Professional Documents
Culture Documents
Textbook Of Interventional Cardiology 8Th Edition Eric J Topol full chapter pdf docx
Textbook Of Interventional Cardiology 8Th Edition Eric J Topol full chapter pdf docx
Textbook Of Interventional Cardiology 8Th Edition Eric J Topol full chapter pdf docx
https://ebookmass.com/product/textbook-of-interventional-
cardiology-7th-edition-eric-j-topol/
https://ebookmass.com/product/oxford-textbook-of-interventional-
cardiology-2nd-edition-redwood/
https://ebookmass.com/product/the-esc-textbook-of-sports-
cardiology-1st-edition-antonio-pelliccia/
https://ebookmass.com/product/textbook-of-pediatric-
rheumatology-8th-edition-ross-e-petty-md-phd-frcpc/
Brocklehurst’s Textbook of Geriatric Medicine and
Gerontology 8th Edition Howard M. Fillit
https://ebookmass.com/product/brocklehursts-textbook-of-
geriatric-medicine-and-gerontology-8th-edition-howard-m-fillit/
https://ebookmass.com/product/abnormal-child-psychology-7th-
ed-7th-edition-eric-j-mash/
https://ebookmass.com/product/an-introduction-to-correspondence-
analysis-1st-edition-eric-j-beh/
https://ebookmass.com/product/principles-and-practice-of-
interventional-pulmonology-2013th-edition/
https://ebookmass.com/product/eric-hobsbawm-a-life-in-history-
richard-j-evans/
Textbook of
INTERVENTIONAL
CARDIOLOGY
8 TH
EDITION
ERIC J. TOPOL, MD
PAUL S. TEIRSTEIN, MD
1600 John F. Kennedy Blvd.
Ste 1800
Philadelphia, PA 19103-2899
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
Practitioners and researchers must always rely on their own experience and knowledge in evaluating and
using any information, methods, compounds or experiments described herein. Because of rapid advances
in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be
made. To the fullest extent of the law, no responsibility is assumed by Elsevier, authors, editors or con-
tributors 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.
Printed in China
William T. Abraham, MD, FACP, FACC, Stephen Balter, PhD Sergio Buccheri, MD
FAHA, FESC, FRCP Professor of Clinical Radiology (Physics) Interventional Cardiologist
Professor of Medicine, Physiology, and in Medicine Cardiac-Thoracic-Vascular Department
Cell Biology Radiology and Medicine Azienda Policlinico-Vittorio Emanuele
College of Medicine Distinguished Columbia University Associate Professor of Cardiology
Professor New York, New York University of Catania
Division of Cardiovascular Medicine Catania, Italy
David T. Balzer, MD
The Ohio State University
Professor, Pediatrics Robert A. Byrne, MB BCh, PhD
Columbus, Ohio
Division of Pediatric Cardiology Interventional Cardiologist
Marcelo Abud, MD Director, Cardiac Catheterization Deutsches Herzzentrum München
Fellow Laboratory Technische Universität
Interventional Cardiology and Washington University School of Munich, Germany
Endovascular Therapies Medicine
Davide Capodanno, MD, PhD
Cardiovascular Institute of Buenos Aires St. Louis, Missouri
Interventional Cardiologist
Buenos Aires, Argentina
Gregory W. Barsness, MD Cardiac-Thoracic-Vascular Department
Jung-Min Ahn, MD Assistant Professor Azienda Policlinico-Vittorio Emanuele
Associate Professor Departments of Internal Medicine, Associate Professor of Cardiology
Department of Cardiology Cardiovascular Medicine, and University of Catania
Asan Medical Center Radiology Catania, Italy
University of Ulsan College of Medicine Director, Cardiac Intensive Care Unit
Ivan P. Casserly, MD
Seoul, Republic of Korea Mayo Clinic
Professor of Medicine
Rochester, Minnesota
Takashi Akasaka, MD, PhD University College Dublin
Department of Cardiovascular Medicine Olivier F. Bertrand, MD, PhD Mater Misericordiae University Hospital
Wakayama Medical University Quebec Heart-Lung Institute Dublin, Ireland
Wakayama, Japan Interventional Cardiology
Matthews Chacko, MD
Quebec City, Canada
Ibrahim Akin, MD Assistant Professor of Medicine
Universitätsklinikum Mannheim Farzin Beygui, MD, MPH, PhD Division of Cardiology
Fakultät Heidelberg Professor of Cardiology Johns Hopkins University and Hospital
Abteilung Kardiologie Interventional Cardiology Unit Baltimore, Maryland
Mannheim, Germany Caen University Hospital
Derek P. Chew, MBBS, MPH, PhD
Caen, France
Waleed Alharbi, MD FRACP, FACC, FESC, FCSANZ
Complex Coronary, Structural and John A. Bittl, MD Professor of Cardiology
Endovascular Interventional Cardiology Interventional Cardiologist Department of Cardiovascular Medicine
Fellow AdventHealth Ocala Flinders University
Prairie Heart Institute Ocala, Florida Network Director of Cardiology
Springfield, Illinois Department of Cardiovascular Medicine
Nyal Borges, MD
Southern Adelaide Health Service
David W. Allen, MD Department of Cardiovascular Medicine
Adelaide, Australia
Assistant Professor of Cardiology Cleveland Clinic
Max Rady College of Medicine Cleveland, Ohio Leslie Cho, MD
University of Manitoba Section Head, Preventive Cardiology &
Vikram M. Brahmanandam, MD
Winnipeg, Manitoba, Canada Rehabilitation
Attending Cardiologist, Assistant
Director, Womens Cardiovascular Center
Alexandra Almonacid, MD Professor of Medicine
Cleveland Clinic
Associate Director Cardiology
Cleveland, Ohio
Beth Israel Deaconess Medical Center Montefiore-Einstein Center for Heart and
Cardiovascular Imaging Core Vascular Care Michael L. Chuang, MD
Laboratory Bronx, New York Assistant Director
Boston, Massachusetts Beth Israel Deaconess Medical Center
Éric Brochet, MD
Cardiovascular Imaging Core
Dominick J. Angiolillo, MD, PhD Cardiology Department
Laboratory
Professor of Medicine Hopital Bichat
Boston, Massachusetts
Director, Cardiovascular Research Paris, France
Program Director, Interventional Antonio Colombo, MD
Cardiology Fellowship EMO-GVM Centro Cuore Columbus
University of Florida College of Medicine San Raffaele Scientific Institute
Jacksonville, Florida Milan, Italy
vi
CONTRIBUTORS vii
Marco A. Costa, MD, PhD Vasim Farooq, MBChB, PhD Mario J. Gössl, MD
University Hospitals Harrington Heart & Newcastle upon Tyne Hospitals Director, Transcatheter Research and
Vascular Institute NHS Foundation Trust Education, Center for Valve and
Case Western Reserve University School Newcastle, United Kingdom Structural Heart Disease
of Medicine Co-chair, Valve Science Center
Miroslaw Ferenc, MD
Cleveland, Ohio Minneapolis Heart Institute
Head of Interventional Cardiology
Abbott Northwestern Hospital
Alain Cribier, MD Division of Cardiology and Angiology II
Minneapolis, Minnesota
Department of Cardiology University Heart Center Freiburg - Bad
Rouen University Hospital Krozingen Nilesh J. Goswami, MD, FACC, FSCAI,
Rouen, France Bad Krozingen, Germany FSVM
Director of Cardiac Catheterization
Fernando Cura, MD, PhD Kenneth A. Fetterly, PhD
Laboratory
Director Medical Physicist
Director of Structural Heart Interventions
Interventional Cardiology and Cardiovascular Diseases
Prairie Heart Institute
Endovascular Therapies Mayo Clinic and Foundation
Springfield, Illinois
Instituto Cardiovascular de Buenos Aires Rochester, Minnesota
Buenos Aires, Argentina Elliott M. Groves, MD, MEng, FACC,
Peter J. Fitzgerald, MD, PhD
FSCAI
Ingo Daehnert, MD, PhD Professor Emeritus of Medicine and
Director, Structural Heart Interventions
Department of Pediatric Cardiology Engineering
Department of Medicine, Division of
University of Leipzig, Heart Center Division of Cardiovascular Medicine
Cardiology
Leipzig, Germany Stanford University School of Medicine
University of Illinois at Chicago
Director, Center for Cardiovascular
Vishal Dahya, MD Chicago, Illinois
Technology
Chief Fellow
Stanford University Medical Center Giulio Guagliumi, MD
Cardiovascular Medicine Fellowship
Stanford, California Cardiovascular Department
Summa Health Heart and Vascular
ASST Papa Giovanni XXIII
Institute Marat Fudim, MD
Bergamo, Italy
Summa Health System Duke University Medical Center
Akron, Ohio Duke Clinical Research Institute Serge C. Harb, MD
Durham, North Carolina Department of Cardiovascular Medicine
Kimberly S. Delcour, DO, FACC
Cleveland Clinic
Director, Cardiac CT Mario J. Garcia, MD
Cleveland, Ohio
Clinical Assistant Professor Chief of Cardiology
Department of Internal Medicine, Medicine Trent Hartshorne, MBBS, FRACP,
Division of Cardiology Montefiore Medical Center FCICM, DDU
Heart & Vascular Center Bronx, New York Cardiologist and Intensive Care Physician
University of Iowa Hospitals & Clinics Intensive Care Consultant
Baris Gencer, MD
Iowa City, Iowa The Alfred Hospital
Cardiology Division
Melbourne, Australia
Robert S. Dieter, MD, RVT Geneva University Hospital
Loyola University Medical Geneva, Switzerland Grant Henderson, MD
Center/Hines VA Fellow, Cardiovascular Medicine
C. Michael Gibson, MD, MS
Maywood, Illinois Cleveland Clinic
CEO of Baim and PERFUSE Research
Cleveland, Ohio
John S. Douglas, Jr., MD Institutes
Professor Professor of Medicine Timothy D. Henry, MD, FACC, MSCAI
Department of Medicine Cardiovascular Division, Department of Medical Director, The Carl and Edyth
Director, Interventional Cardiology Medicine Lindner Center for Research and
Fellowship Program Beth Israel Deaconess Medical Center Education at The Christ Hospital
Emory University School of Medicine Harvard Medical School The Carl and Edyth Lindner Family
Emory University Hospital Boston, Massachusetts Distinguished Chair in Clinical
Atlanta, Georgia Research
Bryan H. Goldstein, MD
Director of Programmatic and Network
Helene Eltchaninoff, MD Associate Professor of Pediatrics
Development Heart and Vascular
Department of Cardiology University of Cincinnati College of
Service Line
Rouen University Hospital Medicine
The Christ Hospital Health Network
Rouen, France The Heart Institute
Cincinnati, Ohio;
Cincinnati Children’s Hospital Medical
Marvin H. Eng, MD Professor of Medicine
Center
Center for Structural Heart Disease University of Minnesota
Cincinnati, Ohio
Division of Cardiology Cedars-Sinai Heart Institute
Henry Ford Hospital Jeffrey Goldstein, MD University of California Los Angeles
Detroit, Michigan Director of Cardiology Department
Howard C. Hermann, MD
Prairie Heart Institute
Zaher Fanari, MD John W. Bryfogle Jr. Professor of
Springfield, Illinois
Heartland Cardiology/Wesley Medical Cardiovascular Medicine
Center Carlos A. Gonzalez Lengua, MD Health System Director for Interventional
University of Kansas School of Medicine Medicine-Cardiology Cardiology Program
Wichita, Kansas Mount Sinai St. Luke’s Hospital Perelman School of Medicine, University
New York, New York of Pennsylvania
Philadelphia, Pennsylvania
viii CONTRIBUTORS
Dominique Himbert, MD David E. Kandzari, MD, FACC, FSCAI John M. Lasala, MD, PhD
Cardiology Department Director, Interventional Cardiology Professor of Medicine
Hopital Bichat Chief Scientific Officer Director, Structural Heart Disease
Paris, France Piedmont Healthcare Washington University School of
Piedmont Heart Institute Medicine
Ravi S. Hira, MD, FACC, FAHA, FSCAI
Atlanta, Georgia St. Louis, Missouri
Assistant Professor of Medicine
University of Washington Samir R. Kapadia, MD Amir Lerman, MD
Seattle, Washington Professor of Medicine Professor
Director, Sones Catheterization Department of Cardiovascular Medicine
Russel Hirsch, MD
Laboratories Mayo Clinic
Professor of Pediatrics
Director, Interventional Cardiology Rochester, Minnesota
University of Cincinnati College of
Fellowship
Medicine Scott M. Lilly, MD, PhD
Department of Cardiovascular Medicine
The Heart Institute Associate Professor
Cleveland Clinic
Cincinnati Children’s Hospital Medical Department of Medicine, Division of
Cleveland, Ohio
Center Cardiology
Cincinnati, Ohio Adnan Kastrati, MD Ohio State University
Professor of Cardiology Columbus, Ohio
Kazuhiro Hisamoto, MD
Deutsches Herzzentrum and 1.
Clinical Assistant Professor Michael J. Lim, MD
Medizinische Klinik rechts der Isar
Department of Cardiothoracic Surgery Interim Director and Associate Professor
Technische Universität
NYU School of Medicine of Medicine
Munich, Germany
New York, New York Cardiology Division
Yuki Katagiri, MD Saint Louis University
Yasuhiro Honda, MD
Department of Cardiology St. Louis, Missouri
Clinical Professor of Medicine
Academic Medical Center
Division of Cardiovascular Medicine William L. Lombardi, MD, FACC, FSCAI
University of Amsterdam
Stanford University School of Medicine Director, Complex Coronary Artery
Amsterdam, Netherlands
Director, Cardiovascular Core Analysis Disease Therapies
Laboratory Athanasios Katsikis, MD, PhD University of Washington Medical Center
Center for Cardiovascular Technology Department of Cardiology Seattle, Washington
Stanford University Medical Center General Military Hospital of Athens
Phillipp C. Lurz, MD, PhD
Stanford, California Athens, Greece
Department of Internal Medicine/
Khalil Ibrahim, MD Dean J. Kereiakes, MD, FACC, FSCAI Cardiology
Department of Cardiology Medical Director, The Christ Hospital Leipzig Heart Center, University
Johns Hopkins School of Medicine Heart and Vascular Center Hospital
Baltimore, Maryland Medical Director, The Christ Hospital Leipzig, Germany
Research Institute
Bernard Iung, MD Kambis Mashayekhi, MD
The Christ Hospital Health Network
Professor of Cardiology Associate Head of Interventional
Cincinnati, Ohio
University of Paris VII Cardiology
Professor of Clinical Medicine
Hospital Doctor Division of Cardiology and Angiology II
Ohio State University
Cardiology Department University Heart Center Freiburg - Bad
Hopital Bichat Morton J. Kern, MD Krozingen
Paris, France Chief of Medicine Bad Krozingen, Germany
Department of Medicine
Hani Jneid, MD, FACC, FAHA, FSCAI Roxana Mehran, MD
VA Long Beach Health Care System
Associate Professor of Medicine The Zena and Michael A. Wiener
Long Beach, California
Director, Interventional Cardiology Cardiovascular Institute
Fellowship Program Ajay J. Kirtane, MD, SM, FACC, FSCAI Icahn School of Medicine at
Director, Interventional Cardiology Associate Professor of Medicine, Mount Sinai
Research Columbia University Medical Center New York, New York
Baylor College of Medicine Chief Academic Officer, Center for
Adrian W. Messerli, MD, FACC, FSCAI
Director, Interventional Cardiology Interventional Vascular Therapy
Associate Professor of Medicine
The Michael E. DeBakey VA Medical Director, NYP/Columbia Cardiac
Director, Cardiac Catheterization
Center Catheterization Laboratories
Laboratories
Houston, Texas New York, New York
Gill Heart Institute, University of
James G. Jollis, MD, FACC Serge Korjian, MD Kentucky
Professor of Medicine PERFUSE Study Group Lexington, Kentucky
Duke University Cardiovascular Division, Department of
Rodrigo Modolo, MD, PhD
Durham, North Carolina Medicine
Department of Cardiology
Beth Israel Deaconess Medical Center
Michael A. Jolly, MD, FACC, RPVI Amsterdam University Medical Center
Harvard Medical School
Interventional Cardiologist Amsterdam, Netherlands
Boston, Massachusetts
OhioHealth Heart and Vascular Department of Internal Medicine
Columbus, Ohio Amar Krishnaswamy, MD Cardiology Division
Program Director University of Campinas (UNICAMP)
Interventional Cardiology Campinas, Brazil
Cleveland Clinic
Cleveland, Ohio
CONTRIBUTORS ix
Gilles Montalescot, MD, PhD Gjin Ndrepepa, MD Marc S. Penn, MD, PhD
Pitié-Salpêtrière University Hospital Professor of Cardiology Director of Research
Institut de Cardiologie Deutsches Herzzentrum München Director of Cardiovascular Medicine
Paris, France Technische Universität Fellowship
Munich, Germany Summa Health Heart and Vascular
Pedro R. Moreno, MD
Institute
The Zena and Michael A. Weiner Franz-Josef Neumann, MD, PhD
Summa Health System
Cardiovascular Institute Endowed Professor of Cardiovascular
Akron, Ohio;
The Marie-Josée and Henry R. Kravis Medicine
Professor of Medicine
Cardiovascular Health Center University of Frieburg
Integrative Medical Sciences
Icahn School of Medicine at Mount Sinai Medical Director
Northeast Ohio Medical University
New York, New York Division of Cardiology and Angiology II
Rootstown, Ohio
University Heart Center Freigurg - Bad
Jeffrey W. Moses, MD
Krozingen Jeffrey J. Popma, MD
Interventional Cardiology
Bad Krozingen, Germany Director, Interventional Cardiology
New York Presbyterian Hospital
Clinical Services
Columbia University Medical Center Christoph A. Nienaber, MD
Medicine (Cardiovascular Division)
New York, New York Imperial College
Beth Israel Deaconess Medical Center
The Royal Brompton & Harefield NHS
Debabrata Mukherjee, MD Professor of Medicine
Trust
Chairman, Department of Internal Harvard Medical School
Cardiology and Aortic Centre
Medicine Boston, Massachusetts
London, England
Chief, Cardiovascular Medicine
Matthew J. Price, MD
Texas Tech University Yoshinobu Onuma, MD, PhD
Assistant Professor
El Paso, Texas Thoraxcenter, Erasmus Medical Center;
Director, Cardiac Catheterization
Cardialysis
Dale J. Murdoch, MBBS Laboratory
Rotterdam, Netherlands
Centre for Heart Valve Innovation Division of Cardiovascular Diseases
St Paul’s Hospital Igor F. Palacios, MD Scripps Clinic
University of British Columbia Associate Professor of Medicine La Jolla, California
Vancouver, Canada Director, Interventional Cardiology
Lorenz Räber, MD, PhD
Fellowship Program
Sahar Naderi, MD Cardiology Department
Director, Interventional Cardiology
Division of Cardiology Bern University Hospital
Research
Kaiser Permanente, San Francisco Bern, Switzerland
Baylor College of Medicine
Medical Center
Director, Interventional Cardiology Vivek Rajagopal, MD
San Francisco, California
The Michael E. DeBakey VA Medical Staff Cardiologist
Srihari Naidu, MD Center Piedmont Heart Institute
Director, Cardiac Catheterization Houston, Texas Atlanta, Georgia
Laboratory
Tullio Palmerini, MD Sunil V. Rao, MD
Division of Cardiology
Unità Operativa di Cardiologia Duke Clinical Research Institute
Winthrop University Hospital
Dipartimento Cardio-Toraco-Vascolare Durham, North Carolina
Mineola, New York
Policlinico S. Orsola
Associate Professor of Medicine Robert F. Riley, MD, MS, FACC, FAHA,
Bologna, Italy
SUNY Stony Brook School of Medicine FSCAI
Stony Brook, New York Duk-Woo Park, MD, PhD Medical Director, Complex Coronary
Associate Professor Therapeutics Program
Craig R. Narins, MD
Department of Cardiology Heart and Vascular Center
Associate Professor of Medicine and
Asan Medical Center The Christ Hospital, Lindner Center for
Surgery
University of Ulsan College of Medicine Research and Education
Divisions of Cardiology and Vascular
Seoul, Republic of Korea Cincinnati, Ohio
Surgery
University of Rochester Medical Center Seung-Jung Park, MD, PhD Madhur A. Roberts, MD
Rochester, New York Professor Interventional Cardiology Fellow
Department of Cardiology Cardiology
Nima Nasiri, MD
Asan Medical Center Westchester Medical Center
Research Fellow in Medicine
University of Ulsan College of Medicine Valhalla, New York
Division of Cardiovascular Medicine
Seoul, Republic of Korea
Beth Israel Deaconess Medical Center Marco Roffi, MD
Boston, Massachusetts Manesh R. Patel, MD Cardiology Division
Department of Medicine University Hospital
Eliano P. Navarese, MD, PhD
Duke University Medical Center Geneva, Switzerland
Interventional Cardiology and
Durham, North Carolina
Cardiovascular Medicine Jason H. Rogers, MD
Mater Dei Hospital and SIRIO Division of Cardiovascular Medicine
MEDICINE Research Network University of California, Davis
Bari, Italy; Sacramento, California
Faculty of Medicine
University of Alberta
Edmonton, Canada
x CONTRIBUTORS
R. Kevin Rogers, MD, MSc, RPVI Danielle N. Sin, MS On Topaz, MD, FACC, FACP, FSCAI
Associate Professor Senior Research Coordinator Professor of Medicine
Program Director, Vascular Medicine & Division of Adult Cardiac Surgery Duke University School of Medicine
Intervention NYU Langone Medical Center Chief, Division of Cardiology
Interventional Cardiology New York, New York Charles George Veterans Affairs Medical
University of Colorado Center
Gagan D. Singh, MD
Aurora, Colorado Asheville, North Carolina
Division of Cardiovascular Medicine
Jennifer A. Rymer, MD, MBA University of California, Davis Mark K. Tuttle, MD
Department of Medicine Sacramento, California Fellow, Division of Cardiovascular
Duke University Medical Center Medicine
Paul A. Sobotka, MD
Durham, North Carolina Beth Israel Deaconess Medical Center
Affiliated Clinical Professor
Clinical Fellow, Harvard Medical School
Bruno Scheller, MD Medicine/Cardiology
Boston, Massachusetts
Clinical and Experimental Interventional The Ohio State University
Cardiology Columbus, Ohio Alec Vahanian, MD, FESC, FRCP (Edin.)
University of Saarland Professor of Cardiology
Nishtha Sodhi, MD
Homburg/Saar, Germany University of Paris VII
Structural Heart Disease & Interventional
Paris, France
Beth A. Schueler, PhD Cardiology Fellow
Professor of Medical Physics Cardiovascular Department Miguel Valderrábano, MD, FACC
Department of Radiology Barnes Jewish Hospital of Washington Lois and Carl Davis Centennial Chair,
Mayo Clinic University Methodist DeBakey Heart and Vascular
Rochester, Minnesota St. Louis, Missouri Center
Associate Professor of Medicine,
Joshua Seinfeld, MD Paul Sorajja, MD
Weill College of Medicine, Cornell
Department of Neurosurgery Roger L. and Lynn C. Headrick Chair,
University
University of Colorado School of Valve Science Center Director
Director, Division of Cardiac
Medicine Center for Valve and Structural Heart
Electrophysiology
Aurora, Colorado Disease
Department of Cardiology
Minneapolis Heart Institute, Abbott
Patrick W. Serruys, MD, PhD Houston Methodist Hospital
Northwestern Hospital
National Heart and Lung Institute, Houston, Texas
Minneapolis, Minnesota
Faculty of Medicine
Birgit Vogel, MD
Imperial College London Sabato Sorrentino, MD, PhD
The Zena and Michael A. Wiener
London, England The Zena and Michael A. Wiener
Cardiovascular Institute
Cardiovascular Institute
Margot M. Sherman Jollis, BS Icahn School of Medicine at Mount Sinai
Icahn School of Medicine at Mount Sinai
Denison University New York, New York
New York, New York
Granville, Ohio
Amit N. Vora, MD, MPH
Goran Stankovic, MD, PhD
Kunihiro Shimamura, MD Duke Clinical Research Institute
Clinic for Cardiology
Department of Cardiovascular Medicine Durham, North Carolina
Department for Diagnostic and
Wakayama Medical University
Catheterization Laboratories Robert Wagner, MD, PhD
Wakayama, Japan
Clinical Center of Serbia Department of Pediatric Cardiology
Satya S. Shreeniva, MD Faculty of Medicine University of Leipzig, Heart Center
Interventional Cardiologist University of Belgrade Leipzig, Germany
The Lindner Research Center Belgrade, Serbia
John G. Webb, MD
Division of Cardiology
Curtiss T. Stinis, MD Centre for Heart Valve Innovation
The Christ Hospital
Director, Peripheral Interventions St Paul’s Hospital
Cincinnati, Ohio
Program Director, Interventional University of British Columbia
Kevin H. Silver, MD Cardiology Fellowship Vancouver, Canada
Director, Coronary Intensive Care Unit Division of Interventional Cardiology
William S. Weintraub, MD
Director, Cardiac Catheterization Lab Scripps Clinic
MedStar Heart & Vascular Institute
Summa Health Heart and Vascular La Jolla, California
Georgetown University
Institute
Matthew Summers, MD Washington, DC
Summa Health System
Fellow Physician
Akron, Ohio Sandra Weiss, MD
Interventional Cardiology
Christiana Care Health System
Mitchell J. Silver, DO, FACC, FSVM, Cleveland Clinic Foundation
Newark, Delaware
RPVI Cleveland, Ohio
Interventional Cardiologist Christopher J. White, MD, MSCAI,
Paul S. Teirstein, MD
OhioHealth Heart and Vascular FACC, FAHA, FESC, FACP
Interventional Cardiology
Columbus, Ohio Professor and Chairman of Medicine
Scripps Clinic
The Ochsner Clinical School, University
Daniel I. Simon, MD La Jolla, California
of Queensland
University Hospitals Harrington Heart &
Chief of Medical Services
Vascular Institute
Ochsner Medical Center
Case Western Reserve University School
New Orleans, Louisiana
of Medicine
Cleveland, Ohio
CONTRIBUTORS xi
Wendy Whiteside, MD Paul G. Yock, MD, MA, AB Khaled M. Ziada, MD, FACC, FSCAI
Assistant Professor of Pediatrics Martha Meier Weiland Professor Professor of Medicine
University of Michigan Division of Bioengineering and Medicine Clinical Chief of Cardiology
Pediatric Cardiology Stanford University Director, Cardiovascular Interventional
C. S. Mott Children’s Hospital Stanford, California Fellowship Program
Congenital Heart Center Gill Heart Institute, University of
Katherine Yu, MD
Ann Arbor, Michigan Kentucky
Fellow
Lexington, Kentucky
R. Jay Widmer, MD, PhD University of Southern California
Assistant Professor of Internal Medicine Los Angeles, California David A. Zidar, MD, PhD
Baylor Scott and White University Hospitals Harrington Heart &
Alan Zajarias, MD
Temple, Texas Vascular Institute
Associate professor of Medicine
Case Western Reserve University School
Mathew R. Williams, MD Co-director, Center of Valvular Heart
of Medicine
Associate Professor of Cardiothoracic Disease
Cleveland, Ohio
Surgery & Medicine Cardiovascular Division
Chief, Division of Adult Cardiac Surgery Washington University school of Andrew A. Ziskind, MD
Director, Interventional Cardiology medicine Senior Vice President, Premier’s
Director, CVI Structural Heart Program St. Louis, Missouri Academic Health System Strategy
NYU Langone Medical Center Premier Inc.
Jeffrey D. Zampi, MD
New York, New York Charlotte, North Carolina
Assistant Professor of Pediatrics
Daaboul Yazan, MD University of Michigan Division of
Research Fellow Pediatric Cardiology
PERFUSE Study Group C. S. Mott Children’s Hospital
Cardiovascular Division, Department of Congenital Heart Center
Medicine Ann Arbor, Michigan
Beth Israel Deaconess Medical Center
Harvard Medical School
Boston, Massachusetts
Preface
The eighth edition of Textbook of Interventional Cardiology has the appropriateness or overuse of procedures. But hopefully, all
been more extensively updated than any previous edition. We of these challenges are outweighed by the immense gratification
have tried to fully capture the excitement and relentless matura- of helping a symptomatic patient with limitations in quality of
tion of the field of interventional cardiology, emphasizing rigor- life get back to his or her baseline. Nowhere in medicine is this
ous evidence-based approaches. New chapters have been added feeling more prevalent than in the transformative field of trans-
to address the diagnosis and treatment of coronary microvascular catheter aortic valve replacement.
disease, percutaneous tricuspid valve repair, and valve-in-valve This book is intended to serve as a resource for the interven-
interventions. Over the years, coronary intervention became tional cardiology community, which not only includes practicing
increasingly predictable and, in many ways, routine, with the cardiologists but also the team involved in procedures, referring
progressive maturation of stents and leaps forward in our adjunct physicians, and those training or who have aspiration to train
pharmacologic therapies. In some ways, the field of interventional in this awe-inspiring field. We have changed authors for many
cardiology lost a bit of its pioneering spark that had so character- chapters to provide a sense of newness and a fresh perspective,
ized this discipline from its inception in the 1980s. In those heady and in every chapter we have sought the authors who are widely
times, performing balloon angioplasty in the coronary artery was regarded as the true experts in the field. Going forward, we fully
unpredictable. The predictability provided by stents was replaced recognize that there needs to be increased cooperativity with car-
with the unpredictability of stent thrombosis. Interventional car- diac surgeons—the rising popularity of hybrid and collaborative
diologists and scientists had to not only rise to the challenge for valve procedures that capitalize on the best parts of percutaneous
each individual patient but also to discover the vital innovations and surgical approaches is clearly indicative of that collaboration.
that would perpetuate the prominence and importance of the We want to express our genuine and deep appreciation to the
specialty. authors from all over the world who have graciously contributed
Currently, the challenges continue, but they have morphed to this new edition. They represent a remarkable brain trust from
considerably. The profile of patients who undergo coronary whom we have learned so much in the review of their input. We
intervention has dramatically increased in complexity to include thank Mary Hegeler at Elsevier for her first-rate, professional
patients with advanced age and those with left main stem lesions, support of this endeavor, and we are especially grateful to the
chronic occlusions, and what would formerly have been consid- cardiovascular community of readers of this book who have sup-
ered prohibitive complexity. What ever happened to patients ported it as the primary reference textbook source for more than
with type A lesions? How can we break the maximal Synergy 30 years. That represents a large sense of responsibility for us
Between Percutaneous Coronary Intervention With Taxus and to maintain, and we hope to have lived up to that and perhaps
Cardiac Surgery (SYNTAX) score barrier for percutaneous cor- exceeded expectations with the eighth edition.
onary intervention? At the same time, the crisis in health care
economics has placed increased burdens on interventional cardi- Paul S. Teirstein, MD
ologists with respect to time, constraints in equipment selection, Eric J. Topol, MD
and fulfilling the responsibility of 24/7 coverage for such emer- La Jolla, California, 2019
gencies as acute myocardial infarction. There is also the incre-
mental pressure from scorecarding initiatives and challenges to
xii
SECTION I Patient Selection 1
KEY POINTS
• Changes in the demographics of patients who present in • Clinical tools based on the Synergy Between Percutaneous
need of revascularization, advances in percutaneous and Coronary Intervention With Taxus and Cardiac Surgery
surgical revascularization techniques, and results from (SYNTAX) trial have evolved from purely anatomic
contemporary studies of percutaneous versus surgical factors (anatomic SYNTAX score) to anatomic factors
revascularization have made it essential that patients be augmented by clinical variables (culminating in the
assessed as individuals prior to selection of a treatment development of the SYNTAX score II) and tools to assess
strategy. a level of reasonable incomplete revascularization that
• Risk stratification plays an important role in the would not have an adverse effect on long-term morbidity
assessment of patients undergoing revascularization. and mortality (residual SYNTAX score). Validation of
many of these newly developed clinical tools is ongoing.
• Clinical tools used to assist the heart team in risk
stratifying patients and deciding the most appropriate • Clinical and anatomic factors have an impact on short-
revascularization modality can be broadly divided into and long-term morbidity and mortality following surgical
assessments based on clinical comorbidities, coronary or percutaneous revascularization and must be considered
anatomy, or a combination of the two. by the heart team in open dialogue with the patient during
the decision-making process.
Revascularization of patients with coronary artery disease (CAD) and consequently patients are more likely to present with more
has progressed exponentially since Andreas Grüntzig1 performed extensive CAD. The Arterial Revascularization Therapies Studies
the first balloon angioplasty in 1977. These developments, which (ARTS) parts I and II were separated by a period of 5 years, and
have been fueled by new technology, have blurred the boundary despite both studies having the same inclusion criteria, patients in
between what was once considered exclusively surgical disease ARTS-II had a significantly greater incidence of risk factors and
and what can be treated percutaneously. Consequently, there is overall increased disease complexity (Table 1.1).4
a greater need than ever to tailor revascularization appropriately, Patient comorbidities must be taken into consideration when
taking into consideration a patient’s comorbidities, coronary assessing patients for revascularization because they have the
anatomy, personal preferences, and individual perception of risk. potential to significantly influence patient outcomes; moreover,
This chapter will explore the increasing requirement for a more they may have a different impact depending on the underlying
individualized assessment of patients undergoing revasculariza- revascularization strategy selected. Notably in patients enrolled
tion, and it will review the clinical tools currently available to in the ARTS-I and II studies, patient age was shown to be a sig-
assist in this decision-making process. nificant independent correlate of major adverse cardiovascular
and cerebrovascular events (MACCEs) who were treated with
CABG.5 More recently, in the randomized all-comers SYNTAX
NEED FOR INDIVIDUALIZED PATIENT ASSESSMENT trial, increasing age was shown to favor PCI over CABG when
A number of confounding factors have made it imperative that adjustments were made for other anatomic and clinical factors.6–8
patients are assessed as individuals prior to the selection of revas- In addition, other anatomic and clinical factors were shown to
cularization strategy. have an impact on long-term mortality, and thereby decision
making between CABG and PCI (SYNTAX score II7,8), and this
topic is discussed later under “SYNTAX-Based Clinical Tools.”
Patient Comorbidities In a collaborative patient-level analysis of 10 randomized
The demographics of patients presenting to tertiary care services trials of patients with multivessel disease (MVD) treated with
in need of revascularization are constantly evolving. This has been PCI using bare-metal stenting (BMS) and CABG, Hlatky and
largely the consequence of increased longevity of the general pop- coworkers9 demonstrated comparable rates of 5-year mortality
ulation, a lower threshold to investigate patients who present with between both treatment groups in patients without diabetes.
symptoms suggestive of obstructive CAD, and increased resources Notably, when patients with diabetes were viewed as a whole,
that have made revascularization via percutaneous coronary inter- mortality was significantly higher in those treated with PCI, even
vention (PCI) or coronary artery bypass grafting (CABG) more after multivariate adjustment (Fig. 1.1). In the Future Revascu-
accessible. Together with increased age, patients in need of larization Evaluation in Patients With Diabetes Mellitus: Opti-
revascularization are currently more likely to have comorbidities mal Management of Multivessel Disease (FREEDOM) trial,10,11
such as diabetes, hypertension, and hyerlipidemia.2,3 These fac- it was shown that in patients with diabetes and advanced CAD,
tors are all implicated in accelerating the progression of CAD, CABG was superior to PCI in that it significantly reduced rates
1
2 SECTION I Patient Selection
TABLE 1.1 Changing Baseline Demographics of Patients Enrolled in Drug-Eluting Stent Trials
All-Comers Studies
SIRTAX24 Leaders25 Resolute27 Arts-I28 Arts-II172 SYNTAX13
Years of Enrollment 2003–2004 2006–2007 2008 1997–1998 2003 2005–2007
Stent Type DES DES DES BMS DES DES
Demographics
Age, years (mean ± SD) 62 ± 11 65 ± 11 64.4 ± 10.9 61 ± 10 63 ± 10 65 ± 10
Diabetes, % 20 24 23.5 19 26 26
Hypertension, % 61 73 71.1 45 67 69
Hypercholesterolemia, % 59 67 63.9 58 74 78
Previous myocardial infarction, % 29 33 28.9 44 34 32
Left ventricular function, % (mean ± SD) 57 ± 12 56 ± 12 61 ± 12 60 ± 12 59 ± 13
Lesion Characteristics (Per Patient)
Multivessel disease, % 59 23 58.4 96 100 92
Bifurcation lesions, % 8 22 16.9 35 34 72
Total occlusions, % 19 12 16.3 3 17 24
SYNTAX score (mean ± SD) 12 ± 7 14 ± 9 15 ± 9 – 21 ± 10 28 ± 12
Mean number of diseased lesions 1.4 1.5 1.5 2.8 3.6 3.6a
Procedural Characteristics (Per Patient)
Mean number of stents 1.2 ± 0.5 1.3 ± 0.7b 11.9 ± 7.5 2.8 ± 1.3 3.7 ± 1.5 4.6 ± 2.3
Total stent length, mm (mean ± SD) 25.9 ± 15.5 24.7 ± 15.5b 34.4 ± 24.5 47.6 ± 21.7 72.5 ± 32.1 86.1 ± 47.9
aTreatedlesions.
bPer
lesion.
BMS, Bare-metal stent; DES, drug-eluting stent; SD, standard deviation; SYNTAX, Synergy Between Percutaneous Coronary Intervention with Taxus and
Cardiac Surgery.
50
P = .049 by log-rank test
40
5-year event rate: 16.3% vs. 10.9% HISTORIC (PRE-SYNTAX) CLINICAL TRIAL RESULTS
Historically, and prior to the publication of the SYNTAX trial,13–
30 16 randomized trials to compare CABG and PCI centered on
two major patient groups: either isolated proximal left anterior
20
descending (LAD) artery lesions or complex CAD (3VD and/or
PCI
10
LM disease). Although results of these studies suggest no differ-
CABG ences were found in the hard clinical outcomes of death and MI
0 between patients treated with PCI or CABG at short- and long-
0 1 2 3 4 5 term follow-up (Table 1.2),9,37–41 there was profound selection
bias in enrollment of patients prior to randomization. Specifically,
Years since randomization
No. at risk between 2% and 12% of screened patients were randomized in
PCI 953 897 845 685 466 243 most trials (Table 1.3), with patients with lesser comorbidities,
B CABG 947 855 806 655 449 238 such as impaired left ventricular function or coronary anatomy
Fig. 1.2 Kaplan-Meier Estimates of the Composite Primary Outcome
(predominantly single- or double-vessel disease) often “cherry-
of death, myocardial infarction (MI), or stroke (A) and death from any
picked” prior to randomization.42–44 Consequently, interpreting
cause (B) truncated at 5 years after randomization in the FREEDOM
and extrapolating these results to routine and contemporary clin-
trial. In FREEDOM, patients with diabetes and multivessel coronary
ical practice has been challenging.
artery disease were assigned to undergo either percutaneous coronary
intervention (PCI) with first-generation drug-eluting stents or coronary SYNTAX Trial
artery bypass grafting (CABG). Patients were followed for a minimum
of 2 years (median among survivors, 3.8 years), and CABG was
The landmark SYNTAX trial13–16 represents the largest (and
shown to be superior to PCI with first-generation drug-eluting stents
only) assessment of revascularization with PCI or CABG in
with significant reduced rates of death (10.9% vs. 16.3%, P = .049)
all-comers with complex CAD. SYNTAX aimed to supply evi-
and MI (6.0% vs. 13.9%, P < .001) but a higher rate of stroke (5.2%
dence to support the somewhat established but non–evidence-
vs. 2.4%, P = .03). (From FREEDOM Trial Investigators. Strategies for
based practice of performing PCI in patients with complex
multivessel revascularization in patients with diabetes. N Engl J Med.
CAD.29 In addition, SYNTAX also sought to identify which
2012;367(25):2375–2384.)
patients should be treated with CABG only. Through an all-
comers design, SYNTAX addressed the limitations of the ear-
lier CABG versus PCI trials, which were plagued by profound
TECHNOLOGIC ADVANCES selection bias as previously discussed (see Table 1.3),43,44 and
in doing so it was anticipated that the results would be more
The introduction in 2002 of DESs revolutionized the practice of relevant to contemporary routine clinical practice. Specifically:
interventional cardiology and was driven primarily through the
dramatic reduction in rates of repeat revascularization.23 The • To ensure results were applicable to routine practice, the
favorable results observed with DES use promptly resulted in an study was designed as an all-comers trial such that there were
expansion of the indications for PCI, such that bifurcation lesions, no specific inclusion criteria other than the need to have re-
chronic total occlusions (CTOs), and MVD were no longer in vascularization of de novo 3VD or unprotected LM CAD (in
the exclusive domain of surgical revascularization, and these were isolation or with CAD). Exclusion criteria were limited to
increasingly treated with PCI. Evidence of this expansion can be prior revascularization, ongoing MI, and patients requiring
seen in the changing baseline lesion characteristics of patients concomitant cardiac surgery.16 In contrast to the earlier stud-
enrolled in all-comers PCI trials such as the Sirolimus-Eluting ies, 70.9% of eligible patients were enrolled.
and Paclitaxel-Eluting Stents for Coronary Revascularization • The previously indicated problem of reporting outcomes from
(SIRTAX) trial,24 the Limus Eluted From a Durable Versus all patients with complex CAD together, irrespective of dis-
Erodable Stent Coating Study (LEADERS),25,26 the Clinical ease severity, was addressed in the SYNTAX trial through the
CHAPTER 1 Individualized Assessment for Percutaneous or Surgical-Based Revascularization 3.e1
A B
9F Outlet area
C
eFig. 1.1 Devices that are increasingly available to provide assistance during high-risk percutaneous coronary
intervention include percutaneous extracorporeal circulatory support devices such as the TandemHeart
(A and B) and the Impella device (C). (A) The TandemHeart removes oxygenated blood from the left atrium and
returns this blood into the peripheral arterial circulation; with the (B) aid of a centrifugal pump. (C) The Impella
left ventricular assist device is a miniaturized rotary blood pump that is placed retrograde across the aortic
valve, and it aspirates (inlet area) up to 2.5 L/min of blood from the left ventricular cavity and subsequently
expels this blood (outlet area) into the ascending aorta. (From Valgimigli M, Steendijk P, Serruys PW, et al. Use
of Impella Recover LP 2.5 left ventricular assist device during high-risk percutaneous coronary interventions;
clinical, haemodynamic and biochemical findings. EuroIntervention. 2006;2[1]:91–100; and Vranckx P, Meliga
E, De Jaegere PP, et al. The TandemHeart, percutaneous transseptal left ventricular assist device: a safe-
guard in high-risk percutaneous coronary interventions. The six-year Rotterdam experience. EuroIntervention.
2008;4[3]:331–337.)
4 SECTION I Patient Selection
1.0
0.8
1.00
Proportion surviving
0.95
0.6
0.90
CABG
0.85
0.4
0.80
0.75 PCI
0.2 0.70
0.00
0 500 1000 1500
0.0
0 500 1000 1500
Days since index revascularization
30-day 1-year 2-year 3-year 4-year
Mortality after CABG, % (95% CI) 2.07 (1.98–2.17) 6.00 (5.58–6.17) 8.76 (8.56–8.94) 12.1 (11.9–12.4) 16.0 (15.7–16.3)
Mortality after PCI, % (95% CI) 1.21 (1.14–1.27) 6.36 (6.22–6.51) 11.2 (11.0–11.4) 16.0 (15.7–16.2) 20.9 (20.6–21.3)
A Relative risk with CABG (95% CI) 1.72 (1.58–1.84) 0.94 (0.91–0.97) 0.78 (0.76–0.80) 0.76 (0.74–0.78) 0.76 (0.75–0.78)
1.0
0.8
1.00
Proportion surviving
0.95
0.6
0.90
CABG
0.85
0.4
0.80
0.75 PCI
0.2 0.70
0.00
0 500 1000 1500
0.0
0 500 1000 1500
Days since index revascularization
30-day 1-year 2-year 3-year 4-year
Mortality after CABG, % (95% CI) 2.25 (2.09–2.41) 6.24 (5.97–6.50) 8.98 (8.68–9.29) 12.4 (12.0–12.8) 16.4 (15.9–16.9)
Mortality after PCI, % (95% CI) 1.31 (1.21–1.41) 6.55 (6.35–6.76) 11.3 (11.0–11.6) 15.9 (15.6–16.3) 20.8 (20.4–21.2)
B Relative risk with CABG (95% CI) 1.72 (1.52–1.89) 0.95 (0.90–1.00) 0.79 (0.76–0.83) 0.78 (0.75–0.81) 0.79 (0.76–0.82)
Fig. 1.3 Incidence of survival in the coronary artery bypass grafting (CABG) and percutaneous coronary
intervention (PCI) cohorts, from unadjusted (A) and adjusted (B) analyses. Cumulative mortality with CABG and
PCI and the relative risk of CABG compared with PCI are shown. Data from the American College of Cardiol-
ogy Foundation and Society of Thoracic Surgeons Database Collaboration on the Comparative Effectiveness
of Revascularization Strategies registry, the American College of Cardiology Foundation National Cardiovascu-
lar Data Registry, and the Society of Thoracic Surgeons Adult Cardiac Surgery Database from 2004 through
2008. (From Weintraub WS, Grau-Sepulveda MV, Weiss JM, et al. Comparative effectiveness of revasculariza-
tion strategies. N Engl J Med. 2012;366[16]:1467–1476.)
CHAPTER 1 Individualized Assessment for Percutaneous or Surgical-Based Revascularization 5
use of the anatomic SYNTAX score (www.syntaxscore.com; TAX score for this group was 26.1 and 28.8 in patients
Fig. 1.5),13,45–49 which enabled CAD complexity to be objec- treated with CABG and PCI, respectively. 1
tively and prospectively quantified. 2. Nested CABG registry (n = 1077 [35.0%]): These patients
• To ensure assessment of patients on an individual level, all had CAD that was considered unsuitable for PCI, clearly
patients eligible for enrollment were discussed by the heart reflected in the high mean SYNTAX score (37.8) for this
team.50 An interventional cardiologist and cardiac surgeon group.
carried out a careful and comprehensive review of the patient 3. Nested PCI registry (n = 198 [6.4%]): These patients were
in terms of their anginal status, comorbidities, and coronary deemed unsuitable for CABG. The commonest reason for
anatomy using the respective Braunwald score, European Sys- this decision was the presence of multiple comorbidities13
tem for Cardiac Operative Risk Evaluation (EuroSCORE), reflected in the mean EuroSCORE, which was 2 points
and SYNTAX score (discussed under “SYNTAX-Based Clin- higher in this group than the mean in the randomized
ical Tools”). The consensus reached from this meeting was group (5.8 vs. 3.8).
subsequently used to allocate the patient into one of the three
arms of the trial. In total, 3075 patients were enrolled into one Overall, SYNTAX failed to meet the prespecified primary
of the following: end point of noninferiority in terms of 12-month MACCEs, a
1. Randomized group (n = 1800 [58.5%]; 897 CABG, 903 composite of death, stroke, MI, and repeat revascularization
PCI): These patients had CAD and were equally suitable (17.8% vs. 12.4%, P = .002). Final 5-year reporting of SYN-
for revascularization with PCI or CABG. The mean SYN- TAX demonstrated significantly higher incidence of MACCE
(i) Death, stroke, or myocardial infarction (ii) Death from any cause
100 25 100 25
20 20
80 15 80 15
14.7%
10 Patients (%) 10 8.2%
Patients (%)
15.4%
60 60
5 5
5.9%
0 0
40 0 6 12 24 36 40 0 12 24 36
6
Hazard ratio, 1.00 (95% CI, 0.79–1.26) Hazard ratio, 1.34 (95% CI, 0.94–1.91)
20 20 P = .11
P = .98
0 0
01 6 12 24 36 01 6 12 24 36
20 20
80 15 80 15
10 10
Patients (%)
Patients (%)
8.3%
60 60
5 2.9% 5 8.0%
2.3%
0 0
40 0 12 24 36 40 0 6 12 24 36
6
Hazard ratio, 0.77 (95% CI, 0.43–1.37) Hazard ratio, 0.93 (95% CI, 0.67–1.28)
20 P = .37 20 P = .64
0 0
01 6 12 24 36 01 6 12 24 36
A PCI 948 930 915 893 839 473 PCI 948 900 882 857 805 452
CABG 957 922 899 880 823 511 CABG 957 879 846 830 776 480
Fig. 1.4 Primary composite end point of death, stroke, or myocardial infarction and its components in the ongo-
ing EXCEL trial at 3 years, indicating similar clinical outcomes (A).17,18 In addition, both percutaneous coronary
intervention (PCI) and coronary artery bypass grafting (CABG) result in similar quality of life (QOL) improvement
at 36 months, with a substantially greater early benefit in QOL seen with PCI at 1 month (B).19 CI, Confidence
interval; SAQ, Seattle Angina Questionnaire. (Reproduced with permission from references 17 and 19.)
Continued
6 SECTION I Patient Selection
CENTRAL ILLUSTRATION: Disease-Specific Health Status After PCI Versus CABG as Measured by the SAQ
100 90
90
80
80
70
70 ∆ = –0.8 ∆ = 16.1 ∆ = 0.7
∆ = 1.5 ∆ = –0.3 ∆ = 1.3
P = .03 P = .63 P = .21 P < .01 P = .24 P = .55
60 60
0 1 12 24 36 0 1 12 24 36
Months Months
100 90
80
95
70
90 60
50
85 ∆ = 3.4
∆ = 1.3 ∆ = 1.0 ∆ = 0.3 40 ∆ = –2.4 ∆ = –1.9
P = .02 P = .08 P = .63 P < .01 P = .02 P = .07
80 30
0 1 12 24 36 0 1 12 24 36
Months Months
B PCI CABG
TABLE 1.2 Summary of Meta-Analyses Prior to Publication of the SYNTAX Trial Reporting Long-Term Outcomes in Patients With Isolated Proximal Left
Anterior Descending Coronary Artery Disease or Multivessel Disease Randomized to Percutaneous or Surgical Revascularization
First Number of Patients POBA/BMS/ Follow-Up Death (PCI vs. MI (PCI vs. Stroke (PCI vs. Repeat Revasc. MACCEs (PCI
Author (PCI/CABG) DES (%) (Months) CABG) CABG) CABG) (PCI vs. CABG) vs. CABG)
Isolated Proximal LAD
Aziz37 1952 (1300/652) 0/91/9 34 2.9% vs. 3.4% 2% vs. 1.1% 2.4% vs. 3.5% 14.3% vs. 4.4%a 21.4% vs.
11.1%a
Kapoor38 1210 (633/577) 22/59/19 60 9.4% vs. 7.2% NA NA 33.5% vs. 7.3%a NA
Multivessel Disease
Hlatky9 7812 (3923/3889) 63/37/0 5.9 10.0% vs. 8.4% 16.7% vs. – 24.5% vs. 36.4% vs.
15.4%b 9.9%a,b 20.1%a
Daemen40 3051 (1518/1533) 4/96/0 60 8.5% vs. 8.2% 2.5% vs. 2.9% 6.6% vs. 6.1% 25.0% vs. 6.3%a 34.2% vs.
19.6%a
Bravata41 9963 (5019/4944) 56/42/2 60 9.3% vs. 11.3% 0.6% vs. 1.2%a 11.9% vs. 10.9% 46.1% vs. 40.1% –
vs. 9.8%a,c
aP < .001.
bComposite with death.
cBalloonangioplasty versus PCI versus CABG.
BMS, Bare-metal stent; CABG, coronary artery bypass grafting; DES, drug-eluting stent; LAD, left anterior descending coronary artery; MACCEs, major
adverse cardiovascular and cerebrovascular events (a composite of death, stroke, MI, and repeat revascularization); MI, myocardial infarction; NA, not
available; PCI, percutaneous coronary intervention; POBA, plain old balloon angioplasty; Revasc., revascularization.
TABLE 1.3 Summary of 15 Randomized Control Trials Comparing Coronary Artery Bypass Grafting Against Percutaneous Coronary Intervention in the
Pre-SYNTAX Era
Trial Number of Patients Screened % Randomized Stent % 3VD Proximal LAD EF >50% % Diabetes
MASS39 142 69 – – 100 100 21
ERACI173 127 9 – 45 – 100 11
1
EAST174 392 4 – 40 70 100 25
GABI175 359 4 – 18 – – 10
CABRI176 1054 3 – 40 – 100 12
BARI177 1829 12 – 41 36 100 24
SIMA178 121 – – – 100 100 11
LAUSANNE179 134 3 – 0 100 – 12
RITA180 1011 4 – 12 – – 6
TOULOSE181 152 3 29 – – 14
AWESOME182 454 – + 45 – – –
ERACI-II183 450 2 + 56 – – 17
ARTS184 1205 5 + 32 – 100 19
SOS185 988 5 + 38 45 100 14
MASS II186 408 2 + 41 – – –
Summary 8826 5 35 41 100 16
3VD, Three-vessel disease; CABG, coronary artery bypass grafting; EF, ejection fraction; LAD, left anterior descending artery; PCI, percutaneous coronary intervention.
From Soran O, Manchanda A, Schueler, S. Percutaneous coronary intervention versus coronary artery bypass surgery in multivessel disease: a current
perspective. Interact Cardiovasc Thorac Surg. 2009;8(6):666–671.
Fig. 1.5 The SYNTAX score algorithm is applied to each individual coronary lesion in a vessel larger than 1.5
mm in diameter that has a stenosis diameter greater than 50%; the individual lesion scores are added together
to give the final SYNTAX score.16,45-47 LAD, Left anterior descending; LCx, left circumflex artery; LM, left main;
RCA, right coronary artery. (Modified from Serruys PW, Onuma Y, Garg S, et al. Assessment of the SYNTAX
score in the SYNTAX study. EuroIntervention. 2009;5:50–56.)
8 SECTION I Patient Selection
25
13.9%
9.7%
11.4%
0 3.8%
0 12 24 36 48 60 0 12 24 36 48 60
Months since allocation Months since allocation
Number at risk
CABG 897 820 810 788 761 606 897 800 784 759 730 575
PCI 903 859 853 832 803 537 903 832 821 792 756 593
25 20.8%
3.7% 16.7%
2.4%
0
0 12 24 36 48 60 0 12 24 36 48 60
Months since allocation Months since allocation
Number at risk
CABG 897 806 790 763 732 579 897 787 776 749 717 566
PCI 903 854 842 815 782 622 903 830 824 792 756 592
37.3%
25.9%
25
26.9%
13.7%
0
0 12 24 36 48 60 0 12 24 36 48 60
Months since allocation Months since allocation
Number at risk
CABG 897 778 760 717 677 532 897 751 739 694 654 512
PCI 903 760 740 688 644 495 903 747 733 681 634 483
Fig. 1.6 Five-year Kaplan-Meier cumulative event curves of major adverse cardiovascular and cerebrovascular
events (a composite of death, stroke, myocardial infarction and repeat revascularization) and its components
among the 1800 patients randomized to percutaneous coronary intervention (PCI) or coronary artery bypass
grafting surgery (CABG) in the SYNTAX trial. (From Mohr FW, Morice MC, Kappetein AP, et al. Coronary artery
bypass graft surgery versus percutaneous coronary intervention in patients with three-vessel disease and left main
coronary disease: 5-year follow-up of the randomised, clinical SYNTAX trial. Lancet. 2013;381[9867]:629–638.)
with PCI compared with CABG (26.9% vs. 37.3%, P < .0001; Fig. 1.7, clinical outcomes between patients treated with PCI
Fig. 1.6).15 and CABG in SYNTAX differed according to the presence of
As indicated earlier, analyses of all patients irrespective of 3VD or unprotected LM CAD. With 3VD, a low SYNTAX
disease severity does not provide adequate information for cli- score (<23) allowed for similar outcomes between CABG and
nicians who are faced daily with patients who display a wide PCI, whereas higher SYNTAX scores (particularly in the high
variety of CAD complexity. To address this limitation of earlier SYNTAX score [>32] group) clearly favored CABG. With
studies, patient outcomes in SYNTAX were stratified accord- unprotected LM CAD, a low-intermediate SYNTAX score
ing to tertiles of the anatomic SYNTAX score. As shown in (<33) allowed for similar outcomes between CABG and PCI,
CHAPTER 1 Individualized Assessment for Percutaneous or Surgical-Based Revascularization 9
Overall Cohort Left Main Coronary Disease Subgroup Three-Vessel Disease Subgroup
Baseline SYNTAX score 0–22 Baseline SYNTAX score 0–22 Baseline SYNTAX score 0–22
1
25
28.6% 30.4% 26.8%
0
0 12 24 36 48 60 0 12 24 36 48 60 0 12 24 36 48 60
Months since allocation Months since allocation Months since allocation
Number at risk
CABG 275 226 221 212 197 154 104 87 86 80 74 56 171 137 135 133 123 98
PCI 299 263 255 237 223 168 118 109 108 98 93 68 181 154 147 139 130 100
Baseline SYNTAX score 23–32 Baseline SYNTAX score 23–32 Baseline SYNTAX score 23–32
Cumulative event rate (%)
50
P = .008 P = .88 P = .0008 37.9%
36.0% 32.7%
25
32.3%
25.8%
22.6%
0
0 12 24 36 48 60 0 12 24 36 48 60 0 12 24 36 48 60
Months since allocation Months since allocation Months since allocation
Number at risk
CABG 300 251 248 230 219 172 92 75 74 66 66 51 208 176 174 164 153 121
PCI 310 257 256 236 221 173 103 91 90 79 78 60 207 166 166 157 143 114
Baseline SYNTAX score ≥33 Baseline SYNTAX score ≥33 Baseline SYNTAX score ≥33
Cumulative event rate (%)
50
P < .0001 P = .003 P = .0005
44.0% 46.5%
41.9%
25
26.8% 29.7%
24.1%
0
0 12 24 36 48 60 0 12 24 36 48 60 0 12 24 36 48 60
Months since allocation Months since allocation Months since allocation
Number at risk
CABG 315 272 267 251 237 185 149 130 127 118 112 86 166 142 141 133 125 99
A PCI 290 224 220 206 188 139 B 135 103 101 95 84 60 C 155 121 119 111 104 79
Fig. 1.7 The evidence supporting the use of the SYNTAX score as a tool to assist in revascularization decisions.
Five-year Kaplan-Meier cumulative event curves of major adverse cardiovascular and cerebrovascular events
(MACCEs; a composite of death, stroke, myocardial infarction and repeat revascularization) among the 1800 pa-
tients randomized to percutaneous coronary (PCI) or coronary artery bypass graft surgery (CABG) in the SYNTAX
trial (A) and stratified by the presence of unprotected left main coronary artery disease (B) or de novo three-vessel
disease (C). In patients with three-vessel disease (C), the incidence of MACCEs at 5-year follow-up was similar
among patients treated with PCI and CABG for low SYNTAX scores (≤22); for all other SYNTAX scores (>22), out-
comes were significantly better following CABG. (B) In patients with unprotected left main coronary artery disease,
clinical outcomes were similar among patients treated by PCI or CABG for all low to intermediate SYNTAX scores
(<33), whereas more complex disease (SYNTAX score >32) fared significantly better with CABG. (From Mohr FW,
Morice MC, Kappetein AP, et al. Coronary artery bypass graft surgery versus percutaneous coronary intervention
in patients with three-vessel disease and left main coronary disease: 5-year follow-up of the randomised, clinical
SYNTAX trial. Lancet. 2013;381[9867]:629–638.)
whereas a high SYNTAX score (>32) clearly favored CABG. subjects in all tertiles of the anatomic SYNTAX score who had
Furthermore, the SYNTAX score II,7,8 essentially the ana- a long-term mortality that favored either CABG, PCI, or both
tomic SYNTAX score augmented with clinical variables shown revascularization modalities (discussed under “SYNTAX-
directly to affect decision making between CABG and PCI, Based Clinical Tools”).7,8,13–16
was developed in the randomized, all-comers SYNTAX trial The results of SYNTAX reiterate the importance of
and allowed for the identification of higher- and lower-risk assessing patients when selecting a revascularization strategy.
10 SECTION I Patient Selection
SYNTAX was able to identify those patients in whom either appropriate revascularization modality, which may be modified
CABG or PCI was appropriate or in whom CABG or PCI was by the heart team consensus. In addition to their role in the
the optimal treatment. Considering the distribution of CAD risk stratification of individual patients, these quantitative risk
in SYNTAX, overall one-third of patients with 3VD/LM dis- scores have increasing use in the wider context of overall health
ease were deemed to have CAD that could be treated safely care. They can provide a vital measure of overall patient care
and effectively with PCI or CABG, whereas in the remaining and can help to identify future directions to further improve
two-thirds, CABG remained the standard of care. Although outcomes. Clinical governance and the increasing requirement
these results helped further delineate the boundaries between to publicly report clinical performance and complications have
a percutaneous and surgical revascularization approach in also propelled the need to risk stratify patients, thereby allow-
patients with complex CAD, the validation of the anatomic ing a useful comparison of performance to be made between
SYNTAX score and development of the SYNTAX score II clinicians and institutions against the standards dictated by
notably facilitated a more objective assessment of patients regulatory authorities.67 In addition, calculation of risk using
by the heart team as discussed later under “SYNTAX-Based accepted risk scores may aid clinicians faced with an increasing
Clinical Tools.” need to be able to justify their clinical decisions to peers, regu-
latory bodies, and patients.
INDIVIDUAL ASSESSMENT FROM A PHYSICIAN’S In comparison with the qualitative risk scores, the use of a
finite number of variables results in these risk scores lacking the
PERSPECTIVE sensitivity to accurately predict risk in an individual, such that
There is no disputing the need for and potential benefits of select- they are more apt at predicting risk for a population of patients
ing a revascularization strategy following an individualized patient with similar comorbidities. The number of variables included in
assessment or risk stratification. Risk stratification is performed the score must strike a balance between sufficient numbers to
routinely and subconsciously by physicians in everyday clinical enable a meaningful prediction of risk to be calculated; however,
practice and is in essence behind all clinical decisions made by a the number must not be excessive so as to prevent use in rou-
physician. Stratification of risk is vital when assessing patients for tine practice. In addition, a minimal number of variables reduces
revascularization because this treatment is only considered appro- the chances of colinearity between independent variables, which
priate when “the expected benefits, in terms of survival or health can result in redundant information being collected62 while also
outcomes (symptoms, functional status, and/or QOL) exceed the increasing the chances of “overfitting” the score and thereby
expected negative consequences of the procedure.”51 However, it reducing the overall applicability and accuracy of the results to
should be emphasized that the SYNTAX-pioneered heart team conventional clinical practice.68
approach, consisting of at least an interventional/clinical cardi- The applicability of a risk score to contemporary practice must
ologist and a cardiac surgeon,50 carries a class I recommendation also take into consideration the time when the score was developed.
in international guidelines for assessing risk and is subsequently Risk scores rely on large patient databases to derive appropriate
the most appropriate revascularization modality in patients with weighting factors for variables in the score to enable the final calcu-
complex CAD.52–55 lation of risk. It follows that they are developed using retrospective
information that may no longer be relevant in the era when the risk
score is being used. For example, the EuroSCORE was developed
Qualitative Versus Quantitative Risk Assessment in 1999; however, there have been calls for its recalibration because
Qualitative risk stratification is subjective and relies on a clini- repeated evaluations indicate that it overestimates risk by a factor
cian’s experience. This subjective qualitative assessment also of 2 to 3, which has largely been attributed to improvements in
allows risk to be calculated and tailored to the expertise of the surgical techniques and lower perioperative mortality in the decade
physician performing the procedure, as opposed to a clinician following its construction.69,70 The updated EuroSCORE II cur-
in another region who may use different techniques and who rently addresses many of the limitations of the original EuroS-
may have different equipment available. In addition, assessments CORE.71–73 The STS score is also derived from a large patient
of patient frailty can be made that are frequently not captured database and is periodically recalibrated to ensure its results are
by conventional risk-scoring systems.56 This assessment does applicable to contemporary practice.74
not require a calculator or computer and can be “computed”
subconsciously very quickly. The major disadvantages of this
method of risk assessment are its dependence on the operator’s
Risk Scores in Contemporary Practice
prior experience, potential personal bias to undertake or with- Numerous risk scores are available to assist clinicians in strati-
hold potential revascularization, and its high interobserver vari- fying risk among patients undergoing revascularization. Some
ability. In addition, influences of local practice often dominate scores are appropriate for patients prior to the selection of a
clinical decision making, irrespective of the revascularization revascularization strategy, whereas some have been validated
guidelines. only in patients undergoing one form of revascularization.
Quantitative risk stratification can be performed using a vari- Nevertheless, the various risk scores can largely be categorized
ety of risk scores that frequently incorporate clinical variables according to the variables—clinical, angiographic, or a combina-
sourced from large patient registries,57–64 with the exception tion of both—used in the overall estimation of risk. Tables 1.4
of the SYNTAX score II,7,8 which was developed in the all- and 1.5 summarize the different risk scores used in contemporary
comers randomized SYNTAX trial to reduce unavoidable (but CABG and PCI practice (excluding SYNTAX-based tools), and
often appropriate) selection bias inherent to all registries no Table 1.6 summarizes SYNTAX-based clinical tools. A selection
matter their size. These risk scores largely incorporate objec- of these is described in more detail later.
tive variables to ensure adequate reproducibility of the score;
however, those risk scores—such as the American College of
Cardiology/American Heart Association (ACC/AHA) lesion
Clinical Scores
score65 or the anatomic SYNTAX score/newly developed These risk scores incorporate only clinical variables and do not
SYNTAX score II,7,8,46 which include angiographic variables— require any data from the angiogram. They offer the advantage
continue to have documented intraobserver and interobserver of being able to be computed relatively quickly, usually at the
variability.47,66 However, these tools do provide a more objec- bedside, and principally include variables that are not subject to
tive assessment of the patient risk and suitability for the most user interpretation, thereby ensuring excellent reproducibility.
CHAPTER 1 Individualized Assessment for Percutaneous or Surgical-Based Revascularization 11
TABLE 1.4 Summary of Contemporary Risk Scores for Assessment of Risk in Patients Undergoing Surgical Revascularization With the Exception of the
Anatomic SYNTAX Score 1
Number of Variables Used to Calculate Score Validated in PCI/CABG
Risk Score Clinical Angiographic PCI CABG
EuroSCORE13,57-59,75,76,80–84 17 0 + +
EuroSCORE II71–73 18 0 − +
ACEF62 3 0 − +
Society of Thoracic Surgery score64,74,87,104 40 2 − +
Anatomic SYNTAX score4,13,47,81,84,94,95,187–194 0 11 (per lesion) + +
SYNTAX-based tools are shown in Table 1.6.
ACEF, Age, creatinine, and ejection fraction; CABG, coronary artery bypass grafting; PCI, percutaneous coronary intervention.
TABLE 1.5 Summary of a Selection of Established and Contemporary Risk Scores Categorized by Anatomic, Clinical, or Combined Types for the
Assessment of Risk in Patients Proposing to Undergo Percutaneous Coronary Intervention
Number of Variables Used
to Calculate Risk
Clinical Risk Score Clinical Angiographic PCI Outcomes (Surgical Outcomes in Italics)
Anatomic Scores
ACC/AHA lesion 0 11 (per lesion) Pre-DES era: predictive of angiographic success of PCI and prognostic effect on early and
classificationa late clinical outcomes. Conflicting results were yielded in the DES era.93,95,195–197
Myocardial Jeopardy Scores
Duke Jeopardy Score 0 Coronary tree divided into six segments: LAD, diagonal, septal perforating branches, LCx, OM, and PDA; a
segment distal to ≥70% is considered at risk. Each segment is assigned 2 points with a maximum of 12
points.96,97b
Myocardial Jeopardy Index 0 Distal terminating portions of LAD, LCx, RCA, and major branch vessels (diagonals, OM, ramus, PDA and
(BARI) LV branches) assigned units of 1, 2, or 3 on the basis of length and vessel size. Septal perforators are
arbitrarily assigned a maximum of 3 units. Extent of jeopardy defined by units jeopardized by ≥50% stenosis
summated and divided by total LV territory.97,98b
APPROACH lesion score 0 Based on principle from autopsy studies that the LAD generally subtends 41% of the LV, with the LCx and RCA
supplying the remainder, dependent on vessel dominance. Score is calculated by percent of myocardium
supplied by a vessel or its branches and jeopardized territories supplied by vessels with ≥70% stenosis
(≥50% in the LMS); the maximum score is 100.97b
Clinical Scores
New Mayo Clinic Risk Scorea 7 0 Procedural death and MACEs for PCI; score has been externally validated for death62,63
(in-hospital death with CABG).87
Parsonnet Score 14 0 Independent predictor of long-term MACEs after LMS PCI in two registry
populations198,199 (operative mortality after open-heart surgery)200
EuroSCORE (additive or 17 0 Evidence for predicting death or MACCEs in high-risk tertiles for PCI79,80,83,201 (operative
logistic) mortality for all forms of cardiothoracic surgery).58,61
NCDR CathPCI Risk Scorea 8 0 Developed from 181,775 procedures performed in Medicare patients; incidence of
in-hospital and 30-day mortality after all PCI patient types internally validated in two
separate cohorts.65
ACEF score (age, creatinine, 3 0 Predictor of cardiac death and MI at 1 year after PCI; inferior to the SYNTAX score at
ejection fraction) predicting overall MACEs and repeat revascularization in two separate populations90,122
(operative mortality in elective cardiac operations).62,88
Combined (Anatomic AND Clinical) Risk Scores
EuroHeart PCI Scorea 10 6 Developed from 46,000 patients from the Euro Heart Survey; in-hospital mortality in all
PCI patient types; internally validated. The score has strong applicability for European
practice.105
New Risk Stratification 17 Angiographic: 33 6-month cardiac death and cumulative MACEs after unprotected LMS PCI; although
Score (NERS) Procedural: 4c internally validated, application to larger all-comers population is required (see text).106
New York PCI Risk Scoreb 8 1 In-hospital death after PCI; developed based on data from 46,090 procedures in 2002
and validated from 50,046 procedures in 2003202; excellent predictive ability in
validation cohort (C-statistic 0.905).
The Texas Heart Institute 8 Angiographic: 2 Predictors of in-hospital MACEs after PCI or CABG; developed in 9494 patients (BMS era)
Risk Scorea Procedural: 1d and validated in 5545 patients (DES era).203
Mayo Clinic Risk Scorea 6 2 In-hospital death, Q-wave myocardial infarction, emergent or urgent CABG or CVA after
PCI; validated using the NHLBI registry.204
aRiskscores that include prediction of in-hospital mortality or MACEs. SYNTAX-based tools are shown in Table 1.6.
bAll
myocardial jeopardy scores were validated in one population-based cohort consisting of more than 20,000 patients and were predictive of 1-year
mortality in patients treated with PCI or medically.96–98
cNeed of intraaortic balloon pump, two-stent technique, intravascular ultrasound guidance.
dNumber of stents.
ACC/AHA, American College of Cardiology/American Heart Association; BMS, bare-metal stent; CABG, coronary artery bypass grafting; CVA,
cerebrovascular accident; DES, drug-eluting stent; LAD, left anterior descending artery; LCx, left circumflex artery; LMS, left main stem; LV, left
ventricular; MACCEs, major adverse cardiovascular and cerebrovascular event; MACE, major adverse cardiovascular event; MI, myocardial infarction;
NCDR, National Cardiovascular Data Registry; OM, obtuse marginal artery; PCI, percutaneous coronary intervention; PDA, posterior descending artery;
NHLBI, National Heart, Lung, and Blood Institute; RCA, right coronary artery.
CHAPTER 1 Individualized Assessment for Percutaneous or Surgical-Based Revascularization 13
TABLE 1.6 Outline of the Anatomic SYNTAX Score and the Progression of SYNTAX-Based Tools
1
Year Structure Remarks
Anatomic SYNTAX 2006 Score of angiographic variables (i.e., anatomic First reported to be useful for decision making between CABG
Score4,13-15,48,55,60,64–72,166 complexity); developed during the design and PCI in the SYNTAX trial in 200913; categories of anatomic
of the SYNTAX trial13,47 as a tool to force complexity (low, intermediate, and high), no clinical variables, no
the heart team to systematically analyze the individual predictions; adding a functional component shown to
coronary angiogram and agree equivalent improve accuracy13; noninvasive multislice computed tomography
anatomic revascularization (CABG and PCI) anatomic SYNTAX score in development,36 with integration of a
could be achieved noninvasive functional component.162
Development Phase: Augmenting the Anatomic SYNTAX Score With Clinical Variables and the Move Toward Individualized Decision Making
ACEF163 2009 Age, creatinine, ejection fraction Predicted individual in-hospital operative mortality post CABG;
shown to be at least comparable to the EuroSCORE (composed
of 17 variables) in predicting operative risk62,88,89; shown to aid in
long-term predictions of mortality after PCI or CABG.62,89
Clinical SYNTAX Score127 2010 Amalgamation of SYNTAX score with modified Similar to the SYNTAX score; categorized patient risk; could only
ACEF score (creatinine replaced with CrCl identify a high-risk group in PCI-treated patients; provided
shown to be more predictive of mortality121) little help in decision making between CABG and PCI; not
individualized.
Global Risk88,123 2010 Amalgamation of SYNTAX score with surgical Similar to the SYNTAX score; categorized patient risk; could identify
EuroSCORE (composed of 17 variables) a low-risk group with comparable outcomes with CABG and
PCI in LM and 3VD patients; not individualized; patients with
a high EuroSCORE were found to have a prognostic benefit in
undergoing CABG compared with PCI irrespective of the SYNTAX
score provided an acceptable threshold of operative risk was not
exceeded.
Logistic Clinical SYNTAX 2011 Combination of age, SYNTAX score, CrCl, Individual 1-year mortality predictions in all PCI patients (STEMI,
Score126,127 and LVEF shown to contain most of NSTEMI) irrespective of clinical presentation (except cardiogenic
the prognostic data for 1-year mortality shock); not designed to help decision making between CABG and
predictions after PCI PCI; cross-validated in seven contemporary stent trials and more
than 6000 patients and further externally validated.123
End Result of This Process Leading to the Development of the SYNTAX Score II
SYNTAX Score II124 2012 Augmenting SYNTAX score with clinical Individualized approach; threshold of the SYNTAX score in guiding
variables; based on the principle that age, decision making between CABG and PCI shown to alter based
CrCl, LVEF, and SYNTAX score contain most on the presence of other risk factors; validated in the DELTA
of the long-term prognostic data in CABG Registry7,8 containing LM and 3VD (25% of the population) with
and PCI patients; additional variables added almost a third (30%) with highly complex disease (SYNTAX scores
that directly influenced decision making ≥33); prospective validation studies are underway in the EXCEL
between CABG and PCI trial (LM), and SYNTAX II trial is ongoing (de novo 3VD).
Use of the SYNTAX Score as an Objective Marker of Completeness of Revascularization
Residual SYNTAX 2012 Recalculation of the SYNTAX score after PCI Developed and validated in the ACUITY146,147 and SYNTAX146 trials;
Score129 a residual SYNTAX score greater than 8 was shown to have an
adverse effect on long-term prognosis at up to 5-year follow-up;
further, prospectively run validation studies are awaited.
Post-CABG SYNTAX 2013 Recalculation of the SYNTAX score after CABG Pilot study in angiographic substudy of the SYNTAX trial
Score147 with points deducted based on the demonstrated the feasibility of this approach in identifying
importance of the diseased coronary artery subjects post CABG with an adverse long-term (5-year)
segment (Leaman score154,155) that has a prognosis109; validation studies are awaited.
functioning bypass graft anastomosed distally
3VD, Three-vessel disease; CABG, coronary artery bypass grafting; CrCl, creatinine clearance; LM, left main; LVEF, left ventricular ejection fraction;
NSTEMI, non–ST elevation myocardial infarction; PCI, percutaneous coronary intervention; STEMI, ST elevation myocardial infarction.
Value of Age, Creatinine, and Ejection Fraction Score From ACEF, a mortality risk can be calculated from a
Ranucci and colleagues62,88 demonstrated in a relatively simple risk graphical relationship of ACEF with an operative risk or an
score consisting of only three clinical variables—age, preoperative equation (Fig. 1.8).62,88 ACEF was developed from an initial
serum creatinine value, and LVEF—a risk score for assessing oper- dataset of 4557 patients and a subsequent validation series of
ative mortality risk in elective cardiac operations. Notably, despite 4091 patients from a single institution. The results demon-
the simplicity of the score, the clinical performance of the Age, strated a similar accuracy and calibration for the prediction of
Creatinine, and Ejection Fraction (ACEF) score appeared to be in-hospital mortality with ACEF when compared with other
comparable with either the additive or the logistic EuroSCORE. more complicated surgical risk scores such as the EuroSCORE
The ACEF score is calculated using the following formula: and the Cleveland Clinic Score. Subsequent validation studies
have shown ACEF to have an accuracy level at least compa-
ACEF = [Age/LVEF (%)] + [1 (if creatinine>2 mg/dL)]
rable with that of the EuroSCORE for operative mortality risk
14 SECTION I Patient Selection
TABLE 1.7 Components of the EuroSCORE and Relevant Weighting Factors of the Additive and Logistic EuroSCOREs
Patient Characteristics Additive Logistic β Coefficient
Age Per 5 years or part thereof over the age of 60 years 1 0.07
Sex Female 1 0.33
Chronic pulmonary disease Long-term use of bronchodilators or steroids for respiratory disease 1 0.49
Peripheral arteriopathy Claudication, carotid stenosis >50%, previous or planned intervention on 2 0.66
the abdominal aorta, limb arteries, or carotidsa
Neurologic dysfunction Severely affected mobility or day-to-day function 2 0.84
Previous cardiac surgery Previous opening of the pericardium 3 1.00
Serum creatinine Preoperatively greater than 200 μmol/L 2 0.65
Active endocarditis Antibiotic therapy at time of surgery 3 1.10
Critical preoperative state Preoperative cardiac arrest, ventilation, renal failure, inotropic support, 3 0.91
intraaortic balloon pump use, ventricular arrhythmiaa
Cardiac-Related Factors
Unstable angina Rest pain that requires IV nitrates 2 0.57
Left ventricular function Moderate (30%–50%) 1 0.42
Poor (<30%) 3 1.09
Recent MI Within 90 days 2 0.55
Pulmonary hypertension Systolic pulmonary pressure greater than 60 mm Hg 2 0.77
Operation-Related Factors
Emergency Operation performed before the start of next working day 2 0.71
Other than isolated CABG Major cardiac procedure other than or in addition to CABG 2 0.54
Surgery on thoracic aorta 3 1.16
Postinfarct septal rupture 4 1.46
Constant β0 −4.79
aAny of these.
The logistic EuroSCORE can be calculated at www.euroscore.org.
CABG, Coronary artery bypass grafting; IV, intravenous; MI, myocardial infarction.
From Singh M, Rihal CS, Lennon RJ, et al. Bedside estimation of risk from percutaneous coronary intervention: the new Mayo Clinic Risk Scores. Mayo
Clinic Proc. 2007;82:701–708; and Singh M, Peterson ED, Milford-Beland S, et al. Validation of the Mayo Clinic risk score for in-hospital mortality after
percutaneous coronary interventions using the National Cardiovascular Data Registry. Circ Cardiovasc Interv. 2008;1:36–44.
National Cardiovascular Database Registry CathPCI Risk- American College of Cardiology/American Heart
Prediction Score Association Lesion Classification System
The NCDR CathPCI risk-prediction score is the most con- The ACC/AHA lesion classification system was one of the first
temporary clinical risk score currently available. It incorpo- angiographic scoring systems developed. Initially devised in 1986
rates information from nine clinical variables (Table 1.9), and modified in 1990, it uses 11 angiographic variables to catego-
which are assigned appropriate weighted values and are then rize lesions into four groups: types A, B1, B2, and C (Table 1.10).
added together to give a final score that can be translated into Historic studies prior to the arrival of DESs indicated that that
risk of in-hospital mortality (Fig. 1.9).63 The score was devel- the ACC/AHA lesion classification did have a prognostic impact
oped using data from more than 180,000 patients from the vol- on early and late outcomes.65,91,92
untary U.S. NCDR database and was validated in more than However, registry data from the DES era has shown con-
400,000 patients from the same database who underwent PCI flicting results. The German Cypher registry (n = 6755) failed
between March 2006 and March 2007. Notably, the C-statistic to show any definite relationship between clinical outcomes and
for the prediction of in-hospital mortality was consistently ACC/AHA lesion class at 6 months.93 These results are at vari-
greater than 0.90 for in-hospital mortality, whereas a lower but ance to the positive relationship identified between ACC/AHA
nevertheless adequate C-statistic of 0.83 was seen for 30-day lesion class and clinical outcomes in smaller studies of patients
mortality. with more complex disease.94,95
CHAPTER 1 Individualized Assessment for Percutaneous or Surgical-Based Revascularization 15
TABLE 1.8 Final Risk Factors by Multivariate Regression for the EuroSCORE II
1
Risk Factor Coefficient Standard Error z P ≥ |z| [95% Confidence Interval]
New York Hospital Association (NYHA)
II 0.1070545 0.1463849 0.73 0.465 [−0.1798547 to 0.3939637]
III 0.2958358 0.141466 2.09 0.037 [0.0185674 to 0.5731042]
IV 0.5597929 0.1697565 3.30 0.001 [0.2270763 to 0.8925095]
CCS4 0.2226147 0.1462888 1.52 0.128 [−0.0641061 to 0.5093356]
IDDM 0.3542749 0.145863 2.43 0.015 [0.0683887 to 0.6401611]
Age 0.0285181 0.0065954 4.32 0.000 [0.0155914 to 0.0414448]
Female 0.2196434 0.0953505 2.30 0.021 [0.0327599 to 0.4065269]
ECA 0.5360268 0.1106046 4.85 0.000 [0.3192458 to 0.7528079]
CPD 0.1886564 0.1232126 1.53 0.126 [−0.0528358 to 0.4301486]
N/M mob 0.2407181 0.1729494 1.39 0.164 [−0.0982564 to 0.5796927]
Redo 1.118599 0.1226272 9.12 0.000 [0.8782539 to 1.3589440]
Renal Dysfunction
On dialysis 0.6421508 0.3083468 2.08 0.037 [0.0378021 to 1.2464990]
CC ≤ 50 0.8592256 0.1446758 5.94 0.000 [0.5756663 to 1.1427850]
CC 50–85 0.303553 0.1240518 2.45 0.014 [0.0604159 to 0.5466901]
Active endocarditis 0.6194522 0.2046001 3.03 0.002 [0.2184433 to 1.0204610]
Critical 1.086517 0.147657 7.36 0.000 [0.797115 to 1.3759200]
Left Ventricular Function
Moderate 0.3150652 0.1036182 3.04 0.002 [0.1119773 to 0.5181530]
Poor 0.8084096 0.1498233 5.40 0.000 [0.5147614 to 1.1020580]
Very poor 0.9346919 0.2917754 3.20 0.001 [0.3628227 to 1.5065610]
Recent MI 0.1528943 0.136257 1.12 0.262 [−0.1141646 to 0.4199531]
Pulmonary Artery Systolic Pressure
31–55 mm Hg 0.1788899 0.1266713 1.41 0.158 [−0.0693812 to 0.4271611]
≥55 0.3491475 0.1676641 2.08 0.037 [0.0205318 to 0.6777632]
Urgency
Urgent 0.3174673 0.1174178 2.70 0.007 [0.0873326 to 0.5476020]
Emergency 0.7039121 0.1719835 4.09 0.000 [0.3668306 to 1.0409940]
Salvage 1.362947 0.33706 4.04 0.000 [0.7023221 to 2.0235730]
Weight of Procedure
1 non-CABG 0.0062118 0.1463574 0.04 0.966 [−0.2806434 to 0.2930670]
2 0.5521478 0.1268137 4.35 0.000 [0.3035975 to 0.8006980]
3+ 0.9724533 0.1463969 6.64 0.000 [0.6855206 to 1.2593860]
Thoracic aorta 0.6527205 0.221183 2.95 0.003 [0.2192097 to 1.0862310]
Constant −5.324537 0.1682446 −31.65 0.000 [−5.65429 to −4.9947830]
CABG, Coronary artery bypass grafting; CC, creatinine clearance; CCS, Canadian Cardiovascular Society; CPD, chronic pulmonary dysfunction; Critical,
critical preoperative state; ECA, extracardiac arteriopathy; IDDM, insulin-dependent diabetes mellitus; N/M mob, neurologic or musculoskeletal
dysfunction severely affecting mobility; MI, myocardial infarction; Redo, previous cardiac surgery.
Weight of procedure: 1, non-CABG, single major cardiac procedure (MCP) that is not isolated CABG; 2, two MCPs; 3+, three or more MCPs. For age, Xi =
1 if patient age is 60 or younger; Xi increases by one point per year thereafter (e.g., age 60 or less, Xi = 1; age 61, Xi = 2; age 62, Xi = 3, etc.).
From Nashef SA, Roques F, Sharples LD, et al. EuroSCORE II. Eur J Cardiothorac Surg. 2012;41(4):734–744.
Specifically, Valgimigli and colleagues94 reported that a higher objectively quantified. This is further discussed in the section on
ACC/AHA lesion score (derived by assigning 1, 2, 3, and 4 points SYNTAX-based scoring tools.
to type A, B1, B2, and C lesions, respectively) correlated with
poor clinical outcomes among 306 patients with 3VD undergoing Myocardial Jeopardy Scores
PCI with a DES. In addition, data from a small registry (n = 255) Myocardial jeopardy scores are a method of estimating the
were potentially predictive of mortality and MACEs in unpro- amount of myocardium at risk based on the assessment of both
tected LM stem PCI at 1-year follow-up.95 the severity of the coronary artery lesion and the volume of
myocardium it supplies. Examples of such scores include the
Anatomic SYNTAX Score Duke Jeopardy Score, the Myocardial Jeopardy Index from the
The anatomic SYNTAX score represents a comprehensive angi- Bypass Angioplasty Revascularization Investigation (BARI) score,
ographic scoring system that allows the complexity of CAD to be and the Alberta Provincial Project for Outcome Assessment in
16 SECTION I Patient Selection
72
70
68 Age (years)
ACEF score = + 1 (if serum creatinine ≥ 2 mg/dL)
66 EF (%)
64
62
60
58
56
54
52
50
48
Predicted mortality rate (%)
46
44
42
40
38
36
34
32
30
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.6 4.8 5.0 5.2
ACEF score
Fig. 1.8 Univariate association (logistic regression) between age, creatinine, and ejection fraction (ACEF)
score—the value of age, creatinine, and ejection fraction—and mortality risk. (From Ranucci M, Castelvecchio
S, Menicanti L, et al. Risk of assessing mortality risk in elective cardiac operations: age, creatinine, ejection
fraction, and the law of parsimony. Circulation. 2009;119[24]:3053–3061.)
Coronary Heart Disease (APPROACH) score (Fig. 1.10; see SYNTAX-based tools to assess completeness of revasculariza-
Table 1.5). The Duke and BARI scores were developed and vali- tion (see Table 1.6) after CABG and PCI have been developed
dated in relatively small populations. All three scores have since and are discussed in the section on SYNTAX-based tools.
been validated in one population-based cohort consisting of more
than 20,000 patients and were predictive of 1-year mortality in
patients treated with PCI or treated medically; within this pop-
Combined Risk Scores
ulation, all three scores also had similar performance measures The previous discussion has reviewed risk scores that rely on
with only minor differences in C-statistics evident.96–98 either clinical or angiographic variables (with the exception
It has since been shown that the Duke Jeopardy Score is an of SYNAX-based clinical tools). There is no disputing that
independent predictor of adverse clinical outcomes, namely death for a complete individualized patient assessment, both fac-
and MI, in medically treated patients with acute coronary syn- tors must be taken into consideration. Moreover, current evi-
dromes (ACSs) at up to 1 year in a post hoc study of the Acute dence indicates that clinical and angiography-based risk scores
Catheterization and Urgent Intervention Triage Strategy (ACU- may be better suited to predict different patient outcomes.
ITY) trial.99 Clinical scores appear to be better at predicting clinical end
The BCIS-1 (balloon pump-assisted coronary intervention points such as death or MI, whereas angiography-based scores
study) Myocardial Jeopardy Score, a variant of the Duke Jeopardy appear to be superior for the prediction of angiographic suc-
Score that has been reported to be simpler to use, has recently cess and the risk of repeat revascularization. Of note, Peterson
been shown to have a strong correlation with the myocardial isch- and coworkers63 observed only a minimal improvement in the
emic burden as assessed by cardiac magnetic resonance perfusion ability of the NCDR CathPCI risk score to predict in-hos-
imaging.100,101 A BCIS-1 Jeopardy Score of 10 to 12 and a revas- pital mortality following the inclusion of angiographic vari-
cularization index (preprocedural minus postprocedural jeopardy ables. These findings are in line with previous reports, which
scores divided by preprocedural jeopardy score, with 1 indicating demonstrated that the MCRS was superior to the ACC/AHA
complete revascularization) of 0 to 0.33 were both shown to be lesion classification in the prediction of death, stroke, MI, and
highly predictive of mortality after contemporary PCI in a single emergent CABG but was inferior for the prediction of angio-
U.K. center experience that involved over 600 patients.102 graphic failure.103
CHAPTER 1 Individualized Assessment for Percutaneous or Surgical-Based Revascularization 17
TABLE 1.9 National Cardiovascular Database TABLE 1.10 American College of Cardiology/American Heart
Association Characteristics of Type A, B, and C Lesions 1
Variable Scoring Response Categories
Type A Lesions (high success, >85%; low risk)
Age <60 ≥60, <70 ≥70, <80 ≥80
Discrete (<10 mm length)
Weighted score 0 4 8 14 Concentric
Cardiogenic shock No Yes Readily accessible
Nonangulated segment (<45 degrees)
Weighted score 0 25 Smooth contour
Prior CHF No Yes Little or no calcification
Less than totally occlusive
Weighted score 0 5 Not ostial in location
Peripheral vascular disease No Yes No major branch involvement
Absence of thrombus
Weighted score 0 5
Type B Lesions (moderate success, 60%–85%; moderate risk)
Chronic lung disease No Yes
Tubular (10–20 mm in length)
Weighted score 0 4 Eccentric
GFR (mL/min) <30 30 to 60 60 to 90 >90 Moderate tortuosity of proximal segment
Moderately angulated segment, 45–90 degrees
Weighted score 18 10 6 0 Irregular contour
NYHA Class IV No Yes Moderate to heavy calcification
Ostial in location
Weighted score 0 4
Bifurcation lesions requiring double guidewires
PCI Status (STEMI) Elective Urgent Emergent Salvage Some thrombus present
Total occlusion less than 3 months old
Weighted score 12 15 20 38
Type B lesions are further subdivided into subtypes B1 (one type B
PCI Status (no STEMI) Elective Urgent Emergent Salvage characteristic) and B2 (two type B characteristics)
Weighted score 0 8 20 42 Type C Lesions (low success, <60%; high risk)
Diffuse (>2 cm length)
CHF, Congestive heart failure; GFR, glomerular filtration rate; NYHA, New
Excessive tortuosity of proximal segment
York Heart Association; PCI, percutaneous coronary intervention;
Extremely angulated segments (>90 degrees)
STEMI, ST elevation myocardial infarction.
Inability to protect major side branches
The risk of in-hospital mortality is derived using Fig. 1.11.
Degenerated vein grafts with friable lesions
Registry risk score from Peterson ED, Dai D, DeLong ER, et al. Contem-
Total occlusion more than 3 months old
porary mortality risk prediction for percutaneous coronary intervention:
results from 588,398 procedures in the National Cardiovascular Data Krone RJ, Laskey WK, Johnson C, et al. A simplified lesion classification
Registry. J Am Coll Cardiol. 2010;55:1923–1932. for predicting success and complications of coronary angioplasty. Reg-
istry Committee of the Society for Cardiac Angiography and Interven-
tion. Am J Cardiol. 2000;85:1179–1184.
Characteristic Points
60% 1
PCI indication
Ongoing STE-ACS 8 50%
Ongoing NSTE-ACS 6
In-hospital mortality
Stabilized after ACS 4 40%
Age 30%
≥60–70 2
≥70–80 3 20%
≥80 6
10%
Haemodynamic instability 11
Ever smoker 1 0%
Diabetes mellitus 3 0 10 20 30 40
Female gender 2
Prior stroke 2 20%
No prior CABG 4
Valvular heart disease 2
In-hospital mortality
15%
BMI < 25 kg/m2 2
TIMI flow 0/1 before PCI 2
10%
3-vessel disease 1
Left main disease 3
Proximal LAD disease 2 5%
Bifurcation lesion 2
Type-C lesion 2
0%
+ 10 12 14 16 18 20 22 24 26 28 30
Euroheart PCI score
Euroheart PCI score
Fig. 1.11 EuroHeart PCI Score–assigned integer scores. ACS, Acute coronary syndrome; BMI, body mass
index; CABG, coronary artery bypass grafting; LAD, left anterior descending; NSTE, non-ST elevation; PCI,
percutaneous coronary intervention; STE, ST elevation; TIMI, thrombolysis in myocardial infarction. (From de
Mulder M, Gitt A, van Domburg R, et al. EuroHeart score for the evaluation of in-hospital mortality in patients
undergoing percutaneous coronary intervention. Eur Heart J. 2011;32[11]:1398–1408.)
Guidewire (the European TOTAL Surveillance Study).112 B recommendation for PCI. In subjects with ULMCA disease
Each vessel segment 1.5 mm in diameter or greater (Fig. 1.13, and low to intermediate SYNTAX scores (<33), a class I LOE A
labeled 1 through 16) with a 50% or more diameter stenosis recommendation is given for CABG and IIb B for PCI. Further-
by visual estimation is awarded a multiplication factor related more, U.S. guidelines currently give surgical revascularization for
to coronary lesion location and severity (see Fig. 1.13A). Fur- ULMCA disease a class I B recommendation54,55 compared with
ther characterization of the coronary lesions leads to the addi- a class I A recommendation in previous guidelines.113
tion of more points (see Fig. 1.13B), which includes features
of total occlusions (duration, length, blunt stumps, presence
of bridging collaterals or side branch), bifurcation (Medina
Functional SYNTAX Score
classification)111 or trifurcation (number of diseased branches PCI guided by the assessment of the functional significance of a
involved), side-branch angulation, aortoostial lesions, severe lesion has been shown to improve clinical outcomes.114–116 The
tortuosity, lesion length greater than 20 mm, heavy calcifica- functional SYNTAX score uses the principle of the functional
tion, thrombus, and diffuse or small-vessel disease. An online assessment of coronary lesions to determine the SYNTAX
SYNTAX score algorithm45 automatically summates each of score, rather than the angiographic determination of the
these features to calculate the total SYNTAX score. SYNTAX score based on visual assessment, as is undertaken
Within the SYNTAX trial,13 the distribution of the SYN- in conventional SYNTAX score calculations. In a retrospec-
TAX score was found to be normal (Gaussian) in the random- tive subanalysis of almost 500 patients (n = 497) from the FFR-
ized PCI and CABG populations with the curves almost being guided arm of the Fractional Flow Reserve Versus Angiography
superimposable on each other (eFig. 1.2). When the scores of the for Multivessel Evaluation (FAME) study, the primary benefit
randomized SYNTAX population were divided into tertiles, the appeared in reclassifying higher-risk groups into lower-risk cat-
upper boundary of the lowest tertile was 22 (low risk), the second egories without any adverse sequelae in terms of MACEs and
tertile ranged from 23 to 32 (intermediate risk), and the lower death or MI at 1 year.36
boundary for the highest tertile was equal to or greater than 33 It should be emphasized that subjects in the FAME study
(high risk). had substantially less complex CAD (mean SYNTAX score
Based primarily on the results of the SYNTAX trial,13–15 cur- 14.8 ± 6.0) compared with the PCI arm of the SYNTAX trial
rent European revascularization guidelines52 give subjects with (mean SYNTAX score 28.4 ± 11.5) and that subjects with LM
3VD and low SYNTAX scores (0 to 22) a class I recommenda- CAD were not investigated. Prospective validation studies of
tion, level of evidence (LOE) A, for CABG and a class IIa LOE the functional SYNTAX score in complex CAD are ongoing
CHAPTER 1 Individualized Assessment for Percutaneous or Surgical-Based Revascularization 19.e1
25
1
CABG RCT
20 PCI RCT
CABG registry
PCI registry
% of patients 15
10
0
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84
SYNTAX score
eFig. 1.2 The distribution of the SYNTAX score in the randomized and nested registry percutaneous coronary intervention (PCI) and coronary artery
bypass grafting (CABG) populations from the SYNTAX trial. Note how the distributions for both the randomized PCI and CABG populations are
almost superimposable on each other, whereas the nested registries are shifted to the right.13,48 RCT, Randomized controlled trials. (Reprinted with
permission from Serruys PW, Onuma Y, Garg S, et al. Assessment of the SYNTAX score in the SYNTAX study. EuroIntervention. 2009;5[1]:50–56.)
20 SECTION I Patient Selection
1.0 + 1.0
+
Cumulative survival rate free from
MACE (%)
0.6 0.6
+
0.4 0.4
0.00 5.00 10.00 15.00 20.00 25.00 30.00 0 500 1000 1500 2000
A Days after stenting procedure (d) B Days after stenting procedure (d)
1.0
1.0 +
+ +
Cumulative survival rate free from
++ +
+
0.6
0.6 MACE (%)
SYNTAX-lower vs. SYNTAX-higher or SYNTAX-lower vs. SYNTAX-higher or
0.4 SYNTAX-intermediate, log rank 0.4 SYNTAX-intermediate, log rank
(mantel-cox): P = .013 (mantel-cox): P < .001
0.00 5.00 10.00 15.00 20.00 25.00 30.00 0 500 1000 1500 2000
C Days after stenting procedure (d) D Days after stenting procedure (d)
Fig. 1.12 Comparison of freedom from in-hospital (A) and overall major adverse cardiac events (MACEs, B)
survival between New Risk Stratification (NERS) groups and in-hospital (C) and overall MACEs (D) survival
among SYNTAX groups. (From Chen SL, Chen JP, Mintz G, et al. Comparison between the NERS [New Risk
Stratification] score and the SYNTAX (Synergy Between Percutaneous Coronary Intervention With Taxus and
Cardiac Surgery) score in outcome prediction for unprotected left main stenting. JACC Cardiovasc Interv.
2010;3[6]:632–641.)
at the time of writing. Namely the SYNTAX II trial, in which variables—age, serum creatinine, and LVEF—for assessing
a noninvasive assessment of the functional SYNTAX is also operative mortality risk in elective cardiac operations (ACEF
being assessed with fractional flow reserve derived from com- score, see Fig. 1.8). Based on the law of parsimony, or “Occam’s
puted tomography angiography,117 and the FAME 3 trial razor,” whereby a simple model can explain a phenomenon
[NCT02100722]).118 with the same level of accuracy as complex models, ACEF was
shown to be least comparable to the EuroSCORE (composed
of 17 variables)58,59 in predicting in-hospital mortality after
Augmenting the Anatomic SYNTAX Score With CABG.62,89
Clinical Factors and the Personalization of Decision The three risk factors used in ACEF are natural continuous
variables that are objectively defined and not subject to personal
Making: Development of SYNTAX Score II estimation (e.g., whether the patient has diabetes or extracardiac
Since the anatomic SYNTAX score was developed, limitations of arteriopathy). In addition, the variables of ACEF were known
this scoring system to aid decision making between CABG and PCI independent risk factors for mortality, and it was subsequently
became evident—namely, the lack of clinical variables and lack of shown that the end-organ manifestations of the risk factor as
a personalized approach to decision making. Following is a brief identified in ACEF were more important for predicting long-
overview of the “development phase” leading to SYNTAX score II, term prognosis rather than the actual presence of the risk fac-
which was designed to overcome these limitations (see Table 1.6). tor.6,88,119,120
Age, Serum Creatinine, and Ejection Fraction Clinical SYNTAX Score/Logistic Clinical SYNTAX Score
As described earlier, Ranucci and devel- colleagues62,88,89 Based on the principle of ACEF, the clinical SYNTAX
oped a simple risk score that consisted of only three clinical score90,121,122 and subsequently the logistic clinical SYNTAX
CHAPTER 1 Individualized Assessment for Percutaneous or Surgical-Based Revascularization 21
1
Points Score 20
0
Sum score
0 5 10 15 20
Sum score
eFig. 1.3 The Logistic Clinical SYNTAX score. CrCl, Creatinine clearance; LV, left ventricular. aSYNTAX-like patient defined as fulfilling the enrollment
criteria for the SYNTAX All-Comers trial: that is, left main stem (isolated or associated with 1-, 2-, or 3-vessel disease) or 3-vessel disease alone.
(From Farooq V, Vergouwe Y, Généreux P, et al. Prediction of 1-year mortality in patients with acute coronary syndromes undergoing percutaneous
coronary intervention: validation of the logistic clinical SYNTAX (Synergy Between Percutaneous Coronary Interventions With Taxus and Cardiac
Surgery) score. JACC Cardiovasc Interv. 2013;6(7):737–745.)
21.e2 SECTION I Patient Selection
Man without COPD or poor LVEF (<30%) Woman without COPD or poor LVEF (<30%)
Man with COPD Woman with COPD
Man with poor LVEF (<30%) Woman with poor LVEF (<30%)
LMS
Age <62 years SYNTAX score <23 Age <62 years SYNTAX score <23–32 Age <62 years SYNTAX score >32
3 Favours CABG Favours CABG Favours CABG
2
0
-1
-2
Favours PCI Favours PCI Favours PCI
-3
Age 62–70 years SYNTAX score <23 Age 62–70 years SYNTAX score 23-32 Age 62–70 years SYNTAX score >32
3 Favours CABG
Favours CABG Favours CABG
2
Log hazard PCI
0
-1
-2
Favours PCI Favours PCI Favours PCI
-3
Age >70 years SYNTAX score <23 Age >70 years SYNTAX score <23–32 Age >70 years SYNTAX score >32
3
Favours CABG Favours CABG Favours CABG
2
Log hazard PCI
0
-1
-2
Favours PCI Favours PCI Favours PCI
-3
A 3VD
Age <62 years SYNTAX score <23 Age <62 years SYNTAX score 23–32 Age <62 years SYNTAX score >32
3
Favours CABG Favours CABG Favours CABG
2
Log hazard PCI
0
-1
-2
Favours PCI Favours PCI Favours PCI
-3
Age 62–70 years SYNTAX score <23 Age 62–70 years SYNTAX score <23–32 Age 62–70 years SYNTAX score>32
3
Favours CABG Favours CABG Favours CABG
2
Log hazard PCI
0
-1
-2
Favours PCI Favours PCI Favours PCI
-3
Age >70 years SYNTAX score <23 Age >70 years SYNTAX score <23–32 Age >70 years SYNTAX score>32
3 Favours CABG Favours CABG Favours CABG
2
Log hazard PCI
1
0
-1
-2
Favours PCI Favours PCI Favours PCI
-3
-3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 3
B Log hazard CABG Log hazard CABG Log hazard CABG
eFig. 1.4 Collective effect of the anatomic SYNTAX score and other anatomic and clinical variables on mortality predictions in the left main stem
(LMS, A) and three-vessel disease (3VD, B) cohorts of the SYNTAX trial (SYNTAX score II). Scatter plots are for illustrative purposes only. The diago-
nal line represents identical mortality predictions for coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI). Individual
predictions plotted to the left of the diagonal line favor CABG (actual percentages shown in top left corner) and to the right favor PCI (actual percentages
shown in bottom right corner). COPD, Chronic obstructive pulmonary disease; LVEF, left ventricular ejection fraction. (From Farooq V, van Klaveren D,
Steyerberg EW, et al. Anatomical and clinical characteristics to guide decision making between coronary artery bypass surgery and percutaneous coro-
nary intervention for individual patients: development and validation of SYNTAX score II. Lancet. 2013;381[9867]:639–650.)
22 SECTION I Patient Selection
12a 7 10
13 +3.5
2 12b
14 10a
16 +2.5
16c 14a
3 16b +1.5
14b 8
4 16a
+1
12a 7 10
13
2 12b
14 10a
14a
3 14b
8
A 15
Dise ent #
segm
on
lusi
ased
occ l
Tota
Do
m ion
ina c at
nc
e if ur
Tr
Diffuse small
SYNTAX Bifurcation
score
us Aor
mb ta o
o stia
Thr l
Se tuos
on
tor
ve ity
Length
ati
ca avy
re
fic
lci
He
B
Fig. 1.13 Coronary tree segments and their importance in supplying blood flow to the left ventricle (vessel
segment weighting factors; Leaman score154,155) based on the presence of a right or left dominant system
(A).109 A multiplication factor of two is used for nonocclusive lesions (50% to 99% diameter stenosis) and five
for occlusive (100% diameter stenosis) lesions. For example, a nonocclusive stenotic proximal left anterior
descending (LAD) coronary artery lesion (segment 6) would have a weighting factor of 3.5 × 2 (7 points) and
an occlusive proximal LAD lesion a weighting factor of 3.5 × 5 (17.5 points). Other adverse lesion character-
istics considered in the SYNTAX score have an additive value (B). #, Segment number. (Images courtesy the
SYNTAX Trial Investigators.)
CHAPTER 1 Individualized Assessment for Percutaneous or Surgical-Based Revascularization 23
–1
–2
0 20 40 60 60 70 80 90 0 30 60 90 120 10 20 30 40 50 60
–1
–2
3VD LMS F M No Yes No Yes
Diabetes
2
1
Log HR
–1
–2
No Yes
Fig. 1.15 Predictor effects for coronary artery bypass surgery (CABG) and percutaneous coronary intervention
(PCI) in the SYNTAX score II. These are represented visually as a log hazard ratio (HR) for CABG and PCI on
the y-axis for each predictor. Each predictor is expressed on the x-axis continuously (upper) or categorically
(lower), for a person of mean baseline characteristics. Diabetes is included (highlighted in red) to illustrate its
absence of interaction when included in the analyses. Note the differing gradients of the hazards for PCI and
CABG, leading to the hazards crossing at an anatomic SYNTAX score of 15. At this crossover point of haz-
ards, the mortality risk is comparable between CABG and PCI. This threshold of crossover of hazards will vary
according to the level of other variables, namely, being lower for female sex, reduced left ventricular ejection
fraction (LVEF) and younger age, and higher for chronic obstructive pulmonary disease (COPD), unprotected
left main coronary artery disease, and older age. As both peripheral vascular disease (PVD; P = 1.00) and
diabetes (P = .67) lacked an interaction effect, as indicated by almost parallel HRs (i.e., a comparable increase
in mortality risk), their presence would have no impact on decision making between CABG and PCI. CrCl, Cre-
atinine clearance. (From Farooq V, van Klaveren D, Steyerberg EW, et al. Anatomical and clinical characteristics
to guide decision making between coronary artery bypass surgery and percutaneous coronary intervention for
individual patients: development and validation of SYNTAX score II. Lancet. 2013;381[9867]:639–650.)
with the PCI strategy adopted in SYNTAX I (Fig. 1.21A). More- (anatomic SYNTAX score >22), in which current revasculariza-
over, clinical outcomes were similar in 3VD patients with a tion guidelines support CABG (see Fig. 1.21B). In addition, the
low anatomic SYNTAX score (≤22), in which revascularization SYNTAX II PCI strategy demonstrated equipoise to CABG at
guidelines currently support PCI or CABG-based revascular- 2-year follow-up for MACCE (exploratory end point) (see Fig.
ization52–55—compared with more anatomically complex 3VD 1.21C). Longer (5 year) follow-up of SYNTAX II is awaited.
LMS 3VD
3
Predicted 4-year mortality Predicted 4-year mortality
favors CABG: 50·1% favors CABG: 84·2%
(11·5% of predictions (40·7% of predictions
lie outside 95% CI) lie outside 95% CI)
2 1
1
Log hazard PCI
–1
–2
Predicted 4-year mortality for Predicted 4-year mortality favors Predicted 4-year mortality for Predicted 4-year mortality favors
CABG and PCI lying within PCI: 49·9% (8·8% of predictions CABG and PCI lying within PCI: 15·8% (0·5% of predictions
95% CI: 79·7% lie outside 95% CI) 95% CI: 58·8% lie outside 95% CI)
–3
–3 –2 –1 0 1 2 3 –3 –2 –1 0 1 2 3
Log hazard CABG
Log hazard CABG
LMS, SYNTAX score <23 LMS, SYNTAX score 23–32 LMS, SYNTAX score >32
3 37·2% (2·7%) 38·3% (10·2%) 68·2% (19·2%)
2
Log hazard PCI
–1
–2
3VD, SYNTAX score <23 3VD, SYNTAX score 23–32 3VD, SYNTAX score >32
3 70·9% (19·2%) 87·1% (37·8%) 95·0% (68·1%)
2
Log hazard PCI
–1
–2
100
80
4-year mortality (%)
60 58.4%
40
31.5%
20
15.1%
6.8%
1.3% 3%
0.6%
0
0 20 40 60 80 100
Total points
CABG PCI
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Points
SYNTAX score
0+ 0 10 20 30 40 50 60
Age (years)
40 50 60 70 80 40 50 60 70 80
CrCl (mL/min)
90 60 30 90 60 30
LVEF (%)
50 40 30 20 50 40 30 20
1 0
Left main
0 1
M F
Sex*
F M
1 1
COPD
0 0
1 1
PVD
0 0
Fig. 1.17 The SYNTAX score II nomogram for bedside application. Total number of points for eight factors can
be used to accurately predict 4-year mortality for the individual patient proposing to undergo coronary artery
bypass grafting (CABG) or percutaneous coronary intervention (PCI). COPD, Chronic obstructive pulmonary
disease; CrCl, creatinine clearance; LVEF, left ventricular ejection fraction; PVD, peripheral vascular disease.
*Because of the rarity of complex coronary artery disease in premenopausal women, mortality predictions in
younger women are predominantly based on the linear relation of age with mortality. The differences in mortality
predictions in younger women between CABG and PCI will therefore be affected by larger 95% CIs than those
in older women. (From Farooq V, van Klaveren D, Steyerberg EW, et al. Anatomical and clinical characteristics
to guide decision making between coronary artery bypass surgery and percutaneous coronary intervention for
individual patients: development and validation of SYNTAX score II. Lancet. 2013;381[9867]:639–650.)
Tools for Assessment of Completeness of trial, complete revascularization was defined as the treatment of any
lesion with more than 50% diameter stenosis in vessels 1.5 mm or
Revascularization larger as estimated on the diagnostic angiogram during the local
Interpreting the long-term prognostic impact of incomplete heart team conference and deemed appropriate for revasculariza-
revascularization in patients with complex CAD has historically tion.144,145 In SYNTAX, incomplete revascularization was shown
remained difficult.143 The lack of standardized definitions of to be linked to adverse long-term clinical outcomes, including
incomplete revascularization has confounded this issue and has mortality, in surgical and percutaneously treated patients (Fig.
made comparisons between studies difficult. In the SYNTAX 1.22). The residual and post-CABG SYNTAX scores were
CHAPTER 1 Individualized Assessment for Percutaneous or Surgical-Based Revascularization 27
0.20
Favors CABG:22.1 % 1
With P < .05: 4.3% With P > .05:
55.2 %
0.15
4-year mortality PCI
0.10
0.05
Favors PCI:77.9%
With P < .05: 40.4%
0.00
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
4-year mortality CABG
Fig. 1.18 First 1000 4-year mortality simulations of the EXCEL trial using the SYNTAX score II. Each dot represents one simulated trial mortality in
both randomization arms based on individual predictions. The diagonal line represents identical mortality for coronary artery bypass grafting (CABG)
and percutaneous coronary intervention (PCI). A dot plotted to the left of the diagonal line favors CABG (actual percentages shown in top left cor-
ner), and to the right favours PCI (actual percentages shown in bottom right corner). Simulated trials with a significant (P ≤ .05) mortality difference
between CABG and PCI are colored black (actual percentage shown in parentheses in respective corners). Simulated trials with a nonsignificant (P
> 0.05) mortality difference between CABG and PCI are colored gray. (Reproduced with permission from Campos CM, van Klaveren D, Farooq V,
et al.; EXCEL Trial Investigators. Long-term forecasting and comparison of mortality in the Evaluation of the Xience Everolimus Eluting Stent vs. Coro-
nary Artery Bypass Surgery for Effectiveness of Left Main Revascularization (EXCEL) trial: prospective validation of the SYNTAX Score II. Eur Heart J.
2015;36[20]:1231–1241.)
designed to quantify the degree of incomplete revascularization and it provides an objective quantitative measure of the degree
and allow for a threshold of incomplete revascularization that and complexity of residual stenosis after revascularization. More
would not have a negative impact on long-term clinical outcomes proximal CAD scores higher on the residual SYNTAX score
(i.e., reasonable incomplete revascularization). because this is dependent on the vessel segment weighting as
previously discussed (see Fig. 1.13), particularly if the obstructive
disease is more complex.146,147
Residual SYNTAX Score Généreux and colleagues146 first demonstrated that a residual
The residual SYNTAX score is based on the principle of being SYNTAX score greater than 8 after PCI was associated with
a measure of the myocardial ischemic burden dependent on the adverse 1-year mortality in a post hoc analysis of the ACUITY
location of the coronary disease, its importance in supplying trial, which consisted of subjects with moderate to high-risk ACS
blood to the myocardium, and the anatomic complexity (e.g., cal- undergoing PCI and substantially less complex CAD (median
cification, bifurcation, long lesion) associated with the obstruc- SYNTAX score 9.0, interquartile range [IQR] 5.0 to 16.0) com-
tive disease. The residual SYNTAX score is essentially the pared with those in the SYNTAX trial (median SYNTAX score
anatomic SYNTAX score recalculated after the PCI procedure, 28, IQR 20.0 to 36.0).
28 SECTION I Patient Selection
SYNTAX Trial II
Inclusion: All-Comers, angiographic, de-novo 3-vessel disease without
left main involvement (visual % diameter stenosis ≥50%)
Screening according to
SYNTAX Score II
Fig. 1.19 Study flow chart of the SYNTAX II trial—using the state-of-art SYNTAX II percutaneous coronary
intervention (PCI) strategy of appropriate patient selection with the SYNTAX score II, newer-generation stent
platform with a biodegradable polymer, contemporary CTO revascularization strategies and functional and
intravascular ultrasound–guided PCI. CABG, Coronary artery bypass grafting; CTO, chronic total occlusion;
FFRCT, fractional flow reserve derived from computed tomography; MSCT, multislice computed tomography.
(From Escaned J, Banning A, Farooq V, et al. Rationale and design of the SYNTAX II trial evaluating the short to
long-term outcomes of state-of-the-art percutaneous coronary revascularisation in patients with de novo three-
vessel disease. EuroIntervention. 2016;12[2]:e224–e234.)
The residual SYNTAX score was subsequently validated in implantation, functionally incomplete revascularization was associ-
the randomized, all-comers SYNTAX trial, which comprised sub- ated with a higher rate of 2-year MACEs compared with functional
jects with complex CAD (ULMCA or de novo 3VD) at the final complete revascularization (functional incomplete revascularization
5-year follow-up.147 The previous findings, of residual SYNTAX vs. functional compete revascularization, 14.6% vs. 4.2%; hazard
score greater than 8 being associated with adverse long-term clini- ratio: 4.09; 95% CI: 1.82 to 9.21; P < .001). In addition, functional
cal outcomes in the ACUITY trial,146 were found to be of equal incomplete revascularization was shown to be an independent pre-
importance in SYNTAX patients who underwent 5-year follow-up. dictor of MACEs (adjusted hazard ratio: 4.17; 95% CI: 1.85 to 9.44;
Notably, as the baseline SYNTAX score increased, the frequency of P < .001), with the newly devised functional residual SYNTAX score
a residual SYNTAX score greater than 8 increased in unison, with appearing to better identify lower and high-risk patients. One of the
an associated increase in long-term mortality (Fig. 1.23). In addi- main limitations of this study was that predominantly low anatomic
tion, progressively higher residual SYNTAX scores were shown complex disease patients were recruited, with only 28.8% having
to be a surrogate marker of sicker patients,145 with greater base- 3VD. At the time of writing, prospective validation of the functional
line clinical comorbidity and anatomic complexity and consequent residual SYNTAX score is awaited from further studies investigating
adverse long-term clinical outcomes, including all-cause mortality. more complex CAD, including the ongoing SYNTAX II trial.
Stratified analyses in the powered subgroups of ULMCA disease In summary, the residual SYNTAX score (with possible func-
and medically treated diabetes showed that a residual SYNTAX tional assessment) allows for the quantification of the degree of
score greater than 8 was associated with adverse long-term clinical revascularization and for determination of an objective level of
outcomes, including mortality. Stratified analyses in subjects with reasonable incomplete revascularization,143 whereby a threshold
reduced LVEF also showed the results to be equally applicable, value could be determined that would not have a negative impact
whereas in subjects with total occlusions, a more modest effect was on long-term mortality and other clinical outcomes.
shown that did not reach statistical significance. The latter perhaps
implied that a higher level of a residual SYNTAX score was required
in patients with total occlusions with MVD and/or that appropriate
Post–Coronary Artery Bypass Grafting SYNTAX Score
viability assessment was required to ensure that revascularization of The CABG equivalent of the residual SYNTAX score, the post-
the total occlusion was appropriate and clinically justified.148 CABG SYNTAX score, has been shown to be linked to adverse
Subsequently, several validation studies from registries have fur- 5-year clinical outcomes, including mortality, in the angiographic
ther supported the use of the residual SYNTAX score in complex substudy of the SYNTAX trial (SYNTAX–LE MANS [Left Main
CAD.149–152 Specifically in one registry of 1043 patients with MVD Coronary Artery Stenting]; Fig. 1.24).154,155 Because of the inher-
and at least one CTO,151 a higher cutoff value for the residual SYN- ently differing mechanisms of treatment of CAD with CABG and
TAX score (≤12) value was demonstrated, in which a reasonable PCI, calculation of the residual SYNTAX score (i.e., the burden
level of incomplete revascularization was achieved that had similar of coronary disease removed by PCI) differs from that of the
outcomes to complete revascularization. These latter findings thus post-CABG SYNTAX score (i.e., coronary disease bypassed with
corroborated the effect of total occlusions on the residual SYNTAX a graft). The basic principle of the post-CABG SYNTAX score is
score in the original SYNTAX trial as discussed in the last paragraph. that it deducts points from the “native” baseline SYNTAX score
More recently, the prospect of a functionally guided incomplete based on the level of “protection” conferred by the bypass grafts,
revascularization has been proposed.153 Notably in 385 patents through deduction of the vessel-segment weighting (Leaman
who underwent three-vessel functional assessment following stent score, see Fig. 1.13)109 that the bypass graft provides (eFig. 1.5).
CHAPTER 1 Individualized Assessment for Percutaneous or Surgical-Based Revascularization 28.e1
20
0
A SYNTAX II SYNTAX I SYNTAX II SYNTAX I
CTO revascularisation in
CTO PCI procedural success rate in
SYNTAX II and SYNTAX I
SYNAX II:87% (n = 108)
P < .0001
100%
80%
n = 94
Success
60%
Failed
87%
n = 14 40%
53%
20%
0%
B SYNTAX II CTO PCI SYNTAX II SYNTAX I
84.1%
40%
84.1% 76.4%
20%
IVUS 4.8%
no IVUS 0%
SYNTAX II SYNTAX I
Fig. 1.20 Impact of the SYNTAX II PCI strategy—coronary physiology (A), chronic total occlusion (CTO) revascularization (B), and intravascular ultrasound
(IVUS) (C)—in the SYNTAX II trial135–137 compared with the original SYNTAX I trial. FFR, Fractional flow reserve; PCI, percutaneous coronary intervention.
Another random document with
no related content on Scribd:
whenever we stopped Mansing pathetically conversed with the dog, who
seemed almost to understand all the coolie was telling him.
It was rather a puzzle to me why this dog followed us so long, for we had so little
food that we could but seldom spare him any. He slept near us at night with his
head on the lap of one of us, and during the march he showed quite sporting
instincts by chasing antelopes and kiang (wild horse) when we encountered
herds of hundreds of them. Curiously enough, when we entered a Tibetan
encampment he always avoided being seen in our company. It seemed almost
as if he realised that we were not welcome guests in the country, and feared the
consequences. Possibly [139]he only temporarily left us to see what he could pick
up in the way of food, but whenever we came across him in the encampment, he
never would show signs of recognition, much less of affection, as was the case
when he would rejoin us some miles beyond on the march, when he made
ample efforts to reingratiate himself. He seemed almost to want to express:
“Sorry I had to cut you in the encampment, but I really had to!”
At last the day came when we were captured, and underwent several kinds of
tortures, as I have already described in In the Forbidden Land. The dog had
vanished, and, to tell the truth, we did not give him much of a thought, as we
were somewhat concerned about ourselves.
Tibetan encampments have no great interest except for the peculiar shape of
the black tents—a pattern of shelter most suitable for the climate of their country.
The two sides of the tent are separate, and when the tent is put up it leaves an
aperture all along its highest ridge. This is for various reasons. First, because
the Tibetans light fires inside their tents, and an opening is necessary to let the
smoke out; also as a means of ventilation, the cold air not penetrating so quickly
as when it comes in at the sides, owing to the warmed atmosphere inside. The
black tents are woven of a coarse and waterproof fabric of yak hair. Through the
slit at the top generally protrude the props of the matchlocks bundled against
one of the tent poles.
Every man in Tibet owns one of these weapons, and is considered a soldier in
time of war.
Interior of a Tibetan Tent, showing Churn for mixing Tea with Butter
The inside of a large Tibetan tent is quaint enough when you have reached it by
skipping over masses of dirt and refuse which surround its outside. Only, when
you peep in, the odour is rather strong of the people, old and young, all since
[141]birth innocent of washing, and the smell of badly-prepared skins, and stores
of chura (cheese). Nor must I forget to mention the wall of yak-dung erected
right round the tent inside to serve the double purpose of protection against the
wind where the tent meets the ground, and of fuel, being gradually demolished
to feed the double mud-stove erected in the centre of the tent. Mud alone is also
occasionally used for the inside wall.
As you know, dung is practically the only fuel obtainable in the highest parts of
Tibet, although occasionally a few low shrubs are to be found. The fuel is
constantly collected and conveyed from one camp to the next, when changing in
order to find more suitable grazing for the sheep and yaks.
The centre mud-stove is built according to the most practical notions to make it
draw properly, and upon it can nearly always be seen one or two large raksangs,
brass vessels in which brick-tea is being stewed and stirred with a long brass
spoon. But the operation of tea-making is rather complicated in Tibet. After the
leaves have been stewed long enough the liquid is poured into a dongbo, or
cylindrical wooden churn, in which have been deposited several balls of butter
with copious [142]sprinkling of salt. A piston which passes through the movable
lid is then vigorously set in action, and when well stirred and steaming the
mixture is served all round and avidly drunk in wooden bowls, one of which
every one carries about the person. Tsamba, a kind of oatmeal, is frequently
mixed with the tea in the bowls, where it is made into a paste with the fingers.
A Little Boy learning to Pray
No matter how much non-Tibetan folks may find merriment in the idea of tea
being brewed with butter and salt, there is no doubt that for a climate like Tibet it
is “the drink” par excellence. It warms, nourishes, and is easily digested. I very
often indulged in the luxury myself, when I could obtain butter, only, my digestion
working rather rapidly owing to the amount of roughing we daily endured, I left
out the salt so that I should not digest the mixture too quickly.
The richer owners of tents generally have a sort of folding shrine, with one or
more images of Buddha, which occupies the place of honour in the tent.
Numerous brass bowls and ornaments are displayed in front of these images
and also offerings of tsamba and butter. Wicks, burning in butter, are
occasionally lighted around and upon the shrine. Decrepit old women seem to
[143]spend most of their time revolving their prayer-wheels and muttering prayers
in front of these altars, and when occasion arises thus teaching little children to
do the same. The younger folk, too, are very religious, but not to the fanatical
extent of the older ones.
It is quite amusing to see little mites—children are always quaint in every country
—try to master the art of revolving the prayer-wheel. It must be revolved from left
to right, to pray in the proper fashion,—not that if you revolved it the other way
you would necessarily be swearing, only, according to the laws of Tibetan
Buddhism, prayers spun in the wrong direction would have no effect and bring
no benefit. In a similar way circumambulations, either round hills for pilgrimages,
or round a tent, or round a sacred lake, must always follow a similar direction to
the revolving of the prayer-wheel.
In Lhassa and many other sacred places fanatical pilgrims make these
circumambulations, sometimes for miles and miles, and for days together,
covering the entire distance lying flat upon their bodies, then placing the feet
where the head was and stretching themselves full length. Inside temples a
central enclosure is provided, round [144]which these circumambulations are
performed, special devotions being offered before Buddha and many of the
other gilt or high-coloured images which adorn the walls of the temple.
As can be seen by the coloured plate illustrating one of these scenes, from the
ceiling of the temple hang hundreds of long strips, Katas, offered by pilgrims to
the temple and becoming so many flying prayers when hung up—for mechanical
praying in every way is prominent in Tibet. There is, after all, no reason why
praying should not be made easy like everything else. Thus, instead of having to
learn by heart long and varied prayers, all you have to do is to stuff the entire
prayer-book (written on a roll in Tibet) into the prayer-wheel, and revolve it while
repeating as fast as you can go these four words: “Om mani padme hum,”—
words of Sanscrit origin and referring to the reincarnation of Buddha from a lotus
flower, literally “O God, the gem emerging from the lotus flower.”
Interior of Tibetan Temple
The temples of Tibet, except in Lhassa itself, are not beautiful in any way—in
fact, they are generally very tawdry and dirty. The attention of the pilgrims is
directed to a large box, or often a big bowl, where they may deposit whatever
[145]offerings they can spare, and it must be said that their religious ideas are so
strongly developed that they will dispose of a considerable portion of their
money in this fashion.
Personally—and I am glad that the few men [146]who know Tibet from personal
knowledge and not from political rivalry agree with me—I believe that the
intrigues of the Lamas with Russia are absolute nonsense. Tibet, it must be
remembered, was not forbidden to Englishmen only, but to everybody from
every side, whether native or white, certain Nepalese and Chinamen, only,
having the privilege of entering the country. It was a fight against Western ways
in general which the Lamas were carrying on, quite successfully owing to the
geographical position of their country, and the natural difficulties of reaching it,
and not a fight against one race more than another. The accounts of the Lhassa
Mission to the Czar were possibly the best diplomatic practical jokes which have
been played upon this credulous country; and the mythical and much-feared
Dorjeff is possibly—at least as far as power is concerned—nothing more than
the creation of hysterical Anglo-Indian officials who, everybody knows, seem to
see the treacherous hand of Russia in everything.
Tibetan Women weaving
Agriculturally, as I have stated, nothing grows there; no very wealthy mines have
so far been discovered, the only mines that are plentiful being of borax, which
has not sufficient market value to pay for the expensive carriage from Tibet to
the coast. Regarded as a climate for a sanatorium for our sick soldiers in India—
for which Tibet is frequently recommended by Anglo-Indian papers—I believe
that such an establishment would be a very quick way of disposing altogether of
all the sick men sent there. And as for such gigantic schemes as the
construction of railways, say from India to the upper waters of the Yangtze-
Kiang, or to Pekin, the expense of taking a railway over the Himahlya range and
keeping it in working order during the wintry months—nine out of twelve—would,
I think, never be remunerative. In Tibet itself the construction of a railway would
be comparatively easy, as great stretches of the country are almost flat. Stations
of imported fuel would have to be provided for the entire distance across Tibet,
and the engines would have to be constructed specially to suit the great altitude.
[148]
For trade and commerce with the natives themselves, the population of the
country is so small, so deplorably poor and so lacking in wants, and the country
is so large that, personally, I do not see how any large commercial venture in
such a country can turn out successful. It is very difficult to get money where
there is none. Small native traders, of course, can make small profits and be
satisfied. Besides, the intercourse between Tibet and the neighbouring
countries, particularly those to the south, can only take place with comfort during
three months of the summer when the high snow-passes are open.
So that, much as I would like to see Tibet open in a proper way to travellers, I
cannot quite understand the necessity of the Government spending millions of
money and butchering thousands of helpless and defenceless natives in a
manner most repulsive to any man who is a man, and of which we can but be
ashamed—and all this to obtain a valueless commercial treaty. It is true, the
Tibetans had been very impudent in every way on our frontier, but for this we
only have to blame ourselves and our incompetent officials. If, instead of giving
way to their bluff, we had kept a firm hand, matters would have been different.
Tibetan Women cleaning Wool
[149]
Even in the case of my capture and torture on my first expedition into Tibet I
never had a feeling of resentment towards the Tibetans for what they did to me.
It was very exciting and interesting for me to endeavour to reach their sacred
city, but I did so at my own risk and against their repeated warnings and threats,
and I got nothing more for it in the end than I expected, in fact, bad as it was,
considerably less. Highly amusing as it was to me to give them endless trouble,
it was undoubtedly equally enjoyable to them to torture me, when once they
succeeded in effecting my capture. Possibly, if I now have any feelings at all
towards the Tibetans, it is a feeling of gratitude towards them for sparing my life
in the end, which, by the way, they came within an ace of taking as they had
promised to do.
As a punishment for what they did to me—because, after all, my men and I
suffered a great deal more than the average man could stand—the Government
of India practically ceded, as we have said, all the rights to Tibet of an immense
district of British territory at the frontier. Can you blame the Tibetans for doing
worse if they had a chance? [150]
[Contents]
CHAPTER XIV
In heart and soul the Tibetan is a sportsman; but if you look for grace in his
movements you will be sorely disappointed. Indeed, more fervour and
clumsiness combined are hardly to be paralleled anywhere. Perhaps the Tibetan
is seen to advantage on his pony, and some of his feats on the saddle I will here
describe.
A Lama Standard-Bearer
Horse races are quite a favourite form of amusement, and are run in a sensible
manner. Only two ponies at a time go round the course, the final race being run
between the winners of the two best heats. Praying is usually combined, in
some form or other, with everything people do in Tibet, and so even races are
run round the foot of an isolated hill or around an encampment of tents; for, as
you know, circumambulation of any kind, if in the right direction, is equivalent to
prayer, and pleases God. Thus, just as with their prayer-wheels, a rotatory
[151]motion is kept up from left to right, so races are run in the same way from the
standpoint of the spectator.
The heavy sheepskin coats worn by the Tibetans are some protection when the
lash is applied, and the pain inflicted is not always in proportion to the noise
made by the blow; but such is not the case when they catch one another across
the face. [152]
The winner is presented with a kata by the umpire—a high Lama or a military
officer, a most picturesque creature in a brilliant red coat and fluffy hat, who has
a peculiar standard with hundreds of long, vari-coloured strips of cloth, or flying
prayers. Sitting on a handsome pony, with gaudy harness of green leather inlaid
with brass, a valuable Chinese rug upon the saddle, and many tinkling bells
round the pony’s neck, the umpire and his pony certainly produce a gay
ensemble. This gentleman takes himself very seriously, and seldom
condescends to smile.
The kata, or “scarf of love and friendship,” which is given to the winner is a long
piece of silk-like gauze, the ends of which have been trimmed into a fringe. As I
have elsewhere described at greater length, these katas play quite an important
part in the social intercourse of Tibetans. They can be purchased or obtained
from the Lamas of any monastery, or where no monastery exists the natives
manufacture them themselves, for they are constantly needed. No gift can be
sent nor accepted without “a veil of friendship” accompanying it, and no stranger
ever enters a tent without offering, with outstretched hands, a kata, which he
quickly lays at the feet of [153]his host. Diminutive katas are enclosed in letters;
sweethearts exchange katas on every possible occasion—until they are actually
married. Polyandry being prevalent in Tibet, when one of the several husbands
returns to his wife after the customary absence, he never fails to bring a kata
with him. Not to offer a kata to an honoured visitor is as palpable a breach of
manners, and as great a slight as can possibly be offered in the Forbidden Land.
Necessarily, when a kata has been blessed by the Lamas, or is won in a race
before high officials, it has additional value, and these simple folks value it more
than a gift of money or food. It is stored away in the tent among the heirlooms,
and is handed down to posterity.
A slightly more difficult feat, very common in a similar form in most countries, is
the picking up of a kata by horsemen at full gallop. One horseman, a high
official, revolves the kata seven times in the wind, and then darts full gallop in
one direction, followed by twenty, thirty, or more horsemen riding wildly, and
each trying to push his neighbours out of the way. The official, some thirty yards
ahead, flies the kata in the wind, and when fancy takes him lets it drop out of his
hand. The kata eventually settles on the ground, and the horde [154]of riders
gallops away from it, yelling and quarrelling. At a signal from the officer the
horsemen turn round and make a dash for the scarf, towards which all the
ponies are converging. Clinging to the saddle with one hand and hanging over,
each rider attempts to pick up the kata without dismounting. Collisions and nasty
falls are numerous, and this sport generally partakes of the character of an all-
round fight among the ponies’ legs. Somebody, however, always succeeds in
picking up the scarf and getting clear of the others, when he triumphantly rides
round the camp fluttering the prize in the wind.
Some of the younger fellows are clever at this sport, and when one rider at a
time does the feat, he seldom misses picking up the kata at the first swoop.
A Race for the Kata
Another exercise consists in bodily lifting a person on the saddle while the pony
is at full gallop. The pedestrian is seized as low near the waist as possible, and
the impetus of the pony’s flight, not the rider’s actual strength, is utilised in
raising the person on the saddle.
Women are scarce in Tibet, and actual raiding parties, I was told, occasionally
take place against neighbouring tribes in order to obtain a fresh supply of wives.
Taking things all round, there are few men and [156]women in Tibet who cannot
ride well, yet there are few who can claim exceptional skill in that line. The
Tibetan generally values his bones too much to indulge in fancy tricks upon his
pony. Some young fellow, more ambitious than others, will master the art of
standing erect upon the saddle while going full speed, his feet being inserted
into the stirrups, which have for the purpose been shortened as high as they
could go. By pressing with his ankles against the saddle he manages to
maintain his balance, in the familiar way of the Cossacks and tribes of Central
Asia, who all excel in this game.
Tibetan saddles, as you know, are in appearance not unlike a cross between a
Cossack saddle and a rude Mexican saddle, and as good as neither, but quite
suited to the country where they are used. Men and women ride astride, with
exceptionally short stirrups, so that the leg is bent at the knee at a right or even
an acute angle. In order to maintain one’s equilibrium when riding fast some
additional stability is obtained by stretching out the arms sideways.
Tibetan Soldier at Target Practice
The blocks of the saddles are of wood imported mainly from India, Nepal, or
China, with bindings of hammered iron or brass, often inlaid with silver and gold.
Lizard skin and coloured leather adorn the front and back of the saddles, and a
substantial pad covers the central part and the otherwise very angular seat. For
extra comfort rugs—occasionally valuable and always decorative in blue and red
tints—are spread, while to leather laces behind the saddle are slung double
bags containing tsamba, chura, or cheese, a brick of compressed tea, and
whatever sundry articles may be used on a journey. [158]The last, but not least
attachment on a Tibetan saddle is a long coiled rope of yak hair with a wooden
peg at the end for tethering the pony at night.
Whatever one may say of Tibetans, the best-inclined could not compliment them
on their shooting. Their matchlocks—their only firearms, made in Lhassa and
Shigatz—are weapons so clumsy and heavy and badly made, that when fired it
is truly more dangerous to be behind them than in front of their muzzle. During
my captivity in Tibet in 1897, indeed, I was fired upon twice—by distinguished
marksmen who took accurate aim only a few paces from me—but neither time
was I hit. Nor in all my experience of Tibet have I any remembrance of ever
seeing a Tibetan hit with a projectile from his matchlock anything which he
intended, although the range was never more than twenty or thirty yards. Few
are the matchlocks in the Forbidden Land which will carry as far as fifty or a
hundred yards.