the development of vagotomy for the treat-
of PUD. The cephalic, gastric, and to a lesser extent the ment of peptic ulcers. Through his animal research, he
intestinal phases are examples of physiologic discovery elucidated the role of acid hypersecretion in the develop-
molding surgical practice. Ivan Pavlov, a Russian physiolo- ment of ulcers and stated “pure gastric juice as it is secreted
t
gist and physician, described the cephalic phase of gastric by the fundus of the stomach has the capacity to destroy
ne
secretion. Through his physiologic studies with dogs, and digest all living tissue, including the wall of the
Pavlov showed that stimulation of the vagus nerves resulted jejunum, duodenum, and even the stomach itself.”13,14
in the secretion of gastric acid. His discovery earned him Despite his understanding of the physiology, he was
k.
the 1904 Nobel Prize in Physiology and Medicine. reluctant to perform a vagotomy on a human because he
The gastric or antral phase of secretion revolves around was unsure that a person could withstand the operation.
oo
the work of the physiologist John Edkins, who injected However, in 1943 he managed a 35-year-old man with
an extract of pyloric mucus membrane “activated by ulcer disease who had failed medical therapy. The patient
hydrochloric acid or boiling” into the jugular vein of cats. was offered a subtotal gastrectomy and the patient promptly
yb
He noted a marked increase in the gastric acid and pepsin declined. The patient stated that both his father and his
secretion. In 1905 he named the active agent “gastrin.” brother had undergone subtotal gastrectomies and that
Further study led to the understanding that gastric disten- his father had died and his brother was miserable as a
tion stimulates the release of gastrin from the antrum, result of the operation.12,15 Dragstedt performed a bilateral
er
resulting in the release of gastric acid.9 This knowledge vagotomy through a left thoracotomy. The patient had
was applied directly to the treatment of PUD and became immediate relief of his symptoms postoperatively. By 1945
rg
the basis for antrectomy. Dragstedt had performed vagotomies on 60 patients and,
The intestinal phase of gastric secretion refers to specific as other surgeons had noted, he began to see postvagotomy
situations when a food bolus, which has not been exposed “pyloric stenosis.” Although he initially performed a
su
to gastric acid, comes in contact with the duodenal mucosa. drainage operation only for patients who were symptomatic
In this situation the stomach is stimulated to secrete acid. from the impaired gastric emptying, he later modified
Physiologically, food boluses are acidified and upon contact his operation and performed abdominal truncal vagotomies
://
with the duodenal mucosa stimulate an inhibitory response. (TVs) with a pyloroplasty concomitantly (Fig. 59.1).
Vagotomy and drainage was gradually accepted because
SURGICAL TREATMENT
tp
early 1900s the standard operation performed for the surgical technique “the most important contribution of
treatment of gastric and duodenal ulcers was either a his career.” George Crile reported the Cleveland Clinic
pyloroplasty without vagotomy or a gastroenterostomy. data: gastrectomy was associated with an ulcer recurrence
These operations were performed regardless of the pres- rate and a mortality rate approximately three times higher
ence or absence of obstructive symptoms, and many than that in the vagotomy group.16 Goligher et al. com-
patients had resolution of their symptoms. pared vagotomy and pyloroplasty with other operations
Gastroenterostomy soon surpassed pyloroplasty as the for duodenal ulceration and found that vagotomy and
treatment of choice. Charles Mayo presented the Mayo pyloroplasty had a recurrence rate at 2 years of 6.3%,
Clinic data on gastroenterostomy for treatment of both vagotomy and enterostomy of 3.6%, whereas gastric resec-
gastric ulceration and duodenal ulceration. His data from tion had 0% recurrence.
647 patients with gastric ulcers showed a mortality rate The surgical dictum of the time was that the treatment
of 3.2% after gastroenterostomy and less than 2% in the of ulcer disease was directed at reducing acid secretion.
2734 patients with duodenal ulceration.11 Eventually, Vagotomy was used to eliminate the cephalic phase of
recurrence rates and marginal ulcer formation were acid secretion, which was considered to be the major
recognized, and surgical management began to change. contributor in duodenal ulceration. Antrectomy was the
Gastric resection gained favor, and the use of gastroen- solution to eliminating the gastric phase of acid secretion,
terostomy alone declined. In 1941 the Mayo Clinic data considered to be a major cause of gastric ulceration.
on the use of subtotal gastrectomy in the treatment of Incorporating both vagotomy and antrectomy would
Surgery for Peptic Ulcer Disease CHAPTER 59 675
t
Proximal gastric vagotomy (PGV), also referred to as
ne
a parietal cell vagotomy or highly SV, does not affect the
distal motor function and thus does not impede gastric
emptying. In the 1970s PGV became the most popular
k.
operation for elective treatment of PUD because of its
lower mortality and morbidity rates and the omission of
oo
Crow's a drainage procedure.18 van Heerden et al. reported the
foot data on 223 patients from 1973 to 1977 at the Mayo Clinic.
With a 39-month mean follow-up, the incidence of post-
yb
operative adverse effects was less than 3%, with 0% deaths
Proximal and a 4.9% recurrence rate. At the end of their study,
gastric vagotomy they concluded that “proximal gastric vagotomy is an
effective, safe, and satisfactory option.”19
er
FIGURE 59.1 Schematic representation of the three standard forms Hoffmann et al. compared TV to SGV and PGV in
of vagotomy. (From Sleisenger MH, Fordtran JS. Operations for a randomized controlled study of 248 patients with an
rg
peptic ulcer disease and early postoperative complications. 11- to 15-year follow-up. They found recurrence rates of
In: Sleisenger MH, Fordtran JS, eds. Gastrointestinal Disease. 5th
28.5% (TV), 37.4% (SGV), and 39.3% (PGV). Although
ed. Philadelphia: Saunders; 1993.)
there was a trend with lower recurrence in the TV group,
su
continuing the “no acid, no ulcer” dogma of the time, of the patients being satisfied. The findings led to the
and eradicated the need for a drainage procedure. conclusion that “none of the three forms of vagotomy can
tp
Hubert et al. from the Mayo Clinic presented their be recommended as the standard operative treatment.”20
results for vagotomy with antrectomy with a mean 17-year As the discovery of H2 blockers and PPIs emerged,
follow-up. They showed an operative mortality rate of surgical indications decreased. The improved understand-
ht
1.1% and an ulcer recurrence rate of 0.7%; the incidence ing of the pathogenesis of ulcer disease with the discovery
of major postoperative complications was less than 1%.17 of H. pylori and understanding the impact of NSAIDs
Extensive controversy revolved around the decision to further decreased the need for operation. Management
perform a vagotomy and pyloroplasty versus a vagotomy became primarily medical in nature, with operation
with antrectomy. The idea that the two procedures both reserved for the emergency treatment of bleeding and
had a place in the treatment of ulcer disease was gradually perforation. This chapter discusses elective operations
accepted. At the time it was well established that for for intractable peptic ulcers and emergency procedures
recurrent disease the optimal operation was vagotomy for complications.
with resection. It was believed that, although both opera-
tions had equivalent results as to resolution of symptoms,
vagotomy with antrectomy was associated with a lower PATHOPHYSIOLOGY OF PEPTIC
recurrence rate, whereas vagotomy with drainage was ULCER DISEASE
associated with a lower mortality rate.
The selective vagotomies (SVs) were thought to be the The decades of surgical therapy dominating the treatment
answer to the question of how to decrease acid secretion of PUD have been followed by a period of potent acid-
and at the same time limit the known and occasionally reducing medication use that has now been replaced with
debilitating postoperative morbidity of the previously a short-term regimen targeting the elimination of H. pylori
discussed operations. The basis behind the more SVs was infection. Discussions of the best operation have been
676 SECTION II Stomach and Small Intestine
t
drug therapy directed against H. pylori has an ulcer recur- to the development of gastric metaplasia in the duodenal
ne
rence rate equivalent to TV with pyloroplasty. Although bulb. This is a necessary forerunner to colonization of
emergency operations for both peptic ulcer bleeding and the duodenal epithelium with H. pylori, because H. pylori
perforation are still occasionally required, even their exclusively binds to the gastric epithelium. The metaplastic,
k.
incidences are on the wane.21 H. pylori-colonized, duodenal epithelium then becomes
The introduction of histamine H2-receptor antagonists more susceptible to acid and pepsin effects and ulceration.
oo
in 1977 radically changed the need for elective surgical After the eradication of H. pylori infection, gastric meta-
therapy of PUD. Yet, it was the discovery of the association plasia in the duodenum does not revert to normal, but
of Campylobacter pyloridis (renamed H. pylori in 1989) with with the elimination of the infection, the risk of ulcer
recurrence is eliminated.24
yb
peptic ulceration by Warren and Marshall in 1982 that
truly revolutionized our understanding of ulcer patho- H. pylori infection impairs the negative feedback of
genesis and its treatment.22 They received the Nobel Prize gastrin release by somatostatin secreted by antral D cells.
for this work in 2005. Epidemiology studies revealed a Somatostatin causes inhibition of gastrin release through
er
strong association between H. pylori infection and both a paracrine effect. Production of alkaline ammonia by
gastric and duodenal ulcer disease. Treatment of the the bacteria on both the surface epithelium and in the
rg
infection resulted in long-term cure of peptic ulcers. antral glands prevents the D cells from properly interpret-
Despite the development of potent antisecretory drugs ing the level of acid present. This leads to improperly low
and treatment for H. pylori infection, PUD remains an levels of somatostatin, and thus loss of gastrin inhibition.
su
important clinical problem because of the widespread Chronic hypergastrinemia caused by H. pylori exerts a
use of NSAIDs. The cause of peptic ulcers is complex and trophic effect and hyperplasia of the acid-secreting parietal
multifactorial, as they result from the interplay of the cells.25
://
effects of gastric acid and pepsin and the gastric mucosal Infection with H. pylori also interferes with the neural
barrier. Any entity that either increases acid and pepsin connections between the antrum and fundus that down-
tp
secretion or weakens the mucosal barrier can result in regulate acid production. This impaired neural control,
ulcers (Box 59.1). coupled with hypergastrinemia, leads to further increases
in acid production. With H. pylori eradication, the hyper-
ht
HELICOBACTER PYLORI AND PEPTIC gastrinemia rapidly resolves. Resolution of acid hypersecre-
ULCER DISEASE tion occurs much more slowly.26
H. pylori is the most common chronic bacterial infection The inflammatory response caused by H. pylori infection
in humans. Once acquired, infection persists and may or of the gastric mucosa leads to cytokine production, mainly,
may not produce gastroduodenal disease. A number of interleukin (IL)-8.27 IL-8 acts as a potent chemotactic and
factors determine whether H. pylori infection causes disease: attracts neutrophils and acute inflammatory cells into the
the pattern of histologic gastritis induced; changes in submucosa. Other cytokines include IL-17 and IL-18. In
homeostasis of gastrin and acid secretion; gastric metaplasia a recent animal model, increased serum level of IL-17
in the duodenum; interaction of H. pylori with the mucosal was found to correlate with severity of gastritis; this cor-
barrier; and the strain of H. pylori present. There is a great relation was not observed with changes in serum level of
deal of variation in the virulence of different strains of IL-8 and IL-18.28
H. pylori. Some genotypes of H. pylori appear to be par- Complex interactions occur between H. pylori and host
ticularly toxic and are more common in patients with defense mechanisms that affect the occurrence of peptic
peptic ulcers. These are vacA and cagA positive.23 There ulceration. Duodenal ulcers appear to be predominantly
is also a genetic predisposition to acquire H. pylori related to increased acid production, whereas in gastric
infection. ulceration, defense mechanism breaches appear to prevail.
H. pylori colonizes the entire gastric epithelium. Despite these differences in mechanisms, H. pylori eradica-
However, the severity of the chronic mucosal inflammation tion effectively cures PUD and prevents relapses. In
Surgery for Peptic Ulcer Disease CHAPTER 59 677
t
and impairs ulcer healing. 31 Acid suppression is the exacerbate the condition. It can develop within hours in
ne
mainstay in the therapy of NSAID-associated ulcer disease. critically ill patients, typically starting in the fundus and
Risk factors that influence PUD in NSAID users include spreading distally. Prior to the development of effective
history of ulcer; advancing age; high-dose NSAIDs; steroids; medical therapy to reduce or eliminate gastric acid, this
k.
aspirin; anticoagulants; and H. pylori infection.32 The use was a feared and highly lethal condition, often requiring
of COX-2 inhibitor and PPI can significantly reduce total or near-total gastrectomy for control in extremely
oo
complications associated with NSAID intake. ill patients. Even with such heroic measures, mortality
was extremely high. With the advent of histamine
LOW-DOSE ASPIRIN AND ULCER DISEASE H2-receptor antagonists and PPI therapy, the primary goal
yb
Even at very low doses (75 mg daily), aspirin decreases of stress ulcer therapy has been to prevent clinically
gastric mucosal prostaglandin levels and can cause sig- important bleeding by identifying those patients at risk
nificant gastric lesions. The effect of aspirin is dose for the development of stress ulceration (Box 59.2) and
dependent, and ulcer complications are twofold to fourfold administering appropriate prophylactic measures. Fortu-
er
higher in patients taking 75 to 300 mg daily compared nately, acid-reducing medication effectively prevents
with controls.33 PPIs, given with low-dose aspirin, can significant bleeding in nearly all patients at risk for stress
rg
significantly decrease the risk of developing peptic ulceration. Esophagogastroduodenoscopy (EGD) is the
ulceration.34 first line of intervention. It aids with the diagnosis. However,
treatment is usually unsuccessful secondary to the diffuse
su
ACID HYPERSECRETORY STATES AND nature of the bleeding. Angiography should be considered
ULCER DISEASE in patients who fail endoscopic intervention. Angiography
Both Zollinger-Ellison (ZE) syndrome as a consequence can facilitate embolization of the bleeding vessel(s), which
://
of gastrinoma, and retained gastric antrum after antrectomy is usually the left gastric artery, or can help reduce the
with gastrojejunal anastomosis (so-called retained excluded rate of bleeding by selective vasopressin infusion. Operative
tp
antrum) result in peptic ulceration secondary to high intervention is considered as the last resort in patients.
levels of gastrin secretion. In cases of retained excluded
gastric antrum, the residual gastric antral tissue is constantly INDICATIONS FOR THE SURGICAL
ht
with those with a persistent infection (97% and 98% vs. dilemma prompted a large number of trials comparing
61% and 65%, respectively).35 H. pylori eradication even these procedures in the surgical literature. Improvements
without concurrent acid suppression therapy heals greater in medical therapy, particularly treatment of H. pylori,
than 85% of duodenal ulcers.36 Confirmation of H. pylori have markedly reduced the risk of ulcer recurrence,
eradication should be strongly considered for all patients rendering much of these data obsolete. Thus surgical
receiving treatment because of the availability of accurate, decision-making has become confusing with little quality
relatively inexpensive, and noninvasive tests. All patients data available from the post–H. pylori era. The choices
with duodenal ulcer(s) should receive antisecretory therapy for surgical intervention for intractable duodenal ulcer
to facilitate ulcer healing; however, the duration of therapy disease include either a vagotomy with or without a drain-
will vary depending upon ulcer characteristics, risk factors age procedure or with a gastric resection.
for recurrent PUD, and the presence of ulcer complica-
tions. In patients with uncomplicated duodenal ulcer who
test positive for H. pylori, PPI, given for 10 to 14 days,
VAGOTOMY
along with the antibiotic regimen to eradicate H. pylori, The rationale for vagotomy is the elimination of direct
is usually adequate to induce healing, and additional PPI cholinergic stimulation of gastric acid secretion. The
therapy is not needed as long as they are asymptomatic released acetylcholine stimulates acid secretion via a
following therapy.37 Thus medical therapy for duodenal specific receptor on the parietal cell. Vagotomy also renders
ulcer has shifted away from an antisecretory/antacid or the acid-producing parietal cells less responsive to hista-
t
surgical approaches to an antimicrobial strategy. mine and gastrin. The distal portion of the anterior and
ne
Most peptic ulcers respond to medical treatment. posterior trunks send branches to the antrum and pylorus
However, in some individuals the ulcer is either refractory that serve a primarily motor function. Gastric motility is
to conventional therapy or recurs following successful affected by the antral and pyloric branches of the vagus
k.
initial treatment. A refractory peptic ulcer is defined as that stimulate peristaltic activity of the antrum and relax-
an endoscopically proven ulcer greater than 5 mm in ation of the pylorus. The celiac branch of the posterior
oo
diameter that does not heal after 12 weeks of treatment vagus mediates small intestine motility, whereas the hepatic
with a PPI. On the other hand, a recurrent peptic ulcer branch mediates bile flow and gallbladder motility.
is defined as an endoscopically proven ulcer greater than TV results in a variety of physiologic alterations in the
yb
5 mm in diameter that develops within 12 months following stomach. Acid secretion is drastically reduced because of
complete ulcer healing documented by repeat endoscopy. diminished cholinergic stimulation of parietal cells, and
Prior to labeling the ulcer disease as intractable, it is the cephalic phase of gastric secretion is essentially
important to rule out the following: eliminated. There is a 75% decrease in basal acid secretion
er
• Cancer by performing endoscopy with adequate biopsy and a 50% decrease in maximum acid output. The
of the ulcer edge and base. increased intraluminal stomach pH leads to elimination
rg
• Gastrinoma by measuring fasting serum gastrin. of the negative feedback on gastrin secretion; therefore,
• Total serum calcium should be measured to screen for this results in increased serum gastrin levels and gastrin
hyperparathyroidism. cell hyperplasia. As a result of loss of reflex relaxation of
su
• Ulcerogenic medication (e.g., NSAIDs, aspirin). the gastric fundus, there is rapid emptying of liquids.
• Persistent H. pylori infection by undergoing additional Similarly, TV affects distal gastric motility, resulting in
tests to confirm eradication. Ideally patients should be difficulty in emptying solids. Because of the latter altera-
://
off of the PPI for at least 2 weeks to reduce false-negative tions, approximately 20% to 30% of patients develop
results.38 gastric atony, which leads to stasis and chronic abdominal
tp
• Chronic smoking, although smoking does not appear pain and distention. For that reason, it is recommended
to be a risk factor for ulcer relapse after H. pylori has that after a TV patients should undergo a drainage pro-
been eradicated. cedure to counteract the nonrelaxing pylorus, which acts
ht
After these have been ruled out and when operative as an obstruction. The various drainage procedures
intervention is being considered, the strategy continues available are discussed later. There are four types of
to be based on reduction of acid secretion. Gastric disten- vagotomy to consider: truncal, selective, proximal gastric, and
tion is an important stimulant of gastrin release by G cells, supradiaphragmatic. Truncal and proximal gastric are
which are mainly located in the antrum; thus decompress- commonly used to treat PUD, whereas selective and
ing the stomach in patients with bleeding ulcers and supradiaphragmatic vagotomies are used infrequently.
gastric outlet obstruction secondary to ulcer is important
to reduce gastrin and hence acid release. Acid release TRUNCAL VAGOTOMY
can surgically be reduced by dividing the vagus (cephalic TV (see Fig. 59.1) involves division of the anterior and
phase), and eliminating hormonal stimulation from the posterior vagal trunks after they emerge below the dia-
antrum (gastric phase). Each of these maneuvers has phragm. The first step is to incise the peritoneal covering
consequences in terms of the normal physiology of the of the gastroesophageal junction. The peritoneum is
upper gastrointestinal tract that tend to be amplified opened horizontally, from the lesser curvature to the
when the procedures are combined, such as with vagotomy cardiac notch at the greater curvature. The surgeon uses
and antrectomy. In the past, the choice of operation thumb and right index finger for blunt dissection to
involved weighing the risk of recurrent ulceration with encircle the esophagus. A Penrose drain is placed around
the possibility of postoperative complications and long-term the lower esophagus to place more effective downward
sequelae (postgastrectomy syndromes). This decision traction on the gastroesophageal junction. When encircling
Surgery for Peptic Ulcer Disease CHAPTER 59 679
the esophagus, the surgeon stays wide of the esophagus branches of the vagus nerve along the lesser curvature
to prevent inadvertent entry into the lumen and to include that innervate the corpus and fundus of the stomach,
the vagal trunks. In the course of this maneuver, the while preserving the hepatic and celiac branches, as well
posterior vagal trunk usually will be palpated as a taut as the distal vagal branches extending to the antrum and
cord anterior to the aorta. A single anterior vagal trunk pylorus. The end result of this procedure is the same
is usually identified in the anterior midportion of the reduction in acid secretion that occurs after TV (basal
esophagus, 2 to 4 cm above the gastroesophageal junction. and stimulated acid secretion are reduced by more than
It is not uncommon for vagal fibers to be distributed 75% and 50%, respectively) but without the troublesome
among two or three smaller cords at this level. These stasis and gastric atony. Because the distal motor nerves
trunks are individually lifted up, and 2- to 4-cm segments are preserved, emptying of solids is normal; however, the
of each are separated from surrounding tissues. A 1- to nerves affecting receptive relaxation are divided, and
2-cm length of nerve is resected and a clip is applied to some rapid emptying of liquids may occur. The alteration
the cut ends of the nerve. The “criminal nerve” of Grassi in liquid emptying is usually minimal. This procedure is
also may be identified wrapping around the cardiac notch associated with the lowest morbidity rate of all vagotomy
from its origin in the posterior trunk and is a common procedures and became the operation of choice in many
cause of incomplete vagotomy. centers despite a reported ulcer recurrence rate of between
The posterior vagal trunk is usually identified along 5% and 20%. A meta-analysis of 12 trials confirmed that
the right edge of the esophagus. If the anterior vagus has PGV has the highest recurrence rate when compared with
t
already been divided, the esophagus is more mobile. This TV with pyloroplasty, but fewer long-term side effects.39
ne
mobility allows downward traction on the gastroesophageal PGV also has been compared with TV in a randomized
junction, causing the posterior vagus to “bowstring” and trial, in which it was shown to have a lower incidence of
making it easier to identify. A 2- to 4-cm segment is sepa- dumping syndrome and weight loss. Although the ulcer
k.
rated from surrounding tissues, its margins marked with recurrence rates were higher with PGV, this was not sig-
clips, and resected. The resected portions of the anterior nificant when prepyloric ulcers (for which PGV is not an
oo
and posterior vagal trunks should be sent to pathology adequate operation) were excluded.40
for frozen section. This procedure completely denervates PGV is a complex and lengthy procedure, and, to help
the stomach and eliminates vagal innervation to the to simplify the procedure, several variations have been
yb
pancreas, small intestine, proximal colon, and hepatobiliary described. They usually consist of a posterior TV and a
tree. Although this procedure significantly reduces acid more selective ablation of the anterior vagal fibers to the
secretion, it also markedly alters gastric motility. As dis- gastric fundus and body. Hill and Baker performed a
cussed earlier, some form of gastric emptying procedure posterior TV with an anterior PGV (Hill-Baker procedure).
er
and ameliorate the increased incidence of gallbladder ulcer surgery, such operations are not commonly used.
stasis, which may lead to increased gallstone formation. However, such approaches are popular for laparoscopic
The vagal fibers are divided distal to the takeoff of the treatment of ulcer disease.
://
technically more demanding than TV and requires a more This procedure is performed primarily for patients for
careful and meticulous dissection. This technique spares whom attempts at complete vagotomy via an abdominal
vagal innervation to the gallbladder and intestine while approach have failed; it is thought that further attempts
ht
completely denervating the stomach. Because the vagal to find the missed trunks in the reoperated abdomen may
pyloric innervation is also eliminated, a drainage procedure be difficult, and thus a thoracic approach is advised. This
is still required. The primary reason for the development operation involves performing a thoracotomy or thora-
of this technique was its presumed lower side-effect profile. coscopy, identifying the two large nerve trunks, and
However, a prospective randomized study failed to show performing a TV.
substantial benefit for SV over TV. 8 The incidence of
diarrhea following an SV was no different when compared
with TV. The introduction of PGV with its lower side-effect
DRAINAGE PROCEDURES
profile and the elimination of the need for a drainage Any patient who undergoes a truncal, selective, or supra-
procedure resulted in a limited use of SV as a therapeutic diaphragmatic vagotomy should undergo a drainage
option. procedure to facilitate gastric emptying. Drainage proce-
dures fall into two categories: pyloroplasties and gastro-
PROXIMAL GASTRIC VAGOTOMY jejunostomy (Fig. 59.2). Pyloroplasty is the preferred
PGV is also known as parietal cell vagotomy and highly approach because it perpetuates the original anatomy, is
SV (see Fig. 59.1). The rationale for PGV is to eliminate a simple procedure, and is associated with less bile reflux
the vagal stimulation to the acid-secreting portion of the than gastrojejunostomy. More than 90% of all drainage
stomach without interrupting motor innervation to the procedures currently performed are variations of
antrum and pylorus. The operation involves severing all pyloroplasty.
680 SECTION II Stomach and Small Intestine
t
walls on each side of the transverse closure.
ne
Finney Pyloroplasty
The Finney pyloroplasty (Fig. 59.4; see also Fig. 59.2) is
k.
Jaboulay Gastroduodenostomy Finney Gastroduodenostomy indicated in the setting of extensive scarring and narrowing
of a significant portion of the duodenal bulb, making a
FIGURE 59.2 Drainage procedures used with truncal or selective
oo
Heineke-Mikulicz pyloroplasty untenable. The pylorus is
vagotomy. (From Matthews JB, Silen W. Operations for peptic
identified and mobilized with a generous Kocher maneuver.
ulcer disease and early operative complications. In: Sleisenger
MH, Fordtran JS, eds. Gastrointestinal Disease. 5th ed.
Traction sutures are placed along the pylorus as described
yb
Philadelphia: Saunders; 1993.) for the Heineke-Mikulicz pyloroplasty. Then a single
inverted U- or V-shaped incision is made though the
prepyloric antrum, the pylorus, and the first part of the
duodenum for a distance of approximately 7 cm in each
er
In 1888 the Heineke-Mikulicz procedure (see Fig. 59.2) The inferior leaf of the stomach is sutured to the inferior
was described independently by two surgeons, Heineke leaf of the anterior duodenal wall and continues to the
and Mikulicz. The technique is popular because it is extent of the incisions laterally on the stomach and
su
technically straightforward, applicable to many clinical duodenum. This is done with a running absorbable suture,
ulcer scenarios, and associated with few complications. and after the inferior leaflet has been approximated to
The Heineke-Mikulicz pyloroplasty is the most commonly the extent of the incision, it is continued back toward the
://
performed drainage procedure, and when conducted pylorus, approximating the superior leaflets in the same
carefully and in a technically sound fashion, obstruction fashion. A layer of Lembert sutures is then placed to invert
tp
or leakage is rare. Patients who are candidates for this the inner layer. The use of this drainage procedure makes
procedure include those who have a mobile, uninvolved a larger lumen possible but is more technically demanding
anterior pylorus; those who have no evidence of a severely compared with the Heineke-Mikulicz technique, involving
ht
distorted or edematous pylorus; and those with small, a great deal more suturing, and has greater potential for
minimally deforming pyloric perforations (massive perfora- complications.
tions make pyloroplasty difficult and somewhat treacher-
ous). The procedure may be performed using a single- or Jaboulay Gastroduodenostomy
double-layer closure. After the pylorus is identified and The Jaboulay drainage procedure (Fig. 59.5; see also Fig.
the duodenum is mobilized with a Kocher maneuver, two 59.2) is the only one of the three described here that
traction sutures are placed in the anterior surface of the does not transect the pyloric muscle. The procedure
pylorus at the 12 and 6 o'clock positions; efforts should involves an anastomosis of the distal stomach to the first
be made to include the pyloric vein of Mayo in these and second portions of the duodenum, thus bypassing
sutures, which is typically found in the inferior-anterior the pylorus. The procedure is indicated primarily for the
position on the pylorus (the vein may be used as a marker severely scarred or deformed pylorus or duodenal bulb
to identify the pylorus location, which is especially useful that would be too difficult and treacherous to incise. After
during laparoscopic procedures) to partially control the carrying out a very extensive Kocher maneuver with
subsequent bleeding. The sutures are elevated, placing thorough mobilization of the second and third portions
gentle tension on the anterior surface of the pylorus. A of the duodenum, an area of the duodenum distal to the
full-thickness longitudinal (horizontal) incision through stenotic/scarred area is chosen, as is an area of the distal
the anterior wall of the pylorus (thus interrupting the stomach just proximal to the pylorus. The duodenum is
circular muscle of the sphincter) is made, starting on the rolled anteriorly onto the stomach, and a posterior row
Surgery for Peptic Ulcer Disease CHAPTER 59 681
Traction suture
Divided pylorus
Serosal
sutures
Heineke-Mikulicz
Kocher
mobilization
t
Inverting
ne
suture
Traction suture
1 2 3
k.
FIGURE 59.3 Schematic representation of Heineke-Mikulicz pyloroplasty. While lifting up on the traction sutures, a longitudinal incision is
made through the pyloric muscles and extended 2 to 3 cm proximally into the stomach and distally into the duodenum (part 1). If the
oo
duodenum is soft, pliable, and minimally deformed, a running closure of the inside layer is begun with absorbable suture in an inverting
fashion (part 2). An outside layer of Lembert silk sutures in an interrupted fashion completes the procedure (part 3). (From Zollinger RM.
Atlas of Surgical Operations. New York: Macmillan; 1975.)
yb
Traction suture
Finney
er
Divided pylorus
Serosal sutures
rg
su
://
Mucosal
tp
suture
ht
4 5
FIGURE 59.4 Schematic of the Finney pyloroplasty. After the careful exploration, closure is initiated by using absorbable suture to begin a
running closure (part 4). A final, interrupted row of silk sutures is then placed in Lembert fashion to complete the pyloroplasty (part 5).
(From Zollinger RM. Atlas of Surgical Operations. New York: Macmillan; 1975.)
of Lembert sutures are placed. Two separate incisions are be performed in a single layer. Use of the Jaboulay pro-
made through the previously chosen sites on the prepyloric cedure has been associated with increased bile reflux as
antrum and the first portion of the duodenum. The the anastomosis is close to the ampulla of Vater.
posterior inner layer of the gastroduodenal anastomosis
is completed with a continuous full-thickness absorbable GASTROJEJUNOSTOMY
suture; the anterior inner layer is completed with a continu- Gastrojejunostomy (see Fig. 59.2) was first performed
ous inverting Connell suture. Finally, anterior Lembert alone in 1881 and was plagued by two problems: marginal
interrupted sutures are placed. The anastomosis can also ulcers (because no vagotomy was performed) and vomiting,
682 SECTION II Stomach and Small Intestine
Angle suture
Jaboulay
Intact Intact
pylorus pylorus
Ulcer
Stomach
Duodenal
incision
incision Serosal
sutures
t
ne
k.
6 7
oo
FIGURE 59.5 Schematic of the Jaboulay gastroduodenostomy. Equal-size incisions are made in the distal stomach and proximal
duodenum approximately 4 to 5 cm in length (part 6). A final anterior, outside layer of interrupted Lembert silk sutures is then placed to
yb
complete the gastroduodenostomy (part 7). (From Zollinger RM. Atlas of Surgical Operations. New York: Macmillan; 1975.)
er
which was thought to be caused by kinking with an exces- GASTRIC RESECTION PROCEDURES
sive length of the afferent limb of jejunum. The two
rg
problems have been overcome with the addition of Although subtotal gastrectomy was used for the treatment
vagotomy and construction of a shorter afferent jejunal of duodenal ulcer disease in the past, currently it is most
segment. Gastrojejunostomy is most commonly indicated commonly used for gastric ulcer and distal gastric malignan-
su
as a drainage procedure when there is duodenal obstruc- cies. A more common gastric resection performed for
tion and the duodenal bulb is so scarred, inflamed, and intractable duodenal ulcer is antrectomy (40% distal
edematous that pyloroplasty is not safe or is excessively gastrectomy) that is combined with a TV or an SV. The
://
technically demanding. This is also the drainage procedure simultaneous effects of vagotomy and antrectomy remove
of choice when vagotomy and drainage is being performed both the cholinergic and gastrin stimulus to acid secretion.
tp
laparoscopically. The jejunum, unlike the duodenum, Basal acid secretion is virtually abolished and stimulated
lacks Brunner glands that secrete alkaline solution and secretion is reduced by nearly 80%. After antrectomy,
protect against stomach pH. Therefore, historically, gastrointestinal continuity must be restored by some form
ht
vagotomy was highly recommended as an adjunct when of reconstruction. The remnant is anastomosed either to
performing a gastrojejunostomy as a drainage procedure the duodenum (Billroth I [B I]) or, after closing the
in the treatment of PUD. This was mainly to reduce duodenal stump, to the jejunum distal to the ligament of
the incidence of marginal ulcers; however, in the era of Treitz (Billroth II [B II]) (Fig. 59.6). B I reconstruction
PPIs, we have learned that lifelong PPI can significantly has several theoretical advantages:
reduce this complication without the morbidity associated 1. Restoring normal GI continuity
with vagotomy.43 Older patients with achlorhydria and 2. Leaving specialized duodenal mucosa next to the gastric
atrophic gastritis make little acid, and a vagotomy may mucosa
not be necessary, especially in the setting of malignant 3. Avoiding problems with an afferent and efferent limb
obstruction. 4. Allowing easier performance of endoscopic retrograde
A variety of postgastrectomy complications may occur cholangiopancreatography (ERCP) and endoscopic
after pyloroplasty or gastroenterostomy, including dumping, examination of the bowel
diarrhea, alkaline reflux gastritis, anemia, and marginal 5. Reduced incidence of gastric cancer in the remnant
ulceration. These may be seen in up to 50% of patients stomach.44
after operation on a temporary basis, but they resolve Despite the theoretical physiologic advantages, no
within 6 to 8 months in most, and only 5% to 7% of important functional differences have ever been demon-
patients have a persistent, symptomatic postoperative strated between these reconstructions. Although studies
complication such as dumping. show a larger fecal fat loss following a B II procedure,
Surgery for Peptic Ulcer Disease CHAPTER 59 683
t
B II (see Fig. 59.8) has a lower incidence of bile reflux; middle colic artery (Fig. 59.10). The dissection should be
ne
consequently, some authors recommend this as the carried approximately 1 cm past the pylorus if a B I
standard reconstruction technique.48 However, other reconstruction is anticipated. If B II is anticipated, the
authors promote the “uncut” Roux-en-Y reconstruction49 dissection need only be carried far enough to comfortably
k.
(Fig. 59.9). place the transverse linear stapler past the pylorus or to
close the duodenum by a handsewn technique. Next, the
oo
lesser curvature is mobilized (Fig. 59.11). The flimsy
tissues of the lesser omentum are divided along the lesser
curvature starting at the incisura and working toward the
yb
pylorus. The right gastric artery is divided and ligated.
The posterior wall of the duodenal bulb is then carefully
dissected off the pancreas. A Kocher maneuver should
be performed prior to distal gastrectomy to minimize
er
FIGURE 59.6 Reconstruction techniques after partial gastrectomy: through the placement of a posterior serosal layer of
Billroth I gastroduodenostomy and Billroth II gastrojejunostomy. interrupted silk sutures (Fig. 59.12). An inner mucosal
(From Matthews JB, Silen W. Operations for peptic ulcer disease closure is initiated with a continuous absorbable suture
://
and early operative complications. In: Sleisenger MH, Fordtran JS, (Fig. 59.13). The mucosal suture continues anteriorly
eds. Gastrointestinal Disease. 5th ed. Philadelphia: Saunders; (Fig. 59.14). Finally, an anterior serosal layer is placed
tp
A B C D E F G H I
FIGURE 59.7 Variations of Billroth I reconstructions. (A) Billroth (1881). (B) Billroth (1881). (C) Kocher (1890). (D) Kutscha-Lissberg (1925).
(E) v. Haberer (1920). (F) v. Haberer (1920), Finney (1923). (G) Winkelbauer (1927). (H) Schoemaker (1911). (I) Harkins, Nyhus (1960).
(From Siewert JR, Bumm R. Billroth I gastrectomy. In: Baker RJ, Fischer JE, eds. Mastery of Surgery. Philadelphia: Lippincott Williams &
Wilkins; 2001.)
684 SECTION II Stomach and Small Intestine
A B C
t
ne
k.
D E F
oo
yb
er
rg
G H I
su
://
tp
ht
J K
FIGURE 59.8 Variations of Billroth II reconstruction. (A) Billroth II. (B) Kronelin. (C) von Eiselberg. (D) Braun. (E) Roux. (F) Roux-en-Y.
(G) Ploy and Reichel. (H) Finsterer-Hofmeister. (I) Balfour. (J) Moynihan. (K) Tanner. (From Wastell C, Davis PA. Billroth II gastrectomy.
In: Baker RJ, Fischer JE, eds. Mastery of Surgery. Philadelphia: Lippincott Williams & Wilkins; 2001.)
For a stapled gastroduodenostomy, a gastrotomy is The anastomosis is inspected to ensure adequate hemo-
created with electrocautery on the anterior surface of the stasis. The anvil is then removed and checked to ensure
stomach at least 3 cm proximal to the staple closure (Fig. that tissue doughnuts from both the duodenum and the
59.16). The end-to-end stapling device, without the anvil, stomach are present. The gastrotomy is closed by the
is passed into the anterior gastrotomy with the rod advanc- application of a TA stapling device or sutured closed in
ing through the posterior gastric wall, again 3 cm proximal two layers (Fig. 59.18). Laparoscopic partial gastrectomy
to the stapled edge. The anvil is introduced into the with B I reconstruction has been described using many
duodenum after placement of a purse-string suture in different surgical techniques. It can be performed using
the end of the duodenum (Fig. 59.17). The end-to-end linear or end-to-end stapling devices depending on the
anastomotic (EEA) stapler is closed, fired, and withdrawn. surgeon’s preference and mobility of the duodenum.
Surgery for Peptic Ulcer Disease CHAPTER 59 685
Ligation of right
gastroepiploic
artery
t
ne
FIGURE 59.10 The gastrocolic omentum is dissected from the
stomach. The dissection begins at the pylorus with ligation of the
k.
right gastroepiploic artery and proceeds cephalad along the
greater curvature. The posterior antrum is then separated from
oo
60 cm the anterior pancreas and base of the transverse mesocolon by
division of fine connective tissue attachments. (From Jones RS.
Gastric resection: Billroth I anastomosis. In: Sabiston DC Jr, ed.
Atlas of General Surgery. Philadelphia: Saunders; 1994:263.)
yb
FIGURE 59.9 “Uncut” Roux-en-Y reconstruction after partial
gastrectomy. A jejunoduodenostomy with a 60-cm efferent limb is
constructed. The afferent limb is occluded with a staple line.
er
1988;166:69.)
su
://
tp
ht
t
FIGURE 59.13 An inner mucosal closure is initiated with a
ne
continuous absorbable suture. (From Jones RS: Gastric resection: FIGURE 59.15 An anterior serosal layer is placed with interrupted
Billroth I. In: Sabiston DC Jr, ed. Atlas of General Surgery. silk seromuscular sutures. (From Zinner MJ. Atlas of Gastric
Philadelphia: Saunders; 1994:268.) Surgery. New York: Churchill Livingstone; 1992. After Gloege. In:
k.
Soybel DI, Zinner MJ: Stomach and duodenum: operative
procedures. In: Zinner MJ, Schwartz SI, Ellis H, eds. Maingot’s
oo
Abdominal Operations. Stamford, CT: Appleton and Lange;
1997:1105.)
yb
er
rg
su
://
tp
Opening gastrotomy
for EEA stapler
ht
t
line down to the origin to allow more mobility of the
ne
Roux limb. A 50- to 70-cm Roux limb is created and a
side-to-side jejunojejunal anastomosis is constructed. The
Securing duodenum over anvil of stapler mesenteric defect is closed using a running 2-0 silk suture
k.
to prevent internal hernias. A mesocolic window large
FIGURE 59.17 The end-to-end stapling device, without the anvil, is enough to accommodate the Roux limb is created to the
oo
passed into the anterior gastrotomy with the rod advancing left of the middle colic vessels. The Roux limb is then
through the posterior gastric wall, again 3 cm proximal to the advanced through the window up to the proximal stomach.
stapled edge. The anvil is introduced into the duodenum after Care must be taken not to twist the mesentery of the Roux
yb
placement of a purse-string suture with an automatic device. The limb when performing this maneuver. Alternatively the
end-to-end anastomotic stapler is closed, fired, and withdrawn. Roux limb may be placed in an antecolic position. The
(From Siegler HF. Gastric resection: Billroth I anastomosis gastrojejunostomy can be performed via handsewn, an
[stapler]. In: Sabiston DC Jr, ed. Atlas of General Surgery. EEA- or a side-to-side linear-stapled anastomosis.
er
not be under tension, the antecolic position will permit CHOICE OF OPERATION FOR INTRACTABLE
emptying as effective as the retrocolic anastomosis. For DUODENAL ULCER
malignant disease, some surgeons favor an antecolic
su
anastomosis to avoid disease recurrence and subsequent As can be seen from the earlier descriptions, a variety of
gastric outlet obstruction. If a retrocolic position is chosen, surgical operations are available for patients with intrac-
the window in the transverse mesocolon should be closed table duodenal ulcer. Reliable data on the results of the
://
following construction of the anastomosis to prevent various procedures for duodenal ulcer were generated
kinking and obstruction of the jejunal limbs. Interrupted by a series of trials during the latter half of the 20th
tp
sutures are placed in seromuscular fashion between the century. Published series generally used different criteria
posterior gastric wall and the antimesenteric border of for patient selection and for estimating the incidence of
the jejunum (Fig. 59.20). Matching incisions are made side effects. Table 59.1 summarizes the data on the three
ht
with electrocautery in the jejunum and stomach, with the most commonly performed procedures: TV and antrec-
latter involving partial excision of the stapled gastric tomy, TV and drainage, and PGV. Mortality and early
closure (Fig. 59.21). The posterior full-thickness suture morbidity were highest for the resection procedures and
line is initiated with a continuous suture of absorbable lowest for PGV, which avoids opening the gastrointestinal
material on a double arm. Corner sutures include the tract. Recurrence rates were significantly lower for
anterior gastric wall, the posterior gastric wall, and the vagotomy, and antrectomy. TV with pyloroplasty is virtually
jejunum. The inner layer, full-thickness suture is continued never indicated as an elective procedure because it has
along the length of the anterior aspect of the anastomosis. both the disadvantages of a high incidence of postgas-
An anterior layer of interrupted silk sutures completes trectomy complications and a high ulcer recurrence rate
the anastomosis (Fig. 59.22). (10% to 15%).
For stapled gastroenterostomy, the jejunal limb is placed Historically, an important factor when considering the
next to gastric stapled line. Traction sutures are placed. choice of operation was the ulcer recurrence rate. However,
A linear stapler is placed through a small gastrotomy and with identification of H. pylori, it is believed that recurrences
small enterotomy to create gastrojejunostomy. The enter- are for the most part eliminated, although no data in this
otomy sites are closed either by TA stapler or handsewn. setting have yet been generated. Because of this, PGV,
Similar to the open technique, a laparoscopic partial which is associated with fewer postoperative sequelae, is
gastrectomy and B II reconstruction procedure is con- the preferred acid-reducing procedure in patients with
ducted in the same order as previously described. intractable ulcer symptoms. One trial randomized 248
688 SECTION II Stomach and Small Intestine
Trimming excess
Completed tissue from
gastrojejunostomy stapled gastrotomy
t
ne
FIGURE 59.18 The anastomosis is
inspected to ensure adequate
hemostasis. The anvil is then removed
k.
and checked to ensure that tissue
doughnuts from both the duodenum
oo
and the stomach are present. The
gastrotomy is closed by the application
of a TA stapling device. (From Siegler
yb
HF. Gastric resection: Billroth I
anastomosis [stapler]. In: Sabiston DC
Jr, ed. Atlas of General Surgery.
Philadelphia: Saunders; 1994:276.)
er
limited.
Site of jejunal
opening for stapler
Electrocautery used to
open stomach for
stapled gastrojejunostomy
1 cm from oversewn
stapled edge of stomach
t
jejunum is apposed to the stomach.
ne
Interrupted sutures are placed in
seromuscular fashion between the
posterior gastric wall and the
k.
antimesenteric border of the jejunum.
(From Jones RS. Gastric resection:
oo
Billroth II. In: Sabiston DC Jr, ed. Atlas
Proximal jejunum of General Surgery. Philadelphia:
Saunders; 1994:284.)
yb
er
rg
su
://
tp
ht
A B
t
whenever possible. If the inflammation and edema of the
ne
duodenum is not a factor, a B I reconstruction can occa-
sionally be performed. However, dissection and anastomosis
can be hazardous, and it is best to leave the ulcer bed in
k.
situ and perform a B II. In such cases, duodenal stump
leak is a major source of morbidity and mortality postop-
oo
eratively. The use of a duodenostomy tube is known to
be a safe and effective means in dealing with a difficult
duodenal stump. This involves insertion of a tube through FIGURE 59.23 Laparoscopic anterior seromyotomy as part of the
yb
the second portion of the duodenum to encourage forma- Taylor procedure. (From Dubois F. New surgical strategy for
tion of a controlled fistula; by doing so, this takes off the gastroduodenal ulcer: laparoscopic approach. World J Surg.
pressure from the stump and allows healing. 2000;24:270.)
In situations in which the duodenum is scarred to the
er
pancreatic capsule, a Nissen closure can be performed. approximately 15 cm from the xiphoid process, slightly
This is performed by first transecting the duodenum. The to the left of the midline. A 12-mm trocar is used for
rg
duodenal stump is then anastomosed to the pancreatic the surgeon’s right hand to accommodate a linear stapler.
capsule or duodenal wall left in place on the pancreatic A third 5-mm trocar is placed in the right subcostal
capsule. Another way of dealing with a difficult duodenal area. After placing the patient in steep reverse Tren-
su
stump is to perform a Bancroft closure. In this method delenburg position, a liver retractor is placed via a 5-mm
of duodenal stump closure, the stomach is transected incision in the subxiphoid area. This will elevate and retract
proximal to the pylorus, where tissue is less fibrotic. The the left lateral segment of the liver. The first portion
://
gastric mucosa in the duodenal stump is then dissected of the duodenum is mobilized and divided using a
away from the submucosa into the duodenum. This is laparoscopic stapler. The staple line can be reinforced with
tp
secured with a purse-string suture, and the seromuscular suture or staple-buttressing material at the surgeon’s discre-
layer is closed over the stump. tion. The proximal stomach is divided using a laparoscopic
stapler. Reconstruction is then performed either with a
ht
seromyotomy is then closed with an overlapping running are now outdated because of the high ulcer recurrence
suture.51 rates.56 Addition of a TV to gastric resection offers no
additional benefit to the patient.57 Low recurrence rates
(5%) and excellent symptomatic relief are usually achieved
ELECTIVE OPERATION FOR INTRACTABLE with a distal gastrectomy alone.
GASTRIC ULCER DISEASE TYPE II GASTRIC ULCER
Although both gastric and duodenal ulcers are peptic Type II gastric ulcers occur in the pyloric channel syn-
lesions, fundamental differences between these entities chronously with ulceration in the duodenum or meta-
affect surgical strategy. The most important difference is chronously with scarring in the duodenum. Therefore
that gastric ulcers more commonly may harbor malignancy patients presenting with a pyloric channel ulcer and
and thus must be excised or generously biopsied. Acid history of duodenal ulcer classifies the gastric ulcer as
hypersecretion, which is important in pathogenesis of type II and should be treated as such. They tend to be
duodenal ulcers, does not have a role in pathogenesis of large, deep ulcers with poorly defined margins. They
many gastric ulcers. In 1965 Dr. H.D. Johnson published frequently occur in younger men and are associated with
a classification system that subsequently was adopted and, increased acid secretion. Preoperative endoscopic examina-
with little modification, is still currently in use. In his 1965 tion of such ulcers must include biopsy of the lesion to
paper, he discusses Dragstedt’s theory that the cephalic rule out an underlying malignancy. Treatment is similar
t
phase of acid secretion is responsible for duodenal to duodenal ulcer, with vagotomy and antrectomy as the
ne
ulceration and that the gastric phase was responsible for preferred approach. PGV alone in type II gastric ulcer is
gastric ulceration. He observed that his theory disregards
the findings that less than half of patients with gastric
k.
ulcers are acid hypersecreters. He argues for a multifacto-
rial pathogenesis for ulcers beyond the hypersecretion of
oo
gastric acid.52 The Johnson classification system, which is
based upon anatomic location and acid-secretory potential,
provides a useful basis for considering operative treatment
Type I
yb
of gastric ulcer (Table 59.2 and Fig. 59.24).
lesser
curve
TYPE I GASTRIC ULCER
Type I ulcers are the most common form and are respon-
er
and antral mucosa near the area of the incisura and occur
in the setting of acid hyposecretion. Distal gastrectomy
with B I or II reconstruction is recommended for most
su
Type II
patients because this approach removes the ulcer and the combined Type III
diseased antrum. Partial gastrectomy also eliminates the gastric and prepyloric
risk of missing a malignancy associated with a biopsy and duodenal
://
discouraged because it leaves the gastric ulcer behind, or nonresective procedures in which the ulcer itself is not
which can harbor malignancy. excised, the Kelling-Madlener procedure, or vagotomy
with pyloroplasty, which has a high ulcer recurrence rate.
TYPE III GASTRIC ULCER The risk of malignant transformation or missed malignancy
Type III ulcers are prepyloric or pyloric channel ulcers. (despite biopsies) is small but real, and thus the nonresec-
They occur in the setting of increased acid secretion and tive procedures are not recommended in this setting.
are also approached in a manner similar to duodenal Although there is no consensus in the literature, some
ulcer and type II gastric ulcer. Type III ulcers are particu- have suggested that for ulcers 5 cm below the cardia, the
larly resistant to both medical therapy and PGV, with Pauchet procedure should be used, whereas for those
recurrence rates ranging from 16% to 44% in various lesions within 2 cm of the cardia, the Csendes procedure
series.39 This finding, plus the observation that these should be attempted (see Fig. 59.25).59 The Csendes
lesions may harbor gastric malignancy, makes vagotomy procedure involves a near-total gastrectomy and a Roux-
and antrectomy the most prudent approach. Early con- en-Y esophagogastrojejunostomy for reconstruction. The
sideration for surgical referral is advisable for resistant principle of this operation is to remove the high gastric
ulcers or those that present with obstructive symptoms. ulcer such that the circumference of the esophageal
Ulcers associated with acid hypersecretion are responsible mucosa remains intact.
for approximately 45% of gastric ulcer disease, with type
II accounting for 25% and type III for 20%. TYPE V GASTRIC ULCER
t
These lesions can occur anywhere in the stomach and
ne
TYPE IV GASTRIC ULCER are induced by the use of medications, such as NSAIDs.
Type IV gastric ulcer is distinguished by its anatomic Prophylactic regimens that have been shown to dramatically
location high along the lesser curvature, close to the reduce the risk of ulcers, especially if the NSAID treatment
k.
gastroesophageal junction. Antral mucosa may extend to cannot be stopped, include the use of PPI and/or pros-
within 1 to 2 cm of the gastroesophageal junction; thus taglandin analogue.60,61 A definitive antisecretory operation
oo
type IV ulcers may simply represent a subset of type I (TV and antrectomy) should be considered if medical
gastric ulcer. Type IV ulcers are associated with gastric treatment fails or if the NSAID treatment cannot be
hyposecretion and present early with dysphagia and reflux. stopped.
yb
Large ulcer size, the degree of surrounding inflammation,
and proximity to the gastroesophageal junction render GASTRIC OUTLET OBSTRUCTION
operative management difficult and potentially dangerous. More than half of gastric outlet obstruction cases are
If the integrity of the distal esophagus can be ensured, caused by malignant disease rather than by chronic PUD.
er
subtotal gastric resection (including the ulcer bed) is Consequently, a careful work-up with biopsy should be
considered optimal. However, lesions close to the cardia done to clarify the diagnosis. Gastric outlet obstruction
rg
pose a particular challenge, and, to help to avoid a total represents 5% to 8% of ulcer-related complications and
gastrectomy and an esophageal anastomosis, other surgical results in an estimated 2000 operations per year in the
approaches have been described. Such alternatives include United States.62 Patients with gastric outlet (pyloric)
su
the Schoemaker procedure (a modification of B I resection obstruction because of a duodenal ulcer typically present
with tube-shaped resection of high gastric ulcers and with symptoms of gastric retention, including early satiety,
anastomosis of the duodenum to the greater curvature bloating, indigestion, anorexia, nausea, vomiting, epigastric
://
side of the stomach; see Fig. 59.7), the Pauchet procedure,58 pain, and weight loss. They are frequently malnourished
a modification of the Schoemaker procedure that involves and dehydrated and have a metabolic alkalosis, which are
tp
a lower gastrectomy and excision of the ulcer (Fig. 59.25), factors that increase the operative risk. Operation in these
ht
A B C
FIGURE 59.25 Operations for a type IV gastric ulcer. (A) Pauchet procedure. (B) Kelling-Madlener procedure. (C) Csendes procedure
(esophagogastrojejunostomy). (Modified from Seymour NE. Operations for peptic ulcer and their complications. In: Feldman M,
Scharschmidt BF, Sleisenger MH, eds. Gastrointestinal Disease. Philadelphia: Saunders; 1998.)
Surgery for Peptic Ulcer Disease CHAPTER 59 693
cases is almost never emergent and should be performed the standard of therapy in this group (in particular, H.
only after the patient has been stabilized and nutritional pylori–negative patients) until further studies define the
and electrolyte abnormalities have been corrected. Naso- role of nonoperative therapy in the H. pylori–positive
gastric decompression is helpful in decreasing gastric patients.
atony and hastening the time to resumption of postopera-
tive oral intake. Nevertheless, operation is generally
indicated if obstruction fails to resolve despite 48 to 72 EMERGENCY OPERATION FOR COMPLICATED
hours of adequate intravenous fluid replenishment, PEPTIC ULCER DISEASE
antisecretory therapy, and nasogastric tube decompression.
In a less acute setting in which the obstruction is incom- Approximately two-thirds of operations for complicated
plete, balloon dilation of the scarred pylorus may be PUD are required because of bleeding and approximately
attempted. Approximately 65% of patients experience one-third are because of perforations.21 Emergency opera-
sustained relief, but many require more than one dilation tions for complicated PUD are most often performed in
session. In patients who fail endoscopic dilation, operation older adults and the sick. Patients may present with
is a reasonable option. The most serious complication of bleeding, perforation, or obstruction. The objectives of
dilation is a perforation. Whenever balloon dilation is operation in these cases are to
being attempted, it is important to rule out an underlying 1. Deal with the complication that necessitated surgical
malignancy. intervention
t
TV with antrectomy is the ideal procedure for this 2. Reduce the risk of future ulcer recurrence
ne
condition. Placement of a feeding jejunostomy tube at 3. Perform a safe, quick, and effective operation
the time of operation is usually recommended, both 4. Minimize long-term effects on the gastrointestinal tract
because of preoperative malnutrition and because the 5. Establish the H. pylori status of the patient
k.
chronic gastric outlet obstruction predisposes to delayed The major intraoperative dilemma is whether to proceed
postoperative gastric emptying. Furthermore, addition of with a definitive antiulcer operation (to reduce the risk
oo
either a retrograde jejunogastrostomy or gastrostomy tube of recurrence) in addition to addressing the specific ulcer
is also beneficial. As a result of the extensive scarring complication. This issue has received considerable attention
present in the duodenum, pyloroplasty of various types over the past several decades but remains unsettled. Shift-
yb
are usually not possible. The Jaboulay side-to-side duode- ing ulcer epidemiology, recognition of the role of H.
noplasty may be used in this setting. In one report of 19 pylori, and improvements in medical therapy have confused
patients treated with this procedure combined with PGV, this issue considerably, and the decision must be individual-
there was a high degree of patient satisfaction (100% ized. However, studies show a trend toward favoring of
er
modified Visick grade I or II), universal weight gain, and less complex procedures in the setting of emergencies
no operative mortality or ulcer recurrence at mean follow- (patching or oversewing of ulcers) and of avoiding
rg
up of 31 months.63 However, these benefits have not been vagotomy or gastric resections.67
noted in all reports. One trial randomized 90 consecutive Omission of an acid-reducing ulcer procedure carries
patients with gastric outlet obstruction secondary to a risk of recurrent ulcer symptoms and complications;
su
duodenal ulcer to PGV with gastrojejunostomy, PGV this risk is variable in the literature but not negligible.
with Jaboulay duodenoplasty, or SV with antrectomy. Evidence suggests that this risk may be considerably
Although there were no differences in the postoperative reduced by treatment for H. pylori postoperatively but
://
course or the reduction in gastric acid secretion, both obviously only if the patient is H. pylori positive. Unfortu-
PGV with gastrojejunostomy and SV with antrectomy nately, there is no reliable, rapid test for H. pylori at the
tp
produced a superior clinical result to PGV with Jaboulay time of operation to help guide this decision making. A
pyloroplasty.64 Although the need for pyloric reconstruction definitive procedure is always more appropriate in the
or bypass would theoretically negate several advantages setting of NSAIDs, especially if the patient is unlikely to
ht
of PGV over other options, preservation of antropyloric be able to stop the treatment because of an underlying
innervation may preserve controlled gastric emptying and medical condition. A definitive procedure is also recom-
minimize bile reflux.65 In the past, simple gastrojejunos- mended if patient has been on antisecretory therapy and
tomy was not recommended because of a nearly 50% developed an ulcer complication despite acid suppression.
recurrence rate of ulcer disease.66 It should be noted that On the other hand, inclusion of an acid-reducing ulcer
these data preceded the era of effective acid-reducing procedure may result in serious gastrointestinal sequelae
medication. With the increased adaptation of minimally in patients who may not have required the intervention.
invasive approaches, gastrojejunostomy with either Definitive operation is generally avoided during emergency
vagotomy or lifelong proton pump inhibitor medication procedures with major underlying medical illness or
has gained resurgence in the management of gastric outlet intraoperative hemodynamic instability.
obstruction.
The role of H. pylori in the pathogenesis of gastric BLEEDING
outlet obstruction has also been evaluated. Studies show Two-thirds of the emergency operations performed for
that the incidence of H. pylori infection in this population PUD are carried out for uncontrolled hemorrhage.
is low (33% to 57%).62 However, in those infected with However, as endoscopic abilities improved, the indication
the organism, eradication therapy and balloon dilation for surgical treatment continued to decrease. As the
may result in long-term symptomatic relief and alleviate indications for surgical interventions have changed dramati-
the need for operation. In general, operation should be cally, so have the operations. Eighty to 85% of bleeding
694 SECTION II Stomach and Small Intestine
t
a bleeding upper gastrointestinal lesion have an endoscopic If EGD has failed to identify the source of hemorrhage,
ne
examination of the stomach, the first and second part of a longitudinal pyloroduodenotomy is necessary to inspect
the duodenum. This procedure enables identification of the duodenal bulb and gastric antrum. The gastroduodenal
the site of bleeding and allows therapeutic attempts at artery is the usual source of bleeding, which should be
k.
stopping the bleeding. Despite endoscopic advances, the controlled by placement of suture ligatures. After the
mortality rate following ulcer bleeding has remained bleeding has been addressed, a definitive acid-reducing
oo
stable at 5% to 10%. Indeed, recent epidemiologic data operation may be performed. With the identification of
suggest that the incidence and mortality rate of bleeding H. pylori, the utility of a vagotomy has been questioned.
duodenal ulcers may be increasing in older women.3 However, the data suggest that, even in the era of H. pylori
yb
However, an estimated 10% to 20% of patients admitted and our ability to eradicate it, a TV perhaps should be
with bleeding peptic ulcers fail medical therapy and performed in those patients with a bleeding duodenal
require urgent surgical intervention. Thus the ability to ulcer. There are several reasons for this recommendation:
predict the risk of rebleeding is important to the endos- 1. Only 40% to 70% of patients with a bleeding duodenal
er
copist and the surgeon because this permits closer monitor- ulcer are positive for H. pylori.
ing of high-risk patients and early involvement of the 2. H. pylori testing in the setting of an acute hemorrhage
rg
surgical team in their management. High recurrent is less reliable, with the CLO (Campylobacter-like organ-
bleeding rates are associated with a spurting vessel, a ism) test having a false-negative rate of 18% versus 1%
visible arterial vessel in the ulcer bed, adherent clot, or in those not actively bleeding.72
su
a large ulcer bed. The Forrest classification was developed 3. If an acid-reducing procedure is not performed, up to
in an attempt to assess the risk of rebleeding based on 50% of patients are at risk of recurrent bleeding.
endoscopic findings (Table 59.3). 4. Conflicting evidence that H. pylori treatment changes
://
Of those patients who have recurrent bleeding, a second the risk of recurrent bleeding.
endoscopic attempt at control of bleeding will fail in 25%, Our inability to determine the H. pylori status in the
tp
requiring an emergency operation. This has stimulated case of acute bleeding and the lack of evidence that
some debate as to the timing of operation for a bleeding treatment of H. pylori alters the risk of rebleeding reinforces
peptic ulcer and the role of a second attempt at endoscopic the need to perform an acid-reducing operation at the
ht
therapy. Randomized prospective studies have shown no time of initial operation; however, if the patient is unstable,
increase in mortality rate in patients who undergo a addition of an acid-reducing procedure to the hemorrhage
second therapeutic endoscopy versus surgery after the control aspect of the operation adds to the operative time
first failed endoscopy. Consequently, most clinicians would and should be avoided. In the latter case, postoperative
encourage a second attempt at endoscopic control.70 long-term PPI therapy and eradication of H. pylori is the
Current indications for operation for peptic ulcer more appropriate course of action.
hemorrhage include Because it is simple to open the pylorus in a longitudinal
1. Hemodynamic instability despite vigorous resuscitation fashion, TV with pyloroplasty is the most frequently used
(>4 units or >6 units taking into consideration the operation for bleeding duodenal ulcer. In most cases the
patient’s age, with more transfusion tolerated for the bleeding will be localized in the first part of the duodenum
younger patient) and the bleeding vessel can be controlled at the time
2. Failure of endoscopic techniques to arrest hemorrhage through the pyloroplasty incision. Upon entering the
3. Recurrent hemorrhage after initial stabilization (with duodenum, the duodenal mucosa is inspected for any
up to two attempts at obtaining endoscopic hemostasis) evidence of active bleeding, ulceration, or induration. If
4. Shock associated with recurrent hemorrhage active bleeding is encountered, this is controlled by digital
5. Continued slow bleeding with a transfusion requirement pressure. This controls the bleeding and gives time for
exceeding 3 units per day fluid resuscitation of the patient. The bleeding vessel is
6. GDU then ligated. This vessel is often the gastroduodenal artery,
Surgery for Peptic Ulcer Disease CHAPTER 59 695
PERFORATION
Smoking and NSAIDs are important etiologic factors for
ulcer perforations, and epidemiologic studies have docu-
mented an increasing rate of perforation, particularly in
older women. The outcome of patients presenting with
a perforated ulcer depends on the following:
1. Time delay to presentation and treatment—data suggest
increasing delays for surgical treatment, in part as a
consequence of more extensive diagnostic work-up
2. Site of perforation—gastric perforation is associated
with a poorer prognosis
3. Patient’s age—older patients who often have associated
comorbidities have a worse outcome
t
4. Presence of hypotension at presentation (systolic blood
ne
pressure <100 mm Hg)
Recent studies comparing nonoperative treatment with
surgical treatment in perforated PUD showed no decrease
k.
in morbidity or mortality with surgical treatment in care-
FIGURE 59.26 Technique of suture control of a bleeding duodenal fully selected groups of patients. The nonoperative
oo
ulcer. After a longitudinal pyloric incision and identification of the approach should be considered only if a water-soluble
bleeding vessel, figure-of-eight sutures are placed at the cephalic contrast study has confirmed that the ulcer is sealed with
and caudal aspects of the ulcer deep enough to occlude the no extravasation of contrast into the peritoneal cavity.
yb
gastroduodenal artery. An additional U stitch is placed to control Such patients should be followed closely with regular
small transverse pancreatic branches from the main vessel. (From physical exams and, if their abdominal exam or laboratory
Debas HT, Mulvihill SJ. Complications of peptic ulcer. In: Zinner findings indicate progressive sepsis, they should undergo
prompt operation. This approach is generally used for
er
which can be ligated both through the lumen as well as comorbidities that increase the risk of surgical intervention.
extraluminally. This vessel at the level of the posterior It should be noted that, although this approach is often
duodenal wall has a T or three-vessel junction. It is impor- used for the older patient with comorbidities, studies
su
tant to suture ligate the gastroduodenal artery superiorly show that the risk of nonoperative treatment failure is
and inferiorly, followed by ligation of the medial transverse highest in older adults and thus close observation of such
pancreatic branches using a U stitch (Fig. 59.26). Care patients is recommended. Because perforated gastric
://
should be taken to avoid injury to the common bile duct ulcers have a higher rate of reperforation and complica-
during suture placement. If no bleeding is encountered tions, nonoperative therapy in situations in which the
tp
upon opening the lumen, the mucosa should be carefully source of the perforation is known to be gastric is not
inspected for an ulcer. If identified, the ulcer base should recommended.
be cleaned to help to identify a visible vessel, which if
ht
10 years after vagotomy with pyloroplasty for perforated are H. pylori positive. In this study, all patients underwent
duodenal ulcer.75 The perioperative mortality was 5.5%, a simple closure of the perforation. The H. pylori-positive
ulcers recurred in 8.8%, and postoperative digestive patients were then randomized postoperatively to a 4-week
sequelae, notably diarrhea and dumping, developed in course of PPIs alone versus H. pylori eradication therapy.
16%. Nevertheless, the overall results were good to excel- The ulcer recurrence rate at 1 year was 5% in the H.
lent in almost 90% of cases. PGV with patch closure does pylori-eradicated group versus 38% in the PPI-treated
at least as well. Boey et al., in a prospective study of 101 group, as determined by repeat endoscopy. Notably, the
patients randomized to simple closure, TV with pyloro- 5% recurrence rate is equivalent to the recurrence rate
plasty, or PGV, showed 39-month recurrence rates of for those who undergo a definitive antiulcer procedure.79
63.3%, 11.8%, and 3.8%, respectively. The operative time These data provide good evidence for the practice of
was significantly more for the PGV, but there were no simple closure of perforated duodenal ulcers in the acute
mortalities in any group. However, the study excluded setting. However, at the time of operation, the H. pylori
older adults (older than age 70 years) and patients with status of the patient is often unknown and, in the absence
preoperative shock; this may account for the low mortality of a reliable intraoperative test, the merits of a definitive
rates.76 Another randomized study by the same group, antisecretory procedure have to be considered. This
comparing PGV with simple closure, documented recur- may be particularly important in those patients with a
rence rates of 10.6% and 36.6% (half requiring surgical previous history of peptic ulcer operation, H. pylori eradica-
intervention) at 3 years. Again there was a sample bias in tion, chronic ulcer symptoms despite use of PPIs, or
t
the group because the unstable and older patients were those on NSAIDs in whom this therapy cannot be discon-
ne
excluded.77 Another series of 107 patients with perforated tinued. In general, simple patch closure is appropriate
pyloroduodenal ulcers documented minimal morbidity, for patients with
low mortality, and excellent patient satisfaction for omental 1. Acute NSAID-related perforation (provided that the
k.
patching and PGV, with a recurrence rate of 3.7% for drugs can be discontinued postoperatively) and for
duodenal ulcer; the recurrence rate for pyloric and patients who have never been treated for PUD but who
oo
prepyloric ulcer was substantially higher at 16%.78 Chronic can be treated with PPIs and H. pylori eradication
pyloroduodenal scarring is considered a relative contra- 2. Perforation in the setting of ongoing shock, delayed
indication to PGV in this setting because it may be associ- presentation, considerable comorbid disease, or marked
yb
ated with delayed gastric emptying after surgery. peritoneal contamination
With the identification of H. pylori, the ideal surgical Fig. 59.27 summarizes the recommended approach to
approach has again been questioned. A study showed a perforated duodenal ulcer. To perform the patch
that 81% of patients with a perforated duodenal ulcer procedure, a midline laparotomy is performed and the
er
rg
No
ht
Yes
Omental patching
t
ne
k.
oo
FIGURE 59.29 The omentum, which has been mobilized on a
vascular pedicle, is secured in place with sutures tied loosely
enough to prevent tissue strangulation. This technique allows
effective closure of the perforation without narrowing the duodenal
yb
lumen. (From Baker RJ. Perforated duodenal ulcer. In: Baker RJ,
Fischer JE, eds. Mastery of Surgery. Philadelphia: Lippincott
Williams & Wilkins; 2001.)
er
placing sutures through the full thickness of the bowel wall in two
steps. This allows the use of smaller, tapered needles, and
Omentum
reduces the risk of inadvertent penetration of the posterior
su
and omentum
Ulcer perforation
plugged with
tp
use is discouraged because it tends to create a negative Even in patients in this high-risk group, who may present
suction vacuum that can interfere with the repair. with shock, there is increasing evidence that definitive
There is a growing body of literature on laparoscopic operation can be tolerated as well as the simpler and
suture patch repair, as well as laparoscopic sutureless quicker patching technique.84,85 Consequently, it is recom-
techniques using fibrin glue to repair the perforated ulcer. mended that patients with a perforated type I gastric ulcer
These studies have demonstrated the feasibility of mini- undergo a partial gastrectomy unless the patient is unstable
mally invasive approaches.80 A comparison of primary with significant comorbidities. Biopsy and patch closure
open and the laparoscopic approach showed similar may be an appropriate approach for treatment of a high
postoperative morbidity and mortality.81 However, patients type IV ulcer, where a more extensive resection may lead
who were treated laparoscopically had a significantly to total gastrectomy in a critically ill patient. Because the
shorter hospital stay.80 The conversion rates for such lapa- pathophysiology of such ulcers does not involve acid
roscopic procedures have been between 15% and 20%. hypersecretion, an acid-reducing procedure is not required.
Factors that necessitated conversion from a laparoscopic Whenever possible, the ulcer should be excised and the
approach included generalized peritonitis, Mannheim stomach closed. It is important to perform an adequate
peritonitis index greater than 21, and a perforation located four-quadrant biopsy of ulcers that are not excised followed
posteriorly.81 The presence of shock on admission is a risk by patch closure.86
factor for severe postoperative complications and generally For type II ulcers, the treatment algorithm should be
an open approach is preferable in these patients. There similar to that for perforated duodenal ulcers because
t
are no absolute contraindications for the laparoscopic the pathophysiology of the disease is very similar. This
ne
approach. means that the ulcers should be patched, H. pylori status
To perform laparoscopic repair of a duodenal ulcer, the of the patient determined with an intraoperative biopsy,
patient should be supine. The operating surgeon can stand and the patient treated appropriately. In addition, it is
k.
either on the patient’s left or between the patient’s legs, important to obtain an intraoperative biopsy to rule out
with the patient placed in split leg position. The patient is malignancy that can be associated with such gastric ulcers.
oo
then placed in reverse Trendelenburg position. Initially, a Similar to a perforated duodenal ulcer, a definitive opera-
diagnostic laparoscopy is performed. After the ulcer size is tion is not required unless the patient has a history of
carefully measured with reference to the 5-mm–diameter recurrent ulcer disease and has been previously treated
yb
working laparoscopic instrument, the perforation is patched for H. pylori. In circumstances in which a definitive pro-
in similar fashion as described for the open approach cedure is deemed appropriate, a PGV or a TV with drainage
with polydioxanone, polyglactin, or silk sutures. The should be considered.
perforation is closed by intracorporeal or extracorporeal Type III ulcers are also thought to have a similar
er
knotting (depending on the surgeon preference) with pathogenesis to duodenal ulcers; however, their treatment
an omental patch (omentopexy) directly applied to the in the case of an acute perforation deserves particular
rg
perforation site. This approach avoids tension on the attention. Patch repair of such prepyloric ulcers is associ-
inflamed ulcer margin and cutting through the tissue. It is ated with a high incidence of gastric outlet obstruction,86
critical during the process to avoid anchoring the posterior and PGV is associated with a high recurrence rate for
su
duodenal wall. This can be avoided by pulling the needle these ulcers. Therefore antrectomy with vagotomy is the
through the perforation and reinserting it through the best surgical approach in this setting. Fig. 59.31 summarizes
perforation to complete the next half of the stitch. It is the proposed surgical approach to a perforated gastric
://
important when tying the knots down to pay attention so ulcer. Gastric ulcers can be managed laparoscopically as
as not to strangulate the omental pedicle. A leak test can described previously for duodenal ulcers; however, these
tp
be performed via a nasogastric tube; however, this step is patients should have a follow-up EGD to rule out cancer
not necessary. Suctioning of the peritoneal fluid is then as the cause of the ulcer.
performed with special attention directed to potential
ht
spaces for fluid collection. The fluid collected can be sent ACKNOWLEDGMENTS
for Gram stain and culture in case the patient develops an
abscess during recovery; however, this should not extend The authors acknowledge these authors of chapters from the
antibiotic coverage beyond 24 hours. Some surgeons seventh edition of Shackelford’s Surgery of the Alimentary
advocate the performance of lavage of the abdominal cavity Tract:
using 3 to 5 L of warm normal saline with the patient in
various positions. Drains are not necessary. Tavakkolizadeh A, Ashley SW. Operations for peptic ulcer.
In: Yeo CJ, McFadden DW, eds. Shackelford’s Surgery of
Perforated Gastric Ulcer the Alimentary Tract. Vol. I. 7th ed. Philadelphia: Elsevier;
A perforated gastric ulcer carries a greater overall mortality 2013:701-719.
that ranges from 10% to 40%, and increases significantly Postier RG, Havron WS III. Vagotomy and drainage. In:
with age (>65 years).82 There has been debate in cases of Yeo CJ, McFadden DW, eds. Shackelford’s Surgery of the
perforated types I and IV gastric ulcers over whether to Alimentary Tract. Vol. I. 7th ed. Philadelphia: Elsevier;
perform a partial gastrectomy or proceed with a simple 2013:720-730.
patching of the perforation. Partial gastrectomy is the Ben-David K, Caban AM, Behrns KE. Gastric resection
preferred approach unless the patient is at unacceptably and reconstruction. In: Yeo CJ, McFadden DW, eds.
high risk because of advanced age, comorbid disease, Shackelford’s Surgery of the Alimentary Tract. Vol. I. 7th
intraoperative instability, or severe peritoneal soilage.83 ed. Philadelphia: Elsevier; 2013:731-748.
Surgery for Peptic Ulcer Disease CHAPTER 59 699
Long ulcer history/ previous Helicobacter pylori treatment Recent onset of ulcer symptoms
t
Type II: Truncal vagotomy and antrectomy or patching
ne
Type III: Truncal vagotomy and antrectomy
k.
FIGURE 59.31 Recommended treatment algorithm for surgical management of perforated gastric ulcers.
oo
REFERENCES 17. Hubert JP Jr, Kiernan PD, Beahrs OH, ReMine WH. Truncal vagotomy
and resection in the treatment of duodenal ulcer. Mayo Clin Proc.
1. Sung JJ, Kuipers EJ, El-Serag HB. Systematic review: the global 1980;55:19-24.
incidence and prevalence of peptic ulcer disease. Aliment Pharmacol 18. Siim C, Lublin HK, Jensen HE. Selective gastric vagotomy and
yb
Ther. 2009;29(9):938-946. drainage for duodenal ulcer: a 10-13-year follow-up study. Ann Surg.
2. Anand BS, Raed AK, Malaty HM, et al. Low point prevalence of 1981;194(6):687-691.
peptic ulcer in normal individuals with Helicobacter pylori infection. 19. van Heerden JA, Kelly KA, Dozois RR, et al. Proximal gastric vagotomy.
er
Am J Gastroenterol. 1996;91(6):1112-1115. Initial experience. Mayo Clin Proc. 1980;55(1):10-13.
3. Groenen MJ, Kuipers EJ, Hansen BE, Ouwendijk RJ. Incidence of 20. Hoffmann J, Jensen HE, Christiansen J, Olesen A, Loud FB, Hauch
duodenal ulcers and gastric ulcers in a Western population: back O. Prospective controlled vagotomy trial for duodenal ulcer. Results
to where it started. Can J Gastroenterol. 2009;23(9):604-608. after 11–15 years. Ann Surg. 1989;209(1):40-45.
rg
4. Bashinskaya B, Nahed BV, Redjal N, Kahle KT, Walcott BP. Trends 21. Zittel TT, Jehle EC, Becker HD. Surgical management of peptic
in peptic ulcer disease and the identification of Helicobacter pylori ulcer disease today—indication, technique and outcome. Langenbecks
as a causative organism: population-based estimates from the Arch Surg. 2000;385(2):84-96.
su
US nationwide inpatient sample. J Glob Infect Dis. 2011;3(4):366- 22. Marshall BJ, Warren JR. Unidentified curved bacilli in the stomach
370. of patients with gastritis and peptic ulceration. Lancet. 1984;1(8390):
5. Wang YR, Richter JE, Dempsey DT. Trends and outcomes of hospi- 1311-1315.
://
talizations for peptic ulcer disease in the United States, 1993 to 23. Rauws EA, Tytgat GN. Cure of duodenal ulcer associated with
2006. Ann Surg. 2010;251(1):51-58. eradication of Helicobacter pylori. Lancet. 1990;335(8700):1233-1235.
6. Irabor DO. An audit of peptic ulcer surgery in Ibadan, Nigeria. West 24. Kokoska ER, Kauffman GL Jr. Helicobacter pylori and the gastroduodenal
tp
incidence, recurrence, risk factors and mortality. Digestion. 2011;84(2): mation and progastrin processing. Scand J Gastroenterol. 1993;28(8):
102-113. 690-694.
8. Beaumont W. Nutrition classics. Experiments and Observations on the 26. McColl KE, Gillen D, El-Omar E. The role of gastrin in ulcer
Gastric Juice, and the Physiology of Digestion. By William Beaumont, pathogenesis. Baillieres Best Pract Res Clin Gastroenterol. 2000;14(1):13-26.
Plattsburgh. Printed by F. P. Allen. 1833. 27. Kim JM, Kim JS, Jung HC, Oh YK, Kim N, Song IS. Inhibition of
9. Modlin IM, Kidd M, Marks IN, Tang LH. The pivotal role of John Helicobacter pylori-induced nuclear factor-kappa B activation and
S. Edkins in the discovery of gastrin. World J Surg. 1997;21(2):226-234. interleukin-8 gene expression by ecabet sodium in gastric epithelial
10. Paulino F, Netto AP. Evolution of the surgical treatment of duodenal cells. Helicobacter. 2003;8(5):542-553.
ulcer. Rev Assoc Med Bras. 1963;9:268-272. 28. Zhao YR, Zhou Y, Lin G, Hu WJ, Du JM. Association between IL-17,
11. Herrington JL Jr. Gastroduodenal ulcer. Overview of 150 papers IL-8 and IL-18 expression in peripheral blood and Helicobacter pylori
presented before the Southern Surgical Association 1888-1986. Ann infection in mongolian gerbils. Jundishapur J Microbiol. 2015;8(8):e21503.
Surg. 1988;207:754-769. 29. Wallace JL. Prostaglandins, NSAIDs, and gastric mucosal protection:
12. Modlin IM, Darr U. The centenary of Lester Dragstedt—fifty years why doesn’t the stomach digest itself? Physiol Rev. 2008;88(4):1547-1565.
of therapeutic vagotomy. Yale J Biol Med. 1994;67(3-4):63-80. 30. Lanas A, Baron JA, Sandler RS, et al. Peptic ulcer and bleeding
13. Gustafson J, Welling D. “No acid, no ulcer” 100 years later: a review events associated with rofecoxib in a 3-year colorectal adenoma
of the history of peptic ulcer disease. J Am Coll Surg. 2010;210: chemoprevention trial. Gastroenterology. 2007;132(2):490-497.
110-116. 31. Wallace JL, McKnight GW. The mucoid cap over superficial gastric
14. Dragstedt LR. Vagotomy for gastroduodenal ulcer. Ann Surg. 1945;122: damage in the rat. A high-pH microenvironment dissipated by
973-989. nonsteroidal antiinflammatory drugs and endothelin. Gastroenterology.
15. Woodward ER. The history of vagotomy. Am J Surg. 1987;153(1):9-17. 1990;99(2):295-304.
16. Stempien SJ, Temkin E, Dagradi A. Duodenal ulcer; gastrectomy 32. Melcarne L, Garcia-Iglesias P, Calvet X. Management of NSAID-
versus vagotomy with accessory procedures. Calif Med. 1956;84(3): associated peptic ulcer disease. Expert Rev Gastroenterol Hepatol.
168-171. 2016;10(6):723-733.
700 SECTION II Stomach and Small Intestine
33. Weil J, Colin-Jones D, Langman M, et al. Prophylactic aspirin and reconstruction after distal gastrectomy for gastric cancer. Br J Surg.
risk of peptic ulcer bleeding. BMJ. 1995;310(6983):827-830. 2016;103(4):337-347.
34. Scheiman JM, Herlitz J, Veldhuyzen van Zanten SJ, et al. Esomeprazole 56. Duthie HL, Moore TH, Bardsley D, Clark RG. Surgical treatment
for prevention and resolution of upper gastrointestinal symptoms of gastric ulcers. Controlled comparison of Billroth-I gastrectomy
in patients treated with low-dose acetylsalicylic acid for cardiovascular and vagotomy and pyloroplasty. Br J Surg. 1970;57(10):784-787.
protection: the OBERON trial. J Cardiovasc Pharmacol. 2013;61:250-257. 57. McDonald MP, Broughan TA, Hermann RE, Philip RS, Hoerr SO.
35. Leodolter A, Kulig M, Brasch H, Meyer-Sabellek W, Willich SN, Operations for gastric ulcer: a long-term study. Am Surg. 1996;62(8):
Malfertheiner P. A meta-analysis comparing eradication, healing 673-677.
and relapse rates in patients with Helicobacter pylori-associated gastric 58. Lewis A, Qvist G. Operative treatment of high gastric ulcer with
or duodenal ulcer. Aliment Pharmacol Ther. 2001;15:1949-1958. special reference to Pauchet’s method. Br J Surg. 1972;59(1):1-4.
36. Lam SK, Ching CK, Lai KC, et al. Does treatment of Helicobacter 59. Csendes A, Braghetto I, Calvo F, et al. Surgical treatment of high
pylori with antibiotics alone heal duodenal ulcer? A randomised gastric ulcer. Am J Surg. 1985;149(6):765-770.
double blind placebo controlled study. Gut. 1997;41(1):43-48. 60. Lai KC, Lam SK, Chu KM, et al. Lansoprazole for the prevention
37. Gisbert JP, Pajares JM. Systematic review and meta-analysis: is 1-week of recurrences of ulcer complications from long-term low-dose
proton pump inhibitor-based triple therapy sufficient to heal peptic aspirin use. N Engl J Med. 2002;346(26):2033-2038.
ulcer? Aliment Pharmacol Ther. 2005;21(7):795-804. 61. Lai KC, Lam SK, Chu KM, et al. Lansoprazole reduces ulcer relapse
38. Chey WD, Spybrook M, Carpenter S, Nostrant TT, Elta GH, Scheiman after eradication of Helicobacter pylori in nonsteroidal anti-inflammatory
JM. Prolonged effect of omeprazole on the 14C-urea breath test. drug users—a randomized trial. Aliment Pharmacol Ther. 2003;18(8):
Am J Gastroenterol. 1996;91(1):89-92. 829-836.
39. Chan VM, Reznick RK, O’Rourke K, Kitchens JM, Lossing AG, 62. Gibson JB, Behrman SW, Fabian TC, Britt LG. Gastric outlet obstruc-
Detsky AS. Meta-analysis of highly selective vagotomy versus truncal tion resulting from peptic ulcer disease requiring surgical intervention
vagotomy and pyloroplasty in the surgical treatment of uncomplicated is infrequently associated with Helicobacter pylori infection. J Am Coll
t
duodenal ulcer. Can J Surg. 1994;37:457-464. Surg. 2000;191(1):32-37.
ne
40. Jordan PH Jr, Thornby J. Twenty years after parietal cell vagotomy 63. Dittrich K, Blauensteiner W, Schrutka-Kolbl C, Hoffer F, Armbruster
or selective vagotomy antrectomy for treatment of duodenal ulcer. C, Vavrik J. Highly selective vagotomy plus Jaboulay: a possible
Final report. Ann Surg. 1994;220(3):283-293; discussion 293-296. alternative in patients with benign stenosis secondary to duodenal
41. Taylor TV, Lythgoe JP, McFarland JB, Gilmore IT, Thomas PE, ulceration. J Am Coll Surg. 1995;180(6):654-658.
k.
Ferguson GH. Anterior lesser curve seromyotomy and posterior 64. Csendes A, Maluenda F, Braghetto I, Schutte H, Burdiles P, Diaz
truncal vagotomy versus truncal vagotomy and pyloroplasty in the JC. Prospective randomized study comparing three surgical techniques
treatment of chronic duodenal ulcer. Br J Surg. 1990;77(9):1007-1009. for the treatment of gastric outlet obstruction secondary to duodenal
oo
42. Oostvogel HJ, van Vroonhoven TJ. Anterior lesser curve seromyotomy ulcer. Am J Surg. 1993;166(1):45-49.
with posterior truncal vagotomy versus proximal gastric vagotomy. 65. Donahue PE, Griffith C, Richter HM. A 50 year perspective upon
Br J Surg. 1988;75(2):121-124. selective gastric vagotomy. Am J Surg. 1996;172:9-12.
yb
43. Coblijn UK, Lagarde SM, de Castro SM, Kuiken SD, van Tets WF, 66. Stabile BE, Passaro E Jr. Surgery for duodenal and gastric ulcer
van Wagensveld BA. The influence of prophylactic proton pump disease. Adv Surg. 1993;26:275-306.
inhibitor treatment on the development of symptomatic marginal 67. Smith BR, Wilson SE. Impact of nonresective operations for com-
ulceration in Roux-en-Y gastric bypass patients: a historic cohort plicated peptic ulcer disease in a high-risk population. Am Surg.
er
for stomach cancer after remote partial gastrectomy for benign troenterology. 1997;44(13):288-293.
conditions. Cancer Res. 1990;50(20):6486-6489. 69. Wang BW, Mok KT, Chang HT, et al. APACHE II score: a useful
45. Moller H, Toftgaard C. Cancer occurrence in a cohort of patients tool for risk assessment and an aid to decision-making in emergency
su
surgically treated for peptic ulcer. Gut. 1991;32(7):740-744. operation for bleeding gastric ulcer. J Am Coll Surg. 1998;187(3):
46. La Vecchia C, Negri E, D’Avanzo B, Moller H, Franceschi S. Partial 287-294.
gastrectomy and subsequent gastric cancer risk. J Epidemiol Community 70. Lau JY, Sung JJ, Lam YH, et al. Endoscopic retreatment compared
Health. 1992;46:12-14. with surgery in patients with recurrent bleeding after initial endo-
://
47. Rathi P, Parikh S, Kalro RH. Giant duodenal ulcer: a new look at a scopic control of bleeding ulcers. N Engl J Med. 1999;340(10):751-756.
variant of a common illness. Indian J Gastroenterol. 1996;15(1):33-34. 71. Lanas A. Editorial: upper GI bleeding-associated mortality: challenges
48. Vogel SB, Drane WE, Woodward ER. Clinical and radionuclide to improving a resistant outcome. Am J Gastroenterol. 2010;105(1):
tp
49. Park JY, Kim YJ. Uncut Roux-en-Y reconstruction after laparoscopic 73. Deleted in review.
distal gastrectomy can be a favorable method in terms of gastritis, 74. Griffin GE, Organ CH Jr. The natural history of the perforated
bile reflux, and gastric residue. J Gastric Cancer. 2014;14(4):229-237. duodenal ulcer treated by suture plication. Ann Surg. 1976;183:
50. Fischer DR, Nussbaum MS, Pritts TA, et al. Use of omeprazole in 382-385.
the management of giant duodenal ulcer: results of a prospective 75. Robles R, Parrilla P, Lujan JA, et al. Long-term follow-up of bilateral
study. Surgery. 1999;126(4):643-648; discussion 648-649. truncal vagotomy and pyloroplasty for perforated duodenal ulcer.
51. Croce E, Olmi S, Russo R, Azzola M, Mastropasqua E, Golia M. Br J Surg. 1995;82(5):665.
Laparoscopic treatment of peptic ulcers. A review after 6 years 76. Boey J, Lee NW, Koo J, Lam PH, Wong J, Ong GB. Immediate
experience with Hill-Barker’s procedure. Hepatogastroenterology. 1999; definitive surgery for perforated duodenal ulcers: a prospective
46(26):924-929. controlled trial. Ann Surg. 1982;196(3):338-344.
52. Johnson HD. Gastric ulcer: classification, blood group characteristics, 77. Boey J, Branicki FJ, Alagaratnam TT, et al. Proximal gastric vagotomy.
secretion patterns and pathogenesis. Ann Surg. 1965;162(6):996-1004. The preferred operation for perforations in acute duodenal ulcer.
53. Csendes A, Burgos AM, Smok G, Burdiles P, Braghetto I, Díaz JC. Ann Surg. 1988;208(2):169-174.
Latest results (12–21 years) of a prospective randomized study 78. Jordan PH Jr, Thornby J. Perforated pyloroduodenal ulcers. Long-term
comparing Billroth II and Roux-en-Y anastomosis after a partial results with omental patch closure and parietal cell vagotomy. Ann
gastrectomy plus vagotomy in patients with duodenal ulcers. Ann Surg. 1995;221(5):479-486; discussion 486-488.
Surg. 2009;249(2):189-194. 79. Ng EK, Lam YH, Sung JJ, et al. Eradication of Helicobacter pylori
54. Nunobe S, Okaro A, Sasako M, et al. Billroth 1 versus Roux-en-Y prevents recurrence of ulcer after simple closure of duodenal ulcer
reconstructions: a quality-of-life survey at 5 years. Int J Clin Oncol. perforation: randomized controlled trial. Ann Surg. 2000;231(2):
2007;12(6):433-439. 153-158.
55. Nakamura M, Nakamori M, Ojima T, et al. Randomized clinical 80. Byrge N, Barton RG, Enniss TM, Nirula R. Laparoscopic versus
trial comparing long-term quality of life for Billroth I versus Roux-en-Y open repair of perforated gastroduodenal ulcer: a National Surgical
Surgery for Peptic Ulcer Disease CHAPTER 59 701
Quality Improvement Program analysis. Am J Surg. 2013;206(6):957-962; 84. Hodnett RM, Gonzalez F, Lee WC, Nance FC, Deboisblanc R. The
discussion 962-963. need for definitive therapy in the management of perforated gastric
81. Muller MK, Wrann S, Widmer J, Klasen J, Weber M, Hahnloser D. ulcers. Review of 202 cases. Ann Surg. 1989;209(1):36-39.
Perforated peptic ulcer repair: factors predicting conversion in 85. Di Quinzio C, Phang PT. Surgical management of perforated benign
laparoscopy and postoperative septic complications. World J Surg. gastric ulcer in high-risk patients. Can J Surg. 1992;35(1):94-97.
2016;40(9):2186-2193. 86. Turner WW Jr, Thompson WM Jr, Thal ER. Perforated gastric ulcers.
82. Hewitt PM, Krige J, Bornman PC. Perforated gastric ulcers: resection A plea for management by simple closures. Arch Surg. 1988;123(8):
compared with simple closure. Am Surg. 1993;59(10):669-673. 960-964.
83. McGee GS, Sawyers JL. Perforated gastric ulcers. A plea for manage-
ment by primary gastric resection. Arch Surg. 1987;122(5):555-
561.
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
CHAPTER
Z
ollinger-Ellison syndrome (ZES) was initially described of the pancreas. Gastrinoma is the most common pan-
in two index cases of refractory ulcer disease and creatic islet cell tumor in patients with MEN1, followed
diarrhea in 1955 at the Ohio State University.1 The by insulinomas. Those with MEN-associated gastrinomas
index patients had recurrent peptic ulceration after typically present at a younger age. Pituitary adenomas
multiple gastric operations, requiring complete gastrectomy occur less commonly (20% to 65%) as do adrenal tumors
to control their symptoms.2 ZES is a rare cause of peptic (10% to 73%) and thyroid adenomas (0% to 10%).11,12
ulcer disease for which a high index of suspicion is required Pancreatic tumors occurring in patients with MEN1 are
to make the diagnosis. The syndrome is characterized by commonly multiple, often requiring different treatment
severe peptic ulcer disease caused by gastrin-secreting strategies than those used for patients with sporadic
t
neuroendocrine tumors, most commonly located in the pancreatic endocrine tumors, which tend to be solitary.
ne
duodenum and pancreas. Frequently gastrinomas are not Thus MEN1-associated gastrinomas are rarely cured with
recognized at initial clinical presentation and are often surgery.
mistreated. Symptoms that should raise suspicion of a
k.
gastrinoma include idiopathic peptic ulcer disease or
long-standing refractory diarrhea. The term gastrinoma is ANATOMY, PATHOPHYSIOLOGY, AND
oo
interchangeably used with ZES. MOLECULAR BIOLOGY
Pancreatic neuroendocrine tumors (PNETs) are rare
cancers occurring in approximately 1000 patients per year Endocrine tumors of the pancreas originate from islet
yb
in the United States, representing 3% of all pancreatic cells. The islets arise from neural crest cells or embryonic
tumors.3 The incidence has increased over the last three foregut endoderm. Histologically PNETs appear similar
decades, likely from improvement in imaging modalities to carcinoid tumors of the gastrointestinal tract. Gastrin
and increased frequency of imaging.4–6 The incidence has is synthesized in the G cells, found predominantly in
er
increased from 17 per 100,000 population in 1973 to 47 the gastric antrum and in smaller numbers in the duodenal
per 100,000 population in 2007.7 There is a male pre- mucosa. Gastrin release is controlled by chemical, neural,
rg
dominance (60%) and the mean age at diagnosis is 50 or mechanical stimuli. Gastrin release is stimulated by
years with most diagnosed between the age of 20 and 60 ingested protein and gastric distention. Calcium, epineph-
years. Survival is significantly longer than those with rine, and achlorhydria are also potent stimuli for gastrin
su
pancreatic adenocarcinoma; however, patients with meta- secretion. Gastrin release is inhibited by beta-blockade
static disease are not likely to be cured. Surgery plays a and atropine.2 Pernicious anemia, atrophic gastritis,
critical role in palliation of symptoms of hormone- and the use of proton pump inhibitors (PPIs) all can
://
producing tumors and for cure if only localized disease cause elevation of serum gastrin and achlorhydria. Acid
is present. Endocrine tumors of the pancreas originate hypersecretion and hypergastrinemia may be found with
tp
from islet cells, hence the traditional name of islet cell many conditions such as Helicobacter pylori infection, gastric
tumor. The tumors are broadly classified as functional outlet obstruction associated with peptic ulcer, retained
and nonfunctional. Nonfunctional tumors compose 60% antrum, short gut syndrome, or renal failure. Despite
ht
to 90% of all PNETs and are more common than functional several advances in the understanding of gastrinomas,
tumors. Nonfunctional tumors are more likely to present the cell of origin remains uncertain. Duodenal gastrinomas
with metastatic disease because there is a lack of symptoms, contain many well-differentiated gastrin-containing
leading them to present late in the disease course.8,9 G cells, and their origin may be the gastrin cells in the
Functional tumors cause symptoms from the hormone duodenal crypts and Brunner glands. Pancreatic gastri-
that the tumor produces. Approximately half of pancreatic nomas are more pleomorphic, with more heterogeneous
endocrine tumors are functional, and of these, about 50% cell arrangements. Although G cells are not normally
are gastrinomas. present in the adult pancreas, it has been proposed that
Gastrinomas can be encountered in a familial pattern multipotent, endocrine-programmed stem cells undergoing
in the setting of multiple endocrine neoplasia type 1 differentiation toward G cells are responsible for pancreatic
(MEN1 or Wermer syndrome), or rarely (10% to 15%) gastrinomas.13–15
with von Hippel-Lindau (VHL) syndrome,10 both autosomal Although gastrinomas tend to be slow growing, 60%
dominant inherited diseases.10 MEN1 is characterized by to 90% of them will have aggressive biology. The liver is
nearly complete penetrance with variable expressivity; it the most common site of metastasis with 70% to 80% of
typically involves the parathyroid glands, pituitary gland, patients diagnosed with liver metastasis at the time of
and pancreas. Ninety to 100% of MEN1 patients will diagnosis. Liver metastasis is considered to be the most
develop primary hyperparathyroidism, and 50% to 75% important predictor for long-term survival, as these patients
of patients develop symptoms from functioning neoplasms tend to have a worse prognosis.16,17
702
Zollinger-Ellison Syndrome CHAPTER 60 702.e1
ABSTRACT
Zollinger-Ellison syndrome is caused by a neuroendocrine
tumor of the pancreas that secretes gastrin, causing massive
gastric acid secretion. Treatment of the acid hypersecretion
may be accomplished with pharmacologic acid inhibition.
The goal of surgical treatment should be removal of all
gross tumor in those without distant metastases. Cure
rates approximate 30% and 20-year survival exceeds 70%
even without cure.
KEYWORDS
Gastrinoma, multiple endocrine neoplasia, hypergastrinemia
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
Zollinger-Ellison Syndrome CHAPTER 60 703
t
more than 70%, followed by heartburn (44%), nausea in 1% to 3% of patients. This renders serum gastrin
ne
(33%), vomiting (25%), and weight loss (17%). Patients measurement a very good screening test for ZES, with a
who present late may have signs and symptoms of the sensitivity that approaches 99%. Several authors indicate
metastatic disease, such as right upper quadrant pain that levels greater than 500 pg/mL, or more than fivefold
k.
secondary to liver disease or bone pain secondary to normal, are highly suggestive of gastrinoma in the presence
metastatic deposits. Occasionally, patients will present of increased gastric acid production.2 Berna et al. studied
oo
with complications of high acid output, such as bleeding 2229 cases from the literature, finding that 57% to 63%
and perforation secondary to severe peptic ulcer disease. of ZES patients had gastrin levels in this range.20 However,
However, widespread use of potent acid suppression two-thirds of gastrinoma patients had fasting serum gastrin
yb
medication has decreased the incidence of such complica- levels that overlapped with levels seen in more common
tions. Zollinger and Grant reported in 1964 that bleeding conditions.
was a major finding in 45% of patients and in 20% it was Of particular interest when evaluating hypergastrinemia
considered massive.19 In contrast, the 2000 NIH study is a condition referred to as retained excluded antrum
er
reported bleeding in only 24% of patients; the incidence syndrome. This rarely occurs in patients who have had a
of complications related to ulcer disease was 44% before partial distal gastrectomy with Billroth II reconstruction.
rg
1980, decreasing to 11% between 1990 and 1999.18 Nev- In this condition, a portion of antrum is left attached to
ertheless, despite widespread awareness of ZES, delay in the duodenum. Because it is disconnected from the
diagnosis still persists; a mean delay to diagnosis of 5.2 proximal stomach and is not exposed to gastric acid, there
su
years was seen in this series.18 is no inhibition of the normal gastrin production in the
residual antrum. This leads to chronic hypertrophy of
the gastrin-producing cells in the retained excluded distal
DIAGNOSIS AND MEDICAL THERAPY
://
prompt workup if exhibiting refractory peptic ulcer disease, preserved in the proximal stomach. These patients have
long-standing diarrhea, absence of H. pylori infection, or elevated fasting gastrin levels, elevated acid production,
failure to improve after treatment for H. pylori and acid and a negative secretin provocation test and may have
ht
suppression therapy. In addition, the presence of neph- medically refractory ulcer disease. This condition may
rolithiasis and hypercalcemia should raise suspicion of require surgical intervention for removal of the retained
possible MEN1. In addition, a careful family history should antrum.
be elucidated to evaluate for MEN1 or other inherited All patients with hypergastrinemia and suspected ZES
conditions related to MEN or VHL. should undergo confirmation with provocative gastrin
Fasting serum gastrin is the appropriate initial diagnostic testing. In addition, those patients with normal fasting
test for patients with suspected ZES; however it is not gastrin levels but still with symptoms suspicious for ZES
sufficient alone to establish the diagnosis as several medical should also undergo provocative gastrin screening. Pro-
conditions may cause hypergastrinemia. Conditions that vocative testing can be performed with secretin, calcium,
cause acid hypersecretion and suppression may cause or meal stimulation. Results of these tests are given in a
hypergastrinemia, all of which are more likely to be the relative rise in gastrin over the baseline value. Secretin
cause of hypergastrinemia than ZES. Most commonly, was first reported to cause stimulation of gastrin in ZES
pernicious anemia, atrophic gastritis, and pharmacologic patients by Hansky et al. in 1971 and was clearly eluci-
acid suppression may cause achlorhydria or reduced acid dated by Isenberg et al. in 1972.21,22 Its effect is mediated
suppression, which can cause hypergastrinemia not due through secretin receptors that are present on gastrinoma
to ZES. Other conditions that may cause fasting hyper- cells. Following an overnight fast, patients are given an
gastrinemia associated with increased acid hypersecretion intravenous (IV) bolus injection of secretin of 0.4 µg/
include H. pylori infection, gastric outlet obstruction kg of body weight over 1 minute. Serial serum gastrin
704 SECTION II Stomach and Small Intestine
levels are then obtained. Importantly, it is not necessary initial management of patients with ZES is the adequate
to discontinue PPIs or histamine-2 (H2) blockers for this suppression of gastric acid output. With current medical
test. Minimal side effects of secretin infusion may include options, it is possible to control the gastric acid secretion
flushing and nausea. Blood draws for determination of very effectively with PPIs. The doses to achieve symptom
gastrin levels are collected and analyzed at 0, 2, 5, 10, 20, relief will be much greater than with typical ulcer disease
and 30 minutes following the administration. Multiple or dyspepsia. Alternatively, the H2-receptor antagonists
definitions for a positive test exist based on the absolute such as cimetidine, ranitidine, and famotidine can be
change in gastrin concentration.2 Based on the absolute used; however, they are less effective than PPIs and require
change in gastrin concentration, four definitions of a larger and more frequent doses. PPI dosing should be
positive test exist: greater than 100 pg/mL, greater than titrated based on symptoms and documented ulcer healing.
110 pg/mL, greater than 120 pg/mL, and greater than Resolution of symptoms may not be a good guide to gauge
200 pg/mL. Percentage of change can also be used, either treatment response. Documentation of ulcer healing, with
greater than 50% or greater than 100%, and maximum serial upper endoscopies is necessary. Alternatively,
gastrin after secretion greater than 186 pg/mL or greater determination of basal acid output (BAO) was suggested
than 335 pg/mL have also been used. To maximize the in the past as a method to individualize the dose of PPI.
sensitivity and specificity of the test, the following change The recommended goal is to achieve a BAO less than
in gastrin concentrations are best used: greater than 10 mEq/h in men and 5 mEq/h in women. Many recent
100 pg/mL (95%, 99.8%), greater than 110 pg/mL studies have brought attention to the safety of long-term
t
(94%, 100%), greater than 120 pg/mL (94%, 100%), PPI use. Many long-term consequences include hypomag-
ne
and greater than 200 pg/mL (87%, 100%).23 At the Ohio nesemia, osteoporosis and bone fractures, increased
State University, we use the 110 pg/mL threshold, as cardiovascular events in patients taking antiplatelet therapy,
proposed by Deveney et al.,24 a level of increase that we community-acquired pneumonia, Clostridium difficile infec-
k.
have found to be accurate in nearly 100% of patients.2 tion, and iron absorption.26–30 Obviously these risks are
Most patients with ZES will experience this increase from mitigated by the importance of controlling acid production
oo
their basal level within 5 minutes of injection. Rarely, false- in this patient population. However, the risks should be
negative or false-positive tests may occur. The false-positive recognized so that preventive measures may be taken to
rate is 0% in nonachlorhydric patients when greater prevent some of these side effects when possible. Another
yb
than 110 pg/mL or greater than 120 pg/mL criteria potential drawback of PPIs is a delay in diagnosis of ZES
are used.23 in the patient who has not had a definitive diagnosis.31
Alternatively, a calcium stimulation test may be used, Finally, somatostatin analogues, such as octreotide, were
as a second-line test in the rare patient with a negative seen to cause significant and long-lasting inhibition of
er
secretin test in whom there is a high degree of suspicion both tumor gastrin release and gastric acid secretion, and
for ZES, or if secretin is not available. The patient is given more recently shown to prolong progression-free survival
rg
a 12-mg/kg infusion of elemental calcium (calcium glu- and believed to stabilize tumor growth.32,33 These properties
conate or calcium chloride) over 3 hours. Gastrin levels render them useful in the management of tumors refrac-
are measured at 0 and 30 minutes and then at 1, 2, 3, 4, tory to conventional acid suppression regimens.34
su
mias occur, so it is rarely used today. It has been recommended that patients undergo plasma
Other less common modalities previously described biochemical evaluation of fasting pancreatic polypeptide
for possible diagnosis include meal stimulation or glucagon (PP), pancreastatin, and chromogranin A (CgA) in addi-
ht
stimulation. Ingestion of food stimulates release of gastrin tion to glucagon and gastrin levels.35 In the past CgA has
from the antrum and duodenum. This test has not been been shown to have modest value in the diagnosis of
shown to be accurate for discrimination of hypergastrin- sporadic PNETs with modest sensitivity and specificity.
emia in ZES. Attempts were made to use glucagon stimula- Recently chromogranin has been shown to be an unreliable
tion when secretin was in short supply because it is in the tumor marker for neuroendocrine tumors (NETs).36–39
same family of GI peptides. In a report of Shibata et al. CgA is recommended by the recent European Neuroen-
after injecting glucagon, patients reportedly had decrease docrine Tumor Society (ENETS) and North American
in serum gastrin levels.25 However this test needs to be Neuroendocrine Tumor Society (NANETS) in the most
further explored to be a viable diagnostic option. recent guidelines as a practical and useful serum marker,
Once the diagnosis of gastrinoma has been confirmed despite the mixed data.40,41 PP levels are not typically
biochemically and physiologically, the next immediate elevated in sporadic gastrinoma. Initially, markedly elevated
step in management is medical control of the excessive fasting plasma PP level in MEN1 patients was 95% sensitive
acid output. Historically, the treatment for ZES was total and 88% specific for the presence of pancreatic islet cell
removal of all acid-secreting tissue by total gastrectomy. tumors detected by imaging, and thus it was thought that
Trying to avoid total gastrectomy led to recurrence, often PP may be a marker for MEN1 in patients with ZES.42
with life-threatening complications. With the advent of However, more recently CgA and PP were not found to
potent acid-reducing medications (first H 2-receptor be independent prognostic markers for overall survival
antagonists and now PPIs), the most critical step in the in a study by Walter et al. These findings were confirmed
Zollinger-Ellison Syndrome CHAPTER 60 705
t
calcium levels should be obtained and hyperparathyroidism FIGURE 60.1 Octreotide scan (somatostatin receptor scintigraphy)
ne
should be treated first before treatment of any pancreatic shows an avid lesion at the tail of the pancreas in this female
endocrine tumor. Hyperparathyroidism should be treated patient with Zollinger-Ellison syndrome. Somatostatin receptor
with subtotal thyroidectomy or total parathyroidectomy scintigraphy can be combined with computed tomography
k.
with autotransplant of parathyroid tissue. This helps reduce scanning. Digital reconstruction of the images provides additional
gastrin by removing the stimulation from the elevated information regarding the lesion and its surrounding organs that
oo
calcium. may assist in preoperative planning.
Suspicion for MEN1 should be considered in patients
with a family history consistent with endocrine tumors of
yb
the pancreas, family members with pituitary or thyroid
disease, kidney stones, young age at diagnosis, endocrine demonstrate a greater degree of enhancement than the
tumor associated with hypercalcemia, multiple NETs, or normal pancreas during the arterial and capillary phases
any patient with ZES.47 of the contrast bolus. This is helpful in identification and
er
Several authors have attempted to describe biologic differentiation of PNETs from other pancreatic tumors
markers and link them to the clinical behavior of endocrine or cancer. Dual-phase magnetic resonance imaging (MRI)
rg
pancreatic tumors and their outcome. Several markers of the abdomen with delayed images may also be helpful
have been indicated as potential predictors of aggressive to delineate the primary tumor or metastatic burden to
biology or metastatic disease such as increased HER2/neu the liver. An octreotide scan (somatostatin receptor
su
expression,17,18 tumor size larger than 2 cm, p16/MTS1 scintigraphy [SRS]) can be of great value in the preopera-
tumor suppressor gene inactivation,19 Ki67 proliferative tive localization of gastrinoma (Fig. 60.1). A prospective
index, and cytokeratin (CK) 19 expression.13 Recently, a study from the NIH compared the imaging methods in
://
study identified chromosomal instability and specific the localization of gastrinomas in 80 consecutive patients
chromosomal alterations to be reliable indicators for with ZES. The authors compared ultrasonography, CT,
tp
metastatic disease and poor tumor-free survival.13 Mutations MRI, selective angiography, and SRS performed using a
in the MEN1 gene are found with considerable variations radiolabeled octreotide colloid solution. The authors
in PNETS, and approximately 37% of gastrinomas harbor concluded that SRS was significantly better than all of the
ht
this mutation.48 The menin gene and its role in the car- conventional imaging methods in the identification of
cinogenesis of the syndrome has been well defined.49,50 gastrinomas later found at surgery, but that scintigraphy
The MEN1 tumor suppressor gene encodes for a still missed 20% of gastrinomas.52 Several studies have
610-amino-acid nuclear protein product called menin. confirmed these findings, whose authors advocated that
Germline mutations in the menin gene include nonsense, octreotide scan should be the initial imaging study in
missense, deletions, or RNA splicing defects. 51 These ZES.52–54 Nevertheless, based on current imaging modalities,
discoveries hold promise because they shed light on the the overall success of preoperative localization approaches
molecular underpinnings of carcinogenesis of gastrinoma. 70% to 80%. Endoscopic ultrasound (EUS) is an alternative
Future goals include developing novel diagnostic and imaging technique when cross-sectional imaging and SRS
perhaps therapeutic tools that will exploit these molecular have not revealed the location of the tumor. In addition,
targets. biopsy can be accomplished at the time of EUS. The
sensitivity of EUS to localize small PNETs is excellent (as
high as 97%) compared with CT (85%) or MRI (70%).55
TUMOR LOCALIZATION The majority of these tumors are found at the gastrinoma
The initial localization test should be cross-sectional triangle, first described by Stabile et al.56 This anatomic
imaging with computed tomography (CT) of the abdomen area is defined by the junction of the cystic duct to the
and pelvis, with fine cuts through the pancreas. Neuro- common bile duct, the transition of the head of the
endocrine tumors are hypervascular and therefore pancreas to the neck of the pancreas, and the transition
706 SECTION II Stomach and Small Intestine
INTRAOPERATIVE IMAGING
Few advances in imaging techniques over the past several
decades have advanced the surgical management of PNETs.
Even with current cross-sectional imaging, preoperative
localization can be difficult. Thus intraoperative imaging
techniques have been developed. Hall et al. combined
t
an intraoperative portable large-field-of-view gamma camera
ne
and a handheld gamma detection probe for 111In-
pentetreotide radioguided localization and confirmation
of gastrinoma in five patients.58
k.
The algorithm for the diagnosis and localization of
gastrinoma is summarized in Fig. 60.3. Localization of
oo
primary gastrinoma will be possible in 60% to 70%
FIGURE 60.2 Gastrinoma triangle. The majority of gastrinomas are
of ZES patients with sporadic gastrinoma. Patients with
found at an anatomic area that is defined by the junction of the
cystic duct and the common bile duct, the transition of the head
negative localization tests may be recommended to have
yb
of the pancreas to the neck of the pancreas, and the transition of exploration as well, but there is a possibility that no tumor
the second portion of the duodenum to the third. may be identified in as many as 15% to 30% of image-
negative patients. The principles of exploration are
summarized in the following section. Patients with MEN-
er
of the second portion of the duodenum to the third associated ZES with negative imaging should not undergo
(Fig. 60.2). exploration because cure is rare and because of meta-
rg
Alternatively, selective secretin arteriography may be chronous primary tumors as a result of the underlying
employed when CT, MRI, SRS, or EUS fail to localize the endocrinopathy.
tumor. This can localize up to 70% of small tumors
su
the superior mesenteric artery, and the proper hepatic cure, prevent disease progression, and prolong survival.
artery with sampling at 0, 20, 40, and 60 seconds. A step-up These patients benefit from a multidisciplinary team that
in hepatic vein gastrin will indicate the dominant blood includes surgeons, gastroenterologists, endocrinologists,
ht
supply of the tumor and its likely location. medical oncologists, and interventional radiologists.
New on the horizon is the 68Ga-DOTATATE positron Sporadic gastrinomas without metastatic disease are most
emission tomography (PET)/CT. With the availability of amenable to surgical resection and cure. The main
PET in recent years, PET tracers labeled with somatostatin principle is to perform a low-morbidity, complete resection
analogues have developed rapidly. With the help of of the disease with preservation of the maximum amount
68
Ge/68Ga generators, which immobilize the parent isotope of normal pancreas. Patients who have a positive preopera-
germanium, the PET radiolabeled tracers can be made tive localization should be offered surgical exploration.
less expensively for clinical practice. Twenty to 30% of patients will have negative localization
PNETs, including gastrinomas, are slow growing and tests; these patients may be offered exploration as well,
hence 18F-fluorodeoxyglucose (FDG) PET/CT is not accepting the possibility that no tumor will be identified
commonly used for initial evaluation. Due to the slow in as many as 15% of imaging-negative patients. When
metabolic activity of PNETs in the initial stages, they unable to locate the tumor preoperatively, the surgeon
are not extremely avid on 18F-FDG PET/CT. Conversely, should perform a thorough intraoperative search to locate
they are avid for 68Ga-DOTATATE, which demonstrate the gastrinoma. The principles of exploration include
high uptake because neuroendocrine tumors express (1) a wide Kocher maneuver to permit careful examination
significant somatostatin 2 receptor. Srirajaskanthan et al.57 of the head of the pancreas and uncinate process,
were the first to compare the 68Ga-DOTATATE to octreotide (2) mobilization of the body and tail of the pancreas to
scan. They evaluated the diagnostic and management permit bimanual palpation, (3) intraoperative ultrasound,
Zollinger-Ellison Syndrome CHAPTER 60 707
Equivocal?
Elevated?
Repeat serum gastrin
level a total of 3 times;
calculate mean
Not elevated?
Secretin stimulation test
Not elevated?
No response; consider
Gastrin 110 pg/mL calcium stimulation
Not gastrinoma
t
• 1st priority: Initiate antiacid regimen (PPI, 40 mg bid)
ne
Gastrinoma
• Screen for MEN1 (serum calcium)
k.
Localization studies:
oo
1. CT of abdomen and pelvis, triphasic, with fine cuts through the pancreas
2. Dual-phase MRI of the abdomen (to evaluate liver disease)
3. Octreotide scan (somatostatin receptor scintigraphy)
yb
Consider:
• Endoscopic ultrasound
• Angiography with selective intraarterial secretin injection and
subsequent measurement of gastrin
er
Head:
Body/Tail: Distal pancreatectomy & • <2 cm: enucleate
splenectomy • >2 cm: PD
ht
FIGURE 60.3 Flow chart depicting the diagnosis and management of Zollinger-Ellison syndrome. CT, Computed tomography; MRI,
magnetic resonance imaging.
708 SECTION II Stomach and Small Intestine
(4) duodenotomy for exploration of the duodenal mucosa, pancreatic body or tail lesions, distal pancreatectomy and
and (5) sampling of lymph nodes in the gastrinoma splenectomy are performed. In our institution, the decision
triangle. A thorough intraoperative ultrasound examina- to extirpate the gastrinoma in a patient with MEN1 is
tion of the pancreas is particularly important in those determined by imaging: (1) image-negative patients are
patients in whom preoperative imaging failed to localize observed and do not undergo exploration given the low
a tumor. cure rates with surgery; (2) image-positive patients with
A duodenotomy is the most critical step of the explora- no distant metastases undergo exploration for surgical
tion to assess for a duodenal primary tumor because the resection because resection has been shown to improve
duodenum is the most common location of gastrinomas, survival independent of a biochemical cure.38,39 The NIH
typically in the first or second portion. Tumors may occur group recommends surgery in patients with index lesions
in both the duodenum and pancreas, so the duodenum larger than 2 to 2.5 cm, whereas others do not use size
should be opened and explored in all cases, even when as a criterion.40 Invasive tumors or those more than 2 cm
a tumor has been found in the pancreas. The mucosa in the head of pancreas should usually be managed by a
should be inspected carefully because the tumors can be pancreaticoduodenectomy with periduodenal lymph node
small (<2 mm) and can be multiple. A lateral incision is dissection.8 This targeted approach avoids the risk of
made in a longitudinal direction, at the junction of the radical surgery and increases survival if R0/R1 resections
first and second portion, which can be extended proximally are accomplished, even without normalization of postopera-
or distally to allow better exposure. Brunner gland hyper- tive gastrin.2,38 In the Ohio State University experience of
t
plasia, which accompanies gastric acid hypersecretion, MEN1 patients operated upon with a curative intent, cure
ne
may produce small misleading nodules in the proximal was achieved in only 6% of patients.38 However, the 10-year
duodenum. If a nodule is identified, it should be excised survival with gastrinoma in MEN1 with R0/R1 resection
locally. Full-thickness excision may be performed for was 90%, compared with only 45% for patients having an
k.
lesions on the lateral duodenum; however, it is not neces- R2 resection or no resection. Because R2 resections do
sary in most cases because local recurrence after local not increase survival, MEN1 patients with extensive meta-
oo
excision is rare. The porta hepatis in the area of the static disease or locoregional spread that precludes
gastrinoma triangle should then be explored and lymph complete resection receive little benefit from surgical
nodes excised, followed by careful palpation and ultraso- resection and are typically not offered surgery.57,59
yb
nographic assessment of the pancreas. Once the tumor
is confirmed by frozen section, the duodenum is closed MANAGEMENT OF METASTATIC DISEASE
in a longitudinal direction. Management of pancreatic
primary tumors depends on the size and location. In Patients with malignant neuroendocrine tumors of the
er
general, pancreatic head tumors smaller than 2 cm may pancreas and duodenum frequently present with liver
be enucleated, unless they are close to the pancreatic metastases. The extent of disease burden in the liver often
rg
duct as determined by intraoperative ultrasound. Lesions dictates the quality and length of survival. Surgical excision
in the head larger than 2 cm or involving the pancreatic of both the primary tumor and liver metastases has been
duct generally require pancreaticoduodenectomy. Lesions reported in the literature; however, its use should be
su
in the body and tail of the pancreas are often near the considered carefully.60 Some authors suggest that select
pancreatic duct. Unless these clearly protrude from the patients may see a survival benefit if a complete or near-
surface of the gland (i.e., are exophytic), they require complete (90% or more) resection of hepatic metastases
://
distal pancreatectomy and splenectomy. Splenic preserva- can be achieved.61,62 Debulking of functional tumors can
tion is not recommended in this setting. be effective palliation of symptoms, but patient selection
tp
Biochemical cure of sporadic gastrinoma is reported must carefully weigh the individual risks and benefits and
in 30% to 50% of patients. However, recurrence has been whether symptoms can be controlled with medical manage-
documented in almost one-third of patients. The average ment. There is, however, controversy regarding whether
ht
time to recurrence is 5 to 10 years. Regardless of achieving surgical resection, with its attendant morbidity, is warranted
biochemical cure, complete resection of all gross tumor in metastatic disease. For nonfunctional tumors, complete
is associated with improved survival. The 10-year disease- resection of both the primary and liver metastases is the
specific survival in patients having R0/R1 resection of goal. With more effective systemic therapies becoming
sporadic gastrinoma is 85% compared to 40% for patients available, resection may become more commonplace even
having R2 resection and 25% for those having no in the setting of metastatic disease. Liver-directed therapy
resection.57 for patients with unresectable liver only or liver-dominant
The role of resection of gastrinoma in MEN1 patients metastases, who present with symptomatic disease is
is less clear. Some authorities consider cytoreduction beneficial for those with greater than 25% liver burden.
important in improving survival and preventing metastatic Transarterial chemoembolization (TACE), radionuclide-
disease gastrinoma in MEN1, even though long-lasting laden spheres (yittrium-90), or local ablative therapy
biochemical cure is rare. However, this recommendation (radiofrequency or microwave ablation) are all effective
is largely based on individual experience and there are liver-directed therapies.63–65 These modalities will not cure
no level I or level II data to support radical surgery in the patient, but they may provide effective cytoreduction
ZES in MEN1. Considering the rarity of cure and the of liver metastases, alleviate symptoms attributable to
indolent natural history of the disease, a targeted approach metastatic disease, and possibly extend survival. However,
with enucleation of tumors, while preserving the head of these modalities have not been compared to one another
the pancreas and duodenum, seems more prudent. For or to best supportive care. Hence, these patients are best
Zollinger-Ellison Syndrome CHAPTER 60 709
t
no growth.68 This growth rate needs to be considered in dictated by the protocol. If the patient is not enrolled in
ne
determining when and in whom antitumor therapy is a clinical trial, then gastrin levels should be determined
initiated, as well as in the assessment of response to biannually and imaging ordered if there is a significant
tumoricidal therapies. In comparison with targeted increase in the gastrin level. Repeat secretin provocative
k.
therapies, the objective response rates with cytotoxic tests are not necessary in this group of patients because
therapies are greater. Capecitabine and temozolomide it provides no additional diagnostic or prognostic informa-
oo
have been shown to have a high and durable response in tion. If there is progression of disease, then consultation
PNETs in a small study.69 Seventy percent of patients had with a medical oncologist is advised and a treatment plan
a radiographic response with median progression-free determined with a multidisciplinary team regarding sys-
yb
survival of 18 months. Given this radiographic response, temic and regional therapies.
this regimen has also been reported in the neoadjuvant
setting.70 PROGNOSIS
Comparisons of predictors of disease-specific survival show
er
TARGETED THERAPY that the odds of death from gastrinoma are equivalent in
There has recently been progress in developing agents sporadic and MEN1 patients.58 Lymph node metastases
rg
that exploit the molecular underpinnings of neuroendo- do not affect survival because its occurrence is not associ-
crine pathophysiology. Everolimus and sunitinib are both ated with poorer disease-specific survival or disease-free
FDA-approved treatments for advanced pancreatic endo- survival. Tumor size is independently predictive of disease
su
crine tumors.71 A randomized controlled trial of everolimus, survival.58 Tumor size is also highly predictive of distant
an oral inhibitor of mammalian target of rapamycin metastases, with the likelihood of such noted to be nearly
(mTor), showed an increase in progression-free survival 60% for primary tumors larger than 3 cm. In addition,
://
from 4.6 months to 11.0 months.72 Sunitinib, a multitar- primary tumor location affects outcomes: patients with
geted tyrosine kinase inhibitor, has also been shown to duodenal tumors have a better prognosis than those with
tp
increase progression-free survival to 11.4 months from pancreatic or combined pancreatic and duodenal prima-
5.5 months and increase in overall survival in patients ries. In terms of completeness of resection, patients with
with metastatic unresectable disease.73 Although both have R0 and R1 resection do far better than those with R2
ht
been shown to increase progression-free survival, response resection or those who cannot undergo resection. In other
rates by Response Evaluation Criteria In Solid Tumors words, patients who undergo surgical exploration resulting
(RECIST) criteria are low. Thus, when response is needed in R2 resection do not survive longer compared with
prior to resection, targeted therapies are not always the patients not undergoing resection.58
best choice given their low radiologic response rate. There
is no current data for the use of these agents in the
adjuvant setting. Continued investigation of these agents REFERENCES
should offer a better understanding of their role in patients 1. Zollinger RM, Ellison EH. Primary peptic ulcerations of the jejunum
with advanced neuroendocrine tumors and could addition- associated with islet cell tumors of the pancreas. Ann Surg.
ally illuminate the molecular basis of these tumors, thus 1955;142(4):709-723 [discussion 724-748].
2. Ellison EC, Johnson JA. The Zollinger-Ellison syndrome: a compre-
expanding the therapeutic armamentarium. hensive review of historical, scientific, and clinical considerations.
Curr Probl Surg. 2009;46(1):13-106.
SOMATOSTATIN ANALOGUES 3. Davies K, Conlon KC. Neuroendocrine tumors of the pancreas. Curr
Both octreotide and lanreotide have been shown to prolong Gastroenterol Rep. 2009;11(2):119-127.
progression-free survival; however overall survival has not 4. Modlin IM, Lye KD, Kidd M. A 5-decade analysis of 13,715 carcinoid
tumors. Cancer. 2003;97(4):934-959.
been significantly increased. These drugs are thought to 5. Cheema A, Weber J, Strosberg JR. Incidental detection of pancreatic
stabilize tumor growth in addition to relieving symptoms neuroendocrine tumors: an analysis of incidence and outcomes.
associated with functional tumors.32,33 Ann Surg Oncol. 2012;19(9):2932-2936.
710 SECTION II Stomach and Small Intestine
6. Yao JC, Eisner MP, Leary C, et al. Population-based study of islet proton-pump inhibitor therapy: a systematic review and meta-analysis.
cell carcinoma. Ann Surg Oncol. 2007;14(12):3492-3500. PLoS One. 2015;10(6):e0128004.
7. Lawrence B, Gustafsson BI, Chan A, Svejda B, Kidd M, Modlin IM. 29. Biswal S. Proton pump inhibitors and risk for Clostridium difficile
The epidemiology of gastroenteropancreatic neuroendocrine tumors. associated diarrhea. Biomed J. 2014;37(4):178-183.
Endocrinol Metab Clin North Am. 2011;40(1):1-18 [vii]. 30. Ngamruengphong S, Leontiadis GI, Radhi S, Dentino A, Nugent
8. Halfdanarson TR, Rabe KG, Rubin J, Petersen GM. Pancreatic K. Proton pump inhibitors and risk of fracture: a systematic review
neuroendocrine tumors (PNETs): incidence, prognosis and recent and meta-analysis of observational studies. Am J Gastroenterol.
trend toward improved survival. Ann Oncol. 2008;19(10):1727- 2011;106:1209-1218 [quiz 1219].
1733. 31. Corleto VD, Annibale B, Gibril F, et al. Does the widespread use of
9. Metz DC, Jensen RT. Gastrointestinal neuroendocrine tumors: proton pump inhibitors mask, complicate and/or delay the diagnosis
pancreatic endocrine tumors. Gastroenterology. 2008;135(5):1469-1492. of Zollinger-Ellison syndrome? Aliment Pharmacol Ther. 2001;15
10. Jensen RT, Berna MJ, Bingham DB, Norton JA. Inherited pancreatic (10):1555-1561.
endocrine tumor syndromes: advances in molecular pathogenesis, 32. Caplin ME, Pavel M, Ćwikła JB, et al. Lanreotide in metastatic
diagnosis, management, and controversies. Cancer. 2008;113(7 enteropancreatic neuroendocrine tumors. N Engl J Med. 2014;371(3):
suppl):1807-1843. 224-233.
11. Burgess JR, David R, Parameswaran V, Greenaway TM, Shepherd 33. Rinke A, Müller HH, Schade-Brittinger C, et al. Placebo-controlled,
JJ. The outcome of subtotal parathyroidectomy for the treatment double-blind, prospective, randomized study on the effect of
of hyperparathyroidism in multiple endocrine neoplasia type 1. Arch octreotide LAR in the control of tumor growth in patients with
Surg. 1998;133(2):126-129. metastatic neuroendocrine midgut tumors: a report from the
12. Skogseid B, Rastad J, Gobl A, et al. Adrenal lesion in multiple PROMID Study Group. J Clin Oncol. 2009;27(28):4656-4663.
endocrine neoplasia type 1. Surgery. 1995;118(6):1077-1082. 34. Ellison EC, O’Dorisio TM, Sparks J, et al. Observations on the effect
13. Solcia E, Capella C, Buffa R, Usellini L, Frigerio B, Fontana P. of a somatostatin analog in the Zollinger-Ellison syndrome: implica-
t
Endocrine cells of the gastrointestinal tract and related tumors. tions for the treatment of apudomas. Surgery. 1986;100(2):437-444.
ne
Pathobiol Annu. 1979;9:163-204. 35. Thakker RV, Newey PJ, Walls GV, et al. Clinical practice guidelines
14. Jonkers YM, Claessen SM, Perren A, et al. DNA copy number status for multiple endocrine neoplasia type 1 (MEN1). J Clin Endocrinol
is a powerful predictor of poor survival in endocrine pancreatic Metab. 2012;97(9):2990-3011.
tumor patients. Endocr Relat Cancer. 2007;14(3):769-779. 36. Kidd M, Bodei L, Modlin IM. Chromogranin A: any relevance in
k.
15. Klöppel G, Willemer S, Stamm B, Häcki WH, Heitz PU. Pancreatic neuroendocrine tumors? Curr Opin Endocrinol Diabetes Obes. 2016;
lesions and hormonal profile of pancreatic tumors in multiple 23(1):28-37.
endocrine neoplasia type I. An immunocytochemical study of nine 37. Paik WH, Ryu JK, Song BJ, et al. Clinical usefulness of plasma
oo
patients. Cancer. 1986;57(9):1824-1832. chromogranin A in pancreatic neuroendocrine neoplasm. J Korean
16. Abood GJ, Go A, Malhotra D, Shoup M. The surgical and systemic Med Sci. 2013;28(5):750-754.
management of neuroendocrine tumors of the pancreas. Surg Clin 38. Hijioka M, Ito T, Igarashi H, et al. Serum chromogranin A is a useful
yb
North Am. 2009;89:249-266 [x]. marker for Japanese patients with pancreatic neuroendocrine tumors.
17. Gibril F, Jensen RT. Advances in evaluation and management of Cancer Sci. 2014;105(11):1464-1471.
gastrinoma in patients with Zollinger-Ellison syndrome. Curr Gas- 39. Qiao XW, Qiu L, Chen YJ, et al. Chromogranin A is a reliable serum
troenterol Rep. 2005;7(2):114-121. diagnostic biomarker for pancreatic neuroendocrine tumors but
er
18. Roy PK, Venzon DJ, Shojamanesh H, et al. Zollinger-Ellison syndrome. not for insulinomas. BMC Endocr Disord. 2014;14:64.
Clinical presentation in 261 patients. Medicine (Baltimore). 40. O’Toole D, Grossman A, Gross D, et al. ENETS Consensus Guidelines
2000;79:379-411. for the Standards of Care in neuroendocrine tumors: biochemical
rg
19. Zollinger RM, Grant GN. Ulcerogenic tumor of the pancreas. J Am markers. Neuroendocrinology. 2009;90(2):194-202.
Med Assoc. 1964;190:181-184. 41. Vinik AI, Woltering EA, Warner RR, et al. NANETS consensus
20. Berna MJ, Hoffmann KM, Serrano J, Gibril F, Jensen RT. Serum guidelines for the diagnosis of neuroendocrine tumor. Pancreas.
su
21. Hansky J, Soveny C, Korman MG. Effect of secretin on serum gastrin neoplasia type 1. Surgery. 1997;122:1012-1019 [discussion 1019–1020].
as measured by immunoassay. Gastroenterology. 1971;61(1):62-68. 43. Walter T, Chardon L, Chopin-laly X, et al. Is the combination of
22. Isenberg JI, Walsh JH, Passaro E Jr, Moore EW, Grossman MI. chromogranin A and pancreatic polypeptide serum determinations
tp
Unusual effect of secretin on serum gastrin, serum calcium, and of interest in the diagnosis and follow-up of gastro-entero-pancreatic
gastric acid secretion in a patient with suspected Zollinger-Ellison neuroendocrine tumours? Eur J Cancer. 2012;48(12):1766-1773.
syndrome. Gastroenterology. 1972;62(4):626-631. 44. Massironi S, Conte D, Sciola V, et al. Plasma chromogranin A
ht
23. Berna MJ, Hoffmann KM, Long SH, Serrano J, Gibril F, Jensen RT. response to octreotide test: prognostic value for clinical outcome
Serum gastrin in Zollinger-Ellison syndrome: II. Prospective study in endocrine digestive tumors. Am J Gastroenterol. 2010;105(9):2072-2078.
of gastrin provocative testing in 293 patients from the National 45. Ahmed A, Turner G, King B, et al. Midgut neuroendocrine tumours
Institutes of Health and comparison with 537 cases from the literature. with liver metastases: results of the UKINETS study. Endocr Relat
Evaluation of diagnostic criteria, proposal of new criteria, and Cancer. 2009;16(3):885-894.
correlations with clinical and tumoral features. Medicine (Baltimore). 46. Gibril F, Schumann M, Pace A, Jensen RT. Multiple endocrine
2006;85:331-364. neoplasia type 1 and Zollinger-Ellison syndrome: a prospective study
24. Deveney CW, Deveney KS, Jaffe BM, Jones RS, Way LW. Use of of 107 cases and comparison with 1009 cases from the literature.
calcium and secretin in the diagnosis of gastrinoma (Zollinger-Ellison Medicine (Baltimore). 2004;83(1):43-83.
syndrome). Ann Intern Med. 1977;87(6):680-686. 47. Waldmann J, Fendrich V, Habbe N, et al. Screening of patients with
25. Shibata C, Funayama Y, Fukushima K, et al. The glucagon provocative multiple endocrine neoplasia type 1 (MEN-1): a critical analysis of
test for the diagnosis and treatment of Zollinger-Ellison syndrome. its value. World J Surg. 2009;33(6):1208-1218.
J Gastrointest Surg. 2008;12(2):344-349. 48. Duerr EM, Chung DC. Molecular genetics of neuroendocrine
26. Hoorn EJ, van der Hoek J, de Man RA, Kuipers EJ, Bolwerk C, Zietse tumors. Best Pract Res Clin Endocrinol Metab. 2007;21(1):1-14.
R. A case series of proton pump inhibitor-induced hypomagnesemia. 49. Starker LF, Carling T. Molecular genetics of gastroenteropancreatic
Am J Kidney Dis. 2010;56(1):112-116. neuroendocrine tumors. Curr Opin Oncol. 2009;21(1):29-33.
27. Melloni C, Washam JB, Jones WS, et al. Conflicting results between 50. Lairmore TC, Chen H. Role of menin in neuroendocrine tumori-
randomized trials and observational studies on the impact of proton genesis. Adv Exp Med Biol. 2009;668:87-95.
pump inhibitors on cardiovascular events when coadministered with 51. Lairmore TC, Quinn CE, Martinez MJ. Neuroendocrine tumors of
dual antiplatelet therapy: systematic review. Circ Cardiovasc Qual the pancreas: molecular pathogenesis and current surgical manage-
Outcomes. 2015;8(1):47-55. ment. Translation Gastrointest Cancer. 2013;3:29-43.
28. Lambert AA, Lam JO, Paik JJ, Ugarte-Gil C, Drummond MB, Crowell 52. Gibril F, Reynolds JC, Doppman JL, et al. Somatostatin receptor
TA. Risk of community-acquired pneumonia with outpatient scintigraphy: its sensitivity compared with that of other imaging
Zollinger-Ellison Syndrome CHAPTER 60 711
methods in detecting primary and metastatic gastrinomas. A prospec- 64. Rhee TK, Lewandowski RJ, Liu DM, et al. 90Y Radioembolization
tive study. Ann Intern Med. 1996;125(1):26-34. for metastatic neuroendocrine liver tumors: preliminary results from
53. Cadiot G, Bonnaud G, Lebtahi R, et al. Usefulness of somatostatin a multi-institutional experience. Ann Surg. 2008;247(6):1029-1035.
receptor scintigraphy in the management of patients with Zollinger- 65. Mazzaglia PJ, Berber E, Milas M, Siperstein AE. Laparoscopic
Ellison syndrome. Groupe de Recherche et d’Etude du Syndrome radiofrequency ablation of neuroendocrine liver metastases: a 10-year
de Zollinger-Ellison (GRESZE). Gut. 1997;41:107-114. experience evaluating predictors of survival. Surgery. 2007;142(1):10-19.
54. Jensen RT, Gibril F, Termanini B. Definition of the role of somatostatin 66. Martin RC, Scoggins CR, McMasters KM. Safety and efficacy of
receptor scintigraphy in gastrointestinal neuroendocrine tumor microwave ablation of hepatic tumors: a prospective review of a
localization. Yale J Biol Med. 1997;70(5-6):481-500. 5-year experience. Ann Surg Oncol. 2010;17(1):171-178.
55. Fujimori N, Osoegawa T, Lee L, et al. Efficacy of endoscopic 67. Moertel CG, Lefkopoulo M, Lipsitz S, Hahn RG, Klaassen D.
ultrasonography and endoscopic ultrasonography-guided fine-needle Streptozocin-doxorubicin, streptozocin-fluorouracil or chlorozotocin
aspiration for the diagnosis and grading of pancreatic neuroendocrine in the treatment of advanced islet-cell carcinoma. N Engl J Med.
tumors. Scand J Gastroenterol. 2016;51(2):245-252. 1992;326(8):519-523.
56. Stabile BE, Morrow DJ, Passaro E Jr. The gastrinoma triangle: 68. Ramanathan RK, Cnaan A, Hahn RG, Carbone PP, Haller DG. Phase
operative implications. Am J Surg. 1984;147(1):25-31. II trial of dacarbazine (DTIC) in advanced pancreatic islet cell
57. Srirajaskanthan R, Kayani I, Quigley AM, Soh J, Caplin ME, Bomanji carcinoma. Study of the Eastern Cooperative Oncology Group-E6282.
J. The role of 68Ga-DOTATATE PET in patients with neuroendocrine Ann Oncol. 2001;12(8):1139-1143.
tumors and negative or equivocal findings on 111In-DTPA-octreotide 69. Sutliff VE, Doppman JL, Gibril F, et al. Growth of newly diagnosed,
scintigraphy. J Nucl Med. 2010;51(6):875-882. untreated metastatic gastrinomas and predictors of growth patterns.
58. Hall NC, Nichols SD, Povoski SP, et al. Intraoperative use of a J Clin Oncol. 1997;15(6):2420-2431.
portable large field of view gamma camera and handheld gamma 70. Strosberg JR, Fine RL, Choi J, et al. First-line chemotherapy with
detection probe for radioguided localization and prediction of capecitabine and temozolomide in patients with metastatic pancreatic
t
complete surgical resection of gastrinoma: proof of concept. J Am endocrine carcinomas. Cancer. 2011;117(2):268-275.
ne
Coll Surg. 2015;221(2):300-308. 71. Devata S, Kim EJ. Neoadjuvant chemotherapy with capecitabine
59. Ellison EC, Sparks J, Verducci JS, et al. 50-Year appraisal of gastrinoma: and temozolomide for unresectable pancreatic neuroendocrine
recommendations for staging and treatment. J Am Coll Surg. 2006; tumor. Case Rep Oncol. 2012;5(3):622-626.
202(6):897-905. 72. Oberstein PE, Saif MW. Update on novel therapies for pancreatic
k.
60. Bloomston M, Muscarella P, Shah MH, et al. Cytoreduction results neuroendocrine tumors. JOP. 2012;13(4):372-375.
in high perioperative mortality and decreased survival in patients 73. Pavel ME, Hainsworth JD, Baudin E, et al. Everolimus plus octreotide
undergoing pancreatectomy for neuroendocrine tumors of the long-acting repeatable for the treatment of advanced neuroendocrine
oo
pancreas. J Gastrointest Surg. 2006;10(10):1361-1370. tumours associated with carcinoid syndrome (RADIANT-2): a ran-
61. Norton JA, Warren RS, Kelly MG, Zuraek MB, Jensen RT. Aggressive domised, placebo-controlled, phase 3 study. Lancet. 2011;378(9808):
surgery for metastatic liver neuroendocrine tumors. Surgery. 2005-2012.
yb
2003;134:1057-1063 [discussion 1063–1065]. 74. Raymond E, Dahan L, Raoul JL, et al. Sunitinib malate for the
62. Chamberlain RS, Canes D, Brown KT, et al. Hepatic neuroendocrine treatment of pancreatic neuroendocrine tumors. N Engl J Med.
metastases: does intervention alter outcomes? J Am Coll Surg. 2011;364(6):501-513.
2000;190(4):432-445. 75. Metz DC, Benya RV, Fishbeyn VA, et al. Prospective study of the
er
63. Chen H, Hardacre JM, Uzar A, Cameron JL, Choti MA. Isolated need for long-term antisecretory therapy in patients with Zollinger-
liver metastases from neuroendocrine tumors: does resection prolong Ellison syndrome following successful curative gastrinoma resection.
survival? J Am Coll Surg. 1998;187:88-92 [discussion 92–93]. Aliment Pharmacol Ther. 1993;7(3):247-257.
rg
su
://
tp
ht
CHAPTER
G
astric surgery for benign and malignant diseases Certain ethnicities are at risk for the development of
has decreased significantly over the last few decades, gastric adenocarcinoma. Japanese, Koreans, Vietnamese,
but the multimodal management of gastric cancer Native Americans, and people of Pacific Island descent
requires that the surgeon be well-versed in the medical are at the greatest risk. In contrast, Filipinos and Caucasians
and surgical facets of care. The purpose of this work is are at the lowest risk, and Chinese, Latinos, and people
to provide an overview of the management of gastric of African descent are at intermediate risk.2
cancer with specific emphasis on the perioperative care. Familial syndromes such as Peutz-Jeghers and familial
adenomatous polyposis (FAP) harbor an increased risk
t
of gastric cancer. Mutations in proteins such as E-cadherin,
EPIDEMIOLOGY
ne
p53, and BRCA2 have also been shown to increase the
In 2016, the American Cancer Society estimated that in potential for the development of gastric adenocarcinoma.1
the United States 26,370 people will be diagnosed with A complete list of risk factors is presented in Table 61.1.
k.
gastric cancer with 10,730 deaths.1 Worldwide, however,
gastric cancer remains the fifth most common cancer and PATHOLOGY
oo
a leading cause of cancer mortality. The incidence of
gastric cancer has considerable geographic variability with Ninety to 95% of gastric cancers are adenocarcinoma with
a significantly higher occurrence in Asia and Latin America the remainder being attributed to lymphoma, gastro-
than in North America and Europe.2 intestinal stromal tumors, and carcinoid tumors.1 Although
yb
The average age of diagnosis in the United States is several histopathology classification systems exist, the most
69 years of age with the majority of patients diagnosed frequently used is the Lauren classification. This classifies
in the seventh decade of life and later. Men are more gastric adenocarcinomas as intestinal (well differentiated)
er
likely to have gastric cancer than women, and Hispanic and diffuse (poorly differentiated).
Americans, African Americans, and Asian/Pacific Islanders Intestinal-type adenocarcinomas are derived from the
rg
are more frequently affected than non-Hispanic whites. gastric mucosa and form glands. They are more frequently
Individuals with lower socioeconomic status are more associated with hematogenous metastases and are observed
likely to be affected in both the United States and in in elderly patients, men, or those individuals in high-risk
su
developing countries.1 populations. They are also more common in the distal
Since 1930, the incidence of gastric cancer has decreased portion of the stomach. This type of adenocarcinoma is
significantly, although the reasons for this change are associated with the risk factors of H. pylori infection,
://
unclear. The incidence of tumors located distally within chronic atrophic gastritis, intestinal metaplasia, and diets
the stomach have decreased, whereas the incidence of high in nitrosamines.5–6
tp
more proximal gastric tumors has increased. Despite the Diffuse-type adenocarcinoma arises from the lamina
decreasing incidence, gastric cancer remains highly lethal propria and spreads through the submucosa. Lymphatic
in the United States with an anticipated overall 5-year metastases are more common with diffuse-type cancers,
ht
survival rate of 29%.1,3 which are more predominant in younger patients and
women. In contrast to intestinal type, diffuse-type adeno-
carcinomas are not gland-forming and are more frequently
RISK FACTORS observed in the proximal stomach. In addition, diffuse-type
Known risk factors for gastric adenocarcinoma include cancers may have transmural extension, develop peritoneal
Helicobacter pylori infection, a history of mucosa-associated metastases, and are more aggressive overall. Linitis plastica,
lymphoid tissue (MALT) lymphoma, the presence of involvement of the entire stomach, is a rare and aggressive
adenomatous gastric polyps, previous gastric operations, form of diffuse-type cancer that constitutes less than 10%
pernicious anemia, atrophic gastritis, intestinal metaplasia, of all gastric adenocarcinomas.5,6
exposure to nitrosamines from cured or smoked foods,
tobacco use, and family history. A relationship between
obesity and gastric cancer has not been identified defini-
DIAGNOSIS
tively, although a hypothesis exists that increasing rates The early signs and symptoms of gastric cancer are non-
of gastroesophageal reflux disease (GERD) associated specific and include nausea and epigastric pain. As a
with obesity may predispose individuals to more proximal result, the majority of cancers are diagnosed at advanced
tumors. In addition, lower rates of gastric cancer are stages. Physical exam findings are also nonspecific, but
observed in individuals with a diet high in fresh fruits may include palpable lymph nodes such as the Sister Mary
and vegetables.1–4 Joseph node in the periumbilical region, the Virchow
712
Gastric Adenocarcinoma CHAPTER 61 712.e1
ABSTRACT
In 2016, the American Cancer Society estimated that in
the United States 26,370 people will be diagnosed with
gastric cancer, with 10,730 deaths. Worldwide, however,
gastric cancer remains the fifth most common cancer and
a leading cause of cancer mortality. Although the incidence
of gastric cancer is declining, it remains a highly lethal
disease. Known risk factors for gastric adenocarcinoma
include Helicobacter pylori infection, a history of mucosa-
associated lymphoid tissue (MALT) lymphoma, the
presence of adenomatous gastric polyps, previous gastric
operations, pernicious anemia, atrophic gastritis, intestinal
metaplasia, exposure to nitrosamines from cured or
smoked foods, tobacco use, and family history. Eradication
and treatment of H. Pylori has assisted in the declining
incidence. The early signs and symptoms of gastric cancer
are nonspecific and include nausea and epigastric pain.
As a result, the majority of cancers are diagnosed at
t
ne
advanced stages. A high index of suspicion is necessary
to diagnose gastric adenocarcinoma at an early stage.
Diagnosis and staging require endoscopy with biopsy,
endoscopic ultrasound, and CT scan to evaluate for
k.
locoregionally advanced or metastatic disease. Diagnostic
laparoscopy and peritoneal cytology are important com-
oo
ponents of staging for advanced tumors. Surgery remains
the only chance for cure, but it must be accompanied by
perioperative chemotherapy or postoperative chemoradia-
yb
tion. Palliation with radiation, chemotherapy, endoscopic
stenting, or surgery are indicated for appropriate patients
with advanced or metastatic disease.
er
KEYWORDS
Gastric adenocarcinoma, gastric cancer
rg
su
://
tp
ht
Gastric Adenocarcinoma CHAPTER 61 713
t
Pernicious anemia Gastric cancers and esophageal cancers are managed
ne
Family history differently with respect to neoadjuvant, adjuvant, and
Hereditary nonpolyposis colorectal cancer surgical therapy. Esophagogastric junction (EGJ) tumors
Li-Fraumeni syndrome are assessed using the Siewert classification, which for
k.
Peutz-Jeghers syndrome gastric adenocarcinoma includes only the Siewert type III
Familial adenomatous polyposis
lesions. These tumors are defined as subcardial carcinomas
oo
PRECURSORS with the tumor epicenter located 2 to 5 cm below the
Adenoma EGJ with infiltration of the EGJ and esophagus from
Atrophic gastritis below. Siewert type I and II lesions are considered esopha-
Dysplasia
yb
geal cancers and their management is beyond the scope
Intestinal metaplasia of this chapter.11
Ménétrier disease Endoscopic screening programs in endemic areas are
recommended and have been shown to diagnose tumors
er
PROTECTIVE
Raw vegetables at an earlier stage. This practice is not applicable to areas
Citrus fruits where the incidence of gastric cancer is low, and thus,
rg
STAGING
node in the left supraclavicular region, or the Blumer The American Joint Committee on Cancer (AJCC) TNM
://
shelf, which is a palpable prerectal drop metastases that system is the most widely used staging system for gastric
may be evident on digital rectal exam. Development of cancer. The system assesses the primary tumor (T), the
tp
a palpable abdominal mass or ascites are findings of presence of lymph node involvement (N), and the presence
advanced disease.7 of metastatic disease (M). T stage is based on the depth
As with all suspected malignancies, a complete history of invasion of the tumor. N stage identifies lymph node
ht
and physical examination are essential. Laboratory values involvement and requires assessment of at least 15 lymph
including a complete blood count and chemistry and nodes. Metastatic disease is identified as distant metastasis
nutritional parameters should be obtained. Many tumor including positive cytology from peritoneal washings.
markers may be elevated in the setting of gastric cancer, Complete descriptions of all TNM levels and stages are
including carcinoembryonic antigen (CEA), CA-125, CA presented in Tables 61.2 and 61.3.13
19-9, and β-HCG. These biomarkers, however, lack suf-
ficient sensitivity and specificity to establish a diagnosis.
Imaging with computed tomography (CT) of the chest,
TUMOR MANAGEMENT
abdomen, and pelvis with oral and intravenous contrast Despite significant advances in the multimodal treatment
should be obtained, but the diagnosis is established of gastric cancer, operative resection remains the best
definitively with an upper endoscopy and biopsy confirma- chance for cure. Survival varies greatly based upon the
tion of an adenocarcinoma. Endoscopic ultrasound (EUS) stage of the tumor at the time of resection (Fig. 61.1).11
accurately assesses the depth of tumor invasion and Preoperative staging with EUS and cross-sectional imaging
enlargement of perigastric lymph nodes, although this is recommended. Once diagnosis and staging are complete,
technology may not be available in all facilities.7–11 formal multidisciplinary evaluation should be performed
Cross-sectional imaging with CT scan or magnetic to determine the optimal treatment strategy.
resonance imaging (MRI) is useful for the evaluation of Stage IA tumors may be managed endoscopically with
metastatic disease. Positron emission tomography (PET) endoscopic mucosal resection (EMR) or endoscopic
714 SECTION II Stomach and Small Intestine
TUMOR
Tx Primary tumor cannot be assessed
T0 No evidence of primary tumor
Tis Carcinoma in situ; intraepithelial tumor without invasion of the lamina propria
T1 Tumor invades lamina propria, muscularis mucosa, or submucosa
T1a Tumor invades lamina propria or muscularis mucosa
T1b Tumor invades submucosa
T2 Tumor invades muscularis propria
T3 Tumor penetrates subserosal connective tissue without invasion of visceral peritoneum or adjacent structures
T4 Tumor invades serosa (visceral peritoneum) or adjacent structures
T4a Tumor invades serosa (visceral peritoneum)
T4b Tumor invades adjacent structures
LYMPH NODES
Nx Regional lymph nodes cannot be assessed
N0 No regional lymph node metastasis
t
N1 Metastasis in 1–2 regional lymph nodes
ne
N2 Metastasis in 3–6 regional lymph nodes
N3 Metastasis in 7 or more regional lymph nodes
METASTASES
k.
Mx Distant metastases cannot be assessed
M0 No distant metastases
oo
M1 Distant metastases (includes peritoneal cytology)
Modified from AJCC Cancer Staging Manual. 7th ed. 2009.
yb
Survival
rate
TABLE 61.3 TNM Staging of Gastric Adenocarcinoma 100
er
Stage T N M
80
Stage 0 Tis N0 M0
rg
Stage IA T1 N0 M0
60
Stage IB T2 N0 M0
T1 N1 M0
su
Stage IIA T3 N0 M0 40
T2 N1 M0
T1 N2 M0 20
://
T2 N2 M0
At Dx 1 2 3 4 5
T1 N3 M0
Stage IIIA T4a N1 M0 IA 100.0 90.2 84.8 79.8 74.8 70.8
ht
In addition, the presence of distant metastases or peri- intragastric margins of 5 cm are recommended; however,
toneal seeding (stage IV tumors) negates the potential a margin of this distance may need to be increased for
for operative cure, and these patients should receive diffuse-type cancers. In addition, an omentectomy and
chemotherapy. lymph node dissection should be conducted.11 Intraopera-
tive frozen section must be performed on the proximal
margin, and intraoperative analysis of the distal margin
MULTIMODALITY THERAPY is highly recommended to ensure the presence of duodenal
Many clinical trials and meta-analyses have evaluated mucosa.
the role of perioperative chemotherapy, preoperative Proximal tumors of the cardia, including Siewert type
chemoradiation, and postoperative chemotherapy or III lesions, are best managed with total gastrectomy with
chemoradiation with conflicting results. Overall, sur- Roux-en-Y esophagojejunostomy reconstruction. This is
vival in patients with stage IB or higher gastric cancer preferred over a proximal gastrectomy with pyloroplasty
is significantly improved with the addition of adjuvant due to the incidence of alkaline reflux esophagitis. Distal
therapy following resection. Both the Macdonald protocol, lesions, including those in the body and antrum, should
5-fluorouracil (5-FU)-based adjuvant chemoradiotherapy, be extirpated via subtotal (or near total) gastrectomy to
and the Medical Research Council Adjuvant Gastric Cancer achieve negative margins as total gastrectomy, despite
Infusional Chemotherapy (MAGIC) trial protocol, with pre- presumed greater margins, does not demonstrate a survival
and postoperative chemotherapy for resectable cancers, benefit, and patients report improved quality of life with
t
demonstrated significant improvement in overall survival a subtotal gastrectomy versus total gastrectomy.18 Recon-
ne
and disease-free survival. Patient factors, including the struction with a Billroth I gastroduodenostomy is the
ability to tolerate the aggressive neoadjuvant protocol reconstruction of choice because it preserves natural
prescribed in the MAGIC trial or the ability to toler- enteric flow. Loop gastrojejunostomy (Billroth II) is an
k.
ate radiation as in the Macdonald protocol should be alternative reconstruction that is performed frequently.
individualized. Tumor-specific factors, including nodal Roux-en-Y gastrojejunostomy is another preferred method
oo
involvement and location within the stomach, should also of anastomotic reconstruction; however it is associated
guide perioperative planning and the appropriateness with the Roux stasis syndrome and poor gastric remnant
of operative resection. Institutional approaches to the function.
yb
treatment of gastric cancer may vary significantly, and the Placement of a temporary jejunal feeding tube to assist
treatment plan should be developed in a multidisciplinary in postoperative nutritional recovery is recommended for
setting prior to initiating any treatment.14 all patients. Nasogastric tube decompression may be
beneficial following partial gastrectomy but is not recom-
er
The MAGIC trial evaluated the role of pre- and postopera- frequently, especially following total gastrectomy with
tive chemotherapy for patients with resectable gastro- esophagojejunostomy.
esophageal cancers. A total of 503 patients with stage II Laparoscopic resection for early gastric cancer has
su
or higher gastric and esophageal cancer were randomized been performed for more than a decade with excellent
in two groups. One group received three cycles of che- results. Several trials have demonstrated its benefits when
motherapy (5-FU, epirubicin, and cisplatin) prior to compared to the open approach, including decreased
://
surgery with curative intent, followed by an additional pain, length of hospital stay, blood loss, and complica-
three cycles of chemotherapy, whereas the other group tions.19 Most recently, the Korean Laparoscopic Gastro-
tp
was treated with surgery alone. There was a significant intestinal Surgery Study (KLASS)-01 trial demonstrated
improvement in 5-year survival (36% vs 23%, P = .009), decreased morbidity and wound infections with the lapa-
a longer interval of progression-free survival, and a decrease roscopic approach without compromising overall survival
ht
in local recurrence rates with pre- and postoperative for stage I gastric cancers.20
chemotherapy compared to surgery alone. In addition,
tumor downstaging was observed.15 LYMPHADENECTOMY
These results were comparable to the Actions Concertées The extent of lymphadenectomy associated with gastrec-
dans les Cancer Colorectaux et Digestifs (ACCORD) 07 tomy has long been controversial. Gastric adenocarcinoma
trial, which demonstrated downsizing of tumor and nodal is accompanied by lymph node metastases in more than
stages with pre- and postoperative chemotherapy. 16 In half of patients at the time of initial presentation or
both of these trials, however, 50% or less of the patients resection. Lymphadenectomy at the time of gastrectomy
completed all cycles of chemotherapy. Thus, the effects has been shown to improve staging accuracy and is the
observed are likely due to the preoperative chemotherapy, standard of care. In 1988, 16 stations of lymph node
thereby demonstrating the significant role of neoadjuvant drainage for the stomach were first described (Table 61.4)
chemotherapy.17 and subsequently expanded by the Japanese.21 Dissection
of stations 1 to 6, a D1 lymphadenectomy, refers to dis-
section of the perigastric lymph nodes. A D2 lymphadenec-
SURGICAL THERAPY tomy includes stations 1 to 11 and involves removal of
Overall survival is improved significantly with complete the perigastric lymph nodes as well as the lymph nodes
surgical resection. Curative intent gastrectomy requires extending along the hepatic, left gastric, celiac, and splenic
an R0 resection. To achieve margin-negative resection, arteries. Traditionally, this has also included a distal
716 SECTION II Stomach and Small Intestine
t
11d Distal splenic artery adenectomy with improved disease-specific survival. Current
ne
12a Hepatoduodenal ligament (along the hepatic artery) NCCN guidelines recommend D2 lymphadenectomy in
12b Hepatoduodenal ligament (along the bile duct) the “hands of experienced surgeons with expertise in the
12p Hepatoduodenal ligament (behind the portal vein)
field, at tertiary centers where gastrectomies are often
k.
13 Posterior surface of pancreatic head
performed.” To decrease morbidity associated with this
14v Superior mesenteric vein
14a Superior mesenteric artery
extended lymphadenectomy, a separate retroperitoneal
oo
15 Middle colic vessels lymphadenectomy may be performed after removal of
16a1 Aortic hiatus the gastric and omental specimens. Splenectomy and
16a2 Abdominal aorta (from celiac trunk to left renal vein) distal pancreatectomy are performed only when necessary
based on tumor involvement.11,24
yb
16b1 Abdominal aorta (from left renal vein to inferior
mesenteric artery) Sentinel lymph node biopsy has been proposed by
16b2 Abdominal aorta (from inferior mesenteric artery to several Japanese authors to determine the need for
lymphadenectomy in early gastric cancers. The relevance
er
aortic bifurcation)
17 Anterior surface of pancreatic head of this procedure to Western gastric cancers has been
18 Inferior margin of pancreas questioned. Early gastric cancers may harbor lymph node
rg
1 2 1 2
ht
4 4
7 10
3 12 3
9 11 10
8
5 5
3 4 3 4
4 4
6 4 4 6 4 4
14
D1 lymphadenectomy D2 lymphadenectomy
FIGURE 61.2 D1 vs D2 lymphadenectomy. Numbers reference lymph node stations as listed in Table 61.4. White numbers refer to D1
lymphadenectomy. Green numbers are additional stations included in D2 lymphadenectomy. (Courtesty Dr. Steven J. Hughes.)
Gastric Adenocarcinoma CHAPTER 61 717
t
with a D2 lymphadenectomy. Thus, adjuvant chemotherapy improved, the overall 5-year survival rate remains poor
ne
does not confer a distinct survival benefit.27 at 29%. Prognosis correlates with the stage of disease at
Gastric cancer is generally radioresistant, and radiation initial presentation (see Fig. 61.1). Neoadjuvant chemo-
therapy is therefore rarely indicated. However, concomitant therapy significantly improves survival and utilization of
k.
adjuvant chemotherapy and radiation have shown a sig- the previously discussed protocols is strongly advocated.
nificant survival advantage. The Macdonald protocol, a Patients presenting with advanced disease may benefit
oo
regimen of 5-FU-based chemoradiotherapy, improves from palliative chemotherapy; however, this has not been
disease-free and overall survival when compared to observa- shown to dramatically alter life expectancy.1,11
tion alone. A significant criticism of this trial is the lack
yb
of appropriate lymphadenectomy at the time of surgical
resection because more than one-half of patients did not
PROPHYLACTIC GASTRECTOMY
even have a D1 resection. Nonetheless, due to the signifi- Overall, 1% to 3% of gastric cancers are hereditary in
cant increase in survival, adjuvant chemoradiation is nature, with the most common type being hereditary
er
in the peritoneal cavity. Systemic chemotherapy has poor due to a germline mutation in CDH1 (E-cadherin).
response rates with peritoneal carcinomatosis. Cytoreduc- Guidelines for screening include:
tive surgery and intraperitoneal chemotherapy have been • a known mutation in a gastric cancer susceptibility
su
studied in select patients with mixed results and currently gene within the family,
are not recommended as the standard of care.30 • gastric cancer in one family member before age 40,
• gastric cancer in two first-degree or second-degree
://
Locoregionally advanced or metastatic gastric cancer does • gastric cancer in three or more first-degree or second-
not benefit from surgical resection. Obstruction and degree relatives regardless of the age of onset,
bleeding tend to be the most common symptoms. Palliative- • gastric cancer and breast cancer in one patient with
ht
intent gastrectomy is rarely performed, but may be one diagnosis before age 50, or
beneficial for uncontrolled bleeding after failure of • gastric cancer in one patient and breast cancer in one
radiation therapy, which is the preferred management first-degree or second-degree relative with one diagnosis
for tumor-related bleeding. Gastric bypass with gastroje- before age 50.
junostomy may be performed for obstruction in an attempt Once the mutation has been diagnosed, prophylactic
to palliate symptoms. However, recent advances in endo- gastrectomy is recommended for asymptomatic carriers
scopic management, including the use of stents, may allow between the ages of 18 and 40. Annual surveillance
for sufficient symptom control without the need for invasive endoscopy with biopsy should be performed for those
procedures. patients who elect not to undergo prophylactic gastrectomy,
Palliative chemotherapy may reduce symptoms and although the efficacy of this practice is not well established
improve survival and quality of life in the setting of due to the diffuse nature of these malignancies.11,31
advanced disease. Multiagent chemotherapy with cisplatin
and fluoropyrimidine are recommended as the first-line
chemotherapy. Trastuzumab may be added in HER2-
SUMMARY
positive cancers. Second-line agents include irinotecan Although the incidence of gastric cancer is declining, it
and docetaxel. The best supportive care to prevent, reduce, remains a highly lethal disease. Eradication and treatment
and relieve suffering and improve the quality of life is of H. pylori has assisted in this decline. A high index of
always indicated.11 suspicion is necessary to diagnose gastric adenocarcinoma
718 SECTION II Stomach and Small Intestine
at an early stage. Diagnosis and staging require endoscopy 15. Cunningham D, Allum WH, Stenning SP, et al. Perioperative che-
with biopsy, EUS, and CT scan to evaluate for locoregionally motherapy versus surgery alone for resectable gastroesophageal
cancer. N Engl J Med. 2006;355(1):11-20.
advanced or metastatic disease. Diagnostic laparoscopy 16. Ychou M, Boige V, Pignon JP, et al. Perioperative chemotherapy
and peritoneal cytology are important components of compared with surgery alone for resectable gastroesophageal adeno-
staging for advanced tumors. Surgery remains the only carcinoma: an FNCLCC and FFCD multicenter phase III trial. J Clin
chance for cure, but it must be accompanied by periopera- Oncol. 2011;29(13):1715-1721.
17. Matuschek C, Bolke E, Peiper M, et al. The role of neoadjuvant
tive chemotherapy or postoperative chemoradiation. and adjuvant treatment for adenocarcinoma of the upper gastro-
Palliation with radiation, chemotherapy, endoscopic intestinal tract. Eur J Med Res. 2011;16(6):265-274.
stenting, or surgery is indicated for appropriate patients 18. Davies J, Johnston D, Sue-Ling H, et al. Total or subtotal gastrectomy
with advanced or metastatic disease. Prophylactic gastrec- for gastric carcinoma? A study of quality of life. World J Surg. 1998;
tomy is indicated for all patients with hereditary diffuse-type 22(10):1048-1055.
19. Lee JH, Han HS, Lee JH. A prospective randomized study comparing
gastric cancer. open vs laparoscopy-assisted distal gastrectomy in early gastric cancer:
early results. Surg Endosc. 2005;19(2):168-173.
REFERENCES 20. Kim W, Kim HH, Han SU, et al. Decreased morbidity of laparoscopic
distal gastrectomy compared with open distal gastrectomy for stage
1. American Cancer Society. Cancer Facts and Figures 2016. Atlanta, GA: I gastric cancer: short-term outcomes from a multi-center randomized
American Cancer Society; 2016. controlled trial (KLASS-01). Ann Surg. 2016;263(1):28-35.
2. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J 21. Schwarz RE, Smith DD. Clinical impact of lymphadenectomy extent
Clin. 2016;66(1):7-30. in resectable gastric cancer of advanced stage. Ann Surg Oncol. 2007;
t
3. Salvon-Harman JC, Cady B, Nikulasson S, Khettry U, Stone MD, 14(2):317-328.
ne
Lavin P. Shifting proportions of gastric adenocarcinomas. Arch Surg. 22. Bonenkamp JJ, Hermans J, Sasako M, et al. Extended lymph-
1994;129(4):381-388, [discussion 388–389]. node dissection for gastric cancer. N Engl J Med. 1999;340(12):
4. Fock KM, Talley N, Moayyedi P, et al. Asia-Pacific consensus guidelines 908-914.
on gastric cancer prevention. J Gastroenterol Hepatol. 2008;23(3):351-365. 23. Wu CW, Hsiung CA, Lo SS, et al. Nodal dissection for patients with
k.
5. Lauren P. The two histological main types of gastric carcinoma: gastric cancer: a randomized controlled trial. Lancet Oncol. 2006;7:
diffuse and so-called intestinal-type carcinoma. An attempt at a 309-315.
histo-clinical classification. Acta Pathol Microbiol Scand. 1965;64:31-49. 24. Songun I, Putter H, Kranenbarg EM, Sasako M, van de Velde CJ.
oo
6. Werner M, Becker KF, Keller G, Höfler H. Gastric adenocarcinoma: Surgical treatment of gastric cancer: 15-year follow up results of the
pathomorphology and molecular pathology. J Cancer Res Clin Oncol. randomized nationwide Dutch D1D2 trial. Lancet Oncol.
2001;127(4):207-216. 2010;11:439-449.
yb
7. Gore RM. Gastric cancer. Clinical and pathologic features. Radiol 25. Tani T, Sonoda H, Tani M. Sentinel lymph node navigation surgery
Clin North Am. 1997;35(2):295-310. for gastric cancer: does it really benefit the patient? World J Gastro-
8. Kodera Y, Yamamura Y, Torii A, et al. The prognostic value of enterol. 2016;22(10):2894-2899.
preoperative serum levels of CEA and CA 19-9 in patients with 26. Paoletti X, Oba K, Burzykowski T, et al. Benefit of adjuvant chemo-
er
gastric cancer. Am J Gastroenterol. 1996;91(1):49-53. therapy for resectable gastric cancer: a meta-analysis. JAMA. 2010;
9. Willis S, Truong S, Gribnitz S, Fass J, Schumpelick V. Endoscopic 303(17):1729-1737.
ultrasonography in the preoperative staging of gastric cancer: 27. Bang YJ, Kim YW, Yang HK, et al. Adjuvant capecitabine and
rg
accuracy and impact on surgical therapy. Surg Endosc. oxaliplatin for gastric cancer after D2 gastrectomy (CLASSIC): a
2000;14(10):951-954. phase 3 open-label randomized controlled trial. Lancet. 2012;379:
10. D’Ugo DM, Pende V, Persiani R, Rausei S, Picciocchi A. Laparoscopic 315-321.
su
staging of gastric cancer: an overview. J Am Coll Surg. 2003;196(6): 28. Macdonald JS, Smalley SR, Benedetti J, et al. Chemoradiotherapy
965-974. after surgery compared with surgery alone for adenocarcinoma of
11. Gastric cancer. Clinical Practice Guidelines in Oncology. National the stomach or gastroesophageal junction. N Engl J Med. 2001;
Comprehensive Cancer Network 2016. http://www.nccn.org/ 345(10):725-730.
://
professionals/physician_gls/pdf/gastric.pdf. 29. Kozak KR, Moody JS. The survival impact of the intergroup 0116
12. Karanicolas PJ, Elkin EB, Jacks LM, et al. Staging laparoscopy in trial on patients with gastric cancer. Int J Radiat Oncol Biol Phys.
the management of gastric cancer: a population based analysis. 2008;72(2):517-521.
tp
J Am Coll Surg. 2011;213(5):544-551. 30. Seshadri RA, Glehen O. Cytoreductive surgery and hyperthermic
13. Washington K. 7th edition of the AJCC cancer staging manual: intraperitoneal chemotherapy in gastric cancer. World J Gastroenterol.
stomach. Ann Surg Oncol. 2010;17(12):3077-3079. 2016;22(3):1114-1130.
ht
14. Knight G, Earle CC, Cosby R, et al. Neoadjuvant or adjuvant therapy 31. Chen Y, Kingham K, Ford JM, et al. A prospective study of total
for resectable gastric cancer: a systematic review and practice gastrectomy for CDH1-positive hereditary diffuse gastric cancer.
guidelines for North America. Gastric Cancer. 2013;16(1):28-40. Ann Surg Oncol. 2011;18(9):2594-2598.
CHAPTER
Postgastrectomy Syndromes
Kristoffel Dumon
| Daniel T. Dempsey
62
U
p to 30% of patients who have had operations However, a small percentage (<5%) of patients after a
on the stomach are afflicted with chronic symp- variety of gastric operations have persistent debilitating
toms that have been relegated to the category symptoms due to the postgastrectomy syndromes discussed
of postgastrectomy syndromes. This convenient classifica- in this chapter. It is important for the managing physician
tion is somewhat of a misnomer because some of these and surgeon to understand the pathophysiology and
patients have not had a gastrectomy (e.g., dumping after treatment options for these conditions. The management
a pyloroplasty or fundoplication), and not infrequently of patients with severe postgastrectomy symptoms can be
the symptom complex for the individual patient does challenging, but appropriate therapy can have a significant
not fit a stereotypical “syndrome.” Most patients with an impact on the patient’s long-term outcome.
t
identifiable postgastrectomy syndrome have one or more
ne
of the following problems: diarrhea, vomiting, abdominal DUMPING SYNDROME
pain, and malnutrition or nutritional deficiency. These
patients have had operations on the stomach for peptic DS is a constellation of GI and vasomotor symptoms, which
k.
ulcer, cancer, obesity, or gastroesophageal reflux disease present postprandially due to rapid gastric emptying. It
(GERD) and represent the subset of gastric surgery patients is caused by loss of pyloric regulation of gastric emptying
oo
with a variety of chronic symptoms that range from annoy- and/or decreased gastric compliance.8 The human stomach
ing to life altering. The evaluation of the most common possesses the remarkable capability of adapting to large
postgastrectomy symptoms and the associated generally volumes of orally administered liquids and solids through
vagally mediated accommodation and receptive relaxation.9
yb
recognized postgastrectomy syndromes are outlined in
Figs. 62.1 to 62.3.1–3 These intragastric contents, usually hypertonic, are then
Because operations on the stomach are frequently acted on by secreted acid and pepsin along with muscular
performed, postgastrectomy syndromes are not uncom- churning to prepare an isosmotic gastric chyme that is
er
mon. The indications for gastric surgery worldwide have slowly discharged into the duodenum for further digestion
changed significantly over the past five decades. Elective and absorption. If there has been a vagotomy, or a portion
rg
gastric surgery for peptic ulcer disease has all but disap- of the stomach has been removed, or the normal pyloric
peared, whereas bariatric surgery in many countries has sphincter has been disrupted or bypassed, the ingested
increased dramatically.4,5 Worldwide, gastric cancer is meal may be incompletely processed by the stomach
su
the third leading cause of cancer death. Gastrectomy and/or prematurely discharged into the proximal small
remains the only potentially curative treatment for most intestine. The flow of liquid out of the stomach is deter-
patients, and cancer is the most frequent indication for mined partly by intragastric pressure and partly by pyloric
://
gastric resection worldwide.6 Therapeutic vagotomy for resistance. Procedures that alter the normal intragastric
peptic ulcer is rarely performed, but partial or complete pressure/volume relationship (proximal gastric vagotomy,
tp
“inadvertent” vagotomy is still quite common during gastric sleeve gastrectomy, fundoplication) or outflow resistance
surgery for obesity and cancer. Laparoscopic fundoplica- (pyloroplasty, gastrojejunostomy [GJ]) predispose to DS.
tion is commonly performed in children and adults for Procedures that alter both have the highest incidence of
ht
GERD and hiatal hernia. It is the most common cause of dumping (gastrectomy, Roux-en-Y gastric bypass [RYGBP]).
dumping syndrome (DS) in children. Dumping symptoms have been reported in up to 70% of
The frequency with which postgastrectomy symptoms Billroth II patients and up to 75% of patients after RYGBP
and syndromes are found depends on how hard they are for obesity.10 Similarly, after gastrectomy for cancer, 67%
sought. For instance, some studies indicate that after partial of patients present with early dumping symptoms and
gastrectomy, the majority of patients suffer from one or 38% with late dumping.11 Depending on the speed of
more upper abdominal symptoms, but clinical experience liquid emptying and the osmolarity of the contents being
teaches that only a small percentage of these patients are discharged, a variety of symptoms may result that have
truly debilitated; most do quite well. The incidence of been referred to as the dumping syndrome. Both an early
clinically significant chronic postgastrectomy complica- and late form of this disorder have been identified. The
tions in patients who undergo subtotal gastrectomy and role of surgically induced microbiome changes in the
Billroth II reconstruction for gastric adenocarcinoma is low etiology of DS is unknown.
(<5%). The incidence is substantially higher in the first Early dumping is more common and includes systemic
postoperative year, but most patients report improvement and abdominal symptoms. Systemic manifestations include
within 1 year after surgery.7 Thus long-term survivors after palpitations, tachycardia, fatigue, a need to lie down follow-
gastrectomy for gastric cancer usually have normal body ing meals, flushing or pallor, diaphoresis, lightheadedness,
weight and lean body mass and satisfactory gastrointestinal hypotension, headache, and possibly syncope. Abdominal
(GI) quality of life. symptoms include early satiety, epigastric fullness or pain,
719
Postgastrectomy Syndromes CHAPTER 62 719.e1
ABSTRACT
Patients who have had operations on the stomach are
afflicted with chronic symptoms that have been relegated
to the category of “postgastrectomy syndromes.” Because
operations on the stomach are frequently performed,
postgastrectomy syndromes are not uncommon. These
patients have had operations on the stomach for peptic
ulcer, cancer, obesity, or gastroesophageal reflux disease
and represent a subset of gastric surgery patients with
a variety of chronic symptoms that range from annoy-
ing to life altering. Most patients with an identifiable
postgastrectomy syndrome have one or more of the fol-
lowing problems: diarrhea, vomiting, abdominal pain, and
malnutrition or nutritional deficiency. The evaluation of
the most common postgastrectomy symptoms and the
associated generally recognized postgastrectomy syndromes
are described in this chapter.
t
KEYWORDS
ne
postgastrectomy syndrome
dumping syndrome
postvagotomy diarrhea
k.
gastric stasis
bile reflux gastritis
oo
alkaline reflux gastritis
afferent loop obstruction
efferent loop obstruction
yb
Roux syndrome
er
rg
su
://
tp
ht
720 SECTION II Stomach and Small Intestine
No Yes
– Oral glucose
Dumping
challenge and/or
unlikely
gastric scintigraphy
+
Dumping treatment
t
Associated
ne
vomiting
Afferent/efferent CT scan, upper GI
loop obstruction barium study, EGD
k.
Gastric stasis EGD
oo
Delayed gastric emptying Gastric scintigraphy
iminodiacetic acid.
tp
diarrhea, nausea, cramps, bloating, and borborygmi. Early distention, hypermotility, and splanchnic blood pooling.
dumping begins within 30 minutes following a meal and This leads to both the GI and vasomotor symptoms that
is attributable to bowel distention, relative hypovolemia, characterize early DS.16,17 Hormones play an important
ht
GI hormone hypersecretion, and autonomic dysregula- role as mediators of this pathophysiologic response (e.g.,
tion.12,13 Late dumping is characterized by symptoms vasoactive intestinal peptide [VIP], serotonin, bradykinin,
that occur 1 to 3 hours postprandially. Symptoms of late norepinephrine).18,19 Late dumping involves a reactive
dumping consist of perspiration, faintness, decreased hypoglycemia brought on by the rapid and high initial
concentration, and altered levels of consciousness, glucose load presented to and absorbed by the small
among others. These symptoms are related to a reactive intestine, leading to a GLP-1–mediated inappropriately
hypoglycemia that occurs 1 to 3 hours postprandially. high insulin response and hypoglycemia.
Patients with late dumping often have early dumping An oral glucose challenge will confirm the diagnosis of
as well.14 DS. Patients fast for 10 hours overnight and then ingest 50 g
Most patients with DS have mild to moderate symptoms, of glucose. It is reported to have a sensitivity and specificity
but some patients have disabling symptoms that may be of up to 100% and 94%, respectively.20 The diagnosis of
severe enough to cause protein-energy malnutrition.15 early DS in a patient with clinical symptomatology may
The differential diagnosis of DS includes gastroparesis, also be confirmed with a scintigraphic gastric emptying
partial small bowel obstruction, anastomotic stricture, study, in which greater than 50% of an isotope-labeled
postvagotomy diarrhea, inflammatory bowel disease, solid meal has emptied within 1 hour.
irritable bowel disease, and bacterial overgrowth. Dumping A variety of gastric procedures may give rise to early
symptoms are triggered by rapid gastric emptying of dumping. More than half of the patients who have had
hyperosmolar voluminous chyme that causes bowel RYGBP for obesity or total gastrectomy experience early
Postgastrectomy Syndromes CHAPTER 62 721
Chronic diarrhea
No Yes
– Oral glucose
Dumping
challenge and/or
unlikely
gastric scintigraphy
+
Dumping treatment
t
Bacterial
ne
overgrowth Breath test (14C–D-xylose Repeated course of
or hydrogen) antimicrobials, correction
Culture of jejunal aspirate of associated nutritional
for bacterial counts deficiencies, probiotics?
k.
oo
Postvagotomy Treat medically
Postvagotomy diarrhea
Surgery if medical Rx failure
yb
Other FIGURE 62.2 Chronic diarrhea. The
er
Cholestyramin
Bile acid malabsorption Antidiarrheal evaluation of diarrhea and the associated
Abnormal peristalsis Octreotide generally recognized postgastrectomy
rg
syndromes.
dumping symptoms, and approximately 30% of those catecholamines, which cause a constellation of symptoms,
su
with partial gastrectomy have early dumping. The risk including tachycardia, tachypnea, diaphoresis, and light-
of problematic dumping after sleeve gastrectomy is low headedness. Late DS is more frequent in patients who
(1.6%), although up to 30% to 40% of patients may had early DS.21 Late DS has been reported in more than
://
have very mild symptoms.21,22 Up to 15% of patients with 50% of bariatric patients after gastric bypass. In some of
pyloroplasty or simple GJ experience early dumping, as these patients, late DS developed 1 to 8 years after the
tp
do 2% of patients with proximal gastric vagotomy and surgery and was significantly more common in patients
fundoplication. The type of GI reconstruction after distal with type 2 diabetes mellitus (44.9% vs. 5.6%). One must
gastrectomy influences the risk of DS because Roux-en-Y rule out an unrelated islet cell tumor as the cause of a
ht
GJ (11%) has a lower rate of dumping than either Billroth severe refractory hypoglycemia by documenting the fasting
I (B1) or Billroth II (B2). DS after B1 (17%) is less than plasma glucose, serum insulin, and C-peptide level. A
after B2, (70%), perhaps because B1 gastric effluent enters prolonged oral glucose tolerance test will also confirm
the duodenum and triggers a neuroendocrine response the diagnosis of late dumping.25 Early dumping tends to
that slows gastric emptying (duodenal brake). The risk of improve with time, whereas late dumping tends to persist
dumping with pyloroplasty and loop GJ are similar, but or exacerbate.
the latter gastric drainage procedure is easily reversible. In most patients with DS, symptoms are not severe and
Dumping symptoms tend to improve with time in most medical management is successful. Consultation with an
patients.7,24 experienced dietitian is helpful. Dietary modification is
Late dumping, less common than the early variant, the first-line treatment for DS. Daily intake should be
is caused by hyperinsulinemic hypoglycemia and occurs divided into at least 6 meals. Liquids and solids should be
2 to 3 hours postprandially. In late dumping the rapid separated. Diets should be high in protein and fat, and
delivery of monosaccharides and disaccharides into the simple sugars should be avoided. Vasomotor symptoms
small intestine causes hyperglycemia. The pancreas is can often be ameliorated if the patient lies down for
subsequently triggered to release insulin by glucagon-like 30 minutes after meals. Elimination of milk and dairy
peptide 1 (GLP-1) and in the process actually “overshoots” products has been successful in many patients.26 Another
so that marked hypoglycemia is induced. This insulin simple therapy is adding dietary fiber. Guar gum and
shock condition stimulates the adrenal glands to release pectin are useful in increasing the viscosity of the food,
722 SECTION II Stomach and Small Intestine
Vomiting
Radiologic
obstruction?
+
–
Afferent/efferent
Consider elective or
loop obstruction,
urgent operation
SBO
Dumping
symptoms?
Yes Oral glucose
t
challenge and/or Dumping treatment
ne
gastric scintigraphy
No
k.
Delayed gastric emptying Gastric scintigraphy
oo
EGD, gastric pH probe,
Bile reflux gastritis
(bilitec probe), HIDA scan?
FIGURE 62.3 Vomiting. The evaluation
of vomiting and the associated
yb
generally recognized postgastrectomy Upper GI barium study
Roux syndrome
syndromes. EGD, Gastric scintigrapy
Esophagogastroduodenoscopy; GI,
er
but their poor taste limits patient compliance.27 Hard refractory to diet therapy. Octreotide can markedly improve
su
candy can be consumed to abort the hypoglycemia of the quality of life in DS patients,37 but the data are limited
late DS. Acarbose is an α-glycosidase hydrolase inhibitor for long-term efficacy. Octreotide alleviates both early
that delays carbohydrate digestion and absorption28–30 and late dumping symptoms through inhibition of
://
and is efficient in the treatment of late dumping. Side hormone mediators. It also delays gastric emptying time
effects include excess flatulence31 and hypoglycemia if and inhibits splanchnic vasodilation.38,39 Short-acting and
tp
carbohydrate absorption is excessively inhibited.25,26,28,29 long-acting octreotide are equally effective in blunting
A number of pharmacologic options exist for the treat- dumping symptoms, but the long-acting preparation
ment of the DS. Tincture of opium is especially effective scored significantly better on quality-of-life measures.13
ht
in relieving diarrhea associated with DS.33 Symptomatic Long-term octreotide therapy loses its efficacy because
treatment of DS may be addressed with over the counter side effects such as diarrhea and steatorrhea as well as
medications. Treatment for diarrhea (e.g., Imodium), cost lead to lack of compliance.26,38,39
nausea (e.g., meclizine, promethazine, proton pump Only a small percentage of patients with dumping
inhibitors [PPIs]), or antigas measures may be helpful. symptoms ultimately require surgery. Most patients improve
Inadequate digestion of nutrients can cause gas and with time (months and even years), dietary management,
bloating when they reach colonic bacteria; therefore and medication. Therefore the surgeon should not rush
probiotics may be a useful adjunct. Anticholinergic agents, to reoperate on the patient with DS. Multidisciplinary
such as dicyclomine, hyoscyamine, and propantheline, nonsurgical management must be optimized first. Before
slow gastric emptying and are also antispasmodic, thus reoperation, a period of in-hospital observation is useful
decreasing abdominal pain related to small bowel motility.34 to define the severity of the patient’s symptoms, and
Diazoxide is a potassium channel activator that inhibits the patient compliance with prescribed dietary and medical
secretion of insulin. Thus diazoxide has showed success therapy. The results of remedial operation for dumping
in recent studies in treating late dumping hypoglycemia are variable and unpredictable. There are a variety of
and can be used when acarbose and lifestyle modifications surgical approaches, none of which work consistently
are insufficient.35,36 well. In addition, there is not a great deal of experience
Octreotide, a somatostatin analogue, should be con- reported in the literature with any of these methods.
sidered for patients with severe postgastrectomy DS Long-term follow-up is rare.
Postgastrectomy Syndromes CHAPTER 62 723
POSTVAGOTOMY DIARRHEA
10 cm
Truncal vagotomy is associated with clinically significant
diarrhea in 5% to 10% of patients. It occurs soon after
surgery and usually is not associated with other GI or
systemic symptoms, a fact that helps to distinguish it
60 cm
t
2% of vagotomy patients. The cause of postvagotomy
ne
diarrhea is unclear. Although rare, it can even occur after
proximal gastric vagotomy or fundoplication, suggesting
A that intestinal vagal denervation may not be the sole
k.
cause. Factors contributing to postvagotomy diarrhea
FIGURE 62.4 Surgical approaches to treat dumping syndrome. include intestinal dysmotility and accelerated transit,
oo
(A) Long-limb Roux-en-Y anastomosis in which the bile acid malabsorption, rapid gastric emptying, altered
jejunojejunostomy is fashioned approximately 60 cm from the microbiome, and bacterial overgrowth. The latter problem
gastrojejunostomy. (B) A 10-cm loop of jejunum is twisted 180 is facilitated by decreased gastric acid secretion and (even
degrees on its mesentery so that its distal end is anastomosed
yb
small) blind loops. Although bacterial overgrowth can be
to the stomach and its proximal end to the small intestine in an confirmed with the hydrogen breath test, a simpler test
antiperistaltic fashion. (Modified from Miller TA, Mercer DW.
is an empirical trial of oral antibiotics and/or probiotics.
Derangements in gastric function secondary to previous surgery.
Some patients with postvagotomy diarrhea respond to
er
be considered for simple takedown of this anastomosis the amount of chenodeoxycholic acid.45 Such findings lend
provided that the pyloric channel is patent endoscopically. support to the hypothesis that bile acid malabsorption may
For dumping following pyloroplasty, pyloric reconstruction contribute to postvagotomy diarrhea in some patients.
://
is described, but modern day experience with this is rare Fat malabsorption should also be considered in the
and today’s surgeon should view pyloroplasty as irreversible. differential diagnosis of postvagotomy diarrhea. This can
tp
Distal gastrectomy with Roux reconstruction (Fig. 62.4A), be caused by acid inactivation of pancreatic enzymes,
or a “duodenal switch” with division of the postbulbar poorly coordinated mixing of food and digestive juices,
duodenum and anastomosis to a Roux jejunal limb, is or bacterial overgrowth. This can be confirmed with a
ht
currently the best option for the uncommon patient with qualitative test for fecal fat. It is best treated with acid
refractory postpyloroplasty dumping. For severe dumping suppression and pancreatic enzyme supplements, and if
after Billroth I or II gastrectomy, conversion to Roux-en-Y appropriate oral antibiotics. Postvagotomy diarrhea usually
GJ should be considered because the motility of the Roux does not respond to these modalities.
limb tends to slow gastric emptying. However, gastric In the rare patient who is debilitated by postvagotomy
stasis and/or marginal ulceration may result particularly diarrhea unresponsive to maximal medical management
in the presence of a large gastric remnant. Lifelong acid for at least 1 year, operation might be considered but
suppression should be considered. The reversed intestinal outcomes can be problematic. The operation of choice
interposition is rarely used currently for DS—and rightly is probably a 10-cm reversed jejunal interposition placed
so. This operation interposes a 10-cm reversed segment in continuity 100 cm distal to the ligament of Treitz (Fig.
of intestine between the stomach and the proximal small 62.5). Another option is the onlay antiperistaltic distal ileal
bowel (see Fig. 62.4B). This slows gastric emptying but graft. Both operations can cause obstructive symptoms
often leads to obstruction, requiring reoperation. Isope- and/or bacterial overgrowth.10
ristaltic interposition (Henley loop) between the gastric
remnant and the duodenum has not been successful in GASTRIC STASIS
sustained improvement of DS. Because pyloric ablation
is a dominant factor in the etiology of postgastrectomy In the rare patient with acute gastric stasis after gastric
dumping, it is not surprising that conversion of Billroth surgery, persistent nausea and vomiting prevent removal
724 SECTION II Stomach and Small Intestine
t
stricture or narrowing. A dilated efferent limb suggests
ne
chronic stasis, either from a motor abnormality (e.g.,
Roux syndrome) or mechanical small bowel obstruction
(e.g., chronic adhesion). If the problem is thought to be
k.
primarily a disorder of intrinsic motor function, newer
techniques, such as electrogastrography and GI manometry,
10 cm
oo
should be considered. However, it should be recognized
that chronic distal mechanical obstruction may result in
disordered motility in the proximal organ.
yb
After mechanical obstruction has been ruled out,
medical treatment is successful in most cases of gastric
FIGURE 62.5 Surgical management of postvagotomy diarrhea: A
motor dysfunction following gastric surgery. Management
consists of dietary modification and promotility agents. One
er
Surgical Care: Physiologic Foundations and Clinical Applications. efficacious in a given patient. Metoclopramide is a dopa-
2nd ed. St. Louis: Quality Medical; 1998:407.) mine antagonist that works on the stomach by facilitating
acetylcholine release from enteric cholinergic neurons.47
su
a gastrostomy may be placed either laparoscopically or acetylcholine release from the mesenteric plexus of the
endoscopically. Alimentation can then be given via a J tube gut.48 Erythromycin is a motilin agonist that works on
tp
extension placed during one of these procedures. If the both the stomach and intestine by binding to motilin
gastric remnant is not of sufficient size to accommodate receptors on GI smooth muscle.49 One of these agents is
these approaches, a decompressing gastric tube can often usually sufficient to enhance gastric tone so that improved
ht
be passed retrograde through the efferent limb and exited gastric emptying results. Intermittent oral antibiotic therapy
through the skin via a Witzel technique. Distal to this may be helpful in treating bacterial overgrowth, with its
placement, another tube may be placed antegrade as a attendant symptoms of bloating, flatulence, and diarrhea.
Witzel feeding jejunostomy. In patients in whom these Probiotics should be tried because alterations in gut
enteral approaches to alimentation are not possible, microbiome are likely.
total parenteral nutrition is an alternative. In any event, Surgery should be considered when chronic postop-
reoperative surgery should generally be delayed for at erative gastric stasis is severe and resistant to medical
least 3 months as the majority of patients will regain management. At operation small bowel obstruction and
satisfactory GI function without surgery. Only after this efferent limb obstruction should always be ruled out.
period should reexploration be considered. Gastroparesis following vagotomy and drainage proce-
Chronic gastric stasis46 following gastric surgery may be dures may be treated with subtotal (75%) gastrectomy.
due to a problem with gastric motor function or be caused Billroth II anastomosis with Braun enteroenterostomy (Fig.
by an obstruction. The gastric motility abnormality may 62.6A) may be preferable to Roux-en-Y reconstruction
have been preexisting and unrecognized by the operating after subtotal gastrectomy in this setting because Roux
surgeon. More commonly it is secondary to some aspect of reconstruction may result in persistent gastric empty-
the operation, such as deliberate or unintentional vagotomy ing problems (Roux syndrome) ultimately necessitating
or resection of the dominant gastric pacemaker. Truncal near-total or total gastrectomy, a nutritionally unattractive
vagotomy is more likely to cause chronic gastric stasis option. Delayed gastric emptying following vagotomy and
Postgastrectomy Syndromes CHAPTER 62 725
t
ne
A B
FIGURE 62.6 (A) The Braun procedure is one of the oldest attempts at bile diversion. The figure on the left shows the original procedure
k.
with Billroth II gastrojejunostomy and “downstream” enteroenterostomy to divert bile distally. (B) A more recent modification (on the right)
adds a staple line distal to the enteroenterostomy in an effort to more completely divert the duodenal contents distally. It has been
oo
designated the “uncut” Roux-en-Y. (Modified from Madura JA. Postgastrectomy problems: remedial operations and therapy. In: Cameron
JL, ed. Current Surgical Therapy. 7th ed. St. Louis: Mosby; 2001.)
yb
drainage or vagotomy and antrectomy may represent an loop by postoperative adhesions; (2) internal herniation,
anastomotic stricture due to recurrent (marginal) ulcer, volvulus, and intussusception of the afferent loop; (3) scar-
or proximal small bowel obstruction. Recurrent ulcer may ring due to marginal ulceration of the GJ; (4) locoregional
er
respond to medical therapy with PPI and abstinence from recurrence of cancer (lymph nodes, peritoneum, gastric
nonsteroidal antiinflammatory drugs (NSAIDs), aspirin, remnant, anastomotic sites); (5) radiation enteritis of the
rg
and smoking. Endoscopic dilation is occasionally helpful. afferent loop; and (6) enteroliths, bezoars, and foreign
Gastroparesis following subtotal gastric resection is best bodies impacted in the afferent loop (Fig. 62.7). In patients
treated with near-total (95%) or total gastric resection with Billroth II anastomoses, afferent loop syndrome is
su
and Roux-en-Y reconstruction. High-frequency gastric seen more commonly in patients with redundant (longer
electrical stimulation (GES) may be an effective treatment than 30 to 40 cm) and antecolic afferent loops, which
for patients with postsurgical gastroparesis who failed are more prone to kinking, volvulus, and entrapment
://
standard medical therapy,50 but long-term follow-up and by adhesions. Improperly closed mesocolic defects may
randomized controlled trials (RCTs) are lacking. predispose to internal herniation of the retrocolic affer-
tp
Afferent loop obstruction, also called afferent loop syn- antegastric positioning after RYGBP. The role of closing
drome, is a mechanical complication that infrequently the mesenteric defect remains unclear.54
occurs following construction of a GJ. The creation of a Although both acute and chronic forms of afferent
GJ leaves a segment of proximal small bowel (duodenum loop syndrome have been described, chronic partial
and proximal jejunum) upstream from the anastomosis. obstruction is the more common clinical manifestation.55
With Billroth II or loop GJ the afferent limb conducts The classic presentation of chronic afferent loop syndrome
bile, pancreatic juices, and other proximal intestinal is postprandial abdominal pain relieved by bilious vomiting,
secretions toward the GJ51; with Roux-en-Y the afferent but the latter may be lacking with Roux-en-Y GJ.
limb conducts the succus toward the jejunojejunostomy A meal elicits pancreatic, biliary, and duodenal secre-
and is also called the biliopancreatic limb. The operations tion into the obstructed afferent limb. As the volume of
most commonly associated with afferent loop obstruction these secretions increases, the obstructed duodenum and
are Billroth II and Roux-en-Y GJ (distal gastrectomy or proximal jejunum become more distended. Eventually the
gastric bypass), and Roux-en-Y esophagojejunostomy (total pressure in the partially obstructed afferent limb overcomes
gastrectomy).52 The incidence of significant afferent loop the obstruction (usually 30 to 60 minutes postprandially),
obstruction after these procedures is low (0.3% to 1.0%) delivering a large volume of bilious secretions into the
and is similar after open and laparoscopic surgery. stomach or Roux limb. This leads to bilious vomiting and
The etiologies of afferent loop obstruction include: prompt relief of the pain, which was caused by the afferent
(1) entrapment, compression, and kinking of the afferent limb distention. Weight loss and anemia are common.
726 SECTION II Stomach and Small Intestine
A B
t
Kinking and Internal herniation
ne
angulation behind efferent limb
k.
oo
yb
er
rg
su
C D E
://
FIGURE 62.7 Causes of afferent loop syndrome include (A) kinking and angulation of the afferent limb, (B) internal herniation of the afferent
limb behind the efferent limb, (C) stenosis of the gastrojejunal anastomosis, (D) redundancy of the afferent limb leading to volvulus, or
ht
(E) adhesions involving the afferent limb. (Modified from Miller TA, Mercer DW. Derangements in gastric function secondary to previous
surgery. In: Miller TA, ed. Modern Surgical Care: Physiologic Foundations and Clinical Applications. 2nd ed. St. Louis: Quality Medical;
1998:402.)
Bacterial overgrowth secondary to afferent limb stasis or more of the following findings: weight loss, upper
may contribute to these problems due to malabsorption abdominal distention, upper abdominal mass, and
of fat and other nutrients, such as vitamin B12 or iron. abdominal tenderness. Peritoneal findings or pain out
If the obstruction is high grade or complete, the dis- of proportion to physical findings are ominous. Rarely
tended afferent loop may not sufficiently decompress. In jaundice, cholangitis, or pancreatitis can confuse the
this scenario, vomiting, if present, will be nonbilious, and clinical picture.
a clinical picture of “closed loop obstruction” manifested Abdominal multiple detector computed tomography
as an acute abdomen will result. If this condition is not (CT) is the diagnostic study of choice. CT appearance
recognized early, the afferent loop may actually perforate of the obstructed afferent loop consists of a C-shaped,
and result in peritonitis. Urgent surgery is necessary to fluid-filled tubular mass located in the midline between
correct this problem. the abdominal aorta and the superior mesenteric artery
Depending on the acuity and severity of the afferent (c-loop sign) with valvulae conniventes projecting into
loop obstruction, physical examination can reveal one the lumen (keyboard sign).56 Adhesions are suspected
Postgastrectomy Syndromes CHAPTER 62 727
t
of treatment for afferent loop obstruction in patients with
ne
curable cancer or benign disease is surgery. At operation
the primary cause of afferent loop obstruction should be
confirmed and treated. This may include resection for
k.
tumor or marginal ulcer, lysis of adhesions, or repair of
internal hernia. Procedures to consider include addition
oo
of a Braun anastomosis in a former Billroth II reconstruc-
tion, excision of the redundant loop and conversion of
Billroth II to Roux-en-Y GJ or Billroth I, and excision of
yb
the redundant loop and reconstruction of the former
Roux-en-Y jejunojejunostomy (Fig. 62.8). Endoscopic
interventions and percutaneous approaches (percutaneous
endoscopic gastrostomy [PEG], balloon dilation, double-
er
with stage IV cancer. These techniques may also be useful Care: Physiologic Foundations and Clinical Applications. 2nd ed.
as temporizing measures in high-risk patients. St. Louis: Quality Medical; 1998:404.)
In contrast to the relatively stereotypical manifestation
su
standing presence of an abnormal amount of duodenal caused by bile salts but acidic pH also inactivates pancreatic
content in the stomach or gastric remnant, a situation enzymes in duodenal refluxate. In a subset of patients,
that often occurs in patients after pyloroplasty or loop GJ bile gastritis leads to dysplasia and some of these patients
with or without gastric resection. A distinction must be progress to gastric cancer (“stump cancer”).
made between histologic bile gastritis, which is present in Clinically significant bile reflux gastritis is not common.
many patients after gastric surgery (up to 85% in Billroth Although many patients have histologic gastritis, the
II patients), most of whom are asymptomatic, and the relationship of chronic gastric mucosal inflammation
presence of clinical bile gastritis leading to significant to symptoms in this setting is problematic. The most
symptoms, a much more unusual situation. Although common symptoms attributed to chronic bile gastritis
both histologic and clinical bile gastritis can occur without are abdominal pain and bilious vomiting. Although the
previous gastric operation (primary bile reflux gastritis), surgeon can almost always eliminate bile from the vomitus,
it is much more common after gastric surgery. Histologic pain and vomiting may persist after remedial operation,
bile gastritis is more common after Billroth II (40% to particularly in patients on chronic narcotics preoperatively.
85%) than Billroth I anastomosis (29% to 48%) or gastric The pain is midepigastric burning abdominal pain, often
drainage operations (pyloroplasty or loop GJ, 15%). Rarely associated with nausea, and characteristically unrelieved by
cholecystectomy is associated with the clinical syndrome,58,59 antacids or acid suppressive medication. Unlike afferent
possibly due to the loss of bile reservoir function, resulting limb syndrome, the pain does not resolve after vomiting.
in a continuous flow of bile into the duodenum with Weight loss and anemia are common.
728 SECTION II Stomach and Small Intestine
60 cm
postoperative anatomy, stomach patency, size of the residual
t
gastric pouch, and the status of the afferent and efferent
ne
limbs. Hepatobiliary iminodiacetic acid (HIDA) scans can
provide a semiquantitative assessment of bile reflux/stasis
in the stomach. Ultrasound and CT scanning may be
k.
useful to exclude pancreatic or biliary causes of symptoms.
Medical management has limited success in relieving
oo
symptoms. Suggested treatment modalities include the
administration of cholestyramine, antacids, H2 blockers,
PPIs, sucralfate, or promotility agents to enhance clearance
yb
of refluxate from the gastric remnant. When these measures FIGURE 62.9 Roux-en-Y gastrojejunostomy for the treatment of
fail, surgery is considered for patients with incapacitating alkaline reflux gastritis. Note the generous distal gastrectomy.
symptoms, a reasonably secure clinical diagnosis, and Adequate Roux length minimizes bile reflux. (From Fromm D.
realistic expectations. Preoperative nutritional support
er
frequently chosen to treat patients with alkaline reflux denum following revision of prior Billroth I or II). The
gastritis (Fig. 62.9).41 Conversion of Billroth I or II to Roux- jejunal segment should be approximately 40 cm in length
en-Y GJ with a 60-cm Roux limb reliably diverts intestinal to minimize enterogastric reflux. Reported advantages
://
contents from the gastric remnant and improves symptoms of the Henley procedure include the duodenal passage
in up to 85% of patients.61,62,65–67 This procedure also of chyme, with the possibility of improving mixing and
tp
PPI might be equally efficacious and preferred given the Although this method achieved satisfactory results in up
gastric emptying issues seen with vagotomy. Although to 70% of patients,69,72 most authors prefer the Roux-en-Y
Roux-en-Y GJ achieves satisfactory symptom relief following GJ because of its technical simplicity.67
surgery, during long-term follow-up epigastric pain can Other procedures described for the management
recur in up to 30% of patients.62 The only symptom that of alkaline reflux gastritis are the biliary diversion and
is consistently relieved is bilious vomiting. the suprapapillary duodenojejunostomy. During biliary
Other less commonly used procedures are the Henley diversion, previous gastric procedures are converted to a
operation (interposition of an isoperistaltic jejunal loop gastroduodenal anastomosis (a Billroth I configuration).
between residual stomach and intestine),69 the Tanner 19 All patients then undergo a choledochojejunostomy with
procedure, the biliary diversion,60 and the suprapapillary a 35- to 40-cm newly created or preexisting Roux-en-Y
duodenojejunostomy (duodenal switch).70 The Roux-en-Y limb. Biliary diversion achieves complete elimination
Tanner 19, which is a modification of the original Roux- of bile salts from the gastric/duodenal lumen by the
en-Y GJ, has some theoretical advantages, but in clinical choledochojejunostomy, and therefore symptoms associated
practice there is no evidence that the Tanner procedure with enterogastric reflux are alleviated.60
is better than the classical Roux-en-Y reconstruction. The The suprapapillary duodenojejunostomy—the so-called
Henley procedure is an infrequently used technique for duodenal switch—proposed by DeMeester et al.73 achieved
the management of alkaline reflux gastritis (Fig. 62.10). good results without altering gastric emptying70,74 in patients
This technique was first described by Henley in 195271 as with primary bile reflux gastritis, a rare entity (Fig. 62.11).
Postgastrectomy Syndromes CHAPTER 62 729
Pylorus
PANCREAS
PANCREAS
40 cm COLON
t
ne
COLON
k.
oo
yb
FIGURE 62.10 Interposition of a 40-cm isoperistaltic jejunal FIGURE 62.11 Duodenal switch procedure as described by
er
segment between the stomach and duodenum to treat alkaline Demeester et al. as a treatment for primary bile reflux gastritis.
reflux gastritis (Henley loop). (From Aronow JS, Matthews JB, The postbulbar duodenum is divided and a 60-cm retrocolic Roux
Garcia-Aquilar J, Novak G, Silen W. Isoperistaltic jejunal limb is anastomosed to the proximal duodenum. (Modified from
rg
interposition for intractable postgastrectomy alkaline reflux Strignano P, Collard JM, Michel JM, et al. Duodenal switch
gastritis. J Am Coll Surg. 1995;180:648.) operation for pathologic transpyloric duodenogastric reflux. Ann
su
Surg. 2007;245:247–253.)
normal gastric reservoir and normal foregut physiology. These clinical observations suggest that the gastric pouch
It may be prudent to add proximal gastric vagotomy or plays a significant role in the etiology of the Roux syndrome,
tp
chronic PPI treatment to minimize the risk of marginal and these factors undoubtedly account for the reported
ulcer with duodenal switch for primary bile gastritis. variability in the incidence of this problem. 75 When
In summary, remedial gastric surgery can be effective performing a distal gastrectomy, the surgeon should be
ht
in ameliorating symptoms of bile reflux gastritis in the cognizant of the factors that predispose to Roux syndrome.
majority of patients, but careful patient selection is essential Symptoms of Roux syndrome include abdominal
to achieve satisfactory results. pain and distention, postprandial bloating, nausea, and
vomiting.76 Typically the vomitus contains solid food
and is nonbilious. Bacterial overgrowth, with diarrhea
ROUX STASIS SYNDROME and nutrient malabsorption, may result. Endoscopically,
After distal gastrectomy with Roux-en-Y reconstruction, the gastric remnant may be dilated with retained food
some patients experience symptomatic delayed gastric and mucosal irritation. The anastomosis is patent and
emptying of solids. This phenomenon has been termed the Roux limb may also be dilated. There is no evidence
the Roux syndrome or Roux stasis syndrome because it has of mechanical obstruction on CT or upper GI series.
generally been attributed to measurable abnormalities Scintigraphy shows markedly delayed emptying of solids.
in Roux limb motility. Peristalsis is abnormal in a Roux Liquid emptying is usually not delayed.
limb, with a significant number of propulsive contractile Most patients with the Roux syndrome can be success-
waves proceeding in the proximal direction (i.e., toward fully managed conservatively with dietary manipulations
the stomach). Presumably this slows gastric emptying, and use of prokinetic agents,77 but some patients require
particularly solid emptying. Interestingly, Roux syndrome is revisional operation in an attempt to relieve debilitating
more common in the presence of a large gastric remnant symptoms and improve nutritional status. Pacing of the
or after vagotomy and quite uncommon after RYGBP. intestine and/or stomach has been investigated as potential
730 SECTION II Stomach and Small Intestine
t
Postgastrectomy syndromes. Surg Clin N Am.
ne
2011;91:1105–1122.) A B C
k.
nonsurgical treatment, but this has not yet been proven traditional syndromes discussed previously. Not infrequently
effective as long-term treatment.78 it contributes to the complexity of the surgical management
oo
The choice of operation for Roux syndrome depends of these postgastrectomy problems. For example, it is
somewhat on the anatomy of the gastric pouch, the status not unusual to find acute or chronic ulceration when
of the Roux limb, and the condition of the patient. The operating for afferent loop syndrome, bile gastritis, and
yb
addition of a feeding jejunostomy is usually prudent. Roux syndrome. The incidence of marginal ulcer ranges
In general, if the original gastrectomy removed less from 0.6% to 25%. It is more common after Roux-en-Y
than half the stomach, consideration can be given to anastomosis (Fig. 62.12C) than after Billroth II (see Fig.
subtotal gastrectomy with anastomosis to a new Roux 62.12B) because the former arrangement lacks the buffer-
er
limb (usually the original Roux should be resected), or ing afferent limb contents that counteract the noxious
with B2 reconstruction and Braun enteroenterostomy. effect of gastric acid on the jejunal mucosa (usually the
rg
However, bile reflux gastritis and esophagitis remain a ulceration is on the jejunal side of the anastomosis).
risk with the latter procedure. Although the addition of Chronic ischemia and permanent suture material may
a “TA” staple line in the afferent limb distal to the Braun also be contributing factors. NSAIDs (including aspirin)
su
anastomosis and proximal to the B2 GJ preserves normal and smoking predispose to marginal ulcer. Incomplete
small bowel motility in the efferent limb and eliminates vagotomy, Helicobacter infection, and hypergastrinemia must
bile from the stomach (the “uncut Roux” procedure [see also be considered.81,82 Presentation may be acute (usually
://
Fig. 62.6B]), the latter effect is transient because the staple perforation) or chronic. Symptoms include abdominal
line inevitably opens up.79–81 pain, vomiting, and various signs and symptoms of chronic
tp
If the original gastrectomy removed more than half the or acute blood loss. In most cases, marginal ulcers can be
stomach, consideration should be given to total gastrectomy adequately treated with PPIs, the elimination of NSAIDs,
with Roux reconstruction. In this setting, if the left gastric Helicobacter treatment, and smoking cessation.
ht
artery is intact, equally good results might be achieved If medical treatment fails and if incomplete vagotomy
with 95% gastrectomy, leaving a small, vertically oriented is diagnosed by history or the sham feeding pancreatic
lesser curvature gastric pouch analogous to that used for polypeptide test, thoracoscopic truncal vagotomy may be
the Roux-en-Y gastric bypass bariatric procedure. A small, an option.83 Hypergastrinemia after distal gastrectomy
horizontally oriented residual gastric pouch should be can be caused by gastrinoma or retained antrum. In the
avoided because of the potential for dilation and stasis. latter there is residual antral tissue left in continuity with
Since the frequency of Roux stasis symptoms may be the duodenal stump after gastric resection with Billroth
higher when Roux-en-Y limbs are longer, the Roux should II anastomosis. The G cells in this retained antral tissue
be limited to 50 cm long in revisional operation for Roux are not exposed to luminal acid, resulting in continuous
syndrome.62,68,79,80 secretion of gastrin and intense stimulation of acid produc-
tion by parietal cells in the proximal gastric remnant. The
exposure of the unbuffered jejunum to this high acid
MARGINAL ULCERS level at the Billroth II GJ results in marginal ulcer (see
Marginal ulceration (i.e., juxtaanastomotic ulceration) is a Fig. 62.12B). Clinical suspicion of retained antrum may be
well-described complication of GJ. Although not typically confirmed by review of previous operative and pathology
considered as one of the postgastrectomy syndromes, reports, barium upper GI study, and/or technetium 99m
it is not uncommon and thus must be considered as scan. Reexcision is curative. Gastrinoma is suspected when
part of the differential diagnosis for many of the more secretin infusion leads to significant further elevation
Postgastrectomy Syndromes CHAPTER 62 731
of gastrin level. CT, endoscopic ultrasound (EUS), and cholecystectomy may be considered if preoperative or
octreotide scan may be helpful, but exploration by an intraoperative evaluation reveals sludge, gallstones, or
experienced surgeon is the best way to find the tumor(s) an abnormal gallbladder, especially if subsequent chole-
if operation is indicated. cystectomy is likely to be difficult. Unless the gallbladder
After operation for marginal ulcer, eventual recurrence is symptomatic, prophylactic cholecystectomy should
is almost inevitable if both smoking and NSAIDs cannot only be considered if it is likely to be straightforward
be eliminated. Vagotomy and/or lifelong PPI should also and the gastric operation has gone well. Studies have
be considered, as well as empiric Helicobacter treatment.84–96 shown that the morbidity of cholecystectomy performed
concurrently with upper abdominal surgery is higher than
that of cholecystectomy alone performed at a subsequent
GALLSTONES intervention. In addition, the results of controlled studies
Gallstone formation following gastric surgery is not uncom- and meta-analyses comparing prophylactic cholecystectomy
mon. It occurs in 10% to 20% of patients and typically to no cholecystectomy remain inconclusive.
presents within 3 years after surgery.97 The pathophysiol-
ogy of increased gallstone formation after gastrectomy
is not completely clear. The type of gastrectomy, the NUTRITIONAL ABNORMALITIES
extent of lymph node dissection, and the method of
digestive reconstruction seem to influence the rate of WEIGHT LOSS
t
postgastrectomy gallstone formation. Vagal denervation Weight loss is common in patients who have had a gastric
ne
is the most thoroughly studied mechanism. Denervation operation for tumor or ulcer. The degree of weight loss
of the gallbladder can occur through truncal vagotomy or tends to parallel the magnitude of the operation. It may
selective division of the hepatic branches of the left vagus. be insignificant in the large person or devastating in the
k.
Also, nodal dissection can lead to transection of the nerves asthenic female. The surgeon should always consider the
controlling gallbladder motility. This may explain the possible nutritional consequences before performing
oo
higher risk and earlier occurrence of gallstone formation a gastric resection for benign disease in a thin female.
after total gastrectomy than after distal gastrectomy98 and The causes of weight loss after gastric surgery generally
after extended lymphadenectomy compared to more fall into one of two categories: altered dietary intake or
limited perigastric lymphadenectomy.86,98 Other factors
yb
malabsorption. Chronic nausea, vomiting, and/or pain
that affect gallstone formation include changes in the frequently lead to decreased food consumption and/or
secretions of gut hormones, including cholecystokinin change in meal composition. Postoperative “blind loops”
(CCK). The exclusion of food passage through the duo- can lead to bacterial overgrowth and alteration in the
er
denum after Roux-en-Y reconstruction results in decreased small bowel (resection and/or Roux limb) can decrease
secretion of CCK with a resultant decrease in gallbladder absorptive surface area, both leading to malabsorption
rg
motility, which explains the higher incidence of stone of protein, fat, and vitamins.
formation when Roux-en-Y reconstruction is performed.98 Chronic postgastrectomy weight loss is most likely due
Other contributing factors include weight loss, leading to decreased intake if a stool stain for fecal fat is negative.
su
to mobilization of cholesterol stores and supersaturation This is the most common cause of weight loss after gastric
of biliary cholesterol, and the presence of edema or surgery and may be due to a combination of specific
inflammation around the bile ducts with resultant biliary factors such as small gastric pouch, postoperative gastric
://
will develop cholelithiasis within 2 years100 and 6% will overgrowth is suspected, a lactulose breath test and/or
ultimately undergo cholecystectomy for symptoms.103 It trial of oral antibiotics or probiotics should be considered.
is more likely to occur after radical gastrectomy (30%) Consultation with an experienced dietitian may prove
ht
than after simple gastrectomy (5%).104 A Roux-en-Y invaluable. Prolonged enteral or parenteral nutritional
reconstruction increases the risk of gallstone develop- support may be necessary in the worse cases.
ment compared with techniques that do not exclude
the duodenum (28%).86,98,105,106 In the first 2 years after ANEMIA
gastric bypass, the risk of cholelithiasis rises to 5.8 times Anemia is a common finding in postgastrectomy patients,
that of the general population. Of note, the incidence of occurring in up to one-third of patients. This is generally
cholelithiasis in the obese female patient is as high as 21% secondary to nutrient malabsorption but can also be
to 38%.107 Gastric bypass with Roux-en-Y GJ is intrinsically caused by decreased nutrient intake or chronic blood
more lithogenic than adjustable gastric banding.108 The loss due to ulcer, tumor, or mucosal inflammation. Iron,
incidence of gallstones after sleeve gastrectomy has not vitamin B12, and folate deficiencies are the most common
been well studied but seems to be similar to that after cause of chronic nutritional anemia after gastric surgery.
gastric bypass.109 For all types of bariatric surgery, the Iron absorption takes place primarily in the duodenum
rapid rate of weight loss110 and the extent of the weight and proximal jejunum and is facilitated by an acidic
loss111,112 lead to the development of cholesterol stones environment in the stomach. Intrinsic factor, essential
in supersaturated bile. for the enteric absorption of vitamin B12, is made by the
There is currently no consensus regarding the need parietal cells of the stomach. Vitamin B12 bioavailability also
for prophylactic cholecystectomy during gastric operation. is facilitated by an acidic gastric environment. Folate-rich
Although routine incidental cholecystectomy is unwise, foods (e.g., green leafy vegetables, fresh fruit, enriched
732 SECTION II Stomach and Small Intestine
bread) may be problematic in patients with a small or associated with symptoms after distal gastrectomy. Am J Gastroenterol.
hypomotile gastric pouch. 2003;98(12):2642-2647.
9. Abrahamsson H, Jansson G. Vago-vagal gastro-gastric relaxation in
With this as background, it is easy to understand why the cat. Acta Physiol Scand. 1973;88(3):289-295.
many patients who have had a gastric operation are at risk 10. Abell TL, Minocha A. Gastrointestinal complications of bariatric
for anemia. Iron deficiency is the most common cause of surgery: diagnosis and therapy. Am J Med Sci. 2006;331(4):214-218.
anemia after gastric surgery, but vitamin B12 and/or folate 11. Mine S, Sano T, Tsutsumi K, et al. Large-scale investigation into
dumping syndrome after gastrectomy for gastric cancer. J Am Coll
deficiency are common. Although patients who have had Surg. 2010;211(5):628-636.
a total gastrectomy will all develop life-threatening B 12 12. Tack J, Arts J, Caenepeel P, De Wulf D, Bisschops R. Pathophysiology,
deficiency without supplemental B12, it may be prudent diagnosis and management of postoperative dumping syndrome.
to monitor all patients after gastric operation for these Nat Rev Gastroenterol Hepatol. 2009;6(10):583-590.
three nutrient deficiencies. Routine supplementation 13. Arts J, Caenepeel P, Bisschops R, et al. Efficacy of the long-acting
repeatable formulation of the somatostatin analogue octreotide in
with oral iron, oral folic acid, and oral or parenteral B12 postoperative dumping. Clin Gastroenterol Hepatol. 2009;7(4):432-437.
should be done after gastric bypass and total gastrectomy. 14. Berg P, Hall M, McCallum RW, Sarosiek I. Dumping syndrome:
updated perspectives on etiologies and diagnosis. Pract Gastroenterol.
2014;38:30-38.
CHRONIC CALCIUM DEFICIT 15. Behrns KE, Sarr M. Diagnosis and management of gastric emptying
disorders. Adv Surg. 1994;27:233-255.
AND OSTEOPOROSIS 16. Gulsrud PO, Taylor IL, Watts HD, Cohen MB, Elashoff J, Meyer JH.
How gastric emptying of carbohydrate affects glucose tolerance and
t
Chronic calcium deficit and osteoporosis may occur after symptoms after truncal vagotomy with pyloroplasty. Gastroenterology.
ne
gastric operation. Calcium absorption occurs primarily in 1980;78:1463-1471.
17. Lipsitz LA, Ryan SM, Parker JA, Freeman R, Wei JY, Goldberger
the duodenum, so any gastric operation that diverts the AL. Hemodynamic and autonomic nervous system responses
food stream away from the duodenum will disturb calcium to mixed meal ingestion in healthy young and old subjects and
k.
homeostasis. This includes simple GJ, distal gastrectomy dysautonomic patients with postprandial hypotension. Circulation.
with Billroth II, and Roux-en-Y GJ (including gastric bypass) 1993;87(2):391.
18. Yamamoto H, Mori T, Tsuchihashi H, Akabori H, Naito H, Tani T.
oo
or Roux-en-Y esophagojejunostomy. Furthermore, any A possible role of GLP-1 in the pathophysiology of early dumping
gastric procedure that predisposes to bacterial overgrowth syndrome. Dig Dis Sci. 2005;50(12):2263-2267.
or inadequate mixing of food and digestive enzymes 19. Ukleja A. Dumping syndrome: pathophysiology and treatment.
yb
may interfere with the absorption of fat-soluble vitamins, Nutr Clin Pract. 2005;20(5):517-525.
including vitamin D. Thus it is likely that both calcium 20. van der Kleij FG, Vecht J, Lamers CB, Masclee AA. Diagnostic value
of dumping provocation in patients after gastric surgery. Scand J
and vitamin D malabsorption contribute to metabolic Gastroenterol. 1996;31(12):1162-1166.
bone disease in patients following gastric surgery. The
er
2012;22(10):1600-1606.
should prompt a study of bone density. Oral calcium and 22. Ramadan M, Loureiro M, Laughlan K, et al. Risk of dumping
vitamin D supplementation may be useful in preventing syndrome after sleeve gastrectomy and Roux-en-Y gastric bypass:
these complications and should be considered in all
su
criteria for diagnosis and identifying new etiologies. Dig Dis Sci.
REFERENCES 2010;55(1):117-123.
26. Didden P, Penning C, Masclee AA. Octreotide therapy in dumping
ht
1. Kitagawa Y, Dempsey DT, et al. Stomach. In: Brunicardi FC, Andersen syndrome: analysis of long-term results. Aliment Pharmacol Ther.
DK, Billiar TR, eds. Schwartz’s Principles of Surgery. New York: Mc 2006;24(9):1367-1375.
Graw Hill; 2015. 27. Leeds AR, Ralphs DN, Ebied F, Metz G, Dilawari JB. Pectin in the
2. Dempsey DT. Reoperative gastric surgery and postgastrectomy dumping syndrome: reduction of symptoms and plasma volume
syndromes. In: Zuidema GD, Yeo CJ, eds. Shackleford’s Surgery of changes. Lancet. 1981;317(8229):1075-1078.
the Alimentary Tract. Philadelphia, PA: Saunders; 2002:161. 28. Imhof A, Schneemann M, Schaffner A, Brändle M. Reactive hypo-
3. Meilahn JE, Dempsey D. Postgastrectomy problems: remedial glycaemia due to late dumping syndrome: successful treatment
operations and therapy. In: Cameron JL, ed. Current Surgical Therapy. with acarbose. Swiss Med Wkly. 2001;131(5-6):81-83.
Phladelphia, PA: Elsevier Mosby; 2004. 29. Yamada M, Ohrui T, Asada M, et al. Acarbose attenuates hypoglycemia
4. Gustavsson S, Kelly KA, Melton LJ 3rd, Zinsmeister AR. Trends from dumping syndrome in an elderly man with gastrectomy. J Am
in peptic ulcer surgery. A population based study in Rochester, Geriatr Soc. 2005;53(2):358-359.
Minnesota, 1956–1985. Gastroenterology. 1988;94(3):688-694. 30. De Cunto A, Barbi E, Minen F, Ventura A. Safety and efficacy of
5. Bardhan KD, Royston C. Time, change and peptic ulcer disease high-dose acarbose treatment for dumping syndrome. J Pediatr
in Rotherham, UK. Dig Liver Dis. 2008;40(7):540-546. Gastroenterol Nutr. 2011;53(1):113-114.
6. Wainess RM, Dimick JB, Upchurch GR Jr, Cowan JA, Mulholland 31. Hasegawa T, Yoneda M, Nakamura K, et al. Long-term effect of
MW. Epidemiology of surgically treated gastric cancer in the United α-glucosidase inhibitor on late dumping syndrome. J Gastroenterol
States, 1988–2000. J Gastrointest Surg. 2003;7(7):879-883. Hepatol. 1998;13(12):1201-1206.
7. Pedrazzani C, Marrelli D, Rampone B, et al. Postoperative com- 32. Deleted in review.
plications and functional results after subtotal gastrectomy with 33. Parrish CR. The clinician’s guide to short bowel syndrome. Pract
Billroth II reconstruction for primary gastric cancer. Dig Dis Sci. Gastroenterol. 2005;29:67.
2007;52(8):1757-1763. 34. Berg P, Hall M, Sarosiek I, McCallum RW. Understanding the
8. Le Blanc-Louvry I, Savoye G, Maillot C, Denis P, Ducrotté P. An etiologies, clinical spectrum, and diagnostic challenge of dumping
impaired accommodation of the proximal stomach to a meal is syndrome. Gastroenterology. 2013;144:S-734.
Postgastrectomy Syndromes CHAPTER 62 733
35. Thondam SK, Nair S, Wile D, Gill GV. Diazoxide for the treatment 62. Ritchie WP. Alkaline reflux gastritis. Gastroenterol Clin North Am.
of hypoglycaemic dumping syndrome. QJM. 2013;106(9):855. 1994;23:281-294.
36. Spanakis E, Gragnoli C. Successful medical management of status 63. Dixon MF, O’Connor HJ, Axon AT, King RF, Johnston D. Reflux
post-Roux-en-Y- gastric-bypass hyperinsulinemic hypoglycemia. Obes gastritis: distinct histopathological entity? J Clin Pathol. 1986;39(5):
Surg. 2009;19(9):1333-1334. 524-530.
37. Li-Ling J, Irving M. Therapeutic value of octreotide for patients 64. Deleted in review.
with severe dumping syndrome—a review of randomised controlled 65. Davidson ED, Herish T. The surgical treatment of bile reflux
trials. Postgrad Med J. 2001;77:441-442. gastritis. Ann Surg. 1986;192:175-187.
38. Hasler WL, Soudah HC, Owyang C. Mechanisms by which octreotide 66. Ritchie WP Jr, Dempsey DT. Postgastrectomy syndromes. In: Moody
ameliorates symptoms in the dumping syndrome. J Pharmacol Exp FG, ed. Surgical Treatment of Digestive Disease. Chicago: Year Book
Ther. 1996;277(3):1359. Medical Publishers; 1990:236-248.
39. Vecht J, Lamers CB, Masclee AA. Long-term results of octreotide- 67. Cabrol J, Navarro X, Sancho J, Simo-Deu J, Segura R. Bile reflux
therapy in severe dumping syndrome. Clin Endocrinol (Oxf). in postoperative alkaline reflux gastritis. Ann Surg. 1990;211(2):239-
1999;51(5):619-624. 243.
40. Richards WO, Golzarian J, Wasudev N, Sawyers JL. Reverse phasic 68. McAlhany JC, Hanover TM, Taylor SM, Sticca RP, Ashmore JD Jr.
contractions are present in antiperistaltic jejunal limbs up to Long-term follow-up of patients with Roux-en-Y gastrojejunostomy
twenty-one years postoperatively. J Am Coll Surg. 1994;178(6):557-563. for gastric disease. Ann Surg. 1994;219(5):451-457.
41. Sawyers JL, Herrington JL Jr. Superiority of antiperistaltic jejunal 69. Aranow JS, Matthews JB, Garcia-Aguilar J, Novak G, Silen W.
segments in management of severe dumping syndrome. Ann Surg. Isoperistaltic jejunal interposition for intractable postgastrectomy
1973;178(3):311-319. alkaline reflux gastritis. J Am Coll Surg. 1995;180(6):648-653.
42. Nunobe S, Sasako M, Saka M, Fukagawa T, Katai H, Sano T. 70. Klingler PJ, Perdikis G, Wilson P, Hinder RA. Indications, tech-
Symptom evaluation of long-term postoperative outcomes after nical modalities and results of the duodenal switch operation
t
pylorus-preserving gastrectomy for early gastric cancer. Gastric for pathologic duodenogastric reflux. Hepatogastroenterology.
ne
Cancer. 2007;10(3):167-172. 1999;46(25):97-102.
43. Ishikawa K, Arita T, Ninomiya S, Bandoh T, Shiraishi N, Kitano S. 71. Henley FA, Hudson RV. Gastrectomy with replacement. A preliminary
Outcome of segmental gastrectomy versus distal gastrectomy for communication with an introduction. Br J Surg. 1952;40(160):118-128.
early gastric cancer. World J Surg. 2007;31(11):2204-2207. 72. Herrington JL, Sawyers JL, Whitehead WA. Surgical management
k.
44. Katsube T, Konno S, Murayama M, et al. Gastric emptying after of reflux gastritis. Ann Surg. 1974;180(4):526-535.
pylorus-preserving gastrectomy: assessment using the 13C-acetic 73. DeMeester TR, Fuchs KH, Ball CS, Albertucci M, Smyrk TC, Marcus
acid breath test. Hepatogastroenterology. 2007;54(74):639-642.
oo
JN. Experimental and clinical results with proximal end-to-end
45. Duncombe VM, Bolin TD, Davis AE. Double-blind trial of chole- duodenojejunostomy for pathologic duodenogastric reflux. Ann
styramine in post-vagotomy diarrhoea. Gut. 1977;18(7):531-535. Surg. 1987;206(4):414-426.
46. Forstner-Barthell AW, Murr MM, Nitecki S, et al. Near-total comple- 74. Hinder RA. Duodenal switch: a new form of pancreaticobiliary
yb
tion gastrectomy for severe postvagotomy gastric stasis: analysis diversion. Surg Clin North Am. 1992;72(2):487-499.
of early and long-term results in 62 patients. J Gastrointest Surg. 75. Kojima K, Yamada H, Inokuchi M, Kawano T, Sugihara K. A com-
1999;3(1):15-21 [discussion 21–23]. parison of Roux-en-Y and Billroth-I reconstruction after laparoscopy-
47. McClelland RN, Horton JW. Relief of acute, persistent postvagotomy assisted distal gastrectomy. Ann Surg. 2008;247(6):962-967.
er
atony by metoclopramide. Ann Surg. 1978;188(4):439-447. 76. Miedema BW, Kelly KA. The Roux stasis syndrome: treatment by
48. Davis RH, Clench MH, Mathias JR. Effects of domperidone pacing and prevention by use of an uncut Roux limb. Arch Surg.
in patients with chronic unexplained upper gastrointestinal 1992;127(3):295-300.
rg
symptoms: a double-blind, placebo-controlled study. Dig Dis Sci. 77. Janssens J, Peeters TL, Vantrappen G, et al. Improvement of gastric
1988;33(12):1505-1511. emptying in diabetic gastroparesis by erythromycin. N Engl J Med.
49. Tack J, Janssens J, Vantrappen G, et al. Effect of erythromycin on 1990;322(15):1028-1031.
su
gastric motility in controls and in diabetic gastroparesis. Gastroenterol- 78. Verne GN, Sninsky CA. Chronic intestinal pseudo-obstruction. Dig
ogy. 1992;103(1):72-79. Dis. 1995;13(3):163-181.
50. McCallum R, Lin Z, Wetzel P, Sarosiek I, Forster J. Clinical response 79. Le Blanc-Louvry IL, Ducrotté P, Lemeland JF, Metayer J, Denis
to gastric electrical stimulation in patients with postsurgical gas- P, Ténière P. Motility in the Roux-Y limb after distal gastrectomy:
://
troparesis. Clin Gastroenterol Hepatol. 2005;3(1):49-54. relation to the length of the limb and the afferent duodenoje-
51. Woodfield CA, Levine MS. The postoperative stomach. Eur J Radiol. junal segment—an experimental study. Neurogastroenterol Motil.
2005;53(3):341-352. 1999;11(5):365-374.
tp
52. Bolton JS, Conway WC 2nd. Postgastrectomy syndromes. Surg Clin 80. Hinder RA, Esser J, DeMeester TR. Management of gastric
North Am. 2011;91(5):1105-1122. emptying disorders following the Roux-en-Y procedure. Surgery.
53. Kimura H, Ishikawa M, Nabae T, et al. Internal hernia after 1988;104(4):765-772.
ht
laparoscopic gastrectomy with Roux-en-Y reconstruction for gastric 81. Turnage RH, Sarosi G, Cryer B, Spechler S, Peterson W, Feldman
cancer. Asian J Surg. 2017;40(3):203-209. M. Evaluation and management of patients with recurrent peptic
54. Al Harakeh AB, Kallies KJ, Borgert AJ, Kothari SN. Bowel obstruc- ulcer disease after acid-reducing operations: a systematic review.
tion rates in antecolic/antegastric versus retrocolic/retrogastric J Gastrointest Surg. 2003;7(5):606-626.
Roux limb gastric bypass: a meta-analysis. Surg Obes Relat Dis. 82. Gurusamy KS, Pallari E. Medical versus surgical treatment for
2016;12(1):194-198. refractory or recurrent peptic ulcer. Cochrane Database Syst Rev.
55. Mitty WF Jr, Grossi C, Nealon TF Jr. Chronic afferent loop syndrome. 2016;(3):CD011523.
Ann Surg. 1970;172(6):996-1001. 83. Ingvar C, Adami HO, Enander LK, Enskog L, Rydberg B. Clinical
56. Zissin R. CT findings of afferent loop syndrome after a subto- results of reoperation after failed highly selective vagotomy. Am J
tal gastrectomy with Roux-en-Y reconstruction. Emerg Radiol. Surg. 1986;152(3):308-312.
2004;10(4):201-203. 84. Tolin RD, Malmud LS, Stelzer F, et al. Enterogastric reflux in normal
57. Zissin R, Osadchy A, Gayer G. Abdominal CT findings of delayed subjects and patients with Billroth II gastroenterostomy. Measure-
postoperative complications. Can Assoc Radiol J. 2007;58:200-211. ment of enterogastric reflux. Gastroenterology. 1979;77(5):1027-1033.
58. Perdikis G, Wilson P, Hinder R, et al. Altered antroduodenal 85. Natomi H, Sugano K, Iwamori M, Takaku F, Nagai Y. Region-specific
motility after cholecystectomy. Am J Surg. 1994;168(6):609-615. distribution of glycosphingolipids in the rabbit gastrointestinal
59. Hubens A, Van de Kelft E, Roland J. The influence of cholecystec- tract: preferential enrichment of sulfoglycolipids in the mucosal
tomy on the duodenogastric reflux of bile. Hepatogastroenterology. regions exposed to acid. Biochim Biophys Acta. 1988;961(2):213-222.
1989;36(5):384-389. 86. Akatsu T, Yoshida M, Kubota T, et al. Gallstone disease after
60. Madura JA, Grosfeld JL. Biliary diversion: a new method to prevent extended (D2) lymph node dissection for gastric cancer. World J
enterogastric reflux and reverse the roux stasis syndrome. Arch Surg. 2005;29(2):182-186.
Surg. 1997;132(3):245-249. 87. Arlt G, Schumpelick V, Kloppel G. [Ulcer risk in the Roux-Y
61. Ritchie WP. Alkaline reflux gastritis: a critical reappraisal. Gut. stomach. An animal experiment study]. Langenbecks Arch Chir.
1984;25(9):975-987. 1984;362(1):43-52.
734 SECTION II Stomach and Small Intestine
88. Oliver JV. Effect of vagus section and isolation of the pyloric antrum 101. Chiloiro M, Pezzolla F, Riezzo G, Maselli MA, Lorusso D. Gallbladder
on the development of the Mann-Williamson ulcer in the dog. Surg motility before and after Billroth II gastric resection. Neurogastroenterol
Forum. 1953;4:336-338. Motil. 1995;7(3):145-149.
89. Kalaiselvan R, Exarchos G, Hamza N, Ammori BJ. Incidence of 102. Masclee AA, Jansen JB, Driessen WM, Geuskens LM, Lamers CB.
perforated gastrojejunal anastomotic ulcers after laparoscopic Effect of truncal vagotomy on cholecystokinin release, gallbladder
gastric bypass for morbid obesity and role of laparoscopy in their contraction, and gallbladder sensitivity to cholecystokinin in humans.
management. Surg Obes Relat Dis. 2012;8(4):423-428. Gastroenterology. 1990;98(5 Pt 1):1338-1344.
90. Patel RA, Brolin RE, Gandhi A. Revisional operations for mar- 103. Gillen S, Michalski CW, Schuster T, Feith M, Friess H, Kleeff J.
ginal ulcer after Roux-en-Y gastric bypass. Surg Obes Relat Dis. Simultaneous/incidental cholecystectomy during gastric/esophageal
2009;5(3):317-322. resection: systematic analysis of risks and benefits. World J Surg.
91. Rasmussen JJ, Fuller W, Ali MR. Marginal ulceration after lapa- 2010;34(5):1008-1014.
roscopic gastric bypass: an analysis of predisposing factors in 260 104. Wu CC, Chen CY, Wu TC, Iiu TJ, P’eng PK. Cholelithiasis and
patients. Surg Endosc. 2007;21(7):1090-1094. cholecystitis after gastrectomy for gastric carcinoma: a comparison
92. Suggs WJ, Kouli W, Lupovici M, Chau WY, Brolin RE. Complications of lymphadenectomy of varying extent. Hepatogastroenterology.
at gastrojejunostomy after laparoscopic Roux-en-Y gastric bypass: 1995;42(6):867-872.
comparison between 21- and 25-mm circular staplers. Surg Obes 105. Nunobe S, Okaro A, Sasako M, et al. Billroth 1 versus Roux-en-Y
Relat Dis. 2007;3(5):508-514. reconstructions: a quality-of-life survey at 5 years. Int J Clin Oncol.
93. Felix EL, Kettelle J, Mobley E, Swartz D. Perforated marginal ulcers 2007;12(6):433-439.
after laparoscopic gastric bypass. Surg Endosc. 2008;22(10):2128-2132. 106. Csendes A, Larach J, Godoy M. Incidence of gallstones development
94. Higa KD, Boone KB, Ho T. Complications of the laparoscopic after selective hepatic vagotomy. Acta Chir Scand. 1978;144(5):289-291.
Roux-en-Y gastric bypass: 1,040 patients—what have we learned? 107. Veyrie N, Servajean S, Berger N, Loire P, Basdevant A, Bouillot JL.
Obes Surg. 2000;10(6):509-513. [Gallbladder complications after bariatric surgery]. Gastroenterol
t
95. Bendewald FP, Choi JN, Blythe LS, Selzer DJ, Ditslear JH, Mattar Clin Biol. 2007;31(4):378-384.
ne
SG. Comparison of hand-sewn, linear-stapled, and circular-stapled 108. Fobi M, Lee H, Igwe D, et al. Prophylactic cholecystectomy with
gastrojejunostomy in laparoscopic Roux-en-Y gastric bypass. Obes gastric bypass operation: incidence of gallbladder disease. Obes
Surg. 2011;21(11):1671-1675. Surg. 2002;12(3):350-353.
96. Capella JF, Capella RF. Gastro-gastric fistulas and marginal ulcers 109. Li VK, Pulido N, Martinez-Suartez P, et al. Symptomatic gallstones
k.
in gastric bypass procedures for weight reduction. Obes Surg. after sleeve gastrectomy. Surg Endosc. 2009;23(11):2488-2492.
1999;9(1):22-27 [discussion 28]. 110. Iglezias Brandao de Oliveira C, Adami Chaim E, da Silva BB. Impact
97. Kodama I, Yoshida C, Kofuji K, Ohta J, Aoyagi K, Takeda J. Gallstones of rapid weight reduction on risk of cholelithiasis after bariatric
oo
and gallbladder disorder after gastrectomy for gastric cancer. Int surgery. Obes Surg. 2003;13(4):625-628.
Surg. 1996;81(1):36-39. 111. Li VK, Pulido N, Fajnwaks P, Szomstein S, Rosenthal R, Martinez-
98. Kobayashi T, Hisanaga M, Kanehiro H, Yamada Y, Ko S, Nakajima Duartez P. Predictors of gallstone formation after bariatric surgery:
yb
Y. Analysis of risk factors for the development of gallstones after a multivariate analysis of risk factors comparing gastric bypass, gastric
gastrectomy. Br J Surg. 2005;92(11):1399-1403. banding, and sleeve gastrectomy. Surg Endosc. 2009;23(7):1640-1644.
99. Deleted in review. 112. Tarantino I, Warschkow R, Steffen T, Bisang P, Schultes B, Thurnheer
100. Fukagawa T, Katai H, Saka M, Morita S, Sano T, Sasako M. Gallstone M. Is routine cholecystectomy justified in severely obese patients
er
formation after gastric cancer surgery. J Gastrointest Surg. 2009;13(5): undergoing a laparoscopic Roux-en-Y gastric bypass procedure? A
886-889. comparative cohort study. Obes Surg. 2011;21(12):1870-1878.
rg
su
://
tp
ht
CHAPTER
Operations for Morbid Obesity
Bruce Schirmer
63
B
ariatric surgery had its beginnings in the 1950s when popularized this modification.5 The Roux-en-Y gastric
malabsorptive operations were performed upon bypass (RYGB) proved to be a very effective operation
patients with severe hyperlipidemia and obesity. for weight loss, as well as treatment of comorbidities
Edward Mason was undoubtedly the father of American associated with obesity. Major proponents of the procedure
bariatric surgery, having first described the gastric bypass during the 1980s included Sugerman, who described its
in 19691 and later the vertical banded gastroplasty (VBG)2 efficacy in treating hypertension, diabetes,6 pseudotumor
in 1981. In between, unfortunately, the field suffered a cerebri,7 and venous stasis ulcers.8 Pories and his group
major initial setback from which it took decades to recover. from East Carolina University were the first to emphasize
The culprit was the performance of the jejunoileal bypass. the beneficial effect of RYGB on the treatment of type
t
The operation was originally designed for hyperlipidemia 2 diabetes.9 Improvements in life expectancy, degree of
ne
and obesity together and was frequently performed in the comorbid medical problems, and cost effectiveness of the
1970s before its untoward side effects, especially hepatic procedure were soon shown in large studies by Christou
failure in a small percentage of patients, were described and MacLean’s group,10,11 Buchwald,12 and Adams.13
k.
and appreciated.3 The subsequent two decades involved Bariatric surgery experienced its only major change in
reversing these operations for many of the patients who operative approach, through the advent of laparoscopy,
oo
had received them, as they manifested problems with mal- slightly later than most general surgery procedures.
absorption of protein, calories, or essential minerals such Wittgrove and Clark14 described the first laparoscopic
as magnesium. Liver failure in 2% of patients, however, gastric bypass in 1995. Soon thereafter, Schauer15 and
later Nguyen and Wolfe16 confirmed the efficacy and
yb
was the true insurmountable danger of this procedure.
Restriction, rather than malabsorption, was then felt benefits of performing RYGB using this approach. Patients
to be the optimal approach for patients with morbid voted rather quickly for adopting this new approach, with
obesity. Various stapling operations of the stomach were the incidence of RYGB in the United States increasing
er
performed to try to limit food intake. Many of these exponentially during the years 1999–2004 (Fig. 63.1).17
suffered from the lack of understanding that a staple line Since then, the incidence of performance of all bariatric
rg
in an intact stomach will, in a high percentage of cases, operations in the United States has risen only by 25% in the
break down and allow passage of luminal contents. This past 10 to 12 years. However, the spectrum of procedures
is the principle used in pyloric exclusion for duodenal and their relative safety of performance have changed
su
injury. Thus many of the patients who had these stapling significantly since the introduction of laparoscopy in the
procedures experienced initial excellent weight loss but first few years of the 21st century.
subsequent regain after the staple line broke down.4 Although the RYGB was the predominant procedure
://
Mason then championed the VBG, which proved to performed in the United States at the start of the 21st
have more durability than simple stapled operations. 2 century, internationally the laparoscopic adjustable gastric
tp
The lesser curvature of the stomach, which is resistant to band (LAGB), introduced by Belachew in 1994,18 was taking
dilation with pressure, was used for the restrictive pouch. the European and Australian continents by storm. LAGB’s
The pouch outlet was reinforced with a circular piece of popularity blossomed in the 15 years following its introduc-
ht
permanent mesh material. Due to its relative technical tion, to the point where in 2009 in the United States it
ease of performance, the operation became immensely rivaled RYGB as the most popular operation performed
popular in the 1980s and was for a period of time the for morbid obesity. Its safety profile was excellent, and it
most popular bariatric procedure performed. This trend proved effective in settings where careful and available
continued until the last decade of the 20th century. By that follow-up for adjustments to the band existed.19 Patients
time the limitations of the VBG had become apparent. viewed it as a less invasive procedure, and its popularity
Patients could not eat a normal healthy diet of vegetables increased accordingly. However, by 2010 the long-term
and fruits, due to the restriction of the band and pouch, efficacy of the LAGB had proven poor in many centers.
and turned instead to a diet of high-calorie liquids and Patients had experienced problems with poor weight loss,
junk food. Weight regain followed. In addition, a percent- frequent prolapse of the band requiring multiple adjust-
age experienced severe gastric outlet obstruction due ments, and overall lack of satisfaction with the amount of
to progressive hypertrophy around the band, requiring weight loss versus the symptoms, cost, and inconvenience
revisional surgery. of maintaining the band in exactly the correct range of
The operation that the VBG was revised to was most restriction. Centers began reporting a high incidence of
often the gastric bypass. Since its introduction in 1969, band removal, with long-term follow-up showing more than
it had evolved quickly from a loop gastrojejunostomy to 50% of bands removed in one center.20 Many patients lost
a Roux-en-Y gastrojejunostomy, due to complications of insurance coverage for adjustments, further decreasing
bile acid reflux esophagitis from the loop. Griffen et al. the efficacy of the band. By 2012 the operation was clearly
735
Operations for Morbid Obesity CHAPTER 63 735.e1
ABSTRACT
The most commonly performed operations both in the
United States and worldwide are the laparoscopic sleeve
gastrectomy and the laparoscopic Roux-en-Y gastric bypass.
Both are associated with excellent durable weight loss
and resolution of associated medical comorbidities.
These operations are furthermore done with increasing
safety, such that they are associated with lower morbidity
and mortality than all intraabdominal operations except
appendectomy. The laparoscopic adjustable gastric
band is rapidly losing popularity due to poor long-term
efficacy, whereas patients choose the duodenal switch
and other malabsorptive operations infrequently. Finally,
new endoscopic and minimally invasive procedures offer
potentially less invasive mechanisms of producing weight
loss, although their long-term efficacy and durability still
are not established.
t
KEYWORDS
ne
Bariatric, gastric bypass, sleeve gastrectomy, morbid obesity,
metabolic surgery
k.
oo
yb
er
rg
su
://
tp
ht
736 SECTION II Stomach and Small Intestine
t
during the past 15 years has improved dramatically. In 2000
ne
TABLE 63.1 American Society of Metabolic and Bariatric the literature reflected the fact that open RYGB had an
Surgery Estimate of Numbers of Bariatric Procedures in accepted mortality rate approaching 1%, with 2% being
United States seen in higher-risk patient populations.26 The profession
k.
then did an outstanding job of quality improvement
2011 2012 2013 2014 2015
through several means. Laparoscopic procedures proved
oo
Total 158,000 173,000 179,000 193,000 196,000 less morbid, especially for long-term wound complica-
RYGB 36.7 37.5 34.2 26.8 23.1 tions and incisional hernias. They also were less morbid
Sleeve 17.8 33 42.1 51.7 53.8 for short-term complications, length of hospital stay,
Band 35.4 20.2 14 9.5 5.7
and mortality.27 The ASMBS and the American College
yb
DS 0.9 1 1 0.4 0.6
of Surgeons (ACS) adopted the concept of centers of
Revisions 6 6 6 11.5 13.6
excellence, and self-review, peer pressure, and external
Other 3.2 2.3 2.7 0.1 3.2
auditing all combined to produce centers that achieved
er
DS, Duodenal switch; RYGB, Roux-en-Y gastric bypass. improved outcomes for bariatric procedures.28 By 2014
the reported national mortality rate for a LRYGB was
rg
Part of the reason for the LAGB decline has been the The LAGB is rapidly declining in popularity, and within
evolution of the laparoscopic sleeve gastrectomy (LSG). a year may be as infrequently performed as LDS, which
Initially described by Gagner and his group as being an has consistently represented 1% to 2% of procedures for
://
effective first step in the two-stage performance of the the past several years.
laparoscopic duodenal switch (LDS) procedure, 21 the
tp
t
ne
FIGURE 63.2 Port placement for lap band.
k.
Relative contraindications to an LAGB would be a high
body mass index (BMI) of more than 50, poor mobility,
oo
lack of ability to exercise, failure to have successfully
dieted and lost more than 25 pounds, small hiatal hernia,
and previous gastric surgery. Absolute contraindications
yb
include previous antireflux surgery, large hiatal hernias,
and esophageal motility disorders.
OPERATIVE STEPS
er
Ports for surgery are generally located above the umbilicus, FIGURE 63. 3 Buckle for lap band.
within 15 to 18 cm of the xiphoid. A 15-mm and 5-mm port
rg
patient’s left side, can optimally help with two left upper that small amount of stomach above the top of the band.
quadrant 5-mm ports. The camera port (12 mm) is placed It is locked onto itself, securing it in place. The band has
just to the left and above the umbilicus. A liver retractor a locking buckle mechanism that allows it to self-secure
://
is placed in the xiphoid region (Fig. 63.2). in this location (Fig. 63.3). The buckle of the band is
After port placement, the surgeon opens the gastro- positioned above the lesser curvature of the stomach.
tp
hepatic ligament in its avascular region proximally. This The fundus is now brought up over the left lateral and
pars flaccida area has been used to describe this technique anterior portions of the band to cover the band and secure
of band placement. The right crus is identified, and an it further into position. Several sutures are required to
ht
area approximately 2 to 3 cm below the gastroesophageal secure the fundus in this location (Fig. 63.4). Care should
junction on the medial border of the right crus is opened be taken to avoid placing the fundus over the buckle of
with a Harmonic scalpel or similar energy device to allow the band because erosion may occur.
the beginning of a tunnel for the lap band posterior to the The tubing is now brought out through the abdominal
stomach. This tunnel is then developed by the surgeon’s wall in a location where the port will be sited. We have
left hand, gently passing a grasper from the right to left favored placing the port just below the costal margin in
crus area posterior to the proximal stomach, hugging the the epigastric region. This location makes the port more
surface of the crura. There is a fair amount of fibrous easily palpable for adjustments. An incision is made on
tissue in this area, and the goal is to develop a tunnel the abdominal wall in the desired area of the port. The
within this tissue that allows for some posterior security tubing is brought out from the abdomen through a stab
of the band, preventing migration in either direction. wound on the medial side of this incision, as far to the
Once the grasper has emerged in the area of the angle end of the incision as possible. This allows the tubing to
of His, the band itself is introduced into the abdomen emerge through the fascia and take a natural slow bend
via the 15-mm port. The tubing end of the band system medially to be joined to the port. Securing the tubing to
is then grasped by the grasper and pulled back to the the port and then the port to the fascia in the incision site
patient’s right crus, through the previously developed completes the operation except to visually confirm that
tissue tunnel. The tubing is pulled through until the band addition of saline to the system causes the band to expand
is partially around the stomach. and not leak. The band is normally placed without any
738 SECTION II Stomach and Small Intestine
COMPLICATIONS
Early postoperative complications after band place-
ment are rare, and discharge the same day is the norm.
However, complications can occur, and these have been
well described in the literature.32 Inaccurate or poor
dissection technique can result in gastric perforation
and postoperative leak. Early band erosion occurs in
approximately 1% or less of cases.
Stenosis at the band site, or excessive band restriction,
will produce nausea and vomiting in most situations but can
also present as new-onset gastroesophageal reflux disease
(GERD) in less stenotic situations. Any such symptoms
of new onset need to be investigated promptly for either
simple excessive restriction or, more commonly, prolapse
of the band. Prolapse occurs when the stomach below
t
the band herniates up into the central lumen of the
ne
FIGURE 63.5 Fundus imbricated over lap band. band and too much stomach is forced into this space.
Complete or partial obstruction results. In severe cases
the prolapse can lead to ischemia and gangrene of the
k.
saline in the system to avoid initial excessive obstruction. prolapsed portion of the stomach. Chronic prolapse is
Fig. 63.5 shows the completed LAGB. seen at times with surprisingly large protrusions of the
oo
distal stomach up and over the edge of the gastric band.
BAND ADJUSTMENTS Technically prolapse is when the stomach herniates up
Postoperative adjustments to the band are necessary to through the band. Slippage is when the band slips down
yb
provide an optimal amount of restriction. When such onto the stomach. Both produce the same end result
adjustments are made, a good rule of thumb is to have mechanically and symptomatically. Prolapse can occur
the patient drink several swallows of water quickly after at any time after the procedure, and its incidence slowly
the adjustment is made. If the patient feels the water rises with duration of the band being in place. Recurrent
er
stop and give a sensation of partial blockage, then the prolapse has been a common reason for band removal
adjustment is too tight and must be loosened. Optimal among patients.
rg
restriction varies from patient to patient, but in general Diagnosis of prolapse is a clinical and, if necessary,
a goal of restriction to one cup of food or less at a meal radiographic one. The signs and symptoms are of obstruc-
and production of satiety for at least a few hours after tion or GERD, with obstruction present in most cases. A
su
eating are the goals of an optimal adjustment. plain film of the abdomen will reveal the band position,
Probably more so than any of the other procedures, the normally at a 7 o’clock to 1 o’clock orientation, flattened
success of LAGB is dependent on patient understanding to horizontal on the film. This is diagnostic for prolapse. If
://
of diet recommendations and compliance. The LAGB doubt exists, a low-volume Gastrografin or barium swallow
should help promote healthy eating habits, and nutritional will confirm the diagnosis.
tp
counseling on a frequent basis, especially early after the Treatment for prolapse initially is removal of all fluid
procedure, is important. in the system. This will, in most cases of acute prolapse,
provide enough reduction in the restriction of the pro-
ht
outpatient procedure under local anesthesia can correct 65 to 70, weight greater than 600 pounds (our cutoff),
this problem. persistent smoking, lack of mobility, and severe medical
problems.
t
of the 21st century, when bariatric centers of excellence Ireland) at approximately 45 to 50 cm distal to the ligament
ne
were formulated and when greater attention to quality (Fig. 63.7). The mesentery is then further divided with
outcomes produced significant improvements in the safety the Harmonic scalpel to obtain as deep a division of the
of the procedure. mesentery as possible without encountering the very large
k.
As with many popular operations, there are considerable vessels at the base of the mesentery. The proximal end of
variations on the theme as to how to best perform LRYGB. the Roux limb is then marked by suturing a small Penrose
oo
The technique of creating the gastrojejunostomy and the drain to it. The Roux limb length is now estimated and
length and location of the Roux limb has varied from
surgeon to surgeon. No optimal technique or configura-
yb
tion has emerged, although some differences have been
shown. Its performance using a laparoscopic approach has
clearly been an improvement over the open approach, as
with all other operations where minimal access has been
er
measured. I usually prefer a 150-cm Roux limb for patients done a centimeter or two above the incisura. However, for
with a BMI greater than 50. As the limb is being measured, very large patients, creating this opening at the incisura is
it is pulled to the left upper quadrant. The jejunum at the advisable because the longer gastric pouch is often needed
desired distance is then placed adjacent to the proximal to allow the Roux limb to easily reach the proximal stomach
jejunum, with the distal end of the proximal jejunum without tension. Once the opening is created, a green
facing to the patient’s right and the proximal end of the load of the GIA stapler is fired from the lesser curvature
Roux limb with the Penrose on it facing up and to the of the stomach to partially divide the stomach body. Then
left. A double-stapled enteroenterostomy is now created I prefer to size the pouch with an Ewald tube (30 French)
with two white loads of the GIA stapler. The stapler defect and place the stapler close to but not directly adjacent to
is sewn closed with running absorbable suture. Then the the tube, which is visible by the contour it creates on the
mesenteric defect at the enteroenterostomy is closed with gastric surface. The stapler, now firing directly cephalad
a running permanent suture. parallel to the lesser curvature, is fired several times until
The transverse colon mesentery is now grasped and the stomach is divided up to the angle of His (Fig. 63.9).
elevated, exposing the lower portion of the mesentery near It is important to exclude the fundus from the proximal
the ligament of Treitz. A defect is made in the mesentery part of the newly created gastric pouch. Care should
to the left and a few centimeters above the ligament of be taken not to catch the Penrose drain in the stapler.
Treitz. This location usually avoids major vessels, but the Similarly, the anesthesiologist needs to double confirm
surgeon must be aware of the vascular anatomy, if visible, there are no temperature probes or orogastric tubes in
t
and cautious not to disrupt it unnecessarily. Openings the stomach other than the Ewald tube.
ne
between mesenteric vessels are easier to find than dealing The Penrose drain is now usually visible in the ret-
with bleeding from major mesenteric vessels. Once the rogastric space. If it is not, the inferior surface of the
mesentery has been opened to expose the lesser sac, the transverse colon mesentery must again be exposed and
k.
posterior surface of the stomach can be seen. It is grasped the Roux limb passed into the retrogastric space again. I
and pulled out of the mesenteric defect a few centimeters, continue to use a retrogastric retrocolic location of the
oo
after which the plane below the stomach is confirmed with Roux limb due to the fact this is the shortest distance
a grasper. Adhesions, if present, are divided to now allow from between jejunum and proximal stomach. A more
the Penrose drain and then the proximal end of the Roux popular approach is to bring the Roux limb directly
yb
limb to be placed into the retrogastric space (Fig. 63.8). anterior to both transverse colon and distal stomach and
Usually if one can pass 4 cm of bowel or more past the cut create the gastrojejunostomy. This approach is technically
edge of the mesentery, that will suffice for later retrieval. easier, except when the mesentery of the Roux limb is
Perhaps one of the most important technical issues now short and there is difficulty in stretching the Roux limb
er
must be strictly obeyed. The Roux limb mesentery must to reach the proximal gastric pouch. For the retrocolic
not have any twists in it. It is very easy to have the bowel retrogastric approach, I now place the proximal suture
rg
twisted a full 180 degrees or more between first passing line of the Roux limb directly adjacent to the distal part
it to the left upper quadrant then retrieving it to pass it of the proximal gastric pouch. The distal 5 cm of gastric
through the transverse colon mesentery. The mesentery pouch is then tacked to the side of the proximal 5 cm of
su
of the Roux limb must be visually confirmed without a the Roux limb with a running absorbable suture.
doubt as being straight and vertical as the limb is passed The gastrojejunostomy is created using a linear stapling
superiorly through the transverse colon mesentery. device. We have found that the linear stapler is associated
://
Now attention is shifted to the stomach. The left lobe with an insignificant incidence of postoperative stenosis,
of the liver is retracted with a laparoscopic liver retractor whereas the circular stapler in our experience yielded
tp
of the surgeon’s choice. The Harmonic scalpel is used a 10% or higher stenosis rate. The Ewald tube serves
to create an opening in the mesentery along the lesser as a good backstop against which to make a gastrotomy
curvature of the stomach. For most patients, this can be in the end of the pouch. An enterotomy in the Roux
ht
FIGURE 63.8 Passing Roux limb. FIGURE 63.9 Creating gastric pouch.
Operations for Morbid Obesity CHAPTER 63 741
t
limb is less difficult. Having pulled the Ewald tube back,
ne
the stapler is advanced into the two lumens to its full
extent and fired (Fig. 63.10). We have not found that
restricting the anastomotic size has any relationship to
k.
long-term postoperative weight loss. The gastric pouch
size must be small, but the anastomosis need not be very
oo
small. The staple defect is closed with a running layer
of absorbable suture and reinforced with a second such
layer. An intraoperative leak test is now performed by
yb
having the anesthesiologist forcefully inject a methylene
blue dye solution into the lumen of the proximal pouch,
after having readvanced the tip of the Ewald tube to
that level.
er
space between the left lateral side of the Roux limb and
the transverse colon mesentery. Fig. 63.12 shows the
completed LRYGB.
ht
t
Small bowel obstruction can be due to simple adhesive of the procedure is not surprising. It is easy for patients to
ne
obstruction or, more dangerously, from internal hernias. understand and relatively straightforward for surgeons to
Internal hernias can lead to strangulation of large sections perform. Because there is no malabsorptive component,
of the small bowel mesentery, with death or short gut patients are less likely to develop nutrient deficiencies.
k.
syndrome resulting. The surgeon who sees a patient with In addition, because the operation does not bypass any
a picture of small bowel obstruction after previous gastric intestinal segments, there are no potential spaces for
oo
bypass is obligated to prove that patient does not have internal herniation. Sleeve gastrectomy also leaves open
an internal hernia and strangulation obstruction. Early the possibility of revision to a full BPD with DS or a gastric
operative intervention in this setting is the standard of bypass if weight loss is inadequate.
yb
care, whereas conservative therapy with nasogastric suction The indications, patient preparation, and positioning
and intravenous fluids may allow strangulation to proceed are similar to gastric bypass. Eligible patients should
to gangrene of the bowel. have a BMI of 40 or a BMI of 35 with obesity-related
Marginal ulcers are of unclear etiology; hence pre- comorbidity, have failed prior attempts at weight loss,
er
vention is difficult. Use of absorbable suture, treatment and have undergone psychologic evaluation.
for Helicobacter pylori, avoidance of nonsteroidal and
OPERATIVE STEPS
rg
unremitting epigastric pain, unrelated to eating. Prophy- induction. The patient is positioned supine or split leg
laxis versus this problem among patients with LRYGB, (French position). A footboard is used to allow for steep
in terms of chronic use of proton pump inhibitor (PPI) reverse Trendelenburg.
://
medications, is controversial. PPI therapy is effective in Port placement is also similar to gastric bypass, although
treating marginal ulcers. Smoking also increases the risk fewer 12-mm ports are needed because stapling is done
tp
of marginal ulcer.38 Most patients will heal marginal ulcers through only one or two ports. Four or five ports are used,
without sequelae. However, persistent or chronic ulcers as well as a stab incision for the liver retractor. We use
may result in obstruction or gastrogastric fistula to the one 15-mm port and one 12-mm port, with the remainder
ht
defunctionalized stomach. These patients require operative being 5 mm. The 15-mm site is later used to extract the
revision. In most cases, re-creation of the gastrojejunostomy specimen (Fig. 63.13).
and resection of the ulcer and any fistula is performed. Retraction of the left lobe of the liver provides exposure
However, in selected cases of chronic or repeated ulcers, of the proximal stomach and the gastroesophageal junc-
takedown of the gastrojejunostomy and performance of tion. The patient is placed in reverse Trendelenburg
a gastrogastrostomy, with gastric narrowing (converting position. The pylorus is identified, and a position on the
to sleeve gastrectomy) or without gastric narrowing, may greater curvature of the stomach 4 cm from the pylorus
be the best option. is selected for the initial dissection.
The most common adverse long-term result after LRYGB This 4-cm margin preserves the antral pump mechanism.
is weight regain. For approximately one-third of patients, Alternatively, dissection is begun on the greater curvature
this is an unfortunate development. For some patients, in the mid-body and carried back to this point. Entrance
situational stresses result in weight gain. Others never to the lesser sac is easier in this position.
establish a firm enough change in eating and exercise Ultrasonic shears or a bipolar energy device is used to
habits to sustain the initial weight loss produced by the enter the lesser sac and divide the gastroepiploic arcade
operation. Options for revisional surgery include perform- proximally from the point of entry along the entire greater
ing further restrictive measures endoscopically or surgically curvature of the stomach, including the short gastric vessels.
or adding a malabsorptive component to the anatomy. Dissection stays very close to the greater curve, dividing
Revisional surgery will be discussed in more detail later. gastroepiploic vessels attached to it (Fig. 63.14). This
Operations for Morbid Obesity CHAPTER 63 743
t
ne
upper quadrant port, and the sleeve resection is begun at
a point 4 to 5 cm proximal to the pylorus. Black (2.3 mm
k.
closed height) or green (2 mm) 60-mm stapler cartridges
are used for the initial fires, transitioning to blue (1.5 mm)
oo
as the tissue thickness decreases at the fundus. We feel
that matching the staple height to the thickness of the
tissue aids in hemostasis. Care is taken not to staple too
yb
close to the incisura angularis because stenosis is common
at this level. Equal portions of anterior and posterior wall
are taken to avoid twisting or spiraling of the staple line.
This is facilitated by stretching the greater curvature as
er
gastrectomy.
Staple line reinforcement is controversial. Approximately
80% of surgeons use some form of reinforcement, with
su
FIGURE 63.14 Dividing tissue along greater curvature. 60% of those using absorbable buttressing material, and the
remainder oversewing the staple line.42 A meta-analysis in
2016 found that use of buttressing material was associated
://
minimizes the amount of fat attached to the gastrectomy with a lower risk of staple line hemorrhage, but there was
specimen and facilitates easier extraction. Division of the not a significant decrease in the rate of leaks. Oversewing
tp
phrenoesophageal ligament completes mobilization of the the staple line showed no clear advantages.43
fundus and allows for detection of a hiatal hernia. If one The integrity of the staple line may be tested with an
is discovered, the distal esophagus is freed of mediastinal air leak or instillation of 50 to 100 mL of methylene blue
ht
attachments and brought into the abdomen. The crura in saline solution. Intraoperative endoscopy may also be
are approximated posteriorly with permanent suture. used, which further facilitates the detection of stenosis or
Dissection of posterior gastric adhesions to the body of intraluminal bleeding. Although common, there are no
the pancreas and retroperitoneum completes mobilization current data supporting the ability of routine intraoperative
of the body of the stomach. leak testing to detect leaks.44
Prior to stapling, any previously placed orogastric tubes Some surgeons perform omentoplasty, suturing the
are removed and the anesthesiologist inserts a bougie. gastric omentum to the staple line or buttressing material
With the assistance and guidance of the surgeon, it is in an effort to reproduce the natural orientation of the
positioned along the lesser curve and directed to the stomach, reduce spiraling of the sleeve, and further buttress
pylorus. The size of the bougie dilator varies from 32 the staple line. Data are lacking to support this practice.
to 50 French, but 36 French is the most common size.42 Drains are not routinely used. The specimen is removed
I use a large-bore gastric lavage tube (Ewald), which through one of the port sites. Use of a 15-mm port site
measures 34 French and is excellent for subsequent leak generally eliminates the need to dilate the site. We close
testing with methylene blue. It is important when using all port sites 12 mm in size or larger. We generally place
a smaller-diameter bougie to leave a 1- to 2-mm distance two sutures in the 15-mm port site and a single suture in
between the bougie and the stapler (Fig. 63.15). This is the 12-mm site, using a transfascial suture passer. Closure
evaluated by the surgeon with the tip of a grasper. An of the 12-mm sites is optional when noncutting trocars
initial linear stapler load is introduced through a right are used.
744 SECTION II Stomach and Small Intestine
t
as 3.8%. Reoperation rates were 1.6%.50
ne
The most common complications that occur after LSG
are staple line leaks and staple line bleeding. Staple line
bleeding is generally reported in the 1% range, with the
k.
use of buttress material decreasing the rate from 1.0% to
0.75%. Staple line leak rate is similarly reported in the
oo
1% range, with buttressing actually potentially increasing
the incidence of leakage from 0.65% to 0.96%.51
Stenosis of the staple line is the only other major com-
yb
FIGURE 63.16 Final configuration of sleeve gastrectomy. (From monly reported complication after LSG. Its incidence is
Poirier RF. Complications of bariatric surgery. In: Adams JG, in the 1% to 2% range generally, with reoperation being
Barton ED, Collings J, eds. Emergency Medicine. Philadelphia: required if dilation does not produce adequate relief of
obstructive symptoms.
er
Saunders; 2008.)
Adjustment to the LSG anatomy can, for some patients,
be difficult. We have observed a small percentage of
rg
resolution of comorbid medical problems. Weight loss After a patient is more than a few months out from
is slightly lower than that seen after gastric bypass but an LSG, the incidence of later complications is relatively
greater than after lap band.27 Excess weight loss varying low. This is especially true when compared with LRYGB,
://
from 51% to 71% at 1 year has been reported in the which has long-term problems of bowel obstruction and
literature after LSG.42,45,46 marginal ulcers. New-onset GERD may arise after LSG,
tp
LSG has been shown to improve comorbid medical with an estimated incidence of 8%.47
problems in a comparable way as gastric bypass. Hutter Data on weight regain after LSG are just now being
et al. 27 showed that resolution of comorbid medical appreciated with most series nearing their 10-year mark
ht
problems for LSG was not significantly different than from the start of performing cases. The early data do
seen after LRYGB, with the notable exception of GERD. not suggest the incidence will be that much higher than
GERD is apparently the Achilles heel of the LSG. One LRYGB.
report suggested that patients with preoperative GERD
did significantly less well after LSG than patients without
GERD, even to the point where morbidity and reoperation
DUODENAL SWITCH
rates were affected.47 There have been two large consensus Marceau24 and Hess25 performed and described this opera-
conferences among experts in the field regarding sleeve tion in response to the high marginal ulceration rate seen
gastrectomy, with results published in the literature. The both after the BPD operation popularized by Scopinaro23
first consensus conference42 confirmed the operation was and his followers. The BPD involved the creation of a
effective for producing weight loss of from 59% excess partial distal gastrectomy, leaving a generous proximal
weight loss at 1 year to 50% at 6 years. Mortality was 0.33% gastric pouch. The ileum, at a point 200 cm proximal
and serious complications of leak (1.1%), bleeding (1.8%), to the ileocecal valve, was divided and anastomosed to
and stenosis (0.9%) occurred rarely. The second consensus the gastric pouch. The remaining proximal bowel was
conference showed that experts felt GERD needed to anastomosed to the ileum either 50 or 100 cm proximal
be thoroughly evaluated prior to sleeve gastrectomy but to the ileocecal valve, depending on the demographics
did not necessarily represent a total contraindication. It of the patient population. In northern Italy, where the
also showed experts were much more likely to consider Italian population ate more protein, Scopinaro created
Operations for Morbid Obesity CHAPTER 63 745
the anastomosis at 50 cm, whereas for southern Italians, type of procedure in preference to other more popular
who ate more pasta and less protein, he created a 100-cm ones could include those who have failed a restrictive
common channel. Marceau and Hess both used the same procedure, are particularly large, or have severe metabolic
distal bowel design but advocated the longer common syndrome or hyperlipidemia or diabetes. Patients who
channel. However, for the DS the stomach was decreased wish to be less restricted in terms of their eating but
in size by creating a sleeve of the lesser curvature of who understand will have more bowel activity due to the
the stomach, which was the first performance of sleeve malabsorption component of the operation would also
gastrectomy. The duodenum was instead divided in the be good candidates for DS.
distal first portion, to which the distal ileum was connected Malabsorptive operations require careful follow-up
to create the shortened alimentary tract. to monitor for potential nutritional complications. DS
The performance of the DS laparoscopically was and patients also optimally need significant vitamin and
remains challenging. Initial reports by Gagner’s group21 mineral supplementation annually, which can represent
showed an increased morbidity and mortality for these a significant out-of-pocket cost. The patient should be
patients, who were typically heavier with correspond- aware of this potential expense and be willing and able to
ingly more associated medical problems, warranting the meet it.
performance of a malabsorptive operation. The separation
of the sleeve gastrectomy as a first-stage procedure prior OPERATIVE STEPS
to later performance of the malabsorptive component The operation begins with the creation of the sleeve
t
of the operation led to the recognition by patients and gastrectomy. Ports are placed in the mid to upper abdomen,
ne
surgeons that the sleeve gastrectomy operation was an as well as a liver retractor placed in the epigastric region.
effective stand-alone procedure. A sleeve gastrectomy is first performed, as described
Variations on the DS have been described that include previously in this chapter. After the sleeve gastrectomy
k.
longer length alimentary tract limbs and longer common is completed, the duodenum is carefully dissected at a
channels. Fig. 63.17 shows the most commonly performed location several centimeters distal to the pylorus. An instru-
oo
version of the DS. Indications for performance of DS ment such as a Harmonic scalpel is helpful for dividing
include the same parameters as for any standard bariatric tissue hemostatically. Creation of a tunnel underneath
and metabolic operation. DS has not proven to be a the duodenum is then followed by placement of a stapler
yb
popular option for patients, whether done open in the across the duodenum, which staples and divides it securely.
past or laparoscopically as at present. Current statistics Attention is now turned to the distal ileum and the
show that DS represents no more than approximately ileocecal area. The distal ileum is then measured backward
1% of bariatric operations performed annually in the from the cecum, and the desired distance of the alimentary
er
United States. Patients who may be candidates for this tract, usually 250 cm but for some surgeons 300 cm, is
determined. At that point the ileum is divided with a
rg
laparoscopic cholecystectomy when performed with LRYGB NOT YET STANDARDIZED PROCEDURES
is quite low.52
t
Patients who undergo laparoscopic DS have similar risks bypass. The operation confers slightly more malabsorption
ne
for early postoperative complications as do those who than LRYGB but is not associated with any significant
undergo LRYGB. The potential for staple line leak, bleed- incidence of protein-calorie malnutrition. Patients with
ing, and anastomotic stenosis and early obstruction are metabolic comorbidities appear to be well served with
k.
comparable and are seen in similar incidence. Late small this operation. Previous bowel resection, malabsorptive
bowel obstruction is as frequent after this procedure as bowel diseases, and significant liver disease may all be
oo
after LRYGB, which has an incidence of from 3% to 7% contraindications to this operation.
in most series. Most surgeons will consider simultaneous
cholecystectomy when performing DS. However, if the OPERATIVE STEPS
yb
gallbladder is not removed, the incidence of gallstone The procedure begins with creation of a proximal gastric
formation postoperatively is significantly increased due pouch that is longer than that created for a gastric bypass.
to the decrease in bile salt pool. A high incidence of The lesser curvature of the stomach is used in its entirety
cholecystectomy is seen in this patient population long- for the pouch. Adequate length of the pouch to prevent
er
t
complication rates for the procedure are generally without endoscopy in the office are currently being trialed
ne
similar to those seen after LRYGB. Kim et al. have shown in the United States.
good efficacy for the procedure.56 A recently published
12-year experience with this operation for 1200 patients OTHER ENDOSCOPIC SUTURING PROCEDURES
k.
showed a 70% long-term excess weight loss and mortality The potential for performing a standard bariatric and
of only 0.16%, with no reported long-term metabolic metabolic operative procedure using a totally endoscopic
oo
complications.57 approach has now arrived. Over the past decade, a number
Currently there are strong advocates for the adoption of of endoscopic procedures have been trialed to reproduce
this procedure as a standardized operation for metabolic the restriction or malabsorption of standard bariatric
yb
and bariatric surgery, but as of yet the accumulated data surgical procedures. Unfortunately, most of these have been
have not swayed the general opinion of experts and abandoned or the instrumentation used to perform them
societies sufficiently to include it as a standard procedure. removed from the market. Restrictive operative procedures
However, the prognosis for its inclusion is improving with using the EndoCinch procedure proved ineffective at
er
improving accumulated data. maintaining suture integrity, and it has been removed
from the market. Diversion of food from duodenal absorp-
SINGLE ANASTOMOSIS DUODENAL SWITCH
rg
anastomosis 200 cm proximal to the ileocecal valve has The AspireAssist device, FDA approved, is essentially a
yielded excellent weight loss (>90% excess weight at 5 gastrostomy tube that is opened postprandially to allow
years) and excellent control of type 2 diabetes (HgbA1c gastric contents to drain out.64
://
<6.0 in 70% of insulin-dependent diabetics and 84% of The most promising of all the current endoscopic
those on oral medications at 5 years).58 These results are procedures is the endoscopic sleeve gastrectomy, or ESG
tp
from a single center, and review of the operation by the procedure. Performed by using the Overstitch device, the
ASMBS Clinical Issues Committee concluded there was procedure reproduces approximately the configuration of
inadequate evidence to recommend the operation as a a sleeve gastrectomy by performing a sutured gastroplasty
ht
LRYGB. Both are associated with excellent durable weight 22. Brethauer SA, Hammel JP, Schauer PR. Systematic review of sleeve
loss and resolution of associated medical comorbidities. gastrectomy as staging and primary bariatric procedure. Surg Obes
Relat Dis. 2009;5:469-475.
These operations are furthermore done with increasing 23. Scopinaro N, Gianetta E, Civalleri D, Bonalumi U, Bachi V. Bilio-
safety, such that they are associated with lower morbidity pancreatic bypass for obesity. Initial experience in man. Br J Surg.
and mortality than all intraabdominal operations except 1979;66:618-620.
appendectomy. The LAGB is rapidly losing popularity 24. Marceau P, Hould FS, Simard S, et al. Biliopancreatic diversion with
a duodenal switch. World J Surg. 1998;22:947-954.
due to poor long-term efficacy, whereas the DS and other 25. Hess DS, Hess DW. Biliopancreatic diversion with a duodenal switch.
malabsorptive operations are chosen infrequently by Obes Surg. 1998;8:267-282.
patients. Finally, new endoscopic and minimally invasive 26. Flum DR, Dellinger EP. Impact of gastric bypass operation on survival:
procedures offer potentially less invasive mechanisms of a population-based analysis. J Am Coll Surg. 2004;199:543-551.
producing weight loss, although their long-term efficacy 27. Hutter MM, Schirmer BD, Jones DB, et al. First report of the American
College of Surgeons Bariatric Surgery Center Network: laparoscopic
and durability still are not established. sleeve gastrectomy has morbidity and effectiveness positioned between
the band and the bypass. Ann Surg. 2011;254:410-422.
REFERENCES 28. Morton JM, Garg T, Nguyen N. Does hospital accreditation impact
bariatric surgery safety? Ann Surg. 2014;260:504-509.
1. Mason EE, Ito C. Gastric bypass in obesity. Surg Clin North Am. 29. O’Brien PE, Dixon JB, Brown W, et al. The laparoscopic adjustable
1969;47:1345-1351. gastric banding (Lap-Band): a prospective study of medium-term
2. Mason EE, Doherty C, Cullen JJ, Scott D, Rodriguez EM, Maher effects on weight, health, and quality of life. Obes Surg. 2002;12:652-660.
JW. Vertical gastroplasty: evolution of vertical banded gastroplasty. 30. Carelli AM, Youn HA, Kurian MS, Ren CJ, Fielding GA. Safety of
t
World J Surg. 1998;22:919-924. the laparoscopic adjustable gastric band: 7-year data from a U.S.
ne
3. Griffen WO, Young VL, Stevenson CC. A prospective comparison of center of excellence. Surg Endosc. 2010;24:1819-1823.
gastric and jejunoileal bypass procedures for morbid obesity. Ann 31. Dixon JB, O’Brien PE, Playfair J, et al. Adjustable gastric banding and
Surg. 1977;186:500-509. conventional therapy for type 2 diabetes: a randomized controlled
4. Deitel M. Overview of operations for morbid obesity. World J Surg. trial. JAMA. 2008;279:316-323.
k.
1998;22:913-918. 32. Allen JW. Laparoscopic gastric band complications. Med Clin North
5. Griffen WO, Bivins MA, Bell RM, Jackson KA. Gastric bypass for Am. 2007;9:485-497.
morbid obesity. World J Surg. 1981;5:817-822. 33. National Institutes of Health Consensus Conference. Gastrointestinal
oo
6. Sugerman HJ, Wolfe LG, Sica DA, Clore JN. Diabetes and hypertension surgery for severe obesity, Consensus Development Conference
in severe obesity and effects of gastric bypass-induced weight loss. Panel. Ann Intern Med. 1991;115:956-961.
Ann Surg. 2003;237:751-758. 34. Kindel TL, Oleynikov D. The improvement of gastroesophageal reflux
yb
7. Sugerman HJ, Felton WL 3rd, Sismanis A, Kellum JM, DeMaria EJ, disease and Barretts after bariatric surgery. Obes Surg. 2016;26:718-720.
Sugerman EL. Gastric surgery for pseudotumor cerebri associated 35. Schauer PR, Bhatt DL, Kiwan JP, et al. Bariatric surgery versus
with severe obesity. Ann Surg. 1999;229:634-642. intensive medical therapy for diabetes—3-year outcomes. N Engl J
8. Sugerman HJ, Sugerman EL, Wolfe L, Kellum JM Jr, Schweitzer MA, Med. 2014;370:2002-2013.
er
DeMaria EJ. Risks and benefits of gastric bypass in morbidly obese 36. Mehaffey JH, LaPar DJ, Clement KC, et al. 10-Year outcomes after
patients with severe venous stasis disease. Ann Surg. 2001;234:41-46. Roux-en-Y gastric bypass. Ann Surg. 2016;264:121-126.
9. Pories WJ, MacDonald KG, Flickinger EG, et al. Is type II diabetes 37. Schirmer B, Erenoglu C, Miller A. Flexible endoscopy in the manage-
rg
mellitus (NIDDM) a surgical disease? Ann Surg. 1992;215:633-643. ment of patients undergoing Roux-en-Y gastric bypass. Obes Surg.
10. Christou NV, Sampalis JS, Liberman M, et al. Surgery decreases 2002;12:634-638.
long-term mortality, morbidity, and health care use in morbidly 38. Azagury DE, Abu Dayyeh BK, Greenwalt IT, Thompson CC. Marginal
su
obese patients. Ann Surg. 2004;240:416-424. ulceration after Roux-en-Y gastric bypass surgery: characteristics, risk
11. Sampalis JS, Liberman M, Auger S, Christou NV. The impact of factors, treatment, and outcomes. Endoscopy. 2011;43:950-954.
weight reduction surgery on health-care costs in morbidly obese 39. Marceau P, Biron S, Bourque RA, Potvin M, Hould FS, Simard S.
patients. Obes Surg. 2004;14:939-947. Biliopancreatic diversion with a new type of gastrectomy. Obes Surg.
://
gastric bypass surgery. N Engl J Med. 2007;357(8):753-761. the super-super obese patient. Obes Surg. 2003;13:861-864.
14. Wittgrove AC, Clark WG, Tremblay LJ. Laparoscopic gastric bypass, 41. D’Hondt M, Venneste S, Pottel H, Devriendt D, Van Rooy F,
Roux en-Y: preliminary report of five cases. Obes Surg. 1994;4: Vansteenkiste F. Laparoscopic sleeve gastrectomy as a single-stage
ht
353-357. procedure for the treatment of morbid obesity and the resulting
15. Schauer PR, Ikramuddin S, Gourash W, Ramanathan R, Luketich quality of life, resolution of comorbidities, food tolerance, and 6-year
J. Outcomes after laparoscopic Roux-en-Y gastric bypass for morbid weight loss. Surg Endosc. 2011;25:2498-2504.
obesity. Ann Surg. 2000;232:515-529. 42. Gagner M, Deitel M, Erickson AL, Crosby RD. Survey on laparoscopic
16. Nguyen NT, Goldman C, Rosenquist CJ, et al. Laparoscopic versus sleeve gastrectomy (LSG) at the Fourth International Consensus
open gastric bypass: a randomized study of outcomes, quality of Summit on Sleeve Gastrectomy. Obes Surg. 2013;23:2013-2017.
life, and costs. Ann Surg. 2001;234:279-289. 43. Wang Z, Dai X, Xie H, Feng J, Li Z, Lu Q. The efficacy of staple line
17. Santry H, Gillen DL, Lauderdale DS. Trends in bariatric surgical reinforcement during laparoscopic sleeve gastrectomy: a meta-analysis
procedures. JAMA. 2005;294:1909-1917. of randomized controlled trials. Int J Surg. 2016;25:145-152.
18. Belachew M, Legrand MJ, Defechereux TH, Burtheret MP, Jacquet N. 44. Bingham J, Lallemand M, Baron M, et al. Routine intraoperative
Laparoscopic adjustable silicone gastric banding in the treatment of leak testing for sleeve gastrectomy: is the leak test full of hot air?
morbid obesity. A preliminary report. Surg Endosc. 1994;8:1354-1356. Am J Surg. 2016;211:943-947.
19. Dixon JB, O’Brien PE. Laparoscopic adjustable gastric banding: 45. Wang S, Li P, Sun XF, Ye NY, Xu ZK, Wang D. Comparison between
outcomes. In: Schauer PR, Schirmer BD, Brethauer SA, eds. Minimally laparoscopic sleeve gastrectomy and laparoscopic adjustable gastric
Invasive Bariatric Surgery. New York: Springer; 2007:189-196. banding for morbid obesity: a meta-analysis. Obes Surg. 2013;23:980-986.
20. Aarts EO, Dogan K, Koehestanie P, Aufenacker TJ, Janssen IM, 46. Peterli R, Wolnerhanssen BK, Vetter D, et al. Laparoscopic sleeve
Berends FJ. Long-term results after laparoscopic adjustable gastric gastrectomy versus Roux-en-Y gastric bypass for morbid obesity—3-year
banding: a mean fourteen-year follow-up study. Surg Obes Relat Dis. outcomes of the prospective randomized Swiss multicenter bypass
2014;4:633-640. or sleeve study (SM-BOSS). Ann Surg. 2017;265:466-473.
21. Ren CJ, Patterson E, Gagner M. Early results of laparoscopic 47. DuPree CE, Blair K, Steele SR, Martin MJ. Laparoscopic sleeve
biliopancreatic diversion with duodenal switch: a case series of 40 gastrectomy in patients with preexisting gastroesophageal reflux
consecutive patients. Obes Surg. 2000;10:514-524. disease: a national analysis. JAMA Surg. 2014;149:328-334.
Operations for Morbid Obesity CHAPTER 63 749
48. Gagner M, Hutchinson C, Rosenthal R. Fifth international consensus gastric bypass: technique, results, and long-term follow-up in 1200
conference: current status of sleeve gastrectomy. Surg Obes Relat Dis. patients. Obes Surg. 2017;27(5):1153-1167.
2016;12:750-756. 58. Sanchez-Pernaute A, Angel Rubio M, Cabrerizo L, Ramos-Levi A,
49. Cho JM, Kim HJ, Lo Menzo E, Park S, Szomstein S, Rosenthal RJ. Pérez-Aguirre E, Torres A. Single-anastomosis duodenoileal bypass
Effect of sleeve gastrectomy on type 2 diabetes as an alternative with sleeve gastrectomy (SADI-S) for obese diabetic patients. Surg
treatment modality to Roux-en-Y gastric bypass: systemic review and Obes Relat Dis. 2015;11:1092-1098.
meta-analysis. Surg Obes Relat Dis. 2015;11:1273-1280. 59. Kim J, ASMBS Clinical Issues Committee. ASMBS statement on
50. Young MT, Gebhart A, Phelan MJ, Nguyen NT. Use and outcomes single anastomosis duodenal switch. Surg Obes Relat Dis. 2016;12:
of laparoscopic sleeve gastrectomy vs. laparoscopic gastric bypass: 944-945.
analysis of the American College of Surgeons NSQIP. J Am Coll Surg. 60. Bazerbachi F, Vargas Valls EJ, Abu Dayyeh BK. Recent clinical results
2015;220:880-885. of endoscopic bariatric therapies as an obesity intervention. Clin
51. Berger ER, Clements RH, Morton JM, et al. The impact of different Endosc. 2017;50:42-50.
surgical techniques on outcomes in laparoscopic sleeve gastrectomies. 61. Imaz I, Martinez-Cervell C, Sendra-Gutierrez JM, Gonzalez-Enriquez
First report from the Metabolic and Bariatric Surgery Accredita- J. Safety and effectiveness of intragastric balloon for obesity. A meta
tion and Quality Improvement Program (MBSAQIP). Ann Surg. analysis. Obes Surg. 2008;7:841-846.
2016;264:464-473. 62. Dastis NS, Francois E, Deviere J. Intragastric balloon for weight loss:
52. Kim JJ, Schirmer B. Safety and efficacy of simultaneous cholecys- results in 100 individuals followed for at least 2.5 years. Endoscopy.
tectomy at the time of Roux-en-Y gastric bypass. Surg Obes Relat Dis. 2009;41:575-580.
2009;5:48-53. 63. ASGE Bariatric Endoscopy Task Force and ASGE Technology Commit-
53. Phillipe A, Topart MD, Becouam G. Revision and reversal for tee, Abu Dayyeh BK, Kumar N, et al. ASGE Bariatric Endoscopy Task
biliopancreatic diversion for excessive side effects or ineffective Force systematic review and meta-analysis assessing the ASGE PIVI
weight loss: a review of the current literature on indications and thresholds for adopting endoscopic bariatric therapies. Gastrointest
t
procedures. Surg Obes Relat Dis. 2015;11:965-972. Endosc. 2015;82:425-438.
ne
54. Rutledge R. The mini-gastric bypass: experience with the first 1,274 64. Thompson CC, Abu Dayyeh BK, Kushner R, et al. Percutaneous
cases. Obes Surg. 2001;11:276-280. gastrostomy device for the treatment of class II and Class III
55. Lee WJ, Ser KH, Lee YC, Tsou JJ, Chen SC, Chen JC. Laparoscopic obesity: results of a randomized controlled trial. Am J Gastroenterol.
Roux-en-Y vs. mini-gastric bypass for the treatment of morbid obesity: 2017;112(3):447-457.
k.
a 10-year experience. Obes Surg. 2012;22:1827-1834. 65. Lopez-Nava G, Sharaiha RX, Neto M, et al. Endoscopic sleeve
56. Lee WJ, Lin YH. Single-anastomosis gastric bypass (SAGB): appraisal gastroplasty for obesity: a multicenter study of 242 patients with 18
of clinical evidence. Obes Surg. 2014;24:1749-1756. months follow-up. Gastroenterology. 2016;150(4 suppl 1):S26.
oo
57. Carbajo MA, Luque-de-Leon E, Jiminez JM, Ortiz-de-Solórzano J,
Pérez-Miranda M, Castro-Alija MJ. Laparoscopic one-anastomosis
yb
er
rg
su
://
tp
ht
CHAPTER
FOREIGN BODY INGESTION it will likely pass spontaneously. However, sharp, long
(>6 cm), or large (wider than 2.5 cm) objects in the
Foreign body ingestion is an unusual occurrence. Foreign stomach should be removed endoscopically. Once in the
body ingestion may be intentional or unintentional. There small bowel, even sharp objects can be watched vigilantly
are several ways to classify and consider foreign body via imaging. These patients must be observed as inpatients
ingestion. One way is by type of object (size and shape). because perforation occurs in 15% to 35% of cases.1
Foreign body ingestion may also be considered according Signs of obstruction or perforation indicate emergency
t
to age groups. It is well established that 80% to 90% of operation.
ne
ingested foreign objects will pass through the gastro-
intestinal (GI) tract without intervention. Endoscopic MAGNETS
removal is required in 10% to 20% of cases and about Several foreign body ingestion types in children require
k.
1% will require surgical intervention.1 special attention. Ingestion of multiple magnets is rare
but dangerous. This foreign body ingestion puts a child
FOREIGN BODY INGESTION IN CHILDREN
oo
at risk for perforation and/or fistula formation.5 A solitary
Witnessed foreign body ingestion in small children is magnet will almost always pass spontaneously. Multiple
generally addressed rapidly. In cases where the ingestion magnets seen on radiographs in the esophagus or stomach
yb
was both unintentional and not realized, clinical presenta- should be removed endoscopically. Multiple magnets in
tion will depend on where the foreign object becomes the small bowel in an asymptomatic child may be followed
lodged. For example, in the pharynx, symptoms are usually with serial plain films. Magnets in adjacent bowel loops
immediate and include choking and hypersalivation. In or a single magnet with another metallic foreign object
er
the esophagus, dysphagia or odynophagia usually occur may erode the adjacent loops, resulting in perforation or
early after ingestion. The treatment is urgent endoscopic fistula formation (Fig. 64.1). These cases require surgical
rg
Hepatology, and Nutrition.2 Foreign objects in the stomach Colorful laundry pods are new items causing problems
most commonly pass into the small bowel. However, if a in small children. The pods look like candy. In this patient
foreign object lodges in the stomach, it may cause nausea group, 4% to 5% require hospitalization. 6 Ingestion
://
and/or vomiting but also may remain without presenting may cause metabolic acidosis with an increased anion
symptoms for a considerable time. Foreign objects in the gap. Symptoms may include nausea, vomiting, and som-
tp
small bowel frequently pass into the colon. Objects in the nolence. These patients may require endotracheal intuba-
small bowel may cause injury at any point, but often tion and ventilator support. At least one death has been
become lodged in the distal ileum due to its small caliber. reported.7
ht
ABSTRACT
Foreign bodies, including bezoars of the stomach and
small bowel, are a fairly common problem encountered
by gastroenterologists and surgeons. Most resolve without
intervention; however, it is important for surgeons to
understand which circumstances require intervention.
Foreign body ingestions and their management vary based
on the specific object ingested as well as the patient
population involved (adult or pediatric). The majority of
ingested foreign objects will pass through the gastro-
intestinal tract without intervention. Ten to 20% of cases
will require endoscopic removal, and about 1% will require
surgical intervention. Similarly, the management of gastric
bezoars, which are collections of undigested materials,
varies based on the composition of the bezoar. Most cases
require endoscopic fragmentation, and very few require
surgical intervention.
t
KEYWORDS
ne
Perforation, bowel obstruction, foreign body, bezoar
k.
oo
yb
er
rg
su
://
tp
ht
Foreign Bodies and Bezoars of the Stomach and Small Intestine CHAPTER 64 751
t
ne
k.
oo
FIGURE 64.1 A 7-year-old boy who had two groups of magnets FIGURE 64.2 Three-dimensional volume rendered computed
surgically removed from the small bowel. Radiograph reveals a tomography image highlighting the burden of foreign material in
central gap between two magnet conglomerates, suggesting the stomach, ascending colon, and rectosigmoid (arrows), most
entrapment of the bowel wall. (With kind permission from of which was obscured on plain radiography. (From Esterson YB,
yb
Guelfguat M, Kaplinskiy V, Reddy SH, DiPoce J. Clinical Vihas P, Nicastro J, Friedman B. Plain radiography may
guidelines for imaging and reporting ingested foreign bodies. AJR underestimate the burden of body packer ingestion: a case
Am J Roentgenol. 2014;203:37.) report. Clin Imaging. 2017;44:57-60.)
er
removal. Batteries distal to the esophagus will most fre- years as a result of increased border security since 9/11.12
rg
quently pass spontaneously. Most commonly the drugs are heroin, cocaine, and
amphetamines. Body packers may use antimotility
ADOLESCENTS drugs (loperamide) during transit and promotility agents
su
In adolescents, esophageal food impaction is most (metoclopramide) once they arrive at their destination.
common.11 These may be precipitated by a sudden event Many drug packers are asymptomatic. In symptomatic
like a motor vehicle accident. Intentional ingestions may patients, symptoms range from nausea and vomiting to
://
be secondary to mental illness or may occur for secondary abdominal pain and obstruction. Computed tomography
gain. One well-recognized example of the former is (CT) scanning is most sensitive for diagnosis; however
tp
trichophagia, the chronic ingestion of their own hair, plain films may be used to monitor the progress of packets
usually seen in teenage girls. This may result in a tricho- (Fig. 64.2).
bezoar, which is covered in detail later in this chapter. In most cases, patients can be given a bowel regimen
ht
t
base of most partial denture prosthetics, can also make
ne
From Pfau PR, Ginsberg GG. Foreign bodies and bezoars. In: Feldman
diagnosis challenging. Metallic struts frequently allow M, Friedman LS, Sleisenger MH, et al, eds. Gastrointestinal and Liver
identification on imaging. Patients may present with Disease. Philadelphia: Saunders; 2002:386.
nonspecific symptoms or may be unable to provide an
k.
accurate history. The most common site of impaction
is the esophagus, for which endoscopy is the treat- anticholinergics and opiates. Gastric surgery, most com-
oo
ment of choice.17 Surgical intervention is required for monly gastrectomy, vagotomy, and restrictive bariatric
dentures that perforate, obstruct, or fail to pass with procedures including Roux-en-y gastric bypass and sleeve
observation. gastrectomy, may also contribute to the development of
Noningested Foreign Bodies. The modern use of biliary
yb
gastric bezoars.
and pancreatic stents represents foreign bodies intention- Bezoars are characterized by their composition as
ally placed by physicians. The use of stents for obstructive phytobezoars, pharmacobezoars, trichobezoars, or lacto-
biliary and pancreatic disorders has increased dramatically bezoars (Table 64.1). Phytobezoars are most common
er
over the past few decades. Stents may be metallic or plastic. and develop from poorly digested fibrous plant or fruit
Stents may be permanent in patients with a short life material (Fig. 64.3). One subtype of phytobezoars that is
rg
expectancy but are more frequently removed when the particularly hard and difficult to treat forms from the
disorder has resolved. Distal stent migration occurs in 5% persimmon fruit and is named the diospyrobezoar.21
to 10% of cases. 18 Most such cases can be managed Pharmacobezoars are composed of medications, espe-
su
expectantly, because the stents pass through the GI tract. cially those with extended release enteric coatings designed
However, small bowel obstruction or perforation as a to resist digestion. Trichobezoars are composed of human
result of the stent mandates surgical intervention. Endo- hair, which is difficult to digest as a result of the enzyme-
://
scopic removal of proximally migrated biliary stents is resistant cuticle. These are most commonly found in
successful in more than 80% of patients.19 With distal young female patients with psychiatric illnesses.
tp
A B
t
ne
k.
C D
oo
FIGURE 64.3 (A) Phytobezoar distal to gastrojejunal anastomosis. (B) Patent mesocolic window. (C) Endoscopy showing endoscopic
fragmentation of bezoar using biopsy forceps. (D) Patent Roux limb after endoscopic fragmentation of bezoar. (From Powers WF,
Miles DR: Phytobezoar causing small bowel obstruction seven years after laparoscopic Roux-en-Y bypass. Surg Obes Related Dis.
2011;7:e3-e5.)
yb
er
MANAGEMENT 2. Kramer RE, Lerner DG, Lin T, et al. Management of ingested foreign
bodies in children: a clinical report of the NASPGHAN Endoscopy
Once a diagnosis is made, treatment may consist of medical Committee. J Pediatr Gastroenterol Nutr. 2015;60(4):562-574.
rg
management, endoscopic removal, or surgery. Bezoars 3. Fontane E. Ingestion of concentrated laundry detergent pods.
that are minimally symptomatic may initially be treated J Emerg Med. 2015;49(1):e37-e38.
with chemical dissolution. Treatment with cola, cellulase, 4. Peters NJ, Mahajan JK, Bawa M, Chabbra A, Garg R, Rao KLN.
su
The addition of promotility agents such as meto- 6. Valdez AL, Casavant MJ, Spiller HA, Chounthirath T, Xiang H,
clopramide (Reglan) has been found to reduce time to Smith GA. Pediatric exposure to laundry detergent pods. Pediatrics.
tp
2014;134(6):1127-1135.
dissolution.25 On occasion, bezoars that are only partially 7. Hernández AR. Infant dies after ingesting detergent pod. Orlando
digested may pass into the small bowel and cause obstruc- Sentinel. 2013.
tion, which often needs to be treated with surgery.
ht
8. Chang YJ, Chao HC, Kong MS, Lai MW. Clinical analysis of disc
Most gastric bezoars require endoscopic therapy to battery ingestion in children. Chang Gung Med J. 2004;27(9):673-677.
9. Litovitz T, Whitaker N, Clark L, White NC, Marsolek M. Emerging
mechanically fragment the bezoar. A water jet is commonly battery-ingestion hazard: clinical implications. Pediatrics.
used, followed by removal with suction, snare, or forceps. 2010;125(6):1168-1177.
Success rates approach 80% to 90% with endoscopic 10. Centers for Disease Control and Prevention (CDC). Injuries from
therapy alone.23 When chemical dissolution and endoscopic batteries among children aged <13 years—United States, 1995–2010.
therapy fail, surgery is required. Associated complications MMWR Morb Mortal Wkly Rep. 2012;61(34):661.
11. Sahn B, Mamula P, Ford CA. Review of foreign body ingestion and
may include refractory bleeding, obstruction, or perfora- esophageal food impaction management in adolescents. J Adolesc
tion. A laparoscopic approach is often initially attempted Health. 2014;55(2):260-266.
and intraoperative endoscopy may be considered to 12. Traub SJ, Hoffman RS, Nelson LS. Body packing—the internal
improve localization. Once a bezoar has been cleared, concealment of illicit drugs. N Engl J Med. 2003;349:2519.
13. Bulstrode N, Banks F, Shrotria S. The outcome of drug smuggling
therapy aimed at prevention of recurrence should be by “body packers”—the British experience. Ann R Coll Surg Engl.
instituted including consideration for psychiatric assess- 2002;84:35.
ment, if appropriate. 14. Steinbach C, Stockmann M, Jara M, Bednarsch J, Lock JF. Accidentally
ingested toothpicks causing severe gastrointestinal injury: a practical
guideline for diagnosis and therapy based on 136 case reports. World
REFERENCES J Surg. 2014;38(2):371-377.
15. Zouros E, Oikonomou D, Theoharis G, Bantias C, Papadimitropoulos
1. Webb WA. Management of foreign bodies of the upper gastrointestinal K. Perforation of the cecum by a toothpick: report of a case and
tract: update. Gastrointest Endosc. 1995;41:36. review of the literature. J Emerg Med. 2014;47(6):e133-e137.
754 SECTION II Stomach and Small Intestine
16. Haidary A, Leider J, Silbergleit R. Unsuspected swallowing of a 22. Byrne WJ. Foreign bodies, bezoars, and caustic ingestion. Gastrointest
partial denture. AJNR Am J Neuroradiol. 2007;28:1734. Endosc Clin N Am. 1994;4:99.
17. Gachaboyov M, Isaev M, Orujova L, Isaev E, Yaskin E, Neronov D. 23. Iwamuro M, Okada H, Matsueda K, et al. Review of the diagnosis
Swallowed dentures: two cases and a review. Ann Med Surg. 2015;4:407. and management of gastrointestinal bezoars. World J Gastrointest
18. Diller R, Senninger N, Kautz G, Tübergen D. Stent migration Endosc. 2015;7:336.
necessitating surgical intervention. Surg Endosc. 2003;17:1803. 24. Ladas D, Kamberoglou D, Karamanolis G, Vlachogiannakos J,
19. Tarnasky PR, Cotton PB, Baillie J, et al. Proximal migration of biliary Zouboulis-Vafiadis I. Systematic review: Coca-Cola can effectively
stents: attempted endoscopic retrieval in forty-one patients. Gastrointest dissolve gastric phytobezoars as a first-line treatment. Aliment Pharmacol
Endosc. 1995;42(6):513-519. Ther. 2013;37:169.
20. Mistry BM, Memon MA, Silverman R, et al. Small bowel perforation 25. Winkler W, Saleh J. Metoclopramide in the treatment of gastric
from a migrated biliary stent. Surg Endosc. 2001;15:1043. bezoars. Am J Gastroenterol. 1983;78:403.
21. Andrus CH, Ponsky JL. Bezoars: classification, pathophysiology, and
treatment. Am J Gastroenterol. 1988;83:476.
t
ne
k.
oo
yb
er
rg
su
://
tp
ht
CHAPTER
Motility Disorders of the Stomach and
Small Intestine 65
Justin Barr
| Rebekah R. White
t
The gastric pacemaker, which is located in the body along Recent research has demonstrated the synergistic effect
ne
the greater curvature, stimulates both the filling and of multiple mechanisms in causing gastroparesis. With
mixing of food in the body and antrum. The antrum gastroparesis, both abnormal peristaltic contractile activity
strongly and periodically contracts against the closed and abnormal electrical slow waves are usually present.
k.
pylorus to grind solid food particles down to a small size. Antral hypomotility, likely a neuropathic process reflecting
The antrum peristalses at a frequency of three cycles per the loss of ICCs, commonly occurs in diabetic patients,
oo
minute and propels small particles and liquids into with evidence linking it to those with Parkinson disease
the duodenum as the pylorus opens. Consequently, the as well.4 Although the fundus has minimal contractile
stomach has three motile regions that coordinate to empty activity, its normally firm tone facilitates food movement
yb
the stomach. The fundus receptively relaxes and subse- to the body; a lax fundus—shown to result from both
quently contracts, and the body then fills and mixes. The vagotomy and diabetes—delays passage of chyme. In
antropyloroduodenal complex triturates and then empties contrast, impaired pyloric relaxation, seen primarily in
into the duodenum as the pyloric sphincter opens. type 1 diabetics, can trap contents in the stomach. The
er
Noncontractile gastric slow waves originate from the stomach and duodenum must act together, with the pylorus
gastric pacemaker at a frequency of three waves per minute and duodenum relaxing as the antrum contracts. Disrup-
rg
and propagate in both circumferential and longitudinal tions in this concert or the various neurohormonal factors
directions.1 The network of interstitial cells of Cajal (ICCs) that coordinate the process also lead to gastric
in the myenteric plexus initiates the slow wave and then dysmotility.
su
the gastric slow wave signal to reach an action potential Americans. The disease affects females more often
level through activation of calcium channels resulting in than males (4 : 1), with the onset of symptoms begin-
tp
muscle contractions. Thus the ICCs in the myenteric ning at an average age of 34 years. There were 16,736
plexus initiate the slow wave frequency in the smooth primary hospitalizations for gastroparesis in 2009 (up
muscle, and the intramuscular ICCs propagate the slow 18-fold from 1994), at a cost of an average $25,000 per
ht
ABSTRACT
Gastrointestinal dysmotility represents a severe constellation
of symptoms resulting from malfunction of the stomach,
small intestine, or large intestine. By definition excluding
organic origins of obstruction, the etiologies of the condi-
tion range from diabetes to Parkinson’s to more exotic
myopathies, although idiopathy remains the most com-
monly cited cause. Patient presentation typically includes
epigastric pain, nausea, vomiting, early satiety, and poor
oral intake; constipation characterizes lower GI dysmotility.
Severity of the symptoms can range from mildly aggravating
to requiring a feeding jejunostomy or even total parental
nutrition for survival. CT scan, EGD, and barium swallow
are helpful to rule out organic causes, but nuclear medicine
emptying studies are required to make diagnosis. Treatment
options remain poor. Symptomatic relief with various
anti-nausea medications can alleviate minor cases. Meto-
clopromide remains the only FDA-approved medication
t
ne
for gastroparesis but comes with well-described, severe
side effects. Other, more exotic interventions like intra-
pyloric botulinum injections and implanted gastric
pacemakers lack convincing evidence of efficacy. More
k.
dramatic, surgical cures like pyloroplasty, jejunostomy
feeding tubes, and even small bowel transplant are indi-
oo
cated in particularly severe cases.
KEYWORDS
yb
Gastric motility, intestinal motility, gastroparesis, meto-
clopramide, gastric electric stimulation, dysmotility,
ileus, chronic intestinal pseudoobstruction
er
rg
su
://
tp
ht
756 SECTION II Stomach and Small Intestine
Serosa
Fundus
Muscularis propria
longitudinal Myenteric
circular plexus
oblique
Submucosal
plexus
Submucosa
Muscularis mucosa
Mucosa
Gastric
pacemaker