of a mesenteric cyst in 1883.

9 In 1890 Carson presented

the first U.S. report of a chylous cyst of the mesentery.11 An
estimated 600 cases of mesenteric cysts were recorded in
the literature by 1950.9 By 1994 that number had grown to
820 patients.15,16 Since then there have been several series
reported in children3,17–19 and adults,20 bringing the total
to about 1000 cases.

Embryology
------------------------------------------------------------------------------------------------------------------------------------------------

Several mechanisms have been suggested to account for the

development of mesenteric and omental cysts including
failure of the embryonic lymphatic spaces to join with the ve-
nous system, deficiency of the normal lymphaticovenous
shunts in perinodal tissue, failure of the leaves of the mesen-
tery to fuse, occult trauma, neoplasia, and localized degener-
ation of lymph nodes.11,21 The most commonly accepted
theory, proposed by Gross,22 is benign proliferation of ectopic
lymphatics in the mesentery that lack communication with the
remainder of the lymphatic system. These cysts are thought to
arise from lymphatic spaces associated with the embryonic

CHAPTER 91
retroperitoneal lymph sac, analogous to cystic hygromas of
the neck arising in association with the jugular lymph sac.23
The role of lymphatic obstruction is questionable because
experimental occlusion of lymph channels in animals fails

Mesenteric and to produce these cysts owing to the rich collaterals in the
lymphatic system.11,21,23 In addition, obstructed lymphatics
have not been demonstrated with lymphangiography.9 As of

Omental Cysts now, there is no conclusive evidence accounting for the

embryologic events leading to the formation of these cysts.24

Richard R. Ricketts
Classification Systems
------------------------------------------------------------------------------------------------------------------------------------------------

Beahrs and colleagues11 classified cystic disease of the mesen-

Mesenteric and omental cysts are rare; the incidence is ap- tery into four categories on the basis of possible etiology: em-
proximately 1 per 105,000 admissions to general hospitals bryonic and developmental cysts, traumatic cysts, neoplastic
and 1 per 20,000 admissions to pediatric hospitals.1,2 Modern cysts, and infective cysts. In this system, mesenteric, omental,
series show that nearly one third of mesenteric cysts occur in retroperitoneal, dermoid, urogenital, and enteric duplication
children younger than 15 years and that one fourth occur in cysts would be classified as “embryonic and developmental.”
patients younger than 10 years.3–5 These cysts are reportedly Traumatic and infective cysts would have no endothelial
more common in females than in males and in white persons lining. “Neoplastic” cysts would include cystic lymphan-
than in nonwhite persons.4,6–10 In pediatric series, however, giomas.24 This classification system creates confusion in ter-
the age range is from birth to 18 years (mean, 4.35 years) minology because most mesenteric and omental cysts in
and the cysts occur slightly more often in males (60%). children are actually benign proliferation of lymphatic tissue
The experience reported in this chapter is with 16 children and thus could be considered “neoplastic” or “embryonic.”
admitted to Children’s Healthcare of Atlanta at Egleston over Some authors differentiate between cystic lymphangiomas
30 years. and mesenteric-omental cysts on the basis of histology.3,21,25–27
Cystic lymphangiomas are simple cysts with an endothelial
cell lining, foam cells, and thin walls that contain small
lymphatic spaces, lymphoid tissue, and smooth muscle. The
History wall of a mesenteric cyst lacks smooth muscle and lymphatic
------------------------------------------------------------------------------------------------------------------------------------------------
spaces, and the lining cells are cuboidal or columnar.21
A mesenteric cyst was first recorded in an autopsy of an Lymphangiomas occur in the mesentery or retroperitoneum
8-year-old boy by the Florentine anatomist Benevieni in and tend to present early in life with acute abdominal symp-
1507.11 In 1842 vonRokitansky12 first described a chylous toms, whereas mesenteric cysts are limited to the mesentery
mesenteric cyst; Gairdner13 reported an omental cyst in and present in adulthood as asymptomatic masses.21,25
1852. The first successful surgery for a cystic mass in the mes- A more recent classification system, that of de Perrot,28 is
entery was performed by the French surgeon Tillaux14 in based on the origin of the cyst: lymphatic origin (lymphangioma);
1880. Pean performed the first successful marsupialization mesothelial origin (mesothelial cyst, cystic mesothelioma);
1165
1166 PART VII ABDOMEN

enteric origin (duplication cysts); and urogenital cysts, cystic TABLE 91-2
teratomas (dermoid), and nonpancreatic pseudocysts (traumatic Differential Diagnosis of Mesenteric and Omental Cysts
and infectious). Intestinal duplication cyst
Because the evaluation and therapeutic objectives for mes- Ovarian cyst
enteric, omental, and retroperitoneal cysts are the same and Choledochal cyst
because they are all anterior extensions of what were retro- Pancreatic, splenic, or renal cyst
peritoneal structures during the embryonic stage,29 they share Hydronephrosis
common characteristics. In this chapter, a “mesenteric cyst” is Cystic teratoma, dermoid
defined as any cyst located in the mesentery; it may or may not Hydatid cyst
extend into the retroperitoneum. It has a recognizable lining of Ascites
endothelial or mesothelial cells.3 An “omental cyst” has the
same histologic characteristics but is confined to the greater
or lesser omentum. Because mesenteric, omental, and retro- bowel, whereas mesenteric cysts can often be enucleated from
peritoneal cysts are all of lymphatic origin, the term “cystic between the leaves of the mesentery.32
lymphatic malformations” may be more appropriate.24 The pathologic features of these cysts can vary consider-
ably. They can be single or multiple, unilocular or multilocu-
lar; they can have serous, chylous, hemorrhagic, or mixed
Spectrum of Disorders and fluid contents; and their lining can vary from a flattened en-
dothelial monolayer to a cuboidal or columnar epithelium
Differential Diagnosis to patchy fibrosis.4,10,26,29,33 Rarely, the cyst wall contains
------------------------------------------------------------------------------------------------------------------------------------------------
calcium.6,9,33 Unlike duplication cysts, mesenteric and omen-
Mesenteric cysts can occur in the mesentery anywhere along tal cysts contain no mucus-producing cells.4 Mesenteric cysts
the gastrointestinal tract from the duodenum to the rectum.11 are most commonly single and multiloculated; the fluid is
They may extend from the base of the mesentery into the generally serous when the cyst involves the distal small bowel
retroperitoneum. Omental cysts are located in the lesser or or colonic mesentery and chylous when it is located in the
greater omentum. Mesenteric cysts are 4.5 times more com- proximal small bowel mesentery.5,9 Omental cysts almost
mon than omental cysts.10 In a thorough review of 162 cases always contain serous fluid.33 The features of the cysts in this
reported between 1950 and 1985, Kurtz and colleagues2 series are shown in Table 91-3.
found that 60% were located in the small bowel mesentery,
24% in the large bowel mesentery, and 14.5% in the retro-
peritoneum. The most common location is in the ileal mesen-
tery. In the colonic mesentery, cysts occur most commonly in Clinical Presentation
the sigmoid mesocolon.4,9 Cyst location in several pediatric ------------------------------------------------------------------------------------------------------------------------------------------------

series is shown in Table 91-1. The clinical presentation of mesenteric and omental cysts can
The differential diagnosis of mesenteric and omental cysts vary from an incidental finding during laparotomy performed
is shown in Table 91-2. Modern imaging studies can usually for another reason to an acute, life-threatening intra-abdominal
determine the organ where cystic lesions within the abdomi- catastrophe. In adults, these cysts are found incidentally in
nal cavity originate. The differentiation between intestinal approximately 40% of patients7,9,10 and present as acute ab-
duplication cysts and mesenteric cysts may be problematic dominal emergencies in up to 60% of patients. The classic pre-
because both are often intimately associated with the bowel sentation is that of a low-grade, partial intestinal obstruction
wall. The former share a common blood supply and muscular combined with a palpable, freely movable abdominal mass.6
layer with the adjacent bowel and have a well-defined mucosal The most common mode of acute presentation in children
layer that mesenteric cysts lack.1,3,5,30,31 At the time of is a small bowel obstruction, sometimes associated with
surgery, duplication cysts require resection of the involved volvulus and intestinal infarction.5,25,33–35 In a series of

TABLE 91-1
Location of Mesenteric and Omental Cysts
Author Date No. of Patients Mesentery Omentum Retroperitoneum
Mollitt5 1978 11 8 0 3
Molander30 1982 6 3 3 0
Radhakrishma35 1989 8 5 3 0
Adejuyigbe34 1990 5 3 2 0
Kosir25 1991 13 9 0 4
Chung1 1991 15 9 2 4
Hebra29 1993 22 13 6 3
Bliss3 1994 10 10 0 0
Senocak17 1994 19 14 5 0
Okur18 1997 10 7 1 2
Egozi19 1997 14 10 2 2
Total 133 91 (68%) 24 (18%) 18 (14%)
 CHAPTER 91 MESENTERIC AND OMENTAL CYSTS 1167

TABLE 91-3 TABLE 91-4

Cystic Lymphatic Malformations of the Abdomen (N ¼ 16) Mode of Presentation and Treatment (N ¼ 16)
Feature Number (%) Feature Number (%)
Age Mode of presentation
< 5 yr 12 (75) Asymptomatic mass, distention 4 (25)
 5 yr 4 (25) Acute (<48 hr) abdominal symptoms 2 (12.5)
Sex Subacute (<2 mo) abdominal symptoms 5 (31.25)
Male 10 (62.5) Chronic (>2 mo) abdominal symptoms 5 (31.25)
Female 6 (37.5) Treatment
No. cysts Enucleation, excision 8 (50)*
Single 11 (69) Intestinal resection 5 (31.25)
Multiple 5 (31) Excision with distal pancreatectomy 1 (6.25)
Type of cyst Marsupialization, cautery 2 (12.5)
Multilocular 14 (88)
Drainage 0
Unilocular 2 (12)
Fluid content *All three omental cysts excised.
Hemorrhagic 7 (44)
Serous 3 (19)
Chylous 2 (12)
In this series, 75% of the patients had abdominal symptoms
Infected 1 (6)
and all patients had findings on abdominal examination. In
Not specified 3 (19)
only two patients (12.5%) did the condition present as an acute
Size (cm)
abdominal emergency.
Smallest 4  4.5  1.7
The following case summaries illustrate the varying presen-
Largest 30  40  10
tations of mesenteric and omental cysts:
Location
1. A 7-day-old female infant presented with a 48-hour history
Mesentery 11 (68.75)
of vomiting and diarrhea and was found to have fever,
Omentum 3 (18.75)
abdominal distention, and erythema of the abdominal
Retroperitoneal 2 (12.5)
wall. Escherichia coli sepsis was documented. The patient
had multiple infected omental and mesenteric cysts invol-
ving the greater omentum and transverse mesocolon, all of
82 children operated on for volvulus, mesenteric cyst was the which were excised. The patient recovered.
cause in 3.54% of cases.36 Small bowel obstruction can also 2. A 3-day-old male infant presented with signs and symp-
develop from acute enlargement of the cyst secondary to hem- toms consistent with a small bowel obstruction. He was
orrhage or from extrinsic compression of the bowel as it is found to have one ileal multiloculated mesenteric cyst with
stretched over an enlarging cyst (Fig. 91-1). The general mode volvulus that required small bowel resection with primary
of presentation of our 16 children is shown in Table 91-4. In anastomosis.
no patient was the cyst found incidentally, although four 3. A 2-year-old boy had an asymptomatic abdominal mass
patients were asymptomatic. In one of these four, a cystic intra- that was found to be a multilocular mesenteric cyst in-
abdominal mass was identified by prenatal ultrasonography. volving the transverse mesocolon and pancreas. Excision

A B
FIGURE 91-1 A, Multiple mesenteric cysts filled with chylous fluid surrounding proximal small bowel. B, Large multilocular mesenteric cyst compressing
overlying bowel and causing chronic partial intestinal obstruction.
1168 PART VII ABDOMEN
of the cyst and distal pancreatectomy were performed
for cure.
4. A 2-year-old girl presented with a 2-week history of
asymptomatic abdominal distention. She was found to
have a single 30- by 40- by 10-cm multilocular mesen-
teric cyst that extended along the base of the mesentery
from the ligament of Treitz to the transverse mesocolon.
Near-total excision of the cyst with marsupialization
of the residual cyst into the peritoneal cavity and scle-
rosis of the wall with tincture of iodine resulted in long-
term cure.
5. A female fetus was found to have an intraabdominal cystic
mass on routine prenatal ultrasonography. After birth, the
infant was jaundiced, which suggested a choledochal cyst;
this was excluded by normal radionuclide scanning. After
phototherapy, the patient remained asymptomatic but had
a palpable, freely movable, nontender abdominal mass.
She underwent elective resection of an omental cyst at
the age of 6 months (Fig. 91-2).
6. A hydropic full-term female infant died from congestive
heart failure. Autopsy revealed a 14-cm multiloculated cyst
in the mesentery.
7. A 12-year-old girl with chronic abdominal pain was found
to have a cystic mass on CT scan. Exploration revealed
intestinal ischemia secondary to compression of the supe-
rior mesenteric artery by several mesenteric cysts. Subtotal
intestinal resection resulted in short-bowel syndrome.
8. A 1-year-old boy with a right-sided reducible inguinoscro-
tal swelling since birth presented with irreducibility of the
swelling of 3 days’ duration. At laparotomy he was found to
have a large mesenteric cyst of the terminal ileum, which
communicated with the inguinoscrotal swelling.37
Complications associated with mesenteric and omental
cysts include intestinal obstruction (most common), volvulus,
hemorrhage into the cyst, infection,38 rupture, torsion of
the cyst,10 obstruction of the urinary or biliary tract,39 and
malignancy.2,5,8,9,11,40,41 The reported incidence of malignant
conditions (sarcoma, lymphangioendothelioma, or, rarely,
adenocarcinoma) is 3%,2 although in a recent adult series of
16 patients (ages 12 to 68 years), there were three malignant
cysts (18.8%).20 No malignant mesenteric or omental cysts
have been reported in children.
FIGURE 91-2 Asymptomatic but palpable, freely movable omental cyst
first seen on prenatal ultrasound.
Diagnosis
------------------------------------------------------------------------------------------------------------------------------------------------
On physical examination, a majority of children with mesen-
teric and omental cysts have abdominal distention with or
without a palpable mass. A definite mass may be difficult to
palpate because of its large size, soft and fluid consistency,
and great mobility. The mass may be huge, filling the abdom-
inal cavity and simulating ascites.17,38,42–45 It is dull to percus-
sion. Omental cysts can sometimes be distinguished from
ascites by the relative sparing of the flanks; ascites causes
the flanks to bulge out, whereas omental cysts do not. If there
is no other reason for ascites (e.g., liver or renal disease),
a mesenteric or omental cyst should be considered.32 If
a definite mass is palpable, mesenteric cysts are generally
movable in the transverse plane, whereas omental cysts are
movable in the transverse and craniocaudal planes.3,9
Modern imaging studies can usually establish (or at least
strongly suggest) the diagnosis of a mesenteric or omental cyst.
A plain abdominal radiograph shows a gasless, homogeneous,
water-dense mass that displaces bowel loops around it.33
Omental cysts may compress bowel loops posteriorly, whereas
mesenteric cysts may be surrounded by bowel loops.31 Fine
calcifications can sometimes be seen.6,33 Abdominal ultraso-
nography and computed tomography (CT) have almost
eliminated the need for contrast studies such as upper gastro-
intestinal series, barium enema, and intravenous pyelography.
Abdominal ultrasonography is currently the imaging proce-
dure of choice.25,30,31 This test reveals a well-circumscribed,
thin-walled, fluid-filled cystic structure that usually contains
thin internal septi and sometimes contains internal echoes
from hemorrhage, infection, or debris (Fig. 91-3).3,30,31
Enteric duplication cysts are thick-walled structures that share
a common muscular wall with the adjacent bowel.31 They also
have a clearly visible mucosal lining on ultrasonography.
CT, with the administration of a gastrointestinal contrast
FIGURE 91-3 Ultrasonogram showing a large, multilocular cystic mass in
the anterior abdomen, consistent with the diagnosis of a mesenteric cyst.
Fine internal echoes represent hemorrhage within the cyst (arrow).
 CHAPTER 91 MESENTERIC AND OMENTAL CYSTS 1169

laparoscopically.15,49,52–54 For those cases requiring a concomi-

tant bowel resection, either an intracorporeal technique or a
laparoscopic-assisted extracorporeal technique can be used.
If enucleation or resection is not possible, the third option
is partial excision with marsupialization of the remaining cyst
into the abdominal cavity. Approximately 10% of patients
require this form of treatment.2 If this procedure is done,
the cyst lining should be sclerosed with 10% glucose solu-
tion,4 electrocautery, or tincture of iodine in an attempt to
minimize recurrence. Percutaneous injection of the lyophi-
lized incubation mixture of group A Streptococcus pyogenes
A OK432 has been successful in the treatment of large, nonre-
sectable lymphangiomas in children.26 The mechanism of
action of OK432 seems to be related to activation of the white
cells (an increased number of natural killer cells and T cells)
and an increase in cytokine-mediated endothelial per-
meability (increased activity of tumor necrosis factor and
interleukin-6), resulting in shrinkage of the cystic spaces.26
This agent may be useful to complement the surgical treat-
ment of mesenteric cysts extending into the retroperitoneum.
Partial excision and drainage are not indicated because of the
high recurrence rate associated with these procedures.2
A simple and universal pathologic classification system that
considers the different varieties of mesenteric cysts has been
proposed by Losanoff and colleagues (Fig. 91-5).26 Types 1
and 2 are easily cured with resection or enucleation, with
or without concomitant bowel resection. Types 3 and 4,
extending into the retroperitoneum, require complex surgical
B procedures and often sclerotherapy as well. Recurrence in
types 3 and 4 is more common than in types 1 and 2.
FIGURE 91-4 A, Computed tomography scan with gastrointestinal con- The surgical treatment used for the current series is shown
trast showing a large cystic mass surrounding a loop of colon. B, Com-
puted tomography scan demonstrating a large unilocular cyst that is in Table 91-4. All three omental cysts were totally excised.
displacing bowel loops laterally and posteriorly. Eleven of the 13 (85%) mesenteric cysts were completely ex-
cised; 2 (15%) were marsupialized into the peritoneal cavity
agent, can show the relation of the bowel to the cystic mass and had the cyst lining sclerosed with electrocautery (1) or
(Fig. 91-4) and demonstrate that the cyst does not arise tincture of iodine (1). Of the completely excised cysts, 5
from any other organ such as the kidney, pancreas, or ovary.9 (45%) required concomitant bowel resection and 1 required
A normal radionuclide scan of the biliary tract excludes distal pancreatectomy. None of the cysts was simply drained.
choledochal cysts from diagnostic consideration. No major short- or long-term complications, deaths, or
recurrences occurred in this series.

Treatment
------------------------------------------------------------------------------------------------------------------------------------------------
Outcome
The goal of surgery is complete excision of the mass. Omental ------------------------------------------------------------------------------------------------------------------------------------------------

cysts are easily removed and almost never require bowel resec- The overall results of treatment in patients with mesenteric and
tion.29,33 Omental cysts can be excised using laparoscopic omental cysts are favorable. The reported recurrence rate ranges
techniques.42,46–48 Partial drainage of the cyst may be neces- from 0% to 13.6%, averaging about 6.1% in a series of 162 adults
sary to confirm the site of origin of the cyst and allow its and children.2 Most recurrences occur in patients with retroper-
removal through the umbilical port.49 The preferred treatment itoneal cysts or those who had a partial excision.2,10,25,29 The
of mesenteric cysts is enucleation.2,9,29,32 In adults, the cyst overall mortality rate in adults and children is 2%.2 Only two
can often be “shelled out” from between the leaves of the mes- deaths have been reported in children since 1950. One was a
entery32; in children, however, bowel resection is frequently 26-month-old girl who presented with small bowel infarction
required to totally eradicate the mass and ensure that the from volvulus associated with a mesenteric cyst,55 and the sec-
blood supply to the bowel is not compromised. A bowel resec- ond was the baby with hydrops reported here.
tion is necessary in only about 33% of adults, but 50% to 60%
of children with mesenteric cysts require resection.2,3,5,18,33,50
For this reason, a mechanical bowel preparation is recom-
mended before surgery, if time permits.1 If bowel resection Summary
is required, intestinal continuity can usually be reestablished ------------------------------------------------------------------------------------------------------------------------------------------------

primarily. The laparoscopic approach to resection of mesen- Mesenteric and omental cysts are rare, but they are more
teric cysts was first described by Mackenzie in 1993.51 In many commonly encountered in children’s hospitals than in adult
cases the cyst must be drained before it can be excised general hospitals. Most cysts are developmental in origin and
1170 PART VII ABDOMEN

L
L L

L
L

1 2 3 4
FIGURE 91-5 Classification of mesenteric cysts. Type 1—pedicled; easily resected. Type 2—sessile in leaves of mesentery; requires bowel resection. Type
3—extending into retroperitoneum; often incompletely resected. Type 4—multicentric; may require complex operations, sclerotherapy, or both. (From
Losanoff JE, Richman BW, El-Sherif A, et al: Mesenteric cystic lymphangioma. J Am Coll Surg 2003;196:598.)

are related to a congenital abnormality of the lymphatic system.
This leads to altered production and flow of lymph in ectopic
lymphatic tissue, which lacks communication with central
channels. The cysts can be located in the mesentery or omen-
tum of any part of the gastrointestinal tract and may extend into
the retroperitoneum. With modern imaging studies, these cysts
can usually be differentiated from cystic lesions of other organs
within the abdominal cavity. They may present incidentally, in-
sidiously, or as an acute life-threatening emergency. In children,
acute presentations from intestinal obstruction with or without
volvulus are relatively common. In adults, less acute modes of
presentation are the rule. The goal of treatment is complete re-
moval of the cyst by either resection or enucleation. In children,
concomitant bowel resection is often required. At times, a por-
tion of the cyst wall must be left in situ and marsupialized into
the peritoneal cavity and sclerosed. The short- and long-term
prognosis of children with mesenteric and omental cysts is ex-
cellent, with a low recurrence rate, few complications related to
treatment, and essentially no mortality.
The complete reference list is available online at www.
expertconsult.com.
 that of plasma. Thus there is normally no significant osmotic
gradient across the sinusoidal membrane.1
The second major source of abdominal lymph fluid is via
the mesenteric venous system. The mean pressure of the
mesenteric capillary is about 20 mm Hg. Intestinal lymph
drains from regional lymphatics into the thoracic duct. The
mesenteric capillary membrane is relatively impermeable to
albumin. Because the concentration of protein in the mesen-
teric lymph is only about 20% that of plasma, there is a signif-
icant osmotic gradient that promotes the return of lymph fluid
into the capillary under normal conditions. In an adult-sized
patient, there is normally less than 150 mL of free fluid in the
abdomen and the normal flow of lymph into the thoracic duct
is about 800 to 1000 mL per day.1
Ascites occurs when there is an alteration in the normal hy-
drostatic, osmotic, and electrochemical forces that determine
fluid balance. Increases in or aberrations of hydrostatic and/or
osmotic pressures of the hepatic and mesenteric capillaries
may cause a net transfer of fluid from blood vessels to lymph
vessels that exceeds the drainage capacity of the lymph system.
Such alterations in hydrostatic pressure may result from
cardiac insufficiency or compromise of blood flow through
the caval or portal venous systems. Changes in colloid pres-
CHAPTER 92 sure may result from hypoproteinemia secondary to liver
compromise or from disturbances in capillary membrane per-
meability from a variety of inflammatory, metabolic, genetic,
or neoplastic causes. Abnormalities in lymphatic drainage
Ascites may result from trauma including surgery and child abuse
or congenital or other anomalies that result in obstruction
of the lymphatic system. Finally, ascites may be caused by
Eugene D. McGahren III direct communication between the urinary tract and the ab-
dominal cavity. In this instance the ascites is partially the urine
itself, as well as the fluid resulting from the inflammation
caused by the urine.1,6,7

Ascites is the pathologic accumulation of excess fluid in the

peritoneal cavity. The word “ascites” is derived from the Greek
askos and askites meaning “bag,” “bladder,” or “belly.”1 Ascites Clinical Features
has been recognized since the time of Hippocrates (400 BC), ------------------------------------------------------------------------------------------------------------------------------------------------

while modern descriptions of ascites in fetuses and infants Abdominal distention is the most common physical finding
date to the seventeenth century.2–5 A wide variety of condi- associated with ascites.5,8 Other physical signs may include
tions, both congenital and acquired, cause ascites in infants bulging flanks, prominent abdominal wall vasculature, an
and children (Table 92-1). The more common causes of ascites everted or protuberant umbilicus if there is an umbilical her-
are addressed in this chapter. nia, or an inverted umbilicus if there is no hernia, inguinal
hernias, scrotal or labial edema, a fluid wave, total body
edema, and shifting dullness to percussion.6–10 Other physi-
cal findings related to the underlying cause of the ascites such
Anatomy and Pathophysiology as a cardiac murmur, enlarged liver, spleen, or kidney, or an
------------------------------------------------------------------------------------------------------------------------------------------------
abdominal mass may be present. Urinary output may be
There are two major sources of abdominal lymph fluid. One is diminished due to intravascular depletion, increased intra-
the flow of blood through the liver. Blood flow from the he- abdominal pressure, or urinary leak.11 Respiratory compro-
patic artery and portal vein perfuses the hepatic sinusoids mise may result from significant fluid accumulation and its
and then exits via the hepatic veins. Pressure in the sinusoids pressure on the diaphragm.2,12 Signs of peritoneal irritation
is typically low because precapillary resistance is greater than suggest a primary inflammatory or infectious etiology of the
postcapillary resistance. The space of Disse is a location in the ascites. Secondary infection of the ascitic fluid, otherwise
liver defined by the hepatocytes on one side and the sinusoidal known as spontaneous bacterial peritonitis (SBP) or bacterial
lining cells on the other side. Hepatic lymph is formed by the ascites (BA), may also occur. Infected ascites may present
filtration of sinusoidal plasma into the space of Disse. It then with fever, nausea, vomiting, or encephalopathy. It is best
drains from the liver via transdiaphragmatic lymphatics to the diagnosed by a paracentesis that demonstrates an elevated
thoracic duct, which then empties into the subclavian vein. polymorphonuclear neutrophil (PMN) count of greater
The sinusoidal endothelium is highly permeable to albumin, than 250/mL. The sample may also identify the offending
and the concentration of protein in hepatic lymph is close to organism.5,13
1171
1172 PART VII ABDOMEN

TABLE 92-1 such as posterior urethral valves.8 Ultrasound cannot distin-

Causes of Ascites guish between types of ascitic fluid, however. Prenatal ultra-
Hepatocellular
sound routinely detects fetal ascites and may aid in prenatal
Glycogen storage disorders, lysosomal storage disorders, galactosemia,
and postnatal strategies to aid the infant.14,15 Computed
alpha-1 antitrypsin deficiency, viral hepatitis, neonatal hepatitis, tomography (CT) may be helpful in determining the cause
congenital hepatic fibrosis, cirrhosis, hepatic or portal vein thrombosis, of ascites, but it does not usually add any information about
Budd-Chiari syndrome, malignancy, venoocclusive disease, biliary the fluid beyond what the ultrasound has obtained.
atresia Ascitic fluid may be obtained by paracentesis and
Biliary analyzed according to the clinical suspicion (Table 92-2).
Bile duct perforation, injury to bile ducts Ultrasound may be helpful in locating a site to be aspirated,
Chromosomal especially if there are scars from previous surgery.1 An aspirate
Trisomy 21, Turner syndrome of 10 to 20 mL should be sufficient for complete analysis.
Macroscopic appearance is assessed and may appear straw
Cardiac
colored, turbid, bloody, or chylous. Laboratory evaluation of
Congestive heart failure, arrhythmias
ascitic fluid should include a cell count with differential,
Infection Gram stain and culture, cytology if indicated, and chemical
Appendicitis, tuberculosis and atypical aflatoxin B, cytomegalovirus, analysis including pH, lactate dehydrogenase, albumin,
parvovirus, syphilis, varicella zoster, enterovirus, listeriosis,
toxoplasmosis, hepatitis A, chlamydia, fungus
protein, amylase, bilirubin, creatinine, and triglyceride
levels.6,7,16
Renal/Urinary
Posterior urethral valves, ureteroceles, ureteral stenosis, urethral
stenosis or atresia, neurogenic bladder, bladder perforation, nephrotic
syndrome, urogenital sinus, cloacal malformation
Gastrointestinal
Hepatocellular Ascites
------------------------------------------------------------------------------------------------------------------------------------------------

Atresia, malrotation, volvulus, necrotizing enterocolitis, intestinal

perforation, enteropathy, meconium peritonitis, lymphangiectasia,
Neonatal and viral hepatitis, biliary atresia, inborn errors of
trauma (including child abuse), congenital obstruction from peritoneal metabolism, congenital hepatic fibrosis, Budd-Chiari syn-
band, gastroschisis, omphalocele, postcardiac surgery, postabdominal drome, and alpha 1 antitrypsin deficiency commonly lead
surgery for tumor, other abdominal surgery, lymphangioma, to cirrhosis, portal hypertension and, subsequently, ascites
intussusception in children. Hepatocellular ascitic fluid is a transudate that
Pancreatic results from increased portal venous pressure, increased intra-
Pancreatitis from any cause, pseudocysts, trauma luminal pressure in the mesenteric capillaries, and loss of
Gynecologic fluid into the peritoneal cavity. Loss of albumin potentiates
Ruptured ovarian cyst, hydrometrocolpos this fluid loss and triggers renal absorption of sodium and
water.16–19 Diagnosis of each of these conditions is based
Miscellaneous
on the results of various tests such as specific biochemical
Ventriculoperitoneal shunt, peritoneal dialysis, pseudomyxoma
peritonei, granulomatous peritonitis, myxedema, neoplasms (e.g., assays for inborn errors, viral cultures, tests for viral antibody
lymphoma, neuroblastoma, ovarian), metabolic storage diseases titers, and liver histologic testing. Analysis of ascitic fluid
should include a serum-ascites albumin gradient (SAAG). A
gradient greater than 1.1 g/dL is 97% accurate in predicting
portal hypertension.19,20 The cell count should be low, and
no bacteria should be visualized, although primary peritonitis
Diagnosis can complicate preexisting ascites.
------------------------------------------------------------------------------------------------------------------------------------------------
Guidelines of therapy for ascites of hepatic cause include
Beyond physical examination, plain radiographs of the abdo- salt and fluid restriction. Sodium intake should be limited
men typically show medial displacement of bowel, as well as to 1 to 2 mEq/kg day for infants and children and 1 to 2 g/
diffuse abdominal opacification with ground-glass appear- day (44 to 88 mEq of sodium/day) for adolescents. Water in-
ance. Separation of the properitoneal fat stripe from the colon take should be no more than 75% of daily maintenance,
and separation of air-filled bowel loops are two other nonspe- although severe water restriction is unnecessary unless there
cific findings. The liver may have a rounded appearance and is significant hyponatremia (<125 mg/dL).8,16,19 Gentle
may appear pushed to the center. Calcifications may be pre- diuresis is commonly undertaken with salt and fluid restric-
sent if the ascites is associated with bowel perforation and tions. The goal of diuresis is to attain an excretion of extracel-
peritonitis, particularly in the prenatal period (Fig. 92-1).6,7,14 lular water and sodium. Spironolactone is a favored initial
Ultrasonography is sensitive for detecting free fluid in the agent due to its potassium-sparing characteristic. Furosemide
peritoneal cavity. Free fluid in the abdomen appears as shifting may be used temporarily with spironolactone at the start of
echo-free zones in dependent locations. Small amounts of diuretic therapy because spironolactone takes a few days to
fluid can be detected in the hepatorenal or splenorenal fossa, be effective.7,16,19 Repeated large-volume paracentesis is re-
as well as in the pelvis. As little as 50 to 100 mL may be visible quired in patients for whom salt and fluid restriction
in the abdomen on ultrasound (see Fig. 92-1).6 Repositioning fails, particularly if there are symptoms of respiratory insuffi-
the patient can help differentiate free from loculated fluid col- ciency. Paracentesis in which the removed ascitic fluid is
lections. Ultrasonography may also be helpful in identifying replaced with salt-poor albumin or dextran has been shown
inciting causes of ascites and is particularly helpful in diagnos- to be safe and effective, especially when combined with
ing urinary tract anomalies that are associated with ascites diuretic therapy and sodium restriction.11,19,21–23
 CHAPTER 92 ASCITES 1173

FIGURE 92-1 A newborn with hepatocellular ascites. The plain radiograph shows diffuse opacification with minimal bowel gas; the ultrasonogram shows
fluid in the hepatorenal fossa.

TABLE 92-2
Laboratory Analysis of Ascites
Type of Ascites Albumin Level Bilirubin Level Triglyceride Level pH Creatinine Level
Hepatocellular 2-3 g/dL Serum level <50 mg/dL 7.35 <1 mg/dL
Biliary 3 g/dL 2-3 Times serum level <50 mg/dL >7.4 <1 mg/dL
Chylous <3 g/dL <1 mg/dL 1000-1500 mg/dL 7.5 <1 mg/dL
Urinary <1 g/dL <1 mg/dL <50 mg/dL <7 5-10 mg/dL

Peritoneovenous shunting for intractable ascites of hepatic perforation is speculative. Distal duct obstruction, pancreatic
and other causes has been described in a few infants and fluid reflux up the common bile duct, congenital weakness of
children with good results.24–26 Limitations attributable to the common bile duct, or a localized mural malformation of
the large size of the pump mechanism and venous tubing have the wall of the common duct and pancreaticobiliary
made the use of such devices less common in pediatric appli- maljunction have been proposed as possible causes.31,37–39
cations. Modifications have been made by manufacturers to It has been suggested that the proposed bile duct wall defects
help customize Denver-type shunts for newborns.25 An intra- that may predispose to perforation may be part of a larger
peritoneal vascular port has been successfully used for pallia- spectrum of defects that includes malformations such as
tive relief of tense malignant ascites caused by hepatocellular choledochal cysts.36 Perforation of the biliary tree in conjunc-
carcinoma.27 A newborn peritoneal dialysis catheter is also tion with choledochal cyst has been reported in case stud-
effective in draining ascitic fluid in a controlled manner. ies.40–42 Biliary ascites may also result from perforation of
Transjugular intrahepatic portosystemic shunt therapy has the biliary tree due to trauma.33,43
been successful in managing ascites in a teenager and may Biliary ascites from spontaneous bile duct perforation
have a role for therapy in selected children.28 typically occurs in infants and toddlers up to about age 2 years
but may occur in older children as well.33,39,44–48 It may also
occur prenatally.45 Most patients are otherwise healthy and
Biliary Ascites have no predisposing conditions associated with the disorder.
------------------------------------------------------------------------------------------------------------------------------------------------
A typical presentation is the development of progressive pain-
Biliary ascites usually results from a spontaneous perforation less abdominal distention with jaundice over a 1- to 4-week
of the common bile duct, most commonly at the junction with period. Vomiting, abdominal pain, and clay-colored stools
the cystic duct.29–33 Rare instances of cystic duct perforation may also be present. The condition is usually indolent in its
have been reported.34,35 Spontaneous biliary duct perforation presentation, although if biliary peritonitis is the initial
is a relatively rare condition with about 150 cases reported in presentation, the patient may be toxic appearing. However,
the literature as of 2003,36 with few reported since. Although most patients can tolerate large amounts of bile in the perito-
many theories have been proposed, the cause of bile duct neal cavity without evidence of infection or peritonitis.33
1174 PART VII ABDOMEN

Physical examination is notable for abdominal distention, Known congenital lymphatic abnormalities that lead to
presence of fluid within the abdomen, and possibly inguinal chylous ascites include atresia or stenosis of the major lacteals
hernias and bile-stained hydroceles.33,36,39,44–47 at the base of the mesentery or the cisterna chyli, mesen-
Biliary ascites should be suspected when the presentation teric cysts, generalized lymphangiomatosis, and lymphan-
and findings noted earlier occur in the absence of any liver gioma.5,8,55–57 The next most common cause of chylous
disease. Ultrasound can confirm the presence of fluid within ascites is thought to be obstruction of lymphatic channels
the abdomen, and nuclear scintigraphy and MRCP can aid by conditions such as intussusception, malrotation, incarcer-
in the identification and localization of a biliary leak.36,46,47,49 ated hernia, primary or metastatic cancer, tuberculosis, gas-
Diagnosis can be definitively established by paracentesis. troschisis, omphalocele, or inflammatory lesions causing
The fluid is bile stained with bilirubin levels of 100 to lymph node enlargement.5,8,9,53,56–60 A large intraabdominal
400 mg/mL, although the level may be lower in more chronic lymphangioma may mimic ascites.61,62 Finally, injury to the
presentations.36 lymphatics is responsible for an additional 15% to 20% of
The primary goal in the treatment of biliary ascites is cases of chylous ascites. This may result from trauma, surgery
typically external drainage.33,36,44,45,47,50 The approach may for a variety of conditions, and child abuse. Ascites may not be
involve exploration of the right upper quadrant with chole- noticed until days or weeks after the inciting event.3,4,58,63–65
cystostomy and cholecystography to document the size and Chylous ascites can occur in children of all age groups,
location of the perforation. If there is evidence of distal ob- although most are infants and toddlers.5,8,53 Abdominal
struction, some advocate giving consideration to a biliary-in- distention is the most common presenting sign. This may
testinal anastomosis but this is not commonly necessary.44,50 If develop gradually or acutely and may be accompanied by
the perforation is confined to the cystic duct, cholecystectomy abdominal pain and respiratory compromise.* Other signs
may provide definitive treatment.35 Additional internal drain- and symptoms may include vomiting, diarrhea, inguinal her-
age procedures such as cholecystojejunostomy or duode- nia, and edema.5,68
notomy with sphincteroplasty are usually unnecessary and The main confirmatory diagnostic study is abdominal
fraught with difficulty because in most cases no intrinsic ob- paracentesis, which reveals a milky white fluid if the pa-
struction exists and inflammation tends to distort the anat- tient is receiving a fat-containing enteral diet. However,
omy.31,51 The region is frequently densely scarred, and a sac the fluid may appear less milky or amber if the child has
filled with thick bile can be mistaken for a choledochal cyst.36 not been receiving enteral feeds. The fluid is characterized
External drainage can be accomplished by placing a Penrose by an elevated triglyceride concentration (often > 1000 mg/dL),
or closed suction drain in the porta hepatis. A cholecystost- a cell differential that is predominantly lymphocytes
omy tube or, less commonly, a T-tube may be useful to help (70% to 90%), and elevated total protein and cholesterol
with decompression and to further assess the biliary tree in concentrations.4,5,6,8,20,53
the future.44,52 External drainage has also been successfully Determining the cause of the chylous ascites can be a chal-
accomplished using percutaneous technique, thus avoiding lenge, especially if the child has no obvious conditions that
laparotomy.47 Alternatively, laparoscopic technique may allow may commonly predispose to a lymph leak. Studies such as
localization of the leak and precise drain placement.44 There ultrasonography, CT, and gastrointestinal contrast studies
has been a recent report of endoscopic retrograde cholangio- may be helpful in identifying predisposing conditions such
pancreatography assisted biliary stent placement facilitating as malrotation, lymphangioma, tumor, or mesenteric cysts.
treatment of spontaneous biliary leak.48 Bile duct stenosis is In newborns in whom ascites has no readily recognizable ex-
the most common complication after external drainage. Portal planation, a congenital lymph channel malformation is the
vein thrombosis, bile leak, and cholangitis have also been most likely diagnosis. Lymphangiography is described in
reported.36,47 If a choledochal cyst is the cause of the biliary older series, but it is difficult to perform in children and it
leak, then excision of the cyst with appropriate enteric-biliary has helped to direct treatment in only a minority of children
anastomosis is indicated.40–42 in whom it has been used.8,20
With adequate external drainage, most patients survive and Surgically correctable lesions such as malrotation, mesen-
need no additional surgical intervention. Eighty percent of teric cysts, intussusception, or incarcerated hernias should
perforations heal within 3 weeks.29,31–33,44,47,51 Antibiotics be corrected. In patients without a surgically correctable
and complete bowel rest with total parenteral nutrition lesion, the initial treatment is aimed at reducing lymph flow
(TPN) are important adjuncts in these patients. Fat-free through the damaged or obstructed lymph channels. This is
enteral infant formulas are also used, but no studies have accomplished by suspending enteral intake and initiating
compared these two nutrition options in patients with biliary TPN. Some patients respond quickly to this intervention, with
ascites. The cholecystostomy and peritoneal drains should resolution of the ascites within 2 weeks. If the patient’s nutri-
remain in place until normal ductal anatomy has been demon- tional and hydration status can be maintained, courses of TPN
strated through the cholecystostomy. of up to 10 weeks can be considered.63,67 Because medium-
chain triglycerides (MCTs) are absorbed directly into the por-
tal system, and therefore do not stimulate lymph flow, they
have been traditionally advocated in the treatment of chylous
Chylous Ascites ascites. Hard data confirming the benefit of MCT formula have
------------------------------------------------------------------------------------------------------------------------------------------------
been lacking. Nonetheless, the use of a low-fat, medium
Chylous ascites is the excessive collection of lymph fluid in the chain-enhanced formula such as Portagen has been reported
peritoneal cavity. In infants and children it is most commonly
idiopathic (45% to 60% of cases) and is presumed to be due to
congenital malformations of the lymphatic channels.4,5,8,53,54 *References 3, 4, 5, 8, 53, 58, 66, 67.
 CHAPTER 92 ASCITES 1175

in multiple successful nonoperative treatment courses and directly into the peritoneal cavity through the uterus and
should be considered as an adjunct to TPN or as the initial en- fallopian tubes.2,6,74–83 A recent case of urinary ascites was
teral formula in patients who respond to TPN.9,63–65,67 Pro- reported with only the presence of vesicoureteral reflux.84
longed use of low-fat infant formulas has been associated Most patients with urinary ascites present soon after birth.
with poor neurologic development, possibly from fatty acid However, prenatal ultrasound may also detect ascites.85–89 On
deficiency.54 Treatment with low-fat formula should therefore physical examination, abdominal distention, frequently with
be limited to 3 to 6 months. Administration of somatostatin palpable flank masses, is the most common presentation.
has been reported to successfully resolve chylous ascites in Respiratory distress and acidosis (absorption of urine from
some cases.69,70 Paracentesis in addition to bowel rest should the peritoneal cavity) may be present. Potter syndrome (mater-
be reserved for patients with respiratory compromise. nal oligohydramnios, renal, pulmonary, and chest wall hypo-
Surgical intervention is warranted if nonoperative plasia) may be present. Prune-belly syndrome may also be a
therapy is not successful after 6 to 10 weeks or if the patient physical finding.2,89,90 Hyponatremia, hyperkalemia, and an
becomes otherwise symptomatic.8,53,54,63 Location of the leak elevated serum blood urea nitrogen (BUN) level and creatinine
may be facilitated by feeding the patient a high-fat diet (usu- level are also common.2,74,91
ally milk) up to 6 hours before surgical exploration. This Workup should begin with abdominal ultrasonography,
results in a copious creamy lymph flow from the leak area. which will demonstrate the ascitic fluid. It will also identify
Addition of Sudan dye to the milk may further aid in identi- most common urinary tract abnormalities that are associated
fying the leak.66,71 The most common location of the lymph with urinary ascites such as hydronephrosis, dilated ureters,
leak is at the base of the superior mesenteric vessels, although and a dilated or thickened bladder.6 Voiding cystourethrogra-
various retroperitoneal and mesenteric lymphatics may be phy (VCUG) demonstrates any obstructing lesions and may
involved.8,12,53,54,66,71 Complete mobilization of the duode- show the point of extravasation.75,76,78 Intravenous pyelogra-
num and the head of the pancreas and a thorough exploration phy (IVP) or renal scanning with diethylenetriamine pentaa-
of the entire retroperitoneum should be performed.56 Ligation cetic acid (DTPA) or mercaptoacetyltriglycine (MAG 3) may
of an identified leaking lymphatic is curative in 85% of provide information about the degree of renal parenchymal
patients.8,67 Recent reports also describe the potential useful- damage caused by long-standing obstruction and may show
ness of laparoscopy in localizing lymph leaks and the use of the point of extravasation.
fibrin glue in the control of lymph leaks.12,66,72 Peritoneove- Treatment ultimately addresses the underlying urinary
nous shunting may be a useful treatment strategy as a primary tract obstruction if present. However, fluid resuscitation,
surgical intervention or as an intervention in those children antibiotics, and distal urinary tract drainage constitute the
in whom surgical exploration has failed to resolve the initial treatment in these patients.75,76,78 Prenatal intervention
ascites.26,63,68,73 In some older series, mortality from chylous for significant urinary obstruction may also be considered.92
ascites was significant, ranging from 24% to 30%.5,8 However, Urinary tract decompression may be accomplished by Foley
with current nutritional support and surgical intervention catheterization, particularly in the case of posterior urethral
capabilities, a successful outcome can now usually be expected. valves. Cystoscopy with valve ablation is then performed
electively. Other decompression strategies may include percu-
taneous nephrostomy or surgical vesicostomy or pyelostomy
depending on the offending lesion and the condition of the
Urinary Ascites patient.75,76 Although there are some who advocate routine
------------------------------------------------------------------------------------------------------------------------------------------------
drainage of the ascites, typically drainage of the ascites or
Urinary ascites is a rare condition, though a common cause of repair of the perforation is not necessary because the urinary
ascites in newborns, accounting for up to one-third of cases ascites is reabsorbed once urinary drainage has been achieved
of isolated ascites.2,6 Males outnumber females by a ratio of and the perforation heals quickly.2,74,76 Drainage of the ascites
7:1.74 Obstruction of the lower urinary tract, particularly from is necessary in the presence of respiratory compromise
posterior urethral valves, accounts for approximately 70% or infection.2 Once the previously mentioned measures have
of cases of urinary ascites. Other obstructive causes of urinary been successfully completed, definitive operation can be un-
ascites include ureteroceles, ureteral stenosis, neurogenic dertaken to address the offending lesion. There is a risk of
bladder, and urethral stenosis or atresia.2,6,74,75,76 Urinary ex- compromised long-term renal function due to the chronic
travasation from a discrete perforation is noted in approxi- obstruction. However, the spontaneous decompression of
mately 65% of patients with an obstructive cause. The the obstruction that results in urinary ascites is felt to be pro-
posterior fornix is the weakest part of the urinary system. Thus tective of renal function. Thus a significant number of
the site of perforation is most commonly located in the renal patients retain good renal function, although this must be
pelvis, although bladder perforation is also common.2,76 monitored on a long-term basis.2,6,74,85,87 With improve-
Other causes of urinary ascites that are not obstructive in na- ments in neonatal care and early recognition of these condi-
ture include trauma to the urachal remnant, often as a result of tions, the mortality rate from urinary ascites has decreased
attempted umbilical arterial catheterization; spontaneous from 70% to 0%.76
rupture of the bladder in infants, sometimes associated with
connective tissue disorders; and, in females with persistent The complete reference list is available online at www.
urogenital sinus or cloacal malformation, leakage of urine expertconsult.com.

CHAPTER 93
Polypoid
Diseases of the
Gastrointestinal
Tract
Joseph L. Lelli, Jr.
Polyps are any masses that project into the lumen of the
gastrointestinal tract. Some masses that appear to be polyps
are subepithelial. True intestinal polyps, however, are of an
epithelial origin. Polyps in children most commonly occur
as an isolated lesion referred to as a juvenile polyp. However,
more rarely, polyps in children can occur as a genetically
related disease with features including multiple polyps
(polyposis) in the gastrointestinal tract, heritability within
a family, and an increased lifetime risk of developing a rela-
ted cancer in the gastrointestinal tract or in other organ
systems. Children with these features are considered to have
a “polyposis syndrome.” Polyposis syndromes in children
are classified into two groups, namely adenomatous poly-
posis syndromes or hamartomatous polyposis syndromes
(Table 93-1).
Polyps are common during childhood, occurring in approx-
imately 1% of preschool and school-aged children.1 Because
of their high incidence, polyps of the gastrointestinal tract
represent the most frequent cause of rectal bleeding in toddlers
and preschoolers aged 2 to 5 years of age. Juvenile polyps are
most common (80%) and are followed by lymphoid polyps
(15%).2 Adenomatous polyps occur in less than 3% of all chil-
dren with polyps.
Juvenile Polyps
------------------------------------------------------------------------------------------------------------------------------------------------
Juvenile polyps were first described by Verse in 1908.3 For
many years, all polyps in children were considered adenomas4
and were often treated with radical procedures. Horrilleno5
first used the term “juvenile polyp” in 1957 to describe a his-
tologically distinctive colorectal polyp that occurred pre-
dominately during childhood. In 1962 Morson6 made an
important contribution when he demonstrated that juvenile
polyps were benign hamartomas, thereby distinguishing
them from potentially malignant adenomas. Less radical treat-
ment of these polyps, however, was slow to follow. The
distinction between isolated juvenile polyps and juvenile
polyposis was first made by McColl and colleagues in
1964,7 and three distinct forms of juvenile polyposis were
further defined by Sachatello in 1972.8 Kaschula in 19719
followed by Enterline in 197610 reported cases of children
with both juvenile and adenomatous polyps. Billingham11
reported the presence of solitary adenomas in children with
juvenile polyps in 1980. These early reports of the coexis-
tence of juvenile and adenomatous polyps led to the identifi-
cation of adenomatous and malignant changes in juvenile
polyps.12–14 However, only polyps associated with juvenile
polyposis have malignant potential in children.15–17 The dis-
tinction between the commonly occurring isolated juvenile
polyps, which are benign, and the rare juvenile polyposis
syndromes, which may be malignant, has become increasingly
important.15,16,18,19
Jass20 has proposed the following criteria for increased risk
for cancer in children with polyps: (1) more than five juvenile
polyps in the colon, (2) polyps throughout the gastrointestinal
tract, and (3) any number of polyps associated with a family
history of juvenile polyposis. Jass’s criteria clarify the three dis-
tinct juvenile polyposis syndromes originally described by
Sachatello in 1972.8 On the basis of Jass’s and Sachatello’s
studies, the following classification of juvenile polyps is most
commonly used:
I. Isolated Juvenile Polyps (nonmalignant): fewer than five
polyps confined to the colon without a family history of
juvenile polyposis.
II. Juvenile Polyposis Syndromes (malignant potential):
1. Diffuse juvenile polyposis of infancy: widespread polyposis
of entire gastrointestinal (GI) tract in patients younger
than 6 months of age.
2. Diffuse juvenile polyposis: multiple polyps throughout
the GI tract, but mostly in the stomach, distal colon,
and rectum, usually occurring in patients 6 months
to 5 years of age.
3. Juvenile polyposis coli: multiple juvenile polyps confined
to the distal colon and rectum in patients 5 to 15 years
of age.
Juvenile polyposis syndromes, Cowden disease, and Peutz-
Jeghers syndrome are all classified as hamartomatous polyposis
syndromes (see Table 93-1).
1177
1178 PART VII ABDOMEN

TABLE 93-1
Classification of Polyposis Syndromes
Classification
Adenomatous polyposis 1. Familial adenomatous
syndromes polyposis (FAP)
2. Gardner syndrome
3. Turcot syndrome
Hamartomatous polyposis 1. Juvenile polyposis
syndromes 2. Cowden disease
3. Peutz-Jeghers syndrome

ISOLATED JUVENILE POLYPS

Pathology
Juvenile polyps, which are also known as retention, inflamma-
tory or cystic polyps, are the most common type of polyp found
in the GI tract and account for 80% of polyps in children. Such
polyps are generally considered hamartomas6 or a malforma- FIGURE 93-2 Gross cross section of juvenile polyp with typical cystic
tion in which normal colonic tissue has become arranged in a “lakes” filled with mucus.
haphazard manner.20
Grossly, the typical polyp has a glistening, smooth, spherical,
reddish head and ranges from 2 mm to several centimeters in
diameter (Fig. 93-1). Polyps will often have an ulcerated surface,
which accounts for the rectal bleeding. A cross section shows
cystic spaces filled with mucus (Fig. 93-2). Juvenile polyps
are typically attached by a long, narrow stalk covered by colonic
mucosa. This stalk predisposes the polyp to torsion, which re-
sults in venous congestion, surface ulceration, bleeding, and
auto amputation.
Microscopically, the surface of the polyp has a single layer
of colonic epithelium. This epithelial layer is often ulcerated or
replaced with granulation tissue. When inflammation occurs,
the epithelium may show a reactive hyperplasia that can
mimic dysplasia or adenomatous changes.21 The main body
of the polyp consists of dilated or cystic epithelial tubules that
are lined by normal colonic epithelium. These tubules and
cystic lakes are embedded in a lamina propria and an abun-
dant, loose, vascular, and fibrous stroma (Fig. 93-3). The FIGURE 93-3 Photomicrograph of typical juvenile polyp. The surface
shows a smooth, flattened, colonic epithelium. Large tubular and cystic
stroma is usually heavily infiltrated with neutrophils, eosino- lakes are present, embedded in an abundant, loose stoma.
phils, lymphocytes, and monocytes. Mitotic figures are rarely
seen. Only two descriptions of malignant changes in a solitary
juvenile polyp are present in the literature,22,23 and according
to Jass,20 only one of these may be a true malignant condition.
In a longitudinal study of patients with solitary polyps, no
increased risk for developing colorectal cancer could be
found.24

FIGURE 93-1 Typical juvenile polyp with its stalk attached.
Etiology
The etiology of juvenile polyps is unknown, but several au-
thors have suggested hereditary,25 genetic,26 hamartomatous
malformation,6 and inflammation.27 Recent data suggest that
juvenile polyps result from a structural rearrangement of the
mucosa secondary to an inflammatory process.28 The initial
event is probably ulceration and subsequent inflammation
of the mucosa, leading to obstruction of regional, small colo-
nic glands of the mucosa. The obstructed glands proliferate,
branch, and dilate, forming a cystic structure. The cystic struc-
ture pushes up the mucosa, leading to further ulceration,
inflammation, and formation of granulation tissue. As the
 CHAPTER 93
cycle continues, an increasingly larger and larger mass pushes
into the lumen. The fecal stream and peristalsis push the mass
down the lumen, causing the stalk to elongate and resulting in
the typical pedunculated appearance of the juvenile polyp.
Incidence
Although the incidence of juvenile polyps is unknown, they
are believed to occur in approximately 1% of all preschool
children.1 Most juvenile polyps appear in the first decade of
life, with the peak incidence between 3 and 5 years of age.1
The polyps are solitary in 50% of cases, with the remainder
having 2 to 10 polyps. The location of these polyps has chan-
ged over the past 10 years. Historically, 70% of the polyps
were found in the rectum. Today, only 40% are found in
the rectum or sigmoid colon, whereas 60% are found evenly
distributed throughout the proximal colon.1 Juvenile polyps
are rarely seen after adolescence.
Clinical Presentation
The most common presenting symptom of a juvenile polyp is
bleeding (93%) that results from ulceration of the polyp sur-
face. Blood loss is usually minimal and appears as bright red
streaks of blood over the surface of the stool. Abdominal pain
(10%), which is believed to be caused by traction on the polyp
from peristaltic activity, and prolapse of the polyp (4%) are
other less common presenting symptoms. Prolapse of the rec-
tum and encopresis have also been reported.1 Many juvenile
polyps will autoamputate, resulting in spontaneous cessation
of rectal bleeding.29
The differential diagnosis of juvenile polyps encompasses
all of the causes of rectal bleeding in toddlers through child-
ren who are 6 years of age. Anal fissures and rectal prolapse
cause rectal bleeding but are easily distinguished from polyps
on physical examination. Bleeding from Meckel diverticu-
lum or duplication of the intestine usually causes more sub-
stantial blood loss than that from a polyp, and the blood
usually commingles with the stool rather than coating it.
Bleeding from an intussusception is accompanied by abdom-
inal pain that is substantially worse than that seen with polyps.
Inflammatory bowel disease is usually accompanied by diar-
rhea, which is not seen with polyps. Blood dyscrasias, such
as Henoch-Schönlein purpura, should also be considered in
the differential diagnosis.
Treatment
The diagnosis and treatment of juvenile polyps requires a
combination of history, digital rectal examination, and colo-
noscopy. The shift of juvenile polyps to the more proximal
colon29 and the concern for the presence of juvenile polyposis
(>5 polyps), with its increased risk of malignancy, mandates
that the entire colon be surveyed. Children with suspected
polyps should have a digital rectal examination initially.
Polyps in the rectum can be easily removed during anoscopy.
Following removal of a rectal juvenile polyp, pancolonoscopy,
in a well-prepared bowel, should be performed to determine
if additional, more proximal polyps are present.30 Children
with juvenile polyposis and adenomatous changes are more
likely to have right-sided polyps31; therefore all polyps should
be removed and undergo histologic evaluation. Complications
following endoscopic removal of polyps are rare.
POLYPOID DISEASES OF THE GASTROINTESTINAL TRACT 1179
Polyposis Syndromes
------------------------------------------------------------------------------------------------------------------------------------------------
Gastrointestinal polyps are the result of a defect in the balance
between cellular growth promotion and cellular growth inhi-
bition. The defect results from either an activated oncogene
that has upregulated a growth-promoting protein or the loss
of function of a growth-inhibiting protein, usually by the
inactivation of a tumor suppressor gene. In the polyposis syn-
dromes there has been inactivation of a tumor suppressor
gene.32,33 A resulting gastrointestinal cancer occurs when
there are additional genetic defects. Polyposis syndromes
can be classified into adenomatous polyposis syndromes
and hamartomatous polyposis syndromes.
Adenomatous Polyposis
Syndromes
------------------------------------------------------------------------------------------------------------------------------------------------
All of the adenomatous polyposis syndromes are distinguished
by the development of a large number of adenomas in the
colon. Additionally, these syndromes are associated with extra-
colonic manifestations such as duodenal polyps; benign soft
tissue tumors such as osteomas of the mandible, long bones,
and skull; and congenital hypertrophy of the retinal pigmented
epithelium (CHRPE) (Table 93-2). Adenomatous polyposis syn-
dromes include familial adenomatous polyposis (FAP), most
common in children; Gardner syndrome; and Turcot syndrome.
Familial Adenomatous Polyposis
------------------------------------------------------------------------------------------------------------------------------------------------
FAP is distinguished by the progressive development of hun-
dreds to thousands of adenomatous polyps in the colon. By
15 years of age 50% of those that carry the FAP gene will have
polyps. The lifetime risk for developing a colorectal cancer is
100%; however, the average age for developing an adenoma in
FAP is 16 years and 39 years for developing a colorectal
cancer.34
TABLE 93-2
Extracolonic Features in Familial Adenomatous Polyposis
Cancer (Lifetime Risk) Other Lesions
Duodenal (1%-5%) CHRPE
Pancreatic (2%) Nasopharyngeal
angiofibromas
Thyroid (2%) Osteomas
Brain (medulloblastoma) (<1%) Radiopaque jaw lesions
Hepatoblastoma (0.7% of children Dental abnormalities
< 5 yr old) Lipomas, fibromas,
epidermoid cysts
Desmoid tumors
Gastric adenomas/fundic
gland polyps
Duodenal, jejunal, ileal
adenomas
CHRPE, congenital hypertrophy of the retinal pigment epithelium.
Modified from Cruz-Correa M, Giardello FM: Diagnosis and management of
hereditary colon cancer. Gasroenterol Clin North Am 2002;31:537-549.
1180 PART VII ABDOMEN

HISTORY
The first case of adenomatous polyposis was recorded by
Covisart in 1847,35 but it was Chargelaigue,36 who, 12 years
later, gave the first definitive account of the disease in a 16-year-
old girl and a 21-year-old man. The first pathologic description
of these colonic polyps was reported by Virchow in 1867,37
with Woodward in 188138 being the first to distinguish
between neoplastic and inflammatory polyps. The first familial
association of these polyps was described by Cripps in 188239
between a 9-year-old boy and his 17-year-old sister. However,
the malignant potential of these polyps was not recognized until
1890 by Handford.4
The first recorded operations for polyposis are credited to
Lilienthal21 in North America and subsequently to Lockhart-
Mummery at St. Mark’s Hospital in London, as reported by
Thompson and Watne.40,41 Lockhart-Mummery, in 1925,42 FIGURE 93-4 Gross specimen of the total colon in a patient with familial
concluded that (1) multiple adenomatous polyps develop adenomatous polyposis demonstrating thousands of adenomatous
in succeeding generations, (2) afflicted individuals develop polyps. Note the large adenoma in the ascending portion of the colon.
cancer at an early age, and (3) the adenomas are an antece-
dent to cancer. In 1927 Cockayne43 described the mode of in-
heritance as Mendelian dominant, which was later confirmed
by Dukes in 193044 and Lockhart-Mummery in 1939.45 Since
then, FAP has become a model for understanding the adenoma-
carcinoma sequence.46
The detection of a deletion in chromosome 5q47 in a patient
with Gardner syndrome has led to characterization of the gene
responsible for FAP, the APC gene.48–57 The 5q deletion or
alteration in the APC gene results in a gene product with a
truncated protein, which subsequently causes the APC gene
to act as a tumor suppressor gene.32,33 Mouse models for
FAP have been developed,58 and these will facilitate research
in gene therapy.

PATHOLOGY
FAP is defined as the presence of at least 100 visible adenoma-
tous polyps in the large intestine.59 However, patients with
fewer polyps have been diagnosed as having FAP, and one pa-
tient who had early onset of colorectal cancer and had a 5q
gene deletion had only one endoscopically detectable polyp.60
Colorectal polyps are the hallmark of FAP. Although 100
polyps is the usual threshold for diagnosing this disorder,
patients coming to colectomy in their teenage years typically
have thousands of polyps (Fig. 93-4). Two types of FAP seem
to exist: the sparse type, which is characterized by hundreds
of polyps, and the profuse type, which is characterized by thou-
sands of polyps. Utsunomiya and colleagues61 have shown that
the history of the two types is different—patients with the
profuse type tend to develop adenocarcinoma at an earlier age.
In FAP the polyps are typically scattered throughout the
colon, which allows diagnosis by sigmoidoscopy. The polyps
vary in size from 1 to 2 mm in diameter, and when present, pe-
dunculated polyps can be 1 cm in diameter or larger (Fig. 93-5).
During duodenoscopy, biopsy of normal-appearing duodenal
mucosa may reveal microscopic adenomas too small to be seen
macroscopically. These microadenomas may consist of only
three or four dysplastic crypts that usually have a tubular
adenomatous structure.62
Adenomatous polyps result from a neoplastic transforma-
tion of at least one epithelial cell that is in the proliferative
portion of a crypt.63 The transformed clone then populates
FIGURE 93-5 Close-up view of colonic mucosa demonstrating a “carpet”
of small 1- to 2-mm adenomas in a patient with familial adenomatous
polyposis.
the upper part of the crypt, which divides by budding.63 Crypt
fusion is then responsible for further growth of the adenoma.
This neoplastic transformation, called dysplasia, implies atyp-
ical morphologic characteristics, with nuclear enlargement
and stratification. Adenomas seem to pass through gradations
of dysplasia until invasive adenocarcinoma develops.64 Exten-
sion of neoplastic cells into the basement membrane of the
colonic epithelium represents carcinoma in situ. As the neo-
plastic cells extend beyond the basement membrane, the
tumor becomes microscopically invasive. Because the colonic
mucosa does not contain lymphatics, metastasis does not
usually occur until the tumor invades through the muscularis
mucosa into the submucosa.
Small intestinal mucosa can also be involved with adeno-
matous polyps. Polyps of the small intestine are frequently
found around the orifice of the common bile duct.65 Polyps
in this region are variable in size, are often microscopic,
and involve only a few crypts. Adenocarcinoma of the duode-
nal papilla and periampullary regions occurs in 2.9% of
patients with FAP65 and has been found in the common bile
duct of these patients.66
 CHAPTER 93
Gastric polyps occur in patients with FAP, but they are usu-
ally benign polyps of the fundic gland.67 These polyps result
from cystic dilation of specialized gastric glands rather than
from dysplasia. No evidence that neoplastic transformation
occurs in polyps of the fundic gland exists.68
ETIOLOGY
The incidence of FAP ranges from 1 in 6000 to 1 in 12,000
births.69–71 FAP is inherited as an autosomal dominant trait
with a moderate incidence (10%) of new mutations.72 The
manifestations of FAP differ greatly, probably as a result of
variation in the mutation of the APC gene. Although the occur-
rence of polyps is strongly related to genetics, the phenome-
non of regression of rectal polyps after colectomy and
ileorectal anastomosis was recognized in 198873,74 and points
to the fact that the luminal environment also plays a role in the
development of polyps.
FAP is caused by a mutation in the APC gene on the long
arm of chromosome 5, where a variable deletion or alteration
of the gene is associated with the disease. The APC gene codes
for a protein product that acts as a tumor suppressor.72
Because of this gene deletion, the probability of cancer (as
diagnosed by biopsy during sigmoidoscopy) by 25 years of
age is 90%.4 In colorectal cancer cell lines that have an intact
chromosome 5 introduced into the cells, the ability of the cells
to induce tumor growth in mice is considerably reduced.58,75
Similar results have been found for other tumor suppressor
genes, namely the TP53 gene on chromosome 17p and the
deleted colorectal cancer gene (DCC) on chromosome 18q.72
It has become apparent there is genetic heterogeneity in FAP.
Certain germline mutations of the APC gene predispose to
attenuated disease expression.77,78 Further evidence indicates
that additional modifier genes influence the severity of FAP.79
CLINICAL PRESENTATION
Although FAP has been recognized in infancy and early child-
hood, it is most frequently identified in early adolescence.
Most patients are asymptomatic, but some present with in-
creased frequency of defecation, rectal bleeding, anemia,
and abdominal pain. Most (90%) are identified by routine
surveillance because of a familial history of adenomatous
polyposis.
Diagnosis is established by sigmoidoscopy and occasion-
ally by air contrast barium enema. Sigmoidoscopy usually
reveals a carpet of polyps that cover the entire surface of the
colon. Biopsy of at least 10 polyps will confirm the diagnosis.
In some patients with FAP, most of the polyps will be located
in the proximal colon. If the diagnosis of FAP represents a new
mutation (no other family members are known to have the
disease), a careful examination of all family members is
required.
Polyps are found in the stomach in up to 50% of patients
with FAP, but only 6% of the polyps are adenomatous,67 with
the rest being fundic gland polyps (hamartomas). Polyps
occur in the duodenum less often than in the stomach; how-
ever, duodenal polyps are much more likely to be adenoma-
tous. Several series have demonstrated that up to 98% of
FAP patients have visible duodenal polyps or at least a histo-
logic abnormality with dysplasia, unicrypt adenomas, or
hyperplasia in the duodenal mucosa.65,67,80 Adenocarcinoma
POLYPOID DISEASES OF THE GASTROINTESTINAL TRACT 1181
of the duodenal papilla or periampullary region event-
ually develops in 2.9% of patients with FAP.65 For these
reasons, upper endoscopy is necessary in all patients with
this disorder.
TREATMENT
Malignant conditions of the colon will occur in all patients
with FAP if left untreated. The average age for developing can-
cer is 39, with 7% developing cancer by age 20 and 15% by
age 25.81 Surgical removal of the entire colonic mucosa will
prevent colorectal carcinoma. Most patients who present with
symptomatic polyps already have a malignant condition.81
Therefore colectomy is recommended at any age, if the child
is symptomatic.
A wide range of surgical options exists. Total proctocolect-
omy with a permanent ileostomy prevents cancer but leaves
a young patient with a permanent abdominal wall stoma.
The physiologic and psychologic impact of a permanent ileost-
omy is substantial in adolescence. This operation can cause
substantial sequelae because of the extensive pelvic dissection,
namely, damage to the nervi erigentes with resultant bladder at-
ony and in males, impotence. For these reasons, total proctoco-
lectomy is not the treatment of choice for FAP in a pediatric
patient.
Total abdominal colectomy with an ileorectal anastomosis
and continued surveillance of the retained rectum has prob-
ably been the most common operation for FAP performed in
the past. St. Mark’s Hospital in London reported results with
215 patients undergoing this procedure.40 Immediate post-
operative complications included prolonged ileus (7%), anasto-
motic breakdown (2%), and bleeding (<1%), for a total
complication rate of 10%. Frequency of defecation at late
follow-up (6.5 years) was three stools per day, and less than
10% reported nighttime soiling. Continence was considered
completely normal in 72%. Forty-four percent (44%) of the
patients, however, required subsequent treatment for their rectal
polyps. Depending on the density of the rectal polyps, the
patients were evaluated every 3 to 6 months, with sigmoidos-
copy and fulguration of any polyps greater than 5mm in diam-
eter. Ten percent developed carcinoma in their rectal stump. The
cumulative risk for rectal cancer was 10% at 50 years of age and
29% by 60 years of age.81
In more recent years, total colectomy with a rectal muco-
sectomy and an ileoanal pouch procedure has become the
preferred operation for children with FAP. The J pouch is
the preferred technique because of simplicity of construc-
tion and sparsity of complications.82 Geiger and colleagues83
reported a novel technique of a double-stapled, ileoanal,
J pouch anastomosis with excellent outcomes. An inverse
relationship exists between the size of the reservoir and the
frequency of defecation.84–87 Other factors that influence
frequency of defecation include inflammation, sphincter func-
tion, and small intestinal motility. Larger reservoirs allow
less frequent defecation but tend to be associated with more
frequent bouts of inflammation of the reservoir (“pouchitis”),
which probably results from stool stasis. In several recent
series that studied 450 patients with reservoirs, the frequency
of defecation ranged from 3.3 to 7.2 with an average of
5.8 stools per day.88–93 Large series of patients with reservoirs,
however, report an average rate of pouchitis of 23% and a
pelvic sepsis rate of 8%.
1182 PART VII ABDOMEN
FOLLOW-UP
In addition to colonic polyps, patients with FAP can develop
a large assortment of benign extracolonic manifestations and
occasionally other cancers (see Table 93–1). Endoscopy of
the upper gastrointestinal tract and sigmoidoscopy should
be performed annually in all patients. Removal of polyps in
the duodenum can be done by endoscopic snaring or by
open duodenotomy. Size larger than 1 cm in diameter, rapid
growth, polyp induration, severe dysplasia, or villous change
suggests the need for a more aggressive intervention. Recent
data have shown that 42% of adenomas of the ampulla of
Vater that are larger than 1 cm in diameter contain a foci
of cancer compared with only 13% of those smaller than
1 cm.94 The incidence of duodenal cancer, however, remains
low at 1% to 5%. Sigmoidoscopy can be done easily in patients
with an ileoanal pull-through. The anal canal and any short seg-
ment of rectal mucosa left behind must be carefully evaluated.
OTHER TREATMENTS
Drugs and several dietary supplements have been used to
treat polyps. These include vitamin C,95,96 sulindac,96 dietary
fiber,97 and calcium.98,99 In a randomized, double-blinded
study of the use of sulindac, inhibition of both rectal and
duodenal polyp growth was observed100 and other studies
have confirmed this finding.71,101 In a randomized, double-
blind, placebo-controlled study, however, standard doses
of sulindac did not prevent the development of adenomas
in subjects with FAP. Rectal cancer has been reported after
prolonged sulindac treatment.
GARDNER SYNDROME
Between 1951 and 1955, Gardner and colleagues102–105
established the association of colonic familial adenomatous
polyposis and the extracolonic findings of multiple osteomas,
fibromas, and epidermoid cysts. They also demonstrated that
the syndrome was inherited in an autosomal dominant pat-
tern.102,104 The natural history and treatment of the colonic
polyps is the same as for those patients with FAP. The osteomas
are most frequently found in the skull and facial bones, and
abnormal dentition with impaction and early tooth decay
and supernumerary teeth are seen.106 Sebaceous cysts are
most commonly found on the legs, followed by the face, scalp,
and arms. Lipomas and fibromas are also seen.
Periampullary cancer was thought to be more prevalent
in Gardner syndrome than in those with FAP, but surveillance
of the duodenum in patients with FAP has shown a high
incidence of duodenal abnormalities in all patients. Because
detailed surveillance of all patients with familial polyposis
has revealed many subtle extracolonic manifestations, the
distinction between Gardner syndrome and FAP is becoming
less clear. FAP is probably an all-encompassing syndrome, in
which patients manifest different signs as the condition
evolves.107
Desmoid tumors of the abdominal wall and mesentery of
the small intestine occur in approximately 20% of patients
with Gardner syndrome. These tumors also occur in patients
who only have the colonic manifestations of FAP and are the
leading cause of death in those who have undergone a pro-
phylactic colectomy. Desmoid tumors are a dense, fibroplastic
proliferation that may remain localized or may become wide-
spread throughout the mesentery or the abdominal wall.
All stages of fibrous dysplasia including fibrosarcoma have
been seen. Most desmoid tumors appear after surgery, usually
within 6 to 30 months.108 Desmoid tumors of the body
wall may be observed initially because many of them remain
static or even regress over several years.108 For those that
continue to grow or reach 10 cm in diameter, complete local
excision is met with a recurrence rate of less than 10%, which
is satisfactory. Most intraabdominal desmoid tumors do not
become evident until they have reached a nonresectable size.
Because of the difficulty in resecting intraabdominal desmoid
tumors, several drugs have been tried with varied rates of
success. Such drugs include steroids; antiestrogen agents (pro-
gesterone, tamoxifen, and toremifene); and nonsteroidal anti-
inflammatory agents.109–115
TURCOT SYNDROME
Turcot, Despres, and St. Pierre116 described two siblings who
initially presented with colonic familial adenomatous polyposis
and eventually died of intracranial brain tumors (glioblastoma
and a medulloblastoma). Turcot believed that the brain tumors
were extracolonic manifestations of FAP. Ependymomas and
carcinoma of the thyroid (which are usually papillary in ori-
gin)107 have also been described to occur in Turcot syndrome.
Hamartomatous Polyposis
Syndromes
------------------------------------------------------------------------------------------------------------------------------------------------
Hamartomatous polyposis syndromes are distinguished by
an overgrowth of cells that are native to the area in which
they normally occur. The overgrowth is in general not considered
to have malignant potential. Several of the syndromes in this
category, however, have an increased lifetime risk of developing
either intestinal or extraintestinal cancers. The hamartomatous
polyposis syndromes include juvenile polyposis syndrome,
Cowden disease, and Peutz-Jegher syndrome.
JUVENILE POLYPOSIS SYNDROME
Juvenile polyposis syndromes are uncommon. These syn-
dromes can be separated into three distinct clinical entities:
diffuse juvenile polyposis of infancy, diffuse juvenile polyposis,
and juvenile polyposis coli.8
Diffuse Juvenile Polyposis of Infancy
Diffuse juvenile polyposis of infancy presents within the first
few months of life,15 usually without any family history.20
Presenting signs include diarrhea, rectal bleeding, intussus-
ception, protein-losing enteropathy, macrocephaly, clubbing
of fingers and toes, and hypotonia.117–121 The extent of
diarrhea, rectal bleeding, and protein-losing enteropathy is
directly related to the number of polyps present. The entire
GI tract is frequently involved. Treatment for this syndrome
is usually aggressive, starting with total parenteral nutrition
and intestinal rest. The latter measure reduces the protein
and blood loss and decreases the incidence of intussuscep-
tion.122 When the nutritional status of the patient has been
stabilized, portions of the intestine with the most polyps are
 CHAPTER 93
removed. Despite appropriate treatment, this disease is almost
universally fatal; only 2 of 12 patients are reported as surviving
beyond 2 years of age.7,17,117–121,123–126
Diffuse Juvenile Polyposis
Children with diffuse juvenile polyposis are usually 6 months
to 5 years of age and present with mild rectal bleeding
and prolapse but may also have protein-losing enteropathy,
intussusception, and malnutrition.15 For this age group it is
important to distinguish juvenile polyposis from familial ade-
nomatous polyposis. Polyps associated with diffuse polyposis
are found throughout the bowel, most often in the stomach,
distal colon, and rectum.15 Endoscopic polypectomies and
segmental bowel resection are used in the treatment of this
disease. These children usually do well, although the need
for recurrent therapy is common.
Juvenile Polyposis Coli
Juvenile polyposis coli occurs in children between the ages of
5 and 15 years of age. This disorder is usually characterized by
bleeding of bright red blood from the rectum, anemia, rectal
prolapse, or all of these disorders. Polyps are usually limited
to the distal colon and rectum.15 Approximately 50% of the
patients will have a family history, indicating an autosomal
dominant pattern of inheritance.17 Associated congenital
defects including cleft palate, malrotation, polydactyly, and
abnormalities of the heart and cranium have been described.20
Pathology Any child with five or more polyps, polyps
throughout the GI tract, or even one polyp if the child has
a family history of juvenile polyposis is considered to have
a juvenile polyposis syndrome.20 Most patients with such
syndromes will have approximately 50 to 100 colorectal
polyps. Patients with diffuse juvenile polyposis of infancy
and diffuse juvenile polyposis will also have polyps in the
stomach, small intestine, or both. The importance of identify-
ing children with a polyposis syndrome lies in the need for
long-term surveillance due to the high risk of carcinoma
(17%) occurring at an early age (mean age at diagnosis of
carcinoma is 35.5 years).15
The gross appearance of a polyp in juvenile polyposis is
the same as one in isolated juvenile polyps. Approximately
20% of the polyps in juvenile polyposis, however, present
grossly as a multilobular mass resembling a cluster of pol-
yps attached to a stalk (Fig. 93-6).18 Histologically, these
polyps demonstrate more epithelium with a villous or papillary
configuration. Epithelial dysplasia can occur in juvenile polyps*
and in coexisting adenomas found in conjunction with juvenile
polyps.9,14,16,18,131–136 Severe dysplasia, which could be con-
sidered carcinoma in situ, has been found in patients with ju-
venile polyps associated with juvenile polyposis syndrome.137
Lobular polyps have a higher propensity for more severe dys-
plasia (47%) than the nonlobular polyps (10%).18 Several re-
ports of infiltrating adenocarcinoma of the colon and rectum
in association with juvenile polyposis exist.8,12,129,131,138–140
According to the St. Mark’s Polyposis Registry in London, the
cumulative risk for cancer in patients with a juvenile polyposis
syndrome by age 60 is 68%.19,128
*References 9, 12–14, 16, 18, 25, 127–132.
POLYPOID DISEASES OF THE GASTROINTESTINAL TRACT 1183
FIGURE 93-6 Gross specimen of colon in a patient with juvenile poly-
posis syndrome. Note small adenomas and multilobular clusters of polyps
attached to stalks.
Treatment The autosomal dominant pattern of inheritance
and the 68% cumulative risk for cancer by age 60 means that
treatment must include long-term follow-up of patients with
polyposis and their family members. Some advocate prophy-
lactic colectomy and rectal mucosectomy with endorectal ileal
pull-through as a primary treatment.129 Others recommend
regular screening (every 2 years) with colonoscopy and ran-
dom biopsy141 and subsequent colectomy if severe dysplasia,
rapid polyp formation, or bleeding occurs.16 One patient had
a colectomy and ileosigmoidostomy at 4 years of age only to
develop inoperable cancer of the sigmoid colon at 27 years
of age.18 First-degree relatives should be screened because
of the familial nature of the disease. The approach to patients
with juvenile polyposis should be similar to that taken for
patients with familial adenomatous polyposis.
PEUTZ-JEGHER SYNDROME
History
The association of intestinal polyps with mucocutaneous
pigmentation spots of the mouth, hands, and feet was first
reported by Peutz in 1921.142 In 1944143 Jegher first reported
two cases. In 1949144 he and his colleagues added 8 more to
12 other cases he had collected from personal communica-
tions. Jegher and colleagues144 subsequently defined the
two main features of the syndrome: melanin spots on the buc-
cal mucosa and lips, with variable melanin pigmentation on
the face and digits, and polyposis of the intestinal tract. In
1954 Bruwer, Bargen and Kierland145 were the first to use
the term Peutz-Jeghers syndrome. In recent years, reports of
intestinal and extraintestinal cancers have led to a reassessment
of the management of these patients.146
Pathology
Peutz-Jeghers syndrome is characterized by melanotic spots,
ranging in color from brown to black, occurring on the lips,
around the mouth, and on the buccal mucosa. These spots
can also be found on the hands, feet, nasal mucosa, and
conjunctivae and in the rectum.146 The pigmented spots are
usually present at infancy and usually fade at puberty.
1184 PART VII ABDOMEN
Although the polyps associated with Peutz-Jeghers syn-
drome can be found anywhere from the stomach to the
rectum, they occur most commonly in the small intestine
(55%). Approximately 30% of these polyps are found in the
stomach and duodenum, and 15% are found in the colon
and rectum.147 Grossly, the polyps range from a few millimeters
to several centimeters and present as smooth, firm, peduncu-
lated lesions that are lobulated, in contrast to juvenile polyps,
which are not lobulated. Peutz-Jeghers polyps are classified his-
tologically as hamartomas of the muscularis mucosa146 and
demonstrate strands of smooth muscle fibers that divide the
polyp into sectors (Fig. 93-7).148–150 Adenomas can occur con-
currently with Peutz-Jeghers polyps.
Etiology/Incidence
Peutz-Jeghers syndrome is rare and has an equal sex distribu-
tion; it has been described in all ethnic groups. Most cases of
Peutz-Jeghers syndrome are inherited in an autosomal domi-
nant pattern, but some develop de novo, most likely represent-
ing new, spontaneous mutations. Recently two groups of
investigators have defined the genetic mutation associated with
Peutz-Jeghers syndrome. In Peutz-Jeghers–affected individuals
many (but not all) have a mutation of a novel serine/threonine
kinase (LKB1 or STK11) with loss of kinase activity.151,152
Routine genetic testing and gene therapy for this disease is
under investigation but currently not available.153
Clinical Presentation
Because of the familial association of Peutz-Jeghers syn-
drome, the syndrome is often revealed in patients through
screening programs. If there is no family history, patients
usually present with crampy abdominal pain related to tran-
sient intussusception of a polyp. Abdominal radiographs will
often demonstrate dilated or partially obstructed small intes-
tine but rarely complete obstruction. Anemia resulting from
occult blood loss and malignant conditions are other present-
ing signs. Thirty percent of patients present with signs and
symptoms in the first 10 years of life, with 50% presenting
by 20 years of age.154,147
Several reports of intestinal tumors in association with Peutz-
Jeghers syndrome have been published.155–160 Malignant
FIGURE 93-7 Photomicrograph of a Peutz-Jeghers polyp demonstrating
a hamartomatous alteration of the muscularis mucosa. Smooth muscle
fibers divide the polyp into sectors.
changes in hamartomatous polyps of Peutz-Jeghers syn-
drome have been commonly reported.101,159,161–167 Separate
adenomatous and carcinomatous changes in hamartomas have
also been reported, which suggests that the adenoma-carcinoma
sequence occurs in the small intestine of patients with Peutz-
Jeghers syndrome.161 A review of 72 patients with the syn-
drome159 showed that 22% developed cancer. Nine cases were
gastrointestinal or pancreatic in origin, and seven were extra-
intestinal. Compared with an age-matched general popu-
lation, patients with Peutz-Jeghers syndrome had a relative
risk fordeath from gastrointestinal cancer alone that was 13 times
greater, and the risk for death from all cancers was 9 times greater.
The chance of dying from cancer by the age of 60 is
approximately 50% in patients with Peutz-Jeghers syndrome.146
Extraintestinal tumors associated with Peutz-Jeghers syn-
drome include ovarian, cervical, and testicular neoplasms.
Reported ovarian tumors include cystadenomas,154,168,169
granulosa cell tumors,135 and sex-cord tumors.170–172
Adenocarcinoma of the cervix can occur and is usually asso-
ciated with an ovarian tumor.174 Sertoli cell tumors of the
testicle have been found and cause gynecomastia in 50% of
cases. These tumors are usually benign but have malignant
potential175–177; thus an orchiectomy is recommended.
Cancer of the breast, thyroid, bile duct, pancreatic, and gall-
bladder have all been described in association with Peutz-
Jeghers syndrome.146
Treatment
Peutz-Jeghers syndrome should be suspected in any child
who presents with colicky abdominal pain or occult anemia
and melanotic pigmented spots. Recommendations for treat-
ment have changed over the past decade because of the
increasing concern of malignancy. The management proto-
col proposed by Phillips and Spigelman146 includes the
following annual evaluations: (1) symptoms related to polyps,
(2) blood count to detect anemia caused by blood loss,
(3) breast and pelvic examinations with cervical smears
and pelvic ultrasonography in girls, (4) testicular examina-
tion with ultrasonography in boys, and (5) pancreatic ultraso-
nography. In addition, esophagogastroduodenoscopy and
colonoscopy are recommended on a biennial basis, along
with small intestine contrast studies. Recently magnetic reson-
ance imaging (MRI) has shown promise as a surveillance
modality for small intestinal screening.178 Mammography is
recommended at 25, 30, 35, and 38 years of age, biennially
until 50 years of age, and then annually.
All polyps larger than 0.5 mm found at endoscopy should
be removed. Laparotomy with intraoperative enteroscopy
is recommended for removal of all small bowel polyps greater
than 15 mm in diameter. The previous practice of radical
intestinal resections should be avoided because of the recur-
rent nature of the polyps and the ensuing short-bowel syn-
drome that can occur. Any intestinal or extraintestinal
tumors should be treated aggressively. Historical data indicate
that the chance of patients with Peutz-Jeghers syndrome dying
by the age of 60 is close to 60% compared with 25% in an
age-matched general population.146
COWDEN DISEASE
Cowden disease is an autosomal dominant condition distin-
guished by multiple hamartomas that affects all three of the
germ layers. Patients with Cowden disease are at an increased
 CHAPTER 93 POLYPOID DISEASES OF THE GASTROINTESTINAL TRACT 1185

risk of developing breast, thyroid, and endometrial neopla-

sias.179,180 Eighty percent of patients present with dermatologic
lesions including facial trichilemmomas, acral keratoses, papil-
lomatous papules, or mucosal lesions. Other major but not
necessarily pathognomonic criteria include breast carcinoma,
thyroid carcinoma, macrocephaly, and endometrial carcinoma.
Only 35% of patients who meet the criteria for Cowden disease
will have gastrointestinal polyposis.7 The polyps are typically
juvenile polyps without any increased risk of developing a
gastrointestinal cancer.

Nonepithelial Polyps
------------------------------------------------------------------------------------------------------------------------------------------------

LYMPHOID POLYPS
History
The first case of lymphoid hyperplasia of the terminal ileum
was described by Marina-Fiol and Rof-Carballo in 1941 as
reported by Patel and Awen.181 Fieber and Schaefer182
reviewed eight cases previously described in the literature
and added four of their own in 1966. Byrne and colleagues183
reported only 44 cases in the world literature; however, lym-
phoid polyps were identified in 60% of children studied by
Franken for remote abdominal complaints.184 Thus the inci-
dence of lymphoid polyps is probably higher than reported,
but clearly most of them are asymptomatic, and, therefore,
never identified.

Pathology
FIGURE 93-8 Barium enema showing multiple, small 1- to 2-mm mucosal
Lymphoid polyps vary in size from a few millimeters to nodules throughout the entire colon and terminal ileum. The central umbil-
3 centimeters in diameter and are usually sessile in form. ication within these nodules seen on this postevacuation film is diagnostic of
The polyps are caused by elevation of hyperplastic submuco- diffuse lymphoid hyperplasia.
sal lymphoid aggregates. The overlying mucosa often becomes
ulcerated, which gives the polyp the volcano-like appearance.
Ulceration of the mucosa leads to occult blood loss. Hyperpla-
sia of the submucosal lymphoid tissue is believed to be caused
by nonspecific infections of childhood.1,2,4,108

Incidence
Lymphoid polyps tend to develop in young children within
the first few years of life as a result of exposure to new bacteria
and viruses.122 The peak incidence is at 4 years of age and
significantly diminishes by 5 years of age.

Clinical Presentation
Anemia resulting from blood loss and occasionally substantial
rectal bleeding are the usual presenting signs of lymphoid pol-
yps. Colonoscopy, air contrast barium enema, or both are the
diagnostic methods of choice for diagnosing lymphoid polyps
of the colon. An air-contrast barium enema will show small, uni-
form, umbilicated, polypoid filling defects that are distinct from
juvenile or adenomatous polyps (Figs. 93-8 and 93-9).183,184
Small elevations of otherwise-normal mucosa are seen on endos- FIGURE 93-9 An upper gastrointestinal series showing diffuse lymphoid
copy, and biopsy will confirm the diagnosis. Histologic evalua- hyperplasia throughout the stomach and small intestine.
tion reveals lymphoid aggregates with large germinal follicles.122
the lesions. Substantial uncontrolled bleeding and intussus-
Treatment
ception may require a more aggressive surgical approach.
Lymphoid polyps are benign, are self-limiting, and tend to
regress spontaneously.42,185 Once a histologic diagnosis is made, The complete reference list is available online at www.
expectant measures will usually be rewarded by regression of expertconsult.com.
 a successfully treated infant with a localized ileal perforation
that was described as a disease resembling NEC is attributed
to Agerty4 in 1943. Subsequently, in 1953 Schmid and
Quaiser5 first used the term necrotizing enterocolitis. In 1964
Berdon6 reported the clinical and radiographic findings of
21 patients with NEC. Then in 1975 Santulli7 first hypothe-
sized that the development of NEC had three essential com-
ponents: injury to the intestinal mucosa, the presence of
bacteria, and the availability of a metabolic substrate (to be
taken as the presence of enteral feedings). This characteriza-
tion remains a central tenet of our understanding of the overall
pathophysiology of NEC (discussed later). Over the past sev-
eral decades the management of infants with NEC has evolved
from aggressive early operation to supportive care with the
increasing realization that most infants can be managed, at
least initially, nonoperatively.
The subsequent seminal work of Bell and colleagues8 cod-
ified a severity-based classification scheme that is widely
accepted due to its simplicity and clinical utility in sugges-
ting therapy on the basis of likely outcomes. Bell’s criteria
(Table 94-1) can be summarized as indicating clinical findings
suspicious for NEC (Bell’s stage I), definitive NEC (Bell’s stage
II), and advanced NEC (Bell’s stage III). In general, a stage I
CHAPTER 94 infant manifests clinical criteria that raise suspicion without
definitive evidence such as pneumatosis intestinalis or bloody
stool. In stage II, or definitive disease, there is nearly always
evidence for pneumatosis intestinalis (Fig. 94-1). The hall-
Necrotizing mark of advanced disease is the appearance of pneumoperito-
neum or other clinical findings to suggest irreversible tissue
damage with perforation. In 1979 the International Classifica-
Enterocolitis tion of Diseases established a code for death from NEC,
thereby allowing more precise epidemiologic and outcome
analyses. Currently, the optimum surgical approach in order
Karl G. Sylvester, Gigi Y. Liu, and Craig T. Albanese to realize the best short- and long-term outcomes in infants
with intestinal perforation secondary to NEC remains the
subject of intense scrutiny via ongoing clinical trials.

Necrotizing enterocolitis (NEC) is an acquired inflammatory Incidence

disease that affects the gut of newborn infants nearly exclu- ------------------------------------------------------------------------------------------------------------------------------------------------

sively. Despite decades of research, NEC remains a leading Although the overall reported incidence of NEC among new-
cause of infant morbidity and mortality in neonatal intensive born infants is relatively low and reported to fall between 5%
care units (NICUs). Increased rates of preterm birth and and 10%.9–11 The true incidence is unknown given a number
advances in neonatal care have contributed to a growing pop- of either early or suspicious cases of NEC that cannot be
ulation of infants at risk for NEC. It is now the most common accurately tabulated. Conversely, the number of infants that
newborn surgical emergency and is associated with significant are under consideration for either having or developing
morbidity and mortality that exceeds all other gastrointestinal NEC can be quite high and is directly related to degree of pre-
(GI) conditions requiring surgical intervention. Although the maturity. Perhaps more puzzling, the incidence of NEC varies
precise pathogenesis remains incompletely understood, clin- significantly within the United States and throughout the
ical progress in recent years portends a shift in focus to pre- developed world. For example, the worldwide incidence
vention and the earlier identification of those infants most of NEC in very-low-birth-weight (VLBW, <1500 g) infants
at risk or with progressive disease. Despite the recent comple- varies from 1% to 2% in Japan, 7% in Austria, 10% in Greece,
tion of two successful prospective trials, the optimum surgical 14% in Argentina, and 28% in Hong Kong.12–16 The reasons
management of advanced disease with perforation remains for these disparities are unclear but are likely multifactorial
controversial. including biologic (e.g., genetic) and environmental (e.g., var-
iation in practice patterns). Irrespective of geographic report-
ing location, it is clear the incidence of NEC varies according
Historical Perspective to degree of prematurity and birth weight. NEC accounts for
------------------------------------------------------------------------------------------------------------------------------------------------
1% to 7% of all NICU admissions in the United States, or 1 to
Dating to at least 1888, there are several reports describing 3 cases per 1000 live births.9,17,18 In VLBW infants, the dis-
pathologic findings of intestinal perforation in neonates as ease occurs in approximately 10% to 12%, but ranges between
the cause of death suggestive of NEC.1–3 The first report of 2% and 22%, depending on the center of inquiry.11,15,19
1187
1188 PART VII ABDOMEN

TABLE 94-1
Modified Bell Staging Criteria for Necrotizing Enterocolitis
Stage Systemic Signs Abdominal Signs Radiographic Signs
IA Temperature instability, apnea, Gastric retention, abdominal Normal or intestinal dilation,
Suspected bradycardia, lethargy distention, emesis, heme-positive mild ileus
stool
IB Same as above Grossly bloody stool Same as above
Suspected
IIA Same as above Same as above, plus absent bowel Intestinal dilation, ileus,
Definite, mildly ill sounds with or without abdominal pneumatosis intestinalis
tenderness
IIB Same as above, plus mild metabolic Same as above, plus absent bowel Same as IIA, plus ascites
Definite, moderately ill acidosis and thrombocytopenia sounds, definite tenderness, with or
without abdominal cellulitis or right
lower quadrant mass
IIIA Same as IIB, plus hypotension, Same as above, plus signs of Same as IIA, plus ascites
Advanced, severely ill, bradycardia, severe apnea, combined peritonitis, marked tenderness, and
intact bowel respiratory and metabolic acidosis, abdominal distention
DIC, and neutropenia
IIIB Same as IIIA Same as IIIA Same as above, plus
Advanced, severely ill, pneumoperitoneum
perforated bowel

Modified from Neu J: Necrotizing enterocolitis: The search for a unifying pathogenic theory leading to prevention. Pediatr Clin North Am 1996;43:409-432 and
Caplan MS, Jilling T: New concepts in necrotizing enterocolitis. Curr Opin Pediatr 2001;13:111-115.

true frequency of NEC because the incidence is often defined

as the total number of cases of NEC divided by the total num-
ber of patients admitted to the NICU. Thus these figures
may include many premature infants who die within the first
several days of life and before enteric feedings and therefore
are unlikely attributable to NEC. In consideration of this, a
notable historical study that excluded early neonatal deaths
and included only infants who had been fed reported an
incidence of 15%.21

Epidemiology and Pathogenesis

------------------------------------------------------------------------------------------------------------------------------------------------

FIGURE 94-1 Plain abdominal radiograph demonstrating extensive
pneumatosis intestinalis (cystic and linear) and an arborizing pattern of
air over the liver shadow representing gas dispersed within the radicles
of the portal venous system.
Several confounding factors may account for the variability
in incidence reporting including the number and survival of
low-birth-weight (LBW) infants, the source of patient refer-
rals (inborn or outborn), and the diagnostic criteria used to
establish definitive NEC. A relatively large multicenter survey
of VLBW infants noted an incidence of 10.1% for definite NEC
(stage II) and 17.2% for suspected NEC (stage I), although
there was considerable intercenter variability.20 In another
study, the incidence of definite NEC versus suspected NEC
was 8.6% and 18.6%, respectively.18 The observed variability
in these and many other series may be underestimating the
EPIDEMIOLOGY
Age and Maturity
NEC is predominantly a disease of premature LBW infants
rather than those who are small for gestational age. It is esti-
mated that only 7% to 13% of all NEC cases occur in full-term
infants.22,23 Kliegman and Fanaroff24 reported that the mean
gestational age of 123 patients with NEC was 31 weeks (average
birth weight, 1460 g). Infants with extremely low birth weight
(ELBW) (<1000 g) and those 28 weeks’ gestational age or youn-
ger are at greatest risk.25,26 In a large multicenter prospective
study from the NICHD Neonatal Research Network involving
4438 infants weighing between 501 and 1500 g, Lemons11
demonstrated an inverse relationship between the incidence
of NEC and birth weight. Specifically, the incidence of NEC
was highest in infants weighing between 501 and 750 g
(14%) and declined with increasing weight: 751 to 1000 g
(9%), 1001 to 1250 g (5%), and 1251 to 1500 g (3%). These
findings have been confirmed by others23,27,28 and extended
to document an inverse relationship between the age at onset
of NEC and gestational age. Infants in whom NEC developed
in the first week of life were more mature (average gestational
age, 36.1 weeks) than those in whom NEC developed after

1 week of age (average gestational age, 33.4 weeks). Complica-
tions were more common, and the mortality rate was higher in
patients with early-onset disease. Wilson29 calculated the birth
weight–specific, weekly attack rate in patients with NEC and
found the risk period for NEC decreased as birth weight
increased. They found a consistent pattern of sharply declining
risk with attainment of age equivalent to 35 to 36 weeks’ gesta-
tion. From these observations the authors speculated that
functional maturation of the GI tract may play a principal role
in determining the risk for NEC.
Feedings
Approximately 90% of NEC cases develop in infants after feed-
ings are initiated.30–32 In a longitudinal cohort study reviewing
the incidence of NEC for a 3-year period before and after imple-
menting a “standardized feeding schedule” for infants weighing
between 1250 and 2500 g and less than 35 weeks’ gestation,
Kamitsuka33 reported an 84% reduction in the risk for NEC.
Other studies have also suggested an association between an in-
crease in the incidence of NEC and advancement of formula
feedings at rates greater than 20 kcal/kg/day.34,35 Despite these
reports, randomized trials failed to demonstrate any difference
in the incidence of NEC related to fast versus slow, early versus
delayed, or continuous versus intermittent bolus feedings.36–39
In a randomized trial involving 185 VLBW infants in which
slow (15-mL/kg/day increments; 10-day schedule to full feed-
ing) and fast (35-mL/kg/day increments; 5-day schedule to full
feeding) feeding advancements were compared, Rayyis39 dem-
onstrated no significant difference in the incidence of NEC
(13% vs. 9%), perforation (4% vs. 2%), and mortality (2%
vs. 3%) between groups. In a review of randomized trials com-
paring continuous versus intermittent bolus tube feeding for
premature infants weighing less than 1500 g, Premji and
Chessell38 found no significant differences in the incidence
of NEC between the two groups.
In a randomized trial investigating the incidence of NEC in
VLBW infants assigned to receive either minimal-volume feed-
ing (20 mL/kg/day) for 10 days before advancing to full-volume
feeding or a standard feeding advancement protocol (starting at
20 mL/kg and increasing by 20 mL/kg/day to full-volume feed-
ing), Berseth40 reported a significantly lower incidence of NEC
in the minimal-volume group than in the standard group (1.4%
vs. 10%). The prolonged use of “trophic feeding” volumes is
thought to trigger maturation of GI structure and function. This
study reinforces previous studies that gut stimulation protocols
are beneficial to VLBW infants41 and that initiation of a minimal-
volume feeding protocol for 7 to 10 days followed by modest
advancement of feeding may greatly reduce the incidence
of NEC.42 However, there are no contemporary studies that
specifically address this issue.
Pharmacologic Agents
Indomethacin is commonly used in premature infants to
treat a hemodynamically significant patent ductus arteriosus
(PDA). Indomethacin blocks prostaglandin synthetase, thus
causing vasoconstriction. Spontaneous gastrointestinal perfo-
ration (SIP) and NEC have been noted in LBW infants treated
with high-dose indomethacin.43,44 It has been postulated that
indomethacin increases mesenteric vascular resistance and
reduces mesenteric blood flow by 16% to 20%.45 Norton46
demonstrated that the use of indomethacin as a tocolytic agent
CHAPTER 94 NECROTIZING ENTEROCOLITIS 1189
was associated with an increased incidence of NEC in babies
delivered before 30 weeks’ gestation (mean age at delivery,
27.6 weeks), although indomethacin did not increase the
incidence of NEC in babies born after 32 weeks’ gestation.
Two randomized controlled trials involving more than 500
LBW premature infants receiving early low-dose indometha-
cin versus placebo for closure of a PDA demonstrated no
difference in the subsequent incidence of NEC.47,48 According
to several recent Cochrane Reviews, ibuprofen appears to be
as effective as indomethacin in leading to a PDA closure
and with fewer reported cases of NEC or SIP.49 Interestingly,
although the continuous infusion of indomethacin for a
PDA leads to fewer alterations in cerebral, renal, and mesen-
teric blood flow compared with bolus infusion, to date, insuf-
ficient evidence exists to demonstrate that this results in a
lowered risk of NEC or SIP.50 Earlier indomethacin may be
associated with increased incidence of SIP but protection
from NEC. Moreover, a PDA itself may predispose to NEC,
independent of indomethacin.
Cytokines and Growth Factors
Cytokines and growth factors play a critical role in mediating
the interaction among enterocytes, endothelial cells, fibroblasts,
and inflammatory cells that together are critical to the overall
cellular pathophysiology of NEC. These soluble factors direct
cellular proliferation, maturation, chemotaxis, and activation
in both the local gastrointestinal milieu and systemically to
effect the onset and progression of NEC (Table 94-2).51–56
Growth Factors
Epidermal growth factor (EGF) is known to be an important
trophic factor for the developing GI tract and has been shown
to be present in high concentration in human breast
milk.57,58 Typically, EGF is secreted into the gut lumen pri-
marily by the salivary and Brunner glands of the duodenum
and binds to EGF receptors that have been demonstrated
throughout the fetal and neonatal intestine, especially on
the basolateral membrane of enterocytes.59–64 EGF enhances
proliferation and differentiation of epithelial cells but also has
significant effects on healing of damaged mucosa and on in-
testinal adaptation after injury.62,65–67 Significantly reduced
levels of salivary and serum EGF have been demonstrated
in premature infants in whom NEC developed versus age-
matched controls.68,69 Furthermore, lower levels of salivary
EGF during the first week of life were associated with an
increased incidence of NEC in a recent clinical trial70 involv-
ing 327 premature and term neonates. Similarly, inactivation
of the EGF receptor in knockout mice has been shown to
result in hemorrhagic enteritis that is histologically similar
to NEC.71
Several animal studies that administered EGF provide in-
sight to the molecular mechanism underlying EGF-mediated
protection against NEC. In studies using a neonatal rat model
of NEC that involved asphyxia and cold stress, enterally
administered supplements of EGF have been shown to signif-
icantly decrease the incidence and severity of NEC in rat
pups72 through down-regulation of the proinflammatory
interleukin-18 (IL-18) and increased production of the anti-
inflammatory cytokine IL-10.73 Supplement of EGF in two
rat models has successfully reduced intestinal epithelial
cell apoptosis in the ileum, decreased intestinal permeability,
1190 PART VII ABDOMEN

TABLE 94-2
Summary of Important Growth Factors and Cytokines Contributing to the Pathogenesis of NEC
Protective
Effects in Trophic Effects
Cytokines Functions Proinflammatory Antiinflammatory Guts in Enterocytes
EGF and • Proliferation and differentiation of epithelial cells No Yes Yes Yes
HB-EGF • Healing of damaged mucosa
Epo • RBC proliferation No No Yes Yes
IL-1b • Macrophage activation, neutrophil recruitment, Yes No Unknown Yes
expression of endothelium adhesion molecules
• Production of IL-6, IL-8, PGE2
IL-4 • T- and B-cell and macrophage regulation Yes Yes Yes Unknown
• Differentiation of CD4 T cell into Th2 cells
IL-6 • Production of acute phase proteins, B-cell growth, T-cell Yes Unknown Unknown Unknown
proliferation, metalloproteinases, and GM-CSF
IL-8 • Attraction of neutrophils and basophils to site of Yes No No Unknown
inflammation
IL-10 • Decreases macrophage activation No Yes Yes Unknown
• Inhibition of proinflammatory cytokine production
IL-11 • Increases megakaryocyte and macrophage production No Yes Yes Yes
IL-12 • Production of IFN-g, Th1 and NK cell proliferation Yes Unknown Unknown Unknown
• Cytotoxic T lymphocyte and Th1 cell differentiation
• Macrophage activation and production of complement-
fixing antibodies
• Up-regulation of IL-18 receptor
IL-18 • IFN-g and B-cell antibody production Yes No Unknown Unknown
• Enhanced NK cell cytotoxic activity
• Activation and migration of neutrophils, phagocytosis,
and integrin expression
NO • Regulation of leukocyte-endothelial interaction and Yes Yes Yes No
platelet aggregation and adhesion
• Apoptosis from peroxynitrite when NO reacts with
superoxide
TNF-a • Cytokine release of IL-1b, IL-6, IL-8, NO, PGE2, matrix Yes Unknown Unknown Unknown
metalloproteases, PAF, and TXA2
• Inhibition of the release of glucocorticoids, TGF-b, and
IL-10
• Apoptosis induction
• Neutrophil activation and recruitment
PAF • Mesenteric vasoconstriction Yes Unknown Unknown Unknown
• Capillary leakage and increased intestinal mucosal
permeability
• Neutrophils and platelet activation
COX2 • Synthesis of proinflammatory prostaglandins Yes Unknown Unknown Unknown

Modified from Ledbetter DJ, Juul SE: Necrotizing enterocolitis and hematopoietic cytokines. Clin Perinatol 2000;27:697.

increased mucin production by goblet cells, and improved
overall intestinal structure.74,75
Heparin-binding epidermal-like growth factor (HB-EGF) is
a member of the EGF family of growth factors. HB-EGF ini-
tially was identified in conditioned medium of macrophage-
like cells as a mitogen for fibroblasts and smooth muscle
cells.76 The presence of HB-EGF has been reported in both
human amniotic fluid and milk.77,78 An HB-EGF knockout
mouse model has shown significantly increased intestinal per-
meability, delayed onset of angiogenesis, and increased inci-
dence and severity of NEC.79,80 According to several animal
studies,78,81–84 HB-EGF has been demonstrated to protect
developing intestinal epithelium from hypoxic necrosis and
cytokine-induced apoptosis, as well as to exert its cytopro-
tective effects via decreased reactive oxygen and nitrogen
species production. Interestingly, simultaneous administra-
tion of both EGF and HB-EGF did not result in any additional
protective effect against NEC.83
Erythropoietin (Epo) is a peptide produced by the kidneys
that regulates red blood cell production in response to anemia.
Since development of the recombinant protein, rEpo has
become widely used in the NICU.85,86 Epo has been found
in human breast milk, and functional Epo receptors have
been demonstrated in fetal and neonatal small intestine, thus

suggesting a possible role in GI development.87–89 In a retro-
spective study comparing 260 VLBW infants who received re-
combinant Epo (rEpo) with 233 matched controls, Ledbetter
and Juul90 demonstrated a significantly lower incidence of
NEC in the rEpo group (4.6%) than in the control group
(10.8%). Studies in neonatal rats given rEpo enterally have
demonstrated a dose-dependent increase in intestinal mucosa
villus surface area and increased cellular proliferation, thus
suggesting a role of rEpo as a trophic factor in the developing
small intestine.91 In a neonatal rat model of NEC involving
exposure to hypoxia and reoxygenation, pretreatment with
intraperitoneal injections of rEpo resulted in significantly
decreased mucosal inflammation and necrosis, which was
suggested to be mediated by decreased production of nitric
oxide (NO).92
Cytokines
Cytokines are endogenous mediators of the inflammatory
cascade. Proinflammatory and antiinflammatory cytokine pro-
duction is tightly regulated by complex feedback mechanisms
to maintain homeostasis.93,94 Overproduction of either may
have significant untoward effects. Overproduction of proin-
flammatory cytokines (IL-1, IL-2, IL-6, IL-8, IL-12, tumor ne-
crosis factor-alpha [TNF-a], interferon) and platelet-activating
factor (PAF) may lead to shock, multiorgan failure, and
death.95,96 Overproduction of antiinflammatory cytokines
(IL-4, IL-10, IL-11) may result in excessive suppression of
immune function.97,98 A number of different inflammatory
mediators have been implicated in the pathogenesis of NEC,
several of which are highlighted later.99,100
IL-1b in the gut can be found in macrophages, neutrophils,
intestinal epithelial cells, endothelial cells, fibroblasts, dendritic
cells, smooth muscle cells, and enteric glia.101,102 Microbial
products, inflammation, and TNF-a trigger its release.101,102
Upon binding to the IL-1 receptor, IL-1b and its receptor acti-
vate the transcription factor NF-kB, which triggers release of
acute phase proteins, IL-6, IL-8, and PGE2.101,102 Elevated
IL-1b has been detected in full-thickness specimens of NEC
intestine.103 Edelson104 detected a greater level of IL-1 recep-
tors late in the course of severe NEC. Both IL-1b and IL-1
receptor may serve as markers for progressive disease.
Levels of IL-6 increase in the presence of microbes, micro-
bial products, TNF-a, and IL-1b.101,105 Upon binding to IL-6
receptors, which are only expressed on hepatocytes and some
leukocytes, IL-6 triggers the STAT-4 pathway, resulting in
production of acute phase proteins, B cell growth, antibody
production, T-cell proliferation, and enhanced activity of he-
matopoietic growth factors such as granulocyte-macrophage
colony-stimulating factor (GM-CSF).101,105 Furthermore, it
leads to the production of tissue inhibitors of metalloprotei-
nases and inhibition of superoxide production.106 Elevated
IL-6 has been reported in the plasma and stool of babies
with NEC,107 but its mRNA expression in surgical intestinal
specimens is not any higher than the controls.105
IL-8 is released in response to LPS, TNF-a, and IL-1b.108
After binding to the chemokine receptors CXCR1 and CXCR2,
which signal through phospholipase C and PI3-kinase, res-
pectively, IL-8 is responsible for neutrophil activation and
migration into tissues, as well as the production of acute phase
proteins.108 In a study evaluating serial serum levels of two
proinflammatory cytokines (IL-1b, IL-8) and two antiinflam-
matory cytokines (IL-1 receptor antagonist [IL-ra], IL-10) in
CHAPTER 94 NECROTIZING ENTEROCOLITIS 1191
infants with NEC, Edelson and colleagues104 demonstrated
significantly higher levels of IL-8 and IL-10 in infants with se-
vere NEC from the onset of disease through 24 hours than in in-
fants with less severe NEC. Nadler109 investigated the pattern of
cytokine expression in infants undergoing resection for severe
NEC and infants undergoing intestinal resection for other in-
flammatory conditions. Significant up-regulation of IL-8 mRNA
was seen in the specimens from infants with NEC as compared
with controls.
IL-12 is released in response to bacteria, bacterial products,
and viruses and exerts its effect on T cells and NK cells upon
binding to its receptor.110 Its immunologic functions include
the following: IFN-g production, Th1 and NK cell prolifera-
tion, cytotoxic T lymphocyte and Th1 cell differentiation,
macrophage activation, and production of complement-fixing
antibodies.110 Halpern111 localized IL-12 via immunohisto-
chemistry to monocytes in the intestinal mucosa and lamina
propria in a neonatal rat model of NEC.
Tissue IL-18 levels peak in response to lipopolysaccharide
(LPS or endotoxin), Fas ligand, gram-positive bacteria exo-
toxins, and IL-12 and subsequently induce the production
of TNF-a and IL-1b.112 Binding to the IL-18 receptor results
in NF-kB activation and promotes Th1 or Th2 lineage matu-
ration depending on the underlying genetic influence and
cytokine environment.113 Heninger114 discovered infants with
stage III NEC or above to have a higher frequency of the
IL-18607 AA genotype, thus suggesting that this variation
may be used for predicting the outcome of NEC.
IL-4, produced by Th2 cells, mast cells, B cells, and stroma
cells, is a key regulator in humoral and adaptive immunity.100,120
It has been demonstrated to possess cytoprotective effects in
human intestinal epithelial cells by reducing bacterial trans-
location, increasing leukocyte superoxide production, and in-
ducing decay-accelerating factor, which protects the host from
the attack of autologous complement activation.115,116 Genetic
studies by Treszl117 revealed that infants with NEC were less
likely to possess the IL-4 receptor a-chain mutant allele com-
pared with infants without NEC. Treszl117 speculated that this
mutant allele is associated with enhanced transduction of IL-4
signals, which shifts the development of lymphocytes to a more
pronounced Th2 state.
IL-10, produced by Th2 cells, monocytes, and B cells, is an
inhibitor of proinflammatory cytokine production and of sev-
eral accessory cell functions of the macrophage, T cell, and
natural killer (NK) cell lines.118 Furthermore, IL-10 has also
been shown to decrease the production of metalloprotei-
nases119 and suppress iNOS mRNA and NO expression in
small bowel, liver, and serum.120 It is postulated that dimin-
ished production of IL-10 in preterm infants resulted in
persistent up-regulation of the inflammatory response and
therefore increased susceptibility in the preterm neonate to
long-term tissue damage after acute inflammatory condi-
tions.121 Although Edelson and colleagues104 found signifi-
cantly higher levels of IL-10 in infants with severe NEC
from the onset of disease through 24 hours than in infants
with less severe NEC, in a neonatal rat hypoxia-reoxygenation
model of NEC, recombinant IL-10 administered subcutane-
ously was found to significantly attenuate the extent of intes-
tinal injury when compared with control animals, thus
suggesting a protective effect of its antiinflammatory proper-
ties.122 Taken together, these findings indicate that IL-10
has the potential to be a marker of severe disease and may
1192 PART VII ABDOMEN
be considered a therapeutic target to function as a strong
cytokine inhibitor. The high level of IL-10 in severe NEC, in
contradistinction to less progressive NEC, suggests a significant
role for antiinflammatory mediators in the pathophysiology of
NEC to dampen the inflammatory response.
IL-11 or adipogenesis inhibitory factor is a member of the
IL-6 type cytokine family that uses the gp130 receptor subunit
for intracellular signal transduction. IL-11 has been shown to
be a pleiotropic cytokine that promotes epithelial regeneration
and enhances adaptation after bowel resection.123 Subcutane-
ous administration of recombinant IL-11 in rats undergoing
placement of a defunctionalized (Thiry-Vella) loop of intestine
or massive small bowel resection has resulted in prevention of
mucosal atrophy in the defunctionalized loop and enhanced
mucosal adaptation and absorptive function in the remaining
intestine after resection.124,125 In addition, Nadler109 noted
significant up-regulation of IL-11 mRNA in infants undergo-
ing resection for severe NEC and found an inverse correlation
between IL-11 expression and the likelihood of pan-necrosis,
thus suggesting that IL-11 secretion may be an adaptive
response to limit the extent of intestinal damage.
TNF-a has many proinflammatory effects including neu-
trophil activation, induction of leukocyte and endothelial
adhesion molecules, and induction of other cytokines such
as PAF.126 TNF-a has been demonstrated to produce profound
hypotension and severe intestinal necrosis similar to NEC in
animals. This effect has been shown to be mediated by PAF
and attenuated by PAF receptor antagonists.127 Confirmatory
studies in human neonates with NEC identified significantly
elevated plasma levels of TNF-a compared with controls.99
Pentoxifylline, a drug that has multiple effects including inhi-
bition of production of TNF-a, was shown to reduce bowel
necrosis in an adult rat ischemia-reperfusion model of bowel
injury, as well as the incidence of NEC in a neonatal rat
model.128 Furthermore, two separate studies129,130 have dem-
onstrated a significant reduction in the severity of NEC in neo-
natal rats after intraperitoneal TNF-a antibody prophylaxis.
However, the A(-308) and A(-238) allele variants of the pro-
moter region of the TNF-a gene, which were reportedly
associated with higher TNF-a production, fail to show any
influence on the risk and course of NEC in VLBW infants.131
A tremendous amount of investigation has been performed
to define the role of nitric oxide (NO) in the pathogenesis of
a number of different inflammatory processes. Evidence
supports a possible dichotomous function of NO as both a
beneficial and a detrimental molecule, especially with regard
to the GI tract.132,133 NO is produced from arginine by three
isoforms of nitric oxide synthase (NOS). NOS-1 (neuronal,
nNOS) and NOS-3 (endothelial, eNOS) are constitutively
present at low levels in the small intestine. NOS-2 (inducible,
iNOS), which is expressed in the myenteric plexus, endo-
thelial cells, gastric epithelial cells, and enterocytes,134,135 is
a form that can be induced in response to inflammatory
cytokines.136,137 The constitutive forms of NOS and constitu-
tive levels of NO have been demonstrated to modulate a
number of important functions in the GI tract including
maintenance of mucosal integrity, regulation of mucosal per-
meability, modulation of water and electrolyte transport,
regulation of blood flow, regulation of motility, and inhibition
of leukocyte adhesion and activation.138–142
Inhibition of NO synthesis in a variety of animal models of
intestinal injury induced by ischemia-reperfusion, LPS, or PAF
has resulted in marked exacerbation of mucosal injury.143–148
Administration of exogenous sources of NO including
L-arginine, sodium nitroprusside, and nitroglycerin greatly
attenuates these detrimental effects.149–152 In a prospective
study of 53 premature infants, Zamora153 demonstrated a
significantly lower plasma arginine level at the time of diag-
nosis in infants in whom NEC developed compared with
controls. In the only randomized, prospective, placebo-
controlled study of VLBW infants assigned to receive either
daily L-arginine supplements (261 mg/kg) or placebo for the
first 28 days of life, Amin154,155 found a significantly lower
incidence of NEC in the supplement group (6.7%) than in
the control group (27.3%). Throughout the study, the group
that received arginine supplementation had significantly
higher mean plasma arginine levels than the control group
did. Interestingly, in both groups the infants in whom NEC
developed had significantly lower plasma arginine levels at
the time of diagnosis than their respective peers did. It is
not known whether the decreased arginine levels represent
increased utilization for NO production, consumption for
the synthesis of other proteins, or decreased enteral absorp-
tion. This study suggests the possible benefits of arginine
supplementation and potentially other NO donors in the
prevention of NEC.
The inducible form of NO synthase (NOS-2, iNOS) is in-
duced in response to inflammatory cytokines. Within several
hours of stimulation, NOS-2 expression and activity within
the intestinal epithelium increase up to 15-fold and result
in the production of large amounts of NO.135 Although the
low levels of NO produced by the constitutive isoforms of
NOS may play a homeostatic role in the GI tract, sustained
release of NO as a result of up-regulation of NOS-2 has been
suggested to have deleterious effects by inducing cellular
injury and failure of the mucosal barrier.135,156,157 This is
thought to occur by the reaction of excess NO with superoxide
(O2
) to produce peroxynitrite (ONOO
), a potent reactive
nitrogen intermediate that may trigger cytotoxic processes
including lipid peroxidation and DNA damage.157–159 In a
prospective study investigating NOS-2 expression in the intes-
tine of infants undergoing resection for NEC, Ford160 demon-
strated marked up-regulation of NOS-2 gene expression in the
intestinal epithelium and increased apoptosis of enterocytes in
the apical villi. In addition, increased levels of nitrotyrosine
residues were detected in the apical villi, thus suggesting that
the mucosal injury and increased apoptosis were mediated
through the formation of NO and peroxynitrite. In a neonatal
rat hypoxia model of NEC in which breast milk–fed ani-
mals were compared with formula-fed animals, Nadler and
colleagues161 demonstrated a significantly higher incidence
of NEC, NOS-2 expression, and enterocyte apoptosis along
with decreased IL-12 mRNA in the formula-fed group than
in the breast milk–fed group. The decreased IL-12 expression
is thought to be mediated by NO and theorized to contribute
to intestinal injury by attenuating bacterial clearance. Further-
more, Whithouse has shown in a rat model that the progres-
sion of NEC to intestinal ischemia is associated with a shift
from nitric oxide to superoxide production by the intestinal
vascular endothelium.162
PAF, an endogenous phospholipid inflammatory media-
tor, has been shown to play an important role in the pa-
thophysiology of intestinal injury in both animal and
human studies.12 PAF has diverse biologic effects including

mesenteric vasoconstriction, capillary leakage, increased intes-
tinal mucosal permeability, and neutrophil and platelet acti-
vation.163–165 Clinically, increased PAF levels have been
demonstrated in formula-fed premature infants, as well as in
those in whom NEC developed.100,166 In a neonatal rat model
of NEC, Caplan167 showed that intestinal PAF concentrations,
intestinal phospholipase A2 (PAF-synthesizing enzyme) mRNA
expression, and intestinal PAF receptor mRNA expression are all
elevated. In other animal experiments, exogenous PAF admin-
istration results in severe bowel necrosis168; endogenous intes-
tinal production of PAF is up-regulated in response to various
stimuli including LPS, hypoxia, and TNF-a127,169–171; and ad-
ministration of PAF receptor antagonist, or PAF acetylhydrolase
(PAF-degrading enzyme), reduces the risk for NEC.170,172,173
In human studies, PAF acetylhydrolase activity has been dem-
onstrated to be present in human breast milk; it has also been
shown to be decreased in neonates, with levels approaching
adult enzyme activity at around 6 weeks of life, and has
been found to be deficient in infants with NEC.99,174,175 In a
prospective study of 164 infants at risk for NEC, Rabinowitz176
monitored serial plasma levels of PAF and PAF-related lipids
(PAF-LL) to investigate the changes that occur with NEC. There
was a significantly higher peak in PAF-LL levels in infants in
whom NEC developed than in controls. In addition, rising
PAF-LL levels were positively correlated with progression of
the severity of NEC; these levels returned to baseline levels
during recovery. With the studies mentioned, a systematic
review56 of serologic tests in diagnosing NEC suggested that
PAF is one of the better performing markers for NEC with a sen-
sitivity of 92% and specificity of 84% despite the considerable
heterogeneity between the studies.
Cyclooxygenase (COX) catalyzes the rate-limiting step of ara-
chidonic acid metabolism into prostaglandins, leukotrienes, and
thromboxanes.177 Two isoforms of the COX enzyme have been
identified. COX-1 is constitutively expressed in many tissues
including the GI tract.178 COX-2 is the inducible form that is
expressed in inflammatory conditions of the GI tract such as
Reduced proteolytic enzymes
Decreased peristalsis
Intestine
immaturity Altered bacterial flora
Too much PAF
PAF
MAMPs
PRRS
TLRs FPRs
Immature mucosal
immune system NFkB MAP
pathway kinases
Exaggerated
inflammatory
response Transcription of
cytokines
Increased intestinal injury and apoptosis
FIGURE 94-2 Schematic summarizing the pathologic features that contribute to the pathogenesis of necrotizing enterocolitis.
CHAPTER 94 NECROTIZING ENTEROCOLITIS 1193
inflammatory bowel disease.179,180 Proinflammatory cytokines
(IL-1, IL-6, TNF-a), as well as the proinflammatory transcri-
ption NF-kB, have been shown to increase COX-2 expres-
sion.181 NF-kB is an important protein in the activation of a
number of inflammatory mediators and cytokines.182 A marked
increase in COX-2 expression has been demonstrated in intes-
tine resected from infants with severe NEC.183 To elucidate
the mechanisms involved in COX-2 expression in NEC,183
Chung demonstrated a coordinated induction of NF-kB activa-
tion and COX-2 expression during the early phases of injury in a
neonatal rat model of NEC.
LPS (endotoxin) is a bacterial product that has the capacity to
produce potent inflammatory responses through the induction
of various proinflammatory cytokines such as PAF and TNF. Re-
cent studies suggest that LPS binds with pattern recognition
receptors on the intestinal epithelial barrier, such as Toll-
like receptors (TLRs), formylated peptide receptors (FPRs), or
nucleotide-binding oligomerization domain-like receptors
(NODs), to trigger mitogen-activated protein kinase (MAPK),
nuclear factor kappa-B (NF-kB), and caspase-dependent
pathways, thus leading to various inflammatory responses
(Fig. 94-2).184 Furthermore, Grishin185 has demonstrated that
LPS stimulates p38 MAPK-dependent expression of cyclooxy-
genase-2 (COX-2) in a rat model of NEC. Systemic injection
of LPS has been shown to produce hypotension, shock, and se-
vere intestinal necrosis.25 It has been used in animal models of
NEC to reliably generate intestinal injury that resembles NEC.186
PATHOGENESIS
Despite many years of extensive investigation and the identi-
fication of several key risk factors, the precise pathogenesis of
NEC remains elusive. The etiology is likely multifactorial
and involves a combination of mucosal compromise, patho-
genic bacteria, and enteral feedings that in a susceptible
host results in bowel injury and an inflammatory cascade
(see Fig. 94-2). Of the risk factors, prematurity is the most
Altered epithelial
membrane, tight
junction, and mucus
proteins
Increased
intestinal
NODs epithelial
permeability
Caspases
Translocation
of
microbes
Apotosis
IFN-1
1194 PART VII ABDOMEN
consistent along with the enteral feeding of formula. Com-
mercially available formulas lack many of the beneficial
properties present in breast milk. These protective agents in-
clude gut trophic hormones, factors that induce intestinal
maturation, factors that enhance colonization by nonviru-
lent bacteria, antiinflammatory mediators, vitamins, antioxi-
dants, and components that support cellular and humoral
immunity.187–190
Evidence for an Impaired Gut Barrier
The preterm GI tract is characterized by an immaturity of cel-
lular and humoral immunity,191,192 increased permeability,193
reduced gastric acid secretion,194 reduced concentration of
proteolytic enzymes,195 incomplete innervation and decreased
motility,196,197 and immaturity of the intestinal epithelium and
microvilli barrier function.198 Decreased barrier function is
evident in otherwise healthy preterm infants by their ability
to systemically absorb and deliver undigested macromole-
cules, whole bacteria, and LPS.199–201 Compromise of the
intestinal epithelial barrier appears to be the first event leading
to activation of the inflammatory cascade.
Reduced mucosal blood flow leading to cellular hypoxia
and injury is one of the most frequently cited etiologic factors
for NEC. Touloukian202 emphasized the high incidence of
perinatal physiologic stressors that may primarily or second-
arily cause intestinal ischemia. Chief among these factors are
hypoxic and hypotensive episodes, exchange transfusions
through the umbilical vein, umbilical artery catheters, cardio-
vascular lesions, and serum hyperviscosity. The “diving reflex”
in which blood is preferentially diverted away from the
splanchnic circulation in order to maintain adequate perfu-
sion of the heart and brain has been hypothesized to occur
during these episodes.203 During periods of low blood flow
and subsequent reperfusion, Parks204 found that a reaction
among xanthine oxidase, hypoxanthine, and molecular oxy-
gen results in a burst of superoxide radical production. These
free radicals may cause damage to cellular and mitochondrial
membranes and alter the permeability of the intestinal muco-
sal barrier. Although animal studies provide support for this
theory, clinical correlation has been lacking, and prospective
clinical trials have not been able to consistently establish an
association between a hypoxic event and the development
of NEC.205
Extensive investigation into the critical role of various
inflammatory mediators in the pathogenesis of NEC has been
conducted. Studies in animal models and human specimens
have identified PAF, LPS, NO, and TNF-a as leading potential
mediators of NEC.160,168,169 Increased mucosal permeabi-
lity and susceptibility allowing translocation of bacteria or
bacterial toxin and activation of the inflammatory cascade
are thought to be the critical steps leading to the final collapse
of intestinal epithelial integrity.206 Recent studies have sug-
gested that the disruption in the intestinal mucosal barrier
results from accelerated apoptosis. All of the key mediators
identified in NEC (PAF, NO, LPS, TNF-a) have been shown
to cause apoptosis of intestinal epithelial cells.160,207–209
In a neonatal rat model of NEC involving formula-fed ani-
mals exposed to hypoxia and cold stress, Jilling210 demon-
strated a marked increase in apoptosis in the epithelial
layer of the NEC group in comparison with mother-fed con-
trols. The accelerated apoptosis was shown to precede gross
morphologic changes in the intestinal epithelium. Administra-
tion of a pan-caspase inhibitor to inhibit intestinal apoptosis
resulted in a significantly reduced rate of epithelial apoptosis,
as well as a decreased incidence of NEC, thus suggesting that
apoptosis was the underlying cause of the subsequent mucosal
damage, ultimately leading to NEC.
Sustained overproduction of NO secondary to up-regulation
of NOS-2 in the GI tract in response to an inflammatory stim-
ulus has been suggested to induce cellular injury and disrup-
tion of the intestinal epithelial barrier through a variety
of mechanisms.156 Ford160 demonstrated up-regulation of
NOS-2 and NO in the intestinal wall of infants with NEC
and increased apoptosis of enterocytes mediated by the po-
tent oxidant peroxynitrite. This accelerated apoptosis is
thought to result in a break in the villus tip of the intestinal
mucosal barrier where bacteria may attach, translocate, and
initiate an inflammatory cascade.211 In addition, peroxyni-
trite has also been shown to inhibit the proliferation of intes-
tinal epithelial cells in a rat model of NEC.212 The data
suggest that in conditions associated with sustained overpro-
duction of NO or peroxynitrite (e.g., NEC), intestinal epithe-
lial barrier dysfunction may result from an imbalance caused
by accelerated epithelial injury and blunted tissue repair
mechanisms.
Role of Infectious Agents
The type of feeding and pattern of intestinal colonization may
determine the risk for development of NEC. Breast-fed infants
become colonized predominantly with bifidobacteria (gram-
positive bacteria) that help control the growth of gram-negative
bacteria.213,214 In contrast, formula-fed infants become colo-
nized predominantly by coliforms, enterococci, and Bacteroides
species.215
The intestinal mucosa serves as a barrier to the largest mi-
crobial challenge to the human body. Beyond serving as a sim-
ple physical barrier, the dynamic interaction between the
intestinal mucosa and colonizing microbes is an integral part
of the innate immune system and is regulated by a system of
pattern recognition receptors (PRR). During gut colonization,
crosstalk between the system of PRR and commensal bacteria
occurs, resulting in intestinal mucosal tolerance or hypore-
sponsiveness toward the enteric microbes in order to establish
a symbiotic relationship. The PRR respond to microbial asso-
ciated molecular patterns (MAMP) (i.e., LPS, flagellin, pepti-
doglycans) in order to modulate the cellular response through
a variety of downstream signaling pathways (see Fig. 94-2).
The PRR system includes the TLRs, which are largely respon-
sible for sampling and interpreting the extracellular environ-
ment, and in this capacity, the TLRs have been implicated as
an integral mechanism to the pathogenesis of NEC.184
Hackam and colleagues have shown that the LPS receptor,
TLR4, is integral to intestinal mucosal repair in rodent
models of NEC through its effects on enterocyte apoptosis
and migration.216 Work by this same group has demonstrated
increased expression of TLR4 in the human intestine that
develops NEC. There is an intense and ongoing interest in this
line of investigation as a primary determinant in establishing a
symbiotic gut-microbe axis, which may provide protection
against the development of NEC.
The importance of bacteria in the pathogenesis of NEC
is supported by the following evidence: (1) NEC occurs in
episodic and epidemic waves in which affected patients are
 CHAPTER 94 NECROTIZING ENTEROCOLITIS 1195

related in place and time or had the same infectious agent217,218;

(2) during clustered occurrences, the identical microorganisms
can be isolated from both afflicted babies and their care-
takers219; (3) NEC can occur in infants with no known risk
factors; (4) NEC can develop several weeks or months after a
perinatal insult, when the GI tract is fully colonized and has
had sufficient time to recover from any perinatal insult220,221;
(5) administration of large doses of vitamin E (interferes with
intracellular killing of bacteria by leukocytes) to premature
infants has been linked with an increase in the incidence of
NEC222; (6) a NEC-like disease occurs in vulnerable hosts after
the ingestion of Clostridium species223; (7) lesions resembling
NEC can be reproduced experimentally with the administra-
tion of LPS169; (8) endotoxemia is demonstrated in 80% of
those with NEC and positive blood cultures for gram-negative
bacteria224; and (9) pneumatosis intestinalis is a common radio-
graphic finding and represents submucosal gas collections
produced by bacterial fermentation.225,226
FIGURE 94-3 Intraoperative photograph of intestine with necrotizing
Unifying Hypothesis for enterocolitis demonstrating areas of hemorrhagic necrosis and gangrene.

Necrotizing Enterocolitis
------------------------------------------------------------------------------------------------------------------------------------------------

Taken together, empiric and experimental data suggest that gangrene, the serosal surface is black or, in the most advanced
NEC occurs in a vulnerable host that has become further com- cases, bland gray to white, given the complete loss of per-
promised at the level of the gastrointestinal tract. Infant prema- fusion. Subserosal gas collections are frequently encountered.
turity and care in the NICU conspire to result in the initiation of The mucosal surface may be ulcerated with wide areas of
bacterial insult or invasion of the immature GI tract, whose key epithelial sloughing. Bloody peritoneal fluid is seen when
functions including barrier function and immune modulation bowel necrosis is present, and brown and turbid fluid is seen
are altered. As a result, the interaction among the enterocyte, when perforation has occurred.
immune effector cells, and resident microbiota initiates an Histologically, enteric inflammation is nearly ubiquitous in
inflammatory cascade that becomes unbalanced, resulting in NEC (Fig. 94-4). The degree and nature, however, vary from
progressive enteric mucosal injury and increased permeability. one area to another. Acute and chronic inflammatory changes
In recent years the key role of the gastrointestinal tract as an im- coexist in 60% of cases. The most common microscopic
mune system organ has been increasingly documented.227,228 lesion is bland or coagulation necrosis of the superficial mucosa
Because the microbiota of the human GI tract has increasingly (89%).231 Edema and hemorrhage of the submucosa follow
been identified as playing a key role in overall human health complete mucosal necrosis (see Fig. 94-4). Pneumatosis intesti-
through this symbiotic relationship, gut colonization during nalis is initially seen in the submucosa and later in the mus-
the newborn period likely presents a particularly vulnerable cularis and subserosa. Bacteria in the bowel lumen and wall
time for innate immune system, human gut, and gut microbial are present in up to 40% of cases and are occasionally found
community dysfunction to occur when this process is disrupted in gas cysts. Transmural necrosis, characterized by hyaline
or delayed. The process of gut colonization has been elegantly eosinophilia and loss of nuclear detail in the muscular layers,
documented to involve both environmental and genetic fac- is present in advanced disease. Epithelial regeneration, forma-
tors as evidenced by the findings in twins of similarities in tion of granulation tissue, and early fibrosis are often present
gut microbiota.229 Moreover, the process is active and dynamic, during the resolution of NEC. This suggests a suppurative
undergoing a significant shift even in well newborns and infants process lasting at least several days. Granulation tissue with
throughout the first year of life. mucosal and submucosal fibrosis may be seen adjacent to areas
of active mucosal and submucosal necrosis and may account, in
part, for late stricture formation. Thrombi are sometimes noted
Pathology in small mesenteric vessels and in small arterioles of the sub-
------------------------------------------------------------------------------------------------------------------------------------------------
mucosa. Small-vessel thrombosis within necrotic tissue is
NEC may involve single (50%) or multiple (discontinuous) considered a secondary change. Large-vessel thrombosis is a
segments of intestine, most commonly in the terminal ileum, relatively rare finding at autopsy.
followed by the colon.230 Involvement of both the large and
small intestine occurs in 44% of cases. Pan involvement (pan-
necrosis, NEC totalis) is a fulminant form of NEC characterized Diagnosis
by necrosis of at least 75% of the gut, and it accounts for 19% of ------------------------------------------------------------------------------------------------------------------------------------------------

all cases of surgically treated NEC and most of the deaths.26

CLINICAL FEATURES
At surgery, the gross appearance of NEC is fairly constant.
The bowel is markedly distended with patchy areas of thin- NEC is commonly heralded by nonspecific clinical findings
ning (Fig. 94-3). The serosal surfaces are typically red to gray that simply represent physiologic instability.231,232 These
and may be covered by a fibrinous exudate. With frank findings include lethargy, temperature instability, recurrent
1196 PART VII ABDOMEN
A from the 1970s.233,235 O’Neill236 demonstrated that in a cohort
B
FIGURE 94-4 A, Histologic section demonstrating early necrotizing
enterocolitis with inflammation and pneumatosis intestinalis in the sub-
mucosa (hematoxylin-eosin stain, 150). B, Histologic section demon-
strating advanced necrotizing enterocolitis with transmural necrosis and
loss of villus and crypt architecture (hematoxylin-eosin stain, 150).
apnea, bradycardia, hypoglycemia, and shock. More specific
symptoms related to the GI tract include abdominal distention
(70% to 98%), blood per rectum (79% to 86%), high gastric
residuals after feeding (>70%), vomiting (>70%), and diarrhea
(4% to 26%). Gross blood in the stool is present in 25% to 63%
of cases and occult blood in 22% to 59%. Rectal bleeding is
seldom massive. Because the spectrum of disease severity var-
ies, physical examination may initially demonstrate only subtle
abdominal distention and minimal tenderness. As the disease
progresses, abdominal palpation may elicit tenderness and
demonstrate palpable bowel loops, a fixed or mobile mass, or
abdominal wall crepitus. Edema and erythema of the abdomi-
nal wall as a result of the underlying peritonitis are present
initially in approximately 4% of cases but are more common
later in the course of the disease. In males, there may be discol-
oration of the scrotum, indicative of perforation. In a small sub-
set of patients, the disease is rapidly progressive and the initial
manifestation is heralded by florid clinical findings and death
within 24 hours.
LABORATORY FINDINGS
Infants with NEC usually have neutropenia, thrombocyto-
penia, and metabolic acidosis. The total leukocyte count may
be elevated, but it is generally low. In one study, 37% of infants
had absolute neutrophil counts less than 1500 cells/mm3.233
Infants with the lowest counts in this study had the worst prog-
nosis. Neutrophil counts less than 6000 cells/mm3 are most
commonly associated with concomitant gram-negative septice-
mia. Some authors234 have advocated that surgical treatment
should be considered in infants older than 34 weeks’ gestation
if they have lower total neutrophil counts, a higher immature
neutrophil number, and a greater immature-to–total neutrophil
ratio at first presentation of NEC.
Thrombocytopenia is nearly universally present and seems
to be associated with gram-negative sepsis and platelet binding
by endotoxin. The incidence of thrombocytopenia in NEC is
65% to 90% and essentially unchanged from the earliest reports
of 40 infants who underwent surgery for NEC, 95% had platelet
counts less than 150,000 cells/mm3. Rowe and colleagues237,238
found that platelet counts less than 150,000 cells/mm3 were
present in patients who had positive cultures for gram-negative
organisms. The nadir platelet count during the course of the
disease was noted to be lower in patients with more severe dis-
ease and in those who died.239 A platelet count less than 104/L
or a rapid fall is a poor prognostic indicator. A retrospective
single-center study of 91 infants by Kenton240 suggests that a
platelet count less than 104/L within 3 days of the diagnosis
of NEC should warrant surgical intervention to decrease the
likelihood of bowel gangrene and its attendant morbidity and
mortality.
EMPIRIC AND EXPERIMENTAL INDICATORS
OF NEC AND ITS SEVERITY
Metabolic acidosis is common (40% to 85% of patients with
NEC) and is believed to result from hypovolemia and sepsis.
It is not a specific indicator of intestinal necrosis. Stool sam-
ples are commonly positive for occult blood and reducing
substances. Book241 reasoned that intestinal mucosal damage
from NEC leads to carbohydrate malabsorption. Poorly dig-
ested carbohydrates pass into the colon, where they are fer-
mented and excreted in stool. The authors tested the stool of
formula-fed infants for reducing substances with reagent
(Clinitest) tablets and found that 71% of formula-fed infants in
whom NEC developed had greater than 2þ reducing substances
in their stool. Colonic bacterial fermentation increases the
local production of D-lactate, which is absorbed and excreted
by the kidneys. Garcia242 could show elevated urinary D-lactate
levels in infants with NEC but not in control infants. With recov-
ery from NEC or administration of enteral antibiotics, D-lactate
excretion decreased. Similarly, hydrogen excretion in the breath
is elevated when fermentation is increased. This test is helpful
in ruling out NEC. A negative result on a breath hydrogen test
is 99% accurate in ruling out NEC.243 Abubacker244 indicated
via a study of 24 infants with NEC that a preoperative blood
lactate level of greater than 1.6 mmol/L carries a poor pro-
gnosis with mortality odds ratio of 22 (CI 1.54 to 314.3,
P ¼ 0.04). Finally, a multicenter study of 473 infants with
NEC by Moss and colleagues245 identified metabolic acidosis

at diagnosis (pH < 7.3 or bicarbonate < 16) as one of the
12 parameters that may assist in predicting the progression
of NEC.
In an uncontrolled retrospective study comparing two
centers’ criteria for surgical intervention, Tepas246,247 found that
if three of seven critical metabolic derangements (positive blood
culture within 96 hours of diagnosis, pH < 7.25 or receiving
bicarbonate, bandemia > 20%, serum sodium < 130 mEq/L,
platelet count < 50,000 cells/mm3, MAP < gestational age or
on any pressors, and absolute neutrophil count < 2000) were
identified, this permitted earlier identification of infants needing
exploration and resulted in better surgical outcomes (mortality
and the need for postoperative parental nutrition) than did
delaying operation until radiographic proven evidence of perfo-
ration by pneumoperitoneum.
C-reactive protein (CRP), an acute phase reactant, has been
measured in an attempt to correlate its level with the presence,
absence, or resolution of the disease.248 CRP may serve as
an early indicator of NEC when levels rise more than 10 mg/L
within 48 hours of the suspected diagnosis (reported sensiti-
vity, 92%; specificity, 81%). Failure of CRP to return to normal
within 10 days was an indicator of abscess, stricture, or septi-
cemia. A systematic review of various biomarkers for NEC56
selected six qualified studies248–253 on CRP that together
indicated CRP is a relatively sensitive but nonspecific marker
for NEC, yielding a combined diagnostic odds ratio (DOR) of
5.82 and an area under the curve (AUC) of 0.70. The diagnostic
odds ratio expresses how much greater the odds of having
the disease are for people with a positive test result rather than
for the people with a negative test result. In this case, although
the DOR is high, its AUC is only 70%, implying a weaker
diagnostic performance.
Intestinal fatty acid–binding protein (I-FABP) is a small
(14 to 15 kd) cytoplasmic protein of mature enterocytes.
It is released into the circulation upon death of enterocytes,
thus representing intestinal injury when detected in high
concentration. Lieberman,254 Edelson,255 and Guthmann256
have demonstrated an elevated level of I-FABP in the serum
of neonates with NEC. Given its small size and its ability
to pass through the glomerulus, urinary I-FABP has been
detected as potential marker of NEC by Derikx.257 Even-
nett258 further characterized the release of urinary I-FABP
by correlating its I-FABP-to-creatinine concentration with
the severity of NEC. The I-FABP-to-creatinine concentration
was significantly higher in infants with extensive disease
than in those with focal disease (7.4 pg/mmol [2.1 to
35 pg/mmol] vs. 1.1 pg/mmol [0.3 to 1.7 pg/mmol], res-
pectively, P ¼ .002). Although studies have shown higher
concentrations of I-FABP in infants with severe NEC,
Thuijls259 concluded urinary I-FABP not to be an effective
screening tool for NEC given its ability to identify merely
one third of neonates with NEC in a group of 226 neonates
with no clinical suspicion of NEC originally. This
study was performed on the basis of a cutoff point of
2.20 pg/nmol creatinine derived from testing urinary I-FABP
on 35 neonates suspected of NEC. However, Thuijls259
repeated the conclusion by Evennett258 that urinary I-FABP
is a promising prognostic marker for NEC.
Calprotectin, a calcium-binding protein found in neu-
trophils and macrophages, is a marker of inflammation of
the gastrointestinal tract. It has been found in stool due to
transepithelial migration of myeloid cells and has been a
CHAPTER 94 NECROTIZING ENTEROCOLITIS 1197
useful marker for exacerbation of inflammatory bowel
disease in children.260 Josefsson found fecal calprotectin to
be elevated greater than 2000 mg/g feces in NEC with per-
foration with microscopic bowel inflammation but less than
2000 in cases with focal disease. Thus similar to I-FABP, it
may be a useful marker for disease severity but not a strong
screening tool.
Microbiology
------------------------------------------------------------------------------------------------------------------------------------------------
It has proven extremely difficult to identify common offending
infectious agents in infants with NEC. Organisms recovered
from the blood and stool of patients with NEC vary depending
on the GI tract flora, the nosocomial flora, the site cultured,
and the duration of previous antibiotic therapy. It is unclear
whether the bacteria cultured represent pathogens causing
NEC or, instead, secondary opportunistic invaders selected
by the antibiotic regimen. Furthermore, lack of sufficiently
matched controls and incomplete bacteriology data on other
patients in similar environments makes study of the micro-
biology of NEC difficult.
BACTERIOLOGY
Bacteriologic data for NEC have primarily been based on
cultures obtained from the blood, stool, and peritoneal cavity.
Blood cultures are positive in 30% to 35% of patients.215
Cultures commonly grow Escherichia coli, Klebsiella pneumo-
niae, Proteus mirabilis, Staphylococcus aureus, Staphylococcus
epidermidis, enterococci, Clostridium perfringens, and Pseudomo-
nas aeruginosa. K. pneumoniae and E. coli cause the majority of
positive blood cultures. The organisms most frequently
cultured from stool specimens are E. coli, K. pneumoniae, Entero-
coccus cloacae, P. aeruginosa, Salmonella species, coagulase-
negative staphylococci (S. epidermidis), C. perfringens, Clostridium
difficile, and Clostridium butyricum.261–263 Peritoneal cultures
most commonly grow Klebsiella species, E. coli, coagulase-
negative staphylococci, Enterobacter species, and yeast.264
FUNGAL CULTURES
In contrast to bacteria, no data implicate a fungus as an
initiating organism in the pathogenesis of NEC. Fungi are
believed to be secondary invaders. Fungal septicemia with
Candida species has been implicated in many late NEC
deaths.265,266 In a retrospective premortem and postmortem
examination of body fluid and tissue cultures from 30 patients
who died of NEC by Smith and colleagues,265 47% of patients
had evidence of fungal infection and colonization was
found in 20%.
CLUSTERED EPIDEMICS
In most large series, sporadic cases have been followed by the
sudden appearance of a cluster of a relatively large number of
cases. In many of the epidemics, no specific pathogen has been
identified. During these episodes, nursery personnel experi-
enced concomitant acute GI illnesses.21,267 In those few studies
in which a specific pathogen is identified, it was in association
with documented contamination of milk formula or milk
fortifier that was used in the NICU.268 Identified species have
1198 PART VII ABDOMEN

included E. coli, Salmonella species, C. difficile, rotavirus, noro-

PORTAL VEIN GAS
virus, torovirus, and a coronavirus-like organism.217,268–274
Portal vein gas (PVG) appears as linear branching radiolu-
cencies overlying the liver and often extending to its periphery
Imaging (see Fig. 94-1). It represents gas dispersed through the fine rad-
------------------------------------------------------------------------------------------------------------------------------------------------
icles of the portal venous system. The presence of portal vein
The cornerstone of the diagnosis of NEC remains plain ante- gas is fleeting, which perhaps accounts for the low reported
roposterior and left lateral decubitus radiographs. Any or all of incidence of 10% to 30%.293 In most series the presence of por-
the following findings are associated with NEC: ileus pattern tal vein gas is associated with a poor prognosis.205 In cases with
(nonspecific bowel distention), pneumatosis intestinalis (lin- pan-intestinal involvement, PVG is present in 61% of patients.
ear or cystic) (see Fig. 94-1), portal vein gas, pneumoperito- The genesis of portal vein gas may involve accumulation of gas
neum, intraperitoneal fluid, and persistently dilated, fixed in the bowel wall as a result of bacterial invasion, dissection into
loops.275 Both pneumatosis and portal vein gas are often fleet- the venous system, and migration to the radicles of the portal
ing signs but are considered pathognomonic in the appropri- vein. Alternatively, it may represent the action of gas-forming
ate clinical setting of prematurity. In more recent years, there bacteria within the portal venous system itself.
has been a trend and tendency toward the increased use of
PNEUMOPERITONEUM
bedside ultrasound with Doppler imaging of the GI tract in
an effort to identify more subtle findings like abdominal fluid, Free air in the peritoneal cavity associated with perforation of
hyperemia, decreased blood flow to the gut, and pneumatosis the intestine can be demonstrated in 12% to 30% of patients.
intestinalis, all of which may not be well seen by plain radio- It is best noted on the left lateral decubitus or cross-table lateral
graph.276–291 Despite this trend, to date there have been few view. Upright radiography is unnecessary. A supine view of the
good comparative studies documenting the clinical utility of abdomen can demonstrate free air by outlining the falciform lig-
ultrasound in the diagnosis of NEC. ament (“football sign”), the umbilical artery, or urachal remnants
or by revealing the “double-wall” sign. This sign refers to visu-
alization of air on both sides of the wall (lumen and peritoneal
BOWEL DISTENTION cavity). In patients who have intestinal perforation proven by
surgery, radiographic evidence shows free air in only 63%, thus
Multiple gas-filled loops of intestine are the earliest and most demonstrating that perforation can occur in a surprisingly high
common radiologic finding in patients with NEC (55% to number of patients without evidence of pneumoperitoneum.
100% of cases).292 As fluid and air accumulate, air-fluid levels One must also recall that pneumoperitoneum may occur with-
are visible on the decubitus view. The degree of dilatation and out intestinal perforation from mechanical ventilation for severe
the distribution of bowel loops are related to the clinical lung disease. In this clinical situation, barotrauma may produce
severity and progression of the disease. In some cases, non- alveolar rupture with air dissection into the abdomen through
specific intestinal dilatation precedes clinical symptoms the mediastinum. The patient’s signs, symptoms, and laboratory
suggestive of NEC by several hours. findings will often differentiate the cause of the air. If one is
unsure, abdominal paracentesis may be performed and any
aspirated fluid analyzed. If there is no ascites, a water-soluble
PNEUMATOSIS INTESTINALIS contrast study via the gastric tube may be performed to rule
out gastric perforation.
Demonstration of pneumatosis intestinalis (intramural gas)
in the appropriate clinical setting is diagnostic of NEC (see
INTRAPERITONEAL FLUID
Fig. 94-1). The air mainly comprises hydrogen, a byproduct
of bacterial metabolism. The frequency of pneumatosis intes- Several plain radiographic findings suggest free fluid in the
tinalis ranges from 19% to 98%, although it may be absent in peritoneal cavity that is amenable to paracentesis: (1) a grossly
up to 14% of patients with NEC (even severe disease).205 distended abdomen devoid of gas, (2) gas-filled loops of bowel
Conversely, extensive pneumatosis may be present with min- in the center of the abdomen surrounded by opacity out to the
imal signs; it often responds promptly to medical manage- flanks, (3) increased haziness within the abdomen, and (4) sep-
ment. Pneumatosis is fleeting, may appear before the onset aration of bowel loops. These findings have been reported in
of clinical symptoms, and is commonly an early rather than 11% of cases. Both ascites and portal vein gas are radiographic
a late finding. Pneumatosis is most frequently noted when findings associated with high mortality rates. Twenty-one per-
infants have been fed (84%), in contrast to unfed babies cent of patients with surgically proven intestinal perforation
(14%).31 Pneumatosis intestinalis is not specific for NEC have ascites. However, 16% of all patients with proven intesti-
and has been noted in infants with enterocolitis of Hirsch- nal perforation have neither ascites nor pneumoperitoneum
sprung disease, inspissated milk syndrome, pyloric stenosis, on plain radiographs.
severe diarrhea, carbohydrate intolerance, and other disor-
ders. Two forms of pneumatosis intestinalis are recognized
PERSISTENT DILATED LOOPS
radiographically: cystic and linear. The cystic form has a
granular or foamy appearance and represents gas in the sub- The “persistent dilated loop sign” is a plain radiographic find-
mucosa. It is often confused with fecal material in the large ing that was described by Wexler294 in a study of five babies
intestine. Linear pneumatosis consists of small bubbles col- with NEC in whom a single loop or several loops of dilated
lected within the muscularis and subserosa to form a thin lin- bowel remained unchanged in position and configuration
ear or curvilinear gas pattern outlining the wall of a segment for 24 to 36 hours. Full-thickness necrosis subsequently
of the intestine. developed in these patients. This finding, however, does not

always indicate bowel necrosis. Leonard295 found a persistent
loop in 33% of 21 patients with proven NEC. Fifty-seven per-
cent of infants with a persistent loop had necrotic intestine at
surgery or autopsy, but necrosis never developed in 43% and
they recovered with nonoperative treatment.
CONTRAST STUDIES
Radiopaque contrast studies of the upper GI tract may occa-
sionally be useful to improve diagnostic accuracy in patients
with equivocal clinical and radiologic signs of NEC. However,
overdiagnosis by contrast radiography is possible and may lead
to unwarranted treatment. Careful attention to the type of con-
trast agent used for the study is critical.294 Barium should never
be used because extravasation of a barium and stool mixture
through a perforation may intensify the peritonitis. Unlike bar-
ium, water-soluble contrast agents are absorbed by both the
bowel and the peritoneal cavity. The practical implication of this
absorption is a transient (6 to 12 hours) increase in urinary spe-
cific gravity. Historically, water-soluble agents were hyperosmolar
and caused dangerously large intraluminal fluid shifts, especially
in premature patients. Current water-soluble agents are non-
ionic, have much lower osmolarity, and produce excellent opa-
cification of the GI tract. NEC is suspected when intestinal
contrast enhancement demonstrates bowel wall loops separated
by edematous walls, an irregular mucosa with ill-defined mar-
gins, mucosal ulceration, bowel wall spiculation, or pneumatosis
intestinalis. Currently, the use of contrast studies is usually
reserved for interrogation of the GI tract after the acuity of
NEC has resolved and in order to examine for stricture formation
during or after recovery. Though advocated by some,296 contrast
enemas should not be performed because of the risk for rectosig-
moid perforation. Unless there is colonic disease or reflux of
contrast into a diseased distal ileum, the contrast enema will
not be diagnostic of NEC. It may, in fact, overdiagnose NEC
because contrast enemas have been shown to produce pneuma-
tosis and transient portal venous air. In an infant who has recently
completed a course of therapy for NEC but has persistent signs of
partial small bowel obstruction or blood-tinged stools, contrast
enemas may be useful in identifying strictures.
ULTRASONOGRAPHY
Ultrasonography (US) has been used to identify necrotic bowel,
intraperitoneal fluid, and portal venous air. However, abdominal
sonography can depict the following findings highly suggestive
of nonviable bowel over plain abdominal radiography: presence
of intra-abdominal fluid, thinning of the bowel wall, reduction
of bowel wall perfusion, abnormal bowel loops, and intermit-
tent gas bubbles in liver parenchyma and the portal venous
system.297 The abnormal bowel loops on US are characterized
by a hypoechoic rim with a central echogenic focus (“target
sign”).278 Appearing as pericholecystic hyperechogenicity,
the intermittent gas bubbles in liver parenchyma and the portal
venous system are believed to represent either pericholecystic
venous gas or extension of the foamy inflammatory infiltrate of
NEC into the pericholecystic space.298,299
Theoretically, US may have significant value if it can identify
patients who require operation in a more sensitive and timely
manner than is otherwise possible with conventional clinical
and radiographic methods. Although Silver299 has correlated
seven sonographic findings with adverse outcomes needing
surgical intervention, more studies are necessary to compare
the sensitivity and specificity, as well as intraobservable and
CHAPTER 94 NECROTIZING ENTEROCOLITIS 1199
interobservable agreements between sonographic and radio-
graphic findings. Thus at this time, the use of US for the diagnosis
of NEC is most applicable to patients with questionable clinical
and radiologic findings or to localize intra-abdominal fluid for
paracentesis.
MAGNETIC RESONANCE IMAGING
Magnetic resonance imaging (MRI) is a noninvasive modality
that has recently been used to identify infants with ischemic
bowel secondary to NEC. Although MRI is capable of demon-
strating the cardinal findings of NEC, its utility is limited.300
Classification
------------------------------------------------------------------------------------------------------------------------------------------------
To select the appropriate treatment (nonoperative vs. opera-
tive) and to determine the impact of therapy on survival and
late outcomes, it is essential that investigators use comparable
criteria for classifying the stages of NEC. Several classification
schemes have been proposed. In 1978 Bell8 introduced the
now most commonly used three-stage classification system
(suspected, definite, and advanced) that categorizes patients
by historical factors, GI manifestations, radiologic findings,
and systemic signs (see Table 94-1). The Bell staging criteria
have been modified220; the three stages are still used, but sub-
sets are included in an effort to identify specific prognostic fac-
tors. Infants with stage I disease have features suggestive of
NEC, patients with stage II disease have definitive NEC with-
out an indication for surgical intervention, and patients with
stage III disease have advanced NEC with evidence of bowel
necrosis or perforation.
Management
------------------------------------------------------------------------------------------------------------------------------------------------
NONOPERATIVE
In the absence of intestinal necrosis or perforation, the main-
stay of treatment for patients with NEC is supportive. Feed-
ings are discontinued, the GI tract is decompressed through
a sump gastric tube, and intravenous fluid resuscitation is ini-
tiated. A complete blood count, platelet count, blood gas anal-
ysis, and CRP and serum electrolyte levels are obtained. Blood
and urine samples are sent for culture, and broad-spectrum
intravenous antibiotic therapy is initiated. Until recently, most
antibiotic regimens included a penicillin, an aminoglycoside,
and an agent effective against anaerobic organisms. It seems
logical that coverage for anaerobic organisms be included be-
cause these infants are usually 1 to 2 weeks old and have or are
undergoing colonization by coliforms. To date, no controlled
study has shown the efficacy of this regimen. The antibiotic
regimen is best tailored not only to the most common organ-
isms found with NEC but also to the nosocomial nursery flora.
Because of recent reports of patients with stool and blood
cultures positive for coagulase-negative staphylococci, some
groups now empirically treat patients with a combination
of vancomycin and gentamicin or vancomycin and a third-
generation cephalosporin.301 The incidence of fungal sepsis
in infants who die of NEC is high, so a strong index of suspi-
cion must be maintained and empirical antifungal therapy
should be considered if the patient’s clinical course is pro-
longed. Historically, close clinical observation consists of fre-
quent physical examination, two-view abdominal radiography
1200 PART VII ABDOMEN
performed every 6 to 8 hours, serum platelet and leukocyte
counts, and blood gas analysis. However, one could question
the utility of the frequent use of plain radiographs at this
interval, given the number of cases that have progressive
NEC without pneumoperitoneum and depending on the
individual practitioners’ trigger for operative intervention.
Patients with definite disease of moderate severity (Bell
stage II) are normally treated by bowel rest, decompression,
and antibiotic therapy for at least 7 to 14 days. A central
venous catheter may be placed for total parenteral nutrition.
If the patient is clinically well, small amounts of formula
may be restarted. The infant is constantly and carefully
monitored for abdominal distention, vomiting, or nonspecific
signs or symptoms of NEC. Once feedings resume, stools are
tested for reducing substances and occult blood. Feedings are
discontinued if the result of either test becomes positive.
Infants who have undergone surgery receive 1 to 2 weeks of
postoperative intravenous antibiotics. Feedings are initiated
when the patient is clinically well and return of bowel function
has been established.
INDICATIONS FOR OPERATION
The principal goals of surgical intervention in the setting of
NEC are to remove gangrenous bowel and preserve intestinal
length.302,303 On the basis of historical clinical experience,
many argue that exploration should not be undertaken until
gangrene is present but should be performed before perforation
occurs. Unfortunately, no combination of clinical examination
or adjunct testing has been shown to have high sensitivity
for intestinal gangrene.304,305 Thus there remains controversy
regarding the indications for operation, the most appropriate
timing of intervention, and the optimal surgical treatment
strategy. The most widely accepted indication is the presence
of pneumoperitoneum. Relative indications include a posi-
tive paracentesis, palpable abdominal mass, abdominal wall
erythema, portal venous gas, fixed intestinal loop, and clinical
deterioration despite maximal medical therapy.
In an attempt to identify characteristics that may serve as
predictors of intestinal gangrene, Kosloske304 reviewed 12
criteria used as indications for operation in 147 patients with
NEC and stratified these criteria according to sensitivity, spec-
ificity, positive/negative predictive value, and prevalence. The
“best” indicators (specificity and positive predictive value
[PPV] approaching 100%, prevalence > 10%) were pneumo-
peritoneum, positive paracentesis (aspiration of > 0.5 mL
brown or yellow fluid containing bacteria on Gram stain),
and portal venous gas. “Good” indicators (specificity and
PPV approaching 100%, prevalence < 10%) were a fixed in-
testinal loop, erythema of the abdominal wall, and a palpable
abdominal mass. A “fair” indicator (specificity of 91%, PPV of
94%, prevalence of 20%) was “severe” pneumatosis intesti-
nalis as graded by a radiographic system. “Poorer” indicators
were clinical deterioration (PPV of 78%), platelet count
lower than 100,000/mm3 (PPV of 50%), abdominal tender-
ness (PPV of 58%), severe GI hemorrhage (PPV of 50%),
and gasless abdomen with ascites (0%). Unfortunately, none
of the indicators had sensitivity greater than 48%.
Instead of looking at indicators separately, Coursey306
combined various radiologic findings into a 10-point Duke
Abdominal Assessment Scale (DAAS) to improve intraob-
server and interobserver agreement on diagnosing severe
NEC. A subsequent validation study by the same authors307
demonstrated that for every one-point increase in the DAAS
score (AUC ¼ 0.83), infants with NEC were significantly
more likely to need surgical intervention. Although the
study did not define a specific score cutoff for surgical inter-
vention, 93% of the operated infants in this study had a score
of 7 or above.
Pneumoperitoneum
Infants in whom pneumoperitoneum develops during the
management of NEC should undergo either laparotomy or
peritoneal drain placement. Unfortunately, pneumoperito-
neum is not always demonstrable in neonates with gut
perforation, with one study reporting that only 63% of infants
with perforation demonstrated free air.308
Paracentesis
A positive result on paracentesis, defined as free-flowing as-
piration of more than 0.5 mL of brown or yellow-brown fluid
that contains bacteria on Gram stain,156 is highly specific for
intestinal necrosis. A negative result on paracentesis is rare
with intestinal necrosis but can occur when a localized,
walled-off perforation is present or a segment of bowel is
injured but not perforated.309 Currently, there is no absolute
indication for paracentesis. Kosloske304 recommends abdom-
inal paracentesis for patients with extensive pneumatosis
intestinalis or for those who do not improve with nonopera-
tive management. If no peritoneal fluid is aspirated, peritoneal
lavage is performed by instilling up to 30 mL/kg of normal
saline solution into the peritoneal cavity, turning the patient
from side to side, and then withdrawing the fluid. Ricketts
and Jerles309 reported a greater than 70% survival rate when
a positive result on abdominal paracentesis was used as the
indication for exploration in 51% of their patients; three
false-negatives occurred. They performed paracentesis when
there was erythema and edema of the abdominal wall, portal
vein gas, a fixed and dilated loop on sequential abdominal
radiography, a fixed and tender abdominal mass, or persistent
clinical deterioration. The indications used for paracentesis in
this report are considered indications for operation by many
surgeons.
Portal Venous Gas
To determine the significance of portal vein gas in relation to
the presence and extent of bowel necrosis and mortality,
Kurkchubasche310 reviewed the experience of Children’s
Hospital of Pittsburgh, as well as the world literature. Of
the 616 patients collected, 118 (19%) had portal vein gas
on plain radiography. Of these 118 patients, 102 underwent
operation, usually 24 hours after the radiographic appear-
ance of portal vein gas. All 102 patients had full-thickness
bowel necrosis, and 52% had necrosis of more than 75%
of the length of the entire small intestine. The overall surgical
mortality rate for patients with portal vein gas was 52%, and
more than 90% of those with pan involvement died. Of
the 15 patients who had portal vein gas and did not undergo
immediate exploration, bowel necrosis requiring subsequent
operation developed in 6 (40%) and 5 of the 6 died. In a
separate study, Rowe26 suggested that intestinal necrosis will
develop in more than 90% of infants with portal venous
gas, with pan involvement developing in 52%. More recently,
in a review of 40 infants with NEC and portal venous gas,

Molik311 reported a 54% overall operative mortality and pan
involvement in 25%.
Fixed Persistent Intestinal Loop
A fixed dilated intestinal loop is defined by persistent location
and configuration for more than 24 hours. In a recent study,
approximately half the patients with evidence of a fixed
dilated loop recover without an operation.312 The finding of
a fixed loop is considered concerning by most observers as
indicating a pregangrenous or severely compromised intesti-
nal loop, yet to date there is little objective evidence to support
that assumption. Moreover, at this time there is no general
consensus on the utility of fixed loop in guiding operative
management.
Clinical Deterioration
Many criteria for operating on children who show continued
clinical deterioration despite adequate supportive therapy
exist,313 but none by themselves are absolute indications.
These criteria include abdominal wall erythema, peritonitis
on physical examination, persistent/increasing acidosis, and
persistent/progressive thrombocytopenia.
Ascites
As mentioned, pneumoperitoneum may not always be evident
with intestinal perforation. A gasless abdomen, suggestive of
a fluid-filled abdominal cavity, may be the only indication
of perforation. Frey308 reported that in 21% of infants with
intestinal perforation, the only radiographic evidence was
ascites. They noted that radiographs are imprecise and that
evidence of ascites in the appropriate clinical setting of NEC
mandates paracentesis for further evaluation.
OPERATIVE MANAGEMENT
Advanced disease requiring surgical intervention develops in
up to 50% of infants with NEC.314,315 Until recently, given
the lack of quality prospective randomized trials, the optimal
surgical treatment strategy for NEC remained controversial.
Surgical goals are to remove gangrenous bowel and preserve
intestinal length. Within this context, a number of different
options exist, but most agree that the surgical approach should
be determined by the extent of intestinal involvement. The
patient’s general condition should be optimized before opera-
tion with aggressive ventilatory support, treatment of shock,
administration of broad-spectrum antibiotics, and correction
of anemia and coagulopathy. Operative procedures may be
performed in the NICU under appropriate conditions without
an increase in complications.316
Primary Peritoneal Drainage
Treatment of intestinal perforation in VLBW infants remains
controversial. In 1977 Ein317 reported the use of peritoneal
drainage (PD) as a means of stabilizing and improving the
systemic status of premature infants with perforation before
laparotomy. Since then, primary treatment with PD has been
used in a variety of settings, and some investigators have sug-
gested that it may serve as definitive therapy. More recently,
several studies have attempted to address this issue more
rigorously.318–333 Currently, most surgeons propose PD as
the initial treatment in ELBW infants (<1000 g at birth) with
perforated NEC to allow resuscitation and stabilization before
CHAPTER 94 NECROTIZING ENTEROCOLITIS 1201
definitive laparotomy. In an attempt to address this issue, two
multicenter prospective randomized controlled trials were
conducted. In the NECSTEPS trial of 117 infants (<34 weeks’
gestation and < 1500 g) by Moss,324 19 of 55 infants with PD
died (34.5%), as compared with 22 of 62 infants with laparot-
omy (LAP) (35.5%, P ¼ 0.92) by 90 days postoperatively.
The percentages of infants who became dependent on total
parenteral nutrition were 17 of 36 (47.2%) in the PD group
and 16 of 40 (40%) in the laparotomy group (P ¼ 0.53).
Although this study was originally designed to detect a reduc-
tion in 90-day mortality from 50% to 25% with a statistical
power of 82%, it failed to enroll the desired number of study
subjects in the time allotted to the study, thus reducing the
power to 77% or less.
Three primary conclusions were drawn from this prospec-
tive randomized trial. The trial found that the type of opera-
tion performed for intestinal perforation in infants with
NEC (1) does not significantly affect 90-day mortality, (2) does
not affect rate of dependence on total parenteral nutrition at
90 days postintervention, and (3) did not affect the total
length of hospital stay. The results of this trial have led to a
follow-up study that seeks to determine the difference in
median to long-term outcomes comparing the two surgical
procedures with an emphasis on neurodevelopmental out-
comes. Two years following the publication of the NECSTEPS
trial results, Rees323 published the results from the European
NET trial that included 69 patients. Similar to the results in
the NECSTEPS trial, there was no significant difference in
the 6-month survival rate between infants undergoing either
laparotomy or PD as primary treatment for perforated NEC.
Six-month survival was 18 of 35 (51.4%) with PD and 21
of 33 (63.6%) with LAP (P ¼ 0.3; Fisher exact text, difference
12%, Cl, (11, 34%). Cox regression analysis showed no signif-
icant difference between groups (hazard ratio for PD 1.6; P ¼
0.3; 95% CI, 0.7 to 3.4). However, in the PD group, delayed
laparotomy was performed in 26 of 35 (74%) patients after a
median of 2.5 days (range, 0.4 to 21) and did not improve
6-month survival compared with primary laparotomy (relative
risk of mortality 1.4; P ¼ 0.4; 95% CI, 0.6 to 3.4). Unlike the
conclusion derived by the NECSTEPS trial, Rees concluded
that PD is ineffective as either a temporizing measure or de-
finitive treatment because of the high percentage of infants
requiring “rescue” laparotomy.
Even though the two prospective studies did not demon-
strate statistically significant differences in mortality between
PD and LAP, the most recent meta-analysis by Sola334 (273 PD,
250 LAP), which selected two clinical trials and three prospec-
tive cohort studies325,326 out of the 12 studies reviewed, indi-
cated an increased mortality of 55% with PD (OR 1.55, 95%
CI, 1.08 to .22, P ¼ 0.02) without statistical heterogeneity.
Upon careful review, four of the five studies showed mortality
rates of PD to be higher than LAP and the authors attributed
the higher rates to more premature and smaller infants
selected for the PD group. Thus controversies persist in initi-
ating PD for stabilization before laparotomy. Currently, in an
effort to address intermediate and long-term morbidity,
a multicenter trial sponsored by the NICHD Neonatal
Research Network was initiated in 2010. The primary objec-
tive of this trial is to determine the rate of primarily neuro-
developmental impairment at 2 years of age in ELBW
infants with NEC who undergo either PD or laparotomy as
primary treatment of perforated NEC.
1202 PART VII ABDOMEN
Laparotomy
At laparotomy, the extent of NEC may be classified as focal,
multifocal, or pan-intestinal (<25% viable bowel). Depending
on the extent of disease and patient characteristics at the time
of surgery, a number of different surgical options may be
undertaken including resection with enterostomy, resection
with anastomosis, proximal enterostomy, the “clip-and-drop”
technique, and the “patch, drain, and wait” technique. The
abdomen is entered via a transverse supraumbilical incision
with precautions taken to not injure the liver. Optionally,
samples of peritoneal fluid may be harvested for culture of
aerobic, anaerobic, and fungal organisms. The entire GI tract
is systematically examined to assess the extent of disease and
viability of the bowel. If clear demarcation of gangrenous
bowel is encountered, it is resected. Marginally viable bowel
can be preserved yet defunctionalized either through a prox-
imal diversion (enterostomy) or clip-and-drop technique
wherein the intestine is closed at either end using hemoclips
in an effort to prevent further passage or leakage of the fecal
stream. These latter techniques are particularly useful if the
overall length of involved intestine is such that the infant
would be rendered “short-gut” if all marginally involved intes-
tinal length is removed. A second-look operative approach to
extensive NEC involvement is currently not widely practiced.
Infants tend to be quite ill with significant ongoing physiologic
instability following most operative interventions for NEC.
At the conclusion of the procedure, one should record the
length of viable intestine remaining and note the presence
or absence of the ileocecal valve.
A rarely cited complication of laparotomy for NEC is spon-
taneous intraoperative liver hemorrhage from injury caused by
retractors or finger dissection. VanderKolk335 reported this
complication in 11.8% of operations for NEC over a 5-year
period. The mean gestational age of those with liver hemor-
rhage was 28 weeks (mean weight, 1262 g). Only one of these
patients survived. The authors identified low mean preopera-
tive arterial pressure and high preoperative fluid administration
over the preceding 24 hours as significant predisposing factors.
This complication occurred shortly after the abdomen was
opened, and the intestine was eviscerated. Liver congestion
was followed by subcapsular hematoma and then free rupture.
Focal Disease
When a single area of bowel is necrotic or perforated, only lim-
ited resection is necessary. Creation of a proximal enterostomy
and distal mucus fistula has been the standard of care. Stomas
can be created either through the ends of the incision or
through a separate exit site on the abdomen. Factors to be con-
sidered when deciding on optimized stoma sites include mes-
entery that has been shortened by inflammation that may
impede exteriorization, placement and fitting of future stoma
bags without leakage, ease of future closure if diverting ends
are placed in close proximity or remotely, and incidence of ad-
ditional complications including wound infection and stoma
prolapse. Most enterostomies are created by suturing the
intestine to the fascia with interrupted sutures. About 2 cm
of bowel are left protruding from the abdominal wall, and
no attempt is made to “mature” the end of the intestine. If
stoma viability is in question postoperatively, a small portion
of the full thickness of the intestine is excised at the bedside
and the cut ends are observed for bleeding.
Resection with primary anastomosis for isolated disease
may be performed in carefully selected patients. Proponents
of primary anastomosis cite the high morbidity associated
with enterostomies in infants and no need for a second oper-
ation.336 To safely perform this technique, the following
criteria must be met: (1) a sharply localized, usually proximal
segment of disease; (2) undamaged appearance of the remain-
ing intestine; and (3) stable overall patient physiology without
evidence of rapidly progressive sepsis or coagulopathy.
Multisegmental Disease (>50% Viable)
If the patient has multiple areas of necrosis separated by
viable bowel, several options are available. Historically, the
surgeon excises each diseased segment individually and cre-
ates multiple stomas rather than performing a massive resec-
tion. Conversely, a single high stoma (proximal jejunum)
may be created and the distal bowel “spliced” together,
thereby avoiding multiple stomas. A proximal jejunostomy
can cause significant fluid and electrolyte loss and peristo-
mal skin complications, although aggressive skin care with
measurement and replacement of stoma losses can avoid
these potential complications. Anastomotic strictures are
not uncommon and are addressed at the time of jejunostomy
closure. Resection plus anastomosis has also gained in-
creased acceptance as a valid treatment option for severe
NEC and for multifocal disease.303 In a study involving 46
infants with multifocal NEC, Fasoli302 reported a higher sur-
vival rate after resection and primary anastomosis (85%) ver-
sus enterostomy (50%).
Moore337 described a controversial approach termed the
“patch, drain, and wait” procedure in 1989. The principles
of this potentially bowel length–preserving method are trans-
verse single-layer suture approximation of perforations
(patch), insertion of two Penrose drains that exit in the lower
quadrants (drain), and a commitment to long-term parenteral
nutrition (wait). The Penrose drains capture fecal fistulas and
function as de facto enterostomies as the peritoneal cavity is
rapidly obliterated by adhesions.338 This procedure does
not address the issue of ongoing sepsis because necrotic bowel
is not resected, the general peritoneal cavity is difficult to
drain, and the thin-walled perforated bowel often cannot
handle suture.
Vaughan339 described a promising novel technique aimed
at avoiding multiple enterostomies, circumventing the com-
plications of a high jejunostomy, and preserving bowel length.
The authors performed the “clip-and-drop-back” technique in
three patients with NEC. In this procedure the obviously
necrotic bowel is removed, and the cut ends are closed with
titanium clips or staples. Reexploration is performed 48 to
72 hours later, the clips are removed, and all segments are rea-
nastomosed without any stomas. In one of the three patients,
reresection was required during the second-look operation,
the bowel ends were clipped again, and a successful primary
anastomosis was performed during a third operation. Follow-
up for this small series was 6 months to 7 years with no
anastomotic complications or delayed reoperations.
Pan Involvement (NEC Totalis, <25% Viable)
Pan involvement develops in 19% of patients310; it poses an
enormous treatment problem and remains a highly controver-
sial management issue. The overriding consideration is to
 CHAPTER 94 NECROTIZING ENTEROCOLITIS 1203

spare as much intestine as possible. Treatment options include and advancements in pediatric anesthesia. The increased sur-
resection of all necrotic bowel with placement of proximal or vival has been most noticeable in infants who weigh less than
multiple stomas or proximal diversion without bowel resec- 1000 g and are less than 28 weeks’ gestational age. In a recent
tion, with plans for a second-look procedure. The decision review of 754 premature infants born between 22 and 25
to forego any treatment is supported by studies that demon- weeks’ gestation, the overall survival rate was 63%, with a
strate a 42% to 100% mortality rate in patients with pan range of 14% at 22 weeks’ gestation to 76% at 25 weeks’ ges-
involvement, with almost all survivors left with short-bowel tation.19 Mortality was still significantly higher in VLBW in-
syndrome. The mortality rate is nearly 100% for infants fants than in larger patients. This is highlighted by studies
who weigh less than 1000 g. that examined the outcome of VLBW infants in comparison
Diverting the intestinal stream by high proximal jejunost- with “standard” premature infants (>1000 g) with pan in-
omy (without bowel resection) may facilitate healing of in- volvement NEC. Snyder345 found that infants weighing less
jured bowel through distal intestinal decompression, a than 1000 g are more likely to require laparotomy (51% vs.
reduction in its metabolic demands, and a decrease in the 34%) and to eventually have pan involvement (10% vs. 4%)
number of bacteria and possibly their byproducts. This tech- than infants greater than 1000 g. Pan involvement was associ-
nique was initially reported by Martin and Neblett340 and ated with 100% mortality in both groups. In a retrospective
involves performing a high jejunostomy without resection, study of 70 infants weighing less than 1000 g with perforated
with plans for a second-look operation after 6 to 8 weeks. NEC, Erhlich331 demonstrated that infant survival was inde-
In a series of 10 patients with pan involvement, Sugarman pendent of the type of surgical treatment (PD vs. laparotomy),
and Kiely341 reported 8 infants surviving to undergo a second but instead was inversely related to the number of comorbid
procedure. Resection of necrotic segments plus anastomosis conditions associated with the patient.
was performed successfully, but the long-term survival rate Taken together, it seems unlikely that the significant dif-
was only 50%. ferences in mortality rates observed in various series are attri-
butable to differences in the effectiveness of the treatment
programs used. In different groups of patients, the disease var-
Stoma Closure and Complications
ies from predominantly localized disease to extensive necrosis.
There is neither an ideal weight and age nor a universally The patient population differs between the various series. The
agreed-upon time at which intestinal continuity should be mortality rate can vary considerably, depending on birth
restored. The principal determinants are time since surgery, weight, coexisting disease, virulence of the disease process,
weight gain, and stoma output and need or effects of TPN and whether the patient is inborn locally versus transferred
on overall metabolic function (i.e., liver). In general, the enter- from another facility that has initiated therapy. The precarious
ostomy may be safely closed anytime after 4 weeks since the state of patients at risk for NEC is emphasized by one series
last operation. Attempted closure at less than 4 weeks postop- that compared patients who had NEC with matched controls,
eratively may be met with a peritoneal cavity that is obliterated in which the mortality rate in controls was 33%.346
by vascular adhesions and resolving inflammation. Before
enterostomy closure, patency of the distal end of the bowel
(i.e., colon) should be confirmed by either a retrograde or
antegrade contrast study to rule out a stricture or strictures Complications
in the distal defunctionalized bowel. A study by Muse- ------------------------------------------------------------------------------------------------------------------------------------------------

meche342 examined the complication rate after stomas were GASTROINTESTINAL

closed less than 3 months after surgery, 3 to 5 months after
surgery, and more than 5 months after surgery; no differences Intestinal Strictures
were found. They also noted no difference in complications The first clinical and radiologic description of intestinal
between patients who underwent closure at a body weight less stricture after recovery from acute NEC was reported in
than 2.5 kg, 2.5 to 5 kg, or greater than 5 kg. 1968 by Rabinowitz.347 The reported overall incidence varies
Although enterostomy in neonates may be lifesaving, it is from 9% to 36%,348,349 and stricture formation is more fre-
also a major cause of morbidity. In recent studies, entero- quent after nonoperative treatment. The incidence of stric-
stomies in newborns had an associated complication rate tures after NEC is increasing as the mortality rate from the
ranging from 34% to 68%.343,344 Complications included disease decreases. Strictures result from fibrotic healing of
wound infections, wound dehiscence, stoma stenosis requir- an area of severe ischemic injury. Regardless of whether the
ing revision, incisional hernia, parastomal hernia, prolapse or stricture follows operative or nonoperative therapy, the most
intussusception, and small bowel obstruction. common site of involvement has been the colon (80%). The
next most common site is the terminal ileum (15%). Sixty per-
cent of colonic strictures involve the left colon, and the most
Survival common colonic site is the splenic flexure (21%). Most pa-
------------------------------------------------------------------------------------------------------------------------------------------------
tients have single strictures, but multiple strictures can occur
Over the past decades, the survival of infants with NEC has (15%).350 An intestinal stricture should be suspected after
progressively improved. This improvement has been attrib- nonoperative management of NEC in an infant with failure
uted to earlier diagnosis and more effective supportive treat- to thrive, rectal bleeding, or bowel obstruction. These signs
ment for premature infants. Effective supportive treatment and symptoms occur in 50% of patients with strictures and
includes improved ventilatory strategies, surfactant therapy, should be evaluated with a contrast enema. If the study
total parenteral nutrition, improved understanding of the demonstrates a stricture in a symptomatic patient, elective
pathophysiology and management of critically ill newborns, resection with anastomosis is usually indicated.
1204 PART VII ABDOMEN
Intestinal Malabsorption and Short-Bowel
Syndrome
Malabsorption may result from a variety of factors including
decreased bowel length, decreased mucosal absorptive area,
enzyme depletion, gut hypermotility, hypersecretion of gastric
acid, bacterial overgrowth, decreased intestinal transit time,
vitamin B12 deficiency, and bile salt deficiency. Short-bowel
syndrome (see Chapter 86) is the most serious long-term GI
complication associated with surgically treated NEC. It occurs
in up to 23% of NEC survivors who undergo surgical
resection.313
Cholestatic Liver Disease
Cholestatic liver disease results from a number of factors but
primarily from prolonged administration of total parenteral
nutrition. It is characterized by direct hyperbilirubinemia, he-
patomegaly, and elevated aminotransferase levels. Although
the condition is multifactorial, the most important contribut-
ing factor is probably prolonged fasting. It has been shown
that the most effective treatment is establishment of early,
small-volume enteral feeding that aids in bowel adaptation
by conferring a trophic effect on the intestinal mucosa and
by stimulating bile flow.
Recurrent Necrotizing Enterocolitis
NEC can recur after operative and nonoperative management.
The incidence of recurrence has been reported to be 4% to
6%.309,351 No consistent association has been noted between
recurrent NEC and the type or timing of enteral feeding, the
anatomic site, or the method of initial management. Various
case reports suggest supraventricular tachycardia, percutane-
ous transluminal angioplasty, and allergic enterocolitis to be
associated with recurrent necrotizing enterocolitis.352–354
Interestingly, Pickard355 proposes that infants with congenital
heart diseases and NEC have a decreased risk of having recur-
rent NEC (OR for CHD 0.58 [95% CI, 0.18 to 1.89], OR for
PDA 0.49 [95% CI, 0.10 to 2.28], OR for all other congenital
cardiac diseases 0.72 [95% CI, 0.15 to 3.34]) through a retro-
spective study of 202 infants with NEC. More than 70% of
patients were successfully treated nonoperatively for recur-
rence by Stringer and colleagues.351
Anastomotic Ulceration
A late complication that may occur many years after resection
for NEC is the development of anastomotic ulceration. Sond-
heimer356 reported six children who underwent ileocolonic
resection and anastomosis in the neonatal period in whom
lower GI bleeding developed at 5 to 13 years of age. Anasto-
motic ulceration was diagnosed by colonoscopy, and treat-
ment entailed ulcer resection in five of six patients.
Recurrence of marginal ulcers developed in four of five pa-
tients who underwent resection. Histologic examination
revealed shallow ulcers penetrating only to the muscularis.
The cause of the ulcers is unknown.
NEURODEVELOPMENTAL COMPLICATIONS
The length of hospitalization of infants has been strongly
associated with developmental progress at 1 to 2 years of
age.349 This probably reflects the adverse effects of medical
and social factors on the developing brain. It is recommended
that developmental screening be performed every 4 months
during the first year and every 6 months during the second
year of life as long-term follow-up data suggest that normal
premature infants and survivors of severe NEC remain at high
risk for neurologic developmental morbidity.
Approximately 50% of infants surviving NEC are neurode-
velopmentally normal.309,346 Historically, it was believed that
any adverse neurodevelopmental outcome in a patient treated
for NEC was due to underlying prematurity and comorbid
conditions rather than NEC itself, but recent evaluations of
surviving infants contest this assumption. Vohr357 studied
the neurodevelopmental, neurosensory, and functional out-
comes of 1151 ELBW (401 to 1000 g) survivors at 18 to 22
months’ corrected age and reported significant deficits in neu-
rologic development (25%), a Bayley II Mental Developmental
Index less than 70 (37%), a Psychomotor Development Index
less than 70 (29%), vision impairment (9%), and hearing im-
pairment (11%). NEC was specifically associated as a risk fac-
tor for both an abnormal neurologic examination and a low
Bayley Psychomotor Development Index. In a study assessing
the effect of NEC on neurodevelopment, Sonntag358 com-
pared VLBW infants with NEC with matched infants without
NEC at 12 and 20 months’ corrected age. Despite normal so-
matic growth in infants with NEC not complicated by short
bowel syndrome, the authors demonstrated significant neuro-
developmental delay at both 12 and 20 months of age. Fifty-
five percent of infants with NEC were noted to be severely im-
paired versus only 22.5% of infants without NEC. Further-
more, Hintz359,360 demonstrated through the National
Institute of Child Health and Human Development Neonatal
Research Network Registry that among the ELBW infants
(weight < 10th percentile for gestational age), infants with
surgical intervention but not medical treatment for NEC are
at significant risk for Mental Developmental Index less than
70 (OR: 1.61, 95% CI, 1.05 to 2.50), Psychomotor Develop-
mental Index less than 70 (OR: 1.95, 95% CI, 1.25 to 3.04),
and neurodevelopment impairment (OR: 1.78, 95% CI, 1.17
to 2.73) compared with infants without NEC.
Prevention
------------------------------------------------------------------------------------------------------------------------------------------------
Attempts to reduce the incidence of or prevent NEC must con-
sider the probable pathogenesis of the disease and some of the
putative perinatal risk factors. Investigations into preventive
measures for NEC including limiting the nosocomial spread
of microorganisms, augmenting host defense, decreasing bac-
terial colonization and overgrowth in the GI tract, and provid-
ing factors that enhance intestinal maturation and attenuate
the inflammatory cascade may be useful. Infection control
measures, breast-feeding, cautious feeding of sick premature
babies, immunoglobulin supplementation of feedings, corti-
costeroid therapy, administration of growth factors, and the
use of inflammatory mediator antagonists are some of the
preventive strategies that have been studied.
INFECTION CONTROL MEASURES
Adoption of infection control measures in the nursery may
limit the incidence and restrict the spread of infections,
thereby potentially eliminating the epidemic waves of NEC.

It has been demonstrated that the initiation of infection con-
trol measures stops epidemics of NEC.219,361 During clustered
occurrences, identical microorganisms can be isolated from
both the afflicted neonates and their caretakers.
AUGMENTATION OF HOST DEFENSE
Oral Immunoglobulin Preparations
The protective immunoglobulins, principally IgA, are defi-
cient in the premature gut. In the absence of breast-feeding,
there are only trace amounts of secretory IgA and gut-
associated IgG and IgM. Secretory IgA acts by binding bacteria
and preventing their attachment to the intestinal mucosa.
Eibl13 demonstrated that enteral administration of an IgG-
IgA preparation decreases the incidence of NEC. Their ran-
domized trial involved feeding 179 high-risk infants weighing
800 to 2000 g a human preparation of IgG and IgA with their
formula, whereas controls received formula alone. Neither
group received breast milk. No cases of NEC developed in
the immunoglobulin group, but 6 cases developed in the 91 con-
trols. In a recent study in rabbits, Dickinson362 demonstrated
that IgA supplementation in feedings prevented bacterial trans-
location by enhancing gut mucosal barrier functions. This effect
was not seen with IgG or lactoferrin. In a randomized double-
blind controlled trial, enteral IgG supplementation in infants
failed to reduce the incidence of NEC.363 A Cochrane review364
of three eligible trials (total of 2095 infants) on oral administra-
tion of IgG or IgG/IgA combination for preventing NEC did not
yield any statistically significant reduction in the incidence of
definite NEC (RR 0.84, 95% CI, 0.57 to 1.25), suspected NEC
(RR 0.84, 95% CI, 0.49 to 1.46), need for surgery (RR 0.21,
95% CI, 0.02 to 1.75), or death from NEC (RR 1.10, 95% CI,
0.47 to 2.59). Similarly, a meta-analysis of three trials using
antistaphylococcal immunoglobulins (INH A-21 and Altastaph)
indicated no significant differences in the risk of staphylococ-
cal infection and NEC between either of the antistaphylococcal
immunoglobulins with placebos.365
Maternal Glucocorticoid Administration
Glucocorticoids have been shown to accelerate epithelial cell
maturation and improve gut barrier function including
reduced mucosal uptake of macromolecules, decreased colo-
nization with aerobic bacteria, reduced bacterial translocation,
and increased activity of enzymes such as lactase, maltase,
sucrase, and Na/K-ATPase.366,367 Glucocorticoids have also
been shown to down-regulate the inflammatory response by
stimulating the enzymatic degradation of PAF.165 These
observations have been made both experimentally and clini-
cally.13,14 Bauer368 retrospectively noted a significant reduc-
tion in the incidence of NEC in babies born to mothers
who received antenatal glucocorticoids for fetal pulmonary
maturation as compared with controls (2% vs. 7%). This large,
multicenter, placebo-controlled trial was well controlled for
many potentially confounding variables. Halac14 conducted
a randomized controlled trial of prenatal glucocorticoid
administration to mothers in preterm labor. The control
mothers received placebos, but their infants received an im-
mediate postnatal course of high-dose dexamethasone for
7 days. The rate of NEC within and between groups signi-
ficantly decreased after prenatal and postnatal steroid treat-
ment. Although postnatal therapy did not decrease the
CHAPTER 94 NECROTIZING ENTEROCOLITIS 1205
incidence as effectively as prenatal therapy did, it improved
the clinical outcome of NEC; however, many potentially con-
founding factors were not assessed in this study. Confirmatory
prospective data and assessment of potential postnatal toxicity
are necessary. In a double-blind randomized controlled
trial comparing perinatal morbidities between antenatal
betamethasone and dexamethasone (299 women),369 no sig-
nificant differences were observed to exist in the incidence of
NEC (0 of 181 for betamethasone, 2 of 178 for dexame-
thasone). Furthermore, another multicenter randomized
controlled trial370 of 502 pregnant women demonstrated no
difference in the incidence of NEC between weekly adminis-
tration and a single course of antenatal corticosteroids
(RR 1.06, 95% CI, 0.44 to 2.56, P value ¼ 0.90).
Breast Milk
Breast milk decreases the risk for a number of neonatal infec-
tions including lower respiratory tract illness, otitis media,
bacteremia, meningitis, and NEC.187 Human milk provides
an array of humoral and cellular antiinfectious factors, growth
factors, and probiotics, as well as essential vitamins and
nutrients. Milk factors include IgA, macrophages, lympho-
cytes, components of the complement system, lactoferrin,
lysozyme, transferrin, interferon, EGF, alpha fetoprotein,
erythropoietin, the probiotics Bifidobacterium infantis and
Lactobacillus acidophilus, PAF acetylhydrolase,174 and several
inflammatory mediators.188,371 Strong evidence exists for the
protective role of secretory IgA, the main immune component
of the enteromammary axis. Breast milk also inhibits the growth
of E. coli by providing an acidic environment, promoting com-
petitive growth of Lactobacillus bifidus, and iron binding (an
element essential for the growth of E. coli).372
Because invasion by infectious agents seems to be a prime
factor in the pathogenesis of NEC, breast milk appears to be
ideally suited to protect the infant against the disease. Admin-
istration of breast milk prevents experimental NEC.372,373
Formula-fed babies have four to six times the incidence of
NEC as breast-fed infants do.374 A meta-analysis done by
Quigley of five trials demonstrated a statistically significantly
higher incidence of NEC in the formula-fed group versus the
donor breast milk group (RR 2.5, 95% CI, 1.2 to 5.1; risk dif-
ference 0.03, 95% CI, 0.01 to 0.06; number needed to harm
33, 95% CI, 17 to 100).375 Subsequently, two more clinical
trials were conducted. The analysis of 1272 infants in the
National Institute of Child Health and Human Development
Neonatal Network Glutamine Trial showed a reduction of
the likelihood of NEC or death after 14 days of diagnosis by
a factor of 0.83 for every 10% increase in the proportion of
total intake of human milk. For every 100 mL/kg increase
in human milk intake with the 14 days of diagnosis, there
is a decreased risk of NEC or death (hazard ratio 0.87, 95%
CI, 0.77 to 0.97).376 Similarly, Sullivan377 showed in a trial
of 207 infants (500 to 1250 g) comparing an exclusively
human milk–based diet with a diet of both human milk
and bovine milk–based products that there was a reduction
in NEC of 50% and in surgical NEC of almost 90% in infants
fed an exclusive human milk diet. The numbers to treat with
an exclusively human milk–based diet to prevent 1 case of
NEC and 1 case of surgical NEC were 10 and 8, respectively.
The last two trials imply a dose effect of human milk in the
reduction of both medical and surgical NEC.
1206 PART VII ABDOMEN
Feeding Practices
There is little disagreement that NEC is more common in fed
infants and that bacterial overgrowth is facilitated by the
substrate provided by formula. Although the potential benefit
of carefully regulated feeding practices to prevent NEC is
widely accepted, randomized trials have failed to demonstrate
any difference in the incidence of NEC related to fast versus
slow, early versus delayed, or continuous versus intermittent
bolus feeding.35–39,378–381
METHODS TO DECREASE INTESTINAL
BACTERIAL COLONIZATION
AND OVERGROWTH
Administration of Probiotics
Probiotic bacteria are defined as “live microbial supplements
that colonize the intestine to provide benefit to the
host.”382,383 The use of anaerobic bacterial supplementation
in the treatment or prevention of GI disease has been well de-
scribed. There has been increasing interest in using probiotics
for the prevention of NEC. Probiotic microorganisms com-
monly used are strains of Lactobacillus, Bifidobacterium, Strep-
tococcus salivarius, and Saccharomyces boulardii. Multiple
randomized, controlled trials have attempted to address this
issue, and several meta-analyses have been published. The
most recent meta-analysis by Deshpande384 analyzed 11
trials385–395 (n ¼ 2176) in which enteral probiotic supple-
mentation was started within the first 10 days and continued
for 7 or more days in VLBW neonates (<1500 g). Compared
with the control group of no probiotics, significant reductions
of risk were noted for developing definite NEC (RR 0.35, 95%
CI, 0.23 to 0.55, P < 0.00001) and dying from all causes (RR
0.42, 95% CI, 0.29 to 0.62, P < 0.00001). Further sensitivity
analysis, examination through the funnel plot, and trial series
analysis supported a 30% reduction in the incidence of NEC
(alpha ¼ 0.05; power of 80%). However, risks for sepsis,
death from NEC, and longer time to full feed (120 to
150 mL/kg per day enteral feeds or as per the prestated
definitions by authors) did not differ significantly between
both groups after adjustments were made for heterogeneity
via a random-effects model.
Rachmilewitz396 and Mackey397 considered applying sim-
ilar concepts used in probiotic feedings by treating infants
with isolated microbe-associated molecular patterns (MAMPs)
to induce protective responses that promote intestinal
health. An MAMP is a molecular sequence or structure in
any pathogen-derived molecule that directly interacts with
TLRs on the intestinal epithelium. It has been shown to effec-
tively improve symptoms of colitis in mice.396 The idea of ad-
ministering inactivated probiotics (heat-killed commensals) or
bioavailable TLR ligands that can potentially induce beneficial
TLR-mediated protective effects without the risk of infection
from administering probiotics is being actively investigated.
Administration of Prebiotics
Besides probiotics, other researchers have suggested adminis-
tering prebiotics, which are nondigestible dietary supple-
ments such as long chain carbohydrates or mucins that
promote proliferation of beneficial commensal bacteria.
A meta-analysis of four trials (n ¼ 126) by Srinivasjois398
examined the efficacy and safety of prebiotic oligosaccharide
supplementation of formula in reducing the incidence of
NEC and sepsis. The duration of the supplementation ranged
from 14 to 33 days. The analysis indicated no NEC in one trial,
but the rest did not report specifically on NEC or sepsis. Two
of the four trials demonstrated a significant increase in bifi-
dobacterial counts in the prebiotic supplemented group.
The major advantage of prebiotic supplements is the lack of
live microorganisms, which reduces the risk of infection that
may exist for the use of probiotics.
Administration of Postbiotics
Similar to prebiotics, postbiotics, which are bacterial
metabolites, may also be a potential treatment for NEC by
generating some beneficial effects on the intestinal flora.
Butyric acid, a short-chain fatty acid produced by commen-
sal bacteria in the colon through anaerobic catabolism of
complex carbohydrates, is a major energy source for colonic
enterocytes and has a widely recognized but poorly under-
stood role in intestinal growth and differentiation,399,400
inflammatory suppression,401 and apoptosis.402 It has been
administered with limited success in human inflammatory
bowel disease.403
Enteral Antibiotics
Nonabsorbable broad-spectrum antibiotics that inhibit bacte-
rial growth have been administered in an effort to prevent
NEC. The use of an enteral aminoglycoside (e.g., kanamycin
or gentamicin) has been proposed as a means to decrease the
incidence of perforation during nonoperative treatment. How-
ever, randomized, controlled studies found no difference in
the clinical course, complications, or mortality rate between
infants who received the antibiotic and those who did
not.404,405 In addition, oral aminoglycosides may be absorbed
across the damaged gut, which can lead to increased serum
levels and thereby potentially contribute to drug toxicity.
Resistant strains of bacteria emerged after treatment with en-
teral kanamycin,404 and there is the omnipresent risk for
promotion of the growth of fungal species. In a recent ran-
domized placebo-controlled study of oral vancomycin in pre-
venting NEC in preterm infants, Siu16 reported a 50%
reduction in the incidence of NEC in the vancomycin group
in comparison with controls. A meta-analysis of five trials
(n ¼ 456) by Bury406 suggested that prophylactic enteral
antibiotics resulted in a statistically significant reduction in
NEC (RR 0.47 [0.28, 0.78]; RD
0.10 [
0.16,
0.04];
NNT 10 [6, 25]) and in NEC-related deaths (RR 0.32 [0.10,
0.96]; RD
0.07 [
0.13, 0.01]; NNT 14 [8, 100]). However,
the summary analysis of three trials gave an increase in the
incidence of colonization with resistant bacteria that was just
significant (RR 1.73 [1.00, 2.97]; RD 0.07 [0.00, 0.13]). Fur-
thermore, a subsequent retrospective cohort analysis of 5693
neonates (<1000 g) from the NICHD database by Cotton407
demonstrated that for every antibiotic treatment day, there
was at least a 4% increase in the odds of an infant having
NEC or dying, especially when antibiotics were initiated in
the first 3 postnatal days for 5 or more days of treatment.
The available studies do not support the routine adminis-
tration of enteral antibiotics to all high-risk premature infants,
many of whom have poor intestinal motility. Effectiveness has
not been proved, and resistant organisms may develop.

Administering specific antibiotics to infants may be indicated
in nurseries in which an outbreak of NEC is associated with a
specific organism.
METHODS TO DETER THE INFLAMMATORY
CASCADE
Inflammatory Mediator Antagonists
The effects of PAF are mediated by receptors, and many com-
pounds that function as receptor antagonists or enzymes that
degrade these proteins have been described. Animal experi-
ments using PAF antagonists or PAF-degrading enzyme (PAF
acetylhydrolase) have demonstrated the capacity to prevent
bowel injury produced by the administration of endotoxin or
hypoxia in rats.47,115 PAF acetylhydrolase is also known to
be present in breast milk. Despite promising results in animal
models, no human trials using PAF antagonists or degrading en-
zymes in the treatment of NEC have been reported.
Arginine
While NO has been linked to the pathogenesis of NEC, it may
be beneficial to neonate intestine by regulating mucosal blood
flow, inflammatory signaling, barrier function, and wound
healing.408 Arginine, a substrate for the production of NO,
has been shown to be in low level in neonates with NEC as
CHAPTER 94 NECROTIZING ENTEROCOLITIS 1207
well.153,409 Thus researchers tested the effect of arginine
supplementation on the incidence of NEC. In the 2007
Cochrane Review155 one trial154 on arginine supplementation
was reported to reduce the incidence of all stages of NEC (RR
0.24, 95% CI, 0.10 to 0.61; RD -0.21, 95% CI, -0.32 to -0.09),
but further studies are necessary to confirm and elucidate
the mechanism of this benefit.
Epidermal Growth Factor
As discussed in previous sections, EGF is a growth factor that
exerts its effects by binding to the EGF receptor. It has multiple
effects including healing of damaged mucosa by inducing mu-
cosal enzyme and trefoil peptide expression and inhibiting ef-
fects on gastric acid secretion.62,65–67 Clark demonstrated
intestinal protection by accelerating goblet cell maturation
and mucin production and normalizing expression of tight
junction proteins in a neonatal rat model. One prospective
randomized controlled study410 of eight neonates with NEC
showed measurable trophic effects on the gastrointestinal mu-
cosa and no significant difference in clinical safety between a
6-day continuous intravenous infusion of EGF and placebo.
The administration of EGF is still at the testing stage. Further
studies are necessary before any clinical use.
The complete reference list is available online at www.
expertconsult.com.

CHAPTER 95
Crohn’s Disease
Obinna O. Adibe and Keith E. Georgeson
Crohn’s disease (CD) is a debilitating chronic inflammatory
bowel disorder with no known cure. The disease affects the
entire digestive tract. The original article by Crohn, Ginzberg,
and Oppenheimer described regional ileitis as the formation “of
ulcers and hyperplastic inflammation in which the terminal
loops of the small intestine, just proximal to the ileocecal
valve, are implicated.”1,2 CD and ulcerative colitis (UC) com-
prise the vast majority of patients exhibiting a wide spectrum
of clinical pathology under the broad heading of inflammatory
bowel disease (IBD). The estimated annual health care cost of
IBD in the United States is about $1.7 billion.3
Approximately 10% of Americans with IBD are children
and adolescents.3
The etiology of CD is unknown. However, it is believed that
its pathogenesis is multifactorial, encompassing environmen-
tal, genetic, and immunologic causes.4 Diagnosis requires a
thorough history and physical examination, along with diag-
nostic imaging, histologic, and laboratory evaluations. Despite
various pathognomonic features such as transmural inflam-
mation, aphthous ulcers, and granulomas, CD may be indis-
tinguishable from UC. In children the reported percentage of
cases of IBD diagnosed as indeterminate colitis ranges from
4% to 23%.5
The natural history of CD is characterized by quiescence
with symptomatic flare-ups leading to severe physical and
social impairment,6 as well as the long-term risk of colorectal
cancer. Both medical and surgical therapies are used to combat
the wide range of symptoms in children and adolescents with
CD. It is imperative, therefore, that this disease is diagnosed
and treated in a multidisciplinary fashion.
Epidemiology
------------------------------------------------------------------------------------------------------------------------------------------------
The worldwide incidence of pediatric CD is 0.2 to 8.5 per
100,000.7,8 A prospective study of a Wisconsin population
reported an incidence of 4.56 per 100,000.9 It is more com-
mon in countries in the northern hemisphere and in industri-
alized nations.7 The highest rates of CD are found in
Scandinavian countries and Scotland, followed by England
and southern Europe.8,10 There has been a noticeable rise
in the incidence of CD after World War II10; Scotland had a
threefold increase in childhood CD between 1968 and
1983,10 and northern France had an increased incidence of
pediatric CD cases, from 2.1 to 2.6 per 100,000 between
1988 and 1999.11 There is a positive family history of CD
in more than 10% of patients,11 with a high rate of concor-
dance for monozygotic twins.7 Although there is a higher
percentage of boys diagnosed with CD in childhood, the
proportions even out in adulthood. Twenty-five to 30% of
patients with CD are younger than age 20 at the time of
diagnosis.7
Etiology
------------------------------------------------------------------------------------------------------------------------------------------------
The etiology of CD remains unknown; however, there seems
to be a complex interaction of environmental, genetic, and im-
mune factors that lead to its development.7 Some experts have
suggested that a triggering event such as a bacterial infection
or other immunogenic stimulus in a genetically predisposed
individual can result in the onset of CD.12
In one recent study, 22% of patients with CD had an affected
family member,10 suggesting a genetic predisposition. Concor-
dance for CD has been reported to range from 36% to 48% in
monozygotic twins, compared with 4% of dizygotic twins.13
The gene for the nuclear oligomerization domain 2 (NOD2)
and the caspase activation and recruit domain 15 (CARD15)
are located on chromosome 16.14 Mutations in NOD2/CARD15
overexpress NF-kb, a critical mediator of the inflammatory
responses that control the transcription of genes for proinflam-
matory cytokines.6,10 Mutations in this gene have been asso-
ciated with ileocecal and stricturing disease phenotypes,6
earlier disease onset, and early need for operative interven-
tion.3,10,12,15–17 Mutations of this gene have been identified
in 30% to 43% of Caucasian North American CD patients.10
Another gene, IBD5, located within chromosome 5q31, has
been associated with CD-linked perianal disease.6,7,10
A recently discovered gene variant of ATG16L1, an auto-
phagy gene, may be associated with CD.16,18,19 This mutation
results in defective macroautophagy in the Paneth cells of the
intestines (within the crypts of Lieberkühn), resulting in the
ineffective elimination of invasive pathogens.19,20 This muta-
tion also results in higher levels of interleukin 1b, making
intestinal cells more susceptible to an abnormal inflammatory
response.21 This ATG16L1 gene variant opens a possible new
mechanism for treating CD patients, focusing efforts on gene
therapy, cytokine neutralization, and Paneth cell rejuvenation.19
1209
1210 PART VII ABDOMEN

Pathology and Clinical Features pathognomonic for CD but are seen in only half of resected
------------------------------------------------------------------------------------------------------------------------------------------------
specimens (Fig. 95-4).22
Within the pediatric population, disease involving both the Common presenting signs and symptoms of CD include
small bowel and proximal colon (ileocecal) occurs in more crampy abdominal pain, diarrhea, decrease in appetite, weight
than two thirds of patients; disease isolated to the small bowel loss, growth delay, and delayed sexual maturation.12 Abdom-
occurs in about 20% of patients and is restricted to the colon inal pain is usually located in the right lower quadrant, asso-
in 10% of patients.11 Endoscopic evidence of inflammation ciated with tenderness and fullness.12 Failure to thrive can be
in patients with ileocolic disease has been noted on upper present in nearly half of children diagnosed with CD.23
endoscopy in up to 40% of patients.12 In 1998 a CD Working Party was organized to classify the
CD initially manifests itself as small mucosal ulcerations behavior of this disorder into three subtypes: (1) nonstric-
(aphthous ulcers).22 These ulcers subsequently coalesce to turing, nonpenetrating; (2) stricturing; and (3) penetrating.24
form linear ulcerations that surround islands of edematous mu- However, the clinical course of CD is often erratic and dis-
cosa, producing the “cobblestone” appearance.22 Mucosal ul- ease phenotypes often change as the duration of disease
cers sometimes achieve transmural penetration, creating lengthens.6 Location of CD in the gastrointestinal (GI) tract
sinuses, abscesses, and fistulas (Fig. 95-1).22 has also been broadly classified into four areas: (1) terminal
The inflammation progresses through all layers of the ileum; (2) colon; (3) ileocolic; and (4) upper GI.24
bowel wall. “Creeping fat” forms from thickening of the mes- Perianal disease may be the primary presenting sign of
entery, with proliferation of the surrounding fat on the serosal CD.12 Anal fissures, perianal fistulas, or abscesses precede
surface. With acute inflammation, the bowel becomes edem- or present simultaneously with diagnosis of CD in many
atous and hyperemic.22 As the inflammation becomes
chronic, fibrotic scarring leads to bowel thickening and
eventual stricture formation (Fig. 95-2).
Histologic examination reveals neutrophilic and eosino-
philic inflammatory cell infiltration of the edematous, fi-
brotic submucosa (Fig. 95-3). Noncaseating granulomas are

FIGURE 95-3 Neutrophilic inflammation with fibrosis within submu-

cosa. (Courtesy Dr. David Kelly, Department of Pathology, The Children’s
Hospital of Alabama.)

FIGURE 95-1 Ileocecal disease with fistula formation. (Courtesy Dr. David
Kelly, Department of Pathology, The Children’s Hospital of Alabama.)

FIGURE 95-2 Disease with stricture. (Courtesy Dr. David Kelly, Department FIGURE 95-4 Noncaseating granuloma. (Courtesy Dr. David Kelly,
of Pathology, The Children’s Hospital of Alabama.) Department of Pathology, The Children’s Hospital of Alabama.)
 CHAPTER 95 CROHN’S DISEASE 1211

patients.8 The incidence of perianal fistulas in children with

CD is approximately 10%.25 Fistulas are classified as simple
or complex. Treatment is discussed later in the chapter.
Musculoskeletal involvement is the most common extrain-
testinal manifestation seen in patients with CD. The incidence
of peripheral arthropathy (arthritis or arthralgia) in CD is
approximately 20%.26 Erythema nodosum presents as tender,
red nodules, usually on the extensor surfaces of the lower
extremities, and is seen in about 10% of patients.26 Pyoderma
gangrenosum manifests itself as erythematous skin nodules
that develop into ulcers, and it occurs in about 5% of patients
with CD.26 Episcleritis and uveitis also occur in about 5% of
patients with CD.26 Episcleritis is a painless hyperemia of
the sclera and conjunctiva, whereas uveitis is a painful man-
ifestation associated with blurring and photophobia. Nephro-
lithiasis, composed of uric acid and calcium oxalate stones,
occurs in up to 5% of patients.26 Hepatobiliary manifestations
associated with CD include primary sclerosing cholangitis,
gallstones, chronic hepatitis, and steatosis.

Diagnosis
------------------------------------------------------------------------------------------------------------------------------------------------

A thorough history often provides the clinician with important FIGURE 95-5 Jejunal ulceration and bleeding via capsule endoscopy.
clues to the diagnosis of CD in children. Intermittent crampy (Courtesy Dr. Shehzad Saeed, Department of Gastroenterology, The Chil-
abdominal pain, food intolerances, diarrhea, fever, family his- dren’s Hospital of Alabama.)
tory of IBD, and perianal disease may be reported.27 Physical
examination should include an abdominal evaluation with and stomach. Endoscopy may reveal skip lesions, rectal spar-
special attention paid to the right lower quadrant for masses. ing, aphthous ulcers, and terminal ileal inflammation. Capsule
A rectal examination may reveal obvious perirectal disease. endoscopy has gained popularity for identification of proxi-
However, a thorough examination is sometimes difficult in mal small bowel disease (Fig. 95-5).32 An upper GI contrast
an anxious pediatric patient. In these cases, an examination study with small bowel follow-through is recommended be-
under anesthesia (EUA) is the gold standard,28 providing fore capsule endoscopy to rule out strictures that may lead
a thorough assessment of the perirectal area. The accuracy to capsule retention. In a retrospective study, capsule retention
of an EUA may be enhanced by the use of an anorectal occurred in 3 of 207 pediatric patients (1.4%) without small
endoscopic ultrasound (EUS) to help detect abscesses.8 bowel abnormalities, compared with 37.5% in patients with
Growth evaluations are an important aspect of the physical known strictures noted on prior small bowel series.32
examination7,12 because weight loss is present in most chil-
dren with CD at diagnosis.7,29 A Tanner assessment of puber-
tal status should be performed to identify delayed sexual Medical Therapy
development. The clinician should also look for signs of ------------------------------------------------------------------------------------------------------------------------------------------------

extraintestinal manifestations (skin rash, joint pain, eye pain). The goals of medical treatment are to (1) achieve remission of
A complete blood count (CBC) should be obtained, specifi- active disease, (2) promote growth through adequate nutrition
cally looking for anemia, leukocytosis, and thrombocytosis.12 and suppression of inflammation, and (3) decrease the need
The erythrocyte sedimentation rate (ESR) and C-reactive protein for surgical intervention.6
(CRP) increase with acute disease activity.27 Baseline nutritional Systemic corticosteroid induces clinical remission and is
laboratory tests (albumin, prealbumin, and transferrin) should standard of care for treatment of moderate to severe CD.6,12
be performed.23 Detection of IgA and IgG anti-saccharomyces Corticosteroids decrease the transcription of proinflammatory
cerevisiae antibody (ASCA) is highly specific for CD in chil- cytokines (IL-1b, IL-6, tumor necrosis factor a [TNF-a]).33 Up
dren.6,7,12,30 Patients presenting with diarrhea should have to 90% of pediatric patients with moderate to severe CD show
stools checked for leukocytes, occult blood, and infectious rapid improvement of symptoms when prednisone or a pred-
etiologies (Clostridium difficile, Giardia, Salmonella, Escherichia nisone equivalent (i.e., prednisolone) of 1 to 2 mg/kg/day is
coli O157:H7, Campylobacter, Rotavirus, Shigella).7,12,27 given (maximum dose 40 to 60 mg/day).33–35 Rectal steroid
An upper GI contrast study with small bowel follow-through therapy as enemas and foam-based preparations may be
assists in assessing existence, location, and extent of the dis- administered for left-sided colitis.12 About 50% of patients
ease.12 Computed tomography (CT) and ultrasound (US) can develop corticosteroid dependency, which is characterized
evaluate for terminal ileal inflammation and intra-abdominal by frequent relapses and a requirement for chronic cortico-
abscesses. In a recent study, gadolinium-enhanced magnetic steroid use.6 Long-term corticosteroid use is associated with
resonance imaging (MRI) was shown to be highly specific and osteopenia, weight gain, glucose intolerance, pubertal re-
sensitive for the identification of proximal small bowel disease.31 tardation, striae, acne, and many other well-described side
Ileocolonoscopy and upper endoscopy are necessary for effects.6,33 It is therefore not recommended for maintenance
direct observation and biopsies of the terminal ileum, colon, therapy.12,36 All patients on corticosteroids should be given
1212 PART VII ABDOMEN
calcium and vitamin D supplements.33 Budesonide, an alterna-
tive steroid therapy with a high affinity for intestinal glucocor-
ticoid receptors, has been reported to have an enhanced first-
pass therapeutic effect with a lesser systemic corticosteroid ex-
posure causing fewer side effects.33 As a timed-release oral
medication, budesonide is indicated for ileal and right-sided
colonic disease,12,34 although it is not as effective as prednisone
in achieving clinical remission in children.6
Mesalamine (an aminosalicylate) is the first-line therapy
for mild to moderate active CD.6 It is not effective in mode-
rate to severe disease or in maintaining remission.12 Asacol
tablets provide mesalamine with an enteric coating that de-
lays systemic release until it reaches the terminal ileum and
colon.33 Pentasa is mesalamine with an ethylcellulose carrier
for time-release throughout the small and large intestine.33
The mechanism of action of mesalamine is largely unknown.
Adverse effects include rash, elevated liver enzymes, and renal
impairment.
Azathioprine (AZA) and 6-mercaptopurine (6-MP) are
thiopurine antimetabolite immunomodulators that inhibit the
inflammatory response via immunosuppression. AZA is con-
verted to 6-MP in plasma. The thiopurines are among the most
effective treatments available for long-term maintenance of
CD.34 They are usually given as an “exit strategy” after induction
of remission with corticosteroids to avoid dependency on the
steroids.6 Thiopurines can reduce the rate of corticosteroid
dependence and resistance from 50% to less than 20%.6 It has
also been shown to decrease the rate of postoperative recurrence
of CD following bowel resection.37 Because of its slow onset of
action (3 to 6 months), thiopurines should be administered early
in the treatment of moderate to severe CD. Adverse effects
include leucopenia, pancreatitis, and hepatotoxicity.8
Methotrexate is a second-line immunomodulator therapy
for patients who do not tolerate 6-MP or AZA.34 It is a dihy-
drofolate reductase inhibitor with corticosteroid-sparing
effects.33 Adverse effects include bone marrow, pulmonary,
and liver toxicity.33 As a known teratogen, women of child-
bearing age should not be administered this drug.
Infliximab is a chimeric (human/murine) monoclonal anti-
body against TNF-a.7 In adults with moderate to severe dis-
ease, infliximab produces excellent symptom improvement
and remission rates for refractory CD and chronic perianal or
abdominal fistulas.12,23,38–40 Infliximab is administered via
infusion at 5 mg/kg given every 8 weeks after an initial loading
period. A multicenter, randomized, open-label study of inflix-
imab in pediatric patients with moderate to severely active
CD demonstrated 88% of patients with a clinical response
and 58.9% with clinical remission compared with placebo.41
Concomitant use of an immunomodulator during therapy
may help maintain clinical response to infliximab.42 Infliximab
has also been shown to decrease corticosteroid use43 and im-
prove height velocity.44 The drug may also be used in the initial
treatment of perianal disease in children and adolescents.8
Crandall and colleagues45 have shown that a majority of
pediatric patients respond to infliximab for perianal disease.
Adverse events from infliximab use include infusion reaction
(flushing, rash, headaches); delayed hypersensitivity reaction;
and an increase in the overall rate of infection (tuberculosis,
histoplasmosis, PCP pneumonia, sepsis).6,8 All patients should
undergo testing for latent tuberculosis before infliximab ther-
apy. Continued use of infliximab may lead to development
of antibodies against itself, causing a diminished clinical
response.46,47 Infliximab is contraindicated in patients with
intra-abdominal or perirectal abscesses.48
Adalimumab is a subcutaneously administered, recombinant,
fully human, immunoglobulin G1 monoclonal antibody that
binds with high affinity and specificity to human TNF-a.49 In
a randomized, phase III, double-blind, placebo-controlled trial,
adalimumab was shown to be effective in induction and long-
term maintenance of clinical remission in adults with CD.49
A multicenter, retrospective study was performed to evaluate
the safety and efficacy of adalimumab in children; it demon-
strated a 70% clinical response rate that sustained at 6- and
12-month follow-ups.47 A prospective study performed in a sin-
gle tertiary institution consisting of 23 patients age 9 to 20 with
moderate to severe CD treated with adalimumab showed a 65.2%
clinical remission and a 91% clinical response for 48 weeks of
treatment.46 Adalimumab may be used to treat those who no
longer respond or have experienced adverse effects to inflixi-
mab.28,47 The recommended dosage for pediatric patients who
weigh more than 15 kg is 20 to 40 kg every other week. The
most common adverse event reported is pain at the injection
site.46 As with infliximab, patients may be required to undergo
testing for latent tuberculosis before initiation of therapy.
Metronidazole and ciprofloxacin appear effective in children
with mild to moderate active ileocolonic CD.7,33 These antibi-
otics may also be used as first-line therapy for the induction
of healing of perianal fistulae.28,48 The addition of infliximab
may promote a synergistic effect in healing the fistulae.28 Met-
ronidazole may delay anastomotic recurrence after surgery50
with long-term use, but relapse often occurs soon after dis-
continuation.33 Dosage of metronidazole should not exceed
40 mg/kg/day. Adverse effects of metronidazole include a metal-
lic taste, glossitis, nausea, pancreatitis, neutropenia, and distal
peripheral sensory neuropathy.8,33 Ciprofloxacin should be
administered cautiously to children due to the reported risk
of Achilles’ tendon rupture in animal studies.33
Enteral therapy (total liquid formula diet) has been used for
both nutritional support and as a method of treatment for CD.23
Exclusive enteral nutrition (EEN) can induce remission in up
to 85% of newly diagnosed patients.29 In 1995 two meta-
analyses of randomized controlled trials comparing EEN with
corticosteroids demonstrated that steroids were more effective
than enteral nutrition for treatment and induction of remission
in patients with CD.51,52 However, a more recent meta-analysis
focusing on pediatric studies suggests that enteral nutrition is as
effective as corticosteroids in both the induction of remission
and the treatment of CD.53,54 This study also documented that
pediatric patients with CD given enteral nutrition often
experience improved growth.34,53 Enteral formulas with trans-
forming growth factor-b (TGF-b) and omega-3 fatty acids have
been shown to increase mucosal healing29,55 and may also have
an immunomodulatory effect.54 The improved growth and nu-
trition status that EEN provides, while avoiding the side effects
of steroids, make it the preferred choice for first-line therapy in
some centers.29 However, because palatability necessitates use
of a feeding tube,4 compliance becomes a major issue. Most
patients relapsed 12 months after discontinuation of EEN.54
Several studies suggest efficacy and safety of stem cell trans-
plantation (SCT) in the treatment of CD.56 In a retrospective
review of 11 IBD patients (7 patients with CD, 4 patients with
UC) who underwent allogeneic SCT for hematologic malig-
nancies (age mean 41, range 27 to 55; median follow-up 34
months), 10 had relief of symptoms after transplantation.57
 CHAPTER 95 CROHN’S DISEASE 1213

In another study, four of five patients with CD and leukemia Many obstructions can be treated medically. Operative interven-
who underwent allogeneic marrow transplantation remained tion is indicated when the risks of continued medical therapy
free of CD after a follow-up of 4.5 to 15.3 years.58 In an outweigh its benefits. The type of operation depends on the
experiment observing the actions of hematopoietic stem cells length of the involved bowel and the number of strictures.
(HSCs) and mesenchymal stem cells (MSCs) in colitis-induced The surgical goal is to alleviate the cause of the mechanical
rats, both HSCs and MSCs migrate to and repopulate damaged obstruction, while minimizing the loss of bowel length7,63 to
GI epithelium.56 In a phase I clinical trial using autologous avoid the numerous complications of short-bowel syndrome.64
HSC transplantation in adult patients with CD, 11 of 12 Strictures up to 10 cm in length may be managed by resection or
subjects showed sustained remission after a median follow- a Heineke-Mikulicz strictureplasty (Fig. 95-6) using a longitu-
up of 18.5 months.59 Another phase I trial suggests that autol- dinal incision through the stricture, with transverse closure of
ogous MSCs harvested from adipose tissue can treat fistulizing the defect. Longer strictures, up to 25 to 30 cm, may be treated
CD.60,61 More rigorous clinical trials, especially in the pediat- by resection or a Finney strictureplasty (Fig. 95-7) using a
ric population, will need to be conducted before this form
of therapy becomes mainstream.

Surgical Therapy
------------------------------------------------------------------------------------------------------------------------------------------------

Unlike UC, there is no surgical cure for CD. Operative inter-

ventions should be reserved for (1) failure of medical therapy,
(2) complications of the disease, (3) severe dysphagia, (4) can-
cer, and (5) stagnated growth and development.62 Medical
failure can be secondary to noncompliance or long-term
steroid dependency.4,62 Many patients with CD will require
surgery at some point in their therapy.4,7
Children often present with right lower quadrant pain and/
or fullness, commonly associated with fever and vomiting.
Imaging studies are indicated to look for a phlegmon, abscess,
perforation, fistula, megacolon, or obstruction caused by stric-
tures. Abscesses are best treated by image-guided drainage
and a course of antibiotics, with a delayed resection if neces-
sary. A few patients present with free perforation,62 requiring
an immediate operation. Fistulas entering the bladder, uterus,
or other intestinal segments are treated with limited bowel
resection and repair of the adjacent organ. Severe acute colitis
and toxic megacolon affects about 5% of CD patients.62 A trial
of intravenous steroids may be cautiously initiated, but
surgical resection is sometimes necessary. FIGURE 95-6 Heineke-Mikulicz strictureplasty. (From Cima RR,
Obstruction is the most common indication for surgery in Pemberton JH: Strictureplasty in Crohn’s disease. In Cameron JL [ed]: Current
children with CD, usually secondary to stricture formation. Surgical Therapy, 9th ed. Philadelphia, Mosby, 2008, p 129.)

FIGURE 95-7 Finney strictureplasty. (From Cima RR, Pemberton JH: Strictureplasty in Crohn’s disease. In Cameron JL [ed]: Current Surgical Therapy,
9th ed. Philadelphia, Mosby, 2008, p 130.)
1214 PART VII ABDOMEN
FIGURE 95-8 Side-to-side isoperistaltic strictureplasty. (From Michelassi
F: Side-to-side isoperistaltic strictureplasty for multiple Crohn’s strictures.
Dis Colon Rec 1996;39:345-349.)
longitudinal incision through the strictured area, followed by a
side-to-side anastomosis. Long segment bowel stenosis caused
by CD may also be treated by the Michelassi technique, where
the stenotic intestine is sectioned transversely at the midpoint of
the stenosis, the mesentery is divided, a longitudinal incision
is made on both loops, the proximal loop is advanced over
the distal loop, and a side-to-side anastomosis is performed
(Fig. 95-8).65 Yamamoto and colleagues66 performed a meta-
analysis of studies on strictureplasty outcomes. They reported
a 13% complication rate for patients who underwent jejunoileal
strictureplasties. Complications included anastomotic leak, fis-
tula, abscess, hemorrhage, wound infection, and bowel obstruc-
tion.66 These complications are similar in type and incidence as
seen after resection.66 Risk factors for these complications in-
cluded preoperative hypoalbuminemia, emergency operations,
low hemoglobin level, and preoperative weight loss.66 Steroid
use was not a risk factor for complications.66 There was no dif-
ference in recurrence rate after strictureplasty versus resection.66
Resection is recommended for diseased bowel or multiple
concurrent strictures greater than 30 cm.
Endoscopic balloon dilatation has been gaining popularity
for treatment of ileal and colonic Crohn strictures. Dilations
are carried out under fluoroscopic guidance using contrast-
filled balloons. Intralesional injections of triamcinolone aceto-
nide (Kenalog) have also been performed.67 Tapering doses of
prednisolone may be administered after dilatations.68 After a
mean follow-up of 81 months, Stienecker and colleagues
reported a success rate of 80% and a perforation rate of 3%
with balloon dilatation.68 Others have reported similar results.67
TABLE 95-1
Type of Surgery in Relation to Extent of Disease, Location of Disease, and Long-term Medical Therapy After the First Year of Follow-up
Procedure >2 Segments
Segmental colectomy 3/54 (5.6)
Total abdominal colectomy 32/49 (65.3)
Total proctocolectomy 47/76 (61.8)
P value <0.0001
From Fichera A, McCormack R, Rubin MA, et al: Long-term outcome of surgically treated Crohn’s colitis: A prospective study. Dis Colon Rectum 2005;48:963-969.
Ileocecectomy is the treatment of choice for patients with
isolated ileocecal disease. It often provides a disease-free inter-
val of at least 2 to 3 years.7 Traditionally, this operation has
been undertaken via an open approach. However, laparo-
scopic resection has gained acceptance in the pediatric surgi-
cal arena. In a retrospective review comparing 12 laparoscopic
and 16 open pediatric resections, the time to full feeds and
hospital length of stay were shorter, while complication rates
were similar.69 In another study of 15 children and adoles-
cents with ileocecal CD, the authors concluded that laparo-
scopic resection is safe and effective.70 In an observational
study from a prospectively collected database of 109 patients,
Soop and colleagues71 reported a complication rate of 11%
after laparoscopic ileocecal resection. Prolonged ileus was
the most common complication. No anastomotic leaks were
reported, and the median hospital length of stay was 4 days.71
Data collected from 2000-2004 via the Nationwide Inpatient
Sample comparing laparoscopic and open ileocecal resections
demonstrated an overall complication rate of 8% with the
laparoscopic approach and 16% with the open approach.72
Finally, a meta-analysis of 14 studies comparing laparoscopic
and open ileocecal resection for CD demonstrated a shorter
hospital length of stay, faster recovery of bowel function,
and decreased morbidity with the laparoscopic approach.73
Resection is the definitive treatment for Crohn colitis
refractory to medical therapy. Operative options depend on
disease location; they include segmental resection, total ab-
dominal colectomy with ileorectal anastomosis, and proctoco-
lectomy with end ileostomy. In 2005 a prospective study
involving 179 patients from the University of Chicago was
reported comparing these three options; after 10 years, a seg-
mental resection was associated with a 50% recurrence rate,
while the recurrence rates for a total abdominal colectomy
and total proctocolectomy were 30% and 20%, respectively.74
Table 95-1 demonstrates the type of surgery, location of
disease, and patients continuing medical therapy 1 year after
their respective surgery.
Although we and others74,75 do not recommend restorative
proctocolectomy with ileal pouch-anal anastomosis (IPAA) in
patients with CD, there is some evidence in favor for this op-
tion. In 2008 the Cleveland Clinic reported their long-term
outcomes of 204 patients with CD who were offered a primary
IPAA.76 In their analysis of a prospectively maintained data-
base, selected patients (highly motivated, preoperative diag-
nosis, isolated colon involvement, no ano-perineal fistulous
disease) achieved an 85% pouch retention rate after a median
follow-up of 7.4 years.
Perianal disease is seen in about one third of children
with CD. It is more commonly associated with colonic CD
than in patients with CD of the small bowel.77 It may be
the only presenting symptom in up to 5% of patients.77 As
Distal Involvement Taking Medication
29/54 (53.7) 30/49 (61.2)
43/49 (87.8) 20/35 (57.1)
73/76 (96.1) 9/43 (20.9)
<0.0001 0.0002

in intra-abdominal CD, perianal disease refractory to medical
therapy requires operative intervention. Perianal CD usually
presents as an abscess or a fistula. Superficial abscesses should
be drained surgically. Simple incision and drainage with irriga-
tion and debridement are recommended.8 The cavity may be
packed with gauze, which is then removed after 24 to 48 hours.
Another option is to insert a small drain into the cavity, which is
secured to the skin with a rapidly absorbing suture, allowing the
drain to fall out with the passage of time. For deep abscesses,
incision and drainage are also required, followed by insertion
of a noncutting seton if a fistula is present or a self-retaining
catheter if a fistula is not identified.8 The catheter may stay in
place for weeks. In most cases, incision and drainage should
be attended by a course of antibiotic therapy.
Perianal fistulas may be classified as low/simple (superficial,
low intersphincteric, low transsphincteric) or high/complex
(high intersphincteric, high transsphincteric, suprasphincteric,
extrasphincteric). The principles of fistula management are
listed in Table 95-2. Evaluation of perianal fistulas in patients
with CD requires an EUA to identify strictures, abscesses,
and/or tracts.63 The accuracy of EUA in detecting fistulas is
about 90%, which can approach 100% with the use of an endo-
scopic ultrasound (EUS).77,78 Sigmoidoscopy is also recom-
mended to assess for any rectal disease. Low/simple fistulas
are treated by fistulotomy,8,63,77 with excellent healing rates
but a recurrence rate of 20%.8,77 For low/simple fistulas
associated with rectosigmoid disease and for high/complex fis-
tulas, placement of a noncutting seton is strongly recommen-
ded8,62,63,77,79 because both fistulotomy and fistulectomy in
these circumstances are associated with high rates of inconti-
nence and nonhealing.8,28 The seton can be left in place indef-
initely. Better control of nonhealing high/complex perianal
fistulas may be achieved via EUS-guided placement of the non-
cutting seton, in combination with an infliximab regimen.28
Alternative treatments for high/complex fistulas include fibrin
glue instillation,80 fibrin glue with adipose-derived stem cell
instillation,81 and intralesional infliximab.82 A diverting colos-
tomy is recommended for patients who have failed medical
TABLE 95-2
Principles of Fistula Management
Define the anatomy of the fistula and, when indicated, assess the
distribution and severity of disease
Drain any associated abscesses
Dilate associated strictures
Attempt to eradicate the fistula tract
Attempt to prevent recurrence
Preserve continence and sphincter integrity
From Lichtenstein GR: Treatment of fistulizing Crohn’s disease.
Gastroenterology 2000;119:1132-1147.
CHAPTER 95 CROHN’S DISEASE 1215
management. An ostomy takedown may be considered after
1 year has elapsed with apparent healing of the fistula.
In a retrospective, multivariate analysis of 100 resective
surgeries in 68 children with CD, there was a 17% symptom-
atic recurrence rate after 1 year, a 38% recurrence rate after
3 years, and 60% after 5 years.83 Recurrence after partial colon
resection was associated with a higher risk of recurrence
than after an ileocecal resection (median, risk-free interval
1.16 years for colonic resection vs. 4.36 years for ileocecal
resection).83 After an ileal or ileocecal resection, second oper-
ations have been reported in 5% of patients after the first year,
11% to 32% of patients after 5 years, and 20% to 44% of
patients after 10 years.84,85 Strictureplasty has been shown
to be both safe and effective in the treatment of CD, while
reducing the risk of short-bowel syndrome. The complication
rate and long-term recurrence rate after ileocecal stricture-
plasty are similar to the rates reported after resection.63
After a colectomy for CD, patients with an end ileostomy
have lower recurrence rates than those with an ileorectal anas-
tomosis.85 Smoking in the postoperative period is a major risk
factor for disease recurrence.84,86 Genetic factors (NOD2/
CARD15 carriers) may also increase the risk of recurrence.84
Medications such as 6-MP, AZA, and metronidazole have been
shown to decrease postoperative recurrence slightly.84,85,87
Patients taking these medications before resection should
continue them postoperatively.83,84 Resection of diseased
bowel often results in a rapid, postoperative increase in linear
growth in children with CD.7,29,83
Cancer
------------------------------------------------------------------------------------------------------------------------------------------------
The ongoing inflammatory process involved in CD pre-
disposes patients to a relatively high risk of malignancy.
A population-based study comparing IBD patients to a normal
population showed an increased incidence rate (IR) of colon
cancer (IR ¼ 2.64, 95% confidence interval 1.61 to 4.12) and
small bowel cancer (IR 17.4, 95% confidence interval 4.16 to
72.9) in patients with CD.88 The risk of future colorectal and
small bowel cancers may increase when IBD presents in the
pediatric years.89 One study demonstrates a 22-fold, lifelong
increase in the incidence of colorectal cancer among men
younger than age 32 at the time of diagnosis.90 Screening
guidelines for colorectal cancer in IBD include (1) a screening
colonoscopy 10 years after onset of symptoms; (2) continued
surveillance every 3 years in the second decade, then every
2 years in the third decade, then every year; and (3) patients
with an IPAA should undergo yearly endoscopic surveillance
of the pouch with biopsies.89
The complete reference list is available online at www.
expertconsult.com.

CHAPTER 96
Ulcerative Colitis
Jeremy Adler, Arnold G. Coran, and Daniel
H. Teitelbaum
Early descriptions of ulcerative colitis (UC)-like illnesses date
back to the late eighteenth century.1 However, the first clear
description of UC is believed to have been made by Wilkis
and Moxon in 1859.2
By the end of the nineteenth century, fecal diversion
became standard practice for severe cases, although improve-
ment was rarely more than transient. In the early twentieth
century, tube appendicostomy and frequent colonic lavage
with bicarbonate solution was recommended.3 Total fecal
diversion by ileostomy, as advocated by Browne in 1913,4
became widely practiced despite significant morbidity and
mortality. Until the mid-1940s, ileostomy appliances were
primitive and routinely associated with soiling and skin irri-
tation. Surgery was usually delayed until the patients were
critically ill; subsequent closure of the ileostomy commonly
resulted in reactivation of disease.
It was not until 1944 that Strauss and Strauss introduced
proctocolectomy with ileostomy for treatment of severe
UC.5 A three-stage operation with ileostomy followed
by subtotal colectomy and sigmoid colostomy and then abdo-
minoperineal rectal resection was recommended by Cattell
in 1948.6 Shortly thereafter, a two-stage operation evolved
with initial ileostomy and simultaneous subtotal colectomy
that, by 1951, was associated with a mortality of only
4.4%.7 Ripstein in 19528 followed by Goligher in 19549
recommended ileostomy with one-stage proctocolectomy,
which is now considered the standard elective operation for
patients with UC.
Although total proctocolectomy with ileostomy provided
a permanent cure for UC, delay in patient acceptance of a
permanent ileostomy resulted in delaying surgery until most
patients had advanced disease, were malnourished, and had
complications from the extended medical therapy. Factors
instrumental in improving the quality of life with an ileostomy
include the eversion technique of ileostomy construction
recommended by Brooke in 195210 and the development of
successful adhesive ileostomy appliances. Subsequent im-
provements in peristomal skin care by Turnbull11 and others,
together with the development of nurses with special expertise
in stomal care, has made it possible for most patients to leave
the appliance attached to the skin for 4 to 6 days. For the goal
of developing a continent ileostomy that required no stoma
appliance and could be irrigated at a convenient place and
time, Kock12 developed the intra-abdominal reservoir with
a nipple valve in the ileostomy stoma in 1969.
The search for a sphincter-saving operation for patients
with UC began more than 50 years ago when Devine of
Melbourne13 recommended a staged ileosigmoidostomy. Aylett
of London14 became the leading proponent of colectomy with
ileoproctostomy, and in 1966 he reported excellent results with
300 patients and a mortality rate of only 5.7%. Parc and col-
leagues15 reviewed 17 studies comprising 1206 patients with
colectomy and ileorectal anastomosis for UC and found that
patients averaged 4.5 bowel movements per day and that
99% were continent. Proctectomy was eventually required in
15%. The risk for subsequent development of cancer in the
retained rectum ranged from 2% to 20%. Ileoproctostomy is
now used only in patients with minimal rectal disease.
In 1947 Ravitch and Sabiston16 modified the previously
unsuccessful procedure of ileoanal anastomosis and added
the technique of rectal mucosectomy. Most patients who
had this procedure had defecation frequency and urgency
of unacceptable degree such that the operation was rarely used
during the ensuing 3 decades.17 In 1964 Soave18 described a
colonic pull-through procedure after distal rectal mucosec-
tomy for Hirschsprung disease and reported good clinical
results. Application of distal rectal mucosal stripping was
combined with a total colectomy and a straight endorectal
ileoanal anastomosis by Safaie-Shirafi and Soper19 in 1973
in four young patients with familial polyposis coli. In 1977
Martin, Le Coultre, and Schubert20 reported satisfactory
results in young patients with UC.
The ileoanal pull-through procedure has undergone many
modifications during subsequent years, the most important of
which was the use of a pouch constructed from the distal
ileum and placed immediately proximal to the anastomosis
to reduce stool frequency and urgency. The development of
the ileal reservoir was based on canine studies by Valente
and Bacon in 195521 with a triple-limb ileal pouch, by Karlan,
McPherson, and Watman22 with a double-barrel isoperistaltic
pouch, and by Ferrari and Fonkalsrud23 with an S-shaped
pouch and an ileoanal anastomosis. In 1978 Parks and
Nicholls24 combined an S-shaped ileal reservoir with rectal
mucosectomy and a pouch-anal anastomosis and reported
acceptable results in adult patients. Total colectomy with
the ileoanal pouch procedure (restorative proctocolectomy)
is now widely accepted as the most favorable option for the
surgical treatment of UC in patients of all ages.25–27
1217
1218 PART VII ABDOMEN

Etiology The muscularis mucosa may be involved, but the inflamma-

------------------------------------------------------------------------------------------------------------------------------------------------
tion does not extend to the muscularis propria. The rectum
UC is a chronic immune-mediated inflammatory condition of is involved in more than 95% of cases, and the inflammation
the colon. The precise etiology of UC remains elusive. Histor- extends proximally in a contiguous manner without skipping
ically, theories of etiopathogenesis involved implicating spe- regions.45,46
cific pathogenic organisms as causative agents. Early theories When the entire colon is present (pancolitis), the most
of infectious etiologies were encouraged by the observation severe changes usually occur in the rectum and sigmoid colon.
of the response of UC to sulfonamides and antibiotics.28 No However, cases of more severe involvement of the proximal co-
organism is consistently found in individuals with UC. And lonic segments have been described and may be consistent with
epidemiologic studies do not support an infectious etiology. UC.46,47 Children are more likely to present with pancolitis at
After the discovery of UC’s responsiveness to adrenocortico- diagnosis than are adults, who are more likely to present with
tropic hormone (ACTH), and later to glucocorticoid steroids, more limited left-sided colitis or proctosigmoiditis.38
research focused on an immunologic etiology of UC.28 The cur- Limited inflammation of the ileum can be seen with UC.
rent understanding of pathogenesis involves an exaggerated The ileal inflammation of UC is generally limited to the distal
immune response to commensal luminal flora.29 10 cm of ileum and consists of villous atrophy and superficial
Early observations of familial patterns of disease have sup- inflammation with crypt abscesses and without granulomas.47
ported the expectation that there is a genetic basis for increased Ileal inflammation with UC is present only in pancolitis and is
susceptibility to UC. Among individuals with UC, 13.8% have a not present in limited left-sided colitis.48 More extensive
first-degree relative with inflammatory bowel disease (IBD), pre- ileal inflammation, or ileal involvement without pancolitis
dominantly UC.1 In twin studies, the rate of UC in a twin sibling or ileal stenosis or the presence of granulomas, should raise
is 14% to 19% when one sibling is affected. The rate is decreased the suspicion of Crohn disease.48
to 0% to 7% when nonidentical twins are considered (either The characteristic microscopic appearance of UC consists
dizygotic twins or nontwin siblings).30 of crypt architectural distortion with crypt abscesses that lead
Recent advances in genetic research have identified numer- to mucosal ulceration with undermining of adjacent mucosa.
ous genetic mutations that are associated with IBD. Early studies Mucosal bridging with pseudopolyp formation often develops
identified an increased risk associated with the presence of in a circumferential pattern, in contrast to granulomatous
HLA-B27. More recently, genome-wide association studies have colitis in which ulceration is usually more linear and located
identified mutations in the nucleotide-binding oligomerization on the mesenteric side of the colonic lumen (Fig. 96-1).
domain containing 2 (NOD2) and interleukin-23 receptor As the disease progresses in the acute phase, the colon dis-
(IL23R) genes, which appear to be associated with an increased tends, peristalsis decreases, and the muscularis becomes thin
risk of developing UC, though the association is not strong.31–33 and diffusely hemorrhagic; severe disease may progress to
acute dilatation, bacteremia, peritonitis, and rarely perforation
(toxic megacolon). The incidence of toxic megacolon is
Epidemiology increased with severe UC. The use of systemic corticosteroids
------------------------------------------------------------------------------------------------------------------------------------------------
and antimotility agents (including narcotics) increase the
The incidence of UC in pediatric populations varies across the probability of developing toxic megacolon.49
globe. Rates in industrialized countries are seen in the range of With chronic UC, however, the muscularis becomes
2.2 to 4.5 per 100,000.34,35 Though there is a clear increasing thickened with flattening of haustral folds and reduction
trend in the incidence of Crohn disease over time, the change of meaningful peristaltic contractions.50–52 The mucosa be-
in incidence of pediatric UC is less clear.36,37 Males and comes atrophic and may become dysplastic in long-standing
females have an equal incidence of UC. Children are more disease. The colonic mesentery becomes shortened, and the
likely to present with involvement of the entire colon (panco-
litis) then their adult counterparts.38 This is especially true in
the youngest cohorts of patients.39,40
Environmental factors that modify the risk of UC include
tobacco smoke and appendicitis. Tobacco exposure is asso-
ciated with a lower incidence of UC in adults, and there is a
higher rate of developing UC in former smokers.41 The same
protective association has been observed of secondhand
smoke in children.42 This is opposite the observed association
of smoking and increased risk in Crohn disease.41 Prior stud-
ies suggested a protective effect of appendectomy on the
development of UC.43 It appears, however, that it may be ap-
pendicitis itself, and not appendectomy, that is associated with
a decreased incidence of the development of UC.44

Pathology
------------------------------------------------------------------------------------------------------------------------------------------------

UC is a chronic inflammatory disease of the rectal and colonic FIGURE 96-1 Open segment of the left colon from a 16-year-old with
mucosa. The inflammation is limited to the superficial chronic ulcerative colitis. Note the presence of large pseudopolyps and
tissue layers including the epithelium and lamina propria. the absence of normal mucosal folds.
 CHAPTER 96 ULCERATIVE COLITIS 1219

serosal surface often develops increased superficial vascularity have pancolitis, those whose symptoms began in childhood,
and an overabundance of adipose tissue. In remission, the and those who have frequent exacerbations of disease.63,64 Ev-
mucosa may return to a nearly normal appearance on micro- idence of dysplasia of the colonic mucosa on biopsy indicates
scopic examination. Mucosal biopsy is helpful in confirming a high risk for subsequent carcinoma.65,66
the diagnosis of UC and in assessing disease activity. Extracolonic manifestations of UC include growth retar-
dation, arthralgias, delay of sexual maturation, skin lesions,
anemia, osteoporosis, primary sclerosing cholangitis, neph-
Clinical Manifestations rolithiasis, uveitis, and stomatitis (Fig. 96-2). Growth retarda-
------------------------------------------------------------------------------------------------------------------------------------------------
tion with delay in bone development frequently accompanies
Approximately 25% of patients diagnosed with UC had the ini- chronic UC in adolescence.34 Delayed sexual maturation
tial presentation of their disease during childhood. Of pediatric may be caused in part by abnormally low levels of urinary
patients diagnosed with UC, by one retrospective estimate 38% gonadotropins.67,68 Growth hormone levels are usually in
were diagnosed before the age of 10 years.38 In a later prospec- the normal range for the patient’s age.
tive multicenter study, of patients with childhood onset, 21.6% Arthralgias occur in approximately 15% of children with
have onset before 6 years of age.40 An additional 46.1% have UC and commonly involve the knees, ankles, wrists, fingers,
the onset of disease between 6 and 12 years of age, and and hips.69,70 Joint symptoms, which occasionally precede
32.3% have onset between 13 and 18 years of age.40 the onset of intestinal symptoms, may be confused with juve-
Symptoms typically present insidiously with persistent di- nile idiopathic arthritis.71 Symptoms are more common with
arrhea followed by the appearance of blood, mucus, and pus active UC. But patients may have symptomatic arthralgias or
in the stools. An intermittent, cramping lower abdominal arthritis with quiescent colitis as well.69
pain, tenesmus, and anemia are common. Anorexia, weight The most common skin lesions are erythema nodosum of
loss, and growth retardation can occur in up to 38% of chil- the lower extremities and occasionally the trunk. Pyoderma
dren with UC.34 The etiology of anorexia is multifactorial gangrenosum (i.e., chronic ulceration of the skin) most often
and may be due to increased circulating cytokines or may occurs on the lower limbs and occurs in less than 5% of chil-
be due to pain associated with peristalsis or tenesmus. Growth dren with UC.72 These lesions are often resistant to local or
failure can also be associated with prolonged use of systemic systemic therapy and frequently do not heal until the patient
corticosteroids.53 goes into remission or the disease is surgically corrected. Pyo-
Children often feel tired as a result of anorexia and/or ane- derma gangrenosum may persist after remission of colitis.
mia. Additionally, there is evidence of disturbed sleep patterns Pyoderma may be successfully treated with infliximab.73 Pri-
in children with UC due to nighttime bowel movements and mary sclerosing cholangitis (PSC) is uncommon in children,
other factors such as an increased incidence of restless leg syn- although it may occur in up to 15% of adult patients with
drome.54 Children with UC may also limit their social interac- chronic UC.74 Approximately 50% to 80% of children with
tions due to fecal urgency and fears of fecal incontinence.55–58 PSC will develop UC.75,76 Development of PSC may precede
Without treatment, most children develop remitting colitis the develop of colitic symptoms. PSC may also develop simul-
with periodic relapses.59,60 Over time, many progress to taneously with UC or may develop after colectomy. Anemia is
chronic colitis with shorter and less frequent remissions. common and usually results from overt or occult blood loss in
Approximately 10% of patients develop chronic symptomatic the stools. Patients may also develop anemia of chronic disease
colitis.38 Rarely, patients may experience a single attack of as well.
colitis with subsequent complete remission (11% to 16% in Osteopenia may occur from decreased calcium absorption
historical literature).59,60 Relapses may be precipitated by associated with diminished uptake of fat-soluble vitamins and
intercurrent infections. Emotional stress does not trigger exac- by increased urinary losses of calcium resulting from steroid
erbations of disease, though stress may worsen perceived therapy.77 However, the exact risk of fracture is difficult to
symptoms.61 estimate in childhood where fractures are not uncommon in
Approximately 10% to 15% of children with UC initially active healthy children. It is unclear whether the risk of frac-
present as fulminating disease with profuse bloody diarrhea, ture is increased in association with UC.78,79 Nephrolithiasis
severe abdominal cramps, fever, and occasionally sepsis, occurs in approximately 8% of patients with UC, largely
necessitating prompt treatment.25 Most of these patients im- because of inadequate fluid intake to compensate for fluid loss
prove with medical therapy, although patients occasionally through diarrhea.80 Uveitis (i.e., inflammation of the iris)
(5%) develop toxic megacolon that requires emergency sur- occurs in less than 2% of patients but may be severe and
gery. Historically, the colectomy rates for pediatric UC were can permanently compromise vision.81,82 Uveitis may present
approximately 50% by 5 years after diagnosis.60 More recent with painless red eyes or may cause photophobia with pain.
estimates of colectomy rates are around 25% after 5 years.38 Aphthous stomatitis is common in patients with UC, particu-
These rates do not take into account any potential change larly during disease exacerbations.
in colectomy rate since the advent of tumor necrosis factor
(TNF)-a antagonist therapy. An accurate estimate of rates of
colectomy after therapy with infliximab remains to be Clinical Examination
determined. ------------------------------------------------------------------------------------------------------------------------------------------------

Carcinoma of the colon or rectum has been reported in 3% Children with active UC may present with a range of com-
of patients during the first 10 years of disease and increases by plaints. In mild colitis, patients may present with diarrhea
approximately 10% to 15% in each subsequent decade.62 Ma- with small amounts of blood. Tenesmus may be present and
lignant conditions may develop even in patients with apparent occasionally may be the primary complaint. With more severe
remission of colitis. Cancer is more common in patients who presentations, patients may have grossly bloody diarrhea.
1220 PART VII ABDOMEN

Carcinoma
Erythema
nodosum

Sclerosing
cholangitis
Hemorrhage

Toxic
megacolon

Arthralgias
Pseudopolyps

Mucosal
ulcerations
Pyoderma
gangrenosa

Nephrolithiasis

Hemorrhoids
Uveitis
FIGURE 96-2 Manifestations of ulcerative colitis in the colon and rectum. Extracolonic disorders occur in more than 60% of children with ulcerative colitis.

With severe or fulminant colitis, children may become pro-

FIGURE 96-3 Appearance of the transverse colon from an 18-year-old
boy with severe ulcerative colitis as viewed through a colonoscope. Note
the mucosal ulceration and loss of mucosal folds.
foundly anemic and may present with pallor, weakness, and
weight loss. Constitutional symptoms such as fever, weakness,
and listlessness may be present. Stools may vary from watery
with scant blood in mild colitis to grossly bloody, liquid stools
with mucus and little solid content.
Children with mild UC or who are in remission may man-
ifest few positive findings for disease on physical examination,
although endoscopy may demonstrate friable and edematous
colonic mucosa with a thin, purulent exudate. Children with
chronic UC often show delayed growth in height, delayed sex-
ual maturation, anemia, and pallor. With long-term corticoste-
roid therapy, children may develop Cushingoid features. With
acute disease, fever, dehydration, and systemic toxicity are
common. Pain is often elicited by palpation over the sigmoid
colon. External hemorrhoids frequently develop from fre-
quency of defecation; however, anal fistulas, fissures, and
abscesses are rare and, if present, suggest Crohn disease. Sig-
moidoscopy or colonoscopy often shows an edematous and
hemorrhagic mucosa that contains superficial ulcers and is
covered with a purulent, bloody exudate (Fig. 96-3).
Anemia from blood loss occurs in approximately two
thirds of patients. The erythrocyte sedimentation rate is typi-
cally elevated as is the C-reactive protein. Hypoalbuminemia
frequently develops secondary to protein loss from weeping
colonic mucosa. Hyponatremia and hypokalemia often
occur with protracted diarrhea. Stool cultures are consistently
negative for pathogenic bacteria and parasites. However
Clostridium difficile infection may coincide with the onset of
colitis, may complicate the clinical picture, and may delay
diagnosis of UC.
 CHAPTER 96 ULCERATIVE COLITIS 1221

Barium enema radiographs in a patient with chronic UC TABLE 96-1

may show a shortened, narrow, and rigid colon with loss of Pediatric Ulcerative Colitis Activity Index (PUCAI)
haustral folds and extensive formation of pseudopolyps from the Hospital for Sick Children, Toronto
(Fig. 96-4). In acute colitis, the intestinal contour may have Item Points
an irregular serrated border from mucosal ulcerations. The
Abdominal Pain
edematous mucosa between areas of ulceration appear as No pain 0
pseudopolyps. Because a contrast enema can stimulate acute Pain can be ignored 5
manifestations of colitis and because more meaningful infor- Pain cannot be ignored 10
mation can be obtained from colonoscopy, contrast enemas Rectal Bleeding
are now used much less frequently. Upper gastrointestinal bar- None 0
ium contrast study with small bowel follow-through is more Small amount only (<50% of stools) 10
Small amount with most stools 20
commonly used to image the small bowel. This is frequently Large amount (>50% of the stool content) 30
performed to assess the small bowel to rule out small bowel Stool Consistency of Most Stools
involvement of Crohn disease and in support of the diagnosis Formed 0
of UC. CT enterography (CTE) can provide more detailed Partially formed 5
assessment of the small bowel; however, CTE typically ex- Completely unformed 10
poses patients to larger doses of ionizing radiation than a small Number of Stools per 24 Hours
0-2 0
bowel follow-through (Table 96-1).83 3-5 5
The Pediatric Ulcerative Colitis Activity Index (PUCAI) was 6-8 10
recently developed for assessing pediatric UC disease activity >8 15
(see Table 96-1). This scoring method was prospectively Nocturnal Stools (Any Episode Causing Wakening)
validated and is useful both as a research tool, as well as a No 0
clinically useful means of assessing patients.84,85 The total Yes 10
score ranges from 0 to 85. A score of less than 10 is consistent Activity Level
No limitation of activity 0
with remission of disease, 10 to 34 indicates mild disease, Occasional limitation of activity 5
35 to 64 moderate disease, and 65 to 85 severe disease. Severe restricted activity 10
A clinically significant response is defined as a PUCAI de- Sum of PUCAI (0-85)
crease of greater than or equal to 20.86 A PUCAI score of
greater than 45 on the third day of illness is predictive of a high

likelihood of failure of systemic corticosteroid therapy. And by

day 5 of illness, a PUCAI score greater than 65 to 70 suggests
the need to escalate medical therapy.87

Medical Management
------------------------------------------------------------------------------------------------------------------------------------------------

Therapy for UC is aimed at symptom relief and prevention

FIGURE 96-4 Barium enema radiograph from a 17-year-old girl with
chronic ulcerative colitis. Note the shortening of the colon and loss of haus-
tral markings, which give the colon a characteristic “lead-pipe” appearance.
of exacerbations of disease. Historically, medical treatment
options were limited to sulfasalazine and corticosteroids.
Systemic corticosteroids are effective at inducing remission
in approximately 54% to 77% of patients.88 However, toxic-
ities prevent steroids from being useful for long-term man-
agement. 5-Aminosalicilates (5-ASAs) are as effective as
sulfasalazine for UC.89 5-ASAs are generally effective for
induction of remission in mild UC.90 And 5-ASAs are better
than placebo at maintaining remission for patients with mild
to moderate disease.49 In cases of limited rectosigmoid
disease, rectal preparations of mesalamine may be effective,
whereas the combination of oral and rectal preparations is
often superior to either one alone.91,92
Azathioprine is not effective for use in acute disease. How-
ever, for patients with steroid dependent UC, azathioprine is
more effective than mesalamine at inducing remission and
allowing patients to maintain remission without systemic
corticosteroids.93
Infliximab has been demonstrated to be effective at induc-
ing remission in 27% to 39% of adults with moderate to severe
UC. In the same study by Rutgeerts and colleagues,94 26% to
37% of adult patients with moderate to severe UC maintained
remission for up to 30 weeks. In pediatric studies, infliximab
has been demonstrated to spare corticosteroid use.53 And
1222 PART VII ABDOMEN
infliximab induces both short-term remission in 82% of pa-
tients and long-term remission in up to 67% of pediatric
patients after 9 months of therapy.95 A recent study by Hyams
and colleagues96 demonstrated maintenance of steroid-free re-
mission in 21% of pediatric patients with UC at 2 years.
In the case of fulminant colitis, Järnerot and colleagues97
demonstrated 17 of 24 (71%) remained in remission after
90 days of infliximab therapy compared with 7 of 21 (33%)
in the placebo control group.
Adherence to medical therapy is critically important to
maintaining long-term remission. Adherence to therapy in pa-
tients with UC who were in remission on mesalamine mono-
therapy has been extensively reviewed. Over 24 months, 89%
of patients who took 80% of their medication remained in re-
mission, although only 39% of those patients who took less
than 80% of their medication remained in remission after
24 months.98
For patients who develop acute exacerbations of disease, a
course of corticosteroids (prednisone) for 2 to 3 months may
induce remission.99 Steroid doses are typically tapered, and
alternate-day therapy is initiated as soon as possible. Pro-
longed use of steroids in young patients is often associated
with side effects including growth failure, osteopenia, and
Cushingoid features. Rectal steroids (hydrocortisone) or
mesalamine enemas may be helpful in treating colitis confined
to the rectum and lower left colon with lower systemic absorp-
tion than systemic steroids. Steroids should be accompanied
by gastric acid suppression with H2-blocking drugs (cimeti-
dine, ranitidine) or proton pump inhibitors (lansoprazole,
pantoprazole, omeprazole) to reduce the incidence of peptic
ulcer or gastritis.
Antidiarrheal medications such as diphenoxylate hydro-
chloride with atropine sulfate (Lomotil) or loperamide hydro-
chloride (Imodium) may reduce the number of bowel
movements and decrease rectal spasm but should be used cau-
tiously because the drugs, as well as opiates, may occasionally
induce toxic megacolon.100 Antibiotics are occasionally used
for treatment of mild symptoms, though there is little evidence
for their efficacy.101,102
Though patients frequently identify dietary factors that
they feel influence their disease activity, there is little evidence
that diet affects the course of UC.103,104 However, at least one
study suggests an association between red meat consumption
and exacerbations of UC.105
Psychotherapy may help patients with chronic UC adjust
to the disease, its complications, and its side effects. Even
more important is the ready availability of a sympathetic
and interested physician and an understanding family on
whom the patient can rely.
During acute exacerbations of UC, most patients require
hospitalization with intravenous fluids and increased doses
of steroids. These measures commonly correct metabolic def-
icits and reduce clinical symptoms but often do not alter the
course of the colitis. Progression of the colitis or failure to re-
spond to therapy is an indication for emergency surgery. If pa-
tients do not tolerate enteral nutrition or if malnutrition is
present, parenteral nutrition may be beneficial. However,
there is little evidence to support the routine recommendation
of bowel rest or hyperalimentation for treatment of UC or for
anticipation of surgery.106–108
Complications following surgical management of UC in
children increase with the duration of the disease and the
length and dosage of immune-suppressing medications.26,109
Surgical treatment in children receiving high doses of cortico-
steroids and immunosuppressive drugs is associated with a
relatively high complication rate that is further increased in
children who have received longer courses of therapy.25
Although intravenous cyclosporine therapy may be rapidly
effective for patients with severe corticosteroid-resistant UC,
the need for colectomy in more than 60% of patients within
6 months after discontinuation of cyclosporine therapy sug-
gests that the drug merely delays colectomy.110 Furthermore,
cyclosporine is associated with a high rate of nephrotoxicity
and other systemic toxicities.
Steroids have been demonstrated to have an adverse effect
on healing of colonic anastomosis.111 Furthermore, aggressive
medical therapy for UC has increased the incidence of emer-
gency-staged colectomy with a resulting increase in morbidity,
hospital stay, and cost and a less optimal functional result.109
Restorative proctocolectomy, however, may be safely carried
out in selected patients requiring emergency surgery for severe
acute UC. In many cases, medical management should be ab-
breviated if prompt control of colitis cannot be achieved.112
Surgical Management
------------------------------------------------------------------------------------------------------------------------------------------------
Although medical therapy often induces remission for varying
periods, UC can be cured by removing the diseased colon and
rectum; up to 40% of patients will require surgical interven-
tion. Because the operative resection has provided excellent
long-term results, colectomy should be seriously considered
for any patient with UC before severe disability and major
complications develop. Elective resection is done for patients
with persistent symptoms of UC despite medical therapy,
growth retardation, severe limitation of activities, and an un-
acceptable quality of life. It is often distressing for both the pa-
tient and family when the child misses many days of school
and must limit participation in physical and social activities
because of severe symptoms. Emergency operation is indi-
cated for patients (approximately 20%) with fulminant disease
that is refractory to medical therapy, extensive rectal bleeding,
or toxic megacolon. Children with UC are more likely to have
an acute onset and more severe symptoms from UC than
adults. Growth potential is the main differentiating consider-
ation in adolescents compared with adults. Thus for children,
consideration should be given for colectomy while the epi-
physes are still open to allow for optimal growth and deve-
lopment. A recent study showed that the cumulative rates
of colectomy were 8% at 1 year, 15% at 3 years, and 20% at
5 years. Predictors of increasing the risk of colectomy were
the presence of extraintestinal manifestations (EIMS) at diag-
nosis and in the first 5 years after diagnosis of UC, resulting in
20% of these children undergoing a colectomy.132
Surgical options should be discussed in detail with the pa-
tient and parents. It may be helpful for the patient to speak to
another child who has been operated on, in order to alleviate
fears and concerns about an ileostomy and to support the
decision. A preoperative discussion with an enterostomal
therapist will help prepare the child and parents for an ileost-
omy. Although care of an ileostomy is usually easily mastered by
a child, the presence of a stoma appliance often creates
embarrassment during physical and social activities. Impo-
tence and bladder dysfunction after proctocolectomy in

children are uncommon; however, these major concerns have
caused many children, parents, and physicians to delay surgical
resection until the colitis becomes debilitating or medical ther-
apy results in systemic complications. Another important issue
to discuss with families is what the immediate and long-term
stooling frequency and patterns are postoperatively (see later).
Anemia, hypoalbuminemia, and electrolyte abnormalities
should be corrected preoperatively. Corticosteroid therapy is
maintained to avoid an acute flare-up; however, should the
amount of steroids remain high, it may influence the surgical
approach. Oral intake is restricted to clear liquids for 48 hours
preoperatively. Cathartics should be avoided. However, a gen-
tle administration of hyperosmolar polyethylene glycol
solutions such as Golytely may be used. Cleansing enemas
are avoided because they may stimulate an acute flare-up of
colitis. Oral antibiotics (erythromycin and neomycin) are
given the day before, and intravenous broad-spectrum,
second-generation cephalosporin antibiotics should be given
within 1 hour of skin incision.
Historically, total proctectomy with permanent ileostomy
has been the standard surgical treatment of UC that is refractory
to medical therapy for more than 50 years. This procedure
cures the disease, is associated with a relatively low rate of com-
plications, and has good long-term results. Nevertheless,
patients of all ages often find the need to wear an ileostomy
appliance for life disturbing, and this is particularly trouble-
some for pediatric patients. Furthermore, approximately
30% of patients with ileostomies experience appliance-related
problems including skin discomfort, skin irritation, leakage,
and odor. In the United States, it is estimated that ileostomy
maintenance costs at least $1000 per year. Currently, few pa-
tients select proctocolectomy with permanent ileostomy when
given a choice of available surgical alternatives. The Kock con-
tinent ileal reservoir with a nipple valve obviates the need for
wearing an ileostomy drainage bag and has been reported to
provide a good alternative to permanent ileostomy for several
years.113 Because of the need for multiple drainages each day,
need for frequent reoperation, and pouch complications re-
lated to the nipple valve and stagnant loop syndrome,114 unless
an ileo-anal pull-through is contraindicated, a Kock pouch is
generally now avoided. Subtotal colectomy with ileorectal
anastomosis should not be considered because active colitis
will typically persist in the rectum of most of these patients
and often eventually requires total surgical removal. The
remainder of this section concentrates on the ileo-anal pull-
through—the standard approach for children and adults today.
APPROACHES TO THE ILEO-ANAL
PULL-THROUGH
The key principle for this procedure is the removal of the en-
tire colon. To avoid an extensive dissection in the deep pelvis,
with potential injury to the nervi erigentes, a mucosectomy is
then carried down through the distal rectum to within a cen-
timeter above the dentate line. Dissection in this fashion al-
lows for avoidance of injury to the internal anal sphincter
and preserves continence. Additionally, this approach leaves
the patient the ability to discriminate between gaseous, liquid,
and solid contents.115 Two general approaches may be taken: a
straight pull-through of the ileum or creation of an ileal pouch
with anal anastomosis. Both are described later, along with
their overall long-term results.
CHAPTER 96 ULCERATIVE COLITIS 1223
STRAIGHT PULL-THROUGH
One of the initial operative approaches was the straight endo-
rectal ileal pull-through technique. Although the initial passage
of stool on closure of an ileostomy will be fairly high and asso-
ciated with moderate degrees of urgency, patients’ symptoms
improve over the first 2 years after pull-through. Reported ex-
perience with the straight endorectal pull-through technique
has produced results roughly similar to those seen with the var-
ious pouch procedures.117,118 The straight endorectal ileal pull-
through technique retains peristaltic contractions and generates
spike waves down to the anal anastomosis. From a technical
standpoint, the procedure is easier than the creation of a pouch
as well. One particular advantage of a straight pull-through is
that if a child has a foreshortened ileal mesentery, the surgeon
may well be faced with a situation where a J pouch (which after
folding the bowel back on itself becomes inherently shorter)
will not afford an adequate length to bring the ileum down
to the anus. In this case a straight pull-through may be an ideal
option and should be considered. Another advantage of a
straight pull-through is the markedly reduced incidence of pou-
chitis (see later).
ILEOANAL POUCH PROCEDURE
The ileoanal pouch was developed to create a reservoir just above
the anus. This reservoir allows for the retention of stool and
fewer number of bowel movements per day compared with
the straight pull-through.27 Regardless of the type of reservoir
used, as long as the lower 4 cm of the rectal muscle is not dam-
aged and the internal anal sphincter is not excessively stretched,
the resting and squeezing pressure of the anal sphincter will
approach normal values within 3 months.120 The number of
bowel movements, however, will remain much higher than nor-
mal values (see Outcomes later).
Four pouch configurations have been used clinically: the
lateral and S-, J-, and W-shaped pouches (Fig. 96-5). Both
the S- and W-shaped pouches are hand sutured and require
a longer operating time than the J-shaped and lateral pouches,
which are usually constructed with a stapling instrument. Sta-
sis is more common in these large pouches, and an irrigating
FIGURE 96-5 Configuration of the two most common pouch
procedures—J (left) and S (right). Note the limitations on length, as well
as a short spout at the base of the S pouch.
1224 PART VII ABDOMEN

catheter is occasionally required for adequate emptying. After just above the umbilicus. The abdominal colectomy proceeds
several months, all pouches commonly enlarge; however, if in routine fashion. The colon is then divided at the recto-
the ileal spout of the S-shaped or lateral pouches is created sigmoid junction. The peritoneal reflection of the rectum is
too long (>2 cm) or if it elongates, pouch enlargement and incised, and the endorectal dissection is begun.
stasis may be pronounced. Endorectal dissection is generally begun with a low-energy
The J-shaped pouch configuration, which was first described cautery in a circumferential fashion. Gaining entry into the sub-
by Utsunomiya and colleagues,123 has been used most widely, is mucosal space may be quite difficult with UC, particularly
the simplest to construct, and produces the fewest long-term when there has been an acute flare-up. Should one encounter
complications in children and adults. The lateral ileal reservoir an undissectable plane, one should restart the dissection
has been used extensively in the past for children and achieves several centimeters more distal. As the rectal muscular cuff is
good long-term results, provided that the ileal spout between developed, countertraction on the mucosal/submucosal tube
the lower end of the pouch and the anus is short.25 The lateral in an upward fashion and outward traction on the muscular
pouch has been used infrequently in recent years, but this cuff helps the dissection, which should be taken down to just
approach, like the straight pull-through, may be used for pa- above the dentate line (Fig. 96-6, A). Placement of a suction
tients with a short ileal mesentery in whom it is difficult to obtain catheter into the rectum (from below) may also make more dis-
sufficient length for a J-shaped pouch (i.e., it cannot reach the tal dissection easier. Once the endorectal dissection is com-
anus without tension). One key consideration of pouch config- pleted, one of the surgeons moves to the foot of the table.
uration is to keep the total length of the pouch short, usually Narrow retractors are placed at the anal mucocutaneous
around 8 cm. This allows for efficient evacuation of stool and junction, and a clamp is inserted into the rectum. An assistant
a presumably lower incidence of pouchitis. working in the abdominal field places the end of the mucosal-
submucosal tube into this clamp. The segment is then everted
outside the perineum. The end of the everted tube is placed in a
Ileoanal Pouch Construction clamp and held on traction by an assistant to assist the anasto-
------------------------------------------------------------------------------------------------------------------------------------------------ mosis. The submucosal-mucosal tube is incised on the anterior
OPEN OPERATIVE APPROACH one half, 1 cm proximal to the dentate line. A Kelly clamp is
inserted into this opening, and the ileum is brought down to
For most patients, the ileoanal pouch procedure is performed this point by grasping two previously placed traction sutures
in two stages. During the first operation, the patient is given (Fig. 96-6, B). Great care must be taken not to twist the bowel
general anesthesia and then placed in the lithotomy position. as it is brought through the muscular cuff. A standard anasto-
A nasogastric tube and bladder catheter are placed. Positioning mosis is performed to the ileum, as either a pouch or straight
for this procedure requires great attention with careful padding pull-through. Rectal examination at this point should show a
of the lower extremities so as to avoid neurovascular injury. well-formed anastomosis 1.5 to 2 cm above the anodermal
In general the area of the peroneal nerve is left completely free, junction. Gloves are then changed, and attention is directed
and weight of the lower extremity is placed on the volar aspect to the abdominal field. The pulled-through ileum is attached
of the feet. A long midline incision is made from the pubis to with seromuscular sutures to the muscular cuff to prevent it

A B
FIGURE 96-6 A, Technique for the endorectal dissection. Note that dissection is assisted with appropriate traction on the cuff and the rectal stump
(arrows). B, The ileum is pulled through an incision in the submucosal tube. Note that care is taken to avoid twisting of the bowel (in this case shown
as a straight pull-through).
 CHAPTER 96 ULCERATIVE COLITIS 1225

from prolapsing in the early postoperative period. In general, a end of the pouch sufficiently out of the anal canal to perform
reperitonealization should be done. A temporizing loop ileost- a hand-sewn anastomosis. Strategies of placing the patient in
omy is typically created in a spot previously marked by an en- reverse Trendelenburg and extensive dissection of the mesen-
terostomal nurse. It is important to carefully inspect this site to teric vessels may help; however, in some cases this may not be
ensure there is no kinking or twisting of this stoma. Addition- sufficient. To address this, the mucosal/submucosal everted
ally, only peritoneal sutures are placed because this will greatly tube is stapled outside of the patient approximately 1 cm
ease takedown of this stoma after healing has taken place. The above the dentate line (Fig. 96-7, A). The residual anorectum
loop may be suspended using a red-rubber-type catheter, and is placed back into the anal canal, and the largest possible cir-
this can be removed in approximately 2 weeks. Once the abdo- cular end-to-end anastomosing (EEA) stapler is inserted. The
men and skin are closed, the stoma is opened. assisting surgeon from the abdominal field places the anvil of
the EEA device into an opening in the end of the J pouch and
secures a purse-string of absorbable, monofilament suture
J Pouch Construction around this opening. Care is taken to ensure that this opening
------------------------------------------------------------------------------------------------------------------------------------------------
is away from mesenteric vessels, which may be injured during
The ileal mesentery is mobilized adjacent to the superior mes- the subsequent anastomosis. The J pouch is guided down, and
enteric vessels into the upper abdomen to provide sufficient an anatomosis is created within the anal canal. This allows for
length for the distal ileum to reach the anus without tension. a much more rapid anastomosis and much less traction to be
A good measure is the ability to easily drag the small bowel placed on the anastomosis (Fig. 96-7, B and C). Both tissue
(after pouch creation) over the most distal portion of the pubic donuts are inspected, and in some patients a sigmoidoscope
symphysis. For the J pouch, the ileum is folded back on itself with air insufflation is performed to assess the integrity of
for a length of 8 to 10 cm. The bowel is opened at the apex of the completed anastomosis. EEA-type anastomoses com-
the J pouch, and an automatic stapling device is fired to create monly narrow down and must be digitally dilated two to three
the anastomosis. In general 2.5-mm staples appear to yield a times during the interim 2 months while the child has a
satisfactory anastomosis. Typically, a small section of ileum protective ileostomy. Once the child begins stooling, the anas-
will remain separated at the proximal side of the J. An addi- tomosis generally stays patent.
tional firing of the stapling device will allow the remaining
portion of the pouch to be anastomosed. Leakage and bleed-
LAPAROSCOPIC APPROACH
ing along the staple lines is not uncommon. Eversion of the
stapled line will allow for inspection and control of any bleed- Minimally invasive surgical (MIS) approaches to a proctoco-
ing. Following this, seromuscular sutures at the top of the lectomy and pull-through have been advanced over the past
suture line will take tension off of this critical area. Note that decade. Two general approaches have been taken. The first
the length of the J pouch has become progressively shortened is to perform the entire proctocolectomy, mucosectomy, and
because longer-length pouches tend to lead to increased stool pull-through via an MIS approach, with the colon being
stasis and higher rates of pouchitis.133 brought out either through the rectal canal or through a small
lower incision.135 This method has the advantage of excellent
cosmetic results but may be time consuming. An MIS alterna-
Modification of the Pull-Through tive approach is to perform the dissection of the colon includ-
------------------------------------------------------------------------------------------------------------------------------------------------
ing mobilization and ligation of mesenteric blood vessels via a
Over the past several years our group and others have made laparoscopic approach, followed by a transverse lower inci-
several modifications to this standard open pull-through pro- sion for the mucosectomy and pull-through procedure.136
cedure.134 One of the major restrictions in performing a In our experience, we have noted a significantly lower opera-
J pouch pull-through is the difficulty in bringing down the tive time and less blood loss with this modified approach.

A B C
FIGURE 96-7 Modification of the pull-through procedure with the use of stapling devices. A, Note the stapling of the submucosal tube just above the
dentate line with a linear stapling device. B, Insertion of an end-to-end anastomosing (EEA) stapler transanally and connection with the J pouch via the
abdomen. As with an open procedure, great care is taken to ensure the J pouch is appropriately oriented during this process. C, Appearance of the anas-
tomosis after firing of the EEA stapler.
1226 PART VII ABDOMEN
Two recent comparisons between open versus laparoscopic
techniques demonstrated similar outcomes between these
two groups,137,138 suggesting that laparoscopy is a viable
and safe approach. Conversion rates are fairly small and were
reported at 7% in a recent pediatric series.139
An initial umbilical port, followed by three to four 5-mm
ports, are placed as shown in Figure 96-8. The umbilical port
initially contains the camera with a 30-degree telescope. This
is followed by an epigastric port and then left and right lateral
ports. The colon, from the terminal ileum to the midrectum, is
mobilized and released from the peritoneal attachments and
the splenic and hepatic flexures. Initial mobilization of the lat-
eral attachments is assisted by “air-planing” the patient to the
contralateral side, with traction on the colon using a blunt
bowel grasper through either the epigastric or the contralateral
trocar sites, and using a cautery scissors via the remaining
trocar site. The ureters are identified early during this dissec-
tion. Depending on the site of mobilization, the camera and
operating ports will vary.
The more time-consuming aspect of the dissection is the
mobilization of the omentum off the transverse colon—a
particular challenge in more obese or high-dose steroid-
dependent children. A pair of 5-mm ultrasonic scissors or
the use of an endoscopic stapling device will help at this point
of the dissection. Care is taken to identify and ligate the
middle colic vessels, which will greatly help bring the trans-
verse colon down to the pelvis. Alternatively, the omentum
may be spared by retracting it superiorly and using electro-
cautery dissection between the stomach and colon.
Once the colon is fully mobilized, a low transverse, supra-
pubic incision is made, predominantly on the left side of the
midline (see Fig. 96-8). The operating surgeon pulls the entire
colon out through this incision and sequentially ligates
remaining mesenteric vessels. The ileum is divided, and an
ileal pouch is then created. The endorectal dissection is then
performed as described earlier.
FIGURE 96-8 Suggested port sites for a laparoscopically assisted
colectomy.
Postoperative Care
------------------------------------------------------------------------------------------------------------------------------------------------
Intravenous steroids are tapered after operation; oral prednisone
can usually be discontinued within 3 to 4 weeks. Most children
are discharged from the hospital by the sixth day after surgery.
A water-soluble contrast enema is performed within 2 months
to ensure that the ileal reservoir has no leaks or sinus tracts.
Most children resume full physical activities within 3 weeks.
Approximately 2 to 3 months after the first operation, the
patient is rehospitalized for ileostomy closure. The ileoanal
anastomosis is dilated under anesthesia.
One to two digital dilations are performed in the clinic be-
fore ostomy closure to prevent narrowing of the anastomosis.
Additional Surgical
Considerations
------------------------------------------------------------------------------------------------------------------------------------------------
PROTECTIVE ILEOSTOMY
The early experience with pull-through operations demon-
strated that a protecting ileostomy that completely diverts the fe-
cal stream for approximately 2 to 3 months minimizes the risk of
pouch leak and pelvic infection. Leaks from the ileal pouch when
not used in conjunction with an ileostomy have been reported to
be more than twice as frequent as when an ileostomy is used.121
On the other hand, the one-stage pull-through operation has been
performed successfully for many years on patients with polypo-
sis coli. Patients with chronic UC who receive long-term steroids
or other medications are often malnourished and frequently have
a suppressed immune response. Such children would probably
be at high risk from a primary procedure. However, patients
with UC who may be considered for the one-stage pull-through
procedure would include those in good general health and ade-
quately nourished, those taking only low-dose steroids and min-
imal immunosuppressive drugs, those who have minimal rectal
inflammation, and those who are not obese.122 Considerable
judgment must be given when deciding whether to perform a pro-
tective ileostomy. High-dose steroids, weight loss, urgent surgery,
or recent (within 3 months) anti-TNF therapy should provide
strong consideration for the performance of a protective ostomy.
Stooling Patterns Post Pull-Through
In general stooling patterns become one of the key outcome
measures in children after a pull-through. Stooling always
takes considerable adjustment. First, nocturnal incontinence
is strikingly common in the first few months, and patients
and families need to be educated about this fact. This appears
to be most common in younger, preadolescent patients and
decreases with the age of the patient and time from surgery.
Frequency of bowel movements may range widely from 6 to
15 motions per day immediately after ostomy closure. Stooling
frequencies decrease over the first 2 to 3 years following
ostomy closure. This is another important factor in the edu-
cation of patients. Generally, after the first 3 years, further
reduction in stooling frequency will not occur.
Straight Versus J Pouch
The question as to whether children fare better or worse with a
straight versus a J pouch reconstruction remains an important
question. A meta-analysis comparing the two procedures in

pediatric patients showed a trend toward lower stool fre-
quency in those children who underwent a pouch. Unfortu-
nately, because of the nature of a meta-analysis review, there
were limitations in being able to comparatively analyze pa-
tients between the selected papers in this meta-analysis.140
A more recent multicenter series retrospectively examined
stooling outcomes.141 This study showed that rate of stooling
was significantly higher in those undergoing a straight pull-
through; however, rates in both groups progressively declined
over time. In fact, by 24 months post-pull-through rates were
fairly similar between the groups (nocturnal and during day
time); however, children undergoing a J pouch had a fivefold
higher rate of pouchitis compared with those with a straight
pull-through (Fig. 96-9, A and B).
Medical and Dietary Management of Stooling
Two strategies may be taken to modify the diet for reducing
stool frequency. The first is an avoidance of highly spicy
foods, chocolate, or acidic foods. Additionally, a bland diet
with reduced simple carbohydrates may prove useful, as is a
strategy of separating liquid intake from solids. One obsta-
cle is that many children tend to lose weight after colectomy,
and one needs to balance such dietary restrictions with the
need to encourage good energy delivery to the child.142
Bulking agents (fiber) are occasionally helpful. Although
they will not reduce the frequency of stooling, they will pro-
vide bulk, which may make stooling more manageable. Initi-
ation of loperamide hydrochloride (Imodium) is often helpful
in decreasing the frequency of stooling. In general this should
not begin until approximately 2 weeks after ostomy closure
but may need to be started earlier in some individuals. Starting
doses should be 2 mg three times a day but may increase to
2 to 4 mg four times a day. Patients and their families need
18
Mean stool frequency (24 hours)
16
14
12
10
8
6
1 3 6 12
A Time post–pull-through (months)
Type of anastomosis
Straight pull-through
J pouch
Error bars: +/– 1.00 SE
FIGURE 96-9 A, Mean 24-hour stooling frequencies for children undergoing a straight and J pouch pull-through. Note the decrease in the frequencies
over the first 2 years. B, Mean rates of pouchitis/enteritis for children undergoing a straight and J pouch pull-through. Note the rates are markedly higher in
those undergoing a pouch reconstruction. (Modified from Seetharamaiah R, West B, Ignash S, et al: Outcomes in pediatric patients undergoing straight
versus J-pouch ileoanal anastomosis: A multi-center analysis. J Pediatr Surg 2009;44:1410-1417.)
CHAPTER 96 ULCERATIVE COLITIS 1227
to be educated about the need to maintain good hydration
and electrolyte intake.
Complications and Outcomes
------------------------------------------------------------------------------------------------------------------------------------------------
Although the mortality rate for children is strikingly low,
morbidity is high. Complications following the ileoanal
pouch procedure are reported to occur in 35% to 65% of
patients.119,126–128 The most common complications are
discussed next.
POUCHITIS
Pouchitis is the most frequent complication following a pull-
through procedure, regardless of whether a straight or pouch
reconstruction was performed, and may be observed in 10%
to 50% of patients.141,143 Pouchitis results in cramping lower
abdominal pain and increased frequency of stooling, often
with watery diarrhea. For cases that present in a nonspecific
manner, the use of fecal calprotectin has been shown by sev-
eral investigators to have a fairly high level of sensitivity in
making the diagnosis.144 With more systemic manifestations,
one may observe fever, malaise, and occasional arthralgias.129
Although pouchitis occurs more frequently within the first 2
years post pull-through, it may be persistent. The etiology of
pouchitis has not been fully elucidated. What is particularly
striking is that those children who undergo a pouch recon-
struction for either multiple polyposis or Hirschsprung dis-
ease rarely develop pouchitis. This suggests that the ileum
has an underlying abnormality that predisposes patients to
pouchitis. One interesting observation is that those patients
who received pre-pull-through calcineurin inhibitors had a
low incidence of pouchitis145; however, this has not been
0.6
Probability of pouchitis/enteritis
0.5
0.4
0.3
0.2
0.1
0
24 12 24 36
B Time post–pull-through (months)
Type of anastomosis
Straight pull-through
J pouch
Error bars: +/– 1.00 SE
1228 PART VII ABDOMEN
subsequently substantiated. Others have found an association
of pouchitis with an abnormal intestinal microbiome, which is
distinct from other patients without pouchitis, or those with
familial adenomatous polyps. This suggests that alteration
in the microbiome may be a causative factor and may allow
future work to be directed at eliminating or treating this
disorder.146
Treatment for pouchitis typically begins with a 2- to 4-week
course of oral metronidazole and confirmation that the
anastomosis is widely patent. More than 90% of children will
respond to this approach. Should this fail, or recurrence
occur, an alternative antimicrobial may be tried; in general
ciprofloxacin is a good alternative. Should this fail, one should
try the addition of daily pouch washouts with tap water or
Rowasa (Solvay Pharmaceuticals, Inc., Marietta, Ga.) enemas.
Once in remission, there is some evidence in adults that the
implementation of probiotics may avoid or decrease the inci-
dence of recurrence.147
For more persistent cases, oral mesalamine and hydrocorti-
sone retention enemas may be added. When the disorder is re-
calcitrant, a diverting ileostomy may become necessary,
particularly in children in whom chronic pouchitis may cause
delayed growth and development. Several recent publications
have shown that the use of probiotics can put patients into
remission after antibiotics have resolved the acute pouchitis ep-
isode. Most of the recent data trends toward supporting the use
of these probiotics148 once the acute pouchitis has subsided.
GASTROINTESTINAL OBSTRUCTION
AND FISTULA FORMATION
Adhesive obstruction of the small intestine occurs in 10% to
30% of patients and remains one of the most common surgical
complications.25,30,117,149 Obstructions due to complications
of the diverting ileostomy are also not uncommon and may be
seen in up to 20% of cases. Not uncommonly these may be
due to a tightening around the fascia or from a food bolus
obstructing the lumen. One should always attempt to clear
this obstruction using a large red rubber catheter through
the proximal limb of the ostomy. Stricture formation at the
ileo-anal anastomosis is common, particularly when an EEA
type anastomosis is performed, or when a diverting ileostomy
is placed. Stricture formations have been reported in 10% to
20% of cases.25,141,143,149 Most of these strictures are readily
dilatable, and once the ileostomy is closed, the passage of stool
often keeps the passage open. However, if there is a devascular-
ization of the ileum, dilations may not succeed. Fistula formation
has been reported in most series.149 In a recent multicenter study,
13% of those undergoing a straight and 5% of J pouch pull-
throughs developed fistulas.141 Fistula formation may occur
to the vagina, skin, perineum, and other portions of the small
bowel. In more than half of these patients, a temporary or perma-
nent ileostomy will be necessary.141 Miscellaneous complica-
tions include hemorrhage and hernia formation.
CONTINENCE
One key aspect of continence is differentiation of daytime and
nighttime continence. A large proportion of children will com-
plain of nighttime accidents, with approximately 40% initially
incontinent.124 This is an important factor to educate patients
and parents on ahead of the pull-through procedure.
Nocturnal incontinence is most prevalent over the first 6
months after a pull-through but will often persist in preado-
lescent children because they typically sleep much more
soundly than older patients. Daytime incontinence is far less
common but has been reported in approximately 5% of all pa-
tients. However, depending on the stringency to which conti-
nence is graded, some degree of decreased continence may be
found in up to 50% of children in some series.150 In one large
series of children, the occurrence of incontinence was
reported at 6% and perioperative complications, particularly
leakage at the anastomosis, were associated with the develop-
ment of lower continence scores.30
VENOUS THROMBOSIS
Postoperative mesenteric vein thrombosis with extension into
the portal venous tree has been described in a number of adult
series151 and has also been reported in children.152,153 The
development of this complication may be due to excessive
tension on the pull-through segment, as well as a proinflam-
matory state that predisposes IBD patients to the development
of lower extremity deep vein thrombosis.154 Strong consi-
deration should be given to placing pneumatic compression
stockings on children during the pull-through and prophylac-
tic low-molecular-weight heparin.
Pull-Through (Pouch) Failure
Approximately 15% to 45% of patients will require a repeat
operation. Furthermore, anywhere from 6% to 25% will
develop a pull-through failure, many of which will require
removal of the pouch and construction of a permanent ileost-
omy.25,124,141,155, For approximately one third of patients
who require pouch removal, the diagnosis will be found to
be Crohn disease.157 Other causes of failure include intracta-
ble stricture; leak; frequent stooling, which results in poor
quality of life; and intractable pouchitis. In these cases a child
should undergo an end ileostomy. Although consideration
should be given to a completion proctectomy in cases of
Crohn disease or intractable stricture, for those with frequent
stooling or pouchitis, the pouch should be retained and an
ileostomy performed, with closure 1 to 2 years after the pa-
tient has recovered from these symptoms. Interestingly, in a
large meta-analysis of pediatric patients having a straight
pull-through, there was a significantly higher rate of pull-
through failure by almost twofold140; however, the cause
of these failures was not identified. Another group that suf-
fers from an increased risk of pouch failure is patients with a
diagnosis of indeterminate colitis. This group of patients may
make up between 5% and 25% of children.158 In general
they will present at a much younger age (some under 2 years),
typically have a more severe diarrhea (often with blood), and
have a more extensive pattern of ulcerations, usually with
rectal sparing.159 This group may well benefit from a total
colectomy with ileostomy and retention of the distal rectum
for future pull-through. Up to one third of patients with in-
determinate colitis will go on to be subsequently diagnosed
with either Crohn disease or UC.158 Because of the future
risk of Crohn disease, one may strongly consider waiting
on the definitive pull-through until this can be determined.
One way to help distinguish these patients is with the use of a
gadolinium magnetic resonance imaging study. Another is
with the use of serum antibodies. Those children who are

p-antineutrophil cytoplasmic antibodies (p-ANCA) positive
and anti-Saccharomyces cerevisiae antibodies (ASCA) negative
will reclassify to UC in 63.6%, whereas those who are
pANCA negative and ASCA positive will reclassify to Crohn
disease in 80%.160 In contrast, another large study of
children with pouch failures found a fairly high incidence
of failures (15%) but was unable to attribute this to the
diagnosis of indeterminate colitis.157 In fact, these authors
associated failure with female gender, perianal disease, Crohn
disease, and early and late complications.
Diagnosis of Crohn Disease After Pull-Through
As suggested in the previous paragraph, it is not uncommon to
secondarily have the diagnosis of UC change to Crohn disease.
In fact, between 5% and 10% of all children undergoing a
proctocolectomy and pull-through will eventually be diag-
nosed with Crohn disease.161 Many of these patients will ini-
tially have a diagnosis of indeterminate colitis; however, others
would have little to no indication of Crohn disease at the time
of pull-through. It is incumbent on the surgeon to make sure
that outside pathology studies are read by their pediatric or
gastrointestinal pathologist and a complete evaluation of the
remainder of the gastrointestinal tract has been thoroughly
evaluated before a pull-through procedure. Postoperative sus-
picion should arise if a child presents with any of the follow-
ing: markedly increased stooling, bloody stools, frequent
pouchitis, weight loss, or most importantly fistula formation
(to perineum, bladder, or vagina). Although some patients
with Crohn disease may remain stable with adequate medical
therapy, up to one third of all pull-through failures are due to
this diagnosis.157
Risk of Postoperative Complications
and Preoperative Medical Therapy
Complications are frequent, and a number of investigators
have attributed some of these complications to the use of
these therapies. Some authors have attributed the use of
corticosteroids to increased infectious complications,162
and others have demonstrated that the association of steroid
use with poor preoperative nutrition is related to an increase
in infectious complications.163 Interestingly, others have
failed to associate steroids, or other conventional therapies
(azathioprine or 6-mercaptopurine) with the development
of such complications.164 Although conventional therapy
has typically involved corticosteroids, recent therapies have
included a number of additional immunosuppressive thera-
pies including cyclosporine A and anti-TNF therapies. Tre-
mendous controversy has arisen in the literature regarding
the risk to secondary septic and leak complications postsur-
gery for both patients with Crohn disease and UC. Although
a meta-analysis failed to show an increased risk,165 more re-
cent publications have identified an increased risk. Unfortu-
nately, virtually all studies are retrospective and in part are
lacking precise details regarding the length of time between
last therapy and performance of surgery or are lacking details
as to the disease severity of the IBD process.166 For UC, two
recent studies, one of which did stratify patients on the basis
of their severity of disease, identified the use of infliximab
as being associated with increased risk of postoperative
CHAPTER 96 ULCERATIVE COLITIS 1229
complications.167,168 Further, many who were on such ther-
apy underwent a three-stage operation due to the surgeon’s
preference. Because virtually all levels of infliximab are out
of a patient’s circulation by 12 weeks, caution should be
taken should a surgical procedure be undertaken before this
time period; however, no data exist on whether there are lon-
ger-term immune-suppressive effects. Additionally, little to
no data exist as to how the pediatric patient population might
be affected by the use of these agents.
QUALITY OF LIFE
Most patients have a remarkably poor quality of life (QOL) be-
fore their pull-through, so it is not surprising that most patients
report a strikingly improved QOL after such a procedure.169
Although large series have reported excellent results,126 it is
important to emphasize that QOL is not ideal after these pro-
cedures. Most patients have urgency, leakage, nocturnal incon-
tinence, pouchitis, and occasionally sexual dysfunction. Each
of these adverse events will significantly reduce QOL. Despite
this, in two series on QOL in children, results showed that QOL
was not significantly different from age-matched U.S. children,
with little or no adverse effects from the surgery.170,171 In fact,
the surgical scar was noted to be a negative factor of significance
in one series and chronic pouchitis was a factor associated
with poorer QOL in the other.
LONG-TERM FOLLOW-UP
Performance of a proctocolectomy and pull-through will
require a child to be followed for the rest of his or her life.
Although 99% of the colon and rectum are removed, a flexible
or rigid sigmoidoscopy to monitor the small amount of
remaining tissue above the dentate line must be done every
3 years to ensure that no secondary dysplastic or malignant
changes develop. In addition, this close follow-up will allow
the surgeon or gastroenterologist to screen for secondary
complications, evaluate the patient’s nutrition, and rule out
extraintestinal manifestations.
Conclusions
------------------------------------------------------------------------------------------------------------------------------------------------
UC has a significant morbidity in pediatric patients. The dis-
ease process requires extensive expertise from the medical,
nutritional, nursing, and surgical specialties. With com-
prehensive and aggressive care, excellent results may be
obtained. Many patients in the pediatric age group will re-
quire surgical care. Although most children do well with a
proctocolectomy and pull-through, realistic outcomes need
to be emphasized to the family. More than half of all patients
will encounter early or late complications after such surgery.
With time, most patients will improve and their lives normal-
ize; however, care following surgery requires patients to
maintain continual follow-up with their medical providers
throughout life.
The complete reference list is available online at www.
expertconsult.com.

CHAPTER 97
Primary Peritonitis
Robert Baird and Jean-Martin Laberge
Primary peritonitis (PP), also referred to as idiopathic or spon-
taneous peritonitis, has been defined as an infectious process
involving the peritoneal cavity without an intra-abdominal
source.1–3 Such infection may seed the abdomen through he-
matogenous or lymphatic routes or by direct extension from
the vagina. With the exception of the ends of the fallopian
tubes, the peritoneal cavity is a completely closed space that
can be penetrated by foreign bodies such as ventriculoperito-
neal shunts and peritoneal dialysis catheters. In prepubertal
and adolescent girls, retrograde spread of fluid out through
the fallopian tubes may account for the presence of an ascend-
ing vulvovaginitis or “swimming pool peritonitis.”4–7
By the early 1900s, up to 10% of abdominal operations in
children fell into the category of PP and carried an associated
mortality of up to 50%.2 In 1964 Conn linked the presence of
intra-abdominal ascites with the bacterial translocation of in-
testinal flora resulting in primary peritonitis.8 Today, PP is a
rare cause of peritonitis in children and represents less than
1% of all pediatric laparotomies because the diagnosis is often
made and treatment frequently does not require an opera-
tion.3 Most cases of PP in the pediatric age group are associ-
ated with nephrotic syndrome (NS) or chronic hepatic states
in which ascites and/or cirrhosis is present. This group in-
cludes infants and children with biliary atresia, cystic fibrosis,
hepatic fibrosis, and lupus erythematosus (Table 97-1).9–11
Organisms isolated from the peritoneal cavity vary according
to the associated conditions (Table 97-2). For instance,
gram-positive organisms including Streptococcus pneumoniae
and group A streptococci are most commonly found in
patients with NS.12–14 The same gram-positive groups are cul-
tured in patients with underlying liver disease, but this patient
population also demonstrates gram-negative isolates includ-
ing Escherichia coli, Klebsiella pneumoniae, and Pseudomonas
species.
Clinical History, Laboratory Data,
and Diagnostic Studies
------------------------------------------------------------------------------------------------------------------------------------------------
Children with PP have acute abdominal pain associated with a
febrile illness, nausea, vomiting, diarrhea, or other viral-like
prodromes. The time course may be more protracted than that
for secondary peritonitis, and diffuse rebound tenderness is
often present. The absence of localized pain may result from
irritation of visceral surfaces.15–16 Diagnostic paracentesis
may be helpful in the presence of ascites, as is sampling of
the dialysate in children with renal failure on ambulatory
peritoneal dialysis programs. The diagnosis of PP is made
when the peritoneal fluid leukocyte count is higher than
500/mm3 and granulocytes predominate (lymphocytes are
usually present in higher numbers in normal peritoneal fluid).
Typical peritoneal fluid in PP also has a pH of less than 7.35
and an elevated lactate level.17 If more than one organism is
present on Gram stain, a perforated viscus should be sus-
pected. Sending a minimum of 10 mL of fluid for Gram stain
and culture analysis will increase the number of positive
isolates, yet cultures may be negative in up to 60% of reported
cases.18 The isolation and sensitivity pattern of specific bacte-
ria decreases the spectrum of antibiotics required and poten-
tially eliminates the use of nephrotoxic drugs in patients with
compromised renal function.13 Samples of blood and urine
should be sent for culture, and radiographs of the chest and
abdomen should be obtained. These studies will help elimi-
nate a perforated viscus or a pneumonic process as the cause
of the abdominal pathology. Routine electrolytes and total
protein levels should be investigated because low serum
protein levels and concomitant decreased opsonins may
contribute to the development of PP.19
Ultrasound studies and computed tomography (CT) of the
abdomen have been used to differentiate primary from
secondary peritonitis (caused by such common pediatric
processes as appendicitis) when the clinical picture is confus-
ing. Double-contrasted CT scan findings in PP include
the presence of a patent appendiceal lumen, diffusely distrib-
uted peritoneal fluid, and secondary enhancement of the
bowel wall.9
Primary Peritonitis in Healthy
Children
------------------------------------------------------------------------------------------------------------------------------------------------
Gross reported 58 cases of PP in a group of previously healthy
infants and toddlers (mean age, 2 years) who had a preexisting
upper respiratory infection.20 The complex of initial symp-
toms included vomiting, fever, irritability, abdominal dis-
tention, and tenderness. S. pneumoniae was the predominant
organism isolated. A later series from the same facility
1231
1232 PART VII ABDOMEN
TABLE 97-1
Conditions Associated with Primary Peritonitis in Childhood
Nephrotic syndrome
Hepatic dysfunction
Adrenogenital syndrome
Cystic fibrosis
Chronic renal failure with the need for chronic ambulatory peritoneal
dialysis
Complications after splenectomy
Diseases requiring long-term steroid administration (systemic lupus
erythematosus, dermatomyositis)
Familial Mediterranean fever
TABLE 97-2
Microorganisms Associated with Primary Peritonitis
Streptococcus species (S. pneumoniae; group A streptococci)
Escherichia coli
Gonococcus
Haemophilus influenzae
Klebsiella pneumoniae
Listeria monocytogenes
Parainfluenza
Salmonella typhi
Serratia marcescens
Yersinia enterocolitica
recorded 33 patients with a mean age of 5 years who were
treated for PP over the next 30 years.21,22 The decrease in
number of affected children was considered the result of in-
creased availability and use of oral antibiotics. Although mul-
tiple potential sources of infection have been implicated
including hematogenous (dental procedures, bacteremia),
lymphatic, gut translocation, and ascending gynecologic pro-
cesses, the etiology remains unclear in most cases. In 1985 two
previously healthy children with dehydration, abdominal
distention, and group A streptococcal peritonitis were
reported. At the time of laparotomy no source of infection
was identified, although one of the children had a concomitant
right-sided diaphragmatic hernia.23 The literature has since
identified a group of healthy children with idiopathic peritoni-
tis and simultaneously positive cultures for group A strepto-
cocci from the trachea, pharynx, or tonsils.24–26 Haemophilus
parainfluenzae was recently isolated during an abdominal
exploration when no intestinal pathology was identified.27
In prepubertal girls with gonococcal vaginitis, 6% may
have evidence of peritonitis. Treatment with parenteral ceph-
alosporins is indicated, and the presence of other associated
simultaneous sexually transmitted diseases should be investi-
gated.6 In earlier reported series of PP, most of the patients
were girls and the vagina was often implicated as the source
of the infections. However, vaginitis was uncommon and
the ultimate source of the infection remained obscure.21
The prepubertal cervix lacks the endocervical glands that
may harbor bacteria, so ascending infections in this age group
would more likely be associated with some traumatic force
that pushes the bacteria up through the vagina, as in sexual
abuse cases or by jumping feet first into a swimming pool
or lake water.5 In more recent series the incidence of PP is
equally distributed between boys and girls.15,28,29 In some
cases the same bacteria causing the peritoneal infection have
also been cultured from the respiratory and urinary tracts,
as well as from the oral cavity.1
In all instances of peritonitis without an overt source of in-
fection, failure to improve with intravenous antibiotics is an
indication for laparoscopy or laparotomy. An appendectomy
can be done safely even in the presence of cloudy exudate,
and the bowel surface can be inspected for secondary causes
of the infection. After the surgical intervention, antibiotics
should be continued until the leukocytosis normalizes and
the ileus resolves.2,16
Nephrotic Syndrome
------------------------------------------------------------------------------------------------------------------------------------------------
PP has been reported in 3% to 17% of children with NS.12,30–35
Before the availability of antibiotic therapy, PP was the leading
cause of death in this group of patients. An increased
susceptibility to infection in these patients may be influenced
by impaired cellular immunity and chemotaxis, decreased
opsonization, and reduced levels of circulating immuno-
globulins. Complement proteins I and B are reduced in the
serum but increased in the urine of patients with peritonitis
and NS.36 The episodes of peritonitis are recurrent in 15% to
26% of cases and have been associated with decreased levels
of circulating IgG.31,35
Large reported series in children indicate that gram-
negative organisms are isolated in 6% to 30% of patients with
NS and peritonitis, whereas S. pneumoniae was cultured in 4%
to 38%.12,31,37 Even when the peritoneal fluid showed no bac-
terial isolates, children symptomatically responded to the sys-
temic administration of intravenous penicillin.10 Initial
treatment of PP should include the administration of antibiotics
to cover both gram-positive and gram-negative bacteria. In
cases in which clinical improvement does not occur within
24 hours after the initiation of therapy, further diagnostic
studies such as abdominal CTor diagnostic laparoscopy should
be employed to eliminate other causes of peritonitis. Abdomi-
nal CT scans may not always successfully differentiate primary
from secondary peritonitis. Laparoscopy and abdominal
irrigation have been successfully used with medical manage-
ment when the clinical course shows no improvement.
In almost 80% of children with NS in whom PP developed,
either steroids were being used as treatment of the condition
or a relapse of NS had occurred.12,34,55 Pneumococcal vacci-
nation (Pneumovax) has been recommended in this group of
patients, yet peritonitis has occurred despite protective immu-
nization.37,38 In the context of the developing world, a recent
report advocates immunization only for the small number of
children who have steroid-dependent or steroid-resistant
nephrotic syndrome.39 Resistant organisms can develop in
children treated with oral penicillin prophylaxis.38
Peritonitis with Peritoneal
Dialysis
------------------------------------------------------------------------------------------------------------------------------------------------
In children maintained on chronic peritoneal dialysis (PD),
peritonitis is the primary complication compromising survival.
The problem has prompted formation of the International

Pediatric Peritonitis Registry, an Internet-based prospective
registry of 47 pediatric centers across 14 countries.35 Perito-
nitis (either bacterial or fungal) occurred at a rate of one
episode per 11.1 to 13.2 patient-months in separate multi-
institutional studies.14,40–42 Gram-positive organisms were
isolated in 44% to 49% of the cases (most commonly S. aureus
and S. epidermidis). Gram-negative bacteria were in 21% to
25% of the cases, and fungi were in 1.8%. Of the symptom-
atic patients, 21% to 31% did not have positive cultures
(Table 97-3).35,40 Gonococcal peritonitis has been reported
in a sexually active adolescent undergoing peritoneal dialysis.7
Risk factors for the development of peritonitis in this group
of children include contamination of the connectors during
dialysate fluid exchange, exit site or tunnel infections, local
trauma to the tunnel site, and nasal colonization with staph-
ylococcal organisms.40 Additional risk factors for the develop-
ment of PD-related peritonitis include the duration of PD
(longer than a year), chronologic age younger than 2 years,
and decreased serum IgG levels.43
Clinical findings include abdominal pain, fever, and an el-
evated neutrophil count in the cloudy dialysate. Once sus-
pected, the fluid should be cultured and both intravenous
and intraperitoneal antibiotics should be instituted, with close
monitoring of antibiotic levels. Because in vitro evaluation
revealed 69% sensitivity of gram-positive organisms to a
first-generation cephalosporin and 80% sensitivity of gram-
negative organisms to a third-generation cephalosporin
worldwide, current treatment guidelines for empiric therapy
include either a first-generation cephalosporin and ceftazi-
dime or a glycopeptide (vancomycin or teicoplanin) with cef-
tazidime.44 Aminoglycosides may play a role in empiric
therapy for high-risk patients, and antibiotic therapy should
be tailored once culture and sensitivity results become
available.
Symptoms should begin to improve within 24 to 36 hours
after initiation of therapy.45 Long-term multi-institutional lon-
gitudinal studies suggest that the PD catheter must be re-
moved to eradicate the infection when Pseudomonas,
Candida, or atypical mycobacterial species are in-
volved.14,40,43,46 There is also a risk of peritoneal membrane
failure when these organisms are involved, thus precluding
the continued use of PD. Peritonitis rates are decreased when
a two-cuff system is used and when the exit site is directed
downward.29,40 Nonetheless, relapsing peritonitis remains a
common clinical problem for these patients. Defined as recur-
rence of peritonitis with the same organism within 4 weeks
TABLE 97-3
Organisms Involved in Peritonitis in Children Maintained
on Chronic Ambulatory Peritoneal Dialysis*
Organism % In patients with indwelling VP shunts and recurrent episodes
Staphylococcus aureus 38.9
Staphylococcus epidermidis 13.2
Group A streptococci 9.7
Pseudomonas species
Other bacteria or fungi
9.0
13.9
No organisms found on culture 13.9
*Data from 144 episodes of peritonitis in 66 children at J.W. Riley Hospital for
Children, Indianapolis.
CHAPTER 97 PRIMARY PERITONITIS 1233
after termination of antibiotic treatment, relapsing peritonitis
was found to occur in 11% of PD patients experiencing an
episode of peritonitis; it significantly decreased the rate of full
functional recovery and increased the rate of necessitating
permanent PD discontinuation.47
Hepatic Dysfunction
------------------------------------------------------------------------------------------------------------------------------------------------
In a child with impaired hepatic function, cirrhosis, and as-
cites, the PP that develops is characteristically associated with
both gram-negative and gram-positive organisms.19,48 The
liver with its rich reticuloendothelial tissue normally filters
the bacteria found in the portal circulation. As cirrhosis
develops, portal venous flow is partially shunted away from
the liver, thereby decreasing the clearance of bacteria and
fungi from both the blood and lymphatic systems. This de-
creased clearance likely allows for the persistence of bacteria
in ascitic fluid.11,18,49–51 Children will have fever and ab-
dominal pain in 50% of PP cases. Paracentesis fluid that
has greater than 500 leukocytes/mm3, bacteria on Gram
stain, and a pH of less than 7.35 had 100% sensitivity and
96% specificity for active PP.13 Patients with cystic fibrosis
and hepatic dysfunction as a result of cirrhosis or impaired
synthetic function are also at risk for episodic PP.52 The treat-
ment of choice for PP in this context is a third-generation
cephalosporin in most pediatric patients. Children at risk
should be maintained on trimethoprim-sulfamethoxazole
for prophylaxis.53 Long-term fluoroquinolone use in infants
and young children is discouraged due to safety concerns re-
garding possible quinolone-induced arthralgia or cartilage
toxicity.
Ventriculoperitoneal Shunts
------------------------------------------------------------------------------------------------------------------------------------------------
Spontaneous bacterial peritonitis has been described in chil-
dren with indwelling ventriculoperitoneal (VP) shunts with-
out evidence of cerebrospinal fluid infection.54 Cultures
from these patients have resulted in a variety of gram-positive
organisms, and improvement in peritoneal irritation occurs
with exteriorization of the shunt tubing. On the other hand,
VP shunt infection from a readily identifiable source remains
a common complication (see Chapter 128); a complete diag-
nostic workup is required including a shunt series, tapping of
the shunt reservoir for culture and sensitivity, head CT scan,
and abdominal imaging depending on presenting complaints.
When shunt fluid appears sterile, Gram stain and culture of
peritoneal fluid may help to differentiate PP or shunt infection
from an intra-abdominal source. Common intra-abdominal
infectious processes like appendicitis must be treated
promptly, and shunt exteriorization strongly considered.
of PP, consideration should be made for elective
appendectomy.
Sterile Peritonitis
------------------------------------------------------------------------------------------------------------------------------------------------
Although PP is traditionally defined as an infectious process,
certain disease entities are associated with peritoneal inflam-
mation without the presence of an offending organism.
1234 PART VII ABDOMEN
SYSTEMIC LUPUS ERYTHEMATOSUS
Systemic lupus erythematosus (SLE) is a rheumatologic
condition characterized by the formation of autoantibodies di-
rected at self-antigens. The pathophysiology is multifactorial,
and the presentation can vary greatly among patients. Serositis
is a frequent manifestation of the disease, and PP has been de-
scribed as the index presentation for SLE.55 Peritoneal inflam-
mation is not a consequence of bacterial contamination, but
rather a reaction to antibody-antigen interaction. A recent re-
view of pediatric-onset SLE demonstrated a 19% rate of ab-
dominal involvement, most commonly due to pancreatitis
or new-onset ascites. Three children in the series, one of
whom was found to have acalculous cholecystitis, underwent
laparotomies for peritonitis before the diagnosis of SLE.56 For
patients with known SLE, high-dose steroids usually prove ef-
fective as first-line therapy for acute abdominal pain once
other surgical pathologies have been ruled out. This can some-
times prove difficult in immunodepressed children, in whom
clinical signs of intra-abdominal perforation or ischemia may
be masked. Clinical vigilance is required, and extensive inves-
tigations are frequently necessary.
FAMILIAL MEDITERRANEAN FEVER
Familial Mediterranean fever (FMF, also known as recurrent
polyserositis) is characterized by transient episodes of pleuritis,
arthritis, and peritonitis and primarily affects Armenians,
Turks, and Jews originating from the Mediterranean area.
Its inheritance is generally recessive, although varying pene-
trance has been reported depending on the genetic mutation.
Nonsense or missense mutations in the MEFV (Mediterranean
fever) gene appear to cause the disease in many cases. MEFV
causes the production of inflammatory proteins, pyrin (de-
rived from the association with predominant fever) or mare-
nostrin (derived from the phrase “our sea,” because of the
Mediterranean heritage of most patients). Frequent episodes
of inflammation of the peritoneum (85%), joints (50%), and
pleura (33%) are associated with high fever and abdominal
tenderness. A recent review has demonstrated that 31% of
patients manifest symptoms by 2 years of age, although this
young cohort is less likely to present with peritonitis.57 In
FMF, the abdominal symptoms begin to abate within 12 hours
of onset, with resolution within 24 to 48 hours. If surgical
intervention is undertaken, the bowel is characteristically
coated with a sterile exudate containing fibrin. As in SLE,
ascitic fluid is characteristically sterile in FMF; antibiotics
do not play a role in the acute setting or as prophylaxis.
The use of colchicine may decrease the severity of painful ep-
isodes and reduce the incidence of long-term adhesive small
bowel disease.58 Five to ten percent of patients have not
responded to colchicine; anakinra (an interleukin-1 receptor
antagonist) can control the acuity of febrile attacks in adoles-
cents.59 Once the diagnosis of FMF has been considered, a
complete family history should be obtained and genetic test-
ing promptly offered. Long-term complications include the
development of adhesive small bowel obstruction in 3% of
pediatric cases.60
The complete reference list is available online at www.
expertconsult.com.
 feeding for neonates with atresia of the duodenum or high
jejunum were introduced next.19,20 In the past 3 decades, endo-
scopic, laparoscopic, and various image-guided approaches have
been added to established open techniques continuously foster-
ing the creation of feeding, venting, decompressing, irrigating,
and special-purpose stomas. Understanding of stomal physiol-
ogy and of specialized enteral and parenteral nutrition, as well
as the diagnosis and management of stoma-related complica-
tions, have paralleled the advances in technique significantly
improving outcome.
Several other factors have contributed to the safety, effec-
tiveness, and ease of care of stomas in adults and children.
Paramount among these is the advent of enterostomal therapy,
which has evolved into a specialty in its own right.21,22 Enter-
ostomal therapists are now an integral part of health care
teams in most medical institutions. Major national and inter-
national ostomy associations23 foster the dialogue among
professionals and provide a wealth of information through
traditional and web-based material including publications
for parents, caregivers, and teenage patients.24,25
Regional and local chapters are involved in establishing non-
medical support systems and guidance to access resources.26
Greater awareness and acceptance of ostomates, as well as
CHAPTER 98 the recognition of their needs and rights among the lay popu-
lation, has also helped to improve their quality of life. The
knowledge and experience derived from enterostomal care
has led to the creation of appliances in a wide variety of types
Stomas of the Small and sizes, manufactured of well-tolerated biomaterials and
complemented by numerous stoma care products.27 Not sur-
prisingly, at times, parents, caregivers, or ostomates contribute
and Large Intestine innovative ideas to the established management techniques.28
In the contemporary clinical setting, primarily because of
earlier diagnosis of certain gastrointestinal anomalies such as
Michael W. L. Gauderer Hirschsprung disease, improved surgical approaches, and peri-
operative care, pediatric surgeons were able to safely perform
more single-stage procedures, thereby decreasing the need for
preliminary decompressing enterostomies (ileostomies and co-
Historical Note lostomies).29–33 Conversely, due to an ever-increasing number
------------------------------------------------------------------------------------------------------------------------------------------------
of children with a variety of complex surgical and nonsurgical
The word stoma originates from the Greek stomoun (to provide pathologies, there has been a greater demand for upper gastro-
with an opening or mouth). Intestinal stomas, considered basic intestinal access for long-term enteral feeding (gastrostomies
surgical procedures, have a long and colorful history.1–4 As a and jejunostomies),34–38 as well as lower intestinal access for
method of treating intestinal obstruction, colostomies date back antegrade enemas (appendicostomies, tube cecostomies, and
to the latter part of the eighteenth century and some of the first tube sigmoidostomies).39–43 Often requiring a team approach,
survivors of this procedure were children with an imperforate the creation, care, and closure of enterostomas continue to
anus.5–7 Despite sporadic early successes, the use of stomas occupy a substantial portion of pediatric surgical practice.
in the large intestine and later the small intestine in children
evolved slowly. Surgeons were understandably reluctant and Child with a Stoma
------------------------------------------------------------------------------------------------------------------------------------------------
even strongly opposed to performing these drastic procedures,
which were associated with major complications. However, as An enterostoma in a child is a major disruption of normality
the experience of surgeons increased toward the end of the and frequently leads to substantial psychologic trauma for the
nineteenth century and beginning of the twentieth century, child and parents. However, most decompressing intestinal
colostomies and occasionally jejunostomies were used to man- stomas in the pediatric age group are temporary and correc-
age a few pediatric conditions. With the advent and the devel- tion of the underlying problem often leads to closure of the
opment of pediatric surgical practice in the mid to late 1900s diverting opening. Although pediatric surgeons continuously
and survival of children with conditions that were formerly search for alternatives to intestinal exteriorization, an appro-
likely to be fatal, the need for stomas increased. Enterostomal priately indicated, properly constructed temporary stoma is
construction techniques, originally developed for adults,8–12 frequently unavoidable and lifesaving. Moreover, in several in-
were modified and adapted for use in children, particularly stances of noncorrectable and crippling pathologic conditions
newborns with congenital intestinal obstruction.13–18 New of the lower intestinal tract, a permanent, well-functioning
techniques that combined proximal decompression and distal stoma contributes to an improved quality of life.44,45

1235
1236 PART VII ABDOMEN

Despite many advances related to enterostomas, their Indications for Enterostomas

placement, care, and closure are still associated with a surpris-
ingly high rate of both early and late complications.46–70 These in Children
facts present the surgeon, the enterostomal therapist, the ------------------------------------------------------------------------------------------------------------------------------------------------

nurses, the parents, and the child with major challenges. Temporary and occasionally permanent stomas of the small
Therefore when the need for a stoma arises, the best results and large intestine are used in the management of a wide
are achieved by carefully evaluating the child’s pathologic variety of surgical and nonsurgical pathologic conditions in
condition and health status, weighing the pros and cons of neonates, infants, and children. With the exception of feeding
diversion, planning ahead (for eventual closure) whenever and antegrade enema access, more than one half of all stomas
possible, and considering both construction and takedown are placed in the neonatal period and another one fourth in
as major interventions. infants younger than 1 year of age.51,52,54,59
In addition to the well-defined guidelines for stomal place-
ment established for adult patients, pediatric factors including
JEJUNOSTOMIES
anatomic and physiologic differences, delicate structures,
growth, and physical and emotional maturity, as well as Indirect access to the jejunum via naso-jejunal or gastro-
preoperative preparation, whenever possible, need to be con- jejunal route is adequate for short- or intermediate-length
sidered.24,25,71 The surgeon and members of the surgical team nutritional support.36 Direct access to the proximal small
must always keep in mind that the quality of life of a patient bowel is most commonly used for long-term enteral alimenta-
with a stoma is largely related to the quality of that stoma. tion as an alternative to a gastrostomy, which is the preferred
route.34,72 The majority of patients requiring a feeding jeju-
nostomy are neurologically impaired children, usually with
Types of Enterostomas complex medical problems associated with foregut dysmoti-
------------------------------------------------------------------------------------------------------------------------------------------------
lity. Some of these may require both a gastrostomy and a
The four basic types of enterostomas, primary purposes, and jejunostomy in their management. Jejunal access can also be
technique options are listed in Table 98-1. Examples of these useful in the care of patients with acute surgical problems
methods are illustrated in Figures 98-1 to 98-3. Options benefiting from early enteral nutrition (e.g., major trauma
for bringing the proximal stoma through the abdominal or burn victims, children needing long-term supplemental
wall and handling the distal stoma are listed in Table 98-2. feedings). Various types of exteriorized jejunal segments were
Examples are found in Figures 98-3 to 98-5. once used in the management of infants with biliary atresia, in
an attempt to reduce ascending cholangitis. However, this
approach is no longer used, in part because of secondary prob-
TABLE 98-1
lems such as bleeding from stomal varices associated with por-
Applications and Considerations for Enterostomas
tal hypertension50 and because the stoma adds complexity to a
future liver transplantation.
Administration of Feedings, Medication, or Both On the other hand, the use of a segment of intestine or
Without entering the jejunal wall: nasojejunal tube, gastrostomy- drainage device interposed between the gallbladder and the
jejunostomy tube34
abdominal wall for partial drainage of bile has been helpful
Direct access through the jejunal wall: tunneled catheter,9 needle
catheter, T-tube,82 button,100 other
in the management of children with some types of genetic
Isolated jejunal loop brought directly to abdominal wall:
cholestatic syndromes.73–76 As with other segments of the
Roux-en-Y108–110 intestine, exteriorization55,77–80 or tube decompression81 is
indicated following bowel resection when a primary anasto-
Proximal Decompression and Distal Feedings mosis is unsafe or impossible (e.g., necrotizing enterocolitis,
Gastrostomy and distal feeding tube, same stoma or separately20,34 midgut volvulus).
Double-lumen tube in dilated proximal jejunum with feeding end
across an anastomosis19; or two single-lumen tubes inserted
separately into divided, closed loops of small intestines81
Divided intestinal segments brought directly to skin level, with ILEOSTOMIES
pouch applied to proximal stoma and feeding catheter inserted
into distal one These stomas are essential in the management of neonates
Access for Antegrade Irrigation
with certain types of distal intestinal obstruction (e.g., long-
Appendix or other intestinal conduit brought to abdominal wall
segment Hirschsprung disease, complex meconium ileus,
for intermittent catheterization5,52,85 gastroschisis with atresia).13,52,54,82 Ileostomies are com-
Catheter, T-tube, skin level device placed in intestinal monly placed to divert bowel contents when reestablishing
lumen37,41,81,88 bowel continuity is precluded by peritonitis, ischemia, or
hemodynamic instability ( e.g., neonatal necrotizing enteroco-
Decompression, Diversion, or Evacuation
litis) (Figs. 98-6 and 98-7).77–79,82 Ileal diversion has tradi-
End stoma, single opening11
tionally been used in the surgical approach to colonic
Double-barrel stoma10,17
pathology (e.g., ulcerative colitis, familial polyposis) as tem-
End stoma with an anastomosis below the abdominal wall13,15
porary, protective, or, at times, permanent stomas.3,4,11,83,84
Loop over a small rod or skin bridge8,14
Less common indications include other forms of inflammatory
Closed loop with catheter81 or open loop with occluding
valve-type device allowing controlled egress bowel disease, rare manifestations of colonic dysmotility, and
Special stomas such as a catheterizable pouch36,47 monitoring of the intestinal graft in patients with small bowel
transplantation.
 CHAPTER 98 STOMAS OF THE SMALL AND LARGE INTESTINE 1237

A B C

D E
F
FIGURE 98-1 Diagrams of select-feeding and decompressing-feeding jejunostomies. A, Tunneled catheter.9 B, Needle catheter. C, T-tube.82 D, Button.100
E, Proximal decompression and distal feeding across an anastomosis.19 F, Temporary decompression feeding using catheters when primary anastomosis
is unsafe and intestinal exteriorization is undesirable or not possible.81

APPENDICOSTOMIES, TUBE CECOSTOMIES,

AND TUBE SIGMOIDOSTOMIES
The main indication for these interventions is to provide
long-term access sites for antegrade intestinal irrigation in
children with colonic motility, anal sphincter problems, and
myelodysplasia.39–42,85–91

COLOSTOMIES
Stomas of the large bowel have the longest history, and exten-
sive experience with these enterostomies has accrued.1–7
Diversion of fecal stream is essential in the treatment of several
congenital hindgut pathologies (e.g., high forms of imperfo-
rate anus,5,6,67 late diagnosis or complicated Hirschsprung
disease,68 complex pelvic malformations,92 colonic atresia93).
Colostomies are also used in patients with severe colonic,
anorectal or perineal trauma,32,94,95 perineal burns,96 and
complications of malignant conditions.58,97 Unlike in the
adult population, in which colorectal cancer is the most com-
mon indication, colostomies are rarely permanent in children.

UROSTOMIES
Exteriorized segments of ileum or colon have been used as
conduits in the management of urinary tract pathologies,
although these diversions are seldom used today. However,
the mobilized appendix, interposed between the bladder FIGURE 98-2 Roux-en-Y feeding jejunostomy with a balloon-type
and the abdominal wall surface, is used in children with skin-level access device.108
1238 PART VII ABDOMEN

A B C

D E F
FIGURE 98-3 Examples of decompressing, diverting, or evacuating stomas. A, End stoma (inset shows typical maturation).11 B, Double-barrel stoma.10
C, End-to-side anastomosis with distal vent for irrigation.13 D, Side-to-end anastomosis with proximal vent.15 E, Loop stoma.8 F, End stoma with closed
subfascial distal of the end of the intestine (inset shows rodless end-loop stoma).

TABLE 98-2 attachment of the bowel to the abdominal wall.36,37,100,101 Di-

Enterostoma Exit rect percutaneous endoscopic jejunostomy (PEJ) is applicable to
Proximal Stoma older patients but difficult in small children due to limita-
Through celiotomy incision tions imposed by the endoscopic equipment.102 A number of
Through separate opening image-guided jejunostomies have been described but have
With proximal and distal limbs close to each other been within the purview of radiologists in a few pediatric
With proximal and distal openings apart centers.37,103,104 Laparoscopic or laparoscopically assisted tech-
Multiple stomas niques are now used with increasing frequency in all age
Variations of the above groups.105–107 Bringing the loop directly to the abdominal wall
and placing a skin-level access device is simple and effective.
Distal Stoma Peristomal leakage is always a concern. An alternative intended
Exteriorization as mucus fistula adjacent to or separate from to decrease this problem is the more complex Roux-en-Y
proximal intestine
approach.108,109 However, this method has a greater potential
Partial closure and placement next to the proximal stoma92
for serious complications such as volvulus and internal
Closure and replacement into abdominal cavity
hernias with intestinal obstruction.110,111
Closure after placement of a catheter for subsequent access
for irrigation or contrast studies
The choice of access device depends on the type of stoma and
the age of the child.27,112 Because straight catheters can be diffi-
cult to immobilize or replace in conventional tunneled jejunos-
various urinary tract dysfunctions to provide a catheterizable tomies, a good alternative is a T-tube for infants (because it does
conduit to the urinary bladder.98,99 not obstruct the narrow lumen) (see Fig. 98-1, C), and an original
type button (Fig. 98-1, D) or other nonballoon skin-level device
for older pediatric patients. Balloon-type devices are suitable for
Choice of Enterostoma the Roux-en-Y loop (see Fig. 98-2). As with a gastrostomy, these
------------------------------------------------------------------------------------------------------------------------------------------------ devices are both replaceable as an office procedure.
FEEDING JEJUNOSTOMY
ILEOSTOMY
Various approaches for establishing direct long-term access to
the jejunum are now available. “Open” placement through a In intra-abdominal interventions requiring intestinal resection,
small, left upper quadrant incision permits excellent identifica- such as neonatal necrotizing enterocolitis, many surgeons prefer
tion of the stoma site in the proximal jejunum, as well as secure to exteriorize a single-end stoma through a counterincision
 CHAPTER 98 STOMAS OF THE SMALL AND LARGE INTESTINE 1239

A B C
FIGURE 98-4 Examples of options for the management of infants after intestinal resection. A, Exteriorization of proximal intestine through a counter-
incision and closure of distal intestine beneath the abdominal wall. B, Same procedures as in A with exteriorization of proximal end of distal intestine
through the wound edge. C, Arrangement after resection of two intestinal segments.

A B
FIGURE 98-5 Sigmoid colostomies. A, Separated stomas. The proximal intestine is at the upper end of the incision, and the mucus fistula is at the
lower one. B, Loop colostomy. The intestine is exteriorized over a rod or skin bridge or with the help of sutures. The circumscribing comma-shaped incision
is used for takedown and pull-through procedures.

(see Figs. 98-4, A, 98-6, and 98-7). A more expedient
alternative is to bring the proximal intestine through the
end of the incision (see Fig. 98-4, B). However, with this
approach, wound complications are more common. In addi-
tion, if the stoma must remain for a prolonged period of
time and the child gains weight, the fold created by the laparo-
tomy incision may interfere with fitting of the stoma appliance
(see Fig. 98-7).
With a healthy distal intestine and anticipated downstream
patency, the distal limb may be closed and placed intra-
abdominally adjacent to the proximal stoma. Otherwise, exte-
riorization as a mucus fistula is prudent (see Fig. 98-4, B).
The use of an exteriorized loop stoma rather than an end
stoma is an alternative in which the intact mesentery pro-
vides maximal perfusion.79 A double-barreled stoma is a
time-honored option.77,78 To save as much intestine as possi-
ble, the placement of multiple stomas may be necessary (see
Fig. 98-4, C). Although some ileostomy types were developed
specifically for newborns with meconium ileus, they are no
longer used. However, T-tube ileostomies have been useful
for the instillation of liquefying solutions.82
In children with ulcerative colitis or familial polyposis, the
enterostomal principles are similar to those established for
adult patients. Choices for a temporary protective diverting
1240 PART VII ABDOMEN
FIGURE 98-6 One-year-old boy with severe necrotizing enterocolitis with
loss of distal ileum and colon down to the peritoneal reflection before rea-
nastomosis. Liquid stools precluded earlier reestablishment of intestinal
continuity. Notice the appliance mark and the appropriate distance from
the incision, the umbilicus, the inguinoabdominal fold, and the right anterior
superior iliac spine.
FIGURE 98-7 Same child as in Figure 98-6 in a sitting position. Notice the
deep crease produced by the transverse supraumbilical incision. A stoma
brought out through such an incision would have precluded proper use of
the pouch, and a revision would have become necessary.
ileostomy include a simple loop, an end (distally closed) loop,
and an end stoma, with the closed distal end under the fascia
(see Fig. 98-3, F).
APPENDICOSTOMY, TUBE CECOSTOMY,
OR TUBE SIGMOIDOSTOMY
The choice of antegrade colonic enema (ACE) depends on the
type of colonic pathology being managed. With normal peris-
talsis, either the right41 or left40,113 colon may be chosen for
access. However, if dysmotility is a concern, access to the right
colon is indicated. If the appendix is present, it is exteriorized
with or without interposition of a “valve” by either an “open,”88
or laparoscopic approach.43,90 If the appendix is no longer avail-
able, the wall of the cecum may be fashioned into a conduit that
is then brought to the skin level.114 Exteriorizing the appendix
at the umbilicus has cosmetic advantages. Either the appendix
or the conduit so constructed is then catheterized to instill
the enema fluid. A simpler technique, especially if there is no
appendix, is the placement of skin-level device in the cecum
by an open115 or percutaneous approach.87 For patients with
normal colonic motility, access to the left colon by means of a
sigmoid irrigation tube can be advantageous.40
COLOSTOMY
Most colostomies fall into three categories: right transverse, left
transverse, and sigmoid. The significant physiologic and ana-
tomic differences among these three must be taken into con-
sideration when choosing the site for the stoma. For infants
with high imperforate anus, the high (proximal) sigmoid is
the preferred site for exteriorization (see Fig. 98-5).62,116 The
main advantages are firmer stools with less tendency for skin
excoriation, less tendency for prolapse, less surface for urine
absorption, and less contamination of the urinary tract in male
children with rectovesical fistula. Sigmoid stomas assist evacua-
tion of meconium from the often dilated distal portion of the
bowel during the initial procedure. The precise site is easily iden-
tified using the pelvic peritoneal reflection as a guide. A further
advantage is that there are no scars in the epigastrium. However,
if the low or mid sigmoid is inadvertently exteriorized, there may
be interference with the blood supply, as well as insufficient
bowel length for the future pull-through.67,116 If the stoma is
placed in the transverse colon, there is always adequate bowel
length for pull-through, and the intestine is easy to mobilize
and has a smaller diameter and no meconium. The disadvantages
of transverse colon colostomy, however, are sizeable: The stools
are looser, skin maceration and dehydration are more common,
there is a greater prolapse rate, and there is an increased possi-
bility of urinary tract problems. In addition, adequate evacuation
of meconium is nearly impossible. Although high sigmoid loop
colostomy is still used (Fig. 98-8), contemporary preference is
for separation of the stomas, particularly in boys (Fig. 98-9).67
In children with Hirschsprung disease requiring a pre-
liminary colostomy, the best site is the dilated segment that con-
tains normal ganglion cells found proximal to the transition
zone. A loop colostomy is usually chosen, although the ten-
dency for prolapse is increased.68 Because most transition zones
are in the sigmoid colon, this lower left quadrant stoma is taken
down at the time of the definitive corrective operation (see
Fig. 98-5, B). If separation of the stomas is chosen, the distal
intestine should not be oversewn in patients with Hirschsprung
disease, particularly if the aganglionic segment is long, because
mucus cannot be appropriately evacuated or washed out.
Although similar data are not available in children, properly
constructed loop colostomies are fully diverting in adults.117
Select Technical Aspects
------------------------------------------------------------------------------------------------------------------------------------------------
Feeding jejunostomies are generally placed in the left upper
abdomen, slightly above the umbilicus, not so cephalic as
to interfere with a possible gastrostomy and/or fundoplication.

FIGURE 98-8 Five-month-old child with high imperforate anus. The
proximal sigmoid loop colostomy is equidistant from the umbilicus, the
anterior superior iliac spine, and the inguinal fold. The original incision
is only slightly longer than the stoma. Notice the raised “spur” between
the two lumina, essential for proper diversion of stool.
FIGURE 98-9 Neonate with high imperforate anus. A divided proximal
sigmoid colostomy was placed. The separation of the bowel ends mini-
mizes the incidence of stoma-related problems.67 The proximal bowel is
slightly everted, and the mucus fistula is flush with the skin. (Courtesy
Dr. Mark Levitt.)
In the “open” technique, the proximal jejunum is approached
through a small, upper left quadrant incision. The ligament of
Treitz is identified, and the catheter or skin-level device is
inserted in the antimesenteric portion of the intestine, 10 to
20 cm distal to the duodeno-jejunal junction. A purse-string
suture of fine multifilament synthetic absorbable material is
placed around the enterotomy and tied. The catheter or
skin-level device is then brought out through a counterinci-
sion. A second purse-string suture, made of monofilament
synthetic absorbable suture is applied, with the sutures alter-
nating between the intestine and the exit site of the catheter
in the abdominal wall. When tied, this second suture approx-
imates the intestinal serosa to the parietal peritoneum in a
watertight manner.100
CHAPTER 98 STOMAS OF THE SMALL AND LARGE INTESTINE 1241
If a PEJ is chosen, the retaining intraluminal bumper must
be size appropriate. Laparoscopic control can be used to
increase the safety of PEJ, particularly in patients with abnor-
mal epigastric anatomy. With laparoscopically assisted jeju-
nostomies, particularly the Roux-en-Y type, proper loop
orientation is essential. To minimize leakage (the most com-
mon problem with jejunostomies), appropriately sized skin-
level devices must be selected. Devices that are too short or
excessive tension on immobilizing crossbars must be avoided
to minimize bowel wall or skin ischemia.
Decompressing ileostomies are usually placed in the right
lower quadrant (see Figs. 98-4 and 98-6). The umbilicus is a
possible site for a stoma118 and is an excellent choice for the dis-
tended proximal intestine in newborns who have gastroschisis
with atresia (Fig. 98-10).
Figure 98-11, A illustrates both appropriate and undesir-
able stoma exit sites in neonates, infants, and small children
(e.g., those with necrotizing enterocolitis). Figure 98-12, B
demonstrates ideal exit sites in older children or adolescents
(e.g., those with ulcerative colitis or familial polyposis). Lap-
arotomy incisions in the lower quadrants should be avoided in
patients who may eventually have long-standing or permanent
stomas because such incisions can create an uneven surface
that interferes with pouch adherence.
When an enterostoma is anticipated, it is important that the
site of the stoma and possible alternatives are marked on the
abdominal wall before any incision is made. This planning is
desirable in both elective and emergency settings. For elective,
long-standing stomas, the best location is determined and
marked the day before the operation (Fig. 98-12, A and B).
The exit site should be located over the convex midportion
of the rectus muscle, away from the incision, umbilicus, bony
prominences, and skin folds. Special attention must be paid in
overweight children because of the deep creases of the abdom-
inal wall. In older children, if a vertical midline laparotomy is
planned, it is advisable to create the opening for the ileostomy
before making the incision. This is done in order to achieve a
FIGURE 98-10 Four-month-old child with gastroschisis and small bowel
atresia during reestablishment of bowel continuity. The dilated and edema-
tous ileus was brought out as an end stoma through the umbilical site. The
proximal closed end of the colon was attached to the side of the ileum
underneath the abdominal wall. This maneuver allows prompt identification
of the distal bowel, minimizing dissection and incision size.
1242 PART VII ABDOMEN

A B
FIGURE 98-11 Ideal sites for stomas. A, Infant. The end stoma can be brought out through a counterincision in the lower right or left quadrants. The sites
marked with an “X” are unsuitable because they are too close to the rib cage, the anterior superior iliac spine, the flank, or the groin. B, Older child or
adolescent. The best site for the stoma is in the mid rectus abdominis muscle in the right lower quadrant. The opposite side is an alternative. Areas marked
with an “X” are unsuitable. The same sites are used with minimally invasive procedures.

straight course for the bowel and avoid distortion of the layers
by traction on the edge of the incision. The entire opening
should be just wide enough to allow for comfortable passage
of the ileum without interfering with the blood supply. In
laparoscopically placed stomas, adequate stoma exit size
and bowel orientation are critical. With either approach, the
bowel is secured intraperitoneally to avoid torsion and inter-
nal hernias and then secured with fine absorbable sutures to
the rectus sheath. Depending on the size of the child, the ma-
tured ileostomy must protrude 2 cm or more to allow proper
pouch fixation. This pouch is applied at the end of the proce-
dure (Fig. 98-12, C). Stomas in neonates, particularly those
established for necrotizing enterocolitis, should not be ma-
tured because this will interfere with the already tenuous
blood supply. The exteriorized end of the stoma is simply
anchored to the skin with four delicate sutures of a synthetic
absorbable material. An antibacterial ointment is applied, and
dressings are avoided. A pouch is applied with the onset of
defecation. In infants, the mucosa grows rapidly over the
exteriorized serosal surface. Deep, full-thickness sutures in
the bowel should be avoided because they may cause a fistula
from the lumen to the peristomal tissue, which will interfere
with stoma pouch adherence.
The preferred colostomy site is the lower left quadrant. The
guidelines for placement are similar to those for ileostomies.
The most common site problem, particularly in newborns, is
that the stoma is placed too caudally, close to the inguino-
abdominal skin folds. When an infant flexes his or her hips,
the resulting folds tend to lift the edges of the stoma appliance,
leading to leakage. In children with imperforate anus, the
meconium-filled sigmoid colon and the pelvic peritoneal ref-
lection are identified once the abdominal cavity is opened.
A suitable portion of the uppermost sigmoid colon is selected,
and the bowel is exteriorized. When the dividing technique is
used (see Fig. 98-5, A), the stomas are placed at each end of
the incision and the intestine is secured with fine, synthetic, ab-
sorbable suture to all layers of the abdominal wall. There should
be enough space between the two openings to permit a good fit
for a pouch over the proximal stoma (or the distal stoma must
be flush with the skin) (see Fig. 98-9). To avoid excessive nar-
rowing of the stoma, an appropriately sized Hegar dilator or
catheter is inserted into the intestinal lumen at the time of
wound closure. End colostomies should only protrude slightly.
With a loop stoma (see Figs. 98-5, B and 98-8), the incision is
the length of that loop or only slightly longer. With the loop
technique, a temporary catheter is placed through the mesen-
tery of the selected segment, which is then lifted above the level
of the skin. Triangulating sutures approximate the two limbs to
each other and to the peritoneum on both sides to prevent in-
ternal hernias. The full circumference of the intestine is then
attached to the peritoneum and fascia. Sutures lift the posterior
bowel wall above the skin level. The intestine is opened
longitudinally, and the edges everted. In all children with
imperforate anus, the distal, meconium-filled segment of intes-
tine is evacuated and flushed out at the time of colostomy
placement. This is important to avoid formation of a fecaloma.
In patients with Hirschsprung disease, the construction of
the loop stoma must also be meticulous and may include
tightening of the distended intestine to decrease the possibility
of prolapse. This is particularly important in the distal segment.
Rods or skin flaps placed under the loop are unnecessary if an
appropriate “spur” between the two openings was created.
To facilitate subsequent takedown, when exteriorizing both
ends of the small or large intestine, these should be kept as
 CHAPTER 98 STOMAS OF THE SMALL AND LARGE INTESTINE 1243

A B

C D

FIGURE 98-12 Eight-year-old boy with intractable ulcerative colitis that

failed to respond to aggressive medical management including corticoste-
roids, azathioprine, and infliximab. He underwent a total abdominal colect-
omy and ileostomy as a preliminary stage to an ileoanal anastomosis.
A, The patient in a supine position. The stoma site had been determined
several days before the procedure. The boy was asked to wear a two-piece
pouch using the selected site as a guide. B, Bringing the already anesthe-
tized patient into a sitting position confirmed the adequacy of the stoma
site away from major abdominal folds. C, Skin barrier with flange (two-piece
system) applied to the ileostoma at the completion of the colectomy. D, Six
weeks later the stoma was well healed with mild residual edema. E, Same
boy wearing a two-piece system with an opaque pouch. He emptied the
pouch three to four times during the day and two to three times evening
and nights. He enjoyed full physical activities including basketball. He did
not use an ostomy belt and changed the skin barrier and pouch every E
3 days. Bowel continuity has since been reestablished.
1244 PART VII ABDOMEN

close as possible without interfering with the pouch. If the dis- application of silver nitrate may be necessary to control granula-
tal limb is placed underneath the abdominal wall, it is tagged tion tissue around the mucosa-skin interface in the early stages.
with a nonabsorbable suture or a metallic clip (or both) and Remaining sutures are often the cause and should be re-
placed as close as possible to the exiting stoma to simplify moved. Routine dilatation of stomas is not recommended. Mal-
identification at reanastomosis. Placing a metallic clip also functioning stomas often require early takedown or revision
helps with radiographic identification of the proximal end before more serious complications occur. Occasional irrigation
of the distal bowel and its patency when a barium enema is of the distal intestine can be useful and help eliminate malodor-
performed before reestablishment of bowel continuity in ous mucus plugs. Dietary and select pharmacologic manipula-
children with necrotizing enterocolitis. tions are helpful in producing firmer stools. Children with high
Several of these techniques can be adapted for a minimally ileostomies must be carefully monitored to prevent electrolyte
invasive approach.43,107,119–123 In addition to smaller access imbalance and insufficient nutrient absorption.49,133,134
sites, laparoscopy can be helpful in the precise identification
of intestinal loops and targeted lysis of adhesions. However,
the surgeon must exercise caution because the advantages
of the laparoscopic approach can be negated by less than
Reestablishing Intestinal
optimal handling of the bowel and exteriorization through Continuity
openings of inadequate size.119 ------------------------------------------------------------------------------------------------------------------------------------------------

Timing of enterostoma closure varies widely depending on the

underlying condition, health status of the child, and presence
Stoma Care
------------------------------------------------------------------------------------------------------------------------------------------------
or absence of stoma-related complications.78,135–139 Unneces-
sary delays in the reestablishment of bowel continuity tend
Proper care of an enterostoma begins with preoperative prepa- to increase morbidity and should be avoided.62,63,137 The more
ration whenever possible.24,25,124 Parents, as well as older proximal the stoma, the earlier it should be closed to decrease
children, must be carefully taught and reassured before leav- metabolic complications.62,133,134 Children who previously
ing the hospital and on subsequent follow-up visits. Ileos- underwent resection because of ischemic intestine must have
tomies and proximal colon colostomies always require a preoperative contrast study of the distal segment to rule out
pouches. With the well-formed stools that result from sigmoid strictures or complete luminal obstruction. Use of routine stud-
stomas, some parents of infants have used a skin barrier and ies in other settings has been questioned.140 Reestablishment of
diapers instead. A plant leaf has been recently suggested as an small bowel continuity generally does not require intestinal
alternative.125 A large selection of stoma appliances are com- preparation. Takedown of a colostomy is preceded by antegrade
mercially available including disposable and reusable pouches intestinal irrigation, supplemented by conventional enemas.
for all ages and sizes, even the smallest premature infants. Skin Although perioperative intravenous antibiotics are routinely
barriers, adhesives, powders, vented pouches, and odor con- administered, the use of intraluminal antibiotic solutions is
trol solutions are among the products that make the care of controversial141 and probably not indicated. When reestablish-
today’s ostomate easier.1,2,27,112,126 ing continuity involves the left colon or sigmoid, the insertion of
Properly fitted appliances should remain in situ for several a soft catheter into the rectum assists the intraoperative identi-
days; 3 days is a reasonable expectation. There are two basic fication of that loop, particularly if the anatomy has been altered
types of pediatric appliances: the one-piece pouching system by peritonitis or a previous operation. Good exposure with full
in which the adhesive skin barrier is already attached to the mobilization of the intestinal ends is important. Laparoscopy
pouch and the two-piece system in which the adhesive skin can be helpful in select patients.142 Although extraperitoneal
barrier is separate from the pouch. In the latter, the pieces closure has been used in children, it is not recommended for
snap together with a flange (see Fig. 98-12, C and E). Because routine use.143 For the intestinal anastomoses, a single-layer
of the holding power of contemporary adhesive skin barriers, technique using fine interrupted sutures of synthetic absorb-
additional fixation with tape or belts is usually not necessary. able suture material provides excellent results. Primary wound
The skin barrier is cut to the proper stoma size with the help closure is generally safe and advantageous.139,144,145 Early post-
of a template provided with the pouches. In addition to operative feeding after colostomy closure is encouraged.146
instructions provided by the surgeon, nurses, and the enteros-
tomal therapist, parents are encouraged to contact one of the
enterostomal societies such as the United Ostomy Association Complications of Enterostomas
of America (an affiliate of the International Stoma Association) ------------------------------------------------------------------------------------------------------------------------------------------------

or the Wound Ostomy and Continence Nurses Society23,127 Problems related to construction, care, and closure of stomas
and make use of the extensive printed and electronically in the small and large intestines are numerous and com-
available educational material. Although “continent stomas” mon. They can lead to significant morbidity and occasional
using special intra-abdominal intestinal pouches,83,128 mag- mortality (Table 98-3). Analysis of pediatric series reveals
netic caps,129 valves,130 inflatable plugs,131 and other forms complication rates that often reach and sometimes exceed
of luminal occlusion132 have been attempted, there is limited 50%.16–18,46–70,132–139
experience with such operations or devices in children. In addition, stoma revision or early takedown is frequently
Candidiasis remains a common problem in the parastomal necessary.47,54,68,135 Complications of enterostomas used for
skin, and local antifungal medication should be used at the ear- feeding are often accentuated by the patient’s underlying
liest sign of irritation. With skin excoriation, the area is exposed disease, particularly in malnourished, neurologically impaired
to air and a synthetic barrier is applied. A hair dryer can be useful. children.57,72,109 Stomas used for evacuation of the small
Mild stomal bleeding is usually self-limiting. Excision and/or intestine are associated with a higher morbidity than are

TABLE 98-3
Common Complications of Enterostomas
Prolapse
Stricture
Retraction
Wound separation, dehiscence
Wound infection, postoperative sepsis
Parastomal hernia
Intestinal wall separation or perforation with catheter change
Exteriorization of wrong intestinal segment or end
Intestinal obstruction (adhesion, internal hernia)
Intestinal torsion with ischemia
Fistula formation
Perforation by feeding or irrigating catheter
Poor appliance fitting and leakage
Psychological trauma
Skin excoriation, candidiasis, dermatitis
Mucosal excoriation and bleeding
Granulation tissue of mucosa-skin interface
Variceal bleeding with portal hypertension
Electrolyte imbalance
Acidosis (caused by urine absorption in the distal loop of intestine)
Fecal impaction (in the distal loop of intestine)
colostomies because these stomas are compounded by fluid,
electrolyte, and absorption losses.39,133,134 Transverse colos-
tomies are more prone to complications than are sigmoid
stomas47,53,62,67,147; in several series, divided colostomies
have been preferred over loop colostomies.62,67
Among the more common serious complications are pro-
lapse (Fig. 98-13), stricture, and retraction.51–57,63,66,67 The
incidence of prolapse in children may exceed 20% and is more
common if the distal loop is exteriorized.51–57,62–69 Stomal
prolapse may be categorized into minor and major forms. Mi-
nor prolapse is associated with a protuberant, edematous
stoma that is still functional. It is usually amenable to nonsur-
gical techniques aimed at decreasing bowel edema, permitting
FIGURE 98-13 Six-month-old child with Down syndrome, congenital
heart, and Hirschsprung diseases. Proximal and distal colon prolapse have
occurred through a left transverse colon loop colostomy. Notice the three
openings at the end of the larger everted loop that correspond to the lumen
of the cecum, the ileocecal valve opening, and the lumen of the appendix.
The problem was managed by anchoring the cecum in the right lower quad-
rant and then securing the distal loop with a U-stitch and two rubber
bolsters.157 A definitive operation followed several months later.
CHAPTER 98 STOMAS OF THE SMALL AND LARGE INTESTINE 1245
manual reduction. The application of table sugar, ionized salt
crystals, hyaluronidase, and hypertonic saline injections have
been shown to effectively diminish bowel edema osmoti-
cally.148–151 In major prolapse there is cyanotic, dusky, and
edematous protruding bowel that may totally occlude the sto-
mal orifice. This requires prompt attention, and general
anesthesia is often necessary for intestinal reduction. Tempo-
rizing measures for control of prolapse include external
devices,152,153 modified purse-string suture techniques154–156
or placement of U-shaped stitches from the lumen of the re-
duced intestine through the abdominal wall over an internal
and external pledget or clothing button.157,158 These pledgets
hold the intestine against the abdominal wall and prevent the
suture from cutting through. Stricture, if mild, may respond
to dilatation. However, if evacuation is decreased and the prox-
imal intestine begins to dilate, revision is advisable. This proce-
dure is usually possible by incising all layers around the
strictured stoma and bringing out healthy, at times dilated,
bowel. The opening should not be excessive, however, because
this might lead to prolapse. If the problem is more complex
or a parastomal hernia is present, the pathologic process is
addressed through a counterincision or laparoscopic repair.159
Retraction of an end stoma results in poor fitting of the appli-
ance and may lead to skin-level stricture and obstruction,
whereas retraction of a loop stoma interferes with proper evac-
uation and leads to filling of the distal intestinal loop with stool.
Stomal bleeding is a rare but serious complication in children
on long-term parenteral nutrition with liver dysfunction lead-
ing to portal hypertension. The varices developing between
the portal and systemic circulations at the level of the stoma
are vulnerable and may result in significant drops in hemoglobin
levels. Following the correction of a possible coagulopathy, most
of these hemorrhages are amenable to direct pressure, suture li-
gation, or application of hemostatic substances.160 Tranexamic
acid, an antifibrinolytic amino acid, has been reported to stabi-
lize formed clots and reduce rebleeding.161 Nonparietal intesti-
nal obstruction from adhesions or internal hernias can occur at
any time,111 even in the immediate postoperative period.162
Intestinal reanastomosis is also associated with a high rate of
complications, most notably wound infection, dehiscence, fis-
tula formation, and intestinal obstruction.56,62,135,136 Among
the various factors contributing to this morbidity are poor
timing, inadequate bowel preparation, technical errors, and
shortcuts such as improper choice of suture material and the ex-
cessive use of the electrocautery. Malnourished, debilitated, ane-
mic patients and those on corticosteroids are at greatest risk for
complications. These contributing factors should be corrected
before reestablishment of intestinal continuity is contemplated.
Although the need, the indications, and various techniques
for establishing stomas of the small and large intestine have
shifted in the past couple of decades, the use of stomas remains
a critical tool in the surgical care of children. Placement and take-
down of any enterostoma should be carefully planned and
considered a major procedure. Requirements include ideal oper-
ating conditions and, in a teaching setting, appropriate super-
vision. Pediatric stomas are quite different from adult stomas,
and the importance of meticulous attention to detail cannot be
overemphasized. Close follow-up is imperative to render these
interventions as beneficial and complication free as possible.
The complete reference list is available online at www.
expertconsult.com.
 clear that many such patients are no longer the subject of in-
dividual reports.
Colonic atresia is reported to occur from 1 in 1498 live
births1 to 1 in 40,000.11 Episodic cases of isolated congenital
colonic atresia are seen annually at most children’s hospitals in
the United States, corresponding to 1 in approximately 20,000
births.5,12 Colon atresia is considered the rarest atresia of the
gastrointestinal (GI) tract, comprising 1.8%13 to 15% of intes-
tinal atresias;5,10,14 these estimates exclude related malforma-
tions of the rectum, which are traditionally considered within
the spectrum of imperforate anus. There is no known gender
or racial predilection of colonic atresia or stenosis. Although a
similar abnormality is commonly inherited in an autosomal
recessive pattern in Holstein cattle,15 no familial association
is known in humans. Hereditary multiple atresias of the intes-
tines typically exclude the colon.
Colonic atresia is associated with several important anom-
alies of the musculoskeletal system (e.g., syndactyly, polydac-
tyly, clubfoot, absent radius)16; eye14; heart14; GI tract16; and
abdominal wall (e.g., gastroschisis, omphalocele, bladder or
cloacal exstrophy).17,18 Facial hemiaplasia, anophthalmia,
and cloacal extrophy,19 as well as facial asymmetry with agen-
esis of the corpus callosum, cerebellar hypoplasia, and enlarge-
CHAPTER 99 ment of the fourth ventricle, are also reported.20 Stickler
syndrome, which is a chondrodystrophy with congenital alter-
ation of type II collagen, has been reported in a patient with
colon atresia.21 In the largest relevant report, 22 of 36 infants
Atresia, Stenosis, with colonic atresia had no associated anomalies.16 Of the
remaining 14 infants in that report, 9 had midline abdominal
wall defects and 5 had jejunal atresia. Davenport reported on
and Other 118 infants with colonic atresia. Twenty-eight percent of the
lesions occurred in the ascending colon, 3% in the hepatic

Obstructions flexure, 23% in the transverse colon, 25% in the splenic flex-
ure, and 20% in the descending and sigmoid colon.22 A recent
report from China reported on three babies with upper rectal
of the Colon atresia.23
Approximately 15% to 20% of infants with colonic atresia
have proximal small intestinal atresia14,16; therefore neonates
Marjorie J. Arca and Keith T. Oldham with intestinal atresia should routinely have distal patency
including the colon evaluated in the operating room or with
preoperative contrast enema. A combination of imperforate
anus and colon atresia have been reported in six patients in
Atresia and Congenital Stenosis the English literature.24–29
Hirschsprung disease is recognized in at least 2% of patients
of the Colon with colonic atresia.30–34 Some authors have hypothesized that
------------------------------------------------------------------------------------------------------------------------------------------------
when a vascular insult occurs before retroperitoneal fixation of
Although Binninger first described colonic atresia in 1673,1 the colon at 11 weeks’ gestation, caudal migration of the myen-
the first survivor was not reported until 1922, when Gaub2 teric neurons is interrupted.31 Therefore it is imperative to rule
treated a child with atresia of the sigmoid colon with a prox- out Hirschsprung disease, typically with suction rectal biopsy,
imal diverting colostomy. In 1947 Potts3 performed a primary before reestablishing intestinal continuity in every patient with
anastomosis of the transverse colon for atresia in a newborn colonic atresia.
who survived. Although colonic atresia is uncommon, colonic
stenosis is exceedingly rare. Only 10 congenital colonic steno-
sis patients have been reported since 1966.4–10 Most cases Embryology
with congenital stenosis and atresia were managed with resec- ------------------------------------------------------------------------------------------------------------------------------------------------

tion of the involved colonic segment and end-to-end anas- The classic etiology considered responsible for intestinal atre-
tomosis, but a staged approach using a decompressive sia is in utero vascular insufficiency after organogenesis.35–37
colostomy in the proximal dilated segment was judged neces- Possible causes of vascular insult include volvulus, intussus-
sary in several patients. All the reported infants with congen- ception, embolic or thrombotic events, and incarceration or
ital colonic stenosis have survived, while mortality in atresia strangulation secondary to hernias or abdominal wall defects.
patients is generally attributed to associated anomalies. It is Maternal use of cocaine, amphetamines, nicotine, and

1247
1248 PART VII ABDOMEN
decongestants has been implicated in intestinal atresia. Focal
resorption of the compromised gut apparently occurs after is-
chemic necrosis. The spectrum of observed abnormalities is
similar to that of small intestinal atresia, in which experimen-
tal work by Louw38 and Barnard and Louw39 confirmed the
pathogenic role of in utero vascular occlusion. The classifica-
tion system of Bland-Sutton40 and Louw38 for small intestinal
atresia is also applied to the colon. Briefly, type I lesions are
intraluminal obstructive webs; type II lesions are blind prox-
imal and distal segments connected by a fibrous cord; and type
III lesions are completely separated segments of intestine, with
an intervening V-shaped mesenteric defect. Type III abnormal-
ities are the most common.14,17 Complex mesenteric defects
and multiple atresias have been described. This classification
system has limited clinical value because surgical principles
and outcome are not dictated by the anatomic categorization
of the lesion.
In the past decade, a possible genetic mechanism for
intestinal atresia including duodenal, small bowel, cecal,
and colonic has been elucidated. Fibroblast growth factor
10 (Fgf10), a protein expressed in the colon and mesenteric
vasculature sites, has been widely implicated in cell survival
and cell growth. Invalidation of the Fgf10 or its receptor,
RFgfr2b, resulted in colon atresia in mice with intact mesen-
teric vasculature.41 This finding suggests that intestinal atresia
may be caused by more than simple vascular insufficiency.
Congenital colonic stenosis is much less common than co-
lonic atresia. Seven of the eight reported cases of congenital
colonic stenosis demonstrated a segmental stricture of the co-
lon, with lengths ranging from 3 to 16.5 cm.4–6,9,10 The cause
of these stenoses has also been attributed to interruption of the
blood supply of the developing colon. A partially obstructing
intraluminal membrane has been described as the cause of
colonic stenosis as well.11,42
A B
FIGURE 99-1 Colonic atresia. A, Plain radiograph of a baby with significant intestinal dilation. Previously injected contrast that was injected per rectum
remains in a diminutive rectal segment. B, Contrast enema in the same patient shows a microcolon. On exploration the atresia was at the level of
hepatic flexure.
Clinical Presentation
------------------------------------------------------------------------------------------------------------------------------------------------
Prenatal ultrasonography typically shows an enlarged loop of
intestine or colon, leading to a presumptive diagnosis of distal
intestinal atresia.11,43 Polyhydramnios is usually not seen be-
cause the relatively distal point of obstruction allows amniotic
fluid to be absorbed in the proximal small intestine.
Clinically, an infant with colonic atresia or stenosis usually
develops a distended abdomen within 24 to 48 hours after
birth. Because the obstruction is distal and usually complete,
abdominal distention is predictably marked and progressive.
Mechanical ventilatory support may be necessary due to mas-
sive intestinal dilatation and abdominal distention. Vomiting
may be delayed in onset, with feculent vomiting as a late man-
ifestation. Infants with colonic atresia typically pass little or no
fecal matter. Rectal examination yields white mucus rather
than bile-stained meconium. In neonates, coexisting cranial,
musculoskeletal, or abdominal wall abnormalities may be
apparent on inspection.
Diagnosis
------------------------------------------------------------------------------------------------------------------------------------------------
Abdominal radiographs of infants with colonic atresia show
multiple distended intestinal loops with air-fluid levels.
Although the small intestine and colon are not readily distin-
guishable in the neonatal period, the degree of distention im-
mediately proximal to the obstruction is often more marked in
the colon than in similarly obstructed small intestine (Fig. 99-1).
In fact, the colonic obstruction may be so profound that it may
be mistaken for a pneumoperitoneum. Pneumoperitoneum is a
sign of proximal colon perforation, usually associated with ab-
dominal tenderness. This is present in approximately 10% of
 CHAPTER 99
infants with this disorder at the time of diagnosis.16 It appears
that this scenario results when a competent ileocecal valve
creates a segment of colon obstructed both proximally and dis-
tally, thereby increasing the risk for perforation.
Contrast enema using isotonic contrast agent establishes
the diagnosis of colonic atresia or stenosis. Complete obstruc-
tion is characterized by a distal “microcolon,” which is dimin-
utive from disuse but is intrinsically normal in most
circumstances. The contrast medium typically fills the lumen
of the distal colon, terminating at the point of obstruction.
Congenital stenosis is characterized by narrowing of the intes-
tine and limited proximal filling with contrast medium; prox-
imal intestinal dilatation is often seen.
A contrast enema is important in the evaluation of any ne-
onate with suspected distal intestinal atresia because the sur-
geon must identify the likely cause of obstruction, as well as
blockages that may be distal to the most proximal atresia be-
fore repair. Retrograde contrast may also differentiate between
a fixed anatomic cause and other causes of distal obstruction
such as meconium plug syndrome, Hirschsprung disease, or
neonatal small left colon syndrome.43 In patients with colonic
atresia or stenosis, no other imaging is necessary to define the
intestinal anatomy.
TREATMENT
Any infant with suspected intestinal obstruction should have
nasogastric tube decompression and resuscitation with intra-
venous fluids. Broad-spectrum antibiotics are generally ad-
ministered before the procedure. Operative management
should not be delayed because of the risks of perforation
and segmental volvulus. The most common finding in colonic
atresia is two blind ends of colon with no intervening mesen-
tery (type III). In the past, emphasis was placed on the rele-
vance of the site of atresia in surgical decision making.14
In contemporary practice, primary anastomosis is preferred
unless there is medical instability, vascular compromise, sig-
nificant mismatch between the two ends, fecal soilage, or
physiologically important comorbidities.
Laparotomy is usually undertaken through a transverse,
supraumbilical position, taking into account the position of
a possible stoma. The obstructing lesion is resected, along
with compromised or excessively dilated proximal intestine
in order to facilitate functional recovery. Preservation of intes-
tinal length is, of course, an important principle for any intes-
tinal atresia, but this can be reasonably moderated for colonic
atresia, in an effort to minimize size mismatch for the anasto-
mosis and facilitate onset of intestinal function. Patency of the
distal segment should be demonstrated unequivocally either
preoperatively or intraoperatively, as noted previously. Intesti-
nal continuity is most commonly reestablished with a single-
layer, end-oblique anastomosis.
Because of the association with Hirschsprung disease, a
suction rectal biopsy to evaluate for ganglion cells should
be performed in the operating room if primary anastomosis
is planned. Similarly, if a colostomy is deemed appropriate,
the myenteric plexus at the colostomy site should be exam-
ined at operation and a rectal suction biopsy done before
reanastomosis.
Coloplasty is usually not recommended, given the excellent
outcome of resection and anastomosis. Complex colonic atre-
sias and those associated with abdominal wall defects may
ATRESIA, STENOSIS, AND OTHER OBSTRUCTIONS OF THE COLON 1249
require other approaches such as proximal stoma formation
and staged repair of the colon.
RESULTS
Complications of colonic atresia and stenosis repair are gen-
erally the same as for other types of colonic surgery. These
include anastomotic stricture, leak, proximal segment dys-
function, adhesive small intestinal obstruction, wound infec-
tion, and stoma-related problems.14,16,44 In a report by
Powell and Raffensperger,14 15 complications occurred in
19 patients treated between 1946 and 1978, with a mortality
rate of 10.5% associated with colonic atresia. DellaVecchia
and colleagues45 reported 21 patients with colonic atresia
and stenosis presenting from 1972 to 1997; 12 of these
patients had complications including wound infection (1),
stoma prolapse (1), prolonged ileus (2), and late adhesive
bowel obstruction (8). In this more contemporary series,
there was no operative mortality and the long-term survival
rate was 100%.
Acquired Colonic Stenosis
------------------------------------------------------------------------------------------------------------------------------------------------
Various inflammatory, infectious, traumatic, and neoplastic
processes may lead to acquired colonic stenosis in infants
and children.46 Of these, neonatal necrotizing enterocolitis
(NEC) is the most common. About 11% to 35% of infants with
this disorder develop intestinal strictures, with about 70%
involving the colon.47–49a Submucosal fibrosis and varying
maturation of intestinal scar are seen on resected stenotic
specimens.50 Butter and colleagues51 performed a retrospec-
tive study of NEC patients with Bell’s stage II or III disease
and found an apparent increase in the overall stricture rate re-
quiring operative intervention, from 46% in 1990-1994 to
69% between 1995 and 1999. Subgroup analysis showed
the increase in stricture was seen in the medically treated in-
fants, with stricture formation increasing from 15% to 48%.
Interestingly, the surgically treated NEC patients had a stable
stenosis rate between the two time points, from 36% to 33%.
The sigmoid colon is the most common site of NEC-associated
stenosis, but the entire intraperitoneal colon and small intes-
tine are at risk. Several lesions may be present and are usually
partially obstructing (Fig. 99-2). Failure to advance oral feed-
ing, bilious vomiting, and abdominal distention are typical.
The diagnosis is established by contrast enema radiography.
Treatment generally consists of segmental resection and pri-
mary reconstruction. Resection with proximal diversion and
staged reconstruction is an alternative in high-risk patients.
Successful balloon catheter dilatation of focal colonic stric-
tures has been reported, but the experience is quite limited
and rarely used in contemporary practice.52,53
Other infectious and inflammatory disease states have been
associated with colonic strictures. They have been reported as
sequelae of hemolytic uremic syndrome54 and methicillin-
resistant Staphylococcus aureus enterocolitis.55 In children with
human immunodeficiency virus, angioinvasive Candida spe-
cies have been reported to cause acute colonic obstruction
and death.56 A recent report of colon stenosis details a sealed
perforation, apparently secondary to Mycobacterial infection,
in a 10-year-old boy, managed with segmental colon resection
and tuberculostatic medications.87 Chronic nonsteroidal
1250 PART VII ABDOMEN
FIGURE 99-2 Contrast enema radiograph in a 3-month-old infant with a
history of necrotizing enterocolitis, treated without surgical intervention.
He developed feeding intolerance, vomiting, and abdominal distention.
A stricture is seen in the descending colon (arrow) and complete obstruc-
tion at the splenic flexure. Segmental resection and primary anastomosis
were curative.
antiinflammatory drug use has been implicated in colonic ste-
nosis in adults,57 but thus far, there are no reports of this prob-
lem in the pediatric population.
Neoplasia may cause colonic obstruction by development of
intraluminal lesions (polyps, neurofibromas,58,59 adenocar-
cinoma1,60,61,91–93); extrinsic compression (lymphoma, retro-
peritoneal sarcomas or teratomas); or intramural infiltration
(tuberous sclerosis, neurofibromatosis).62 None of these is com-
mon. In all cases the underlying disease should be controlled, if
possible. In general, focal benign lesions are resected, with pri-
mary reanastomosis of the colon. Congenital infantile fibrosar-
coma of the colon has been reported in newborns, with
perforation as the presenting sign.64,90 Malignant and complex
benign lesions may require fecal diversion, staged procedures,
or more complex reconstruction of the colon.
Adenocarcinoma of the colon occurs rarely in young pa-
tients. A significant percentage of patients have predisposing
factors such as ulcerative colitis and familial adenomatous
polyposis.65 Obstructive pain is the presenting complaint in
97% of patients younger than 21 years with adenocarci-
noma.66 Overall survival in children is approximately 25%;
in adults, survival approaches 50%. Delay in diagnosis,67,68
advanced disease stage at presentation,69 and unfavorable his-
topathologic characteristics66 all contribute to the observed
lower survival rate in children. LaQuaglia and colleagues35
reviewed 29 pediatric patients with adenocarcinoma and con-
cluded that children are more likely to have poor-prognosis
lesions such as “signet ring” or undifferentiated lesions
(69%), compared with the moderately differentiated or carci-
noma in situ (31%) found in adults. Histopathologic charac-
teristics and lymph node involvement were the most
important predictors of childhood mortality in patients who
had resection. Patients with high-grade lesions or involved re-
gional nodes should be considered for adjuvant therapy in ad-
dition to resection.35
Functional Colonic Obstruction
------------------------------------------------------------------------------------------------------------------------------------------------
Various disease processes are associated with functional co-
lonic obstruction. These forms of colonic obstruction are dis-
cussed in detail in Chapters 101 and 102 but are noted here as
part of the differential diagnosis of colonic obstruction in in-
fants and children.
Intestinal neuronal dysplasia is a disorder of the myenteric
nervous system that has a clinical presentation similar to that
of Hirschsprung disease.70 While controversy over nomencla-
ture and treatment continues, type A intestinal neuronal dys-
plasia is characterized by decreased sympathetic innervation
of the intestine, with moderately increased parasympathetic
nerve fibers. Type B IND B has histopathologic involvement
of the submucosal ganglia, where there is an increase of gan-
glionic (nerve) cells in the ganglion with more than eight nerve
cells per ganglion. Nerve cells are smaller than normal and are
grouped into spherical giant ganglia. A minimum 20% of giant
ganglia in 30 serial sections must be found to make a diagnosis
of type B IND.71 Intestinal dysmotility is manifested by various
degrees of abdominal distention or constipation.
Ogilvie syndrome is an acute distention of the colon, often
occurring in nonambulatory and medically complex patients.
There is absence of an identifiable mechanical cause. It is
usually seen in adults but may also occur in children and
adolescents. The cecum and the right colon are frequently in-
volved. Risk factors for developing Ogilvie syndrome include
electrolyte abnormalities and the use of certain medications,
particularly narcotics, that impair intestinal motility. Disten-
tion of the cecum impedes the natural propulsive activity
of the colon. Cecal distention may cause ischemic necrosis.
Nasogastric suction, discontinuation of relevant medications,
and colonoscopic decompression are the mainstays of therapy.
Recently, the use of neostigmine (0.36 to 0.45 mg/kg), an ace-
tylcholinesterase inhibitor, has been reported in the pediatric
literature.72,89 Oral erythromycin may also have efficacy.73
Chronic idiopathic intestinal pseudo-obstruction occurs in
children and adults.74,75 Abnormalities of the intestinal
smooth muscle (myopathy), myenteric plexus (neuropathy),
or a combination of both may occur.
In neonates, a visceral myopathy known as the megacystis-
microcolon-intestinal hypoperistalsis syndrome is characterized
by an anatomically patent intestinal tract with normal gan-
glion cells but clinical evidence of obstruction secondary to
ineffective peristalsis. This syndrome is often fatal because de-
finitive medical or surgical therapy is not available.44,46,76,77,93
The meconium plug syndrome and the small left colon syn-
drome appear to be specific but related entities in the spec-
trum of functional neonatal colonic obstruction. For clarity,
they are considered separately as follows.
MECONIUM PLUG SYNDROME
Meconium plug syndrome is a relatively common cause
of functional colonic obstruction in neonates (see also
Chapter 101).78,88 Affected infants may be premature but are
generally normal otherwise. It has been suggested that the syn-
drome results from an immature myenteric nervous system
causing ineffective peristalsis79 and excessive water absorption,
thereby producing desiccated meconium plugs that obstruct
the colon. The pathogenesis is unproved but is thought to be
 CHAPTER 99 ATRESIA, STENOSIS, AND OTHER OBSTRUCTIONS OF THE COLON 1251

abnormalities need to be ruled out in a patient with meconium

plugs—Hirschsprung disease and cystic fibrosis. In a retrospec-
tive survey encompassing 13 years, only 13% of patients with
colonic meconium plugs were found to have Hirschsprung dis-
ease.80 In the same analysis, no patient had cystic fibrosis, in
contradistinction to other reports that showed up to 40% asso-
ciation between cystic fibrosis and meconium plug syndrome.
The authors concluded that patients with isolated colonic plugs
are less likely to have cystic fibrosis.36 Suction rectal biopsy may
be performed to identify Hirschsprung disease, although re-
serving this procedure for children who continue to have diffi-
culties after evacuation of the plug is a reasonable approach.
Sweat testing for cystic fibrosis may be reserved for infants with
distal small bowel meconium plugging. Neonatal genotypic
screening for the most common mutations for cystic fibrosis
has become routine in most of the United States.

SMALL LEFT COLON SYNDROME

FIGURE 99-3 Radiograph of a term infant with the meconium plug syn-
drome. Contrast enema radiograph demonstrates many intraluminal filling
defects (meconium plugs) throughout the colon (arrow).
related to temporary dysmotility. The clinical presentation is the
same as that for distal intestinal obstruction. A water-soluble
contrast enema is usually both diagnostic and therapeutic,
but repeated enemas may be necessary for resolution
(Fig. 99-3). Surgical management is rarely indicated, but rectal
examination may be curative. Classically, two associated
Small left colon syndrome refers to a dysfunctional, diminu-
tive left colon that causes transient obstructive symptoms.69,81
Approximately 40% to 50% of infants with this disorder have
diabetic mothers, almost all of whom are insulin dependent.
In a retrospective study of infants born of mothers with either
gestational or pregestational diabetes, Ellis noted a 4.7% inci-
dence of small left colon syndrome.82 Davis and Campbell81
reported that 50% of asymptomatic infants of diabetic
mothers also have radiographic evidence of a small left colon.
The obstruction is typically partial and involves the descend-
ing colon distal to the splenic flexure (Fig. 99-4). The degree

A B
FIGURE 99-4 A, Plain radiograph of an infant born to a diabetic mother illustrating a distal intestinal obstruction. B, Barium enema radiograph shows
typical findings of the small left colon syndrome. Note the distal microcolon (asterisks), the splenic flexure transition (arrows), and the dilated transverse
colon proximal to the point of obstruction (arrowheads). (Courtesy D. Frush, MD)
1252 PART VII ABDOMEN

of obstruction varies. Proximal perforation, especially of the

cecum, has been described.83 The pathogenesis is thought
to be related to hypoglycemia-induced release of glucagon
in the infant. For reasons that are unclear, the neonatal left co-
lon seems to be particularly sensitive to the smooth muscle
constrictive effects of glucagon and other agents.69 The prob-
lem is typically transient and resolves 24 to 48 hours after
birth, corresponding to normalization of the infant’s endoge-
nous insulin output. A contrast enema is diagnostic and may
be therapeutic by hastening resolution of the functional ob-
struction. The radiographic appearance of small left colon
syndrome is quite similar to that of Hirschsprung disease be-
cause of the splenic flexure transition zone, so suction rectal
biopsy is routinely performed to eliminate this possibility. It
is prudent to also rule out cystic fibrosis.

Meconium Ileus
------------------------------------------------------------------------------------------------------------------------------------------------

Meconium ileus is the intestinal obstructive entity associated

FIGURE 99-5 Contrast enema radiograph illustrates inspissated stool
in the transverse colon (arrow). The patient was an adolescent with cystic
fibrosis who had recurrent obstructive episodes that were consistent with
meconium ileus.
with cystic fibrosis and is discussed in detail in Chapter 83.
Exocrine pancreatic insufficiency and water-deficient intesti-
nal secretions are features of cystic fibrosis that combine to
create thick, viscid stools that can impact and obstruct the
small intestine or colon. This is classically seen in newborns
but may also be a problem in adolescents with cystic fibrosis,
whose compliance with exocrine enzyme replacement therapy
may be erratic. The diagnosis is often revealed by the history
and plain radiography but can be confirmed by contrast en-
ema (Fig. 99-5). The preferred treatment is colonic irrigation
in an antegrade or retrograde direction and the resumption
of pancreatic enzyme replacement. Dilute (5% to 10%) N-
acetylcysteine may be used occasionally. Chronic problems
may require gastrostomy or cecal access for formal, periodic
irrigation of the intestine.

Miscellaneous Causes
of Obstruction
------------------------------------------------------------------------------------------------------------------------------------------------

Several rare causes of colonic obstruction merit brief mention.

FIGURE 99-6 Contrast enema in a child with sigmoid volvulus. The arrow
depicts the point of obstruction.
Among these are cecal or sigmoid volvulus, a wide variety of
rare neoplastic lesions, anomalous congenital peritoneal
bands,63 hindgut duplication,84 and congenital segmental di-
latation (ectasia) of the colon.27 Segmental dilatation of the co-
lon is an idiopathic problem of infants and children and
presents as partial colonic obstruction. It is a focal, functional
defect in which the involved segment of colon lacks peristalsis
and has no taeniae coli. Because both the proximal and distal
colon are functionally and anatomically normal, surgical re-
section is curative.85
Segmental volvulus of the colon involves either the sigmoid
colon or the cecum (Fig. 99-6). The typical patient is a child or
adolescent with a long-standing history of constipation or
pseudo-obstruction. The clinical presentation is abdominal
distention, nausea, and vomiting. It may be difficult to distin-
guish from exacerbation of pseudo-obstruction in certain pa-
tients. Cecal volvulus has been reported in two children with
Cornelia de Lange syndrome.86
 CHAPTER 99
The diagnosis is established with a contrast enema.
Operative intervention, usually to resect the involved seg-
ment, is typically required. Endoscopic decompression
can be useful to decompress the involved colon and convert
this to a more elective operation, but this can be prohibi-
tively difficult.
In the mid-1990s, long-segment colonic fibrosis was
described in patients with cystic fibrosis. Colonic injury was
determined to be caused by the ingestion of high doses of
pancreatic enzymes (>19,000 units/kg/day).85
Resection of involved bowel was required in many of
these patients. Alteration of the dosage of pancreatic enzymes
has decreased the occurrence of this complication in recent
years.
ATRESIA, STENOSIS, AND OTHER OBSTRUCTIONS OF THE COLON 1253
Summary
------------------------------------------------------------------------------------------------------------------------------------------------
The causes of colonic obstruction in infants and children are
diverse. Generally, anatomic obstruction is easily addressed by
surgical resection with primary anastomosis. Functional ob-
structions have a varied pathogenesis, and a nonsurgical
approach is generally preferred. The clinical outcome of
anatomic colonic obstruction is excellent in contemporary
pediatric surgical practice; the exceptions are children with
concurrent illness or peritonitis with sepsis at presentation.
The complete reference list is available online at www.
expertconsult.com.
 the retrocolic or retrocecal position. In cases of malrotation
or situs inversus, the malpositioned appendix may give rise
to signs of inflammation in unusual locations.
The appendix averages 8 cm in length but can vary from
0.3 to 33 cm. The diameter of the appendix ranges from 5
to 10 mm. Its blood supply is the appendiceal branch of the
ileocolic artery, which passes behind the terminal ileum.
The base of the appendix arises at the junction of the three tae-
niae coli, a useful landmark in locating an elusive appendix.
Its colonic epithelium and circular and longitudinal muscle
layers are contiguous with the cecal layers. A few submucosal
lymph follicles are present at birth. These increase to approx-
imately 200 by age 12 and reduce abruptly after the age of 30.
Suppression of cecal development results in appendi-
cular hypoplasia or agenesis. Appendiceal duplication has a
reported incidence of 4 in 100,000.9,10 The duplicated appen-
dix may be partial (bifid appendix) or full, and it may have
separate or common orifices in the cecum, which may also
be duplicated.
The function of the appendix is unknown. Primates have
an appendix, but most mammals do not. Curiously, rabbits
do have an appendix and it is believed to be an important site
for the immunologic development of B cells.11
CHAPTER 100
Spectrum of Disease
------------------------------------------------------------------------------------------------------------------------------------------------

Appendicitis The geographic variation in the incidence of appendicitis

is widespread. Appendicitis occurs less frequently in less
industrialized countries.12 Over the past few decades, the world-
James C. Y. Dunn wide incidence has steadily decreased.12–15 In the United States
over 70,000 children are diagnosed with appendicitis annually,
or approximately 1 per 1000 children per year. The lifetime risk
for appendicitis is 9% for men and 7% for women.16 About one
third of patients with appendicitis are younger than 18 years of
In 1886 Fitz1 coined the term appendicitis. Morton is credited age. Appendicitis occurs more commonly in whites and during
with performing the first deliberate appendectomy for a the summer months. The peak incidence occurs between ages
perforated appendix in the United States in 1887.2,3 In 1889 11 and 12. Although the disorder is uncommon in infants,
McBurney4 reported his treatment of appendicitis with appen- perforated appendicitis can occur even in premature infants.
dectomy before rupture and described “the seat of greatest Perforation may also be the end result of another disease process,
pain . . . has been very exactly between an inch and a half and as is seen in neonates with Hirschsprung disease.17 Although
two inches from the anterior spinous process of the ilium on a the diagnosis and treatment have improved, appendicitis con-
straight line drawn from the process to the umbilicus.” From then tinues to cause significant morbidity and still remains, although
on, this location was known as the McBurney point. Modern sur- rarely, a cause of death.18,19
gical care and antibiotic therapy have turned this once frequently Many terms have been used to describe the varying stages
fatal disease rarely so today. Appendicitis remains the most of appendicitis including acute appendicitis, suppurative
common acute surgical condition of the abdomen5; many aspects appendicitis, gangrenous appendicitis, and perforated appen-
of the treatment of appendicitis remain controversial. dicitis. These distinctions are imprecise, with considerable
variability among surgeons and among surgeons and pathol-
ogists. Clinically the only relevant distinction is between
Embryology and Anatomy simple and complicated appendicitis. In general, acute and
------------------------------------------------------------------------------------------------------------------------------------------------
suppurative appendicitis tend to have a simple clinical course,
During embryogenesis, the appendix first becomes visible whereas gangrenous and perforated appendicitis tend to have
during the eighth week of gestation as a continuation of the a complicated course.
inferior tip of the cecum.6,7 The appendix rotates to its final Whether chronic and recurrent cases of appendicitis exist
position on the posteromedial aspect of the cecum, about has been debated for decades. Literature contends that they do
2 cm below the ileocecal valve, during late childhood. The var- exist and should be considered in the differential diagnosis of
iability in this rotation leads to multiple possible final posi- recurrent lower abdominal pain.20,21 Inflammation of the
tions of the appendix. The appendix is intraperitoneal in appendix does not inevitably lead to perforation because spon-
95% of cases, but the exact location varies widely.8 In 30% taneous resolution does occur.22,23 Antibiotics may also assist
of cases the tip of the appendix is in the pelvis, in 65% it is the resolution of the inflammation. Recognition of this may
behind the cecum, and in 5% it is truly extraperitoneal in contribute to the decreasing number of appendectomies.14,15
1255
1256 PART VII ABDOMEN

Presentation
------------------------------------------------------------------------------------------------------------------------------------------------
obstruction, which leads to tissue ischemia, infarction, and
gangrene. Bacterial invasion of the wall of the appendix then
Traditional teaching is that appendicitis evolves as a con- occurs. Fever, tachycardia, and leukocytosis develop as a
tinuum from simple inflammation to perforation, typically consequence of mediators released by ischemic tissues, white
occurring after 24 to 36 hours of symptoms, and subsequent blood cells, and bacteria. When the inflammatory exudate
abscess formation occurring over a period of 2 to 3 days.24,25 from the appendiceal wall contacts the parietal peritoneum,
Nevertheless, the variability of the clinical presentation of somatic pain fibers are triggered and the pain localizes near
appendicitis leads to laparotomies that do not reveal an the appendiceal site, most typically at the McBurney point.
inflamed appendix. Clinical experience and advances in Pain occasionally occurs only in the right lower quadrant
imaging methods have improved the diagnostic accuracy without the early visceral component. With a retrocecal or
but are not foolproof. The clinical presentation of appendicitis pelvic appendix, this somatic pain is often delayed in onset
can be understood in terms of its pathophysiology. because the inflammatory exudate does not contact the pari-
Appendicitis results from luminal obstruction followed by etal peritoneum until rupture occurs and infection spreads.
infection. This process was first described by van Zwalenberg26 Pain of a retrocecal appendix may be in the flank or back.
in 1905 and experimentally confirmed by Wangensteen27 in A pelvic appendix resting near the ureter or testicular vessels
1939. Wangensteen showed that the human appendix can cause urinary frequency, testicular pain, or both. Inflam-
continues to secrete mucus even when intraluminal pressures mation of the ureter or bladder by an inflamed appendix
exceed 93 mm Hg. Although it is clear that luminal obstruction can also lead to pain on micturition or the deceptive pain of
causes appendicitis, the cause of the obstruction is not always a distended bladder secondary to urinary retention.
clear. Inspissated and sometimes calcified fecal matter, known Further breakdown of the appendiceal wall leads to per-
as a fecalith, often plays a role.28 Fecaliths can be surgically foration with spillage of infected intraluminal contents with
found in approximately 20% of children with acute appendici- localized abscess formation or generalized peritonitis. This
tis and are reported in 30% to 40% of children with perforated process depends on the rapidity of progression to perforation
appendicitis.29,30 The presence of fecaliths can often be and on the patient’s ability to mount a response and contain
documented radiographically. Hyperplasia of appendiceal the spilled contents of the appendix. Signs of perforated
lymphoid follicles frequently causes luminal obstruction, appendicitis include a temperature higher than 38.6 C,
and the incidence of appendicitis closely parallels the amount leukocyte count greater than 14,000,46 and the presence of
of lymphoid tissue present.31 Causes of local or generalized more generalized peritoneal signs. Other reported risk factors
reaction of lymphatic tissue such as Yersinia, Salmonella, and include the male sex, extremes of age, and such anatomic fac-
Shigella32–35 can lead to luminal obstruction of the appendix, tors as a retrocecal position of the appendix.25,47 However,
as can parasitic infestations by Entamoeba, Strongyloides, perforated and nonperforated appendicitis may be entirely
Enterobius vermicularis, Schistosoma, or Ascaris species.36–38 separate entities.13 Spontaneous resolution of appendicitis
Enteric and systemic viral infections such as measles, chicken does occur. Patients may be asymptomatic before perforation,
pox, and cytomegalovirus may also cause appendicitis.39,40 and symptoms may be present for longer than 48 hours with-
Patients with cystic fibrosis have an increased incidence of out perforation. In general, however, the longer duration of
appendicitis, which presumably results from alterations in symptoms is associated with a greater risk for perforation.
the mucous-secreting glands.41,42 Carcinoid tumors can ob- Constipation is unusual, but the sensation of rectal fullness
struct the appendix, especially when they are located in the or tenesmus is common. Diarrhea occurs more frequently in
proximal third. Foreign bodies such as pins, vegetable seeds, children than in adults and can result in a misdiagnosis of
and cherry stones have been implicated as causes of appendi- gastroenteritis. Diarrhea is typically of short duration and
citis for more than 200 years. Trauma has also been reported as often results from irritation of the terminal ileum or cecum;
a cause,43 as has psychologic stress44 and heredity.45 however, it may indicate a pelvic abscess.
Initially the patient may describe mild gastrointestinal Younger children typically present with complicated appen-
symptoms before the onset of pain (e.g., decreased appetite, dicitis due to their inability to give an accurate history and phy-
indigestion, or subtle changes in bowel habits). Anorexia is sicians’ low index of suspicion that leads to misdiagnosis.48–50
a helpful sign, particularly in children, because a hungry child The most frequent presenting symptom in preschool children
rarely has appendicitis. Any severe gastrointestinal symptoms is vomiting, followed by fever and abdominal pain.51 Perfora-
before the onset of pain, however, should suggest an alterna- tion is almost always the finding at laparotomy, and these
tive diagnosis. Distention of the appendix results in activation children may have associated small bowel obstruction second-
of its visceral pain fibers. Typical early visceral pain is non- ary to extensive inflammation in the terminal ileum and cecum.
specific in the periumbilical region. This initial pain is poorly
localized as a deep, dull pain in the T-10 dermatome. The con-
tinued distention of the appendiceal wall elicits nausea and
vomiting, which typically follow the onset of pain within a
Diagnosis
------------------------------------------------------------------------------------------------------------------------------------------------

few hours. Nausea is common, but vomiting is typically not

PHYSICAL EXAMINATION
severe. The appearance of these symptoms before the onset
of pain casts doubt on the diagnosis. As with most disease processes, much can be learned before
The obstructed appendix is a perfect breeding ground the patient is touched. Children with appendicitis usually
for trapped bacteria. As intraluminal pressure increases, lie in bed with minimal movement. A squirming, screaming
lymphatic drainage is inhibited, leading to further edema child rarely has appendicitis. An exception to this is the child
and swelling. Finally, the increase in pressure causes venous with retrocecal appendicitis and subsequent irritation of the

ureter presenting with pain similar to renal colic. Older chil-
dren may limp or flex the trunk, whereas infants may flex the
right leg over the abdomen. A recall of localized pain elicited
by bumps in the road on the ride to the hospital is helpful.
Before starting palpation of the abdomen, it is useful to ask
the child to point with one finger to the location of the abdom-
inal pain. With the knees bent to relax the abdominal muscles,
gentle palpation of the abdomen should begin at a point away
from the location of perceived pain. Palpating the abdomen in
an area remote from the site of pain may elicit tenderness in
the right lower quadrant (Rovsing sign of referred pain), indi-
cating peritoneal irritation. Younger children may be more
cooperative if their hand or the stethoscope is used for palpa-
tion. The stethoscope can have several roles in the evaluation
of a patient who potentially has appendicitis, the least impor-
tant of which is auscultation. Although patients often have
diminished or absent bowel sounds, this is not uniform and
auscultation of the abdomen is of little benefit. However, aus-
cultation of the chest to examine for lower respiratory infec-
tion is useful because right lower lobe pneumonia can
mimic appendicitis. Cutaneous hyperesthesia, a sensation
derived from the T10 to L1 nerve roots, is often an early
although inconsistent sign of appendicitis. Lightly touching
the patient with the stethoscope creates this uncomfortable
sensation.
Localized tenderness is essential for diagnosis and is noted
either on palpation or percussion. Tenderness can be mild and
even masked by more generalized abdominal pain, especially
during initial stages. The McBurney point is the most common
location. Retrocecal appendicitis may be detected by tender-
ness midway between the twelfth rib and the posterior supe-
rior iliac spine. Pelvic appendicitis produces rectal tenderness.
A child with malrotation will have localized tenderness that
corresponds to the position of the exudative drainage from
the inflamed appendix.
As the disease progresses to perforation, peritonitis ensues.
The pattern of pain depends on the location of the appendix.
Perforation may result in temporary relief of symptoms as the
pain of the distended viscus is relieved. Initially, peritonitis is
reflected as local muscular rigidity. This progresses from
simple involuntary guarding to generalized rigidity of the
abdomen. Other signs include rigidity of the psoas muscle
(demonstrated by right hip extension or raising the straight
leg against resistance) or of the obturator muscle (demon-
strated by passive internal rotation of the right thigh), both
of which indicate irritation of these muscles due to retrocecal
appendicitis. Other tests of peritoneal inflammation such as
rebound tenderness are seldom necessary for diagnosis and
cause unnecessary discomfort.
The routine use of rectal examination in the diagnosis of
appendicitis has recently been questioned.52–54 Pain during
this examination is nonspecific for appendicitis. If other signs
point to appendicitis, the rectal examination is unnecessary.
However, it may be a helpful diagnostic maneuver in question-
able cases such as when a pelvic appendix or abscess is sus-
pected or when uterine or adnexal pathologic conditions
are being considered.
If appendicitis is allowed to progress, two results are
possible: (1) diffuse peritonitis and shock will occur or
(2) the infection will become isolated and an abscess will form.
Diffuse peritonitis is more common in infants, probably be-
cause of the absence of omental fat. Older children and
CHAPTER 100 APPENDICITIS 1257
teenagers are more likely to have an organized abscess. The
physical examination in cases of an organized abscess reveals
a boggy, tender mass over the abscess.
A frequently unreported but critical aspect of the evalua-
tion is serial examinations done by the same person. The safety
and efficacy of serial observation was first reported by White
in 197555 and has since been reinforced by other studies.
Surana56 reported a prospective study showing no increase
in morbidity with appendectomy after active observation in
a hospital compared with urgent appendectomy. When the
diagnosis is unclear, serial abdominal examinations permit
the physician to decrease the number of unnecessary laparot-
omies without increased risk to the patient.
LABORATORY STUDIES
Much has been discussed concerning the laboratory findings
of appendicitis. Total leukocyte and neutrophil counts have
been extensively investigated.57–59 The sensitivity of an
elevated leukocyte count ranges from 52% to 96% and that
of a left-shifted neutrophil count ranges from 39% to 96%.
The latter is of better diagnostic value, but misinterpretation
of the values is still common. Normal leukocyte count occurs
in 5% of patients with appendicitis. Greater specificity and
sensitivity have been reported using a neutrophil-lymphocyte
ratio greater than 3.5.60
In the majority of children with suspected appendicitis, the
combination of clinical history, physical findings, and labo-
ratory studies should provide sufficient data for making the
diagnosis. Nevertheless, misdiagnosis leading to negative ap-
pendectomy ranging from 10% to 30% has been reported.61
An appendicitis score based on weighing eight clinical factors
(localized right lower quadrant tenderness, leukocytosis, pain
migration, left shift, fever, nausea-vomiting, anorexia, perito-
neal irritation) was proposed to improve the diagnostic accu-
racy.62,63 In prospective evaluations of children with acute
abdominal pain, the sensitivity of the scoring system ranged
from 76% to 100% and its specificity ranged from 79% to
87%.64,61 In cases where the diagnosis is equivocal, serial
observation is warranted and imaging studies may be useful.
IMAGING STUDIES
Imaging studies have variable success in improving diagnostic
accuracy. Plain radiography can be helpful. Fecaliths are
present in 10% to 20% of patients and may be an indication
for surgery when symptoms are present. An abnormal gas
pattern in the right lower quadrant, lumbar scoliosis away
from the right lower quadrant, and obliteration of the psoas
shadow or fat stripe on the right are also helpful. A chest
radiograph to rule out pneumonia may be indicated.
A barium enema contrast radiograph may show absent
or incomplete filling of the appendix, irregularities of the
appendiceal lumen, and an extrinsic mass effect on the cecum
or terminal ileum. The sensitivity and specificity of this tech-
nique are low,65 and it is best used in the diagnosis of non-
specific abdominal pain.
In skilled hands, ultrasonography has proven to be an
effective diagnostic aid. A prospective study showed that
ultrasonography was more accurate than the surgeon’s initial
clinical impression.66 Most studies demonstrate a sensitivity
greater than 85% and a specificity greater than 90%.67
1258 PART VII ABDOMEN

Demonstration of a noncompressible appendix that is 7 mm TABLE 100-1

or larger in anteroposterior diameter is the primary criterion Differential Diagnosis of Acute Appendicitis
for the diagnosis. The presence of an appendicolith is helpful. Appendix
Such techniques as graded compression, self-localization,68 Appendiceal tumor, carcinoid tumor
and transvaginal or transrectal ultrasound approaches69 have Appendiceal mucocele
also improved results. Crohn disease
Computed tomography (CT) has become more widely
used in the diagnosis of appendicitis.70–73 The findings of Cecum and Colon
an enlarged appendix (>6 mm), appendiceal wall thickening Cecal carcinoma
(>1 mm), periappendiceal fat stranding, and appendiceal Diverticulitis
wall enhancement are useful diagnostic criteria.74,75 The sen- Crohn disease
sitivity of CT scans is over 90%, and its specificity is over Intestinal obstruction
80%.76 Comparisons of ultrasonography and CT have shown Stercoral ulcer
that the latter is more sensitive, whereas the former is more Typhlitis (leukemic, amebic)
specific.67,70,77 These two imaging modalities, however, Hepatobiliary
should be employed only if the diagnosis is uncertain. In a Cholecystitis
protocol that evaluated children with equivocal clinical find- Hepatitis
ings for appendicitis, the combination of pelvic ultrasound Cholangitis
followed by limited CT with rectal contrast, if necessary, Small Intestine
yielded a sensitivity of 94% and a specificity of 94%.78 The Adenitis
same protocol reduced the negative appendectomy rate from Duodenal ulcer
12% to 6% at the same institution during the study period but Gastroenteritis
performed imaging in almost 80% of children with suspected Intestinal obstruction
appendicitis. The perceived improved diagnostic accuracy led Intussusception
to a dramatic increase in the number of CT performed in the Meckel diverticulitis
pediatric population,79–81 even though there is no good Tuberculosis
evidence that supports the routine use of CT in the diagnosis Typhoid (ulcer perforation)
of appendicitis.82 In addition to the hospital resource utiliza-
Urinary Tract
tion and the delay in surgical care, the potential cancer risk
Hydronephrosis
associated with ionizing radiation from CT should be
Pyelonephritis
considered.79
Ureteral or renal calculus
Wilms’ tumor
DIFFERENTIAL DIAGNOSES Uterus, Ovary
Acute appendicitis can mimic virtually any intra-abdominal Ectopic pregnancy
process.83 The differential diagnosis of appendicitis is listed Ovarian torsion
in Table 100-1. Consideration of these other disease processes Ruptured ovarian cyst
before surgery is as important as it is to examine for them Salpingitis
carefully when the patient is under anesthesia and during Tubo-ovarian abscess
surgery in the case of a normal appendix. Other
The clinical diagnosis of appendicitis is challenging be- Cytomegalovirus infection
cause many symptoms of appendicitis are nonspecific and Diabetic ketoacidosis
the presentations can be variable. Acute gastroenteritis is a Schönlein-Henoch purpura
common cause of abdominal pain in children. It is usually Kawasaki disease
due to a viral illness and is self-limited. The symptoms include Burkitt lymphoma
watery diarrhea, crampy abdominal pain, fever, nausea, and Omental torsion
vomiting. Constipation is another common pediatric problem Rectus sheath hematoma
and may cause abdominal pain, nausea, and vomiting. Pain is Pancreatitis
usually persistent but not progressive. History and a plain Parasitic infection
radiograph will suggest the diagnosis. Urinary tract infection Pleuritis
will also cause fever, nausea, and vomiting. A urinalysis Pneumonia
should be obtained if urinary symptoms are present. Porphyria
Despite advances in diagnostic imaging, operation done for Psoas abscess
appendicitis does not always reveal an inflamed appendix. Sickle cell disease
Formerly accepted rates of laparotomy that did not reveal Torsion of appendix epiploica
appendicitis range from 15% to 40%.25 These rates are not
supported by the recent literature that report negative appen-
dectomy rates to be less than 10%.84–87 When a normal quadrant should be performed to look for other causes of the
appendix is encountered, most surgeons recommend that it symptoms. The terminal ileum may demonstrate mesenteric
be removed to allow for pathologic examination and to avoid adenitis, enlarged lymph nodes in the ileal mesentery that
potential confusion if the patient experiences right lower may be secondary to an upper respiratory infection. These pa-
quadrant pain in the future. An exploration of the right lower tients may have abdominal pain, fever, and nausea, but their
 CHAPTER 100 APPENDICITIS 1259

tenderness is not as well localized as that in appendicitis. Lym- effectiveness of other antibiotic combinations are usually mea-
phocytosis may be noted on the blood count differential. sured against this empiric regimen. More recently, it has been
A search for a Meckel diverticulum should be done, but it shown that a single or double broad-spectrum antibiotic is
rarely causes pain. Painless bleeding and obstruction are the equally effective for the treatment of complicated appendicitis.
more common presenting symptoms. If the patient has Crohn There is a trend toward decreasing the duration of antibi-
disease, the appendix should not be removed if it or the cecum otic therapy.99 Only perioperative antibiotics are required for
is involved in the disease process because removal is associ- simple appendicitis. The recommended duration is from a sin-
ated with a high incidence of subsequent fistula formation.88 gle, preoperative dose to 24 hours of postoperative antibiotic
The diagnostic accuracy for appendicitis is lowest among therapy for simple appendicitis.100,101 For complicated ap-
young women because of the variety of gynecologic condi- pendicitis, recent studies have suggested that as little as 48
tions that can cause low abdominal pain. Ectopic pregnancy hours of coverage is adequate.69 Others suggest that treatment
should be considered in teenage girls with low abdominal be continued as clinically indicated using the leukocyte count
pain. They may present with vaginal bleeding, amenorrhea, and presence of fever as guides.102–104 There is also a trend to
dizziness, nausea, and vomiting. Rupture of ovarian cysts use oral antibiotics instead of intravenous antibiotics when gas-
and ovarian torsion may also present with low abdominal trointestinal function returns. A prospective, randomized
pain. Sexually active girls with pelvic inflammatory disease study demonstrates equivalency between a 10-day course of
may present with low abdominal pain, vaginal discharge, intravenous antibiotics and a 10-day course of combined intra-
and adnexal enlargement. Most will have cervical motion venous and oral antibiotics for complicated appendicitis.105
tenderness and will respond to antibiotics. Operative inter-
vention may be indicated for those who do not respond or
APPENDECTOMY
persistent abscess.
Carcinoid tumors are present in less than 1% of patients The most widely accepted treatment of appendicitis is appen-
undergoing appendectomy.89 Most appendiceal carcinoid tu- dectomy. Randomized trials that compared medical therapy
mors lack the serotonin-containing cells that are typical of with appendectomy in adults with appendicitis showed that
midgut carcinoid tumors, so they are rarely symptomatic medical therapy is associated with a 10% to 20% chance of
and typically present incidentally at appendectomy.90 Most recurrence but has lower rates of complications. There is a
are benign, and simple appendectomy is curative.91,92 Con- trend away from performing immediate operation.106,107
troversy surrounds the proper surgical management of poten- There was no increased rate of complications noted between
tially malignant carcinoid tumors. The consensus is that a group of patients diagnosed with acute appendicitis and op-
carcinoid tumors larger than 2 cm in diameter, those that have erated upon within 6 hours of admission and those with de-
obviously metastasized, and those located at the base of the lays between 6 and 18 hours of admission.55,56 Nevertheless,
appendix require right hemicolectomy,90,93 whereas those the majority of pediatric surgeons will perform appendectomy
that are smaller than 1 cm in diameter and have not metasta- within 8 hours.94
sized at the time of diagnosis are treated by appendectomy In the open technique, a transverse or oblique right lower
alone. Treatment of tumors that are 1 to 2 cm in diameter quadrant incision is made through the McBurney point
remains controversial. Moertel91 believes that a conservative (Fig. 100-1). The muscles of the abdominal wall are usually
surgical procedure is all that is required regardless of tumor split. After the abdomen is entered, the cecum and appendix
size or location as long as metastases are not present. are mobilized and the appendix is brought out through the in-
cision. The mesoappendix is then divided, and the base of the
appendix is ligated. A short base is left to avoid inflammation
Treatment in the stump.108 The stump is managed by simple ligation,
------------------------------------------------------------------------------------------------------------------------------------------------
ligation with inversion using a purse-string or Z-stitch suture,
Although it is generally agreed that the treatment for appendi- or inversion without ligature. Simple ligation can be done
citis is appendectomy, the details of the management vary quickly and may reduce adhesions.109 Inversion theoretically
considerably.94 For example, surgical techniques such as the leads to better hemorrhage control, a doubly secure closure,
laparoscopic approach, the use of drains, the necessity of peri- and less chance of contamination; however, it can create arti-
toneal irrigation, the handling of the appendiceal stump, and facts on future contrast examinations and can cause intussus-
the closure of the incision continue to be debated. The need ception.110 For simple appendicitis, irrigating the wound is
for interval appendectomy after initial nonoperative manage- unnecessary. The wound is closed in layers, and no drains
ment of an appendiceal phlegmon is unclear. The choice of are placed. A normal diet can be given soon after appendec-
antibiotics and the length of its use vary considerably from tomy, and the patient can be discharged in 1 to 2 days. If a nor-
surgeon to surgeon. mal appendix is found, the peritoneal cavity should be
inspected for inflammatory bowel disease, mesenteric adeni-
ANTIBIOTICS tis, Meckel diverticulitis, or, in females, pathologic conditions
of the ovary.
The use of antibiotics for the treatment of appendicitis is “Endoscopic” appendectomy was first described in
clearly beneficial.95 Intraoperative cultures have not been 1983.111 Laparoscopic appendectomy can be done by a lapa-
shown to alter the treatment outcome.96,97 The best regimen roscopic-assisted technique in which the appendix is mobi-
and duration of antibiotics use is a subject of continued lized laparoscopically using one or two ports and is drawn
controversy. A 10-day course of intravenous ampicillin, genta- through a small abdominal opening and removed by standard
micin, and clindamycin or metronidazole is the gold standard open technique.112,113 Alternatively, the appendix can be
for the treatment of complicated appendicitis,98 and the removed entirely laparoscopically. Three trocars are usually
1260 PART VII ABDOMEN

Mesoappendix

Exterior oblique
Rectus

D Inflamed appendix

Cecal
taenia
Interior oblique
muscle Transverse
C abdominal muscle
Exterior oblique
B fascia

Crushing clamps
FIGURE 100-1 A, A transverse in- applied
cision is made in the right lower
quadrant over the lateral muscula-
E
ture. B, The external oblique fascia
is incised exposing the internal obli-
que fascia and muscle. C, The trans-
verse abdominal muscle and
peritoneum are opened, and the ce-
cum is identified. D, An inflamed ap-
pendix is identified, and the
mesoappendix is isolated, clamped,
divided, and tied. E, A purse-string
suture is placed in the cecal wall.
F, The base of the appendix is
crushed and tied, and the appendix
is excised. G, The appendiceal F
stump is inverted into the cecal wall,
and the purse-string suture is tied.
(From Rowe M, O’Neill JS, Grosfeld
JL: Essentials of Pediatric Surgery, G
Philadelphia, Mosby, 1995.)
 CHAPTER 100 APPENDICITIS 1261

employed: one at the umbilicus for the scope, one in the
suprapubic area, and one in the left lower quadrant
(Fig. 100-2), although a single-incision multiport approach
may also be employed. The appendix is found by following
the cecum, and the mesoappendix is grasped near the tip to
lift the appendix toward the abdominal wall. A window is
made in the mesoappendix near the base to allow its division
by applying electrocautery, clips, staples, or the harmonic scal-
pel. Many variations of ligating and removing the appendix
have been described.85,114 The simplest technique applies
an endoscopic stapler to the base of the appendix, and the
appendix is delivered through the umbilical trocar site. There
is now early experience with single port/incision techniques.
Despite prospective, randomized trials that compared open
and laparoscopic appendectomy, the advantages and dis-
advantages of laparoscopic appendectomy continue to be
debated.115–124 Advantages claimed include shorter hospital-
izations, decreased postoperative pain, decreased wound
complications, increased ability to diagnose uncertain cases,
surgical ease in an obese patient, and faster postoperative
recovery.84,85,125–129 Disadvantages are a higher cost because
of equipment needs and longer time for surgery, increased
training and experience required for surgeons and ancillary
support staff, increased incidence of finding a normal
appendix, and an increased incidence of intra-abdominal
infection.84,130–133 Although the conclusions regarding the

FIGURE 100-2 A, Three ports are placed for laparoscopic appendectomy. The umbilical port is 12 mm in size to accommodate the endoscopic stapler.
The other two ports are 3 or 5 mm in size for dissecting instruments. B, The appendix is lifted upwards by a grasper placed on the mesoappendix, and a
window is made at the base of the mesoappendix by a dissector. The mesoappendix is divided by electrocautery or harmonic scalpel. C, The scope is
switched to the left lower quadrant port to allow the endoscopic stapler to come through the umbilical port to divide the appendix.
1262 PART VII ABDOMEN
advantages of this technique over the open technique vary
widely, laparoscopic appendectomy is a safe and effective
means of performing an appendectomy and its utilization
has increased dramatically over the past decade.
Treatment of patients with complicated appendicitis is
more controversial. Due to social, cultural, economic, and
medical influences on the diagnosis and treatment of this dis-
ease process, perforation rates vary from 16% to 57% in dif-
ferent institutions. There is no consensus on the optimal
treatment of patients with complicated appendicitis. Opinions
range from nonoperative treatment to aggressive surgical
resection with antibiotic irrigation, drainage of the peritoneal
cavity, and delayed wound closure.134–138 Weiner139 reported
no significant differences in the number of hospital days, cost
of treatment, or overall complication rates using initial non-
operative treatment of complicated appendicitis followed by
interval appendectomy in 8 weeks. In another study that ex-
amined initial nonoperative therapy for complicated appendi-
citis confirmed by imaging, 22% of the patients were
converted to appendectomy because of small bowel obstruc-
tion.140 Operative treatment remains the standard approach
because of the difficulty in determining whether perforation
has occurred before exploration.
The operative procedure for complicated appendicitis is
appendectomy. Controversy continues regarding the details
of the procedure: whether to drain the peritoneal cavity,
whether to close the wound or leave it open with delayed
closure, whether to irrigate the peritoneal cavity and, if so,
whether to use antibiotic solutions. Drains have been de-
scribed as both increasing infectious complications and pre-
venting them.29,30,46,141–143 Most studies do not support
the use of drains, with the possible exception of retrocecal ab-
scesses that cannot be properly debrided. Delayed wound clo-
sure is not supported by the literature46,85–87,98 and does not
seem to be warranted because the wound infection rate asso-
ciated with appendectomy is less than 3%. Irrigation remains
controversial. Putnam, Gagliano, and Emmons87 suggest that
irrigation prolongs ileus and may cause small intestinal
obstruction and report excellent results without irrigation.
Other recent studies support saline irrigation of the peritoneal
cavity with or without antibiotics.29,46,85,86,98
Management of patients with a palpable abdominal mass is
another controversial topic. It occurs in a small but significant
fraction of patients with complicated appendicitis, especially
in young children after perforation. Some advocate immediate
appendectomy,144 whereas others perform the procedure only
if a mass is confirmed with the patient under anesthesia. If an
operation is done, care should be taken to avoid damage to
adjacent structures subject to inflammatory processes such
as the small intestine, the fallopian tubes and ovaries, and
the ureter. Surana145 and Nitecki146 recommend treatment
with intravenous antibiotics until the leukocyte count is
normal and the patient remains afebrile for 24 hours. If the
patient’s condition worsens or the mass enlarges on serial ul-
trasonography, the mass is drained percutaneously, followed
by interval appendectomy. Interval appendectomy prevents
repeated episodes of appendicitis and affords the surgeon
the opportunity to evaluate the patient for other conditions
that can masquerade as an appendiceal mass. Whether an in-
terval appendectomy is necessary is also debated.146–152
Nitecki116 has suggested that interval appendectomy is unnec-
essary because only 14% of patients have recurrent symptoms,
and recurrence within 2 years after initial diagnosis is uncom-
mon. The current standard of treatment is conservative man-
agement with interval appendectomy after 8 to 12 weeks.
Complications
------------------------------------------------------------------------------------------------------------------------------------------------
The incidence of complications increases with the degree of
severity of the appendicitis. The complications include wound
infection, intra-abdominal abscess formation, postoperative
intestinal obstruction, prolonged ileus, and rarely entero-
cutaneous fistula. Wound infection is the most common
complication, but the rate has fallen from 50% to less than
5%, even in complicated appendicitis.29,30,47,85–87 Intra-
abdominal abscess formation is also more common in compli-
cated appendicitis but is still less than 2%. The abscess can be
drained percutaneously under CT guidance or transrectally in
the operating room, although others have advocated more
conservative management.153–155 Postoperative intestinal
obstruction occurs in 1% of patients with complicated
appendicitis, which often requires operative adhesiolysis.156
Enterocutaneous fistula is a rare complication and will usually
respond to nonoperative management. Suppurative pylephle-
bitis is a particularly serious, although rare, complication.157
Sepsis and multisystem organ failure can occur in young
children who had prolonged illness before diagnosis. Major
complications including postoperative intestinal obstruction
and intra-abdominal abscess formation have also fallen to
an incidence of less than 5%.
An unresolved issue is the effect of complicated appendi-
citis on fertility in females; available studies contradict each
other. Puri, McGuinness, and Guiney158 report that compli-
cated appendicitis before puberty plays little if any role in
the cause of tubal infertility, whereas Mueller159 reports that
the condition is associated with a fourfold risk for tubal
infertility. The consequences of complicated appendicitis
may be mitigated through both public and medical education
that ensures prompt, early treatment before perforation.
Outcomes
------------------------------------------------------------------------------------------------------------------------------------------------
The mortality rate for complicated appendicitis has dropped
to nearly zero. Antibiotics have markedly decreased the
incidence of infectious complications. Although the length
of hospitalization and the morbidity of patients with compli-
cated appendicitis still far exceed those with simple appendi-
citis, the overall morbidity in children with complicated
appendicitis is less than 10%.
The widely varied postoperative management of appen-
dicitis is beginning to be addressed by implementing evi-
dence-based clinical pathways.160,161 Prompted primarily
by economic pressures, there is increasing scrutiny of patient
treatment and outcome.162–164 Early outcome research has
shown that hospitals that perform fewer than one appendec-
tomy per week are associated with higher likelihood of
misdiagnosis.165 There are also reports that suggest better
outcome in younger children with appendicitis when they
are cared for by pediatric surgeons.166,167 The combination
of surgical evaluation, prompt operation when the diagnosis
is clear, a period of observation if the diagnosis is equivocal
followed by imaging if necessary, and care provided by

experienced clinicians and institutions will lead to the best
outcome for children with appendicitis.
The complete reference list is available online at www.
expertconsult.com.
SUGGESTED READINGS
Eriksson S, Granström L. Randomized controlled trial of appendicectomy
versus antibiotic therapy for acute appendicitis. Br J Surg 1995;82:166.
Fraser JD, Aquayo P, Sharp SW, et al. Accuracy of computed tomography
in predicting appendiceal perforation. J Pediatr Surg 2010;45:231.
Hernanz-Schulman M. CT and US in the diagnosis of appendicitis:
An argument for CT. Radiology 2010;255:3.
CHAPTER 100 APPENDICITIS 1263
Jen HC, Shew SB. Laparoscopic versus open appendectomy in children: Out-
comes comparison based on a statewide analysis. J Surg Res 2010;161:13.
Mason RJ. Surgery for appendicitis: Is it necessary? Surg Infect (Larchmt)
2008;9:481.
Ponsky TA, Hafi M, Heiss K, et al. Interobserver variation in the assessment of
appendiceal perforation. J Laparoendosc Adv Surg Tech A 2009;19:S15.
Solomkin JS, Mazuski JA, Bradley JS, et al. Diagnosis and management of com-
plicated intra-abdominal infection in adults and children: Guidelines by the
Surgical Infection Society and the Infectious Diseases Society of America.
Surg Infect (Larchmt) 2010;11:79.
St Peter SD, Aquayo P, Fraser JD, et al. Initial laparoscopic appendectomy ver-
sus initial nonoperative management and interval appendectomy for perfo-
rated appendicitis with abscess: A prospective, randomized trial. J Pediatr
Surg 2010;45:236.
Strouse PJ. Pediatric appendicitis: an argument for US. Radiology 2010;255:8.
 ultimately bore his name.2 Until the beginning of the twenti-
eth century, most children with congenital megacolon died,
presumably from malnutrition and enterocolitis. Because
the underlying pathologic abnormality was not recognized,
surgeons who operated on these children usually resected
the dilated proximal bowel with or without primary anasto-
mosis, with mixed results.3
The absence of ganglion cells in the distal colon of a child
with Hirschsprung disease was first noted by Tittel in 1901.
Over the following decades numerous papers were published
to document abnormalities of innervation within the colon
and recognize the absence of ganglion cells that is now patho-
gnomonic of Hirschsprung disease. The first surgical recogni-
tion of aganglionosis as the cause of congenital megacolon was
by Ehrenpreis in 1946. In a landmark paper, Whitehouse and
Kernohan summarized the literature and presented a series of
cases of their own, which documented that the aganglionosis
within the distal colon or rectum was the cause of the
functional obstruction.4
In 1949 Swenson published a paper in the New England
Journal of Medicine recommending rectosigmoidectomy with
preservation of the sphincters as the optimal treatment of this
disease.5 This operation was originally performed without a
CHAPTER 101 decompressing colostomy.6 However, technical difficulties in
small infants and the debilitated and malnourished state in
which most children presented caused most surgeons to adopt
a multistaged approach with colostomy as the initial step,6 an
Hirschsprung approach that became the standard of care for decades. In
recent years, improvements in surgical technique and earlier
suspicion and diagnosis of the disease have resulted in an
Disease evolution toward one-stage and minimal access procedures.
These advances have resulted in significantly improved
morbidity and mortality in infants with Hirschsprung disease.
Jacob C. Langer

Etiology and Genetics

Hirschsprung disease is a developmental disorder of the in- ------------------------------------------------------------------------------------------------------------------------------------------------

trinsic component of the enteric nervous system that is char- Ganglion cells are derived from the neural crest. By 13 weeks
acterized by the absence of ganglion cells in the myenteric and postconception, the neural crest cells have undergone a pro-
submucosal plexuses of the distal intestine. Because these cells cess of migration through the gastrointestinal tract from prox-
are responsible for normal peristalsis, patients with Hirsch- imal to distal, after which they differentiate into mature
sprung disease present with functional intestinal obstruction ganglion cells.7 In infants with Hirschsprung disease this pro-
at the level of aganglionosis. In most cases the aganglionosis cess is disturbed, so the ganglion cells are absent in the distal
involves the rectum or rectosigmoid, but it can extend for bowel. There are two theories as to why this occurs. The most
varying lengths, and in 5% to 10% of cases can involve the en- prevalent is that the neural crest cells never reach the distal
tire colon or even a significant amount of the small intestine. intestine because they either mature or differentiate into gan-
The incidence of Hirschsprung disease is approximately 1 in glion cells earlier than they should. Data supporting this the-
5000 live-born infants. ory come from spontaneously occurring animal models of
aganglionosis8 and from studies of normal neural crest cell mi-
gration done in chick embryos and human fetuses.9,10 The
second theory is that the ganglion cells do reach their destina-
History tion but fail to survive or proliferate. Data in support of this
------------------------------------------------------------------------------------------------------------------------------------------------
theory include animal studies suggesting that there are at least
The condition of “congenital megacolon” has been recognized two sources of neural crest cells (vagal and sacral), with migra-
for centuries. The first description of this condition was in the tion both proximally and distally. In addition, a number of
seventeenth century by Frederick Ruysch, who described a studies have suggested that the smooth muscle and extracel-
5-year-old child dying from an intestinal obstruction, followed lular matrix in the aganglionic bowel provides an inhospitable
by another account of a child with congenital megacolon by microenvironment for neuronal growth.11,12 It is likely that
Battini in 1800.1 It was not until 1887 that Harald Hirsch- Hirschsprung disease is actually a heterogeneous condition
sprung, a pathologist at Queen Louise Children’s Hospital with multiple genetic causes and etiologic mechanisms, so
in Copenhagen, described two cases of the condition that each of these theories may be true in individual cases.
1265
1266 PART VII ABDOMEN

The heterogeneous nature of Hirschsprung disease is sup- Hirschsprung disease from more common causes of constipa-
ported by increasing evidence that mutations in a variety of tion. Clinical features that point to this diagnosis include
genes may be responsible.13–15 The most commonly identified failure to pass meconium in the first 48 hours of life, failure
gene is the RET proto-oncogene, which was first identified in to thrive, gross abdominal distention, and dependence on
studies of Mennonite populations. RET encodes a tyrosine enemas without significant encopresis.21
kinase receptor, and many mutations of this gene and related
genes such as neurturin and glial cell line–derived neuro-
trophic factor (GDNF) have been identified in association with Enterocolitis
Hirschsprung disease. It remains unclear how these mutations Approximately 10% of children with Hirschsprung disease
result in aganglionosis, but there is some evidence that early present with fever, abdominal distention, and diarrhea due
neuronal cell death may be a prominent mechanism.16,17 to Hirschsprung-associated enterocolitis (HAEC), which
RET abnormalities are more commonly found in familial may be chronic, or may be severe and life-threatening. Be-
and long-segment disease. Mutations in the endothelin family cause Hirschsprung disease is generally thought of as causing
of genes, particularly endothelin-3 and the endothelin-B constipation, presentation with diarrhea may be confusing
receptor, are also commonly associated with Hirschsprung and the diagnosis may not be considered. A careful history
disease, although many of these children also have other including the failure to pass meconium and the presence of
abnormalities of neural crest–derived tissues, the most common intermittent obstructive episodes should lead to investigation
of which is Shah-Wardenberg syndrome. There is evidence for Hirschsprung disease.
from animal models that mutations in the endothelin and The etiology of HAEC is controversial. The most common
SOX-10 genes may produce early maturation or differentia- theory is that stasis caused by functional obstruction due to
tion of neural crest cells, which decreases the number of the aganglionic bowel permits bacterial overgrowth with sec-
available progenitor cells and prevents the neural crest cells ondary infection. Infectious agents such as Clostridium difficile
from migrating any further.18,19 Other genes that have been or Rotavirus have been postulated as being causative, but
associated with Hirschsprung disease include S1P1 (now there are few data to support a specific pathogen.22 There is
known as ZFHX1B), and Phox2B. some evidence implicating alterations in intestinal mucin
Hirschsprung disease is also associated with a number of production and alterations in the mucosal production of
syndromes for which the precise genetic basis of the aganglio- immunoglobulins in children with Hirschsprung-associated
nosis has not yet been elucidated. These include Trisomy-21, enterocolitis, which presumably results in loss of intestinal bar-
congenital central hypoventilation syndrome, Goldberg- rier function and allows bacterial invasion.23,24
Shprintzen syndrome, Smith-Lemli-Opitz syndrome, neurofi-
bromatosis, neuroblastoma, and a variety of other congenital Associated Conditions
anomalies.
Hirschsprung disease is associated with a variety of other con-
genital abnormalities, the presence of which should increase
the clinician’s level of suspicion (Table 101-1). These include
Diagnosis
------------------------------------------------------------------------------------------------------------------------------------------------
malrotation, genitourinary abnormalities, congenital heart dis-
ease, limb abnormalities, cleft lip and palate, hearing loss, mental
CLINICAL PRESENTATION retardation, and dysmorphic features. In addition, Hirschsprung
disease may be part of a large number of recognized syn-
Neonatal Obstruction
dromes such as trisomy 21, a variety of neurocristopathies,
Approximately 50% to 90% of children with Hirschsprung and congenital central hypoventilation syndrome.
disease present during the neonatal period with abdominal
distension, bilious vomiting, and feeding intolerance sugges-
TABLE 101-1
tive of distal intestinal obstruction. Delayed passage of meco-
Congenital Anomalies and Conditions Commonly Associated
nium beyond the first 24 hours is characteristic but is only with Hirschsprung Disease
present in approximately 90% of children with Hirschsprung
disease. In some patients cecal or appendiceal perforation may Down syndrome (trisomy 21)
be the initial event.20 Plain radiographs usually show dilated Neurocristopathy syndromes
bowel loops throughout the abdomen. The differential diag- 1. Waardenberg-Shah syndrome
nosis includes intestinal atresia, meconium ileus, meconium 2. Yemenite deaf-blind-hypopigmentation
plug syndrome, or a number of other less common conditions. 3. Piebaldism
4. Other hypopigmentation syndromes
Chronic Constipation Goldberg-Shprintzen syndrome
Smith-Lemli-Opitz syndrome
Some patients present later in childhood, or even during Multiple endocrine neoplasia 2
adulthood, with chronic constipation. This is most common Congenital central hypoventilation syndrome (Ondine curse)
among breast-fed infants, who typically develop constipation Isolated congenital anomalies
around the time of weaning. Although most children who pre- 1. Congenital heart disease
sent after the neonatal period have short-segment disease, this 2. Malrotation
history may also be found in those with longer segment or 3. Urinary tract anomalies
even total colonic involvement, particularly if the child has 4. Central nervous system anomalies
been exclusively breast-fed. Because constipation is frequently 5. Other
seen in childhood, it may be difficult to differentiate

RADIOLOGIC EVALUATION
For the neonate with a clinical picture and plain radiographs
suggesting distal neonatal bowel obstruction, the first step in
the diagnostic pathway is a water-soluble contrast enema. The
pathognomonic finding of Hirschsprung disease on contrast
enema is a transition zone between the normal and aganglio-
nic bowel (Fig. 101-1, A), although approximately 10% of
neonates with Hirschsprung disease may not have a demon-
strable radiologic transition zone.25 It is important to use
a water-soluble material because the enema may potentially
be a definitive treatment for other conditions in the differential
diagnosis such as meconium ileus and meconium plug syn-
drome. In older children an unprepped barium enema should
be done rather than a water-soluble contrast study. The ab-
sence of a transition zone is less common in this age group
but may still be present due to a short aganglionic segment.
In both neonates and older children, the most important view
is the lateral projection, in which a rectal transition zone will
be most evident (Fig. 101-1, B). Other findings on the contrast
enema that are suggestive of Hirschsprung disease include a
reversed recto-sigmoid index (Fig. 101-1, B) and retention
of contrast in the colon on a 24-hour postevacuation film.
ANORECTAL MANOMETRY
The recto-anal inhibitory reflex (RAIR) is defined as reflex re-
laxation of the internal anal sphincter in response to rectal dis-
tension and is present in normal children but absent in
children with Hirschsprung disease. The RAIR can be docu-
mented using anorectal manometry by inflating a balloon in
the rectum while simultaneously measuring the internal
sphincter pressure. Anorectal manometry is not widely avail-
able for neonates and is often operator dependent. In older
children the test is technically easier, but false-positive results
may occur due to masking of the relaxation response by con-
traction of the external sphincter, as well as artifacts created by
movement or crying. Anorectal manometry is most useful in
the evaluation of an older child with chronic constipation,
where documentation of a normal RAIR effectively rules out
Hirschsprung disease and avoids the need for a rectal biopsy.
A B
FIGURE 101-1 A, Water-soluble contrast enema demonstrating a transition zone at the splenic flexure. B, The lateral view is the most important one to
identify a low transition zone. In this case the recto-sigmoid index, consisting of the ratio of rectal diameter (R) to sigmoid diameter (S), is less than 1.0. D.
Retention of contrast on a 24-hour postevacuation film.
CHAPTER 101 HIRSCHSPRUNG DISEASE 1267
RECTAL BIOPSY
Definitive diagnosis of Hirschsprung disease is based on his-
tologic evaluation of a rectal biopsy, which remains the gold
standard diagnostic technique. The definitive finding that de-
fines Hirschsprung disease is absence of ganglion cells in the
submucosal and myenteric plexuses (Fig. 101-2, A). Most pa-
tients will also have evidence of hypertrophied nerve trunks
(Fig. 101-2, B), although this finding is not always present,
particularly in children with total colonic disease or a short
aganglionic segment. Because there is normally a paucity of
ganglion cells in the area 0.5 to 1 cm above the dentate line,
the biopsy should be taken at least 1 to 1.5 cm above it. How-
ever, a biopsy too proximal may miss a short aganglionic seg-
ment. Most surgeons use a suction biopsy technique, which is
associated with a low risk of perforation or bleeding. For chil-
dren in whom the suction biopsy yields an inadequate spec-
imen and in older children in whom the mucosa is too thick
for a suction biopsy, punch biopsies or full-thickness biopsies
provide more tissue and deeper levels.
In many centers the routine hematoxylin and eosin staining
is supplemented by staining for acetylcholinesterase, which
has a characteristic pattern in the submucosa and mucosa
in children with Hirschsprung disease (Fig. 101-2, C). Other
pathologists choose not to use acetylcholinesterase staining,
believing that it is too subjective and does not add any infor-
mation to the hematoxylin and eosin. A number of newer
stains have recently been shown to have additional value for
the diagnosis of Hirschsprung disease. The most accurate
appears to be immunochemical identification of calcitonin,
which is almost always absent in patients with Hirschsprung
disease (Fig. 101-2, D).26
Occasionally a premature infant will develop distal intesti-
nal obstruction, and the possibility of Hirschsprung disease
will be raised on the basis of clinical and radiologic parame-
ters. Early rectal biopsy in these children is not recommended
for two reasons: (1) The pathologist may have difficulty recog-
nizing ganglion cells due to their immaturity, and (2) it may be
difficult to obtain enough tissue without increasing the risk of
complications in a small premature infant. It is best in this
situation to decompress the rectum using stimulations and/or
irrigations and wait until the child is closer to term before
1268 PART VII ABDOMEN

A B C1

C2 D1 D2
FIGURE 101-2 Pathologic findings in children with Hirschsprung disease. A, Absence of ganglion cells in the myenteric plexus. B, Hypertrophied nerve
trunks. C, Cholinesterase staining in normal colon and colon affected by Hirschsprung disease. D, Calretinin staining in normal colon and colon affected by
Hirschsprung disease. (D, Courtesy Dr. Raj Kapur.)

doing the rectal biopsy. Although some surgeons believe that be administered, and a nasogastric tube should be inserted.
Hirschsprung disease is not seen in premature infants, this Children with associated abnormalities such as cardiac disease
condition has been well-documented in a number of series. or congenital central hypoventilation syndrome must be
investigated and managed before definitive surgical repair.
Children with enterocolitis or those in whom immediate
Preoperative Management surgery cannot be done for other reasons should undergo
------------------------------------------------------------------------------------------------------------------------------------------------
decompression of the colon using digital rectal stimulation,
In most cases the treatment of Hirschsprung disease is surgi- irrigations, or occasionally an emergency stoma.
cal. However, there are a number of important preoperative Once a child has been resuscitated and stabilized, opera-
interventions that must be considered before definitive surgi- tion can be done semielectively. While waiting, most children
cal intervention. The first priority is resuscitation, particularly can be discharged home on breast milk or an elemental for-
in neonates with intestinal obstruction or children presenting mula, in combination with rectal stimulations or irrigations.
with enterocolitis. In both groups, intravenous fluids and In the older child with an extremely dilated colon, operation
broad-spectrum antibiotics against enteric organisms should should be delayed until the diameter of the colon has
 CHAPTER 101 HIRSCHSPRUNG DISEASE 1269

decreased sufficiently to do a pull-through safely. This can some- the anus while preserving normal sphincter function. The
times be accomplished with weeks or months of irrigations, but most commonly performed operations are the Swenson,
some of these children may require a colostomy in order to Duhamel, and Soave procedures (Fig. 101-4), although a
adequately decompress the dilated colon (Fig. 101-3). number of other operations such as the Rebhein and State pro-
Some authors have advocated nonoperative long-term cedures have been described and are still done in some cen-
management of short-segment Hirschsprung disease using ters. Although many publications in the literature report
enemas and laxatives. Others have suggested that simple results after each of these operations, few randomized or prop-
myectomy may be adequate.27 However, these techniques erly controlled prospective studies exist. Because of this lack of
do not provide a good quality of life for most children with evidence, it is fair to say that all are acceptable alternatives and
Hirschsprung disease, and most pediatric surgeons recom- that the best operation for an individual patient is the one that
mend a pull-through procedure. the surgeon has been trained to do and does frequently.

SWENSON PROCEDURE
Pull-through Procedure for Swenson provided the first description of a surgical approach
Hirschsprung Disease to Hirschsprung disease in the late 1940s. Swenson’s goal was
------------------------------------------------------------------------------------------------------------------------------------------------
removal of the entire aganglionic colon, with an end-to-end
The goals of surgical management for Hirschsprung disease anastomosis above the anal sphincter. The operation was orig-
are to remove the aganglionic bowel and reconstruct the intes- inally done through a laparotomy, with the anastomosis being
tinal tract by bringing the normally innervated bowel down to performed from a perineal approach after eversion of the

A B
FIGURE 101-3 Value of a decompressing colostomy to decrease the size of the dilated sigmoid colon before pull-through surgery. A, Before colostomy.
B, Six months after colostomy.

A B C
FIGURE 101-4 The three most commonly performed operations for Hirschsprung disease. A, Soave. B, Swenson. C, Duhamel. þ ¼ ganglionic
bowel,  ¼ aganglionic bowel.
1270 PART VII ABDOMEN
aganglionic rectum. Safe performance of the operation re-
quires careful attention to dissection tightly on the rectal wall,
in order to avoid injury to deep pelvic nerves, vessels, and
other structures such as vagina, prostate, vas deferens, and
seminal vesicles. Despite the theoretical risks inherent in
the deep pelvic dissection, long-term outcome studies of the
Swenson procedure report excellent functional results with
respect to continence, urinary, and sexual function.28
SOAVE PROCEDURE
The Soave procedure was designed to avoid the risks of injury
to pelvic structures inherent in the Swenson procedure by
doing a submucosal endorectal dissection and placing the
pull-through bowel within a “cuff” consisting of aganglionic
muscle. In the initial description of the operation, the
pulled-through colon was left hanging out through the anus.
This exteriorized bowel was excised, and the anastomosis
done, at a second operation several weeks later. This was sub-
sequently modified by Boley, who performed the procedure in
a single stage.29 Over the years there has been controversy
regarding how long the cuff should be, as well as whether it
should be split or a segment excised. Despite claims by some
authors that the Soave procedure is more likely to result in
long-term issues with constipation due to incomplete excision
of the aganglionic rectum,30 most late follow-up studies have
reported similar outcomes to that seen with the Swenson
procedure.31
DUHAMEL PROCEDURE
The Duhamel procedure involves bringing the normal colon
down through the bloodless plane between the rectum and
the sacrum and joining the two walls to create a new lumen,
which was aganglionic anteriorly and normally innervated
posteriorly. In the initial description, two Kocher clamps were
used to join the walls and were left in for a week. More
recently, surgical staplers were used instead. The Duhamel
procedure has several potential advantages over the Swenson
or Soave procedures. It is believed to be easier and safer, with
less pelvic dissection than the other two operations; it has a
large anastomosis, which decreases the risk of anastomotic
stricture; and the presence of a “reservoir” makes it appealing
for children with longer aganglionic segments.
A B
FIGURE 101-5 Laparoscopic pull-through. A, Confirmation of the pathologic transition zone. B, Laparoscopic view of the completed pull-through
from above. (Courtesy Dr. Steve Rothenberg.)
Role of Colostomy
------------------------------------------------------------------------------------------------------------------------------------------------
Swenson’s initial operation was described as a one-stage
procedure, but relatively high incidence of stricture, leak, and
other adverse outcomes led him and others to recommend a
routine preliminary colostomy, followed by a period of growth
and a subsequent reconstructive operation.32 This approach
became surgical dogma, which was reinforced by the fact that
many children with Hirschsprung disease presented late with
malnutrition and dilated colon, so a colostomy was a life-
saving procedure. In the 1980s, however, a number of surgeons
reported series of single-stage pull-through procedures even in
small infants.33,34 Over the next 10 to 15 years, one-stage
operations became increasingly popular and many reports
documented the safety of this approach, suggesting that a
single-stage procedure avoids the known morbidity of stomas
in infants and is also more cost-effective.35–37 It is important
to remember, however, that a stoma may still be indicated
for children with severe enterocolitis, perforation, malnutri-
tion, or massively dilated proximal bowel, as well as in situa-
tions where there is inadequate pathology support to reliably
identify the transition zone on frozen section.
Minimal Access Approaches
------------------------------------------------------------------------------------------------------------------------------------------------
LAPAROSCOPIC PULL-THROUGH
With the advent of laparoscopic surgery in the late 1980s,
minimal access techniques became increasingly applied to
pediatric surgical diseases. The first minimal approach to
pull-through surgery for Hirschsprung disease was described
by Georgeson in 1995,38 who described a laparoscopic pull-
through for Hirschsprung disease, involving laparoscopic bi-
opsy to identify the transition zone, laparoscopic mobilization
of the rectum below the peritoneal reflection, and a short
mucosal dissection through a perineal approach (Fig. 101-5).
The rectum is then prolapsed through the anus, and the anas-
tomosis is done from below. This procedure has been
associated with a shorter hospital time, and both early and mid-
term results appear to be equivalent to those reported for the
open procedures.39 Laparoscopic approaches have been also
described for the Duhamel and Swenson operations, with
excellent short-term results reported.40,41
 CHAPTER 101 HIRSCHSPRUNG DISEASE 1271

TRANSANAL (PERINEAL) PULL-THROUGH

The transanal pull-through procedure uses the same mucosal
dissection from below as the Georgeson operation, but with-
out laparoscopic mobilization of the rectum. The mucosal in-
cision is made 0.5 to 1 cm above the dentate line, depending
on the size of the child, and the mucosa is stripped from the
underlying muscle as in the Soave operation. The rectal mus-
cle is then incised circumferentially, and the dissection is con-
tinued on the rectal wall, dividing the vessels as they enter the
rectum. The entire rectum and part of the sigmoid colon can
be delivered through the anus. When the transition zone is
reached, the anastomosis is done from below (Fig. 101-6).
In patients with a more proximal transition zone (usually
above the proximal sigmoid colon), laparoscopy or a small
umbilical incision can be used to mobilize the left colon
and/or splenic flexure to achieve adequate length. A transanal
approach can also be used if the patient has already had a co-
lostomy, by using the stoma as the end of the pull-through
bowel and performing the rectal excision using the transanal
technique.
The transanal approach has a low complication rate, re-
quires minimal analgesia, and is associated with early feeding
and discharge.42–45 Although there have not been any studies
comparing the transanal and laparoscopic approaches, the
transanal pull-through can be done by any pediatric surgeon,
including those without laparoscopic skills, and by pediatric
surgeons in parts of the world where access to appropriately
miniaturized laparoscopic equipment is limited.
A number of ongoing controversies surround the transanal
pull-through. The first is whether the pathologic transition
zone should be defined before beginning the anal dissection.
This was not done in the early descriptions of the operation
and continues to be omitted by many surgeons. The main
rationale for this practice is the known inaccuracy of the
contrast enema in predicting the level of aganglionosis, with
approximately 8% of children who have a rectosigmoid tran-
sition zone on contrast study having a more proximal transi-
tion zone on histology.46 This step is particularly important for
surgeons who do a different operation for long-segment dis-
ease than they do for rectosigmoid disease. The preliminary
biopsy can be done using a laparoscopic approach, or through
a small umbilical incision, both of which can also be used to
mobilize the splenic flexure in children with higher transition
zones.47 The advantage of the umbilical approach is that it can
be done by any surgeon, anywhere in the world, and does not
require laparoscopic skills or equipment. Evidence would
suggest that a preliminary biopsy to determine the pathologic
transition zone does not have a deleterious effect on post-
operative outcomes such as time to feeding, pain, or length
of hospital stay.45,48
There is also controversy about whether the transanal pull-
through is best done in the prone or supine position.49 The
prone position provides excellent visualization and is familiar
to most pediatric surgeons because of their experience with it
in the repair of anorectal malformations. The supine position
has the advantage of access to the peritoneal cavity for initial
colonic biopsy or for mobilization of the colon or a stoma if
necessary. Finally, there is controversy about the length of
the rectal cuff. The initial description of the transanal pull-
through involved a long cuff, with a submucosal dissection
extending into the peritoneal cavity. Many surgeons continue
to do the operation this way, and most advocate division of the
cuff to prevent narrowing. Other surgeons have modified the
procedure by doing a short submucosal dissection for a few
centimeters, and others have abandoned the submucosal
dissection entirely and do what is essentially a transanal

A B C

D E
FIGURE 101-6 The transanal Soave pull-through. A, An umbilical incision is used for a preliminary biopsy. A Heger dilator is used to push the sigmoid
into the umbilical incision. B, Eversion sutures are placed, and a nasal speculum is used to provide exposure to the anal canal. A circumferential incision is made
5 mm from the dentate line. C, The submucosal dissection is carried 2 to 3 cm. D, Once the muscle cuff has been divided circumferentially, the dissection
is carried proximally, staying right on the colonic wall. E, The bowel is divided at least 2 cm above the biopsy showing ganglion cells, and the anastomosis
is performed. Care must be taken to do the anastomosis to the rectal mucosa, not to the transitional epithelium, or normal sensation will be lost and the risk
of incontinence will be increased.
1272 PART VII ABDOMEN

Swenson procedure.50 The advantage of a shorter cuff is a Once the level of aganglionosis has been identified, most
lower rate of narrowing and potentially a lower risk of surgeons create a stoma, wait for permanent sections, and
postoperative obstructive symptoms and enterocolitis.48 do a definitive reconstructive procedure at a later time. Al-
though primary pull-through without ileostomy for total co-
lonic disease has been reported, this approach requires a
Surgical Approach to Long- high degree of confidence in the pathologist because it re-
quires doing a total colectomy on the basis of frozen sections
Segment Hirschsprung Disease alone. In addition, many surgeons believe that the results of
------------------------------------------------------------------------------------------------------------------------------------------------
pull-through surgery are better once the stool has thickened,
Long-segment Hirschsprung disease is usually defined as a which usually occurs in the first few months of life.
transition zone that is proximal to the midtransverse colon. Three types of operations are available for reconstruction in
The most common is total colonic aganglionosis, which usu- children with long-segment Hirschsprung disease: straight
ally also includes some of the distal ileum. In rare cases most of pull-through, colon patch, and J-pouch construction. Straight
or the entire small bowel is aganglionic. Long-segment disease pull-through procedures involve bringing the normally inner-
is more likely to be associated with a positive family history51 vated ileum to just above the anal sphincter, using any one of
and is more likely to be diagnosed prenatally.52 Contrast the standard techniques (Swenson, Duhamel, or Soave). Co-
enema typically shows a shortened, relatively narrow colon lon patch procedures involve a side-to-side anastomosis be-
(“question mark colon”) (Fig. 101-7),53 and there may also tween normally innervated small bowel and aganglionic
be a transition zone in the small bowel. The rectal biopsy colon, using the small bowel for motility and the colon as a
shows absence of ganglion cells, but in many cases there are reservoir for storage of stool and absorption of water. The
no hypertrophic nerves or abnormalities of acetylcholinester- Martin procedure consists of a Duhamel reconstruction that ex-
ase staining. tends proximally to involve the entire left colon (Fig. 101-8, A).
Early resuscitation and management is similar to that de- Kimura, using the rationale that the right colon is better at water
scribed for standard Hirschsprung disease. Sequential colonic absorption than the left colon, advocated a staged procedure, in
biopsies are done looking for ganglion cells on frozen section. which the right colon is anastomosed side-to-side to the ileum.
These can be done through a standard laparotomy, laparosco- The “ileo-colon” is then disconnected from the right colonic
pically, or through an umbilical incision, which in a newborn blood supply after several months and anastomosed above the
can be used to access all parts of the colon. Traditionally, many anal sphincter (Fig. 101-8, B). The J-pouch procedure is done
surgeons started with an appendectomy, assuming that lack of commonly for children and adults with ulcerative colitis and
ganglion cells in the appendix would be diagnostic of total familial polyposis syndrome, and some pediatric surgeons
colonic disease. However, this may result in a false-positive have reported the use of this operation for children with
diagnosis of total colonic Hirschsprung disease because there long-segment Hirschsprung disease.55
may be a paucity of ganglion cells in the appendix in children There are no prospective or well-controlled series reporting
with shorter segment disease.54 long-term results of surgery for long-segment Hirschsprung
disease. Although the colon patch procedures theoretically
result in decreased stool output due to better water absorp-
tion, the aganglionic colon gradually tends to dilate and many
of these patients develop severe enterocolitis, which requires
removal of the patch or a permanent stoma. Children under-
going straight pull-through tend to experience gradually
decreasing stool frequency over time, with an acceptable
quality of life.56–58
Near-Total Intestinal Aganglionosis
Rarely, almost the entire intestinal tract of a patient is agangli-
onic, usually leaving 10 to 40 cm of normally innervated jeju-
num. In most of these cases, there is not enough functional
small bowel to support enteral nutrition and intestinal failure
results. These children require total parenteral nutrition from
birth, a situation that has been associated with a high risk of
mortality from liver failure. The surgical approach at the time
of the first laparotomy is to determine the extent of aganglio-
nosis on the basis of frozen sections and to bring out a stoma at
the most distal point that has normally innervated bowel.
Some surgeons prefer to bring out a more distal stoma, but this
approach may increase the risk of chronic intestinal obstruc-
tion and bacterial overgrowth. A central venous catheter
should be inserted for parenteral nutrition, and a gastrostomy
should be considered for continuous “trophic” feeding of
FIGURE 101-7 Contrast enema in a child with total colonic Hirschsprung
breast milk or elemental formula.
disease. There is no transition zone in the colon, and the colon is foreshor- The management of these children is similar to the manage-
tened with a “question-mark” configuration. ment of any child with intestinal failure.59 Strict attention to
 CHAPTER 101 HIRSCHSPRUNG DISEASE 1273

C
FIGURE 101-8 Operations for long-segment Hirschsprung disease. A, Martin procedure. B and C, Kimura procedure.

prevention of sepsis, treatment of bacterial overgrowth, use of
trophic feeds, and prevention of TPN-related cholestasis are
all extremely important. Recent experience with omega-3
lipids has resulted in encouraging trends toward prevention
and treatment of this problem.60
A number of surgical options are available for children with
near-total aganglionosis.61 For children who develop signifi-
cant proximal dilatation of the normally innervated bowel,
tapering, imbrication, or bowel-lengthening procedures
such as the Bianchi or serial transverse enteroplasty (STEP)
1274 PART VII ABDOMEN

procedure may be used.62,63 Zeigler has popularized a tech- OBSTRUCTIVE SYMPTOMS

nique known as “myectomy-myotomy,” in which a length of
aganglionic small bowel distal to the transition zone un- There are a range of obstructive symptoms that can be seen
dergoes myectomy.64 Although a few successful cases using after a pull-through. Abdominal distension, bloating, vomit-
this technique have been reported, most surgeons have not ing, or ongoing severe constipation may be present immedi-
found it to be successful and have noted high rates of postop- ately after surgery or may develop later after an initial
erative complications. For children with ongoing liver failure, period of normal bowel function. There are five major reasons
small bowel or combined small bowel-liver transplantation for persistent obstructive symptoms following a pull-through:
may offer the only chance for survival. A recent report from mechanical obstruction, recurrent or acquired aganglionosis,
Paris documented extremely good results in 12 patients, many disordered motility in the proximal colon or small bowel,
of whom underwent successful pull-through surgery follow- internal sphincter achalasia, or functional megacolon caused
ing their intestinal transplant.65 by stool-holding behavior (Table 101-2). The clinician will
have much greater success in managing these difficult patients
if an organized approach to this problem is taken. One
Postoperative Care
------------------------------------------------------------------------------------------------------------------------------------------------
proposed algorithm is shown in Figure 101-9.71
Mechanical Obstruction
Most children undergoing a laparoscopic or transanal pull-
through for standard Hirschsprung disease can be fed imme- The most common cause of mechanical obstruction after a
diately, and most can be discharged within 24 to 48 hours. pull-through is a stricture, which usually occurs after a Swen-
The anastomosis should be calibrated with an appropriately son or Soave procedure (Fig. 101-10, A). Patients undergoing
sized dilator or finger 1 to 2 weeks after the procedure. Al- a Duhamel procedure may have a retained “spur” consisting of
though many surgeons instruct the parents to dilate the anas- the anterior aganglionic bowel, which may fill with stool and
tomosis on a daily basis, others have found it to be obstruct the pulled-through bowel (Fig. 101-10, B). In other
unnecessary in most cases and instead perform weekly cali- cases, there may be obstruction secondary to a twist in the
bration for a period of 4 to 6 weeks. It is important for the pulled-through bowel (Fig. 101-10, C) or narrowing due to
parents to protect the buttocks with a barrier cream because a long muscular cuff in children who have had a Soave
at least 50% of children will have frequent stools and peri- procedure.
neal skin breakdown postoperatively. Fortunately, this prob- Obstruction can be identified using a combination of dig-
lem tends to resolve over time. ital rectal examination and a barium enema. Initial manage-
As with any operation, children undergoing a pull-through ment of anastomotic stricture consists of repeated dilatation
may develop a wound infection or intra-abdominal bleeding. using a finger, dilator, or radially dilating balloon. Some au-
In addition, anastomotic complications such as leak or stric- thors have advocated innovative techniques for recalcitrant
ture may occur. Intestinal perforation can occur at a proximal strictures including antegrade dilatation over a string using
biopsy site due to back pressure from anal sphincter spasm or Tucker dilators72 and the use of intralesional steroid.73 In
due to unrecognized cautery injury. Bowel obstruction can re- some cases the stricture cannot be successfully dilated, and
sult from intra-abdominal adhesions, a twist in the pull- revision of the pull-through is necessary. This is best done
through bowel, or a muscular cuff that has rolled down and using the Duhamel technique, although other operations have
constricted the pull-through bowel. In rare cases, rectovesical also been advocated.74–76 Duhamel spurs can be resected from
or rectovaginal fistulas have developed after pull-through above or managed by extending the staple line from below,
surgery. Close monitoring for and early treatment of these with or without laparoscopic visualization. Twisted pull-
complications is imperative. In addition, children with throughs and narrow muscular cuffs usually require surgical
Hirschsprung disease can develop enterocolitis, even in the intervention, typically a repeat pull-through. In some cases,
early postoperative period. The family and the primary care a muscular cuff can be divided laparoscopically without
physician should be educated about the signs and symptoms having to re-do the entire pull-through.
of enterocolitis, and the family must be told to bring the child
to the hospital if there are any signs suggestive of this problem Persistent or Acquired Aganglionosis
because children can become very sick and even die from This rare problem may be due to pathologist error,77 a tran-
enterocolitis.66 sition zone pull-through,78 or ganglion cell loss after a
pull-through.79 Repeat rectal biopsy, above the previous anas-
tomosis (it must be posterior in the case of an initial Duhamel
Long-Term Outcomes
------------------------------------------------------------------------------------------------------------------------------------------------
procedure), should be done to determine whether there are
Long-term problems in children with Hirschsprung disease
include ongoing obstructive symptoms, soiling, and enteroco- TABLE 101-2
litis.67 Quite often an individual child may have a combination Causes of Obstructive Symptoms Following Surgery
of problems. Although early reports suggested that long-term for Hirschsprung Disease
issues were rare after surgical treatment of Hirschsprung dis- Mechanical obstruction
ease,68 it is now clear that these complications are more com- Recurrent or residual aganglionosis
mon than previously recognized.31,69,70 It is important for the Motility disorder involving the ganglionated bowel
surgeon to follow these children closely, at least until they are Internal anal sphincter achalasia
through the toilet training process, in order to identify and Functional megacolon (stool-holding behavior)
provide early treatment for these problems.
 CHAPTER 101 HIRSCHSPRUNG DISEASE 1275

Physical exam and barium enema

No stricture Stricture

Rectal biopsy
Dilate or redo
pull-through
Ganglion cells present
No ganglion cells

Motility workup
Redo pull-through

Normal Abnormal focal Abnormal generalized

Try botox Resect

Bowel regimen
or stoma

Clinical response No clinical response

Repeat botox or
myectomy

FIGURE 101-9 Algorithm for the investigation and management of the child with obstructive symptoms following a pull-through.

A B C
FIGURE 101-10 Causes of mechanical obstruction after a pull-through. A, Stricture following a Soave procedure. B, Anterior aganglionic “spur” following
a Duhamel procedure. C, Twisted transanal pull-through.

normal ganglion cells present in all patients with persistent
obstructive symptoms after surgery. The pathology from the
original operation should be reviewed to ensure that there
was normal innervation at the proximal margin, and in some
cases further sections should be done circumferentially from
the resection margin because the transition zone is often asym-
metrical in children with Hirschsprung disease.80 The best
treatment for persistent or acquired aganglionosis in most
cases is a repeat pull-through, which can be done using either
a Soave or Duhamel approach.75
Motility Disorder Children with Hirschsprung disease
often have associated motility disorders including an in-
creased incidence of gastroesophageal reflux and delayed
gastric emptying,81 small bowel dysmotility, and disordered
colonic motility. Some cases are more focal, usually involving
the left colon. In some cases the disordered motility may be
associated with histological abnormalities such as intestinal
neuronal dysplasia (discussed in greater detail later).
In children who have been shown not to have a mechanical
obstruction and who have normal ganglion cells on rectal
1276 PART VII ABDOMEN
biopsy, investigations for motility disorders should be under-
taken. This can include a radiologic shape study, radionuclide
colon transit study,82 colonic manometry,83 and laparoscopic
biopsies looking for evidence of intestinal neuronal dyspla-
sia.84 If a focal abnormality is found, consideration should
be given to resection and repeat pull-through using normal
bowel. If the abnormality is diffuse, the appropriate treatment
is bowel management and the use of prokinetic agents. Some
children with particularly dysmotile colon will benefit from
placement of a cecostomy for antegrade colonic enemas.85
Internal Sphincter Achalasia This term refers to the lack of
a normal RAIR that is present in all children with Hirsch-
sprung disease (as described earlier in “Diagnosis”). It is
unclear why only some children develop obstructive symp-
toms from this nonrelaxation, while others function normally
in the postoperative period. It is also unclear why most chil-
dren eventually “grow out” of this problem over time, usually
by the age of 5 years. Internal sphincter achalasia is a diagnosis
of exclusion, which is made after ruling out mechanical ob-
struction, aganglionosis, and dysmotility. The diagnosis can
be confirmed by demonstrating a clinical response to intra-
sphincteric botulinum toxin.86 However, the response to bot-
ulinum toxin is not related to the absolute value of the internal
anal sphincter pressure (i.e., patients with a higher resting
pressure are not necessarily more likely to benefit from botu-
linum toxin).
The standard treatment of internal sphincter achalasia has
been internal sphincterotomy or myectomy,87,88 but because
this problem tends to resolve on its own in most children
and there is concern about sphincter-cutting operations
exacerbating future soiling issues, we prefer to use “che-
mical sphincterotomy” with intrasphincteric botulinum
toxin.86,89,90 In many cases repeated injection of botulinum
toxin or applications of nitroglycerine paste or topical nifedi-
pine are necessary while waiting for resolution of the problem.
Functional Megacolon Functional megacolon is the result
of stool-holding behavior, a common cause of constipation
that some authors claim affects up to half of normal children
at some time during their first few years of life.91 This
condition is probably even more common in children with
Hirschsprung disease because of their predisposition to con-
stipation, which leads to hard painful stools, withholding be-
havior, and a resulting vicious cycle.92 The treatment for this
problem is a bowel management regimen consisting of
laxatives, enemas, and behavior modification including
support for the child and family. In some severe cases of
obstructive symptoms, the child may be best served by use
of a cecostomy and administration of antegrade enemas, or
even by the creation of a proximal stoma. In many cases the
cecostomy or stoma can ultimately be reversed when the child
reaches adolescence.
Fecal Soiling
There are three broad causes of soiling after a pull-through:
abnormal sphincter function, abnormal sensation, or
“pseudo-incontinence” related to abnormal rectal function
or obstipation (Table 101-3). Abnormal sphincter function
may be due to sphincter injury during the pull-through or
to a previous myectomy or sphincterotomy and can usually
be identified using anorectal manometry. Two forms of
TABLE 101-3
Causes of Soiling Following Surgery for Hirschsprung Disease
Abnormal sensation
1. Inability to feel rectal distension
2. Loss of transitional epithelium
Abnormal sphincter function
“Pseudo-incontinence”
1. Associated with severe constipation
2. Associated with hyperperistalsis of the pulled-through bowel
abnormal sensation exist. The first is lack of sensation of a
full rectum, which is also identifiable using anorectal manom-
etry by expanding a balloon in the rectum and asking the
child to state when he can feel it. The other type of sensation
that may be abnormal is the ability to detect the difference
between gas and stool, which is dependent on intact transi-
tional epithelium in the anal canal. This sensation may be im-
paired if the anastomosis is done too low and the transitional
epithelium is damaged. This problem is usually evident on
simple physical examination. Neither sphincter weakness
nor abnormal sensation is amenable to a surgical solution,
and most of these children are best managed using a bowel
routine that may include a constipating diet, rectal enemas,
or antegrade enemas through a cecostomy. Biofeedback train-
ing has been advocated, especially for those children with
sphincter weakness. In some cases the child is best served
by a colostomy.
If both the sphincter and sensation are intact, the
most common cause of soiling after a pull-through is “pseudo-
incontinence.”93 Some patients have severe obstipation with a
massively distended rectum and develop overflow of liquid
stool around the fecal mass. Others simply leak small amounts
of stool through the day, creating “skid marks” in the under-
wear on a constant basis. Other children suffer from hyper-
peristalsis of the pulled-through bowel, which results in
inability of the anal sphincter to achieve control despite
normal sphincter function.83
Successful management depends on a clear understanding
of the underlying basis for the soiling, which requires a clear
history and physical examination, as well as investigations
such as abdominal radiograph, barium enema, anorectal ma-
nometry, and in some cases colonic manometry. Children with
severe constipation will benefit from laxative therapy. How-
ever, if the sphincter or sensation, or both, are inadequate,
passive laxatives such as lactulose or PEG 3300 will make
the problem worse and the child should instead be treated
with stimulant laxatives such as senna or enemas. On the other
hand, children with stool-holding behavior who have a nor-
mal sphincter and sensation will often experience exacer-
bation of the behavioral problem by rectal enemas or any
other kind of anal manipulation. Children without constipa-
tion who have hyperperistalsis of the pulled-through bowel or
abnormal sphincter function or sensation will benefit from a
constipating diet and medications such as loperamide. Chil-
dren with slow transit constipation or stool-holding behavior,
on the other hand, will benefit from a high-fiber diet and
passive laxative therapy. The treatment of soiling must be
based on a clear understanding of the child’s underlying
problem.

Enterocolitis
Enterocolitis may be present both before and after surgical
correction of the disease, and it can be severe or life threaten-
ing. HAEC is more common in children diagnosed at a youn-
ger age,94 those with longer segment disease, and those with
trisomy 21. The clinical features of HAEC are generally agreed
on and include fever, abdominal distention, diarrhea, elevated
white blood cell count, and evidence of intestinal edema on
abdominal radiograph. Because there is overlap between
HAEC and other conditions such as obstructive symptoms
and gastroenteritis, there has been confusion in the literature
as to the exact definition and the true incidence of the condi-
tion. A recently developed HAEC score may be useful in the
future in both the clinical setting and in research into this area
(Table 101-4).95
The treatment of postoperative HAEC involves nasogastric
drainage, intravenous fluids, broad-spectrum antibiotics, and
decompression of the rectum and colon using rectal stimulation
or irrigations. The risk of HAEC can be minimized by using pre-
ventive measures such as routine irrigations96 or chronic ad-
ministration of metronidazole or probiotic agents, particularly
in those who are thought to be at higher risk for this complica-
tion on the basis of clinical or histologic grounds. Because en-
terocolitis is the most common cause of death in children with
Hirschsprung disease and can occur postoperatively even in
children who did not have it preoperatively, it is extremely im-
portant that the surgeon educate the family about the risk of this
complication and urge early return to the hospital if the child
should develop any concerning symptoms.66
Long-Term Outcomes
Despite the relatively common occurrence of postoperative
problems, most children with Hirschsprung disease overcome
these issues and do well. Obstructive symptoms, soiling, and
TABLE 101-4
Hirschsprung-Associated Enterocolitis (HAEC) Score*
History
Diarrhea with explosive stool
Diarrhea with foul-smelling stool
Diarrhea with bloody stool
Previous history of enterocolitis
Physical examination
Explosive discharge of gas and stool on rectal examination
Distended abdomen
Decreased peripheral perfusion
Lethargy
Fever
Radiology
Multiple air-fluid levels
Dilated loops of bowel
Sawtooth appearance with irregular mucosal lining
Cutoff sign in recto-sigmoid with absence of distal air
Pneumatosis
Laboratory
Leukocytosis
Shift to left
*A score of 10 or higher was associated with a positive diagnosis of HAEC by an
international panel of experts.
CHAPTER 101 HIRSCHSPRUNG DISEASE 1277
enterocolitis, in the absence of an ongoing source of obstruc-
tion, usually resolve after the first 5 years of life. Studies of
teenagers and adults with Hirschsprung disease suggest that
sexual function, social satisfaction, and quality of life all
appear to be normal in the vast majority of patients once they
reach their late teens.69,97
Several populations of children have less optimistic out-
comes. Children with long-segment disease appear to have
a higher risk of enterocolitis, incontinence, and dehydration
than children with shorter-segment disease. Children with
Hirschsprung disease associated with Down syndrome have
a greater risk of enterocolitis and incontinence.98,99 Finally,
prognosis may be poor in children with other types of comor-
bidity such as those with congenital central hypoventilation
syndrome, congenital heart disease, and syndromes that are
associated with mental retardation or other forms of disability.
“Variant” Hirschsprung Disease
------------------------------------------------------------------------------------------------------------------------------------------------
Some children present with signs and symptoms suggestive of
Hirschsprung disease but have ganglion cells present on rectal
biopsy.100 There is a significant amount of controversy sur-
rounding the definitions and features of many of these condi-
tions,101 and in some cases their existence has even been
called into question.
INTESTINAL NEURONAL DYSPLASIA
This condition was first described by Meier-Ruge in 1971. Two
types are usually described.102 Type A is less common and is
characterized by diminished or absent sympathetic innerva-
tion of the myenteric and submucosal plexuses, as well as hy-
perplasia of the myenteric plexus. Type B consists of dysplasia
of the submucous plexus with thickened nerve fibers and
giant ganglia, increased acetylcholinesterase staining, and
identification of ectopic ganglion cells in the lamina propria.
Type B can occur on its own or can be present in the nonagan-
glionic bowel in children who also have Hirschsprung disease.
2 In addition, intestinal neuronal dysplasia may be either diffuse
2 or focal. The reported incidence of IND varies significantly
1 from one center to another, largely due to differences in
1 definition of the condition.
Despite multiple publications on the topic of IND, this
2 topic still stimulates controversy among pediatric surgeons
2 and pediatric pathologists.103 The diagnostic criteria have
1 changed over time, and there is disagreement among pathol-
1 ogists about the diagnosis both in general terms and with re-
1 spect to individual patients.104 Sophisticated histologic
techniques including special stains and the use of thick sec-
1 tions are often believed to be necessary for an accurate diag-
1 nosis.105 In addition, there is some evidence that the
1 histologic finding of IND may in some cases be secondary
1 to chronic obstruction rather than the cause of it, and in many
1 cases there may not be good correlation between the histologic
finding of IND and the motility function of the bowel.106
1 Hypoganglionosis
1
This is a rare form of dysganglionosis, which is characterized by
sparse and small ganglia, usually in the distal bowel, often as-
sociated with abnormalities in acetylcholinesterase distribution.
1278 PART VII ABDOMEN
The appropriate treatment is to resect the abnormal colon and
perform a pull-through procedure, much as one would do for
a child with Hirschsprung disease.107 It is important to differ-
entiate this condition from immature ganglia, which is seen in
preterm children who present with a picture of distal intesti-
nal obstruction resulting from underdeveloped colonic motil-
ity. The colonic motility is self-limited and should not be
treated surgically.108
Internal Sphincter Achalasia
Some children have normal ganglion cells on rectal biopsy but
lack the RAIR on anorectal manometry. These children will
sometimes develop obstructive symptoms or severe constipa-
tion that mimics Hirschsprung disease. Similarly to the situa-
tion mentioned previously in which obstructive symptoms
continue after surgery for Hirschsprung disease, this condi-
tion has been termed internal sphincter achalasia.109 The diag-
nosis is made using anorectal manometry and rectal biopsy.
The initial treatment is a bowel management regimen, consisting
of diet, laxatives, and enemas or irrigations. If this is unsuccess-
ful, many surgeons have advocated the use of anal sphincter
myectomy.110,111 Because the constipation associated with this
condition usually improves over the first 5 years of life, the
same rationale has been used to advocate temporary or revers-
ible sphincter-relaxing measures such as botulinum toxin,112
nitroglycerine paste,113 or topical nifedipine.
Ultrashort-Segment Hirschsprung Disease
There is much confusion in the literature regarding this
condition and how it is defined. Some authors use this term to
describe children with normal ganglion cells on rectal biopsy,
but with absence of the RAIR (which is synonymous
with the definition of internal sphincter achalasia). We prefer
to reserve it for children who have a documented aganglionic
segment of less than 1 to 2 cm. In children with thiscondition, the
findings of hypertrophic nerves and abnormal cholinesterase
staining may be absent.114 The treatment of ultrashort-segment
Hirschsprung disease is controversial. Some authors advocate
simple anal sphincter myectomy,115,116 and some prefer excision
of the aganglionic segment and pull-through reconstruction.
Desmosis coli
This is a condition described by Meier-Ruge characterized by
total or focal lack of the connective tissue net of the circular
and longitudinal muscles and the connective tissue layer of
the myenteric plexus, without any abnormality of the enteric
nervous system.117 Patients with this condition present with
chronic constipation. In one family Hirschsprung disease
and desmosis coli coexisted,118 although in most cases they
are completely separate entities.
The complete reference list is available online at www.
expertconsult.com.

CHAPTER 102
Intestinal
Dysganglionosis
and Other
Disorders of
Intestinal Motility
Prem Puri
The typical patient with Hirschsprung disease (HD) is a new-
born presenting with delayed passage of meconium and ab-
dominal distension or a young child presenting with severe
chronic constipation. The diagnosis is confirmed by the histo-
logic and histochemical evaluation of suction rectal biopsies,
which demonstrate absence of ganglion cells in the submu-
cosa and increased acetylcholinesterase (AChE) activity in
the lamina propria. However, there are a number of patients
who clinically resemble HD despite the presence of ganglion
cells in rectal biopsy. Various terms have been used to describe
these conditions: “chronic idiopathic intestinal pseudo-
obstruction,” “pseudo-Hirschsprung disease,” “neonatal intes-
tinal pseudo-obstruction,” and “intestinal hypoperistalsis
syndrome.” There is a growing interest to further investigate
these relatively rare functional bowel disorders because the di-
agnosis and management of these patients continues to be a
challenge for clinicians. Over the past 3 decades our group
has focused its research interest into delineating variant HD
on the basis of specific histochemical, immunohistochemical,
and electron microscopic studies.
Between 1981 and 2009, full-thickness bowel biopsy or
resected surgical specimens from 178 patients (75 boys and
103 girls) with clinical symptoms suggesting HD were exam-
ined in our research laboratory. Their ages ranged from 1 day
to 9 years. One hundred and three patients had onset of
symptoms in the neonatal period. Ninety-one patients were
from Ireland, and in 87 patients biopsy material was sent to
our laboratory from other countries. Table 102-1 shows
various functional bowel disorders diagnosed using different
histologic techniques. The vast majority of variant HD cases
include intestinal neuronal dysplasia, isolated hypogang-
lionosis, internal sphincter achalasia, and megacystis-
microcolon-intestinal hypoperistalsis syndrome (MMIHS).
Rarer motility disorders include immature ganglia, absence
of argyrophil plexus, and smooth muscle disorders.
Intestinal Neuronal Dysplasia
------------------------------------------------------------------------------------------------------------------------------------------------
HISTORY AND PATHOGENESIS
Intestinal neuronal dysplasia (IND) was first described by
Meier-Ruge in 1971 as a malformation of enteric plexus.1
The first association between IND and HD was reported by
my colleagues and me in a 5-year-old Arab boy who had
rectosigmoid aganglionosis and IND of descending and trans-
verse colon.2 In 1983 Fadda and colleagues3 classified IND
into two clinically and histologically distinguished subtypes.
Type A, which occurs in less than 5% of cases, is characterized
by congenital aplasia or hypoplasia of the sympathetic inner-
vation and presents acutely in the neonatal period with epi-
sodes of intestinal obstruction, diarrhea, and bloody stools.
Type B is clinically indistinguishable from HD, is characterized
by a malformation of the parasympathetic submucous and
myenteric plexuses, and accounts for more than 95% of cases
of IND.4,5 IND occurring in association with HD is of type B.
Since its original description, little has been written about IND
type A. IND has become synonymous with IND type B.
Many investigators have raised doubts about the existence
of IND as a distinct histopathologic entity.6,7 It has been sug-
gested that the pathologic changes seen in IND may be part of
normal development or may be a secondary phenomenon in-
duced by congenital obstruction and inflammatory disease.6,8
On the other hand, the literature contains several familial
cases of IND suggesting that genetic factors may be involved
in this condition. Martuccielo and colleagues9 observed three
families with multiple IND cases, and Moore and colleagues10
and Kobayashi and colleagues11 reported IND in monozygotic
twins. The strongest evidence that IND is a real entity stems
from the animal models. Hox11L1 is a homeobox gene in-
volved in peripheral nervous system development and is
reported to play a role in the proliferation or differentiation
of neural crest cell lines. Two different Hox11L1 knockout
mouse models have been generated. In both cases, homozy-
gous mutant mice were viable but developed megacolon at
the age of 3 to 5 weeks. Histologic and immunohistochemical
analysis showed hyperplasia of myenteric ganglia, a pheno-
type similar to that observed in human IND type B.12,13
1279
1280 PART VII ABDOMEN
TABLE 102-1
Variant Hirschsprung Disease (1981-2009)
Intestinal neuronal dysplasia
Isolated hypoganglionosis
Internal sphincter achalasia
Megacystis-microcolon-intestinal hypoperistalsis syndrome
Rarer intestinal motility disorders
Total 178
However, the mutation screening of this gene in 48 patients
with IND did not show any sequence variant, either causative
missense mutation or neutral substitution.14
In 2002 Von Boyen and colleagues15 reported abnormali-
ties of the enteric nervous system in heterozygous EDNRB-
deficient rats resembling IND in humans. They showed that
a heterozygous 301-base-pair deletion of the EDNRB gene
led to abnormalities of the enteric nervous system. Malfor-
mations of the enteric nervous system observed in þ/sl rats
included hyperganglionosis, giant ganglia, and hypertrophied
nerve fibers in the submucous plexus resembling the his-
topathologic features of IND type B in humans. These findings
support the concept that IND may be linked to a genetic
defect. However, no mutations of the EDNRB gene were
detected in a small series of IND patients.16
INCIDENCE
Intestinal neuronal dysplasia is the most commonly encoun-
tered variant of HD.5 The incidence of isolated IND has varied
from 0.3% to 40% of all suction rectal biopsies in different
centers.5,17 The incidence varies considerably among different
countries. IND immediately proximal to a segment of agan-
glionosis is not uncommon and often presents as persistent
obstructive symptoms after a pull-through operation for
HD.18 Some investigators have reported that 25% to 35% of
patients with HD have associated IND.3 However, others have
rarely encountered IND in association with HD. The uncer-
tainty regarding the incidence of IND has resulted from the
considerable confusion regarding the essential diagnostic
criteria. The diagnostic difficulty is centered on a wide
variability encountered in the literature, not only in terms
of age of the patient, the type of specimen examined, and
the stains performed but also in the diagnostic criteria used.
CLINICAL PRESENTATION
The characteristic clinical pattern of IND can be found in in-
fants younger than 1 year old with a history of constipation
and abdominal distention, thus mimicking HD. Montedonico
and colleagues19 classified IND according to the severity of
histochemical changes in rectal biopsies. The criteria used
for the diagnosis of severe IND included hyperplasia of sub-
mucous plexus, giant ganglia, ectopic ganglia, and increased
acetylcholinesterase (AChE) activity in the lamina propria or
around submucosal blood vessels. Any biopsy that showed
giant ganglia of the submucous plexus with only one of the
other criteria was considered mild. Montedonico and col-
leagues reported that the patients with severe IND begin
their symptoms at an earlier age than those with mild IND
(5.2  112 months vs. 17.5  23 months).
The incidence of associated anomalies in IND has been
reported to be between 25% and 30%.9,19 The common asso-
76
ciated anomalies include anorectal malformations, intestinal
18
malrotation, MMIHS, congenital short bowel, hypertrophic py-
46
loric stenosis, necrotizing enterocolitis, and Down syndrome.
12
26 DIAGNOSTIC CRITERIA
Since the first description in 1971, most controversy sur-
rounding IND has been regarding which histologic diagnostic
criteria are required for definitive diagnosis. The presently rec-
ognized diagnostic criteria, previously reported by our group
and supported by others, are hyperganglionosis and giant
ganglia, in addition to the presence of at least one of the fol-
lowing on suction rectal biopsy: ectopic ganglia in the lamina
propria and increased AChE-positive nerve fibers around sub-
mucosal blood vessels and in the lamina propria.4,5,17,20 How-
ever, Lumb and Moore21 believe that giant ganglia can be a
normal feature in normal bowel, having identified them in
segments of adult bowel removed during surgery for colorectal
carcinoma. To overcome the confusion in diagnostic criteria,
Borchard and colleagues4 produced guidelines for identifying
IND in mucosal rectal biopsies. These comprised two obliga-
tory criteria (hyperplasia of the submucous plexus and an in-
crease in AChE-positive nerve fibers around submucosal
blood vessels) and two additional criteria (neuronal heteroto-
pia and increased AChE activity in the lamina propria). In our
experience, hyperganglionosis and giant ganglia are the most
important features for the diagnosis of IND in suction rectal
biopsies, except in the newborn, when hyperganglionosis is
a normal finding.22
DIAGNOSIS
Suction rectal biopsy is the principal method for the diagnosis
of disorders of intestinal innervation. It is necessary to include
a sufficient amount of submucosa in the suction biopsy
specimens. Traditionally, hematoxylin and eosin and AChE
histochemical staining (Figs. 102-1 and 102-2) in suction rec-
tal biopsy specimens have provided the basis for the diagnos-
tic evaluation of IND. However, there has been discussion
about whether AChE histochemistry is sufficient for the accu-
rate diagnosis of IND and other additional staining techniques
have been proposed. Meier-Ruge and colleagues suggest
lactate dehydrogenase histochemistry.5,23 In my laboratory,
neuronal markers currently being used are reduced nicotin-
amide-adenine dinucleotide phosphate (NADPH) diaphorase
histochemistry and immunohistochemistry using antibodies
raised against neural cell adhesion molecule, protein gene
product 9.5 (PGP9.5), S-100, peripherin, and synaptophy-
sin.5,17 Ganglion-cell counting can be difficult. Standard
immunohistochemical techniques, using antibodies raised
against neuron-specific enolase or PGP9.5, which are com-
monly used to display the enteric nervous system, are less suit-
able for cell counting because they stain not only the cell
bodies but also the axonal processes. Cuprolinic blue staining
has been proposed as the method that stains the largest num-
ber of ganglion cells.5 Furthermore, this method stains only
the cell bodies and not the axons, which makes it relatively
easy to distinguish the individual cells.
Barium enema in IND does not show any specific radio-
logic features other than rectosigmoid distention. Similarly,
 A B

C
FIGURE 102-1 Acetylcholinesterase staining of suction rectal biopsy. A, Normal biopsy showing a submucosal ganglion. B, Biopsy from a patient with
intestinal neuronal dysplasia showing hyperganglionosis and giant ganglia. C, Ectopic ganglia in the lamina propria and increased AChE activity in the
lamina propria in a biopsy specimen from a patient with intestinal neuronal dysplasia.

A B
FIGURE 102-2 Hematoxylin and eosin staining of suction rectal biopsy from a patient with intestinal neuronal dysplasia. A, Hyperganglionosis. B, Giant
ganglia.
1282 PART VII ABDOMEN
anorectal manometry may show the rectosphincteric reflex to
be present, absent, or atypical.
CORRELATION BETWEEN HISTOLOGIC
FINDINGS AND CLINICAL SYMPTOMS
Several investigators have raised doubts about the existence of
IND as a distinct histopathologic entity.6,7 One reason for this
has been the weak correlation found between the severity
of clinical symptoms and the histologic findings. It has been
suggested that the histologic findings of IND can be a part
of normal development or secondary to prolonged constipa-
tion. Recently, Montedonico and colleagues19 correlated spe-
cific clinical, radiologic, and manometric findings with the
severity of histopathologic findings in 44 patients with IND
treated at their institution over 20 years. They classified
IND according to the severity of histochemical alteration into
two groups: mild IND and severe IND. The criteria used for
the diagnosis of severe IND included hyperplasia of the sub-
mucous plexus, giant ganglia with more than seven ganglion
cells, increased AChE activity in the lamina propria or sur-
rounding submucosal blood vessels, and heterotopic neuronal
cells in the lamina propria. Any biopsy specimen that showed
giant ganglia of the submucous plexus with only one of the
other elements was considered mild IND. According to their
results, the characteristic clinical pattern of severe IND can be
found in infants younger than 1 year old with a history of con-
stipation and abdominal distention, thus mimicking HD, with
absence of internal sphincter relaxation in anorectal manom-
etry but who have a normal barium enema. The median age at
presentation of severe IND was 5 months, and similar results
have been reported in other series.3 These authors found a
correlation between the histologic severity of IND and the
clinical symptoms, suggesting that IND is a distinct entity. Al-
though many cases of IND are clinically indistinguishable
from HD, barium enema findings in IND are often equivocal
or show slight to moderate rectosigmoid distention but lack
the typical narrow segment of aganglionosis.
MANAGEMENT
Current treatment of IND type B is in the first instance con-
servative, consisting of laxatives and enemas. In the majority
of patients the clinical problem resolves or is manageable in
this way because maturation of nerve cells is often observed
in IND. If bowel symptoms persist after at least 6 months of
treatment, internal sphincter myectomy should be con-
sidered. Resection and a pull-through operation are rarely
indicated in IND. The indication for pull-through should
not be determined on the basis of histopathologic findings
alone; rather, the decision must be based on the individual
patient’s clinical symptoms.
OUTCOME
Gillick and colleagues24 reported results of treatment in
33 patients with IND observed for periods ranging from
1 to 8 years (mean 2.4 years). Twenty-one (64%) patients
had a good response to conservative management and cur-
rently have normal bowel habits. Twelve patients (36%)
underwent internal sphincter myectomy after failed conserva-
tive management. Seven of these patients now have normal
bowel habits. Two patients were able to stay clean with regular
enemas. Three patients who continued to have persistent con-
stipation after myectomy and underwent resection of redun-
dant and dilated sigmoid colon now had normal bowel habits.
Isolated Hypoganglionosis
------------------------------------------------------------------------------------------------------------------------------------------------
Isolated hypoganglionosis (IH) has been classified as a “hypo-
genetic type” of intestinal innervation disorders.4 Clinically,
IH resembles classical HD: Patients present with severe consti-
pation or pseudo-obstruction. Due to the fact that IH is one of
the rarest subtypes of intestinal innervation disorders, ac-
counting for only 5% of all cases,4 the number of reported
cases in the literature is limited. Dingemann and Puri recently
reviewed 92 patients with IH reported in the English literature
between 1978 and 2008 and critically analyzed the current
state of the epidemiologic, diagnostic, and therapeutic features
of this rare neuronal intestinal disorder.25
EPIDEMIOLOGY
In the reported review the overall male-to-female ratio of IH
was 3:1. This is similar to the overall male-to-female ratio
of HD, which is commonly considered to be 4:1. Although
29 (32%) patients were diagnosed in the newborn period,
the median age at diagnosis of the reported patients with IH
was 4.85 years. This is due to the fact that, in some patients,
diagnosis was made as late as the age of 17 years. This differs
significantly from HD. Although late diagnosis is also possible
in HD, more than 90.6% of patients are diagnosed in the
newborn period. This late diagnosis of IH might reflect the
difficulties in diagnosing IH in rectal suction biopsies but also
the rareness of the disease.
CLINICAL PRESENTATION
Even though the median age at diagnosis is significantly
higher in patients with IH than in those with HD, the symp-
toms of IH resemble classical aganglionosis. The symptoms
reported in all IH included intractable constipation, ileus,
and enterocolitis. As in HD, enterocolitis of the newborn
remains the most serious complication of IH, as 6 of the
7 reported IH related deaths were neonates with enterocolitis.
HISTOPATHOLOGIC APPROACH
The diagnosis of IH by means of rectal suction biopsy is
difficult.26 As recommended by the interdisciplinary consen-
sus conference,27 a full-thickness bowel is essential for the di-
agnosis of IH. Besides standard hematoxylin and eosin
staining, histochemical evaluation of AChE was performed
by 91% of the authors in the review by Dingemann and Puri
to visualize the characteristic changes for IH such as low
mucosal activity of AChE, deficiency of nerve cells in the
myenteric plexuses, and hypertrophy of muscularis mucosae
and circular muscle layers. Morphometric measurements in
IH using AChE staining, as suggested by Meier-Ruge and
colleagues,23 represent one of the cornerstones of diagnostic
criteria for the disease. It has been shown that plexus area
and nerve cell number are dramatically decreased, and the
distance between ganglia is almost doubled.
 CHAPTER 102 INTESTINAL DYSGANGLIONOSIS AND OTHER DISORDERS OF INTESTINAL MOTILITY 1283

A B
FIGURE 102-3 Nicotinamide-adenine dinucleotide phosphate—diaphorase staining of the myenteric plexus. A, Normal colon showing normal myen-
teric ganglia and a large number of nitrergic nerve fibers in the circular and longitudinal muscle. B, Rectal biopsy in isolated hypoganglionosis patient
showing small myenteric ganglia and reduced nitrergic innervation. (See Expert Consult site for color version.)

Several additional neuronal markers have been used to Internal Anal Sphincter
diagnose IH. Rolle and colleagues used NADPH diaphorase
staining, which not only allows one to characterize the nitrer- Achalasia
gic innervation of the muscle but also differentiates mature ------------------------------------------------------------------------------------------------------------------------------------------------

fully developed ganglia from the immature ganglia in IH
(Fig. 102-3).
SURGICAL MANAGEMENT
According to literature, the treatment of hypoganglionosis is
similar to HD involving resection of the affected segment
and pull-through operation. In the review of 92 patients with
IH reported by Dingemann and Puri, 54 patients received
resection of affected bowel and pull-through procedures of
different types; 11 underwent ileostomy, colostomy, or jeju-
nostomy alone; and 2 underwent sphincter myectomy.
In 25 patients, the operative treatment was not clearly stated.
The treatment of choice remains resection of the affected
segment. The exact method used must always be tailored to
the extent of affected bowel, the localization of the disease,
and presumably it will also depend on the surgeon’s pref-
erence in this rare disease.
OUTCOME AND COMPLICATIONS
The overall mortality of the patients included in this review
was 8%. Six of the seven patients who died were newborns
suffering from severe enterocolitis. The other patient died
due to total parenteral nutrition–associated complications
during follow-up. During a postoperative follow-up of 7
months to 12 years, typical complications reported were sim-
ilar to those in HD. Enterocolitis, chronic constipation, over-
flow encopresis, and the need for redo pull-through for
residual disease were reported.
The internal anal sphincter (IAS), a specialized smooth muscle
continuation of the circular muscle layer of rectum, plays a
significant role in the maintenance of anorectal continence
and in the pathophysiology of incontinence and consti-
pation.28 The IAS relaxes in response to rectal distension, a
phenomenon called the rectosphincteric inhibitory reflex, which
is mediated by intramural nerves descending from the rectum
to the IAS. The IAS receives adrenergic, cholinergic, and non-
adrenergic noncholinergic (NANC) innervations.29 Several
investigators have reported that IAS relaxation is brought
about by the activation of intramural NANC nerves. Nitric
oxide (NO) is now recognized as a potent mediator of non-
adrenergic noncholinergic inhibitory nerves, which regulate
smooth muscle relaxation in the mammalian gastrointestinal
tract including IAS.
Internal anal sphincter achalasia (IASA) is a clinical
condition with presentation similar to HD but with the pres-
ence of ganglion cells on suction rectal biopsy. The diagnosis
of IASA is made on anorectal manometry, which shows the
absence of rectosphincteric reflex on rectal balloon inflation.
Previously, IASA has been referred to as ultrashort segment
HD. The ultrashort segment HD, which is a rare condition,
is characterized by an aganglionic segment of 1 to 3 cm long
and normal acetylcholinesterase (AChE) activity in the lamina
propria and increased AChE activity in the muscularis muco-
sae.30 Many patients who are considered to have ultrashort
HD on abnormal anorectal manometric findings show pres-
ence of ganglion cells and normal acetylcholinesterase (AChE)
activity in suction rectal biopsies. Many investigators have
therefore suggested that the term IASA is more suitable
1284 PART VII ABDOMEN
because it reflects more accurately failure of relaxation of
the internal sphincter, which is the causative factor in this
condition.31
INCIDENCE
The exact incidence of isolated internal anal sphincter
achalasia is not known. De Caluwe and colleagues30 reported
an incidence of 4.5% among 332 children who were
investigated for severe chronic constipation.
PATHOGENESIS
The exact pathogenesis and pathophysiology of IASA is not
fully understood. Altered intramuscular innervation has been
reported in IASA, and this is believed to be responsible for
the motility dysfunction seen in these patients. Hirakawa and
colleagues32 reported absence of nitrergic innervation within
the IAS muscle in patients with IASA and suggested that
nitrergic nerve depletion may play an important role in the
development of IASA (Fig. 102-4). Because nitrergic nerves
regulate smooth muscle relaxation, their deficiency or absence
in IASA may be responsible for the spasm or increased tone in
the IAS in these patients.33 Oue and Puri reported defective
innervation of the neuromuscular junction (NMJ) of the IAS
in patients with IASA. If the NMJ is abnormal, the neuro-
transmitter chemicals synthesizing neurons cannot be trans-
mitted to muscle cells, thereby causing motility dysfunction.29
Altered distribution of c-kit positive interstitial cells of Cajal
(ICCs) has been reported in the internal sphincter of patients
with internal sphincter achalasia, which may further contribute
to motility dysfunction in these patients. Piotrowska and col-
leagues reported deficiency or absence of ICCs in the IAS myect-
omy specimens obtained from patients with IASA at the time of
internal sphincter myectomy.34 The functions of ICCs include the
generation of electrical pacemaker activity, generation of slow
A B
FIGURE 102-4 Nicotinamide-adenine dinucleotide phosphate–diaphorase staining of the internal sphincter. A, Normal specimen showing a large num-
ber of nitrergic nerve fibers. B, Internal sphincter achalasia patient showing nitrergic nerve depletion in the sphincter muscle. (See Expert Consult site for
color version.)
waves, and neurotransmission between the enteric nervous sys-
tem and smooth muscle cells. It has been proposed that ICCs in
certain regions of the gut may not act as pacemakers, but as stretch
receptors.32,35 The lack or deficient expression of NO and ICCs
in the IASA may lead to defective generation of nitric oxide–
mediated pacemaker activity causing motility dysfunction.
DIAGNOSIS
Patients with IASA have clinical presentation similar to HD.
The vast majority of patients present with severe constipation
with or without soiling. About one third of the patients give
history of abdominal distension. Laxatives usually fail to im-
prove constipation in patients with IASA. Definite diagnosis
of IASA is based on anorectal manometry, which shows
absence of rectosphincteric reflex on rectal balloon inflation
and the presence of marked rhythmic activity of the internal
anal sphincter presence of ganglion cells and normal acetyl-
cholinesterase activity in the suction rectal biopsy.
TREATMENT
Posterior internal sphincter myectomy has been recom-
mended as the treatment of choice for patients with internal
sphincter achalasia. The myectomy is performed posteriorly
starting at the level of the pectinate line, and a 5- to 10-mm
wide strip of smooth muscle is resected extending proximally
for varying lengths ranging from 15 to 50 mm.30 De Caluwe
and colleagues reported bowel function in 15 consecutive pa-
tients with IASA 2 to 6 years after posterior internal sphincter
myectomy. At the time of follow-up, seven patients had regular
bowel motions and were not on any laxatives. Six patients had
normal bowel habits but were on small doses of laxatives. One
patient was able to stay clean with a regular enema regimen.
One patient required resection of dilated and redundant sig-
moid colon and now has normal bowel habits with laxatives.
 CHAPTER 102 INTESTINAL DYSGANGLIONOSIS AND OTHER DISORDERS OF INTESTINAL MOTILITY
Recently Heikkinen and colleagues36 reported long-term
follow-up in 10 IASA patients, 7 to 16 years after internal
sphincter myectomy. Three of the ten patients in their study
needed laxatives at the time of follow-up, and one patient
had required resection of the dilated and redundant rectosig-
moid 2½ years after myectomy. The remaining six patients had
one to two bowel motions daily without the help of laxatives.
Four of their patients suffered from soiling-related social
problems.
Recently intrasphincteric injection of botulinum toxin has
been used to treat patients with IASA. This is thought to work
by interfering with the Ach release at the neuromuscular junc-
tion (NMJ) and thus inhibiting sympathetic stimulation to the
IAS. Intrasphincteric injection is thought to produce a local
and transient chemical denervation of the sphincter. This
treatment modality has been found to be safe, but the effects
are transient.37 The IAS is injected in four quadrants, 25 i.u.
per quadrant at the level of the dentate line. Irani and
colleagues injected Botox in 24 patients with IASA and found
that the duration of response following Botox injection
was variable, ranging from 1 month to longer than 1 year.
Foroutan and colleagues38 compared efficacy of intrasphinc-
teric Botox injection and internal sphincter myectomy in
patients with IASA and found that Botox injection was as effec-
tive as myectomy. Further studies with longer follow-up are
necessary to determine the effectiveness of this treatment
modality in the management of internal sphincter achalasia.
MEGACYSTIS-MICROCOLON-INTESTINAL
HYPOPERISTALSIS SYNDROME
MMIHS is a rare congenital and generally fatal cause of
functional intestinal obstruction in the newborn. This syn-
drome is characterized by abdominal distension caused by
a distended nonobstructed urinary bladder, microcolon,
and decreased or absent intestinal peristalsis.27 Usually in-
complete intestinal rotation and shortened small bowel are
associated.
PATHOGENESIS
MMIHS was first described in 1976 by Berdon and colleagues27
and, to date, 182 cases have been reported in the literature.39
The etiology of this syndrome remains unclear. Several hypo-
theses have been proposed to explain the pathogenesis of
MMIHS: genetic, neurogenic, myogenic, and hormonal origin.
Histologic studies of the myenteric and submucosal
plexuses of the bowel of MMIHS patients have found normal
ganglion cells in the majority of the patients, decreased in
some, hyperganglionosis, and giant ganglia in others.39 Re-
cently, Piotrowska and colleagues34 reported absence of inter-
stitial cells of Cajal (ICCs) in the bowel and urinary bladder of
patients with MMIHS. ICCs are pacemaker cells that assist
active propagation of electrical events and neurotransmission,
and their absence may result in hypoperistalsis and voiding
dysfunction in MMIHS. Puri and colleagues40 showed, in
1983, vacuolar degenerative changes in the smooth muscle
cells (SMCs) with abundant connective tissue between muscle
cells in the bowel and bladder of patients with MMIHS and
suggested that a degenerative disease of smooth muscle cells
could be the cause of this syndrome. Several subsequent
reports have confirmed evidence of intestinal myopathy in
1285
MMIHS. Other investigators have reported absence or marked
reduction in a-smooth muscle actin and other contractile
and cytoskeletal proteins in the smooth muscle layers of
MMIHS bowel.41 Contractile and cytoskeletal proteins are im-
portant structural and functional components of SMCs and
play a vital role in the interaction of the filaments in smooth
muscle contraction.
PRENATAL DIAGNOSIS
Puri and Shinkai reviewed 182 cases of MMIHS reported in
the literature.39 In 54 cases ultrasound findings associated
with MMIHS were described. The most frequent finding
was enlarged bladder (88%), with hydronephrosis seen in
31 patients (57%). Normal amniotic fluid volume was
revealed in 32 cases (59%), increased volume occurred in
18 (33%), and volume decreased in 4 (7%). In three cases
(5%) abdominal distention caused by dilated stomach was
detected. Three cases of oligohydramnios during the second
and early third trimesters were reported, probably related to
the functional bladder obstruction.
Serial obstetrical ultrasonography showed that the earliest
finding in MMIHS is enlarged bladder, detectable from
16 weeks of gestational age. A later finding is hydronephrosis,
caused by the functional obstruction of the bladder. Usually
polyhydramnios develops late, appearing during the third
trimester.
CLINICAL PRESENTATION
Of the 182 cases reported in the literature, sex of the patient
was mentioned in 149 patients. Ninety-eight were females and
43 were males. In four cases, pregnancy was terminated after
ultrasonography detected MMIHS, which was confirmed at
autopsy in all cases. The duration of pregnancy was reported
in 98 cases. Fifty-eight patients (59%) were born at term,
25 (25.5%) at 36 to 39 weeks of gestation, 12 (12%) at
32 to 35 weeks, and 3 (3%) at 31 weeks and less. Dystocia
delivery caused by abdominal distention was reported in
eight cases. In four cases Caesarean section was required,
and in four cases the bladder was so distended that the baby
could only be delivered vaginally after removal of 250, 500,
650, 500 mL of urine, respectively, from fetal bladder by
paracentesis. The mean birth weight was normal (3 kg) for
gestational age.
The clinical symptoms of MMIHS are similar to other
neonatal intestinal obstructions. Characterized by abdominal
distention, bile-stained vomiting, and absent or decreased
bowel sounds, abdominal distention was a constant and early
finding. A consequence of the distended, nonobstructed
urinary bladder was relieved by catheterization. Of 182 cases
61 had bilious vomiting, and failure to pass meconium was
clearly reported in only 23 cases. The majority of patients were
not able to void spontaneously.
In the review by Puri and Shinkai,39 19 sets of siblings
affected with MMIHS were reported. Eighteen families had
two affected siblings and one had three. Four sets of affected
siblings occurred to consanguineous parents. In another case
an affected child was born to a member of the family reported
by Penman, and consanguinity was also present in these
parents. In three further cases an elder sibling of the affected
child died just after birth because of intestinal obstruction30 or
1286 PART VII ABDOMEN
multiple abnormalities; in another case a sibling of the patient
was affected by prune-belly syndrome. The occurrence of
MMIHS in 19 sets of affected siblings and consanguinity in
four sets of parents suggest an autosomal recessive pattern
of inheritance.
RADIOLOGIC FINDINGS
In the vast majorities of the 182 cases, radiologic evaluation
usually suggested the diagnosis of MMIHS. Plain abdominal
films showed either dilated small bowel loops or a gasless
abdomen with evident gastric bubble. An enlarged urinary
bladder was present in all patients who had cystography or ul-
trasonography (Fig. 102-5). Intravenous urography or ultra-
sonography detected unilateral or bilateral hydronephrosis
in 84 patients. Forty-four patients had an upper gastrointesti-
nal series both before and after laparotomy: hypoperistalsis or
aperistalsis in stomach, duodenum, and small bowel was a
constantly detected symptom. In three cases reverse peristalsis
from small bowel into the stomach was also observed. In two
cases hypoperistalsis was associated with gastroesophageal
reflux, and in one case the esophagus was aperistaltic. Barium
enema showed microcolon in all 71 patients in whom this
study was performed; in 39 cases malrotation was associated
(see Fig. 102-5).
SURGICAL OR AUTOPSY FINDINGS
Megacystis and microcolon were the two most frequent find-
ings at surgery or autopsy and were present in all patients. In
the review by Puri and Shinkai,39 short-bowel syndrome was
found in 37 cases, dilated proximal small bowel in 19, seg-
mental stenosis of the small bowel in 3, duodenal web in 1,
and Meckel diverticulum in 1. Malrotation was found in a total
A B
FIGURE 102-5 Megacystis-microcolon-intestinal hypoperistalsis syndrome patient. A, Voiding cystourethrogram showing enlarged urinary bladder.
B, Contrast enema showing microcolon.
of 81 cases. Although surgical management was not men-
tioned in several reports, 93 patients (70%) underwent one
or more surgical procedures. A variety of interventions were
performed: gastrostomy, jejunostomy, ileostomy, cecostomy,
segmental resection of jejunum and ileum, lysis of adhesions,
and internal sphincter myectomy. Surgical manipulation of
the gastrointestinal tract has generally been unsuccessful,
and in most patients total parenteral nutrition was required.
In 37 patients vesicostomy was performed to decompress
the urinary tract and to preserve renal function.
HISTOLOGIC FINDINGS
Histologic studies of the myenteric and submucous plexuses
were reported in 93 out of 182 cases. In 72 the ganglion cells
were normal in appearance and number, and in the remaining
21 cases the various neuronal abnormalities reported included
hypoganglionosis, hyperganglionosis, and immature ganglia.
The majority of reports do not mention the histologic
findings in the muscle layers of bowel and bladder wall.
Nevertheless, some authors found significant abnormalities
in SMCs. In nine cases thinning of the longitudinal muscle
was found on light microscopy. Electron microscopy showed
vacuolar degeneration in the center of the smooth muscle of
the bowel in 11 cases and of the bladder in 8 cases. Connective
tissue proliferation was found in the bowel in nine cases and
in the bladder in eight cases. In three more cases the blad-
der showed elastosis. In two patients electron microscopy
revealed vacuolar degeneration of smooth cells in the muscle
layers of the bowel and the bladder in addition to neuronal
abnormalities. Other investigators have reported absence
or marked reduction in a-smooth muscle actin and other
contractile and cytoskeletal proteins in the smooth muscle
layers of MMIHS bowel.41
 CHAPTER 102 INTESTINAL DYSGANGLIONOSIS AND OTHER DISORDERS OF INTESTINAL MOTILITY
OUTCOME
The management of patients with MMIHS is frustrating.
A number of prokinetic drugs and gastrointestinal hormones
have been tried without success. Surgical manipulation of the
gastrointestinal tract has generally been unsuccessful. The
outcome of this condition is generally fatal: Only 23 of the
182 reported patients were alive, the oldest being 18 years
1287
old. Twenty-one of the 23 patients were being maintained
by total or partial parenteral nutrition. The need for surgical
intervention should be made carefully and individualized,
in that most explorations have not been helpful and are
probably not necessary.
The complete reference list is available online at www.
expertconsult.com.
 exposure eventually led to the posterior sagittal anorecto-
plasty,15,16 which was rapidly adopted because it allowed full vi-
sualization of the sphincteric complex and more clearly showed
the relationship of the rectum to the urologic system and the sur-
rounding structures.

Incidence
------------------------------------------------------------------------------------------------------------------------------------------------

No specific cause of anorectal malformation has been described.

The average incidence worldwide is 1 in 5000 live births,17
although the condition is more common in certain geographic
areas. Some families have a genetic predisposition, with anorec-
tal malformations being diagnosed in succeeding generations.18
In addition, imperforate anus occurs in association with several
syndromes.19–20 A slight male preponderance exists.
The most common defect in females is rectovestibular
fistula, whereas the most common defect in males is rectoure-
thral fistula. Cloacal malformations are more common than for-
merly thought, most likely because they were previously
misdiagnosed as rectovaginal fistulas.21 Imperforate anus with-
out fistula occurs in 5% of patients, but interestingly, half of
them also have Down syndrome. Patients with Down syndrome
CHAPTER 103 and anorectal malformations have this type of defect 95% of
the time.20

Anorectal Classification
------------------------------------------------------------------------------------------------------------------------------------------------

The classification system shown in Table 103-1 is anatomically

Malformations descriptive with therapeutic and prognostic implications.

Marc A. Levitt and Alberto Peña Embryology

------------------------------------------------------------------------------------------------------------------------------------------------

The cloaca in the embryo is a cavity into which opens the

hindgut, tailgut, allantois, and later, the mesonephric ducts.
History
------------------------------------------------------------------------------------------------------------------------------------------------
The cloaca is first formed at around 21 days’ gestation; it is
U shaped, with the allantois lying anteriorly and the hindgut
Throughout history, surgeons have attempted to treat infants posteriorly. The septum in the middle grows downward and
and children with anorectal malformations. The written fuses with lateral folds (Rathke plicae) until it joins the cloacal
accounts describe few patients, so it is likely that most patients membrane. In this 6-week process, a urogenital cavity anteri-
died without treatment.1–2 Successful maneuvers involved orly and an anorectal cavity posteriorly are created. Rapid
rupturing an obstructing membrane with a finger or the point growth of the genital tubercle changes the shape of the cloaca
of a knife3 and evolved to making an incision to find the bowel and the orientation of the cloacal membrane, which is
in the perineum4 or in the hollow of the sacrum. The first in- displaced posteriorly. The cloacal membrane breaks down at
guinal colostomy was performed in 1783,5–6 but most infants 7 weeks’ gestation, thereby creating two openings: the urogen-
died, and colostomy was thought of as a method of last resort. ital and anal openings.22–24 The muscles that surround the
In 1835 Amussat mobilized the bowel through a perineal rectum develop at the same time and are seen in the sixth
incision and sutured it to the skin, thus performing the first and seventh weeks of gestation,25 and by the ninth week,
anoplasty7 and avoiding the strictures that were common with all relevant structures are in place. At this stage, differentiation
earlier procedures. In the 1700s and 1800s, several authors into male or female external genitalia has not yet occurred.
suggested that the peritoneum be opened if the bowel was not
encountered from below.8–11 In the mid-1900s, single-stage
abdominoperineal procedures became popular12–13 and usually Associated Malformations
involved resection of the rectosigmoid. Shortly after these ------------------------------------------------------------------------------------------------------------------------------------------------

reports, Stephens described a procedure that emphasized pas- Most babies (50% to 60%) with anorectal malformations have
sage of the rectum within the puborectalis sling.14 Until the early one or more abnormalities that affect other systems.26 Higher
1980s, this approach with its modifications was used, but it abnormalities are associated with more malformations. Many
involved blind dissection near the posterior urethra. A pro- are incidental findings, but others, such as cardiovascular
gressive increase in the length of the perineal incision to gain defects, may be life threatening.

1289
1290 PART VII ABDOMEN

TABLE 103-1
Classification of Anorectal Malformations A
Males Females
Perineal fistula Perineal fistula
Rectourethral fistula Vestibular fistula
Bulbar Persistent cloaca
B
Prostatic 3 cm common channel
Rectobladder neck fistula >3 cm common channel
Imperforate anus without fistula Imperforate anus without fistula
Rectal atresia Rectal atresia C
Complex defects Complex defects

CARDIOVASCULAR ANOMALIES
Cardiovascular anomalies are present in approximately one A Normal ratio: BC/AB = 0.74
third of patients,27–29 but only 10% of these require treatment.
The most common lesions are atrial septal defect and patent A
ductus arteriosus, followed by tetralogy of Fallot and ventri-
cular septal defect. Transposition of the great vessels and
hypoplastic left heart syndrome are rare. B

GASTROINTESTINAL ANOMALIES
A wide spectrum of associated gastrointestinal anomalies has
C
been described. Tracheoesophageal abnormalities occur in
about 10% of cases.26–27 Duodenal obstruction caused by
atresia or malrotation has been reported to have an incidence
of 1% to 2%.27,30 Hirschsprung disease has been found in
only 3 patients in our series of 2100. Overdiagnosis of this as-
sociation is probably due to the high incidence of constipation
seen in patients with anorectal malformations.31

SPINAL, SACRAL, AND VERTEBRAL B Normal ratio: BC/AB = 0.77

ANOMALIES
Lumbosacral anomalies such as hemivertebrae, scoliosis,
butterfly vertebrae, and hemisacrum are common. The most
frequent spinal problem is tethered cord,32–35 but spinal
lipomas, syringomyelia, and myelomeningocele are also
commonly seen. The more complex the anorectal anomaly is,
the more likely is the presence of an associated spinal and verte-
bral anomaly. These problems are found by ultrasound or mag-
netic resonance imaging in one third to half of patients with
anorectal malformations.4,36–38 It is still unclear how many of
these lesions are clinically significant and which lesions require
prophylactic surgery.35,39 A sacral defect, particularly hemisa-
crum, in association with imperforate anus and a presacral mass
is known as the Currarino triad40 and has a strong familial ten-
dency.41–43 The sacral ratio is a valuable prognostic tool
(Fig. 103-1) because it quantifies the degree of sacral hypodeve-
lopment. Patients with ratios less than 0.4 are universally incon-
tinent. Ratios that approach 1.0 usually predict a good prognosis.
GENITOURINARY ANOMALIES
Urologic abnormalities predominate.44–45 Improved imaging
has shown an increasing range of upper and lower urinary
tract anomalies. The incidence ranges from one third to half
of patients in several series and increases with increasing
complexity of anorectal defect.30,46 Vesicoureteric reflux is
the most common anomaly found, and renal agenesis and
FIGURE 103-1 Anterior-posterior (A) and lateral (B) sacral ratios.
dysplasia are the next most frequent findings.46–47 Cryptor-
chidism is reported in up to 20% of males with imperforate
anus,47–48 and hypospadias occurs in approximately 5%.48
GYNECOLOGIC ANOMALIES
Associated gynecologic anomalies have emerged as a significant
concern in patients with anorectal malformations.33,49–50 In
the newborn period, hydrocolpos can lead to urinary obstruc-
tion or can cause pyocolpos. Müllerian anomalies may become
manifest later when teenagers have obstruction of menstrual
flow. Large intra-abdominal collections and peritonitis can de-
velop, or patients may have amenorrhea because of absence of
the müllerian structures. Asymmetric obstruction to menstrual
flow can occur as a result of an anatomic abnormality at any
level such as an obstruction of the tube, atresia of the cervix,
or a blind hemivagina.33,51 Uterine malformations (predomi-
nantly bicornuate uterus and uterus didelphys), vaginal abnor-
malities (particularly a vaginal septum), and associated vaginal
atresia33,49 occur in approximately one third of patients.
A search for hydrocolpos needs to be performed in the newborn
period. During definitive repair or at the time of colostomy
closure, inspection of the intra-abdominal gynecologic struc-
ture is vital to prevent immediate and long-term sequelae.
The full obstetric impact of these anomalies is not yet known.
 CHAPTER 103 ANORECTAL MALFORMATIONS 1291

Anorectal Anatomy and at the appropriate time. Distention of the rectum produces a
vague sensation of fullness or even colicky pain (proprio-
Pathophysiology ception) but does not provide specific information concerning
------------------------------------------------------------------------------------------------------------------------------------------------
the physical characteristic of the contents. Little is known
Bowel control implies the ability to detect and retain flatus and about the mechanism that triggers peristalsis of the rectosig-
stool until the appropriate time for evacuation. It is the result moid to defecate, but it is clear that the degree of rectal fullness
of a complex interplay among sphincter function, anorectal has an important role. Thus when the time comes, the recto-
sensation, and colonic motility. All these factors are affected sigmoid generates waves of peristalsis aimed at emptying the
in children with anorectal malformations.52 lumen. Individuals can restrain this process temporarily by
using their voluntary sphincter. With a conscious decision
made to allow the stool to come out, the sphincter is relaxed
SPHINCTER MECHANISM and the individual waits for the next peristaltic wave. Normal
The muscle groups of the sphincter mechanism form a funnel- defecation allows massive emptying of the rectosigmoid,
like structure in the pelvis. These muscles are innervated by the followed by another resting period of about 24 hours, during
pudendal nerve, both motor to the voluntary muscles and sen- which the rectosigmoid again acts as a reservoir.
sory to the skin around the anus and anal canal. They are derived Children with anorectal malformations have a spectrum of
from the sacral plexus roots S2 to S4, as well as the autonomic rectosigmoid motility disorders. Those patients subjected to
nervous system via the nervi erigentes, from the same segments surgical techniques that preserve the rectosigmoid usually
of the spinal cord. The exposure afforded by the posterior sagittal suffer from constipation. Constipation, one of the most important
approach shows that the junction of the levator musculature functional sequelae, is probably the result of hypomotility of the
with the fibers about the anal dimple is defined by a vertical rectosigmoid, which is self-perpetuating and self-aggravating
group of striated muscle fibers called the muscle complex. Electri- to the point that if left untreated, megasigmoid develops.31
cal stimulation of the upper end of the levator group pulls the In extreme cases, fecal impaction and encopresis, or overflow
rectum forward. Stimulation of the muscle complex (vertical pseudoincontinence, may develop. It seems that constipation
fibers) elevates the anus, and stimulation of the parasagittal fi- is worse with lower defects. Knowing this and the fact that
bers closes the anus. Children with anorectal malformations hypomotility can begin a vicious cycle leading to megasigmoid,
have varying degrees of striated muscle development from it is vital that the clinician avoid the cycle of hypomotility,
almost normal-appearing muscle to virtually complete absence. constipation, and megasigmoid, with aggressive patient follow-
up plus dietary, mechanical, and pharmacologic treatment.
SENSATION AND PROPRIOCEPTION Children with anorectal malformations who have lost their
rectosigmoid suffer from the opposite problem (i.e., tendency
Under normal circumstances, the anal canal is an exquisitely to have diarrhea). These children have no reservoir capacity,
sensitive area. It allows the individual to discriminate solids, are highly sensitive to certain foods, and frequently suffer from
liquids, and gas. The overwhelming majority of children with incontinence.
anorectal malformations is born without an anal canal and
therefore lacks this anal canal sensation. There is, however,
proprioception, which is described as a vague feeling that
Clinical Findings and Initial
is perceived when the rectum is distended, simultaneous with Management
stretching of the voluntary muscle that surrounds the rectum. ------------------------------------------------------------------------------------------------------------------------------------------------

Proprioception seems to be present to a variable degree in most Figure 103-2 shows the decision-making algorithm for the
patients. Thus these patients lose control when they suffer an initial management of male patients.
episode of diarrhea but have the ability to be toilet-trained
when they form solid stool and learn to perceive it.
Newborn male—anorectal malformation
Perineal inspection
COLONIC AND RECTOSIGMOID MOTILITY
* Spine * Sacrum
Normally it takes between 3 and 6 hours for the gastric contents * Kidney U/S * Spinal U/S
20–24 hrs
to reach the small bowel. The intestinal contents reach the cecum * Urinalysis * Cardiac echo
in a liquid state. It then takes about 20 to 24 hours for that fecal * R/O esophageal atresia
Reevaluation and cross-table lateral film
material to reach the rectum and become formed stool with the
absorption of water that occurs in the colon. The rectosigmoid
acts as a reservoir and holds the fecal material for a variable pe-
riod of time. The anal canal (below the pectinate line) is usually Perineal fistula Rectal gas below Rectal gas above
empty because of the action of the surrounding sphincteric coccyx coccyx
mechanism. Occasionally, however, there are peristaltic waves No associated defects Associated defects
Abnormal sacrum
in the colon that push the fecal material toward the anus. The Flat bottom
rectal contents move distally and touch the exquisitely sensitive Anoplasty Consider PSARP
tissue of the anal canal, thereby providing valuable information with or without
related to the nature of the rectal contents (solid, liquid, or gas). colostomy
Depending on the surrounding social circumstances, the Colostomy
individual may contract the sphincteric mechanism to push FIGURE 103-2 Algorithm for the treatment of a male newborn with an
stool or gas back into the rectum and then voluntarily relax anorectal malformation.
1292 PART VII ABDOMEN

A clinician is called to see a newborn male with an anorectal
malformation. First the clinician must perform a thorough
perineal inspection, which usually provides the most impor-
tant clues about the type of malformation that the patient has
(Fig. 103-3). It is important to not make a decision about a
colostomy or a primary operation before 20 to 24 hours of
age. The reason for waiting is that significant intraluminal pres-
sure is required for meconium to be forced through a fistula,
which is the most valuable sign of the location of distal rectum
in these babies. If meconium is visualized on the perineum, it
is evidence of a rectoperineal fistula. If there is meconium in
the urine, the diagnosis of a rectourinary fistula is obvious.
Radiologic evaluations do not show the real anatomy before
24 hours because the rectum is collapsed by the muscle tone
of the sphincters that surround its lower part. Therefore
radiologic evaluations done too early (before 24 hours) will
likely reveal a “very high rectum” and therefore yield a false
diagnosis.
During the first 24 hours, the newborn should receive in-
travenous fluids, antibiotics, and nasogastric decompression
to prevent aspiration. The clinician should use these hours
to evaluate for the presence of associated defects such as car-
diac malformations, esophageal atresia, and urologic prob-
lems. An echocardiogram can be performed, and the baby
should be checked for the presence of esophageal atresia.
A plain radiograph of the lumbar spine and sacrum should
be taken to evaluate for hemivertebrae and sacral anomalies.
A spinal ultrasound helps screen for tethered cord and other
spinal problems. Ultrasonography of the abdomen evaluates
for the presence of hydronephrosis.

A B

FIGURE 103-3 External appearance of the perineum of male patients with anorectal
malformations. A, Perineal fistula with “bucket handle.” B, Rectourethral bulb fistula. C,
C Flat bottom consistent with poor pelvic musculature.
 CHAPTER 103 ANORECTAL MALFORMATIONS 1293

If the baby has signs of a perineal fistula, an anoplasty can

be performed without a protective colostomy in the newborn
period. If the baby is ill from associated problems, is very pre-
mature, or if the clinician chooses to wait until the baby is a
little older, the fistula can be gently dilated. The repair in such
cases should not be delayed more than several months. After
24 hours, if no meconium is seen on the perineum or in the
urine, a cross-table lateral x-ray film with the baby in prone
position should be obtained. If the gas in the rectum is located
below the coccyx (Fig. 103-4) and the baby is in good condi-
tion with no significant associated defects, depending on the
surgeon’s experience, a posterior sagittal operation without
a protective colostomy can be considered. A more conser-
vative alternative would be to perform a colostomy, with the
definitive repair planned for a second stage.
If rectal gas is seen above the coccyx (Fig. 103-5) or the
patient has meconium in the urine, significant associated
defects, and/or an abnormal sacrum or a flat bottom, a colos-
tomy is recommended with postponement of the main repair
for a subsequent operation. This can be performed 2 to 3
months later after a distal colostogram delineating the anat-
omy is performed, provided that the baby is gaining weight
normally.
Performing the definitive repair at 2 to 3 months of age has
important advantages for the patient including less time with an
abdominal stoma, less size discrepancy between the proximal FIGURE 103-5 Cross-table lateral radiograph, high rectum.
and distal bowel at the time of colostomy closure, easier anal
dilation, and no recognizable psychologic sequelae from pain-
ful perineal maneuvers. In addition, at least theoretically,
placing the rectum in the right location early in life may colostogram. While looking for the rectum, the surgeon
provide an advantage in terms of the potential for acquired local might inadvertently find and injure the urethra, an ectopic
sensation. ureter, the bladder neck, the vas deferens, or the seminal
The potential advantages of an early operation must be vesicles.56
weighed against the possible disadvantages of an inexperi- Figure 103-6 shows a decision-making algorithm for the
enced surgeon unfamiliar with the minute anatomic structures initial management of female patients. As in males, the most
of an infant’s pelvis. important step in diagnosis and decision making is the peri-
A trend to repair these defects without a protective colos- neal inspection (Fig. 103-7). The first 24 hours should also
tomy53–55 must be balanced against the concern that such a be used to rule out serious associated defects. Perineal inspec-
repair without a colostomy is done without precise anatomic tion may disclose the presence of a solitary perineal orifice.
information about the patient’s specific type of anorectal This finding establishes the diagnosis of a cloaca. The clinician
defect. The most catastrophic complications seen in patients should know that such a patient has a high likelihood of a uro-
operated on without a colostomy occur in patients in whom logic defect. The presence of hydrocolpos should be ruled out
the surgeon did not have a properly done preoperative distal by ultrasound.
Such a baby requires a colostomy. It is important to perform
the colostomy proximally enough in the sigmoid to allow repair
of the malformation without interference from the colostomy.
In other words, the surgeon must leave enough redundant, dis-
tal rectosigmoid to allow for the pull-through and even a vag-
inal replacement if that is necessary. During opening of the
colostomy, it is mandatory that the hydrocolpos be drained
when present. If the hydrocolpos is not large enough to reach
the abdominal wall above the bladder, to be sutured to the skin,
it can be drained with a curled rubber tube. Because a signi-
ficant number of these patients have two hemivaginas, the
surgeon must be certain that the tube inserted into the hydro-
colpos is draining both hemivaginas. Occasionally, one has to
open a window in the vaginal septum to drain both with a single
tube. At times the hydrocolpos is so large that it may even
produce respiratory distress. Rarely, some patients with cloaca
are unable to empty their bladder because the common channel
is almost atretic. In such rare circumstances, the baby may
FIGURE 103-4 Cross-table lateral radiograph, reachable rectum. require a vesicostomy. Endoscopy is best done outside of the
1294 PART VII ABDOMEN

Algorithm for female newborn with anorectal malformation

Rule out serious, - Sacrum - Kidney and abdominal U/S

potentially lethal - Esophagus - Spinal U/S (tethered cord)
associated defects - Lumbar spine - Cardiac echo

Perineal inspection

No visible fistula (<10%)

Single perineal
orifice 24 hrs
Cloaca Perineal Vestibular
- Urologic fistula fistula Cross-table
evaluation lateral X-ray
- Rule out
hydrocolpos Rectum High
Colostomy below rectum
Drain coccyx
Anoplasty or Colostomy
hydrocolpos
dilatations or primary repair*
Urinary Colostomy
diversion
(if necessary) FIGURE 103-6 Algorithm for the
treatment of a female newborn with
*Depending on the experience of the surgeon and general condition of the patient an anorectal malformation.

newborn period when the baby is larger and the perineum is
less swollen, to determine the anatomy, particularly the length
of the common channel, which helps with surgical planning.
Perineal inspection may show the presence of a perineal fis-
tula, and an anoplasty can be performed without a colostomy.
The presence of a vestibular fistula is the most common
finding in female patients. This malformation may be repaired
during the neonatal period without a protective colostomy;
however, these operations represent the most common source
of complications in such patients. The decision to repair this
malformation primarily or to open a colostomy should be based
on the experience of the surgeon. Colostomy is still the most
effective way to protect these patients from a perineal infection
of dehiscence.
Occasionally (<10% of cases), no fistula is visible and
there is no meconium apparent even after 24 hours of obser-
vation. This small group of patients requires a cross-table lateral
radiograph. If the radiograph shows gas in the rectum located
near the skin, the patient probably has imperforate anus
with no fistula. If the patient is in good condition, one can
perform a primary operation without a colostomy. Many of
these patients with no fistula also have Down syndrome.20
When patients are subjected to primary repair of a vesti-
bular fistula or perineal fistula later in life, strict preoperative
bowel irrigation 24 hours in advance should be performed.
Our routine is to keep such patients with nothing by mouth,
on parenteral nutrition, for 7 days to decrease the passage
of stool.
With emerging advancements in perinatology and prenatal
ultrasound techniques, anorectal malformations are more
commonly being suspected.57 Sonographic findings such as
a dilated rectum or hydrocolpos or demonstration of an asso-
ciated anomaly such as an absent kidney, a vertebral anomaly
such as a hemisacrum, or an orthopedic problem such as a
missing radius can make the perinatologist suspicious that
the fetus may in fact have an anorectal malformation.
With improving technology, it is likely that this area of diag-
nosis and even fetal intervention for, perhaps, a massive
hydrocolpos will continue to advance.58–59
Abdominal ultrasonography to evaluate for associated uro-
logic problems and echocardiography should be performed
during the initial newborn evaluation. It is also convenient
at this time to perform a spinal ultrasound to evaluate for as-
sociated spinal anomalies such as myelomeningocele or a teth-
ered cord. Such an ultrasound is an excellent screening
examination for these anomalies, but it must be performed
before sacral ossification, which occurs at 3 months of age.
If screening is done later, magnetic resonance imaging is
required. From a plain anteroposterior and lateral radiograph
of the sacrum, a sacral ratio can be calculated (see Fig. 103-1),
which can be helpful in predicting the prognosis for fecal
continence with low ratios predictive of poor potential for
bowel control and high ratios (>.7) predictive of good poten-
tial. If a hemisacrum is identified, a presacral mass may be
present and must be looked for by magnetic resonance imag-
ing. During the assessment conducted before performance
of either anoplasty or colostomy, the baby is maintained on
intravenous fluids, with a nasogastric tube in place to de-
compress the stomach. No oral feeding is allowed. Antibiotics
are given at the time of surgery.
LIMITED POSTERIOR SAGITTAL
ANORECTOPLASTY
When a low anomaly (perineal fistula) is diagnosed, the fistu-
lous track to the perineum is always located anterior to the
sphincter mechanism. A limited posterior sagittal anorecto-
plasty can be performed in the newborn period. In newborns
who cannot undergo such a procedure because they are ill,
have significant associated anomalies, or the surgeon prefers
to wait, dilatation of the fistulous tract is appropriate with a
plan for repair in the future preferably within 2 to 3 months.
Occasionally, such babies escape detection in the newborn
period and delayed primary repair is then appropriate. It is
advantageous to perform this operation in the first 48 hours
of life because the meconium is sterile. When such an ope-
ration is delayed and performed at several months of life,
 CHAPTER 103 ANORECTAL MALFORMATIONS 1295

Fistula

A B

C
FIGURE 103-7 External appearance of the perineum of female patients with anorectal malformations. A, Perineal fistula. B, Vestibular fistula. C, Cloaca.

a complete bowel preparation including a postoperative pe-
riod of parenteral nutrition without oral intake is recom-
mended to reduce the risk of perineal infection.
For a limited posterior sagittal anorectoplasty, the baby is
placed prone. In males, a urinary catheter is inserted. The dis-
tal end of the rectum is intimately attached to the posterior
urethra, and urethral injury must be avoided. Multiple 6-0 silk
sutures are placed at the mucocutaneous junction around the
fistula orifice. An incision divides the posterior sphincter in
half and is continued circumferentially around the fistula.
While traction is maintained on the bowel, a circumferential
dissection is performed to mobilize the bowel and reposition
it within the limits of the sphincter. Mucosa is sutured to
skin with fine, absorbable sutures under slight tension. The
perineal body is reconstructed.
COLOSTOMY
Colostomy is usually performed as a first stage in a newborn
with a high anomaly. A descending colostomy is a relatively
trouble-free stoma (Fig. 103-8). It also leaves the sigmoid co-
lon entirely free for reconstruction. It is fashioned by making
an oblique left lower quadrant incision. The stoma is divided,
with the bowel openings placed at opposite ends of the
1296 PART VII ABDOMEN

FIGURE 103-9 Distal colostogram showing a rectourethral bulbar fistula.

FIGURE 103-8 Descending colostomy.

wound. The proximal part of the stoma is at the start of the

sigmoid colon, just after the colon comes off its left retroper-
itoneal attachment. The distal part of the stoma, in the upper
sigmoid, is made intentionally small in an attempt to prevent
prolapse. The distal segment should be copiously irrigated to
remove all meconium. It should be placed far enough away
from the functional stoma so that the stoma bag does not cover
it. This way no spillage downstream can occur, and the stoma
bag stays on more easily.
A colostomy can result in significant morbidity, so it is vital
to construct it meticulously.60 Transverse colostomies often
lead to prolapse, make cleaning the distal segment difficult,
and make delineation of the downstream rectourinary
anatomy during the distal colostogram challenging.61 Such
colostomies also lead to megarectosigmoid and a left micro-
colon in the defunctionalized segment, particularly when
the colostomy is left in place for many months. A colostomy
that is placed too low in the sigmoid can limit the pull-
through. A loop colostomy leads to distal contamination
of the urinary stream if there is a fistula and to a dilated mega-
rectosigmoid from stool spillage.
FIGURE 103-10 Distal colostogram showing a rectobladder neck fistula.

MANAGEMENT AFTER COLOSTOMY

Before proceeding with definitive reconstruction, the anatomy
Anorectal Reconstruction
------------------------------------------------------------------------------------------------------------------------------------------------

of the anorectal malformation is delineated by high-pressure

distal colostography (Figs. 103-9 and 103-10).61 Patience is
required when performing this study as the rectum is slowly
distended with contrast. The muscles of the pelvis are nor-
mally contracted, and thus the distal rectum is compressed
and can at first appear flat at the pubococcygeal line. With
gentle pressure, this flat line begins to bulge, and it is at this
point that the examiner can visualize the true extent of the
rectum and the presence of a rectourinary fistula.
BASIC PRINCIPLES
All defects can be repaired using a posterior sagittal approach.
Approximately 10% of male patients (those with a rectum-
bladder neck fistula) and about 40% of female patients with
a cloaca may, in addition, require an abdominal approach
for mobilization of a high rectum or vagina. The patient is
placed in a prone position with the pelvis elevated. A Foley
catheter is inserted into the bladder before the patient is

positioned. A coudé catheter is ideal. The catheter sometimes
goes into the rectum through the fistula rather than into the
bladder. To avoid misplacement, the surgeon can direct the
catheter with a lacrimal probe inserted in the tip of the Foley
catheter to function as a guide. Occasionally, the catheter must
be positioned intraoperatively once the fistula is exposed. Spe-
cial care must be taken to cushion all the pressure areas of the
baby’s body (Fig. 103-11). Two small bolsters are used in front
of each deltopectoral groove to avoid hyperextension of the
neck and shoulders. Electrical stimulation of the perineum
allows evaluation of the strength of the sphincter contraction.
A midline incision is made, and the sphincteric mechanism is
divided exactly in the midline, with equal amounts of muscle
on each side. The rationale for this approach is based on the
fact that important nerves or vessels do not cross the midline.
Sharp Weitlaner retractors are used to achieve good exposure.
The retractors should be placed superficially to avoid pressure
damage to the muscles.
POSTERIOR SAGITTAL ANORECTOPLASTY
FOR ANOMALIES IN MALES
Rectourethral Fistula (Bulbar and Prostatic)
The size of the incision must be adapted to the specific anat-
omy of the patient (Fig. 103-12). No deleterious effects are
caused by extending the incision, provided that it remains
in the midline. The incision usually extends from the lower
portion of the sacrum through the center of the anal dimple
and sometimes extends to the perineal body. Below the skin
and running parallel to the midline are the parasagittal muscle
fibers. Medial to these muscle structures and located within
the limits of the anal dimple are other muscle structures that
run perpendicular. These structures extend from the skin to
the levator muscle and are called the muscle complex. Deeper
to the parasagittal fibers, the ischiorectal fat becomes evident.
Below this is the levator, which is also divided in the midline.
The crossing of the muscle complex fibers with the parasagit-
tal muscle structures defines the anterior and posterior limits
of the new anus. These limits can be determined most clearly
with use of an electrical stimulator.
FIGURE 103-11 Patient in prone position with proper padding.
CHAPTER 103 ANORECTAL MALFORMATIONS 1297
In boys with a rectourethral bulbar fistula, on opening the
incision the bowel is evident. In boys with a rectoprostatic
fistula, the rectum is much smaller and located much higher
in the incision just under the coccyx. The distal colostogram
provides information that is valuable at this point for deter-
mining where the rectum can be found (see Fig. 103-9).
In the case of a rectobladder neck fistula, the rectum is not
visible through a posterior sagittal approach and should not
be searched for (see Fig. 103-10).
Temporary silk sutures are placed in the posterior rectal
wall for traction, and the rectum is opened exactly in the
midline. The incision in the rectal wall is extended distally.
As the rectum is being opened, more sutures are placed to
hold the edges of the rectal wall and improve exposure of
the rectal lumen. The incision must be extended distally until
the fistula site is found (Fig. 103-13).
Because there is no distinct plane of separation between the
rectum and urethra, the surgeon must be extremely careful
while separating the rectum from the urinary tract. Multiple
6-0 silk sutures are placed above the fistula site in a semicir-
cular fashion (Fig. 103-14). These sutures allow the surgeon
to exert uniform traction on the rectum while dissecting and
separating it from the urethra. Dissection of the lateral aspects
of the rectum first helps to delineate the anterior plane. The
dissection then proceeds in the submucosal plane, anteriorly,
for approximately 5 to 7 mm until the rectum and urethra
gradually separate and become independent. The surgeon
must be careful in this anterior plane because this is where
the vas deferens and seminal vesicles can be injured if the dis-
section is too deep. A single suture is placed at the fistula site
so that it can easily be identified and closed later. Once the
rectum is fully separated from the urinary tract, circumferen-
tial perirectal dissection is performed to gain enough rectal
length to reach the perineum. To achieve this, the fascia that
surrounds the rectum must be divided including blood vessels
and nerves attached to the rectum. This is a key plane to iden-
tity. While the surgeon is applying traction to the rectum, the
dissection continues in a circumferential manner. The vessels
that hold the rectum are coagulated and divided until enough
rectal length has been achieved. The rectum has an excellent
intramural blood supply; even with a high prostatic fistula,
sufficient length can be obtained without making the rectum
ischemic. When the rectal wall is injured, however, this blood
supply is damaged and ischemia may occur. Therefore every
effort must be made to perform this dissection as close as pos-
sible to the rectal wall but without interfering with the intra-
mural blood supply. Once enough length has been achieved to
perform a tension-free bowel-to-skin anastomosis, the size of
the rectum must be evaluated and compared with the available
space. If necessary, the rectum can be tapered by removing
part of its posterior wall and reconstructed by closing the
posterior incision with two layers of long-term absorbable
interrupted sutures. The anterior rectal wall that used to be
attached to the posterior urethra may require a few stitches
to reinforce the wall by bringing the seromuscular layer
together. The urethral fistula is closed with interrupted
absorbable suture.
The perineal body is reconstructed by bringing together the
anterior limits of the external sphincter, which was previously
marked with temporary silk sutures. The levator muscle is su-
tured behind the rectum (see Figs. 103-12 and 103-15), and
the posterior edge of the muscle complex is sutured together
1298 PART VII ABDOMEN

A B C D E

J I
H F
G
FIGURE 103-12 Essential steps for posterior sagittal anorectoplasty in male patients. A, Planned posterior sagittal incision. B, Posterior sagittal approach
with the parasagittal fibers and ischiorectal fat split in the midline. C, Posterior rectal wall exposed. D, Posterior rectal wall opened in the midline. E, Pos-
terior rectal wall opened going anteriorly until the rectourethral fistula is identified. F, Separation of the rectum from the posterior urethra with dissection
above the fistula. G, The rectum fully mobilized and in this case tapered. Sutures are placed anteriorly to close the perineal body. H, The rectum pulled
through and placed within the limits of the sphincter mechanism. I, Closure of the levator and tacking of the posterior edge of muscle complex to the
posterior rectal wall. J, Closure of the posterior sagittal incision and completed anoplasty.

in the midline while taking, with the same stitches, part of the
posterior rectal wall to anchor the rectum. This helps to pre-
vent prolapse (Fig. 103-16). The ischiorectal fossa and the
subcutaneous tissue are then reapproximated, and the skin
is closed. The anoplasty is performed by placing the rectum
within the limits of the sphincter with the use of interrupted
long-term absorbable sutures (Fig. 103-17).
Rectobladder Neck Fistula
In patients with a rectobladder neck fistula, an abdominal
approach via either laparoscopy or laparotomy, in addition
to the posterior sagittal approach, is necessary to mobilize
the rectum. The fistula is separated from the back of the blad-
der neck. Laparoscopy is well suited for this stage. The chal-
lenge thereafter is often mobilization of adequate rectal length
to reach the perineum. Such mobilization, plus the bulbous
distension of the rectum, makes the laparoscopic stage tech-
nically demanding. In general, for higher lesions such as a
rectoprostatic urethral fistula, a laparoscopic approach has
utility62,63 because sufficient mobilization is often a challenge
through the posterior sagittal incision. Lower lesions, how-
ever, such as a rectourethral bulbar fistula or the no-fistula
defect, can be more safely repaired with a posterior sagittal
approach without laparoscopic assistance because the rectum
lies below the peritoneal reflection.
Before beginning operation on higher lesions, the entire
body surface from chest to toes is prepared so that the surgeon
can work simultaneously in the abdomen and the perineum.
The abdomen is entered via either laparoscopy or laparotomy
(Fig. 103-18). The sigmoid is identified and dissected down
to the bladder neck. The distal rectum is usually located
approximately 2 cm below the peritoneal floor and therefore
requires minimal pelvic dissection. It is important to remem-
ber that the vas deferens and ureters run close to the rectum;
thus the dissection must be carried out as close as possible to
the rectal wall. The rectum narrows rapidly and opens into the
bladder neck in a T-shaped manner. It is relatively easy to sep-
arate the rectum from the urinary tract because the fistula
forms less of a common wall the higher the malformation.
The rectum is divided at its most distal part, and the fistula
is closed with absorbable suture.
In this type of high defect, mobilization of the rectum is
sometimes limited by branches of the inferior mesenteric
vessels. To gain length surgeons have traditionally been ad-
vised to divide the inferior mesenteric vessels at their origin
from the aorta. This technique is contraindicated here because
this is often the only blood supply of the rectosigmoid—the
 CHAPTER 103 ANORECTAL MALFORMATIONS 1299

FIGURE 103-13 Operative view, posterior sagittal approach. The rec-

tourethral fistula is seen with a metallic probe in it.
FIGURE 103-15 Levator muscle sutured behind the rectum.

FIGURE 103-14 Operative view. Fine silk sutures are placed through the FIGURE 103-16 Sutures on the posterior edge of the muscle complex
mucosa of the cephalad hemicircumference of the fistula. anchor the rectum.
1300 PART VII ABDOMEN

A B
FIGURE 103-17 A, Repair of perineal body and anoplasty. B, Anoplasty performed with 16 long-term absorbable sutures.

A B
FIGURE 103-18 Operative view of a rectobladderneck fistula showing the rectosigmoid ending at the bladder neck. A, Laparoscopic view. B, View via
laparotomy.

opening of the colostomy may have interfered with the conti-
nuity of the arcade that connects the middle colic vessels with
the inferior mesenteric vessels. Therefore the blood supply of
the distal part of the rectum may depend solely on the inferior
mesenteric vessels, and therefore preservation of the main
trunks is crucial to preserve the intramural blood supply of
the rectum. Meticulous dissection along the rectum is re-
quired to gain length (Fig. 103-19). In extremely high defects,
the rectum can be elongated by rectoplasty. Sometimes the
distal rectum is wide and must be tapered. With the legs lifted
up, a midsagittal incision is made and the presacral space
dissected. Then the rectum is pulled through. An anoplasty
should be performed as previously described. Tacking of
the rectum to the muscle complex is important because these
patients have poor pelvic musculature and are particularly
prone to prolapse.
Imperforate Anus Without a Fistula
In patients with an imperforate anus and no fistula, the blind
end of the rectum is located at the same level as in a patient
with a bulbar urethral fistula. Because the rectum is still inti-
mately attached to the posterior urethra, separation from the
urethra requires careful dissection. The rest of the repair is
performed as described for rectourethral fistulas.

Divide
FIGURE 103-19 Ligation of the peripheral branches of the inferior
mesenteric vessels with preservation of the main trunks, in order to gain
length on a high rectum.
Rectal Atresia and Stenosis
Repair of rectal atresia and stenosis involves connecting the
blind dilated end of the rectum proximally with the anal canal
distally. Both structures are usually separated by a few milli-
meters of fibrous tissue. The rectum must be sufficiently mo-
bilized to allow the performance of an end-to-end anastomosis
to the anal canal. The wound is then closed by reconstructing
all the muscle structures, as previously described. Because the
anal canal is normal, these patients have excellent potential for
bowel control. This defect in particular is associated with a
presacral mass, often a teratoma; full evaluation is essential.
POSTERIOR SAGITTAL ANORECTOPLASTY
FOR ANOMALIES IN FEMALES
Vestibular Fistula
The complexity of this defect is frequently underestimated.
Patients with a vestibular fistula are born with excellent poten-
tial for bowel control. Thus every effort should be directed at
successful reconstruction with a single operation; wound de-
hiscence and infection may compromise the final functional
result. A protective colostomy minimizes these complications,
but if the surgeon has adequate experience with this malfor-
mation, a primary repair can be performed in the newborn
period. Occasionally, such patients escape detection in the
newborn period and undergo reconstruction several months
or even years later. A primary repair without colostomy under
these circumstances can also be performed. However, to avoid
a perineal infection, meticulous technique, excellent bowel
preparation, 1 week without oral intake, and parenteral
nutrition postoperatively are vital.
With the patient in prone position, a midline incision is
performed, but it is not usually necessary to extend the inci-
sion as high as the middle portion of the sacrum. The midline
incision continues around the fistula into the vestibule and
CHAPTER 103 ANORECTAL MALFORMATIONS 1301
multiple 5-0 sutures are placed circumferentially at the fistula
site. While traction is placed on these sutures, the rectum is
dissected in a circumferential manner. The posterior rectal
wall can easily be identified, and the dissection must start from
the posterior aspect and be extended laterally. The last step,
separation of the rectum from the vagina, is the most delicate
part of the dissection. The common wall between the rectum
and vagina in this kind of defect is long and extremely thin.
Once fully separated, the surgeon finds an areolar plane be-
tween the two structures. At this point branches of the hem-
orrhoidal vessels found on the lateral aspects of the rectum can
be ligated with cautery, thereby mobilizing the rectum as pre-
viously described, to gain enough length to perform a tension-
free bowel-to-skin anastomosis. The limits of the sphincteric
mechanism are electrically determined and marked with tem-
porary silk sutures. The perineal body is then reconstructed
by bringing together the anterior limit of the sphincter com-
plex. The anterior edge of the muscle complex is reapproxi-
mated taking bites of rectal wall, as described previously for
males. The levator muscle is not usually exposed and thus
does not need to be reconstructed. The anoplasty and wound
closure are performed as described for males. For the ex-
tremely unusual case of a rectovaginal fistula, the technique
is the same except that the rectum requires more mobilization
to reach the perineum.
Cloaca
This female malformation is managed by a variation of the
posterior sagittal anorectoplasty. It is considered separately
because of its complexity. The reconstruction requires a deli-
cate, meticulous technique, as well as creativity, imagination,
and a resourceful, experienced surgeon. Repair of this defect
has three main goals: achievement of urinary control, bowel
control, and sexual function. Because a spectrum of defects
exists, these goals are achieved in some cases; in others,
however, the defect is repaired anatomically, but the patients
are left with functional sequelae that require extensive medical
assistance to provide a good quality of life. During surgical
exploration, the surgeon must be prepared to find bizarre an-
atomic arrangements of the rectum, vagina, and urinary tract
(Fig. 103-20). A long midsagittal incision is made, extending
from the middle portion of the sacrum through the sphincter
mechanism and down into the single perineal opening.
Before undertaking repair of a cloaca, the surgeon should
perform an endoscopy with the specific purpose of determin-
ing the length of the common channel. A contrast study of
each of the three systems is valuable and can be done in three
dimensions (Fig. 103-21). There are two well-characterized
groups of patients with a cloaca.64 These two groups represent
different technical challenges and must be recognized preop-
eratively. The first group consists of patients with a common
channel shorter than 3 cm; fortunately, this represents the
majority. These patients can usually be repaired via a posterior
sagittal approach alone, without laparotomy.
The second group consists of patients with longer common
channels. These patients usually need laparotomy, followed
by a decision-making process that usually requires intricate
vaginal maneuvers or vaginal replacement, tricks to mobilize
a high rectum, knowledge of bladderneck anatomy, and deci-
sion making related to ectopic ureters if found. This group
should be cared for at centers with special dedication to the
repair of these defects.
1302 PART VII ABDOMEN

Vagina

Urethra

Common
channel

A B
FIGURE 103-20 A, Posterior sagittal view of a typical cloaca with a common channel of 2 cm with a visible vaginal septum. B, A cloaca with a very large
vagina (hydrocolpos). The rectum opens higher in the vagina and is not seen here.

Cloacas with a Common Channel Shorter than 3 cm The

FIGURE 103-21 Three-dimensional cloacagram.
incision extends from the middle portion of the sacrum down
to the single perineal orifice. The sphincter mechanism is
divided in the midline. The first structure that the surgeon will
find after division of the sphincter mechanism is the rectum.
The rectum is opened in the midline, and silk stitches
are placed along the edges of the posterior rectal wall. The
incision is extended distally through the posterior wall of
the common channel. The entire common channel is exposed
to allow confirmation of its length under direct vision
(Fig. 103-22). The rectum is then separated from the vagina
(see Fig. 103-22). This is performed in the same way as for
repair of a rectovestibular fistula because the rectum and
vagina share the same common wall already described.
In patients with a vaginal septum, the rectum must be
searched for at the posterior aspect of this septum.
Once the rectum has been completely separated from the
vagina, a maneuver called total urogenital mobilization is
used.65 Before this advance, the vagina was separated from
the urinary tract, which was a technically challenging maneu-
ver associated with significant morbidity.
Total urogenital mobilization consists of mobilization of
both the vagina and urethra as a unit without separating them.
After the rectum has been separated, multiple silk stitches are
placed through the edges of the vagina and the common
channel to apply uniform traction on the urogenital sinus
(see Fig. 103-22). Another series of fine stitches are placed
 CHAPTER 103 ANORECTAL MALFORMATIONS 1303

B C

FIGURE 103-22 A, Total urogenital mobilization (cloaca with a short common channel, <3 cm). B, The rectum is separated from the urogenital sinus.
C, Sutures are placed around the urogenital complex for uniform traction to facilitate mobilization of the urogenital sinus.

across the urogenital sinus approximately 5 mm proximal to the
clitoris. The urogenital sinus is transected between the last row
of silk stitches and the clitoris. The dissection takes advantage of
the fact that there is a natural plane of separation from the pu-
bis. Rapidly and in a bloodless field, one can reach the upper
edge of the pubis. There, a fibrous, avascular structure that
gives support to the vagina and bladder is identified.
This structure is called the suspensory ligament of the ure-
thra and bladder. This ligament is divided, which immediately
provides significant mobilization (between 2 and 3 cm) of the
urogenital complex. One can then dissect the lateral and dor-
sal walls of the vagina to gain an extra 5 to 10 mm. This dis-
section is enough to repair about 60% of all cloacas and is a
reproducible maneuver. It has the additional advantage of
preserving an excellent blood supply to both the urethra
and vagina. It places the urethral opening in a visible location
and provides a smooth urethra that can easily be catheterized.
What used to be the common channel is divided in the mid-
line to form two lateral flaps that are sutured to the skin to
create the new labia. If a vaginal septum is present, it is re-
moved. The vaginal edges are then mobilized to reach the skin
to create the introitus. The limits of the sphincter are electri-
cally determined. The perineal body is reconstructed by bring-
ing together the anterior limit of the sphincter. The rectum is
placed within the limits of the sphincter in the manner previ-
ously described. Patients can eat the same day, and the pain is
usually easily controlled. Patients are discharged 48 hours
postoperatively.
1304 PART VII ABDOMEN
Cloacas with a Common Channel Longer than 3 cm When
endoscopy and a cloacagram show that the patient has a long
common channel, the surgeon must be prepared to face
several significant technical challenges.
Patients with a long common channel should receive a
total-body preparation because it is likely that laparotomy will
be required. The rectum is separated from the vagina and ure-
thra via a posterior sagittal incision or via laparotomy when it
is located very high. The presence of a long common channel
(>5 cm) indicates there is no way that total urogenital mobi-
lization will be sufficient to repair the malformation, and
therefore it is advisable to leave the common channel in place,
which can be used as urethra for intermittent catheterization.
In this situation the vagina should be separated from the uri-
nary tract by placing multiple 6-0 silk sutures through the
vaginal wall to try to create a plane of dissection between
the vagina and the urinary tract. This is best done through
the abdomen when the vagina(s) are high. In such a high com-
mon channel case, when the vaginas are low and closer to the
perineum, the surgeon may need to do a total urogenital mo-
bilization and then deliver that mobilized urogenital complex
into the abdomen. Sometimes dissection in the abdomen then
allows this complex to reach the perineum. Often, though, at
this point, the urinary tract and vaginas must be separated and
the neourethra tubularized. These are delicate maneuvers.
Often, the bladder must be opened in the midline, and feeding
tubes placed into the ureters to protect them. In these types of
malformations there is an extensive common wall between
vagina and bladder. Both ureters run through that common
wall, and therefore during separation of the vagina from the
urinary tract, the ureters must sometimes be skeletonized
and thus need to be protected. Once the separation has been
completed, if the vagina does not reach, the surgeon has to
make decisions about the vaginal reconstruction on the basis
of specific anatomic findings.
Within the abdomen, the patency of the müllerian struc-
tures are investigated by passing a No. 3 feeding tube through
the fimbriae of the fallopian tubes and injecting saline. If one
of the systems is not patent, the atretic müllerian structure
should be excised, with great care taken to avoid damage to
the blood supply of the ipsilateral ovary. When both müllerian
structures are atretic, they should be left in place. The patient
must be monitored closely and further investigated with
ultrasound when she enters puberty.
Vaginal Switch Maneuver A specific group of patients are
born with hydrocolpos and two hemivaginas. The hemivagi-
nas are large and the two hemiuteri are separated, the distance
between them being longer than the vertical length of both
hemivaginas. In these cases it is ideal to perform a maneuver
called a “vaginal switch” (Fig. 103-23).
One of the hemiuteri and the ipsilateral fallopian tube is
resected with particular care taken to preserve the blood sup-
ply of the ovary. The blood supply of the hemivagina of that
particular side is sacrificed. The blood supply of the contralat-
eral hemivagina is preserved and provides for both hemivagi-
nas. The vaginal septum is resected, and both hemivaginas are
tubularized into a single vagina by taking advantage of the
long lateral dimension of both hemivaginas. Then, what used
to be the dome of the hemivagina where the hemiuterus was
resected is turned down to the perineum. This is a useful
maneuver that can be performed only when these specific
anatomic characteristics are encountered.
Vaginal Replacement In a patient with a small vagina(s) or
in the rare case of an absent vagina, a vaginal replacement is
required. In such cases the choices are rectum, colon, or small
bowel. When the patient has internal genitalia or an upper
blind vagina, the upper part of the bowel used for replacement
should be sutured to the vaginal cuff. When the patient has no
internal genitalia (no vagina and no uterus), a vagina is created
and left with its upper portion blind and used only for sexual
purposes.
Rectum: This form of vaginal replacement is feasible only in
patients who have a good size rectum that is large
enough to be able to divide it transversely or longitudi-
nally into a portion with its own blood supply that will
form a new vagina and another portion with enough cir-
cumference to reconstruct an adequately sized rectum
(Fig. 103-24). The blood supply of the rectum will be
provided transmurally from branches of the inferior
mesenteric vessels. This can also work for patients with
a rectosigmoid long enough to allow the distal portion to
be used as a new vagina with its own blood supply and
the upper portion to be pulled down as the new rectum.
Colon: The colon is an ideal substitute to replace the vagina
(Fig. 103-25). However, sometimes the location of the
colostomy interferes with this type of reconstruction.
The left colon or sigmoid work well because their
arcades reach the perineum nicely. Sometimes taking
the colostomy down and using its distal segment for
the vaginal graft is a helpful maneuver.
Small Bowel: When the colon is not available, the most
mobile portion of the small bowel is used for vaginal
reconstruction. The mesentery of the small bowel is lon-
gest in an area located approximately 15 cm proximal to
the ileocecal valve, which is the best portion of the small
bowel for vaginal replacement. A portion of the ileum is
isolated and pulled down while preserving its blood
supply (Fig. 103-26).
In the highest type of cloaca one may find two little hemi-
vaginas attached to the bladder neck or even to the trigone of
the bladder. In these cases the rectum also opens in the trigone
(Fig. 103-27). Separation of these structures is done abdom-
inally. Unfortunately, when the separation is completed, the
patient is frequently left with either no bladder neck or a
severely damaged one. At that point the surgeon must have
enough experience to make a decision whether to reconstruct
the bladder neck or close it permanently. In the first situation,
most patients will require intermittent catheterization to
empty the bladder. A vesicostomy is created in such cases,
and the patient will require a continent diversion type of
procedure and possibly a bladder neck procedure and/or
bladder augmentation, done at the age of urinary continence
(3 to 4 years old). In this particular type of malformation, the
patient usually also needs a vaginal replacement, which
should be performed in one of the ways previously described.
At the time of colostomy closure, endoscopy should be per-
formed to be sure that the repair is intact and that there is no
prolapse, stricture, or urethrovaginal fistula. If the cloaca re-
pair did not require laparotomy, the time of colostomy closure
is the opportunity to investigate the patency of the müllerian
structures.
 CHAPTER 103 ANORECTAL MALFORMATIONS 1305

R. tube
R. hemiuterus

Divide Vaginal septum

R. ovary L. ovary
R. giant
hemivagina
hydrocolpos

Communication with
urinary tract
and/or rectum

A Perineum

Preserved
l. hemiuterus
Preserved
ovary

(R. hemihysterectomy)
Resected
vaginal septum Preserved
blood supply

R. hemivagina
switched down

B
FIGURE 103-23 A, Very large hemivaginas with a vaginal septum and long common channel. B, Vaginal switch maneuver.

Inferior
mesenteric
vessels

Rectum
(intramural
blood supply)

Future vagina New vagina

A Line of divison B
FIGURE 103-24 A and B, Vaginal replacement with the rectum.
1306 PART VII ABDOMEN

GENERAL PRINCIPLES OF POSTOPERATIVE

FIGURE 103-25 Vaginal replacement with the sigmoid colon.
CARE
In the absence of a laparotomy, oral feedings may begin when
the child is awake. Antibiotics are given for 48 hours. In males
who had a rectourethral fistula, the urinary catheter should be
left in place for 7 days. In patients who have undergone cloaca
repair, the urethral catheter is left in place for 2 to 3 weeks un-
til the urethral orifice is visible.
A dilatation program is begun 2 weeks after surgery. The
anus is calibrated, and a dilator that fits snugly is initially used
to dilate the anus twice a day. Every week, the size of the di-
lator is increased by one unit until the desired size is reached.
The optimal size of dilator is shown in Table 103-2. Once the
correct size is reached, the colostomy can be closed, which is
usually 8 to 12 weeks after the reconstruction.
Dilatations must continue after closure. Once the dilator
can be inserted easily, the schedule is reduced to once a day
for 1 month, twice a week for 1 month, once a week for
1 month, and then once a week for 3 months.
After the colostomy is closed, the patient may have multiple
bowel movements and perineal excoriation may develop.
A constipating diet may be helpful in the treatment of this
problem. After several weeks the number of bowel movements
decreases, and most patients will essentially become consti-
pated. This constipation must be anticipated and proactively
treated.31 After 3 to 6 months, a more regular bowel move-
ment pattern develops. A patient who has one to three bowel
movements per day remains clean between bowel movements

A C
FIGURE 103-26 Vaginal replacement with small bowel. A and B, Using the portion of ileum with the longest mesentery. C, Pulling the small bowel down
as a neovagina (the insert shows an anastomosis to the upper part of the vagina).

Bladder
Right hemivagina
FIGURE 103-27 Extremely long common channel–hemivaginas and rec-
tum connected to the bladder neck.
and shows evidence of a feeling of pushing during bowel
movements. This type of patient has a good bowel movement
pattern, generally has a good prognosis, and is trainable.
A patient with multiple bowel movements or one who passes
stool constantly without showing any signs of sensation or
pushing typically has a poor functional prognosis for bowel
control. The original malformation type, as well as the quality
of the sacrum and spine, predict the potential for voluntary
bowel movements.
In cloacas the baby is watched to see whether she is
capable of emptying her bladder. If she cannot pass urine
and efficiently empty the bladder, she needs intermittent
catheterizations.
Outcomes
------------------------------------------------------------------------------------------------------------------------------------------------
COMPLICATIONS
All the usual postoperative complications may occur after
surgery of this nature. In addition, some specific problems
develop both early in the postoperative course and later.66
Wound infection and retraction are known to occur. When
the infection affects only the superficial layers of the wound, is
not accompanied by dehiscence of the structures pulled
through, and is promptly treated, it is likely that no functional
sequelae will result. On the other hand, infection accom-
panied by dehiscence may reach catastrophic proportions
TABLE 103-2
Anal Dilatation Program
Patient Age
1-4 mo
4-8 mo
8-12 mo
1-3 yr
3-12 yr
>12 yr
CHAPTER 103 ANORECTAL MALFORMATIONS 1307
and leave sequelae that include incontinence, strictures, ac-
quired atresias, recurrent fistulas, and severe pelvic fibrosis.
We have reoperated on many such patients who suffered
these kinds of complications at another institution and were
referred to us for a secondary procedure.67–68
Rectum
Infections and dehiscence have occurred mainly in patients
operated on primarily, without a colostomy. Therefore al-
though there is the benefit to operate earlier, primarily, and
without a colostomy, it must be remembered that a colostomy
is a valuable adjunct in the management of these defects.
Surgeons must make decisions regarding this issue on the
basis of their personal experience.
Determining the precise causes of these complications is
difficult; however, it seems the main contributing factors are
fecal contamination, ischemia, and suture line tension (often
from the incomplete mobilization of structures).
Long common channel Rectal or vaginal strictures (or both) are usually due to
the previously mentioned complications, although they
may occur independently. An anal dilatation program is
recommended to avoid strictures; however, these maneuvers
prevent only minor, ringlike strictures. Difficult anal dilata-
tions usually reflect a major problem related to ischemia or
tension that will result in a long narrow stricture or even
acquired atresia.
Rectal mucosal prolapse is not uncommon after recon-
struction. The rate of prolapse after posterior sagittal anorec-
toplasty in the authors’ series was less than 5%.69 Mucosal
prolapse is uncommon after posterior sagittal anorectoplasty,
probably because of several key technical steps: (1) tacking of
the posterior rectal wall to the posterior edge of the muscle
complex, (2) tapering a dilated rectum if necessary, and
(3) performing the anoplasty under slight tension so that after
the sutures of the anoplasty are cut, the rectum retracts slightly
with no mucosa being visible. Mucosal prolapse is managed by
full-thickness trimming and is performed when prolapse
causes excess mucus production or ulceration or interferes
with anal sensation.
A review of patients operated on at other institutions has
revealed significant urologic injuries in male patients who
underwent repair of anorectal malformations.56 The posterior
sagittal approach, when performed without a good preopera-
tive distal colostogram, was the most important source of these
complications. Urethral, ureteral, vas deferens, and seminal
vesicle injuries can occur. The laparoscopic approach when
done for malformations in which the rectum ends below
the peritoneal reflection risks leaving behind a posterior ure-
thral diverticulum (the original distal rectum). Postoperative
neurogenic bladder in male patients who undergo a techni-
cally correct operation for the treatment of anorectal mal-
formations must be extremely unusual because in our series
it happens only in patients with an abnormal sacrum or
spine.70 Otherwise, we believe that it may reflect a poor sur-
gical technique with denervation of the bladder and bladder
neck during the repair.
Hegar Dilator Constipation is the most common sequela after surgical
Size No. 12 repair of anorectal malformations.31 The lower the malfor-
Size No. 13 mation, the more likely the development of constipation.
Size No. 14 A vicious cycle ensues with megarectosigmoid leading to more
Size No. 15 constipation. Constipation that is not properly managed
Size No. 16 will lead to more megarectosigmoid, resulting in overflow
Size No. 17 pseudoincontinence. Patients appear to be incontinent, but
if their constipation is managed appropriately, they become
1308 PART VII ABDOMEN

continent. Such patients sometimes benefit from sigmoid
resection to reduce their laxative requirement.71–74
Every effort must be made to avoid this cycle and maintain
a collapsed and clean bowel from the moment the baby is
born. Transverse colostomies left for a long period of time
lead to severe megarectosigmoid. Loop colostomies may also
contribute to distal fecal impaction, dilation, and subsequent
constipation. Adequate treatment of constipation starting
after colostomy closure is vital.
CONTINENCE
Long-term follow-up of our series suggests that good bowel
control is achieved in about 75% of patients.
Our results according to the type of abnormality are
shown in Tables 103-3 through 103-6. Patients with low
anomalies have done extremely well; those with high lesions
and those with associated spinal or sacral problems have
done less well.
All patients with anorectal malformations, regardless of the
complexity of the defect, can be kept completely dry of urine
and clean of stool after the age of 3, either because they achieve
bowel and urinary control or because we implement a pro-
gram to keep them artificially dry and clean. The majority
can have voluntary bowel movements, sometimes with the
help of laxatives. For the others, we implement a bowel
management program that uses enemas to ensure adequate
emptying for a 24-hour period of cleanliness. With the ratio-
nal administration of bowel irrigation, diet, and drugs, most
patients, including those with inherent fecal incontinence,
are able to remain clean for 24 hours.75–76 Only patients
with loose stool secondary to an absent or a short colon need
a permanent colostomy.
Patients suffering from fecal incontinence are evaluated and
classified into those with constipation or those with increased
motility (tendency to have diarrhea). In the first group the sa-
line enema must be of large volume, with additives (e.g., glyc-
erin, soap, phosphate) to help empty the colon. This program
takes advantage of the decreased bowel motility in constipated
patients; they remain clean for the next 24 hours. No laxatives
or diets are given as part of this protocol. The second group
(patients who suffer from increased bowel motility because
of loss of the rectal reservoir) require a constipating diet,

TABLE 103-3 TABLE 103-5

Constipation and Type of Defect Soiling and Type of Defect
Patients Patients with Soiling
with
Constipation Defect Cases n %
# of Patients
Defects with Defect n % Perineal fistula 57 9 16%
Vestibular fistula 135 49 36%
Atresia or stenosis 10 7 70% Atresia or stenosis 10 4 40%
Bulbar fistula 110 65 59% Bulbar fistula 105 51 49%
Perineal fistula 72 42 58% Imperforate anus without fistula 39 20 51%
Vestibular fistula 141 77 55% Cloaca common channel  3 cm 91 57 63%
Imperforate anus without fistula 41 20 49% Prostatic fistula 110 86 78%
Prostatic fistula 112 47 42% Vaginal fistula 5 4 80%
Cloaca common channel  3 cm 97 38 39% Cloaca common channel > 3 cm 55 46 84%
Cloaca common channel > 3 cm 60 17 28% Bladderneck fistula 48 43 90%
Vaginal fistula 5 1 20% Total 655 369 56%
Bladderneck fistula 50 7 14%
Total 698 321 46%

TABLE 103-6
TABLE 103-4 Totally Continent* Patients and Type of Defect
Voluntary Bowel Movement and Type of Defect
Totally Continent
Patients with VBMs
Defect Cases n %
Defect Cases n %
Perineal fistula 52 43 83%
Atresia or stenosis 11 11 100% Atresia or stenosis 9 5 56%
Perineal fistula 58 56 97% Vestibular fistula 135 86 64%
Vestibular fistula 146 131 90% Imperforate anus without fistula 36 18 50%
Imperforate anus without fistula 40 31 78% Bulbar fistula 101 46 46%
Bulbar fistula 112 89 79% Cloaca common channel  3 cm 91 32 35%
Cloaca common channel  3 cm 99 65 66% Vaginal fistula 5 1 20%
Prostatic fistula 109 71 65% Prostatic fistula 105 19 18%
Vaginal fistula 5 3 60% Cloaca common channel > 3 cm 52 6 12%
Cloaca common channel > 3 cm 69 24 35% Bladderneck fistula 46 3 7%
Bladderneck fistula 49 10 20% Total 632 259 41%
Total 698 491 70%
*Voluntary bowel movements and no soiling.
 CHAPTER 103 ANORECTAL MALFORMATIONS 1309

FIGURE 103-28 Diagram showing appendicostomy.

medication to decrease bowel motility, and a smaller daily sa-

line enema.
The treatment is adjusted by trial and error over a period of
1 week, with 95% of patients remaining clean.75 Bowel man-
agement is started just before the patient has to go to school
and join his or her classmates, who are already wearing regular
underwear. Most patients and parents are happy with the im-
plementation of this program, but when the patient reaches
the age of 7 to 12 years or sometimes younger, more indepen-
dence is usually desirable.
At that point, creation of a continent appendicostomy
(Malone procedure) is beneficial.77–79 This operation
creates a communication between the abdominal wall
and the cecum through the patient’s appendix. A valve
mechanism is created by plicating the cecum around the
appendix, which allows catheterization of the cecum but
FIGURE 103-29 Neoappendix made from a tabularized flap of colon.
prevents leakage of stool. Patients can administer their
own enema while sitting on the toilet. The operation con-
sists of the plication and then the connection of the ap- Certain patients in whom the rectum was mislocated dur-
pendix to the umbilicus to make it inconspicuous. This ing the original operation may be candidates for a reoperation.
is accomplished with a V-to-V anastomosis between the This is recommended only in patients who were born with
appendix and the umbilical skin (Fig. 103-28). A signifi- a good sacrum, good sphincters, and a malformation with a
cant number of patients do not have an appendix. In such good prognosis. The results of this procedure vary, with
cases we make an appendix with a tubularized flap of the worthwhile continence achieved in more than half of
cecum and then plicate the cecum around it (Fig. 103- patients.68,80–81
29). The stoma is also exteriorized through the umbilicus. Urologic problems in male patients and in female patients
Most patients who have undergone this operation express without cloacas are rare.82 In patients with a cloaca, those with
a great deal of satisfaction. a common channel shorter than 3 cm require intermittent
Some patients benefit from an enema program early on, but catheterization one third of the time. Patients with common
if they have potential for bowel control, when they are a little channels longer than 3 cm require intermittent catheterization
older, a laxative trial can be used to help their colon empty, or a continent diversion 70% to 80% of the time.64
with careful radiologic monitoring. This may allow them to
demonstrate the capacity for voluntary bowel movement The complete reference list is available online at www.
and thus eliminate the need for enemas. expertconsult.com.
 external sphincter and perianal skin by muscular strands. It is
involuntarily in a tonic state of contraction, providing at least
85% of the resting anal canal pressure that keeps the anal canal
closed.4 The IAS can increase contraction only slightly, but it
relaxes completely by reflex in response to rectal distention.5
This reflex, mediated by the myenteric plexus, is intrinsic and
thus remains independent from extrinsic innervation.
Both sympathetic and parasympathetic nerves innervate
the internal sphincter.6 Sympathetic innervation is primarily
excitatory, contracting the IAS, and is supplied by the lumbar
splanchnic nerves from L2 to L4 ganglia and by the hypogastric
nerves from the inferior mesenteric ganglion. a-Adrenergic
stimulation is excitatory, but b-adrenergic stimulation is
inhibitory. The continuous tonic state of IAS appears to be
mediated by excitatory fibers of both adrenergic and choliner-
gic innervation. Parasympathetic innervation comes from the
S2 to S4 ganglia through the pelvic nerves and is generally
inhibitory, causing the IAS to relax.7,8
The IAS also has an important intrinsic innervation, which
is responsible for reflex relaxation in response to rectal disten-
tion. These intrinsic nerves lie in intramural plexuses and in
the myenteric (Auerbach), deep submucosal (Henke), and
submucosal (Meissner) plexuses and are nonadrenergic and
CHAPTER 104 noncholinergic in origin.6,7
The external anal sphincter is a striated muscle that is
partially under voluntary control. The muscle surrounds the
anus below the dentate line and is attached anteriorly to the
Other Disorders perineal body and posteriorly to the anococcygeal raphe.
The external anal sphincter has traditionally been divided in
three anatomic components: subcutaneous, superficial, and
of the Anus and deep. It is innervated by rami of the pudendal and perineal
nerves, which originate from sacral roots 2 to 4.9

Rectum, Anorectal Although capable of phasic contraction, the external anal

sphincter may also have tonic contraction.10 Unlike the IAS,
the tonic contraction of the external sphincter depends
Function on extrinsic efferent innervation. The phasic contractions
are born voluntary and involuntary; thus the external anal
sphincter provides part (10% to 15%) of the resting anal
Risto J. Rintala and Mikko P. Pakarinen sphincter tone.11
The pelvic floor muscles consist of the levator ani muscle
group, one part of which, the puborectal muscle, forms a sling
around the anorectal junction. Although a striated muscle, the
puborectalis remains in tonic contraction.12 This creates an
Anatomy angle between the anus and the rectum, the anorectal angle,
------------------------------------------------------------------------------------------------------------------------------------------------
which is 80 to 90 degrees at rest and during defecation more
The anal canal measures 2.5 to 4 cm in length (Fig. 104-1).1 It obtuse, at 100 to 105 degrees.13 With coughing, straining, or
is the area between the anal verge and the junction of the strat- other sudden increases in intra-abdominal pressure, the pub-
ified cuboidal and columnar epithelium (dentate line).2 It is orectalis automatically increases its tonic contraction, hereby
lined by squamous epithelium in the lower anal canal below maintaining the sharp anorectal angle that plays a significant
which sebaceous glands and hair follicles arise but changes to role in fecal continence.12 The levator ani muscle complex that
stratified cuboidal and, finally, columnar epithelium of the contracts in a vertical plane is shaped like a funnel that sur-
rectum. Sensitive sensory receptors in this epithelium and rounds the rectum and tapers down to the anal canal, where
in the more proximal anal mucosa respond to a variety of its fibers merge with the voluntary anal sphincter.
stimuli, permitting discrimination of solid, liquid, and gas.
The mucosa of the rectum meets the epithelial lining of the
anal canal at the dentate or pectinate line, marking the oral
extension of the anal columns and valves. High in the anal Physiology of Anal Continence
canal, the mucosa forms four to six longitudinal folds. ------------------------------------------------------------------------------------------------------------------------------------------------

The smooth muscle of the internal anal sphincter (IAS) The adaptive compliance of the colon with the retentive mech-
is continuous with the inner circular muscle of the rectum. anism of the anorectum is required for fecal continence. The
It becomes more prominent low in the anal canal. It is approx- simple anatomic impedence to the descending fecal mass
imately 3 cm long and 5 mm thick.3 IAS is bound to voluntary occurs because the distal end of the gastrointestinal tract is
1311
1312 PART VII ABDOMEN

Sensory Elements Motor Elements

Rectal wall
and stretch
receptors
Rectal wall

Levator ani
Puborectalis
Puborectalis

“Striated muscle complex”

“Striated muscle
complex” Internal sphincter
Anal mucosa Deep
External
Superficial
sphincter
Subcutaneous
FIGURE 104-1 The main ana-
tomic components of the rectum
and anus.

not a straight tube. The angulations in the sigmoid colon and and is sampled by the exquisitely sensitive sensory receptors
the valves of Houston in the rectum impede the caudal move- in the anal mucosa, which allows evaluation of stool consis-
ment of fecal contents. The 80-degree anorectal angle also tency and recognition of flatus. The need to defecate is appre-
assists fecal continence.14 The longitudinal mucosal folds in ciated, but this urge can be voluntarily inhibited. The external
the anus also contribute to continence of semisolid or fluid anal sphincter and puborectalis can be contracted voluntarily
feces by the way they fold together, even when the internal to abort the fecal expulsion. The rectum and colon stretch
sphincter partially relaxes. compliantly, decreasing the intrarectal pressure and the urge
The anal sphincters play an important role in maintaining to stool.
continence, and both the voluntary external sphincter and the As more stool is delivered to the rectum, the sensation in-
involuntary internal sphincter have a resting tone. Fecal con- creases and rectal waves intensify. The urge to defecate is
tinence requires a normally functioning IAS, external sphinc- increased, and the reflex inhibition of the sphincters becomes
ter complex, and levator funnel, as well as intact sensory input greater. The reflex suppression of the tonic contraction of the
from the rectum and anal canal.15,16 The rectum has stretch external anal sphincter and puborectalis begins; if time and
receptors, which have an increasing sensitivity the more distal place are appropriate, voluntary defecation can occur.
they are sited. The anal canal is richly endowed with sensory Young children have a shorter intestinal transit time than
receptors for most modalities of sensation.17 These receptors older children, and this correlates with the more frequent
allow us to distinguish between flatus, fluid, and feces. bowel movements in infants.20 Infants who are fed by breast
Gross continence, the ability to hold large volumes of solid milk or cow’s milk–based formula have one to seven bowel
or liquid feces, is a function of the intact anorectal angle and movements per day.21 In children between 1 and 4 years of
tonic contraction of both external and internal sphincter sys- age, 85% pass stool once or twice per day.22 Normal English
tems. Fine continence—the control of small volumes of feces school children23 have mean transit times of 26 hours and
or flatus—is the function of the coordinated action of the constipated children of 80 hours. Healthy young adults have
sphincters. The distal 2 cm of the anal canal above and at transit times from 30 to 48 hours.24
the dentate line is the critical site of fine continence.18 This
is the site where the “sampling reflex,” discrimination between
solid stool, liquid, or gas, is initiated.19 Constipation
------------------------------------------------------------------------------------------------------------------------------------------------

Constipation in children is a common condition, especially in

Physiology of Defecation the Western world. Unresolved constipation can lead to fecal
------------------------------------------------------------------------------------------------------------------------------------------------
retention and impaction, and finally to overf1ow incon-
The defecation sequence is a complex combination of invol- tinence. In the overwhelming majority of patients the exact
untary actions and voluntary accomplishments that are con- cause of constipation remains obscure; in this case the condi-
trolled by higher cerebral centers. The rectum fills gradually tion is called functional constipation.
by colonic action. The distension of the rectum stimulates
stretch receptors in the rectal wall and levator and induces
DEFINITIONS
an initial contraction of the voluntary sphincter complex,
retrograde emptying of the distal rectum, and relaxation of The consensus Rome III classification is most commonly used
the IAS. This causes a sensation to desire to defecate that in the diagnosis of functional constipation in children.25
increases in intensity as the rectum fills. Stool is temporarily Functional constipation in children is defined as constipation
allowed to contact the sensitive mucosa of the anal canal not associated with congenital abnormalities, acquired
 CHAPTER 104 OTHER DISORDERS OF THE ANUS AND RECTUM, ANORECTAL FUNCTION
diseases, or medication. The following criteria apply for chil-
dren of all ages:
Two or more of the following must exist:
1. Two or fewer defecations in the toilet per week
2. At least one episode of fecal incontinence per week (after
acquisition of toileting skills)
3. History of retentive posturing or excessive volitional stool
retention
4. History of painful or hard bowel movements
5. Presence of a large fecal mass in the rectum
6. History of large-diameter stools that may obstruct the toilet
Both constipation and soiling are common. Soiling has
been reported in approximately 3% of children older than 4
years of age, and constipation accounts for at least 3% of all
medical and 25% of pediatric gastroenterology referrals.26–29
More than 50% of constipated children have a familial inci-
dence, and most studies suggest a male predominance. The
reported ratios range between 1.5:1 and 3:1.30,31
Because constipation during the neonatal period, usually
associated with distention and vomiting, is never functional,
anatomic or mechanical obstruction must be suspected. Most
(94% to 98%) full-term and most (76%) preterm normal
babies pass meconium during the first 24 hours after birth.
All normal babies (100% of full-term and 99% of preterm)
have a first stool with the first 48 hours of life.32 During the
first year of life, symptomatic constipation warrants an evalu-
ation and organic causes for constipation should be ruled out.
During infancy, constipation is often initiated after dietary
manipulations, often the change from breast-feeding to bottle-
feeding or the introduction of solid foods.33 Specifically, in
children anal fissure or perianal dermatitis with group A Strep-
tococcus causes a vicious cycle of stool withholding and painful
defecation and chronic constipation. A suggested alternative
etiology is cow’s milk protein intolerance.34 Dietary fiber is
poorly associated with constipation except in older children.35
Constipated parents are more likely to have constipated
children.36 Slow transit constipation is a major problem in
adults and is probably important in a subset of children with
constipation.37,38 Psychologic problems are common in con-
stipated children, but they are secondary to constipation in the
majority of cases.30,39
ACUTE CONSTIPATION
Acute constipation may be secondary to inactivity, changes of
environment or diet, or an anal fissure. Presentation as acute
abdominal pain is common. Acute constipation presenting
with abdominal pain is usually relieved by one single enema.
The management of acute constipation is usually straightfor-
ward, especially in infants and toddlers. Adding more water
to the diet and restriction of cow’s milk intake usually relieve
the symptoms. Older children and those who have an acute
anal fissure require bulk laxatives for a variable period of time.
CHRONIC CONSTIPATION
Persistent constipation, which does not rapidly respond to
dietary manipulation or simple laxative treatment, can be
defined as chronic. A child with chronic constipation com-
monly presents with fecal soiling. Organic causes that should
be taken into account in the diagnostic workup for chronic
constipation are summarized in Table 104-1.
1313
TABLE 104-1
Differential Diagnosis of Chronic Constipation: Organic
and Acquired Causes
Hirschsprung Disease and Allied Disorders
Internal sphincter achalasia
Intestinal neuronal dysplasia (hyperganglionosis)
Hypoganglionosis
Congenital Anomalies
Anal stenosis
Anterior perineal anus
Acquired Diseases
Chronic anal fissure
Chronic anal fistula
Crohn disease
Associated with Systemic Disease
Hypothyreosis
Hypercalcemia
Cerebral palsy and other neurologic impairment conditions
Uremia
Psychiatric Disease
Depression
Anorexia nervosa
Primary encopresis
Medication
Anticonvulsive drugs
Psychiatric drugs
Anticholinergic drugs
Studies of children with constipation and overflow soiling
show that 40% were never completely toilet trained and that
enuresis is an associated problem in more than 30%.28
Chronic constipation and soiling typically present between
ages 2 and 4 years40; however, up to 40% of the patients have
the onset of symptoms during the first year of life. The diag-
nosis of chronic constipation relies on history and clinical
examination. A detailed bowel history should focus on the
age at which constipation first occurs; the frequency and de-
scription of stools; and therapeutic interventions previously
attempted including any medications taken. Anticonvulsants,
diuretics, antacids, and supplemental iron preparations are
frequently associated with constipation. Family history can
be important, particularly a history of Hirschsprung disease,
cystic fibrosis, or familial constipation. The incidence of
clinical findings in patients with chronic constipation is
summarized in Table 104-2.
On clinical examination the child’s abdomen is usually
nontender and rarely distended. Stool masses are frequently
palpable above the pubic symphysis and in the lower left ab-
domen. The perineum must be inspected carefully for the po-
sition and condition of the anus and the perianal skin, as well
as for soiling marks. The normal position of the anus must be
defined because malpositioning is a well-recognized cause of
constipation.41,42 Reisner and colleagues43 defined normal
values by using a ratio of the midanal to fourchette distance
and the fourchette to coccyx distance in the female and mid-
anal to posterior scrotum distance and the posterior scrotum
to coccyx distance in the male. If the ratio is less than 0.34 in the
female (usual values: newborn, 0.44; age 4 to 18 months, 0.40)
1314 PART VII ABDOMEN
TABLE 104-2
Clinical Features of Severe Functional Constipation
Age at Onset
0-1 yr 20-25%
1-5 yr 70%
> 5 yr 10-15%
Boys-to-girls 1-3:1
Previous attempts to treat 80-90%
Soiling 70-75%
Unsuccessful toilet training 70-80%
Pain at defecation 70-80%
Abdominal pain 50-60%
Occasional blood in stool 25-30%
Poor appetite 20-30%
Wetting 20-30%
Primary psychopathology 15-20%
Rectal prolapse 3%
Data from references 28, 30, 36, 52, and 53.
and less than 0.46 in the male (usual values: newborn, 0.58;
age 4 to 18 months, 0.56) the child should be investigated
carefully, especially if constipation is present. To differentiate
between anterior anus and perineal ectopic anus is often dif-
ficult, especially if the anterior position of the anus is associ-
ated with stenosis. Final diagnosis may require muscle
stimulation in general anesthesia.
Perianal sensation to rule out neurologic disorders can
be evaluated by stroking the perianal tissue gently with a
cotton-tipped applicator, watching for the anal puckering,
and ascent of the perineum. The child’s underwear should also
be examined for soiling.
It is advisable to perform a digital rectal examination at least
once in a child with chronic constipation to rule out organic
obstructing causes. The fecal mass that fills the rectum may
have developed to a fecaloma that can be stone hard. The
rectorectal space must be examined to feel masses within
the hollow of the sacrum. The closing reflex should be seen
when the finger is withdrawn. The absence of either the per-
ineal cutaneous or closing reflexes suggests an underlying
neurologic disorder.
If the clinical history and physical examination do not
suggest organic etiology, a trial of medical treatment can be
initiated. A small infant with a history of neonatal symptoms
and early onset constipation should undergo barium enema
without bowel preparation. An abnormal barium enema
should be followed by rectal biopsy and possibly anorectal
manometry to rule out Hirschsprung disease and allied disor-
ders. In an older child without any history of significant
neonatal constipation a primary barium enema or biopsies
are not necessary44,45 before the onset of medical manage-
ment. Barium enema and other imaging such as plain abdom-
inal radiography, transit time studies, or magnetic resonance
imaging (MRI) are indicated in patients who have poor
response to appropriate medical treatment or abnormal clin-
ical findings such as neurologic symptoms. Patients who have
a poor response to optimal medical therapy should undergo
also rectal biopsy and anorectal manometry to rule out rare
forms of dysganglionoses such as hypoganglionosis, intestinal
neuronal dysplasias, and internal sphincter achalasia.
Barium enema usually shows dilatation of the rectosigmoid
that extends to the anal canal. This rules out classic Hirsch-
sprung disease but not internal sphincter achalasia.46,47 Plain
abdominal radiography may reveal spinal vertebral anomalies
in patients who have clinical findings suggesting a neurologic
disorder. In case of vertebral anomalies an MRI study should
be performed to rule out intraspinal pathology such as tether-
ing and intraspinal lipomas.48 Transit time studies by radi-
opaque markers37,49 or radioisotopes38 have been used to
assess colonic motility in patients with poor response to stan-
dard medical management. These studies have revealed that
slow-transit constipation occurs in a significant proportion
of patients with recalcitrant constipation. The main method-
ological problem with transit studies in children is the lack
of standardized methods and normal values in healthy
children.
Anorectal manometry is useful only for the diagnosis of
Hirschsprung disease and internal sphincter achalasia. In both
these conditions the rectoanal relaxation reflex is missing.
In patients suffering from functional constipation, anorectal
manometry does not improve diagnostic or therapeutic
accuracy. Some constipated children with normal histology
also have manometric evidence of anal outlet obstruction
and paradoxically contract the sphincter complex with
straining, so-called rectoanal dyssynergy.46,50–52
MANAGEMENT OF CHRONIC IDIOPATHIC
CONSTIPATION
The general approach to the management of a child with
chronic constipation includes the following steps39,53:
• Provide parental counseling and education
• Determine whether a fecal impaction is present
• Disimpact the fecaloma if present
• Initiate oral medication
Counseling and Education
Counseling and educating the family are the first steps in the
management. The pathogenesis of constipation needs to be
explained to the parents. If the patient has fecal soiling, the
involuntary nature of overflow incontinence needs to be clar-
ified to the parents. Parents are encouraged to maintain a
consistent supportive attitude during the long period of treat-
ment. Education of the child and parents, emotional support,
and a commitment to continue to see the family until normal
stooling is established are all important elements of the pro-
gram.51,53,54 The patient and the parents need to be aware that
it may take 6 to 12 months or even years before a normal
stooling pattern is achieved.
Disimpaction
If the patient has fecal impaction, disimpaction is necessary
before initiation of oral maintenance therapy. A fecal mass
can be identified by physical examination in the lower abdo-
men, rectal examination, or radiographic methods. A typical
symptom of fecal impaction is overflow incontinence.
Disimpaction has been traditionally accomplished with bowel
washouts, but oral medication is effective, too.55,56
Oral disimpaction can be accomplished by high doses of
stimulant laxatives, docusate, mineral oil, and polyethylene
glycol-electrolyte (PEG) solutions.54,56 Osmotic laxatives such
 CHAPTER 104 OTHER DISORDERS OF THE ANUS AND RECTUM, ANORECTAL FUNCTION
as lactulose or sorbitol can be used in combination with
other medication. Oral disimpaction is often associated with
abdominal pain and colic, as well as an initial increase in fecal
soiling.
Rectal washouts usually work faster than oral disimpaction.
It is, however, invasive and painful, especially in patients who
have associated anal pathology. Therefore rectal disimpaction
is contraindicated in children with anal fissure. Saline, docu-
sate, mineral oil, or phosphate enemas are recommended
by different investigators.53,57,58 When rectal disimpaction
is used, it is essential that the number of enemas is kept min-
imal. Usually one to three washouts are required for complete
disimpaction. In recalcitrant cases manual evacuation under
general anesthesia may be considered.
Maintenance Therapy
The goal of maintenance therapy is to produce one to two soft
stools per day and prevent recurrent fecal impaction. This
ensures that the vicious cycle of hard stools and painful
defecation will be abolished. Treatment consists of dietary
interventions, laxatives, and behavioral modification. Dietary
changes are universally advised, particularly increased intake
of fluids and fiber. The role of fiber may not be significant in
children.35 Too much milk is discouraged.59
There is no evidence that any particular drug treatment
regimen is superior. Good compliance with the selected treat-
ment is more essential. In the early phases of management
after the disimpaction, more effective medication is often re-
quired. Stimulant laxatives (Senna, sodium picosulfate)
should be used for short periods, as short as possible, and
should be replaced by less harmful therapies. Osmotic laxa-
tives (lactulose, docusate, PEG) can be safely used for months
and years. A recent meta-analysis has shown that PEG is
superior to lactulose for childhood chronic constipation.60
Having achieved regular pattern of bowel movements, the
medication needs to be continued as long as needed. This
usually means months and sometimes years. A good thumb
rule is that the treatment continues as long as the patient
has had symptoms before the management. The dosage of
the medication needs to be tapered regularly; the goal is to
use a minimum effective dose.
Behavioral Therapy
Behavioral therapy varies with the age of the patient. In infants
and toddlers, behavioral therapy has no role; it is important
that too early and aggressive toilet training is discouraged.
In young children younger than 2 to 3 years of age, the toilet
should be avoided and diapers reinstituted as long as the child
is ready to go back to potty or toilet.53 Older children may
benefit from regular toileting routines after a major meal dur-
ing the day. Praise, rewards, and a diary may contribute to a
successful outcome.55
Long-Term Outcome
There are few studies of the long-term results of the manage-
ment of chronic idiopathic constipation.36,61 The reported
final cure rates after 5 years of follow-up range between
50% and 70%. A recent longitudinal follow-up study showed
that one third of children with chronic constipation continue
to have severe complaints of constipation beyond puberty.62
1315
Surgical Options for Chronic Constipation
The majority of children with chronic constipation improve
by appropriate medical therapy or with time. Most patients
also comply even with aggressive medical therapy. However,
a few children, despite optimal and maximal medical manage-
ment, have persistent constipation, abdominal symptoms, and
soiling throughout and beyond childhood. There are no con-
trolled studies in children devoted to the surgery of pediatric
idiopathic constipation.
For pediatric patients who are refractory to all medical ther-
apy, there are some viable surgical options. It is essential that
organic causes of constipation are ruled out. The MACE pro-
cedure63 that has been originally used for neuropathic incon-
tinence or incontinence following management of anorectal
anomalies has been successfully used also for idiopathic con-
stipation.64,65 The MACE procedure can be easily reversed,
which may make it an attractive alternative for enemas and
laxatives, especially for pubertal and adolescent patients. Re-
section of the megarectum and megasigmoid has been per-
formed for patients with functional constipation, but there
is a complete lack of controlled studies and longer-term
follow-up data. Moreover, high failure rates have been
reported following colon resections.66 Colostomy has been
used in selected patients with significant patient and parent
satisfaction.66,67
Functional Fecal Soiling Without
Constipation
------------------------------------------------------------------------------------------------------------------------------------------------
There is a small subgroup of children with fecal soiling who
are otherwise healthy without constipation or any other
organic or neuropsychiatric underlying cause for the inconti-
nence.68–71 A typical age of presentation is between 4 and
8 years of age. The main symptom is involuntary passage of
variable amounts of stool to the underwear one to several
times a day. The patients rarely soil at school or during sport
and social activities. Most have regular bowel actions. At least
one third of these patients suffer from daytime or nighttime
wetting.
The authors have encountered an increasing number of
such patients during the past decade.72 Most of these patients
had had extensive but unsuccessful therapies to treat consti-
pation. No signs of neuropsychiatric disorders or myelopathy
in MRI were observed. Patients had normal barium enemas
and spinal radiographs. All had developed normally and went
to normal schools. The incidence of daytime or nighttime en-
uresis was 40%. The only measurable functional abnormality
was an isolated impairment of rectal sensibility at anorectal
manometry; sphincter performance was comparable with con-
stipated control individuals. Treatment consisted of counsel-
ing, toilet training, and dietary modification. Some patients,
especially those with accompanying enuresis, benefit from
anticholinergic treatment with oral oxibutynin hydrochloride
(5 to 15 mg/day). Most patients improved with decreased
frequency of soiling. All the patients who have reached ado-
lescence have experienced striking improvement of soiling.
The most important issue with this subgroup of patients with
encopresis is to keep in mind that childhood fecal soiling
is not always related to constipation.
1316 PART VII ABDOMEN

circumferential.76 Rectal prolapse and mucosal ectopia were

Rectal Prolapse much more common before the era of the PSARP procedure.77
------------------------------------------------------------------------------------------------------------------------------------------------
Rectal polyps may be a leading point for a prolapse.
Rectal prolapse is a relatively common, usually self-limiting The diagnosis of rectal prolapse is usually based on history.
condition in children. The peak incidence is between 1 and Most commonly a rosette of rectal mucosa is noted after defeca-
3 years of age. Prolapse can be either partial or complete. In tion. The child complains that something comes out of the anus.
partial prolapse the rectal mucosa protrudes only about 1 to Usually the prolapse reduces spontaneously but must be some-
3 cm from the anal verge with characteristic radiating folds times reduced manually. In mild forms the prolapse comes out
from the center of the anal aperture. In complete prolapse, occasionally following major straining or during diarrheal ill-
the full thickness of the rectum is involved; 5 cm or more ness. The problem is more annoying and worrisome for the
of the rectum protrudes, and the prolapse is distinguished patient and parents if the prolapse occurs after every defecation.
by the circular folds of the mucosa (Fig. 104-2). There is sig- Usually the prolapse cannot be provoked when the child is
nificant controversy as to whether rectal prolapse in children brought to consultation. Rectal examination is indicated to
is partial or complete. rule out rectal polyps and ulcers. If there is a history of rectal
bleeding, colonoscopy may be necessary to look for higher
polyps or other lead points. Dynamic defecography is war-
PATHOGENESIS AND DIAGNOSIS
ranted at least when a prolapse is associated with rectal ulcer,
The vast majority of patients suffering from rectal prolapse do suggesting intussuscepting prolapse of the sigmoid colon or
not have any predisposing factors. The children suffering from accompanying enterocele.
idiopathic rectal prolapse are usually otherwise healthy. The
role of constipation as an etiologic factor is controversial; only
3% of patients suffering from severe chronic constipation have
TREATMENT
rectal prolapse.30
Several organic conditions predispose to rectal prolapse. In acute prolapse, reduction may occur spontaneously on
Cystic fibrosis is associated with rectal prolapse.73 More than standing up. If not, the prolapse must be reduced as soon
a fifth of the patients with cystic fibrosis develop rectal pro- as possible. The parents often rush the child to a hospital
lapse.74,75 The neuropathic causes of complete rectal prolapse when the prolapse appears for the first time. The tip of the her-
excluding myelomeningocele are rare. Nevertheless, paralysis niated bowel can usually be gently pushed into the anus.
of the levator ani with raised intra-abdominal pressure leads to If edema has formed, a gentle squeezing pressure may be
procidentia and prolapse. In ectopia vesicae there is wide required. Reduction technique must be taught to the parents.
separation of the symphysis pubis and the puborectalis mus- There is spontaneous cure in most cases of recurrent pro-
cle, and this wide hiatus predisposes to prolapse of the pelvic lapse.78 In many cases the prolapse reduces spontaneously.
organs including the rectum. Besides severe malnutrition, In cases without spontaneous reduction the parents can
connective tissue diseases (e.g., Ehler-Danlos) and behavioral reduce the prolapse gently if appropriately instructed. Accom-
disorders (e.g., Asperger) predispose to rectal prolapse. panying constipation is treated with laxatives when present.
Iatrogenic full-thickness rectal prolapse may occur follow- Local transanal treatments such as injections of the prolapse,
ing pull-through operations for high anorectal anomalies. multiple linear thermocauterization to the mucosa, excision of
Much more common is mucosal prolapse that is usually not redundant mucosa, or insertion of a subcutaneous suture
around the anus are not tested in controlled trials.
Operation is indicated in rare cases with intractable prolapse
and may be considered in patients who are not spontaneously
cured in 12 to 18 months of follow-up. Patients older than
4 years of age require surgery much more often than younger
children. There are several surgical methods that have been
used with success for recurrent prolapse. We prefer laparo-
scopic suspension of the rectum to anterior sacrum with rou-
tine suture closure of the space between the rectum and the
vagina or the urinary bladder in order to avoid development
of enterocele.79 An additional resection of the sigmoid colon
may be performed in intussuscepting prolapse of the sigmoid
colon and in recurrent cases. Laparoscopic approach has been
successful in nearly 20 patients that have required surgery. The
procedure is associated with minimal postoperative pain and
short hospital stay. Patients benefit from laxative therapy during
the early postoperative period. Posterior sagittal approach with
muscle repair and suspension of the rectum to the sacrum,80–83
posterior rectal plication,84 and Ekehorn rectosacropexy85
are also reported to be associated with a high cure rate.
Secondary operation is indicated for iatrogenic prolapse
after a pull-through operation in symptomatic patients.
Typical symptoms include bleeding and leak of mucus.
FIGURE 104-2 Rectal prolapse in a 2-year-old boy. In patients with mucosal prolapse treatment involves excision
 CHAPTER 104 OTHER DISORDERS OF THE ANUS AND RECTUM, ANORECTAL FUNCTION 1317

of the mucosal ectopia and reconstruction of a skin-lined anal stools streaked with bright-red blood. Anal fissure is often
canal with a local skin flap. Patients with a complete prolapse but not necessarily associated with constipation, which is
require a repeat posterior reconstruction of the levator funnel caused by fear of painful defecation.
and external sphincter complex, as well as rectal suspension. The diagnosis is made by direct inspection. The typical
longitudinal tear distal to the dentate line can be visualized
by retracting the perianal skin gently away. No further diag-
nostic modalities are necessary. The most common location
Anal Fissure of idiopathic anal fissure is posterior midline, but especially
------------------------------------------------------------------------------------------------------------------------------------------------
in infants it may be found anywhere in the anal circumference.
Anal fissure is a longitudinal tear or ulcer in the distal anal In female infants a common site of anal fissure is the anterior
canal epithelium extending to the anal verge. Most acute midline. A sentinel pile or skin tag at the area of the fissure
fissures heal spontaneously within a few weeks, but a pro- is associated with chronic or subchronic fissure. Atypical
portion become chronic. Anal fissure is the most common fissures may be multiple and often off the midline and are
cause of hematochezia in childhood, and it is one of the most commonly large and irregular. Atypical appearance of fissure
common lesions to consider in the differential diagnosis of should initiate further investigations including biopsy,
anal pain. Anal fissures are common, although their exact cultures, and colonoscopy to rule out Crohn disease, immu-
incidence in children is unknown. nodeficiency states, tuberculosis, venereal infection, and
malignancies.
PATHOGENESIS
TREATMENT
Pathogenesis of idiopathic anal fissure is still incompletely
understood, and it may differ in adults and children.86 Anal Most idiopathic anal fissures in children heal without any
fissures in childhood are often associated with secondary specific treatment in a few months.89,90 Only symptomatic
constipation due to painful passage of stools. The classic con- fissures require treatment. If fissure is associated with consti-
cept of mechanical tear caused by hard stools as a primary pation and/or painful defecation, stool softening with dietary
causative factor may be too simple and outdated. However, de- modification and bulk laxatives is indicated. Lubricants ease
liberate avoiding of defecation does cause rectal distension painful passage of stools. The goal is to interrupt the vicious
and leads to decreased rectal sensation, which in turn, results circle of painful defecation, fecal retention, hard stools, and
in infrequent, bulky, and hard stools that prevent healing of prevention of healing of fissure. As expected, most fissures re-
fissure. Fear of painful defecation may lead to fecal retention spond promptly to stools softening and heal in several
and gives rise to a vicious circle. weeks.89 Hematochezia stops when fissure heals. Occasion-
There is a widely accepted theory on the pathogenesis of ally a child presents with typical history after the symptoms
anal fissure in adults.86,87 According to this theory increased have disappeared and the fissure has healed. Initially, these
internal sphincter pressure and muscle spasm lead to im- patients may be treated expectantly unless no abnormal
paired tissue perfusion and finally epithelial ulceration. Spasm clinical signs are present and hematochezia has not recurred.
of internal anal sphincter is so severe that the pain caused by Botulinum toxin injection into sphincter muscles in order
fissure is thought to be due to ischemia. The most common to overcome increased pressure is a novel treatment for
site of idiopathic anal fissure is posterior midline, which is less chronic fissures. Quick and effective healing has also been
vascularized than other areas of the anal canal. Anal canal rest- reported in children.91 We use botulinum toxin injections into
ing pressure is increased in patients with anal fissure. Decrease the internal part of the sphincter complex with a dose of
of anal canal pressure after surgical or pharmacologic sphinc- 15-25 U, depending on the patient’s age, into each of the four
ter relaxation is accompanied with improved perfusion of quadrants. Usually healing occurs in several weeks and injec-
anoderm and healing of chronic fissures. Currently, it is un- tions may be repeated in refractory cases. After encouraging
known whether this theory also applies to pediatric patients. initial results in adults, several recent randomized placebo-
In children a vast majority of idiopathic anal fissures heal controlled trials have assessed efficiency of topical glyceryl
without any specific therapy. This may be due to relatively trinitrate in anal fissure in children.89,90,92,93 Two studies
better tissue perfusion of the anal canal, greater regenerative reported faster healing of fissures and relief of symptoms in
capacity in general, or different pathogenesis in children than children treated with glyceryl trinitrate,90,93 whereas no ben-
in adults. An unhealed fissure may become inflamed due to efit was found in one.89 Few children experienced temporary
bacterial infection and chemical and mechanical irritation. incontinence, and none reported headache during glyceryl
As a result of long-standing inflammation, chronic anal fissure trinitrate treatment.89,90,93 Taken together, topical glyceryl
may have hypertrophied anal papilla proximally and a sentinel trinitrate for anal fissures is only marginally better than
skin tag distally. This kind of chronic anal fissure is only rarely placebo.92
seen in children and should raise the suspicion of underlying Surgical therapies reported for treatment of anal fissure
Crohn disease.88 in children include fissurectomy, anal dilatation under general
anesthesia, and lateral internal sphincterotomy.94,95 Lateral
DIAGNOSIS subcutaneous sphincterotomy also appears to be an effective
procedure in children.94 Fissure cure rates (80%) after fissur-
Anal fissure may occur in any age. Typical age of presentation ectomy combined with laxatives are comparable with simple
is around 2 years. Most often anal fissures present with bright laxative therapy.89,95 Anal dilatation causes unpredictable
red rectal bleeding that may be associated with painful defe- degree of sphincter damage and should be avoided. In adult
cation. The child may cry with bowel movements and have patients, lateral sphincterotomy is associated with an
1318 PART VII ABDOMEN

incontinence rate of 10%.96 Anal stretch has a significantly

higher risk of minor incontinence than sphincterotomy.96
Outcomes of different surgical procedures for fissures are
poorly characterized in children. Thus lateral internal sphinc-
terotomy should be reserved for those rare children whose
anal fissure has progressed to a real chronic fissure after treat-
ment with botulinum toxin has failed. In these cases a biopsy
should be obtained to rule out possible Crohn disease.

Perianal Streptococcal
Dermatitis
------------------------------------------------------------------------------------------------------------------------------------------------

Perianal streptococcal dermatitis is a common cause of anal

complaints (e.g., pain, itching, anal discharge, constipation)
in preschool-aged children. The clinical finding of sharply
demarcated wet perianal erythema is usually characteristic
(Fig. 104-3). A positive culture of group A or B b-hemolytic
streptococci from a perianal swab confirms the diagnosis.
The treatment includes oral antibiotics for 10 to 14 days
with penicillin V or cephalosporin and topical antimicrobial FIGURE 104-4 Typical perianal abscess in a male infant.
therapy.97
subcutaneous and straight. It usually traverses from the
Perianal Abscess and Fistula affected crypt through the subcutaneous external sphincter
to the perianal skin. The fistulas are usually distributed evenly
in Ano around the anal circumference. Multiple lesions occur in 15%
------------------------------------------------------------------------------------------------------------------------------------------------
to 20% of cases (Fig. 104-5).103,104
Perianal abscess is not uncommon in small infants. The vast The traditional management of perianal abscess is incision
majority of patients are male.98 A congenital etiology has been and drainage. This is associated with a significant recurrence
suggested for infant anal fistulas.99,100 A proposed relation- rate.104,105 However, at the end of the day most abscesses are
ship to androgens resulting in congenital deep, epithelialized cured by expectant treatment only.104,105 Fistula-in-ano has
crypts may explain the predominant occurrence in male been considered as an indication for surgical treatment. The
infants.100,101 A typical presentation is a perianal abscess in traditional method has been identification of the affected anal
a child younger than 12 months (Fig. 104-4), in the majority crypt and fistulotomy by deroofing of the fistula tract. Recur-
before the age of 6 months. After initial incision the condition rent fistulas occur in 10% to 20% of cases. Recent reports
may progress to or recur as an anal fistula. The incidence of have advocated expectant treatment for asymptomatic
fistula formation in patients with perianal abscess may be as
low as 10% to 20%.102,103 The fistula may present alone with-
out preceding perianal abscess. The fistula is typically

FIGURE 104-5 Bilateral fistula-in-ano. The probe passes straight from

FIGURE 104-3 Perianal streptoccoccal dermatitis. the skin opening to the corresponding anal crypt.
 CHAPTER 104 OTHER DISORDERS OF THE ANUS AND RECTUM, ANORECTAL FUNCTION 1319

fistula-in-ano; most fistulas heal within 12 to 24 months

without further sequelae.104,105
Fistulas in older children or adolescents are cryptoglandu-
lar or associated with inflammatory bowel disease, mainly
Crohn disease. Crohn disease should be ruled out in all
children who present outside the typical age groups, infants
and adolescents. Treatment of adolescent fistula-in-ano
should be along the same lines as in adults. The fistulous tract,
once identified, is either incised and left open to granulate or
excised with primary closure of the defect.

Vascular Malformations
------------------------------------------------------------------------------------------------------------------------------------------------

The classifications of intestinal vascular malformations have

been confusing because different labels are laid on similar le-
sions without knowledge of their clinical and biologic proper-
ties. The classification of vascular malformations by Mulliken
and Glowacki106 is based on biologic properties of the lesion.
The classification distinguishes between hemangiomas that
are true neoplasms and usually regress spontaneously, and
FIGURE 104-6 Hemorrhoids in a 5-year-old boy, without any underlying
vascular malformations, which are nonproliferative lesions conditions. Note the typical adult type piles at 4, 7, and 11 o’clock.
that do not regress. Intestinal bleeding from vascular malfor-
mations in children is rare. The most common sites are distal
colon and rectum.107 The vascular anomalies in the distal from the anal opening. A more common finding, especially
bowel are usually venous malformations and not hemangi- in constipated children, is a prominent venous plexus around
omas as previously thought.108 In the minority of cases the the anal opening. This is a common source of rectal bleeding
vascular lesion is part of a systemic disease such as Klippel- in constipated children. The venous sinuses of this venous
Trenaunay (congenital varicose veins, cutaneous hemangiomas, plexus may rarely thrombose, causing similar symptoms as
and ectatic hypertrophy of the lower limbs) and Osler-Rendu- in adults with acute thrombosis of external hemorrhoids.
Weber (multiple telangiectasia) syndromes. Adult-type hemorrhoids begin to occur in adolescents
A typical symptom of rectal vascular malformations is re- and may be complicated by thrombosis of the external part
current hematochezia that may sometimes be profuse. The pa- of piles.
tient may also present with hemorrhoids. The diagnosis of In otherwise healthy children hemorrhoids do not require
vascular anomalies may be difficult. In the distal bowel endos- surgical therapy. Major bleeding does not occur. Symptomatic
copy may show findings that suggest localized inflammation. patients with portal hypertension may require treatment.
Dilated vessels are rarely visible. MRI and angiography are the Banding or sclerotherapy controls the symptoms in most
best methods to diagnose and localize intestinal vascular le- cases.112
sions. Nonoperative methods may be used to control bleeding,
but permanent cure is best achieved by complete resection of
the lesion. Lesions that extend to the low rectum and anal ca-
nal are best treated by endorectal pull-through and colo-anal Solitary Rectal Ulcer
anastomosis.109,110 This sphincter-saving operation eradicates ------------------------------------------------------------------------------------------------------------------------------------------------

bleeding episodes for long periods of time, if not permanently. Solitary rectal ulcer syndrome (SRUS) is a chronic, benign
disorder characterized by hematochezia, mucous discharge, te-
nesmus, and local perianal pain. It is rare in children but should
be kept in mind in patients with local anal symptoms.113,114
Hemorrhoids In children the macroscopic finding at endoscopy is
------------------------------------------------------------------------------------------------------------------------------------------------
a thickened and edematous lesion that may have an ulcerative
Hemorrhoids are uncommon in children unless in those with or polypoid appearance in the anterior rectal wall 2 to 5 cm
portal hypertension. One third of the children with portal hy- above the anal verge. We have encountered 12 cases in the past
pertension have hemorrhoids. Hemorrhoids are more com- 15 years. Some patients have a feeling of incomplete evacuation
mon in patients with extrahepatic portal hypertension than with frequent visits to the toilet associated with chronic
in those with intrahepatic disease.111 Symptomatic hemor- straining at stool. Rectal or internal rectosigmoid prolapse
rhoids, however, are uncommon. Hemorrhoids in children has been reported to occur in a significant percentage of
may be associated with rectal vascular malformations. cases.40 According to our experience, conservative manage-
Clinically detectable internal hemorrhoids with external ment with stool softeners and local steroid suppositories is
extension occur also in otherwise healthy children but are rare successful in most children. Open or laparoscopic recto-
(Fig. 104-6). We have seen approximately one to two healthy pexy113,115 has been suggested to correct the external or inter-
children with internal and external hemorrhoids per year nal rectal prolapse that is often associated with SRUS; three
during the past 15 years. Usually there are no symptoms, of our patients have required rectopexy for recalcitrant
but the child or parents have noticed something to protrude symptoms.
1320 PART VII ABDOMEN

Infantile Proctocolitis
------------------------------------------------------------------------------------------------------------------------------------------------
Sexual Abuse
------------------------------------------------------------------------------------------------------------------------------------------------

Apart from anal fissure, the most common cause of hemato- Child sexual abuse is a common and worldwide problem. In
chezia in infants younger than 3 months of age is eosinophilic Europe the incidence appears to be 6% to 36% of girls and
proctocolitis. The infantile proctocolitis typically presents at 1% to 15% of boys younger than 16 years.122 Sexual abuse
the age of 3 to 4 weeks with fresh blood streaks mixed in rarely causes death, but its consequences can be serious and
stools. The stools often contain mucus. Usually there are no persist through adulthood. In child sexual abuse the signifi-
other symptoms, and the growth and development of the cance of findings in the vagina and hymen are well recognized.
infant is normal.116 Colonoscopy shows colitis that is often In the context of anal abuse the findings are more contro-
patchy and rarely extends beyond the left colon. Histology versial and difficult to verify.
reveals marked eosinophilic infiltrate. Some patients may also Hobbs and Wynne123 stated that “dilatation and reflex anal
have an elevated eosinophilic count in peripheral blood. dilatation” are not seen in normal children and are signs of
Allergic etiology has been suggested because of these findings abuse. Reflex anal dilatation (even gross) was found on routine
but is found in a minority of cases.117 The condition is self- examination in 28 of 200 children (14%) examined consecu-
limiting, and symptoms usually subside within a few weeks. tively in community health clinics, a pediatric outpatient
Diet change has been also reported to be helpful in reversing clinic, a district hospital, and a rectal clinic.124 Constipation
the symptoms.118 or feces in the rectum, without associated sexual abuse,
may often produce gaping of the anus on separation of the but-
tocks.125 Hobbs and Wynne126 also suggested that an anal fis-
sure was a sign of abuse in 53% of 143 cases quoted. Pierce127
found that the majority of children with strong or definite his-
Proctalgia Fugax tory of anal abuse had anal fissures or scars. However, an anal
------------------------------------------------------------------------------------------------------------------------------------------------
fissure is far too common a finding in routine clinical practice
Proctalgia fugax (PF) is a benign painful rectal condition that to be regarded as a cause for suspicion of sexual abuse.128
is defined as intermittent, recurring, and self-limiting pain Obvious laceration, bruising around the anus or inner thighs,
in the anorectal region in the absence of organic pathology. and scratches, especially with a suspicious or definitive history of
There was also thought to be a male preponderance; however, abuse, are more conclusive. This sort of forensic evidence is
more recently a female preponderance is reported.119 The rarely found in children because there is usually a significant
onset of symptoms is usually at school age or adolescence. time lapse between the abuse and medical examination.
The etiology of proctalgia fugax is unknown. Spasm of the anal If a child displays total passiveness and indifference to
sphincter complex is the most commonly suggested etiologic physical examination, or to inspection and examination of
mechanism. the anus, with no attempt to withdraw or tighten the striated
Rome III criteria define proctalgia fugax as recurrent muscle complex when a rectal examination is attempted, the
episodes of pain that localize to the anus and lower rectum. clinician should be alert to the possibility of sexual abuse.
The episode lasts from seconds to minutes, and there is no Perianal warts caused by the human papillomavirus should
pain between episodes.120 The condition is uncommon in always raise the possibility of sexual interference because this is
children; the authors see approximately one to two new the most likely mode of transmission.129,130 Although many stud-
patients per year with proctalgia fugax. It is essential to rule ies suggest a high association with sexual abuse (60% to 90% of
out organic causes for anorectal pain before diagnosis of proc- cases), recent studies have questioned this association, particu-
talgia fugax is made. Proctalgia fugax in children is a self- larly in infants. Parents of children with condylomata often have
limiting condition that, however, may need treatment if warts on nongenital skin, and vertical transmission between
episodes of pain are frequent and long lasting. mother and child has been well documented. The presence of con-
There no evidence-based management protocol for dyloma acuminata in infants younger than 1 year of age may
proctalgia fugax, but in most cases topical treatment with represent vertical transmission from the mother; older children
glyceryl nitrate cream or oral spasmolytics is helpful. Other must be treated with a high index of suspicion for child abuse.
options are oral clonidine or inhaled salbutamol. In recal-
citrant cases local botulinum toxin injections may be The complete reference list is available online at www.
helpful.121 expertconsult.com.
 Diagnostic uncertainty made early surgery for biliary
atresia an area of controversy during the 1960s. Thaler and
Gallis felt that surgical manipulation was deleterious in cases
of neonatal hepatitis and recommended a waiting period of
4 months before exploration for possible biliary atresia.6,7
Other authors also reported that there was a possibility of
spontaneous resolution of biliary atresia, further diminishing
the potential merits of early operation. Finally, some believed
that improved outcomes may be obtained after waiting for the
course of the disease to progress with the hope that any ductal
structures would enlarge, making repair more successful.8–11
Outcomes for infants with “noncorrectable” biliary atresia
remained dismal.
In the late 1950s Kasai and his colleagues in Japan noted
that bile flow from the porta hepatis was possible after excision
of the entire fibrotic extrahepatic biliary tree.12–14 By surgi-
cally connecting the porta hepatis to the intestine (the duode-
num in early cases), Kasai achieved the first long-term “cure”
of a patient deemed to have the noncorrectable form of biliary
atresia. Despite these encouraging initial results, Kasai’s he-
patic portoenterostomy did not gain immediate acceptance.
Surgeons both in Japan and in the United States were initially
skeptical of his results.15,16 Over time, however, the outcomes
CHAPTER 105 associated with Kasai’s hepatic portoenterostomy were con-
firmed17,18 and in its most recent form, using a Roux-en-Y
portoenterostomy, this procedure has become the standard
operative treatment for biliary atresia.
The Jaundiced Because many children develop progressive liver failure
despite a technically well-executed hepatic portoenterostomy,
salvage therapies are often necessary. Liver transplantation,
Infant: Biliary popularized by Starzl in the early 1960s,19 is the only readily
available salvage treatment for children with biliary atresia

Atresia and progressive liver dysfunction. In fact, biliary atresia

is the most common diagnosis leading to liver transplantation
in children.
Robert A. Cowles
Incidence, Demographics,
and Classification
History ------------------------------------------------------------------------------------------------------------------------------------------------

------------------------------------------------------------------------------------------------------------------------------------------------
Jaundice is a common finding in the newborn nursery and
Biliary atresia was first described by John Thompson in in young infants. In a majority of cases this hyperbiliru-
1892.1 He detailed 49 clinical cases of congenital biliary ob- binemia is due to elevated unconjugated bilirubin and
struction and documented the autopsy findings in each case. resolves spontaneously with no need for surgical consultation
In 1916 Holmes, a pediatrician and pathologist at Johns or intervention. It is important, however, to identify the rare
Hopkins University, reported his own case, reviewed previ- infant who has persistent, pathologic jaundice beyond
ously reported cases, and concluded that successful surgical 2 weeks of life. These neonates should be presumed to have
treatment for congenital atresia of the bile ducts was theoret- biliary obstruction due to biliary atresia or choledochal cyst
ically possible in at least 16% of cases.2 Ladd was the first to or a cholestatic process due to a myriad of underlying causes,
report successful surgical correction of biliary atresia in cases and a careful evaluation should be promptly undertaken.
where either the gallbladder or the common bile duct com- Biliary atresia occurs in between 1 in 10,000 and 1 in
municated with the liver.3 This success led him to recom- 16,700 live births and this incidence has been consistent
mend a more aggressive surgical approach to cases of over time.20–22 A slight predominance of females exists with
presumed biliary atresia. Despite this, a majority of infants a female-to-male ratio ranging between 1.4 to 1.7 to 1. No
do not have the favorable anatomy that was described by clear genetic factors have been associated with biliary atresia;
Holmes and Ladd and subsequent large series showed low however, the incidence of the disease appears to be higher in
success rates for surgery in cases of biliary atresia. From Asia than in Europe or North America.
the late 1940s to the early 1960s several creative surgical ap- Several systems have been proposed to classify the anatomy in
proaches attempting to obtain bile drainage from the liver cases of biliary atresia. The Japanese Association of Pediatric
were used, but these were ultimately unsuccessful and sub- Surgeons proposed that cases should be classified according
sequently abandoned.4,5 to the location of atresia. In their system, type I anatomy is
1321
1322 PART VII ABDOMEN

associated with atresia at the level of the common bile duct; TABLE 105-1
in type II, atresia is at the level of the hepatic duct; and Congenital Anomalies Associated with Biliary Atresia
in type III, the most frequent type, atresia occurs at the porta Malrotation
hepatis.22 The common anatomic variations are shown in Preduodenal portal vein
Figure 105-1. Polysplenia
Interrupted inferior vena cava
Azygous continuation
Etiology
------------------------------------------------------------------------------------------------------------------------------------------------
Cardiac malformations

The exact etiology of biliary atresia is unknown and likely

multifactorial. Theories implicating genetic, inflammatory,
and infectious causes have been presented, but none has
been proven. The higher incidence of biliary atresia in certain Embryology
populations makes a genetic cause plausible. Moreover, ------------------------------------------------------------------------------------------------------------------------------------------------

the observation that up to 20% of biliary atresia cases The biliary system originates from the hepatic diverticulum of
are associated with other congenital malformations22 (Table the foregut at 4 weeks’ gestation. This structure differentiates into
105-1) implies a global developmental abnormality that cranial and caudal components, which give rise to the intrahe-
may be under genetic control. Despite this, the occurrence patic and extrahepatic bile ducts, respectively. It is during this
of biliary atresia in twins is exceedingly rare, familial patterns period that the bile ducts undergo recanalization, eventually
have not been seen, and a clear genetic cause has not been leading to an intact biliary tree. Errors in the recanalization
found. process constituted early theories regarding the etiology of
Theories suggesting an acquired, infectious etiology are biliary atresia, but this is no longer believed to be correct.
supported by several findings. First, several viruses such as
reovirus and rotavirus have been proposed as possible infec-
tious agents responsible for the development of biliary atre- Pathology
sia.23,24 Animal models of perinatal viral infection produce ------------------------------------------------------------------------------------------------------------------------------------------------

biliary atresia,25 although consistent viral isolation has not The pathologist plays a central role in the diagnosis of
been possible in human cases.26 In addition, given the high biliary atresia and provides an important assessment of the struc-
prevalence of these viruses, it would be expected that the tures present at the fibrous biliary remnant. Inflammation is in-
incidence of biliary atresia would be higher if, in fact, viral variably seen in liver biopsies and in resected specimens.32–35
infection was causative. Second, it appears that many cases In the appropriate clinical scenario, a percutaneous liver
of biliary atresia are acquired rather than congenital.27 Up biopsy can reliably aid in the diagnosis of biliary atresia. The
to 60% of infants who are found to have biliary atresia have finding of bile ductular proliferation in the liver biopsy is con-
been documented as having pigmented stools sometime sidered diagnostic for biliary atresia.36,37 Associated findings
during the postnatal period.28 Third is the epidemiologic often include bile stasis, periportal inflammation, identification
finding of seasonal clustering of biliary atresia cases during of giant cells, and varying degrees of fibrosis.
the winter months in some studies.29 Histologic evaluation of the fibrous remnant can reveal
Other proposed factors associated with biliary atresia patency or partial patency of ductal structures or complete
include bile duct ischemia, abnormal bile acid metabolism, absence of these structures. This may depend on whether
pancreaticobiliary maljunction, and the effect of certain excision of the remnant occurred before or after inflam-
environmental toxins.30,31 matory obliteration of the extrahepatic ducts. Identifiable

A B C
FIGURE 105-1 Variants of ductal anatomy in biliary atresia. A, The most common pattern. Ducts are in continuity and obliterated from the porta hepatis
to the common bile duct. B, The gallbladder, cystic duct, and distal common bile duct remain patent, but the hepatic ducts and porta are fibrotic. A chol-
angiogram in this case would show duodenal opacification, but no contrast would enter the intrahepatic ducts. C, Fibrous gallbladder with poor-quality
tissue at the porta hepatis. (From Altman RP, Lilly JR, Greenfeld JR, et al: A multivariable risk factor analysis of the portoenterostomy [Kasai] procedure for
biliary atresia: Twenty-five years of experience from two centers. Ann Surg 1997;226:348, discussion 353. Courtesy Lippincott & Wilkins.)
 CHAPTER 105 THE JAUNDICED INFANT: BILIARY ATRESIA 1323

structures that can be found in the biliary remnant include

bile ducts, collecting ductules, and biliary glands.38 Of these,
the ductules are ultimately responsible for bile drainage
after the portoenterostomy procedure and, over time, these
can form stable bile conduits (Fig. 105-2).

Clinical Evaluation
------------------------------------------------------------------------------------------------------------------------------------------------

SIGNS AND SYMPTOMS

The combination of progressive jaundice, acholic stools, dark
urine, and firm hepatomegaly in an infant should raise the
suspicion for the presence of biliary atresia. Jaundice involves
direct hyperbilirubinemia rather than the indirect hyperbilir-
ubinemia seen in neonatal jaundice. In most cases, the perina-
tal course is unremarkable and meconium is described as
normal. In more than 50% of cases, initial stools are consid-
ered pigmented.28 Over time, stools take on a lighter color
and become acholic due to biliary obstruction (Fig. 105-3).
Eventually, signs of advanced liver disease such as palpable
hepatomegaly and splenomegaly, ascites, failure to thrive, FIGURE 105-3 Gross appearance of acholic stool from an infant later
and malnutrition become present. If not treated, biliary atresia confirmed to have biliary atresia.
is fatal within the first 2 years of life with one study reporting a
survival rate of less than 10% at 3 years.20
BLOOD TESTS
DIAGNOSIS
Standard serum “liver function tests” are uniformly obtained,
Children with persistent jaundice, especially those with an but specificity is lacking. In biliary atresia both the direct and
elevated conjugated bilirubin, should be evaluated promptly indirect bilirubin levels are elevated. The transaminases are
to exclude the diagnosis of biliary atresia. No single test can also mildly to moderately increased. Alkaline phosphatase
reliably be used to differentiate biliary atresia from other levels are often elevated in infants and children due to contri-
causes of jaundice in the infant. For this reason, a combination bution from bone remodeling. For this reason, the gamma-
of data obtained from a complete clinical assessment, blood glutamyl transpeptidase (GGTP) level can be used as a more
tests, imaging studies, and pathologic evaluation is often specific indicator of hepatobiliary disease. Unless decompen-
used to arrive at a provisional diagnosis in almost all cases. sated liver disease is present, tests of hepatic synthetic func-
The final diagnosis, however, is always confirmed at surgical tion such as the clotting cascade and serum albumin are
exploration. The following evaluation is recommended as a normal.
useful guide. In order to rule out conditions that can mimic biliary
atresia, screening for perinatal infections due to members of
the TORCH family (Toxoplasmosis, Other viruses, Rubella,
Cytomegalovirus, and Herpes Simplex Virus) should be
performed. In addition, screening for the presence of alpha-1
antitrypsin deficiency should occur.

ULTRASOUND
Abdominal ultrasound is an easily obtained, simple, and non-
invasive imaging study that can aid in the evaluation of infants
suspected of having biliary atresia.39,40 It should be obtained
in the fasting state to allow for filling of the gallbladder,
if patent, and the use of Doppler techniques can improve
the accuracy of the study.40 The echotexture of the liver, the
presence of ascites, the patency of the hepatic vasculature,
and the anatomy of the biliary structures can be assessed.
Although the abdominal ultrasound does not secure a diagno-
sis in most cases, on occasion it can streamline the subsequent
evaluation and management in specific situations. For exam-
ple, the presence of a hilar cystic structure in the clinical
FIGURE 105-2 Schematic showing biliary ductules draining into the setting of obstructive jaundice is sufficient information to pro-
Roux limb as postulated to occur after successful hepatic porto- ceed with surgical exploration with a presumptive diagnosis
enterostomy. (Courtesy M. Kasai, MD) of cystic biliary atresia or choledochal cyst with obstruction.
1324 PART VII ABDOMEN

In either scenario, relief of biliary obstruction should be assessment or, on rare occasions, at exploration when the
provided as soon as feasible. Alternatively, an ultrasound diagnosis cannot be confirmed preoperatively.
appearance of dilated proximal bile ducts suggests the pres-
ence of inspissated bile syndrome, which may be treated by
observation or, if needed, operative cholangiogram with or Other Tests
without biliary irrigation. ------------------------------------------------------------------------------------------------------------------------------------------------

In almost all infants with biliary atresia, however, the A combination of the previously described tests constitutes
ultrasound reveals that the gallbladder is either shrunken or an adequate evaluation for biliary atresia, and operative
normal appearing and the bile ducts are not easily delineated. therapy can be planned on the basis of these data. In some
In selected centers with extensive experience in the use of centers, however, other diagnostic modalities have been pro-
ultrasound in biliary atresia, an abnormal “cord” can be posed. These adjunctive tests may be performed in certain
appreciated in the area of the portal plate.41 circumstances but should not be considered part of the rou-
tine evaluation for biliary atresia.
Intubation of the duodenum via the nasoduodenal route with
aspiration or prolonged collection of duodenal fluid can exclude
HEPATOBILIARY SCINTIGRAPHY
biliary atresia if bile-stained fluid is obtained.45 Although simple,
Hepatobiliary scintigraphy relies on the use of isotopes of this test is invasive, subjective, and similar in concept to the
technetium 99m to assess excretion of bile from the liver into DISIDA scan, which is easily obtained in most centers.
the small intestine and therefore biliary patency. The term Endoscopic retrograde cholangiopancreatography (ERCP)
“HIDA” (hydroxy iminodiacetic acid) scan is often used, but is a technique that is widely applied in the evaluation of biliary
the technetium-labeled compound diisopropyl iminodiacetic diseases in adults and older children. With improvements in
acid (DISIDA) is more effective in the presence of significant endoscopic instrumentation, small side-viewing endoscopes
cholestasis and therefore more commonly used. The use- that can be used in infants are available. Some authors have
fulness of all hepatobiliary scintigraphy is diminished in the proposed ERCP as a useful adjunct in avoiding unnecessary
presence of severe jaundice, and this may cause errors in exploration for presumed biliary atresia.46 Because other
interpretation. less-invasive tests yield an accurate diagnosis in almost all
If time allows, all jaundiced infants undergoing hepatobili- cases, however, the practical use of ERCP is actually quite lim-
ary scintigraphy should be pretreated with phenobarbital ited and is not currently recommended. Magnetic resonance
(5 mg/kg/day) for 5 days before the study.42 Presence of iso- imaging (MRI) techniques such as magnetic resonance cho-
tope in the intestine immediately confirms patency of the bil- langiopancreatography (MRCP) have superior accuracy in
iary system, and the diagnosis of biliary atresia can be the diagnosis of biliary atresia.47 MRI and MRCP, however,
excluded. Excretion of isotope may be delayed, however, in are expensive, sometimes require sedation, and have not been
the presence of liver dysfunction. For this reason, a delayed routinely used in most centers.
assessment of isotope excretion at 24 hours is warranted.
When no isotope is seen in the intestine after 24 hours, biliary
obstruction is presumed and the diagnosis of biliary atresia
must be further pursued.
Treatment
------------------------------------------------------------------------------------------------------------------------------------------------

PREOPERATIVE CARE
Infants with presumed biliary atresia should be prepared for
Liver Biopsy exploration soon after diagnostic testing has been completed.
------------------------------------------------------------------------------------------------------------------------------------------------
Standard presurgical measures should be taken and, in gen-
A percutaneous liver biopsy can help differentiate biliary eral, children can be admitted on the day of surgery. A bowel
atresia from other cholestatic conditions with a high degree preparation is not required, but a dose of broad-spectrum an-
of reliability and should be considered the most accurate tibiotics should be administered before making the abdominal
nonsurgical diagnostic test.43,44 It is the most invasive of incision. Although most infants diagnosed with biliary atresia
the diagnostic modalities but can be performed safely by an will have normal coagulation studies, poor absorption of the
experienced pediatric hepatologist, or, if needed, by the sur- fat-soluble vitamin K can theoretically render them function-
geon. Typically, examination of several portal tracts by a ally vitamin K deficient. If possible, preoperative oral sup-
well-trained pathologist will reveal important findings that plementation with fat-soluble vitamins (A, D, E, and K) or
can confirm or exclude biliary atresia. The presence of varying an intramuscular injection of vitamin K (1 mg) should be
degrees of inflammation with ductular proliferation is consid- considered.
ered compatible with the diagnosis of biliary atresia because
these findings are not seen in other nonobstructive cholestatic
SURGICAL TECHNIQUE
syndromes. Further findings of bile stasis with plugging and
giant cell transformation further support the diagnosis of bil- The Roux-en-Y hepatic portoenterostomy procedure (Kasai
iary atresia. On the basis of the appearance of the liver biopsy, Procedure) is the standard initial operation for treatment of
bile duct paucity syndromes can be readily differentiated from infants with biliary atresia. The operation involves excision
biliary atresia. In contrast, however, it can be difficult to of the entire extrahepatic biliary tree with transection of the
differentiate between parenteral nutrition-associated cholesta- fibrous portal plate near the hilum of the liver. Bilioenteric
sis and biliary atresia on the basis of liver biopsy alone. This continuity is then reestablished with a Roux-en-Y limb.13
distinction should be made on the basis of the overall clinical The ultimate goal of the procedure is to allow drainage of bile

from the liver into the Roux limb via microscopic ductules
in the portal plate. The conduct of the operation is described
as follows.
The infant should be placed supine and on an operating
table that will permit operative cholangiography, if deemed
necessary. The exploration begins via a right upper abdominal
incision. The left upper quadrant is examined, first searching
for the spleen. Absence of the spleen or the finding of poly-
splenia can alert the surgeon to the presence of important
associated anomalies such as malrotation, preduodenal portal
vein, and interrupted inferior vena cava with azygous contin-
uation. During evaluation of the left upper quadrant, adequate
position of the tip of the nasogastric tube is also confirmed and
the tube is secured by the anesthesiologist.
Next, the liver, biliary structures, and porta are inspected.
The liver in biliary atresia can appear nodular and fibrotic with
a greenish color. This finding is not common in neonatal
hepatitis or bile duct paucity syndromes, where the liver is
smooth and dark brown in color. Many infants with biliary
atresia have a contracted, fibrotic gallbladder. If a rudimentary
fibrous gallbladder is noted at initial exploration and if it
clearly has no lumen, then the diagnosis of biliary atresia
has been confirmed and the operation can proceed with
dissection of the portal plate. If the gallbladder is normal-
appearing or if it is felt to have a lumen, then additional intrao-
perative diagnostic maneuvers are warranted before dissecting
the portal plate. In this situation, a purse-string suture can be
placed in the fundus of the gallbladder and the fluid within the
lumen of the gallbladder aspirated with an angiocath. If clear
fluid (white bile) returns, then our approach has been to
proceed with portal plate dissection without a cholangiogram.
If, however, the aspirated fluid is darker in color or if there
is any ongoing question regarding the diagnosis, then a chol-
angiogram should follow.
Although simple in concept, the cholangiogram can be
difficult to perform and interpret successfully during this
exploration. The following steps can be used to maximize
the diagnostic yield of the intraoperative cholangiogram. First,
a secure purse-string suture should be placed. A second
purse-string suture can also be placed to prevent contrast leak-
age. Either an angiocath or a laparoscopic cholangiogram
catheter can be placed into the lumen of the gallbladder before
tying the purse-string suture(s). Diatrizoic acid (Hypaque or
Gastrografin) is diluted 1:1 with normal saline and injected
as the contrast agent via the cholangiogram catheter to assess
for patency or obstruction of the biliary tree. Real-time, live
fluoroscopy facilitates rapid intraoperative interpretation of
the study. If contrast flows freely into the duodenum and into
intrahepatic bile ducts, then patency of the biliary tree has
been established and the diagnosis of biliary atresia excluded
(Fig. 105-4). In this scenario, a wedge liver biopsy should
be performed, the cholangiogram catheter removed, the
cholecystotomy closed, and the operation concluded.
On occasion, contrast will flow freely into the gallbladder
and down the distal common bile duct into the duodenum but
not proximally into the intrahepatic bile ducts. This finding
should be confirmed by occluding the distal common bile
duct with an atraumatic “bulldog” clamp and reinjecting the
cholangiogram catheter. This maneuver may encourage
preferential filling of any patent proximal ducts that may have
initially failed to opacify with contrast material when the distal
resistance to flow was low. It is important to inject contrast
CHAPTER 105 THE JAUNDICED INFANT: BILIARY ATRESIA 1325
FIGURE 105-4 Cholangiogram obtained during abdominal exploration
for presumed biliary atresia. The gallbladder (GB) appeared normal, and an
intraoperative cholangiogram revealed patency of the intrahepatic ducts
to the duodenum (Duo). A liver biopsy was performed, and the procedure
was concluded.
gently because a speedy injection under pressure is likely to
cause leakage of contrast around the purse-string suture,
resulting in an obscured and confusing cholangiogram. Failure
to delineate patent intrahepatic and extrahepatic biliary struc-
tures mandates that the surgeon proceed with portal dissection.
Regardless of the presence of a patent gallbladder or distal
common bile duct, a direct Roux-en-Y hepatic portoenterost-
omy affords outcomes that are superior to other forms of
reconstruction such as the portocholecystostomy.48 Some
surgeons gain exposure by dividing both triangular ligaments,
thereby exteriorizing almost the entire liver.49,50 This step is
not necessary and likely disrupts lymphatics that may be re-
sponsible for the development of ascites. We have found that
upward retraction of the liver to expose the porta affords
excellent visualization of the critical structures and can be
accomplished by assigning an assistant for retraction in the
right upper surgical field or by carefully placing a fixed metal
retractor with upward traction.
The peritoneum overlying the hepatoduodenal ligament is
opened to allow identification of the structures in this area.
The fibrous remnant of the distal common bile duct is often
present here. It can be identified in the anterolateral aspect
of the hepatoduodenal ligament, isolated and divided. Care
should be taken to avoid injury to aberrant hepatic arteries
that may be nearby. With traction on the cut end of the oblit-
erated distal common bile duct, the fibrous biliary remnant
can be dissected toward the porta. It is often necessary to
dissect the right and left hepatic arteries away from the field
in order to preserve them. During this dissection, the gallblad-
der remnant is also dissected away from the liver in continuity
with the rest of the extrahepatic biliary remnant. As dissection
continues proximally, the biliary remnant develops into a cone
of fibrotic tissue that is located at the bifurcation of the main
portal vein into its left and right branches. This constitutes the
most important landmark during the dissection of the portal
plate and should be the goal of every dissection.51,52 In this
1326 PART VII ABDOMEN

location it may be necessary to control several small portal

vein branches to avoid inadvertent injury with subsequent
bleeding. The absence of identifiable biliary remnant tissue
or the finding of poor quality tissue at the expected location
of the portal plate sometimes confuses the dissection. As
stated earlier, the bifurcation of the portal vein should be
used as a landmark and is especially important in such cases
where the biliary remnants and portal plate are difficult to
identify clearly.
Once the fibrous cone and portal plate region have been
identified, the fibrous cone is placed on gentle traction and
transected with sharp scissors or a knife. It is not beneficial
to cut deeply toward the liver parenchyma because this may
result in more scar formation and inhibition of bile drainage.
Bleeding at the transected portal plate is controlled with pres-
sure and placement of a surgical sponge in the area. Use of
cautery on the portal plate is discouraged and should be min-
imized because this structure contains the fine ductules
needed for success of the procedure. The resulting surgical
specimen should be marked and submitted for routine path-
ologic evaluation. Careful measurement of the diameter of any
biliary ductules by an experienced pathologist should be
requested because this may provide important prognostic in-
formation. Although advocated by some,53 we have not found FIGURE 105-5 Artist’s rendition of the completed hepatic portoenter-
frozen section to be helpful in guiding the level of transection. ostomy procedure.
With a completed portal plate dissection, the operation
shifts to the construction of the Roux limb. The proximal je-
junum is identified and transected about 10 centimeters distal
to the ligament of Treitz. The distal end, destined for the right valves28,55–57 have been subsequently abandoned either due
upper quadrant, is oversewn and the Roux limb is measured to to associated complications or lack of effectiveness.58–60
40 to 50 centimeters.54 At this location, an end-to-side jejuno- Similarly, use of the appendix as a conduit between the liver
jejunostomy is created with interrupted absorbable sutures. and the small intestine has been proposed but its use has
The oversewn end of the Roux limb is carefully brought been limited with some reports suggesting an inferior surgical
into the right upper quadrant via a small defect created in outcome.61
the avascular portion of the transverse mesocolon. With the widespread application of minimally invasive
The side of the Roux limb is opened, and the portoenter- techniques, even to the most complex operations, the laparo-
ostomy is created. The posterior suture line is placed first scopic Kasai procedure has been described and used at
using 6-0 absorbable suture with knots tied inside the lumen. several centers worldwide.62,63 Most reports, however, in-
Placement of these sutures must be exact with care taken to volve single cases or small series of carefully selected patients.
avoid injuring or impinging on the portal tissue. Once this A recent prospective trial confirmed the feasibility of the
posterior row is complete, the anterior sutures can be placed operation but revealed a significantly poorer outcome at
using similar care and precision. When complete, the entire 6 and 24 months, causing the trial to be stopped.64 Outcomes
surface of the portal plate must be contained within the jejunal of other series have been variable,65 and therefore the lapa-
lumen of the Roux limb. In this way, any bile drainage via roscopic Kasai procedure currently lacks support among
biliary ductules at the portal plate will be contained within general pediatric surgeons including those who specialize in
the Roux limb and proceed distally. minimally invasive pediatric surgery. Laparoscopic techniques
Before closure, it is advisable to close the mesenteric may be used to perform the exploration and cholangiogram.
defect created by construction of the Roux limb and to anchor Some investigators have proposed that primary liver
the Roux limb to the edges of the defect in the transverse transplantation be considered the initial treatment for infants
mesocolon. These maneuvers are aimed at preventing internal with biliary atresia, citing deleterious effects of the Kasai
herniation and at keeping the Roux limb in the right upper procedure on subsequent liver transplantation, if needed.
quadrant and without tension. A small closed suction drain This approach, however, would subject a percentage of chil-
is placed near the portoenteric anastomosis, and a wedge liver dren to the dangers of transplantation and its associated
biopsy is routinely performed but likely not absolutely short- and long-term complications who would have been
necessary. The abdomen is then closed using standard tech- cured by the Kasai procedure. For this reason, primary liver
niques. The completed operation is shown in (Fig. 105-5). transplantation has been reserved for selected cases such as
delayed diagnosis with severe liver failure where a Kasai
procedure would be risky and have a high failure rate.66–68
SURGICAL CONTROVERSIES
For all other children with compensated liver function and a
Much of the controversy surrounding surgery for biliary timely diagnosis, the Kasai procedure and liver transplant
atresia has subsided. Previously reported techniques with are considered by most to be sequential, complementary
stomas to exteriorize the Roux Limb28 or antireflux procedures.69

POSTOPERATIVE CARE
Drainage of gastric secretions with a nasogastric tube should
continue for the first 48 hours, and the tube can usually be re-
moved at this point. By the second or third postoperative day,
infants often have already passed gas and stool and an oral diet
can be started. Intravenous antibiotics are administered until
the child begins feeding, and this is subsequently converted
to long-term oral antibiotics that are continued at discharge.
The closed suction drain is removed before discharge, typically
on the fifth postoperative day. Antibiotics, a choleretic agent,
fat-soluble vitamin supplements, and an oral steroid taper
have been the routine for our group (Table 105-2), although
the support for the use of steroids is not universal as discussed
in the next section.
STEROID CONTROVERSY
Steroids are known to have antiinflammatory and immuno-
suppressive effects. Their use in patients with biliary atresia
with its significant inflammatory component is therefore
logical. In addition, steroids have been postulated to have a
choleretic effect, making them even more attractive in this pa-
tient population. Despite this, definitive evidence for the ben-
eficial effects of adjunctive steroids in biliary atresia has been
elusive. First, attempts to prove the choleretic action of ste-
roids were not successful.70 Second, many studies comment-
ing on the efficacy of steroids are retrospective and contain
confounding factors that make it impossible to comment on
the isolated effects of steroid treatment. Finally, studies often
appear underpowered to allow definitive conclusions.71,72
For these reasons, it is not surprising that one study found that
perioperative steroids were used in less than 50% of patients
in the United States between 2003 and 2008.73
The paucity of data regarding the utility of steroids may be
changing, however. A recent randomized, double-blind,
placebo-controlled trial of corticosteroids after hepatic por-
toenterostomy revealed that steroids resulted in accelerated
clearance of jaundice but did not result in improved survival
of the native liver.74 A current study by the biliary atresia re-
search consortium in the United States is also attempting to
address this question systematically. The results are pending.
Adverse sequelae of steroid use are not commonly
reported. Fluid retention and appetite suppression are minor
side effects, but major steroid-related complications such as
infection and wound breakdown have not been problematic.
Given that the evidence supporting steroid use is currently
stronger than the evidence against it, our group has incor-
porated a protocol for postoperative steroid administration
(see Table 105-2).
TABLE 105-2
Postoperative Medical Regimen After Hepatic
Portoenterostomy
Ursodiol (Actigall)—10-15 mg/kg/dose BID day
Trimethoprim-Sulfamethoxazole—2.5 mg/kg/day based on
Trimethoprim component dosing
Vitamins ADEK drops—1 mL/day
Prednisone—2 mg/kg/day and taper over 6 weeks
CHAPTER 105 THE JAUNDICED INFANT: BILIARY ATRESIA 1327
Outcomes
------------------------------------------------------------------------------------------------------------------------------------------------
Over the past 50 years, the hepatic portoenterostomy proce-
dure has dramatically improved the outlook for infants diag-
nosed with biliary atresia. Before the introduction of the
hepatic portoenterostomy, few surgical options were available
for treating infants with biliary atresia. A review of 89 patients
who were explored for biliary atresia before the advent of
formal surgical correction revealed a “cure” rate of 1.1%
(1 of 89) and a known mortality of 94% (84 of 89).75 The cur-
rent 30-day mortality is low, ranging between 0% and 5%,
reflecting the safety of the hepatic portoenterostomy.61,69
An important initial observation is the color of the stool
after hepatic portoenterostomy. Pigmented stools either imme-
diately or within the first 10 to 14 days after surgery suggest
successful bile flow from the liver to the intestinal tract. This
occurs in two thirds of patients, on average.59,61,69 Of these
patients who have documented bile flow, half will continue
to drain bile well, with efficient clearance of jaundice and
normal development. This is considered a “cure,” and liver
transplantation is not necessary.
In the remaining patients who initially had good bile drain-
age, ongoing inflammation and scarring of the liver will lead to
progressive liver failure. Jaundice, which may have initially
cleared completely or partially, returns. Signs of portal hyper-
tension and failure to thrive eventually appear, and liver
transplantation is therefore required. In this patient popula-
tion that had initial successful palliation, liver transplantation
occurs at a mean age of 5.4 years.69
The remaining patients (15% to 30%) who continue to
have acholic stools after hepatic portoenterostomy never expe-
rience clinically relevant bile drainage. Jaundice continues to
worsen, and the liver fails within months. In these infants an
evaluation for liver transplantation, either cadaveric or living
related, should commence before the arrival of true end-stage
liver disease.
The 10-year survival after hepatic portoenterostomy has
improved over time. Initial improvements were likely due to
refinement of the portoenterostomy procedure, while more
recent gains have been made in the technique of liver trans-
plantation and in immunosuppressive strategies and medi-
cations. A biliary atresia registry report from the United
States in 1990 found 5- and 10-year survival rates of 48%
and 30%, respectively,20 while a report from our center in
1997 found a 20-year survival rate of 49%.69 A recent multi-
center study in the United States reported a 2-year overall
survival rate of 91% with a 56% native liver survival and
40% liver transplant rate.76 Studies from other countries
including Japan quote similar statistics.21,22,56,61,77
Long-term patient outcomes beyond liver function are
available. Abnormalities in menstruation and pubertal dis-
orders occur and are closely related to liver function. These
findings suggest that hepatic function is deteriorating and
transplantation will eventually be necessary. When transplan-
tation occurs, menstrual abnormalities have been shown to
improve.78 Pregnancy can be challenging for patients who
have been treated with hepatic portoenterostomy for biliary
atresia. Progression of liver disease with development of portal
hypertension has been documented during gestation.78 New-
borns can be small but are generally healthy. The postpar-
tum period has also been documented as complicated.79
1328 PART VII ABDOMEN
These data suggest that great care should be taken in counsel-
ing former biliary atresia patients regarding pregnancy, and
careful monitoring is indicated when pregnancy is being
considered.
Psychosocial outcomes have been reported in long-term
survivors of the hepatic portoenterostomy.78 This study of
44 patients with long-term follow-up revealed that almost
all patients (93%) are functioning well in society with high
levels of employment and education. Psychologic morbidity
was present, however, in 18% of this population.
OUTCOME VARIABLES
A variety of factors appear to play a role in the outcome after
hepatic portoenterostomy for biliary atresia. Age at operation,
operative technique, severity of liver disease, gross and micro-
scopic aspects of the biliary tree and portal plate, and the
presence of comorbid conditions are a few of the many
variables that can affect outcome.
Strong evidence suggests that age at operation is an impor-
tant factor in outcome, and this is supported by multiple
reports.69,80–84 The exact cutoff, however, where outcome
from surgery becomes predictably poor, is unknown. This
may be due to the existence of several distinct pathologic
forms of biliary atresia. In fact, recent studies suggest that
age at surgery may be a factor only in certain subsets of
children such as those with cystic biliary atresia or those with
biliary atresia splenic malformation as compared with those
who present with isolated biliary atresia.85 In our series, in-
fants operated on at 0 to 49 days had equivalent outcomes
when compared with those who underwent operation at 50
to 70 days of age. Those explored at 71 days or older had a
statistically higher rate of failure when compared with both
of the younger groups.69 In the Japanese experience, the better
outcomes were seen until the age of 90 days with subsequent
worsening outcome after 90 days.22 Therefore it is the goal at
most centers to accurately obtain a provisional diagnosis of
biliary atresia and successfully operate before 70 to 90 days
of life. It is important to note, however, that hepatic portoen-
terostomy is not contraindicated after 90 days of age. A study
of older children (100 days) revealed acceptable survival
rates, supporting the use of the procedure, if possible, even
in older infants.86
The gross appearance of the biliary tree appears to influ-
ence outcome. Patients with a patent gallbladder had better
outcomes when compared with those with complete fibrosis
or complete absence of the fibrous cone of the portal plate.16
The size of ductules that are present at the level of the portal
plate seems to be related to prognosis. It is logical that larger
ductules allow for more efficient drainage of bile and therefore
would result in an overall better outcome. Our series found
that bile drainage was universal in patients with ductules
greater than 150 micrometers, occurred in 86% of those with
ductules less than 150 micrometers, but was only seen in 12%
of those with no identifiable ductules.69 Other studies have
supported these findings.34,38,87–90
The experience of the surgeon and the center has been
suggested as a factor in outcome of hepatic portoenterostomy.
Our study revealed inferior outcomes in patients who had
surgery in the 1970s, possibly reflecting a learning curve
regarding surgery and the care of these patients.69 Recent
studies in the United Kingdom91 and in France92 also suggest
that experienced centers and centralization of care for pa-
tients with biliary atresia produce superior outcomes. These
national health policies, however, have not been implemented
in many countries and long-term outcome differences are
pending.
Complications
------------------------------------------------------------------------------------------------------------------------------------------------
NUTRITIONAL COMPLICATIONS
Nutrition with a well-functioning hepatic portoenterostomy
follows a normal course. Specialty formulas and vitamin
supplements, if used, can often be discontinued after several
months without harm. In the face of ongoing or progressive
liver dysfunction, however, weight gain can be affected and
metabolic complications may occur.93 Nutritional parameters,
vitamin levels, and growth should be monitored over the long
term. In fact, some authors suggest that growth failure is an
important marker for liver disease and an indication for liver
transplantation in biliary atresia.94
CHOLANGITIS
Cholangitis is the most common complication of the
hepatic portoenterostomy and occurs in 33% to 60% of
patients.54,77,95 Cholangitis occurs most frequently during
the first several years after initial surgery. Because cholangitis
causes liver injury and promotes accelerated development
of cirrhosis, prevention and active treatment are important
adjuncts during the postoperative period.
The exact cause of cholangitis is unknown, although reflux of
intestinal contents via the Roux limb, portal venous infection,
impaired lymphatic drainage at the porta, and bacterial trans-
location have been proposed. Intrahepatic biliary stasis due to
the fibrotic ductal obstruction is also felt to play a role in the
pathogenesis of cholangitis. The offending organisms have
been identified as enteric gram-negative and anaerobic bacteria,
further supporting an enteric source for the condition.96
The clinical signs of cholangitis include fever, diminished bile
flow, jaundice, and sometimes abdominal pain. Blood testing will
reveal leukocytosis, elevated serum bilirubin, elevated hepatic
enzymes (transaminases, alkaline phosphatase, and GGTP),
and inflammatory parameters such as C-reactive protein.
Treatment of cholangitis is aimed at the presumed infection
and the ongoing inflammation. The mainstay of treatment
includes intravenous fluids, broad-spectrum intravenous
antibiotics including coverage of anaerobes, and choleretic
agents. Steroids are indicated for their ability to decrease
inflammation and edema of the periductular areas.97
Prevention of cholangitis is an important goal. An ade-
quately constructed Roux limb is a surgical maneuver that aids
in prevention of cholangitis. Similarly, postoperative use of
oral antibiotics is generally felt to be efficacious as a preventa-
tive measure. Studies have shown that use of trimethoprim-
sulfamethoxazole or neomycin appeared to delay the onset
and reduce the number of episodes of cholangitis compared
with historic controls.98 The use of steroid pulses during
the postoperative period has also decreased the frequency of
this complication.99
A surgical approach to revise a failing hepatic portoenter-
ostomy with recurrent jaundice or unrelenting cholangitis

has been attempted.48,100 Hasegawa reported the results of
attempted revision of the hepatic portoenterostomy in
25 patients who did not drain effectively after the initial
operation.101 Only five patients had improvement in jaundice,
a similar result to a series from the United States in which two
of seven children were helped by a re-do portoenterostomy.48
A similarly low success rate has been shown for surgical
revision of the hepatic portoenterostomy in children who
develop severe, intractable cholangitis.22,102 Given these dis-
couraging outcomes, the possibility that repeated operations
can negatively affect the eventual success of liver transplanta-
tion, and in an era where liver transplantation has improved
dramatically, operative approaches to rescue a failed or failing
portoenterostomy are not recommended.
PORTAL HYPERTENSION
Portal hypertension occurs in 34% to 76% of children and
young adults after hepatic portoenterostomy and can occur
despite a seemingly excellent result from the point of view
of bile drainage.103,104 Ongoing intrahepatic inflammation,
often manifested as recurrent cholangitis, is likely responsible
even when jaundice clears.
The most common finding in patients with portal hyper-
tension is ascites, occurring in more than 60%. Esophageal
varices, the most feared sequela of elevated portal pressure,
occur in nearly half of patients who are followed for more than
3 years. Surveillance endoscopy therefore is indicated,
especially in cases where progressive liver dysfunction is
suspected. No clear recommendations are available for the
frequency of surveillance endoscopy; however, yearly or semi-
annual examinations have been recommended. When varices
are identified endoscopically, prophylactic sclerotherapy has
shown benefit in randomized trials.105
Bleeding from esophageal varices can be massive and life
threatening. Treatment consists of endoscopic sclerotherapy,
often requiring multiple sessions.77 Other interventions
that can aid in treatment include use of octreotide or beta-
adrenergic blockers or variceal ligation.106–108 Although con-
cerning, variceal bleeding should not prompt urgent
consideration of liver transplantation and does not signify
the arrival of end-stage liver disease.77
Portal hypertension–associated hypersplenism occurs in
16% to 35% of patients after portoenterostomy, and the asso-
ciated thrombocytopenia can complicate episodes of gastroin-
testinal bleeding from varices or other causes.109 Although
splenectomy, portosystemic shunting, and splenic emboliza-
tion have been described,110 evaluation for liver transplan-
tation is the most prudent option in patients who develop
this serious secondary complication of portal hypertension.
LIVER TRANSPLANTATION
Liver transplantation is covered in a separate section of this
textbook. Nevertheless, it is mentioned here because liver
failure due to biliary atresia represents the most common
indication for liver transplantation in the pediatric age group.
The general indications for liver transplantation in biliary
atresia include (1) failure of initial portoenterostomy with
no bile drainage and progressive liver disease; (2) episodic
CHAPTER 105 THE JAUNDICED INFANT: BILIARY ATRESIA 1329
or inefficient bile drainage with slow deterioration of liver
function and development of growth failure; and (3) develop-
ment of one or more complications of chronic liver disease
such as cholangitis or portal hypertension that cannot be easily
managed despite an apparently functional portoenterostomy.
Thankfully, death on the transplant waiting list is uncom-
mon. Improvements in allocation of organs using the pediatric
end-stage liver disease system, based on several markers of
severity of disease, result in transplantation in the most se-
verely affected first. In addition, two techniques have resulted
in an increase in the number of available organs. First was the
technique of splitting or rescuing the size of organs for use in
children,111 and the second is the development of living-
related liver transplantation.112 These techniques have been
applied in liver transplantation programs worldwide with
excellent success.113–115
With continual improvement in the outcomes of pediatric
liver transplantation, some have questioned whether primary
liver transplantation without an initial attempt at hepatic
portoenterostomy is indicated.68,116,117 At this juncture,
however, most believe that a well-executed hepatic porto-
enterostomy is the best initial surgical treatment for biliary
atresia. This opinion is based on several important observa-
tions: (1) Nearly half of infants who undergo hepatic portoen-
terostomy obtain bile drainage and maintain either excellent
or adequate liver function after surgery. Even in those who
eventually progress to end-stage liver disease, transplantation
is delayed by having undergone hepatic portoenterostomy
first; (2) The improved but still insufficient supply of donor
organs would make primary liver transplantation logistically
challenging; and (3) A subset of children who would have
been “cured” by the hepatic portoenterostomy would be un-
necessarily exposed to the morbidity (surgical, infectious,
and oncologic) and mortality of liver transplantation. At this
point, therefore, liver transplantation remains the most impor-
tant rescue therapy for children with biliary atresia after
portoenterostomy with 5-year survival rates expected to
exceed 90%.118,119
Acknowledgments
The author would like to acknowledge R. Peter Altman, MD, for his priceless
guidance and education on the treatment of infants with biliary atresia.
The complete reference list is available online at www.
expertconsult.com.
SUGGESTED READING
Altman RP, Lilly JR, Greenfeld J, et al. A multivariable risk factor analysis of the
portoenterostomy (Kasai) procedure for biliary atresia: Twenty-five years of
experience from two centers. Ann Surg 1997;226:348.
Davenport M, Caponcelli E, Livesey E, et al. Surgical outcome in biliary atresia:
Etiology affects the influence of age at surgery. Ann Surg 2008;247:694.
Davenport M, De Ville de Goyet J, Stringer MD, et al. Seamless management of
biliary atresia in England and Wales (1999-2002). Lancet 2004;363:1354.
DeRusso PA, Ye W, Shepherd R, et al. Growth failure and outcomes in infants
with biliary atresia: A report from the Biliary Atresia Research Consortium.
Hepatology 2007;46:1632.
Kasai M, Kimura S, Asakura Y, et al. Surgical treatment of biliary atresia.
J Pediatr Surg 1968;3:665.
Kuroda T, Saeki M, Nakano M, Morikawa N. Biliary atresia, the next genera-
tion: A review of liver function, social activity, and sexual development in
the late postoperative period. J Pediatr Surg 2002;37:1709.
1330 PART VII ABDOMEN

Majd M, Reba RC, Altman RP. Hepatobiliary scintigraphy with
99mTc-PIPIDA in the evaluation of neonatal jaundice. Pediatrics
1981;67:140.
Nio M, Ohi R, Miyano T, et al. Five- and 10-year survival rates after surgery for
biliary atresia: A report from the Japanese Biliary Atresia Registry. J Pediatr
Surg 2003;38:997.
Schneider BL, Brown MB, Haber B, et al. A multicenter study of the
outcome of biliary atresia in the United States, 1997 to 2000. J Pediatr
2006;148:467.
Serinet MO, Broué P, Jacquemin E, et al. Management of patients with biliary
atresia in France: Results of a decentralized policy 1986-2002. Hepatology
2006;44:75.
 decade of life. Roughly 20% remain undiagnosed until later
in childhood or adulthood, and the remaining 20% to 25%
of cases are diagnosed prenatally. Prenatal diagnosis of chole-
dochal cysts is increasing in frequency,2 as is the number of
cases diagnosed in adulthood in the United States and Europe,
perhaps due to higher index of suspicion or improved imaging
techniques.
The etiology of choledochal cyst remains unknown but is
commonly accepted to be congenital in nature. Distal obstruc-
tion, weakness of the duct wall, or a combination of the two
are the predominant hypotheses.4–6 The etiologic basis for
choledochal cyst is likely multifactorial. In 1852 Douglas
reported that there is potential congenital weakness of the
common bile duct in the common fusiform type.5 In 1936
Yotsuyanagi reported that the defect is due to failed formation
in the early embryologic development of the biliary system
from an overproduction of epithelial cells, leading to a dilated
common bile duct.7 This hypothesis is now largely regarded as
unlikely, however.
In 1916 Kozumi and Kodama recognized an anomalous
junction between the bile and pancreatic ducts during an au-
topsy case with choledochal cyst.8 In 1969 Babbitt described
the “long common channel” theory, also known as pancreati-
CHAPTER 106 cobiliary maljunction (PBM), which is commonly seen in asso-
ciation with choledochal cyst.9,10 PBM is a rare congenital
anomaly described as a proximal insertion of the pancreatic
duct into the common bile duct as a result of incomplete
Choledochal Cyst migration of the choledochopancreatic junction into the duo-
denal wall, creating a “long common channel.” In 1984 Todani
and colleagues were able to show PBM through analysis of
Kelly D. Gonzales and Hanmin Lee endoscopic retrograde cholangiopancreatography (ERCP).11
In fetal development, PBM creates a nidus for reflux of pancre-
atic enzymes into the common duct that causes damage to the
ductal wall and leads to cyst formation. Due to the reflux, PBM
is also commonly associated with carcinoma of the bile duct
Choledochal cyst is a congenital anomaly involving cystic di- and gallbladder. Distal obstruction at the level of the duo-
lation of various ducts of the biliary tree. They are relatively denum is an additional factor leading to damage of the ductal
rare in the Western world but more common in Asia. Several wall and causing formation of a saccular dilation.11–15 Inves-
hypotheses exist as to the etiology of choledochal cysts includ- tigators have proposed various mechanisms related to distal
ing pancreaticobiliary maljunction. Three patterns of pre- ductal stenosis such as an abnormal ductal insertion into
sentation have emerged since Alonso-Lei and colleagues the duodenum, distal common bile duct stenosis, persistent
published their landmark paper describing their classification epithelial web, valve dysfunction, or neuromuscular irregular-
of ductal dilation. The first is a cystic mass in the abdomen ity of the sphincter.13 PBM has a reported 3% incidence.16
identified prenatally, the second is jaundice presenting in in- PBM is commonly associated with choledochal cysts and
fancy, and the third is ascending cholangitis, obstructive carcinoma of the biliary duct and gallbladder. The majority
jaundice, or pancreatitis presenting later in childhood or in of choledochal cysts are associated with PBM; however,
adulthood. Regardless of the time of presentation, surgical PBM can be seen without an associated choledochal cyst in
intervention is indicated to mitigate potential damage to the 20% to 30% of cases. Carcinoma of the gallbladder and, less
liver, prevent jaundice and pancreatitis, and prevent cancer. frequently, the bile duct remains an association with PBM even
without ductal dilation.16
Other types of choledochal cysts, specifically choledocho-
Epidemiology and Etiology cele and Caroli disease, have different proposed etiologies.
------------------------------------------------------------------------------------------------------------------------------------------------
Choledochocele has two main variations as described by
The incidence of choledochal cyst ranges from 1 in 100,000 to Manning and colleagues.17 In the most common variation,
150,000 live births in Western populations with the incidence the common bile duct and pancreatic duct enter the duo-
in the United States reportedly as high as 1 in 13,500. Chole- denum separately. In the second variation, a common bile duct
dochal cysts are significantly more common in Asia, with rates diverticulum arises at the level of the ampulla of Vater with the
as high as 1 in 1000 in Japan. There is a well-documented pancreatic duct entering the distal common bile duct in its
female dominance (3 to 4:1) that contributes to the belief that normal anatomic variation. The formation of the diverticulum
choledochal cyst is sex-linked.1–3 Familial cases have been is thought to be due to either an obstruction of the ampulla
recognized, but no genetic basis has been evident. The major- and a low insertion or a congenital duodenal duplication.
ity (60%) of choledochal cysts are diagnosed in the first Caroli disease is believed to arise from an arrest in the twelfth
1331
1332 PART VII ABDOMEN

week of gestation in the ductal plate at the level of the large pancreaticobiliary system or termination of the junction
ducts of the hilum. This is distinct from Caroli syndrome, in migration as once thought.24
which there is arrest of the small ductules of the periphery.18,19
Caroli disease has been associated with polycystic kidney dis-
ease.1,20–22 Overall, the etiologic background of choledochal Anatomic Classification
cyst is congenital in nature, resulting from reflux of pancreatic ------------------------------------------------------------------------------------------------------------------------------------------------

enzymes causing damage to the bile duct wall and complete or The original anatomic classification of choledochal cysts
partial distal obstruction of the bile duct. included types I, II, and III.4 After review of cholangiograms,
Todani and colleagues broadened the classification into five
types with some subtypes (Fig. 106-1).25,26 Type I is by far
Embryology the most common, accounting for 90% to 95% of cases,4
------------------------------------------------------------------------------------------------------------------------------------------------
and constitutes the cystic/saccular or fusiform dilation of
In normal development the fourth week of gestation marks the the common bile duct.27 Type I is divided into three subtypes
formation of a hepatic diverticulum extending from the ven- (types Ia, Ib, and Ic). Type Ia consists of cystic dilation of the
tral aspect of the foregut. Thus the biliary structures are endo- entire common bile duct. Type Ib is cystic dilation of a segment
dermal in nature. The diverticulum progresses to cranial and of the common bile duct, and type Ic is fusiform dilation of the
caudal buds. The liver and extrahepatic biliary tree form from common bile duct. Type II is a diverticulum of the common
the cranial bud and the caudal bud becomes the superior and bile duct with no dilation of the common bile, extrahepatic,
inferior buds. The gallbladder and cystic duct are derived from or intrahepatic ducts. Type III, also referred to as a choledocho-
the superior bud, and the inferior bud gives rise to the right cele (Fig. 106-2), usually has a normal common bile duct and
and left ventral pancreas. There is some controversy regarding main pancreatic duct with cystic dilation of the distal common
whether there is a solid phase to the biliary tree early in ges- bile duct that is either intraduodenal or intrapancreatic in
tation and eventual recanalization or whether the biliary tree location. The ducts may either enter the choledochocele sep-
has a continuous lumen.23 At the sixth week, the ventral pan- arately or in union at the wall of the duodenum, but are usu-
creatic bud and common bile duct rotate around the duode- ally stenotic at their openings due to chronic inflammation.
num clockwise by 180 degrees. The common bile duct at this Type IV is composed of multiple cysts located intrahepatically,
point enters the duodenum at the left posterior surface. In the extrahepatically, or in both locations. Type V comprises single
seventh week the main pancreatic duct (Wirsung duct) and or multiple intrahepatic cysts without extrahepatic duct
common bile duct junction ends in the developing duodenum dilation. Type V cysts in conjunction with hepatic fibrosis
as closed cavities through elongation to form the ampulla of are commonly referred to as Caroli disease.
Vater. The junction retracts in the eighth week of gestation
to reside in the submucosa of the duodenal wall. A concentric
ring of mesenchyme forms around the junction of the pancre- Pathology
atic and biliary ducts, beginning the formation of the sphincter ------------------------------------------------------------------------------------------------------------------------------------------------

of Oddi. In the twelfth week of gestation, the main pancreatic The gross pathology of choledochal cysts is largely described
and common bile ducts are obliquely arranged in the duo- in the anatomic classification section earlier. As described,
denum. PBM is believed to arise from a misarrangement of choledochal cysts can occur in the intrahepatic ducts, the
the pancreatic and common bile duct and not due to the common bile duct, intraduodenally, or in a combination of

Type Ia Type Ib Type Ic

Type II Type III Type IV Type V

FIGURE 106-1 Classification of the five major choledochal cyst forms and some subtypes.

A Type 1 Type 2
B Prenatal Diagnosis
FIGURE 106-2 Two types of choledochoceles. (A from O’Neill JA:
Choledochal cyst. Curr Probl Surg 1992;29:374. Used with permission.)
locations. The dilation is primarily fusiform. Pancreatico-
biliary maljunction is common, as is stricture of the common
bile duct. Sludge, cholelithiasis, or choledocholithiasis is
commonly found with choledochal cysts.
The histopathology depends on patient age and severity
of symptoms. The extrahepatic ducts are thickened with
dense connective tissue intermixed with smooth muscle
strands. Inflammatory reaction is usually present but may
be minimal in the infantile stage. Generally, the inflammatory
reaction increases with age and may lead to ulceration in
the mucosa and submucosa. The inflammation can extend
through the cyst leading to possible adherence to adjacent
structures. The most marked inflammation is noted in the
intrahepatic cysts. The lining of the cyst is not composed of
typical biliary mucosa but rather is relatively acellular and
may lack a typical mucosal lining. In the wall of the cyst,
mucin-secreting cells are seen in intramural glands. A gastro-
intestinal mucosa-type composition of immunoreactive
gastrin- and somatostatin-containing cells may also be found
in the cyst wall due to repeated destruction and regrowth
causing epithelial metaplasia. Small areas of columnar epithe-
lium and bile ducts can be noted in the cystic wall of young
patients.28 The degree of epithelial metaplasia generally in-
creases with age, as does the risk for adenocarcinoma.1,29–31
These histologic changes are commonly seen in pancreatico-
biliary maljunction as well. As mentioned, the distal common
bile duct has various degrees of stricture, with a high-grade
stricture commonly seen in the infantile stage. Older patients
have stenosis of the distal common bile duct but not complete
obstruction.
CHAPTER 106 CHOLEDOCHAL CYST 1333
The histologic changes of choledochocele differ from those
of the other forms of choledochal cysts that were described
earlier. As these ducts are primarily intraduodenal, the cystic
lining takes on a duodenal lining rather than biliary. The
common bile duct, although normal on gross pathology,
reveals inflammatory change on histologic examination.
Sludge and stones may also be identified in the common bile
duct extending into the extrahepatic and intrahepatic ducts,
especially in older patients.
Liver pathology in a newborn is usually normal or reveals
mild proliferation of intrahepatic ducts. Mild periportal fibro-
sis is often noted on liver histology of older patients. Congen-
ital hepatic fibrosis has also been described in some patients.32
Chronic inflammation is the likely cause of carcinoma in
the choledochal cyst wall found in older patients. PBM with
or without the presence of choledochal cysts increases the risk
factor for carcinoma of the biliary tract.28,33,34 Adenosqua-
mous carcinoma is the most common type of carcinoma found
in the wall, although small-cell carcinoma has also been
reported. The choledochal cyst wall is the primary location
of carcinomas, but the gallbladder and pancreatic head are
other common sites. The location of carcinoma in the head
of the pancreas is usually associated with the presence of
PBM.2,10,35
------------------------------------------------------------------------------------------------------------------------------------------------
Choledochal cysts are being diagnosed with increasing fre-
quency on prenatal anatomic ultrasounds in the second and
third trimesters.36–42 The typical finding on high-resolution
ultrasound is that of a cyst in the porta hepatis. The differential
diagnosis includes hepatic cysts, duodenal atresia, mesenteric
or omental cysts, intestinal duplication, gallbladder dupli-
cation, and ovarian cysts.43 Because choledochal cysts are rare
anomalies, prenatal diagnosis can be challenging.44 Having a
skilled and experienced sonologist performing the prenatal
ultrasound has, in our experience, been critical in increasing
both the sensitivity and specificity of prenatal ultrasound for
choledochal cysts.
Fetal magnetic resonance imaging (MRI) has also been
performed to aid in the diagnosis.45 Our group has not found
that MRI adds significantly to the information found on
ultrasound, but as both imaging techniques evolve, they
may increase diagnostic yield in a complementary fashion.
As discussed in further detail later in this chapter, choledo-
chal cysts historically presented in two broad categories:
(1) the infantile form indicated by obstructive jaundice and
(2) the adult form generally presenting with obstructive
jaundice, pancreatitis, or ascending cholangitis.46 Prenatally
diagnosed choledochal cysts could fall into either of these cat-
egories. The true natural history of choledochal cysts diag-
nosed prenatally is impossible to accurately delineate due to
widespread support for early surgical management to avoid
the high risk of potential complications from untreated chole-
dochal cysts. Clearly, those with an infantile form need surgi-
cal excision and reconstruction within the first several weeks
of life. Many of these neonates may have choledochal cysts in
conjunction with complete or nearly complete biliary obstruc-
tion. Some categorize these patients as having biliary atresia
in association with choledochal cyst, whereas others may term
1334 PART VII ABDOMEN

these patients as having what was historically called surgically Bile stasis, sludge, stone formation, inflammation, and recur-
correctable biliary atresia. Some overlap between the two rent superinfection have all been identified as complications
diseases is likely. leading to intermittent jaundice and abdominal pain in the
In those patients with prenatally diagnosed choledochal right upper quadrant. Mucous or protein plugging is likely
cysts that are asymptomatic, opinions differ in management. due to chronic inflammation leading to formation of albumin-
Some have advocated for repair within the first several months rich exudates or dysplastic epithelium with hypersecretion
of life to avoid potential complications of cholangitis and of mucin.53 Persistent bacterial colonization of intrahepatic
hepatopathy.39,47 These papers compare infants diagnosed ducts in type IV and V cysts is exacerbated by bile stasis,
prenatally who underwent surgery within the first 6 months sludge, and stone formation and is one theory for recurrent
of life to children diagnosed later who underwent surgery at cholangitis. Secondary biliary cirrhosis is noted in 40% to
an average age of 4 to 6 years. In one series, children in the 50% of patients due to obstruction and inflammation and
older group had greater hepatic fibrosis, as well as a higher in- leads to portal hypertension in some patients.48 In cases of
cidence of preoperative clinical manifestations of choledochal choledochal cyst not diagnosed until adulthood, patients
cysts. However, no long-term follow-up was conducted to may present with cholelithiasis and symptoms mimicking
provide data on long-term liver function and quality of life biliary colic or cholecystitis. If imaging has been inadequate,
in order to compare these two different cohorts. In our center, the diagnosis of choledochal cyst may only be made intrao-
our general recommendation has been to perform laparo- peratively by visual inspection or cholangiography. In some,
scopic operation for asymptomatic choledochal cysts diag- cholecystectomy may be performed without diagnosis of the
nosed prenatally within the first year of life. choledochal cyst. Pain similar to recurrent pancreatitis has
been described by some patients, which has led to a misdiag-
nosis of abnormal amylase clearance; however, after further
Clinical Presentation work-up, an elevated serum amylase is noted.54,55 However,
------------------------------------------------------------------------------------------------------------------------------------------------
these patients may truly be having pancreatitis as a result of
Patients with choledochal cysts can manifest clinical symp- mucous plugging in the PBM.
toms at any time during their life, with 80% of patients being
symptomatic before the age of 10 years.48 Abdominal pain,
jaundice, and a palpable right upper quadrant abdominal Diagnosis
mass is the classic triad described for patients with choledo- ------------------------------------------------------------------------------------------------------------------------------------------------

chal cyst. The triad is only reported in about 20% of patients

LABORATORY STUDIES
diagnosed. Two of the three symptoms are seen in two thirds
of patients at the time of diagnosis.48 Some have applied a clas- Laboratory data are usually obtained in patients with chole-
sification according to age at presentation. The infantile form dochal cysts in the diagnostic workup but are reserved primar-
occurs before 12 months of age, and these patients tend to pre- ily for evaluating the clinical condition of the patient and not
sent with obstructive jaundice, acholic stools, and hepatomeg- diagnosis. Serum markers of obstructive jaundice (e.g., conju-
aly similar to biliary atresia.49 The adult form occurs anytime gated hyperbilirubinemia and increased serum alkaline phos-
after 12 months of age and usually has a greater number of phatase) are usually present. An elevated conjugated bilirubin
symptoms including fever, nausea, vomiting, and jaundice.4 is typical in the infantile form but may or may not be present
As previously mentioned, choledochal cysts diagnosed prena- in the adult form due to intermittent or incomplete biliary ob-
tally may fall into either category. struction. In chronic cases an abnormal coagulation profile
Signs of hepatic fibrosis may be present in the infantile may be evident due to hepatic injury.
form, and these patients benefit the most from early treat-
ment. Delayed treatment of asymptomatic infants is advo- IMAGING STUDIES
cated by some; however, the risk of progressive hepatic
fibrosis exists. In the infantile form, patients with the fusi- Diagnosis of choledochal cysts is made by imaging studies. An
form type of choledochal cyst tend to present with transient abdominal ultrasound is the first imaging modality used be-
jaundice. Perforation of the choledochal cyst is rare (1% to cause it is noninvasive, inexpensive, and provides excellent
12%) and thought to be due to a fragile cystic wall from in- detail of the portal structures. Choledochal cyst on ultrasound
flammation, increased ductal pressure, or increased intra- is commonly identified as a cystic mass that is not part of the
abdominal pressure. The site of rupture is at the low-flow gallbladder in the right upper quadrant and is usually in
region of the junction of the cystic and common bile ducts. the porta hepatis. The cyst needs to be in continuity with
These patients present with abdominal pain, sepsis, and the biliary tree to be diagnosed as a choledochal cyst because
peritonitis.50 other cystic abnormalities such as pancreatic pseudocyst, bil-
The adult form usually presents after 2 years of age. These iary cystadenomas, or echinococcal cysts can exist in this
patients tend to have a fusiform dilation of the common bile area.18 The diagnosis of choledochal cyst on ultrasound
duct without complete obstruction of the distal common bile carries a 71% to 97% sensitivity.56 Ultrasonography is an ex-
duct and present with intermittent symptoms from a mucous cellent screening tool and is used by some as the only imaging
plug or biliary sludge. Patients with the adult form are more study in infants.
likely to present with the classic triad of abdominal pain, jaun- In the event that choledochal cyst is suspected on ultra-
dice, and a palpable right upper quadrant abdominal mass sound, a technectium-99 HIDA scan may provide more defin-
compared with the infantile form. itive data. Photopenia is initially evident at the cyst followed
Symptoms arise from biliary obstruction leading to ascend- by filling and delayed emptying. The HIDA scan has varying
ing cholangitis and complications of pancreatitis.3,27,46,48,51,52 sensitivities, with 100% for type I cysts and 67% for type IV.56
 CHAPTER 106 CHOLEDOCHAL CYST 1335

A HIDA scan has decreased sensitivity for intrahepatic cysts.

A HIDA scan may also be helpful for distinguishing between
choledochal cyst and biliary atresia. Biliary atresia on a HIDA
scan is characterized by lack of contrast emptying into the
duodenum, whereas a choledochal cyst will have contrast
entering the duodenum. A HIDA scan may also be useful in
the diagnosis of cystic rupture, in which case contrast would
empty into the peritoneal cavity.48
An abdominal computed tomography (CT) scan may be
another useful modality. A CTcan show the intrahepatic ducts,
distal common bile ducts, and the pancreatic duct, features
that are not reliably identified on ultrasound, making CT scan
highly useful in identifying type IV and type V cysts. The full
anatomy of the biliary tree can be delineated using CT cholan-
giography, a valuable modality in preoperative planning.
The sensitivity of CT cholangiography is 93% for biliary tree
visualization and diagnosing lithiasis, 90% for diagnosing
choledochal cysts, and only 64% for characterizing the
pancreatic duct.57 The risk of using CT or CT cholangiography
is contrast toxicity that can cause nephrotoxicity or hepato-
toxicity and radiation exposure.
Invasive imaging studies also play a role in diagnosis
and evaluation of the biliary anatomy. These modalities FIGURE 106-3 A magnetic resonance cholangiopancreatography of a
include cholangiography through endoscopic retrograde patient with type I choledochal cyst and an accessory pancreatic duct
entering the duodenum.
cholangiopancreatography (ERCP), percutaneous transhepa-
tic cholangiography (PTC), and intraoperative cholangiogram.
Cholangiography will delineate the anatomy and locate abnor-
malities such as PBM and filling defects caused by stones,
stenosis, or carcinomas.58,59 The risk of cholangitis and pan-
creatitis after invasive cholangiography is higher than in the
general population, with reported incidences as high as
87.5%48 due to the frequency of PBM, dilated ducts, and
dysfunctional sphincters in patients with choledochal cyst.
Due to recurrent inflammation and scarring, ERCP can be
challenging with respect to cannulization of the ampulla.
A high volume of dye load is often required for adequate
visualization, especially in the setting of large cysts.48
Magnetic resonance cholangiopancreatography (MRCP) is
now considered the gold standard for imaging choledochal
cyst, especially given the complications and concerns with in-
vasive cholangiography (Fig. 106-3). The diagnostic sensitiv-
ity of MRCP is 90% to 100%.60 Administration of secretin
increases pancreatic secretion and dilates the pancreatic duct.
Thus some centers administer secretin before MRCP to in-
crease diagnostic yield.61 As the technology for MRCP has im-
proved, smaller ductal structures are being visualized with
finer detail. This has allowed for increased diagnosis of PBM
and pancreatic divisum and usually gives accurate information
as to the location of insertion of the pancreatic duct. The type
and extent of the choledochal cyst are well visualized, and im- FIGURE 106-4 Three-dimensional reconstruction of a magnetic reso-
ages can be reconstructed in three dimensions (Fig.106-4). nance cholangiopancreatography image of a choledochal cyst with mild
MRCP avoids radiation associated with CT scan and ERCP dilation (6 mm) of the right and left extrahepatic ducts. Pancreatic duct
and avoids the potential post-ERCP complications of cholan- is seen entering the normal caliber distal common bile duct.
gitis and pancreatitis.
Choledochoceles (type III cysts) can be diagnosed with var- as a sphincterotomy. Endoscopic ultrasound is also well-
ious imaging modalities such as upper gastrointestinal series suited to diagnosing choledochoceles because this modality
(UGIS), endoscopy, ERCP, MRCP, and CT cholangiography. can reach close to the cyst for appropriate visualization. The
On UGIS for choledochoceles, a filling defect is noted at the best imaging techniques for Caroli disease (type V cysts) are
level of the cyst entering the duodenum. The papilla will have ultrasound, CT, and MRI.48 The “central dot sign” has been
a smooth bulging on endoscopy and ERCP. The advantage of described for Caroli disease but is not pathognomic.62 On
ERCP over MRCP and CT cholangiography for choledocho- HIDA scan a beaded appearance due to intraductal bridging
celes is the ability to perform a therapeutic procedure such may be indicative of Caroli disease.
1336 PART VII ABDOMEN

Our preference for imaging for all types of choledochal In previous series of choledochal cysts, diagnosis was often
cysts is ultrasound followed by MRCP. If the MRCP does delayed, leading to patient presentation with repeated
not give sufficient anatomic detail to guide therapy, episodes of cholangitis. Delayed diagnosis led to inflammation
we perform either an ERCP before operation or an intraopera- of the cyst and adherence to the periportal structures, partic-
tive cholangiogram at the time of cyst resection. ularly the portal vein. In patients such as these, for whom
excision of the full thickness of the cyst may carry an unac-
ceptably high complication rate, the technique described by
Surgical Management Lilly can be performed70,71: The cyst can be incised anteriorly
------------------------------------------------------------------------------------------------------------------------------------------------
away from the hepatic artery and portal vein, allowing the
The management of types I and IV choledochal cysts has inner epithelial lining of the entire choledochal cyst to be
evolved from either internal or external drainage to cyst excised while avoiding the difficult and potentially dangerous
excision and hepaticojejunostomy to a Roux-en-Y limb plane of dissection between the outer cyst wall and the portal
(Fig. 106-5). Due to the relatively complex nature of cyst vein. This removes the potentially premalignant aspects of
excision, early attempts at treatment focused on external drain- the cyst while minimizing potential operative complications.
age to avoid complications of biliary tract obstruction. How- Type II choledochal cysts are rare but appear to have a low
ever, an external biliary fistula proved to have high morbidity malignant potential. This type requires simple cyst resection
and mortality.4 As surgical techniques advanced, subsequent and is easily amenable to a laparoscopic approach because
attempts focused on cyst duodenostomy or cyst jejunostomy complicated dissection and reconstruction are not required.
to a Roux-en-Y limb to internalize biliary drainage.15,27,63,64 In fact, the first laparoscopic treatment for a choledochal
However, complications of cholangitis and progressive liver cyst in the pediatric population was reported in a child with
damage and the high risk of cancer in the choledochal cyst a type II cyst.72
led to the current surgical strategy of cyst excision with Type III choledochal cysts, or choledochoceles, are intra-
Roux-en-Y jejunostomy as advocated by Kasai and colleagues65 duodenal or intrapancreatic dilations of the distal common
and Ishida and colleagues.66–68 This operative approach elim- bile duct, as outlined previously. Management has tradi-
inates the potentially premalignant epithelial cyst lining and tionally been operative marsupialization of the cyst, usually
also separates the pancreatic drainage from the biliary drainage. through a transduodenal approach.73 Increasingly, however,
This strategy is summarized in Figure 106-6. This technique is choledochoceles are being treated by sphincterotomy or cyst
essentially identical to our laparoscopic technique for treat- marsupialization during an ERCP.74–76 The incidence of
ment for type I and type IV cysts and is described in detail later cancer is unclear due to the rarity of this type of cyst but is
in the laparoscopic section. Some surgeons prefer cyst resection likely significantly elevated.77 It is unclear to what extent
with hepaticoduodenostomy, which yields similar outcomes to drainage with either an operative approach or via ERCP
that of drainage with a Roux-en-Y limb.69 minimizes the malignant potential for choledochoceles.

Cyst excision,
Cyst duodenostomy hepaticojejunostomy

Roux-en-Y
cyst jejunostomy

FIGURE 106-5 Surgical operative management of choledochal cyst and the types of anastomoses that can be created.
 CHAPTER 106 CHOLEDOCHAL CYST 1337

FIGURE 106-6 Diagram reveals

preoperative anatomy (A) and post- A
operative anatomy (B) after cyst exci-
B
sion and Roux-en-Y reconstruction.

For patients with type IV disease, most surgeons recom-

mend cyst resection to the hepatic ducts, leaving in place
the dilated intrahepatic ducts because they may decrease in
size without distal obstruction. However, careful long-term
follow-up is required to monitor the relative size of the intra-
hepatic ducts and for the possibility of symptomatic lithiasis
and cholangitis. Surgical treatment of type V (Caroli disease)
is challenging. Segmental resections can be performed
if the disease is localized to a portion of the liver. Liver trans-
plantation has also been performed.78,79
We and others have reported laparoscopic excision of
choledochal cysts with either Roux-en-Y reconstruction or
hepatico-duodenostomy.80–83 The largest series, from Tang
and colleagues,83 was a retrospective review of 62 children
undergoing laparoscopic resection of choledochal cysts. They
reported outcomes comparable with those of published open
series with a low rate of complications including only one case
of a biliary leak. Long-term outcome for laparoscopic surgery
for choledochal cyst in the pediatric population has not been FIGURE 106-7 Port placements for laparoscopic choledochal cyst exci-
well-reported, and data are similarly scarce for outcomes of sion and Roux-en-Y hepaticojejunostomy.
open surgery. A recent series from the long-term follow-up
data reported by Ono and colleagues is discussed in “Outcome left-handed working port for the primary surgeon. A 5-mm
and Complications” later and provides the best overall long- port is placed in the left midclavicular line just superior to
term data from open cases.84 the level of the umbilicus for the right-handed working port
Techniques for laparoscopic repair including ours are all for the primary surgeon. The fourth port is located in the left
similar. This next section describes in detail our technique anterior axillary line just below the costal margin. This port is
for laparoscopic excision of type I choledochal cysts with used by the assistant to retract the gallbladder and liver supe-
Roux-en-Y reconstruction, as previously reported.81 riorly. The location of this port, although it may seem counter-
Detailed preoperative workup including an MRCP is essen- intuitive, has proven to be the optimal location for retraction
tial in determining the type of cyst and the insertion of the and affords adequate distance from the other instruments.
pancreatic duct(s) for safe resection. Particularly in instances Placing two ports in the right abdomen, as one would typically
in which the pancreatic duct inserts near the distal extent of do for a cholecystectomy, results in inadequate distance
the choledochal cyst, care must be taken to avoid the pancre- between instruments placed through these ports. After lapar-
atic duct in the distal dissection. Intraoperative cholangio- oscopically confirming the diagnosis of a choledochal cyst,
graphy may also be performed by placing a catheter in the a cholangiogram is performed, if necessary, by placing an
gallbladder if MRCP is inadequate or not available. The place- angiocatheter into the gallbladder and installing water soluble
ment of four ports for the procedure is illustrated in contrast (Fig. 106-8).
Figure 106-7. A 5-mm port is placed through the umbilicus The gallbladder is then divided at the cystic duct junction
for the camera. Typically, we prefer a 4-mm or 5-mm with clips or ties (Fig. 106-9). This allows the assistant to
30-degree endoscope due to the improved optics as compared retract the gallbladder superiorly through the left subcostal
with smaller endoscopes. A 3-mm port is placed in the right port and expose the porta hepatis. If the cyst is large enough
abdomen between the midaxillary line and the midclavicular that the hepatic hilum is obscured, it may need to be decom-
line just inferior to the level of the umbilicus. This is the pressed. Next, the hepatic duct is divided just superior to the
1338 PART VII ABDOMEN
FIGURE 106-8 Intraoperative cholangiogram during laparoscopic cyst
excision and Roux-en-Y hepaticojejunostomy revealing the gallbladder
(GB) and the choledochal cyst (CY).
FIGURE 106-9 Intraoperative laparoscopic view of the gallbladder (GB)
and cystic duct in relation to the choledochal cyst.
most cephalic extent of the choledochal cyst. Care should be
taken to identify the right and left branches. In some instances,
separate anastomoses may be required from the right and left
ducts to the Roux-en-Y limb. Dividing the hepatic duct early
in the operation allows the surgeon to manipulate both the
superior and right lateral aspects of the choledochal cyst to
aid in the dissection. The cyst is then carefully dissected away
from the hepatic artery and the portal vein. At this point, the
cyst is attached only to the distal common bile duct within the
pancreas. Dissection continues distally, dissecting closely on
the cyst wall to avoid damage to the duodenum or pancreas
until the cyst begins tapering in size to become a normal-sized
common bile duct. Again, the MRCP can be valuable in guid-
ing the distal dissection with respect to the location of the
pancreatic duct insertion. The common bile duct is excised,
and the distal common bile duct is ligated with a tie or
occluded with a clip. If the cyst is not leaking bile, we prefer
to place the cyst on the lateral aspect of the liver and remove it
subsequently along with the gallbladder.
We then begin the creation of the Roux-en-Y limb. The
ligament of Treitz is identified by elevating the transverse
colon. The area of the jejunum that will be divided for the
Roux-en-Y limb is identified approximately 15 cm from the
ligament. We prefer to mark this area by placing vessel loops
of different colors separated by 1 to 2 cm. Marking the jeju-
num in this manner allows easy identification of the proximal
and distal aspects of the jejunum when the bowel is exterior-
ized. Next, the umbilical incision is extended to approxi-
mately 1.5 cm and the marked jejunum is exteriorized.
A 30- to 40-cm Roux-en-Y limb is then created in the standard
fashion, the bowel is then returned intracorporeally, and the
umbilical port is replaced. The Roux-en-Y limb is brought
to the porta hepatic in a retrocolic fashion, and then an
end-to-side anastomosis is performed from the hepatic duct
to the Roux-en-Y limb. We use fine absorbable monofilament
suture material. We have found that using a continuous suture
on the posterior aspect of the anastomosis and interrupted su-
tures on the anterior and lateral aspects is simplest. All knots
are tied on the outside of the anastomosis to prevent the
potential for a focus of formation or collection of biliary
sludge on the knots. Finally, a cholecystectomy is performed
and the gallbladder and the choledochal cyst are removed
through the umbilical port site.
Outcome and Complications
------------------------------------------------------------------------------------------------------------------------------------------------
In past operative treatment of cystenterostomy, there was
a high rate of complications such as anastomotic stricture,
recurrent ascending cholangitis, bowel obstruction, portal
hypertension, and malignancy. However, the advancement
of surgical treatment to include cyst excision resulted in min-
imal morbidity and mortality and reduced the number of late
complications. The most common late complication continues
to be anastomotic stricture. Early diagnosis and cyst excision
results in low complication rates in most experienced centers.
The type of internal drainage after cyst excision has not
been shown to affect postoperative complications. The tech-
nique of Roux-en-Y hepaticojejunostomy is favored by
most, although comparable results can be achieved by hepa-
ticoduodenostomy. Either procedure can be performed lapar-
oscopically as well.
Although late complications are reduced with current sur-
gical management, studies suggest that long-term follow-up to
17 years or beyond is indicated due to the potential for prob-
lems such as anastomotic stricture, cholangitis, intrahepatic
stone formation, and malignancy.85 This is particularly impor-
tant for patients with type IV disease because malignancy can
occur in incompletely resected cystic hepatic ducts or recur-
rent cysts.85 The risk of malignancy is greatly reduced after
cyst excision but is still elevated as compared with the general
population. Intrahepatic stone formation in the intraoperative
setting has been evaluated and reported in the literature and
seen in patients who showed no stone formation in the preop-
erative course. Stone formation has been reported to occur

anywhere from 3 to 22 years postoperatively.86 However, if the
duct is patent and there is no stenosis of the hepaticojejunos-
tomy, stones are likely to pass spontaneously. Intrahepatic
stones usually present in cases of stenosis that initially cause
bile stasis and lead to stone formation.85 Todani and col-
leagues87 reported a 25-year review with the identification
of biliary complications primarily associated with either
anastomotic stricture or primary ductal stricture and recom-
mended a wide hepatic hilum anastomosis to prevent biliary
complications.
The long-term complication of malignancy can be avoided
with cyst excision. In addition, any cause of chronic inflamma-
tion and stasis, as seen with PBM, has to be treated appropri-
ately.34 The incidence of malignancy of the gallbladder and
bile ducts remains high in patients with PBM compared with
the general population and occurs at a younger age. Further
studies reviewing the genetic changes of PBM have revealed
KRAS gene mutations early in epithelial hyperplasia and meta-
plasia. In the late progression of adenocarcinoma of the biliary
tract, inactivation of DPC4 gene occurs.28,44,88
CHAPTER 106 CHOLEDOCHAL CYST 1339
Conclusion
------------------------------------------------------------------------------------------------------------------------------------------------
The management of choledochal cysts is performed pre-
dominantly by pediatric surgeons. Generally considered a
congenital abnormality, most cases are diagnosed in infancy
with many now being diagnosed prenatally, though presenta-
tion may be delayed until adulthood. Because of the high
risk of serious sequelae including malignancy, early surgical
excision is warranted even in asymptomatic patients. Ade-
quate imaging with MRCP helps map the extent of the dilated
biliary tree, as well as the insertion of the pancreatic
duct(s), ensuring safe resection. Increasingly, surgery is being
performed with minimally invasive techniques with good
outcomes. Because of the rare nature of this disease, exten-
sive knowledge of the different variants and experience with
advanced biliary tract surgery is critical in attaining good
outcomes.
The complete reference list is available online at www.
expertconsult.com.
 circulation of bile. The administration of fat is believed to ame-
liorate the deleterious effects of the amino acids in the TPN.
Because many infants receive TPN and only a few develop
gallstones, other factors are probably necessary for the devel-
opment of cholelithiasis in this patient population. Septi-
cemia, dehydration, chronic furosemide therapy, cystic
fibrosis, short-bowel syndrome, and ileal resection for necro-
tizing enterocolitis also may be important contributing
factors.6,8–10 The cause of gallstone formation is probably
multifactorial. Stasis, sepsis, a lack of enteral feedings, and
some alteration of the enterohepatic pathway for the recircu-
lation of bile are all important for stone development. More-
over, neonates and, particularly, premature infants are
susceptible to the cholestatic effects of TPN because of the im-
maturity of their enterohepatic circulation of bile salts. Finally,
the composition of gallstones in children may be different than
that found in adults. Whereas stones in adults are primarily
cholesterol in composition, calcium carbonate and black
pigment stones are often found in pediatric patients, especially
those younger than 10 years of age.11–13
Causes associated with gallstones in older children include
the use of oral contraceptives, cystic fibrosis, pregnancy,
obesity, and ileal resection.14–17 Cholelithiasis has also been
CHAPTER 107 reported in children undergoing cardiac transplantation
who are receiving cyclosporine and in patients who have
previously required extracorporeal membrane oxygenation
as a newborn.18–21
Gallbladder HEMOLYTIC CHOLELITHIASIS
Disease and Although the incidence of nonhemolytic cholelithiasis is in-
creasing at many centers, hemolytic cholelithiasis (usually sec-

Hepatic Infections ondary to sickle cell disease [SCD]) remains the most common
cause of gallstones in children at many academic institutions,
especially in urban areas where an active sickle cell program
George W. Holcomb III and Walter S. Andrews exists. The prevalence of pigment gallstones associated with
sickle cell disease appears to be age dependent. Fifty percent
of patients with sickle cell anemia develop gallstones by 20
years of age.22–24 Interestingly, in one study, patients with sickle
cell disease and gallstones had rates of bilirubin production,
Gallbladder Disease
------------------------------------------------------------------------------------------------------------------------------------------------
bilirubin clearance, and plasma levels of unconjugated biliru-
bin similar to patients with sickle cell hemoglobin and no gall-
Cholelithiasis and biliary dyskinesia are being increasingly stones.25 Thus the researchers suggested that excessive
diagnosed in children. Whether their incidence is actually unconjugated hyperbilirubinemia alone is not sufficient to pro-
escalating or the diagnostic accuracy is improving because duce pigment gallstones. In that study, enlarged fasting and
of the increasing use of ultrasonography (US) and cholescinti- postprandial gallbladder volumes were found in patients
graphy remains unclear. Moreover, hemolytic disease is no who developed stones. It was hypothesized that stasis occurs
longer a prerequisite for the development of gallstones.1–3 as a result of incomplete emptying and leads to sludge and later
to gallstone formation. In addition, genotyping may become a
screening tool for predicting SCD children who are most likely
NONHEMOLYTIC CHOLELITHIASIS to develop gallbladder disease.26 Because gallbladder sludge is
frequently documented in patients with sickle cell anemia,
Several factors are known to cause cholelithiasis in children. elective cholecystectomy has been recommended when
By far, the most commonly reported condition associated with evidence suggests the presence of sludge, with or without
the development of nonhemolytic cholelithiasis, particularly stones.27 In one study of 35 patients with sickle cell disease
in neonates and infants, is the use of total parenteral nutrition and biliary sludge, 23 (65.7%) went on to develop gallstones.28
(TPN).4–6 Up to 43% of children receiving long-term TPN
develop gallstones.7 The mechanism of gallstone formation
COMPLICATIONS
related to TPN remains unclear but may be associated with
changes in bile composition caused by the amino acid Patients with SCD represent the largest group of patients at
infusion. Also, the lack of enteral feeding leads to ineffective risk for postoperative complications after cholecystectomy.
gallbladder contraction and results in reduced enterohepatic In a report of 364 cases from a national sickle cell disease
1341
1342 PART VII ABDOMEN

study group, the total complication rate was 39%, with sickle
cell events representing 19%; intraoperative or recovery room
problems, 11%; transfusion complications, 10%; postopera-
tive surgical events, 4%; and death, 1%.29 The open operation
was performed in 58% of the patients, and the laparoscopic
route was used in 42%. The complication rates were similar be-
tween the two groups. From that study, it was believed that the
incidence of sickle cell events may be higher in patients who
were not preoperatively transfused. Moreover, the laparoscopic
approach is now recommended for this patient population.30
The acute chest syndrome can be seen in up to 20% of
sickle cell patients undergoing abdominal operation. The
laparoscopic approach did not decrease the incidence of
this complication.31 Meticulous attention to perioperative
management, transfusion guidelines, and pulmonary care
may reduce the incidence of the acute chest syndrome.
Two other hemolytic conditions commonly associated with
gallstones are hereditary spherocytosis and thalassemia.
The incidence of cholelithiasis in patients with hereditary
spherocytosis ranges from 43% to 63% and is slightly more
common in girls than boys.32,33 Because of this association,
US is recommended before elective splenectomy to determine
whether concomitant cholecystectomy should be performed
at the time of the splenectomy. Previously, patients with
thalassemia major comprised a group of children at risk for
cholelithiasis.34 However, the incidence of cholelithiasis has
markedly decreased in this patient population. This reduction
has been attributed to a hypertransfusion regimen that blocks
the bone marrow so that the fragile cells of thalassemia major
are not produced.35 Currently, patients with this disorder are
rarely seen with cholelithiasis.
cholecystectomy, 22% presented initially with complica-
tions related to gallstones.19 Seven presented with acute
cholecystitis, six presented with jaundice and pain, five de-
veloped gallstone pancreatitis, and four had acute biliary
colic without evidence of cholecystitis. Others have also
reported a large number of patients presenting with symp-
tomatic choledochal obstruction.36,37 In such patients,
management should be initially directed at trying to reduce
the inflammation in the region, followed by laparoscopic
cholecystectomy several days later.
ACALCULOUS CONDITIONS
Hydrops of the gallbladder, acalculous cholecystitis, biliary
dyskinesia, and gallbladder polyps are being seen more
frequently. Hydrops is characterized by massive distention
of the gallbladder in the absence of stones, infection, or con-
genital anomalies. It has been most frequently reported in as-
sociation with Kawasaki disease and is usually due to a
transient obstruction of the cystic duct or to increased mucus
secretion by the gallbladder resulting in poor emptying.38–42
With additional gallbladder distention, further angulation
of the cystic duct may increase the obstruction.43 Conserva-
tive treatment is initially recommended in this setting.
Appropriate antibiotics for septic patients and early initia-
tion of enteral feedings to stimulate gallbladder emptying
often lead to resolution of this condition. If serial US exam-
inations show progressive gallbladder distention with increas-
ing pain, or if the gallbladder appears gangrenous,
cholecystectomy is recommended.

CLINICAL PRESENTATION
Gallbladder disease in infants usually occurs during a systemic
illness or following TPN use. Some infants will also have
undergone ileal resection because of acquired diseases such
as necrotizing enterocolitis or congenital anomalies. Jaundice
may occur before gallstones are detected with US. Most of
these stones are composed of calcium bilirubinate.
Gallstones that develop between 2 and 12 years of age are
primarily composed of mixtures of calcium bilirubinate, with
varying amounts of calcium carbonate and cholesterol. Opaque
gallstones are sometimes detected on plain radiographs
(Fig. 107-1), but the diagnosis is usually determined with US.
Unfortunately, the diagnosis may be delayed in this young age
group despite abdominal pain, nausea, emesis, and, occasion-
ally, fever. Abdominal pain may vary from the typical right
upper abdominal location to vague and poorly localized pain,
especially in younger children. Older patients are usually more
specific in localizing their pain to the right subcostal region.
The symptoms and physical findings in teenagers are
similar to those found in adults. The pain is usually described
as right subcostal with some radiation to the subscapular area.
Nausea and vomiting commonly occur, and intolerance to
fatty foods is often mentioned.
The primary reason for elective cholecystectomy in
patients with cholelithiasis is to prevent associated compli-
cations. Four major complications can occur: cholecystitis,
FIGURE 107-1 This 9-year-old girl developed nausea and right upper ab-
jaundice, cholangitis, and biliary pancreatitis. The inci- dominal pain. This plain radiograph revealed calcified gallstones (arrow)
dence of acute cholecystitis seems to be increasing. In that had settled in the dependent portion of the gallbladder on this up-
one report of 100 patients undergoing laparoscopic right abdominal film.
 CHAPTER 107
Acalculous cholecystitis commonly occurs in association
with severe illness such as sepsis, burns, or trauma that results
in dehydration, hypotension, and ileus. In this setting, TPN is
often administered and, if prolonged, may be followed by de-
creased gallbladder contractility with progressive distention,
stasis, and possible infection. In a small report of 12 patients,
the investigators used daily US criteria for clues to indicate the
need for cholecystectomy. In the three patients who under-
went operation, there was progressively increasing gallbladder
wall thickness and distention along with pericholecystic fluid.
In the other patients, the daily US examinations found pro-
gressive improvement from the previous day’s findings. These
patients all recovered uneventfully.44
Biliary dyskinesia is now a common diagnosis in chil-
dren.45–56 In some centers, it has become the most common
reason for cholecystectomy.57–59 This disorder is character-
ized by poor gallbladder contractility and the presence of
cholesterol crystals within the bile. It should be considered
when patients present with typical biliary pain but no evi-
dence of gallstones on US. Gallbladder contractility can be
assessed with radionuclide scanning during cholecystokinin
(CCK) injection. Recently, Lipomul has been shown to provide
similar findings.49 Most surgeons use a gallbladder ejection
fraction of less than 35% as an indicator for cholecystectomy
in a patient with symptoms.53,59–62 Laparoscopic cholecystec-
tomy has been shown to be an effective treatment for this
disorder, with expected resolution of symptoms in more than
80% of the patients using this benchmark.* Chronic cholecys-
titis is often documented on histologic examination of the gall-
bladder specimen.63 A few centers use a lower benchmark for
operation. In a review of 51 patients who underwent laparo-
scopic cholecystectomy for symptoms consistent with biliary
dyskinesia, it was found that nausea, pain, and a decreased
gallbladder ejection fraction of less than 15% most reliably
(93% PPV, 81% NPV) predicted which children will benefit
from cholecystectomy for this condition. In that report, chil-
dren with an ejection fraction greater than 15% did not have
predictable resolution of symptoms.46
In a recent report from our institution, we have found that
children with biliary dyskinesia have a marked increased
number of mucosal mast cells in the gallbladder mucosa com-
pared with patients with stone disease.64 In a follow-up study,
a moderate to high degree of mast cell activation was also
found in children with both biliary dyskinesia and gall-
stones.65 These findings provide biologic credibility for cho-
lecystectomy in both patient groups.
Gallbladder polyps are being described more frequently in
children.66,67 Cholecystectomy is advisable if there are biliary
symptoms or if the polyp is greater than or equal to 1 cm.68,69
Several unusual conditions merit attention. First, partial
external biliary diversion interposing a jejunal loop between
the gallbladder and abdominal wall has been described to treat
the intractable pruritus in patients with progressive familial
intrahepatic cholestasis.70 We have used this technique as well
with resolution of the pruritus. Thus in these children, chole-
cystectomy should be avoided. Second, ventriculo-gallbladder
shunts have been performed in patients with a scarred perito-
neal cavity from multiple previous operations or severe
peritonitis.71 We have performed one ventriculo-gallbladder
shunt with a good result.
*References 45–48, 50, 51, 53, 59, 61–62.
GALLBLADDER DISEASE AND HEPATIC INFECTIONS 1343
RADIOGRAPHIC EVALUATION
Although a plain abdominal radiograph is often the initial
imaging study used to evaluate abdominal pain in children, it
is rarely helpful with gallbladder disease unless the gallstones
are calcified (see Fig. 107-1). The incidence of radiopaque stones
has been reported to be as high as 50% in patients with hemolytic
disorders compared with approximately 15% in adolescents
with cholesterol stones.72 Oral cholecystography is rarely used
today because of the superior accuracy of US. Real-time US
has an accuracy of approximately 96% for gallbladder disease
and is effective in determining hepatic and common bile duct
involvement, the presence of thickening of the gallbladder wall,
and any abnormalities in the liver or head of the pancreas.73 The
most useful procedure for diagnosing acute cholecystitis is cho-
lescintigraphy that uses technetium-99m-labeled iminodiacetic
acid analogues. With this study, the gallbladder is not visualized
in patients with acute cholecystitis. In critically ill patients, espe-
cially infants who are fasting, have severe associated illnesses, or
are receiving TPN, there may be a false-positive result.
Intravenous morphine may be useful in this setting because it
causes spasm at the sphincter of Oddi, resulting in increased
bile duct pressure. The latter, in turn, enhances visualization
of the gallbladder and helps reduce the rate of false-positive
studies.74 Cholecystokinin-assisted or Lipomul-challenged
cholescintigraphy is an accurate predictor of biliary dyskinesia
and also suggests the likelihood of symptomatic relief with
cholecystectomy.45,46,48,49,75,76
NONSURGICAL TREATMENT
Nonsurgical treatment for gallstones in children or adults is
of historical interest only. Previously, attempts at oral dissolu-
tion therapy or extracorporeal shockwave lithotripsy were
attempted in adults with minimal success.43,77,78 In one adult
study evaluating lithotripsy, the duration of treatment, the
high cost, and the 50% risk of recurrent gallstones within
5 years were significant disadvantages.77 In these studies,
children were not included for a variety of reasons including
the fact that pigmented stones would not be amenable to these
therapies. Currently, observation is not recommended for
children older than age 2 or 3 years with gallstones. In some
reports, gallstones that developed in infants secondary to
prolonged TPN have been found to resolve spontane-
ously.79–82 Thus a 6- to 12-month period of observation in
this age group is not unreasonable after cessation of the
TPN and initiation of enteral alimentation.
In a study of 55 patients with biliary dyskinesia comparing
observation versus laparoscopic cholecystectomy, there was
no difference (74% vs. 75%) in resolution of pain at 2 years
between groups.83 However, of those patients whose
symptoms improved, those who underwent laparoscopic
cholecystectomy had a much quicker resolution of pain com-
pared with those who were observed.
SURGICAL TREATMENT
Surgical treatment for symptomatic cholelithiasis consists of
either cholecystectomy or cholecystolithotomy. In a prelimi-
nary report, one group reserved laparoscopic cholecysto-
lithotomy for symptomatic cholelithiasis in patients before
the onset of puberty.84 However, long-term outcome data from
1344 PART VII ABDOMEN
this approach are lacking. In another report of 10 patients un-
dergoing cholecystolithotomy over 25 years with a mean fol-
low-up of 5 years, 30% of the patients undergoing
cholecystolithotomy were reported as having recurrent right
upper abdominal pain and recurrent stones.85 Because of
the high recurrence rate, these authors believe that cholecys-
tectomy is preferred in children.
The laparoscopic approach has become the standard
method for cholecystectomy in children over the past 20 years.
The major advantages of this approach include decreased dis-
comfort and reduced length of hospitalization, an improved
cosmetic result, and a faster return to routine activities such
as work, school, play, or participation in athletic activities.
These advantages result from less muscle disruption from
the small incisions and reduced trauma to the tissue, leading
to less discomfort and ileus than with the open approach.
In most reports, children without complications undergoing
laparoscopic cholecystectomy are ready for discharge on ei-
ther the first or the second postoperative day.19,53,75,86–88
However, there are recent reports of laparoscopic cholecystec-
tomy being performed in children as an outpatient proce-
dure.59,89 Patients with sickle cell disease require special
preoperative care to prevent postoperative complications.
Several authors have reported favorable results with cho-
lecystectomy in this patient population and have emphasized
the need for preoperative transfusion.29,31,90–92 From recent
reports, the laparoscopic approach does not appear to be more
hazardous for these patients and may be preferred.19,29–31,91
Several special circumstances merit attention. The first is
the child with hereditary spherocytosis who is undergoing
splenectomy. In these patients, a gallbladder US is recom-
mended before the splenectomy. If gallstones are present, then
cholecystectomy should be performed at the time of the
splenectomy, regardless of whether the laparoscopic or open
approach is used. Whether splenectomy protects these
patients from the future development of cholelithiasis is
unclear. However, in a study of 17 patients undergoing
splenectomy alone in which cholelithiasis was not seen at
the time of the splenectomy, none of the patients subsequently
developed symptoms of cholelithiasis with a mean follow-up
of 15 years.93 Thus prophylactic cholecystectomy at the time
of splenectomy is probably not indicated in patients with
hereditary spherocytosis who do not have gallstones.
The second situation is that of the patient who presents
with known or suspected choledocholithiasis in addition to
cholelithiasis. A number of management strategies for such
patients are available.94–97 Options include preoperative
endoscopic retrograde cholangiopancreatography (ERCP)
with sphincterotomy and stone extraction, laparoscopic or
open common duct exploration at the time of laparoscopic
cholecystectomy, or postoperative endoscopic sphincterotomy
with stone extraction, which is the approach commonly used
in adults. This decision analysis appears to be related to two
factors: the surgeon’s experience with laparoscopic common
duct exploration and the availability of ERCP and sphincter-
otomy in children and adolescents at the treating institution.
For those pediatric surgeons who are facile with laparoscopic
common duct exploration, this may be an attractive approach
with postoperative endoscopic sphincterotomy reserved for
the unsuccessful cases. However, for most pediatric surgeons,
this is probably not the best option because postoperative
endoscopic sphincterotomy is a rarely performed procedure
in many pediatric surgical centers. Thus in this setting, it
would appear best to perform preoperative ERCP with sphinc-
terotomy and stone extraction if stones are found. If success-
ful, the surgeon can then proceed with uncomplicated
laparoscopic cholecystectomy in the next few days. However,
if the endoscopic sphincterotomy and stone extraction are not
successful, then the surgeon will know at the time of the
operative procedure whether a laparoscopic or open common
duct exploration is necessary (Fig. 107-2). Both approaches
have been used at our institution.
A final preoperative concern is whether to perform an
intraoperative cholangiogram. In the early to mid-1990s,
when there was not much experience with laparoscopic cho-
lecystectomy in children, it was suggested that most, if not
all, children undergo a cholangiogram for surgeon-training
purposes and to ensure that the correct anatomy has been vi-
sualized. A third reason was to evaluate for the presence of
common duct stones, although this is often suspected preop-
eratively either from symptoms, US, or laboratory studies. As
pediatric surgeons have gained more familiarity with the tech-
nique of laparoscopic cholecystectomy, routine intraoperative
cholangiography for surgeon-training purposes does not ap-
pear necessary once the surgeon has become familiar with
the technique. Thus in our minds, intraoperative cholangiog-
raphy is useful primarily to ensure that the correct anatomy is
visualized and to evaluate for choledocholithiasis. We often
obtain a US evaluation a few days before the laparoscopic
Suspected Choledocholithiasis
ERCP
Stone(s) Confirmed No Stones
Endoscopic Sphincterotomy
&
Stone Removal
Not Successful
successful
Laparoscopic or Open
Laparoscopic
Cholecystectomy and
Cholecystectomy
Common Duct Exploration
FIGURE 107-2 This algorithm depicts the authors’ preferred strategy for
managing patients with suspected choledocholithiasis. Preoperative en-
doscopic retrograde cholangiopancreatography (ERCP) is favored with
stone extraction and sphincterotomy, if necessary, before the laparoscopic
cholecystectomy. If the ERCP does not show evidence of choledocho-
lithiasis, then laparoscopic cholecystectomy is performed. The reason this
approach is favored is that, at the time of the laparoscopic cholecystec-
tomy, the surgeon knows whether a laparoscopic or open choledochal
exploration is necessary.
 CHAPTER 107 GALLBLADDER DISEASE AND HEPATIC INFECTIONS 1345

TABLE 107-1 Another modification in the authors’ technique over the

Financial Analysis of Preoperative Ultrasonography Versus past several years merits mention. For many laparoscopic op-
Intraoperative Cholangiography for Detection of erations including cholecystectomy, we have adopted a “stab
Choledocholithiasis at Children’s Mercy Hospital, Kansas incision” technique for instruments that are not continually
City, MO
exteriorized and reinserted into the abdominal cavity during
Immediate Charges Intraoperative Charges the operation. In a study of 511 patients undergoing a variety
Preoperative Cholangiography of laparoscopic procedures in which only one or two Step
Evaluation with
Ultrasound (Covidien, Norwalk, Conn.) cannulas were used, a reduction
of $187,180 was noted in the patient charges.99 If the Ethi-
Ultrasound study $307.67 15 minutes $734.25 con (Ethicon Endosurgery, Cincinnati) disposable cannulas
(including radiologist’s operating room
fee) time were used, patient savings of about $123,000 would have
C-arm with $365.41 been realized. For cholecystectomy, the instruments placed
radiologist’s fee through the two right-sided incisions are not routinely chan-
Sterile drape for $20.00 ged. Thus these two sites are optimal for the stab incision
C-arm technique (Fig. 107-4). The telescope is introduced through
Cholangiocatheter $83.50 the umbilical cannula, and the main working port is in the
Contrast agent for $40.00 upper abdomen, so two Step cannulas are still used in these
cholangiogram sites.
Total $307.67 Total $1,243.16

cholecystectomy to confirm the presence of gallstones and to
evaluate for common duct obstruction. This approach reduces
the operative time and is cost effective (Table 107-1). Thus at
the time of the operation, the primary need for
cholangiography is to ensure the anatomic roadmap. If the
anatomy appears clear at the time of the laparoscopic opera-
tion and there is no suspicion of common duct stones, routine
cholangiography is not performed. If cholangiography is
necessary, we still prefer the Kumar clamp technique, al-
though a number of other techniques in which a catheter is
introduced into the cystic duct through a transcystic incision
are certainly appropriate.1,98 Fluoroscopy is also useful
because it is more time efficient than static radiography and
allows dynamic assessment of the biliary tree (Fig. 107-3).
LAPAROSCOPIC CHOLECYSTECTOMY
Four-Port Technique
The patient is placed supine on the operating table with two
video monitors situated at the head of the table. After induc-
tion of anesthesia, an orogastric tube is inserted for gastric de-
compression and the urinary bladder is evacuated using a
Credé maneuver. A two-cannula and two “stab-incision” tech-
nique is used, but the location of the incisions and the cannu-
las depends on the patient’s age and size (Fig.107-5). For
infants and small children, the right lower abdominal incision
can be positioned in the inguinal crease region and the epigas-
tric cannula should be situated more on the patient’s left to al-
low adequate working space between the instruments.
A 10-mm incision is made in the umbilicus, through which
a 10-mm cannula is inserted into the abdominal cavity and ab-
dominal insufflation is initiated. After creation of an adequate

FIGURE 107-3 The Kumar clamp technique for cholangiography in infants and small children is depicted. Note the clamp across the infundibulum and
the sclerotherapy needle (arrow), which exits the side arm of the clamp and is introduced into the proximal infundibulum. The advantage of this technique
is that lateral incision and cannulation of the small cystic duct are not necessary.
1346 PART VII ABDOMEN
FIGURE 107-4 This operative photograph depicts the stab incision tech-
nique in a 5-year-old undergoing laparoscopic cholecystectomy. On the
patient’s right side, the two 3-mm instruments were introduced through
the stab incision technique without the use of cannulas. A 5-mm cannula
has been placed in the epigastric region, and a 10-mm cannula is intro-
duced into the umbilicus. This is the site of extraction of the gallbladder.
One advantage of the stab incision technique is cost savings because two
additional cannulas are not necessary.
pneumoperitoneum, a 5- or 10-mm 45-degree telescope is
inserted through the umbilical cannula and connected to
the camera. The image is then displayed on the video moni-
tors. The right lower abdominal stab incision is created with
a No. 11 blade, and a locking, grasping forceps is introduced
for retraction of the gallbladder superiorly over the liver by the
assistant. Usually this is a 5-mm instrument, but a 3-mm in-
strument can be used in small children. Through the right
upper abdominal stab incision, a nonlocking, grasping forceps
is inserted for lateral retraction of the infundibulum of the
gallbladder by the operating surgeon. The epigastric port is
usually 5 mm in diameter and is the main working site. Occa-
sionally, in larger adolescents, a 10-mm port may be necessary if
a 10-mm endoscopic clip is required to completely occlude the
cystic duct. Once the location of the cannulas is individualized
according to the patient’s size, the remaining principles of the
procedure are similar to those used in adults.
For improved exposure, the patient and table are usually
rotated into reverse Trendelenburg and left-dependent
FIGURE 107-5 Placement of the cannulas and instruments in patients undergoing laparoscopic cholecystectomy depends on the patient’s age and size.
In an infant, 3- and 5-mm ports/instruments can be used. Also, the ports/instruments should be separated as much as possible to create an adequate
working space for the operation (left). In a prepubertal child the port/instruments should be widely spaced as well. A 3-mm instrument can often be used
in the surgeon’s left hand. In younger patients in this age range, a 5-mm umbilical cannula may be possible. However, for most patients, a 10-mm umbilical
cannula is necessary to extract the gallbladder through this incision (middle). In a teenager, the port placement more closely resembles that seen in
adults (right).
positions, which helps the adjacent viscera fall away from
the surgical area. In addition, lateral retraction of the infundib-
ulum is important because the cystic duct is then positioned at
more of a 90-degree orientation to the common duct rather
than an oblique or even parallel orientation, which occurs
without such lateral retraction (Fig. 107-6). This parallel ori-
entation may lead to misidentification of the cystic and com-
mon bile ducts resulting in injury to the common duct.
The initial surgical maneuver is to expose the cystic duct
and identify the cystic artery. Adhesions between the duode-
num and stomach often require lysis for access to the infun-
dibulum and triangle of Calot. Once the cystic duct is
identified, two options are available. One is to proceed with
cholangiography as previously discussed. If the anatomy is
clear and there is no evidence of choledochal obstruction
on a preoperative US, we proceed with ligation and division
of the cystic duct. Usually 5-mm clips are adequate, although
a 10-mm clip may be required in larger patients. Two clips are
placed on the cystic duct approximately 5 mm from its inser-
tion into the common bile duct (Fig. 107-7). Another one or
two clips are placed on the cystic duct near the infundibulum
to prevent spillage of stones from the gallbladder. The cystic
duct is then divided, and the cystic artery is similarly ligated
and transected. Once these two structures are divided, the
gallbladder is detached in a retrograde manner using one of
several instruments: the hook cautery, spatula cautery, or en-
doscopic scissors attached to the cautery.
Before the gallbladder is completely separated from the
liver bed, the triangle of Calot should be carefully inspected
to ensure that all clips are secure and that there is no evidence
of bleeding. After complete detachment of the gallbladder, the
telescope is rotated from the umbilical port to the epigastric
port. If the epigastric port is 5 mm, then a 5-mm telescope
is used to visualize the gallbladder and a locking, grasping in-
strument is introduced through the umbilical cannula to
secure the gallbladder. The gallbladder is then extracted
through the umbilical port. In older patients, the gallbladder
may be too large for removal without further incising the um-
bilical fascia. This opening should be enlarged in patients with
large gallbladders to prevent rupture with spillage of stones
and bile during removal of the gallbladder. These gallstones,
if spilled, should be removed because complications from
retained stones have been reported in adults.15,100,101 After
the gallbladder is extracted, the area of dissection is again
 CHAPTER 107
FIGURE 107-6 During the initial stages of the laparoscopic cholecystec-
tomy, it is vitally important to retract the infundibulum directly lateral so
as to create a 90-degree angle between the cystic duct (solid arrow)
and common duct. Note the cystic artery (dotted arrow) cephalad to the
cystic duct.
carefully inspected to ensure adequate hemostasis. All irrigant
is evacuated, and bupivacaine is injected into the incisions for
postoperative analgesia. In smaller patients, the fascia sur-
rounding the 5-mm cannula sites should be closed. In larger
patients, fascial closure for the 5-mm ports or stab incisions is
usually unnecessary, but the fascia surrounding the 10-mm
cannula sites should be approximated carefully to prevent
herniation. For the stab incisions, skin closure is usually all
that is necessary. The patients are usually hospitalized after
the procedure and discharged the next morning.
Single-Site Umbilical Laparoscopic
Cholecystectomy
Single-site umbilical laparoscopic surgery (SSULS) is being
used more and more frequently for common surgical proce-
dures such as cholecystectomy and appendectomy. It has also
been used for splenectomy, pyloromyotomy, and ileocecec-
tomy for Crohn disease. When compared with traditional
FIGURE 107-7 After identification and isolation of the cystic duct, the
duct is doubly ligated with two clips proximal and two clips distal. The duct
will be divided between the two middle endoscopic clips.
GALLBLADDER DISEASE AND HEPATIC INFECTIONS 1347
three- and four-port/incision laparoscopic surgery, the only
advantage of SSULS appears to be cosmesis. SSULS has be-
come an attractive alternative between traditional laparo-
scopic surgery and natural orifice transluminal endoscopic
surgery (NOTES) because there are a number of real and po-
tential complications with access to the abdominal cavity with
NOTES. Similar to SSULS, the only advantage of NOTES over
traditional laparoscopic procedures is cosmesis.
For single-site laparoscopic cholecystectomy, an umbilical
incision of approximately 2 cm is necessary. For this particular
indication, our group uses either the SILS port (Covidien,
Norwalk, Conn.) or the Tri-Port (Olympus America, Center
Valley, Pa.). The SILS port is a foam port with three channels
through which the telescope and instruments are introduced.
A fourth instrument is then introduced along the side of the
foam port for retraction of the gallbladder (Fig. 107-8, A).
There is also a channel for insufflation. The TriStar port is
designed for three instruments, but a fourth 3-mm instrument
can be introduced through one of the insufflation portals
(Fig. 107-8, B). As with other SSULS procedures, it is helpful
to have a long telescope so that the telescope/camera holder
can stand away from the operating surgeon. Once the tele-
scope and instruments are introduced, the operation proceeds
A
B
FIGURE 107-8 For single-site laparoscopic cholecystectomy, an umbili-
cal incision of approximately 2 cm is necessary. Our group uses both the Cov-
idien SILS port (A) and the Olympus Tri-Port (B). The SILS port is a foam port
with three channels through which the telescope and instruments are
inserted. A fourth instrument (arrow) is then introduced along the side of
the SILS port for retraction of the gallbladder over the liver. The Tri-Port is
designed for single-site umbilical surgery as well. We insert a fourth 3-mm in-
strument through one of the insufflation ports (dotted arrow), if necessary.
1348 PART VII ABDOMEN
exactly like a traditional four-port laparoscopic cholecys-
tectomy. The instrument used to grasp the dome of the gall-
bladder is introduced at approximately 9 o’clock in the SILS
or TriStar port, and another instrument is used to retract
the infundibulum. With the SILS port, this instrument is in-
troduced along the side of the port in the umbilical incision.
The telescope should be angled at 30 to 45 degrees. As with
traditional laparoscopic cholecystectomy, it is important to re-
tract the infundibulum of the gallbladder laterally to create a
right-angle orientation of the cystic and common bile ducts.
Following identification and dissection of the cystic duct
and cystic artery, the duct is ligated with endoscopic clips
(Fig. 107-9, A). In a similar fashion, the cystic artery is divided
as well (Fig. 107-9, B). After division of the cystic duct and
cystic artery, the gallbladder is dissected free from its liver at-
tachment using electrocautery. A number of instruments can
A approach for cholecystectomy (Fig. 107-10).
B complicated group when compared with the elective patients.
FIGURE 107-9 Once the telescope and instruments are introduced, the
operation proceeds exactly like the traditional four-port laparoscopic chole-
cystectomy. It is important to retract the infundibulum of the gallbladder lat-
erally to create a right-angle orientation of the cystic and common bile ducts.
In the top photograph (A), the endoscopic clips are being applied to the cystic
duct. Note the right-angle orientation of the cystic and common bile ducts
(asterisk). Following ligation and division of the cystic duct (B), the cystic artery
(arrow) is well visualized and will be similarly ligated and divided.
FIGURE 107-10 This teenager underwent a single-site umbilical laparo-
scopic cholecystectomy and was discharged the following day. She has
achieved a nice cosmetic appearance at 1 month postoperatively. (She re-
moved her belly-button ring for this photograph.)
be used for this purpose including the hook cautery, spatula
cautery, or a Maryland dissector attached to cautery. After
complete mobilization of the gallbladder, it is then grasped
with an instrument placed through one of the channels in
the umbilical port and exteriorized through the umbilical in-
cision along with the port. Before extracting the gallbladder, it
is helpful to assess for bleeding and to irrigate/suction because
it would be necessary to reintroduce the umbilical port to rec-
reate the pneumoperitoneum to perform these functions after
removing the gallbladder.
The umbilical fascia is then closed with 0 absorbable suture
in an interrupted or continuous fashion. We prefer to close the
skin with interrupted 5-0 plain sutures, which is our usual
practice for umbilical skin closure for traditional laparoscopic
surgery. A nice cosmetic result is achieved using this SSULS
CHILDREN’S MERCY HOSPITAL EXPERIENCE
Despite the widespread use of laparoscopic cholecystectomy
in children, there are surprisingly few reports in the literature
describing a series as large as 100 patients.19,37,53,58,61,94 In a
1999 review of 100 patients undergoing laparoscopic chole-
cystectomy at Vanderbilt University Medical Center, one of
the authors (GWH) described 78 children who underwent
an elective operation and 22 who required an urgent opera-
tion for acute complications of cholelithiasis.19 In that series,
19 patients had hemolytic disease and two patients had biliary
dyskinesia. The six patients who presented with jaundice
underwent ERCP before the laparoscopic cholecystectomy.
Stones were extracted in two patients, and the other studies
were normal. In that report, two patients required laparo-
scopic choledochal exploration. The operating time and post-
operative hospitalization were significantly longer for the
No significant complications occurred such as the need for
reoperation, injury to the choledochus or other viscera, bile
leak, or retained choledocholithiasis.
Recently, our group reported a 6-year experience with
traditional four-port laparoscopic cholecystectomy at Chil-
dren’s Mercy Hospital (Table 107-2).53 Between September
2000 and June 2006, 224 patients underwent laparoscopic
 CHAPTER 107 GALLBLADDER DISEASE AND HEPATIC INFECTIONS 1349

TABLE 107-2 cholecystectomy. Operative time is the primary outcome var-

Patients Undergoing Laparoscopic Cholecystectomy at iable. Using an alpha of 0.05 and a power of 0.80, a total of
Children’s Mercy Hospital (September 2000 to June 2006) 60 patients will be enrolled. To date, 42 patients have parti-
Symptomatic gallstones 166 (29) Mean Age (Yr) 12.9 (0-21) cipated in this trial.
(Hemolytic Disease)
Biliary dyskinesia 35 Mean weight
(Kg)
58.3
(3-121)
Hepatic Infections
------------------------------------------------------------------------------------------------------------------------------------------------

Gallstone pancreatitis 7 Mean 77

operating time (30-285) Various hepatic infections occur in infancy and childhood
(min) (Table 105-3). The focus in this section of the chapter is on
Concomitant 6 Major 1 hepatic infections that often require surgical consultation
splenectomy complications for either diagnosis or treatment. This group includes hepatic
Calculous cholecystitis 5 abscess and nonviral causes of liver infection. In most pediat-
Miscellaneous 5 ric series, more than half of liver abscesses are found in
224 patients younger than 5 years of age and more than 25% occur
during the first year of life.102–104
From St. Peter SD, Keckler SJ, Nair A, et al: Laparoscopic cholecystectomy in the
pediatric population. J Laparoendosc Adv Surg Tech A 2008;18:127-130.
PATHOPHYSIOLOGY
cholecystectomy.53 The mean age was 12.9 years (range 0 to
21 years) with a mean weight of 58.3 kg (range 3 to 121 kg). Bacteria and other organisms may enter the liver through var-
One hundred and sixty-six children had symptomatic gall- ious routes, the most common being the biliary tract, the por-
stones, 35 children had biliary dyskinesia, seven patients tal vein, and the hepatic artery. Also, organisms may directly
presented with gallstone pancreatitis, six patients were under- invade the liver during penetrating trauma. In one report, a
going splenectomy and found to have gallstones, five patients liver abscess occurred in 4 of 16 patients with unruptured
had calculous cholecystitis, and one patient each had choledo- post-traumatic intrahepatic hematomas.105
cholithiasis, gallbladder polyp, acalculous cholecystitis, and The clinical signs and symptoms of liver abscess are usually
congenital cystic duct obstruction. In this series there were only nonspecific. The most common signs and symptoms are fever,
29 patients with hemolytic disease. Eighteen patients had sickle abdominal pain, and hepatomegaly. Abnormalities in leuko-
cell disease, and 11 had hereditary spherocytosis. The mean cyte count, liver enzyme levels, or bilirubin levels vary in de-
operative time (excluding patients undergoing a concomitant gree. Symptoms may last from days to months. The diagnosis
operation) was 77 minutes (range 30 to 285 minutes). of an hepatic abscess with imaging techniques has improved
Due to preoperative concerns about choledocholithiasis, a markedly with the availability of US and computed tomogra-
preoperative ERCP was performed in 17 patients. Stones were phy (CT) (Figs. 107-11 and 107-12).102,103
retrieved endoscopically in eight of these patients. Addition-
ally, an operative cholangiogram was performed in 38 patients
PYOGENIC LIVER ABSCESS
and common duct stones (CBD) were identified in 9 patients.
CBD stones were cleared intraoperatively in five patients, Pyogenic liver abscess (PLA) is uncommon in children, with a
while the other four patients required postoperative endos- reported incidence varying from 3 to 25 per 100,000 pediatric
copy and sphincterotomy to retrieve the stones. Because of hospital admissions.106 Its incidence has decreased since
a postoperative rise in direct bilirubin, two patients who the introduction of antibiotics. In contrast to adults, in whom
did not have an intraoperative cholangiogram underwent a
postoperative ERCP. The ERCP was normal in both cases.
There were no conversions, ductal injuries, bile leaks, or
mortality. However, one sickle cell patient developed a postop-
erative hemorrhage, which required laparotomy for control. In-
terestingly, biliary dyskinesia was diagnosed in only 10% of the
first 30 patients in this series but was diagnosed in 40% of
the last 30 patients studied. The mean ejection fraction in these
patients was 21%. All of these patients had improvement in
their symptoms following the laparoscopic cholecystectomy.
Over the past 10 years, laparoscopic cholecystectomy has
become the accepted standard for removal of the gallbladder
in children and adolescents. Reported complications have
been rare. There are no reports of choledochal injury in
children, although some have undoubtedly occurred. As
additional experience is gained, pediatric surgeons may use
intraoperative US rather than cholangiography to assess the
correct anatomy and the presence of choledocholithiasis.
SSULS is also being used for gallbladder removal at our
institution. A prospective, randomized trial is under way FIGURE 107-11 Computed tomography scan demonstrating several
at our hospital comparing single-site umbilical laparoscopic hepatic abscesses (defects) in a child with staphylococcal infection related
cholecystectomy with traditional four-port laparoscopic to chronic granulomatous disease.
1350 PART VII ABDOMEN
FIGURE 107-12 Sonogram demonstrating several hepatic abscesses
(echolucent areas) in the same patient as in Figure 107-13.
biliary tract disease is the primary cause of hepatic abscess,
pyogenic liver abscess in children can be caused by a number
of different conditions. Historically, the most frequent cause of
hepatic abscess in children was perforated appendicitis. The
development of hepatic infection in this setting was presumed
to be due to pylephlebitis of the mesenteric and portal veins
with ascending migration of bacteria into the liver. An intra-
abdominal source of bacteria is still a common etiology of
PLA, but liver abscesses are also seen in children with chronic
malnutrition, granulocyte dysfunction, sickle cell disease, and
with congenital or acquired immunosuppression.
Chronic granulomatous disease of childhood is one of the
more common immunodeficiencies. Also, acquired impair-
ment in host resistance, such as that caused by chemotherapy
or transplantation immunosuppression, can predispose pa-
tients to hepatic abscess formation. In immunocompetent
children, causes for PLA include an intestinal source (i.e.,
appendicitis, Crohn disease), chronic cholangitis with intra-
hepatic abscess formation in patients with biliary atresia or
choledochal cysts, umbilical vein catheterization (neonates),
and systemic bacteremia of any cause.107 Abdominal trauma,
either penetrating or blunt, with liver injury is also associated
with an increased incidence of hepatic abscess.108,109
Abscesses in such cases are presumed to be secondary to
bacterial seeding of the devitalized liver parenchyma and
are often associated with a hematoma. It must be remembered,
however, that liver abscesses do occur in otherwise normal
children without any obvious predisposing conditions.110,111
In 20% of these patients, the infectious source of the hepatic
abscess is never found.112
The diagnosis of a PLA is often dependent on its being con-
sidered as part of the differential diagnosis. The low incidence
of liver abscess in childhood leads to a low index of suspicion
and sometimes delays the diagnosis for weeks to months. The
overall mean age of presentation for PLA is reported to be 7.5
years with children from developing countries presenting at
younger ages and children from developed countries present-
ing at older ages.113 The classic signs and symptoms of fever,
abdominal pain, jaundice, and liver enlargement occur infre-
quently, which is why the diagnosis is often missed. These
symptoms can develop slowly and, if associated with weight
loss, can lead initially to the suspicion of cancer. Usually
the presenting symptoms are nonspecific, with the major
symptom being abdominal pain.106,110 The duration of symp-
toms is highly variable, but usually patients present between
1 week and 1 month after they become ill. Most patients are
febrile, but an enlarged liver is rarely found on physical exam-
ination. Right upper quadrant tenderness is only found in
about 40% of patients.107 Hematologic studies are usually
not helpful because only mild leukocytosis is seen. Elevated
liver function studies are found in only one half of the patients.
Inflammatory markers, however, are frequently elevated in
patients with a PLA.113
When the bacteriology of pyogenic abscesses is examined
in children, Staphylococcus aureus is the primary infecting
agent in 37% to 50% of the children.106,110,112,114 Most
hepatic abscesses, however, tend to be polymicrobial with
gram-positive cocci and gram-negative rods including Escher-
ichia coli, Klebsiella, Enterobacter, and Pseudomonas. Recently,
anaerobes have also been cultured from these abscesses in
at least 20% of the cases.114–116 Blood cultures are positive
in about 50% of patients. However, blood cultures often grow
fewer bacterial types than are subsequently recovered from di-
rect cultures of the abscess. As a cautionary note, in a third of
the blood cultures, the organisms that were recovered from
the blood were not subsequently recovered from the cultures
of the liver abscess. Unfortunately, whereas Gram stains of a
direct aspirate from a liver abscess are highly accurate for
gram-positive cocci, they have only a 50% sensitivity for
gram-negative organisms.117 Fungal abscesses have been
found with an increasing incidence in children who become
febrile and neutropenic while being treated for acute leuke-
mia.118 One prospective series reported a 29% incidence of
fungal hepatic and/or splenic abscesses in this population.119
A specific ultrasonographic appearance for hepatic and
splenic candidiasis has been described.120
Neonates comprise another special subpopulation of pa-
tients. In the perinatal period, direct transplacental bacterial
seeding of the liver can occur. This seeding may be associated
with premature rupture of membranes. Omphalitis, with or
without catheterization of the umbilical vein, has been rarely
associated with bacterial hepatitis or abscess. Most of the time
these abscesses are miliary. There is a disproportionately high
representation of gram-negative organisms in the neonatal
hepatic abscess. Because they are usually miliary, they are
not amenable to percutaneous drainage. Therefore high-dose,
long-term antibiotic treatment is necessary.
In those neonates who presented with a solitary hepatic
abscess, the offending organism is usually S. aureus. These pa-
tients present with an enlarging right upper quadrant mass
that is often confused with a tumor. Treatment of these solitary
abscesses has often been resection because of the difficulty in
distinguishing between an abscess and a solid tumor.121
RADIOLOGIC EVALUATION
The radiologic studies that are most helpful in the diagnosis of
PLA are abdominal US and CT. US is usually preferred as the
initial screening tool because of the lack of need for sedation
and the absence of ionizing radiation. CT, however, is more
sensitive and can discover lesions that are missed by US.
In addition, CT is usually the preferred radiologic modality
for localization of the abscess for percutaneous drainage.106
 CHAPTER 107
On radiologic evaluation, PLAs can present as either a solitary
lesion or as multiple lesions. There is also a predisposition for
the lesions to be in the right lobe of the liver.122
TREATMENT
The initial treatment of a PLA should be with broad-spectrum
antibiotics with coverage for gram-positive, gram-negative,
and anaerobic organisms. As part of the gram-negative spec-
trum, Pseudomonas coverage must be included. The preferred
therapeutic and diagnostic intervention is percutaneous
drainage/aspiration of the abscess under either CT or US guid-
ance. After the abscess has been drained and specific bacteria
have been recovered, antibiotic therapy can then be tailored to
cover these organisms. The recommended duration of intrave-
nous antibiotic therapy is 4 to 6 weeks, followed by oral an-
tibiotics for an additional 4 to 8 weeks.104 The complication
rate of percutaneous drainage has been reported to be about
4%, with a failure rate of 15%.123 However, if the patient does
not respond to percutaneous drainage or if percutaneous
drainage is not possible, then an operative approach is war-
ranted.110,124 One study reported the laparoscopic approach
for drainage of a pyogenic liver abscess.125 With an aggressive
approach to treatment of pyogenic liver abscess, the mortality
of this condition has dropped from 36% in the 1970s to 15%
in the late 1980s and 10% now.
One complication of PLA that has been reported in the
adult population is endophthalmitis. In one series, there
was a 3.5% incidence of this complication, which can lead
to a variety of devastating visual problems, ranging from
decreased visual acuity to complete blindness.126 The best re-
sults for recovery were associated with initiation of treatment
within 24 hours of any visual symptoms. Therefore a child
who appears to be having any visual difficulties with a pyo-
genic abscess should undergo an ophthalmologic evaluation.
CAT-SCRATCH DISEASE
Cat-scratch disease is a subacute illness preceded by exposure
to an infected cat. It is characterized by regional lymphadenitis
but may involve other organ systems including the liver, brain,
and bone. This disorder was first described as oculoglandular
syndrome in 1889, and the association with a scratch by a cat
was reported in 1950 by Debre.127 Since then, various infec-
tious organisms have been postulated as the etiologic agent.
Cat-scratch disease is now known to be caused by Bartonella
henselae and is seen throughout North America with most
cases occurring in late summer to fall. The disease primarily
affects children and adolescents (75% to 80% of cases). A pri-
mary lesion develops at the site of inoculation in most cases,
usually within 1 to 2 weeks of inoculation. This lesion is
typically a single papule 2 to 5 mm in diameter. Regional
lymphadenopathy develops approximately 3 weeks later
and frequently involves axillary, inguinal, and cervical nodes.
The primary skin lesion has often resolved by the time the
adenopathy occurs. Other associated symptoms are fever,
anorexia, malaise, fatigue, and headache. Traditionally the
diagnosis has been confirmed by acute/convalescent IgG
titers. Recently, polymerase chain reaction (PCR) of tissue,
lymph node, or peripheral blood has been found to be both
sensitive and specific for B. henselae, as well as having a rapid
turn-around time.128
GALLBLADDER DISEASE AND HEPATIC INFECTIONS 1351
It is estimated that 5% to 10% of patients present with sig-
nificant systemic symptoms or atypical manifestations of the
disease. In these patients, the typical regional adenopathy
may be absent.129 Patients who are immunosuppressed are
also at risk for cat-scratch disease. There have been several
recent reports documenting cat-scratch disease causing liver
masses in patients after liver transplantation.130–132
Liver involvement is the third most common clinical man-
ifestation of cat-scratch disease after fever and lymphadenop-
athy.133 It is usually diagnosed by abdominal CT, which is
performed as part of the evaluation for fever of unknown
origin. This test typically shows multiple lesions in both lobes
of the liver and, sometimes, associated splenic involve-
ment.127,134 The diagnosis is confirmed by liver biopsy, either
open or guided by laparoscopy. The histologic features consist
of a granulomatous reaction with necrosis. Cultures of the
lesions are usually negative for bacteria and fungi. Treatment
of cat-scratch disease remains controversial because most
untreated cases resolve without sequelae. Patients with
marked symptoms or visceral involvement are usually treated
with antibiotics (e.g., gentamicin, ciprofloxacin, rifampin, or
azithromycin).135 Currently azithromycin is the agent of
choice for uncomplicated disease. For hepatic involvement,
a multidrug regimen (gentamicin and rifampin) has been used
for 2 to 3 weeks.136
PERIHEPATITIS
In the pediatric population, bacterial liver infection with either
Neisseria gonorrhoeae or Chlamydia trachomatis tends to occur
most frequently in teenage girls. The initial pelvic infection
travels cephalad in the abdomen, resulting in the perihepatitis.
If the perihepatitis is untreated, then the classic “violin string”
adhesions develop between the liver and the peritoneal cavity
that are associated with the Fitz-Hugh-Curtis syndrome.
In addition to the typical symptoms of pelvic inflamma-
tory disease including dysmenorrhea, vaginal discharge,
and lower abdominal pain, these patients also present with
severe pleuritic right upper abdominal pain that often radi-
ates to the right shoulder. It is not unusual for the patient to
complain of left upper quadrant pain as well. In these pa-
tients, the liver is often tender to palpation and a rub may
be audible near the costal margin. Pelvic findings of acute sal-
pingitis are usually present. Laboratory findings include leu-
kocytosis, elevated aminotransferase levels, and occasional
hyperbilirubinemia.
Perihepatitis can mimic a variety of other conditions in the
upper abdomen including cholelithiasis, hepatitis, a perfo-
rated peptic ulcer, appendicitis, or pancreatitis. In female ad-
olescents with a history of pelvic inflammatory disease,
perihepatitis must be part of the differential diagnosis when
these patients present with right upper quadrant pain.
Recently, dynamic abdominal CT scanning including the arte-
rial phase has been reported as an important tool in the diag-
nosis of this syndrome.137 If the arterial phase of the scan
shows hepatic capsular enhancement (secondary to increased
blood flow in the inflamed hepatic capsule), then the diagno-
sis of Fitz-Hugh-Curtis syndrome can be made without
further invasive studies.
In patients in whom the diagnosis is uncertain or in
patients who have persistent symptoms, diagnostic laparos-
copy can allow direct visualization of the perihepatic space.
1352 PART VII ABDOMEN
If found, the classic adhesions can be lysed, which often
results in a significant improvement in pain.138
Recent reports have implicated C. trachomatis, in addition
to N. gonorrhoeae, as a cause of this syndrome.137,139 Therefore
a treatment regimen that is effective against both chlamydial
infection and gonococcus is necessary. This would include
tetracycline or doxycycline or erythromycin for patients in
whom tetracycline is contraindicated.140
AMEBIC ABSCESS
The incidence of amebic liver abscess in the United States
seems to have gradually increased. Hepatic involvement by
Entamoeba histolytica is primarily a disease of adults but does
occur in younger patients, most commonly in early child-
hood.141 Poor sanitation and crowded conditions help foster
the development and spread of amebiasis. In patients with
amebic colitis, trophozoites from colon ulcers travel through
the portal vein to the liver, where they initiate a necrotizing
process with subsequent abscess formation. In the adult pop-
ulation, liver involvement with an amebic abscess occurs in
10% to 50% of adults with a 90% male predominance. In chil-
dren, however, liver involvement occurs in only 1% to 7% of
the patients with an equal distribution between males and
females.142–144 Corticosteroids and immunosuppression
seems to predispose patients to amebic abscess formation.
More than 80% of amebic liver abscesses occur in the right
lobe of the liver. Symptoms include fever, chills, nausea,
weight loss, and abdominal distention. On physical examina-
tion, the children tend to have hepatomegaly, but abdominal
pain and liver tenderness are relatively infrequent. Breath
sounds are often decreased in the lower right chest, and
right-sided pleural effusions are common. Clinical jaundice
is uncommon. A history of exposure or travel to endemic areas
(Mexico, South America, or Southeast Asia) is often a better
indicator of amebiasis than are the imaging studies.145 In
the United States the typical patient is a child of immigrants
who come from an area that is endemic for amebiasis but
are not necessarily living in poverty.146
Laboratory findings include leukocytosis without eosino-
philia, mild anemia, and normal or mildly abnormal liver
function studies. Abdominal US and CT are useful for defining
the location and number of lesions. The characteristic US
appearance is a hypoechoic, thin-walled abscess that some-
times contains air. CT shows a nonspecific picture of a well-
circumscribed, low-density lesion that is usually solitary
and has predominance for the right lobe of the liver. Also, with
modern imaging techniques, it is now clear that multiple ame-
bic abscesses can be found within the liver.147
The diagnosis of an amebic liver abscess relies on the iden-
tification of a space-occupying lesion in the liver and the pres-
ence of positive amebic serology. Amebic serology (indirect
hemoglobin assay or agar gel immunodiffusion precipitant)
has been found to be highly sensitive (>94%) and highly spe-
cific (>95%) for the diagnosis of amebic liver abscess.148 PCR
analysis for E. histolytica and E. dispar has become available
and may prove to be more sensitive and specific than the
previous assays.149 If it is not possible to distinguish clinically
between a pyogenic versus an amebic liver abscess, if the
abscess does not respond to appropriate medical treatment,
or if serologic tests for ameba are inconclusive, then aspiration
of the abscess is indicated for culture.
The current treatment for an amebic abscess is oral metro-
nidazole (30 to 50 mg/kg/day divided 3 times a day for 7 to 10
days). Most patients will respond to this therapy with resolu-
tion of their fever and abdominal pain. Currently there is no
indication for aspiration of an uncomplicated amebic abscess
because the majority of these will resolve with medical manage-
ment alone.147,150–152 If the patient does not respond to med-
ical management within 4 days, if the liver abscess is greater
than 5 cm in diameter, if it is located in the left lobe of the liver,
or if it is believed to be in danger of rupture, then aspiration and
percutaneous drainage is advisable. In addition, percutaneous
catheter drainage is also indicated in patients who have had
rupture of an amebic abscess into the right pleural space. Total
resolution of the abscess may take months to years and can be
followed by US.
The most common complication of an amebic liver abscess
is erosion through the diaphragm and rupture into the right
chest. The rupture rate of a right lobe amebic liver abscess
is relatively low, but the rupture rate of a left lobe abscess
has been reported to be as high as 22%.153 Intraperitoneal
rupture of an amebic abscess carries a 20% to 75% mortality,
and the patients who are at the highest risk are those who
present with acute peritonitis.154,155 Intraperitoneal rupture
results in a clinical picture of peritonitis and shock. If this de-
velops, fluid resuscitation, antibiotics, antiamebic drugs, and
surgical exploration are necessary. Erosion into the pericardial
sac is less frequent but may cause cardiac tamponade. Erosion
into intra-abdominal organs including the stomach has also
been reported.146,156
There have been attempts at developing a vaccine for
E. histolytica in an attempt to eradicate amebiasis. However,
unfortunately in some cases, a natural infection with E. histo-
lytica does not seem to result in long-term immunity. In fact,
recurrent amebic liver abscesses are a well-known problem in
areas where E. histolytica is endemic.157–159
ECHINOCOCCAL CYST
Hydatid disease is generally caused by infection with the larval
stage of the dog tapeworm Echinococcus granulosis. The
ingested ova burrow through the intestinal mucosa and travel
to the liver through the mesenteric veins. A few ova may
bypass the liver and be trapped in the lung. The reported peak
incidence of echinococcal infection in children is between
5 and 15 years of age.160,161 The liver is involved in about
60% of the cases, the lung in about 20%, and other organs
(kidney, brain, bone, or muscle) in about 20% of the cases.
A solitary cyst will develop in one organ about 80% of the
time.162 The host organ separates itself from the parasite by
forming a pericyst, which is a capsule of native connective tis-
sue. For most surgical procedures, the pericyst is left behind
and the intraluminal contents of the cyst are removed.
The usual presenting symptoms include right upper quad-
rant pain, a mass, and right pleuritic type chest pain in about
one half of the patients. Interestingly, about a fourth of the pa-
tients are asymptomatic at presentation. Untreated cysts tend
to grow until they reach the surface of the liver, where they can
then spontaneously rupture, resulting in dissemination of the
parasite into the peritoneal cavity and possible anaphylaxis.163
In one fourth of the cases, the cyst can become secondarily
infected with bacteria.147 Echinococcal liver abscess is best di-
agnosed by serologic testing. The indirect hemagglutination
 CHAPTER 107 GALLBLADDER DISEASE AND HEPATIC INFECTIONS 1353

assay is the most specific test, but it lacks sensitivity for echi-
nococcal disease. Results are considered positive for tissue in-
vasion at a titer of 1:512, suspicious at a titer of 1:128, and
negative at a titer of 1:32. Additional tests include an en-
zyme-linked immunosorbent assay (ELISA), which has been
reported to be positive in 96% of patients, or a compliment
fixation test, which has an 89% positivity rate.162 Sunita
and colleagues have also reported excellent results using urine
and saliva for ELISA antibody testing. The combination of two
positive serologic tests provides a laboratory diagnosis in more
than 94% of patients with echinococcal disease.163a
Diagnostic aspiration is indicated if there is concern that
the cyst could be a pyogenic or amebic abscess, but it is not
indicated for echinococcal disease. Abdominal US or CT is
the diagnostic imaging modality of choice.164 A plain film
of the abdomen or abdominal CT may show calcification of
the rim of an echinococcal cyst or daughter cysts that is not
found in amebic liver abscess (Fig. 107-13). FIGURE 107-13 Computed tomography scan demonstrating several
The definitive treatment of hepatic hydatid disease is surgi- hepatic abscesses with one (arrow) containing daughter cysts charac-
teristic of hydatid disease.
cal. Medical treatment of hydatid cyst includes the second-
generation anthelmintic albendazole. Albendazole has been
shown to be most effective in decreasing recurrent disease when capitonnage), placement of an omental flap over the residual
it is given both preoperatively and postoperatively.161,166,167 cavity (cystostomy with omentoplasty), or total excision of
Successful operative management starts with isolation of the entire cyst (cystectomy).167,168 When cystostomy is
the liver from the rest of the abdomen, usually with sponges performed, the cyst wall needs to be carefully inspected for
soaked in hypertonic saline or a povidone-iodine (Betadine) any open biliary radicals. If found, they are individually
solution to prevent spillage of the cyst material. The cyst fluid sutured closed.161 Recurrence rates after surgical excision of
is then aspirated. The cyst is then opened 2 to 3 cm and the hepatic hydatid cysts have been reported to range from
remaining contents including the laminated membrane are 1.1% to 9.6%.169 When surgery and albendazole treatment
removed. The cavity is then flushed with 20% saline; however, are combined, the hepatic recurrence rate can be lowered
the effectiveness of this step has been questioned.168 The cav- to 2%.161,165–168
ity then needs to be obliterated using one of several different
techniques: closed suction drainage (cystostomy with drain- The complete reference list is available online at www.
age), obliteration with absorbable sutures (cystostomy with expertconsult.com.

CHAPTER 108
Portal
Hypertension
Riccardo Superina
A man who thinks that a science can perform what is outside
its province, or that nature can accomplish unnatural
things, is guilty of ignorance more akin to madness than to
lack of learning. Our practice is limited by the instruments
made available by Nature or by Art. When a man is attacked
by a disease more powerful than the instruments of
medicine, it must not be expected that medicine should
prove victorious.
—Hippocrates
The treatment of portal hypertension in children is always
evolving as the nature of underlying disorders is better under-
stood. Treatment is often successful, Hippocrates notwith-
standing, and the strategy usually involves medical, surgical,
or combined therapy. This improved outlook is a departure
from past years because of the improved understanding of
the underlying pathophysiology. More effective imaging tech-
niques and a wider array of options in pharmacologic manage-
ment have resulted in opportunities to apply both established
and newer surgical techniques precisely and deliberately to
address underlying disease processes that cause portal
hypertension.
Historical Background
------------------------------------------------------------------------------------------------------------------------------------------------
The Italian physician and pathologist Guido Banti first de-
scribed the relationship among splenic enlargement, anemia,
and cirrhosis of the liver. The first splenectomy done for
hypersplenism was done in Florence in 1903 on his advice.
The constellation of symptoms later became known as Banti
syndrome.1
Earlier, Nikolai Eck, a German anatomist, had produced a
connection between the portal vein and the inferior vena cava
in a dog.2 Later studies described the association between the
direct communication of portal and systemic venous blood
and the onset of encephalopathy associated with hyperammo-
nemia and protein ingestion.3–5
The first applications of portosystemic shunting in patients
were reported in the 1950s and 1960s. More selective shunt-
ing procedures were devised when it was discovered that
diversion of mesenteric blood flow into the systemic circu-
lation could precipitate severe encephalopathy. Nonshunt
procedures aimed at interrupting the blood flow to gastric
and esophageal varices were also described.
Finally, with the advent of liver transplantation, portal
hypertension caused by advanced liver disease was treated
by replacement of the diseased organ, rather than palliative
procedures that produced significant morbidity and often
hastened the demise of the patient.
Embryology and Anatomy
------------------------------------------------------------------------------------------------------------------------------------------------
The portal vein develops through a complex system involving
the formation and involution of primitive vessels, followed by
circulatory changes at birth that affect the flow of blood in the
mature portal circulation (Fig. 108-1).
The left and right omphalomesenteric veins drain the de-
veloping gut as it is formed from the yolk sac during the fourth
to sixth weeks of embryonic life. At the same time, the umbil-
ical veins transport blood from the placenta to the embryo.
Blood from both the umbilical and the omphalomesenteric
veins drains through the nascent hepatic sinusoids in the
developing liver and into the hepatic veins back to the devel-
oping heart. Some of the blood from the umbilical veins
bypasses the liver and drains into the sinus venosus through
the ductus venosus. Although the omphalomesenteric veins
communicate by a series of intermediary vessels early in
embryonic life, the right one involutes and disappears by
the sixth week of gestation. The left one develops into the
permanent portal vein, which drains the mesenteric venous
bed through the superior mesenteric vein and its branches.
The right umbilical vein also involutes, and the left remains
patent until shortly after birth, when it thromboses and
becomes the ligamentum teres.
The portal vein is formed by the confluence of the splenic
and superior mesenteric veins posterior to the head of the
pancreas. The coronary vein draining the gastric venous bed
inserts into the portal vein at or just distal to the splenomesen-
teric confluence, and the inferior mesenteric vein drains into
the splenic vein anywhere along its length, also behind the
pancreas.
1355
1356 PART VII ABDOMEN
Venae revehentes
Anterior detached
portions of umbilical
veins
Stomach
Venae
advehentes
Bileduct
Right umbilical
vein
Portal vein
Obliterated portions
Vitelline veina
of venous rings
FIGURE 108-1 The portal vein forms from the union of the two vitelline
veins draining blood from the yolk sac. The left vitelline vein joins with the
splenic vein to form the extrahepatic portion of the portal vein. The intra-
hepatic portal vein forms from the umbilical veins. In addition, the left
umbilical vein communcates with the sinus venosus, allowing placental
blood to bypass the liver during intrauterine life. The right vitelline and
umbilical veins involute.
Despite the relative variability in the anatomy of the liver’s
arterial supply, the portal vein anatomy varies little. The prin-
cipal extrahepatic vein divides into a left and a right branch.
The left branch supplies the left lobe as it courses under
the surface of segment IV of the liver; turns anteriorly in the
recessus of Rex among segments II, III, and IV; and ends by
dividing into the branches that supply those segments. The
right branch divides into the posterior and anterior sectoral
branches at or just before the liver plate at the capsule.
A cavernous venous malformation may be the result of a
disordered sequence of biologic steps. Thrombus in the
postnatal umbilical vein may also propagate into the hepatic
portal vein and occlude flow in the extrahepatic portion of
the vein. Both processes may lead to the syndrome of extra-
hepatic or prehepatic portal hypertension.
Definition
------------------------------------------------------------------------------------------------------------------------------------------------
Portal hypertension occurs whenever the resistance to the flow
of blood from the mesenteric venous circulation through or to
the liver increases. The causes of portal hypertension are most
commonly obstructive in nature. That is, there is an impedi-
ment to the forward flow of blood from the omphalomesen-
teric vascular bed through the liver and toward the heart.
However, portal hypertension can also be caused by an
increase in the volume of blood going to the liver or by a direct
communication between the arterial and venous supplies to
the liver that exposes the venous bed to abnormally high
pressures.
Portal hypertension may be defined as pressure in the
portal venous bed that exceeds 5 to 8 cm H2O or a pressure
gradient of more than 5 cm H2O between the hepatic veins
and the portal circulation. When the pressure in the portal
circulation exceeds those values, a series of physiologic
changes leads to the symptoms common to all forms of portal
hypertension, regardless of the cause.
Ductus
venosus Collateral Circulation
------------------------------------------------------------------------------------------------------------------------------------------------
Liver
As a consequence of the resistance to flow and the increase in
pressure in the mesenteric venous circulation that includes the
Pancreas superior mesenteric, inferior mesenteric, splenic, and coro-
nary veins, blood that normally goes to the liver from the por-
Left umbilical tal vein must find alternative or additional routes back to the
vein heart. The abnormal communications that form between the
mesenteric and systemic venous systems are called varices
Duodenum or “shunts” because blood is shunted away from the liver
and back to the heart. The hemorrhoidal plexus in the rectum,
the paraumbilical network between the portal vein and the
paraumbilical veins that form through the recanalization of
A. Boutwell
the umbilical vein and ligamentum teres (forming the “caput
medusa” around the umbilicus), and the communications in
the gastroesophageal area between the coronary and splenic
veins through the esophageal and paraesophageal veins back
to the hemiazygous system are the typical areas where the ab-
normal shunts between the systemic and mesenteric venous
systems occur.
Other locations of spontaneous shunts are between the
veins of the colon and duodenum and the left renal vein;
the accessory portal system of Sappey, whose branches in
the round ligament unite with the epigastric and internal
mammary veins through the diaphragmatic veins to unite with
the azygos vein; the veins of Retzius, which connect the intes-
tinal veins with the inferior vena cava and its retroperitoneal
branches; the inferior mesenteric veins and the hemorrhoidal
veins that open into the hypogastrics; and, rarely, the patent
ductus venosus, affording a direct connection between the
portal vein and the inferior vena cava.
Causes: A Spectrum of Disorders
------------------------------------------------------------------------------------------------------------------------------------------------
Portal hypertension can be divided into two categories:
(1) portal hypertension from hepatocellular injury and liver
fibrosis and (2) portal hypertension from primary vascular
causes (Table 108-1).
HEPATOCELLULAR INJURY
Table 108-2 illustrates some of the common liver disorders in
children that can lead to portal hypertension. Hepatocellular
injury from a variety of toxins leads to cell death, initiation and
progression of fibrosis, and ultimately cirrhosis. Progressive
injury to hepatic tissue and replacement of normal liver with
fibrous tissue lead to increased resistance to blood flow
through the liver. The process of deposition of collagen in re-
sponse to hepatocyte injury appears to be a complex process
mediated through the transcription of proinflammatory cyto-
kines such as osteopontin and through stellate cells.6 The stel-
late cell, a normal constituent of hepatic sinusoids, is the
primary source of excess extracellular matrix proteins in liver
fibrosis.7 It can change its phenotype from one that is fairly
quiescent during normal liver homeostasis to one that

TABLE 108-1
Classification of Portal Hypertension
Type Description
Hepatocellular Intrinsic liver disease with increased liver fibrosis
(see Table 108-2)
Vascular
Prehepatic Extrahepatic portal vein thrombosis
Cavernous transformation of portal vein
Extrinsic compression of portal vein
Posthepatic Budd-Chiari syndrome
Intrinsic web
Stenosis of hepatic vein orifice
High-flow Arteriovenous communication—intrahepatic or
(hyperkinetic) extrahepatic
Congenital
Acquired
TABLE 108-2
Hepatocellular Diseases Leading to Portal Hypertension
Biliary atresia
Postinfectious cirrhosis
Congenital hepatic fibrosis
Congenital disorders of bile acid metabolism
Sclerosing cholangitis
Autoimmune hepatitis
Drug toxicity
Metabolic diseases (e.g., alpha-1 antitrypsin deficiency)
undergoes myofibroblastic differentiation or activation during
episodes of hepatocyte injury. Dysregulation in the activity
and number of stellate cells through faulty apoptosis or per-
petuation of the insult that led to hepatocyte injury leads to
ongoing deposition of proteins in the extracellular compart-
ment of the liver. Ultimately, excess deposition of collagen
in sinusoids results in portal hypertension.
VASCULAR CAUSES
Prehepatic Portal Hypertension
Prehepatic portal hypertension is caused by obstruction at the
level of the portal vein proximal to its branching at the liver
plate. Extrahepatic portal vein thrombosis has been recog-
nized and treated in both adults and children; it may be a con-
genital lesion or acquired at a later time. Even though the
symptoms may be dramatic, patients, particularly children, tol-
erate the bleeding from varices well because of the well-
preserved hepatic function.8 Occlusion of the main trunk of
the portal vein may lead to recanalization of the vein and its
transformation into a series of smaller collateral veins that as-
sume the appearance of a venous cavernoma visible on ultraso-
nography or CT scanning (Fig. 108-2). In most instances the
intrahepatic architecture of the portal vein is preserved.
Thrombosis of the portal vein is associated with omphalitis, in-
strumentation and cannulation of the umbilical vein at birth in-
cluding umbilical vein catheters for intravenous access,9,10
sepsis and dehydration in infancy, and mass lesions that exert
CHAPTER 108 PORTAL HYPERTENSION 1357
S
FIGURE 108-2 Preoperative computed tomography scan of a child with
extrahepatic portal vein thrombosis showing the typical cavernoma, the
tangle of varices in the hilum of the liver (white arrow) that replaces the
normal portal vein. Also of note is the typically small size of the intrahepatic
portal vein branches (black arrow) and the large spleen (S) that appears
larger than the liver.
extrinsic compression on the vein. These lesions may be inflam-
matory or malignant; examples include reactive enlargement of
the lymph nodes in the hilum of the liver from histoplasmosis
or compression from non-Hodgkin lymphoma involving the
hilum of the liver.
Unlike children with preexisting liver disease, children
with extrahepatic portal vein thrombosis are usually com-
pletely well before the sudden onset of symptoms. More than
50% of children present with unheralded hematemesis or
hematochezia from gastroesophageal varices.11 Others are di-
agnosed with splenomegaly of unknown cause and may be re-
ferred to an oncologist for suspected hematologic malignancy.
Posthepatic Portal Hypertension
Posthepatic portal hypertension is caused by occlusion of
large or small veins draining blood from the liver into the in-
ferior vena cava. Obstruction at this level leads to passive con-
gestion of the liver and necrosis of the hepatocytes in the
central areas of the hepatic lobule. Acute venous obstruction
may lead to massive hepatic necrosis and cell death unless it is
relieved in an urgent fashion. The Budd-Chiari syndrome,
sometimes referred to as Chiari disease, was first described
in 1845.12,13 It is uncommon in children.14 Outflow occlu-
sion is caused by congenital venous webs in the hepatic veins,
hydatid disease,15 myeloproliferative diseases, hypercoagu-
lable states, and increased estrogen levels, most commonly
from oral contraceptives. It can also occur after liver transplan-
tation (Fig. 108-3).
Microvascular nonthrombotic occlusion of hepatic venules
is termed venoocclusive disease, and it is being seen with in-
creasing frequency in patients following bone marrow trans-
plantation.14,16 Cytoreductive regimens involving busulfan
and cyclophosphamide are thought to induce obliterative
phlebitis in the small veins of the liver, and damage to the en-
dothelial cells is induced through the depletion of endogenous
1358 PART VII ABDOMEN
FIGURE 108-3 Acute hepatic venous obstruction after split liver trans-
plantation, caused by a blood clot in the left hepatic vein of a segment
II-III transplant that resulted in hepatic necrosis. The child was
retransplanted.
glutathione S transferase stores within the cells.17 Other
causes of venoocclusive disease such as ingestion of herbal teas
containing pyrrolizidine alkaloids have been described. Other
diverse agents including contrast media, estrogen, and thiogua-
nine have also been associated with the development of venooc-
clusive disease. This disease has a high mortality rate, and
treatment is limited to withdrawal of the offending agents,
if possible, coupled with implementation of thrombolytic ther-
apy and transjugular intrahepatic portosystemic shunting if
feasible.18–20
High-Flow States
A congenital or acquired communication between an artery
and vein in the mesenteric circulation is a rare but known
cause of portal hypertension (Fig. 108-4). Portal hypertension
from such abnormal communications between arteries and
veins in the portal circulation is termed hyperkinetic21 because
of the hyperdynamic circulation responsible for the increased
portal pressure. Patients have well-preserved liver function
and present with bleeding from gastrointestinal (GI) varices,
ascites, or splenomegaly. The site of the communication be-
tween the artery and vein is quite varied. The splenic and gas-
tric arteries have been reported as sites of these fistulas, as well
as the more common site of the hepatic artery within the pa-
renchyma of the liver. Arteriovenous fistulas causing portal
hypertension have been reported in all age groups.22–24 Most
often they are congenital, but some are thought to be acquired
through a pathologic process or iatrogenically.25,26 Patients
may present with varied symptoms that depend on the site
of the arteriovenous communication. Bleeding from varices
may be severe, and because of the exposure to arterial pres-
sures, the usual therapeutic maneuvers may be inadequate
to control it.27
Treatment of high-output fistulas requires interruption of
the abnormal communication between the artery and vein. Al-
though surgical procedures including hemihepatectomy22
and surgical ligation of extrahepatic fistulas27,28 have been de-
scribed as a means of removing the fistula, access by radiologic
FIGURE 108-4 Contrast flow directly from the smaller hepatic artery (A)
into the larger portal vein (P) without traversing the parenchyma of the
liver. This child had an arteriovenous fistula of unknown etiology resulting
in esophageal and gastric varices. The fistula was treated by embolizing
the branch of the hepatic artery.
means and the use of selective transcatheter embolization to
interrupt flow has become the treatment of choice for this
condition.29,30
Clinical Presentation
------------------------------------------------------------------------------------------------------------------------------------------------
GASTROINTESTINAL HEMORRHAGE
Bleeding from the GI tract is one of the most common, dra-
matic, and ominous signs of portal hypertension. Bleeding
most commonly occurs from varices in the distal esophagus
and gastric cardia. Although the incidence of bleeding in
adults with portal hypertension from nonvariceal sites may
be as high as 25%, there is no evidence that the same holds
true in children. In one study of children presenting with up-
per GI bleeding, the vast majority of unselected patients had
bleeding from varices secondary to portal hypertension, with
no other source.31 The importance of portal hypertension as a
cause of upper GI bleeding in children may vary in different
parts of the world. Peptic ulcer disease is still a frequent find-
ing at endoscopy in children in the Western Hemisphere.32 In
children with portal hypertension of any cause, upper GI
bleeding is almost universally caused by variceal hemorrhage,
although portal hypertensive gastropathy may cause less acute
bleeding. Therefore variceal hemorrhage is usually the only
source of bleeding in children with portal hypertension,
and it is the most common cause of bleeding from the upper
GI tract in nonhospitalized children older than 2 years.
Variceal hemorrhage may take the form of hematemesis,
hematochezia, melena, or chronic anemia. Bleeding most
commonly originates from the lower esophagus and the cardia
of the stomach. Thin-walled varices that contain blood under
raised venous pressure are eroded from gastric acid or medi-
cations such as aspirin or nonsteroidal antiinflammatory
drugs, and blood fills the stomach. In the absence of advanced
liver disease, the blood clots and gathers in the stomach,
producing gastric distention and eventually hematemesis.

The bleeding may be sudden and dramatic and is usually not
accompanied by abdominal pain. A previously well child can
vomit large quantities of altered or fresh blood. Although this
can be extremely frightening to both the child and the parents,
the bleeding usually stops spontaneously. Varices may be
present in the lower GI tract as well. Rectal bleeding is less
common and occurs from rectal and sigmoid varices. Melena
or the passage of gross blood through the rectum or accom-
panying a bowel movement may occur in the absence of
hematemesis.
Although most patients with portal hypertension have
varices, not all varices bleed. Predicting the risk of bleeding
from varices is based on Laplace law, which states that the
transmural tension in the wall of a vessel is directly related
to the pressure inside the vessel and inversely related to
the thickness of the vessel. Unfortunately, intravariceal
pressure cannot be easily measured33; therefore the risk of
bleeding in adults has been linked to the size of the varix,
the severity of liver disease, and the appearance of red wale
markings on the vessel wall, which may indicate epithelial
thickness.34 More recently, detailed measurements of the
cross-sectional diameter of the varices have been correlated
with the risk of bleeding35 in adults. Similar studies are nec-
essary in children.
Predicting which patient is likely to bleed from varices is
important in determining treatment options, particularly for
those who may be candidates for prophylactic therapy of
any kind.36 It is estimated that 30% to 50% of patients bleed
from varices once they are diagnosed37 and that 10% to 50%
of patients with liver disease die every year after the first
hemorrhage.38
Aggressive therapy is indicated after the first bleed in pa-
tients with or without liver disease. This may include medical
therapy in addition to local control of varices with banding or
sclerotherapy, portosystemic shunting in patients with stable
and well compensated cirrhosis, or liver transplantation in
those with more advanced disease. There may be a role for
noninvasive or even invasive surgical therapy in those who
are at high risk for bleeding but have not yet bled, provided
the morbidity and mortality associated with bleeding can be
reduced with an acceptably low risk of complications.
In children with portal hypertension from extrahepatic
portal vein thrombosis, mortality from bleeding is much less
than in those with liver disease and bleeding is generally well
tolerated. However, the same principles apply: early inter-
vention and implementation of a long-term plan to reduce
hospital admissions and invasive procedures. Because these
children have a vascular problem rather than hepatic disease,
they are more amenable to surgical intervention and generally
obtain excellent long-term results.
Portal Hypertensive Gastropathy
The gastric mucosa develops distinct pathologic changes in
the presence of portal hypertension and liver disease. The
symptoms include chronic blood loss and iron deficiency
anemia, but occasionally portal gastropathy may also cause
acute GI bleeding and hematemesis. The diagnosis is generally
made during endoscopy, when the severity of the lesion can
also be graded. The severity of the lesion has been correlated
with the severity and frequency of bleeding and the necessity
of treating the consequent anemia.
CHAPTER 108 PORTAL HYPERTENSION 1359
Hypersplenism
Silent splenomegaly without any other history suggestive of
portal hypertension is often the first sign of a serious under-
lying disorder. Splenomegaly may be detected in infancy or
early childhood and can easily be misinterpreted as a sign
of hematologic malignancy, particularly in children with extra-
hepatic portal vein thrombosis and no other stigmata of liver
disease. With the advent of Doppler ultrasound examinations
of the abdomen, portal vein thrombosis is more readily diag-
nosed and can be appropriately monitored. In addition to
splenic enlargement, patients may present with severe throm-
bocytopenia and leukopenia from splenic sequestration of
platelets and white blood cells. Children are often prevented
from engaging in sporting activities for fear that even slight
trauma to the abdomen may result in splenic rupture and
catastrophic intra-abdominal bleeding, exacerbated by raised
venous pressure and low platelet counts. It is hard to find
documented cases of splenic rupture in which the underlying
portal hypertension contributed to massive bleeding and
death.39 Splenic artery aneurysm is a rare complication in
patients with portal hypertension and spontaneous hemor-
rhage, and suggested treatments have been reported.40
Encephalopathy
Encephalopathy is extremely unusual in the absence of signs
of advanced liver disease such as jaundice and low levels of
liver-dependent coagulation factors or low albumin levels.
Learning disabilities and behavioral abnormalities are mani-
festations of encephalopathy in children, in contrast to the
traditional signs of disorientation, memory loss, and drowsi-
ness commonly seen in adults. Children may also have accom-
panying hyperammonemia.41
Bleeding from Nongut Sites
Severe thrombocytopenia can lead to hematuria, menorrhagia
in adolescent girls, epistaxis, and hematochezia. In severe
cases, spontaneous intracranial bleeding can also occur, with
serious neurologic consequences. Liver-dependent coagula-
tion factor deficiencies also lead to bleeding, typically from
the genitourinary tract or from the nose. In cases of advanced
liver disease, hemorrhagic complications in the lungs may
cause severe respiratory compromise.
Ascites
Children with ascites may have advanced liver disease
with synthetic failure. Ascites may be accompanied by a low
serum albumin level and decreased plasma oncotic pressure.
Dilated lymphatics in the abdomen from increased hydrostatic
pressure in all portal tributaries can lead to transudation of
fluid across capillary membranes and accumulation of fluid
in the abdominal cavity. Additional mechanisms include
increased nitric oxide production in capillaries, causing
vasodilatation, and an increase in renal tubular absorption
of sodium in patients with decompensated cirrhosis. Ascites
can also occur in the setting of Budd-Chiari syndrome
and outflow obstruction. Ascites is an alarming symptom
that may herald the onset of liver decompensation on a
wider scale.
1360 PART VII ABDOMEN

Pulmonary Disorders Jaundice

Pulmonary hypertension can occur in children with both Jaundice and portal hypertension usually do not occur to-
cirrhotic and noncirrhotic forms of portal hypertension.42,43 gether unless there is advanced liver disease. Children with
Pulmonary hypertension may be secondary to an increase in extrahepatic portal vein obstruction may have mild elevation
vasoactive substances that exert a vasoconstrictive or direct of the total bilirubin level, but jaundice is not a clinical feature.
toxic effect on pulmonary vessels.43a An increase in blood If jaundice is a prominent symptom in a child with portal hy-
levels of prostaglandin F2a, thromboxane, and angiotensin I pertension, it may indicate that the liver function is severely
in patients with portal and pulmonary hypertension suggests compromised and consideration should be given to a liver
that these agents play a role in mediating pulmonary hyperten- transplant evaluation.
sion by reaching the pulmonary circulation in blood shunted
Abdominal Examination
around the liver. In experimental studies, portal hypertension
induced by portal vein ligation was associated with increased Patients with portal hypertension often have protuberant ab-
amounts of inducible nitric oxide synthetase and heme oxyge- domens. Causes include splenomegaly, ascites, and enlarge-
nase-1 messenger RNA and protein in the lungs, compared ment of the liver. Distended abdominal veins are a frequent
with animals in the control group.44 This suggests that the finding, and the direction of blood flow away from the abdo-
pulmonary pathology seen with portal hypertension is associ- men or umbilicus is easily demonstrable. In patients with con-
ated with the production of nitric oxide and carbon monoxide genital hepatic fibrosis, the combination of enlarged liver and
in the lung. spleen and often large polycystic kidneys contributes to mak-
Patients with both pulmonary and portal hypertension may ing the abdomen extremely protuberant.
develop pulmonary symptoms before there is any evidence of
Encephalopathy
bleeding from portal hypertension. Symptoms may include
exertional dyspnea or chest pain, or there may be no symp- Clinically evident encephalopathy is a sign of advanced liver
toms other than unheralded syncope or sudden death.45 disease. Signs of encephalopathy may be hard to discern or
Treatment of portal hypertension may be problematic in chil- quantify in young children. Advanced psychometric testing
dren with advanced pulmonary hypertension, and medical may detect subclinical encephalopathy that may be manifest
treatment for both conditions may be limited in scope and as impaired attention span, poor school performance, or be-
duration.46 havioral issues that may be present even in children with
The hepatopulmonary syndrome includes persistent no underlying liver disease or patients with well-compensated
hypoxemia from arteriovenous shunting in the lungs. Both cirrhosis.49 Although advanced encephalopathy may be exac-
pulmonary hypertension and arteriovenous shunting are po- erbated or unmasked by acute bleeding because of the in-
tentially reversible after the portal hypertension and shunting creased protein load in the gut, it usually signifies advanced
are eliminated, although in cases of cirrhosis, a liver transplant hepatocellular disease that may require the patient to be eval-
may be necessary.47,48 uated for liver transplantation.50

Diagnosis
------------------------------------------------------------------------------------------------------------------------------------------------ LABORATORY INVESTIGATIONS
The diagnosis of the underlying cause of portal hypertension Laboratory tests are useful adjuncts in evaluating patients with
depends on the synthesis of the clinical information gathered portal hypertension. Tests may indicate the severity of the
from the parents and the child and the results of imaging tests accompanying liver disease or may indicate the cause of the
and laboratory investigations. underlying liver disease in many instances. In the acute setting
of a GI bleed, a complete blood count and electrolyte levels
will be necessary as a guide to blood and fluid replacement.
CLINICAL INFORMATION A complete metabolic panel including renal function tests is
helpful in guiding fluid resuscitation and to assess the under-
Hepatomegaly lying status of the kidneys. In patients with established cirrho-
An enlarged liver is usually a sign of hepatocellular disease, sis from underlying systemic diseases like cystic fibrosis,
although patients with cirrhosis may present with impalpable blood gases may be necessary. Low blood glucose levels
shrunken livers. If the liver is hard or nodular to palpation, may indicate impending liver decompensation or an underly-
this is further evidence that the cause of the portal hyper- ing glycogen storage disorder.
tension is a diseased liver. Children with congenital hepatic Children with long-standing portal hypertension may have
fibrosis often present with enlarged livers that may extend nothing more than thrombocytopenia and leukopenia, partic-
as far as the iliac crest on palpation. Children with venous out- ularly those with extrahepatic portal vein thrombosis. An in-
flow obstruction may also present with easily palpable liver crease in the direct bilirubin fraction, a low albumin level, and
enlargement. a long prolongation in prothrombin time all indicate signifi-
cant hepatocellular disease and possible cirrhosis as the
Splenomegaly underlying cause of the portal hypertension. A raised serum
Enlargement of the spleen is a common finding in children ammonia level indicates significant portosystemic shunting
with portal hypertension and is often the first abnormality and may occur with all forms of portal hypertension, although
found on a routine physical examination before the onset of severely increased ammonia levels are generally only seen in
bleeding. Splenomegaly occurs in all forms of portal hyper- patients who have decompensated cirrhosis or those in fulmi-
tension and is not helpful in diagnosing the cause. nant hepatic failure.

Children with extrahepatic portal vein thrombosis may
have laboratory evidence of disordered synthesis of liver-
dependent coagulation factors in both the procoagulant and
plasminolytic pathways, but this is generally not evident
clinically without detailed testing.51
ENDOSCOPY
Endoscopy is indispensable in establishing the cause of GI
bleeding and to confirm the presence of varices in the esoph-
agus and stomach. In cases of acute bleeding and after
stabilization in the intensive care unit, upper GI endoscopy
provides direct confirmation of the bleeding source. Varices,
if present, will be most prominent in the distal third of the
esophagus and in the cardia of the stomach. Endoscopy
should be done only by those who are well trained in the pro-
cedure and familiar with the equipment. Endoscopy provides
an opportunity to intervene therapeutically, as well as to rule
out other sources of bleeding such as peptic ulcers, Mallory-
Weiss tears of the esophagus, or hemorrhagic gastritis.
If endoscopy is done before the onset of bleeding because
of suspected portal hypertension, the appearance of the vari-
ces provides valuable clues about the likelihood of future
bleeding. Variceal diameter greater than 5 mm, the appearance
of red wale markings, and more advanced liver disease based
on Child-Pugh class indicate a greater chance for future
bleeding and may justify the institution of prophylactic treat-
ment.52–54
IMAGING
The first imaging study in any child who presents with hema-
temesis should be abdominal ultrasonography. The status of
the portal vein is one of the most important radiologic con-
siderations in arriving at a diagnosis. Cavernous transforma-
tion of the portal vein and portal vein thrombosis are best
diagnosed by ultrasonography with Doppler interrogation
of the vessels of the liver. A complete evaluation of the
intra-abdominal vasculature including the hepatic veins, the
patency of the splenic and superior mesenteric veins, and
the inferior vena cava is possible. In addition, information
about the size of the spleen can be obtained to confirm the
clinical examination. Liver parenchymal abnormalities such
as nodularity, inhomogeneity, or the presence of cysts can
be seen. The appearance of the kidneys can yield useful diag-
nostic clues because children with congenital hepatic fibrosis
may also have autosomal dominant polycystic kidney disease
and other renal abnormalities.
Computed tomography (CT) and magnetic resonance (MR)
angiography are excellent diagnostic tools and have sup-
planted conventional digital angiography for most purposes.
Both modalities provide excellent information about all the
intra-abdominal vessels and detailed information about the
liver anatomy including the bile ducts. CT-angiography has
several advantages; it can be done more quickly and is less
prone to image degradation from motion artifact than is MR
angiography, unless the MR study is done under general
anesthesia with ventilatory arrest for the duration of the study.
Conventional angiography is an invasive modality that has
few indications as a diagnostic tool in children with portal hy-
pertension. The exceptions are rare cases of unusual vascular
malformations such as arteriovenous communications in the
CHAPTER 108 PORTAL HYPERTENSION 1361
abdomen or liver that may best be delineated by angiography.
Also, angiographic visualization is essential as a prelude
to angiographic embolization or stenting of a vascular
abnormality.
Splenoportography is a technique that was written about
extensively in the past but is almost never used for diagnostic
purposes today.
Transjugular hepatic venography is a test that can be used
to measure free and wedged hepatic vein pressures indicating
the gradient across the hepatic veins. This technique is limited
in small children where the size of the child and the veins
makes the performing of this test more challenging. Wedge he-
patic vein pressure can rule out parenchymal disease and high
sinusoidal pressure as a contributing factor to portal hyperten-
sion. Sinusoidal pressures are normal in cases of prehepatic
portal venous obstruction.
Retrograde transjugular portal venograms can also be used
to obtain high-quality definition of the intrahepatic portal vein
(Fig. 108-5). Hepatic vein angiography can also be used ther-
apeutically for the placement of a transjugular intrahepatic
portosystemic shunt (TIPS) or the dilatation of congenital or
acquired strictures causing venous outflow obstruction.
LIVER BIOPSY
Many children who present with portal hypertension do not
need a liver biopsy to establish a diagnosis. For example, in
children with biliary atresia a biopsy is not necessary as an
adjunctive test after the onset of portal hypertension. In chil-
dren with a known underlying metabolic disease such as
cystic fibrosis or alpha-1 antitrypsin deficiency, a biopsy will
not necessarily add useful information to aid in formulating
a treatment plan.
However, if a biopsy is considered helpful, it can be done
percutaneously in most cases. In children with mild coagulo-
pathies that can be corrected with vitamin K, fresh frozen
plasma, and platelets, the procedure can be performed with
close monitoring, where any bleeding from the biopsy site
can be treated promptly. Children with more profound
FIGURE 108-5 A catheter has been advanced into the left hepatic
vein, and contrast is injected under pressure filling the entire intrahepatic
portal vein.
1362 PART VII ABDOMEN

coagulopathies, as manifested by a prothrombin time in MEDICAL MANAGEMENT

excess of 20 seconds, should have a transjugular biopsy in OF BLEEDING VARICES
the interventional radiology suite or an open biopsy in the
operating room. In general, the limiting factor in the radiolo- Patients admitted with acute variceal hemorrhage require in-
gist’s ability to do a transjugular biopsy is the patient’s size. tense resuscitation with blood and crystalloids, replacement of
The difficulty and technical limitations are greatest in infants coagulation factor deficiencies with fresh frozen plasma, and
and babies. control of the bleeding. Bleeding from varices may be exacer-
bated by disseminated intravascular coagulation, fibrinolysis,
and a decrease in circulating platelet counts. The availability of
Treatment exogenous recombinant factor VII to treat those with compro-
------------------------------------------------------------------------------------------------------------------------------------------------
mised liver synthetic function has enhanced the care of these
The treatment of a child with portal hypertension is predi- patients. Factor VII replacement decreases the amount of
cated first on the underlying pathophysiology of the disease sodium and fluid necessary to correct the liver-dependent fac-
and then on the severity of the symptoms. tor deficiencies, although its use is limited to some extent by
Children with liver disease and portal hypertension do not its cost.58
tolerate bleeding as well as those with essentially normal livers Patient care is best delivered in the intensive care unit,
whose portal hypertension is caused by primary vascular mal- where vital signs including central venous and arterial pres-
formations. Acute hemorrhage adversely affects liver function, sures, hourly urine output, and oxygen saturation can be
which may already be abnormal. The prognosis of a child with monitored. Monitoring of mental status is also essential in
liver disease and portal hypertension depends entirely on the those with cirrhosis because bleeding into the GI tract may ex-
hepatic reserve. Children with relatively stable cirrhosis and acerbate or unmask encephalopathy.
preserved synthetic function require a completely different Cessation of bleeding is the most important therapeutic
therapeutic approach from those with hypoalbuminemia, goal, along with the replacement of extracellular fluid and
growth failure, and coagulopathy. The Child-Pugh score55 blood. In the past, Sengstaken-Blakemore tube tamponade
was devised in an attempt to categorize patients according of varices was the most effective method of short-term control
to degree of hepatic reserve and thus arrive at a prognosis of blood loss. Pharmacologic therapy has proved to be just as
for survival if left untreated or with surgery.56 More recently, effective in adults,59 rendering the balloon method almost ob-
a pediatric-specific score (pediatric end-stage liver disease solete. Pharmacologic control of variceal bleeding has im-
[PELD]) has been developed to stratify children on liver trans- proved greatly with the use of intravenous octreotide, the
plant waiting lists in a way that accurately reflects the severity octapeptide analogue of somatostatin. It has a longer half-life
of the underlying liver disease.57 Children with age-appropriate than somatostatin and fewer side effects. Octreotide has
growth, normal coagulation, and no jaundice have low PELD largely replaced vasopressin in the management of acute var-
scores; their treatment for the complications of portal hyper- iceal bleeding because of the severe vascular side effects of the
tension is different from that of children with high scores, who latter drug.
would be better served by liver transplantation. Octreotide and other somatostatin analogues such as
In children with Child-Pugh scores of 5 or 6 or PELD vapreotide reduce hepatic blood flow and wedged hepatic
scores lower than 10 who have a predicted 1-year survival vein pressure and constrict splanchnic arterioles by a direct
of more than 90%, it is justified to consider complex surgi- effect on arteriolar smooth muscle. Somatostatin analogues
cal therapy to alleviate the symptoms of portal hypertension. are effective in reducing and temporarily stopping the bleed-
The expectation is that these children will survive the sur- ing from portal hypertension and can be used in the initial 24
gery and have a good long-term prognosis without the need to 72 hours while awaiting direct endoscopic management of
for liver transplantation in the short term. For children with varices.60–63 More recently, terlipressin, a vasopressin deriva-
portal hypertensive bleeding from varices and who also have tive, has also been proven effective in the acute bleeding situ-
ascites, advanced coagulopathy or poor synthetic function ation, although experience in children is limited.62,64,65
as indicated by hypoalbuminemia, endoscopic banding or Octreotide is started as a continuous infusion at 1 to 2 mg/kg/hr,
sclerosis of esophageal varices in addition to aggressive up to a maximum of 100 mg/hr, and continued for as long as
medical therapy of portal hypertension is indicated as they symptoms of bleeding persist. Antibiotics and acid suppression
await transplantation. Operation is not indicated in children are also recommended.62
with advanced liver disease because it would be poorly tol- The long-term medical management of children with portal
erated and not address the underlying issue of poor hepatic hypertension includes the use of nonselective beta blockers
reserve. such as propranolol or nadolol, either alone or in combination
Pediatricians and pediatric gastroenterologists closely fol- with nitrate vasodilators such as isosorbide-5-mononitrate.
low all patients with portal hypertension and liver disease. Some authors, however, have shown that local control of var-
When portal hypertension becomes symptomatic, it is rarely ices with banding is equivalent to long-term pharmacologic
a surprise and the parents are usually well versed in the med- management. In a majority of patients, surveillance endoscopy
ical issues. Under ideal circumstances, surgery can be consid- and intermittent banding may be sufficient,37 with equivalent
ered for the control of symptoms when bleeding is imminent survival and rebleeding rates in endoscopically managed and
or likely to recur. Selective shunting is generally the procedure pharmacologically managed groups. The role of beta blockers
of choice, as discussed later. in reducing splanchnic blood flow and wedge hepatic vein
In contrast, patients with noncirrhotic forms of portal hy- pressure is well documented; their use may decrease the inci-
pertension can develop symptoms quite unexpectedly and as dence of recurrent bleeding by as much as 50% and lessen the
the first manifestation of a serious underlying condition. need for liver transplantation in patients with liver disease.66

Assessment of wedge hepatic vein pressure reduction has been
promoted as the best way to determine the impact of medica-
tions on portal pressure and to guide and refine pharmaco-
logic management,67 but these studies are all adult based
and may represent a less practical approach in children.68
PROPHYLACTIC TREATMENT OF VARICES
Intervention of any kind before the onset of bleeding is con-
troversial. However, the benefit to the patient of preventing a
first bleed may be considerable and justifies early prophylactic
management. Studies in both adults and children have dem-
onstrated that prophylactic treatment of varices by pharmaco-
logic means and with endoscopic ligation reduces the
frequency of bleeding but may not diminish mortality in pa-
tients with advanced liver disease without transplanta-
tion.69,70 Primary procedures for the prevention of an initial
bleed are more controversial if there is no other primary goal
such as the treatment of severe hypersplenism.
ENDOSCOPIC SCLEROSIS OR BANDING
OF VARICES
Surgeons were involved in the earliest attempts to control var-
iceal bleeding by injecting sclerosing solutions into the tissue
around a varix or directly into a varix to obliterate the vessel
lumen.71 This method has proved to be effective and safe in
children in the acute phase of bleeding.52–54,72–81 Sclerosing
solutions include sodium morrhuate, ethanolamine, sodium
tetradecyl sulfate, and polidocanol. Long-term follow-up in
children after sclerotherapy for the control of initial bleeding
has demonstrated success in almost 90% of patients. Initial
sclerotherapy should be followed by repeat endoscopy at reg-
ular intervals until all the varices are obliterated or too small to
inject and then at longer (yearly) intervals for at least 4 years to
check for reappearance of varices. Generally, two to three in-
jections at 1 mL per injection are required for each varix, up to
a maximum of 10 to 15 mL per session. The endoscopist
should proceed circumferentially from the distal esophagus
at the gastroesophageal junction to the more proximal
esophagus.
Sclerotherapy is associated with a fairly high incidence of
complications, both major and minor, in at least one third
of patients. Acute complications include chest pain, esopha-
geal ulceration, and mediastinitis; chronic ones include
esophageal strictures from fibrosis after multiple injection
sessions.
Esophageal banding has become an increasingly popular
method of treating varices. In one prospective trial in children
with portal hypertension from extrahepatic portal vein throm-
bosis, banding was found to be a more effective, more rapid,
and safer method of reducing the chance of bleeding from var-
ices.82 The incidence of complications and of long-term
rebleeding was lower with banding. New technology, such
as the multiband ligator scope, has made it possible to apply
bands to multiple varices; this method was first used in adults
but has recently been extended to children.83,84 This tech-
nique increases the speed and safety with which esophageal
varices can be ligated in children as young as 3 months old;
obliteration of varices was accomplished in almost 100% of
children after only two sessions.
CHAPTER 108 PORTAL HYPERTENSION 1363
Esophageal banding in concert with pharmacologic control
has become the procedure of choice in the early therapy of
bleeding esophageal varices. Subsequent sessions are aimed
at ligating residual varices or varices that arise after the larger
ones are tied off. Ligation is effective in most children and is
not associated with many of the adverse side effects of
sclerotherapy.
INJECTION THERAPY FOR GASTRIC
VARICES (TIPS)
With the advent of more effective means of controlling esopha-
geal varices, bleeding from varices lower in the GI tract has be-
come more problematic. Injection with N-butyl-2-cyanoacrylate
has had modest success in adults with bleeding from gastric
varices and results in children have been encouraging.85,86
Gastric varices may coexist with esophageal varices in up to
15% to 20% of patients, or they may be isolated. Gastric varices
may emerge after the obliteration of esophageal varices, or they
may persist after esophageal varices have been ligated. Experi-
ence in adults suggests that in patients with gastric varices,
cirrhosis, and advanced liver disease, bleeding can be controlled
with minimal morbidity, but there is a high incidence of rebleed-
ing at 1 year after treatment.87 In adults, injection of gastric
varices results in relatively poorer control of bleeding in patients
with extrahepatic portal vein thrombosis (35%) than in those
with liver disease (75%).88 Given the fact that extrahepatic portal
vein thrombosis is a more important cause of bleeding varices in
children than in adults, injection therapy for gastric varices in
pediatric patients is not justified, except in the setting of a formal
study to determine its safety and efficacy compared with surgery
or other medical therapies.
TRANSJUGULAR INTRAHEPATIC
PORTOSYSTEMIC SHUNTS
The percutaneous insertion of vascular stents to create chan-
nels between the portal vein and hepatic veins within the pa-
renchyma of the liver heralded a new way of treating patients
with advanced liver disease and portal hypertension.89,90
The main indication for TIPS is variceal hemorrhage recalci-
trant to more conservative therapy with endoscopy or octreo-
tide. It is usually reserved for patients with advanced liver
disease and serves as a bridge to transplantation.91 Other in-
dications include refractory ascites, hepatic venous outflow
obstruction in both transplant and nontransplant patients,
and hepatorenal syndrome.
In children, the experience is limited. TIPS may act as a
bridge to transplantation in children with portal hypertensive
gastropathy or in those whose variceal bleeding is difficult to
control by other means. The TIPS method has been used in
children with cystic fibrosis,92 biliary atresia,93 and congenital
hepatic fibrosis94 with apparent success and in infants as
young as 1 year.
Its primary limitation is the high rate of shunt thrombosis.
Vigilance is required to monitor shunt patency and to declot
the shunt when it is thrombosed. The other limitation is the
aggravation of preexisting encephalopathy in approximately
20% of patients and the hastening of liver failure in patients
whose livers are already borderline; TIPS facilitates the
1364 PART VII ABDOMEN
shunting of blood without the benefit of hepatic clearance in
these extremely debilitated patients.95
Hepatic encephalopathy is less common after TIPS in chil-
dren than in adults, but the incidence of shunt occlusion is
higher because of the smaller shunt diameter.
SHUNT SURGICAL PROCEDURES
Early shunt operations diverted all the mesenteric blood flow
into the systemic circulation, as first described by Eck.2 Later
authors reported the application of portacaval shunting in
the setting of severe bleeding from portal hypertension in
adults with advanced liver disease,55,96–99 as well as in
children with prehepatic and intrahepatic causes of portal
hypertension.100–104
Shunts can be described as nonselective or selective. Selec-
tive shunts preserve the majority of portal or mesenteric blood
flow to the liver while shunting blood from high-pressure gas-
troesophageal varices into the low-pressure systemic venous
circulation. In essence, selective shunts divide the portal cir-
culation into two separate entities, although it is impossible
to fully divide all communications. In contrast, nonselective
shunts divert a large proportion of mesenteric blood flow away
from the liver so that the entire GI tract including the spleen
and pancreas is decompressed. Nonselective shunts can be
further subdivided into total or partial portal diversions.
Nonselective Shunts
Total Diversion The end-to-side portacaval shunt as first
described by Eck is the classic example of a total portal diver-
sion. Portal blood is completely redirected into the inferior
vena cava below the liver, and the hepatic end of the portal
vein is oversewn. That operation is almost never done in
children and bears no further discussion. All nonselective
large-caliber shunts deprive the liver of virtually all mesenteric
venous blood flow and increase the danger that the undesir-
able side effects of portosystemic shunting will be produced.
Encephalopathy, pulmonary hypertension, and formation of
regenerative nodules in the liver are some of the long-term
side effects of complete mesenteric venous diversion away
from the liver and can happen even in children with ostensibly
normal liver function and no intrinsic liver disease.
Proximal Splenorenal Shunt The proximal splenorenal
shunt is another example of a shunt that results in total diver-
sion of mesenteric venous blood into the systemic circulation.
The pancreas is mobilized cephalad, exposing the underling
splenic vein. All branches between the pancreas and the
splenic vein must be tied off meticulously. The splenic vein
is divided close to the spleen, and the spleen is removed.
The left renal vein is isolated, and the mesenteric end of the
splenic vein is sewn to the side of the left renal vein so that
all the blood from the superior and inferior mesenteric veins
is shunted into the systemic venous circulation through the
left renal vein.105,106 This procedure almost invariably in-
cludes a direct attempt to ligate the varices of the esophagus
and stomach in the region of the coronary vein.
Although the operation is effective in relieving the symp-
toms of portal hypertension, it should be used sparingly
because it not only exposes the child to the potential draw-
backs of complete mesenteric diversion, as outlined earlier,
but also results in the loss of the spleen.
Mesocaval Shunt Wide-diameter mesocaval shunts also
completely divert mesenteric blood away from the liver. This
procedure can be done directly as a side-to-side anastomosis
between the two veins or with the interposition of a short au-
tologous vein graft or prosthetic graft.107,108 Use of an inter-
position graft is the more common method. Exposure of
both the superior mesenteric vein at the root of the bowel mes-
entery and the inferior vena cava below the duodenum is
required.
Mesocaval shunting has been used in children in a wide va-
riety of settings and diseases, with uniformly acceptable re-
sults.109–111 Some prefer this method because it allows an
anastomosis between two large-diameter vessels and because
the length of the anastomosis can be increased to some extent
to facilitate the creation of a large venous fistula, which is tech-
nically easier to perform than a proximal splenorenal shunt. In
addition, in contrast to the proximal splenorenal shunt, the
spleen is preserved, which is thought to be important in pre-
venting postsplenectomy sepsis.
Side-to-Side Portacaval Shunting The side-to-side porta-
caval shunt allows blood from the intestine and spleen to
flow easily into the vena cava. In addition, and unlike with
the end-to-side portacaval shunt, the hepatic end of the por-
tal vein is changed into an outflow tract. In cases of Budd-
Chiari posthepatic portal hypertension, the liver is decom-
pressed by this operation; this may result in long-term
palliation of the disease, with arrest or delay in the progres-
sion of hepatic fibrosis and ultimate failure.14,112,113 Other
effective operations for Budd-Chiari syndrome are mesocaval
and central splenorenal shunts, which allow portal blood in
the liver to empty in a retrograde fashion through the patent
portal vein.
Partial Diversion Partially diverting shunts depend on a
fixed, narrow communication between a vessel in the portal
circulation and one in the systemic venous circulation. The
Sarfeh shunt can divert enough blood from the portal circula-
tion to drop mesenteric pressure below 12 cm H2O, thus
decompressing the varices and reducing the chance of bleed-
ing while maintaining enough pressure in the portal bed to al-
low hepatopetal flow and hepatic blood flow preservation.114
Although there may be less encephalopathy with the Sarfeh
shunt because of sustained hepatic portal flow, it is associated
with a higher incidence of thrombosis and recurrence of hem-
orrhage and rarely used in children.115
The relationship between deprivation of portal blood to the
liver and encephalopathy has long been noted following porta-
caval shunt surgery. Sarfeh was able to determine the diameter
shunt that would preserve forward flow in the portal vein, de-
compress the portal circulation so that bleeding from varices
decreased, and minimize the incidence of encephalopathy.
Selective Shunts
To diminish the encephalopathy that followed portacaval
shunting, Warren and Zeppa116 described a shunt between
the portal end of the splenic vein and the side of the renal vein
(Fig. 108-6). The splenic confluence with the portal vein was
ligated, and the coronary vein was also interrupted. In this
manner, the gastroesophageal varices were decompressed
across the short gastric vessels and the spleen was decom-
pressed into the renal vein. Long-term studies have shown

Ligated coronary vein
Spleen
Portal vein
Splenic
Ligated stump
vein
of splenic vein
Superior Kidney
mesenteric vein
Renal vein
FIGURE 108-6 Schematic drawing of the distal Warren shunt. Note that
the coronary vein has been ligated. The varices are both decompressed by
allowing them to drain via the short gastric vessels and splenic vein into
the left renal vein and by decreasing blood flow to the varices by tying
off the blood that reaches them through reverse flow in the coronary vein.
that the disconnection between the lower-pressure splenic cir-
culation and the higher-pressure hepatic one is not always
maintained, and the connection may re-form over time. Hepa-
topetal flow may decrease and be redirected to the coronary
circulation around the stomach; as a result, the hepatic flow
decreases, just as in more central shunts.117,118 In this case,
there would be both angiographic and endoscopic evidence
of gradual conversion of a selective shunt to a less selective
shunt over time. A more complete pancreaticosplenic dis-
connection may avoid the reestablishment of communication
between the splenic and hepatic circulations. Long-term
mortality and rebleeding rates in adults are equivalent with
selective and nonselective shunts,119 but the rate of encepha-
lopathy is reduced with selective shunts.120,121
In children, the selective shunt as described by Warren or
with minor modifications has been used successfully for the
treatment of bleeding varices, as well as for the treatment of
hypersplenism.122 Long-term patency rates exceeding 90%
have been reported in most series.
Distal splenorenal shunting is used primarily in children
with extrahepatic portal vein thrombosis, stable Child-Pugh
classification A or B cirrhosis, or less common forms of intra-
hepatic portal hypertension such as congenital hepatic fibrosis
with well-preserved liver function but symptomatic variceal
bleeding.
The preservation of neurocognitive function is particularly
important in children; thus all forms of nonselective shunting
should be avoided whenever possible. Encephalopathy in
children may manifest as learning disorders or behavioral
abnormalities that are not usually attributed to portosystemic
shunting.123
Hypersplenism may be particularly problematic in children
and has led to splenectomy or splenic embolization in some
children100,124–132 with or without a central splenorenal
shunt.
The distal splenorenal shunt has been beneficial in chil-
dren with advanced hypersplenism as a more physiologic
CHAPTER 108 PORTAL HYPERTENSION 1365
alternative to the splenectomy.133 Spleen size regresses, and
the platelet and leukocyte counts return toward normal.
Following splenic decompression by shunting, hematologic
indices may continue to improve for years after the procedure.
Splenectomy and splenic embolization were advocated in
the past for hypersplenism. However, these procedures are
not without complications and may, in the long run, exacer-
bate bleeding from gastroesophageal varices and leave the
child prone to postsplenectomy sepsis. In addition, splenec-
tomy removes the possibility of later distal splenorenal shunt-
ing if the child continues to suffer from variceal bleeding.
Pediatric surgeons should preserve the spleen whenever
possible, leaving splenectomy as the last resort in children
who have no other options.
MESENTERIC-TO-LEFT PORTAL VEIN BYPASS
The treatment of extrahepatic portal vein thrombosis has been
evolving since the advent of mesenteric vein-to-left portal
vein bypass in 1992. The operation was originally described
as a way to revascularize liver transplant grafts after post-
transplant portal vein thrombosis,134,135 but the indications
were later expanded to treat children with idiopathic portal
vein thrombosis.136 A growing number of children have been
reported in the literature from a small group of institutions
worldwide.11,47,137,138
Essentially, the operation relieves portal hypertension by
redirecting blood from the obstructed mesenteric system to
the still patent intrahepatic portal vein. The prerequisites
for a successful bypass operation are threefold: (1) no intrinsic
liver disease, (2) a patent intrahepatic portal tree, and (3) a
suitable vein in the mesenteric circulation to function as a suit-
able inflow tract for mesenteric blood.
In portal vein thrombosis, the extrahepatic portal vein is
replaced by a network of small collaterals that supply the liver
with a small amount of mesenteric blood and keep the intra-
hepatic portal circulation patent. The intrahepatic portal tree
in these children is patent but hypoplastic.
The preoperative workup includes CT-angiography to as-
sess the caliber and patency of the intrahepatic portal vein,
a liver biopsy to rule out intrinsic liver disease, and a coagu-
lation workup to rule out a hereditary hypercoagulable
state.139 It was once thought that hereditary hypercoagulable
states, particularly protein C deficiency, often may have caused
the thrombosis of the portal vein.140–142 However, it has been
reported that deficiencies not only in the plasminolytic path-
way but also the procoagulant pathway may be secondary to
the hepatic deprivation of portal blood and are reversible after
the successful restoration of portal flow.51
The operation starts by dissecting out the intrahepatic por-
tal vein in the recessus of Rex, where the round ligament in-
serts between segments III and IV of the liver. Portions of
segments III and IV are removed. All the branches to segments
II, III, and IV are controlled. If it is suitable, the superior
mesenteric vein is dissected out at the root of the mesentery.
Alternative sources of blood for the bypass including the
splenic, inferior mesenteric, and coronary veins have been
described.143
Once the inflow vein has been isolated, the jugular vein
(either right or left) is removed and sewn first to the intrahe-
patic portal vein and then to the superior mesenteric vein.
The vein graft usually goes through the mesocolon and the
1366 PART VII ABDOMEN
lesser sac, up behind the stomach, through the lesser omen-
tum, and up to the liver. In children with malrotation or
other congenital disorders, the route of the graft may vary,
depending on the child’s individual anatomy. CT or MR scan-
ning in the postoperative period demonstrates the patent
shunt (Fig. 108-7).
Unlike portosystemic shunting, the mesenteric-to-left
portal vein bypass, or Rex shunt, is restorative rather than pal-
liative. It restores portal flow to the liver and relieves the symp-
toms of portal hypertension. With proper patient selection,
patency rates have been in excess of 90%.11 Patency depends
primarily on the quality of the intrahepatic portal vein and the
number of branches that allow for sufficient runoff to permit
blood flow to be sustained. The intrahepatic portal vein may
be difficult to visualize by preoperative imaging, and a final
assessment may be possible only at the time of surgery.
NONSHUNT OPERATIONS
Nonshunt operations for portal hypertension are generally
limited to a direct attack on the varices by operative interrup-
tion and ligation, as well as esophageal transection to interrupt
the intramural varices within the esophagus.144–147 In the ab-
sence of suitable shunt options, these procedures are useful
alternatives.
Sugiura and colleagues147 described an operation in which
all perigastric and periesophageal vessels were separated from
the incisura of the stomach to the midesophagus at the level
of the inferior pulmonary vein. In addition, the esophagus
LPV
G ventional radiologic techniques can study the vessel and the
SV
REX
dSMV
FIGURE 108-7 The postoperative appearance of a meso-Rex bypass. The
blood flows from the superior mesenteric vein (dSMV) and recruits blood
from the splenic vein (SV), the coronary vein (G). The vein graft (Rex) flows
cephalad into the left portal vein (LPV). The white arrows indicate a mild
narrowing near the anastomosis between the jugular vein graft and the
intrahepatic portion of the left portal vein.
was divided and reattached. This radical approach to portal
hypertension presented an alternative to shunting procedures,
but it may have additional morbidity. A modified Sugiura
operation in which the esophagus is transected and reanasto-
mosed by a stapling device and in which the spleen is pre-
served has been described in children. As experience with
shunting and other vascular surgery in children has increased,
the use of these effective but palliative and unphysiologic
operations has decreased.
Complications of Surgery
------------------------------------------------------------------------------------------------------------------------------------------------
SHUNT THROMBOSIS
Any vascular operation, especially on the venous system, has
the potential for thrombosis. Prophylaxis against thrombosis
may be instituted in the immediate postoperative period with
low-dose heparin and antiplatelet agents such as aspirin or
dipyridamole and continued for 3 to 6 months. Once a shunt
is thrombosed, it may not be salvageable, and other solutions
such as devascularization must be sought if bleeding recurs.
Careful follow-up, antithrombotic prophylaxis, and good
operative technique to begin with should result in a long-term
patency rate of greater than 90%.
ANASTOMOTIC STENOSIS
Anastomoses can develop narrowing that can cause sluggish
flow in the vessels, possibly thrombosis, and recurrence of
the original signs and symptoms. Stenoses have characteristic
appearances on Doppler ultrasound examinations that in-
clude a narrow area at the anastomosis, poststenotic dilatation
of the vessel secondary to turbulent flow, and acceleration of
the velocity of blood flow through the stenotic area. If the ste-
nosis is believed to be of hemodynamic significance, then it
must be dilated either by radiologic or surgical means. Inter-
stenotic area, assess the hydrostatic gradient on either side
of the anastomosis, and dilate the narrow area by balloon di-
latation. The effect of the ballooning can also be assessed by
remeasurement of the pressure gradient. Placement of an
endovascular stent is also a possibility in recalcitrant strictures
(Fig. 108-8). We prefer to avoid stents in smaller children be-
cause, once placed, a stent may act as a fixed stenosis as the
child grows and cannot be removed.
ASCITES
Ascites is common after shunt operations because of disrup-
tion of the retroperitoneal lymphatic channels. In most cases
this resolves spontaneously. Oral diuretics may be necessary
for a short time. A reduced-fat diet may also be helpful be-
cause all the ascites is chylous in nature. Rarely, paracentesis
may be necessary if the ascites is tense and persistent. If so,
parenteral nutrition may be used to eliminate any chylous
flow for a short time until the lymphatic leak seals. If ascites
does not resolve with these measures, it may indicate that
the hepatic reserve is small and transplantation may be
necessary.
 CHAPTER 108 PORTAL HYPERTENSION 1367

INTERMITTENT BLEEDING FROM

ESOPHAGEAL OR GASTRIC VARICES
Today, because of the effectiveness of endoscopy and medica-
tions, few patients with extrahepatic portal vein thrombosis
present with acute, poorly controlled variceal bleeding requir-
ing surgery. Patients are presenting with greater frequency
with advanced hypersplenism and well-controlled varices.
Controlling variceal bleeding does, however, require numer-
ous endoscopies, occasional transfusions, and treatment with
beta blockers and vasodilators.
As surgical options have become more effective and the re-
sults more predictable, definitive surgery for extrahepatic por-
tal hypertension should be offered at an earlier point in the
disease process rather than waiting until all other therapy
has failed.148 This is especially relevant with the advent of
the meso-Rex bypass because this operation restores normal
portal pressure and relieves hypersplenism by redirecting
mesenteric blood back to the liver. The portal vein delivers
FIGURE 108-8 A transhepatic catheter is entering the liver from the right 50% of the oxygen and 80% of the total blood supply
side of the patient. It has deployed a clearly visible endovascular stent used to the liver. After a successful meso-Rex bypass, the flow in
to treat a stricture in the vein graft or at the anastomotic suture line that the vein graft increases over time as evidenced by the
was resistant to simple balloon dilatation. rapid expansion in portal vein branches inside the liver
(Fig. 108-9, A and B). The effects of restoring portal blood
to the liver are not yet apparent and may have far-reaching
metabolic consequences during the child’s early development.
The correction of coagulation parameters back to normal
ESOPHAGEAL STRICTURES after portal flow restoration may be only one of many amelio-
After devascularization procedures accompanied by esopha- rations in hepatic function that take place. The common belief
geal transection, it is not uncommon for the esophagus to nar- that a child with portal vein thrombosis and portal hyper-
row at the site of transection. Overt leaks are an uncommon but tension has “normal” hepatic function may not be correct.
serious complication that requires thoracostomy drainage, pro- As experience with the meso-Rex bypass increases, more
longed antibiotics, and possibly reoperation. Fortunately, most evidence of the metabolic sequelae of the restoration of portal
strictures are manageable by esophageal dilatation, although flow to the liver has become apparent. Both clinical and
some may require repeated sessions to achieve satisfactory experimental data point to improved growth, protein synthe-
patency. sis, and neurocognitive function following a meso-Rex by-
pass.49,51,149–151 The long-term consequences of portal vein
thrombosis on bile and bile ducts are also not well under-
stood, but evidence suggests there is an increased incidence
Outcome of cholelithiasis and biliary cirrhosis from uncorrected portal
vein thrombosis.152,153
------------------------------------------------------------------------------------------------------------------------------------------------

The outcome of treatment in a child with portal hypertension Recently, the conclusions of a consensus conference held to
depends on the cause of the hypertension and the underlying formulate surgical guidelines in the treatment of children with
state of the liver. In the past decade, the care of these patients portal vein thrombosis were published. These take into ac-
has improved considerably because of better medical and sur- count both the developing comfort in doing this procedure,
gical options and the availability of liver transplantation in as well as the new data that indicate the meso-Rex bypass
children with poor hepatic reserve. not only relieves the symptoms of portal hypertension as well
as other procedures but also restores to normal many of the
metabolic parameters of the liver that have been affected by
the long-standing deprivation of portal blood flow.148
EMERGENCY BLEEDING FROM VARICES
Selective shunting, through either a distal splenorenal
Medical control of portal hypertension has improved to the shunt or a coronary-to-vena cava graft, to correct portal hyper-
point that emergency shunts in children are the exception tension from portal vein thrombosis has been established as a
rather than the rule. Medical therapy and endoscopic control procedure with minimal morbidity and no mortality when
of esophageal varices by ligation is excellent. Emergency done by experienced surgeons. Patency rates exceed 95%,
shunts, when necessary, are effective in controlling hemor- and hospital stays of less than a week are common. In the long
rhage, with good long-term patency rates reported. However, run, selective shunting may be better for children than re-
the patency rate of emergency shunts is lower than the 90% or peated hospital admissions for banding and long-term medi-
greater patency rate reported for elective operations. Emer- cation use to control bleeding. In addition, advanced
gency mesocaval shunting is seldom reported today, owing hypersplenism occurs in many of these children when defin-
to the availability of medical and endoscopic means of control- itive surgical therapy is delayed. Even when variceal banding
ling acute variceal hemorrhage. is instituted prophylactically, bleeding may shift away from the
1368 PART VII ABDOMEN
A treatment with beta blockers. Long treatment may also include
B Almost all shunts have been shown to decrease the severity
FIGURE 108-9 These two images of the liver from the same patient be-
fore (A) and after (B) the meso-Rex bypass illustrate the expansion in the
intrahepatic portal vein size that occurs quickly after restoration in mesen-
teric venous flow (arrows in both images point to the portal vein in the
equivalent area). The change in portal vein size is indicative of an increase
in the volume of blood flowing through the liver.
esophagus and into the stomach in many children over time.
Surgical therapy (Rex shunt or distal splenorenal shunt)
controls bleeding, reduces hospital admissions, decreases
the need for medications, and relieves hypersplenism in a
physiologic way that does not expose the child to the lifelong
risk of overwhelming sepsis after splenectomy.
Shunting for portal hypertension caused by intrinsic liver
disease is a more complex proposition because it has to take
into account the health of the liver. Selective shunting is excel-
lent at decompressing the child’s portal system, with minimal
morbidity in patients with low Child-Pugh or PELD scores. Al-
though the theory behind the distal splenorenal shunt is that it
separates the splenic and gastric venous drainage from the in-
testinal and mesenteric venous flow, careful studies have
shown that separation between the two systems is not always
complete and may disappear over time as the body forms col-
lateral veins to replace the ones that were interrupted. Never-
theless, selective shunts that attempt to preserve hepatopetal
flow result in less encephalopathy and better hepatic function
than do nonselective shunts in both children and adults.
If a child with well-preserved liver function and portal hy-
pertension cannot have a selective shunt because of a previous
failed shunt or because of unfavorable anatomy, it may be bet-
ter to consider a devascularization procedure to minimize
the potential for encephalopathy.154,155 The only role for
nonselective shunts in children may be the rare child with
well-preserved liver function and obstructive venopathy that
cannot be stented or opened by an interventional radiologic
approach. In this case the risk of minimal encephalopathy
may be outweighed by the dangers of recurrent bleeding.156
GASTROPATHY
The treatment includes all measures generally applied to
bleeding from esophageal varices except that endoscopic treat-
ment is not possible. Measures include drug therapy with
somatostatin analogues, acid suppression, antibiotics,62 and
surgical relief of the portal hypertension. This has been shown
to reverse the gastropathy.157 The choice of operation of
course depends on the cause of the portal hypertension and
the venous anatomy.
HYPERSPLENISM
As the treatment for esophageal variceal bleeding has im-
proved, more children are presenting with bleeding from gas-
tric varices, portal hypertensive gastropathy, or complications
of advanced and debilitating hypersplenism. There is no
medical therapy for hypersplenism other than stimulating
the bone marrow with granulocyte colony-stimulating factor
(GCSF).158,159 In fact, GCSF may transiently lower the platelet
count while it helps with the leucocytes.160 Surgery is the only
alternative.
of hypersplenism when done primarily for the indication of
bleeding from symptomatic varices. The same two principles
should govern the choice of surgery done primarily for
hypersplenism: splenic preservation and the avoidance of
encephalopathy. The distal splenorenal shunt has been shown
to ameliorate hypersplenism in children. In children with
extrahepatic portal vein thrombosis, the Rex shunt has had ex-
cellent results in terms of resolving hypersplenism and also
improving the neurocognitive parameters associated with por-
tosystemic shunting.
Summary
------------------------------------------------------------------------------------------------------------------------------------------------
A child with portal hypertension presents a unique challenge
to the surgeon. The number of treatment options has in-
creased over the past 2 decades and includes a wider variety
of effective medications that reduce the pressure in the mes-
enteric tree and diminish the blood flow in the portal circula-
tion. These options have greatly decreased the need for
emergency surgery in children with portal hypertension.
Figure 108-10 summarizes the treatment options discussed
in this chapter on the basis of the underlying cause of portal
hypertension.
Ancillary interventions such as endoscopic ligation of var-
ices and radiologically guided placement of intrahepatic stents
have also reduced the reliance on operative control of bleeding
 CHAPTER 108 PORTAL HYPERTENSION 1369

Portal Hypertension Treatment Process

Portal
Hypertension

Intrinsic Liver Extrahepatic Outflow

Disease Portal Venous Obstruction
Obstruction

Poor hepatic Good hepatic

Poor hepatic
function (Child function (Child Liver biopsy to rule Stable hepatic
function and/or
class C/D, PELD class A/B, PELD out liver disease function
widespread necrosis
>15) <15)

Treat bleeding Treat bleeding Treat bleeding

complications complications complications Treat medically or List for liver
medically and medically and medically and by stenting transplantation
endoscopic ligation endoscopic ligation endoscopic ligation

Rex shunt or DSRS Consider Consider TIPS

List for liver DSRS for gastric varices, for recurrent bleeding, side-to-side
transplantation advanced hypersplenism while waiting for
hypersplenism, or portacaval shunt transplant
or recurrent esophageal gastric varices
varices

Consider TIPS
while waiting for
transplant

FIGURE 108-10 Treatment algorithm according to the underlying cause of portal hypertension. Portal hypertension accompanied by advanced liver
dysfunction from either venous outflow obstruction or intrinsic liver disease is ultimately treated by transplantation, with or without a transjugular intra-
hepatic portosystemic shunt (TIPS) procedure. If the liver dysfunction is acceptable and not rapidly progressive, an appropriate shunt may be a good long-
term solution. For symptomatic portal hypertension from extrahepatic portal hypertension, mesenteric-to-left portal vein bypass is the procedure of choice,
followed by distal splenorenal shunt (DSRS). PELD, Pediatric end-stage liver disease.

in patients with advanced liver disease. Such interventions in-
crease the chance that these patients will survive to receive a
liver transplant.
Although the need for surgical procedures has been re-
duced, surgical results have improved so that both shunt
and nonshunt procedures offer a more permanent solution
to bleeding and hypersplenism in patients with prehepatic
and posthepatic portal hypertension and in those with
well-compensated cirrhosis. With the decreased morbidity
associated with portal hypertension surgery, the excellent
long-term patency rates reported even in infants and small
children after shunt surgery, and the increased number of sur-
gical options available for children, surgery for portal hyper-
tension offers a good alternative to medical treatment, even
early in the course of treatment.
The complete reference list is available online at www.
expertconsult.com.
 The tip of the right ventral bud lobule eventually rests behind
the superior mesenteric artery. During the seventh week of
gestation, the smaller ventral bud fuses with the proximal part
of the dorsal pancreas.
Secretory acini appear during the third month as clusters of
cells around the ends of the ducts from which they are derived.6
Pancreatic development involves a process in which two mor-
phologically distinct tissue types must derive from one simple
epithelium. These two tissue types, exocrine (including acinar
cells, centro-acinar cells, and ducts) and endocrine cells, serve
disparate functions and have entirely different morphology. In
addition, the endocrine tissue must become disconnected from
the epithelial lining during its development.7,8

PANCREATIC ANOMALIES
Pancreatic anomalies are numerous and are listed in Ta-
ble 109-1.

ANNULAR PANCREAS
During the sixth week of gestation, the common duct and
the right ventral bud lobule are carried in a dorsal direction
CHAPTER 109 around the circumference of the duodenum. The longer duct
of the dorsal pancreas anastomoses with that of the ventral
pancreas to establish the main pancreatic duct. Develop-
mental errors at this stage are believed to cause annular pan-
The Pancreas creas. This anomaly occurs in 1 in 20,000 births and is more
frequent in males than females with a ratio of 2:1. The appear-
ance of an annular pancreas in a woman and her child,9 in
N. Scott Adzick siblings, and in members of two successive generations10 sug-
gests a possible hereditary link, but this has yet to be delin-
eated. Annular pancreas can cause duodenal stenosis or
obstructive symptoms in the first days of life.11
Embryology
------------------------------------------------------------------------------------------------------------------------------------------------ PANCREAS DIVISUM
The pancreas originates from two endodermal buds that arise The adult pancreas is derived from ventral and dorsal pancre-
from the caudal part of the foregut. One is the ventral bud, atic buds that fuse by the end of the eighth week of gestation.
which arises from the base of the hepatic diverticulum and Most of the gland is derived from the dorsal bud and includes
is closely related to the bile duct; the other arises from the dor- the superior anterior part of the head, body, and tail; it is
sal surface of the duodenum immediately opposite and in a drained by the duct of Santorini through the minor papilla.
slightly rostral direction to the liver diverticulum. Both fuse The ventral bud becomes the posterior and inferior part of
at approximately the sixth or seventh week of gestation. The the head. The ventral pancreatic bud is drained by the duct
dorsal bud appears before the ventral bud, grows faster, and of Wirsung into the major papilla of Vater. With fusion of the
becomes larger.1,2 An axial duct is formed and opens into ventral and dorsal pancreas and their duct system, the duct
the gut tube by the beginning of the sixth week of gestation, of Wirsung becomes the major duct, and the duct of Santorini
and short branches appear soon thereafter.3 usually regresses with the minor papilla in approximately 90%
The ventral bud of the pancreas transiently consists of two of people. Failure of the two ductal systems to unite results in
lobules4; small ducts form within these lobules and open into the anatomic variant known as pancreas divisum. In this disor-
the hepatic diverticulum close to the gallbladder. The ventral der the duct of Wirsung is small, and the duct of Santorini
bud is carried away from the duodenum by elongation of the becomes the major ductal system and maintains communica-
proximal part of the hepatic diverticulum from which it arises. tion with the duodenum through the minor papilla. This con-
This proximal part is called the common bile duct. The left ven- figuration is found in 4% to 11% of cadavers and in 3.25% of
tral bud lobule later regresses completely. Because of rapid patients undergoing endoscopic retrograde cholangiopancrea-
growth of the duodenum, which is limited to the left half of tography (ERCP).
its circumference,5 the right ventral bud and the developing
common bile duct rotate backward and become situated close
CONGENITAL SHORT PANCREAS
to the posterior and inferior surface of the dorsal pancreatic
bud. The outcome of this rotation is to bring the opening of Congenital shortening of the pancreas is an unusual anomaly
the common bile duct to the same side as the dorsal bud so that has been described in individuals with polysplenia syn-
that the duct lies ventral to and below the opening of its duct. drome,12 as well as being an isolated anomaly.13 It has also

1371
1372 PART VII ABDOMEN

TABLE 109-1
Errors in Pancreatic Development
Pancreatitis
------------------------------------------------------------------------------------------------------------------------------------------------

Aplasia and hypoplasia Although pancreatitis is uncommon during childhood, it must

Hyperplasia and hypertrophy be considered in every child with unexplained acute or recur-
Dysplasia rent abdominal pain. The prognosis in children is generally
Ductal anomalies good, with the exception of pancreatitis occurring with
Pancreas divisum multiorgan failure. Regardless of the cause, certain common
Annular pancreas features are found in all types of pancreatitis. Patient manage-
Rotational anomalies ment must be highly individualized because numerous
Heterotopic pancreas disease entities can cause pancreatitis.
Cysts
Arterial, vascular, and lymphatic anomalies ACUTE PANCREATITIS
Neoplasm
Cystic fibrosis Clinically, acute pancreatitis is defined as a single episode or re-
current episodes of abdominal pain associated with elevated
serum pancreatic enzyme levels. The morphologic correlate is
acute focal or diffuse swelling and inflammation of the pancreas.
Resolution of symptoms and normalization of blood biochem-
been seen in conjoined twins with a shared duodenum. In istry and anatomic abnormalities follow an acute episode. A
this condition the pancreas appears blunted and lacks tissue continuum exists between acute and chronic pancreatitis; as a re-
in the region of the body and tail, which is believed to be sult, recurrent episodes of acute pancreatitis may evolve into the
caused by agenesis of the dorsal pancreatic bud. Pancreatic typical clinical and morphologic features of chronic pancreatitis.
function generally remains normal despite partial absence
of the pancreatic parenchyma.14 Etiology
Although pancreatitis in adults is usually related to alcohol
abuse or biliary tract disease, in children the causes are more
Surgical Anatomy diverse.17–19 Most cases of acute pancreatitis in children result
------------------------------------------------------------------------------------------------------------------------------------------------
from systemic infection, trauma, choledocholithiasis, ano-
The pancreas is composed of a head, body, and tail, although malies of the pancreaticobiliary duct system, and drugs
these areas have no distinct anatomic borders. The head sits to (Table 109-2). Less common causes include idiopathic
the right on L2, the body covers L1, and the tail rises to T12 on
the left. The abdominal aorta and vena cava provide some TABLE 109-2
cushioning effect against the vertebrae; however, certain forces Causes of Acute Pancreatitis in Children
(such as seat-belt and steering wheel injuries) can cause blunt Systemic infection
trauma to the body of the pancreas. Mumps
The pancreas is retroperitoneal. Peritoneal reflections Rubella
from the transverse mesocolon, small intestinal mesentery, Coxsackie B virus
and the splenorenal and phrenocolic ligaments are present Trauma
and establish mesenteric planes for the direct spread of extra- Blunt abdominal trauma
vasated pancreatic enzymes.15 These anatomic relationships Iatrogenic (e.g., surgery, ERCP)
are important for understanding the pathophysiology of Anomalies of the pancreaticobiliary duct system
the various radiographic signs seen in patients with acute Pancreaticobiliary malunion
pancreatitis. Pancreas divisum
The splenic artery provides most of the blood supply to the Hypertensive sphincter of Oddi
pancreas and lies along its superior border with the splenic Choledochal cyst
vein. It seems that the blood supply is generally greatest to Choledocholithiasis
the head of the pancreas. The head of the pancreas shares Drugs
its blood supply with the duodenum through the gastroduo- Azathioprine
denal artery and superior and inferior pancreaticoduodenal Tetracycline
arteries, which may complicate surgery on the head of the L-Asparaginase
pancreas. Blood from the pancreas drains in a posterior direc- Immunosuppressants
tion into the superior mesenteric vein and splenic vein via Valproic acid
small, delicate branches. Metabolic abnormalities
Several anomalies of the pancreatic (third) part of the com- Hypertriglyceridemia
mon bile duct have been noted: (1) The bile duct may be partly Hypercalcemia
covered by a tongue of pancreas (44%), (2) the bile duct may Cystic fibrosis
be completely covered by the pancreas (30%), (3) the poste- Liver transplantation
rior surface of the bile duct may be uncovered (17%), and (4) Miscellaneous
the bile duct may be covered by two tongues of the pancreas Idiopathic
(9%).16 Surgeons must be aware of these variants during sur-
gery involving the head of the pancreas. ERCP, Endoscopic retrograde cholangiopancreatography.

disease, metabolic disorders, and miscellaneous conditions
such as familial pancreatitis and Crohn disease.
Six percent to 33% of cases of pancreatitis are related to
disorders of the biliary or pancreatic duct,20 with choledo-
chal cyst and cholelithiasis being the most common. A cho-
ledochal cyst is often associated with malunion of the
pancreatic and biliary ducts (pancreaticobiliary malunion),
and a long common channel may form and permit reflux
of bile into the pancreatic duct.21 The common channel is
seen easily on ERCP; this channel is often dilated and
contains protein plugs or stones, which may cause obstruc-
tive pancreatitis or jaundice.
Cholelithiasis in children is frequently associated with
various hemolytic disorders including spherocytosis, alpha-
thalassemia, and sickle cell disease. It may also be caused
by obesity and the use of total parenteral nutrition.22 The role
of pancreatic duct anomalies such as pancreatic divisum and
stenosis of the ampulla of Vater has only recently been appre-
ciated.23–26 Pancreaticobiliary malunion not associated with
a choledochal cyst is drawing clinical attention because it
tends to be overlooked on ultrasonography; most patients
with the disorder have recurrent abdominal pain as a result
of pancreatitis.24,26–28 Tagge and colleagues26 found ductal
anomalies in 6 of 61 (10%) patients with pancreatitis. They
noted that patients with ductal abnormalities usually had
recurrent episodes and that relapsing or chronic pancreatitis
was also associated with ductal anomalies.
Severe abdominal trauma may cause immediate acute
symptoms.29 However, less severe trauma may result in
a delayed onset and subacute manifestation. Child abuse
should be considered in all children with acute pancreatitis
who have a vague history.19 Finally, trauma may also be
iatrogenically induced during the course of abdominal surgery
or ERCP.30
Drugs cause 8% to 25% of the cases of pancreatitis. Drugs
that cause pancreatitis most frequently include didanosine,
azathioprine, mercaptopurine, L-asparaginase, and valproic
acid.20,31,32 Recently, liver transplantation has been reported
as a possible cause of acute pancreatitis. The incidence of this
complication and the mortality rate in children were 1.9% and
43%, respectively.33 A strong association between the devel-
opment of acute pancreatitis after liver transplantation and
several risk factors such as the diagnosis of fulminant hepatic
failure, retransplantation, extensive dissection at the time of
liver transplantation, and the use of infrarenal arterial grafts
has been reported.34
Certain variants of the trypsinogen gene and cystic fibrosis
transmembrane regulation (CFTR) gene may cause acute
recurrent pancreatitis. Many patients with the diagnosis of
idiopathic pancreatitis have compound heterozygote geno-
types in which both alleles of the CFTR gene are abnormal,
thus constituting a variant of cystic fibrosis. There are
also numerous reports of an association between chronic
pancreatitis and a number of CFTR mutations both in
patients with cystic fibrosis and in carriers with a single
mutation.35
In certain conditions, destruction of the pancreas begins
antenatally. Infants with Shwachman-Diamond syndrome,
an autosomal recessive disorder mapped to the centromeric
region of chromosome 7,36 have pancreatic insufficiency at
birth, generally with low serum trypsinogen levels, indicative
of nearly total exocrine pancreatic atrophy by birth.
CHAPTER 109 THE PANCREAS 1373
Diagnosis
The diagnosis of acute pancreatitis is based on the clinical
history, physical examination, laboratory test results, and find-
ings of diagnostic imaging investigations. Determination of am-
ylase isoenzymes has also been used to increase diagnostic
accuracy by identifying the tissue from which the amylase orig-
inates. Serum lipase levels may also complement pancreatitis
testing and increase the yield of positive diagnoses.37 A new
scoring system for acute pancreatitis in children has been pro-
posed that has better sensitivity than the Ranson and Glas-
gow scores.22 The parameters are age (<7 years), weight
(<23 kg), admission white blood cell count (>18,500), admis-
sion lactate dehydrogenase level (>2000), 48-hour trough
Ca2þ level (<8.3 mg/dL), 48-hour trough albumin level
(<2.6 g/dL), 48-hour fluid sequestration (>75 mL/kg/48 hr),
and 48-hour rise in blood urea nitrogen (>5 mg/dL).
Plain radiographs of the chest and abdomen are obtained to
exclude intestinal perforation. Occasionally radiopaque gall-
stones, a gas-filled right colon, or a distended loop of small
intestine (sentinel loop) may be seen. Left basal pleural effu-
sion is relatively common, and mottling of the lung fields is
a sign of systemic cytokine release. Radiographic studies
enhanced by water-soluble contrast of the upper gastrointes-
tinal tract are occasionally useful, particularly in cases of
trauma when injury to the duodenum or small intestine is
suspected. Ultrasonography (US) and computed tomography
(CT) are useful for detecting pancreatic abnormalities.
Trauma-induced injuries to abdominal organs, particularly
the pancreas, are readily detected on CT.
ERCP is rarely indicated for acute pancreatitis, but it even-
tually becomes essential in any child who has pancreatitis with
an unclear cause.23,26 This technique is useful in cases of
relapsing pancreatitis associated with pancreaticobiliary mal-
union.21,24 ERCP is an invasive method that can aggravate
pancreatitis and is often not an option during the acute phase
of pancreatitis. Magnetic resonance cholangiopancreatogra-
phy (MRCP) is a noninvasive method of obtaining images of
the pancreaticobiliary tract. MRCP images the common bile
duct in more than 96% of patients and detects common bile
duct stones with a sensitivity of 71% to 100%, thus exceeding
the sensitivity of US (20% to 65%) and CT (45% to 85%).
Visualization of the smaller pancreatic duct is successful in
more than 80% of patients.37 Miyano and colleagues38 have
previously evaluated the efficacy of MRCP in pediatric patients
with acute pancreatitis. Patients were divided into two groups.
Group 1 consisted of seven patients in whom choledochal
cysts were sonographically diagnosed, and group 2 consisted
of nine patients with no obvious cause of acute pancreatitis.
Pancreaticobiliary malunion was detected in six of seven in
group 1 and in one of nine in group 2. Pancreatic divisum
was detected in one patient in group 1 but could not be con-
firmed in any of the group 2 patients. Dilatation of the main
pancreatic duct was detected in one in group 1 and in three in
group 2. These findings indicate that MRCP is a potentially
useful method for identifying and ruling out structural abnor-
malities of the pancreaticobiliary tract in children.
Treatment
Treatment of acute pancreatitis has two primary goals: (1)
to minimize any causative factors and (2) to provide metic-
ulous supportive care including liberal use of analgesics,
1374 PART VII ABDOMEN
administration of parenteral fluids, maintenance of nutrition,
prevention of infection, and inhibition of endocrine and
exocrine activity.
Parenteral analgesia with narcotics or nonsteroidal antiin-
flammatory agents is generally required, even in mild cases
of pancreatitis, because the pain can be extreme. It is tradi-
tional to administer meperidine rather than morphine
because the latter produces ampullary spasm. Although na-
sogastric suction is usually initiated to reduce vomiting
and abdominal distention, the value of nasogastric decom-
pression is questionable, unless the patient is vomiting.
Aspiration of gastric acid may reduce pancreatic exocrine
secretion by limiting the release of secretin. Oral feeding
must be withheld to reduce pancreatic stimulation. Total par-
enteral nutrition should be initiated early to avoid malnutri-
tion.39 In addition, early placement of a central venous
catheter in patients with severe disease will provide access
for aggressive intravascular volume support and nutrition.
The hematocrit and serum levels of glucose and calcium
should be measured, and hourly urine output should be
monitored carefully.
Although the advantages of prophylactic antibiotics have
not been proved, patients with necrotizing pancreatitis may
benefit.40 Other treatment strategies involving somatostatin,
glucagon, anticholinergics, histamine blockers, and protease
inhibitors have been recommended, but to date they have
not shown conclusive benefit.
Standard medical treatment must be used in patients with
acute pancreatitis in an attempt to control the disorder before
surgical intervention, which is rarely required in children but
is occasionally performed if the diagnosis is uncertain or a
complication develops after an acute episode such as a pseu-
docyst or an abscess. Patients with acute pancreatitis associ-
ated with underlying pancreaticobiliary disease generally
require surgical correction of the underlying condition before
cure can be expected.
In the management of severe acute pancreatitis, a major
decision is whether and when surgery for pancreatic necrosis
or infection is necessary. Infection of necrotic pancreatic tissue
is the major risk factor for mortality in severe acute pancrea-
titis and is an indication for surgery. On the other hand, there
is support for nonsurgical conservative management of sterile
pancreatic necrosis including antibiotic treatment. The
reported death rate is 1.8% with sterile pancreatic necrosis
and 24% with infected necrosis.41 Early surgery has been
associated with increased mortality and should be delayed,
if not avoided.42
CHRONIC RELAPSING PANCREATITIS
Chronic relapsing pancreatitis is characterized by recurrent
episodes of upper abdominal pain associated with varying
degrees of pancreatic exocrine and endocrine dysfunction.
Relatively few case series of chronic relapsing pancreatitis
in childhood exist, although it is more frequently being
suspected in children with recurrent episodes of abdominal
pain.43–45 The disease produces a wide variation of pro-
gressive and irreversible structural changes in the pancreas.44
At present, the most controversial aspect of treatment
of this disorder is the choice of an appropriate surgical
procedure.
Causes and Pathophysiology
In children the most common causes of chronic relapsing
pancreatitis are trauma, heredity, systemic disease, and mal-
formations of the pancreaticobiliary duct such as pancreas
divisum, annular pancreas, and choledocholithiasis. In addi-
tion, various unusual conditions including metabolic disease,
endocrine disorders, and inflammatory bowel disease13 may
cause the disorder. Causes of chronic relapsing pancreatitis
are listed in Table 109-3.
Certain congenital anomalies affect the pancreas and sur-
rounding tissue, several of which are associated with chronic
relapsing pancreatitis. Obstruction of pancreatic flow caused
by a stenotic papilla of Vater and bile reflux resulting from a
common channel are believed to be responsible for the resul-
tant inflammation, which may be reversible if the obstruction
is relieved. A relatively common cause of relapsing pancreatitis
is pancreas divisum,46 in which most of the pancreatic fluid
flows through the minor duct of Santorini, with the possibility
of a relative obstruction to flow because of anatomy and flow
dynamics. Some series have reported an increased incidence
of pancreas divisum in patients with idiopathic chronic pan-
creatitis.46 Annular pancreas, which results from an error of
rotation or fixation of the embryologic pancreatic primor-
dium, has also been associated with chronic pancreatitis
(Fig. 109-1).63 Pancreaticobiliary malunion with or without
choledocholithiasis may likewise result in relapsing chronic
pancreatitis.24,27
In the past few years, several genes have been identified as
being associated with hereditary and idiopathic chronic pan-
creatitis: PRSS1, CFTR, and SPINK1. SPINK1 mutations were
found predominantly in patients without a family history.
These mutations were most common in those with idiopathic
chronic pancreatitis, whereas patients with hereditary chronic
pancreatitis predominantly had PRSS1 mutations.47
TABLE 109-3
Causes of Chronic Relapsing Pancreatitis in Children
Congenital anomalies of the pancreatic duct
Pancreaticobiliary malunion
Pancreas divisum
Annular pancreas
Biliary tract disorders
Choledocholithiasis
Cholelithiasis
Trauma
Systemic infection
Hereditary pancreatitis
Hyperlipoproteinemia
Hyperparathyroidism and hypercalcemia
Cystic fibrosis
Inborn errors of metabolism
Chronic inflammatory conditions
Crohn disease
Ulcerative colitis
Systemic lupus erythematosus
Chronic fibrosing pancreatitis
Juvenile tropical pancreatitis
Idiopathic chronic pancreatitis

FIGURE 109-1 Preoperative endoscopic retrograde cholangiopancrea-
tography in a patient with an annular pancreas. A complex anomaly of
the pancreatic duct with dilatation (arrowheads) and stenosis (arrows) is
shown. The patient underwent a longitudinal pancreaticojejunostomy
(Puestow) procedure.
Diagnosis
The degree of permanent damage to the pancreas may be
assessed by blood tests (pancreatic enzymes), stool tests (pan-
creatic enzymes, fecal fat), and noninvasive tests of pancreatic
function such as the pancreatic stimulation (secretin) test.
In 30% to 50% of adolescents and adults with chronic pan-
creatitis, plain radiographs of the abdomen reveal pancreatic
calcification. Calcification is diagnostic of chronic pancreati-
tis, even if clinical evidence of pancreatic disease is absent.
However, the incidence of calcification in children with
chronic relapsing pancreatitis is low, except in younger
children with hereditary pancreatitis, in whom the incidence
is high.
US is ideal for examining the pancreas in children. Dilata-
tion of the pancreatic or biliary tracts may be noted, and cal-
cification and complications such as pseudocysts, abscesses,
calculi, and ascites can be seen. CT is useful for visualizing
the size of the pancreas and its ducts (Fig. 109-2) and for
detecting small calculi that may be missed on plain radio-
graphs and US. ERCP is an important tool in the diagnosis
and management of chronic relapsing pancreatitis in adults
and children.23 Pancreaticobiliary malunion with or without
dilatation of the bile duct may be detected.24 ERCP is up to
90% accurate in diagnosing ductal abnormalities. However,
as described earlier, MRCP has been greatly refined in the past
decade. MRCP is considered to be equivalent to ERCP for the
diagnosis of many pancreatic and biliary conditions and is
preferable because it is noninvasive and safer.38
Surgical Treatment
Surgery for chronic relapsing pancreatitis is done to relieve pain,
treat complications, or both. Surgery on the pancreas is
generally of three types: (1) sphincteroplasty45; (2) pancreatic
drainage via longitudinal pancreaticojejunostomy (Puestow),48
end-to-end pancreaticojejunostomy (Duval),49 or the Frey pro-
cedure50,51; or (3) partial or total pancreatectomy. Several groups
of investigators have attested to the success of sphincteroplasty
in controlling the symptoms of chronic pancreatitis in carefully
CHAPTER 109 THE PANCREAS 1375
FIGURE 109-2 Computed tomographic scan of a patient with an irreg-
ularly dilated pancreatic duct (arrows). The patient underwent a longitudi-
nal pancreaticojejunostomy (Puestow) procedure.
selected patients. Correct use of this procedure, however,
demands that intrapancreatic ductal obstruction be ruled out
by pancreatography. O’Neill and colleagues45 reported good
results in six of seven children who underwent sphincteroplasty
for chronic pancreatitis.
The general indication for the use of a direct ductal decom-
pression procedure is evidence of pancreatic ductal ectasia
with multiple intrapancreatic duct strictures. An important
secondary feature of drainage procedures is preservation of
existing endocrine function by avoiding major pancreatic re-
section. Longitudinal pancreaticojejunostomy (Puestow tech-
nique) in adults has been successful in eliminating or
ameliorating pain from chronic pancreatitis in 70% to 97%
of patients in series reported 3 decades ago.52,53 In children,
several authors have shown that either the Puestow proce-
dure43,54–56 or the Frey procedure51 improves pancreatic
function, decreases hospitalization, and increases body weight
toward ideal. Others have reported that distal pancreatectomy
and pancreaticojejunostomy are effective.57 Tailored organ-
preserving resective pancreatectomy can also be performed
with low morbidity and mortality in pediatric patients with
chronic pancreatitis and not responding to medical manage-
ment. Total pancreatectomy is not generally indicated in chil-
dren.58 The individual procedure for chronic pancreatitis
should be tailored to the pancreatic ductal anatomy.
PANCREAS DIVISUM
The embryology of this entity was described in the first section
of this chapter. Pancreas divisum is suspected on MRCP or
ERCP when the duct of Wirsung fails to be visualized after in-
jection of the major papilla or when it is attenuated and rudi-
mentary. MRCP is a useful and noninvasive diagnostic tool for
pancreas divisum (Fig. 109-3), but ERCP can provide detailed
information on the pancreatic duct. If pancreas divisum is sus-
pected after injection of the papilla of Vater, an attempt should
be made to find and cannulate the minor papilla, but this is
possible in less than half the cases because of its small size
and the angle at which it meets the duodenum. If the minor
papilla can be entered, ERCP will demonstrate that the duct
1376 PART VII ABDOMEN
FIGURE 109-3 Magnetic resonance cholangiopancreatography in a pa-
tient with recurrent pancreatitis. The duct of Santorini is the dominant
duct, and it extends the entire length of the body and tail of the pancreas
(arrowheads).
of Santorini is the dominant duct and that it extends the entire
length of the body and tail of the pancreas. Imaging of the duct
of Santorini occasionally demonstrates dilatation, irregularity,
or stricture suggestive of chronic pancreatitis.
Autopsy studies have suggested that pancreas divisum,
which is found in 4% to 11% of patients, is the most common
congenital anomaly of the pancreas. When absence of the duct
of Wirsung is associated with pancreas divisum, approxi-
mately 10% of affected patients have anomalous anatomy
resulting in most or all of the pancreatic secretions draining
by means of the accessory papilla.59,60
Pancreas divisum should not be regarded as a disease.
However, if the orifice of the accessory papilla is stenotic, pan-
creatitis can result. Stenosis of the minor papilla is probably
developmental because the orifice is usually small with no
evidence of inflammation. The incidence of pancreas divisum
may be significantly higher in patients with unexplained
recurrent pancreatitis, and it was detected in up to 25% of
patients. Neblett and O’Neill reported that pancreas divisum
was identified in 7.4% of all children with pancreatitis and
19.2% of children with relapsing or chronic pancreatitis. In
10 patients with pancreas divisum, 8 had complete pancreas
divisum and 2 had incomplete variants.61
The primary goal of treatment of pancreas divisum associ-
ated with pancreatitis is to establish adequate drainage of the
duct of Santorini. The progression of disease to pancreatic in-
sufficiency can be arrested when the obstruction is relieved
early. Correction may not only preserve pancreatic function
but may also help ensure that normal growth and deve-
lopment occur.
Several reports have shown that adequate drainage of the
duct of Santorini can be achieved with accessory papilla sphinc-
teroplasty.46,59–61 A report by Adzick and colleagues delineated
the technical details of the operative procedure.46 Endoscopic
sphincterotomy has been performed, but restenosis and recur-
rence of symptoms have been reported. If chronic pancreatitis
has developed in the presence of a dilated duct, longitudinal
pancreaticojejunostomy should be considered. Neblett and
O’Neill61 reported that 8 of 10 patients with pancreas divisum
underwent surgery: 7 underwent transduodenal sphinctero-
plasty of the accessory papilla, along with sphincteroplasty of
the major papilla in 2 (plus septoplasty in 1). Three patients
underwent longitudinal pancreaticojejunostomy, as a primary
procedure in one patient with mid-ductal stenosis and in two
because of recurring pancreatitis after sphincteroplasty without
endocrine and exocrine pancreatic insufficiency. Surgical inter-
vention is directed toward relief of ductal obstruction and may
involve accessory duct sphincteroplasty alone or in conjunc-
tion with major sphincteroplasty and septoplasty. Patients with
more distal ductal obstruction or ductal ectasia may benefit
from pancreaticojejunostomy. In an extremely rare case, Miyano
and colleagues treated a patient with coexistence of pancreas
divisum, choledochal cyst, and pancreaticobiliary malunion.
Figure 109-4 shows the preoperative ERCP and intraoperative
cholangiogram. This patient underwent complete excision of
the choledochal cyst with a Roux-en-Y hepaticojejunostomy,
followed by transduodenal papilloplasty to allow complete
drainage of the common channel.62
Cysts and Pseudocysts
------------------------------------------------------------------------------------------------------------------------------------------------
Pancreatic cysts are relatively unusual in children. The causes
of these cysts tend to be diverse and include both congenital
and acquired conditions. Pancreatic cysts can be classified as
congenital and developmental, retention, enteric duplication,
and pseudocysts.
CONGENITAL AND DEVELOPMENTAL
FORMS
Congenital and developmental cysts of the pancreas are rare and
may be encountered in a fetus, infant, child, or adult. Only 25
cases have been reported in the pediatric literature to date.64–67
These cysts have a female preponderance. They are most com-
mon in the body and tail of the pancreas, are more often uniloc-
ular than multilocular, and are lined with epithelium. The cysts
are usually filled with a cloudy, yellow sterile fluid that has no
enzyme activity, and they are remarkably free of adhesions and
infection. They have been reported to occur simultaneously in
other organs; for example, von Hippel-Lindau disease is charac-
terized by hereditary cerebellar cysts, retinal hemangiomas, and
cysts of the pancreas and other organs.68 The cysts that occur
with cystic fibrosis are not considered pancreatic pseudocysts
and are not discussed in this chapter.
Clinically, congenital and developmental cysts are detected
in the prenatal period as an incidental sonographic finding or
as a cause of polyhydramnios. After birth, they may be man-
ifested as asymptomatic abdominal distention, as vomiting or
jaundice caused by extrinsic pressure on neighboring organs,
or as an asymptomatic mass. Surgical treatment consists of
total excision of cysts located in the pancreatic body or tail
and either internal drainage or complete resection of those
in the pancreatic head.
RETENTION CYSTS
Retention cysts of the pancreas are rare and believed to result
from chronic obstruction of the gland. They contain cloudy
fluid composed of pancreatic exocrine secretions and a high
concentration of pancreatic enzymes. Such cysts are lined with
ductal epithelium unless it has been destroyed by chronic di-
latation or inflammation from enzyme exposure.

A B
FIGURE 109-4 A, Preoperative endoscopic retrograde cholangiopancreatography. B, Intraoperative cholangiogram in a patient with a choledochal cyst
associated with pancreas divisum. The septum (arrowheads) in the intrapancreatic bile duct is also seen.
ENTERIC DUPLICATIONS
Enteric duplication involving the pancreas is rare and usually
associated with gastric duplication.69 This duplication is most
likely to result from failure of regression of an enteric divertic-
ulum formed from the pancreatic duct. Many of the reported
cases of enteric duplication involving the pancreas communi-
cate with the pancreatic duct, are lined with gastric-type epi-
thelium, and contain ectopic pancreatic tissue in their walls.
Pancreatic duplication cysts have been noted on fetal US.70
The most common symptom in these patients is recurrent
abdominal pain, often postprandial. Pancreatitis associated
with enteric duplication cysts is believed to be caused by
obstruction of the pancreatic ducts by viscous secretions from
the cyst or by blood and debris from peptic ulceration within
the cyst. In patients without pancreatitis, pain may be caused
by tension on the wall of the cyst as a result of accumulation
of secretions and muscular contraction.71 CT may help identify
the location and size of duplication cysts, as well as any edema
of the pancreatic head. ERCP or MRCP is useful to outline
ductal anatomy and demonstrate any communication between
the duplication cyst and the pancreatic duct to aid in planning
the surgical approach. To date, virtually all duplication cysts
reported in the literature have been treated by extirpation,
and some have required pancreaticoduodenectomy.69,72
PSEUDOCYSTS
Clinical Features
Pancreatic pseudocysts are localized collections of pancreatic
secretions that do not have an epithelial lining and develop
after pancreatic injury, inflammation, or duct obstruction.
The most common causes of pseudocysts in children are
CHAPTER 109 THE PANCREAS 1377
trauma and infection.73 Drug-induced acute pancreatitis such
as with valproic acid is also a rare cause of pancreatic pseudo-
cyst.74 These cysts typically lie in the lesser sac behind the
stomach and are composed of a fibrous capsule surrounded
by inflamed connective tissue. The capsule of the cyst may also
be formed by neighboring tissue such as the stomach, duode-
num, colon, small intestine, or omentum. Pseudocyst fluid is
clear or straw colored in most cases and may contain tooth-
paste-like debris. The amylase level of cyst fluid is typically
higher than 50,000 Somogyi U/mL.
Diagnosis
The presence of a pancreatic pseudocyst is suggested by a his-
tory of blunt abdominal trauma; an illness resembling pancre-
atitis, possibly followed by a symptom-free interval of weeks
to months; or palpation of a mass in the epigastrium or left
upper quadrant. Abdominal pain is the most common symp-
tom, with jaundice, chest pain, signs of gastric obstruction,
vomiting, gastrointestinal hemorrhage, weight loss, fever,
and ascites also being features.
US, CT, and magnetic resonance imaging (MRI) are helpful
and accurate in diagnosis. These studies are also invaluable for
evaluating the thickness of the cyst wall and for observing
changes in the cyst during the ensuing period of treatment
(Fig. 109-5). ERCP is often useful because it can definitively
determine the status of the pancreatic duct and thus guide
surgical interventions.75,76 Rarely, pancreatic pseudocysts
can extend into the mediastinum in children.77
Treatment
Optimal management of a pancreatic pseudocyst remains
controversial. Treatment options range from conservative
medical management to surgical drainage.78 Conventional
1378 PART VII ABDOMEN
A B
FIGURE 109-5 Computed tomographic scans of a post-traumatic
pseudocyst in the abdomen of a 12-year-old boy. A, The pancreas body
is interrupted (dotted circle). B, A large cyst (arrowheads) developed 2
weeks after injury. C, Percutaneous external drainage (arrow) was
performed.
management involves supportive therapy over a 6-week wait-
ing period, during which time either the cyst resolves sponta-
neously or the cyst wall undergoes fibrous maturation, thereby
permitting internal surgical drainage to the stomach or jeju-
num. This 6-week interval is what has traditionally been ac-
cepted, but CT may demonstrate thickening of the cyst wall
sufficient to hold sutures as early as 3 to 4 weeks. Surgical
drainage is usually performed in adults when necessary but
is controversial in children because some pancreatic pseudo-
cysts in this group resolve without surgical intervention and
have a low risk for recurrence. Pseudocysts from nontraumatic
etiologies are more likely to require surgical intervention,
whereas traumatic pseudocysts are more amenable to nono-
perative treatment.79 Octreotide acetate, a long-acting ana-
logue of somatostatin, may facilitate medical management of
pancreatic pseudocysts.80
A significant risk for complications such as infection or ma-
jor hemorrhage in untreated pseudocysts or persistence of
severe symptoms may be an indication for earlier intervention.
If the patient cannot withstand major surgery, external drain-
age is preferred. There is significant evidence indicating that
internal drainage, especially transgastric cystogastrostomy or
Roux-en-Y cystojejunostomy, is effective in the treatment of
pancreatic pseudocyst. Roux-en-Y cystojejunostomy is the
most widely used internal drainage procedure for this prob-
lem and is associated with the lowest rate of complications
and recurrence.81
C
In contrast to this approach, others82,83 have reported that
percutaneous drainage and endoscopic transmural drainage
are safe and efficient procedures. However, it is generally rec-
ognized that external drainage carries a higher risk for compli-
cations such as fistula formation and a higher recurrence rate
than internal drainage does. Internal drainage procedures can-
not be accomplished until the wall of the pseudocyst has
matured. Recent studies have shown that internal drainage
should be performed in patients with pseudocysts that are
more than 6 weeks old and have a diameter larger than
5 cm because a large proportion of pseudocysts regress during
the first 6 weeks after diagnosis and the risks associated with
managing the pseudocyst are reduced. Cystogastrostomy has
been performed laparoscopically.84
Though less commonly used than drainage procedures
into the stomach or jejunum, cystoduodenostomy is effective
when a cyst is closely adherent to the duodenum. Either
pancreaticoduodenectomy or distal pancreatectomy, as indi-
cated by anatomy, is effective in treating patients with
pseudocysts under ideal circumstances. Distal pancreatic
resection is indicated in patients with a pseudocyst in the
body or tail of the pancreas, particularly if associated with
multiple small cysts. Pseudocysts involving the head and un-
cinate process of the pancreas that are not amenable to inter-
nal drainage are rare and may require proximal pancreatic
resection, but such resection should be done only as a last
resort.
 CHAPTER 109 THE PANCREAS 1379

Hyperinsulinism
------------------------------------------------------------------------------------------------------------------------------------------------
disease, and about 45% have diffuse disease. Distinguishing
focal from diffuse disease is of importance in guiding the ex-
Congenital hyperinsulinism (HI) is a rare derangement of glu- tent of surgical resection. Patients with diffuse disease often
cose metabolism, which carries an estimated incidence of 1 to require near-total pancreatectomy, which has the long-term
1.4 in 50,000 live births, leading to about 80 to 120 new cases risk of diabetes mellitus. Conversely, babies with focal disease
in the United States each year. Higher rates of 1 in 2500 live can be cured with a selective partial pancreatectomy with
births have been reported in areas of high consanguinity such little risk of subsequent diabetes.
as the Arabian Peninsula. Pancreatectomy for management of At the Congenital Hyperinsulinism Center at CHOP, we use
persistent infantile hypoglycemia was first performed at the a multidisciplinary approach (pediatric endocrinology, radiol-
Children’s Hospital of Philadelphia (CHOP) in 1950. Inappro- ogy, pathology, and surgery) for patients with HI to distinguish
priate oversecretion of insulin is the hallmark of HI. The old focal from diffuse disease, localize focal lesions, treat focal dis-
term “nesidioblastosis” should be discarded. HI is the most ease with partial pancreatectomy, and treat medically refrac-
common cause of persistent hypoglycemia in neonates and tory diffuse disease with near-total pancreatectomy. During
can lead to seizures and irreversible brain damage.85 the past 12 years, more than 250 patients (median age 10
weeks) with HI have been treated with pancreatectomy at
CHOP. This section on hyperinsulinism will be based largely
DIAGNOSTIC APPROACH on the clinical experience at CHOP. We have also crafted an
Genetics educational DVD regarding the management of congenital
hyperinsulinism that is available at the CHOP website
Molecular biologic studies have shown that abnormalities of
(www.chop.edu).
the KATP channel, which are encoded by the sulfonylurea re-
ceptor 1 (SUR1) and Kir6.2 genes, are responsible for altered
Diagnosis and Medical Management
control of insulin secretion.85 In response to elevated glucose
by Pediatric Endocrinology
levels, the KATP channel closes, depolarizing the beta-cell
membrane and initiating calcium-dependent release of insulin Babies with HI present with severe and persistent hypogly-
from the beta-cell storage granules. Uncontrolled insulin cemia manifested by seizures, lethargy, apnea, and other
secretion may occur if either the SUR1 or Kir6.2 proteins symptoms resulting from neuroglucopenia.85 The diagnosis
are defective. The SUR1/Kir6.2 form of HI may not be con- of congenital HI is established if fasting hypoglycemia (glu-
trolled with medical therapy such as diazoxide, which acts cose < 50 mg/dL) occurs simultaneously with an inappro-
on SUR1 to suppress insulin secretion, and pancreatectomy priately elevated plasma insulin (>2 mU/mL), low plasma
is often necessary.86 In contrast, surgery is not usually neces- beta-hydroxybutyrate (<2 mmol/L) and free fatty acids
sary in other genetic forms of HI that result from mutations of (<1.5 mmol/L), and an inappropriate glycemic response to in-
glucokinase or glutamate dehydrogenase genes that are travenous glucagon (>30 mg/dL rise in serum glucose level).
responsive to diazoxide treatment. Medical therapy to maintain euglycemia is standardized and
Neonates with HI may have either diffuse involvement of involves high continuous intravenous infusions of glucose
the pancreatic beta-cells or focal adenomatous islet cell hy- as measured by the Glucose Infusion Rate (which is the
perplasia. Mutations of the SUR1/Kir6.2 complex are involved amount of glucose infused in mg/kg/min), frequent oral feed-
in both of these types. Recessive mutations cause diffuse HI, ings, and administration of diazoxide, glucagon, and octreo-
whereas loss of heterozygosity together with inheritance of a tide. Early efforts to distinguish focal from diffuse disease
paternal mutation causes focal adenomatous HI. Patients with involved the injection of intravenous calcium and tolbutamide
diffuse disease have recessively inherited mutations of the (a sulfonylurea) to elicit different types of insulin responses by
SUR1/Kir6.2 complex, whereas patients with focal disease have focal and diffuse disease, but the results were not predictive
normal beta cells, as well as a focal clone of abnormal beta enough to be clinically useful.
cells that are homozygous for the SUR1/Kir6.2 mutation. The Preoperative assessment of babies with HI reveals that they
focal lesions arise by a two-hit loss-of-heterozygosity mecha- are large, are often fluid overloaded due to high intravenous
nism.87 First, there is a specific loss of maternal alleles of the glucose requirements, have hepatic enlargement due to stea-
imprinted chromosome region 11p15 in cells from the focal tosis, may be anemic due to frequent blood sampling, and
lesion but not in the surrounding normal pancreatic cells. Sec- have oral aversion. They are predisposed to central venous
ond, there is a transmission of a mutation of SUR1/Kir6.2 in line sepsis both preoperatively and postoperatively. Octreotide
the paternal chromosome 11p; focal lesions have been linked is the mainstay of medical therapy for HI but can rarely lead to
to non-Mendelian expression of paternally transmitted SUR1 necrotizing enterocolitis because octreotide reduces splanch-
mutation in which there is duplication and reduction to homo- nic blood flow in a dose-dependent manner.88
zygosity of the mutant paternal allele. In the future, molecular
Localization Procedures Performed by Radiology
biology testing of peripheral leukocytes may help differentiate
focal from diffuse disease. However, the search for mutations is Diagnostic radiology tests such as US (both preoperative and
currently of limited use in clinical practice because the process intraoperative), MRI, CT, contrast angiography, and radiola-
takes many weeks and not all mutations are known. beled octreotide scans have all been unsuccessful in identify-
One of the big challenges in diagnosis has been that the ing focal lesions. For insulinoma localization in adults,
diffuse and focal forms of HI are clinically identical. Patients intraoperative saline injection into the pancreas followed by
with either focal or diffuse disease are usually large for gesta- tissue aspiration with rapid insulin measurements has been
tional age, reflecting the effects of HI on fetal growth. We have helpful, but this localization technique is untenable in the
found that approximately 55% of our patients have focal fragile neonatal pancreas.
1380 PART VII ABDOMEN
Two interventional radiology tests have been used in an at-
tempt to differentiate focal from diffuse disease. The Arterial
Stimulation with Venous Sampling (ASVS) technique involves
selective pancreatic angiographic stimulation and venous
sampling using intra-arterial calcium, which stimulates ab-
normal islet cells to release insulin. An immediate rise in
insulin from stimulation in only one artery suggests focal
HI in the corresponding area of the pancreas (gastroduo-
denal artery—pancreatic head; superior mesenteric artery—
uncinate process and neck; splenic artery—pancreatic body
or tail), whereas an insulin rise in all three areas suggests dif-
fuse HI. We and others have also used transhepatic portal
venous catheterization and selective sampling of the pancre-
atic veins (THPVS). Both techniques require that the patient
be off all glycemic medications (5 days for diazoxide, 1 to 2 days
for octreotide) before catheterization under general anesthesia.
THPVS requires that glucose levels be maintained at 50 mg/dL
during the procedure as compared with 60 to 80 mg/dL
for ASVS. For THPVS, the pancreatic venous insulin levels
are compared with simultaneously drawn plasma levels of
insulin and glucose. Both ASVS and THPVS are technically
demanding and have limited specificity and sensitivity for
distinguishing between focal and diffuse disease. These tech-
niques are being replaced by a new PET-CT scan technique
using 18-Fluoro-L-DOPA.
Neuroendocrine cells have an affinity for taking up and
decarboxylating amino acid precursors such as L-dihydroxy-
phenylalanine (L-DOPA). Decarboxylation of the L-DOPA to
dopamine in islet cells allows meaningful localization by
means of PET scanning, using the radioactive isomer 18-
Fluoro-L-DOPA. The isotope is manufactured by the Cyclo-
tron Facility at the University of Pennsylvania on the day of
the PET scan because of the isotope’s short half-life, and it is
administered to patients under an Investigational New Drug
(IND) program approved by the U.S. Food and Drug Admin-
istration. Hopefully, 18-Fluoro-L-DOPA will become commer-
cially available in the future so that it can be used in other
medical centers. The PET results are dramatic and visually
spectacular for preoperative localization of a focal lesion
(Fig. 109-6).89 In more than 140 PET scans for HI, we have
found that PET-CT scans read as showing a focal lesion have
been 100% accurate in localizing a lesion. However, in about
20% of PET scans that were interpreted as showing diffuse dis-
ease, the patient at operative exploration will prove to have a
FIGURE 109-6 Positron emission tomography scan is coregistered with
a computed tomography scan that shows a focal lesion (arrow) in the head
of the pancreas. The kidneys excrete the 18-FluoroDOPA and also light up.
focal lesion that was usually small. This false-negative rate
should decrease with greater clinical experience.
Histopathology
A focal lesion is characterized by a tumor-like proliferation of
islet cells that push exocrine elements aside or haphazardly in-
corporate them (Fig. 109-7). Unlike insulinomas, the focal le-
sion retains the lobular architecture of the normal pancreas,
and exocrine elements usually remain within the lesion.
The lesions often have irregular borders, and the endocrine
cells frequently have enlarged nuclei. Islets outside the lesion
appear normal. Patients with diffuse disease have abnormal is-
lets containing 5% to 10% of cells with enlarged nuclei present
throughout the pancreas. After the surgery, all frozen samples
are processed for routine histology and confirmation of find-
ings on the basis of paraffin-embedded sections and insulin
immunohistochemistry.
Surgery
Open operations are approached in a similar manner using a
transverse supraumbilical laparotomy.90–92 The pancreas is
exposed by an extended Kocher maneuver, entry into the
lesser sac, and mobilization of the inferior border of the pan-
creas. It is not necessary to mobilize the spleen. The pancreas
is inspected under 3.5 loupe magnification in an attempt to
visualize a focal lesion, and the pancreas is palpated. If no focal
lesion is seen, then 2- to 3-mm-diameter biopsies are taken
each from the pancreatic head, body, and tail. Patients with
suspected diffuse HI have intraoperative biopsies to confirm
the diagnosis and then undergo near-total pancreatectomy.
Near-total pancreatectomy (95% to 98%) involves resection
of the entire pancreas leaving only a tiny residual piece of
the pancreas between the common bile duct and the duode-
num. The intrapancreatic course of the common bile duct
should be completely dissected for an adequate near-total
pancreatectomy to be performed.90 For children with diffuse
disease treated by near-total pancreatectomy, a gastrostomy
tube is also placed to make it easier to administer supplemen-
tal glucose or night-time feedings if necessary.
When the biopsies demonstrate normal pancreatic histol-
ogy, a further search for the focal lesion using the preoperative
localization data is conducted. Additional biopsies of suspi-
cious areas are obtained until the focal lesion is diagnosed
by frozen section. Expert pediatric pathologic interpretation
is vitally important. Focal lesions tend to be less than 10
mm in size (although they can be much larger) and are fre-
quently irregularly shaped. Some lesions have octopus-like
tentacles that make imperative the intraoperative confirmation
of clear margins by frozen section analysis. Although focal le-
sions may maintain a lobular structure similar to that of the
normal pancreas, subtle visual clues (ranging from a slightly
reddish color to a marble-like appearance) may permit visual
detection of the lesion intraoperatively and accurate preoper-
ative localization studies greatly facilitate the visual search for
a focal lesion. In some cases the lesion will feel firmer than the
surrounding normal pancreas. However, a tiny focal lesion can
be buried within the pancreas and be impossible to see or feel.
Greater operative experience has led to more frequent intrao-
perative visualization or palpation of a focal lesion. Insulino-
mas differ from focal lesions because they are usually
straightforward to identify intraoperatively and occur in older
children.
 CHAPTER 109 THE PANCREAS 1381

A B

C D
FIGURE 109-7 A, Cut surface of pancreatectomy specimen through a focal lesion (marked by suture). B, Chromogranin A immunolabeling highlights the
architecture of the focal lesion (left) and adjacent normal pancreas (right). Comparison of cytologic features of the endocrine tissue in the focal lesion (C) and
normal pancreas (D). Note enlarged islet cell nuclei in C.

Once the focal lesion is identified, a partial pancreatec-
tomy is performed using frozen sections of margins to ensure
a complete resection. For periductal lesions in the body and
tail, a distal pancreatectomy is performed. With pancreatic
head lesions close to the common bile duct or pancreatic
duct, it can be tricky to excise all of the lesion, particularly
if there are tentacles of diseased tissue that emanate from
the lesion. To ensure complete lesion resection in these
challenging cases, the surgeon should remove most or all
of the pancreatic head followed by Roux-en-Y pancreaticoje-
junostomy to drain the remaining pancreatic body and tail.
In this way, the endocrine and exocrine function of the
remaining normal pancreas is saved. In babies the pancreatic
duct on the cut surface of the transected pancreatic body
is not visible, so the end of the Roux-en-Y jejunal limb is
meticulously anastomosed to the capsule of the pancreatic
body with fine interrupted monofilament suture to effecti-
vely dunk the cut end of the pancreas into the small
bowel lumen. Rarely, a focal lesion in the head will extend
into the duodenal wall in which case a Whipple procedure
may be necessary. Because PET scan localization of focal
lesions has proven to be so accurate, focal lesions in the
body and tail are now resected using laparoscopic tech-
niques. The drawback to the laparoscopic approach is
that there is little tactile feedback to help locate a nonvisible
focal lesion.
Because more than 50% of focal lesions involve the pancreatic
head, subtotal (50% to 75%) distal pancreatectomy is inade-
quate therapy in many of these cases. Our experience with
several referrals who underwent subtotal pancreatectomy
elsewhere with the focal lesion remaining within the residual
pancreas are good examples of this potential pitfall.
1382 PART VII ABDOMEN

Postoperative Care and Follow-up from serous cystadenomas, which are benign; the latter tu-
mors are rich in glycogen and contain little or no mucin.103
Postoperative management has been standardized by a clinical
care pathway including the use of the Glucose Infusion Rate to
quantitate the patient’s glucose requirement.85 Glucose Infu- Serous Cystadenoma and Cystadenocarcinoma Serous cys-
sion Rate (GIR) is calculated as % dextrose  IV rate  tic neoplasms of the pancreas are rare in children and adults,
0.169/Wt in kilograms. For the initial postoperative period, are more often observed in females, and are found mainly
blood glucose values are determined hourly. The GIR begins in the body and tail of the pancreas. These tumors consist
at 2 mg/kg/min immediately postoperatively, is increased to predominantly of small cysts (microcystic adenomas).104 Cal-
a GIR of 5 on the morning of postoperative day 1, and then cification is often demonstrated on CT and US. Serous cysts do
usually advances to a GIR of 8 by the evening of the first post- not need to be excised unless they create a mechanical
operative day. It is not unusual for an intravenous insulin in- obstruction because nearly all of them are benign with no
fusion to be necessary for the first few postoperative days. malignant potential.105 Biopsy is required. Serous cystadeno-
After hospital discharge, a complete response at follow-up is carcinoma is virtually nonexistent in children, and only a few
defined as no requirement for glycemic medications, no con- cases have been reported in adults.
tinuous tube feedings, no diabetes mellitus, and the ability to
tolerate an 18-hour fast without hypoglycemia. Mucinous Cystadenoma and Cystadenocarcinoma Mucin-
In our entire experience, 95% of babies with the focal form ous cystic neoplasms of the pancreas are usually large and
of HI are cured after limited pancreatectomy. The vast majority often multilocular. They form papilla lined with columnar,
had a less than 50% pancreatectomy. For babies with diffuse mucin-producing epithelium; are more often observed in
HI treated with near-total pancreatectomy (95% to 98%), females; and are found mainly in the body and tail of the pan-
about one third require no glycemic medications, one third creas.106 Other features are the association of large blood ves-
require insulin to treat diabetes, and one third require a gly- sels within the capsule and the presence of subepithelial
cemic medication (usually octreotide). Long-term follow-up is hemorrhage. A common differential diagnosis is pancreatitis
necessary for all of these children, particularly with regard to with pseudocyst formation. Mucinous cystadenocarcinoma
neurodevelopmental issues. of the pancreas represents 1% of all malignant conditions of
the pancreas, occurs at an earlier age than conventional solid
pancreatic tumors do, and seems to be approximately half as
Neoplasms common as cystadenoma. The clinical and radiologic manifes-
------------------------------------------------------------------------------------------------------------------------------------------------
tations of mucinous cystadenocarcinoma are similar to those
Pancreatic neoplasms are relatively rare in infants and chil- of mucinous cystadenoma. It is not always possible to differ-
dren. They can be cystic, solid, and benign or malignant entiate mucinous cystadenoma from cystadenocarcinoma
and may or may not be hormonally active. In 1990 Grosfeld93 pathologically. It is also common to find apparently benign
reported 13 cases of pancreatic tumor in children including epithelium103 in the same tumors as malignant epithelium,
5 insulinomas, 2 mucinous cystadenomas, 2 rhabdomyosar- thus suggesting that a malignant focus may develop within
comas, and 4 carcinomas. In 1992 Jaksic reported 6 cases a mucinous cystadenoma. Therefore mucinous cystadenoma
of pediatric pancreatic tumors,94 and now more than 200 should be considered a premalignant lesion that should be
cases have been reported in the English literature. More completely excised after incomplete excision or marsupia-
recently, pancreatic tumor series have been reported by lization.98,103 However, the prognosis for this form of pancre-
Shorter (17 cases),95 Perez (58 cases),96 and Yu (18 cases).97 atic malignancy is significantly better than that for the
Unlike malignant pancreatic tumors in adults, tumors in common typical solid ductal adenocarcinoma because of its
children and adolescents are usually resectable and long-term slow growth and lack of metastatic potential.
survival is likely.
Papillary-Cystic Epithelial Neoplasm
The first cases of papillary-cystic endothelial tumors of the
CYSTIC NEOPLASMS
pancreas were reported by Franz in 1959.107 Since these tu-
Cystic neoplasms of the pancreas are relatively rare.98,99 How- mors were described as solid and cystic acinar cell tumors
ard100 classified these lesions into (1) cystadenoma and of the pancreas by Kloppel and colleagues,108 increasing num-
cystadenocarcinoma, which include benign (microcystic) cysta- bers have been reported in children. They occur predomi-
denoma, benign and malignant mucinous (macrocystic) cys- nantly in girls and young women and are manifested as
tadenoma and cystadenocarcinoma, papillary-cystic epithelial large, encapsulated masses, usually with extensive necrosis
neoplasm, and acinar cell cystadenocarcinoma (not reported and varying amounts of cystic change. Histologically, they
in children) and (2) teratomatous cysts. are composed of solid areas of rather small cells with pseu-
dorosette formation reminiscent of endocrine tumors and cys-
tic areas with papillary structures.109 Immunohistochemically,
Cystadenoma and Cystadenocarcinoma
the tumor cells of solid and cystic pancreatic neoplasms
Cystadenoma of the pancreas is rare in children and adults. contain periodic acid-Schiff-positive granules and often have
Only six cases including one in a newborn have been reported progesterone receptors. Immunologic hallmarks are immuno-
in the pediatric population to date.93,101,102 Cystadenoma and reactivity with alpha-antitrypsin, alpha-antichymotrypsin,
cystadenocarcinoma should be divided into two groups. phospholipase A2, and neuroendocrine markers such as
Because large mucinous cystic adenomas have considerable neuron-specific enolase and synaptophysin. However, they
malignant potential, these tumors should be distinguished are generally free of the usual markers of pancreatic carcinoma

such as carcinoembryonic antigen, CA19-9, and tissue pep-
tide antigen.110 These tumors seem to be associated with a
much better prognosis than the usual type of pancreatic
carcinoma but still have malignant potential.111 The tumor
follows a benign course after resection in most cases. Metasta-
sis or recurrence has occurred in 5% of cases in Japan.112
Thus complete extirpation is necessary because of the slow
tumor progression associated with metastatic disease.113 Liver
metastases have been treated with resection and liver
transplantation.114
Other Cystic Neoplasms in Children
In 1992 Flaherty and Benjamin115 reported a case of multicys-
tic pancreatic hamartoma in a 20-month-old girl. In 1990
Mester and colleagues116 reviewed 10 cases of cystic teratoma
of the pancreas including one of their own. Cystic teratomas
were usually seen as benign extragonadal germ cell tumors in
younger patients; 5 of 10 patients were younger than 11 years.
HORMONALLY ACTIVE TUMORS
Endocrinologically active tumors are usually identified by
their symptoms, and most of them originate in the islet
cells. Beta-cell tumors secrete insulin; alpha cell tumors, gluca-
gon; gamma cell tumors, gastrin; delta cell tumors, somato-
statin; and delta1 cell tumors, vasoactive intestinal polypeptide
(VIP) and possibly substance P and secretin. Glucagonomas
and somatostatinomas have not been identified in children.
Insulinoma
Insulinomas are the most common tumor arising from islet
cells and are usually benign (>90%), solitary (80%) lesions
that occur in children older than 4 years. A plasma insulin-
to-glucose ratio greater than 1 is diagnostic (<0.4 is nor-
mal),117 and the ratio increases with fasting. Concomitant
measurement of C peptide levels may be used to exclude fac-
titious hypoglycemia. CT, US, arteriography, and transhepatic
portal venous sampling can be used to localize an insulinoma.
The tumor is usually discrete and well encapsulated, and most
can be enucleated. The introduction of endoscopic and intrao-
perative US has allowed obscure lesions to be identified. If all
methods of tumor localization are unsuccessful, distal pancre-
atectomy with careful sectioning of the gland is advisable. In
that circumstance, measurement of intraoperative insulin
levels is recommended to avoid missing lesions. Virtually all
infants and children with insulinoma can be cured.
Multiple endocrine neoplasia type I (MEN-I) is character-
ized by endocrine tissues in the gut and pancreas including
insulin-secreting tumors, but expression is usually delayed be-
yond the first decade of life. MEN-I is rare, with a frequency
not exceeding approximately 1 in 10,000. The gene for MEN-I
has been localized to the long arm of chromosome 11 and
functions as a tumor suppressor gene, unrelated to the func-
tional alterations in the SUR1/Kir6.2 components of KATP on
the short arm of chromosome 11.118
Gastrinoma
The second most commonly reported pancreatic hormonal
tumor is the gamma cell tumor. The clinical syndrome related
to gastrinoma is often referred to as Zollinger-Ellison syndrome,
a condition characterized by hypergastrinemia with severe
peptic ulcer disease. This syndrome is rare in children.
CHAPTER 109 THE PANCREAS 1383
Gastrinoma is part of the MEN-I syndrome and is commonly
malignant, multicentric, and metastatic at discovery.93 The di-
agnosis is usually made after recurrent episodes of peptic
ulceration associated with an elevated gastrin level (>500
pg/mL). Calcium infusion and secretin simulation tests are
used for diagnosis. CT, percutaneous transhepatic venous
sampling, and gastrin assay have been useful for localization.
Although total gastrectomy was originally the most common
method of treatment, the development of inhibitors of gastric
acid secretion such as H2 blockers and omeprazole has
changed the direction of treatment of gastrinoma substantially.
Children with gastrinoma generally require lifelong medical
treatment, and somatostatin, although beneficial because it
decreases gastric acid secretion, also inhibits growth hormone
secretion, so it cannot be used long-term. Surgical treatment
should thus be aggressive, especially if a solitary tumor is
found with no evidence of metastasis at the time of laparot-
omy. However, total gastrectomy may still be required for pa-
tients in whom medical treatment has failed or those with
residual tumor or metastatic disease.93
VIPoma
VIPoma in children is far more commonly related to other
neurogenic neoplasms such as neuroblastoma or ganglioneur-
oma than to tumors of primary pancreatic origin. Only a few
cases of VIPoma in children have been reported. The VIP that
is produced by a pancreatic VIPoma probably originates from
neural cells in the islets.117 Patients have profuse, watery
diarrhea associated with hypokalemia and hypochlorhydria
(WDHA syndrome), and metabolic acidosis and prerenal azo-
temia secondary to dehydration are often present. Although
medical therapy includes the use of streptozotocin and so-
matostatin, long-term use of somatostatin is undesirable. Sur-
gical extirpation of the tumor-bearing gland is recommended
whenever possible because 50% of pancreatic VIPomas are
malignant.93
CARCINOMA
Carcinoma of the pancreas is common in adults but rare in
children. The clinical features are different in that obstructive
jaundice is the primary finding in adults, whereas an abdom-
inal mass is usually the initial feature in children. Pancreatic
carcinoma in children can be divided into four groups: is-
let cell carcinoma, adenocarcinoma, pancreatoblastoma, and
miscellaneous lesions.
Islet Cell Carcinoma
Islet cell carcinoma in children may or may not be functional.
Functioning beta-cell and non–beta-cell islet cell carcinoma
occurs in infants and children with hypoglycemia in conjunc-
tion with Zollinger-Ellison syndrome, but it is not associated
with a palpable mass. A beta-cell carcinoma is usually discov-
ered during laparotomy for insulinoma. Four functional islet
cell carcinomas have been described in children; all were
treated by surgical extirpation, with good long-term sur-
vival.119–121 Seven of the eight children with Zollinger-Ellison
syndrome described by Wilson had non–beta-cell carcinoma
with metastatic disease.122
Nonfunctioning islet cell carcinomas are more common in
children than adults. Because these tumors are usually dis-
covered by palpation of an abdominal mass, they are often
1384 PART VII ABDOMEN
diagnosed late, and many affected infants and children have
distant metastatic disease at the time of diagnosis. Most non-
functional tumors are large, solitary lesions that can appear in
any location within the gland. In general, although most of
these tumors grow slowly, aggressive surgical therapy is war-
ranted because of their malignant potential.123
Adenocarcinoma
Adenocarcinoma of the pancreas is rare in children. Vejcho124
reviewed 37 cases of adenocarcinoma in 1993. The clinical
manifestations differ from those of adults in that the in-
cidence of pain with jaundice is lower.121 Abdominal pain
and a palpable epigastric mass are the primary findings in
children. Classification is still controversial and is based on
tumor activity and histopathologic and immunohistochemi-
cal findings. Kloppel classified adenocarcinoma into three
types: acinar cell carcinoma, solid and cystic tumor (papillary
cystic tumor), and pancreatoblastoma.108 Acinar cell tumors
are proportionately more common in children.111 It seems
that children with an acinar cell tumor have a somewhat
better prognosis than do those with an adenocarcinoma of
the duct cell type. The prognosis for children with adenocar-
cinoma of the duct cell type is discouraging and similar to
that for adults. CT and sonography are useful for diagnosis.
For localized lesions, pancreaticoduodenectomy seems to be
associated with a favorable prognosis. Infants and children
tolerate radical resection of the pancreas somewhat better
than adults do and have a lower mortality rate and better
long-term survival.
Pancreatoblastoma
125
In 1977 Horie reported two cases of pancreatic carcinoma,
which they termed pancreatoblastoma, an infantile type of
pancreatic carcinoma. In 1984 Buchino126 reviewed eight
patients with pancreatoblastoma, six of whom survived after
surgical extirpation. Pancreatoblastoma is the most common
pancreatic neoplasm in young children. Pancreatoblastomas
have a better prognosis than adenocarcinoma because they
have an encapsulated organoid structure that does not directly
interfere with the main duct system. The criteria for diagnos-
ing pancreatoblastoma have been described.126–128 Although
there is no agreement on the pathology of these rare lesions,
terminology based on recognizable lines of differentiation
seems preferable (e.g., islet cell, duct cell, acinar cell, undiffer-
entiated lesions).111 The recurrence risk after resection is
high. These tumors can be responsive to chemotherapy and
radiation, but the appropriate role for these modalities is
unknown.
Miscellaneous Carcinomas
The following miscellaneous carcinomas in children have
been reported in the literature: one case of carcinoma simplex;
one case of medullary carcinoma129; six cases of sarcoma
(two lymphosarcomas,129 two with sarcomatous degeneration
from cystadenoma,93,130 and two rhabdomyosarcomas93);
four cases of undifferentiated carcinoma121,126; and two cases
of cylindrical cell adenocarcinoma.131
Acknowledgments
This revised chapter based on the version written by Takeshi Miyano that ap-
pears in the sixth edition.
The complete reference list is available online at www.
expertconsult.com.
 bacterial infections prompted pediatric surgeons to study
the role of nonoperative management of splenic injuries,2 a
development that evolved into the preferred method for
treating both childhood and adult splenic injuries. Another
significant development in the surgical management of splenic
disorders was the initial report of laparoscopic splenectomy in
1991 by Delaitre and Maignien.3 This technique has quickly
become the preferred technique for splenectomy in children.

Embryology and Anatomy

------------------------------------------------------------------------------------------------------------------------------------------------

The splenic primordium is first recognized at the 8- to 10-mm

stage as a mesenchymal bulge in the left dorsal mesogastrium
between the stomach and pancreas. By the 10- to 12-mm stage
a true epithelium is noted, and sinusoids are present with
communication to capillaries. By 4 months of fetal life, the
spleen produces red and white cells; however, this function
ceases later in gestation, and the spleen is rarely the site of clin-
ically significant hematopoiesis in childhood. The anatomic
arrangement of the spleen is consistent with the various func-
tions of the spleen. Blood enters the spleen through splenic
CHAPTER 110 artery segmental vessels that branch into trabecular arteries;
and, after further bifurcations, small arteries enter the white
pulp. The white pulp consists of lymphocytes and macro-
phages arranged as a germinal center around the central artery.
The Spleen Branches off of the central artery deliver suspended particulate
material into the white pulp, an arrangement that may facili-
tate antibody formation in response to particulate antigens.4–7
Katherine A. Barsness and Marleta Reynolds Approximately 20% of the volume of the spleen is white pulp,
and the remainder is the red pulp, consisting of the endothe-
lial cords of Billroth.8 Blood passing through the spleen tra-
verses in either an “open” or “closed” fashion. Ten to 20
percent of blood passes through the closed system directly
Splenic physiology can be broadly categorized into red blood from capillaries into venous sinusoids. The remainder passes
cell maintenance and reservoir and immunologic functions. directly into the red pulp. These cells and other particulate
Children frequently require splenectomy for disease processes material must pass through in an open fashion within the
involving abnormalities in red cell deformability or the pres- cords of Billroth and then migrate into the sinusoids, which
ence of antiplatelet antibodies resulting in excessive splenic enter into trabecular veins.
red cell destruction or platelet sequestration. The vital role
of the spleen as an immunologic organ has not only led to non-
operative management of splenic trauma but also affected the Function
timing of elective splenectomy and renewed interest in partial ------------------------------------------------------------------------------------------------------------------------------------------------

splenectomy for certain disorders. The technique of splenec- The function of the spleen is closely related to the anatomic
tomy has changed significantly with the advent of advanced arrangements and can be divided into red cell maintenance,
laparoscopy. The technique of laparoscopic splenectomy immune function, and reservoir. Derangements in the sub-
and the available evidence comparing it with open splenec- strate entering the spleen often lead to excessive function of
tomy are discussed in the latter part of this chapter. one of the splenic roles, thus leading to the need for splenec-
tomy. The spleen destroys red cells at the end of their life span
and repairs other damaged red cells. Because the spleen selec-
History tively removes abnormal cells such as spherocytes, removal of
------------------------------------------------------------------------------------------------------------------------------------------------
the spleen increases the life span of these cells. As blood passes
The role of the spleen has been the subject of debate for more through the spleen, Howell-Jolly bodies (nuclear remnants),
than 300 years. Billroth is credited with developing the open Heinz bodies (denatured hemoglobin), and Pappenheimer
circulation theory of the splenic microcirculation, and the en- bodies (iron granules) are removed, and after splenectomy
dothelial cords of the red pulp bear his name. The recognition they are noted in red cells on a peripheral smear.
of the spleen as a functional organ in health and disease has The immune function of the spleen is characterized by spe-
been well documented in the twentieth century. One of the cific and nonspecific responses. The specific immune function
more important observations was that of King and Shumaker of the spleen is primarily related to antigen processing. Anti-
in 1952, who noted the susceptibility of infants to infection gens come into contact with macrophages and helper T cells in
after splenectomy.1 The protective role of the spleen in the area of the arterioles of the white pulp. T-cell populations
1385
1386 PART VII ABDOMEN

respond with synthesis of cytokines as part of the cellular

response, and activated T cells then circulate to modulate
the response. A humoral response can occur as antigens come
in contact with macrophages and helper T cells and are then
transmitted to antibody synthesizing cells.9
The nonspecific function involves removal of the particu-
late matter from the blood primarily by macrophages. The
spleen is also the principal source of nonspecific opsonins,
which further activate the complement system, facilitating
the destruction of organisms.
The spleen provides another aspect of the immune func-
tion, by simply serving as a biologic filter. If little antibody
is available for opsonization of bacteria, the spleen assumes
a greater role in clearance of bacteria. This may be the basis
for the age-related differences in postsplenectomy infections
because young children lack adequate antibody response.5
The spleen also serves as a reservoir, which in humans is pri- FIGURE 110-1 Computed tomography scan of a wandering spleen in a
marily limited to platelets. The proportion of platelets in the 7-year-old boy with intermittent abdominal pain and an abdominal mass.
spleen is increased in cases of splenomegaly, occasionally A laparoscopic splenopexy was used to secure a left upper quadrant
leading to thrombocytopenia. The spleen also serves as fixation.
a reservoir for factor VIII.

closed around the splenic hilum and then tacked to the

Anatomic Abnormalities diaphragm and posterolateral abdominal wall.
------------------------------------------------------------------------------------------------------------------------------------------------

ACCESSORY SPLEENS SPLENIC CYSTS

Accessory spleens represent the most common anatomic ab- Primary splenic cysts have an epithelial lining, frequently with
normality and are present in 15% to 30% of children. They a trabeculated internal appearance. They most likely originate
most likely originate from mesenchymal remnants that do from inclusion of the surface mesothelium into the splenic pa-
not fuse with the main splenic mass. The majority (75%) renchyma. They may be asymptomatic or may present as
are located near the splenic hilum or adjacent to the tail of pain,26 rupture,27 abscess,28 or symptoms due to gastric com-
the pancreas; a few are found along the splenic artery, in pression. Symptomatic cysts are usually greater than 8 cm.26
the omentum, mesentery, and retroperitoneum and have even Small (<5 cm) simple cysts can be observed; however, larger,
been noted in the scrotum.10 Of those with accessory spleens, enlarging, or symptomatic cysts require definitive treatment.
86% have one, 11% have two, and 3% have three or more,11,12 Percutaneous aspiration and sclerosis have been used, but re-
with the hilum the most common site of multiple accessory currence is common and partial splenectomy has been re-
spleens. Surgeons must be cognizant of these locations and quired.29 Alcohol sclerosis of a congenital cyst has been
routinely check for their presence because a missed accessory reported with success, although the length of follow-up was
spleen can be a cause of recurrence of immune thrombocyto- not specified.30 Marsupialization can be associated with recur-
penic purpura (ITP) or hereditary spherocytosis. Recurrence rence if an adequate segment of cyst is not removed.31,32 Lap-
with ITP is frequently early,13,14 whereas recurrence with aroscopic partial cyst excision and cyst resection with partial
hereditary spherocytosis has been reported as long as 31 years splenectomy33–40 have been reported but are accompanied
later.15 with a 64% to 88% recurrence rate. Partial splenectomy with
excision of the cyst at the margin with the spleen can be per-
WANDERING SPLEEN formed with an electrosurgical device41–43 or with the use
of stapling devices in an open40,44 or laparoscopic ap-
Wandering spleen is characterized by lack of ligamentous proach.40,45,46 Use of cautery,46 sutures, and omental packing
attachments to the diaphragm, colon, and retroperitoneum. of the defect may also be useful.43 Splenic pseudocysts lack an
The embryologic basis of this is probably related to failure epithelial lining and most frequently occur after trauma.
of development of the splenic ligaments from the dorsal mes- Enlarging cysts and symptomatic cysts should be excised.47
entery.16 Children with this condition most often present with
diffuse abdominal pain secondary to torsion and infarction,
although they may also present with episodic pain and an
ASPLENIA AND POLYSPLENIA
abdominal mass (Fig. 110-1).17–22 Although splenectomy is
required for infarction, splenopexy is preferred in nonis- Asplenia and polysplenia syndromes have many similar
chemic and incidentally detected cases. Splenopexy has been features. Congenital asplenia is usually noted with complex
performed with various techniques including placement in an congenital heart disease, as well as bilateral “right-sidedness”
extraperitoneal pocket,21 use of Dexon mesh basket,22,23 su- including bilateral three-lobed lungs, right-sided stomach,
ture splenopexy,22,24 and colonic displacement with gastro- and a central liver.5 Intestinal malrotation has also been ob-
pexy.22,25 Laparoscopic detorsion and splenopexy can be served.48 Howell-Jolly bodies are noted on peripheral smear,
performed with the use of an absorbable mesh bag that is and infants have the risk of overwhelming infection.
 CHAPTER 110 THE SPLEEN 1387

Polysplenia in which the spleen is divided into multiple children with chronic ITP (with bleeding complications) resis-
splenic masses is often associated with biliary atresia. Other tant to first-line therapies are often referred for splenectomy.55
associated features included preduodenal portal vein, situs Unfortunately, predicting a response to splenectomy has been
inversus, malrotation, and cardiac defects.49 These children difficult. In one study, response to corticosteroids, IVIG, or
have adequate splenic immune function. both predicted a 97% response to splenectomy, whereas fail-
ure to respond to either predicted a 70% failure rate to sple-
SPLENIC GONADAL FUSION nectomy.56 Another study noted the response to IVIG to be a
better predictor of response to splenectomy than corticoste-
Splenogonadal fusion occurs as a result of early fusion roids.57 The most recent study directly challenged these find-
between the spleen and left gonad.50 The remnant may be ings with a 100% response rate among children who were
continuous with a fibrous band or discontinuous with splenic nonresponsive to corticosteroids, with postsplenectomy
tissue attached to the gonad. Another abnormality identified platelet counts inversely related to peak platelet response to
has been ectopic splenic tissue in the scrotum apart from steroids.58 Despite the widely disparate literature, there is a
a gonadal fusion. sustained response rate as high as 80% for children with
chronic ITP treated with splenectomy.59

Indications for Splenectomy

------------------------------------------------------------------------------------------------------------------------------------------------
SICKLE CELL DISEASE
HEREDITARY SPHEROCYTOSIS Sickle cell disease occurs as the result of a substitution of
valine for glutamic acid in the b chain of normal hemoglobin
Hereditary spherocytosis is the most common inherited red A, resulting in hemoglobin S. These red cells become rigid
cell disorder among northern European descendants. Approx- when oxygen saturation decreases, with subsequent capillary
imately 75% of affected children have an autosomal dominant occlusion and shortened red cell life. Children with homozy-
inheritance pattern; the remainder are new mutations or au- gous sickle cell anemia and with hemoglobin sickle cell
tosomal recessive. The attendant membrane defects in ankyrin disease have significant rates of painful crisis and acute chest
or spectrin proteins result in poorly deformable spherocytes. syndrome. Splenic sequestration with rapid drop in hemoglo-
The degree of splenic hemolysis can range from mild to severe. bin and platelet counts can occur and, if severe or recurrent,
Most children are noted to have anemia, an elevated reticulo- merit splenectomy to reduce the chance of further episodes.
cyte count, and a mild elevation of bilirubin. The presence of Postoperatively, children with sickle cell disease are at risk
spherocytes along with a positive osmotic fragility test con- for acute chest syndrome (ACS). ACS is characterized by bas-
firms the diagnosis. In affected children, an aplastic crisis ilar pulmonary infiltrates and is thought to be secondary to
can occur secondary to parvovirus B19 infection, with sup- polymerized deoxygenated sickle hemoglobin, causing local
pression of red cell production and a resultant low hematocrit tissue infarction in the lung. Treatment is supportive.
due to ongoing destruction.51 Splenectomy is usually required
for moderate to severe anemia. Gallstones are common, in-
THALASSEMIA
creasing in frequency with the age of the child and degree
of hemolysis,52 and thus a gallbladder ultrasound is required The thalassemias are a group of disorders characterized by ab-
before splenectomy. Hereditary elliptocytosis is a similar normal production of a or b chains of hemoglobin. Thalasse-
disorder associated with spectrin defects, but most patients mia major (Cooley anemia; b-thalassemia) results in the most
are asymptomatic and without significant hemolysis. severe clinical anemia among this group of disorders. Some
children develop significant splenomegaly with sequestration.
IMMUNE THROMBOCYTOPENIC PURPURA Splenectomy has been used to decrease the need for transfu-
sions in children with severe anemia and those with significant
ITP occurs when antiplatelet autoantibodies, usually IgG, splenomegaly and platelet or white cell sequestration.
bind with platelets, leading to destruction in the reticuloendo-
thelial system. The thrombocytopenia can be transient or can
GAUCHER DISEASE
persist in a chronic phase. The decision to treat a patient with
ITP is based primarily on bleeding complications, not on an Gaucher disease is a metabolic disorder characterized by de-
absolute platelet count. Childhood ITP is usually a self- ficiency of the enzyme b-glucocerebrosidase, which results
limited, acute disorder, and splenectomy is only required in in excessive glucocerebroside in macrophages of the spleen,
chronic cases. First-line therapies include short-course liver, bone marrow, and lungs. It is an autosomal recessive dis-
corticosteroids, which may function by inhibiting the reticu- order found most commonly in Ashkenazi Jews. Splenomeg-
loendothelial binding of platelet-antibody complexes; intrave- aly can be severe, and both partial and total splenectomy have
nous immunoglobulin (IVIG), which competitively inhibits been used to alleviate the hypersplenism and increase the red
the Fc receptor binding of platelets by macrophages; and cell, leukocyte, and platelet counts.60 However, recurrence
Rho (D) immunoglobulin in Rh-positive children, which has been reported after partial splenectomy.
binds to red cells that then saturate the splenic capacity and
spares the antibody-coated platelets. Second-line therapies
include rituximab, which binds the CD20 antigen on B lym-
ABSCESS
phocytes and is presumed to decrease autoantibody produc- Splenic abscesses are rare but have been noted with increasing
tion, and splenectomy. With poor long-term response and frequency with the increase in the number of immunocom-
concerning drug toxicities associated with rituximab,53,54 promised children. Organisms are frequently fungal or
1388 PART VII ABDOMEN

mycobacterial.61 Children present with fever, pain, and posi- as reflected in lower pain medication use; shorter length of
tive blood cultures. An isolated splenic abscess may be treated stay; and improved cosmesis are the main benefits compared
with percutaneous drainage and intravenous antibiotics.62 with open splenectomy. The laparoscopic procedure has a
However, diffuse or multifocal abscesses may be better served steep learning curve for those unfamiliar with advanced lap-
by splenectomy. aroscopic techniques, can be difficult in cases of splenomega-
ly, and is associated with longer operative times than the open
procedure. The efficacy of accessory spleen detection has
Splenectomy been questioned in adult studies,13,14 although comparison
------------------------------------------------------------------------------------------------------------------------------------------------ of open and laparoscopic pediatric series has found similar
PREOPERATIVE IMMUNIZATION detection rates.65–70
The technique of laparoscopic splenectomy has evolved
The risk of postsplenectomy sepsis remains a concern for any with the development of smaller instruments and advanced
asplenic individual. Preoperative immunization has become technology such as the Harmonic scalpel and the Ligasure.
the standard of care for all patients being considered for elec- Although preoperative splenic artery embolization was
tive splenectomy. In case of emergent splenectomy, postoper- reported in many initial adult series,71 complications such
ative immunization is still strongly encouraged. Current as pain, abscess, and retroperitoneal necrosis72,73 have oc-
recommendations include polysaccharide pneumococcal, curred. Some authors have noted no difference in the risk of
conjugate Haemophilus influenzae type b, and polysaccharide bleeding or transfusion requirement. Embolization for spleens
meningococcal vaccinations.63 All vaccinations should be between 20 and 30 cm in length and open splenectomy if the
completed at least 2 weeks before planned splenectomy. In ad- organ is larger than 30 cm have been recommended, but most
dition, yearly influenza vaccination is strongly recommended childhood spleens in cases of splenomegaly are smaller, so this
to reduce the incidence of secondary bacterial pneumonia. appears less relevant in children.73 Hand-assisted procedures
have also been described in adults74 for large spleens, but
OPEN SPLENECTOMY the accessory incision prolongs the stay75 and also appears to
have no role in children.
A left upper quadrant subcostal incision provides excellent Although some initial reports used an anterior approach,
exposure for most spleens even in cases of splenomegaly. most surgeons currently use a lateral approach, as depicted
The spleen is retracted medially, and the splenorenal, spleno- in Figure 110-2. The patient is placed approximately 45 de-
phrenic, and splenocolic ligaments are divided. This can grees left side up with slight elevation of the right flank to in-
usually be accomplished with a relatively small incision, crease access on the left side. The table is rotated to the
and the spleen is removed from the abdominal cavity by patient’s left side for trocar placement and then to the right
delivering the lower pole first, taking advantage of the concave to achieve a right lateral decubitus position for the procedure.
medial surface of the spleen. The gastrosplenic ligament and The general principle is for the assistant to use two grasping
short gastric vessels are divided, followed by the hilar vessels. instruments through the upper abdominal midline sites (see
If the pancreas is close to the hilum, it may be necessary to A and B in Figure 110-2) to elevate the spleen and provide
divide the segmental vessels. A careful search is made for ac- traction on surrounding tissues. In young children and older
cessory spleens, and, if present, they are removed. An alterna- patients with small spleens, 3-mm instruments are used
tive open approach uses a lateral muscle-splitting incision,64 at those sites and placed directly into the abdominal cavity
which has been associated with a 2.7-day length of stay, through small stab incisions without the use of trocars. If
comparable with some laparoscopic series.65–67 the spleen is large, at least one of the upper midline trocars
must accommodate 5-mm instruments. For most of the pro-
LAPAROSCOPIC SPLENECTOMY cedure the surgeon uses the Harmonic scalpel or Ligasure
through the left lower quadrant incision. The umbilical port
Technique
is most commonly a 15-mm port to accommodate the
The laparoscopic technique for splenectomy was initially 15-mm diameter endosurgical bag and the endovascular sta-
reported in 1991 3 and over the subsequent 5 years has be- pling devices. Use of the 5-mm 30-degree telescope allows its
come the preferred method at most institutions. Less pain, use through the left lower quadrant site, which is necessary

c
a b

FIGURE 110-2 Patient position

and port size and location for lateral
approach: a and b, 3 or 5 mm mid-
line; c, umbilical 12 or 15 mm; d, 5
mm left lower quadrant.
 CHAPTER 110 THE SPLEEN 1389

when stapling and for placement of the endosurgical bag
through the umbilical site.
The splenocolic ligament is divided, allowing the splenic
flexure to fall away from the spleen (Fig. 110-3, A). The gas-
trosplenic ligament is then divided (see Fig. 110-3, B). The di-
vision of the most superior aspect is difficult, owing to the
close proximity of the spleen and stomach, and care must also
be taken to avoid injury to the diaphragm. Accessory spleens
identified at the hilum can often be removed en bloc with the
spleen and others from the lesser sac or omentum removed
separately, usually through the umbilical trocar.
The spleen is then held up from the superior and inferior
poles and the hilum approached, often dividing the inferior
set of segmental vessels with either an electrosurgical device
or with clips and scissors. The surgeon must decide between
the use of clips or an endovascular stapler for the remaining
segmental or polar vessels. If the vessels are individually
divided, the surgeon must be prepared, should bleeding
occur, to control the vessel proximally with a grasper and
either clip the vessel or staple the remainder of the hilum. If
the stapler is used, the splenorenal ligament can be divided
so that the posterior arm of the stapler can easily pass posterior
to the hilum. Care must be taken at this step to avoid injury to
the tail of the pancreas (see Fig. 110-3, C).
The endosurgical bag is placed through the umbilical port;
the spleen is then placed in the retrieval bag, and the neck of
the bag delivered out the umbilicus. A finger and ring forceps
are placed through the external opening and used to fracture
the splenic capsule and remove splenic fragments (see
Fig. 110-3, D). The use of an ultrasonic morcellator and lipo-
suction device76 to fragment the spleen has been described.
Transvaginal removal of the bag has been described.77 In some
cases of splenomegaly it may be difficult to place the spleen
in the bag. Hebra and colleagues78 used a coring device to

A B

D
FIGURE 110-3 Division of the splenocolic (A) and gastrosplenic (B) ligaments with the Harmonic scalpel. The endovascular stapler is used to divide the
splenic hilum (C) and, after placement in the retrieval bag, delivered out the umbilical port (D). Finger fracture and forceps are used to remove splenic
fragments.
1390 PART VII ABDOMEN
remove portions of the spleen before its placement in a bag.
This is not recommended in cases of hereditary spherocytosis
or ITP in which splenosis could potentially lead to recurrent
anemia or thrombocytopenia.
At completion, the fascia at the umbilical site and, if con-
venient, the anterior fascia layers at the 5-mm sites are closed.
The orogastric tube and Foley catheter used intraoperatively
are removed at the end of the procedure. Diet is advanced
as tolerated the day of surgery, and pain control is achieved
with intravenous morphine and nonsteroidal antiinflamma-
tory agents, followed by oral pain medications when the
patient is tolerating liquids. The techniques of laparoscopic
splenectomy are applicable to other procedures including
partial splenectomy for disease or lesions such as cysts, sple-
nopexy for wandering spleen, and treatment of traumatic
splenic injuries.79,80 Single incision or single port techniques
are being developed.
Conversion
Conversion to open splenectomy is most frequently due to
splenomegaly or bleeding and is reported in 1.3% to 2.8%
in several large series.68,81–84 Giant spleens with malignancy
are more difficult situations, and one author noted a 41%
conversion rate in adults with malignancy compared with a
3% conversion rate for benign conditions.85 In another adult
series, Mahon and Rhodes noted that their conversion rate was
0% for spleens weighing less than 1 kg and 60% for those
weighing more than 1 kg.86 A lower conversion rate with
increasing surgeon experience has been observed in some
series.72,87
Operative Time
Comparative series have uniformly observed longer operative
time with the laparoscopic technique,65–70 and several have
noted shorter times with increased surgeon experience.82,88
In a review of 112 laparoscopic splenectomies, a longer oper-
ative time (115 vs. 98 minutes) was noted in the first 50
procedures compared with the next 62 procedures but the dif-
ferences were not significant.82
Pain
The pediatric reports do not use pain scores but rely on pain
medication usage as a reflection of pain. All available compar-
ative studies report lower morphine equivalent dose usage for
pain control for laparoscopic compared with open splenec-
tomy, but the difference did not reach statistical significance
in two of the five studies.66,69,70,89,90 This is consistent with
an adult study that noted equal quality of life and control of
the hematologic disease with less pain in the laparoscopic
group.91
Length of Stay
Length of stay ranges from 1.3 to 3.6 days, with more studies
reporting 1.7 to 1.8 days.68–70 All pediatric comparative stud-
ies have noted a shorter length of stay with the laparoscopic
technique (laparoscopic, 1.3 to 3.6 days; open, 2.5 to 4.9 days),
with the reduction ranging from 18% to 56%.65–70 The under-
lying disorder can affect length of stay because infants and
children with sickle cell disease have been noted to require a
longer stay than those with ITP and hereditary spherocytosis
(2.5 vs. 1.4 days).70 This has been due to the occurrence of
postoperative acute chest syndrome.
Splenomegaly alone, in children, does not appear to ad-
versely affect length of stay because splenomegaly with hered-
itary spherocytosis is the most frequent indication for
splenectomy and is associated with a short length of stay
(1.4 days). A meta-analysis of primarily adult reports (2940
patients; 2119 laparoscopic procedures, 821 open proce-
dures) noted longer mean operative time (laparoscopic, 180
minutes; open, 114 minutes; P < 0.0001) but a shorter hos-
pital stay (laparoscopic, 3.6 days; open, 7.2 days; P < 0.001)
with laparoscopic splenectomy.92 Unproven benefits of the
laparoscopic approach concern the child’s return to full activ-
ity and the effect of return to school on the parent’s return to
work. Children undergoing laparoscopic splenectomy have
been observed to return to full activity 1 to 5 weeks earlier
than with the open procedure.67 In addition, parents who
had themselves undergone splenectomy view the laparoscopic
technique better than the open technique.
Cost
The presumption of most authors is that the shorter length of
stay with laparoscopy will offset the higher operative charges
related to the longer operative time and the use of disposable
materials. Data from comparative studies note three studies
with higher67,90,93 and three with lower65,69,70 hospital
charges with laparoscopy. The method by which individual
hospitals determine charges or cost for supplies, operating
room time, and inpatient days makes comparison difficult.
The available cost data do not appear to confer an advantage
to either procedure.
Complications
A meta-analysis of 26 comparison studies (adult and pediat-
ric) noted a total complication rate of 15.5% for laparoscopic
and 26.6% for open splenectomy (P < 0.0001).92 The laparo-
scopic group had fewer pulmonary, wound, and infectious
complications (P < 0.0001) but had more hemorrhagic com-
plications when conversions for bleeding were included.
PARTIAL SPLENECTOMY
Total splenectomy with removal of accessory spleens elimi-
nates splenic hemolysis and sequestration for congenital he-
molytic anemias. However, concerns of overwhelming
postsplenectomy sepsis, particularly in children younger than
5 years of age, have led to the exploration of partial splenec-
tomy. This has primarily been used for hemolytic anemias
such as hereditary spherocytosis but has also been used for
Gaucher disease,60 Hodgkin disease staging,94 trauma,95
hypersplenism with cystic fibrosis,96 as well as excision of
splenic cysts,46 hamartomas,97 and hemangiomas.98 When
this technique is used for splenic volume reduction, 85% to
95% of the disease-affected splenic mass is removed, leaving
approximately 10% to 25% of a normal splenic remnant.
Anatomic studies of the splenic blood supply indicate that
86% of spleens have two lobar vessels and 12% have three
lobar arteries.99 In addition, 77% have either four or five seg-
mental vessels, thus allowing preservation of either an upper
or lower pole segmental vessel supplying 20% to 25% of the
spleen. The spleen is divided with either a stapling device, cau-
tery, or the Harmonic scalpel, and hemostasis is achieved with
cautery, argon beam coagulation, or topical agents as needed.
If the spleen is totally mobilized, the remnant can be fixed to
 CHAPTER 110 THE SPLEEN 1391

the retroperitoneum to prevent torsion. This procedure can be Antibiotic prophylaxis is recommended for all asplenic in-
performed laparoscopically using the Harmonic scalpel to both dividuals in the immediate postoperative period. Penicillin
divide the spleen and provide hemostasis (Fig. 110-4). prophylaxis is given to all children, except for those who
Long-term follow-up has noted increased red blood cell are penicillin allergic. In penicillin-allergic children, options
half-life, higher hemoglobin level, and lower transfusion re- are limited. Trimethoprim-sulfamethoxazole can be used, rec-
quirement, reticulocyte count, and bilirubin levels.100–103 ognizing higher pneumococcal resistance with this antibiotic.
Preservation of splenic phagocytic function has been demon- There are no universal recommendations on the duration of
strated by elimination of Howell-Jolly bodies on peripheral antibiotic prophylaxis, but most recommend prophylaxis for
smear and decreased numbers of pitted red cells.101 Other pa- at least 2 years after splenectomy.63
rameters have included normal IgM and IgG levels of specific an-
tibody titers to Streptococcus pneumoniae. Splenic regrowth
occurs in all patients with hemolytic anemias, although the de-
POSTSPLENECTOMY SEPSIS
gree of regrowth does not correlate with hemolysis. Laparo-
scopic partial splenectomy has higher intraoperative blood The occurrence of overwhelming postsplenectomy infection
loss, morphine use, and longer times to full oral intake, with lon- (OPSI) was first reported by King and Shumacker in 1951
ger hospital stays when compared with laparoscopic total sple- in a series of five infants younger than 6 months of age under-
nectomy in the same institution.104 Complications of postpartial going splenectomy for hereditary spherocytosis.1 All five
splenectomy hemolysis can include ongoing development of developed serious infections, and two died. Since then numer-
cholelithiasis (7% to 22%), recurrent anemia with need for ous reports have documented this risk and the increased risk
transfusion, or subsequent total splenectomy.101–103,105 Caution (60- to 100-fold) in children younger than 5 years of age, as
must be exercised in the assumption of normal splenic function well as during the first few years after splenectomy. The etiol-
because one patient with evidence of residual splenic function ogy is probably related to decreased clearance of encapsu-
by scan after partial splenectomy for trauma died of an lated bacteria111 and decreased immunoglobulin levels.112
S. pneumoniae infection 13 years later.106 This patient had not The spleen is particularly important in infancy and childhood
received pneumococcal vaccine. Preoperative immunization is because the phagocytic function in infection occurs almost
recommended for all partial splenectomy candidates, and post- exclusively in the spleen.113
operative antibiotics are administered at least until splenic A cumulative review of the literature reported in the
immunologic competence is noted. 1970s114 identified the incidence of OPSI at 3.75%, with a
mortality of 1.7%. The rate in pediatric cases was 4.1% com-
pared with 1.9% in adults and a mortality rate of 1.8% in chil-
Postoperative Considerations dren compared with 1.1% identified in adults. Other studies
------------------------------------------------------------------------------------------------------------------------------------------------
in the 1980s and 1990s had a wide range of OPSI, and as of the
Complications are relatively rare after splenectomy. Thrombosis mid-1990s the overall incidence of OPSI ranged from 0.13%
of the splenic, portal, and mesenteric veins has occurred to 8.1% in children younger than 15 years of age compared
after splenectomy, with a reported incidence of 1.6% to with 0.28% to 1.9% in adults.115 S. pneumoniae was causative
11%.107–110 Symptoms include fever, vomiting, and/or ab- in 50% to 90% of all infections and is responsible for 60% of
dominal pain as early as 2 days postoperatively. The diagnosis all fatal infections.114,116–118 H. influenzae accounted for 32%
can be made by Doppler ultrasonography or contrast- of mortality,119 and the meningococcus and group A Strepto-
enhanced CT of the abdomen. Treatment consists of antiplate- coccus were also significant pathogens.
let and antithrombotic therapy. Recannulization of the portal The risk of late infectious mortality also varied with the dis-
vein can occur with prompt diagnosis and treatment of the ease state with a rate of 0.51% for spherocytosis and 2.67% for
thrombus, but long-term complications of portal hypertension ITP.120 Others also noted a higher mortality in children youn-
have also been described.110 ger than 4 years of age (mortality: 8.1% in those younger than
4 years of age, 3.3% in older patients) with most deaths within
4 years of splenectomy.121,122 In view of this, penicillin
prophylaxis was advocated. The increased risk in children rel-
ative to adults was documented with an incidence of fatal
OPSI of 3.77% in children compared with 0.39% in adults.123
Many of these early reports were performed before widespread
vaccination and use of prophylactic antibiotics, and more
recent studies have somewhat lower rates of OPSI of 3.5%
to 3.8%.124,125 Vaccination has been shown to decrease the
risk of bacteremia.126
A more recent population-based study out of Denmark de-
termined that among all patients undergoing splenectomy
(adults and children), excess risk of bacteremia was greatest
in the first 90 days after splenectomy, occurring in 10% of
all patients.127 The risk remained higher than the general
population for the first year. Different than previous studies,
the Denmark study found enteric rods to be the predominant
FIGURE 110-4 Laparoscopic partial splenectomy using the ultrasonic organism isolated from the blood in early and late post-
dissector. splenectomy infections.
1392 PART VII ABDOMEN
A comparison study of an early era of splenectomy with no
immunizations or prophylactic antibiotics compared with a
later era with a 70% immunization and a 100% prophylaxis rate
noted a decrease in the infection rate (6% to 3.8%) and mortal-
ity (3.9% to 0.9%).124 Even in this more recent study the rate of
infection was related to age at splenectomy (birth to 5 years,
13.8%; >5 years, 0.5%). The historical mortality associated
with OPSI has been 50% to 70%,115,118,128,129 but more re-
cently a mortality of around 10% has been observed.124,130
The complete reference list is available online at www.
expertconsult.com.
 called glial cell line–derived neurotrophic factor (GDNF). This
protein binds to a tyrosine kinase c-ret receptor on the caudal
wolffian duct, inducing the formation of the ureteric bud. The
ureteric bud elongates and penetrates the nearby mesenchy-
mal blastema, where it undergoes multiple generations of
bud divisions. Mesenchymal (blastemal) cells cluster around
the tip of each terminal branch and differentiate into epithe-
lial, stromal, and endothelial cells, the building blocks of
the metanephric kidney. The epithelial cells first form into a
hollow ball (renal vesicle), which elongates into a prenephron
tubular structure that eventually attaches to the terminal
branches of the ureteric bud, the future collecting ducts.1
Decreased ureteric bud branching will be associated with
fewer clusters of mesenchymal cells and therefore fewer neph-
rons (i.e., hypoplasia). Poor communication between the
branching bud tips and the clusters of mesenchymal cells or
absence of continuity between the collecting ducts and devel-
oping nephrons can lead to persistence of immature structures
(i.e., dysplasia). Overexpression or lack of expression of spe-
cific growth factors can lead to hyperproliferative renal disor-
ders such as Wilms’ tumor and renal cystic disease. For
example, the aforementioned gene IGF2 is overexpressed in
most sporadic Wilms’ tumor and WT1, a tumor suppressor
CHAPTER 111 gene, is mutated in many syndromal Wilms’ tumors.

Renal Dysgenesis
Renal Agenesis, ------------------------------------------------------------------------------------------------------------------------------------------------

Maldevelopment of the kidney that affects its size, shape, or

structure is referred to as renal dysgenesis. There are three prin-
Dysplasia, and cipal forms: hypoplasia, dysplasia, and cystic dysplasia (as
well as various other forms of renal cystic disease). Hypoplasia

Cystic Disease refers to a kidney or a segment of a kidney with a decreased

number of nephrons. The hallmark of dysplasia is the pres-
ence of primitive ducts, structures thought to represent earlier
Kenneth I. Glassberg and Grace Hyun stages of development and characterized by a surrounding
collar of smooth muscle and collagen but lacking elastin.2
Sometimes cysts appear in dysplastic kidneys or in areas of
dysplasia. The cysts may be microscopic or macroscopic, focal
or diffuse. When the entire kidney is involved with both
Renal Development dysplasia and cysts, it is referred to as a multicystic dysplastic
------------------------------------------------------------------------------------------------------------------------------------------------
kidney. Multicystic dysplastic kidneys may involute with time
More than 400 genes have been identified as playing a role in to become a “nubbin” of tissue; this is referred to as renal
renal development. A defect in any of the more important aplasia. The latter represents a tiny cluster of cells with no re-
genes can lead to specific abnormalities of development such niform shape. Kidneys that are severely dysplastic have either
as renal dysplasia, hypoplasia, or cystic disease. The kidney immature or absent collecting systems.
develops as a result of cross-talk between the wolffian duct
(later, the ureteric bud) and the future metanephric blastema;
each sends messages to the other, sometimes simultaneously Renal Agenesis
and sometimes in response to the message received. Each ------------------------------------------------------------------------------------------------------------------------------------------------

step of renal development is dependent on the expression A defect of the wolffian ducts, the ureteric bud, or the meta-
of different genes and their protein products. For example, nephric blastema can lead to absence of kidney development,
the mesenchymal cells of the metanephric blastema persist or renal agenesis. Renal agenesis occurs bilaterally in 1 of every
and proliferate in response to insulin-like growth factor-2 4000 births, with a predominance in males.3 Because of the
(IGF-2). Expression of the Wilms’ tumor 1 gene (WT1) sup- lack of urine production, oligohydramnios develops. Affected
presses IGF-2 and allows a cascade of events to occur includ- infants are born with immature lungs, pneumothorax, and
ing development of the ureteric bud from the wolffian duct Potter facies (hypertelorism, prominent inner canthal folds,
and conversion of mesenchymal cells into epithelial cells. This and recessive chin). Bilateral renal agenesis can also appear
latter conversion is initiated by the PAX2 gene. as part branchio-oto-renal syndrome.4 It is universally fatal.
Ureteric bud formation from the wolffian duct occurs in Unilateral renal agenesis is compatible with life; its incidence
response to a protein produced by the mesenchymal cells is 1 in 450 to 1000 births.5 It is difficult to accurately determine

1395
1396 PART VIII GENITOURINARY DISORDERS

the true incidence of unilateral renal agenesis because in some TABLE 111-1
conditions, such as a multicystic kidney, the organ can regress Classification of Cystic Disease of the Kidney
into a nubbin of tissue that does not appear on imaging studies. Genetic
Unilateral renal agenesis is associated with other urologic Autosomal dominant (adult) polycystic kidney disease
abnormalities including primary vesicoureteral reflux (28%), Autosomal recessive (infantile) polycystic kidney disease
obstructive megaureter (11%), and ureteropelvic junction Juvenile nephronophthisis–medullary cystic disease complex
obstruction (3%).6 Either unilateral renal agenesis or renal Juvenile nephronophthisis (autosomal recessive)
ectopia can be associated with ipsilateral müllerian defects Medullary cystic disease (autosomal dominant)
and vaginal agenesis; this association is referred to as Mayer- Congenital nephrosis (familial nephritic syndrome; autosomal
Rokitansky-Küster-Hauser syndrome. recessive)
Familial hypoplastic glomerulocystic disease (autosomal dominant)
Multiple malformation syndromes with renal cysts (e.g., tuberous
Hypoplasia and Hypodysplasia
------------------------------------------------------------------------------------------------------------------------------------------------
sclerosis, von Hippel-Lindau disease)
Nongenetic
With hypoplasia, an entire kidney or a segment of one has a de- Multicystic kidney (multicystic dysplastic kidney)
creased number of nephrons. A hypodysplastic kidney also has Benign multilocular cyst (cystic nephroma)
a decreased number of nephrons; it may have a reniform shape Simple cyst
but contains areas of dysplasia. Kidneys associated with ectopic Medullary sponge kidney
orifices may be totally dysplastic, hypoplastic, or hypodysplas- Sporadic glomerulocystic kidney disease
tic. Ectopic orifices may be secondary to an abnormal ureteric Acquired renal cystic disease
bud site on the wolffian duct and are associated with abnormal Caliceal diverticulum (pyelogenic cyst)
renal development.

sclerosis (TS) . Although multicystic kidneys often present as

an abdominal mass, they are rarely removed. The large cysts in
Renal Cystic Disease ADPKD have been treated with laparoscopic unroofing, but this
------------------------------------------------------------------------------------------------------------------------------------------------
author is unaware of any unroofing procedures performed in
Renal cysts may be congenital, sporadic, or acquired; single or children.
multiple; unilateral or bilateral. The majority arise from neph-
rons and collecting ducts. Multicystic dysplastic kidneys con-
tain cysts that form from tubular structures before the
nephrons develop. Benign multilocular cysts are a form of
Autosomal Dominant (Adult)
neoplasia and must be differentiated from possible malignant Polycystic Kidney Disease
tumors. The origin of simple cysts is not clear. Most develop in ------------------------------------------------------------------------------------------------------------------------------------------------

kidneys that are otherwise normal. ADPKD was formerly referred to as adult polycystic kidney dis-
Renal cysts vary in size from microscopic to macroscopic. ease because it is seen predominantly in adults; however, it
All cysts are lined by epithelial cells. Some actually represent may manifest in childhood or even in utero. ADPKD is the
ectatic tubules or collecting ducts and remain continuous with most common cause of renal failure, accounting for 10% to
the nephron. Others may pinch off from the nephron and be- 15% of patients requiring kidney dialysis and transplanta-
come isolated structures. According to Gardner,7 any duct that tion.9 Its incidence ranges from 1 in 500 to 1 in 1000 individ-
is dilated to four times the normal diameter is considered a cyst. uals. The condition affects both kidneys, although it may
The classification system used in this chapter is based present predominantly on one side before it manifests on
on that proposed by the Committee on Terminology, Nomen- the contralateral side.
clature, and Classification of the Section on Urology of the
American Academy of Pediatrics, which divides the various
GENETICS
conditions into genetic and nongenetic disease.8 Some mod-
ifications based on more recent findings have been added ADPKD has been associated with three different mutations:
(Table 111-1).1 PKD1 on chromosome 16 accounts for approximately 90%
From the outset, it must be clarified that although the terms of cases,10,11 PKD2 on chromosome 4 accounts for 5% to
multicystic and polycystic both mean “many cysts,” they describe 10% of cases,12,13 and PKD3 has not yet been characterized.
different conditions. The former refers to a dysplastic entity, Penetrance is thought to be 100%, and 50% of the offspring
and the latter refers to a number of separate entities, most of of affected individuals have the potential of developing the
which are inherited, most of which are associated at some point disease. ADPKD will manifest in 96% of affected individuals
in life with cysts in communication with nephrons or collecting by age 90 years.12,14
ducts, but not all of which are associated with dysplasia. The
most common conditions associated with the term polycystic
CLINICAL FEATURES
kidney disease are autosomal recessive polycystic kidney disease
(ARPKD) and autosomal dominant polycystic kidney disease Because 50% of them will eventually develop the disease, the
(ADPKD). These conditions may progress to renal failure. offspring of patients with ADPKD are screened by ultrasonog-
The entities with more important surgical implications in- raphy. By age 25 years, at least 85% of individuals will have
clude the various forms of multilocular cysts and the neoplasms cysts, but most will be asymptomatic.15 Family members of
associated with von Hippel-Lindau disease and tuberous individuals with ADPKD are unable to obtain life insurance
 CHAPTER 111 RENAL AGENESIS, DYSPLASIA, AND CYSTIC DISEASE 1397

before age 25 years because of the possibility of developing the

GENETICS
disease. The majority of individuals present between 30 and
70 years of age with findings that include microscopic and The disease is transmitted by PKHD1, a gene located on chro-
gross hematuria, flank pain, hypertension, and renal colic sec- mosome 6, with possible involvement of the mTOR pathway,
ondary to either clots or stones. In adults, the disease is asso- as recently described.17,18 Both the severe and mild forms of
ciated with colonic diverticula, hepatic cysts, and berry ARPKD are associated with this mutation. Because it is an au-
aneurysms. Hepatic cysts appear much more frequently in tosomal recessive trait, one of four offspring is affected and
adults than in children and are more common in females. neither parent shows evidence of the disease.
ADPKD occasionally presents in utero or in infancy; when
it does, 50% of the kidneys are large and usually contain
CLINICAL FEATURES
macrocysts. Sometimes the large cysts are not apparent, and
the kidneys are hyperechogenic due to the reverberation of ul- For infants with ARPKD identifiable in utero or at birth, death
trasound waves caused by the multiple interfaces between di- in the first 2 months of life from uremia or respiratory failure is
lated ducts and tubules; the latter presentation is more typical not unusual. If they survive the first month of life, their
for ARPKD (Fig. 111-1). Eventually, the cysts enlarge to mac- chances of living for a year with proper supportive therapy
roscopic size. When prenatal ultrasonography identifies a pa- are improved. Eventually, all infants with ARPKD develop re-
tient suspected of having ADPKD or ARPKD, genetic or nal failure (Fig. 111-2). In children who present later in life,
ultrasound studies of the parents must be obtained. It is im- hypertension presents later and is more controllable and
portant to monitor these children for hypertension because progression to renal failure is slow.
control of blood pressure delays the onset of renal failure. Newborns with ARPKD may have significant nephromegaly.
Those with elevated serum creatinine levels, increased urinary The kidneys can be so large that they compromise breathing or
protein excretion, and high blood pressure at a young age tend cause a difficult delivery. When severe, oligohydramnios may
to decline faster and go on to renal failure. Almost all patients be present, as well as Potter facies and deformities of the limbs.
who present in utero or in the first year of life eventually de- Rarely, the kidneys are so enlarged that the diaphragm is signif-
velop renal failure, but with close monitoring, the onset can be icantly elevated, resulting in respiratory compromise; in these
delayed. cases, nephrectomy may be required. Respiratory care helps
extend the child’s life.
Those with advanced congenital hepatic fibrosis may have
Autosomal Recessive (Infantile) portal hypertension, esophageal varices, and hepatosplenome-
galy. Variceal bleeding due to portal hypertension may require
Polycystic Kidney Disease a splenorenal shunt. The kidneys in utero are typically large
------------------------------------------------------------------------------------------------------------------------------------------------
and hyperechogenic on ultrasound examination. Hyperecho-
This entity is no longer referred to as infantile polycystic kidney genicity is due to the presence of numerous small microscopic
disease because it is now realized that some patients can pre- cysts and dilated ducts and tubules within the kidney, creating
sent as adolescents or in their early 20s. ARPKD is a relatively reverberations of the sound waves. With time, macrocysts de-
rare disease occurring in approximately 1 of every 40,000 live velop. As these patients get older, their kidneys get smaller,
births.16 All patients have some degree of congenital hepatic contrary to the sequence of events in ADPKD. On histologic
fibrosis. In general, younger patients with severely affected examination, the collecting ducts are elongated and very
kidneys have the mildest congenital hepatic fibrosis. dilated. There is no cure for the condition.

B
FIGURE 111-1 A, Multiple cysts are seen throughout the left kidney on this ultrasound study in a newborn with autosomal dominant polycystic kidney
disease. B, Newborn with autosomal recessive polycystic kidney disease. Note the increased echogenicity and the small cysts. (Courtesy Walter Berdon, MD.)
1398 PART VIII GENITOURINARY DISORDERS

A B

C D
FIGURE 111-2 Autosomal recessive polycystic kidney disease. A, Gross specimen with diffuse, small subcapsular cysts. B, Radially arranged, elongated,
“cystic,” dilated collecting ducts. C, On low-power microscopic view, dilated collecting ducts are seen. D, On intravenous urography, the characteristic
delayed “sunburst” pattern is identified, secondary to contrast pooling in the dilated collecting ducts.
 CHAPTER 111 RENAL AGENESIS, DYSPLASIA, AND CYSTIC DISEASE 1399

Multiple Malformation Fortunately, AMLs can be diagnosed via computed tomogra-

phy (CT) scan. The presence of fat, 20 HU or lower, within
Syndromes with Renal Cysts a renal lesion on a CT scan is pathognomonic for an AML, ex-
------------------------------------------------------------------------------------------------------------------------------------------------
cluding the diagnosis of RCC.26–28 Life-threatening hemor-
A number of multiple malformation syndromes are character- rhage is the most significant complication of AMLs. Risk
ized by renal cysts. Two that are of particular interest to factors have been identified specifically, tumors larger than
surgeons are TS and von Hippel-Lindau disease. 4 cm, pregnancy, multicentricity, and TS. Therefore proactive
intervention should be considered in these patients.29 Fur-
TUBEROUS SCLEROSIS thermore, a nephron-sparing approach, by either selective
embolization or partial nephrectomy, is clearly preferred in
TS is associated with the classic triad of epilepsy (80% of patients with small AMLs requiring intervention because of
cases), adenoma sebaceum (75% of cases), and mental retar- symptoms, in patients with TS or multicentric AML, and in
dation (60% of cases).19 Adenoma sebaceum consists of flesh- patients for whom preservation of renal function is critical.
colored papules of fibroma located on the malar area of the
face. The hallmark lesion is a superficial cortical hamartoma
VON HIPPEL-LINDAU DISEASE
of the cerebrum, which looks like a hardened gyrus, suggest-
ing the appearance of a tuber (root). Hamartomas affect other This autosomal dominant condition is associated with cerebel-
areas of the body as well, especially the kidneys and eyes. lar hemangioblastomas, retinal angiomas, cysts of the pan-
Other kidney lesions include angiomyolipomas and cysts. creas, epididymis and kidney epididymal cystadenoma,
pheochromocytoma, and clear cell renal cell carcinoma. Its
Genetics
incidence is approximately 1 in 35,000.30,31
The trait is transmitted in 25% to 60% of cases and occurs ei-
Genetics
ther sporadically or as a genetic condition with variable or in-
complete penetrance. Two genes have been identified as being The VHL gene is a tumor suppressor gene located on chromo-
responsible for the autosomal dominant transmission of some 3 and has a dominant transmission; 50% of offspring of
TS.20,21 TSC1 is located on chromosome 9, and TSC2 on chro- affected persons can expect to develop the disease. Individuals
mosome 16. The latter gene defect is located at a site contig- are born heterozygous for the disease initially. When the sec-
uous with PKD1, the mutation most commonly associated ond allele of the VHL gene mutates in a specific organ, the typ-
with ADPKD.50 The fact that both TSC gene mutations result ical lesions develop in that site. Different mutations of the VHL
in similar clinical manifestations suggests that both genes are gene have been found in cases of renal cell carcinoma not
involved in a common developmental pathway. When either associated with von Hippel-Lindau disease.
gene does not produce its normal protein product, similar re-
Clinical Features
sults occur. Both are considered tumor suppressor genes
because a mutation of either can result in tumor development. The mean age of presentation is 35 to 40 years; rarely, von
Hippel-Lindau disease presents in childhood. Renal cysts are
Clinical Features
the most common renal lesion and are the first manifestation
Twenty percent of patients develop renal cysts, and these usu- of the disease in 76% of patients.32 Pheochromocytomas
ally manifest before 3 years of age. One third of children who occur in 10% to 17% of affected individuals but seem to be
present with cysts are younger than 1 year. The disease may be more common in specific families.33,34 Some patients present
detected on prenatal ultrasonography because of the presence with seizures because of hemangioblastomas of the central
of cysts. When it does present in utero, there is a greater like- nervous system. Because the cells lining the cysts are hyperplas-
lihood that it is secondary to a mutation of the TSC2 gene on tic, these patients have a high risk of developing renal cell
chromosome 16. In such cases, defects of the adjacent PKD1 carcinoma (RCC).31 In adulthood, the possibility of bilateral
gene responsible for ADPKD may also be present. The renal renal cell carcinoma should be considered, even if it is not
cysts rarely cause a problem. In the past, many of these pa- bilateral initially. Annual or biannual examinations of the kid-
tients died early of central nervous system lesions; however, ney, usually by CT, are recommended after age 30 years, partic-
they are now living longer, and in the future, it may be discov- ularly when small tumors, representing benign adenomas, are
ered that renal cysts cause compromised kidney function. On present. RCC is responsible for 30% mortality rate for those
histologic examination, the cysts have a unique appearance patients with VHL.35
and are lined by hypertrophic, hyperplastic eosinophilic cells.
In theory, these hyperplastic cells may account for the 2%
incidence of renal cell carcinoma in patients with TS. Patients
as young as 7 years have been reported with this tumor. Multicystic Dysplastic Kidney
------------------------------------------------------------------------------------------------------------------------------------------------

This condition represents an extreme form of dysplasia involv-

ing the entire kidney (Fig. 111-3). When occurring in one
ANGIOMYOLIPOMA
pole of a kidney, it is likely associated with a duplicated sys-
Renal angiomyolipoma (AML) is a benign hamartomatous tem. When patients have the disease bilaterally, it is incompat-
neoplasm consisting of varying amounts of mature adipose tis- ible with life and is therefore associated with stillbirth,
sue, smooth muscle, and thick-walled vessels and occurs in oligohydramnios, and Potter facies. The condition is usually
40% to 60% of patients with TS.22–24 These lesions are rarely associated with atretic ureters and no renal pelvis. When a re-
seen before 6 years of age and are common after age 10.25 nal pelvis is present, the entity is referred to as a hydronephrotic
1400 PART VIII GENITOURINARY DISORDERS
FIGURE 111-3 Typical multicystic kidney with atretic ureter. Note the
“bunch of grapes” appearance. (From Shah SH, Glassberg KI: Multicystic
dysplastic kidney disease. In Gearhart JP, Rink RC, Moriquand PDE [eds]:
Pediatric Urology. Philadelphia, WB Saunders, 2001, p 279.) (See color plate.)
form of multicystic kidney (Fig. 111-4). When the kidney ap-
pears solid, with fewer cysts, the disease is referred to as solid
cystic dysplasia. Typically, however, the appearance is that of a
bunch of grapes, and on ultrasound studies the kidney has a
nonreniform shape; the cysts have a haphazard appearance,
with no central or medial cyst, suggesting the presence of a
renal pelvis (Fig. 111-5). This form of multicystic kidney must
be differentiated from hydronephrosis, in which there is a cen-
tral or medial cyst—the largest—and there is communication
with peripheral cysts, representing dilated calices. In the past,
it was thought that the cysts do not communicate; however,
injection of contrast into one cyst invariably demonstrates
communication between cysts (Fig. 111-6).15
FIGURE 111-4 Hydronephrotic form of multicystic dysplastic kidney.
The largest cyst is located centrally and medially, and the smaller cysts
are at the periphery.
FIGURE 111-5 Typical ultrasound appearance of a multicystic kidney.
Three large cysts are seen, with some smaller cysts in between. Little pa-
renchyma is identified.
CLINICAL FEATURES
Multicystic dysplasia is the most common form of renal cystic
disease, and it along with ureteropelvic junction obstruction
are the most common causes of a palpable abdominal mass
in an infant. It is important to evaluate the contralateral system
because 3% to 12% of infants with multicystic kidney disease
have a contralateral ureteropelvic junction obstruction, and
18% to 43% have contralateral vesicoureteral reflux.11,36–38
Because of the latter finding, voiding cystourethrography
should be obtained in all infants diagnosed with the entity.
Historically, these kidneys were detected in the first year of life
by palpation. Before the use of ultrasonography, most multi-
cystic kidneys were removed because it was hard to confi-
dently differentiate them from renal tumors and because of
rare reports of Wilms’ tumors being identified in multicystic
kidneys. To help differentiate a multicystic kidney from a
FIGURE 111-6 Multicystic dysplastic kidney. Contrast material injected
into one cyst is identified moments later in all cysts, demonstrating a duct-
like communication between them. Although the cysts do not appear to
communicate on ultrasonography, they are invariably found to communi-
cate when contrast material is injected into excised specimens.
 CHAPTER 111
A B
FIGURE 111-7 A, Multicystic kidney identified on ultrasonography in a newborn. B, At age 9 months, the cysts have shrunken, and the kidney is barely
visible. (From Shah SM, Glassberg KI: Multicystic dysplastic kidney disease. In Gearhart JP, Rink RC, Moriquand PDE [eds]: Pediatric Urology. Philadelphia, WB
Saunders, 2001, p 284.)
hydronephrotic kidney, a dimercaptosuccinic acid (DMSA)
scan is obtained. Little, if any, function exists in a multicystic
kidney, whereas a hydronephrotic kidney rarely has no func-
tion. The decision must be made whether a multicystic dys-
plastic kidney requires removal. With ultrasonography and
other imaging studies providing a more definitive evaluation,
the incidence of nephrectomy for an inconclusive diagnosis
has fallen dramatically. Most cases of Wilms’ tumor in multi-
cystic dysplastic kidneys have occurred in kidneys that were
not previously known to be multicystic. Thus it is unclear
whether the multicystic dysplastic kidney led to the develop-
ment of the Wilms’ tumor. Two reports in the literature that im-
ply that these kidneys should be removed describe a total of
120 patients, 5 of whom were found to have nephrogenic rests
of nodular renal blastoma.39,49 One had Wilms’ tumorlets in
the hilar region. However, renal blastomas have been reported
to occur in normal kidneys as well, so it is not clear that the
presence of these rarely seen tumorlets in multicystic kidneys
implies a greater risk of developing Wilms’ tumor. According to
Beckwith,40 only 5 of 7500 Wilms’ tumor specimens in the
Wilms’ Tumor Registry occurred in multicystic kidneys.
Beckwith calculated that multicystic kidneys have a fourfold
increased risk of developing Wilms’ tumor, and this low inci-
dence does not justify prophylactic nephrectomy. He pointed
out that even if Wilms’ tumor did develop, the current survival
rate is now greater than 90% with treatment. Zerres and col-
leagues17 and Aslam and colleagues41 suggest conservative
therapy for multicystic kidneys as long as there is follow-up
ultrasound surveillance every 3 months until 8 years of age.
They calculated a $2000 to $5000 cost for such surveillance,
in comparison with a simple nephrectomy at $5000 to $7000.
There have been isolated case reports of hypertension de-
veloping in association with a multicystic kidney, with re-
moval of the kidney curing the hypertension. In the
National Multicystic Kidney Registry of the Section on Urology
of the American Academy of Pediatrics, most neonatal cystic
kidneys became smaller or stayed the same size over a 5-year
period, with a small percentage becoming larger.16 Therefore
many kidneys that become smaller will disappear from view
on ultrasonography. This does not mean that the kidney itself
has disappeared; it implies that the cysts have disappeared
RENAL AGENESIS, DYSPLASIA, AND CYSTIC DISEASE 1401
(Fig. 111-7). The cells that lined the cysts persist, however;
this amorphous group of cells forms a nubbin of kidney with-
out any function (renal aplasia).
Although it is customary to follow multicystic kidneys with
ultrasound studies every 3 to 6 months during the early years
of life and at progressively longer intervals until 8 years of age,
it is not clear whether it is necessary to follow a patient whose
kidney disappears on imaging studies, or even when those
findings persist. Because these cells remain without the cyst
fluid, their presence, by itself, should not determine whether
to obtain ultrasound follow-up.
Benign Multilocular Cyst
(Cystic Nephroma)
------------------------------------------------------------------------------------------------------------------------------------------------
A benign multilocular cyst is a neoplasm of the kidney that is
not associated with renal dysplasia. The lesions are large and
circumscribed by a thick capsule. There is usually normal renal
parenchyma adjacent to the lesion. The loculi, which represent
cysts within the lesion, can measure a few millimeters to a few
centimeters in size and do not intercommunicate. The cysts are
lined by cuboidal or low columnar epithelial cells. The septa of
a benign multilocular cyst are composed of fibrous tissue, and
no poorly differentiated tissues or blastema cells are present.
It is important to note that when a multilocular cystic lesion
is identified on an imaging study, particularly ultrasound, it is
impossible to determine whether it is a benign multilocular
cyst, a multilocular cyst with partially differentiated Wilms’ tu-
mor, a benign multilocular cyst with nodules of Wilms’ tumor,
or a cystic Wilms’ tumor. Thus multilocular cystic lesions can
be considered a spectrum of entities, and there have been no
reports of one entity converting into another (Fig. 111-8).
Because imaging studies cannot differentiate them, it is essen-
tial that these lesions be removed (Fig. 111-9). If they are lo-
cated in one pole of the kidney and there is a large remaining
portion of normal parenchyma present, a partial nephrectomy
should be considered. Even when the lesion is not a typical
benign multilocular cyst, there is little chance of the disease
being aggressive. None of these conditions involves expansive
nodules, and metastasis rarely if ever occurs.42 The lesion
1402 PART VIII GENITOURINARY DISORDERS

should not be confused with a Wilms’ tumor that contains ne-

Benign Multilocular Cyst
crotic areas, which can simulate cysts on ultrasonography; the
latter represents a more aggressive tumor.43 The age distribution
is bimodal, occurring primarily in boys during the first 2 to 3
Multilocular Cyst years of life, and again after the fourth decade, usually women.42
with Children tend to present with an asymptomatic abdominal mass
Multilocular Cyst
Nodules of detected on routine physical examination, whereas symptomatic
with
Wilms’ Tumor
Partially Differentiated presentation is more common in adults.44
Wilms’ Tumor

Cystic Wilms’ Tumor

Simple Cyst
------------------------------------------------------------------------------------------------------------------------------------------------

FIGURE 111-8 Spectrum of multilocular cystic lesions in childhood. Be-
cause imaging studies cannot differentiate one type of lesion from an-
other, the diagnosis must be made under the microscope.
Simple cysts are discrete findings within the kidney that usu-
ally appear oval to round with a smooth border. The cysts usu-
ally contain clear or straw-colored fluid, and when they
manifest in utero, they usually disappear by birth. When they
appear during childhood, they should be evaluated just as in

A B

C
FIGURE 111-9 Multilocular cyst with partially differentiated Wilms’ tumor. A, Ultrasound image identifies a mass in the kidney with cystlike structures
within it. B, On contrast-enhanced CT, a large, smoothly outlined renal mass is seen with fine septa within it; note the caliceal system compressed on its
edge. C, Cross section of the mass reveals noncommunicating loculations. D, On microscopy, nests of blastema cells with tubular elements are seen be-
tween the loculi, findings typical of multilocular cyst with partially differentiated Wilms’ tumor.
 CHAPTER 111 RENAL AGENESIS, DYSPLASIA, AND CYSTIC DISEASE 1403

adults with simple renal cysts. Because cysts were once con-
sidered such an infrequent finding in children, and because
the ability to clearly define a cyst was limited, most simple re-
nal cysts were operated on in the past. In children, renal cysts
are rarely symptomatic. When more than one simple renal cyst
is present in a kidney, the patient should be followed and the
contralateral kidney carefully evaluated with imaging studies,
because unilateral multiple cysts may represent an initial
asymmetric presentation of ADPKD.
The sonographic criteria for a classic benign simple cyst are
as follows:
1. Sharply defined, thin, distinct wall with smooth and dis-
tinct margin
2. Absence of internal echoes
3. Good transmission of sound waves through the cyst with
acoustic FIGURE 111-10 Two simple renal cysts in a 7-year-old. Note the white-
4. Spherical or slightly ovoid shape (Fig. 111-10) ness (hyperechogenicity) just outside the kidney on ultrasonography. This
If all these criteria are met, the chance of malignancy is neg- is referred to as acoustic enhancement, a finding typical of simple cysts. It
ligible.44 Sometimes there is a cluster of renal cysts at one pole occurs secondary to increased sound wave reverberation through the cyst,
and it interfaces with the kidney. It is important not to confuse this finding
of a kidney, and this must be differentiated from a multilocular with an atypical, asymmetric presentation of autosomal dominant polycys-
cystic lesion. When the cysts are all simple, the condition may tic kidney disease.
be referred to as unilateral renal cystic disease, a variant of simple
cysts.45 Although follow-up recommendations have not been
established with evidence-based reports, it seems reasonable
to follow these patients with sequential ultrasonography with narrow neck. It has also been referred to as a pyelogenic cyst;
decreasing frequency, mainly to ensure the stability of the simple that term, however, is best used when the lesion communi-
cyst and screen for an asymmetric presentation of ADPKD. cates with the renal pelvis. These diverticula have the same lin-
ing as calices, with a smooth layer of transitional epithelium.
Caliceal diverticula are benign lesions, and small diverticula
Acquired Renal Cystic Disease are usually asymptomatic and discovered as an incidental
------------------------------------------------------------------------------------------------------------------------------------------------
finding. The diagnosis is best made by excretory urography
Acquired renal cystic disease is common in dialysis patients or CT with delayed images. Whereas simple cysts do not take
with chronic renal failure. There are few reports on acquired up contrast, calyceal diverticula will fill with contrast specifi-
renal cystic disease in children, but as in adults, the incidence cally on the delayed images. Asymptomatic patients do not
increases with the length of time on dialysis.31 Young patients require treatment. Over time these diverticula tend to progres-
exposed to dialysis may develop the condition later in child- sively distend with trapped urine. Infection, hematuria, flank
hood. Once renal cysts develop, the kidneys should be fol- pain, milk of calcium (crystallization of calcium salts without
lowed regularly with imaging studies to rule out tumors. actual stone formation), and true stone formation are compli-
After renal transplantation, the cysts regress in size. New cysts cations of stasis or obstruction that can produce symptoms46
can form in transplanted patients, but they tend to be small. and are indications for surgical intervention. Percutaneous re-
moval of the stones and ablation of the mucosal surface47 and
ureteroscopic enlargement of the diverticular communication
Caliceal Diverticulum with removal of the stones48 are the modern treatment options
------------------------------------------------------------------------------------------------------------------------------------------------
available.
A caliceal diverticulum is a cystic, smoothly outlined sac, usu-
ally located at one of the poles, most frequently the upper one. The complete reference list is available online at www.
It communicates with the pelvicaliceal system by means of a expertconsult.com.
 embryonic remnants left such as the paradidymis in the male and
the paroöphoron in the female.3 The metanephros develops
from the portion (metanephric blastema) of the nephrogenic
cord caudal to the mesonephros.
The ureteral buds arise from the mesonephric duct near its
junction with the cloaca during the fourth to fifth week of ges-
tation. The cranial end of the ureter then ascends to meet the
nephrogenic cord, which begins to develop into the meta-
nephros and continues its cephalad migration. The cranial
end of the ureteral bud begins a series of branchings to form
the renal pelvis, the calices, and a portion of the collecting
ducts.4 The cephalad migration of the kidneys ends at the
eighth week of gestation, as well as their axial rotation of
90 degrees medially. During their ascent, the kidneys receive
their blood supply from the neighboring vessels. Blood is ini-
tially supplied by the middle sacral artery, then the common
and inferior mesenteric arteries, and finally the aorta. Renal
development is closely related to the location of the origin
of the ureteral buds.5 The ureteral ducts (which in the first
phases of development are permeable until the 35th day) sub-
sequently undergo a process of obstruction and recanalization
of their lumen until the embryo is 41 days of age.6,7 The mus-
cular and neurologic construction and maturation of the uri-
CHAPTER 112 nary excretory system are a slow phenomenon (Fig. 112-2)
that, when impaired, may affect the entire anatomic organiza-
tion of the urinary tract. Any delay or mislocation of these
complex events may lead to the following renal anomalies8:
Renal Fusions and renal agenesis, multicystic dysplastic kidney and other renal
cystic diseases, small congenital kidney, supernumerary kid-
ney, and renomegaly. Anomalies related to abnormal ascent
Ectopia and abnormal fusion9 are discussed in the following sections.
Many genes are likely to control all these events, although few
are identified.10 Renal metanephric development results from
Pierre Mouriquand and Nicoleta Panait the expression of many genes in the ureteral bud and meta-
nephric blastema with each sending messages to the other to
induce organogenesis.11 The transcription factor BF212 stud-
ied in the mouse model seems to play an important role in this
embryonic construction.
Embryology
------------------------------------------------------------------------------------------------------------------------------------------------

The embryology of congenital anomalies of the kidneys in- Anomalies Related to Abnormal
volves two protagonists and three events. The two protagonists
are the nephrogenic material (pronephros, mesonephros, Ascent of the Kidneys (Ectopic
metanephros) and the ureteral buds (wolffian structures).
The three events are induction of metanephric tissue (renal
Kidneys)13
------------------------------------------------------------------------------------------------------------------------------------------------

construction), ascent of the kidneys (cephalad migration),

PELVIC KIDNEY
and positioning of the kidneys (partition and rotation). Inci-
dents occurring during these three phases (first 8 weeks of ges- A pelvic location (sacral or pelvic) below the aortic bifurcation
tation) may affect the presence, number, and location of the is the most common site of ectopic kidneys.14 A pelvic kidney
kidneys. (Fig. 112-3) is generally unilateral with a slight predilection
The nephrogenic cord is divided into three parts (Fig. 112-1)1: for the left side. It is found bilaterally in 10% of cases.15 In ad-
The most cranial segments collectively constitute the proneph- dition to its abnormal location, a pelvic kidney is frequently
ros; the intermediate segments, the mesonephros; and the small, with an irregular shape, variable rotation, and extrarenal
most caudal segments, the metanephros or kidney. This clas- collecting system.
sification is purely topographic, and the nephrogenic material
arises from the same source and exhibits identical properties
LUMBAR OR ILIAC ECTOPIC KIDNEY
throughout its craniocaudal extent.2 The pronephros (eight-
somite–stage embryo) degenerates completely (day 24 to 25); These kidneys are fixed above the crest of the ileum but are
however, its duct contributes to the formation of the meso- below the level of L2 and L3. Such kidneys can be difficult
nephros. In an embryo with 23 somites, the mesonephros pro- to differentiate from an ptotic kidney, which has a normal ure-
duces small quantities of diluted urine and forms mesonephric teral length, is not malrotated, is mobile, and can usually be
vesicles and excretory tubules, which also degenerate with a few manipulated into its normal position.15
1405
1406 PART VIII GENITOURINARY DISORDERS

⫻106 mature nephrons

renal mass grammes

FETAL
RENAL Nephrogenesis
complete
DEVELOPMENT continued growth

1.2 Collecting ducts

B complete 30
C

Ureteric bud

Histologic
First nephron connexions
bladder-urethra

20
PUV
Type I
0.6

PUV
Type II
10

0.0
10 20 30 40
Pro Weeks
MesoNeph

FIGURE 112-2 Chronology of development of the urinary tract. PUV,

posterior urethral valves.

DIAGNOSIS OF ECTOPIC KIDNEYS

FIGURE 112-1 Schemes showing early development of the human
nephros. A, Ten-somite embryo. B, Twenty-somite embryo. C, Twenty-
five-somite embryo. The Roman numerals indicate the level in somites.
The circles indicate nephrotomes that do not contribute to the nephros
(based on Torrey85). (From Hamilton WJ, Boyd JD, Mossman HW [eds]:
Human Embryology, 3rd ed. Cambridge, England, Heffer, 1962.)
THORACIC KIDNEY
Thoracic kidneys (Fig. 112-4, A and B) are rare (1 in
16,000)16–19 and may be related to delayed mesonephric invo-
lution (i.e., persistence of the nephrogenic cord).20 Thoracic
kidneys may be associated with cardiovascular, pulmonary,
diaphragmatic,21,22 and spinal anomalies.23,24
Ectopic kidneys may be an incidental finding that requires no
surgery or may be associated with impaired drainage or reflux
that may require surgical correction. Isotope studies, antenatal
and postnatal ultrasonography, magnetic resonance imaging
(MRI) (Fig. 112-5), and intravenous urography usually allow
identification of the ectopic renal parenchyma. These investi-
gations can be complemented by retrograde contrast studies
when required. Computed tomography (CT) has been
reported as a possible method for identifying ectopic renal
tissue, especially in the thoracic cavity.17 An ectopic kidney
can be suspected antenatally, especially if the abnormality is
associated with some degree of dilatation.

ASSOCIATED ABNORMALITIES
Anomalies Related to Abnormal
Ectopic kidneys are often associated with genital and contra-
Fusion of the Kidneys
------------------------------------------------------------------------------------------------------------------------------------------------

lateral urinary abnormalities such as the absence of a vagina,25

retrocaval ureter,26 bicornuate uterus, supernumerary kid-
ney,27 and contralateral ectopic ureter.28 An ectopic kidney
can be a component of a more complex syndrome such as
the Mayer-Rokitansky-Küster-Hauser syndrome,29–31 Fanconi
anemia,31,32 or conjoined twins.33 Cephalad kidneys have
also been reported in patients with omphaloceles.34
HORSESHOE KIDNEY
A horseshoe kidney is the most common fusion defect of the kid-
neys. The incidence of horseshoe kidneys (Figs. 110-6, A and B
and 110-7, A and B) varies from 1 in 400 35 to 1 in 1800 36
autopsies. The condition is more common in male patients.
Horseshoe kidney and other renal fusion abnormalities such

FIGURE 112-3 Pelvic kidney. (Courtesy Professor I. Gordon, Department
of Radiology, Great Ormond Street Hospital, London.)
as crossed ectopia have been reported to occur in more than one
member of a family including apparently identical twins. These
findings suggest a genetic influence.37–39 In 95% of cases, the
lower poles of the two kidneys are joined by a bridge of renal
tissue that can be either normal renal tissue or dysplastic or fi-
brous tissue. In approximately 40% of cases, the isthmus lies
at the level of L4, just beneath the origin of the inferior mesen-
teric artery; in 20% of cases, the isthmus is in the pelvis; and in
the remaining cases, it lies at the level of the lower poles of nor-
mally placed kidneys.25 Although the isthmus usually passes an-
terior to the great vessels, it may pass posterior to the aorta or
inferior vena cava.40 It is postulated that the inferior mesenteric
artery obstructs the isthmus and prevents further ascent. A few
horseshoe kidneys have fusion at their upper poles. Ureters arch
anteriorly to pass over the isthmus.41 The blood supply of horse-
shoe kidneys varies considerably, a fact that must be borne in
mind at the time of surgery: 30% of patients have a renal blood
supply that consists of one renal artery for each kidney. However,
the supply may be asymmetric, with duplicate or even triplicate
vessels. The isthmus may be supplied by the aorta or by the renal,
inferior mesenteric, common or external iliac, or sacral arteries.
In addition, the entire blood supply may enter through the isth-
mus.42 Duplication of the horseshoe kidney and communicating
renal duplication have also been mentioned in previous re-
ports.43,44 One part of the horseshoe kidney may be dysplastic
and nonfunctioning.45
Associated abnormalities are common and occur in 78% of
cases.46,47 The most common involve the central nervous sys-
tem, gastrointestinal tract, and skeletal and cardiovascular sys-
tems. In children with trisomy 18, the incidence of fused
kidneys is 20%.48,49 More than 60% of patients with Turner
syndrome have renal abnormalities50 including horseshoe
kidneys (7%).51,52 Horseshoe kidneys can be complicated
later in life by aortic aneurysms.53
CHAPTER 112 RENAL FUSIONS AND ECTOPIA 1407
B
FIGURE 112-4 A and B, Thoracic kidney (intravenous urography).
(Courtesy Professor I. Gordon, Department of Radiology, Great Ormond
Street Hospital, London.)
Horseshoe kidneys are diagnosed with the same investiga-
tions mentioned earlier; isotopic studies usually give the clear-
est information. On intravenous urography, the diagnosis is
clearly indicated by malrotation of each kidney with a projec-
tion of the calices medially to the renal pelvis, a renal axis that
is more vertical or shifted outward, and the low position of the
excretory cavities. The course of the ureters varies, but their
upper segments seem to be laterally displaced by the paren-
chymal bridge. Association with a dilated renal pelvis is com-
mon and due to either anterior projection of the ureter (which
can be kinked) or obstruction of the pyeloureteric junction
caused by an anomaly of the junction itself or the disposition
of the blood vessels around the junction. It is usual in these
cases for the horseshoe kidney to be associated with some de-
gree of impaired urine flow, and thus the diagnosis can be sus-
pected antenatally. Vesicoureteric reflux is also commonly
associated with horseshoe kidneys (10% to 80% of
cases).43,52,54
1408 PART VIII GENITOURINARY DISORDERS
FIGURE 112-5 Magnetic resonance image of a pelvic kidney (arrow).
(Courtesy Professor J. P. Pracros, Department of Radiology, Debrousse
Hospital, Lyon, France.)
A B
FIGURE 112-6 Horseshoe kidney. A, Intravenous urogram. B, Dimercaptosuccinic acid [DMSA] scan. (Courtesy Professor I. Gordon, Department of
Radiology, Great Ormond Street Hospital, London.)
The prognosis of patients with horseshoe kidneys depends
on the associated anomalies. Most horseshoe kidneys are
asymptomatic throughout life. However, associated hydrone-
phrosis, reflux, lithiasis, or dysplastic tissue may lead to surgical
intervention because of recurrent urinary tract infections, pain,
or hematuria. A horseshoe kidney carries a conceivably in-
creased risk for nephroblastoma and an approximately three-
fold to fourfold higher risk for cancer of the renal pelvis.55,56
Overall, the risk of Wilms tumor in a horseshoe kidney is
approximately twice that of the normal population. Tumors
that arise, mostly in the bridge of a horseshoe kidney, can mimic
the symptoms of an intra-abdominal disease process. A paren-
chymatous renal isthmus that is the result of an abnormal
migration process may be predisposed to the development of
renal cell carcinoma.51 The risk for primary renal carcinoid
tumor is even greater, and it might arise from neuroendocrine
cells within foci of metaplastic or teratomatous epithelium
within the kidney.57 Association with ganglioneuroblastoma
has also been reported.58 Although these tumors have a higher
risk in a horseshoe kidney, the frequency does not justify
routine surveillance. An infected caliceal diverticulum can be
associated with this malformation.59 Xanthogranulomatous
pyelonephritis in a horseshoe kidney is another possible,
though a rare complication that represents a response of the
parenchyma to chronic infection.60
Robotic, laparoscopic, or open correction61–63 of uretero-
pelvic junction obstruction is the most frequent indication
for surgical intervention in a patient with a horseshoe kidney.
Division of the isthmus is no longer recommended in most
cases because the isthmus probably does not contribute to
the obstruction.64 If the position of the kidney is within nor-
mal limits and the procedure is unilateral, the traditional

A B
FIGURE 112-7 A and B, Horseshoe kidney pathology specimens. (Courtesy Professor A. R. Risdon and Dr. G. Anderson, Department of Histopathology,
Great Ormond Street Hospital, London.)
anterolateral extraperitoneal approach is recommended. If the
kidney has a more awkward position or if the procedure is bi-
lateral, a transperitoneal approach is preferable. Ureterocali-
costomy is an excellent alternative for achieving dependent
drainage of the urinary tract.65 Endopyelotomy is another al-
ternative but is best reserved for older children66 because the
recurrence rate is relatively high. Urolithiasis develops in 20%
of patients with a horseshoe kidney15 and can usually be treated
by extracorporeal shock wave lithotripsy or percutaneous sur-
gery. Laparoscopic heminephrectomy for benign disease of a
horseshoe kidney is an elegant option for experienced sur-
geons.67 Experienced surgeons are required because of the in-
creased incidence of visceral and vascular surgery associated
with these anomalies.61,63
CROSSED RENAL ECTOPIA
Crossed renal ectopia (Fig. 112-8, A and B) is the second most
common fusion anomaly after horseshoe kidney.15 Its inci-
dence is around 1 in 7000 autopsies. There are four varieties
of crossed renal ectopia: (1) with renal fusion (85% of cases),
(2) without fusion (<10%), (3) solitary, and (4) bilat-
eral.1,68,69 It has a slight male preponderance, and crossing
from left to right occurs more frequently than from right to
left. The fusion is usually located between the upper pole of
the crossed kidney and the lower pole of the normally posi-
tioned kidney (unilateral fused type). The renal pelvises re-
main in their anterior position with incomplete rotation. In
the sigmoid or S-shaped kidney, the fusion is the same but
both kidneys have completed their rotation; thus the two renal
CHAPTER 112 RENAL FUSIONS AND ECTOPIA 1409
pelvises face in opposite directions. Various degrees of fusion
have also been reported rarely (e.g., lump kidney, L-shaped
kidney, and superior ectopic kidney).15 Pseudo-crossed renal
ectopia secondary to ureteropelvic obstruction has likewise
been reported.70 Intravenous urography, isotopic studies, ul-
trasonography, MRI, and CT allow definition of the anatomic
type of ectopia. This process is often difficult, however. The
vascular supply varies and may also cross the midline.71 Most
cases of crossed renal ectopia are discovered incidentally and
are asymptomatic. When surgery is indicated, laparoscopy
may have some targeted indications for crossed fused renal
ectopia.72
Associated anomalies are common including orthopedic
and skeletal abnormalities, imperforate anus, cardiovascular
anomalies,1,73 genital anomalies,68,74 spina bifida,86 vesi-
coureteric reflux, reflux nephropathy,75 crossed ectopic ure-
ter, multicystic dysplasia,76 renal tumors,77,78 incarcerated
ureter,79 retrocaval ureter,26 and testicular tumor.80 Renal
ectopia may also be a component of more complex syndromes
such as vertebral abnormalities, anal atresia, cardiac abnor-
malities, tracheoesophageal fistula and/or esophageal atresia,
renal agenesis and dysplasia and limb defects (VACTERL) syn-
drome81; agenesis of the corpus callosum82,83; and caudal re-
gression. The complexity of some of these anomalies can
explain some of the complications reported such as accidental
ureteral ligation during repair of an inguinal hernia in a patient
with crossed fused renal ectopia.84
The complete reference list is available online at www.
expertconsult.com.
1410 PART VIII GENITOURINARY DISORDERS

A B

B C D
FIGURE 112-8 A, Crossed renal ectopia with renal fusion. (From Professor A. R. Risdon and Dr. G. Anderson, Department of Histopathology, Great
Ormond Street Hospital, London.) B, The four different types of crossed renal ectopia: with fusion (A), without fusion (B), solitary (C), and bilateral (D). (From
Ritchey M: Anomalies of the kidney. In Kelalis PP, King LR, Belman AB [eds]: Clinical Pediatric Urology, 3rd ed. Philadelphia, WB Saunders, 1992.)

CHAPTER 113
Ureteropelvic
Junction
Obstruction
Travis W. Groth and Michael E. Mitchell
Hydronephrosis is defined as dilation of the renal collecting
system. Clinically, this may result from obstruction or reflux
of urine. The physiology of obstruction (hydronephrosis) is
complex resulting from interactions between glomerular he-
modynamics and alterations in tubular function.1 In children,
hydronephrosis is typically discovered during maternal-fetal
ultrasound and accounts for approximately 0.5% to 0.6% of
all uropathies seen in the neonatal period. However, one in
five neonates with the prenatal diagnosis of hydronephrosis
will demonstrate spontaneous resolution of the hydronephro-
sis.2 In the neonates with persistent hydronephrosis, uretero-
pelvic junction (UPJ) obstruction represents 44% of all
postnatal causes of hydronephrosis.3 UPJ obstruction is,
therefore, the leading cause of postnatal hydronephrosis.
The natural history of hydronephrosis secondary to UPJ
obstruction varies. High-grade UPJ obstruction results in hy-
drostatic distention, increased intrapelvic pressure, and poor
outflow of urine. Chronic increases in intrapelvic pressure can
result in irreversible damage to the kidney. However, with
low-grade UPJ obstruction, the developing kidney can remain
in a homeostatic state and often shows temporal growth and
improvement. Between these two extremes there exists a spec-
trum of pathology—hence the debate regarding the clinical
management of kidneys with UPJ obstruction. The inevitable
questions remain, “which patient requires surgical intervention,
and which can be observed?”
Etiology/Incidence
------------------------------------------------------------------------------------------------------------------------------------------------
The incidence of UPJ obstruction is 1 in 1250 births.4,5 It oc-
curs more commonly in males (M/F ¼ 2:1). There is a predi-
lection for the left side in children (66%), whereas the reverse
is true for UPJ obstruction of adults.6–8 Bilateral cases of UPJ
obstruction occur in 10% to 36% of patients, with the highest
percentage occurring in the younger age group.6,9
Koff described two types of UPJ obstruction, intrinsic and
extrinsic, discovered at the time of surgical exploration
(Fig. 113-1).10 The most common UPJ obstruction is the in-
trinsic type, classically known as the “adynamic” segment. This
adynamic segment interferes with the proximal ureteral peri-
stalsis. At surgery, these segments are of variable length and
usually narrower than the rest of the ureter. Additionally, these
segments are poorly distensible but are typically probe patent.
Histologically, they are deficient in circular muscle fibers,
which tend to be disorganized and dysmorphic.11,12
The second type of UPJ obstruction is the extrinsic type, typ-
ically related to such mechanical factors as crossing vessels,
adhesive bands, arteriovenous malformations, and uretal
folds.13,14 Extrinsic factors create an abrupt angulation by
kinking or compressing the UPJ, which leads to obstruction
of urine flow and progressive hydronephrosis. Similarly,
Östling described pleatlike ureteral folds that occur at the level
of the UPJ (Fig. 113-2).14 Teleologically, these folds are thought
to allow the dramatic increase in axial growth that children
experience during the first 2 years of life. Östling folds are typ-
ically not felt to be obstructive and tend to disappear with
patients’ increase in linear growth; however, if persistent, they
can result in obstruction.
Aside from the intrinsic and extrinsic types of UPJ obstruc-
tion there are intraluminal causes of proximal ureteral obstruc-
tion, which include ureteral valves and benign fibroepithelial
polyps.15,16 Both lead to partial, intermittent obstruction and
typically require surgical intervention (Fig. 113-3).17
Another way to characterize types of obstructive processes is
by their volume-dependent (extrinsic) and pressure-dependent
(intrinsic) characteristics.10,18 Experimental and clinical obser-
vations led Koff to conclude that these parameters, or a combi-
nation thereof, may explain the many differences observed in
the clinical spectrum of UPJ obstruction.18
Embryology
------------------------------------------------------------------------------------------------------------------------------------------------
During human embryogenesis, there are three sequential and
interdependent kidney systems: the pronephros, mesoneph-
ros, and metanephros or definitive kidney. The earliest and
most transient kidney is the pronephros. Sequentially the pro-
nephros is replaced by the mesonephros by the third week
of gestation. The mesonephros elongates to form a tubular
structure, which eventually forms the caudal end of the meso-
nephric, or wolffian, duct. At the fifth week of gestation, the
1411
1412 PART VIII GENITOURINARY DISORDERS
Exit flow Exit flow
Simple Complex
Pressure Pressure
FIGURE 113-1 Diagrammatic representation of simple and complex
pressure-flow patterns in human hydronephrosis. (From Koff SA: Patho-
physiology of ureteropelvic junction obstruction: Clinical and experimental
observations. Urol Clin North Am 1990;17:269.)
dorsomedial wall of this ductal system gives rise to a divertic-
ular outgrowth called the ureteric bud. The “ureteric bud”
seems to have an important bidirectional inductive role, which
results in differentiation of the mesonephric duct (to form the
trigone, posterior bladder neck and proximal urethra) and to
initiate the development of the metanephrosis into the kidney.
Therefore just as the mesonephric system moves into the blad-
der to form the trigone, posterior bladder neck, and urethra,
the ureteric bud elongates in an ascending fashion toward the
primitive metanephric anlagen. On contact, the two structures
initiate a cascade of events known collectively as nephrogenesis.
The ureteric bud ultimately forms the ureter, renal pelvis, and
calyceal system.
In this developmental process, mechanical obstruction
may occur at three different anatomic sites: the pelvocaliceal,
ureteropelvic, and ureterovesical junctions.19 Each of these
sites represents a confluence of two or more embryologic
structures (i.e., ureter and bladder and ureter and kidney).
The cellular and molecular mechanisms are incompletely un-
derstood at this time; however, complex cellular “crosstalk”
and induction initiated in a bidirectional manner normally
FIGURE 113-2 Percutaneous nephrostogram demonstrating Östling’s
fetal folds. (Courtesy Children’s Hospital and Regional Medical Center,
Seattle.)
FIGURE 113-3 Ureteropelvic junction obstruction secondary to stenosis:
simple, segmented, serpiginous, and shelving. (Courtesy Dr. F. Douglas
Stephens.) (From Fung LCT, Lakshmanan Y: Anomalies of the renal collect-
ing system: Ureteropelvic junction obstruction [pyelocalyectasis] and in-
fundibular stenosis. In Belman AB, King LR, Kramer SA [eds]: Clinical
Pediatric Urology, 4th ed. London, Martin Dunitz, 2002, p 561.)
culminate in appropriate anatomic junctions. However, minor
imprecision because of the inherent variables of the process
can result in two of the more common pathologic conditions
in pediatric urology, namely, vesicoureteral reflux (VUR) and
UPJ obstruction.
At one time it was thought that bidirectional recanalization
of the ureter explained the frequent obstruction at the last seg-
ments to recanalize the UPJ and ureterovesical junction
(UVJ).20 More recently, however, it has been shown that bidi-
rectional recanalization does not occur. Recanalization occurs
only at the middle portion of the ureter.21 None of the current
experimental models or theories of ureteral development ex-
plain why the UPJ is the most common site of congenital ure-
teral obstruction. This continues to be an active area of
research. For example, a new study using an animal model
of obstructive nephropathy has reported a possible genetic
component of hydronephrosis.22 Chang and colleagues22 de-
scribe a mouse model in which the gene encoding the calci-
neurin B type 1 isoform (Cnb1) has been deleted from the
mesenchyme lining the urinary system. Hydronephrosis de-
velops in these animals because of the lack of peristalsis. This
study not only points to the renal pelvis as a regulator of peri-
stalsis but may also explain congenital obstruction in which
no obvious physical obstruction is identified.
Recently, Wang and colleagues23 examined the role of bone
morphogenic protein 4 (BMP4) signaling by examining a
BMP4 murine knock-out mouse model. Antagonism of BMP
signaling with Nogin protein inhibited ureteral smooth mus-
cle formation around the ureter and localized antagonism of
BMP signaling demonstrated similar histologic findings as in-
trinsic UPJ obstruction from an adynamic segment.23 Even
though these data support a molecular cause for UPJ obstruc-
tion, there has yet to be one molecular theory explaining how
and why different UPJ obstructions form.
Clinical Features
------------------------------------------------------------------------------------------------------------------------------------------------
UPJ obstruction can affect infants, adolescents, or adults. The
presentation of each differs. Typically, infants are asymptom-
atic and are now diagnosed by prenatal ultrasonography.
Thirty years ago, however, most infants with UPJ obstruction
were initially diagnosed with a palpable abdominal mass at
initial evaluation or, less often, on workup for urinary tract in-
fection, failure to thrive, feeding difficulties, and nephrolithia-
sis.7 With the advent of prenatal fetal ultrasonography, most
infants born with UPJ obstruction are detected antenatally.
The other population in which UPJ obstruction is detected

consists of children with a variety of symptoms. Accordingly,
there seems to be a bimodal age distribution of UPJ obstruc-
tion: asymptomatic postnatal infants and symptomatic school-
aged children.
ANTENATAL MANIFESTATIONS
At the eighth week of intrauterine life, fetal urine is produced by
the mesonephros. Once this transient kidney regresses, urine is
made by the metanephros and continues to be made through-
out gestation. Fetal urine production increases from 5 mL/hr
at 20 weeks gestation to a rate that may approach 50 mL/hr
at 40 weeks of gestation.24 Thus maternal-fetal sonography
should initially be performed between 16 and 20 weeks of
gestation when adequate urine production has been estab-
lished.25 Most centers performing maternal-fetal ultrasound
can detect normal kidneys at 17 to 19 weeks, whereas markedly
dilated collecting systems can be identified as early as 12 to 14
weeks.26
The first report of antenatally detected urinary tract anom-
alies was in 1970.27 Since that time, the American Institute for
Ultrasound in Medicine’s Bioeffects Committee has concluded
that prenatal ultrasound screening is safe. This endorsement
has led to the common practice by obstetricians, as well as
the expectation by the general population, that screening ul-
trasound be performed during the second trimester. Despite
the fact that protocols for maternal-fetal ultrasonography
continue to evolve in this country, they still depend in part on
specific conventions and random institutional idiosyncrasies.4
Prenatal ultrasound can detect hydronephrosis and abnor-
mal anatomy, but it is not specific for the diagnosis of partic-
ular disease processes. Sonographic features suggestive of UPJ
A B
C D
FIGURE 113-4 Renal ultrasound scans representing the Society for Fetal Urology (SFU) grading schema. A, SFU grade 1 hydronephrosis. B, SFU grade 2
hydronephrosis. C, SFU grade 3 hydronephrosis. D, SFU grade 4 hydronephrosis. (Courtesy Children’s Hospital of Wisconsin, Milwaukee.)
CHAPTER 113 URETEROPELVIC JUNCTION OBSTRUCTION 1413
obstruction include (1) unilateral or, less frequently, bilateral
pelviectasis; (2) normal amniotic fluid volume; (3) no evi-
dence of ipsilateral ureteral dilation; and (4) normal thickness
of the bladder wall and normal cycling of the bladder. In cases
in which the classic features of hydronephrosis are estab-
lished, the diagnosis of UPJ is straightforward. Differentiation
between true obstructive uropathy and transient hydrone-
phrosis of the neonate is not possible. Additionally, prenatal
ultrasound cannot distinguish between UPJ obstruction and
multicystic dysplastic kidney (MCDK). Sequential renal ultra-
sound and other complementary radiographic imaging studies
may be required for follow-up.
The Society of Fetal Urology (SFU) organized consensus
guidelines for grading different degrees of hydronephrosis
(Table 113-1) (Fig. 113-4A to D).28,29 Once these in utero
criteria are established, a prenatal diagnosis of hydronephro-
sis with a possibility of UPJ obstruction should be enter-
tained by the physician and reported to the parents. In
TABLE 113-1
Society of Fetal Urology (SFU) Grading of Hydronephrosis
SFU Description
Grade
1 Slight splitting of the central renal complex without
calyceal involvement, normal parenchyma
2 Splitting of the central renal complex with extension to
nondilated calyces
3 Wide splitting of the renal pelvis, dilated outside the renal
border, calyces uniformly dilated, normal parenchyma
4 Large, dilated calyces (may appear convex); thinning of the
parenchyma to 50% of the ipsilateral (normal) kidney
1414 PART VIII GENITOURINARY DISORDERS

male fetuses, however, bilateral hydronephrosis should be TABLE 113-2

considered to be a different and more urgent problem Symptoms of Ureteropelvic Junction Obstruction in Children
relating to bladder outlet obstruction, possibly caused by Episodic flank pain
posterior urethral valves. Episodic abdominal pain  vomiting
Cyclic vomiting
PRENATAL COUNSELING Urinary tract infections
Nephrolithiasis
Parents should be advised of a presumptive diagnosis of Hematuria
hydronephrosis and counseled appropriately. Education and Anorexia
communication are paramount in the process of prenatal Asymptomatic
counseling. Fortunately, the majority of antenatally detected
genitourinary abnormalities are unlikely to require postnatal
surgical intervention.30,31 In fact, only 1% to 25% require
surgery at 4 years of follow-up.32,33 However, referral to a duplicated collecting systems (12% incidence); posterior
pediatric urologist for prenatal counseling is highly recom- urethral valves (9% incidence); and ectopic ureter, urethral
mended. The survival rate for fetuses found to have unilateral atresia, sacrococcygeal teratoma, and hydrometrocolpos. Non-
hydronephrosis secondary to obstruction is virtually 100%. obstructive causes of hydronephrosis include VUR (14% inci-
Postnatal follow-up is essential to track progression or res- dence), physiologic dilation, prune-belly syndrome, renal
olution of hydronephrosis, as well as establish the manage- cystic diseases, and megacalicosis (Table 113-3).35,36
ment principles with the family. For this reason, referral to Although it is easy to diagnose hydronephrosis, it is, unfor-
a pediatric urologist is highly recommended. tunately, difficult to prove obstruction. Furthermore, there has
yet to be one diagnostic method that has been able to differ-
entiate significant obstruction from insignificant obstruction
POSTNATAL MANIFESTATIONS
in an infant. None of the existing imaging modalities can ac-
Although often asymptomatic, children with UPJ obstruction curately predict which hydronephrotic kidney is at risk for
may have variable symptoms such as episodic flank, abdom- progressive damage and loss of renal function. With this lim-
inal pain, or less commonly a urinary tract infection. Cyclic itation in mind, sequential radiographic studies can be used to
abdominal pain, often associated with vomiting, is a classic define change that may indicate physiologically significant
finding (Dietl crisis*) usually observed in older patients. UPJ obstruction that might require surgical intervention.
UPJ obstruction can also be an unsuspected diagnosis made
after evaluation for vague and often chronic abdominal symp-
INTRAVENOUS UROGRAPHY
toms, frequently referred from the pediatric gastroenterologist
after an ultrasound had been performed. UPJ obstruction may Historically, intravenous urography (IVU) was the radio-
be associated with renal calculi with an incidence of nephro- graphic modality of choice for noninvasive assessment of
lithiasis 17 times higher in patients with UPJ obstruction.34 the urinary tract. As an imaging study, it combines anatomic
Hematuria may also be the initial symptom in UPJ obstruction. accuracy with qualitative information regarding renal function
Trivial trauma can cause friable mucosal vessels in the dilated and obstruction. IVU is now infrequently used in the assess-
collecting system to rupture. Left-sided UPJ obstruction can ment of a pediatric patient with obstructive uropathy and has
be associated with the insidious manifestation of anorexia been replaced by sonography and scintigraphy. Obstruction,
and failure to thrive. These children are typically found to on IVU, of the kidney may be inferred from delay in the
have a left UPJ obstruction. The dilated left renal pelvis creates appearance of contrast material or a negative nephrogram, a
a mass effect on the stomach that results in early satiety, an- delay in drainage, dilution of contrast medium, or uniform
orexia, and vomiting. Still other children with UPJ obstruction cortical loss.
may have profound urosepsis. Snyder and colleagues6
reported that urinary tract infection is the initial sign in
30% of children with UPJ obstruction beyond the neonatal
period (Table 113-2).

TABLE 113-3
Diagnosis
------------------------------------------------------------------------------------------------------------------------------------------------
Differential Diagnosis of Hydronephrosis
Obstructive Nonobstructive
Because hydronephrosis can be caused by multiple etiologies,
it is helpful to try and categorize hydronephrosis into obstruc- Ureteropelvic junction obstruction Vesicoureteral reflux
tive versus nonobstructive causes. Obstructive causes of Ureterovesical junction obstruction Physiologic dilatation
hydronephrosis include UPJ obstruction (44% incidence); Multicystic dysplastic kidney Prune-belly syndrome
UVJ obstruction (21% incidence); MCDK, ureterocele, and Ureterocele Renal cystic diseases
Duplicated collecting system Megacalycosis
Posterior urethral valves
Ectopic ureter
*Jozef Dietl (1804-1878), a Polish physician, described a syndrome character-
ized by violent paroxysms of colicky flank pain, nausea, chills, tachycardia, Urethral atresia
oliguria, transient hematuria or proteinuria, and a palpable enlarged tender Sacrococcygeal teratoma
kidney that was attributed to acute hydronephrosis caused by kinking or vas- Hydrometrocolpos
cular obstruction of the ureter of a floating kidney.
 CHAPTER 113 URETEROPELVIC JUNCTION OBSTRUCTION 1415

A B
FIGURE 113-5 A, Renal ultrasound representing an obstructed ureteropelvic junction with an intrarenal pelvis. B, Renal ultrasound of a nonobstructed
extrarenal pelvis. (Courtesy Children’s Hospital and Regional Medical Center, Seattle.)

RENAL ULTRASOUND
Renal ultrasound is a widely available, relatively inexpensive,
noninvasive, safe test that provides adequate anatomic vi-
sualization without radiation exposure. Given these exceptional
qualities, it is widely used both in maternal-fetal sonography and
for postnatal imaging. Consequently, renal sonography has
resulted in the increased diagnosis of prenatal and postnatal di-
lation of the upper urinary tract. In fact, renal ultrasound is the
most commonly performed initial study for the postnatal eval-
uation of neonates who have been discovered prenatally to have
hydronephrosis. Renal ultrasound is highly accurate in the di-
agnosis of hydronephrosis. It provides information on findings
characteristic of UPJ obstruction including pelviectasis and
caliectasis, absence of ureterectasis, normal bladder filling
and emptying (cycling), and normal bladder thickness. In uni-
lateral cases of hydronephrosis, ultrasound may delineate pa-
renchymal atrophy on the affected side in comparison with
the contralateral side. FIGURE 113-6 Renal ultrasound of normal infant kidney with lucent pyr-
amids, which can be confused with calyceal dilation. (Courtesy Children’s
Twenty-five years ago, the general consensus among pedi- Hospital of Wisconsin, Milwaukee.)
atric urologists had been that the amount of hydronephrosis
did not correlate with the degree of obstruction. Ransley
and colleagues33 finally established this fact. In their study, the ability to track the axial growth of the kidneys as an indi-
progressive hydronephrosis and deterioration in renal func- cator of progression of UPJ obstruction. In cases of unilateral
tion were uncommon in neonates and infants with a maximum UPJ obstruction, the growth rate of the contralateral normal
anteroposterior renal pelvic diameter of less than 10 mm and kidney can be used as a comparison standard.38 With the
no evidence of infundibular or caliceal dilation (SFU grade 1 use of renal growth–function nomograms, compensatory hy-
hydronephrosis). All the patients initially in the nonoperative pertrophy in the contralateral “normal” kidney may be indic-
group who eventually required surgery had a prenatal antero- ative of progressive obstruction in the hydronephrotic kidney.
posterior renal pelvic diameter of greater than 12 mm. How- Conversely, a slower than normal growth rate (compensatory
ever, renal pelvic diameter, alone, was found to be a poor hypertrophy) in the normal kidney suggests rapid recovery of
positive predictor of outcome because only 34% of such pa- function (or even supranormal function) in the unilaterally
tients required pyeloplasty. In some of these patients, perhaps, hydronephrotic kidney and would imply that obstruction is
such mild degrees of hydronephrosis are probably physio- not present. Koff’s data, however, have not been corroborated
logic, secondary to the highly compliant nature of the fetal because of operator variability in renal length measurements.
(and neonatal) renal pelvis. An extrarenal pelvis or lucent pyr- Serial ultrasound studies are important for implementing
amids of infancy can be mistaken for pathologic hydronephro- any protocol for UPJ obstruction, but they can be problematic
sis (Fig. 113-5A and B, Fig. 113-6); however, caliceal dilatation because they require a waiting period during which progres-
is typically not present in these patients. An extrarenal pelvis, sive renal injury can occur. For example, cortical thinning, an
therefore, is considered to be a variation of normal. important finding, may develop but is typically a late irrevers-
Koff and colleagues37 proposed the use of serial renal ultra- ible finding. Focal cortical loss is easier to detect, though less
sound to predict progression in unilateral cases of UPJ ob- common than global cortical thinning. Reflecting on Koff and
struction. In addition to being able to monitor the temporal colleagues’ study, it is logical to conclude that in patients with
behavior of UPJ obstruction, serial renal ultrasound provides unilateral hydronephrosis, contralateral compensatory renal
1416 PART VIII GENITOURINARY DISORDERS
hypertrophy may indicate a greater risk for continued deteri-
oration of the hydronephrotic kidney.37 However, reliable
sonographic diagnosis of renal hypertrophy requires multiple
sequential measurements of renal length or volume, thus lim-
iting its applicability for guiding early intervention.39
In summary, renal ultrasound is highly accurate in the di-
agnosis of hydronephrosis but cannot diagnose obstruction.
The degree of hydronephrosis neither specifically indicates
the presence or absence of obstruction nor predicts whether
the hydronephrosis will improve or progress.40 A case in point
is the neonate, in whom hydronephrosis (by ultrasound) may
be transient and is dependent on the degree of hydration and
bladder fullness. Most limiting is the fact that ultrasound
cannot provide information regarding renal function.
TECHNETIUM 99m RENAL SCINTIGRAPHY
Differential renal function is one of the most important param-
eters used in defining significant obstruction and defining the
need for surgical correction of congenital UPJ obstruction. It is
the present study of choice for estimation of overall and dif-
ferential renal function. Three radiopharmaceuticals are pri-
marily used in renal scintigraphy, and their characteristics are
linked to their biologic activity. Technetium 99m-diethylene-
triaminepentaacetic acid (99mTc-DTPA) and technetium 99m-
mercaptoacetyltriglycine (99mTc-MAG-3) are preferentially con-
centrated by the kidney and freely filtered by the glomerulus.41
DTPA is neither secreted nor resorbed by the renal tubules,
whereas MAG-3 is secreted by the tubules. Because each is
nearly completely excreted, they can be used to estimate differ-
ential renal function and urinary drainage. The third agent,
technetium 99m-dimercaptosuccinic acid (99mTc-DMSA), is
Normal pattern
Diuretic injection
Activity
Activity
Kidney
Ureter
Activity
Activity
Diuretic injection
0 5 10 15 20 25
Time (min)
FIGURE 113-7 Lasix renography, classic patterns: normal, dilated nonobstructed, and obstructed. (From Fung LCT, Lakshmanan Y: Anomalies of the
renal collecting system: Ureteropelvic junction obstruction [pyelocalyectasis] and infundibular stenosis. In Belman AB, King LR, Kramer SA [eds]: Clinical
Pediatric Urology, 4th ed. London, Martin Dunitz, 2002, p 594.)
tightly bound to renal tubular cells and is, therefore, useful
for the detection of differential renal function and clinically
significant cortical lesions such as renal scars.
At least three consensus statements have been published to
decrease the wide clinical variance in protocols: (1) the Society
of Nuclear Medicine’s Nuclear Medicine Procedure Guidelines
for Pediatric Diuretic Renography, (2) the Well Tempered Diuretic
Renogram, and (3) the consensus statement from the Ninth
International Meeting of the Society of Radionuclides in
Nephrourology, Committee on Diuretic Renography.42–44
They are all similar in methodology and interpretation. Basi-
cally, a collecting system without significant obstruction will
have a clearance half-time (i.e., the time for half the radiophar-
maceutical to clear) after furosemide administration of less
than 10 minutes. Longer than 20 minutes is abnormal and as-
sociated with significant obstruction. Clearance half-times bet-
ween 10 and 20 minutes are considered indeterminate. The
half-time should not be the only criterion on which to define
obstruction. As in most studies, interpretation requires the
use of other available data: curve analysis, differential renal
function, and, of course, the clinical context (Fig. 113-7).
Currently, renal scintigraphy is the most popular modality
for determining the functional significance of UPJ obstruction,
mainly because several investigators have shown that most
cases of severe hydronephrosis (SFU grades 3 to 4) demon-
strate obstruction on diuretic renography.28,45 As a result of
these studies, the differential renal function of a hydronephro-
tic kidney, as determined by 99mTc-DTPA or 99mTc-MAG-3
renal scan, creates an arbitrary threshold for surgical inter-
vention.46 Thirty-five percent is the pivotal function around
which intervention or observation is determined. If the
initial ipsilateral differential renal function is greater than
Dilated nonobstructed pattern Obstructed pattern
Diuretic injection
Diuretic injection
Activity
Kidney Kidney
Ureter
Diuretic injection
Activity
Diuretic injection
Ureter
0 5 10 15 20 25 0 5 10 15 20 25
Time (min) Time (min)

35%, the neonate may be monitored conservatively by renal
ultrasound every 3 months. In contrast, if the differential renal
function is less than 35% on the initial study or if there is a
decremental change in function of 10% by repeat renal scan,
consideration should be given to surgical intervention.47
Renal scintigraphy is the study of choice for the estimation
of overall and differential renal function, except in patients
with poor or immature renal function and those with capa-
cious collecting systems. Other determinates that affect renal
scintigraphy include the region of interest, time of measure-
ment, state of hydration, bladder fullness, reservoir effect, type
of protocol, renal response to furosemide, and the concept of
supranormal function (Fig. 113-8).42,43,48–51 Supranormal
function is defined as greater than 55% of differential renal
function in the hydronephrotic kidney in children with uni-
lateral hydronephrosis. It has not been confirmed whether
supranormal differential renal function is an artifact or a true
finding, nor is there a consensus on how to manage these pa-
tients (Fig. 113-9). Oh and colleagues83 examined supranor-
mal function in human subjects and determined that
supranormal function of the ipsilaterally obstructed kidney
may not be true hyperfunction but merely a pathologic com-
pensatory mechanism in response to chronic obstruction. For
this reason, they warned that interpretation of differential re-
nal function should be made with this caveat in mind. Khan
and colleagues52 believe that supranormal function is an arti-
fact that can be avoided by using MAG-3 and appropriate
computer software to account for multiple algorithms.
VOIDING CYSTOURETHROGRAPHY
It is well known that VUR, the most common problem of
the lower urinary tract in children, can coexist with UPJ ob-
struction, the most common pediatric problem of the upper
FIGURE 113-8 MAG-3 lasix renal scintigraphy of a patient with ureteropelvic junction obstruction with delayed washout pattern and excretion of con-
trast with preserved renal function on the right side. (Courtesy Children’s Hospital of Wisconsin, Milwaukee.)
CHAPTER 113 URETEROPELVIC JUNCTION OBSTRUCTION 1417
urinary tract.53 The exact reason for this association has not
yet been established.54 However, what is known is that
approximately 9% to 14% of patients with UPJ obstruction
have VUR. Historically, voiding cystourethrography (VCUG)
is the standard of practice for the clinical evaluation of all
infants with prenatal hydronephrosis, regardless of age or
gender.55,56
However, several investigators have challenged this
standard.57 In a retrospective review of 106 patients who
underwent pyeloplasty for UPJ obstruction, low-grade reflux
coexisting with UPJ obstruction spontaneously resolved
after pyeloplasty. Additionally, they found that all cases of
high-grade VUR coexisting with UPJ obstruction were easily
detected by renal ultrasonography. They concluded that
VCUG should be limited to children with UPJ obstruction
who also have a dilated ureter on ultrasonography. However,
patients with a history of febrile UTI and hydronephrosis all
should undergo VCUG regardless of dilatated ureter on ul-
trasonography. Caution should be practiced when evaluating
children with this diagnostic dilemma because the approach
to management is subtle yet complex.
MAGNETIC RESONANCE UROGRAPHY
Most recently, dynamic, gadolinium-enhanced magnetic reso-
nance urography (Gd-MRU) has been advocated for its non-
ionizing radiation and its unparalleled, three-dimensional
views of anatomic obstruction. Wen and colleagues58 used
Gd-DTPA-enhanced magnetic resonance imaging in rats and
determined that it could distinguish between an obstructed
and nonobstructed collecting system. Using a similar tech-
nique, Kirsch recently reviewed a cohort of 200 children
with hydronephrosis who underwent systematic evaluation
1418 PART VIII GENITOURINARY DISORDERS
FIGURE 113-9 MAG-3 lasix renal scintigraphy: Supranormal renal function of a 6-month-old baby who had the SFU grade 3 hydronephrosis, found postnatally
to have high-grade left ureteropelvic junction obstruction. (Courtesy Children’s Hospital of Wisconsin, Milwaukee.)
FIGURE 113-10 Three-dimensional maximal intensity projection recon-
struction after Gd-DTPA infusion showing classic left ureteropelvic junction
obstruction. (From Perez-Brayfield MR, Kirsch AJ, Jones RA, Grattan-Smith
JD: A prospective study comparing ultrasound, nuclear scintigraphy and
dynamic contrast-enhanced magnetic resonance imaging in the evalua-
tion of hydronephrosis. J Urol 2003;170:1331.)
with Gd-MRU.59 When compared with sonography and renal
scintigraphy, Gd-MRU provided superior anatomic images of
the genitourinary system and excellent spatial resolution
(Fig. 113-10). The study has been shown to be robust and re-
producible with a linear regression coefficient of 0.99. The
limitations of the study are that it requires general anesthesia
or sedation and involves cumbersome interpretation of the im-
ages.60 Although the technique is not yet widely available,
MRU has been considered by some investigators to be a more
reliable determinant of renal anatomy and function in cases of
UPJ obstruction.
RETROGRADE PYELOGRAPHY
This radiographic study is rarely necessary to diagnose UPJ
obstruction. In a series of 108 children undergoing pyelo-
plasty, retrograde pyelography (RPG) did not change the sur-
gical approach or planned procedure in any of the cases.61
However, RPG is still commonly performed before open pye-
loplasty. The reason for this is simple. During the diagnostic
process, information regarding the exact anatomic location
of the UPJ in relation to other anatomic structures is lacking.
Therefore to precisely identify the location of obstruction and
perhaps exclude the presence of another distal obstruction,
RPG is performed (Figs. 113-11 and 113-12).62

FIGURE 113-11 Left retrograde pyelogram demonstrating left uretero-
pelvic junction obstruction. Note the dilation of renal pelvis and small jet of
contrast at the ureteropelvic junction. (Courtesy Children’s Hospital of
Wisconsin, Milwaukee.)
PRESSURE-FLOW STUDY
The perfusion provocation study, introduced by Bäcklund and
colleagues63 in 1965 and popularized by Whitaker in 1973, is
the standard pressure-flow study and is simply referred to as
the Whitaker test (Fig. 113-13).64 It detects the resistance of
the ureter to a known flow rate by simultaneously measuring
manometric pressure between the renal pelvis and bladder.
The test is unaffected by the kidney’s GRF, and its antegrade
pyelography offers excellent anatomic location of the sus-
pected obstructing segment. The study is most useful when
FIGURE 113-12 Left retrograde pyelogram demonstrating left uretero-
pelvic junction obstruction along with infantile appearing proximal ure-
ter. (Courtesy Children’s Hospital of Wisconsin, Milwaukee.)
CHAPTER 113 URETEROPELVIC JUNCTION OBSTRUCTION 1419
FIGURE 113-13 Whitaker pressure perfusion study. (From Fung LCT,
Lakshmanan Y: Anomalies of the renal collecting system: Ureteropelvic
junction obstruction [pyelocalyectasis] and infundibular stenosis. In Bel-
man AB, King LR, Kramer SA [eds]: Clinical Pediatric Urology, 4th ed.
London, Martin Dunitz, 2002, p 596.)
there are equivocal studies regarding obstruction in a hydro-
nephrotic kidney. High renal pelvic pressure relative to blad-
der pressure at a given flow rate confirms the presence of
obstruction. According to this test at a flow rate of 10 mL/min,
a differential pressure between the renal pelvis and the
urinary bladder of less than 13 cm H2O is considered to be
normal. Arbitrary values of 14 to 20 cm H2O indicate mild
obstruction; values in excess of 35 cm H2O indicate severe ob-
struction. The Whitaker test requires percutaneous access to
the kidney and usually requires general anesthesia in the pe-
diatric population. Consequently, it is undesirable for a rou-
tine test. It is typically indicated for equivocal cases of
obstruction and is, therefore, rarely performed. In addition,
the Whitaker test requires a rigid protocol and exacting mea-
surements to avoid technical errors. The most common error
is failure to achieve an equilibrium (what goes in comes
out). A physiologic pressure/flow test would be the ideal
means of diagnosing “obstruction,” but unfortunately one test
does not presently exist.
BIOCHEMICAL MARKERS
A biomarker is something that can be used as an indicator of a
disease. Researchers have used proteomics, which is defined
as the characterization of protein content within tissue, cells,
or fluid, to try to find new biomarkers. It has been well estab-
lished that ureteral obstruction results in renal injury on a cel-
lular level.65 Studies have led investigators to surmise that there
might be potential biochemical markers detected in the urine
that may be predictive of significant postnatal renal injury in
patients with elevated hydronephrosis.66 Prospective markers
include urinary sodium, calcium, and b2-microglobulin, or ap-
propriate urinary proteins (proteomics) unique to obstruction.
1420 PART VIII GENITOURINARY DISORDERS
Clearly, it has been demonstrated that persistently elevated col-
lecting system pressure induces renal abnormalities including
an increase in tubular pressure, decrease in renal blood flow,
decrease in GFR, and renal tubular dysfunction.67,68
Recently, there have been several potential biomarkers that
have been shown to be increased in the urine of children with
UPJ obstruction. These include transforming growth factor b1
(TGF-b1), monocyte chemotactic protein 1 (MCP-1), and
endothelin-1.69–73 Also, urinary epidermal growth factor
(EGF) was shown to decrease during obstruction.72 Fung
and Atala, using a nonobstructed porcine model in which
urine drained against a predetermined pressure gradient,
demonstrated a progressively elevated level of urinary N-
acetyl-b-D-glucosaminidase, an indicator of acute renal cell
injury.74 There has yet to be any prospective studies or clin-
ical studies that demonstrate if these biomarkers will be of
clinical value. At present, however, because of the lack of stan-
dard biochemical or physiologic markers to measure renal ob-
struction, serial radiographic studies are necessary to evaluate
the function and anatomy of a kidney with suspected UPJ
obstruction. Proteomics has great potential to help discover
new biomarkers that could in theory dramatically change
how patients with hydronephrosis are followed and treated.
It may be that the profile of a marker of different urinary pro-
teins will reveal a reproducible and dependable measure of
ongoing renal injury related to obstruction. Because these
markers require vigorous clinical validation, which has been
limited to date, and need to be reproducible, their future role
in management remains undetermined.
PREOPERATIVE VASCULAR IMAGING
Once the diagnosis of UPJ obstruction has been determined
from the history and basic radiographic studies, there is often
a question regarding the possibility of involvement of a cross-
ing polar vessel in the obstructive process. This particular an-
atomic problem is primarily important in older children or
adults with UPJ obstruction because it can influence the sur-
gical approach and require radiographic studies focused on
defining the vascular anatomy.
The incidence of crossing vessels in patients with UPJ ob-
struction is inconsistent in the literature and ranges from 11%
to 79%, mainly because of the variable definitions used to de-
scribe UPJ obstruction.7,75 Polar vessels crossing immediately
adjacent to the UPJ are designated as “crossing vessels.” They
are thought to exacerbate rather than initiate the obstructive
process, although this remains a controversial issue.76 Cross-
ing vessels have the potential to cause significant morbidity if
not recognized before an endourologic procedure. Several im-
aging modalities can be used to search for a crossing vessel in-
cluding digital subtraction intra-arterial angiography, helical
computed tomography, contrast-enhanced color Doppler im-
aging, endoluminal sonography, and magnetic resonance an-
giography. The anatomic peculiarity of polar crossing vessels
makes vascular imaging important, especially when minimally
invasive techniques are proposed for the management of UPJ
obstruction.77 An extensive historical review addressing
crossing vessels has been published.78 Among pediatric urol-
ogists, the concept of preoperative vascular imaging to verify
a crossing vessel is controversial, if not moot. Surgical manage-
ment of children with UPJ obstruction is open surgery,
laparoscopy, or robotic surgery, but not endopyelotomy, which,
if used blindly, could inadvertently disrupt an unsuspected
crossing vessel. To this date, endourologic procedures are lim-
ited in children for several reasons: (1) the available endouro-
logic equipment is not small enough to accommodate children
younger than 3 years; (2) most pediatric surgeons are trained to
perform an open dismembered pyeloplasty, with the recogni-
tion that laparoscopy is an excellent choice in the proper hands;
(3) both open and laparoscopic approaches would identify an
anterior crossing vessel intraoperatively; and (4) the open
surgery, laparoscopy, or robotic surgery for UPJ obstruction
success rate is much greater than the reported success rate
for endopyelotomy as a primary procedure in children.
Management
------------------------------------------------------------------------------------------------------------------------------------------------
The management of possible UPJ obstruction remains contro-
versial because the natural history of UPJ obstruction still re-
mains unclear in children. Management of unilateral
hydronephrosis cannot be based on either the degree of pelvic
dilation on ultrasound or obstruction on diuretic renogra-
phy.45,79 Several investigators have demonstrated that nearly
50% to 80% of all cases of prenatal hydronephrosis will re-
solve spontaneously with postnatal follow-up.80,81 Others,
however, have demonstrated progressive hydronephrosis with
diminution of function in follow-up 3 years postpartum.82
These different outcomes are unsettling and, as yet, unan-
swered. Currently, watchful waiting and early intervention
are two main options regarding the management of UPJ
obstruction.
WATCHFUL WAITING
Conservative management of unilateral UPJ obstruction in-
volves sequential radiographic studies and cautious patience.
The fact that so many algorithms have been devised for the
management of UPJ obstruction is a sure indication that no
single algorithm is absolute. In general, however, the typical
evaluation of an infant with the prenatal diagnosis of hydro-
nephrosis begins with a directed physical examination and
postnatal ultrasound. The physical examination should deter-
mine the child’s vital signs, general health, and overall appear-
ance. Respiratory distress or vomiting may be associated with
a palpable abdominal mass in some cases. In most instances,
however, an abdominal mass is not identified.
Postnatal ultrasound should be performed at least 2 days
after birth because relative dehydration of the fetus and phys-
iologic oliguria may result in a false-negative study after birth.
The one exception to this 2-day rule is a male neonate with
bilateral hydronephrosis and a thickened bladder wall, which
would be suggestive of a more urgent problem such as bladder
outlet obstruction secondary to posterior urethral valves.
Renal sonography suggesting UPJ obstruction should be
followed by additional diagnostic studies to confirm or ex-
clude the obstruction. Further radiologic tests should be
based on other possible causes of hydronephrosis. Several
conditions can produce dilatation of the collecting system
with or without urinary tract obstruction such as physiologic
pelviectasis, UPJ obstruction, duplication of the collecting sys-
tem with an ectopic ureter (or an associated ureterocele),
 CHAPTER 113 URETEROPELVIC JUNCTION OBSTRUCTION 1421

megaureter, MCDK, UVJ obstruction, posterior urethral
valves, prune-belly syndrome, renal cystic diseases, parapelvic
renal cysts, and megacalicosis. In addition, the diagnosis of
hydronephrosis may be falsely diagnosed because of the sono-
lucent appearance commonly seen in the neonatal renal med-
ullary and pyramids. This curious finding resolves with renal
parenchymal maturation, which usually occurs within the first
6 months of life. Follow-up ultrasound reveals no apparent
evidence of hydronephrosis (Fig 113-14).
Postnatal renal ultrasound can eliminate several of the
aforementioned diagnoses. If a dilated ureter is demonstrated
on postnatal ultrasound, the diagnosis of UPJ obstruction is
less likely and the differential diagnosis would include ectopic
ureter, megaureter, UVJ obstruction, posterior urethral valves,
vesicoureteral reflux, or prune-belly syndrome (bilateral).
Sonographic evidence of multiple, noncommunicating cysts
is typical of parapelvic cysts or MCDK and not UPJ obstruc-
tion. Parapelvic cysts are rare in the pediatric population
and should not be mistaken for hydronephrosis.83 MCDK
can be confirmed by nuclear scintigraphy, typically a DMSA
scan, which reveals no evidence of ipsilateral renal function.
Historically, if the postnatal sonogram confirms character-
istic findings consistent with hydronephrosis, the newborn
was placed on a once-daily prophylactic dose of oral antibi-
otics with amoxicillin, 15 mg/kg, but the necessity of this
has come into question and frequently these patients will be
followed without prophylactic antibiotics, especially in the
circumcised male. The use of prophylactic antibiotics in in-
fants with postnatal hydronephrosis in the absence of VUR
continues to be controversial.84–86
OPEN SURGERY
The goal of surgery for UPJ obstruction is to preserve renal
function by assisting unobstructed drainage of the renal pel-
vis. Several techniques have been designed for this purpose.
The first is the flap techniques, which include Y-V-plasty,
spiral flap, or Scardino-Prince vertical flaps. These flaps all
use renal pelvic tissue to augment the narrowed UPJ segment
(Fig. 113-15). The second type of technique, the dismem-
bered pyeloplasty (Anderson-Hynes pyeloplasty), is the most
popular.8 It involves complete removal of the narrowed (dys-
functional) segment, tailoring of the renal pelvis (if necessary),
and reapproximation of the ureter to the renal pelvis in a de-
pendent position. Because neither technique sufficiently treats
all the possible anatomic variations, it is imperative to be fa-
miliar with all techniques. The anatomy of the renal pelvis,
in association with other intraoperative findings, dictates
the technique of choice. Dismembered pyeloplasty is appro-
priate for most repairs of intrinsic and extrinsic obstruction.
The flap technique may be useful in the patient with a long
narrowed segment. Each technique can provide a solution
to UPJ obstruction with success rates in excess of 90%
to 95%.13,87

Prenatal hydro.
Possible in-utero
↓Amniotic fluid intervention if
early enough
Unilateral hydro. Bilateral hydro. Severe hydro.
Birth
US + VCUG
US @ US (+/–) Low-moderate- US Evaluate for PUV
2–5 days at 2–5 days grade US

Hydro. PUV:
US 4–8 weeks not from Foley
PUV or catheter
reflux placement
and PUV tx
Low-grade Moderate Severe
hydro. hydro. hydro.

Follow w/
US 3–6 ↑Hydro. MAG-3 lasix renal
months scan, VCUG, US

FIGURE 113-14 A flow chart for

workup and management of US variable (3–6) months for 1–2 years
prenatal-detected hydronephrosis Repeat renal scan in 1 year (+/–)
(hydro.) without ureteral dilation. (hydro. not from VUR)
Low-grade hydro., SFU grade 1;
moderate hydro., SFU grade 2; PUV,
posterior urethral valves; renal scan,
MAG-3 lasix renal scan; severe hy- Low-grade Surgery:
dro., SFU grades 3-4; US, renal and hydro. or Worsening hydro.
bladder ultrasound; VCUG, voiding resolution Decrease in renal function
cystourethrogram; VUR, vesicouret- observe Symptomatic
eral reflux. Nonresolution (+/–)
1422 PART VIII GENITOURINARY DISORDERS

To perform a pyeloplasty, the child is given adequate
general anesthesia, with the option of using a regional block.
We routinely perform an RPG (retrograde pyelogram) at the
onset of the procedure. Although some have challenged the
necessity of this, we believe an RPG helps to characterize
the lesion and distal ureter and to assist in defining the appro-
priate site for optimum exposure. After RPG, a Foley catheter
is placed in a sterile manner and the child is repositioned for
surgery with respect to the particular approach that is
planned.
A number of approaches can be used to obtain access to the
retroperitoneal space to perform an open pyeloplasty includ-
ing anterior subcostal, flank, and dorsal lumbotomy. In infants
we prefer an anterior subcostal, muscle-splitting approach,

a'
b' 1
2
3
4
c' b' 5
a'
c'

D
FIGURE 113-15 A, Dismembered pyeloplasty. B, Anderson-Hynes pyeloplasty. C, Foley Y-V-plasty. D, Variation of the Foley Y-V-plasty for proximal ure-
teral high insertion.
 CHAPTER 113 URETEROPELVIC JUNCTION OBSTRUCTION 1423

Left kidney Possible excision

of redundant pelvis

c no. 6
1 ureteral
a a
catheter
Robinson
2 catheter
u.p. c
angle b b b-c
b
normal lumen
F

FIGURE 113-15—CONT’D E, Spiral flap. F, Scardino-prince vertical flap pyeloureteroplasty for long proximal ureteral narrowing. (From Fung LCT,
Lakshmanan Y: Anomalies of the renal collecting system: Ureteropelvic junction obstruction [pyelocaliectasis] and infundibular stenosis. In Belman AB,
King LR, Kramer SA [eds]: Clinical Pediatric Urology, 4th ed. London, Martin Dunitz, 2002, p 612.)

which provides excellent exposure through a small incision.
(In older children, a flank incision might be preferential given
the lateral displacement of the peritoneum and the frequently
thicker muscle bundles.)
The child is then placed in a modified supine position with
the ipsilateral side at an approximately 15- to 20-degree angle
to the operating table. Small children may have both upper
extremities at their sides. The operating table is flexed at the
level of the child’s anterior superior iliac spine, which exposes
the anterior abdominal area and flank. Approximately two fin-
gerbreadths below the subcostal region, a 3- to 4-cm incision
is made along Langer’s lines. A “muscle-splitting” technique is
used to separate the muscle layers without transecting them.
This technique provides adequate exposure and reduces post-
operative pain. After the fascia is opened, the peritoneum is
reflected medially. The Gerota fascia is incised in a plane ver-
tical to the patient. A self-retaining ring retractor provides
adequate exposure to the kidney and renal pelvis.
Once the kidney and renal pelvis have been exposed, care-
ful inspection of the anatomy should allow a decision to be
made regarding the best technique to use. A dismembered
pyeloplasty is performed for most UPJ obstructions. However,
there are three anatomic situations that may be treated appro-
priately with three different flap procedures: Foley Y-V-plasty,
spiral flap, or Scardino-Prince vertical flaps. The first situation
occurs when the obstructing segment is longer than 1.5 to
2 cm. The second situation occurs in a child with a small extra-
renal pelvis, and the renal pelvis does not need to be reduced.
The last situation, in which a Foley Y-V-plasty may be the pre-
ferred technique, is when the ureter has a high insertion into
the renal pelvis. In these cases, the pelvis can be funneled to
construct a dependent drainage system.
When a decision is made to perform a dismembered pye-
loplasty, the renal pelvis should be evaluated for possible re-
duction in size. The ureter is transected between stay
sutures at the UPJ. Care is taken to not excessively dissect
the blood supply from the proximal ureter. Stay sutures are
strategically positioned in a “diamond” pattern, and tenotomy
scissors are used to excise redundant renal pelvis within the
borders of the stay sutures. The narrowed ureteral segment
should be incised laterally along the longitudinal axis of the
ureter to the point of healthy ureteral tissue of normal caliber.
The proximal ureter at the UPJ obstruction level should be ex-
cised from and removed before reapproximation. The anasto-
mosis is begun at the inferior apex of renal pelvis and opened
ureter, with continuous running 7-0 absorbable monofila-
ment suture used in the infant and 5-0 to 7-0 absorbable su-
tures used in older patients. During the anastomosis, care
should be taken to include adequate adventitial tissue with
less mucosal tissue to provide a watertight anastomosis.
A temporary ureteral stent may be used when performing
the anastomosis to reduce the risk for obstruction as a result
of “back-walling.” Alternatively, a double-J ureteral stent may
be positioned in the distal end of the ureter and bladder with
the Seldinger technique. This is typically removed in 4 to 6
weeks.
The dorsal lumbotomy approach provides excellent expo-
sure, especially in the rare case of bilateral UPJ lesions
(Fig. 113-16). As in the subcostal approach, a muscle-splitting
technique can be used to access the retroperitoneum through
a relatively short distance. The dorsal lumbotomy approach
requires that the ureteral anatomy be precisely defined before
embarking on the technique because it provides limited expo-
sure to the distal part of the ureter.
1424 PART VIII GENITOURINARY DISORDERS
2 Longitudinal
Head roll thoracic rolls
(doughnut)
Transverse
subcostal roll
Supine position
A Vertical incision
Vertebral body
Sacrospinalis
Psoas major
Quadratus lumborum
Latissimus dorsi
Skin
C Renal exposure
FIGURE 113-16 Dorsal lumbotomy approach. (From Sheldon CA, Duckett JW: Infant pyeloplasty. AUA Update Series 1988;7[Lesson 37]:293.)
MINIMALLY INVASIVE SURGERY FOR
URETEROPELVIC JUNCTION OBSTRUCTION
IN CHILDREN
Unlike in adults, open pyeloplasty is considered the gold stan-
dard in the management and treatment of UPJ obstruction.
The success rate has been reported to be between 90% and
100%.88,89 However, open surgery can result in postoperative
morbidity because of pain, prolonged length in recovery, and
significant scar formation. Minimally invasive surgery evolved
to decrease the morbidity of pyeloplasty while maintaining the
efficacy of treatment. Minimally invasive techniques to correct
UPJ obstruction in adults include balloon dilation, percutane-
ous antegrade endopyelotomy, retrograde cutting-wire balloon
endopyelotomy, ureteroscopic retrograde endopyelotomy, and
laparoscopic pyeloplasty.
Techniques such as antegrade and retrograde endopyelot-
omy, although less invasive, have a significantly decreased suc-
cess rate when compared with open surgery, with a reported
success rate of 61% to 89%. They can also carry a risk of bleed-
ing in patients with crossing vessels.90,91 The endopyelotomy
success rate decreases to 39% to 44% with UPJ obstruction
associated with crossing vessels.92,93 Laparoscopic pyelo-
plasty was first introduced in adults in 1993 and has a success
rate approaching 95%.94,95 In the adult population, studies
have demonstrated decreased pain and decreased length of
hospital stay.96 Laparoscopy in children has shown similar
success rates, but the benefit from recovery for the younger
patient has been less well established.
Pediatric laparoscopic pyeloplasty faces the same chal-
lenges as the other minimally invasive techniques in chil-
dren.97 Much of pediatric urologic surgery is reconstructive
rather than extirpative, which limits the ability of many prac-
titioners to gain experience in basic laparoscopic skills. Peters
Transverse Sacrospinalis
incision
(Langer's)
Quadratus
(standard) lumborum
B Topographic anatomy
Spinous Incision
process
Gerota's fascia
Kidney
reported the first case of pediatric laparoscopic pyeloplasty in
1995.98 There have been several series demonstrating compa-
rable success rates with laparoscopic dismembered pyelo-
plasty to open pyeloplasty in the pediatric population.99–106
Recently, several authors have demonstrated that laparo-
scopic pyeloplasty is a viable option in infants and children
younger than 2 years of age and not solely limited to older
children.107,108
The two main laparoscopic techniques to pyeloplasty are
transperitoneal versus retroperitoneal approaches. The trans-
peritoneal approach is used more frequently in the pediatric
patient because it allows for increased working space along
with the benefit of easy identification of crossing vessels
and readily identifiable anatomic landmarks. The two main
procedures used to treat UPJ obstructions laparoscopically in-
clude dismembered pyeloplasty and Foley Y-V flap for UPJ ob-
structions secondary to high inserting ureters. There have
been a limited number of reports of higher success rates with
a dismembered approach when compared with a nondismem-
bered approach in the laparoscopic setting and thus is the
more commonly used.103
Various techniques are used to achieve pneumoperitoneum
including the Veress needle technique versus the open Hasson
technique. The patient is placed at 30 to 45 degrees in a mod-
ified flank with the affected side elevated. Peritoneal access is
obtained through the umbilicus via Hasson technique. The
abdomen is then insufflated to 10 to 12 mm Hg to allow for
adequate pneumoperitoneum. The 5- or 10-mm camera port
is placed at the umbilicus, and two 5-mm (in infants some-
times 3-mm ports can be used instead) working ports are
placed in the midline between the umbilicus and xyphoid
and in the midclavicular line 2 cm below the umbilicus on
the affected side. Initially, the colon is then mobilized medially
by incising the line of Toldt and reflecting the colon to gain
 CHAPTER 113 URETEROPELVIC JUNCTION OBSTRUCTION 1425

access to the retroperitoneum and kidney. On the left side, a defines a successful pyeloplasty as improvement in SFU-
transmesenteric approach can sometimes be undertaken. A graded hydronephrosis, temporal axial growth of the ipsilat-
ureteral stent is placed either antegrade during laparoscopy eral kidney, and a gradual increase in renal parenchyma.
or in a retrograde fashion during initial cystoscopy at the start Diuretic renography should demonstrate improved urinary
of the procedure depending on surgeon preference. The type drainage and differential renal function. In older children, suc-
of absorbable suture varies between 5-0 and 4-0 depending on cessful pyeloplasty is determined by the absence of symptoms.
the patient’s age and surgeon’s preference. There is a signifi- Depending on the definition, the success rate of pyelo-
cant learning curve to laparoscopic pyeloplasty, and it is tech- plasty for UPJ obstruction is as high as 98%.13,113,114 The suc-
nically challenging secondary to the intracorporeal suturing cess rate is so high, in fact, that one investigator suggested that
and knot tying required. This is why it has been relegated a satisfactory diuretic renogram 3 months after pyeloplasty
to specialized centers and not universally adopted. eliminated the need for further urologic follow-up. Other in-
The recent implementation of robotic laparoscopic systems vestigators claim that ideally, follow-up should be extended to
has increased the availability of laparoscopic reconstructive 2 years, which would include the period when initial symp-
surgery including pyeloplasties. The most commonly used toms of recurrence are most likely to take place.115 Subse-
robotic system is the da Vinci robotic system (Intuitive Sur- quently, renal ultrasound is performed annually for 2 years.
gical, Sunnyvale, Calif.). The benefits of robotic laparoscopy Of course, after pyeloplasty, primary care physicians should
include increased magnification, three-dimensional vision, monitor any patient with renal scars on a preoperative DMSA
articulating robotic Endowrist with six degrees of articulation, scan by annual blood pressure reading. We recommend renal
tremor filtering, and an ergonomic surgeon’s position. These ultrasound 3 months after pyeloplasty and diuretic renography
benefits to robotic laparoscopy assist dissection and suturing, if improvement is not obvious by ultrasound.
allowing for greater applications with decreased learning Complications of pyeloplasty can be categorized as acute
curve in reconstructive surgery. The patient is placed at 30 and late. Acute complications have a higher incidence in in-
to 45 degrees in modified flank with the affected side elevated. fants.6 Postoperative pyelonephritis is most commonly seen
Peritoneal access is obtained similar to conventional laparos- in children who had a positive intraoperative urine culture.
copy through the umbilicus. The 12-mm camera port is It is treated with intravenous antibiotics. Delayed opening
placed at the umbilicus, and two robotic working ports of the ureteropelvic anastomosis typically occurs as a result
(5 mm or 8 mm) are placed in the midline between the um- of edema and may lead to a prolonged leak around the anas-
bilicus and xyphoid and in the midclavicular line 2 cm below tomotic site. Interestingly, a transanastomotic ureteral stent
the umbilicus. The absorbable suture used varies with age but does not reduce the frequency of this complication. Patients
ranges between 7-0 and 4-0 similar to sutures used in an open with persistent hydronephrosis or symptoms after pyelo-
repair. Robotic surgery has the potential to advance minimally plasty may be salvaged with placement of a percutaneous
invasive surgery by allowing urologists with limited laparo- nephrostomy tube. In some of these children, a secondary
scopic experience to rapidly master the endocorporeal skills procedure may be necessary such as revision pyeloplasty
necessary to treat UPJ obstruction.109,110 or nephrectomy. Late complications can be manifested as
clinical symptoms or progressively worsening radiographic
studies. In either case, evaluation should determine whether
Outcome optimizing drainage via a nephrostomy tube or ureteral stent
------------------------------------------------------------------------------------------------------------------------------------------------
improves the clinical symptoms or renal function. Typically,
A successful pyeloplasty is most specifically measured by pres- surgery for a failed pyeloplasty should not be considered
ervation of or improvement of renal function. Several investi- for at least 2 months. There are five basic approaches to
gators have reported improvement in renal function in the treatment of secondary UPJ obstruction: revision pyelo-
children with obstruction detected early and surgery per- plasty, ureterocalicostomy, endopyelotomy, laparoscopy, and
formed before renal impairment occurs.1,111 Meticulous at- nephrectomy.116
tention to surgical principles contributes to operative
success. The use of proper pediatric instruments is important. Summary
Atraumatic tissue handling, careful dissection, and preserva- ------------------------------------------------------------------------------------------------------------------------------------------------

tion of ureteral blood supply all contribute to a successful UPJ obstruction is the most common congenital urinary ob-
anastomosis. The ureteropelvic anastomosis should be water- struction, and management remains complex because of the
tight and tension free. The retroperitoneal space should have variability of the lesion and the unpredictability of the mani-
adequate passive drainage. festations. In many cases, observation alone is a viable option
Finally, the routine use of ureteral stents and percutaneous because the obstruction is not progressive. In other cases,
nephrostomy tubes after an open pyeloplasty remains contro- radiographic studies reveal renal deterioration or progressive
versial.112 The most common reasons to leave a stent in place hydronephrosis, and surgery is required. Unfortunately, the fac-
after pyeloplasty are to ensure proper urinary diversion, main- tors that determine which kidneys require surgery have not
tain ureteral caliber, and ensure anastomotic alignment. The been clearly defined at this point. However, with future devel-
main reasons against leaving a temporary stent in situ are that opments in biochemical markers and improvements in ra-
they are foreign bodies and may be a nidus for infection, they diographic studies, timing of surgical intervention will be
may erode through the anastomotic site, and in children, defined earlier so that renal deterioration in children with
internalized stents require general anesthesia for removal. suspected UPJ obstruction might be prevented.
The goal of pediatric pyeloplasty is to reduce hydronephro-
sis and preserve renal function. This goal is evaluated by radio- The complete reference list is available online at www.
graphic studies and clinical considerations. Renal ultrasound expertconsult.com.
 hypertension, and decreased renal function. On the basis of
several studies, the incidence of renal scarring in young chil-
dren with febrile UTI may be as high as 40% to 50%.1 Among
infants between the ages of 2 months and 2 years who present
with fever, approximately 5% are diagnosed with a UTI.2,3 Age
and gender are significant factors in the incidence of UTI in
the pediatric population. The incidence of UTI is greater in
girls than in boys; in prepubertal girls, the incidence is
reported to be 3%, whereas in prepubertal boys it is only
1%.1 Some studies have demonstrated that the incidence of
UTI in uncircumcised males, particularly boys younger than
1 year, is 5 to 20 times higher than in circumcised males.4,5
Over the past few years, there have been advances in the un-
derstanding of the pathogenesis of UTI and the identification
of risk factors that predispose patients to renal damage.

PATHOGENESIS
There are four major pathways by which bacteria gain access
to the urinary tract: ascending, hematogenous, lymphatic,
and direct extension. Hematogenous access is most com-
monly seen in immunocompromised children or in neo-
nates, owing to their immature immune systems. Direct
CHAPTER 114 extension can occur in the setting of fistulas from the vagina
or the gastrointestinal tract to the genitourinary tract. The
ascending route is considered the most common pathway,
with bacteria colonizing the periurethral area and then enter-
Renal Infection, ing via the urethra. Host factors including the ability to resist
infection are important elements in the pathogenesis of UTI.
There is a higher incidence of UTI in infancy, likely due to
Abscess, the immaturity of the neonatal immune system and the
higher periurethral bacterial colonization in the first year
of life.6 Bacteria from the gastrointestinal tract colonize the
Vesicoureteral periurethral mucosa and may ascend into the bladder and
then the kidneys (Table 114-1).7 Certain bacterial traits such

Reflux, Urinary as the presence of fimbriae mediate bacterial adherence to

uroepithelial cells and increase bacterial virulence.
Differentiation of pyelonephritis from lower UTI is difficult
Lithiasis, and Renal because there is poor correlation between clinical symptoms
and localization of bacteria to the upper versus lower tract. In
older children, however, the diagnosis of pyelonephritis may
Vein Thrombosis be suggested by fever, flank pain, and systemic symptoms. Per-
sistent fever for more than 48 hours after the initiation of
appropriate antibiotics may be indicative of upper tract infection.
Leslie T. McQuiston and Anthony A. Caldamone Infants with a febrile UTI should be considered to have pyelone-
phritis and evaluated and treated accordingly. Alternatively, if
one has access to dimercaptosuccinic acid (DMSA) nuclear
scanning at the time of the illness, more specific identification
of pyelonephritis is possible, which may modify the evaluation
Urinary Tract Infection, Renal and treatment. Studies have demonstrated that 40% to 60% of
Abscess, and Vesicoureteral children with a febrile UTI have a positive DMSA scan.8,9
Reflux
------------------------------------------------------------------------------------------------------------------------------------------------ RISK FACTORS
The urinary tract is the second most common site of infection Although bacterial virulence plays an important role in the de-
in infants and children after the respiratory tract. The true velopment of UTI, a variety of host characteristics may predis-
incidence of urinary tract infection (UTI) in children is diffi- pose to UTI including age, gender, colonization factors,
cult to determine because of the challenges of diagnosis. constipation, genitourinary abnormalities, sexual activity, ge-
Young children may present with only fever or failure to netic factors, and iatrogenic factors (Table 114-2).10 The most
thrive, with no specific urinary symptoms or signs. Diagnosis important factors influencing prevalence are age and gender.
and treatment of UTI are of particular importance in children, Males are five to eight times more likely than girls to have UTIs
given the long-term complications including renal scarring, during the first 3 months of life.11 Beyond that age, girls have a

1427
1428 PART VIII GENITOURINARY DISORDERS

TABLE 114-1 than 60% of children younger than 1 year are found to have
Pathogens Associated with Infections VUR after their first UTI.3 This percentage decreases to 30% in
Organism Comment
2- to 3-year-olds and is 5% or less in adults.3 As children grow,
the submucosal tunnel elongates and the ratio between the
Escherichia coli Accounts for 70%-90% of infections submucosal tunnel length and the ureteral diameter increases.
Pseudomonas aeruginosa Most common nonenteric gram-negative In addition, filling and voiding pressures change over time,
pathogen
which can also improve VUR. These factors, along with
Occasionally seen in
immunocompromised patients
changes in bladder dynamics and voiding patterns, account
Enterococcus Most common gram-positive pathogen
for the spontaneous resolution rate of VUR. VUR is graded I
Group B streptococci Occasionally seen in neonates
to V on the basis of the International Reflux Grading System.
Staphylococcus aureus Suggests additional site (abscess,
Grades I and II generally resolve or improve over time, whereas
osteomyelitis, bacterial endocarditis) grades IV and V typically do not resolve spontaneously and
Proteus mirabilis Boys > 1 yr old may require intervention.
Candida, coagulase- Seen after instrumentation of urinary tract Functional abnormalities such as a neuropathic bladder
negative staphylococci predispose to UTI as a result of increased urinary tract pres-
Klebsiella Occasionally seen in sures, incomplete bladder emptying, and increased frequency
immunocompromised patients of instrumentation. UTI is also more common in patients
with bladder and bowel dysfunction (BBD). These children
Modified from Handel LN, Caldamone AA: Urinary tract infections in the
pediatric population. Lebanese Med J 2000;52:194. commonly have incomplete bladder emptying, urinary sta-
sis, and abnormally high urinary tract pressures, as well as
bladder instability, infrequent voiding, Hinman syndrome,
greater incidence of UTIs than boys. Symptomatic infections and constipation. Periurethral colonization and preputial
are 10 to 20 times more likely in preschool girls than boys.5 colonization also may increase the risk of UTI. Preputial aer-
UTI often serves as a marker for anatomic abnormalities of obic bacterial colonization is highest during the first month
the genitourinary tract. It is important to identify these abnor- after birth, decreases after 6 months, and is uncommon after
malities early because, if uncorrected, they may lead to recur- 5 years of age.10,14
rent infection and possible loss of renal parenchyma.
Obstructive malformations such as ureteropelvic junction
CLINICAL PRESENTATION
obstruction, posterior urethral valves, ectopic ureters, ureter-
oceles, and urethral diverticula can increase the risk of UTI. Children with UTIs, especially those younger than 2 years, do
Renal abnormalities such as papillary necrosis, nonfunction- not always present with the typical symptoms of dysuria, fre-
ing kidneys, and unilateral medullary sponge kidney also quency, or pain, and the physical examination may be of lim-
predispose patients to UTI. ited value. Thus it is important to maintain a high index of
Vesicoureteral reflux (VUR), which is the abnormal flow of suspicion in infants and young children. In children younger
urine from the bladder to the kidney, allows bacteria from the than 2 years, parents may describe nonspecific symptoms
bladder to gain access to the kidney. By itself, VUR does not such as fever, fussiness, vomiting, or diarrhea; patients may
cause UTI; bacteriuria must be present in the setting of present with failure to thrive or difficulty feeding. In infants,
VUR for UTI to develop. The exception to this rule is a child often the only presenting symptom is fever. Children 2 to
with massive VUR who may have a significant postvoid resid- 5 years of age often present with fever and abdominal pain co-
ual from refluxed urine. VUR occurs in approximately 1% of incident with upper respiratory infection. After toilet training,
the general population and is estimated to occur in 20% children may start to demonstrate more specific symptoms
to 35% of children evaluated for bacteriuria.12,13 More such as fever, dysuria, flank pain, and urgency.

TABLE 114-2 DIAGNOSIS

Risk Factors for Urinary Tract Infection
Accurate diagnosis of UTI in children is critical. The method of
Anatomic abnormalities collection can affect the accuracy of diagnosis (Table 114-3).
Vesicoureteral reflux In toilet-trained children, clean-catch specimens have a contam-
Obstruction ination rate of approximately 30%.15 The method least likely to
Other: diverticulum, labial adhesion be contaminated is suprapubic aspiration; this method, how-
Female gender ever, is often not favored by parents or by primary practitioners.
Uncircumcised male Suprapubic aspiration may be contraindicated in patients with
Voiding dysfunction certain anatomic abnormalities of the genitourinary tract.
Constipation Transurethral bladder catheterization is the preferred
Instrumentation method, with a lower false-positive rate than perineal bag col-
P-fimbriated bacteria lection. The sensitivity of positive urine culture is 95%, and
Toilet training the specificity is 99%.2 In children not yet toilet trained, per-
Sexual activity, pregnancy ineal bag collection is more common but has a high probabil-
ity of contamination. False-positive results occur in 85% of
From American Academy of Pediatrics, Committee on Quality Improvement,
Subcommittee on Urinary Tract Infection: Practice parameter: The specimens collected in this manner.2 Perineal bag collection
diagnosis, treatment, and evaluation of the initial urinary tract infection in may be used when there is low suspicion of UTI. Current
febrile infants and young children. Pediatrics 1999;103:843. recommendations are urethral catheterization or suprapubic
 CHAPTER 114 RENAL INFECTION, ABSCESS, VESICOURETERAL REFLUX, URINARY LITHIASIS, AND RENAL VEIN THROMBOSIS 1429

TABLE 114-3
Criteria for the Diagnosis of Urinary Tract Infection
Method of Collection Colony Count (Pure Culture) Probability of Infection (%)
Suprapubic aspiration Gram-negative bacilli: any number >99
Gram-positive cocci:
>a few thousand
Transurethral catheterization >100,000 95
10,000-100,000 Infection likely
1000-10,000 Suspicious, repeat
<1000 Infection unlikely
Clean-voided (boy) >10,000 Infection likely
Clean-voided (girl) 3 specimens >100,000 95
2 specimens >100,000 90
1 specimen >100,000 80
50,000-100,000 Suspicious, repeat
10,000-50,000 If symptomatic, suspicious, repeat; if asymptomatic, infection unlikely
<10,000 Infection unlikely

aspiration following a positive urinalysis or a positive urine

culture from a bag specimen to eliminate the possibility of
contamination.
On urinalysis, the combination of urine nitrite, leukocyte
esterase, bacteria, and white blood cells on microscopy yields
a sensitivity of 99.8% and a specificity of 70%.2,15 However,
UTI cannot be ruled out if any of these findings is absent on
urinalysis or microscopy. A urine culture is the only way to
accurately diagnose a UTI; however, prompt treatment be-
fore final culture results are obtained is recommended due
to the increased risk of renal scarring in children younger
than 5 years, as well as the possibility for rapid clinical
deterioration.

IMAGING
The role of imaging after a UTI is controversial.10,13,16 The pri- FIGURE 114-1 Sagittal ultrasonography of a kidney demonstrating
mary goal of imaging is to identify children who are at risk for hydronephrosis of the renal pelvis and calices. (Modified from Handel
subsequent UTIs and renal damage. Imaging includes an an- LN, Caldamone AA: Urinary tract infections in the pediatric population.
atomic evaluation of the kidney and bladder, as well as a func- Lebanese Med J 2004;52:194.)
tional study to identify the presence of VUR. The current
American Academy of Pediatrics guidelines recommend renal
and bladder ultrasonography (US) for all children younger
than 2 years with a first UTI.2 The National Institute for Health
and Clinical Excellence (NICE) recommends renal and blad-
der US for all children younger than 3 years of age, but only
for those with atypical or recurrent UTI for those older than
3 years of age.17 The purpose of US in these cases is to assess
for the presence of any congenital structural abnormalities of
the urinary tract such as hydronephrosis (Fig. 114-1).
Additionally, US allows the visualization of any renal paren-
chymal abnormalities such as cysts and identifies the presence
of perinephric fluid collections. Ureteral dilatation, bladder
wall thickening (Fig. 114-2), a duplicated collecting system
(Fig. 114-3), ureteroceles (Fig. 114-4), bladder diverticula,
and calculi are also identifiable with US. In cases of acute
pyelonephritis, US may demonstrate focal or diffuse renal en-
largement, as well as abnormal cortical echogenicity. Multiple
studies have demonstrated obstructive lesions in 5% to 10% of
patients and VUR in 20% to 50% following a UTI.18 Thus
imaging should be considered after a first UTI. Imaging for FIGURE 114-2 Transverse ultrasonography of the bladder demonstrat-
VUR using cystourethrography (VCUG) is the standard of care ing posterior wall thickening (arrows).
1430 PART VIII GENITOURINARY DISORDERS

Upper pole Lower pole decreases the probability that there is high-grade VUR.8 Chil-
dren with a negative scan may not require further evaluation
unless UTI recurs. Further studies are necessary to clarify the
best evaluation strategy to detect children who are at risk of
progressive renal damage from UTI. Currently based on age
and gender, the recommendation is US and VCUG for all boys
with their first UTI regardless of age and for girls younger than
5 years with their first UTI. Additionally, for girls older than 5
years, imaging should be done in those with pyelonephritis or
recurrent UTIs (Table 114-4). The type and timing of imaging
may depend on the patient’s response to initial antibiotic ther-
apy and the availability of DMSA scintigraphy.
The most common anomaly associated with UTI in chil-
dren is VUR, which is noted in about 20% to 35% of children
evaluated after UTI. VCUG is used to detect the presence of
VUR (Fig. 114-5). Standard VCUG is performed by instilling
iodinated contrast into the bladder and then imaging during
filling and voiding. The American Academy of Pediatrics rec-
ommends VCUG as the initial study of choice in VUR, espe-
cially in male patients to rule out posterior urethral valves
FIGURE 114-3 Sagittal ultrasonography of a kidney demonstrating a (Fig. 114-6).2 Previous recommendations were that VCUG
duplicated collecting system. should be done 4 to 6 weeks after UTI because of possible
false-positive results owing to inflammation of the bladder
in infants and children younger than 3 years or not yet toilet wall causing VUR. More recent studies have shown no differ-
trained. In older children, the data to support VCUG are less ence between early (within 7 days of infection) and late
clear because it is not certain, on the basis of available data, VCUG.20 VCUG can be performed once the child is afebrile,
that the treatment of children with VUR provides a clinically and cultures are negative to prevent the risk of infected urine
significant benefit with respect to reduction in long-term renal being forced in a retrograde fashion during the study.
damage.19 The “Top-Down Approach” has suggested that ini-
tial DMSA scintigraphy may help to identify children at high-
est renal risk and therefore those who need further TABLE 114-4
investigation by VCUG.9 This approach is based on evaluation Pediatric Genitourinary Imaging
with DMSA scan as close to the acute episode of febrile UTI as
possible, preferably within the first 10 days of diagnosis, and Study Findings Indications
further imaging with VCUG for those with a positive scan. Renal/ Abscess All boys with first UTI
Several studies have shown that a negative DMSA scan bladder Presence of 2 kidneys
US Renal size
Presence of
Bladder Ureterocele hydronephrosis (see
Fig. 114-1)
Bladder wall thickening
(see Fig. 114-2)
Postvoid residual
Duplicated collecting
system (see Fig. 114-3)
Ureterocele (see
Fig. 114-4)
VCUG Vesicoureteral reflux All boys with first UTI; girls
(see Fig. 114-5) <5 yr with first UTI
Posterior urethral
valves (see Fig. 114-6)
MAG3 Renal function Hydronephrosis on US, no
diuretic reflux on VCUG
scan
Renal obstruction
DMSA Pyelonephritis To diagnosis acute
renal pyelonephritis (“top-down”
scan approach)
To assess for renal scarring

DMSA, dimercaptosuccinic acid; MAG3, mercaptoacetyltriglycine; US,

ultrasonography; UTI, urinary tract infection; VCUG, voiding
cystourethrography.
FIGURE 114-4 Transverse ultrasonography of the bladder demonstrat- Modified from Handel LN, Caldamone AA: Urinary tract infections in the
ing a left ureterocele. pediatric population. Lebanese Med J 2004;52:194.
 CHAPTER 114 RENAL INFECTION, ABSCESS, VESICOURETERAL REFLUX, URINARY LITHIASIS, AND RENAL VEIN THROMBOSIS
FIGURE 114-5 Voiding cystourethrogram demonstrating vesicoureteral
reflux.
Renal cortical imaging with dimercaptosuccinic acid
(DMSA) is both sensitive and specific for the detection of py-
elonephritis in children.18,21 Areas with decreased uptake
may represent acute pyelonephritis or renal scarring. Before
the availability of DMSA, imaging techniques demonstrated
that approximately 17% of children with bacteriuria had renal
scarring; current studies using DMSA demonstrate that the
percentage is likely twice that.1 The use of DMSA should be
limited to those patients with suspected acute UTI in whom
the diagnosis is unclear but there is a high suspicion of
FIGURE 114-6 Voided cystourethrogram demonstrating posterior ure-
thral valves (arrow).
1431
pyelonephritis. DMSA can also be used to assess for renal scar-
ring after UTI if the results would change the management
of the patient.
Mercaptoacetyltriglycine (MAG3) or diethylenetriamine
pentaacetic acid (DTPA) with a diuretic is used to assess renal
function and excretion in patients with hydronephrosis with-
out VUR and to determine the presence or absence of obstruc-
tion. Both false-positives and false-negatives are possible, so
caution must be used when interpreting the study and clinical
correlation is essential.
Renal abscess is a potential complication of pyelonephritis
regardless of the underlying cause. Abscess is more common
in compromised patients. Gram-positive organisms such as
Staphylococcus aureus and streptococci may be the causative
agents. The diagnosis is best made by US or computed tomog-
raphy (CT). Treatment may involve long-term antibiotics, per-
cutaneous drainage and antibiotic instillation, or open surgical
drainage. Sequential follow-up with radiographic imaging is
essential.
TREATMENT
Because of the risk of renal scarring in children, prompt diag-
nosis and treatment are critical (Fig. 114-7). The treatment
strategy depends on various factors, particularly the child’s
age and the severity of illness. Children younger than 4 weeks
with fever are at increased risk for bacteremia and sepsis and
may require hospitalization for parenteral antibiotics. Simi-
larly, any child younger than 5 years with a suspected UTI
who appears systemically ill should be hospitalized. The
American Academy of Pediatrics recommends that children
who are dehydrated, are unable to tolerate oral medication,
and appear toxic should receive intravenous antibiotics.2 Ad-
ditionally, children who are immunocompromised, are
suspected of having urosepsis or bacteremia, or have failed
oral therapy should be hospitalized. If the child can tolerate
oral antibiotics and has reliable parents who will maintain
follow-up and contact with the physician, the child may be
managed as an outpatient.
Initial antibiotic treatment needs to be broad spectrum to
cover the more common pathogens. Once culture results
and sensitivities are complete, treatment can be tailored spe-
cifically to cover the identified organism. In patients who have
had previous UTIs, the prior culture results should be consid-
ered when selecting the initial therapy. For children who are
hospitalized, parenteral treatment such as a combination of
aminoglycosides and ampicillin or cephalosporins is contin-
ued for 48 to 72 hours until the child is afebrile and stable
and sensitivities are available. Once the child is stable, oral an-
tibiotics can be initiated on the basis of sensitivity results. Most
studies recommend treating febrile UTIs in young children for
7 to 10 days in total,22,23 although more recent studies suggest
that shorter-duration therapy may be equally effective.24–26
The American Academy of Pediatrics recommends that the
treatment duration be 7 to 10 days for uncomplicated UTI
and 14 days for children with presumed renal involvement
or who are toxic.2 In infants younger than 6 weeks, ampicillin
and gentamicin or ampicillin and a cephalosporin are good
initial choices. The exception is ceftriaxone, which can
increase the risk of kernicterus in young infants. For children
likely to be infected with Pseudomonas or Enterococcus, ampi-
cillin and gentamicin are the agents of choice because
1432 PART VIII GENITOURINARY DISORDERS

Child with fever

Yes Toxic No

Oral antibiotics
Obtain urinalysis by most convenient method
IV antibiotics and consider hospitalization
Obtain urine via SPA or transurethral cath; urine culture
UA positive for LE, nitrite or
WBCs

Obtain urine via SPA or transurethral cath for

culture
Culture positive

No Yes

No UTI
Antibiotics for 7-14 days based on culture results
Clinical response in
48 hours

No Yes

US now

VCUG when convenient US and VCUG when convenient

FIGURE 114-7 Algorithm for management of urinary tract infection (UTI). LE, leukocyte esterase; SPA, suprapubic aspiration; UA, urinalysis; US, ultraso-
nography; VCUG, voiding cystourethrography; WBC, white blood cell.

cephalosporins have no effect on enterococci. Gentamicin limiting its use in children with symptomatic UTI. Generally,
levels should be monitored closely, particularly in children the antibiotic used for prophylaxis should be different from
with impaired renal function. the antibiotic used to treat the acute infection.
In patients who are clinically stable, a broad-spectrum oral
antibiotic is recommended. In a well-appearing child, a ceph-
TABLE 114-5
alosporin, trimethoprim-sulfamethoxazole (TMP-SMZ), or
Antibiotics Used to Treat Pediatric Urinary Tract Infections
nitrofurantoin is a common choice. TMP-SMZ should not
be used in children younger than 2 months because of the in- Method of
creased risk of kernicterus and blood dyscrasias. Nitrofuran- Delivery Antibiotic Additional Information
toin, which is excreted in the urine and does not achieve Parenteral Ampicillin and Check gentamicin levels
therapeutic concentrations in the bloodstream, should not gentamicin
be used in febrile infants with UTI or in children in whom Ampicillin and No Enterococcus or
there is concern about renal involvement. The fluoroquino- cephalosporin Pseudomonas coverage
lones have recently been authorized for pediatric use.11,27 Ceftriaxone/ Contraindicated in
cefotaxime infants < 2 mo old
Table 114-5 provides a summary of the commonly used
Oral TMP-SMZ Contraindicated in
antibiotics. infants < 2 mo old
Low-dose antibiotic prophylaxis should be considered in Amoxicillin Increasing resistance by
all children with a febrile UTI awaiting a full evaluation with Escherichia coli
renal US and VCUG. Children with VUR, recurrent UTIs, or Nitrofurantoin Inadequate if renal
partial obstruction or who are immunocompromised should involvement
be considered for prophylactic antibiotics, although the clin- Cephalosporins No Enterococcus or
ical benefit remains in question.17,19 The ideal prophylactic Pseudomonas coverage
agent provides high urinary antibiotic excretion with low se- Ciprofloxacin Recently approved in children
rum levels. Amoxicillin, nitrofurantoin, and TMP-SMZ are the TMP-SMZ, trimethoprim and sulfamethoxazole.
most commonly prescribed agents. However, Escherichia coli Modified from Handel LN, Caldamone AA: Urinary tract infections in the
has demonstrated increasing resistance to amoxicillin, thus pediatric population. Lebanese Med J 2004;52:194.
 CHAPTER 114 RENAL INFECTION, ABSCESS, VESICOURETERAL REFLUX, URINARY LITHIASIS, AND RENAL VEIN THROMBOSIS
In patients with normal studies and recurrent UTIs—
defined as three or more infections within a 6-month
period—prophylactic antibiotics may be indicated. In these
cases, one must consider other factors such as the patient’s
age, as well as the patient’s biologic predisposition for UTIs
and voiding pattern. For children started on prophylactic an-
tibiotics, the rate of infection usually decreases; however, the
risk of developing another UTI returns to baseline once pro-
phylactic antibiotics are discontinued. These patients should
be carefully screened for bladder and bowel dysfunction.
In cases of asymptomatic bacteriuria, current recommenda-
tions include strict follow-up without antibiotic treatment.
VUR treatment depends on several factors: age at presenta-
tion, grade of reflux, presence or absence of voiding dysfunc-
tion, comorbidities, family compliance with a treatment
protocol, and choice. It is helpful to divide VUR into primary
and secondary categories. Primary reflux refers to a congenital
abnormality of the vesicoureteral junction resulting in a fore-
shortened ureteral tunnel and poor detrusor backing of the
ureter. The lack of detrusor backing prevents the ureter from
being collapsed as the bladder fills (Fig. 114-8). Secondary re-
flux refers to other anatomic or functional abnormalities that
may lead to reflux such as bladder outlet obstruction or an
abnormality of the vesicoureteral junction (Table 114-6).
In the majority of patients with VUR, antibiotic prophy-
laxis, UTI surveillance, a regular voiding pattern, constipation
management, and regular follow-up US and VCUG are the
mainstays of therapy.28 Indications for surgical correction
are debatable, but most would agree with the following:
breakthrough upper UTIs, renal scarring developing on anti-
biotic prophylaxis, poor renal growth, failure to follow a
treatment plan, high-grade VUR (except in young infants), a
fixed anatomic defect of the ureterovesical junction, and per-
sistence of VUR followed for 2 years or longer associated with
symptoms of BBD or UTI.28
VUR in a child with a recognized bladder and bowel dys-
function requires special mention. It is clear that if the bladder
and bowel dysfunction (including constipation, if present) is
managed well, the incidence of breakthrough UTIs and renal
scarring and the need for surgical correction of VUR are sig-
nificantly reduced.28–31 Voiding dysfunction is a common
cause of failed ureteral reimplantation.32
Normal Reflux
FIGURE 114-8 Anatomy of the normal and refluxing ureterovesical junc-
tion (UVJ). Note the lack of detrusor backing of the refluxing UVJ.
1433
TABLE 114-6
Causes of Secondary Reflux
Anatomic
Posterior urethral valves
Ureterocele
Diverticulum
Ectopic ureter
Prune-belly syndrome
Bladder exstrophy
Functional
Voiding dysfunction
Neuropathic bladder
Myelodysplasia
Sacral agenesis
Many surgical techniques exist using intravesical and extra-
vesical approaches. They all require the establishment of an
intramural ureter with sufficient muscular backing to provide
adequate ureteral wall coaptation with filling and contraction
of the bladder such that retrograde flow of urine is prevented
(Figs. 114-9 to 114-11). This approach requires an average
hospital stay of 48 hours or less and has a complication rate
of about 4%, 2% to 2.5% due to ureteral obstruction and
2% to 2.5% due to persistent VUR. The Pediatric Vesicouret-
eral Reflux Guidelines Panel delineated an open surgical
success rate of 98% and a rate of 50% to 92% with endoscopic
correction of VUR.28
In 1981 Matouschek33 introduced the concept of endo-
scopic correction of urinary incontinence and VUR by polyte-
trafluoroethylene (Teflon) paste injection. This method was
developed and popularized by Puri and O’Donnell, initially
in a piglet model in 1984 and then applied to children with
VUR.34,35 This procedure causes little postoperative discom-
fort and can be performed as an outpatient procedure
(Figs. 114-12 and 114-13). Owing to the potential risks of
particle migration and malignancy induction,36 Teflon is not
in current use.32 Many other injectable materials have been in-
vestigated and applied to animal models and humans for VUR
correction. Success rates approach 75% to 80%, depending on
the grade of reflux and the number of injections.37–40 Today,
the most commonly used materials worldwide are dextranomer
macrospheres (Deflux, Oceana Therapeutics, Edison, N.J.)
and polydimethylsiloxane with a water-based biocompatible
carrier gel.
Urinary Lithiasis
------------------------------------------------------------------------------------------------------------------------------------------------
HISTORY AND INCIDENCE
Urinary stones, in both the bladder and the kidney, have been
reported in Egyptian mummies dating from 4800 BC,41 and
urolithiasis has long been studied in an effort to understand
its pathogenesis and to refine treatment strategies. Although
urolithiasis has been recognized in children for centuries,
the clinical picture, evaluation, and management continue
to evolve.
Approximately 7% of all stones occur in children younger
than 16 years.42 Traditionally, urolithiasis was characterized
by bladder calculi in children of developing countries; the
1434 PART VIII GENITOURINARY DISORDERS

FIGURE 114-9 Leadbetter-Politano

A B
C D
ureteral reimplantation. A, The ureter
is catheterized through a midline
bladder incision. B, The ureter is dis-
sected from the surrounding detru-
sor until it is totally mobilized. C, A
neohiatus is created by passing a
right-angle clamp into the bladder.
A subepithelial tunnel is created,
and the detrusor defect is closed.
D, The ureter is then passed through
the tunnel, and its orifice is matured.
(From Rowe M, O’Neill JA, Grosfeld JL,
et al: Essentials of Pediatric Surgery.
St Louis, Mosby-Year Book, 1995.)

A B
C D
FIGURE 114-10 Cohen cross-trigonal ureteral reimplantation. A, Both
ureters are catheterized. B, Both ureters are fully mobilized, as in the
Leadbetter-Politano procedure. C, The detrusor is approximated around
the ureter to create a normal neohiatal caliber, and two transtrigonal tun-
nels are created. D, The ureters are then passed through the respective
tunnels, and the orifices of the ureters are matured. (From Rowe M, O’Neill
JA, Grosfeld JL, et al: Essentials of Pediatric Surgery. St Louis, Mosby-Year
Book, 1995.)
incidence of upper tract calculi, occurring mainly in industri-
alized areas, was much lower in children than in adults. Also,
in comparison with adult stone formers, children were more
likely to demonstrate risk factors outside the metabolic realm
such as UTI, anatomic abnormalities, and surgical alterations
in the urinary tract. Currently, the incidence of upper tract cal-
culi in children without these predisposing factors is on the
rise worldwide,43,44 and the paradigms are changing. The im-
portance of metabolic evaluation in children with urolithiasis
has been shown, even in countries thought to have primarily
endemic bladder stone disease based on diet. With the intro-
duction of smaller endoscopic instruments and the refinement
of extracorporeal shock wave lithotripsy (ESWL) technology,
treatment of pediatric stone disease now closely parallels stone
management in adults.
SPECTRUM OF DISORDERS
Although there are a number of contributing factors, the cen-
tral concept in urolithiasis is urinary supersaturation. When a
solute is added to a solvent, it dissolves until a certain concen-
tration is reached, at which point the solution is saturated. Be-
yond this point, the solute may form crystals in the solution,
and those crystals may aggregate. Supersaturation occurs
when this point is surpassed and crystal precipitation occurs
in the urine in the form of nucleation, the basis of urinary
stones. Crystallization and aggregation must also occur for
stones to form. These processes are influenced by the presence
of inhibitors and promoters. Citrate, magnesium, pyrophos-
phate, glycosaminoglycans, nephrocalcin, and Tamm-Horsfall
proteins are inhibitors of crystallization and aggregation.
Bacterial infections and anatomic abnormalities such as ob-
struction or stasis may encourage crystal aggregation and
retention, thus increasing the risk of clinically significant uro-
lithiasis.45,46 Multiple studies of urolithiasis in children have
shown metabolic abnormalities in up to 92% of patients, with
hypercalciuria and hypocitraturia the most common.46–52
Interestingly, even in children with urolithiasis and anatomic
obstruction, the prevalence of urinary metabolic abnormali-
ties is high.45,49,53 The prevalence of infection-related stones
is especially high in children younger than 6 years.54 Ap-
proximately 12% of children with urolithiasis have no identi-
fiable risk factor.55 Urinary stone disease can, therefore, be
classified as metabolic, anatomic, infectious, or idiopathic,
on the basis of underlying factors. The relative contribution
of these factors, which may overlap in some cases, is shown
in Table 114-7.
Stones can also be classified on the basis of their location in
the upper urinary tract (kidneys and ureters) or lower urinary
tract (bladder and urethra) and as symptomatic or asymptom-
atic. These classifications may influence the decision-making
process with regard to treatment options.
CLINICAL PRESENTATION
In adults, upper tract calculi present in a characteristic fashion
in the form of renal colic. This severe, intermittent, refractory
pain, often accompanied by nausea and vomiting, is a less
common presentation in children. Although classic renal colic
does occur in children, more commonly, stones involving
both the upper and lower tract are detected during the radio-
logic evaluation for UTI or hematuria. Younger children in
 Anchoring sutures

A B C

FIGURE 114-11 Extravesical detrusorrhaphy—a modification of the Lich-Gregoir procedure. A, The ureter is mobilized for several centimeters externally.
The detrusor is then incised in a caudal direction and carried around the ureteral insertion. B, The ureter is left attached only to the underlying bladder
mucosa. C, The distal ureter is advanced and fixed with two sutures. Then the detrusor is approximated over the ureter to recreate the tunnel, which is now
longer than before. (From Rowe M, O’Neill JA, Grosfeld JL, et al: Essentials of Pediatric Surgery. St Louis, Mosby-Year Book, 1995.)

Ureteral
orifice
Ureteral
orifice

Endoscopic Injection Ureter

needle material

B
A

Endoscopic
needle

Crescent or
volcano-like
appearance
Narrowed
ureteral
orifice

Crescent or
volcano-like
appearance Ureter
C

Injection
material

Withdrawn
endoscopic
D
needle
FIGURE 114-12 Endoscopic technique for reflux connection. A, The needle is placed at the 6 o’clock position, bevel up. B, Slow injection in subureteral
space. C, Postinjection appearance. D, Appropriate position of injected material. (From Russinko PJ, Tackett LD: Endoscopic correction of reflux. In Calda-
mone AA [ed]: Atlas Urol Clin North Am 2004;12:55.)
1436 PART VIII GENITOURINARY DISORDERS

B
FIGURE 114-13 Endoscopic injection. A, Left and right ureteral orifices before injection. B, Left and right ureteral orifices demonstrating the crescent
appearance after injection.

TABLE 114-7
Clinical Diagnoses in Children with Urolithiasis
Mayo Clinic North America Europe
(N ¼ 221) (N ¼ 492)* (N ¼481)*
No. of Patients No. of Patients No. of Patients
Diagnosis (%) (%) (%)
Metabolic disorder 115 (52.0){ 162 (32.9){ 59 (12.3)
Idiopathic hypercalciuria 38 39 36
Immobilization 8 39 4
Uric acid stones 8 22 2
Endemic stones (urate) 0 10 0
Cystinuria 15 15 9
Hyperoxaluria 25 12 5
Renal tubular acidosis 4 10 2
Other 38{ 25 1
Developmental anomalies of genitourinary tract 66 (29.9){ 160 (32.5) 145 (30.1)
Infection
Primary 41 (18.6){ 21 (4.3) 209 (43.5)
All causes — 215 (43.7) 358 (74.4)
Unknown cause 55 (24.9) 139 (28.3) 68 (14.1)

*Data from Polinsky MS, Kaiser BA, Bacuarte HJ: Urolithiasis in childhood. Pediatr Clin North Am 1987;34:683.
{
Metabolic disorders often coexisted with genitourinary tract anomalies and infections.
{
More than one disorder in some patients.
From Liebermann E: Importance of metabolic contributions to urolithiasis in pediatric patients [editorial]. Mayo Clin Proc 1993;68:313.
 CHAPTER 114 RENAL INFECTION, ABSCESS, VESICOURETERAL REFLUX, URINARY LITHIASIS, AND RENAL VEIN THROMBOSIS 1437

1.20 should be obtained to assess for microscopic hematuria, urine

pH, crystals in the sediment, and UTI. Baseline serum electro-
Other
1.00 lytes, blood urea nitrogen, creatinine, calcium, uric acid,
Infection
phosphorus, and magnesium levels should also be obtained.
A 24-hour urinalysis for stone risk assessment should be
Fraction of patients

0.80 obtained after treatment of the presenting stone.

X-ray finding
0.60
Pain
0.40
0.20 Hematuria
0-5 6-11
Age at onset, yr
FIGURE 114-14 Clinical features at the time of initial assessment of 221
children and adolescents with urolithiasis who were examined at the Mayo
Clinic between 1965 and 1987. (From Milliner DS, Murphy ME: Urolithiasis
in pediatric patients. Mayo Clin Proc 1993;68:241.)
particular may also present with nonspecific abdominal pain
accompanied by microscopic hematuria or, less frequently,
with outlet obstruction (Fig. 114-14).
DIAGNOSIS
History and physical examination are key components to rule
out other causes of abdominal and back pain. The history
should include a dietary and urologic history. One clue to
the presence of urolithiasis is a positive family history for kid-
ney stones, which may be present in up to 37% of children
with stone disease.56 Other underlying medical conditions
that may predispose children to urolithiasis include hyper-
parathyroidism, renal tubular acidosis, inflammatory bowel
disease, seizure disorders treated with a ketogenic diet,57 cys-
tic fibrosis, and Dent disease.58 Urinalysis and urine culture
Radiologic evaluation may include plain abdominal radiog-
raphy, US, intravenous pyelography, and noncontrast CT. In
adults, the standard imaging study is noncontrast CT because
of its sensitivity and specificity. In children, however, these
studies are limited, although some authors recommend CT
as the first-line imaging study.59,60 Others recommend initial
abdominal US because of the varied presentation of urolithia-
sis in children and because of concerns about radiation expo-
sure.61,62 Plain abdominal radiography may not be helpful in
12-16
the initial diagnosis but may assist in planning treatment on
the basis of the ability to visualize the stone.
TREATMENT
Four main factors affect initial treatment decisions: the clinical
scenario, stone composition, stone size, and stone location
(Table 114-8). In the setting of potential urinary tract obstruc-
tion with fever or complete obstruction, urgent relief of the ob-
struction is warranted, either by placement of a ureteral stent
or by percutaneous nephrostomy. The decision between ure-
teral stent placement and percutaneous nephrostomy is de-
pendent on physician preference, available resources, and
patient size and anatomy. In children, nephrostomy drainage
may be performed without general anesthesia and may be the
procedure of choice. After relief of the obstruction and subse-
quent treatment of infection, definitive therapy directed at
stone clearance can be undertaken.
The most common composition of upper tract calculi is cal-
cium oxalate. The stones are radiopaque and do not respond
to dissolution therapy. Calcium oxalate monohydrate stones
may be resistant to ESWL, whereas calcium oxalate dihydrate

TABLE 114-8
Factors Affecting the Treatment of Stones
Factor Treatment Consideration
Clinical scenario
Bilateral obstruction (
Obstruction of a solitary kidney For all scenarios, urgent relief of obstruction via stent or nephrostomy
Fever/UTI with potential obstruction
Intractable pain
Stone composition
Uric acid Consider chemodissolution
Struvite Continue antibiotic therapy throughout treatment
Cystine Responds poorly to ESWL
Calcium oxalate Radiopaque; may respond well to ESWL
Stone size
<4 mm Approximately 90% chance of spontaneous passage
4-6 mm Approximately 50% chance of spontaneous passage
>6 mm Approximately 10%-20% chance of spontaneous passage
Stone location
Renal ESWL or PCNL
Proximal ureteral Ureteroscopic extraction (antegrade) or ESWL
Distal ureteral Ureteroscopic extraction (retrograde) or ESWL
Bladder Cystolithotomy or cystolitholapaxy

ESWL, extracorporeal shock wave lithotripsy; PCNL, percutaneous nephrostolithotomy; UTI, urinary tract infection.
1438 PART VIII GENITOURINARY DISORDERS
calculi, which are more common, generally respond well. Uric
acid calculi are radiolucent and form at a low urinary pH. They
are often identified by their characteristic Hounsfield units on
noncontrast CT. Uric acid stones may be dissolved and pre-
vented by alkalinization of the urine (pH > 6.5) using sodium
bicarbonate or potassium citrate. If symptoms are present, a
ureteral stent or nephrostomy tube may be used for temporary
relief during dissolution therapy. Struvite stones are usually
associated with infection, and antibiotics should be continued
throughout treatment. Because struvite stones are generally
large or staghorn in shape, percutaneous nephrostolithotomy
(PCNL) is often the first-line treatment for these stones. Cys-
tine stones are typically difficult to treat and frequently recur.
Small renal cystine stones may be treated by ESWL, whereas
larger stones will likely require PCNL or ureteroscopic
extraction.63
Stone size can be used to predict whether the stone will
pass without intervention. In adults, the likelihood of passing
a stone less than 4 mm is 90%, a stone between 4 and 6 mm
has an approximately 50% rate of spontaneous passage, and
one greater than 6 mm has a 10% to 20% rate. Interestingly,
small stones and stone fragments after ESWL have shown
the same passage rate in the pediatric population, theoretically
because of ureteral pliability and peristalsis.46,64,65 Therefore
supportive care in the form of vigorous hydration (oral or in-
travenous, as needed) and analgesic therapy is a reasonable
first step in a child with a small stone in the absence of fever
or complete ureteral obstruction. Studies demonstrate that
partial obstruction is well tolerated in the short term; thus this
treatment may be continued for 3 to 4 weeks on an outpatient
basis to allow spontaneous passage of the stone. In the case of
an obstructing stone 4 mm or greater in size, the likelihood of
spontaneous passage is significantly lower, and intervention
may be indicated sooner. Nonobstructing stones can be trea-
ted electively. Stones larger than 2 cm have demonstrated a
poor stone-free response to ESWL, and a National Institutes
of Health Consensus Conference in 1998 recommended that
stones larger than 2 cm be approached by PCNL as first-line
treatment. “Sandwich therapy,” which is a combination of PCNL
and ESWL, is an addition to the treatment armamentarium
for large calculi.
Calculi are most commonly located within the renal pelvis
or calices, proximal ureter, distal ureter, or bladder. Each lo-
cation lends itself to different treatment approaches. For renal
calculi, ESWL or PCNL is suitable. ESWL or ureteroscopic
stone extraction is an appropriate choice for ureteral stones,
whereas cystolitholapaxy or cystolithotomy is recommended
for bladder calculi.
ESWL, first described in the early 1980s,66 is noninvasive
and well-tolerated but requires general anesthesia in many
young children and special precautions in infants. Children
who weigh as little as 6.8 kg have been successfully treated.
This procedure may be suitable for proximal ureteral stones
and even some distal ureteral stones; however, treatment of
calculi more distal than the midureter is contraindicated in fe-
males because of the unknown effects of shock waves on the
developing ovary. Stone-free rates after one to two treatments
generally range from 70% to 97% (mean 85%).67–71 Compli-
cations of ESWL therapy include incomplete fragmentation
of the stone, retained fragments, Steinstrasse, perinephric he-
matoma, fever, renal colic, and abrasion or ecchymosis at the
site of shock wave entry and exit.72 Although the distant and
long-term effects of ESWL are unknown, short-term follow-up
studies suggest no demonstrable complications with regard to
renal function or hypertension.42,72–74
PCNL requires nephrostomy tube placement, general anes-
thesia, and inpatient hospitalization. Once the nephrostomy
tube is placed, the tract is dilated to a size appropriate for
the nephroscope and electrohydraulic or ultrasonic lithotrip-
tor.75 It is a good choice for a large renal stone associated with
hydronephrosis or one refractory to ESWL treatment (cystine or
calcium oxalate monohydrate). PCNL and ESWL can be com-
bined for optimal treatment in some cases, so-called sandwich
therapy. Stone-free rates may approach 100%; however, multi-
ple procedures may be required to render a child stone free.68,76
Complications of PCNL include perforation of the collecting
system, bleeding, extravasation of irrigant, pneumothorax,
intestinal injury, and retained fragments.
Ureteroscopic stone extraction in children has become fea-
sible with the development of progressively smaller uretero-
scopes and working instruments. Ureteroscopic treatment of
ureteral calculi may be approached in an antegrade or retro-
grade fashion, and the stone may be extracted intact or frag-
mented using a laser, ultrasound, or hydraulic lithotriptor
(Swiss Lithoclast). Success rates for ureteroscopic stone ex-
traction may exceed 95%.68,69,77–80 An indwelling ureteral
stent may be left in place for 24 to 72 hours to prevent obstruc-
tion secondary to ureteral spasm or edema. Complications
include ureteral perforation, ureteral stricture, reflux, proxi-
mal migration of the stone, and loss of the stone through a
perforated ureter.
Despite the success of minimally invasive treatment for pe-
diatric stone disease, open surgical treatment is still required
in up to 17% of patients,69 which may result in decreased re-
nal function in up to 45%.81 Anatomic abnormalities such as
ureteropelvic junction obstruction or obstructed megaureter
may be addressed concurrently with stone treatment and must
be dealt with eventually to prevent recurrence and optimize
renal function.82,83
Cystolitholapaxy, or transurethral lithopexy of bladder
stones, has been the preferred approach for all but large blad-
der calculi. The stone is fragmented using the electrohydraulic
lithotriptor, and the fragments are irrigated from the bladder.
Bladder calculi are becoming more common in children who
have undergone augmentation cystoplasty, and in this popu-
lation, open removal of the intact calculi may be a more pru-
dent approach to minimize the risk of recurrence due to
retained fragments, although this remains controversial.84–86
RECURRENCE
Children who present with urolithiasis are at risk for recur-
rence for a longer time than adults. Thus the cumulative likeli-
hood of recurrent stone disease is higher in children. Therefore
a thorough metabolic evaluation is strongly encouraged in chil-
dren after their first presentation with urolithiasis. A 24-hour
urinalysis for stone risk should be obtained including, at a min-
imum, urinary volume, pH, and calcium, creatinine, uric acid,
citrate, oxalate, and magnesium levels. The cornerstones for
preventing stone recurrence as the child enters adulthood are
the ability to render the patient stone free, elucidate and treat
metabolic abnormalities, control urinary infection, and correct
anatomic anomalies.
 CHAPTER 114 RENAL INFECTION, ABSCESS, VESICOURETERAL REFLUX, URINARY LITHIASIS, AND RENAL VEIN THROMBOSIS 1439

Renal Vein Thrombosis after a predisposing illness, or it may develop in an older child
------------------------------------------------------------------------------------------------------------------------------------------------
who has preexisting renal disease (nephrotic syndrome). Pre-
Renal vein thrombosis is a variable clinical disorder related to sentation in older children generally includes acute flank pain,
the acuteness and extent of venous occlusion of the main or hematuria, and a palpable mass.
intrarenal veins. Rayer87 first described renal vein thrombosis
more than 150 years ago. For many decades, the condition DIAGNOSIS
was rarely diagnosed clinically but was discovered at operation
or autopsy. Once the condition was clinically recognized, treat- Renal vein thrombosis is frequently associated with thrombo-
ment initially consisted of simple nephrectomy; this procedure cytopenia (platelet count < 75,000 cells/mm3) as a result of
was advocated until the early 1960s.88,89 In 1964 Stark90 platelet entrapment in the renal thrombus. A consumptive
reported excellent recovery after conservative management, coagulopathy characterized by a prolonged prothrombin time
suggesting that aggressive surgical intervention was unneces- and increased fibrin degradation products may also be pre-
sary. The use of anticoagulants or thrombolytic agents still re- sent. Leukocytosis is often present, and the presence of
mains controversial because of the limited number of pediatric anemia varies. Serum electrolyte levels vary, depending
cases and the absence of controlled trials evaluating the risks on predisposing illness and renal function. Urinalysis usually
and benefits of these therapeutic interventions. shows blood and protein. Screening for prothrombotic
abnormalities should be considered.

CAUSE IMAGING
Renal vein thrombosis is a disease of infancy; approximately An abdominal radiograph may show an enlarged renal
80% of patients present during the first month of life.91 It re- shadow or have a lacelike pattern, similar to the renal vascular
sults from diminished renal blood flow, increased blood viscos- tree, but it is usually normal.111 Renal Doppler US is the test of
ity, and hypercoagulability. The neonatal kidney is particularly choice to evaluate the renal mass and image the renal vein and
vulnerable because of its low renal perfusion pressure and inferior vena cava. Diagnostic features include thrombus in
double intrarenal capillary network. Many factors also lead the vein and renal enlargement.112 Renal function can be
to hemoconcentration, with a sluggish venous flow that pre- monitored initially and during recovery with radionuclide
disposes patients to thrombosis. In addition, up to 50% of scintigraphy using technetium 99m DMSA or technetium
these patients may have prothrombotic abnormalities.92,93 99m MAG3. CT, magnetic resonance imaging, or inferior
The thrombotic process may originate in the main renal vein venacavography may be indicated to determine the extent of
or in the small intrarenal veins.94 Thrombosis may be unilateral a thrombus (with inferior vena cava involvement) or to rule
or bilateral, but thrombosis from the inferior vena cava is rare.95 out rare causes of thrombus or thrombosis (e.g., external
Clinical conditions predisposing neonates to thrombosis in- venous compression or thrombus from a renal tumor).
clude preterm delivery92; perinatal asphyxia, polycythemia, and
cyanotic congenital heart disease96; maternal diabetes97,98; sep-
ticemia and congenital nephrotic syndrome99,100; cytomegalovi- TREATMENT
rus infection101; shock, diarrhea, and dehydration102; activated
protein C resistance92; factor V Leiden heterozygosity93; and ma- Because of the relative infrequency of renal vein thrombosis
ternal use of diuretics and corticosteroids.103 Renal vein throm- and the lack of controlled trials, no general consensus on op-
bosis has been noted in association with adrenal hemorrhage timal treatment exists. Although nephrectomy and thrombec-
and has been recognized on prenatal US.104–106 tomy were previously performed during the acute phase, this
In infants older than 1 month, renal vein thrombosis was rarely necessary on an emergent basis and is now limited
is usually associated with hypovolemia such as with gas- to patients with anatomic obstruction of the inferior vena
trointestinal fluid losses or burns99,102,107 or renal disease such cava.90 Management of renal vein thrombosis should include
as nephrotic syndrome.108,109 Renal vein thrombosis may oc- a multidisciplinary team including neonatologist or pediatri-
cur secondary to prolonged inferior vena cava catheterization, cian, radiologist, hematologist, and nephrologist. Initial treat-
with thrombus extending into the renal vasculature.108 ment should include correction of and therapy for all
predisposing conditions, aggressive fluid and electrolyte re-
pletion, and prevention of propagation of the venous throm-
bus. The use of anticoagulants or thrombolytic agents is
CLINICAL PRESENTATION
controversial because of the absence of controlled trials. In
Clinical presentation varies, depending on the age, extent, and a retrospective study of 16 patients with renal vein thrombo-
acuity of the thrombosis. In the neonatal period, males are af- sis, renal function was normal in 6 of 9 patients treated with
fected predominantly (62%) and most cases are unilateral heparin or enoxaparin or both, compared with 0 of 7 patients
(70%) and on the left side (63.6%).110 Renal vein thrombosis who did not receive anticoagulation therapy.93,113 In a review
may present prenatally or in a healthy newborn with no pre- of the English-language literature from 1992-2006, no differ-
disposing illness. In neonates, classic features include a palpa- ence in the rate of significant renal atrophy (70.6% overall)
ble mass and gross hematuria; however, these signs may be was seen between groups treated with supportive care only
found in only 13% of patients at the time of diagnosis.101 (72.5%) and heparin (75.3%).110 A reasonable approach in-
Caval thrombosis may be suspected when the lower extremi- cludes aggressive supportive measures in cases of unilateral
ties are swollen and the superficial abdominal veins are di- renal vein thrombosis without caval extension, especially
lated. Renal vein thrombosis may also present in infancy for those at increased risk of hemorrhagic complications.
1440 PART VIII GENITOURINARY DISORDERS
For unilateral renal vein thrombosis with caval extension or
bilateral renal vein thrombosis, heparin therapy may be con-
sidered because of the risk of pulmonary emboli or renal fail-
ure.91 Thrombolytic therapy may be considered for bilateral
renal vein thrombosis and renal failure. Although heparin
has been used successfully, hemorrhagic complications have
been reported in neonates.114,115 Low-molecular-weight hep-
arin may be a safer alternative, with careful attention to spe-
cial dosing requirements in preterm and full-term
newborns.93,116 Thrombolytic therapy with streptokinase
and urokinase by systemic and regional routes has been
reported.95,99,114 Potential complications with the systemic
use of streptokinase include hemorrhage and allergic reac-
tions; these occur less often with selective regional use.
Route, rate, and length of infusion of thrombolytic agents dif-
fer among various reports; thus standardized protocols are
difficult to establish.
OUTCOME
With improved recognition and management of renal vein
thrombosis and the various predisposing conditions, the over-
all survival rates have improved to about 80% to 85% or
higher.91,110 Most deaths are due to underlying disease rather
than renal infarction. Long-term morbidity including condi-
tions such as hypertension, renal atrophy, and chronic UTI
has not been systematically assessed. In terms of renal function,
the outcome can vary from full recovery to nonfunction. Peri-
odic monitoring with radionuclide scintigraphy permits de-
tailed follow-up evaluation of renal recovery. Renin-mediated
hypertension in association with a small, atrophic kidney or
recurrent UTIs localized to the shrunken renal unit are reason-
able indications for nephrectomy.
The complete reference list is available online at www.
expertconsult.com.
SUGGESTED READING
Urinary Tract Infection, Renal Abscess, and Vesicoureteral Reflux
Clark CJ, Kennedy WA, Shortliffe LD. Urinary tract infection in children:
When to worry. Urol Clin North Am 2010;37:229.
Merguerian PA, Sverisson E, Herz D, McQuiston L. Urinary tract infections
in children: Recommendations for antibiotic prophylaxis and evaluation.
An evidence based approach. Curr Urol Rep 2010;11:98.
National Institute for Health and Clinical Excellence. Urinary tract infection in
children. London: NICE, 2007. Available at http://www.nice.org/uk/
nicemedia/pdf/CG54NICEguideline.pdf.
Peters CA, Skoog SJ, Arant BS, et al. Summary of the AUA Guideline on Man-
agement of Primary Vesicoureteral Reflux in Children. J Urol 2010;184:
1134.
Pohl HG, Belman AB. The “top-down” approach to the evaluation of children
with febrile urinary tract infection. Adv Urol 2009; 783409, Epub 2009
Mar 30.
Urinary Lithiasis
Hoppe B, Kemper MJ. Diagnostic examination of the child with urolithiasis or
nephrocalcinosis. Pediatr Nephrol 2010;25:403.
Straub M, Gschwend J, Zorn C. Pediatric urolithiasis: The current surgical
management. Pediatr Nephrol 2010;25:1239.
Renal Vein Thrombosis
Lau KK, Stoffman JM, Williams S, et al. Neonatal renal vein thrombosis: Re-
view of the English-Language literature between 1992 and 2006. Pediatrics
2007;120:e1278.

CHAPTER 115
Ureteral
Duplication and
Ureteroceles
Ramnath Subramaniam
A duplex (duplicated) system is a kidney with two pelvicali-
ceal systems. Complete duplication refers to a kidney with
two ureters that drain separately into or below the bladder.
Incomplete duplication refers to a kidney with two ureters that
fuse into a unit proximal to the bladder and then drains into
the bladder through a single orifice. A bifid system is a form of
incomplete duplication with two renal pelves that join into a
single ureter.
The upper or lower pole ureter refers to the ureter drain-
ing the upper or lower pole (moiety) of the kidney, respec-
tively. The upper or lower pole orifice refers to the ureteral
orifice associated with the upper or lower pole ureter, respec-
tively. An ectopic ureter refers to a ureter that drains into an
abnormal site.
The greater majority of duplicated systems are incomplete,
and these forms rarely give rise to clinical problems. In con-
trast, complete duplication anomalies are rare, affecting less
than 0.1% of individuals (most females), and are more com-
monly of clinical significance.1 Patients with urinary symp-
toms such as those associated with urinary tract infection
(UTI), have about a 4% incidence of ureteral duplication.2
Two thirds of children with duplex systems who present with
UTI have vesicoureteric reflux (VUR).3 This is particularly true
of complete duplex systems, which are associated with higher
grades of reflux and renal dysplasia with poor renal function
in the affected moieties.4 A significant familial predisposition
exists—between 12.5% and 30% of siblings of affected pa-
tients have duplications.5–7 It has been suggested that the
mode of inheritance is autosomal dominant with incomplete
penetrance.5,8
Less common types of ureteral duplications are (1) the
“inverted Y” ureter, consisting of a single pelvis and a proximal
ureter that bifurcates distally into two ureters ending with two
separate orifices, one of them being frequently ectopic; and
(2) the “blind-ending” duplication of the ureter, which occurs
when one limb of a bifid ureter does not drain a portion of
renal parenchyma.9,10
A ureterocele is a cystic dilatation of the lower end of the
ureter where it joins the epithelium of the lower urinary tract.
An intravesical ureterocele lies entirely within the bladder,
whereas an ectopic ureterocele has a portion that lies below
the bladder neck.
Embryology
------------------------------------------------------------------------------------------------------------------------------------------------
Ureteral duplication is the consequence of the abnormal
development of the ureteral bud from the mesonephric
(wolffian) duct. Premature branching of the bud causes in-
completely duplicated ureters. Complete ureteral duplication
occurs when two separate ureteral buds arise independently
from the mesonephric bud and induce the development of a
duplex kidney.
The cloaca is divided into the urogenital sinus anteriorly
and the alimentary tract behind. Around the eighth week of
gestation, the distal end of the mesonephric duct plus a short
segment of this duct above the ureteric bud expands and, by
differential growth, is incorporated into the lower end of the
posterior wall of the developing bladder and urogenital sinus.
At this stage the mesonephric duct lies below and medial to the
ureteral orifice and is referred to as the wolffian duct. As a re-
sult of this “migration,” the normal ureteral orifice is situated at
the superolateral angle of the bladder trigone (Fig. 115-1, A).11
In the male the mesonephric duct becomes the epididymis,
vas deferens, and seminal vesicle and enters the posterior ure-
thra above the external urethral sphincter at the verumonta-
num. The associated ureteric bud must, therefore, also join
the developing bladder and urethra above the external sphinc-
ter. The testis that initially arises from a region medial to the
developing kidney becomes attached to the mesonephric
duct. As it descends to the deep inguinal ring, the vas deferens
(mesonephric duct) crosses the origin of the ureter superiorly
as it arches medially and inferiorly to the seminal vesicle at the
level of the verumontanum.11
In the female the mesonephric duct becomes the Gartner
duct, which lies close to the lateral wall of the vagina. The
paramesonephric ducts (müllerian ducts) develop medial to
the mesonephric duct. At the caudad end of the embryo, these
incorporate the distal end of the remnants of the mesonephric
duct where they penetrate the cloaca near the midline. Thus
owing to differential growth, the ureteric bud may migrate
1441
1442 PART VIII GENITOURINARY DISORDERS

a b

c d

Key
Upper pole ureter
Lower pole ureter
Mesonephric duct

a b

c d

Key
Upper pole ureter
Lower pole ureter
Mesonephric duct

FIGURE 115-1 A, The normal relationship between the mesonephric duct and the ureteric buds and the subsequent positions of their orifices within the
bladder (Weigert Meyer law). Also note the area where the buds come in contact with the renal blastema. B, Compare with A and you will notice that
the upper ureteric bud reaches the renal blastema late and into the periphery, thus inducing dysplasia of the upper renal moiety (shaded); also note
the position of the upper ureteric orifice in this scenario is caudal in comparison with the situation in A.

a b
c d
FIGURE 115-1—CONT’D. C, Compare with A and you will notice the lower ureteric bud reaches the renal blastema in the periphery and thus induces
abnormal dysplastic development of the lower renal moiety; also note that the position of the lower ureteric orifice in this case is superolateral in com-
parison with the situation in A. Here the submucosal tunnel of the lower pole ureter is relatively short and therefore more prone to reflux.
with the associated paramesonephric duct structures to sites
such as the vestibule, vagina, cervix, uterus, and rectum.11
The induction of the metanephric blastema by the ureteral
bud in a central position is critical to normal renal devel-
opment. If the ureteral bud meets the metanephric blastema
either too caudal or too cranial along the metanephric blas-
tema, renal dysplasia will result.12 The more superior of the
two ureteric buds may make contact with the blastema of
the upper pole of the kidney too late to induce normal devel-
opment. In clinical practice, renal dysplasia affecting the up-
per pole of the kidney is often encountered in duplex systems
(see Fig. 115-1, A and B).
During development of the trigone, the most cranial ureter
that drains the upper moiety of the duplex kidney rotates in-
ward on its long axis and crosses the lower pole ureter. There-
fore the upper pole ureter opens in the urogenital sinus in a
more distal and medial position than the lower pole ureter that
opens more proximally. This relationship between the two
ureteral orifices is constant and is called the Weigert-Meyer
rule (Fig. 115-1, A).
When the two ureters originate close to each other and in a
near normal position, both ureteral orifices open in the trigone
(see Fig. 115-1, A). Conversely, when the two ureteral buds
originate at widely separate positions on the mesonephric
duct, the upper pole ureter, which is located at a more cranial
position on the mesonephric duct, is incorporated into the
urogenital sinus at a later stage of development. As a result,
the ureteral orifice is situated in an ectopic position, inferior
to the trigone (see Fig. 115-1, B). In the male in this situation,
the location of the ureteral orifices can be anywhere along the
lower mesonephric duct or its derivatives below the bladder
neck in the posterior urethra but always above the external
urethral sphincter. Therefore urinary incontinence due to
CHAPTER 115 URETERAL DUPLICATION AND URETEROCELES 1443
Key
Upper pole ureter
Lower pole ureter
Mesonephric duct
duplex systems with ectopia of the upper pole ureter does
not occur in boys. In girls, however, the situation is different.
During development, the mesonephric duct derivatives can
open anywhere caudally from the developing bladder,
commonly into the vault of the vagina.
The lower pole ureter, which joins the mesonephric duct
closer to the urogenital sinus, is incorporated into the devel-
oping trigone at an early stage and consequently migrates lat-
erally and cranially, leaving a relatively short intravesical
tunnel (Fig. 115-1, C). VUR into the lower pole ureter may
be present because of the short length of the intramural
tunnel.
Occasionally, in either complete or incomplete duplication,
one of the ureters fails to meet the developing blastema of the
metanephros. In this case, a blind-ending ureter occurs, with
no attachment to the kidney.10 Rarely, an inverted Y abnormal-
ity is encountered. It is presumed to be caused by double ure-
teric buds arising from the mesonephric duct caudally and
fusing cranially before merging above with the blastema of
the metanephros.9
Recently, there has been a shift from classic theories to the
cell biology view of congenital anomalies of the kidney and
urinary tract (CAKUT).13 The process of ureteral budding
and metanephric differentiation into the final kidney is under
the control of the renin-angiotensin system through the Agtr-2
receptor, which is intensely expressed in the mesenchymal
cells surrounding the wolffian duct at the time of the initial
budding of the ureter. Most recently it has been shown that
ectopic budding and ureteral duplication may occur in Agtr2
null mice.14 The formation of a double collecting system has
also been observed in another group of experimental animals
that are heterozygous mutants for the bone morphogenetic
protein 4 (BMP4). BMP4 is a member of the transforming
1444 PART VIII GENITOURINARY DISORDERS

growth factor-beta (TGF-b) superfamily and is credited to be

part of the navigating system that allows the budding of the
ureter from the mesonephric duct at a precise site.13 Abnor-
malities in the number and site of the initial ureteral buds pre-
ceding a duplex collecting system and ectopic ureteral orifice
with concurrent renal abnormalities are constantly observed
in mice that are homozygous mutants for the FOXC1 gene,
which encodes for a transcription factor playing an essential
role in embryonic development.13

Incomplete Duplicated Systems

------------------------------------------------------------------------------------------------------------------------------------------------

Uncomplicated duplex kidneys may be detected incidentally

on ultrasound scans done for unrelated reasons. Bifid pelvis
is never responsible for any symptoms. Incompletely dupli-
cated ureters are also often completely asymptomatic but in
rare cases are associated with distinct clinical problems.
Three clinical problems may affect an incompletely dupli-
cated ureter: (1) VUR, (2) ureteropelvic junction obstruction,
and (3) retrograde ureteral peristalsis (yo-yo phenomenon).
Treatment depends on the function of the affected renal
moiety and on the length of the lower limb. If the function
is poor, a partial lower pole nephroureterectomy is the appro-
priate choice, removing the lower moiety and its ureter down
to the junction with the upper pole ureter. If the lower moiety
is worth saving, the type of repair depends on the length of the FIGURE 115-2 Drooping lily appearance of a refluxing right lower pole
ureter. If the ureter is very short, the upper and lower pelvis ureter. Note that the contralateral ureter is also refluxing with all its calyces
intact, but the calyces are missing on the right, giving it a characteristic
may be joined side by side (pyelopyelostomy) to create a single lower pole appearance.
pelvis drained by the upper ureter. Conversely, if the lower
ureter is long, a conventional dismembered pyeloplasty can
be done, but, if the junction between the two limbs is close Cystourethrogram often reveals the pelvis in a lower than
to the bladder, a ureteroureterostomy with excision of the dis- normal position with an abnormal renal axis. The upper caly-
tal ureter is a better solution in order to avoid the possibility of ces are missing, and the lower ones often show signs of
uretero-ureteral reflux. chronic reflux nephropathy, resulting in the appearance of a
Even when uretero-ureteral reflux (yo-yo reflux) is demon- drooping lily (Fig. 115-2).
strated, it is difficult to relate the frequently vague complaints The affected lower pole on dimercaptosuccinic acid
of the patient to the peristaltic disorder. Clinical judgment is (DMSA) renal scan may be nonfunctioning or poorly function-
required before proceeding to a surgical repair, which may ing, indicating severe dysplastic changes or parenchymal scar-
consist of a proximal pyelo-ureteral anastomosis with removal ring due to repeated episodes of UTI. In this relatively rare
of one ureter or, if the ureteral junction is close to bladder wall, case, the treatment of choice is lower pole partial nephrectomy
excision of the common limb and reimplantation of the two and ureterectomy (Fig. 115-3). It is recommended that the
ureters into the bladder side by side. lower pole ureter be removed completely down to the bladder
wall in order to avoid leaving a refluxing stump. In most cases,
however, the lower pole is worth saving and several options
are available for treatment.
COMPLETE DUPLICATION ANOMALIES Contrary to popular belief, a significant proportion of
refluxing units associated with duplex systems resolve spon-
Vesicoureteric Reflux
taneously with time. This is more likely with grade 1 and 2
The most frequent anomaly detected in completely duplicated VUR (up to 85%) than with higher grades.16–19
ureters is VUR. As a rule, it always affects the lower pole ureter Provided the child remains asymptomatic (as with primary
either independently or with VUR into the upper pole ureter VUR), antibiotic prophylaxis is justified initially. These pa-
because it opens in the trigone in a more cranial and lateral tients must be followed carefully to assess the kidneys for scar-
position with a shorter submucosal tunnel, causing incompe- ring and its progression. VUR can be assessed using either
tence of the ureterovesical junction. When VUR rarely occurs indirect MAG3 cystography or a micturating cystourethro-
into both moieties, the ureteral orifices are usually close gram (MCUG). If progressive renal scarring with deteriorating
together and in a lateral ectopic location. renal function is identified, or if symptomatic upper renal
The most common presentation is UTI, with more than UTIs cannot be controlled by antibiotic prophylaxis, operative
60% of children with UTIs and a duplex kidney having vesi- intervention is indicated.20,21
coureteral reflux.15 Therefore when ultrasound reveals a du- Endoscopic treatment of VUR in a duplicated system is not
plex kidney in a child with a history of a UTI, a micturating as successful as with a single-unit VUR, with a permanent res-
cystourethrogram is mandatory. olution rate of only 46%.22 However, it is still worthwhile to
 CHAPTER 115
A B
FIGURE 115-3 A case of poorly functioning lower pole (A) on dimercaptosuccinic acid scan and refluxing on micturating cystourethrogram (B). The
specimen after lower pole heminephroureterectomy (C); note that the refluxing ureter has been removed as low as possible to avoid a refluxing stump.
consider it as an effective option in selected cases.17,23 In du-
plex systems in which the ureteral orifices are close together,
the implant is injected under the upper pole moiety orifice
(inferomedial). The needle is advanced below both orifices,
and the injection is commenced under direct vision until both
moiety orifices are compressed. When the ureteral orifices are
separated by more than a few millimeters, they can be individ-
ually injected; however, caution is required when placing the
injection under the lower pole orifice because the upper pole
ureteral lumen may be entered in error.
The procedure of choice, when the refluxing ureter is not
excessively dilated, is “double barrelled” or common sheath
ureteroneocystostomy. When the ureteral orifices are situated
relatively normally on the trigone, both should be mobilized
from the bladder base together, even if only the lower pole ureter
is refluxing. This is necessary because the ureters share the same
adventitial sheath and blood supply.24 Attempts to mobilize only
the affected ureter will result in ischemia of one or both ureters.
Alternatively, both ureters can be mobilized to a level above the
common sheath. The ureters can then be rerouted through the
bladder wall using either an intravesical cross-trigonal technique
or an extravesical approach. The objective in both instances is to
create a sufficiently long submucosal tunnel for the ureters
(preferably three times their diameter), in order to create a
satisfactory “flap valve” antireflux mechanism.
When one or both ureters are massively dilated, they can be
safely reduced in size, plicating or excising the side away from
the common wall.25 When both ureters need to be reduced in
size, excisional tapering is a better choice in order to avoid an
excessive bulk of tissue that may make subsequent uretero-
neocystostomy difficult.
Other surgical options, when only one ureter needs to be
addressed, include uretero-ureterostomy. Recent literature in-
dicates it is effective and has a low incidence of complica-
tions.26–28 The size of the ureter does not seem to be a
problem with this technique, which has also been performed
via an inguinal incision with good results.29 It is important
URETERAL DUPLICATION AND URETEROCELES 1445
C
to make sure there is no reflux in the ureter draining into the
bladder. There is a theoretical risk of inducing yo-yo reflux.
Such a risk can be avoided with pyelo-ureterostomy, which
is a highly successful procedure, but a second lower abdominal
incision is required to completely remove the refluxing stump
or the stump can be dealt with by endoscopic correction.
Ectopia of the Upper Pole Ureter
Another common anomaly in complete duplications is ectopia
of the upper pole ureter.
Approximately 80% of all ectopic ureters occur in dupli-
cated systems. The upper pole drained by an ectopic ureter
commonly has severe dysplasia. In addition, findings consis-
tent with obstructive changes are the rule in the male, where
ectopic ureters are always obstructed. In females, obstruction
affects only ureters ending inside the urinary tract.
Ureters ending on the bladder neck or in the posterior urethra
are obstructed when the bladder neck is closed but may occa-
sionally reflux during micturition; in such cases UTI is common.
In the male, ectopic ureters generally cause infections and are
identified during the evaluation for UTI. Epididymo-orchitis
is a frequent presentation when the ureter opens in the genital
tract in association with the vas or seminal vesicle.30–32
Many infant girls with ureters opening inside the urinary
tract and, as a consequence, obstructed, present with UTI.
Conversely, an ectopic ureter opening in the müllerian deriv-
atives, having its orifices outside the realm of the urinary
sphincter, causes urinary pseudo incontinence. A constant
dribbling of urine in the interval between normal micturition
is frequently observed.33 This symptom is generally over-
looked until the girl is fully toilet trained and has normal void-
ing habits. If the ectopic upper pole function is poor and little
urine is produced, dribbling may not be observed and the pa-
tient may present with constant vaginal discharge. When the
urine pools in a dilated ureter, especially at night when the pa-
tient is in a recumbent position, urinary losses may assume
different features and mimic stress or urge incontinence.
1446 PART VIII GENITOURINARY DISORDERS

Physical examination is often completely normal. Rarely,
close and careful scrutiny of the external female genitalia re-
veals that the ectopic ureter is located in the urethro-vaginal
septum. In addition, a dilated ureter is occasionally palpated
during rectal examination.
Diagnosis of a suspected ureteral ectopia is straightforward
when the ectopic ureter is massively dilated and may be easily
identified on an abdominal ultrasound or when the upper
pole functions sufficiently to excrete contrast material on
intravenous urography (IVU).
Difficulties arise when the upper pole function is so poor
that contrast is not excreted and the ureter is not dilated or
when the ectopic ureter is associated with other malforma-
tions like massive reflux in the ipsilateral lower pole ureter
or contralateral pathology that confuses the picture.
Ultrasound, the first-line investigation, may reveal a dilated
duplex ureter and upper pole or, when the ureter is not
dilated, show changes in the echogenicity of an otherwise
normal upper pole that help in defining the correct diagnosis.
MCUG should be included in the diagnostic workup.
Associated reflux either in the ipsilateral lower pole ureter
or in the ectopic upper pole ureter may be demonstrated; in
the latter case MCUG delineates the anatomy of the ureter,
making more sophisticated imaging studies unnecessary. An
ectopic ureter should always be suspected in cases of isolated
reflux into the upper pole of a duplicated system, which most
commonly occurs on the voiding phase of the MCUG. Excre-
tory urography is useful for defining the anatomy in a ureter
associated with a functioning renal unit.34 However, when the
affected moiety is nonfunctioning, the findings are more sub-
tle and changes on ultrasonography, DMSA scans, and excre-
tory urography are easily missed.19,34 Magnetic resonance
imaging (MRI) scanning is particularly useful in these circum-
stances and provides both anatomic detail and functional
analysis (Fig. 115-4)35–38; however, anesthesia is required
in children.
The choice between the different treatment options largely
depends on the function of the affected segment, on the coex-
istence of associated malformations, and, to a lesser degree, on
the size of the affected ureter.

A B

C D
FIGURE 115-4 A case of ectopic ureter causing incontinence in a young girl. The ultrasound shows a dysplastic upper pole (A) with a dilated ureter
behind the bladder (B). Excretory urogram is unremarkable (C), but magnetic resonance imaging (MRI) shows the ureter opening well below the bladder (D).
This girl had bilateral ectopic upper pole ureters as diagnosed on the MRI.

In the great majority of cases when the involved upper pole
contributes to less than 5% total function, open surgical, lap-
aroscopic, or, even better, retroperitoneoscopic removal of the
upper pole along with its ureteral segment is the treatment of
choice.
Partial nephrectomy involves identifying the demarcation
between the normal lower pole and the nonfunctioning upper
pole and excising the ureter of the affected renal unit inferiorly
as near the bladder as possible without compromising the
lower pole ureter. The potential problems of excising the
lower end of the affected ureter are well recognized and are
discussed in more detail later.
The operation may be undertaken by either open or endo-
scopic techniques. The important steps are to clearly identify
the ureter and pelvis of the affected upper pole moiety. After
dividing the blood supply to the upper pole of the kidney, it is
then possible to identify the limit of the affected upper pole,
and a relatively bloodless plane can be developed between the
abnormal and normal renal tissue without entering the caliceal
system of the lower pole.39 The ureter can then be excised
down to the pelvic brim.
This upper tract approach lends itself well to minimally in-
vasive surgery because access to the kidney and ease of defining
the vascular anatomy, particularly with the retroperitoneo-
scopic approach, are excellent.40–43 Secondary surgery at the
bladder level to remove the stump or to reimplant the refluxing
lower pole ureter is necessary only in a small number of
cases.44,45
If the upper pole retains some function and is considered
worth saving, then proximal ureteropyelostomy or distal
uretero-ureterostomy is advisable. In a series of 11 infants
who underwent ureteropyelostomy for obstructed duplicated
ureters with satisfactory results, only 3 children needed sec-
ondary surgery at the bladder level for persisting reflux in
the lower pole ureter.46 When ipsilateral lower pole reflux is
present, a “double-barrelled” ureteroneocystostomy may allow
resolving both obstruction and reflux with a single operation.
It requires a meticulous and tedious extravesical dissection in
order to isolate enough length of both ureters and perform
a safe reimplantation. Moreover, both ureters are usually mark-
edly dilated and need tapering before ureteroneocystostomy.
A possible, easier alternative is to perform a distal end-to-side
anastomosis of the obstructed ureter to the refluxing one and
to reimplant the lower pole ureter into the bladder.
However, in the great majority of cases an upper tract ap-
proach is strongly recommended, either by an upper pole par-
tial nephrectomy or a ureteropyelostomy. If VUR is present
preoperatively or develops postoperatively, it can be safely
managed conservatively with antibiotic prophylaxis or treated
endoscopically at a later stage. With such an approach the
number of surgical procedures and complications can be sub-
stantially reduced, allowing, when possible, the preservation
of a functioning upper pole.
Ureteropelvic Obstruction
In duplex systems, the obstruction usually affects the lower
pole and is more common in boys.47,48 The decision to oper-
ate is guided by the size and function of the renal unit in-
volved. An increase in hydronephrosis or decreased
function of the affected part of the kidney on isotope study
is an indication to intervene. However, these parameters are
sometimes difficult to assess objectively in a growing child.
CHAPTER 115 URETERAL DUPLICATION AND URETEROCELES 1447
Loin pain in older children, typically preadolescents, associ-
ated with objective evidence of ureteropelvic junction ob-
struction on MAG3 isotope scan, is a common indication
for operation. The surgical options are pyeloplasty or pyelour-
eterostomy to the unaffected upper pole ureter, with excision
of the distal ureter of the affected lower pole.47–49 Partial
nephroureterectomy is undertaken when the function of the
affected renal unit is severely impaired.
Ureteroceles
------------------------------------------------------------------------------------------------------------------------------------------------
The term ureterocele describes the cystic dilatation of distal
intravesical portion of the ureter. A single-system ureterocele
is associated with a kidney with only one ureter, whereas a du-
plex system ureterocele is associated with the upper pole of a
kidney with a complete ureteral duplication.
Most frequently the pelvicaliceal system drained by the
ureterocele is also obstructed. The malformation may be fur-
ther complicated by VUR into the ipsilateral lower pole or into
the contralateral ureter or by ureteral obstruction.
The incidence of ureteroceles is approximately 0.02% of in-
dividuals with 80% occurring in females.50,51 A slight major-
ity are left sided and bilateral in 10%.52 Some 80% of
ureteroceles are associated with the upper pole of a complete
duplication,53 and in most series 60% to 80% are ectopic.54
Bilateral ureteroceles or those associated with contralateral
duplicity occur in 15% of cases.55 Single-system ureteroceles
are uncommon, occur most often in males, and can be asso-
ciated with other anomalies including abnormalities of the
kidneys such as fusion, ectopia,56 or multicystic dysplasia.57
CLASSIFICATION
The Committee on Terminology of the Urologic section of the
American Academy of Pediatrics, proposed a simple and widely
adopted classification, whereby ureteroceles contained entirely
within the bladder are named orthotopic or intravesical and a ure-
terocele that has a portion permanently located outside the
bladder is called ectopic.58 In 1971 Douglas Stephens classified
the ureteroceles associated with ureteral duplication according
to the position and/or presence of intrinsic obstruction of the
ureteral orifice as shown in Table 115-1.59
Stenotic ureteroceles are completely located inside the
bladder and have a small, stenotic, often pinpoint orifice; they
are usually tense and show a well muscularized wall with pre-
dominantly longitudinal muscle fibers. The nonobstructed va-
riety is rather rare; in such cases the ureterocele is visible only
when a peristaltic wave fills it.
TABLE 115-1
Douglas Stephens Classification Ureteroceles
Intravesical
Extravesical
Stenotic 40%
Sphincteric 40%
Nonobstructed 5%
Sphinctero-stenotic 5%
Cecoureterocele 5%
Blind ureterocele 5%
1448 PART VIII GENITOURINARY DISORDERS
In the sphincteric subtypes the ureteral orifice is wide
and often gaping, but it is located inside the bladder neck
or urethra and is obstructed by the contraction of sphincteric
muscle. These ureteroceles decompress during voiding.
Sphinctero-stenotic ureterocele is similar to the previous
one, but the orifice is both ectopic and stenotic. Because a
sphinctero-stenotic ureterocele does not decompress during
micturition, it may ball valve and obstruct the bladder outlet.
The meatus of the cecoureterocele is in the bladder, but a
tongue of the ureterocele extends down in the urethra. Ureter-
oceles are described as blind when no kidney or upper pole
associated with ureterocele can be demonstrated.
In duplex kidneys the upper pole drained by the uretero-
cele is affected by some degree of dysplasia in 43% to 73% of
cases. In 20% the dysplasia is severe, affecting more than 25%
of the upper pole.60,61 The contribution of the affected upper
pole to the global renal function ranges from 4% to 8%. Mea-
surable loss after upper pole partial nephrectomy is, on aver-
age, around 1% while functional gain after endoscopic
incision is around 2%.62
CLINICAL PRESENTATION
Prenatal diagnosis of ureterocele represents approximately
15% of all antenatal diagnosis of duplex kidneys as the num-
ber of neonates with prenatally detected ureteroceles has in-
creased from 2% to 28% in the past 20 years.63 Prenatal
ultrasound screening of the kidneys often identifies some de-
gree of hydronephrosis, which is most pronounced with ureter-
oceles.64 Although there is no evidence that treatment of these
prenatally diagnosed children affects the prognosis for the
kidney, symptomatic UTIs can be reduced in frequency.65–67
In the postnatal period, infection is the most common pre-
sentation for ureteroceles in both sexes; in infants symptoms
may vary from a life-threatening gram-negative sepsis to a fe-
brile illness with gastrointestinal symptoms or failure to thrive.
In older boys and girls symptoms are more specific and point to
a urinary tract infection: fever, malaise, dysuria, foul-smelling
urine, and back pain.
Other presenting symptoms in both girls and boys may be
hematuria after minimal trauma or a palpable abdominal mass
representing the bladder or the obstructed urinary tract.
In girls an interlabial mass or acute urethral obstruction
due to the ball valve effect of a ureterocele prolapsing in the
urethra is another possible presentation. Rarely, the most
A B
FIGURE 115-5 Ultrasound images of a dysplastic dilated upper pole (A) with corresponding dilated ureter and ureterocele in the bladder (B).
distal portion of a cecoureterocele can protrude from the ure-
thra as a pink, interlabial mass without obvious urethral ob-
struction or retention and, consequently, such cases do not
need emergency treatment.
Prolapsing ureteroceles are the most common cause of
acute urethral obstruction in girls,68 but this event has been
also reported in males,69 even if much less frequently. Patients
present in obvious pain with an acute urinary retention and a
palpable bladder. A purple red or frankly necrotic round mass
may be seen protruding from the labia. In such cases emer-
gency treatment to decompress the ureterocele and remove
the bladder outlet obstruction is warranted. In extreme cases,
when the ureterocele prolapses and obstructs the urethra, it
may affect the function of all the renal units.69,70
DIAGNOSIS
The diagnostic workup relies on the use of ultrasound,
MCUG, and renal isotope scans.
Excretory urography, the mainstay of ureterocele diagnosis
until a decade ago, is much less commonly used now. Com-
puted tomography scan and magnetic resonance urography
may be useful, especially the latter, when the urinary tract
anatomy is not clear.
Ultrasound of the urinary tract is the first investigation and
generally depicts clearly the ureterocele as a sonolucent round
image that sits on the bladder base and occupies a portion of
the bladder. One or more dilated ureters can be seen behind
the bladder (Fig. 115-5). Ultrasound gives valuable informa-
tion on the presence of unilateral or bilateral renal duplicity
and on the dilatation of the collecting systems, and it helps
in identifying which kidney is drained by the ureterocele. Cor-
tical cysts and severe parenchymal thinning, but not increased
echogenicity, suggest renal dysplasia that is confirmed by
isotope scan.71
An MCUG is an essential part of a ureterocele evaluation.
The ureterocele is seen in the first films as a negative shadow
in the bladder, with a rim of contrast around it. If the uretero-
cele is not tense, it may be obscured with the progressive filling
of the bladder because it may be compressed or, if it is small,
may be obscured by the contrast medium around it. MCUG
may also reveal VUR in the lower pole ureter (50%) or in the
contralateral ureter (25%).53,72,73 In less than 10% of cases, re-
flux can be identified in the ureterocele if it is ruptured or if it
has a large open meatus placed on the bladder neck that allows
 CHAPTER 115 URETERAL DUPLICATION AND URETEROCELES 1449

FIGURE 115-6 Voiding cystourethrogram demonstrating retrograde

prolapse of a ureterocele into the upper pole ureter and vesicoureteral re-
flux into the lower pole ureter on the left side in a complete duplex system.
This appearance may be confused with a bladder diverticulum. FIGURE 115-7 Excretory urogram appearances of a ureterocele in the
bladder as a negative shadow. Please note bilateral duplex with nonfunc-
tioning right upper pole corresponding to the ureterocele. The lower pole
reflux during voiding.33,73 The MCUG is also useful to ascer- of the right and both the poles on the left are functioning. The lower pole
on the right is dilated and possibly refluxing (back flow of contrast during
tain the degree of the detrusor backing for the ureterocele. If the study).
the detrusor support is poor, the ureterocele may evert during
micturition, mimicking a bladder diverticulum (Fig. 115-6).
During voiding, the ureterocele may also be seen prolapsing The function of the pole or of the kidney associated with
through the urethra and obstructing the urinary flow. the ureterocele is best assessed by a DMSA renal scan that
IVU was the most important diagnostic step in the past. shows the quality and quantifies the amount of functioning
More recently, however, the progress of ultrasonography, renal tissue (Fig. 115-8).
DMSA, and MRI has made this examination obsolete in most
cases. Nevertheless, when the anatomy is confusing, excretory
urography may still play a role.
If the renal parenchyma associated with the ureterocele re-
tains some function, which most commonly occurs in single
systems, a characteristic “cobra head” or “spring onion” defor-
mity of the intravesical ureter is produced due to opacified
urine in the ureterocele being surrounded by a radiolucent
halo that represents the wall of the ureter. More commonly,
when the ureterocele is associated with the upper pole of a du-
plex kidney, function is absent or minimal in 90% of cases;
therefore the radiographic signs are mainly negative, reflecting
the displacement of the lower pole renal unit by the hydrone-
phrotic upper pole ureter. The lower pole pelvis is often later-
ally and downward displaced, producing the characteristic
“drooping lily” appearance; the number of calices is reduced;
and the upper calices are missing (Fig. 115-7).
At the bladder level a negative shadow may be seen, sug-
gesting the presence of a ureterocele (see Fig. 115-7). The
shadow may vary from a large, tense, round shadow occupy-
ing most of the bladder to a minor irregularity in the base of
the bladder.
Excretory urogram also shows the condition of the contra-
lateral kidney collecting system that may be duplex (see FIGURE 115-8 Dimercaptosuccinic acid isotope scan demonstrating
Fig. 115-7) or may be obstructed at the bladder level by a large poorly functioning upper pole moiety in a left-sided complete duplex sys-
and tense ureterocele compressing the ureteral orifice. tem. The upper pole ureter was associated with a ureterocele.
1450 PART VIII GENITOURINARY DISORDERS
PRINCIPLES OF TREATMENT
The goals of the ureterocele treatment are to preserve renal
function by providing unobstructed drainage of all function-
ing renal tissue, removing any potential source of infection
and treating VUR; prevent bladder outlet obstruction and/or
urinary incontinence; and prevent and treat any bladder wall
deficiency (e.g., diverticula, poor detrusor backing).
The surgical strategy should be optimized in order to attain
these goals with minimal surgical morbidity. Few cases of prenatal
treatment of a prolapsing obstructing ureterocele with anhydram-
nios have been recently reported with a view to restore amniotic
volume.74–76 In a small number of prenatally detected cases of
ureteroceles with no obstruction or reflux, a nonoperative watch-
ful waiting approach has been attempted successfully.65,77,78
Postnatally, the treatment options are endoscopic incision,
upper pole partial nephrectomy (upper tract approach), com-
plete reconstruction at the bladder level, or nonoperative
(conservative) treatment. These methods cannot be compared
easily, and the choice must be appropriately applied to pa-
tients with different clinical presentations. Skill in all methods
of management is important, and an understanding of when to
apply these methods is critical because the management of
each patient with a ureterocele must be individualized.79
The factors that influence treatment choice are type of pre-
sentation (prenatally diagnosed or symptomatic), age of the pa-
tient, type of ureterocele (intravesical versus ectopic), differential
function of the renal moieties, and presence of VUR and UTI.
ENDOSCOPIC DECOMPRESSION
Wide, open deroofing of the ureterocele was performed in the
past. The problem with this technique is that if the ureterocele
arises from the urethra, epithelial remnants that have not been
completely excised may result in urethral obstruction owing to
the valvelike nature of these folds.80 In addition, with this type
of operation, postoperative VUR is almost inevitable. There-
fore deroofing usually requires a subsequent antireflux proce-
dure that is normally best undertaken as part of the initial
surgery. This more aggressive approach is potentially difficult
and is associated with increased morbidity, especially in the
very young child with a small bladder.81
Endoscopic techniques have been used for some time to sim-
ply incise the ureterocele. At any age endoscopic incision has
the advantage of being a simple, quick, and minimally invasive
procedure that can be performed with a minimal hospital stay.
This can take the form of either diathermy incision or sim-
ple puncture of the cyst.82–84 During this procedure it is im-
portant not to overfill the bladder; otherwise, the ureterocele
may be compressed, making it difficult to identify. If this oc-
curs, loin compression on the affected side may assist in refill-
ing the ureterocele. The planned decompression point on the
ureterocele is important and, if possible, should not overlie
the ureteral lumen where it is attached to the bladder base. Re-
cently a novel technique of fulgurating the internal layer of the
ureterocele has been described to achieve and maintain col-
lapse of the ureterocele.85 This technique is challenging and
tedious, especially when the postoperative outcomes follow-
ing endoscopic incision depend on anatomic and functional
characteristics rather than the technique used.86
Simple decompression obviates obstruction and allows a
better evaluation of the function of the kidney or pole involved
by the ureterocele. It often reduces the size of the ureter so that
secondary procedures (such as excision of the ureterocele and
ureteroneocystostomy), if required, are less difficult and may
reduce the need for ureteral tapering. In addition, associated
VUR into the lower pole ureter may subside spontaneously.72
Endoscopic incision of intravesical ureterocele is the de-
finitive treatment in 77% to 93% of cases; therefore it can
be considered to be the initial treatment for this type of ure-
terocele.79 A recent meta-analysis looking at endoscopic
management of ureteroceles confirmed what others have
suggested that the risk of reoperation after endoscopic inci-
sion of ectopic ureteroceles was significantly higher than
with intravesical ureteroceles.73,87
There is no consensus about the effectiveness of endoscopic
incision in ectopic ureteroceles because it rarely represents the
definitive treatment modality88,89; seldom ameliorates the func-
tion of the involved upper renal pole; and, if an iatrogenic reflux
is created in the ureterocele, it may unnecessarily commit the
patient to future lower tract surgery. Alternative treatment op-
tions are represented by the upper tract approach, which in-
cludes either an open or retroperitoneoscopic upper pole
partial nephrectomy and ureterectomy, leaving the upper ure-
teral stump open in order to decompress the ureterocele, or,
if the upper pole is considered worth being kept, a uretero-
pyelostomy or uretero-ureterostomy, joining the upper pole
ureter to the lower one.
The preservation of the upper pole affected by an ectopic
ureterocele is rarely warranted because its function is often
minimal, the severity of renal dysplasia being directly propor-
tional to the degree of ectopia of the ureterocele.62 There is lit-
tle hope of recovery of function after relief of obstruction in
most cases because the upper pole is often dysplastic with ir-
reversible changes, neither progressive nor amenable to any
type of treatment.
The upper tract approach may be recommended when
there is no associated VUR and the choice between removing
and keeping the upper pole is dictated by its relative function.
With this approach, the rate of secondary procedures on
the bladder ranges from 15% to 20% if preoperative VUR
was absent.90,91
When VUR is associated with an extravesical ureterocele,
no matter which approach is used first, a second stage at
the bladder level to remove the ureterocele and reimplant
the ureters is necessary in the majority of cases. As the number
of renal units affected by VUR in a duplex system with an ec-
topic ureterocele increases, the higher the incidence of surgery
at the bladder level.61,92
The reoperation rate in patients with duplex system ectopic
ureterocele and VUR varies from approximately 50% to 100%
after endoscopic incision93 and from 84% to 90% after upper
pole partial nephrectomy.90
BLADDER BASE RECONSTRUCTION
As mentioned earlier, bladder base reconstruction including
excision of ureterocele and ureteroneocystostomy may be re-
quired. This usually produces a significant defect that neces-
sitates a precise anatomic reconstruction; otherwise,
disruption of the bladder neck and posterior urethra could
lead to incontinence. Fortunately, it is unusual for the ureter
to open directly into the urethra; this most often occurs at a

level at or just above the bladder neck. Recently, it has been
suggested that the ureterocele can be excised or marsupialized
without much effect on clinical outcome.94 The ureterocele,
however, often involves the urethra, requiring that the epithe-
lial edges of the cyst be excised completely and then meticu-
lously repaired. Otherwise, obstruction of the urethra may
result from either the residual “valvular” remnants of the ure-
terocele or stricturing due to excessive scarring.80 Therefore it
is usually necessary to mobilize the unaffected lower pole ure-
ter with that associated with the ureterocele. If the upper pole
ureter is to be excised, mobilization of both ureters should be
extended above the common adventitial sheath. Alternatively,
the two ureters can be separated by preferentially taking the
adventitia of the affected ureter, thus preserving the blood
supply to the “normal” ureter. In any case the defect in the
bladder base and outlet must be repaired without narrowing
the outlet. The remaining lower pole ureter can then be reim-
planted using either a cross-trigonal or an extravesical method
as described earlier.
Proponents of complete primary lower urinary tract recon-
struction early in the neonatal period or infancy argue that the
need for secondary surgery in patients affected by duplex sys-
tem ectopic ureteroceles with preoperative VUR ranges from
0% to 32%.95–97 They have reported that this approach in pa-
tients with ectopic ureterocele with VUR appears to have bet-
ter results than a staged approach with initial endoscopic
treatment. Moreover, they conclude and have been supported
by others that extensive reconstructive bladder surgery in ne-
onates and infants does not lead to bladder function deterio-
ration at a later age.98 However, severe bladder dysfunction
was observed in 7 of 10 girls who underwent bilateral ureter-
ocele repair in their first 2 years of life,81 suggesting that due to
the gross abnormality of the bladder base in duplex systems,
there is a significant risk of damaging the bladder outlet in the
course of mobilizing the ureters. The difficulty is in determin-
ing whether these children’s problems are a primary abnor-
mality of bladder function or congenital and unrelated to
surgical intervention.99,100
CHAPTER 115 URETERAL DUPLICATION AND URETEROCELES 1451
Summary
------------------------------------------------------------------------------------------------------------------------------------------------
Ureteral duplications and ureteroceles represent a spectrum of
often overlapping embryologic and anatomic abnormalities.
Most unobstructed or nonrefluxing duplications are of no
clinical significance along with incomplete duplications;
others are associated with significant morbidity. Complete du-
plications are often asymptomatic but can be associated with
VUR or ureteroceles with a predominance in girls.
VUR is the most common anomaly of duplications, occur-
ring in about 65% of those with symptomatic UTIs; the reflux
is usually into the lower pole ureter. Ectopia of the ureter is not
uncommon, and a detailed history is often useful in pointing
toward the cause of incontinence. Specialized imaging tech-
niques are often required, however, to identify the precise an-
atomic abnormality; these include ultrasonography, isotope
scanning, MCUG, excretory urography, and MRI.
On the basis of these data, a planned, individualized
approach to each case can be taken, with an emphasis on pre-
venting deterioration in renal function, controlling symptoms,
and avoiding surgical interventions that could risk bladder
function.
Acknowledgments
I would like to acknowledge the contribution of Dr. Victor Boston, the previ-
ous author of this chapter, because parts of the previous version and some fig-
ures are retained. I would also like to thank my friend and colleague Dr. Emilio
Merlini for allowing me to refer to previous educational materials on duplica-
tion anomalies that we produced together.
The complete reference list is available online at www.
expertconsult.com.
SUGGESTED READINGS
Thomas DFM, Duffy PG, Rickwood AMK: Essentials of Paediatric Urology.
2nd ed. London, Informa Health Care, 2008.
Walsh PC, Retik AB, Vaughan Jr ED, Wein AJ, eds. Campbell’s textbook of
urology. 7th ed. Philadelphia: WB Saunders; 1998.
 discussed in this chapter can have a combination of these
symptoms. For example, parents of a boy with obstructing
posterior urethral valves may not report that their child is suf-
fering from urinary retention but rather that he is experiencing
urinary frequency and urgency (secondary to increased blad-
der irritability from overdistention). It should be established
whether the child has had a urinary tract infection (UTI)
and, if so, whether there has been associated fever. This latter
symptom will help distinguish an episode of cystitis from the
more significant episode of pyelonephritis. The physician
should inquire as to whether the child suffers from constipa-
tion in that it may reflect a form of retentive behavior (as in the
case of dysfunctional elimination) or neuropathic bowel dys-
function (as in the various forms of spinal dysraphism). As
pertains to the latter diagnosis, it is also important to record
any symptoms of lower extremity weakness or change in
sensation.
The physical examination of the child with bladder com-
plaints should include evaluation of the abdomen, back, gen-
italia, and lower extremities. In examining the abdomen,
special note should be made of palpable feces and whether
the bladder is palpable and/or percussible, especially if the
child has just voided. What has become a common pediatric
CHAPTER 116 urology office procedure and an extension of the physical ex-
amination is a postvoid bladder scan to check for residual
urine.
Examination of the back is critical. Cutaneous findings of a
Disorders of hair patch or nevus overlying the lumbar spine reflect abnor-
mal migration of the neuroectoderm, which can often signal
abnormal formation of the lumbosacral spinal cord. A dimple
Bladder Function over the lumbosacral spine may likewise signal such an abnor-
mality (Fig. 116-2). An asymmetric gluteal crease is a further
sign of potential problems of the spinal cord. This asymmetry
Martin Kaefer reflects differing degrees of impaired innervation to the two
groups of gluteal muscles. For example, if the right gluteal
nerves are affected to a greater degree than those on the left,
then a greater degree of atrophy may ensue and the crease
would deviate to the right. With respect to this last outward
sign of occult spinal cord dysraphism, it is critical to note that
Disorders of bladder function can range from the most mun- gluteal asymmetry may be an acquired finding as the child ma-
dane problem of dysfunctional voiding to complex disorders tures and grows in axial length. Therefore the back should be
involving neuropathic bladder dysfunction. The typical symp- examined not only on the initial patient evaluation but during
tom with which the patient presents is that of wetting. Al- return visits in children with continued voiding complaints.
though of great concern to parents and child alike, this Figure 116-3, A and B, demonstrates the concept of spinal
relatively benign-appearing outward sign of bladder dysfunc- cord tethering. Patients who possess any of these cutaneous
tion may in the extreme case reflect a far more insidious and findings should undergo a magnetic resonance imaging
occult problem involving injury to the lower and upper uri- (MRI) of the lumbosacral spine.
nary tracts. This chapter begins with the physician’s initial as- Examination of the male genitalia should include palpation
sessment of the child with voiding symptoms. Radiographic along the entire urethra. Induration may be a sign of urethral
and dynamic studies relevant to bladder dysfunction are cov- inflammation and stricture. Position of the testicles should be
ered next. Finally, specific disorders and their treatment are noted because there is an increased incidence of cryptorchi-
discussed. dism in children with myelodysplasia.1,2 The anus should
be examined to ensure that there is good tone and proper lo-
History and Physical cation on the perineum. Finally, any deficits in lower extremity
reflexes or motor strength should be noted in that they are out-
Examination ward signs of neuromuscular disorders, especially disorders of
------------------------------------------------------------------------------------------------------------------------------------------------
the lumbosacral spinal cord.
The evaluation of the child with bladder dysfunction should A urinalysis should be obtained in every child with
begin with establishing both day and nighttime voiding pat- voiding symptoms. The presence of red cells, white cells,
terns. Voiding symptoms have classically been divided into ir- glucose, and protein should be noted. A low specific gravity
ritative and obstructive, examples of which can be found in may reflect poor concentrating ability secondary to renal
Figure 116-1. It should be noted that nearly every disorder dysfunction.
1453
1454 PART VIII GENITOURINARY DISORDERS

Voiding Symptoms (e.g., detrusor sphincter dyssynergy) form of bladder outlet

obstruction (Fig. 116-4).3 Calculation of postvoid residual
Irritative urine volume is an excellent means of assessing whether the
Dysuria patient can efficiently empty the bladder.
Urgency
Frequency
VOIDING CYSTOURETHROGRAPHY
The instillation of radiographic contrast into the bladder
Obstructive
provides information regarding bladder, bladder neck, and
Retention urethral anatomy. Figure 116-5 demonstrates the classic ra-
Hesitancy
Staccato voiding (starting and stopping) diographic findings seen in neuropathic bladder dysfunction.
Straining to void Because of chronic high-pressure urine storage, the bladder
Feeling of incomplete emptying wall becomes thick and multiple diverticuli may form. In
many conditions the voiding phase provides the most useful in-
FIGURE 116-1 Irritative and obstructive voiding symptoms. formation because it is only during this time that the bladder
neck and urethra can be properly evaluated. The importance
of obtaining oblique imaging during voiding cannot be overem-
phasized because this may be the only image that reveals ure-
thral pathology. A classic example of this is the patient with
posterior urethral valves. The bladder neck and urethra proxi-
mal to the site of obstruction are markedly dilated, and the blad-
der itself can demonstrate findings of trabeculation and
diverticuli formation. Vesicoureteral reflux is a common finding
in both anatomic and functional causes of bladder dysfunction.

URODYNAMIC EVALUATION
The urodynamic evaluation is the one test that provides in-
depth functional information regarding the bladder. It estab-
lishes bladder capacity, bladder filling pressures, the patient’s
perception of bladder filling, and the ability of the bladder
neck to empty efficiently and in proper coordination with
external sphincter relaxation.
Normal values for various urodynamic parameters have been
published. Two parameters are especially useful to mention.
Early formulas for calculating expected bladder volume gave
reasonable estimates of bladder capacity. One of the most com-
monly used is Age (years) þ 2 ¼ Capacity (in ounces).4 How-
ever, data in the studies were obtained in large part from patients
undergoing evaluation for functional bladder pathology.5 More
recently, Kaefer and colleagues, in measuring the bladder capac-
ity of 2000 healthy children without bladder pathology, demon-
FIGURE 116-2 Cutaneous findings suggestive of occult spinal cord teth- strated that bladder capacity was not a linear function of age.6
ering: picture of a 10-year-old child with neuropathic bladder who was The curvilinear relationship 4.5  age0.4 ¼ capacity (ounces)
found on physical examination to have a cutaneous lumbar birthmark, a
lumbar dimple and asymmetry of the gluteal cleft. Magnetic resonance
is similar to the relationship between age and other morpho-
imaging of the lumbosacral spinal cord revealed an intraspinal lipoma metric parameters (e.g., height and weight). This nonlinear re-
responsible for cord tethering. lationship can be approximated by two practical linear formulas
that have excellent predictive value when applied prospectively
(2  Age [years] þ 2 ¼ capacity [ounces] for children younger
Radiographic and Dynamic than 2 years old, and Age [years]  2 þ 6 ¼ capacity [ounces]
for those older than 2 years old).
Assessment of Bladder Neurologic modulation combined with the viscoelastic
Dysfunction properties of the healthy detrusor muscle allow the bladder
------------------------------------------------------------------------------------------------------------------------------------------------ to maintain fairly constant pressure throughout the filling
BLADDER ULTRASOUND phase. As a result, intravesical pressures in the healthy bladder
remain at or below 5 to 10 cm H2O pressure until capacity is
Although it is the most basic of radiographic imaging modal- achieved. When various pathologic processes alter the compo-
ities, the bladder ultrasound can provide many clues about sition of the bladder wall (e.g., collagen deposition following
whether there is significant bladder pathology. In the absence mechanical outlet obstruction) or affect neurologic control
of infection, a thickened bladder wall may reflect a com- of the bladder (e.g., loss of upper or lower motor neuron func-
pensatory response on the part of the bladder to either tion in myelomeningocele) compliance can be adversely af-
an anatomic (e.g., posterior urethral valves) or functional fected, resulting in increased intravesical storage pressures.
 CHAPTER 116 DISORDERS OF BLADDER FUNCTION 1455

A Healthy B Tethered
FIGURE 116-3 A, Normal spinal cord. Spinal column increases in axial length at a greater rate than does the spinal cord. As a result, the spinal cord rises to
a higher position. B, Abnormal spinal cord anatomy with spinal cord lipoma. As the spine grows, the spinal cord is unable to float upward and the cord is
stretched. This stretching results in a relative state of hypoxia in the area of the distal cord and subsequent nerve injury to the distal spinal cord segments.

If intravesical storage pressures reach levels higher than 40 cm
H2O, renal injury is likely to ensue.7
Before undergoing placement of the urodynamic catheter,
the child is asked to void into a toilet that measures urinary
flow rate. The catheter is then placed through the urethra
(or alternatively through a suprapubic site if one is available),
and the intravesical pressure is measured. The bladder is then
emptied and residual urine determined, yielding a pressure at
residual volume.8 This measurement may prove useful in de-
termining whether the bladder is experiencing high resting

FIGURE 116-4 Bladder ultrasound from a child with posterior urethral FIGURE 116-5 Typical voiding cystourethrogram from a child with neu-
valves. Note the markedly thickened bladder wall (left arrow) and the dila- ropathic bladder dysfunction secondary to myelomeningocele. The
tation of the distal ureters (right arrow). “Christmas tree” appearance is a result of greater bladder wall thickening
at the dome relative to the bladder base.
1456 PART VIII GENITOURINARY DISORDERS

pressures. Next, electromyographic (EMG) electrodes are
placed on the perineum either in the form of patch surface
sensors or needle sensors, the latter of which is placed directly
into the external sphincter complex. These electrodes record
the electrical activity of the muscle in order to determine the
function of the external sphincter during bladder filling and
emptying. Under normal conditions the external sphincter
is active during periods of urine storage. Just before voiding
there should be silencing of external sphincter activity as
the bladder neck relaxes (Fig. 116-6, A). Failure of the sphinc-
ter to relax before a voiding contraction is termed detrusor
sphincter dyssynergy (DSD) and is commonly noted in patients
with neuropathic bladder dysfunction (Fig. 116-6, B), as well
as in patients with functional urinary incontinence.

Bladder neck Bladder neck Bladder

closed OPEN empty

Ext. sphincter Ext. sphincter Ext. sphincter

CLOSED OPEN CONTRACTS

EMG

100

Bladder
Pressure (cm H2O)

50
Voiding

0
A Synergic voiding

FIGURE 116-6 A, Coordinated synergic voiding: Electrical activity of the external urethral sphincter is chronically active. Just before experiencing a detru-
sor contraction, the activity of the external sphincter is silenced, providing a marked decrease in urethral resistance and thereby facilitating emptying.
Intravesical pressure is shown in the bottom panel.
 CHAPTER 116 DISORDERS OF BLADDER FUNCTION 1457

Bladder neck Bladder neck Residual

closed still closed volume

Ext. sphincter Ext. sphincter Ext. sphincter

CLOSED CLOSED CLOSED

EMG

100

Bladder
Pressure (cm H2O)

50
Voiding

0
B Dyssynergic voiding

FIGURE 116-6—CONT’D B, Dyssynergic voiding: In cases of neuropathic bladder dysfunction discoordination between bladder and sphincter activity
can often be appreciated. As the bladder begins to contract the activity of the external sphincter actually increases resulting in high outlet pressures and
ineffective voiding. Intravesical pressure shown in bottom panel.

The bladder is next filled slowly with normal saline to de-

termine bladder capacity, and the pressure-volume relation- Neuropathic Causes of Bladder
ship is measured in order to establish bladder wall
compliance. Percent expected bladder capacity is calculated
Dysfunction
------------------------------------------------------------------------------------------------------------------------------------------------

by dividing the patient’s actual capacity by the expected nor-
mal bladder capacity for age.9 During the filling phase, any un-
inhibited contractions should be noted because they are a
reflection of bladder irritability. Figure 116-7 shows a compar-
ison between pressure tracings from a compliant bladder
with those of a poorly compliant, irritable bladder. During
the filling phase, the leak point pressure is also noted.
NEUROPATHIC BLADDER SECONDARY
TO MYELODYSPLASIA
Myelodysplasia, defined as abnormal development of the
spinal canal and spinal cord, is the most common etiology
of neuropathic bladder dysfunction in children. A genetic
component appears to be partially responsible for this
1458 PART VIII GENITOURINARY DISORDERS

40

Pressure (cm H2O)

30

20

10

0.0
50 100 200 240
Volume (ml)

40
Pressure (cm H2O)

30

20

10

0.0
50 75 120
Volume (ml)
FIGURE 116-7 Representation of typical urodynamic tracings from a compliant bladder and a poorly compliant bladder. Upper tracing demonstrates a
compliant bladder: As the bladder fills, numerous factors serve to maintain low intravesical pressure until a voiding contraction is initiated. Lower tracing
demonstrates a poorly compliant bladder with detrusor hyperreflexia. As the bladder fills, intravesical pressure rises precipitously. Irritability of the dys-
functional detrusor muscle is reflected in numerous small pressure spikes (i.e., uninhibited contractions).

disorder. In a family with one child with myelodysplasia, there

FIGURE 116-8 Photograph of spinal dysraphism.
exists a 2% to 5% chance that each subsequent sibling will
suffer from the same condition. Early studies reported the
incidence at 1 in 1000 live births.10 However, over the past
20 years there has been a steady decrease in this rate. Reasons
cited for this decrease include the widespread perinatal sup-
plementation of folic acid (a metabolite important for proper
spinal cord formation) and pregnancy termination.11,12 Die-
tary supplementation with folic acid can reduce the incidence
of myelodysplasia by approximately 50%.
The most common forms of myelodysplasia consist of pro-
lapse of the meninges (meningocele) and in many cases neural
tissue (myelomeningocele) beyond the confines of the bony
vertebral canal. Initial urologic evaluation of the child found
to have spinal dysraphism (Fig. 116-8) should include a renal
ultrasound, voiding cystourethrogram, and determination of
postvoid residuals by clean intermittent catheterization
(CIC). The renal ultrasound will detect the presence of hydro-
nephrosis. Up to 15% of newborns will be found to have an
abnormal urinary tract; 3% have hydroureteronephrosis sec-
ondary to spinal shock from the closure procedure; and
10% have abnormalities that developed in utero as a result
of abnormal lower urinary tract function in the form of outlet
obstruction, thus resulting in high intravesical pressures.13,14
The renal ultrasound serves the additional purpose of evalu-
ating for renal fusion anomalies, which are known to be more
common in patients with myelodysplasia.15

The voiding cystourethrogram serves the purpose of detect-
ing vesicoureteral reflux, which is present in 3% to 5% of new-
borns with myelodysplasia.16 Reflux is typically seen in
children with detrusor hypertonicity or detrusor sphincter dys-
synergy. New reflux will develop in approximately 30% of chil-
dren with unfavorably high-pressure bladder dynamics if
proper treatment is not taken to lower intravesical pressures.
As mentioned earlier, several postvoid residuals (PVRs) are
obtained in order to determine the efficiency of emptying. If
the postvoid residuals average greater than 10 mL, the parents
are taught CIC to be performed three times daily after discharge
to home. Alternatively, because it may be difficult to obtain
exact PVRs in newborns, random catheterized volumes can
be measured. Random catheterized volumes consistently below
expected bladder capacity imply efficient bladder emptying.
Follow-up urologic evaluation consists of a urodynamic
evaluation at 3 months of life. Performing urodynamics at
an earlier time point is often discouraged due to the fact that
the infant may experience a degree of spinal shock from spinal
closure that may last for up to 2 months. The urodynamic
evaluation serves two purposes. First, it serves as a baseline
against which all future urodynamic evaluations can be com-
pared. Changes in the urodynamic profile may be the first in-
dication (often before lower extremity function changes)
that postmyelomeningocele closure spinal cord tethering is
occurring and that surgical intervention may be required.
The second purpose of the urodynamic evaluation is to de-
termine the overall storage characteristics of the bladder and
sphincteric function. Three specific combinations of bladder
contractility and external sphincter activity are seen. Bauer
demonstrated that 19% of patients will demonstrate synergic
voiding, 45% will have dyssynergic voiding, and the remain-
ing 36% will suffer from complete denervation.17 This system
of classification is of great importance in predicting long-term
renal and bladder function and thereby determining who will
benefit from aggressive measures to minimize progressive uri-
nary tract injury. Bauer has shown that within the first 3 years
of life, patients with a dyssynergic voiding pattern have a
much higher incidence of urinary tract deterioration. He
found that 71% of newborns with DSD had urinary tract de-
terioration on initial assessment or subsequent studies,
whereas only 17% of synergic children and 23% of completely
denervated individuals developed similar changes. Notably,
the small percentages of children with synergic or denervated
voiding patterns that did go on to demonstrate urinary tract
deterioration all had subsequent conversion to a high outlet
resistance.
Frequent follow-up of patients with neuropathic bladder
dysfunction secondary to myelomeningocele is mandatory
because bladder dynamics (and thereby the effect on the up-
per urinary tract and continence) can change with time. Many
reasons for a change in bladder dynamics exist, the most
important being spinal cord tethering.18
The primary goal of treatment of the neuropathic bladder
is the preservation of renal and bladder function. An addi-
tional goal of achieving urinary continence is addressed once
the child reaches the age at which his or her peers are achiev-
ing this developmental milestone. CIC and the use of anti-
cholinergic medications are the cornerstones of medical
therapy to provide a low-pressure storage environment for
urine. Edelstein and colleagues19 demonstrated that if one ag-
gressively and proactively treated high-risk patients who are
CHAPTER 116 DISORDERS OF BLADDER FUNCTION 1459
defined as having either high outlet resistance, detrusor
sphincter dyssynergy, and/or bladder hypercontractility (so-
called hostile bladder dynamics) with CIC and anticholinergic
medications that the rate of upper urinary tract injury could
be substantially reduced. Kaefer and colleagues20 subse-
quently demonstrated that long-term bladder function is sim-
ilarly preserved in a greater number of patients if such
measures are taken (Fig. 116-9). In this later study the inci-
dence of long-term bladder deterioration requiring surgical
bladder augmentation was reduced 2.5-fold. If these medical
measures do not adequately reduce intravesical pressures and
progressive hydronephrosis is noted in the first few years of
life, then surgical intervention in the form of a cutaneous vesi-
costomy is indicated.
Urinary incontinence is likewise treated initially with con-
servative measures (i.e., CIC and anticholinergic medications)
to decrease intravesical pressures. If such measures do not re-
sult in adequate improvement in bladder capacity and compli-
ance, it may become necessary to enlarge the bladder by
means of enterocystoplasty.21,22 Additional surgical maneu-
vers may be required to provide adequate outlet resistance
in order to achieve a state of urinary continence.23
OCCULT SPINAL CORD TETHERING
A number of congenital defects affect the formation of the spi-
nal column yet do not result in an open vertebral canal. These
conditions occur in approximately 1 of every 4000 live
births.24 These lesions may result in no obvious outward neu-
rologic signs, but in many there is a cutaneous abnormality
overlying the lower spine (see earlier section on physical ex-
amination). There is an increased incidence (up to 50%) of
these conditions in individuals with anorectal malforma-
tions,25,26 with earlier reports demonstrating that the inci-
dence varied proportionately in relation to the height of the
Expectant Prophylactic
11 16 15
3
− + − +
Augmentation
FIGURE 116-9 Incidence of bladder deterioration relative to treatment
regiment in myelomeningocele patients. The expectant group comprised
patients with hostile bladder dynamics noted at birth who were started on
clean intermittent catheterization and anticholinergic medications to
lower bladder pressure only after signs of upper tract deterioration or
worsening incontinence became evident. On average treatment was initi-
ated at age 4.1 years in this group. The prophylactic group comprised pa-
tients with hostile bladder dynamics noted at birth who were treated with
measures to lower intravesical pressures beginning in the first 3 months of
life. The greater than twofold higher incidence of bladder augmentation in
the expectant group is highly statistically significant (41% vs. 17%). (From
Kaefer M, Pabby A, Kelly M, et al: Improved bladder function after prophy-
lactic treatment of the high risk neurogenic bladder in newborns with mye-
lomeningocele. J Urol 1999;162(3 Pt 2):1068-1071.)
1460 PART VIII GENITOURINARY DISORDERS

rectal lesion.27 However, one recent study suggests that the

incidence may not differ significantly between patients with
supralevator and infralevator lesions.25 All patients with a
cutaneous abnormality overlying the lower spine or an anor-
ectal malformation should undergo an MRI to evaluate for
an intraspinal lesion.28 In the first 6 months of life it is also
possible to evaluate for spinal cord tethering with a spinal
ultrasound.29 When using ultrasound, one expects to see
the conus medullaris at the level of the L2 vertebrate. In addi-
tion, the nontethered spinal cord should move with respira-
tion. After 6 months of life, ossification of the vertebral
segments limits the ability of ultrasound to penetrate to the
level of the spinal cord and achieve adequate visualization
of the distal spinal cord.
Urologic symptoms of occult spinal cord tethering may in-
clude difficulty with toilet training, urinary incontinence after
an initial period of dryness (especially during the pubertal
growth spurt when the most stress is put on the tethered
cord), recurrent urinary infections, and/or fecal soiling. A
Although the majority of newborns evaluated for occult
spinal cord tethering have a perfectly normal physical exam-
ination, urodynamic testing will reveal abnormal lower uri-
nary tract function in about one third of babies younger
than 18 months of age.30 In contrast, practically all individuals
older than 3 years of age who have not been operated on or in
whom an occult dysraphism has been belatedly diagnosed
have abnormal urodynamic profiles. Truncated cord
Once it has been shown on MRI that there are signs of spi-
nal cord tethering (e.g., the conus medullaris lies below the
second lumbar vertebrae or there is a significant intraspinal
lipoma), then a judgment is made as to the need for neuro-
surgical intervention. Often, the first sign of neurologic
impairment is an abnormal urodynamic profile. Therefore
urodynamics are performed on all patients with positive
MRI findings. If a decision is made not to perform spinal cord
tether release, then follow-up with biannual renal ultrasound Sacral agenesis
and yearly urodynamic evaluations is indicated so that any
future neurologic deficits can be identified early and appropri-
ate intervention can be instituted in a timely fashion before B
irreversible nerve injury occurs.
FIGURE 116-10 A, Photograph of buttocks of patient with sacral agen-
esis. Note flattened buttocks and a low, short gluteal cleft. B, Magnetic res-
SACRAL AGENESIS onance imaging of the lumbosacral spine in patient with sacral agenesis.
Note sharp cutoff of the conus at T12 (arrow).
Patients with sacral agenesis frequently suffer from neuro-
pathic bladder dysfunction. The etiology of sacral agenesis
is still uncertain. It is, however, known that insulin-dependent have no sign of bladder dysfunction.35 The number of affected
mothers have a 1% chance of giving birth to a child with this vertebrae does not correlate with the type of motor neuron le-
disorder and that 16% of children with sacral agenesis have a sion. In contrast to the lesions noted in myelodysplasia or oc-
diabetic mother.31,32 Because these children have normal sen- cult spinal cord anomalies, the injury is most frequently found
sation and little or no orthopedic deformity in the lower ex- to be stable. Although it is rare to see signs of progressive de-
tremities, the underlying lesion is often overlooked. Up to nervation such as a changing urodynamic pattern in these
20% of children with this condition escape detection until children as they grow, there can occasionally be a progressive
the age of 3 or 4 years.33 Typically, the only outward sign of renal impairment.36,37 Anticholinergic agents are used to treat
this condition is flattened buttocks and a low, short gluteal uninhibited contractions and/or relax the hypertonic detrusor,
cleft (Fig. 116-10, A). The diagnosis is confirmed with a lateral whereas CIC is used in patients with impaired bladder
film of the lower spine, which reveals the missing sacral ver- emptying.
tebrae. MRI of the lumbosacral spine, which is recommended
in all patients, consistently reveals a sharp cutoff of the conus
CEREBRAL PALSY
at T1234 (Fig. 116-10, B). Urodynamic testing shows that an
almost equal number of individuals manifest a primarily up- Cerebral palsy is the term used to describe a nonprogressive
per (i.e., hypertonic, hyperreflexic) or lower (i.e., atonic) mo- injury of the brain (e.g., hypoxic) occurring in the perinatal
tor neuron type lesion (35% vs. 40%, respectively), while 25% period that frequently produces a neuromuscular disability.38
 CHAPTER 116 DISORDERS OF BLADDER FUNCTION 1461

Its incidence is approximately 1 in 1000 births. However, the (e.g., short stature). Today, with the widespread use of prena-
incidence of cerebral palsy appears to be increasing as smaller tal sonography, the majority of boys with bladder outlet
premature infants survive in intensive care units. Although a obstruction are identified before delivery.44
high percentage of children will be found to have uninhibited Anatomic obstruction of the urethra results in pathologic
contractions on urodynamic evaluation, the majority of chil- changes at both the level of the kidney and the bladder. The
dren with this diagnosis have the capacity to develop total uri- bladder wall hypertrophies, diverticuli can form, and the
nary control.39 Incontinence is most commonly related to the posterior urethra becomes variably dilated (Fig. 116-11).
physical handicap, making it difficult for the child to get to the Vesicoureteral reflux is seen in between 30% and 50% of
bathroom in time. For this reason continence is often achieved patients. In approximately one third of these cases the reflux
at a later than expected age, once the child has become adept resolves once the obstruction has been relieved.45 However,
at transferring. Properly addressing the issue of adequate ac- even in the absence of vesicoureteral reflux the upper uri-
cess to handicapped facilities may be extremely helpful in nary tract may still suffer injury due to lower urinary tract
many of these children. dysfunction that arises as a result of bladder wall thickening.
As a result of the many physical issues facing the child with
cerebral palsy, urodynamic evaluation is typically reserved for
those children who appear to be trainable, do not seem to be
hampered too much by their physical impairment, and have
not achieved continence by late childhood. For similar reasons,
upper and lower urinary tract imaging is not recommended
unless UTI has occurred.
Treatment focuses on providing an appropriate setting to
allow the patient to easily access a toilet where they can prop-
erly relax while voiding. Individualized orthotics and upper
body stabilization can help in this regard. If urodynamic data
reveal uninhibited bladder activity, one can use anticholiner-
gic medications but residual urine must be monitored closely
to ensure complete evacuation with each void. CIC may be re-
quired for those who cannot empty their bladder. These chil-
dren are also prone to severe constipation largely due to their
poor fluid intake and limited physical activity. Insofar as con-
stipation can adversely affect bladder stability and emptying,
treatment to improve fecal elimination can prove beneficial in
treating incontinence in these children. In a small subset of
patients who do not respond to conservative measures, selec-
A
tive dorsal rhizotomy has improved bladder capacity, reduced
the number of uninhibited contractions, and increased
compliance.40

Anatomic Causes of Bladder

Dysfunction
------------------------------------------------------------------------------------------------------------------------------------------------

POSTERIOR URETHRAL VALVES

The most common form of anatomic bladder outlet obstruc-
tion in the pediatric population is posterior urethral valves
(PUVs) in boys. Other etiologies of bladder outlet obstruction
include urethral stricture, anterior urethral diverticulum, and
an obstructing ureterocele situated at the bladder outlet.41–43
The incidence of posterior urethral valves is between 1 in
5000 and 1 in 8000 male births. Children present in a variety
of ways depending on the degree of obstruction. Newborn
children with severe outlet obstruction from posterior urethral
valves typically present with a palpable midline abdominal
mass (a distended bladder) and/or ascites (from urinary leak B
at the level of the kidney). In the most extreme cases children FIGURE 116-11 Voiding cystourethrograms in children with posterior
may experience respiratory distress from pulmonary hypopla- urethral valves. A, Severe case in which there is dramatic bladder decom-
sia due to the lack of adequate amniotic fluid volumes during pensation with multiple small and large diverticuli (uppermost arrow) and
lung development. Children who are identified later in life as dilated posterior urethra (middle arrow). The site of valvular obstruction is
shown by the inferior arrow. B, Mild case in which there is little evidence of
having PUVs generally do not suffer from as severe an obstruc- bladder deterioration. Upper arrow points to the distended posterior ure-
tion. These children typically present with voiding dysfunc- thra. Lower arrow points to the negative impression resulting from the pos-
tion, UTI, and on rare occasion with signs of renal failure terior urethral valve itself.
1462 PART VIII GENITOURINARY DISORDERS

Both cellular hypertrophy and the increased deposition of ex-

tracellular matrix are seen histologically in a bladder that is of Nonneuropathic, Nonanatomic
lower capacity and decreased compliance.
Endoscopic destruction of the valve leaflets can be per-
Causes of Bladder Dysfunction
------------------------------------------------------------------------------------------------------------------------------------------------

formed safely in nearly all term infants. An antegrade ap-
proach to valve destruction has been used in patients whose
urethra is too small to accept a cystoscope.46 Although cautery
electrodes are the most frequently used device for disrupting
the obstructing leaflets, cautery hooks, the neodymium:
yttrium laser, and the antegrade withdrawal of a balloon cath-
eter have also been used effectively for this purpose.47–49
Once the urethral obstruction has been relieved, bladder
dysfunction may persist. Most often the bladder dysfunction
manifests itself as daytime urinary incontinence that persists
after the child has reached school age. Initially it was felt that
persistent incontinence was due to injury to the urethral
sphincter during endoscopic ablation of the valves and/or
abnormal bladder sphincter development as a result of the
initial obstruction. However, urodynamic assessment of
these children has clearly shown that persistent bladder dys-
function is frequently seen well after the valves have been
ablated. Several investigators have demonstrated that there
are three primary abnormalities of detrusor function: blad-
der hypertonia, detrusor hyperreflexia, and myogenic
failure.50–53 Furthermore, evidence suggests that urody-
namic patterns can change as the child matures. A study
by Holmdahl and colleagues54,55 has demonstrated that
many children who are initially shown to have a hypercon-
tractile, poorly compliant bladder as newborns will later
have a urodynamic pattern that is more consistent with myo-
genic failure. Many children with bladder dysfunction sec-
ondary to posterior urethral valves also appear to have
abnormal bladder sensation and an inability to sense when
their bladder is full.56 Adding to the problem of persistent
bladder dysfunction is the common finding of renal injury
and associated poor urinary concentrating ability. The large
volumes of urine that result put a further stress on the
dysfunctional lower urinary tract.
The treatment of bladder dysfunction in children with pos-
terior urethral valves depends on the severity and form of
bladder dysfunction, as well as the efficiency of bladder emp-
tying. Patients with high-pressure voiding dynamics often
benefit from the use of anticholinergic medications to improve
compliance. In younger patients with an impaired ability to
empty the bladder spontaneously, CIC may be required. Day-
time catheterization can often be avoided in older children if
they can adhere to a strict schedule of timed voiding. Recently
Koff and colleagues have advocated the use of a nighttime in-
dwelling catheter to provide optimal bladder drainage. By im-
proving nighttime drainage, the authors have documented
significant improvement in hydronephrosis and bladder com-
pliance.56,57 Although conservative measures are usually ef-
fective in modifying bladder dynamics (i.e., keeping bladder
volumes sufficiently low to maintain acceptable bladder pres-
sures), a small number of patients may still require bladder
augmentation to improve bladder volume and compliance.52
Patients with posterior urethral valves do not have altered ure-
thral sensation, and many of these children will find catheter-
ization through the penis painful. Creation of a continent
catheterizable channel to the anterior abdominal wall can
therefore improve the quality of life and result in improved
compliance with catheterization in these children.58
DYSFUNCTIONAL ELIMINATION
SYNDROMES
Bladder dysfunction in the absence of anatomic bladder outlet
obstruction or neurologic disease is frequently termed dysfunc-
tional voiding. This term encompasses children who fit along
a spectrum of relative urinary retention. At the most severe
end of this spectrum are children who retain urine to such a
great degree that they demonstrate altered bladder anatomy, up-
per urinary tract dilatation, and scarring, which is virtually
indistinguishable from that seen in neuropathic bladder dys-
function (the so-called non-neurogenic neurogenic bladder)
(Fig. 116-12).59,60 The severity of the radiographic findings
leads the physician to obtain an MRI of the lumbosacral spine,
which by definition shows no abnormalities suggestive of spinal
cord tethering. Urodynamic evaluation classically demonstrates
a low-capacity, high-pressure bladder. Treatment of Hinman
syndrome parallels that used in true neuropathic bladder dys-
function. An attempt should be made to lower intravesical pres-
sures with the combined use of anticholinergic medication and
CIC. If conservative measures fail to adequately reduce storage
pressures, bladder augmentation is performed. Unlike patients
with true neurologic lesions, many of these children will find
catheterization painful and will benefit from creation of a con-
tinent catheterizable channel to the anterior abdominal wall.
At the opposite end of the spectrum is a large population of
otherwise healthy children who, as a result of relative urinary
retention experience, increased irritative voiding symptoms
and a higher propensity toward developing UTIs. In the past
decade it has become clear that constipation can be seen in
many of these individuals and that a large fecal burden can have
a significant impact on bladder function. As a result, the more
encompassing term dysfunctional elimination syndrome (DES)
has largely supplanted the term dysfunctional voiding.
FIGURE 116-12 Voiding cystourethrogram from a child with Hinman
syndrome. Note the significant irregularity of the bladder outline second-
ary to bladder wall thickening.
 CHAPTER 116 DISORDERS OF BLADDER FUNCTION 1463

FIGURE 116-13 The effect of constipation on bladder function. Left, Stool in the rectum may push up on the posterior aspect of the bladder, resulting in
bladder instability. Right, Severe forms of constipation have the potential to externally compress the bladder neck and inhibit bladder emptying.

Children affected by this condition do not empty their

bowel or bladder as often or as completely as they should,
or they require high bladder pressures to empty their blad-
ders. As a result of poor bladder emptying, the bladder is con-
stantly distended and as such leaves the patient prone to urge
incontinence. If children fail to empty completely, they will
reach a state of fullness more often and this may be reflected
in the symptom of urinary frequency. The maintenance of a
high postvoid residual leaves the patient more prone to UTIs
and incontinence. Constipation can contribute bladder irrita-
bility from the mass effect on the posterior aspect of the blad-
der and in severe cases may also inhibit bladder emptying
(Fig. 116-13). Poor bladder emptying has been shown by
Dohil and colleagues61 to improve following bowel treatment
in constipated children, with postvoid residual urine drop-
ping from 66% at baseline assessment to 21%. Chronic con-
stipation is associated with significant hypertrophy of the
internal anal sphincter and abnormal anal sphincter EMG
activity.62 This may generate increased urethral sphincter
and pelvic floor activity and explain the association between
voiding dysfunction and incomplete voiding.
Radiographic findings that are highly suggestive of this di-
agnosis include a large stool burden on the plain abdominal
film or evidence of a full rectosigmoid segment behind the
bladder on ultrasound, a large postvoid residual on bladder
ultrasound, and the “spinning top” image of the bladder neck
on voiding cystourethrogram (Fig. 116-14). FIGURE 116-14 Voiding cystourethrogram from a child with dysfunc-
Treatment of the child with DES includes (1) timed voiding tional elimination syndrome demonstrating a full bladder with a small di-
verticulum and a distended proximal urethra (uppermost arrow). The
in which the child is asked to void every 2 hours, (2) double lowermost arrow depicts the site of the external sphincter. Because of fail-
voiding in which the child is asked to attempt to void a second ure to relax, the result is proximal urethral ballooning (the so-called spin-
time after emptying the bladder (in an attempt to minimize the ning top deformity).
chances of carrying a large postvoid residual), and (3) estab-
lishing good bowel emptying habits. The latter behavior can these conservative measures fail to result in adequate relaxa-
be supported by increasing dietary fiber, using stool softeners, tion during voiding/defecating, biofeedback may prove bene-
and taking plenty of time to defecate completely. Providing ad- ficial. Many ingenious methods for performing biofeedback
equate foot support can be beneficial in achieving a state of including the use of videogames have been developed.63
balance and thereby enhancing adequate relaxation. Should Although biofeedback has proven to be highly beneficial in
1464 PART VIII GENITOURINARY DISORDERS
many patients, this modality may require multiple office visits
with skilled nursing care, which can often be difficult due to
time and/or financial constraints. An alternative that has re-
cently gained much attention is treatment with a-adrenergic
inhibitors. The exact mechanism and site of action for a-blocker
therapy in children with poor bladder emptying remains in
question. There are documented reports of a1-adrenergic recep-
tors at the bladder outlet and in the proximal urethra.64,65 It is
felt that the primary mechanism of action is smooth muscle re-
laxation at the base of the bladder and decreased outlet re-
sistance in the proximal sphincter complex. Both tamsulosin
and doxazosin have been studied in the pediatric setting.66
Selective a-blocker therapy appears to be effective for impro-
ving bladder emptying in children with wetting, recurrent in-
fection, and increased PVR urine. Early treatment of increased
postvoid residual urine volumes with a-blockers may signifi-
cantly reduce the number of patients who will need biofeedback
therapy. The more commonly reported side effects of these med-
ications are nasal congestion, dizziness, and postural hypoten-
sion. We recommend administering the medication before
bedtime so as to minimize the potential for experiencing dizzi-
ness during the waking hours. It is also imperative that the
blood pressure be evaluated 3 days following initiation of the
medication to ensure that the child is not experiencing ortho-
static hypotension.
OVERACTIVE BLADDER SYNDROME
Although the majority of children who present to the pediatri-
cian with wetting and urinary frequency will have a retentive
behavior as the etiology, a smaller population will demonstrate
these symptoms due to a state of relative detrusor overactivity
in the absence of demonstrable neurologic impairment. As a
result, bladder filling even to moderate volumes will often trig-
ger a bladder contraction. This group of children does not
demonstrate retentive bladder or bowel behavior and carries
minimal postvoid residual when evaluated by bladder ultra-
sound. An effective treatment for this condition is the use of
anticholinergic medications. Anticholinergic medications are
competitive inhibitors of acetylcholine that block the neuro-
transmitter’s muscarinic effects. Anticholinergic medications
are typically titrated to effect. Common side effects consist
of dry mouth, flushing, and constipation. Constipation espe-
cially may pose a problem because children with detrusor
overactivity have a predilection for constipation, and the de-
velopment of constipation may aggravate detrusor overactivity
and thus counteract the beneficial effects of the drug. Oxybu-
tynin (0.2 mg/kg twice daily or three times daily) has been the
most often used anticholinergic medication. The newer anti-
muscarinic drug tolterodine has, in adults, shown a more fa-
vorable therapeutic profile, with the same clinical efficacy and
a lesser frequency of side effects compared with oxybutynin.67
Tolterodine has also been shown to be safe and effective in
children.68 A commonly used dosage for tolterodine in chil-
dren is 0.1 mg/kg twice daily. Sustained release formulations,
which can be taken once daily, are also a choice for children
who can tolerate medication in pill form. In patients who ex-
perience minimal side effects, studies have shown increased
effectiveness by doubling the recommended daily dosage.69,70
Overactive bladder symptoms are typically self-limiting with
even the most recalcitrant cases showing resolution within
an 18-month period.71 Failure to improve on anticholinergic
medication may warrant a more thorough investigation of the
patient in the form of a urodynamic evaluation.
NOCTURNAL ENURESIS
Whether a true problem of bladder function, a failure of ad-
equate nocturnal urinary concentrating ability, or an issue re-
lated to a relatively poor state of nocturnal arousal, nighttime
wetting is often discussed in chapters on bladder dysfunc-
tion.72 Various series have reported the incidence of isolated
nocturnal enuresis as between 6% and 10% in children
who have reached age 7 years. The spontaneous cure rate in
children between the ages of 5 and 20 is 15% annually.73 Most
children who are enuretic eventually obtain normal control.
Nocturnal enuresis is a genetically complex and heteroge-
neous disorder. Genetic factors play an important role in the eti-
ology. A positive family history may be elicited in more than 50%
of cases. Within families different members can show the same
or different forms of wetting. Studies evaluating the incidence of
nocturnal enuresis in twins have shown 46% for monozygotic
pairs and 19% for dizygotic pairs.74 The most common mode
of transmission appears to be autosomal dominant.
Various pathophysiologic mechanisms appear to be at play
including nocturnal polyuria, bladder overactivity at night
with a small functional bladder capacity, and disorder of
arousal.75 Furthermore, these groupings show considerable
overlap. In humans there is a marked circadian rhythm of
urine production so that there is a nighttime reduction in di-
uresis of up to 50% of daytime levels.76 In children this is con-
trolled by increased nocturnal release of arginine vasopressin,
as well as angiotensin II and aldosterone.77,78 It has long been
known that children with nocturnal enuresis have signifi-
cantly larger nocturnal urine production than nonenuretic
children. This group of patients has a favorable response to
the arginine vasopressin analogue dDAVP.79 Although a cer-
tain subset of enuretics has a smaller nocturnal functional
bladder capacity, the incidence of functional abnormalities
seems rather low in children with isolated nocturnal enuresis.
Whether wetting is primarily due to a problem of urinary con-
centrating ability, functional bladder capacity, or other as of yet
unrecognized etiology, a prerequisite for wetting is failure to
awake when micturition is imminent. This has caused many
to conclude that sleep disturbance per se is the major patho-
physiologic factor in enuresis, and it is still a widely held belief
that enuretics are deep sleepers. However, a number of studies
have been unable to convincingly show abnormalities in sleep
patterns.
Treatment of nocturnal enuresis takes many forms. If pre-
sent, daytime symptoms should be treated. The child should
be asked to void regularly and not to delay urinating after
experiencing a sense of bladder fullness. Symptoms of urgency
and frequency, as well as constipation, should be addressed.
A rate of enuresis cure up to 72% has been reported after con-
stipation has been adequately addressed.80 To minimize the
possibility of nocturnal polyuria, fluids should be limited sev-
eral hours before bedtime. A low calcium and sodium dietary
content of the afternoon and evening meals may also be useful.
The child should be instructed to void before bedtime. This
helps to minimize the nocturnal bladder volume.
Enuresis that persists after age 6 may be treated with med-
ication and/or a bed alarm device. Desmopressin (dDAVP) is
easy to administer and has an immediate effect on urinary

volume. Dosing is between 0.2 and 0.6 mg orally. Intranasal
dDAVP has recently lost its indication for nocturnal enuresis
by the U.S. Food and Drug Administration due to the side ef-
fect of water intoxication. In most trials response to desmo-
pressin (defined as >50% reduction of the number of wet
nights) was noted in 60% to 70% of patients.81 Another phar-
macologic option is the tricyclic antidepressant imipramine
(Tofranil), which exerts a direct effect on bladder smooth mus-
cle and has sympathomimetic effects. Unlike anticholinergic
medications, which have a relatively short half-life, the blood
levels of imipramine build up over a period of several days.
The true beneficial effect of this medication may therefore
not become apparent for up to 2 weeks. Approximately
50% of enuretic children improve with imipramine, yet a sig-
nificant number will relapse after treatment is discontinued. It
has been reported that only 17% of children who were dry
during imipramine medication stay dry 6 months after cessa-
tion of the medication.82 The major drawback to imipramine
CHAPTER 116 DISORDERS OF BLADDER FUNCTION 1465
therapy is its cardiotoxic side effects, even in therapeutic
doses.83 Although anticholinergic medications should not
be expected to be effective in true isolated nocturnal enuresis,
this medication can occasionally be of help.
Alarm therapy is the most effective method for treating noc-
turnal enuresis. A meta-analysis of the world literature
revealed an average success rate of 68%.72 Efficacy is reported
to increase with duration of therapy to as high as 90% by
6 months.84 Although most children will not awaken, they
will stop emptying the bladder at the onset of the alarm. It
therefore requires that a parent assist the child in awakening
and proceeding to the toilet to finish voiding. Once the under-
garments have been changed, the alarm should be reset. The
exact mechanisms of alarm treatment are not known, although
it is clearly an operant type of behavioral approach.
The complete reference list is available online at www.
expertconsult.com.

CHAPTER 117
Reconstruction of
the Bladder and
Bladder Outlet
Eugene Minevich and Curtis A. Sheldon
Effective surgical management of the abnormal bladder re-
quires a thorough understanding of both normal micturition
and dysfunctional voiding (see Chapter 116). Urinary conti-
nence is a complex balance between intravesical pressure
and bladder outlet resistance. Several variables determine
intravesical pressure, only one of which is compliance, which
reflects both muscular tone and interstitial elasticity. This mea-
sure is not static but rather varies with bladder capacity. Con-
sequently, urine output, postvoid residual volume, and
frequency of bladder emptying determine the working com-
pliance range of the bladder. Pathologic alterations (e.g., un-
inhibited detrusor contraction or interstitial fibrosis) may
markedly alter compliance and resultant intravesical pressure
characteristics.
Bladder outlet resistance also depends on several variables
and must be highly coordinated in order to prevent inconti-
nence but allow voiding to occur. Reflex contraction of the
sphincteric mechanism helps prevent incontinence after sud-
den increases in intra-abdominal pressure. Conversely, blad-
der outlet sphincteric mechanisms reflexively relax at the
time of detrusor contraction. At this point, volitional voiding
ensues. In addition, voluntary contraction of the striated mus-
cle of the urinary sphincter adds supplementary support to
prevent incontinence. In some circumstances, however, reflex
contraction of the bladder sphincteric mechanism or inappro-
priate voluntary contraction (e.g., dysfunctional voiding) can
result in dangerously high intravesical pressures. Furthermore,
the bladder outlet sphincteric mechanism may fail to relax dur-
ing attempts at voiding, a condition known as detrusor-sphincter
dyssynergy. Pathologic bladder outlet resistance may be caused
by congenital anatomic obstruction (e.g., posterior urethral
valves) or acquired lesions (e.g., stricture).
The structural consequences of unbalanced voiding are
highly variable and ominous. Bladder consequences include
hypertrophy of detrusor musculature with trabeculation. Sac-
culation and diverticula may develop, leading to urinary tract
infection (UTI) and urolithiasis. Interstitial fibrosis may also
occur and cause a progressive loss of bladder compliance.
Of particular concern are the consequences of high bladder
pressure on renal function. Even without vesicoureteral re-
flux, a close correlation exists between intrapelvic renal pres-
sure and intravesical pressure.1 Renal pelvic pressures greater
than 40 cm of H2O deform the renal papillae, which may distort
the orientation of the tubules draining into the papillae and
result in intrarenal reflux.2 This increases the vulnerability of
the kidney to pyelonephritis.
In patients with neurogenic bladder (such as those with
myelodysplasia), bladder pressure and renal prognosis are
closely correlated. Spontaneous urine leakage through the
bladder outlet mechanism at a bladder pressure greater than
40 cm H2O has been associated with poor prognosis for the
upper urinary tract.3 A similar correlation with the maximal
urethral pressure on urodynamic testing has also been demon-
strated.4 Furthermore, elevated intravesical pressure has been
shown to adversely affect the renal allografts.5
Otherwise healthy children with increased intravesical
pressure have an increased risk for UTI and vesicoureteral re-
flux (VUR).6 Anticholinergic medications, which reduce intra-
vesical pressure, can dramatically alleviate reflux and UTI.7,8
The surgeon must realize that the cause of bladder injury
cannot be reliably determined from end-stage abnormalities
because the ability of various primary diseases to cause
bladder injury overlaps considerably. Secondary changes
due to chronic retention, infection, stones, or surgery also
confound the issue. Therefore one cannot predict the degree
of bladder function on the basis of knowledge of such pri-
mary diseases as myelodysplasia, posterior urethral valves,
prune-belly syndrome, and imperforate anus. Measurement
of physiologic properties of the bladder by formal urody-
namic investigation is essential in planning any major surgical
intervention.
Although it is accepted that bladder dysfunction can dam-
age the kidney, it is less well recognized that renal disease can
contribute to bladder injury. Polyuria caused by primary renal
disease can substantially alter bladder function. It can result in
a hypertensive bladder with extensive detrusor hypertrophy
or, if decompensated, a massively dilated, flaccid, hypotonic
bladder that cannot empty completely. These effects can also
accompany secondary renal disease (e.g., polyuria accompa-
nying posterior urethral valves). Such lesions injure the renal
medulla and lead to a loss of concentrating capacity; the
resulting polyuria may further damage the bladder.
1467
1468 PART VIII GENITOURINARY DISORDERS
Bladder outlet obstruction is a major factor in progressive
bladder and kidney injury. Structural lesions such as posterior
urethral valves or urethral strictures are particularly problem-
atic. Also important is detrusor-sphincter dyssynergy. This
condition may be volitional or nonvolitional. With the latter,
the sphincteric periurethral muscle inappropriately contracts
(rather than relaxes) during detrusor contraction. In the for-
mer, the child closes the bladder outlet to prevent inconti-
nence that results from uninhibited detrusor activity or to
delay the need for voiding when the bladder has become over-
filled. Such activity may also be used to prevent painful void-
ing. This activity is often manifested clinically by the
occurrence of Vincent curtsy, in which the child sits on a foot,
squats, or squeezes the legs together to help prevent voiding.
The “valve bladder” best exemplifies the interplay of these
factors (Fig. 117-1). Even after valve resection, the bladder
may still operate at high pressures because of residual bladder
wall changes such as altered compliance and uninhibited
detrusor contractions. The bladder may not empty completely,
and substantial VUR may be present. During voiding, urine is
emitted through the urethra but also refluxes into the kidneys.
When the patient completes voiding, this urine immediately
drains back into the bladder. This drainage, along with
elevated postvoid residual urine volume caused by abnormal
bladder function, forces the bladder to continuously work
within a range that results in high storage pressure with
subsequent adverse effects on the kidneys.
Recent advances in surgical technique, the successful appli-
cation of intermittent catheterization to the reconstructed uri-
nary tract, and the lessons learned from the pioneering work
on urinary undiversion9 allow even the most anatomically dev-
astated children to be reconstructed for continence, as well as
preservation of renal function. Such reconstructive principles
may now be applied to virtually all urinary tract anomalies
with a good expectation of success. Reconstructive options
are presently available even for children with end-stage renal
Infection
Renal injury
Polyuria
Early refill
Reflux Immediate
refill
Postvoid
residual
Voluntary
dyssynergy
FIGURE 117-1 Pathophysiology of the valve bladder. (From Minevich E, Sheldon CA: Structural disorders of the bladder, augmentation. In Grosfeld JL,
O’Neill J, Fonkalsrud E, Coran A [eds]: Pediatric Surgery. St. Louis, Mosby, 2006.)
disease for whom renal transplantation will ultimately be re-
quired. Reconstruction involves many challenges in achieving
the goals of low-pressure storage of urine and complete blad-
der emptying. Most frequently, such reconstruction is neces-
sary for treatment of bladder exstrophy; posterior urethral
valves; and neurogenic bladder related to myelomeningocele,
other sacral dysraphisms, or the VATER complex (vertebral de-
fects, imperforate anus, tracheoesophageal fistula with esopha-
geal atresia, and radial and renal dysplasia). Other disease states
necessitating reconstruction include urogenital sinus and cloa-
cal anomalies, cloacal exstrophy, and bilateral single ectopic
ureters. All of these are commonly associated with abnormal
storage pressures of urine or problems with emptying.
Disorders of Bladder and Urethra
------------------------------------------------------------------------------------------------------------------------------------------------
ANATOMIC DISORDERS OF BLADDER
AND URETHRA
Primary structural abnormalities are important and are best
exemplified by posterior urethral valves (Fig. 117-2), the most
common cause of structural lower urinary tract obstruction in
male infants. As a consequence of obstruction, the proximal
urethra is elongated and dilated, and the bladder neck is rel-
atively narrowed because of secondary detrusor hypertrophy.
The distal urethra is of normal caliber, resulting in an abrupt
transition between the dilated proximal urethra and the thin
anterior urethra. The bladder exhibits detrusor hypertrophy
and is irregular because of trabeculation, sacculation, and
the presence of diverticula. The upper urinary tracts typically
show severe hydronephrosis with or without vesicoureteral
reflux. Renal impairment is frequent and often presents from
birth; recovery is incomplete even after valve ablation.
Most patients are recognized during infancy, and more than
two thirds are identified within the first year of life. With the
Hypertrophy,
irritability
Bladder
pressure
Working
volume
Pressure
Volume
Residual
obstruction
 CHAPTER 117
FIGURE 117-2 Radiographic appearance of posterior urethral valves.
(From Minevich E, Sheldon CA: Structural disorders of the bladder, augmen-
tation. Grosfeld JL, O’Neill J, Fonkalsrud E, Coran A [eds]: Pediatric Surgery.
St. Louis, Mosby, 2006.)
advent of perinatal ultrasonography, in utero diagnosis has be-
come common. A newborn may have an enlarged palpable
bladder or kidneys as a result of hydroureteronephrosis. An
infant may have UTI or failure to thrive because of azotemia,
whereas in older children the initial manifestation may be in-
continence. Diagnostic evaluation includes ultrasonography
of the upper urinary tract and radiographic voiding cystoure-
throgram (VCUG). Renal function can be evaluated by renal
scanning. Urodynamic investigation may later be indicated
if hydronephrosis, reflux, azotemia, or incontinence persists
in a child whose valves have been ablated successfully.
Infants should initially be stabilized by catheter drainage of
the bladder, intravenous resuscitation, and respiratory sup-
port when needed. Urine is drained with a 5- or 8-French in-
fant feeding tube because larger catheters or catheters with
balloons may cause intense detrusor contraction, which
may impede urine flow through the ureterovesical junction.
Most patients are subsequently treated by valve ablation. In
premature infants with a urethra too small for endoscopic ful-
guration, either antegrade ablation or a vesicostomy may be
created; with current miniaturized cystoscopes, however,
these procedures are rarely necessary. Patients with massive
reflux, particularly when accompanied by azotemia, are also
candidates for vesicostomy.
The rare infant with profound azotemia who is unrespon-
sive to catheter drainage may benefit from upper urinary tract
diversion by way of cutaneous pyelostomy. Such low-pressure
drainage may allow sufficient return of renal function so that
dialysis or renal transplantation can be deferred until the child
is older. Both cutaneous vesicostomies and cutaneous pyelos-
tomies are readily reversed. It is important to avoid fulguration
of valves in a dry urethra (e.g., in the patient who has under-
gone diversion by vesicostomy or pyelostomy), which may
result in urethral stricture.
Anterior urethral valves, Cowper duct cysts, urethral
polyps, urethral duplication (Fig. 117-3), and urethral diver-
ticula can produce structural changes in the bladder that are
related to obstruction. Acquired obstructive lesions include
stricture and meatal stenosis. Exstrophic bladder abnormality,
epispadias, and hypospadias are other structural lesions,
discussed elsewhere in the textbook.
RECONSTRUCTION OF THE BLADDER AND BLADDER OUTLET 1469
FIGURE 117-3 Urethral duplication. (From Sheldon CA, Bukowski TP:
Male external genitalia. In Rowe MI, O’Neill J, Grosfeld JL, et al [eds]: Essen-
tials of Pediatric Surgery. St Louis, 1995, Mosby-Year Book, p 779.)
NEUROGENIC DISORDERS OF THE BLADDER
AND URETHRAL FUNCTION
The surgeon is commonly faced with a primary abnormality of
bladder innervation (a neurogenic bladder). Myelodysplasia,
an open dystrophic state, is particularly common. Myelome-
ningocele, in which neural tissue and the meninges protrude
beyond the confines of the vertebral canal, is the most com-
mon defect. The neurologic effect of these entities on the lower
urinary tract varies and cannot be predicted by the observed
level of the anomaly; urodynamic testing is, therefore, essen-
tial. The risk for upper urinary tract deterioration can be pre-
dicted urodynamically. Both a leak point pressure exceeding
40 cm H2O3 and the presence of dyssynergy10 have been
clearly shown to be associated with poor prognosis for the up-
per urinary tract (Figs. 117-4 and 117-5). The risk for deteri-
oration exceeds 70% in the presence of dyssynergy;
sphincteric synergy or sphincteric denervation is associated
with an incidence of deterioration of 15% and 25%, respec-
tively. Unfortunately, the neurologic defect is often a dynamic
one in which dyssynergy, if not present initially, may develop
over time. The development of dyssynergy is often related to
tethering of the spinal cord, which may result in spinal injury
or spinal root injury as the child grows. The greatest risk oc-
curs within the first 2 years of life, especially the first year, but
such injury can occur throughout childhood.
Occult (closed) spinal dysraphisms include tethered cords,
intradural lipomas, dermoid cysts or sinuses, diastematomye-
lia, and cauda equina tumors. These children commonly have
urologic manifestations as UTIs and incontinence. Fortunately,
however, cutaneous manifestations are obvious in most
patients. Such lesions as cutaneous dimpling, the presence of
a skin tag, a subcutaneous lipoma or a patch of hair over the bony
sacrum, dermal vascular malformation, and pigmentation may
1470 PART VIII GENITOURINARY DISORDERS

Vesicoureteral 81% COMBINED ANATOMIC AND NEUROGENIC

80 reflux
DISORDERS
Ureteral
dilatation 68%
A third group of infants requiring bladder reconstruction are
those with both functional and structural bladder abnormal-
60 ities such as imperforate anus or cloaca. In these cases, a full
anatomic assessment must be done early in life, which at a
minimum requires renal ultrasonography, radiographic
Percent

40
20
10%
0%
<40 cm H2O >40 cm H2O
FIGURE 117-4 Risk for upper urinary tract injury as a function of urethral
opening pressure. (From McGuire EJ, Woodside JR, Borden TA, Weiss RM:
Prognostic value of urodynamic testing in myelodysplastic patients. J Urol
1981;126:208.)
indicate an underlying spinal disorder. The presence of these
findings in any child with urinary incontinence or other evi-
dence of bladder dysfunction requires exclusion of occult spinal
diastrophism. These diagnoses may be confirmed by urody-
namic investigation and spinal magnetic resonance imaging
(MRI) or, in the first 6 months of life, ultrasonography of the
spinal cord. Abnormal urinary tract function occurs in approx-
imately 40% of patients.11
Another important cause of neurogenic bladder is sacral
agenesis, often seen in imperforate anus or in the infant of a
diabetic mother. This condition is characterized by congenital
absence of all or part of two or more sacral segments. The le-
sion may be suggested on physical examination by the pres-
ence of an abnormal gluteal cleft or detection of an
incomplete sacrum on direct palpation. The diagnosis may
be confirmed by anteroposterior and lateral lumbosacral
spine radiographs. Approximately 75% of such patients have
abnormal urinary tract function.12
p<0.004
Deterioration
No deterioration
60
No. of children
VCUG, and ultrasonographic assessment of the spinal cord.
The combination of both anatomic and functional abnormal-
ities creates a high risk for substantial urologic morbidity.
A study by McLorie and colleagues involving 484 consecutive
patients revealed a high risk for end-stage renal disease.13
Nonfistula genitourinary anomalies were encountered in
60% of patients with high lesions and 20% of those with
low lesions. Substantial upper urinary tract abnormalities
were encountered in more than one third of cases, and
bilateral upper urinary tract abnormalities occurred in 14%.
Our review of 90 consecutive cases of imperforate anus
showed an 18% incidence of significant neurovesical dysfunc-
tion.14 The greatest risk was encountered in patients with a
high imperforate anus. However, neurovesical dysfunction
was also encountered in some patients with low lesions. In
many patients, a tethered spinal cord must be addressed be-
cause of the risk for progressive injury to innervation of the
bladder, urethra, anorectum, and lower extremities.
The importance of these anomalies was demonstrated in a
review of 23 of our patients with imperforate anus recon-
structed for either upper urinary tract preservation or achieve-
ment of continence.15 As shown in Figure 117-6, 9% of
patients presented with end-stage renal disease, 65% had sig-
nificant renal abnormalities, and 57% had vesicoureteral re-
flux. Ureteral anomalies were encountered in 30% of
patients, and neurovesical dysfunction was encountered in
70%. A wide spectrum of urologic reconstructive procedures
was required for these patients (Fig. 117-7). Of particular
note, 43% required ureteral reimplantation, 43% required
bladder augmentation, 35% required a Mitrofanoff neoure-
thra, and 22% required bladder neck reconstruction. The
urethra in boys with imperforate anus may be difficult to cath-
eterize because of urethral irregularity related to the congenital
rectourethral fistula. This, along with intact urethral sensa-
tion, may necessitate the creation of a Mitrofanoff neourethra
to assist intermittent catheterization.
Other lesions, described elsewhere in this text, may also
require bladder reconstruction. Patients with urogenital
sinus anomalies, in which the bladder and vagina drain
through a common channel, often have associated neurove-

45
sical dysfunction. In addition to urethral and vaginal recon-
struction, these patients may have an incompetent bladder
outlet, necessitating reconstruction. As with other conditions
30 associated with neurovesical dysfunction, bladder augmenta-
tion may be required. Patients with exstrophy or cloacal
15
exstrophy characteristically require complex urinary tract
reconstruction. Patients with the prune-belly syndrome can
usually be managed medically but sometimes require surgical
0 intervention. These patients typically have a hypotonic blad-
Dyssynergy Synergy Absent
activity der that operates at low pressure but empties incompletely, is
highly irregular, and is prone to infection. Some patients,
FIGURE 117-5 Urinary tract deterioration as a function of dyssynergia,
synergy, and absent sphincteric activity. (From Bauer SB: Early evaluation
however, lose capacity and compliance over time. Conse-
and management of children with spina bifida. In King LR [ed]: Urologic Sur- quently, urodynamic investigation is often necessary to
gery in Neonates and Young Infants. Philadelphia, WB Saunders, 1988, p 256.) facilitate therapy.
 CHAPTER 117 RECONSTRUCTION OF THE BLADDER AND BLADDER OUTLET 1471

25

20

70%
65%
15
Number of Patients

Ectopia 57%
Mal-
rotation

10
Dys-
plasia
39%
30% Isolated
vaginal 26%
5 UPJ
Mega- 17% Penile
Agenesis 9% ureter Classic
Urethral
Ectopic Cloacal Stricture
0
r

it.
D

VD

a
r.
al

te
R

ac
R

st

en
en

re
VU
ES

Ex

lo
G
U
R

C
KIDNEY COLLECTING BLADDER PERINEUM
SYSTEM
FIGURE 117-6 Genitourinary anomalies associated with imperforate anus. ESRD, End-stage renal disease; NVD, neurovesical dysfunction; UPJ, uretero-
pelvic junction; VUR, vesicoureteral reflux. (From Sheldon CA, Gilbert A, Lewis AG: Surgical implications of genitourinary tract anomalies in patients with
imperforate anus. J Urol 1994;152:196.)

Reconstructive Philosophy pressure or precipitating incontinence. The second compo-

------------------------------------------------------------------------------------------------------------------------------------------------
nent is adequate bladder outlet resistance to maintain urinary
The goals of urinary reconstruction are to preserve the upper continence. Third, there must be a convenient, reliable mech-
urinary tract, attain socially acceptable continence, and max- anism for bladder (reservoir) emptying. Ideally, this should be
imize the child’s ease of care and potential for self-care. Pa- achieved by spontaneous voiding; otherwise, intermittent
tients at risk for upper urinary tract injury are evaluated catheterization is necessary. The native urethra may be an ac-
during the newborn period, and therapy to protect the upper ceptable conduit for this maneuver, although should catheter-
urinary tracts is initiated at that time. Urinary incontinence is ization of it prove excessively difficult or uncomfortable
evaluated and treated just before school age in order to make (preventing patient compliance), an alternative catheterizable
social integration easier. In general, an attempt is made to conduit may be necessary. Finally, unobstructed and nonre-
manage the urinary tract with medical therapy before sub- fluxing sterile upper tract drainage of urine into the bladder
jecting the child to surgical intervention. Such therapy may (reservoir) is desirable in order to protect the upper tracts.
include intermittent catheterization to facilitate bladder emp-
tying, anticholinergic therapy to diminish intravesical pres-
sure, and a-adrenergic agents to enhance bladder outlet Compensating for Inadequate
resistance. Surgical therapy is undertaken only if medical ther-
apy is unsuccessful. When contemplating urinary tract recon-
Bladder Capacity or Compliance
------------------------------------------------------------------------------------------------------------------------------------------------

struction, meticulous preoperative evaluation is critical, and it

PHYSIOLOGIC CONSIDERATIONS
is essential to tailor the reconstruction to the individual needs
of the patient. Bladder capacity and compliance are increased by bladder
Four components of balanced urinary tract function must augmentation. Indicators that augmentation is necessary in-
be achieved to ensure long-term success with urinary recon- clude clinical symptoms, such as incontinence caused by
struction.16 The first component is that of adequate bladder bladder dysfunction unresponsive to medical therapy, and up-
(reservoir) capacity and sufficient compliance to provide per urinary tract deterioration due to inadequate low-pressure
low-pressure storage. Optimal bladder capacity should allow storage volume. Also suggestive is a measured bladder capac-
a 4-hour catheterization or voiding interval during the day ity that is significantly less than that expected for the patient’s
and an 8-hour interval at night without reaching excessive age. Caution must be exercised because with an incompetent
1472 PART VIII GENITOURINARY DISORDERS

25

20

70%
15
Number of Patients

43% 43%
10
Tapered Colon
Ileal 39% 30%
35%
Appen- Hypospadias
dix Epispadias
22%
5 Urethroplasty
17% Gastr. Ileal Closure 17%
13% Kropp 13%
Classic Vag.
Ureter Y-D-L
4% Cloac.
0
.

s.

p.

h.

r.

c.

VU

it.
hr

st

gm

itr

ec

st
an

si

en
m

at
ro

M
ep

Ex

R
Ve
ei

Au
Tr

G
U

BN

PS
N

KIDNEY COLLECTING BLADDER PERINEUM

SYSTEM
FIGURE 117-7 Surgical procedures performed in patients with imperforate anus. Augm., Augmentation; BNRec., bladder neck reconstruction; Cath., in-
termittent catheterization; Exstr., exstrophy; Gastr., gastrocystoplasty; Mitr., Mitrofanoff neourethra; Nephr., nephrectomy; Reimp., reimplantation; Trans.,
renal transplantation; Urost., ureterostomy; Vag., vaginoplasty; Vesic., vesicostomy; Y-D-L, Young-Dees-Leadbetter procedure. (From Sheldon CA, Gilbert A,
Lewis AG: Surgical implications of genitourinary tract anomalies in patients with imperforate anus. J Urol 1994;152:197.)

bladder outlet, the bladder may drain at low pressure, making
determination of functional bladder volume more difficult.
Performing cystometrography with a Foley catheter balloon
that occludes the bladder neck may provide more reliable
data on functional bladder capacity. The generation of pres-
sure exceeding 35 to 40 cm H2O with urine volumes equal
to those anticipated during 4 hours of urine production dur-
ing the day or 8 hours of urine production during the night
during maximal medical therapy further suggests that bladder
augmentation should be considered.
Several important reconstructive concepts are pertinent to
bladder augmentation. The first regards management of the re-
cipient bladder. If bowel augmentation is performed, with the
detrusor essentially left intact to generate high pressure, the
former will act urodynamically as a capacious diverticulum.17
The problem can be avoided by an extended sagittal opening
of the bladder from the level of the bladder neck anteriorly
to the trigone posteriorly (“clam cystoplasty”). Essentially, this
is a reconfiguration of the bladder from a sphere into a flat
plate so that the detrusor is no longer capable of generating a
contraction that produces a significant pressure elevation.18,19
Just as pressure generated by the bladder detrusor is an
important contributor to the pressure generated in an aug-
mented urinary reservoir, so also is the pressure generated
by the tubularized bowel segment itself. With peristaltic con-
tractions, pressures ranging from 60 to 100 cm H2O may be
encountered.20 This observation led Kock to develop his
concept of turning the intact bowel into a reservoir incapable
of effective peristalsis by creating a “pouch.”21 Opening the
bowel along its antimesenteric border and closing it with dis-
ruption of the circular muscle (“detubularization”) inhibits
peristalsis. Once unable to undergo peristalsis, the reservoir
dilates and stores urine at a low pressure (Fig. 117-8).22,23
Additionally, there is a significant increase in the geometric
capacity of the intestinal segment.23 A third important con-
cept is accommodation (Fig. 117-9). It is well known that
the reconstructed bladder will gradually enlarge over time.
At a constant pressure, a structure with a larger radius will
accommodate a greater volume—again, an advantage of
detubularized bowel segments.
Several considerations must be entertained when choosing
an augmentation donor site. Anatomic considerations such as
mobility of the blood supply favor the use of ileum, sigmoid,
the ileocecal region, and the greater curvature of the stomach.
The ability to implant a ureter or a Mitrofanoff neourethra may
also be a consideration. Additionally, it may be important to
avoid the peritoneal cavity so that the option of performing
peritoneal dialysis or placement of a ventricular peritoneal
shunt is preserved. Such considerations favor the use of
ureteral augmentation or autoaugmentation.
The choice of augmentation donor site may be limited by
the patient’s primary disease. Patients with short gut may
not tolerate a loss of the ileocecal region or a significant length
of ileum. Patients with borderline fecal continence (such as
 CHAPTER 117 RECONSTRUCTION OF THE BLADDER AND BLADDER OUTLET 1473

100 of ileum 20 to 40 cm in length is isolated and incised along

Cecal its antimesenteric border. It is reconfigured as a “cup patch”
reservoir and anastomosed to the bladder plate with running 3-0
50
pressure Vicryl suture (inner-layer interlocking). Bowel continuity
cm H2O is reestablished by end-to-end anastomosis. Advantages of
0 this procedure include its technical simplicity. However,
100 200 300 400 500 antirefluxing implantation of the ureter or Mitrofanoff neour-
A Volume ml ethra into the ileal segment is less reliable than implantation
100 into the native bladder or other augmentation donor
Cecal segments.
reservoir
50 Other techniques of bladder augmentation or replace-
pressure ment using small bowel include the Camey procedure
cm H2O
(Fig. 117-11),26 Kock pouch (Fig. 117-12),21 and ileal neo-
0 bladder.27,28 These procedures are less successful in achiev-
100 200 300 400 500
ing continence and have a significant rate of complications
B Volume ml
and reoperation.
FIGURE 117-8 Urodynamics of cecal reservoir. A, A tubular bowel pro-
duces high-pressure peristaltic waves. B, A cup patch bowel with disrupted
peristalsis stores large volume without pressure rise. (From Goldwasser
HR, Webster GD: Augmentation and substitution enterocystoplasty. J Urol ILEOCECAL SEGMENT PROCEDURES
1986;135:221.)
The ileocecal bowel segment has been favored by urologists
for bladder reconstruction because of the natural configura-
those with imperforate anus or myelodysplasia) may not tol- tion of the cecum, which gives it the appearance of an ideal
erate loss of the ileocecal valve or the water reabsorptive substitute for the bladder.29 This technique is technically sim-
capacity of the right colon. Metabolic consequences may as- ple to perform and also has the major advantage of allowing
sume an overriding influence: The risk of absorptive acidosis antirefluxing implantation of massively dilated ureters into
and growth retardation, which may be exacerbated by chronic the ileum. Reflux is prevented by the ileocecal valve, bolstered
renal insufficiency, may favor the use of autoaugmentation, by intussusception. Unfortunately, the intussusception antire-
ureteral augmentation, or gastrocystoplasty techniques. Be- flux mechanism is inherently unstable. Consequently, various
cause the reconstruction must be tailored to the individual surgical modifications of the ileocecal valve have been intro-
needs of the patient, the surgeon must be familiar with a wide duced in an effort to try to lessen the incidence of reflux.
variety of reconstructive alternatives and prepare the patient Of greater impact is the fact that loss of the ileocecal valve from
accordingly including bowel preparation even when gastro- fecal continuity risks devastating fecal incontinence in patients
cystoplasty, autoaugmentation, or ureteral augmentation is with marginal anorectal continence (e.g., myelomeningocele
anticipated. or the VATER complex).
When the cecal or ileocecal segments have been used intact
for bladder augmentation, nighttime incontinence has been a
SMALL BOWEL PROCEDURES significant problem in most series.30 This problem most likely
reflects peristaltic waves in the intact bowel segment because
Ileocystoplasty (Fig. 117-10) is one of the most commonly enuresis is rare when the cup patch technique is used. Other
used bladder augmentation techniques.24,25 The bladder is continent diversions using the ileocecal valve have included
prepared by a “clam” cystoplasty incision, and a segment the Maintz pouch,31 the Penn pouch,32 the Indiana pouch,33
and the Florida pouch.34 Of these techniques, the Indiana
pouch has been applied most frequently in pediatric practice
and has met with variable results.35

GF/cm
T= 50 LARGE BOWEL PROCEDURES
cm
D= 2
P = 100 Mathisen36 reported sigmoid augmentation of the bladder
P =100 performed in a manner similar to that described for ileocysto-
plasty earlier. This procedure, too, is relatively simple to per-
D =12 cm form but allows a better antirefluxing implantation of the
ureter or Mitrofanoff neourethra into the tenia. This technique
did not appear to differ from other bowel segments with re-
spect to ability to empty, infections, or surgical complica-
tions.37 In patients who have previously undergone
reconstruction for imperforate anus, this procedure may inter-
rupt blood supply to the rectum because the anorectum de-
FIGURE 117-9 Inflated surgeon’s glove illustrates the LaPlace relation- pends on a descending blood supply. Positive experiences
ship. Although pressure (P) is equal throughout, tension (T) is greater in
portion with greater diameter (D). (From Hinman F: Selection of intestinal
with construction of a colonic neobladder have been
segments for bladder substitution: Physical and physiological characteris- reported,38,39 although nocturnal incontinence continues to
tics. J Urol 1988;139:522.) be a problem in up to 33% of patients.39
1474 PART VIII GENITOURINARY DISORDERS

C
FIGURE 117-10 Technique of “clam” ileocystoplasty. A, Small, high-pressure bladder. B, Bladder has been prepared by a long, longitudinal incision, and
a segment of ileum has been detubularized and fashioned into a cup patch. C, Cup patch is anastomosed to bladder. (From Minevich E, Sheldon CA:
Urinary tract reconstruction for continence and renal preservation. In Ziegler MM, Azizkhan RG, Weber TR, et al [eds]: Operative Pediatric Surgery. Norwalk,
Conn, Appleton & Lange, 2003, p 915.)

A B
FIGURE 117-11 Camey enterocystoplasty: A, A 35- to 40-cm segment of intact ileum is anastomosed to the urethral stump to create a continent in-
testinal reservoir. B, Ureters are sutured into a 3- to 4-cm trough in the bowel mucosa in each limb of the reservoir to create effective antireflux flap valves.
(From Minevich E, Sheldon CA: Urinary tract reconstruction for continence and renal preservation. In Ziegler MM, Azizkhan RG, Weber TR, et al [eds]:
Operative Pediatric Surgery. Norwalk, Conn, Appleton & Lange, 2003, p 916.)
 CHAPTER 117 RECONSTRUCTION OF THE BLADDER AND BLADDER OUTLET 1475

FIGURE 117-12 Kock pouch. A 70-cm segment of ileum is reformed into A

a peristaltic pouch with two nipple valves. The most recent modification
involves fixation of a nipple valve to the reservoir wall to change it to a
fixed flap valve. (From Minevich E, Sheldon CA: Urinary tract reconstruc-
tion for continence and renal preservation. In Ziegler MM, Azizkhan RG,
Weber TR, et al [eds]: Operative Pediatric Surgery. Norwalk, Conn, Appleton
& Lange, 2003, p 917.)

GASTRIC SEGMENT PROCEDURES

The work of Mitchell and colleagues40 ushered in the
modern era of the use of stomach in urinary reconstruction
(Fig. 117-13). They demonstrated gastrocystoplasty to be
highly successful, versatile, and well tolerated even in the face
of azotemia. This procedure is associated with a reduced risk
for infection, mucus production, and urinary stone formation
and provides excellent compliance characteristics. Implanta-
tion of a ureter or Mitrofanoff neourethra is technically easy.
Our long-term follow-up with gastrocystoplasty or gastric
neobladder reveals a continence rate of 91%, stable renal func-
tion in all patients, and upper tract deterioration in only one
patient who became noncompliant with intermittent catheter-
ization.41,42 Additionally, gastrocystoplasty has proved to be
an excellent alternative for patients with end-stage renal dis-
ease facing subsequent transplantation.41,42 The gastric or B
composite neobladder has been successfully used for recon- FIGURE 117-13 Gastrocystoplasty. A, Development of right gastroepi-
struction43 with the native urethra,44 the orthotopic ureteral ploic pedicle and isolation of wedge of gastric fundus. B, Mobilization
of right gastroepiploic pedicle through retroperitoneal plane into augmen-
neourethra,41,45 and the orthotopic appendiceal neoure- tation position. The stomach is closed. (From Minevich E, Sheldon CA: Uri-
thra.41 The major complication and limiting application of nary tract reconstruction for continence and renal preservation. In Ziegler
the gastrocystoplasty is the development of hematuria-dysuria MM, Azizkhan RG, Weber TR, et al [eds]: Operative Pediatric Surgery.
syndrome, which can be refractory to medical management Norwalk, Conn, Appleton & Lange, 2003, p 918.)
and may require reversal of the augmentation.
sufficiently dilated ureter is rarely available for this procedure.
In addition, some reports indicate that the long-term ability of
URETERAL AUGMENTATION ureteral augmentation to maintain low storage pressure and
AND AUTOAUGMENTATION adequate capacity may be suspect.50

Ureteral augmentation46 (Fig. 117-14) and autoaugmentation47

(Fig. 117-15) hold great promise because of their ability to prevent
absorptive metabolic disorders and to be performed via an entirely
Procedures to Correct Deficient
extraperitoneal approach.48 However, these procedures are more Bladder Outlet Resistance
likely to fail to attain adequate capacity and compliance because of ------------------------------------------------------------------------------------------------------------------------------------------------

an inherent restriction in the availability of surface area.49 In Urinary continence is maintained by a complex relationship
case of ureterocystoplasty the ureter is made available either between bladder outlet resistance and pressure. In order to
by nephrectomy or transureteroureterostomy. Unfortunately, a maintain dryness, the bladder outlet resistance must exceed
1476 PART VIII GENITOURINARY DISORDERS

A C

FIGURE 117-14 Operative stages of ureteral bladder augmentation. A, Normal blood supply to ureter. B, Ureteral detubularization following mobiliza-
tion. C, Reconfiguration of ureter into U-shaped patch. D, Anastomosing ureteral patching to native bivalved bladder. (From Minevich E, Sheldon CA:
Urinary tract reconstruction for continence and renal preservation. In Ziegler MM, Azizkhan RG, Weber TR, et al [eds]: Operative Pediatric Surgery. Norwalk,
Conn, Appleton & Lange, 2003, p 919.)

intravesical pressure not only at rest but also during changes
in posture, coughing, sneezing, and straining. Bladder outlet
reconstruction is necessary in patients with incontinence despite
low-pressure storage of urine in the bladder. The bladder outlet
in patients who have a high-pressure or low-capacity bladder is
more difficult to assess because incontinence may be caused by
these storage characteristics alone. The likelihood of a compe-
tent sphincteric mechanism in this setting is suggested by a
stress leak point pressure greater than 100 cm H2O.51 Urethral
pressure profilometry and static leak point pressure have not
proven to be reliable, independent indicators of sphincteric
competence. Most surgical interventions designed for the
 CHAPTER 117
A B
C
FIGURE 117-15 Autoaugmentation. A, Detrusor incised. B, Detrusor stripped from intact bladder epithelium. C, Epithelium bulges with bladder filling.
(From Minevich E, Sheldon CA: Urinary tract reconstruction for continence and renal preservation. In Ziegler MM, Azizkhan RG, Weber TR, et al [eds]:
Operative Pediatric Surgery. Norwalk, Conn, Appleton & Lange, 2003, p 920.)
achievement of continence include procedures to lengthen the
urethra, suspend the bladder neck, or compress or completely
close the urethra.
Efforts to proximally lengthen the existing urethra through
tubularization of the posterior detrusor grew out of the early
work of Young.52 In the procedure, which was later modified
by Dees53 and Leadbetter,54 the urethra is lengthened by tubu-
larization of a long (4- to 5-cm) segment of the posterior blad-
der wall. Two triangular sections of urothelium are excised,
and the resultant urothelial strip is approximated over a
small catheter (8- or 10-French)55 to fashion the neourethra.
The adjacent detrusor is approximated to itself over this
mucosal tube to add muscular support (Fig. 117-16). Anterior
urethral suspension is usually performed by suture fixation
and may be supplemented by the placement of a compressive
fascial sling. Several variations in technique have been
described.56,57
The specific goals of this procedure are to achieve conti-
nence and allow spontaneous voiding if the detrusor can con-
tract. Intermittent catheterization must also be attainable if the
detrusor cannot effectively contract to empty the bladder. Ben-
efits of this technique are that it may allow spontaneous
RECONSTRUCTION OF THE BLADDER AND BLADDER OUTLET 1477
voiding, uses native tissue, and affords a “pop-off” mecha-
nism. The procedure does, however, reduce bladder capacity
and may make the urethra difficult to catheterize. Further,
catheterization may injure the continence mechanism and
result in reconstructive failure.
The placement of an adjuvant Mitrofanoff neourethra in
patients undergoing Young-Dees-Leadbetter bladder neck re-
construction allows a channel for intermittent catheterization,
which has been useful for all such reconstructions.58 With
time, as the patient learns to void through the reconstructed
urethra, the Mitrofanoff neourethra can be removed in a sim-
ple outpatient surgical procedure or, because it does not leak,
it can be left in situ.
The Kropp procedure (Fig. 117-17) involves the creation of
an extended length of urethra by tubularization of an anterior
strip of bladder wall that is left in continuity with the urethra.
This tube is then implanted suburothelially into the bladder.59
This procedure allows reliable continence and uses native tis-
sue. It can, however, lead to difficulty in catheterization. The
Pippi Salle modification,60 which uses an onlay detrusor strip
(with preservation of the posterior bladder strip), may permit
easier catheterization. Neither the Kropp tube nor the Salle
1478 PART VIII GENITOURINARY DISORDERS
A B
FIGURE 117-16 Young-Dees-Leadbetter bladder neck reconstruction. (From Sheldon CA, Bukowski TP: Bladder function. In Rowe MI, O’Neill JA, Jr.,
Grosfeld JL, et al, [eds]: Essentials of Pediatric Surgery. St Louis, Mosby-Year Book, 1995, p 742.)
onlay allows a “pop-off” mechanism; thus the potential for
spontaneous voiding is eliminated and bladder capacity is
reduced.
Another option for the creation of bladder outlet resistance
is the fascial sling. A free fascial graft is frequently used; the
graft may be placed transvaginally with an endoscopically
assisted technique61 or by open retrourethral dissection.62 Al-
ternatively, the fascial strip may be developed in such a way
that it remains attached to the ipsilateral pyramidalis
(Fig. 117-18).63 The latter approach is particularly applicable
to children who have a denervated but anatomically intact
sphincteric mechanism and require concomitant bladder aug-
mentation. These procedures are technically fast and relatively
simple to perform, involve native tissue, and are readily re-
versible. They can also be used as an adjunct to the Young-
Dees-Leadbetter procedure. Spontaneous voiding is possible
if angulation is not excessive, and a “pop-off” mechanism is
maintained. Excessive angulation may, however, interfere with
spontaneous voiding or catheterization.
The artificial urinary sphincter (Fig. 117-19) has been
shown to be effective for compressing the urethra and thus
contributes to bladder outlet competence. The most popular
model consists of a cuff placed around the bladder neck or
bulbar urethra, a reservoir placed intra-abdominally, and an
activating pressure bulb located in the scrotum or labia. Con-
trolled pressure is maintained in the cuff until the pump is
squeezed, transferring fluid from the cuff into the reservoir
balloon and permitting bladder emptying to take place.
A delay-fill resistor in the control mechanism provides 1 to
2 minutes of lowered intraurethral pressure before automatic
refilling of the cuff takes place from the reservoir balloon.
Pressure-regulating balloons of various pressure ranges are
available, and a 60 to 70 cm H2O balloon is generally selected
for pediatric reconstruction.64
Several series have documented the utility of the artificial
urinary sphincter in patients who have bladder augmenta-
tion.65–68 The advantage of this procedure is that it allows
spontaneous voiding, but there are several disadvantages.
Multiple mechanical problems have occurred in patients with
the artificial sphincter in place. The most common problems
have been fluid leaks from the cuff or tubing kinks requiring
surgical revision. The most serious complications are erosion
of the sphincter into the urethra or the development of infec-
tion around the cuff. The latter problems generally require re-
moval of the device. Consequently the artificial sphincter is
recommended primarily for those patients who have a chance
of maintaining spontaneous voiding.
Finally, surgical urethral closure may be necessary. This is
obviously a highly effective means of ensuring urethral conti-
nence but should be a last resort.69,70 Simple oversewing of
the mucosa at the level of the bladder neck is insufficient be-
cause recanalization occurs. Urethral division, which allows
muscular approximation (proximally and distally), eliminates
this risk. This procedure requires the placement of an alter-
nate catheterizable conduit as discussed later. The advantages
of this procedure are that it is definitive and technically sim-
ple. It does, however, place the patient at increased risk for
bladder calculi and is difficult to reverse. In addition, the lack
 CHAPTER 117 RECONSTRUCTION OF THE BLADDER AND BLADDER OUTLET 1479

A C

B D
FIGURE 117-17 The Kropp procedure. A detrusor tube is created (anterior shown, posterior tube also possible) and tunneled submucosally in the bladder
to create a competent flap valve. (From Minevich E, Sheldon CA: Urinary tract reconstruction for continence and renal preservation. In Ziegler MM, Azizkhan
RG, Weber TR, et al [eds]: Operative Pediatric Surgery. Norwalk, Conn, Appleton & Lange, 2003. p 925.)

of a “pop-off” mechanism increases the risk for bladder rup- that cannot be overcome by intravesical (intrareservoir) pres-
ture if the patient does not comply with regular catheterization sure. The success of these procedures in terms of continence
to empty the bladder. relies on attaining controlled reservoir-neourethra balance.
Neourethral resistance to reservoir outflow must be sufficient
to exceed both resting and intermittently elevated intravesical
Providing for Alternate pressure associated with gravity (upright posture), as well as
episodic additive intra-abdominal pressure spikes (coughing,
Continent Urine Drainage sneezing, straining, and sudden postural changes). The crea-
------------------------------------------------------------------------------------------------------------------------------------------------
tion of neourethral resistance must be complemented by low
The procedures discussed in the preceding section require the intravesical (intrareservoir) pressure. This may entail bladder
creation of a continent catheterizable conduit that connects augmentation or replacement by bowel and should include
the bladder to the skin of the abdominal wall or perineum. reconfiguration by detubularization. A large capacity is imper-
An alternative to catheterization of the native urethra is indi- ative, as is intermittent catheter drainage before the low com-
cated when a urethra is difficult to catheterize because of pliance portion of the reservoir’s pressure-volume curve is
(1) tortuosity from either previous urethral surgery or congen- entered.
ital irregularity or (2) physical limitations of the patient that Alternate techniques include those associated with a nipple
impede access to the native urethra. Procedures directed at valve (procedures involving the ileal-cecal junction or ileal-
urethral functional replacement are based on the creation of ileal intussusception) or those associated with a flap valve
a tubular conduit of sufficient length that is exposed to exter- (Fig. 117-20). The nipple valve is inherently unstable because
nal compressive forces, thereby providing outlet resistance wall tension causes distraction of the base of this continence
1480 PART VIII GENITOURINARY DISORDERS

mechanism over time, often resulting in a loss of effective- The Mitrofanoff neourethra is an example of a flap valve
ness of the valve and thus incontinence. The flap valve mech- mechanism that is particularly applicable to children.71 In this
anism is considerably more durable because the continence procedure, a continent, catheterizable tubular conduit (neour-
mechanism is a stable component of the reservoir wall. ethra) connecting the urinary bladder to the skin is achieved.
This provides a one-way flap valve mechanism that permits
a catheter to be easily passed into the bladder. The flap valve
is also a secure mechanism to prevent incontinence
(Fig. 117-21). The appendix, the most common type of
Mitrofanoff tube used, is removed from its cecal origin and
the appendiceal mesentery is preserved. A submucosal plane
is developed in the bladder by either detrusor incision or
cystotomy with creation of a long submucosal tunnel. The ce-
cal end of the Mitrofanoff neourethra is exteriorized to the
skin; a U-insertion flap technique is used to help minimize
the risk for stomal stenosis. The Mitrofanoff neourethra con-
cept has proven extremely versatile and has been implanted
into the bladder, colon, and stomach with equal efficiency.58,72
Fascial sling Multiple sites for exteriorization are possible including the
lower abdomen, umbilicus, and perineum.58 The versatility
A of this technique has been enhanced by extending the length
of the appendix using tubularized cecum.58,73,74 Mitrofanoff’s
concept and extension of these principles have permitted
successful continent reconstruction of the lower urinary tract
in a wide variety of situations.71,75,76
B Fascial sling
If the appendix is unavailable, a tapered segment of ileum
Urethra (over a 12- to 14-French catheter) can be used, although cur-
FIGURE 117-18 Placement of a fascial sling to enhance bladder outlet rently a transverse retubularized segment of ileum is more
competence. (From Sheldon CA: Urinary reconstruction (rather than diver- commonly used.71,77,78 The length of these ileovesicostomies
sion) for continence in difficult pediatric urologic disorders. Semin Pediatr is limited by the circumference of the bowel segment used,
Surg 1996;5:10.) which is inadequate in some cases. Casale introduced a tech-
nique that allows a doubling in length of the continent
conduit.79 The ureter also provides a source for a Mitrofanoff
conduit when available (this conduit can be created if ne-
phrectomy had been or is being done or if a transureteroure-
terostomy is being performed).80 Because an ileal conduit is
readily constructed, a transureteroureterostomy is not recom-
Pressure mended unless otherwise indicated. If the ureteral segment is
regulating refluxing, concomitant ureteral reimplantation may be re-
balloon quired. Other continent, catheterizable mechanisms are avail-
able including the Benchekroun procedure,81 the Indiana
pouch,33 the Kock pouch,82 and the hemi-Kock.83 Although
used less commonly in children than the Mitrofanoff flap
valve, they may prove useful in selected circumstances.
The most common complication of the Mitrofanoff conduit
Artificial
is that of stomal stenosis, requiring stomal revision.76,84,85
urinary Mitrofanoff neourethras that could not be negotiated or were
sphincter lost due to ischemic necrosis were only rarely encountered.

Interface with Fecal

Incontinence
------------------------------------------------------------------------------------------------------------------------------------------------

The urinary tract should not be reconstructed without consid-

eration of anorectal function. Achievement of urinary conti-
Control
nence in the patient who will still require a diaper for fecal
assembly incontinence can hardly be considered a success. Further,
and pump persistent fecal soilage may potentiate the risk of UTI and pro-
FIGURE 117-19 Artificial urinary sphincter. (From Sheldon CA, Bukowski gressive deterioration of the reconstructed bladder and upper
TP: Bladder function. In Rowe MI, O’Neill JA, Jr., Grosfeld JL, et al [ed]: Essentials urinary tracts. Therapy for urinary continence can signifi-
of Pediatric Surgery. St Louis, Mosby-Year Book, 1995.) cantly compromise gastrointestinal function. For example,
 CHAPTER 117 RECONSTRUCTION OF THE BLADDER AND BLADDER OUTLET 1481

A B
FIGURE 117-20 A, Flap valves. B, Nipple valves. Nipple valves are continent because the nipple is circumferentially compressed by pressure within the
reservoir. Unfortunately, the intrareservoir pressure also has a laterally destructive force on the base of the nipple, causing a shortening or total effacement
of the valve with loss of the continence mechanism. Flap valves are continent because the submucosal segment is compressed by filling of the reservoir (as
for a reimplanted ureter for vesicoureteral reflux). Unlike nipple valves, flap valves are stable because they are fixed to the wall of the reservoir. Thus res-
ervoir filling does not tend to cause loss of the continence mechanism. (From Minevich E, Sheldon CA: Urinary tract reconstruction for continence and renal
preservation. In Ziegler MM, Azizkhan RG, Weber TR, et al [eds]: Operative Pediatric Surgery. Norwalk, Conn, Appleton & Lange, 2003, p 928.)

A C

B D
FIGURE 117-21 Mitrofanoff procedure. A, The appendix has been mobilized on its mesentery, and cecal segment is closed. B, Extravesical dissection
shows mucosal orifice in which the distal end of the appendix will be implanted. C, The detrusor is closed over implanted appendix, and its proximal end is
then brought to the skin to serve as a catheterizable stoma. D, This diagram depicts the resulting continent flap-valve mechanism of the Mitrofanoff pro-
cedure; a rise in intravesical pressure compresses the conduit against the detrusor, thereby occluding its lumen and achieving continence. (From Minevich
E, Sheldon CA: Urinary tract reconstruction for continence and renal preservation. In Ziegler MM, Azizkhan RG, Weber TR, et al [eds]: Operative Pediatric
Surgery. Norwalk, Conn, Appleton & Lange, 2003, p 929.)
1482 PART VIII GENITOURINARY DISORDERS

anticholinergic therapy may result in severe constipation, and
ileocecocystoplasty may facilitate fecal incontinence.
The management of intractable fecal incontinence may be
addressed at the time of urinary reconstruction and thus
should be evaluated before this undertaking. Because most
patients can be managed nonsurgically by dietary restrictions,
bulking agents, behavior modification, and expansion
enemas, an exhaustive trial of these interventions should be
undertaken before surgical reconstruction. In the patient who
is refractory to these treatments, the antegrade continence
enema (Fig. 117-22) has been shown to be an excellent alter-
native.86–89 This procedure is indicated when nonoperative
management is ineffective or when expansion enemas prove
effective but cannot be administered by the patient.90
The technique is similar to the Mitrofanoff neourethra. The
appendix or a retubularized ileal segment91 is implanted into a
An incision is made along the anterior tenia of the cecum;
an anastomosis is performed between it and the conduit;
and a mucosal defect is created distally in the mucosal trough.
The muscle is covered over the conduit while the other end is
exteriorized to skin. Once daily a catheter is inserted into the
continent cecostomy while the patient sits on the commode.
An antegrade enema consisting of tap water and table salt is
administered by gravity flow. If this is unsuccessful, additives
of mineral oil, polyethylene glycol (MiraLax), or glycerin can
be added to the irrigant daily.92
Urinary Reconstruction and
End-Stage Renal Disease
------------------------------------------------------------------------------------------------------------------------------------------------

tenia in the cecum, and the other end is exteriorized to skin. Congenital urologic disease is reported to occur in 20% to
The appendix or ileal segment is prepared in a manner 30% of pediatric end-stage renal disease patients.5,93 The ad-
identical to that described for the Mitrofanoff procedure. verse effects of congenital urologic disease on the success of

Cecum
Ileal mesentery

Catheter

Appendiceal
mesentery

Appendix

GIA stapler

Ileum C
B

Continent cecostomy

FIGURE 117-22 Technique for performing continent cecostomy for antegrade continence enema. A, Anatomy of the appendix and the anterior cecal
tenia. B, Appendix is mobilized on its vascular pedicle. C, Alternatively, a segment of ileum is tapered and mobilized on its vascular pedicle. D, Final ap-
pearance of continent cecostomy. (From Sheldon CA: Urinary reconstruction [rather than diversion] for continence in difficult pediatric urologic disorders.
Semin Pediatr Surg 1996;5:14.)
 CHAPTER 117
renal transplantation are demonstrated by the increased
incidence of UTI, allograft dysfunction, technical complica-
tions, and graft loss in such patients. UTI can be particularly
deleterious to both graft and recipient in the face of immuno-
suppression. The risk of UTI is higher in transplant recipients
with posterior urethral valves as the etiology of their renal fail-
ure.94,95 An analogous problem is seen in patients whose pri-
mary disease is vesicoureteral reflux and who continue to have
reflux into the native kidneys at the time of transplantation.96
These data demonstrate that patients with a history of sig-
nificant urologic disease must be extensively evaluated before
transplantation.97 Significant vesicoureteral reflux should be
resolved either by ureteral reimplantation or nephrectomy.
Whenever possible, nephrectomy is avoided to allow the pro-
duction of urine output so that a functionalized bladder is
maintained and dialysis therapy is still an option, either before
transplantation or after, should the graft fail. Any endogenous
erythropoietin production will be maintained, and this may
help avoid presensitization from frequent transfusions. In ad-
dition, the ureters are preserved, which may be necessary
for subsequent transplant reconstruction should a ureteral
complication occur. Indications for nephrectomy include
the presence of recurrent UTIs, UTI-prone upper tract ana-
tomy, refractory hypertension, severe proteinuria, and severe
polyuria.
When necessary, nephrectomy is performed via lateral
flank incisions or dorsal lumbotomy incisions, through which
the peritoneal cavity is rarely, if ever, entered. Such procedures
can be performed without interrupting peritoneal dialysis. Al-
ternatively, a laparoscopic approach is applicable. If bilateral
nephrectomy is considered necessary and preemptive (to
avoid dialysis) transplantation is desired, one may remove
the worst-functioning kidney first, and if the second kidney
provides adequate function, dialysis can be avoided. At the
time of transplantation the remaining kidney can be removed
via the transplant incision.
Elevated intravesical pressure must be stabilized by either
pharmacologic measures or intermittent catheterization (or
both) and, when necessary, augmentation cystoplasty. When
undertaking pretransplant surgery it is important to plan
the surgical incision so that it minimizes interference with cur-
rent or anticipated peritoneal dialysis, as well as the subse-
quent transplantation itself. The best means for avoiding
loss of access to the peritoneal cavity for purposes of dialysis
is an extraperitoneal approach to the urinary tract, which
can readily be achieved for nephrectomy and ureteral
reimplantation.
Surgical reconstruction of a dysfunctional bladder can be
successfully performed and has been demonstrated to allow
transplantation with acceptable results.98,99 Such an approach
allows optimal allograft preservation and appliance-free con-
tinence and is preferable to the alternative of transplantation
into an intestinal conduit. Transplantation is applicable to
even the most anatomically complicated child with end-stage
renal failure.100,101 The principles of transplantation into a
reconstructed bladder include the following: (1) Avoid a dry
reconstruction. The capacity and compliance of a recon-
structed bladder can be lost if the interval to transplantation
is long. This can be avoided by undiversion of native kidneys
(if present) or by using a living-related donor allograft where
timing can be optimized. (2) The risk of reabsorptive acidosis
occurring either before transplantation, with drainage of the
RECONSTRUCTION OF THE BLADDER AND BLADDER OUTLET 1483
native kidneys into the reconstructed bladder, or after trans-
plantation if the allograft fails over time can be minimized
by avoiding the use of intestine and colon for bladder recon-
struction. Specifically, gastrocystoplasty, augmentation with
the ureter and pelvis, and autoaugmentation all have the po-
tential for minimizing reabsorptive acidosis. (3) Autoaugmen-
tation and augmentation using the ureter have the potential
for being performed retroperitoneally, thereby avoiding
interference with present or future peritoneal dialysis or with
ventricular peritoneal shunts. (4) Whenever possible, ureteral
implantation into the native bladder is preferable. If not
possible, the colon, an ileocecal segment, or the stomach is
the only source of augmentation tissue that will permit
reliable antirefluxing anastomosis. (5) During the perfor-
mance of such transplantations, the blood supply of the
augmented bladder and any associated Mitrofanoff neourethra
must be meticulously identified and preserved.
Complications
------------------------------------------------------------------------------------------------------------------------------------------------
ACUTE ABDOMINAL SURGICAL ILLNESS
Acute abdominal surgical illness is a grave concern in the pa-
tient who has undergone urinary tract reconstruction, partic-
ularly when associated with bladder augmentation or creation
of an intestinal reservoir. The most common causes of an acute
surgical abdomen in this setting are perforation of the aug-
mented bladder or intestinal reservoir or small bowel obstruc-
tion. Even though the etiology remains conjectural, the
majority of perforations have been associated with augmenta-
tion of a remnant of neurogenic bladder.102 More than two
thirds of such patients have been on intermittent catheteriza-
tion, and total continence appears to be a common factor. Al-
though the clinical findings are usually those of an acute
abdomen, the symptoms may be quite nonspecific and a high
index of suspicion is essential. It is important to be aware of
the fact that the rupture may occur many years after recon-
struction. Altered sensation in patients with dysraphic states
or spinal cord injury and steroid administration in renal trans-
plant patients may confound the diagnosis. In establishing the
diagnosis a cystogram is essential but is associated with a sig-
nificant false-negative rate.103,104 A computed axial tomo-
graphic study of the abdomen (with contrast in the bladder
or reservoir) may be the most accurate method of making
the diagnosis. However, any patient with an augmented
bladder on intermittent catheterization who has abdominal
pain, fever, or vomiting should be presumed to have a bladder
perforation unless the symptoms can be conclusively attrib-
uted to another etiology. Exploratory laparotomy may be
required to make the diagnosis. Less common, but impor-
tant, etiologies of the acute abdomen include small bowel ob-
struction, pseudomembranous enterocolitis, toxic shock
syndrome, and ventricular-peritoneal shunt complications.37
The performance of laparotomy following urinary tract
reconstruction is of critical concern. Certainly, elective lapa-
rotomy should be preceded by formal bowel preparation in
the face of augmentation or continent diversion. The surgeon
should have access to a catheter in the bladder to allow
insufflation and deflation for identification purposes, and a
catheter should be placed in any catheterization conduit
1484 PART VIII GENITOURINARY DISORDERS
such as a Mitrofanoff neourethra. Efforts must be directed at
identification and preservation of mesenteric blood supply
to any gastric or intestinal segments used in reconstruction.
Whenever possible, an experienced reconstructive urologist
should be present.
METABOLIC COMPLICATIONS
Metabolic alterations may be encountered when gastrointesti-
nal segments are incorporated into the urinary tract.105,106
These metabolic derangements are due to solute flux, both ac-
tive and passive, between the urine and blood across the gas-
trointestinal segment wall. The character and severity of such
derangements depend on the nature of the segment used, the
absorptive surface area, the dwell time, and the metabolic re-
serve of the individual patient. Compensatory mechanisms for
metabolic changes are provided by the kidneys, liver, and
lungs. Significantly compromised function of any of these or-
gan systems may exacerbate an underlying metabolic defect.
Syndromes include alterations in acid-base status, disorders
of serum electrolyte composition, hyperammonemia, and
bone demineralization.
Systemic acidosis may result from the incorporation of je-
junal, ileal, or colonic segments into the urinary tract. Jejunal
conduits have been found to be associated with acidosis in
20% to 40% of instances.107,108 In the jejunum, passive diffu-
sion of solutes occurs along their concentration gradients. The
passage of hypertonic urine into a jejunal segment will result in
a loss of sodium, chloride, and water, resulting in hyponatre-
mia, hypochloremia, and volume contraction, with subsequent
contraction acidosis.109 Additionally, diminished renal blood
flow results in secondary hyperaldosteronism, resulting in a
more hypertonic urine and hyperkalemia. The latter is further
aggravated by the potassium shift as a result of acidosis.
The metabolic consequences of interposing ileal and co-
lonic segments within the urinary tract relate to the active se-
cretion of sodium (in exchange for hydrogen) and bicarbonate
(in exchange for chloride), as well as the reabsorption of
ammonium, hydrogen, and chloride. Ammonium absorption
appears to be quantitatively the most important, explaining
many of the abnormalities encountered when ileal or colonic
segments have an interface with urine. Hydrogen ion, gener-
ated from ammonium, is buffered by serum bicarbonate pro-
ducing water and CO2. The latter is readily eliminated by the
lungs and results in a chronic compensatory respiratory
alkalosis. Additional buffering is provided by bone, resulting
in a variable degree of demineralization and secondary hyper-
parathyroidism. This is manifest by hypercalciuria, hyper-
phosphaturia, hyperoxaluria, hypocitraturia, hypocalcemia,
and hypomagnesemia. Osmotic diuresis and acidosis combine
to result in total-body potassium depletion.
Metabolic alkalosis is a unique complication of gastrocys-
toplasty. Though uncommon,42 hypokalemic-hypochloremic
metabolic alkalosis has been reported.110,111 Excessive
bicarbonate absorption is postulated to occur secondary to
the combination of mineralocorticoid excess and potas-
sium/chloride depletion.
Hypokalemia, hypocalcemia, and hypomagnesemia are
significant potential sequelae of incorporating intestinal seg-
ments within the urinary tract.106 Sufficiently severe hypoka-
lemia to result in muscular paralysis has been reported.
Although hypocalcemia is rarely severe enough to be
symptomatic, it may present with irritability, tremors, tetany,
and coma and may even prove fatal. Hypomagnesemia is also
rarely severe enough to be symptomatic, but it may be man-
ifested as altered sensorium, personality changes, delirium,
psychosis, weakness, tremors, tetany, and seizures and may
likewise be fatal.
Hyperammonemia complicating urinary tract reconstruction
with intestine may cause altered sensoria and coma. As previ-
ously noted, ammonium ions are actively absorbed from intes-
tinal segments and may be present in large amounts in urine
because of their generation by renal tubules and production
from urea by urea-splitting organisms.
Perhaps the most overriding concern of incorporating
intestinal segments is the effect on childhood growth and
development. Several studies have provided data strongly
suggestive of defective linear growth in such cases.112,113 This
concern is particularly worrisome in patients with diminished
renal function. Here, the metabolic insult is more likely and
more severe because growth and development are often
already significantly impaired.
Strong evidence exists for a primary effect of incorporat-
ing intestinal segments into the urinary tract on bone miner-
alization.114 Metabolic acidosis results in defects in bone
mineralization, bone disease, and linear growth failure
through decreased renal tubular calcium reabsorption, de-
pressed intestinal absorption of calcium and phosphorus,
and vitamin D metabolism. Treatment with alkalinizing
agents has been partially successful in preventing or reversing
demineralization disease.
HEMATURIA-DYSURIA SYNDROME
Although rare in our own experience42 the hematuria-dysuria
syndrome is an important complication of bladder reconstruc-
tion using stomach.115 This syndrome is characterized by se-
vere pain and urinary bleeding as a result of urothelial erosion
from acid secreted in the urine after gastrocystoplasty.116 En-
doscopic evaluation of children with the hematuria-dysuria
syndrome suggests greater involvement of the urethra than
the bladder itself. Three major factors appear to be of impor-
tance in the genesis of this complication: acid hypersecretion,
profound oliguria,41 and bladder neck incompetency. True
acid hypersecretion appears to be quite rare and, presumably,
the predominant mechanism for this would be hypergastrine-
mia that has been reported in some instances.117 H2 receptor
blockers and proton-pump inhibitors such as omeprazole
have been shown to be reasonably effective in this syndrome.
ALTERED GASTROINTESTINAL FUNCTION
The incorporation of intestinal segments into the urinary tract
may result in significant alterations in gastrointestinal tract
function. Alterations in gastric function have been reported
following gastrocystoplasty.118 Functional alterations reported
include weight loss, feeding intolerance, dumping syndrome,
delayed gastric emptying, and esophagitis. Our own review of
gastrocystoplasty with emphasis on long-term follow-up
(minimum follow-up of 5 years) failed to demonstrate any sig-
nificant incidence of altered gastric function or altered acid
base status in 44 consecutive patients.42 Technical emphasis
must be placed on avoiding the vagus nerves, avoiding signif-
icant dissection in the region of the gastric pylorus, and
 CHAPTER 117
avoiding traction-distortion of the angle of His (e.g., by gastro-
stomy tube), which may predispose to gastroesophageal junc-
tion incompetence.
Several potentially important sequelae of intestinal malab-
sorption may accompany intestinal resection including diar-
rhea, vitamin deficiency, and fecal incontinence. Diarrhea is
most frequently a result of alterations in bacterial colonization
and impairment of bile acid reabsorption (with or without ac-
companying steatorrhea).119 Malabsorption from ileal resec-
tion is directly related to the length of resection. Resections
of greater than 100 cm of ileum (adult equivalent) diminish
bile acid reabsorption to a degree that cannot be compensated
by increased hepatic synthesis. As a result, the bile salt pool is
diminished and steatorrhea develops. Another important se-
quela of the diminished bile salt pool in this population is cho-
lelithiasis, which, like urolithiasis, is clinically seen at a
significantly increased incidence following ileal resection. Di-
arrhea induced by altered bile acid reabsorption (with or with-
out accompanying steatorrhea) is further enhanced by the
rapid emptying of ileal contents into the colon, causing a
tendency for osmotic diarrhea.
Although most of the data involve adult patients, it is
estimated that 10% of children undergoing resection of ileo-
cecal valve segments will experience chronic diarrhea. This
may be resolved after restoring intestinal continuity by return-
ing the ileocecal segment to its normal position within the
gastrointestinal tract.120
RECONSTRUCTION OF THE BLADDER AND BLADDER OUTLET 1485
Despite these data, the use of intestinal and gastric seg-
ments appears to be extremely well tolerated in most children.
It would, however, appear prudent to avoid the removal of
large segments of ileum or removal of the ileocecal valve for
purposes of urinary reconstruction, particularly in those al-
ready compromised. Such patients include those with preex-
isting malabsorption or short-gut syndromes and patients
with marginal fecal continence, in whom fecal soilage may be-
come incapacitating by loss of stool consistency. The latter pa-
tients include those with myelomeningocele and imperforate
anus, who commonly require urinary tract reconstruction.
MALIGNANCY
A majority of our understanding of urointestinal malignancy
following reconstruction comes from the experience with
ureterosigmoidostomy.121,122 The incidence of malignancy
developing in conduits and continent diversion has been
comparably small.123–125 In most cases there is a long interval
between surgery and the onset of cancer. This latency interval
has several important implications: A carcinogenic effect
is clearly implied. Moreover, there is a potential for early
surveillance diagnosis. Clearly, long-term follow-up is man-
datory.
The complete reference list is available online at www.
expertconsult.com.

CHAPTER 118
Incontinent and
Continent Urinary
Diversion
Audrey C. Rhee, Elizabeth B. Yerkes,
and Richard C. Rink
Urinary diversion can be either incontinent or continent. With
incontinent diversion the reservoir is external to the body (os-
tomy bag or diaper in children) and is emptied routinely when
full. With continent urinary diversion the urinary reservoir is
routinely emptied by clean intermittent catheterization (CIC).
Classically when the reservoir has been made entirely of gas-
trointestinal segments, it has been known as a continent urinary
diversion (CUR). In children, the primary diagnoses requiring
urinary reconstruction are spinal dysraphism, exstrophy, pos-
terior urethral valves, or complex cloacal anomalies. In these
disorders, the diseased native bladder can be incorporated
into the reconstruction so that true CURs are rare in children.
However, most will require intestinocystoplasty and a
catheterizable abdominal wall channel and therefore they
are “continent reservoirs” and are considered as such in this
chapter. The term undiversion used in this chapter previously
refers to the situation in which a child was previously treated
with an incontinent urinary diversion and is converted to a
continent diversion. Due to the infrequency of this situation,
it is not covered in this chapter.
Historically permanent incontinent urinary diversion was
used to protect the kidneys from the deleterious effects of in-
creased bladder and renal storage pressures or to manage uri-
nary incontinence. In more contemporary practice, medical
management of bladder dysfunction with CIC1 and anticho-
linergic agents has limited the indications for surgical inter-
vention. Urodynamic expertise has allowed identification
and aggressive management of the high-risk patients.2 The de-
velopment of surgical techniques to allow storage of urine at
safe pressures has virtually eliminated the use of permanent
forms of incontinent urinary diversion, but this still remains
an appropriate option for some patients. Temporary diversion
is still required for select patients but is rarely first-line
management for the bladder or the upper urinary tract.
Incontinent Urinary Diversion
------------------------------------------------------------------------------------------------------------------------------------------------
Although eventual urinary continence is desirable for all chil-
dren, protection of the integrity of the upper urinary tract is
the primary concern. Several forms of incontinent urinary di-
version are still occasionally used on either a temporary or per-
manent basis to prevent upper urinary tract deterioration or to
simplify day-to-day care of patients with limited capability for
self-care.
Temporary urinary diversion is most commonly performed
in infants or very young children, and the urine is easily col-
lected in the diaper. Most permanent incontinent diversions
require an ostomy appliance for hygienic collection of the
urine. Compared with the alternative of continence by clean
intermittent catheterization, the incontinent stoma carries a
stigma and body image issues that are better avoided in the
pediatric population. For select patients and families, how-
ever, this remains a reasonable option.
CUTANEOUS VESICOSTOMY
Cutaneous vesicostomy is the most common form of inconti-
nent urinary diversion in children today, but it is rarely used as
first-line therapy and has limited indications. Infants with a
neuropathic bladder and unsafe storage pressures may require
a vesicostomy if CIC and anticholinergic medications fail or if
the family is unable or unwilling to execute these medical
measures. Inability to incise posterior urethral valves safely
due to the small urethral caliber in some premature infants
is another indication for cutaneous vesicostomy. In these pa-
tients the vesicostomy is closed when the child has grown
enough for incision of the valves. Young infants with high-
grade vesicoureteral reflux and multiple breakthrough urinary
tract infections on antibiotic prophylaxis may benefit from
vesicostomy.3–4 Improvement in the efficiency of both bladder
and upper tract drainage reduces the incidence of urinary in-
fection. In persistent cloaca refractory hydronephrosis, hydro-
colpos or urinary infection despite intermittent catheterization
of the common channel is another uncommon indication for
vesicostomy.5
Ideally the vesicostomy provides low-pressure drainage of
the bladder contents with minimal residual volume. Multiple
series have confirmed the benefits of vesicostomy.3–4,6–9
Improvement in upper urinary tract dilation and stabilization
or improvement of renal function is achieved. Reflux may
1487
1488 PART VIII GENITOURINARY DISORDERS
resolve, or the degree of ureteral dilation may improve
enough that ureteral tapering is not required at the time of
reimplantation.3–4 Although bacterial colonization of the
open system is common, symptomatic infections and urosepsis
are reduced. One concern about prolonged vesicostomy
drainage in the non-neuropathic population is failure of
the bladder to develop normally in the absence of cyclic
filling and emptying.10–11 Several studies have refuted this
concern.4,12–13
Once the urinary tract is stabilized and the long-term po-
tential of the bladder becomes clearer, the vesicostomy is eas-
ily closed. It may be closed primarily or in combination with
continent diversion. Some patients and families are not appro-
priate candidates for vesicostomy closure or find that vesicos-
tomy and diaper drainage is an acceptable low-maintenance
permanent option. Vesicostomy, with or without the use of
an ostomy appliance, can also be performed in adult patients
who are not candidates for continent diversion.9
Two techniques for cutaneous vesicostomy have been used
for many years. Lapides described elevation of an anterior blad-
der wall flap with deep insertion of an abdominal skin flap to
fashion the cutaneous vesicostomy.14 The more common tech-
nique was described by Blocksom in 195715 and modified by
Duckett (Fig. 118-1).16 The Blocksom vesicostomy is fashioned
through a small transverse incision halfway between the umbi-
licus and the pubis. The fascia is incised, and the peritoneum is
pushed superiorly off the dome of the bladder. The urachal
remnant is divided, and the dome of the bladder is pulled
up to the skin. The fascia is secured to the bladder wall to form
a 24-Fr defect, and the bladder is matured as a flush stoma.
FIGURE 118-1 The Blocksom vesicostomy, showing the incision in the
dome of the bladder, the fixation to the overlying fascia, and the everted
stoma.
Complications of cutaneous vesicostomy include stomal
stenosis, prolapse, peristomal dermatitis, and bladder calculi.
Stomal stenosis is more common with a stiff or thick bladder
wall as in posterior urethral valves. Failure to fashion the vesi-
costomy from the dome of the bladder may allow the posterior
bladder wall to prolapse through the vesicostomy. This is one
advantage of the Blocksom technique over the Lapides vesi-
costomy. Prolapse is managed by manual reduction with tem-
porary catheter drainage and revision of the low vesicostomy.
Dermatitis is particularly bothersome in some patients and is
managed with application of a zinc oxide barrier cream or top-
ical treatment for obvious candidiasis. Bladder calculi are un-
usual provided that the vesicostomy is functioning well and
the upper tracts are drained.
CUTANEOUS URETEROSTOMY
Ureterostomy for temporary supravesical urinary diversion is
rarely performed today. It has been used historically for drain-
age of a profoundly dilated upper urinary tract in the face of
urosepsis17 or to maximize drainage of the renal units in cases
of severe obstruction. Percutaneous or endourologic drainage
of these systems is possible now in children, but tube size
limits the utility of this in very small infants.18 Low diversion
can be performed as a loop or end cutaneous ureterostomy
(Figs. 118-2 and 118-3), and subsequent takedown involves
ureteroneocystostomy. The diversion allows the caliber of
the ureter to decrease, thereby reducing the complexity of
the reimplantation. Diversion at the level of the kidney is per-
formed through a flank incision and can be performed as a
FIGURE 118-2 The loop ureterostomy is easily taken down and there-
fore allows for an excellent temporary diversion.
 CHAPTER 118 INCONTINENT AND CONTINENT URINARY DIVERSION 1489

Another method to treat distal ureteral obstruction is cre-

ation of a freely refluxing end-to-side or side-to-side ureteral
reimplant.23 This alternative approach, while obviating the
continual drainage of urine via a cutaneous stoma, remains
controversial.

CUTANEOUS PYELOSTOMY
As with cutaneous ureterostomy, proximal diversion by pye-
lostomy has limited indications in current practice. The con-
troversies surrounding this type of diversion are similar. With
direct drainage of the renal pelvis, pyelostomy may be used for
temporary diversion in select cases of obstruction at the ure-
teropelvic junction but could also be used as discussed earlier
for posterior urethral valves. Pyelostomy requires a dilated
extrarenal pelvis to create a tension-free anastomosis to the
skin. Takedown of the pyelostomy at the time of a definitive
procedure is relatively simple due to reliable blood supply
to the renal pelvis.

ILEAL CONDUIT/COLON CONDUIT

Permanent supravesical urinary diversion was popular many
decades ago as reliable means to manage neuropathic bladder,
bladder outlet obstruction, and severe urinary incontinence.
Conduit diversions have been created with most intestinal
segments, although the ileum (Fig. 118-4) and colon are used
FIGURE 118-3 The end ureterostomy requires formal reimplantation to most commonly. Both simple refluxing and tunneled nonre-
reverse. fluxing ureterointestinal anastomoses have been used. The
primary advantage of the colon conduit over an ileal conduit
loop ureterostomy, Sober ureterostomy, or end ureterostomy. is the reliable antireflux reimplantation achievable with colon.
A loop ureterostomy can easily be taken down and is the more The risk of upper tract deterioration after ileal conduit di-
appropriate choice for temporary diversion. End ureterostomy version has exceeded 50% in some series.24–26 Although this
just below the ureteropelvic junction makes reconstitution of is not a tremendous concern in the older patient after cystec-
the ureter more difficult and should be reserved as a rare per- tomy for bladder carcinoma, the long-term impact on the kid-
manent form of diversion in cases of severe bladder dysfunc- neys is extremely relevant in the pediatric population. The
tion.18 Sober ureterostomy, with its proximal diverting limb ability to create a nonrefluxing anastomosis between the ureter
and reanastomosis of the distal limb to the renal pelvis, is rarely
used today and is technically much more involved and may
therefore be less appropriate for diversion during acute illness.
On the other hand, it offers the advantage of diversion and
continued antegrade drainage to cycle the bladder.19 In all
cases, care should be taken to preserve the medial blood sup-
ply and adventitial vessels and to create a tension-free anasto-
mosis to the skin. Given the profoundly dilated nature of the
ureter in most of these cases, stomal stenosis is rarely an issue.
Supravesical diversion may be warranted in select posterior
urethral valve patients when renal function fails to improve af-
ter bladder drainage.19 Persistent ureterovesical junction ob-
struction is the concern. Ureterostomy and pyelostomy
would maximize upper tract drainage, but percutaneous
nephrostomy is a less invasive temporary means to assess the
potential of the affected kidney.20 Patients who demonstrate
improvement in renal function after supravesical diversion
may do so because of better drainage or just due to normal renal
maturation.21 As with cutaneous vesicostomy, there is concern
that the valve bladder will fail to develop normally if the urine is
diverted. In patients with a solitary functioning kidney, a loop
ureterostomy or an end ureterostomy would defunctionalize
the bladder. Several studies suggest that bladder dysfunction
is due to bladder wall abnormalities from infravesical obstruc- FIGURE 118-4 The ileal conduit is the standard permanent supravesical
tion rather than supravesical diversion of urine.13,22 diversion for patients whose bladder is absent.
1490 PART VIII GENITOURINARY DISORDERS
and colon gives colon conduits a theoretical advantage over
ileal conduits. Unfortunately, deterioration of the upper tracts
occurs in 26% to 48% of cases and ureteral reflux persists in
8% to 58% of colon conduits.27 Anastomotic stricture
occurs with similar frequency with ileum and colon (9%
to 22%).24,27
Koch and colleagues24 compared a cohort of myelomenin-
gocele patients with ileal conduit diversion to another group
managed with CIC. The ileal conduit group suffered renal de-
terioration, nephrolithiasis, and pyelonephritis significantly
more than the catheterization group. Bone density was de-
creased in both groups and the incidence of fractures was
equal, but the diversion group had impaired linear growth,
more spinal curvature, and more complications from orthope-
dic procedures. The metabolic alterations associated with in-
corporation of intestinal segments into the urinary tract do
occur with incontinent diversions.24
Other complications of conduit diversion included stomal
stenosis, stomal prolapse, peristomal hernia, and peristomal
dermatitis. Stomal stenosis occurred in 8% to 61% of cases.
Stoma complications are best prevented with careful surgical
technique including preservation of intestinal perfusion, cre-
ation of a straight and adequate caliber fascial window, and fix-
ation of the conduit to the fascia. A protuberant rosebud stoma
enhances the fit of the stoma appliance and prevents stenosis.
Bacteriuria is nearly universal but does not cause upper
tract damage if the ureterointestinal anastomosis and the con-
duit itself function appropriately.27–28 Stasis of colonized
urine, however, may contribute to upper tract stone formation
and may have other implications as well. Nitrosamines pro-
duced by the mixture of urine and bacteria appear to have
some role in induction of malignancy in the isolated intestinal
segment.27,29
Austen recently reviewed the world literature and identi-
fied 81 tumors collectively in a variety of continent and incon-
tinent diversions with incorporated bowel segments. This
total is exclusive of the large number of malignancies reported
after ureterosigmoidostomy. Neoplasia was reported in 12 ileal
conduits and 5 colon conduits with a mean latency of 22 and
10 years, respectively. Colon conduit tumors were exclusively
adenocarcinoma, but transitional cell carcinoma, squamous
cell carcinoma, and carcinoid and anaplastic tumors were also
found in ileal conduits. Chronic inflammation of the intestinal
segment may play a role in the malignant changes. Early yearly
surveillance for tumor is recommended.30
Preoperative preparation for conduit diversion involves a
full mechanical and antibiotic bowel preparation and paren-
teral antibiotic to sterilize the urine on the day before surgery.
Patients with severe neuropathic bowel may actually require
more than one day of preparation for adequate cleanout. It
is critical that an enterostomal therapist mark potential stoma
sites bilaterally after examining the patient in multiple posi-
tions. The abdominal contour can change significantly in mye-
lomeningocele patients when seated, and an ill-fitting ostomy
appliance will be a lasting source of dissatisfaction.
INCONTINENT ILEOVESICOSTOMY
Incontinent ileovesicostomy, initially described by McGuire
and colleagues in 1994,31 has primarily been used in the adult
population with bladder and sphincter dysfunction secondary
to spinal cord injury and multiple sclerosis. Patients had failed
other efforts to manage the bladder pressures and urinary in-
continence and refused or were not candidates for continent
diversion. Many suffered severe complications from use of a
chronic indwelling urethral catheter. Several series report ex-
cellent results with regard to preservation of renal function,
incidence of urinary tract infection and urolithiasis, and sto-
mal complications.32–35 The ileovesicostomy functions as a
pop-off valve to maintain safe bladder storage pressures. Sto-
mal leak-point pressures are low, and a moderate residual
urine remains.35
Incontinent ileovesicostomy is an option to consider in
select pediatric patients with limited dexterity and social
support. Although it does require a permanent stoma, pres-
ervation of renal function appears superior when compared
with standard conduit diversion. In addition, the disruption
of the ureterovesical junction is not required. Parents may
have difficulty accepting permanent diversion even if the
child has limited potential to perform self-catheterization
in the future. Some families are not ready to assume the care
required for bladder augmentation or continent urinary
reservoir, despite the fact that the child needs reconstruction
to preserve the upper urinary tracts. In these instances,
incontinent ileovesicostomy with both an efferent inconti-
nent limb and a large open patch of ileum to augment the
bladder can be considered. Consideration could be given
to creation of a Mitrofanoff catheterizable channel at the time
of the initial procedure. Should the family later embrace
continence and the long-term care of bladder augmentation,
the ileal chimney can be amputated to leave a continent
diversion.36
Thorough preoperative bowel preparation and sterilization
of the urine are required. A stoma site is carefully preplanned
for the right lower quadrant to ensure optimal fit of the stoma
appliance. The bladder is bivalved in the coronal plane in
preparation for anastomosis to the isolated ileal segment.
The proximal end of the ileum is opened widely on its antime-
senteric border and anastomosed to the bladder in a running
fashion. The length of the ileal segment can be tailored to ac-
count for body habitus and for the need to augment the native
bladder capacity.32 A rosebud stoma is fashioned in the right
lower quadrant, and a 22-Fr catheter is left in place for 3
weeks to maintain the caliber of the ileovesical anastomosis.
If significant augmentation is performed, a suprapubic
catheter should be left as well to maximize perioperative
drainage. In select patients with perineal excoriation from
urinary incontinence, a bladder neck or urethral procedure
in conjunction with an ileovesicostomy may be used to create
outlet resistance.
Continent Urinary Diversion
------------------------------------------------------------------------------------------------------------------------------------------------
Many different types of continent urinary diversions have
been developed to improve or replace the native lower urinary
tract. The chosen technique depends largely on the surgeon’s
experience and preference, but the primary pathology plays a
large role in any given patient. Reconstruction is tailored to the
specific anatomy and related functional deficits. A fundamen-
tal difference exists in pediatric lower urinary tract reconstruc-
tion versus that of adults. In adult reconstruction, a particular
type of reconstruction (i.e., Kock pouch, Indiana pouch) may
 CHAPTER 118
be chosen preoperatively. In children, the principles of a low-
pressure reservoir with a means of catheterization without
leakage or reflux must be achieved from the tissues available
at this time of reconstruction. Whenever possible, enhance-
ment of the native bladder and outlet is preferred to avoid
the potential complication associated with ureteral anastomo-
sis to a CUR. CUR would be performed if the native bladder is
absent or not salvageable. Patients with bladder agenesis syn-
dromes, the most severe bladder exstrophy, and pelvic organ
destruction secondary to trauma or malignancy would be
candidates for CUR.
Continent urinary diversion is now common, but one must
remember that it requires a broad spectrum of surgical skills
and must address the varied underlying diagnoses. By defini-
tion, a reservoir that must be emptied by intermittent catheter-
ization is created. This may include the augmentation of an
intact but severely diseased bladder or the creation of a reser-
voir completely from heterologous tissue. The surgical goals
and principles are similar: to create the ideal storage reservoir
for urine that can be easily emptied with the minimum of com-
plications. Because the gastrointestinal (GI) tract is plentiful
and accessible, it is the most common source of tissue. The
physiologic variability along the GI tract allows the surgeon
to tailor the reconstruction to the patient. However,
these complex surgeries are not without complications, and
these have to be understood and anticipated.
BLADDER AUGMENTATION
Bladder augmentation is an essential component of the pedi-
atric urologist’s surgical armamentarium. It has a prominent
role in the management of many lower urinary tract (LUT) dis-
orders including neurogenic bladder, posterior urethral
valves, and bladder exstrophy. The modern application often
includes the use of a continent catheterizable abdominal wall
stoma based on the Mitrofanoff principle. This is generally
performed using either the appendix (appendicovesicostomy)
or reconfigured ileum (Monti) and may require an additional
procedure to improve the continence mechanism at the blad-
der neck. Intermittent emptying via the native urethra remains
an option.
As previously discussed, children with severe LUT dys-
function were historically treated with urinary diversion.
Lapides’ introduction of clean intermittent catheterization en-
sured that any child could safely empty his or her bladder.1,37
This brought the dawning of lower urinary tract reconstruc-
tion and allowed the urinary tract to remain intact and avoid
the use of external collecting devices. Advances in urodynamic
monitoring improved our ability to predict who would best
benefit from bladder reconstruction and therefore prevent re-
nal deterioration.2 Parallel medical advances ensured the sur-
vival of these complex patients and allowed for the shift of
focus toward continence as an important aspect of their
quality of life.38–39 Recent advances with cutaneous catheter-
izable channels continue to simplify care and increase its
popularity.40
The primary goal of lower urinary tract reconstruction is to
store urine safely without leakage. Adequate storage requires a
low-pressure, highly compliant reservoir of adequate volume,
whereas continence results from limited contractility and an
effective sphincter mechanism.
INCONTINENT AND CONTINENT URINARY DIVERSION 1491
The vast majority of augmentations are performed with
gastrointestinal segments because bowel segments are readily
available and easily configured. Ileum is currently the most
popular segment,41–43 but sigmoid is often used. Stomach
and ileocecal segments have been used but have only a small
role in contemporary management.44–47 Their abundance en-
sures that adequate capacity is obtained while detubulariza-
tion and the natural viscoelastic properties allow for the
low-pressure reservoir.48–49 However, the secretive50–51 and
absorptive52 nature of this tissue is also responsible for most
of the common complications associated with this procedure.
Requirements for a successful augmentation include proper
patient selection,53 the ability and willingness to perform
CIC,50,54 proper selection of augmentation material, and
recognition and treatment of complications.
PATIENT EVALUATION AND SELECTION
The patient, family, and entire health care team must be in-
volved in the decision to pursue major surgery. The common
risks of incorporating GI segments into the urinary tract in-
clude mucus production, urinary tract infection, bladder
and renal calculi, and metabolic changes. Life-threatening
risks such as malignant degeneration and spontaneous perfo-
ration must be explained to the family and child, and they
must be understood.50,51,54 Everyone involved in caring for
the child must be committed to the appropriate postoperative
care including mandatory use of CIC and regular bladder
irrigation.50,54,55
Clinical evaluation requires a thorough history and physi-
cal examination. Particular attention should be paid to latex
allergy because this can be a fatal complication. Preoperative
evaluation should include a renal-bladder ultrasound (RBUS)
and a voiding cystourethrogram (VCUG). Serum chemistries
are required because impaired renal function may change
the gastrointestinal segment used.54 Initially, it was suggested
that a creatinine clearance of less than 60 mL/minute might be
a contraindication for the use of ileum as a continent urinary
reservoir,56–57 but augmentation often stabilizes renal func-
tion and reconstruction has been done before renal transplan-
tation. Urodynamic studies are necessary to determine outlet
resistance, bladder capacity, and storage pressures. Cystos-
copy will confirm anatomy and ensure that no untoward
findings that may hinder reconstruction are present.50
The net result of these investigations is a comprehensive
surgical plan. The exact procedure will depend on which seg-
ment is to be used and whether the patient requires concom-
itant ureteric reimplantation, bladder neck procedure, and a
catheterizable channel. Patients with a neurogenic bladder
may also undergo an antegrade continence procedure to assist
evacuation of their neurogenic bowel.58
Preoperative preparation usually involves bowel prepara-
tion, as described previously for conduit diversion. Perioper-
ative antibiotics and a sterile urine are required and are
particularly important if the patient has a ventriculoperitoneal
shunt.59
Long-term follow-up is required with all patients. After the
postoperative visit, and assuming the absence of complica-
tions, the patient should be seen annually with a history, phys-
ical examination, creatinine and electrolytes, an ultrasound,
and an annual cystoscopy 5 to 10 years out from surgery.50,54
1492 PART VIII GENITOURINARY DISORDERS
GASTROINTESTINAL CYSTOPLASTY
Most children undergoing lower urinary tract reconstruction
for hostile bladder dynamics can have their native bladder
preserved but need to have it augmented to lower intravesical
pressures, limit contractility, and improve compliance. Gas-
trointestinal segments are generally used, but this requires
reconfiguration and detubularization to prevent their own
inherent contractile properties.44,48,49,60 Maximum storage
capacity requires approximation of the spherical shape.48
A widely bivalved bladder enables this and helps to prevent
the augment behaving as a diverticulum (Fig. 118-5). The
volume of the sphere is maximized by folding the ileum into
a U or an S shape, which increases the potential radius and
volume (Fig. 118-6).54 This reconfiguration, as well as detu-
bularization along the antimesenteric border, is critical to dis-
rupting intestinal contractions because the intact intestine
can create pressures of 40 to 100 cm H2O.61–63 The recon-
structive surgeon should err on the side of a larger rather
than smaller augmentation.50 Regardless of the bowel seg-
ment selected, a water-tight anastomosis using absorbable su-
ture is done. A suprapubic tube is placed into the native
bladder and secured to the abdominal wall, and a perivesical
drain is placed.
FIGURE 118-5 The offset cystoplasty allows for more mobility of the
detrusor muscle and provides better access to the muscular backing that
is essential for implantation of the catheterizable channel.
FIGURE 118-6 The ideal bladder augmentation will increase both capac-
ity and compliance, which are required for continence and renal
protection.
ILEOCYSTOPLASTY
Ileum has become the segment of choice due to its inherently
low contractility, abundance, and ease of manipulation.41–43
In children 20 to 30 cm are harvested, with the distal margin
located 15 to 20 cm from the ileocecal valve to prevent vitamin
B12 and bile salt malabsorption.49,64 Ileum is known to
produce less mucus than colon, and we noted a lower rate
of perforation when compared with sigmoid.35 It has the dis-
tinct disadvantage of being difficult to create submucosal
tunnels for ureteral reimplantation or catheterizable channel
placement.
SIGMOID CYSTOPLASTY
Advantages of the sigmoid include its proximity to the bladder
and its marked dilation in the neuropathic population. Ap-
proximately 15 to 20 cm are isolated and irrigated with an an-
tibiotic solution. Due to the extreme contractile nature of the
sigmoid, complete detubularization is essential.46 Spontane-
ous perforation rates have been shown to be higher with sig-
moid as opposed to ileal augmentations,65 presumably due to
their increased contractile pressure.66
ILEOCECAL CYSTOPLASTY
The main advantage of the ileocecal segment is the consistent
blood supply. Two main techniques exist, each with multiple
variations. Either both the ileal and cecal segment are
tabularized and reconfigured together, or solely the cecum
is tubularized and the ileum used to create a continent stoma47
or for ureteric replacement.54 This segment is infrequently
used in the neuropathic population because loss of the ileo-
cecal valve can result in intractable diarrhea.67
GASTROCYSTOPLASTY
The stomach is much less absorptive than other intestinal seg-
ments, and its secretion of hydrogen ions may be beneficial in
patients with chronic renal failure and metabolic acidosis.44,68
A 10- to 15-cm wedge from the greater curvature is mobilized
along the right gastroepiploic vessel and passed through the
mesentery of the transverse colon to the bladder. Originally
felt to be an option for all augmentation candidates, its role
is now limited primarily due to the hematuria dysuria
syndrome, which occurs in up to 70% of patients.
Autoaugmentation
------------------------------------------------------------------------------------------------------------------------------------------------
Described by Cartwright and Snow,69 in order to decrease
intravesical pressures and increase bladder capacity, a large
diverticulum is created at the dome of the bladder by a detrusor-
ectomy while leaving the bladder mucosa intact. Unfortunately,
although the initial reports of urodynamic improvements
seemed promising, the results were not as durable as those
patients treated with an enterocystoplasty and seem to do better
in adults than in children.70–72 In general, autoaugmentation
will decrease intravesical pressures. It will not reliably increase
capacity that is often necessary in children. It does remain an
option for a select group of patients.
 CHAPTER 118 INCONTINENT AND CONTINENT URINARY DIVERSION 1493

SEROMUSCULAR SEGMENTS (WITH

UROTHELIAL LINING)
To avoid the incorporation of intestinal mucosa with the uri-
nary tract, Shoemaker and Marcucci described using seromus-
cular segments of bowel overlying an autoaugmented
bladder.73 Most experiences have demonstrated that whether
the segment is facing the bladder lumen or is reversed, signif-
icant contracture develops.74–75 Most recently, some have
demonstrated success with demucosalized colon over urothe-
lium with findings of increased bladder capacity and low fill-
ing pressures.76 Long-term results and more experience are
necessary to help determine the efficacy of this approach.

CONTINENT CATHETERIZABLE CHANNELS

Bladder augmentation and bladder neck surgery markedly de-
crease the patient’s ability to empty spontaneously while in-
creasing the consequences of incomplete emptying. Not
adhering to a strict CIC schedule can result in an increased
risk of urinary tract infection, bladder stones,77 and spontane-
ous bladder perforation.78–79 In 1980 Mitrofanoff introduced
the principle of a continent channel by using the vermiform
appendix and implanting it submucosally into the bladder.80
What is now known as the “Mitrofanoff Principle” states
that any supple tube implanted submucosally with sufficient
muscle backing acts as a flap valve and results in a reliable
continence mechanism (Fig. 118-7).
The Mitrofanoff principle has been widely embraced and ap-
plied to a variety of tissues because the appendix (Fig. 118-8) is
not always appropriate or available, especially if a simultaneous
Malone antegrade continence enema (MACE) procedure is be-
ing performed. The use of stomach, colon, bladder, ureters,
and fallopian tubes have all been reported with good suc-
cess.81–92 However, the utility of the tubularized ileal channel
was introduced by Yang87 and Monti82 and has assumed a lead-
ing role in genitourinary reconstruction (Fig. 118-9).84,93
The continent catheterizable channel requires a supple
tube, straight path, and short intra-abdominal segment. The
appendicovesicostomy requires full mobilization of the right
colon to ensure adequate mobility. Amputation should in-
clude some cecum because this allows for a larger-caliber cu-
taneous stoma or can be tubularized to increase stomal
length.81,94 The appendix is then carefully mobilized with
the mesoappendix to ensure adequate blood supply. The
FIGURE 118-7 The Mitrofanoff principle states that a supple tube will
compress with increasing intravesical pressure and provide continence.
A
B
FIGURE 118-8 A, The appendix is readily accessible and mobilized with
maintenance of its vasculature. B, It is implanted easily beneath the blad-
der mucosa to provide a reliable catheterizable channel.
Monti-Yang technique requires the isolation of 2 cm of ileum,
opening it along its antimesenteric border and retubularizing
it over a 12-Fr catheter. The channel is then preferentially
implanted into the submucosa of the bladder because its thick
muscle is ideal for a continent valve; however, stomach and
the tenia coli have been successfully used. Reimplantation into
ileum is the most challenging and requires the creation of a
seromuscular trough. The stoma site can be hidden within
the umbilicus or placed in the right lower quadrant, the latter
being favored as a shorter and more direct route. Great care is
taken to ensure a catheter, usually 12-Fr, passes easily and that
the channel is as straight and tension free as possible. When
satisfied, the surgeon fixes the channel to the abdominal wall
with a permanent suture. Many ingenious skin flap techniques
are used to minimize the chance of stomal stenosis, all insert-
ing into the spatulated channel. An indwelling catheter is
left for 2 to 3 weeks, and the first catheterization is usually
performed in the clinic setting.
Stomal continence is excellent, and rates of 90% to 99% are
reported in the two largest published series.84,95 Complica-
tions pertain primarily to difficulties with catheterization,
most commonly at the skin level, but may also occur deeper
within the channel itself. Stomal stenosis is reported to occur
in 5% to 25%,95–96 with a lower rate potentially seen with
tubularized ileum.84
1494 PART VIII GENITOURINARY DISORDERS
3.5 cm
12-14 cm
FIGURE 118-9 Normally, a 2-cm segment of ileum can be opened along
its antimesenteric border and tubularized for an effective channel. A
slightly longer piece (3.5 cm) can be partially divided, both ends tubular-
ized, and reanastomosed for a spiral Monti, for situations demanding more
channel length.
Several authors have described continent catheterizable
tubes fashioned from native bladder. Casale’s original descrip-
tion of the “intravesical channel”97 was modified by Rink.98
Although continence rates of 100% were reported, the
technique is hampered by a 45% incidence of stomal stenosis.
Although still a useful technique if an intraperitoneal pro-
cedure can be avoided, it has been largely replaced by the
appendicovesicostomy and reconfigured ileovesicostomies.
These channels provide a more convenient and more so-
cially acceptable means of catheterization. It is especially ben-
eficial for those with a sensate urethra and in the obese or
wheelchair-bound female. It also assists caregiver comfort
and, most importantly, patient independence.
CONTINENT URINARY RESERVOIRS
If the entire bladder has been replaced by nonurothelial tissue
(usually gastrointestinal) and requires emptying by means
other than the urethra, it is referred to as continent urinary di-
version. Although popular and with many refinements over the
past several decades, it is not commonly used in pediatrics
today because generally the bladder can be incorporated into
the reconstruction.50
A few diagnoses such as bladder agenesis or malignancy
resulting in cystectomy will require complete bladder
replacement. When required, however, it allows the complex
patient to achieve continence and enjoy the benefits of such.
Patient selection for the procedure is as important as with
bladder augmentation because noncompliance will result in
the same myriad of complications, despite a technically per-
fect operation. Preoperative evaluation is similar to the blad-
der augmentation patient, with the acquisition of sufficient
clinical, metabolic, anatomic, and functional data to develop
a detailed surgical plan.
The goals of the continent urinary diversion include the
creation of a reservoir of adequate capacity and compliance,
nonrefluxing ureteric implantations, a continent cutaneous
stoma, and a minimum of complications.
Although the gastrointestinal tract again provides for the
reservoir, several key differences exist with an augmentation.
Colon and stomach have assumed a larger role because the
musculature of the tenia coli and stomach allow for more re-
liable nonrefluxing ureteric and catheterizable channel anas-
tomoses. The three most common types of continent
reservoir in children are (1) a reservoir fashioned solely from
ileum (Kock pouch), (2) a gastroileal composite, and (3) an
ileal reservoir (Indiana pouch).95 Each has a myriad of varia-
tions described, all attempting to decrease complications and
assist construction. However, because published reports in
children are usually limited to small numbers, each reservoir
having subtle variations and no published direct comparisons,
it is impossible to determine an ideal reservoir.55,95,99 There-
fore the reconstructive surgeon must be familiar with several
techniques because each has advantages that may require ex-
ploitation with any clinical situation.
KOCK POUCH
Kock first described a continent ileostomy in 1971100–101 and
from this developed the Kock pouch in 1982.102 This continent
ileal reservoir was among the original nonorthotopic
bladder substitutions used and remains popular today.103–105
It requires the construction of an efferent nipple for continence
and an afferent limb for the prevention of reflux. This tech-
nically demanding aspect usually requires the use of stapling
devices, and these can be the source of its most significant
complications (stone formation) and long-term failure rate.106
Approximately 80 cm of ileum is harvested in adults, with
the proximal 15 to 20 cm used in an isoperistaltic fashion to
create the antireflux mechanism. The distal 12 to 15 cm are
used for the continent cutaneous stoma. These lengths are de-
creased appropriately depending on the size of the child. The
bowel is intussuscepted through its full thickness, and this is
secured by three rows of gastrointestinal staples, although
absorbable mesh has also been described.107 Significant mod-
ifications include removing the distal 6 staples from the
device,106 stripping the distal mesentery,108 and the use of
absorbable staples.109
GASTROILEAL POUCH
The advantages of both gastric and ileal segments can be max-
imized while offsetting their complications by incorporating
them both into a composite reservoir. The metabolic derange-
ments complement each other, as the acidic gastric secretion
will neutralize the absorptive properties of the colon or
 CHAPTER 118
ileum.110–112 Furthermore, in patients at high risk for short
gut syndrome, as in cloacal exstrophy, it allows a minimum
of valuable absorptive tissue to be lost.50 However, it is a more
complex reconstruction, requiring two anastomoses and lon-
ger operating time, but is a valuable alternative for select
patients.
INDIANA POUCH
The ileocecal segment has been popularized due to the poten-
tial continence or antireflux mechanism inherent to the ileo-
cecal valve. Many surgical techniques have been described,
in the attempt to maximize cutaneous continence, simplify
the surgical procedure, and minimize complications. The In-
diana group modified the pouch described by Gilchrist and
Merricks with a unique plication method that has retained
its popularity. An Indiana pouch uses the detubularized ce-
cum as its reservoir, with ureters implanted in a nonrefluxing
manner into the tenia. The ileum is then plicated to the ileo-
cecal valve, which is reinforced by Lembert sutures (Fig. 118-10).
Continence has been reported in 95% to 99%,113–114 in part due
to its large capacity.115 Complication rates have been acceptable,
as low as 0.6%,113 but not all surgeons have reported such
success.116
COMPLICATIONS FROM THE
INCORPORATION OF INTESTINE
IN THE URINARY TRACT
Despite the popularity of using the gastrointestinal tract as a
urinary reservoir, the potential complications are numerous
and can be significant. A broad categorization would include
(1) complications due to structural defects, (2) complications
secondary to the loss of the intestinal segment, (3) complica-
tions due to secretions, and (4) complications from
absorption.
FIGURE 118-10 The Indiana pouch uses the ileocecal valve as a conti-
nence mechanism and the cecum for urinary storage.
INCONTINENT AND CONTINENT URINARY DIVERSION 1495
STRUCTURAL COMPLICATIONS
Structural complications include the requirement of a second
augmentation, spontaneous perforation, and long-term malig-
nancy potential. A secondary augmentation may be required
in up to 6% of patients,66 and most are due to high intravesical
pressures from persistent bowel contractility.117–118 The hall-
mark clinical signs are incontinence or hydronephrosis with
screening ultrasound, and this is confirmed with urodynamic
studies.
A spontaneous perforation can be a fatal complication and
must not be underestimated. Early urine leaks are likely due to
technical error, but late complications usually originate in the
bowel, approximately 1 cm from the anastomotic line.50 Pa-
tients often present later in the course, as their neurologic def-
icit impedes symptoms. Therefore sepsis and death are
realistic possibilities, and a high index of suspicion must pre-
vail. Diagnosis is with a CT cystogram and treatment is usually
by laparotomy and primary closure,78 but conservative treat-
ment with catheter drainage has been successful in select
patients.119
The development of an adenocarcinoma is concerning due
to the well-documented occurrence following ureterosigmoi-
dostomy.120–124 It arises from the ureterointestinal junction
and occurs in these patients with a 7000-fold risk over the
general population.123 Although a tumor has been reported
as soon as 3 years after augmentation,125–126 the mean latency
is 21.5 to 26 years.30,123 At our institution, 3 malignancies
have occurred in our series of more than 500 augmentations.
All patients died of metastatic disease, with a mean time from
augmentation to diagnosis of 19 years.127 Husmann and col-
leagues128 have suggested that patients with neurogenic blad-
ders, exstrophy, and other congenital anomalies requiring CIC
are at higher risk for malignancy regardless of whether enter-
ocystoplasty is performed. The risk of carcinoma (both adeno-
carcinoma and transitional cell) in this patient population is
increased with exposure to other carcinogens (e.g., tobacco)
and immunosuppression.129 Although most surgeons appre-
ciate the risk, definitive guidelines for surveillance have not
been established. Annual cystoscopy has been recommended
to begin 3 to 10 years after augmentation.30,50
LOSS OF INTESTINAL SEGMENT
Removal of an intestinal segment places the patient at risk for a
bowel obstruction, and this occurs in approximately 3% of
cases.50,60,65,130 Because many of these are due to internal her-
niation through the pedicle of the augment, Leonard and col-
leagues131 have recommended early exploration to minimize
the risk of augmentation ischemia.
The removal of a gastrointestinal segment for CUR or aug-
mentation is usually well tolerated; however, bowel dysfunc-
tion has been reported in 10% to 54% of patients.132–133
Removal of the ileocecal valve can result in problematic diar-
rhea and rectal incontinence, especially in the neurogenic
population.134–135 Loss of the distal ileum can also result in
diarrhea, in these cases due to an interruption in the entero-
hepatic circulation,136 but can usually be prevented by leaving
the distal 15 to 20 cm intact. Treatment of mild cases is usually
successful with anion-exchange resins.54
Vitamin B12 deficiency can develop in up to 35% of patients
following an 80-cm small bowel resection for a Kock
1496 PART VIII GENITOURINARY DISORDERS
pouch,137 but it had not been demonstrated following ileocys-
toplasty in the past.138–139 More recently, our group in Indiana
has demonstrated that vitamin B12 deficiency often occurs fol-
lowing standard ileocystoplasty using 25 cm or less of small
bowel.140 Of patients greater than 7 years from ileocystoplasty,
21% had low values of vitamin B12 and 41% were found to be
low-normal. Vitamin B12 levels should be routinely tested, and
if found to be low, patients can be adequately treated with oral
therapy.141
COMPLICATIONS DUE TO SECRETIONS
The acidic nature of gastric secretions can result in a hematuria-
dysuria syndrome (HDS) in 9% to 70% of patients.50,131,142–143
This troublesome complication can result in bladder spasms,
suprapubic pain, dysuria, gross hematuria, and excoriation of
genital skin. This is especially troublesome in sensate patients,
where nearly 75% will have symptoms.54,144–145 Treatment of
mild cases is with histamine blockers or proton pump inhibi-
tors144 but may also require bicarbonate irrigations.146–147
The net acid loss can result in a profound hypokalemic, hypo-
chloremic, metabolic alkalosis and can be especially severe with
a coexisting gastroenteritis.
Mucus production continues with all other bowel segments,
although less so with ileum,65,148–149 and can result in incom-
plete bladder emptying. Mucus production also predisposes
the patient to urinary tract infection (UTI) and bladder stone
formation.50 Bladder stones occur in approximately 7% to
52% of augmentations,77,150–151 with the two largest series re-
cording an incidence of 10% to 15%.151–153 The increased risk
may be due to increased levels of calcium and phosphate in mu-
cus.154 Their struvite composition implies a common etiology
with a urease-producing bacteria. The systemic acidosis com-
mon to the augmented population decreases stone inhibitors
and promotes stone growth.77 Prevention is aimed at regular
CIC and daily bladder irrigations. Treatment is amenable to
open or endoscopic means,77,150,152,155 with open surgery
reserved for the larger stones.
Bacteriuria is nearly universal in any patient performing
CIC, especially when combined with an enterocystoplasty. A
UTI most frequently presents with malodorous urine, but
symptoms may include hematuria, incontinence exacerba-
tion, suprapubic pain, or increased mucus production.50 A
symptomatic urinary tract infection is reported by Rink and
colleagues65 in 22.7% of patients with an ileal augmentation,
but in only 8% of patients with a gastrocystoplasty. Overall oc-
currence of a febrile UTI was reported to be 14%. Treatment of
asymptomatic bacteriuria is not indicated unless culture indi-
cates a urease-producing organism or a virulent organism.
However, treatment may decrease the risk of stone formation.
COMPLICATIONS CAUSED BY ABSORPTION
The use of bowel for a urinary reservoir can be associated with
profound metabolic changes due to its absorptive nature. Co-
lon and ileum readily absorb ammonium, hydrogen ion, and
chloride, and this will result in a hyperchloremic metabolic
acidosis. This is tolerated in many patients with normal renal
function156 but may require medical therapy in others. The
extent of ion absorption depends primarily on intestinal con-
tact area and length of time for contact; therefore significant
acidosis should prompt an investigation into incomplete emp-
tying.50 Although not all patients will be frankly acidotic,
nearly all will have a rise in their serum chloride levels, albeit
still in the normal range.60 The acidosis prompts mobilization
of buffers and can result in bone demineralization. Although
some believe that somatic growth impairment occurs, it re-
mains controversial. This has been demonstrated in animal
models,157–159 as well as in patients with bladder exstrophy,
where there has been a reported 15% to 20% decrease in their
overall height.160 However, in the much more prevalent mye-
lodysplastic population the clinical correlation has been more
difficult to prove. The acidosis will also result in hypocitra-
turia and increase the risk of both renal and bladder stone
formation.77
Medications such as Dilantin and methotrexate are readily
absorbed across the bowel, and levels must be closely moni-
tored.161 Glucose is also absorbed, making urinary monitoring
of hyperglycemia less reliable.162
Summary
------------------------------------------------------------------------------------------------------------------------------------------------
The description by Lapides of CIC revolutionized lower uri-
nary reconstruction. It opened the door for primary recon-
struction in children, allowing the urinary tract to remain
intact. It virtually eliminated permanent incontinent urinary di-
version with the social stigma of ostomy drainage, leaving only
rare indications for temporary incontinent diversion. The Mitro-
fanoff principle allowed for an even more aggressive approach to
achieving complete dryness in these children. Although there
have been enormous gains in independence and social well-
being in the affected children, the reconstructions have pro-
vided an entire new set of complications, some of which are
potentially lethal. It is clear that the most important aspect is
patient and family motivation. They must be willing and able
to catheterize at 4-hour intervals, daily, forever. These recon-
structions require a team approach, from the family, surgeon,
and nursing, and they require lifelong follow-up evaluation.
The complete reference list is available online at www.
expertconsult.com.
 despite the dilation (nonobstructive nonrefluxing megaureters).
Megaureters may be classified on the basis of pathophysiology
according to an international system proposed in 1977.1
When the abnormality is intrinsic in the lower ureter, the
megaureter may be a unilateral problem, especially with a typ-
ical obstructive megaureter. These cases classically feature an
abnormal terminal ureter with a variety of histologic and ul-
trastructural abnormalities.2,3 Some have referred to this ab-
normality as an adynamic segment, but it is not comparable
with the adynamic segment of lower colon in Hirschsprung
disease or the lower esophagus in achalasia. The ureteral ori-
fice often looks normal in obstructive megaureter, and true
stenosis is rarely noted. Typically, a ureteral catheter can be
passed through the segment without difficulty. The ureteral or-
ifice of a refluxing megaureter is usually abnormally dilated.
Indeed, an endoscope can often be passed readily up many
such megaureters. In primary reflux, there exists an inade-
quate valvular mechanism related to the diameter of the ureter
and the length of submucosal ureter within the bladder wall.
Ureterocele and ectopic ureter are commonly associated
with megaureter, especially in the female infant with a duplex
collecting system. Rarely, megaureter is the result of excessive
urine flow, such as that seen in diabetes insipidus. The prune-
CHAPTER 119 belly syndrome features abnormal development of the entire
urinary tract including the ureters, which are typically elon-
gated and tortuous. Microscopic section of those ureters
may show decreased musculature and excessive connective
Megaureter and tissue in the ureteral wall, often affecting the lower ureter more
than the mid or upper segments.
Megaureters are typically found in two patient populations.
Prune-Belly Newborns with the problem may be identified in follow-up of
antenatally detected hydronephrosis. The relative incidence of

Syndrome megaureter compared with other congential upper tract

anomalies is higher in this group than in older children pre-
senting with clinical problems. The most common presenta-
Mark C. Adams and W. Hardy Hendren III tion in older children is urinary tract infection, although
some children may have intermittent pain from obstruction.
Infants may show failure to thrive.
Whenever megaureters are encountered during urologic
evaluation, the entire urinary tract must be evaluated to deter-
mine whether they are a primary or secondary problem. The
Megaureter kidneys and ureters are often first visualized by ultrasonogra-
------------------------------------------------------------------------------------------------------------------------------------------------
phy. Intravenous pyelography may be useful; however, the di-
Megaureter describes a ureter that is abnormally dilated. The agnosis of obstruction is most often made by nuclear imaging
ureter can sometimes also be greatly elongated and tortuous. with diuretic washout. The drainage curve after diuresis is in-
In general, the two principal pathologic types of megaureter formative, but care must be taken that the area of interest in-
are those that are obstructed and those with massive vesicour- cludes the entire upper system (ipsilateral kidney and ureter).
eteral reflux (Fig. 119-1). There may be both reflux and ob- Drainage of radionuclide out of the kidney into ureter and not
struction present, as in a megaureter ending ectopically in bladder can lead to a false-negative examination. Particularly
the bladder neck or urethra, which may be obstructed during in newborns evaluated for antenatally detected hydronephro-
the resting state but allow reflux during voiding when the sis, the function of the kidney should be considered in inter-
bladder neck opens. In primary cases, the bladder is normal preting nuclear renography, and significantly decreased
and the abnormality is located in the distal ureter. In other relative function is generally considered an indication for sur-
cases, megaureter may be secondary to bladder dysfunction. gical intervention in an otherwise asymptomatic child. Any
For example, impressive megaureters are seen in some males decline in function (>10%) on serial renography is a strong
with high-grade urethral obstruction from posterior urethral indication for repair. Antegrade pyelography performed by
valves. Ablating the valves can decrease intravesical voiding inserting a needle into the kidney and infusing contrast me-
pressures and allow compensation of the ureters in time. Sim- dium is occasionally helpful in defining anatomy and demon-
ilarly, a neuropathic bladder secondary to meningomyelocele strating differential pressures between the upper tract and
can cause physiologic obstruction of the bladder outlet with bladder (Whitaker test). Voiding cystourethrography is man-
secondary megaureter. Occasionally, megaureters may repre- datory to determine if reflux is present and will demonstrate
sent a dysmorphic anomaly without obstruction or reflux the ureteral anatomy if there is reflux. In addition, in males it
1497
1498 PART VIII GENITOURINARY DISORDERS
Megaureter
Obstructive
FIGURE 119-1 Types of megaur-
eter and scheme of lower repair.
will demonstrate any urethral pathology. Urodynamic study
may be a valuable adjunct in assessing these cases if any blad-
der abnormality is suspected. Temporary placement of an in-
dwelling urethral catheter may also differentiate between
primary and secondary megaureter on serial imaging. Cystos-
copy should define the appearance of the orifices but is gen-
erally only performed at the time of reconstructive surgery. As
noted, the orifices in obstructive megaureter usually look nor-
mal; in refluxing megaureter, they are usually widely dilated
with a “gopher hole” appearance.
The advent and widespread use of intrauterine ultrasonog-
raphy has allowed early diagnosis of many cases of megaureter.
Affected babies should have selective urologic investigation
after birth. Obstructive megaureters without reflux may im-
prove spontaneously. Thus in asymptomatic infants with mild
or moderate obstructive megaureter and normal relative renal
function, serial observation may be considered for a period
of time to see whether spontaneous improvement occurs.4,5
If improvement is not clear cut or if the patient develops
symptoms, operative repair should be performed. Retrovesical
ureteral diameter is a reasonable predictor of the likelihood
of resolution of hydronephrosis or the necessity of surgical in-
tervention in the absence of reflux. This is helpful in counseling
parents.6
If the primary problem is a neuropathic bladder, anticho-
linergic therapy and intermittent catheterization may result
in improvement in ureteral dilation. If the patient has severe
urethral valves, fulgurating the valves may suffice. Valves
occur in a spectrum of severity.7 Thus treatment may be valve
fulguration in some patients, valve fulguration followed by
early megaureter repair in others, and, occasionally, valve
ablation and immediate megaureter reconstruction.8
In summary, indications for repair of a primary mega-
ureter vary with the etiology. In cases of primary obstructive
Repair
Diverticulum
is common
Lower ureter
shortened,
tapered,
reimplanted
with long tunnel
Refluxing
megaureter, poor renal function (35% to 40% of total function
by renography), a severely scarred kidney, symptoms, or de-
creasing function on serial studies are generally considered as
clear indications for intervention. Bilateral megaureters or one
involving a solitary renal unit threaten total renal function and
should be treated more aggressively. Finally, failure to improve
after a reasonable period of observation may, at times, be an ac-
ceptable indication for surgical repair. Primary reflux into a
megaureter, except in a newborn with normal renal function,
is unlikely to resolve and usually will eventually require repair.
Febrile urinary tract infection is another indication for surgical
correction of primary megaureter whether reflux or obstruction
is present. Secondary reflux may resolve with treatment of blad-
der outlet obstruction or bladder dysfunction. Intervention,
when indicated, is virtually always open repair because the re-
sults are good even when dealing with large ureters in neonates.
Endoscopic dilation followed by temporary internal stenting
may result in temporary improvement in function or drainage
but is unlikely to be lasting in effect.9 Diversion by cutaneous
ureterostomy has occasionally been used for an infected system
or when renal function in a neonate is so poor that it is not
clear whether repair or nephrectomy is more appropriate,
or in premature infants who are symptomatic or have signif-
icantly compromised renal function. In those settings, tempo-
rary percutaneous nephrostomy drainage may be just as
beneficial and less morbid. As such, cutaneous ureterostomy
is rarely necessary.
SURGICAL TECHNIQUE
In most cases, only the lower ureter requires repair.10–14 Al-
though the upper ureter may be tortuous, it often straightens
out in time if the lower ureter is repaired and obstruction or
reflux is relieved. This type of surgery requires gentle and

meticulous technique based on the principles of standard anti-
reflux repair. The ureter should be handled minimally with
forceps to avoid injury to its delicate wall. The ureter should
be mobilized enough to taper and reimplant it into the bladder
without angulation or tension. Its blood supply must be pre-
served. The ureter should enter into the bladder through the
fixed posterior wall at a point at which it will not be obstructed
or angulated when the bladder fills. The ratio of tunnel length
to the diameter of the ureter must be high enough to prevent
reflux, usually 4 or 5 to 1.
Repair of the Lower Megaureter
The surgical steps are shown in Figure 119-2. A transverse
suprapubic skin incision is usually made, and the midline fas-
cia opened vertically. A vertical fascial incision may be prefer-
able to give easy access not only to the bladder but also the
middle and lower third of the ureter. Exposure or mobilization
of the middle third of the ureter is rarely necessary in routine
cases but may be accomplished by medial reflection of the
colon. The Denis Browne universal retractor provides ideal
exposure.
As shown in Figure 119-2, A, a catheter is sewn into the
ureteral orifice for manipulation. After the orifice is circum-
scribed, ureteral mobilization begins intravesically and con-
tinues until mobilization is no longer easily accomplished
through the bladder hiatus. Muscular and vascular attach-
ments to the ureter within the intramural tunnel are divided
and then cauterized on the bladder side but not the ureter.
Shifting outside the bladder and dissecting up along the hypo-
gastric vein, the lateral umbilical ligament is encountered and
divided to allow upward and medial retraction of the perito-
neum. After dissection cephalad along the anterior aspect of
the ureter, it can be pulled through the hiatus and exposed
in the gutter. Paravesical dissection, which can injure the
bladder nerve supply, should be minimal. As the ureter is
mobilized upward, all periureteral tissue should be swept
toward the ureter to maintain its collateral blood supply
(Fig. 119-2, B). In dissection of the ureter from peritoneum
located medial to it, the peritoneum, not the ureter, should
be skeletonized. In the male, the vas deferens should be
identified where it lies on the peritoneum. Mobilization of
the ureter is often easier if it is filled with saline.
After careful mobilization, periureteral tissue is under-
mined on the lateral aspect of the ureter and then opened
to expose the wall to be tapered (Fig. 119-2, C). During the
first 20 years of experience with repairing megaureters, which
began in 1959, special megaureter clamps were used to ex-
clude redundant circumference as an aid in tapering ureters.
Some surgeons continue to find them useful. In more recent
years, we have not used such clamps because the vascular
periureteral tissue can be laid back, preserved, and then
closed over the trimmed ureter more easily without clamps.
The segment to be tapered is marked and trimmed appropri-
ately (Fig. 119-2, D). Excision of redundant ureteral wall pos-
teriorly places the eventual ureteral closure against the bladder
muscle rather than next to mucosa and can usually be
achieved. Occasionally, however, the dominant vessels in
the ureteral adventitia are found there and should be avoided
and preserved. It is important not to make the ureter too nar-
row, and it should be noted that closing the ureter uses 3 to
4 mm of ureteral circumference. Closure should be performed
over a 10-Fr catheter except in infants, in whom 8-Fr may
CHAPTER 119 MEGAURETER AND PRUNE-BELLY SYNDROME 1499
be more appropriate. Closure should never be tight over the
catheter. The tapering should begin proximally enough that it
begins well cephalad to the eventual bladder hiatus. The taper
should begin gradually to avoid creation of an outpouching in
the ureter that can behave as a diverticulum.
The ureter is closed with a running, locking suture
(Fig. 119-2, E). It is helpful to interrupt the last few sutures
in order to allow shortening, if necessary, during reimplanta-
tion. The periureteral tissue that had been saved is then closed
to seal the primary suture line. Both layers should be closed
with fine absorbable suture. It can be helpful to retain the dis-
tal tip of the ureter as a handle, trimming it off after placing the
first anchoring sutures during reimplantation. The adynamic
segment of an obstructed megaureter should be excised in this
manner.
Selection of the new hiatus through which to bring the ure-
ter (Fig. 119-2, F) is of prime importance. One mistake is to
make the hiatus too lateral, which can result in angulation of
the ureter and obstruction when the bladder fills. If the hiatus
is located in the back wall of the bladder, it will not do so. The
new hiatus must be large enough that it does not compress the
ureter. Incising the lower rim of bladder muscle at the new hi-
atus helps create a smooth course for the ureter to enter the
bladder. Care must be taken to avoid injury to the vas in males
at this point in the repair.
After the ureteral hiatus has been prepared, the mucosa can
be opened widely to prepare a bed in which to place the ureter
(Fig. 119-2, G). Opening the mucosa is often more accurate
than tunneling beneath intact mucosa. There is no disadvan-
tage from this approach if the suture line in the ureter lies pos-
teriorly against bladder muscle. In some bladders (such as in
the prune-belly syndrome), elevating mucosa is difficult and
requires a tedious sharp dissection. The old muscular hiatus
is closed precisely. If a paraureteral diverticulum is present be-
fore surgery, the bladder mucosa must be dissected free from
the original hiatus to eliminate the diverticulum and the
muscular defect closed.
The ureter is then sutured in its bed (Fig. 119-2, H). The
distal anchoring sutures at the orifice should include trigone
muscle and mucosa and full thickness of ureter. The remain-
der of the orifice is completed with mucosal sutures. The ta-
pered back wall of the ureter lies against the muscle of the
bladder.
Figure 119-2, I shows the completed repair. The trigone
mucosa is closed over the tapered ureter. A 5-Fr feeding cath-
eter is passed through the opposite bladder wall and up the
operated ureter to the kidney for 10 to 12 days of drainage af-
ter surgery. Alternatively, double J stents can be used, with the
disadvantage of often requiring an additional anesthetic for re-
moval. Nephrostomy drainage is generally not necessary. In
patients with bilateral megaureters, we routinely repair both
lower ureters simultaneously, although this involves a long op-
eration. The technique of megaureter repair as described ear-
lier is our preference for these cases. Some surgeons favor a
cross-trigonal reimplantation technique, but it should only
be considered if the bladder floor is wider than long so as
to provide a long intramural tunnel. Refluxing megaureters
have been repaired extravesically as well with the same success
rates.15,16 Likewise, primary megaureters have been repaired
laparoscopically with and without robotic assistance17 but
should only be approached in that manner if the same surgical
principles can be applied equally well.
1500 PART VIII GENITOURINARY DISORDERS

Retracting Skeletonize
1 Initial peritoneum peritoneum,
Hypogastric
mobilization art. and vein not the ureter
is intravesical
Pe
rito
ne
um
2
Lateral umbilical
ligament is divided
Paravesical web

4
…and ureter
pulled
extravesically
3 P.V. web intact
Scissor dissection Divide attachments far from
anterior to ureter Ureter filled
with saline ureter, sweeping periureteral
A B tissue toward it

Peritoneum
Periureteral
tissue
opened
... and mark segment
to excise (not too much!)
Vas

Bla
dde
r

Ureter mobilized with Excise

all surrounding tissue Spermatic longitudinal
vessels strip

... and periureteral tissue Trim further

C undermined and opened D if needed

Pull up on Midline Peritoneum

Tip partially traction
transected
Close ureter with
running, locking suture
New hiatus
(avoid peritoneum)
Last few sutures
interrupted
Close periureteral tissue
as second layer

E F
FIGURE 119-2 Repair of the lower ureter. A, Initial intravesical mobilization. B, Extravesical mobilization. C, Opening periureteral tissue. D, Trimming of
lower ureter. E, Closing lower ureter. F, Preparing new hiatus.
 CHAPTER 119 MEGAURETER AND PRUNE-BELLY SYNDROME 1501

Open
mucosa New hiatus Close
widely (lower rim incised) old
hiatus
Old hiatus

G
H First suture
placed

New
hiatus

Mucosa
closure

Deep anchoring
sutures
New hiatus
in back wall

Reimplant with
Cath to kidney suture line in back
I
FIGURE 119-2—CONT’D. G, Creating bed for tapered reimplant. H, Suturing tapered ureter in its tunnel. I, Completed repair.

For bladder drainage after the repair, it is best to use a Lembert sutures along the clamp marks (Fig. 119-3), which
straight rather than Foley catheter through the urethra so that served to imbricate the ureteral wall and reduce lumen size.
the balloon does not compress or irritate the ureteral repair. Kalicinski and colleagues19 placed a running horizontal mat-
Only rarely is a suprapubic tube used, for example, in a patient tress suture, which excluded the redundant portion of lumen.
undergoing incontinence surgery on the bladder outlet simul- A second running suture was then used to fold the accessory
taneously. Ten to 12 days after surgery, contrast studies are per-
formed by injecting the ureteral catheters. This testing rules
out a leak, which should be rare. The catheters are then re-
moved, one side at a time if both sides have been repaired.
Measuring the urine output after each ureteral catheter is
pulled indicates whether each kidney is draining.

Imbrication
The classic description of megaureter repair involves excision
of a part of the ureteral wall to reduce circumference, and this
is our preference. Plication or folding may be considered for
a moderately dilated ureter, although such techniques may
create excessive bulk in the wall of the ureter when severely
dilated or thickened. In either type of plication, atraumatic
clamps are briefly placed onto the ureter to mark redundancy
around an appropriately sized catheter. Again, the clamps
FIGURE 119-3 Ureteral plication performed with small interrupted
should be placed to avoid major blood supply in the adventi- absorbable sutures placed in Lembert fashion. (From Keating MA, Retick
tia. After removal of the clamps, the ureteral wall is plicated AB: Management of failures of ureteroneocystostomy. In McDougal WS (ed):
by one of two techniques. Starr18 described placement of Difficult Problems in Urologic Surgery. Chicago, Year Book, 1989, p 131.)
1502 PART VIII GENITOURINARY DISORDERS
FIGURE 119-4 A running horizontal mattress suture isolates redundant
lumen. The redundant portion is then folded and approximated to the
main ureteral wall. (From Kalicinski ZH, Kansy J, Kotabinsia B, et al: Surgery
of megaureters: Modification of Hendren’s operation. J Pediatr Surg
1977;12:183.)
channel over against the wall of the ureter (Fig. 119-4). His-
tologic examination demonstrates that the folded segment un-
dergoes progressive obliteration over months, although the
second lumen remains patent.20
Endoscopic Injection
Since early reports of subureteric injection for reflux using
polytetrafluoroethylene (Teflon),21 endoscopic treatment has
been used for refluxing megaureters. The procedure is not in-
dicated for obstructive megaureters. Higher grades of reflux
have proven harder to stop endoscopically just as they are with
open surgery. The success rate with a single injection for reflux
A B
FIGURE 119-5 Endoscopic injec-
tion. A and B Orifice with hydro-
distention. C, Needle for injection
inserted in intraureteric submucosal
space. D, Coaptation after injection.
(Pictures courtesy Anthony Calda- C D
mone, MD.)
may approach 60% to 80% with a normal-sized ureter. The
failure rate increases significantly by as much as 20% or more
when treating a refluxing megaureter, although success may
increase with a second or third injection.22 Recurrence rates
are relatively low with long-term follow-up.23 Alternative bio-
compatible, biodegradable materials have been evaluated with
the most extensively used material being a dextranomer/hya-
luronic acid copolymer (Deflux, Oceana Therapeutics, Inc.,
N.J.).24 The material is easier to use and inject than Teflon
and has been used with similar effect for all grades of reflux
including megaureters.25–28 Ureteral hydrodistention at the
time of injection and use of an increased volume of injected
material within the submucosal intraureteral space may result
in more effective placement and less caudal migration of the
bioimplant (Fig. 119-5).26,29 Endoscopic injection has been
used for complex patients with reflux including those with
secondary reflux, duplication anomalies, and reflux after
transurethral incision of ureterocele.27,29 The least effective
site for injection appears to be reflux into ectopic ureters.
Open dissection of a ureter previously injected with any ma-
terial may be more difficult if necessary. In general, the ureter
with bioimplant should be mobilized together initially and the
distal involved segment of ureter then excised.
Direct repair of megaureters has worked well for 45 years in
our experience, and we continue to prefer it. The few cases in
which we have used injection therapy with a dilated ureter are
megaureters that have some residual reflux despite previous
surgery and a long intramural tunnel.
Repair of the Upper Ureter
If the primary pathophysiology was ureterovesical obstruction
or severe reflux alone, upper ureteral obstruction after lower
repair will be rare. Several points should be stressed regarding
the upper ureter. First, tortuosity can considerably improve

A B
FIGURE 119-6 Refluxing megaureters. A, Massive preoperative reflux at age 4 weeks. Previously placed nephrostomy tubes were unnecessary.
B, Intravenous pyelogram 13 years after bilateral megaureter repair. Note proximal improvement after distal ureteral repair alone.
after successful repair of the lower ureter that was once
obstructed or refluxing. Kinks can straighten, and dilation
can recede (Fig. 119-6). If the patient is doing well clinically,
it is best to wait for several months or longer to make any de-
cision about proximal ureter. The upper ureter may be dilated
immediately after repair of the lower end because of edema
surrounding the reimplantation. Imaging studies of the kidney
should be repeated to rule out a true obstruction. If the upper
ureter remains persistently tortuous and dilated from docu-
mented obstruction, it should be tailored as shown in
Figure 119-7. The tailoring sometimes merely resects a kink,
which can cause poor drainage, particularly at the ureteropel-
vic junction. In extreme cases, tailoring includes reducing the
caliber of the upper ureter. The ureter is incised longitudinally,
in situ, and its width is trimmed appropriately. Clamps are not
necessary to taper the upper ureter. Temporary urinary drain-
age with a nephrostomy catheter or indwelling ureteral stent
should be used.
It is interesting to watch a dilated ureter fluoroscopically. If
the walls of the ureter cannot coapt, active peristalsis may be
present but ineffective. The urine churns back and forth with
ureteral peristaltic waves instead of being passed effectively
into the bladder. Figure 119-8 shows a ureteral peristaltic
study in a dilated ureter. Before surgery, peristaltic waves
could be seen at various levels of the ureter, but they were
of low amplitude and ineffective. After surgery, peristaltic
waves in the tapered lower 10 cm of ureter were excellent
and effective but not in the persistently dilated ureteral seg-
ment 15 cm above the lower end of the ureter. In some cases,
the upper ureter has been observed for a long time before
being tapered because of continued dilation and poor empty-
ing. In one case in which the upper ureter was repaired 8 years
CHAPTER 119 MEGAURETER AND PRUNE-BELLY SYNDROME 1503
after the lower ureter was repaired, the upper tract hydro-
nephrosis improved remarkably and creatinine clearance
increased substantially.
COMPLICATIONS
Surgical success without obstruction or reflux when reim-
planting a normal-sized ureter into a normal bladder should
approach 97%. The success rate of megaureter repair is not
quite as high. Figure 119-9 shows the principal complications
that can occur with repair of the megaureter.30 When the or-
ifice is too lateral, the tunnel too short, or the distal ureter too
broad, reflux can result. A rare cause of reflux is a fistula at the
top of the ureteral tunnel potentially caused by a stiff ureteral
stent that can erode the ureter at the point at which it turns to
exit the bladder. A soft No. 5 infant feeding tube should pre-
vent this problem. The most common cause of persistent post-
operative obstruction is by compression of the ureter at its
new hiatus through the bladder wall often due to a hiatus that
is located too far laterally on an unstable part of the bladder.
This complication underscores the importance of having the
new hiatus in the back wall of the bladder, not its side wall.
In some cases, fixation of the bladder and hiatus with a psoas
hitch is helpful to prevent this complication. A diverticulum is
sometimes seen at the site of the new hiatus if it is left too large.
The complication of fibrosis of the distal ureter can be avoided
by meticulous handling and preservation of the blood supply
to the ureter during mobilization, not making it too narrow
when trimming it, and avoiding the temptation to taper so
extensively that the blood supply to the lower ureter is
jeopardized.
 Subcostal flank incision

Rib 12
Initial
transection

A Kidney

C
B Mobilize tortuous ureter,
preserving blood supply medially Second transection
Anastomosis of ureter
begun
Open ureter
longitudinally

Forceps

FIGURE 119-7 Straightening and Excision

of edges
tapering of the upper part of the D
ureter. Trimming the upper ureter
Traction
can improve its emptying and is per-
suture
formed in one quarter of cases in
our experience, particularly those
that remain tortuous and show evi-
dence of obstruction, especially at
the ureteropelvic junction. In some,
the upper ureter remains dilated
and does not empty well. E

Megaureter Pre-repair:

10 cm above 5 cm ⫻ 2 3 cm ⫻ 2
orifice ⫻ 2

After Lower Repair (Lower 10 cm of ureter):

15 cm ⫻ 1 10 cm ⫻ 1 5 cm ⫻ 1 3 cm ⫻ 1 2 cm ⫻ 2
(above tapering)

Waves of 2–3 mm Hg every 10–20 seconds

in tapered segment of ureter
FIGURE 119-8 Peristaltic study. Peristaltic waves in the lower ureter were ineffective before surgery but were excellent after the lower ureter was tapered.
Peristalsis remained ineffective in the dilated upper ureter above the level of tapering at the time of this study.

Reflux
2
1 Fistula
Orifice too lateral
Obstruction
4
Angulation 5
Diverticulum
B
FIGURE 119-9 The most common complications in failed surgery for megaureter.
In an experience with tapering 404 ureters in 294 patients
(217 male and 77 female), which included 75 infants younger
than 1 year of age, the success rate was 93% in obstructive
megaureter but 83% when the problem was massive reflux.
Success was higher with a normal bladder (130 patients) than
with a markedly abnormal bladder (164 patients), such as in
cases with urethral valves, neurogenic bladder, ureterocele,
prune-belly syndrome, and exstrophy. Most patients who
underwent a second operation on the ureter eventually
achieved success (Fig. 119-10). Others have reported similar
results,12,19,31,32 and problems may occur whether excision
or folding techniques are used. Consistently, problems with
secondary reflux and obstruction have been noted more com-
monly after megaureter repair for reflux variants as compared
with obstructive ones. Increased collagen deposition and abnor-
mal collagen to muscle orientation have been noted in refluxing
megaureters relative to obstructive ones above the distal ady-
namic segment.33 Such intrinsic differences may be responsible
for the difference in results for the two types.
CONCLUSIONS
The function of the ureter is to transport urine from the kidney
to the bladder at low pressure. It cannot do this effectively
when the ureter is so dilated that its walls do not coapt with
peristaltic waves. When ureters store urine rather than trans-
port it, there is little question that stasis favors infection. This
is particularly true if an inadequate ureterovesical junction al-
lows reflux of urine. Megaureters may be repaired with con-
sistent, good results. The large ureter should be handled
minimally and mobilized carefully so as to preserve blood
CHAPTER 119 MEGAURETER AND PRUNE-BELLY SYNDROME 1505
3
Tunnel too short
6
Fibrosis
supply. The distal adynamic segment of ureter when obstruct-
ing should be completely excised. More proximal dilated ure-
ter can then be tailored to normal size. We favor excisional
tapering to achieve this, although folding techniques may
be useful if the ureter is less dilated or thickened. The recon-
structed ureter can then be reimplanted into bladder using the
principles of standard antireflux surgery. The results of such
repair are slightly more prone to complications of obstruction
and, particularly, persistent reflux than are those done for
reflux into a ureter that is not initially dilated. Those compli-
cations are also more likely to occur when repair is performed
for reflux rather than obstruction. Longer follow-up and
further experience at more centers should help define the
role of endoscopic injection in the treatment of refluxing
megaureters.29
Prune-Belly Syndrome
------------------------------------------------------------------------------------------------------------------------------------------------
Some of the most unusual and impressive megaureters are
found in patients with the prune-belly syndrome, a term used
by William Osler in 1901 to describe the appearance of the
abdominal wall in patients with congenital deficiency of the
abdominal wall musculature.34 The skin usually has an irreg-
ularly wrinkled appearance similar to that of a prune. The con-
dition has also been called the triad syndrome because there
are three major features: deficiency of abdominal muscles,
hydroureteronephrosis, and cryptorchidism.35,36 Its cause is
unknown. It has been suggested that early urinary tract ob-
struction37 or urinary ascites38 might be causative or that a
primary mesodermal error might lead to the abdominal wall
1506 PART VIII GENITOURINARY DISORDERS
Before
Prior ops.:
3 bilat. reimplants
2 Bl. Neck revisions
Reduction cystoplasty
Internal urethrotomy
Obstructed
ux
Refl
Tic
Type I valves
FIGURE 119-10 A 14-year-old boy presented after three previous ureteral reimplantation operations but was left with a short obstructed megaureter on
the right and massive reflux on the left. The left ureter was satisfactory for tapering and a fourth reimplant, and extra length was gained by sliding the kidney
down and hitching the bladder upward.81 Blood supply to that ureter was enhanced by maintaining the spermatic vessels with it. The right side was drained
by transureteropyelostomy.
and genitourinary findings.39 Almost all patients are male, but
the condition has occurred in females.40 We have cared for 48
patients with prune-belly syndrome since 1968, only 3 of
whom were female. It has been described in a pair of twins,41
one infant from a pair of twins, and in consecutive siblings
with a mosaic chromosome abnormality.42 Such cases suggest
a genetic influence, but no clear pattern of inheritance has
been identified. Like most pathologic entities, the prune-belly
syndrome ranges from mild to severe in the degrees of both
abdominal wall and urinary tract abnormality.
ABDOMINAL WALL
The abdominal walls of four patients with the prune-belly syn-
drome are shown in Figure 119-11. In a typical case the ab-
dominal wall is lax and protuberant, and close inspection
shows coarse wrinkling of the skin. The abdominal wall
may be so thin that loops of dilated ureter and intestine are
easily visible and peristalsis can be seen or palpated. A protu-
berant abdomen and flaring rib margins are common, and the
lower sternum is often depressed. As the child becomes older,
the skin often becomes smoother; however, the lower abdo-
men may be remarkably protuberant in a potbellied appear-
ance. Histologic examination of the abdominal wall has
revealed both absence and hypoplasia of muscle, particularly
in the lower, central area.43 Normal relationships of muscle
groups may be preserved, or they may be fused as a single fi-
brotic layer. Innervation does appear to be intact.44 Poor sup-
port of the lower chest wall leads to an ineffective cough
mechanism in many of these children and may make them
After
Transuretero-
pyelostomy
Lower pole rotated
medially and pexed Kidney moved
Spermatic down and
vessels pexed
6 cm cross trigone
tunnel
Spermatic
Psoas vessels
hitch divided
Tic repaired
Catheter to
each kidney
TUR valves
more prone to respiratory infections. The lax abdominal wall
may also contribute to poor bladder and bowel function. Inter-
estingly, wound healing does not appear compromised and the
appearance and function of the abdominal wall may be im-
proved by surgical repair. The earliest and most simple tech-
niques involved a midline incision and closure after
trimming redundant skin and thinned abdominal wall from ei-
ther side. Randolph and colleagues45 described a nearly trans-
verse incision extending from the tip of the lowest rib on either
side to the symphysis pubis with central excision. Techniques
reported by Ehrlich46,47 and Monfort48,49 (Fig. 119-12) pre-
serve a central musculofascial strip, which is overlapped by bet-
ter lateral fascia and skin. Both provide adequate exposure for
orchiopexy or genitourinary reconstruction at the same time.
KIDNEYS AND URETERS
Renal architecture and function vary widely in prune-belly
syndrome. In some cases the kidneys are severely dysplastic
and the infants die soon after birth of renal failure
(Fig. 119-13).50,51 At the other end of the spectrum are pa-
tients with relatively normal kidneys. Some element of renal
dysplasia is found in more than half of patients with prune-
belly syndrome,51 and the severity is a major determinant of
patient prognosis. The degree of dysplasia may vary greatly
from side to side in the same patient. It tends to be worse
in patients with bladder outlet obstruction (urethral atresia)
and imperforate anus.52
This wide variability in severity is seen not only microscop-
ically in the renal parenchyma but also in the collecting system

A B
C D
FIGURE 119-11 Variable appearance of the abdominal wall in the
prune-belly syndrome. A, Neonate with typical features of severe prune-
belly syndrome. Note the sagging flanks and loops of bowel visible be-
neath extremely thin abdominal wall with deficient musculature. B, Mild
case of prune-belly syndrome in an 8-week-old boy. Note the slight wrin-
kling of the skin of the lower abdomen and underlying distended bladder.
The abdominal wall is lax but more developed than the patient shown in
A. C, An 18-year-old patient with typical protuberant abdomen seen
in prune-belly syndrome. Note inward protrusion of the lower sternum,
which is common in these patients. A long-standing nephrostomy and
cystostomy have been present since infancy. The urinary tract was subse-
quently reconstructed. D, A 16-year-old patient with severe wrinkling of
the abdominal wall. The patient had long-standing end ureterostomies
(arrow). Marked kyphoscoliosis is present. An undiversion operation was
performed subsequently.
of the kidney, which can be dilated and have a peculiar dys-
morphic appearance (Fig. 119-14). The ureters are character-
istically dilated, elongated, and tortuous to a much greater
degree than is seen in males with even high-grade obstruction
from urethral valves. Some patients have scant muscle in the
ureteral wall; others have well-developed muscle. In general,
the upper ends of the ureters have better preserved architec-
ture than do the lower ends.53 Collagen is often increased and
muscle mass decreased in the ureteral wall, especially when
there is massive reflux.54 In some patients the ureteropelvic
junction is extremely tortuous but not obstructed. In others,
ureteropelvic junction obstruction is present. Peristalsis is
poor in some ureters and active but ineffective in others be-
cause the walls of the ureters cannot coapt. Repeat urinary in-
fection can lead to inflammation and further fibrosis of the
ureter, aggravating an already compromised situation. The
hydronephrosis at the level of the kidney is often less remark-
able than that of the distal ureter and bladder, and the paren-
chyma is sometimes better preserved than might be expected
from the degree of distal dilation. Up to 85% of patients with
CHAPTER 119 MEGAURETER AND PRUNE-BELLY SYNDROME 1507
prune-belly syndrome have vesicoureteral reflux.55 The reflux
may be difficult to grade because of the dysmorphism but of-
ten involves a massive volume of urine that may interfere with
bladder dynamics and make the patient prone to urinary tract
infection.
BLADDER
In these patients the bladder usually has a large capacity. Most
bladders feature a large diverticulum-like dome involving the
area of the urachus (see Fig. 119-14). In those patients with
urethral atresia, the dome may drain through a patent urachus
at the umbilicus. Endoscopy shows that the wall of the bladder
is usually smooth, not trabeculated. The ureteral orifices are
often lateral and somewhat craniad. The orifices are often
widely dilated, consistent with the massive reflux seen in most
cases. The bladder neck is abnormally wide and funnels down
into a dilated prostatic urethra. Grossly, the bladder wall is
usually thick and quite vascular. The thickness of the bladder
wall is not the result of hypertrophied muscle. Histologic ex-
amination reveals muscle and abundant collagen and fibrous
tissue. The bladder will usually function as an adequate, if not
enlarged, reservoir with good capacity and compliance, but it
may not empty efficiently. Massive reflux, diminished detrusor
contractility, decreased sensation of fullness, poor abdominal
wall support, and diverticulum-like malformations at the
dome all may serve to prevent normal emptying. Despite those
problems, about half of such patients can void spontane-
ously.56 Both spontaneous improvement and deterioration
in the quality of voiding have been noted. Reducing the size
of the bladder by excising the dome should be considered
in some cases.57,58 Domectomy can improve emptying and
decrease postvoid residual urine in select cases.80 Some chil-
dren with prune-belly syndrome benefit from bladder empty-
ing by intermittent catheterization.56
URETHRA
In males with prune-belly syndrome, the dilation of the pos-
terior urethra may be secondary to congenital deficiency of
surrounding prostate tissue. The prostatic urethra appears
heart shaped on voiding cystography. The external urethral
sphincter is just below the apex. The narrowing of the pros-
tatic urethra at the level of the external sphincter does not nec-
essarily denote urethral obstruction, but some male patients
have urethral atresia.59 Those with urethral atresia who sur-
vive usually have a patent urachus that affords decompression
of the urinary tract and allows for amniotic fluid needed for
pulmonary development. Congenital urethral stenosis is also
sometimes present. In the 45 males we have treated, true ure-
thral valves were seen in 6; 20 had valvelike narrowing, usu-
ally at the level of a rudimentary verumontanum. Patients with
poor bladder emptying may benefit from incision of such folds
(Fig. 119-15). Besides atresia, the anterior urethra may be ab-
normal in the form of congenital megalourethra, the fusiform
variant of which may distort development of all corpora in the
phallus. Anterior urethral anomalies generally require surgical
repair when present. Atresia may be repaired by open urethro-
plasty (Fig. 119-16), although progressive dilation may at
times be effective.60 Megalourethra is corrected by open tailor-
ing of the urethral mucosa, which can be covered by the
investing, periurethral tissue in several overlapping layers.
1508 PART VIII GENITOURINARY DISORDERS

A B

C D

E F
U

FIGURE 119-12 Surgical technique for Monfort abdominoplasty.48,49 A, Delineation of redundancy by tenting up abdominal wall. B, Skin incisions are
outlined with a separate circumscribing incision to isolate the umbilicus. C, Skin (epidermis and dermis only) is excised with electrocautery. D, Abdominal
wall is incised at the lateral border of the rectus muscle on either side, creating a central musculofascial plate. E, Adequate exposure is provided for con-
comitant transperitoneal genitourinary procedures by retracting the central strip laterally. F, Completion of abdominoplasty by scoring of the parietal
peritoneum underlying the lateral abdominal wall musculature with electrocautery.
 CHAPTER 119 MEGAURETER AND PRUNE-BELLY SYNDROME 1509

G H

FIGURE 119-12—CONT’D. G, The edges of the central plate are sutured to the lateral abdominal wall musculature along the scored line. H, Lateral flaps
are brought together in the midline, with closed suction drains placed between the lateral flaps and the central plate. Skin is brought together in the
midline, enveloping the previously isolated umbilicus. (From Woodard JR, Perez LM: Prune-belly syndrome. In Marshall FF (ed): Operative Urology.
Philadelphia, WB Saunders, 1996.)

TESTES
Bilateral undescended testes lie in the abdomen in most males
with prune-belly syndrome. The testes are usually smaller
than normal and lie well inside the internal inguinal ring,
much like ovaries in females. The spermatic vessels are shorter
than normal, and there is often dysjunction of the epididymis
from the testis. The vasa may be tortuous, and atresia of the
vas is not uncommon. The pelvic peritoneum adjacent to
the vas deferens usually has an abundant blood supply, which
makes the Fowler-Stephens method of orchidopexy possi-
ble.61 This technique involves dividing the spermatic vessels
in the abdomen above the testis while mobilizing the testis
with a broad strip of pelvic peritoneum and collateral blood
supply surrounding the vas deferens. The goal of orchidopexy
is to place the testes in the scrotum, where they can be mon-
itored for development of malignancy and for psychologic
reasons. Fertility has not been reported in these patients,
although Leydig function leading to secondary sexual devel-
opment with penile growth and libido is usually normal. Most
surgeons have found that bringing a high testicle into the
scrotum is easier in the young infant, in whom the relative dis-
tance is shorter.62 Laparoscopic orchidopexy has been used
for these patients63 and may be particularly useful for a staged
Fowler-Stephens technique. Access technique, insufflation
pressure, and port stabilization may have to be modified
because of the floppy abdominal wall.64

A B
FIGURE 119-13 Prune-belly syndrome in a neonate who died at 1 week of age with uremia. A, Cystogram showing large, elongated bladder with
diverticulum-like dome, often seen in prune-belly syndrome. Massive reflux into dilated, convoluted ureters. B, Postmortem specimen of urinary tract.
Note attachment of bladder to navel and cystic, dysplastic kidneys.
1510 PART VIII GENITOURINARY DISORDERS

A B
FIGURE 119-14 Typical severe prune-belly case. A, Cystogram. Note large bladder, with appearance of an inverted light bulb and massive reflux into
convoluted ureters. When straightened, the ureters were longer than the patient’s legs. B, Cystourethrogram. Note smooth wall of bladder, open bladder
neck, and heart-shaped urethra.

OTHER ANOMALIES CLINICAL MANAGEMENT

Many patients with prune-belly syndrome have musculoskel-
etal or orthopedic abnormalities that require treatment.65,66
Most of these patients have malrotation with nonfixation of
the colon, but intestinal obstruction with midgut volvulus is
extremely rare. Imperforate anus may be seen in prune-belly
syndrome. All of the females we have treated for prune-belly
syndrome have had cloacal malformations. Pulmonary hypo-
plasia is seen in some cases, and such severe cases are also
prone to pulmonary infection. Pectus excavatum is common,
and the abdominal wall musculature is weak. Various types
of congenital heart disease may occur in association with
prune-belly syndrome.
A newborn with prune-belly syndrome should be evaluated
carefully because of the potential for sepsis. Any study that in-
volves catheterization should be performed under cover of ap-
propriate antibiotics and with meticulous aseptic technique.
The dilated urinary tract, which may contain a large volume
of urine and drain poorly, does not tolerate the introduction
of bacteria. In some babies with poor renal function and dys-
plastic kidneys, early death from renal failure may be predes-
tined due to associated pulmonary hypoplasia. Measurement
of serum creatinine levels and, particularly, creatinine clear-
ance identifies infants who have a reasonable measure of
kidney function.

A B
FIGURE 119-15 Voiding urethrograms. A, Preoperative study showing typical dilated proximal urethra and wide bladder neck, with abrupt cutoff of
contrast distal to rudimentary veru. There was a urethral valve. B, Postoperative study 7 months later showing free flow and normal caliber of the entire
urethra.
 CHAPTER 119
A B
FIGURE 119-16 Cystourethrograms. A, Preoperative. Note hypoplastic urethra (small arrows) and unusual obstruction of a megaureter by ureteral valve
(large arrow). B, Postoperative study after long preputial patch graft to urethra.
Prune-belly syndrome in babies represents one of the most
controversial conditions in pediatric urology. At one end of the
therapeutic spectrum are those who advocate watchful wait-
ing and minimal surgical intervention, citing a delicately
balanced uropathy.67–69 Others favor major reconstructive
surgery when indicated.8,57,70–76 It should be emphasized that
reconstructive surgery in an infant with prune-belly syndrome
should not be undertaken unless expert anesthesiology and
medical support are available and the surgeon has extensive
experience doing such surgery. Furthermore, a major recon-
structive operation should not be considered unless the baby’s
overall status including pulmonary function and absence of
established urinary tract infection is reasonable. Woodard
and Zucker have suggested that an aggressive approach must
be considered with caution because the surgery is difficult and
the patients are vulnerable to pulmonary complications.76
Much of this controversy centers on the appropriate role for
reconstruction of the urinary tract. There is little question that
the intra-abdominal testes should be brought down by orchio-
pexy at an early age. Likewise, many patients and families will
request reconstruction of the abdominal wall, for psychologic
reasons if none other. It is our preference to perform such re-
pairs at the same time as urinary tract reconstruction if neces-
sary. If the patient will tolerate it from a pulmonary standpoint,
we favor one longer reconstruction to repair everything rather
than putting the same patient through multiple surgeries.
When initial evaluation demonstrates clear obstruction, cor-
rection should be performed whether it involves ureteropelvic
junction, ureterovesical junction, or bladder outlet (Fig. 119-17).
Most patients have some underlying renal dysplasia, and second-
ary insult from obstruction should not be allowed. It is our expe-
rience that most patients with this syndrome and vesicoureteral
reflux eventually undergo repair. Classic grading systems for
reflux may not apply to such patients; however, reflux in patients
with prune-belly syndrome is often massive in volume. Early
observation while the patient is on daily antibiotics to prevent
infection may be appropriate and allow evaluation of bladder
emptying or other issues. If reflux persists, correction is appro-
priate. When patients with reflux suffer breakthrough urinary
MEGAURETER AND PRUNE-BELLY SYNDROME 1511
tract infection or deterioration in renal function, repair of the
reflux should be undertaken.
In considering the urinary system in patients with this syn-
drome, it may be useful to categorize them into three groups as
described by Smith and Woodard.77 The categories roughly
correlate with the degree of renal dysplasia present. Group I
patients usually have both severe renal dysplasia and pulmo-
nary hypoplasia and do not survive. Urologic intervention will
not rescue such patients. Group III patients generally have a
mild or incomplete form of the syndrome and well-preserved
renal function. Although some such patients may have im-
pressive dilation of the urinary tract, little urologic interven-
tion may be required if the patients do not suffer recurrent
urinary tract infection. Definite indications for intervention
in group II patients with intermediate degrees of dysplasia
and renal dysfunction are less clear. These patients typically
have full-blown features of the syndrome and a stagnant uri-
nary system at best. Clearly, some such patients will progress
to renal insufficiency with or without surgical interven-
tion.69,78 So many variables are present in patients with
prune-belly syndrome that it remains difficult to predict the nat-
ural history for a given patient early in presentation. Careful sur-
veillance is critical, and we favor surgical intervention in group
II patients if they suffer from recurrent infection or deterioration
of renal function. Those two problems often go hand in hand.
We believe that correction of obstruction or reflux, optimizing
bladder emptying, and avoidance of infection may prevent renal
insufficiency in some patients with marginal function or, at
least, prolong the time until they suffer insufficiency. It is clear
that the scientific evidence to support such intervention, or
argue against it, is largely anecdotal at this point.
We have performed reconstructive surgery on 40 patients
with prune-belly syndrome; 12 were primary cases
(Figs. 119-18 and 119-19), and 28 were secondary cases that
had been previously treated with surgery elsewhere. Twenty-
four of the latter 28 patients had undergone urinary diver-
sion.72,79 This illustrates how often urosepsis occurs in
patients with prune-belly syndrome and has led to various
types of surgery for diversion in the past (Fig. 119-20). We
1512 PART VIII GENITOURINARY DISORDERS

A B
FIGURE 119-17 Intravenous pyelograms from same patient shown in Figure 119-16. A, Ureteral valve in right megaureter (arrow). B, Intravenous
pyelogram 3 years after extensive undiversion.

A B
FIGURE 119-18 Infant with prune-belly syndrome. A, Preoperative cystogram. There is massive bilateral reflux into elongated and tortuous ureters,
especially on the right side. Note the dysmorphic collecting system of the left kidney. Total reconstruction was performed. B, Intravenous pyelogram
at 7 years of age. The patient is well with no urinary tract symptoms. Note the contrast visible in the left ureter, which is of normal caliber. It is difficult
to imagine that the great improvement in the appearance of the urinary tract does not equate with a better prognosis for the patient.
 Before After

L. pyeloplasty via
2-day-old boy
window in
Prune-Belly Syndrome
peripelvic tissue
Good kidneys

TUU

UPJ kinked Gonad vessels

remain with ureter
Psoas hitch
Left ureter
Huge Long tunnel a. Lower end
ureter reimplant shortened and
reimplanted
b. TUU in mid ureter
c. Pyeloplasty at
Intra- upper end
abdominal Great care with
testes TUR diaphragm blood supply
(cuts at 5, 7, 12)
"Prune bladder"
(smooth, thick walled, R. testis down with Cath. to
Lateral, open orifices each kidney
fibrous, vascular) vas plus intact vessels
with massive reflux
L. testis down
Wide bladder neck on pedicle of vas

Sphincter
Diaphragm valve ar veru Cath. to bladder

FIGURE 119-19 Infant with prune-belly syndrome. Anatomy preoperatively and postoperatively.

Prune Belly Syndrome Age 10 years OP2 2 mos. after stricture repair

Poor Good Transuretero-

kidney kidney pyelostomy

Kidney moved Closure

Ureterostomies down and medially ureterostomies
since birth and pexed
Shaded areas Short spermatic vessels
Short spermatic to resect retained with ureters
vessels for collateral Tapered
Urachal Cystostomy tube
extension
Psoas
Vas hitch

Ureteral Long tunnel

valves Tube to each reimplant of
High
kidney tapered ureter
testes
Closed
High, lateral orifices. hiatus
OP1
Bladder thick, fibrous, Reflux
Repair of stricture
vascular, nontrabeculated
Veru 4 mm
(typical prune)
opened Bilat. Orchidopexy
Sphincter Patch (on pedicle of
vas only)
Ample prepuce
for graft 15 cm stricture extending 18 French Button
from tip of penis to above
membranous urethra Long, wide patch

FIGURE 119-20 1-year-old boy with prune-belly syndrome, unusual anatomic findings, and long-standing ureterostomies. With operation 1, an extensive
urethroplasty was performed with a long onlay of prepuce. Two months later, undiversion with urinary reconstruction and bilateral orchiopexies was per-
formed. The patient voids normally and is continent. Renal function has been normal (serum creatinine 0.9 mg/dL and creatinine clearance 106 L/m2 per
day). Pyelograms are shown in Figure 119-17.
1514 PART VIII GENITOURINARY DISORDERS

generally favor reconstruction over diversion in such patients selective cases by aggressive surgical reconstruction is the op-
as long as their general condition allows it. Twelve of these pa- posite of the “watchful waiting” espoused by others. Surgical
tients had only one kidney because of previous nephrectomy. intervention is actually a conservative method in that it at-
Thirty-five of the 40 patients are alive (time since surgery, 1 to tempts to decrease stasis, urinary tract infection, and renal fail-
27 years). Eight patients have had renal transplantation. Eight ure, the natural history of many patients with the syndrome.
more will probably require renal transplantation, given their Reconstruction is demanding and technically difficult in these
current renal function. Five patients have died. One infant patients. It also requires extensive anesthesiology expertise to
died in the hospital because of blood loss secondary to hemo- maintain metabolic hemostasis in a young child undergoing
philia and inadequate monitoring. Another infant died at major reconstruction. This surgery should be performed in
home 2 months after surgery of pulmonary aspiration. Three centers where surgeons concentrate on major reconstructive
late deaths occurred: one 3 years after renal transplantation; surgery and have access to all of the requisite back-up systems.
another 10 years after successful transplantation, from an au-
tomobile accident; and one 11 years after reconstruction from The complete reference list is available online at www.
septicemia during dialysis. It should be noted that this series is expertconsult.com.
a select group and does not necessarily reflect rates of death
from renal dysplasia in primary infants with prune-belly
syndrome.

Conclusions
------------------------------------------------------------------------------------------------------------------------------------------------

Some of the most impressive megaureters are found in patients

with prune-belly syndrome, and they are often then associated
with other clinical problems. The philosophy of treating
 HISTORICAL PERSPECTIVES
The earliest account of bladder exstrophy can be found on As-
syrian tablets, dating back to 2000 BC. Von Grafenberg first de-
scribed the medical condition in 1597, and Mowat is credited
with providing a complete description of bladder exstrophy in
1748. It was not until 1780, however, that Chaussier first
coined the term “exstrophie.”1 Early management of bladder
exstrophy included the application of an external urinary re-
ceptacle to the surface of the exposed bladder,2 ureterosigmoi-
dostomy,3 transplantation of the bladder trigone into the
rectum,4 and coverage of the exposed bladder with lateral skin
flaps.5 These methods were fraught with continued urinary
incontinence and/or urosepsis. Contributions by Coffey,
Nesbitt, Leadbetter, and Clarke improved the technique of
ureterosigmoidostomy; however, associated morbidities in-
cluded infection, electrolyte abnormalities, and malignancy.6
Complete urinary diversion into the colon or alternate conduit
was preferentially used to provide continence and minimize
infection, but the problems of anatomic reconstruction and
sexual function persisted.
Paralleling efforts to develop improved methods of urinary
diversion were attempts at successful bladder closure. Trende-

CHAPTER 120 lenburg described bilateral sacroiliac osteotomies and the ap-
plication of a pelvic sling to support bladder and abdominal
wall closure in 1892.7 The first case of successful closure
and continence in a female patient with bladder exstrophy

Bladder and Cloacal was not reported until 1942 by Young.8 Michon subsequently
reported successful reconstruction in a male patient 6 years
later.9 Despite these accounts of functional closure, a 1970 re-
Exstrophy view of 329 cases by Marshall and Muecke concluded that
only 19% of patients undergoing total reconstruction had
fair-to-satisfactory results.10 These unfavorable outcomes
Lynn L. Woo, John C. Thomas, and John W. Brock III were reported by others, spurring efforts to improve methods
of surgical repair.11,12

EPIDEMIOLOGY
The authors would like to acknowledge Romano T. DeMarco
and James A. O’Neill, Jr. for their previous contributions to The incidence of bladder exstrophy is estimated at between 1
this chapter. in 10,000 and 1 in 50,000 live births13 with a higher male-
to-female ratio of between 2.3:1 and 4:1.14 Familial recur-
rence is approximately 1 in 100.15 On the basis of a survey
of 2500 indexed cases, familiar occurrence was found to be
Bladder Exstrophy 1 in 275.14 Multiple reports of bladder exstrophy among
------------------------------------------------------------------------------------------------------------------------------------------------
identical twins have demonstrated variability in involvement
Bladder exstrophy is a rare midline defect and exists as part of of one or both twins. Subsequent siblings and the offspring
a larger spectrum of abdominal-pelvic fusion abnormalities, of individuals with bladder exstrophy may be at increased risk
known collectively as the exstrophy-epispadias complex of being affected.14,15 However, no clear pattern of inheritance
(EEC). Presentation of EEC can range from isolated glanular has been characterized and no specific genetic or environmen-
epispadias to cloacal exstrophy, in which several other organ tal factor that predisposes to bladder exstrophy has yet been
systems may be affected. In the case of bladder exstrophy, the identified.
open bladder can be seen everting through a lower abdominal
wall defect. This is accompanied by epispadias, a widened pu-
EMBRYOLOGY
bic diastasis, and an anus that is anteriorly displaced. Over the
past 2 decades, continued improvements in the methods of The underlying embryologic defect shared by bladder exstro-
functional bladder closure have dramatically increased recon- phy and other variants of the EEC is due to abnormal devel-
structive success rates; however, achieving the ultimate goals opment of the cloacal membrane, a bilaminar structure
of adequate bladder capacity, urinary continence, and a good composed of endoderm and ectoderm that overlies the cloacal
cosmetic outcome remain challenging. cavity at the caudal end of the germinal disk.16 In normal

1515
1516 PART VIII GENITOURINARY DISORDERS

Mesonephric
duct
Allantois

Primitive Bladder
urogenital
Phallus
sinus
Mesonephric Perineal
Cloacal duct body
membrane Ureteric bud
Urorectal
Hindgut Anorectal
Urorectal septum Ureter
A septum B canal C
FIGURE 120-1 Division of the cloaca in the urogenital sinus and rectum. A, At the end of the fifth week. B, At 7 weeks. C, At 8 weeks. (Modified from Sadler
TW: Langman’s Medical Embryology, 8th ed. Philadelphia, Lippincott Williams and Wilkins, 2000, p 316.)

development, lateral ingrowth of mesoderm occurs between
the two layers of the cloacal membrane during the fourth
and fifth weeks of gestation. This results in formation of the
lower abdominal wall and pelvis. Subsequent caudal growth
of the urorectal septum results in its fusion with the cloacal
membrane, thus fully separating the cloaca into the bladder
anteriorly and the rectum posteriorly (Fig. 120-1). The paired
genital tubercles, which will give rise to the phallus, migrate
medially to fuse in the midline. Normal perforation of the clo-
acal membrane occurs after fusion with the urorectal septum,
at approximately the sixth week, resulting in formation of
separate urogenital and anal openings.17
Migratory failure of the lateral mesodermal folds and ab-
normal overdevelopment of the cloacal membrane have both
been proposed as potential causes of the prevention of normal
mesodermal ingrowth to the cloacal membrane.16,18 The lack
of adequate mesodermal reinforcement is thought to result in
premature rupture of the cloacal membrane, the timing of
which determines the extent of the abdominal wall defect
and degree/severity of urogenital tract involvement.19 Rupture
of the cloacal membrane after fusion with the urorectal sep-
tum results in bladder exstrophy, whereas rupture before fu-
sion gives rise to the more severe presentation of cloacal
exstrophy (see later discussion).
submucosal tunnel in the bladder wall.22 With continued ex-
posure and chronic inflammation, the exstrophied bladder
becomes thickened and polypoid (Fig. 120-3). Long-term
exposure may eventually result in a fibrotic, rigid bladder
plate that is ultimately unsuitable for closure.
GENITAL DEFECTS—MALE
In the male infant, the open and everted urethral plate can be
seen joining the exposed bladder. The penis is characteristi-
cally short with a flattened, everted glans. The prepuce is lo-
cated on the penile ventrum (Fig. 120-4). The ejaculatory
ducts are typically normal and exit at the exposed verumon-
tanum in the posterior urethra. The base of the penis and scro-
tum are widely separated, with lateral displacement of the
corporal bodies and neurovascular bundles. Historically it
was believed that the individual corpora were of normal cal-
iber and appeared shortened because of their attachment to
the widened pubic diastasis and associated dorsal chordee.

CLINICAL PRESENTATION
In general infants with bladder exstrophy are born full term,
without coexisting anatomic anomalies. At birth, an everted
posterior bladder plate of varying size is seen in the midline
of the lower abdomen. The mucosa of the exposed bladder
in the newborn is typically smooth and pink. The umbilical
cord exits from the superior-most border of the bladder plate,
and a small umbilical hernia may be present (Fig. 120-2). In
addition, there is significant widening of the pubic symphysis
and the anus is anteriorly displaced. The levator ani complex
is also divergent, leading to an inherent weakness in the pelvic
floor and a tendency toward rectal prolapse and varying de-
grees of fecal incontinence. Associated inguinal hernias are
common and have been reported in 82% of boys and 10%
of girls.20 The upper urinary tract is usually normal, though
renal anomalies including ectopic, horseshoe, hypoplastic,
dysplastic kidneys, and megaureters may be observed.21 Vesi- FIGURE 120-2 Typical findings of classic bladder exstrophy in a new-
born male. The bladder plate is small, and a small hernia is evident at its
coureteral reflux occurs in the vast majority of children after superior border. The penis is foreshortened, with widely splayed corporal
bladder closure, secondary to an exaggerated lateral course bodies and glanular separation. The urethral plate is short and located on
of the ureters within the pelvis and lack of adequate the anterior surface of the split phallus.

FIGURE 120-3 When the bladder has been exposed for at least 1 week
after birth and the mucosa is subjected to continued exposure and inflam-
mation, polypoid excrescences typically appear, as in this female infant.
More recently, an MRI-based study by Silver and colleagues of
adult men with exstrophy and age-matched controls found
that although the length of the posterior corporal bodies
was the same between groups, anterior corporal length in
men with exstrophy was nearly 50% shorter than that of con-
trols.23 Therefore the penis appears shortened not only second-
ary to corporal divergence, dorsal chordee, and abnormal
crural attachments to the corpora cavernosa but also because
of an inherent deficiency of corporal tissue (Fig. 120-5). The
testes may appear to be undescended, but in most cases they
are actually retractile and will eventually reside in the scrotum
without the need for formal orchiopexy. Should it be required,
orchiopexy is performed in conjunction with inguinal hernia
repair.
FIGURE 120-4 Classic exstrophy in the male. The penis is pulled downward to expose the dorsal aspect, revealing the urethral plate leading to the
exposed bladder. 1, umbilical cord; 2, bladder mucosa; 3, paraexstrophy tissues; 4, left ureteric orifice; 5, verumontanum; 6, urethral plate; 7, glans penis.
CHAPTER 120 BLADDER AND CLOACAL EXSTROPHY 1517
GENITAL DEFECTS—FEMALE
The clitoris is bifid, with divergence of the mons pubis, labia,
and clitoral halves (Fig. 120-6). The urethra and vagina are
shortened, and the introitus is anteriorly displaced. The vag-
inal orifice is often stenotic. The uterus and adnexa are typi-
cally normal, though vaginal and uterine duplication have
been reported.24,25 Uterine prolapse occurs commonly in
female patients, secondary to the inherent weakness in pelvic
floor support.
PELVIC DEFECTS
Some degree of widening of the pubic symphysis is present in
all cases of bladder exstrophy and contributes to outward ro-
tation and eversion of the pubic rami at their junctions with
the ischial and iliac bones (Fig. 120-7). Using computed to-
mography (CT), Sponseller and colleagues further character-
ized the pelvic anatomy of a large group of exstrophy
patients, noting a significantly increased distance between
the triradiate cartilages (31%), external rotation of the anterior
pelvis (18%), and 30% shortening of the pubic rami.26 On the
basis of three-dimensional models generated by CT, Stec and
colleagues observed that among children with exstrophy, the
levator ani muscles were more posteriorly positioned and out-
wardly rotated. Furthermore, the puborectal sling had a more
flattened configuration and supported twice the body cavity
area in exstrophy patients.27 As mentioned previously, these
pelvic floor defects predispose to pelvic organ and rectal
prolapse in this patient population.
PRENATAL DIAGNOSIS
The use of prenatal ultrasound (US) and MRI has improved
the antenatal diagnosis of bladder exstrophy, allowing for ap-
propriate parental counseling and planning of postnatal man-
agement. The prenatal diagnosis of bladder exstrophy may be
suggested on US by failure to visualize the bladder in the pres-
ence of normal kidneys and amniotic fluid.28–30 In a review of
prenatal US studies from 25 women who delivered infants
1518 PART VIII GENITOURINARY DISORDERS

1
2
11
9
3
FIGURE 120-5 Penile configura-
tion in classic bladder exstrophy. 4
Normal male perineum (A) and with 7
bladder exstrophy (B). Note the loss
of the normal triangular shape of 7 8
8
the perineum and widening of the 6
pubic symphysis. In the setting of
exstrophy, the corpora cavernosa
are widely separated and are intrin- 6
sically shorter. 1, corpus caverno- 9
sum of the penis; 2, glans penis;
3, corpus spongiosum; 4, bul-
bospongiosus muscle; 5, ischium; 5 5
6, ischiopubic ramus; 7, pubis; 8,
ischio-cavernosus muscle and crus 10 10
of penis; 9, urogenital diaphragm;
10, anus and external anal sphincter. A B

with exstrophy, Gearhart and colleagues observed the follow- be accomplished in a multi- or single-stage (complete) repair.
ing features: absent bladder (71%), lower abdominal bulge Multiple contemporary approaches including the modern
(47%) and anteriorly displaced scrotum with small phallus staged reconstruction of exstrophy (MSRE) and complete
in male fetuses (57%), low-set umbilical cord (29%), and primary reconstruction of exstrophy (CPRE) along with War-
abnormal widening of the iliac crest (18%).31 saw,32 Erlangen,24 Mainz,33 and Kelly34 techniques have been
published; however, for the purposes of this review, only the
major principles of MRSE and CPRE are discussed.
SURGICAL RECONSTRUCTION
Surgical management of classic bladder exstrophy consists of
functional closure of the native bladder, closure of the epispa-
dic urethra and genitalia, and creation of a continence mech-
anism to allow for proper urine storage. Reconstruction may

FIGURE 120-6 Typical appearance of classic bladder exstrophy in a fe- FIGURE 120-7 Plain radiograph of a neonate with bladder exstrophy
male. Note the widely divergent labia and anterior displacement of the demonstrates the soft tissue mass effect of the exposed bladder, the wide
vaginal introitus and anus. diastasis of the symphysis pubis, and the posterior rotation of the
acetabula.

INITIAL MANAGEMENT
At birth, the umbilical cord should be ligated with a silk suture
to avoid irritation of the bladder surface from the traditional
plastic clip. The exposed bladder mucosa should be moistened
with saline and protected with a nonadherent sheet of plastic
wrap (e.g., Saran Wrap). A complete physical examination is
performed to rule out associated anomalies and to assess the
size of the bladder plate and extent of the genital defect.
Renal US may be obtained to exclude hydronephrosis and/or
other upper tract abnormalities. Prophylactic antibiotics
should be administered.
MODERN STAGED RECONSTRUCTION
OF BLADDER EXSTROPHY
A three-stage approach for the treatment of bladder exstrophy
was first pioneered by Jeffs and Cendron in the 1970s,35,36
and continued improvements in technique have contributed
to increased success of the procedure.37–39 Stage 1 is per-
formed at birth to protect the upper urinary tracts and assist
later continent reconstruction. It consists of early closure of
the bladder, posterior urethra, and abdominal wall with or
without osteotomy. The primary objective of functional clo-
sure is to convert the bladder exstrophy into a complete epis-
padias.40 Stage 2 is performed in later infancy and involves
repair of the epispadias, with the goal of optimizing genital
function and appearance, as well as increasing outlet resis-
tance to promote bladder growth. Stage 3 is undertaken before
school age and consists of bladder neck repair for continence
and ureteral reimplantation for vesicoureteral reflux.
Primary functional closure is generally undertaken in the
neonatal period, which offers several potential advantages.
The pliability of the pelvic ring, in infants younger than
72 hours old, may obviate the need for osteotomy; early
closure prevents further exposure and scarring of the bladder
plate; there is theoretically less opportunity for bacterial
colonization of the plate with decreased risk of postoperative
infection. Alternatively, delayed closure in combination with
pelvic osteotomy may be performed, allowing for patient
growth and a period of time out of the hospital before
reconstruction.
Stage 1: Functional Closure
At the time of surgery, the patient should be prepped widely
including the entire body anteriorly and posteriorly below the
nipple line so that intraoperative turning is easier. Cardiopul-
monary monitoring and adequate intravenous access are crit-
ical. Intraoperative and postoperative analgesia may be
afforded by means of an epidural catheter. The most common
technique of functional closure is based on descriptions by
Jeffs and colleagues36,41,42 and Duckett and Caldamone.43
Figure 120-8 depicts the initial incisional template for bladder
closure in the female infant, and Figure 120-9 details the com-
plete sequence of stage 1 closure in the male. Traction sutures
are placed into the glans penis, and ureteral catheters are se-
cured in each ureteral orifice. An incision is made around the
periphery of the exstrophic bladder plate, and a plane of dis-
section is established between the rectus fascia and bladder.
The umbilical cord is excised, and umbilicoplasty may be per-
formed during or after the initial procedure. Dissection is con-
tinued toward the pubis, and the incision is then extended
CHAPTER 120 BLADDER AND CLOACAL EXSTROPHY 1519
FIGURE 120-8 Typical findings of classic bladder exstrophy in a female
and the incision template used for repair.
distally to the verumontanum on both sides of the prostatic
urethra, leaving a wide strip of bladder neck and urethral
plate.
A major objective of primary closure in the male patient is
to place the bladder and prostate deep within the pelvis to
achieve a more normal anatomic position.44,45 In some boys,
this maneuver results in inadequate length of the urethral
plate for subsequent penile reconstruction. Duckett therefore
proposed transection of the urethral plate, distal to the veru-
montanum, with the development of lateral paraexstrophic
skin flaps, which could be then rotated medially to bridge
the gap between the transected edges of the urethra.46 Al-
though this technique allows for better mobilization of the
bladder, Gearhart and colleagues41,47 reported a 40% compli-
cation rate associated with the use of paraexstrophy flaps. Ure-
thral stricture is the most common complication and may be
secondary to local tissue ischemia. The routine use of such
flaps appears to be decreasing, though their application re-
mains a viable option when a short urethral plate prohibits
adequate bladder mobilization.48
If the urethral plate is left intact, it should be mobilized to
the level of the prostate to create as much urethral length as
possible. Following complete mobilization of the bladder,
the corpora cavernosa are dissected off the inferior pubic rami
bilaterally, taking care to preserve the neurovascular bundles
and avoid penile devascularization. This maneuver aids in pe-
nile lengthening, primarily through release of dorsal chor-
dee.23 After placement of a Malecot suprapubic tube and
exteriorization of the ureteral catheters, the bladder is closed
anteriorly in the midline and the urethra tubularized over a
10- to 12-Fr sound. The first-stage repair thus results in an
isolated penopubic epispadias, which is generally incontinent.
Closure of the pelvic ring is required to assist in abdominal
wall closure. Pubic approximation without ancillary osteot-
omy may be possible in the immediate newborn period, when
the bones are still malleable; however, osteotomy is generally
required after 3 days of age. Although multiple techniques are
1520 PART VIII GENITOURINARY DISORDERS

A B C D

E F G H

I J K
FIGURE 120-9 Sequence of repair of bladder exstrophy in a male. A, Completion of the dissection around the periphery of the bladder and the urethral
plate. B, Inversion of the bladder plate and approximation of the corpora as a first stage in epispadias repair. Also note the inferior paraexstrophy incisions.
C, Further closure of the skin over the corpora and their partial freeing from the pubis. D, Placement of a suprapubic drainage tube. E, Further closure of the
skin inferiorly, with approximation to the urethral plate. Creation of the paraexstrophy flaps is now evident. F, The urethral plate is prepared for tubular-
ization over a catheter. G, The urethral plate is now tubularized, and ureteral catheters are placed bilaterally and brought out on each side of the bladder.
The bladder is also in the process of being tubularized. H, Completion of tubularization of the bladder and urethra, and location of the various drainage
tubes. I, After two-layer closure of the bladder and urethral plate, the bladder is reduced into the pelvis and fixed with sutures. J, Sutures are placed to
encourage approximation of the pubic halves. K, Drainage tubes are brought out superiorly, and fascia, subcutaneous tissue, and skin are approximated.
Approximation of the pubis helps protect the bladder closure and the abdominal wall closure.

described, combined bilateral anterior innominate and verti-
cal iliac osteotomy is most frequently used to assist symphy-
seal approximation and medial rotation of the pelvic bones.49
Fixator pins are then placed into the iliac wings and lower
osteotomized segments. Our group generally used bilateral
anterior iliac osteotomies (Fig. 120-10). Closure of the pelvic
ring is performed using a large-sized, monofilament suture,
taking care to place the knot anteriorly to avoid erosion into
the soft tissue below. The newly closed bladder and urethra
can now be covered by reapproximation of the rectus fascia
and skin, with externalization of tubes and drains. External
fixators are applied to the pins to hold the pelvis in the correct
configuration. Lower extremity traction is applied to keep the
legs still and prevent destabilization of the pelvis (Fig. 120-11).
The external fixator remains in place for 4 to 6 weeks after
surgery, allowing for callus formation at the osteotomy sites.
As an alternative to external fixation, immobilization may also
be accomplished through the application of a spica cast, which
envelops the hips and lower extremities. The cast remains in
place for 4 to 6 weeks.
B pelvis, and the pubic halves are
CHAPTER 120 BLADDER AND CLOACAL EXSTROPHY 1521
The technique for initial closure in a female patient is sim-
ilar to that described previously. The traction suture is initially
placed anterior to the vagina, which is fully mobilized, as the
neourethra is tubularized. The vagina is then repositioned to
create a more caudal angle of entry.
Postoperatively, the patient is maintained on antibiotic pro-
phylaxis. Parenteral nutrition may be used initially in order to
avoid abdominal distension. Close attention must be given to
patient positioning and fixator pin sites to minimize the risk of
skin ulceration and nerve injury.
Stage 2: Epispadias Repair
The second stage of repair is generally undertaken between
6 and 12 months of age. It centers on reconstruction of the
phallus, with repair of epispadias and urethroplasty. This
may further optimize bladder capacity, through an increase
in outlet resistance.50 Although many techniques have been
used, the method described by Cantwell and later modified
by Ransley has been shown effective in accomplishing urethral
relocation to the penile ventrum, correction of chordee, and a
FIGURE 120-10 A, If the pubis
cannot be approximated in a new-
born or if the procedure is under-
taken later, when the pelvis is less
flexible, pelvic osteotomy is re-
quired. The location of an anterior il-
iac osteotomy is shown. B, External
fixator pins are used to hold the
brought together in the midline.
1522 PART VIII GENITOURINARY DISORDERS
FIGURE 120-11 An infant in Bryant’s traction. The patient must remain
immobilized for 4 to 6 weeks postoperatively.
low fistula rate (Fig. 120-12).51,52 This technique involves full
mobilization of the corpora and dorsal urethral plate, which is
dissected free and tubularized. Correction of dorsal chordee is
achieved by incising each corpora transversely, creating
diamond-shaped defects and anastomosing their dorsal me-
dial aspects over the tubularized urethra. A ventral meatotomy
is then performed at the tip of the glans to produce a more
anatomically normal meatal position.53 Other described
modifications of the Cantwell-Ransley technique include full
detachment of the urethral plate from the corporal bodies,
leaving only the distal-most 1 cm of urethra attached to the
glans tip. The urethra is then tubularized, the corpora are
A B
FIGURE 120-12 Steps in the Cantwell-Ransley epispadias repair. A, The urethral plate is dissected from the corpora and is tubularized, taking care to
preserve the lateral blood supply of the urethra and the neurovascular bundles. B, Corporotomies are created at the midphallus, and the urethra is trans-
posed to the ventral surface. C, The corpora cavernosa are rotated medially and reapproximated at the corporotomy sites, pulling the corporal bodies
inward and providing coverage of the neourethra. This procedure permits further urethral lengthening, approximation of the corpora with preservation
of the blood supply, and full coverage of the urethra.
incised transversely at the point of maximum dorsal curvature,
and the corporocavernostomy defects sutured together, thus
covering neourethra. This maneuver places the urethra ven-
trally between the corporal bodies, causes downward deflec-
tion of the penis, and also provides extra length. If chordee is
correctable by simple corporal rotation, bilateral corporo-
tomies are not required. The glansplasty is completed in
two layers, resulting in a ventrally placed neourethra.54 Fur-
ther modifications to the technique of epispadias repair in-
clude the use of full penile disassembly described by
Mitchell and Bagli (Fig. 120-13).55
Stage 3: Bladder Neck Reconstruction
and Ureteroneocystostomy
The final stage of exstrophy repair involves the construction of
a urinary continence mechanism and is generally undertaken
around 4 years of age.56 During this interval, the patient is
monitored periodically with renal US to evaluate the adequacy
of upper tract drainage. Bladder capacity is also assessed be-
fore bladder neck surgery. Jeffs and colleagues57 reported that
a bladder capacity of greater than 60 mL typically allows for
adequate functional storage capacity without the need for
concomitant bladder augmentation, although others have
reported median capacities of greater than 85 mL to be more
predictive of achieving continence.58 Regardless, it is generally
agreed that continence is highly dependent on the size of the
original bladder plate, successful initial bladder closure, and
an adequate preoperative bladder capacity.
The Young-Dees-Leadbetter technique of bladder neck re-
pair remains the most common approach to bladder neck
reconstruction (Fig. 120-14). After opening the bladder, the
ureters are first mobilized and reimplanted into a more ceph-
alad position by either cross-trigonal or cephalotrigonal ure-
teroneocystotomy.22,59 This procedure not only corrects
vesicoureteral reflux, which occurs in virtually all cases of
closed bladder exstrophy, but also allows for creation of pos-
terior bladder plate flaps for the bladder neck reconstruction.
A recent report by Braga and colleagues also describes
C

B bic symphysis is approximated using PDS sutures, and the
A are rotated medially and reapproximated (Fig. 120-16).
C D
FIGURE 120-13 Complete penile disassembly technique. A, The corpo-
ral bodies and the hemiglans are separated. B, The urethra is tubularized
and moved ventrally. C, The corpora are reapproximated dorsally. D, Glans
closure is performed distally to complete the repair. (Modified from
Mitchell ME, Bagli DK: Complete penile disassembly for epispadias repair:
The Mitchell technique. J Urol 1996;155:300.)
successful bilateral ureteral reimplantation at the time of pri-
mary bladder closure.33
A strip of posterior bladder plate 2 cm  4 cm is marked
off, and the triangles of bladder laterally are demucosalized.
The strip of bladder is tubularized, and the triangles of de-
nuded muscle are mobilized laterally to provide coverage of
the neourethra. If prior urodynamic evaluation has demon-
strated inadequate bladder capacity, augmentation cystoplasty
with a bowel segment may be performed at this setting.
SINGLE-STAGE RECONSTRUCTION:
COMPLETE PRIMARY REPAIR OF EXSTROPHY
Recently, Mitchell and Grady minimized the number of re-
quired operations by combining bladder closure with epispa-
dias repair at birth in a technique known as complete primary
repair.60 Major potential benefits of this approach include the
earlier creation of bladder outlet resistance, theoretically lead-
ing to normal cycling and improved bladder capacity and
functionality as the patient grows. Major principles of CPRE
include total penile disassembly and division of the intersym-
physeal band, which enables posterior positioning of the
CHAPTER 120 BLADDER AND CLOACAL EXSTROPHY 1523
bladder, bladder neck, and urethra. As described by Grady
and Mitchell,60 CPRE begins with intubation of each ureteral
orifice with ureteral catheters. Traction sutures are placed into
each hemiglans, the bladder plate is circumscribed, and dis-
section is continued inferiorly along the ventral aspect of
the penis. The urethral plate is mobilized off the penis, which
is fully disassembled into separate right and left corporal bod-
ies and the spongiosum-containing urethra. The intersymphy-
seal band is incised (Fig 120-15), which allows the bladder
unit to be positioned deep within the pelvis. A suprapubic
tube is left in place, the ureteral catheters are externalized,
and the bladder is closed. Similar to the staged approach, pel-
vic osteotomy may be required for abdominal closure. The pu-
abdominal wall is closed. The urethral plate is tubularized
and transposed to the penile ventrum, and the corporal bodies
Because of the new posterior positioning of the bladder unit,
urethral length is often inadequate to reach the glans, and
a hypospadiac meatus is left for future reconstruction
(Fig. 120-17). Penile shaft coverage is achieved through the
use of ventral rotational penile skin flaps. Postoperatively,
the patient remains immobilized as previously described in
the technique of staged closure. Additional procedures to cor-
rect for residual hypospadias, vesicoureteral reflux, and
incontinence may be required as the child grows.
URINARY DIVERSION
Urinary diversion, in the form of a bowel conduit or reservoir,
may ultimately be required for patients with insufficient blad-
der plate or after reconstructive efforts have been unsuccessful
and is discussed in Chapter 118.
OUTCOMES AND COMPLICATIONS
The most devastating complication of bladder closure is dehis-
cence. Major contributing factors include wound infection,
abdominal distension, bladder prolapse, and loss of ureteral
and/or suprapubic catheters within 6 days of closure.61 Uri-
nary diversion, reclosure of the bladder as a urethral tube
for later augmentation, or delayed repair of the bladder may
be performed. If not performed in the initial setting, pelvic
osteotomy is frequently necessary for successful reclosure.
Urinary incontinence remains a significant problem for up
to 30% of bladder exstrophy patients. In the case of bladder
neck incompetence, injectable bulking agents, bladder neck
sling or artificial urinary sphincter have all been applied. Blad-
der neck reconstruction or formal closure of the bladder neck,
with the creation of a catheterizable channel, can also be per-
formed. In cases where incontinence is secondary to insuffi-
cient bladder capacity, augmentation cystoplasty remains the
most viable treatment option.
Following epispadias repair, the most common complica-
tion is urethrocutaneous fistula, which ranges from 2% to
26% in modern series.51,54,62
The incidence of adenocarcinoma of the bladder in adults
with bladder exstrophy has been estimated to be 250 times that
of the normal population and is likely due to chronic inflam-
mation, infection, and metaplasia of an exposed bladder
plate.63 A series by Woodhouse and colleagues, however, re-
cently documented an 800-fold risk in the incidence of bladder
1524 PART VIII GENITOURINARY DISORDERS

Bladder

Suspensory sutures
Urethra

Lateral view

FIGURE 120-14 Young-Dees-

Leadbetter procedure for bladder
exstrophy repair. A, The ureters are
appropriately reimplanted to avoid
reflux. Triangular areas at the blad-
der base are then denuded, and
the remaining muscle is tubularized
over a catheter. This serves to
lengthen the urethra and provides
sufficient pressure to encourage
the development of improved blad-
der capacity without causing ure-
thral obstruction. B, The bladder is
reclosed, and the bladder neck is
further supported by sutures, which B
secure it to the pubis anteriorly.

malignancy among those with a history of bladder exstrophy by
age 40.64 The development of adenocarcinoma and transitional
cell carcinoma of the bladder is also a potential risk in those
patients who have undergone augmentation cystoplasty.65,66
Fertility in patients with bladder exstrophy and epispadias
was studied by Shapiro and colleagues,14 who surveyed 2500
patients. Among these, 38 men had successfully fathered chil-
dren and 131 women had given birth. Diminished fertility
Bladder augmentation is preferred in patients without ade-
quate bladder capacity, and bladder neck closure with creation
of a continent catheterizable stoma may be performed when
other continence procedures have failed.
Cloacal Exstrophy
------------------------------------------------------------------------------------------------------------------------------------------------

rates among males may be secondary to retrograde ejaculation,
though libido and erectile function appear to be normal
according to a report by Woodhouse and colleagues.67 Female
patients face a significant risk of uterine prolapse.
CONCLUSION
Contemporary reconstructive techniques for the repair of
bladder exstrophy have resulted in acceptable function and
cosmesis for the majority of patients with classic bladder
exstrophy. Overall continence rates range from 70% to 80%.
Cloacal exstrophy is a rare condition occurring in 1 of 200,000
to 400,000 live births68 and comprises the most severe defor-
mation along the EEC spectrum, which includes both epispa-
dias and classic bladder exstrophy. Cloacal exstrophy is also
referred to as the OEIS complex (omphalocele, exstrophy, im-
perforate anus, and spinal defect) when other malformations
of the urogenital, gastrointestinal, skeletal, and neurospinal
axis are present.
Although first described by Littre in 1709, historic survival
rates were dismal secondary to sepsis or fluid, electrolyte, and
nutritional derangements from short gut syndrome or
 CHAPTER 120 BLADDER AND CLOACAL EXSTROPHY 1525

Cross section
Correction of Exstrophy
Plane of Cross section

Incision through
perineal membrane

Bladder
Symphysis
Prostate
Urethra
Corpus spongiosum
Perineal
membrane
Sup. Transverse
perineal muscle
Anus

FIGURE 120-15 Pelvic view of male exstrophy repair as described by Grady and Mitchell. Aggressive dissection along each side of the urethra and division
of the intersymphyseal band allow posterior positioning of the bladder in the pelvis. (Modified from Grady RW, Mitchell ME: Complete primary repair of
exstrophy. J Urol 1999;162:1416.)

intestinal obstruction.69 The first successful repair was
reported in 1960 by Rickham, who recommended staged sur-
geries for reconstruction.70 Advances in neonatal care and sur-
gical technique have resulted in present-day survival rates that
exceed 90%, and principle goals of treatment are now directed
toward improving quality of life in these patients.68,71–75
EMBRYOLOGY AND GENETICS
The underlying defect in cloacal exstrophy is thought to be re-
lated to abnormal development and premature rupture of the
cloacal membrane, as described earlier in the bladder exstro-
phy section. In the setting of cloacal exstrophy, it has been pos-
tulated that membrane rupture occurs within the first 8 weeks
of gestation. Confirmation of this theory is difficult, however,
given no embryologic stage similar to cloacal exstrophy exists
in normal development.76 Disruption of the cloacal membrane,
as the principle underlying abnormality, has been supported by
surgically induced exstrophy in animal models.19,77 The pre-
vailing developmental theories are further clouded by several
recent reports documenting rupture as late as 26 weeks.78,79
Rupture at 5 weeks gestation, as traditionally postulated, would
cause anterior herniation of the bladder and small bowel, which
would prevent normal midline fusion of the hindgut, bladder
plate, genital tubercles, and müllerian ducts, thus resulting in
the typical anatomic presentation of two open bladder halves
separated by a strip of exstrophied cecum, hemiphallic halves
with a widely separated pubic diastasis, an underdeveloped
and blind-ending distal hindgut with imperforate anus, and
an omphalocele of varying size (Fig. 120-18).

A B
FIGURE 120-16 A, Closure of the urethral plate and bladder as a continuous unit. B, Placement of the urethra ventral to the corporal bodies by posi-
tioning the bladder, bladder neck, and urethra posteriorly in the pelvis. (Modified from Grady RW, Mitchell ME: Complete primary repair of exstrophy. J Urol
1999;162:1417.)
1526 PART VIII GENITOURINARY DISORDERS
FIGURE 120-17 At completion of complete primary closure, placement
of the bladder and urethral units deeper within the pelvis along with ven-
tral transposition of the urethra may result in a hypospadiac meatus, which
can later be reconstructed with formal urethroplasty.
No single environmental exposure or consistent genetic de-
fect in the etiology of cloacal exstrophy has yet been identified.
Thauvin-Robinet and colleagues recently identified an unbal-
anced translocation between chromosomes 9q and Yq, and other
studies have implicated mutations in homeobox genes such
as HLXB9 and HOX, which are involved in the development
of embryonic mesoderm.80–82 Although there have been multi-
ple reports of cloacal exstrophy among members of the same
family, these are generally anecdotal and have involved multigen-
erational relatives or nontwin siblings.83–85 Multiple instances of
affected monozygotic twins have been reported, however,
which lends support to an underlying genetic cause.76,86–88
FIGURE 120-18 Typical presentation of cloacal exstrophy in a male in-
fant. O, omphalocele; hB, hemibladder; Ce, exstrophied cecum; hP, hemi-
phallus; hS, hemiscrotum; I, prolapsed ileal segment.
ASSOCIATED ANOMALIES
Unlike classic bladder exstrophy, cloacal exstrophy is typically
associated with a variety of other anatomic defects (Table 120-1).
GASTROINTESTINAL
Ileocecal exstrophy with an associated omphalocele, hindgut
remnant, and imperforate anus is the most common clinical
presentation.54 Omphaloceles are present in 88% to 100%
of infants and generally contain portions of small bowel
and/or liver.54,89 Other findings include intestinal duplication
anomalies, gastroschisis, ectopic anus, colonic exstrophy, and
malrotation.90,91 Short gut syndrome may be a significant
source of morbidity among patients with cloacal exstrophy
and is observed in 25% of cases.71 The risk of short gut
syndrome is markedly increased in patients subjected to
ileostomy placement as the initial intestinal diversion proce-
dure.92,93 Furthermore, the phenomenon may occur even
in the presence of normal bowel length, implicating an inher-
ent absorptive abnormality of the intestine.71,72,90
GENITOURINARY
Abnormalities of the upper urinary tract have been reported in
41% to 66% of cases.72,89 Unilateral renal agenesis, pelvic kid-
ney, and/or hydronephrosis were observed in up to 33% in
Diamond’s 1990 series.89 Less commonly reported findings
include horseshoe kidney, fusion anomalies, and ureteral ab-
normalities.72,89,90 Complete separation or even absence of
the phallic/clitoral halves may be observed, and the scro-
tum/labia are widely separated. Male infants frequently have
undescended testes with associated bilateral inguinal hernias,
whereas failure of müllerian duct fusion in females results in
varying degrees of uterine and vaginal duplication anomalies
in the majority of patients.72,89
TABLE 120-1
Anomalies Associated with Cloacal Exstrophy
Gastrointestinal
Omphalocele
Imperforate anus, anal atresia/stenosis
Short gut syndrome
Intestinal malrotation
Intestinal duplication
Genitourinary
Unilateral renal agenesis
Pelvic kidney
Ureteral duplication
Hydronephrosis
Bilateral cryptorchidism, inguinal hernias
Uterine duplication
Vaginal duplication
Central Nervous System
Spinal dysraphism
Skeletal
Vertebral (absent, extra, hemi)
Club foot
Other lower limb (absence, shortening)
Hip subluxation

CENTRAL NERVOUS SYSTEM
Some form of spinal dysraphism including tethered cord,
myelomeningocele, or lipomyelomeningocele is present in
nearly all patients, with recent reports ranging from 64% to
100%.74,84,91,94 Neurologic impairment is variable and may
affect bladder function, urinary continence, lower extremity
movement, and erectile function. Detailed postmortem micro-
dissection studies have demonstrated both aberrant origin and
vascular supply of the pelvic autonomic nerves,95 and these
nerves are at additional risk of iatrogenic injury during oper-
ative repair. Other reported abnormalities include periventri-
cular leukomalacia, hydrocephalus, hypoplastic cerebellum,
and Chiari malformation.84
SKELETAL
Abnormalities of the spine, pelvis, and limbs have all been ob-
served in the setting of cloacal exstrophy. Spinal anomalies, ex-
cluding myelodysplasia, have been reported in 22% to 60%
and consist mainly of absent or extra vertebrae, scoliosis,
and kyphosis.90,96,97 The pelvic deformity is similar to that
of classic bladder exstrophy but typically more severe with sig-
nificant widening of the pubic diastasis, external angling of the
posterior and anterior segments, and external rotation and ab-
duction of the iliac wings.70 A review by Jain and Weaver
found a 17% to 26% incidence of associated lower limb abnor-
malities.98 Certain limb malformations like club foot and equi-
novarus deformities can be seen in association with
myelomeningocele, which often accompanies cloacal exstro-
phy; however, a variety of true limb malformations including
hypoplasia, absence, split foot, and ectopic or additional digits
have also been observed.98
PRENATAL DIAGNOSIS
Early prenatal diagnosis allows time for thorough parental
counseling and allows for consideration of pregnancy termi-
nation. Prenatal diagnosis was first reported by Meizner and
Bar-Ziv in 1985,99 and since then, several authors have pro-
posed criteria for the prenatal diagnosis of cloacal exstrophy.
Principle findings include failure to visualize the urinary
bladder along with a large midline anterior abdominal wall
defect and/or lumbosacral myelomeningocele.87,100–102 The
prolapsed ileal segment, which may appear as an “elephant
trunk–like” mass on US, has also been reported as a path-
ognomic finding.103 From a review of 22 cases, Austin and
colleagues102 developed a list of major and minor criteria
for prenatal US diagnosis on the basis of the frequency with
which abnormalities were observed. Major criteria were those
seen in greater than 50% of cases and included nonvisualiza-
tion of the bladder (91%); a large, midline, infraumbilical
anterior abdominal wall defect or cystic anterior abdominal
wall structure (82%); omphalocele (77%); and myelomenin-
gocele (68%). Minor criteria were observed in less than 50%
and consisted of lower extremity defects (23%), renal anom-
alies (23%), ascites (41%), widened pubic arches (18%), nar-
row thorax (9%), hydrocephalus (9%), and single umbilical
artery (9%).102
CHAPTER 120 BLADDER AND CLOACAL EXSTROPHY 1527
SURGICAL REPAIR
Immediate Postnatal Management
After delivery and stabilization of the newborn, exposed or-
gans and mucosal surfaces including the omphalocele, blad-
der, intestine, and myelomeningocele should be protected
by enclosing the infant’s lower torso in a bowel bag or by first
moistening surfaces with saline and covering with sterile plas-
tic wrapping.90 Urologic examination should attempt to note
genetic sex and size of hemibladder plates. Baseline renal
function, electrolyte, and hematologic status should be deter-
mined. Karyotyping can be performed if gender has not been
previously determined or is not obvious on examination. Ini-
tial imaging should include plain films of the chest and spine
along with head, abdominal, renal, and spinal US. In the ab-
sence of obvious spinal dysraphism, magnetic resonance im-
aging (MRI) may be advisable for detection of occult lesions.
Consultation should also be made to general surgery, neuro-
surgery, and orthopedics for operative planning. Once the ini-
tial evaluation has been completed, discussion may be had
with the parents regarding gender assignment, surgical recon-
struction, potential functional deficits, and overall expected
quality of life.
Principles of Repair
The surgical management of cloacal exstrophy is typically un-
dertaken in the newborn period (48 to 72 hours) as a com-
bined effort between pediatric surgery and urology. In the
setting of associated spinal dysraphism, neurosurgical consul-
tation and closure should be undertaken as soon as the infant
is medically stable. Early operation minimizes bacterial colo-
nization of exposed viscera and may decrease the need for pel-
vic osteotomy.68,104 The traditional approach of staged repair
has been thoroughly described by Gearhart and Jeffs.40,94
Complete primary repair has also been reported by Howell
and colleagues,68 Zderic and colleagues,105 Hendren,106
and most recently by Mitchell and Plaire.75 It is generally
agreed that an individualized approach toward reconstruc-
tion, whether in a single-staged or multistaged procedure, re-
sults in the best long-term outcomes.107 The main goals of
reconstruction include secure abdominal wall and bladder
closure, preservation of renal function, prevention of short
gut syndrome, creation of functionally and cosmetically ac-
ceptable genitalia, and attainment of urinary and fecal
continence.73,94
Although various operative algorithms have been published,
all approaches begin with initial separation of the intervening
cecal plate from the two bladder halves, closure of the ompha-
locele, and hindgut preservation (Fig. 120-19).68,71,72,75,107
In the past, the bowel was initially diverted through the creation
of loop or end-ileostomies, and the hindgut segment was uni-
formly discarded. This practice has since fallen out of favor in
order to maximize the absorptive capabilities of the intestinal
tract.72,74 Currently, after tubularization of the exstrophied ce-
cum, it is recommended that the hindgut segment be brought
out as an end-colostomy.92 In the rare instance when the hind-
gut remnant is not used, it may be left as a mucous fistula for
use in future urologic or vaginal reconstruction.90
The omphalocele is reduced to assist abdominal wall clo-
sure; however, in cases of large omphaloceles, complete initial
reduction may not be possible. In this setting, a silo device
1528 PART VIII GENITOURINARY DISORDERS
A B
FIGURE 120-19 Repair of cloacal exstrophy. A and B, The bowel and bladder mucosa are separated, and the ileocecal junction is tubularized and brought
out as an end-colostomy. The bladder halves are turned in, as in a complete exstrophy repair, with approximation of the pubic rami. C, Alternatively, the
bladder halves are approximated in the midline and left open for staged repair if tubularization is not possible. The omphalocele is also closed.
may be used or the omphalocele may be allowed to re-
epithelialize, converting it to a ventral hernia, which may be
repaired at a later time (Fig. 120-20).94
The hemibladders are dissected and then reapproximated
in the midline. In infants with few other associated malforma-
tions and who are medically stable, complete closure of the ab-
dominal wall, bladder, and phallic halves may be undertaken at
this point in a single-stage procedure with or without pelvic
osteotomy. If this is not possible at the initial setting, the two
bladder halves can first be joined in the midline, recapitulating
the appearance of classic bladder exstrophy, which can then be
repaired in a staged fashion as described in the previous section.
Genital reconstruction consists of bringing the phallic
halves together to create an appearance congruent with the
assigned gender. In the male infant with cloacal exstrophy,
the phallic halves are characteristically diminutive, widely
separated, and asymmetric. Historically, genetically male in-
fants were routinely assigned to female gender at the time of
initial closure, undergoing orchiectomy and feminizing geni-
toplasty.108 Recent data regarding gender identity outcomes in
gender-reassigned cloacal exstrophy patients has suggested an
FIGURE 120-20 If initial closure of the omphalocele is not possible, it
may be allowed to epithelialize, converting the omphalocele to a ventral
hernia, which can be repaired at a later time.
Ileum
Colon
C
inherent preference toward male behaviors and sexual identi-
ties in these patients.75,105,109 It remains a topic of continued
study and debate. Gender reassignment has since been largely
abandoned in the current management of cloacal exstrophy,
though functional and aesthetic phallic reconstruction re-
mains challenging. Vaginal reconstruction is necessary in fe-
males and in gender-reassigned males and is accomplished
through the use of bowel or skin grafts.
Gastrointestinal reconstruction, in the form of a pull-
through procedure, may be performed in select patients, some
time after initial diversion and abdominal closure. The deci-
sion is based on the potential for fecal continence and may
be influenced by colonic length, ability to form solid stool,
and the presence of anal stenosis versus imperforate anus.
Like those with classic bladder exstrophy, these patients
will also require the creation of antireflux and urinary conti-
nence mechanisms. The presence of myelodysplasia in these
patients usually necessitates augmentation cystoplasty with
a bowel segment and intermittent catheterization in order to
achieve continence. Continence procedures include creation
of a neourethra, construction of a catheterizable abdominal
stoma with concomitant bladder augmentation, and/or blad-
der neck closure, the selection of which is influenced by the
presence of short gut syndrome, manual dexterity, degree of
mobility, and patient motivation.94
POSTOPERATIVE CONSIDERATIONS
Given the high incidence of short gut syndrome, fluid and nu-
tritional status must be carefully monitored and the initial use
of total parenteral nutrition (TPN) is advocated.110 The keys to
postoperative success are similar to those for repair of classic
bladder exstrophy. Patients are immobilized in some type of
traction device. In the setting of pelvic osteotomy, an external
fixator is left for 4 to 6 weeks postoperatively. Broad-spectrum
antibiotics are administered to minimize risk of wound in-
fection and urosepsis. In contrast to patients with classic
bladder exstrophy, the presence of associated myelodysplasia
in cloacal exstrophy generally precludes use of an epidural
catheter. Pain control in cloacal exstrophy patients can be
challenging, and the involvement of the pediatric pain service
is recommended. Finally, the importance of limiting abdomi-
nal distension to ensure successful abdominal closure and

adequate drainage of both ureteral and bladder catheters
cannot be understated.61,111,112 Following repair, close mon-
itoring of the upper tracts by US is mandatory to observe for
adequate renal growth and to detect evidence of obstruction
or VUR, which has been reported in 50% to 60% of cloacal
exstrophy patients after staged or complete primary
repair.60,107
CONCLUSION
For the past 20 years, survival among patients with cloacal
exstrophy has exceeded 90%.68,72,74,105 Death is typically re-
lated to complications related to extreme prematurity, renal
agenesis, or other complex malformations that are incompat-
ible with life. It is interesting to note that cardiovascular anom-
alies are rarely observed in the setting of cloacal exstrophy. The
various complications related to the management of patients
with cloacal exstrophy are similar to those of patients with
classic bladder exstrophy, as described in the previous section.
Compared with those with classic exstrophy, however, cloacal
exstrophy patients face additional challenges of achieving
bowel and bladder continence secondary to the need for anal
reconstruction and the associated defect of spinal dysraphism.
It must be stressed that multiple operations are the rule, and
these patients will likely face significant medical, psychologic,
and social challenges throughout their lives. Advancements in
CHAPTER 120 BLADDER AND CLOACAL EXSTROPHY 1529
medical and surgical management continue to improve func-
tional and quality of life outcomes in these patients, but it is
important that these individuals remain under the care of a
multidisciplinary team of providers who can offer medical
care, psychologic support, and lifelong follow-up.
The complete reference list is available online at www.
expertconsult.com.
SUGGESTED READINGS
Ebert AK, Reutter H, Ludwig M, Rösch WH. The exstrophy-epispadias com-
plex. Orphanet J Rare Dis 2009;4:23.
Gearhart JP, Mathews R. The Exstrophy-Epispadias Complex: Research Con-
cepts and Clinical Applications. New York: Kluwer Academic Publishers;
1999.
Hernandez DJ, Purves T, Gearhart JP. Complications of surgical reconstruction
of the exstrophy-epispadias complex. J Pediatr Urol 2008;4:460–466.
Husmann DA. Surgery insight: Advantages and pitfalls of surgical techniques
for the correction of bladder exstrophy. Nat Clin Pract Urol 2006;3:
95–100.
Ludwig M, Ching B, Reutter H, Boyadjiev SA. Bladder exstrophy-epispadias
complex. Birth Defects Res A Clin Mol Teratol 2009;85:509–522.
Woodhouse CR, North AC, Gearhart JP. Standing the test of time: Long-term
outcome of reconstruction of the exstrophy bladder. World J Urol 2006;24:
244–249.
Woo LL, Thomas JC, Brock JW. Cloacal exstrophy: A comprehensive review of
an uncommon problem. J Pediatr Urol 2009 Oct; (Epub ahead of print).

CHAPTER 121
Hypospadias
Laurence S. Baskin
Hypospadias is one of the most common congenital anoma-
lies, occurring in approximately 1 in 250 newborns, or
roughly 1 in 125 live male births.1,2 Hypospadias can be
defined as an arrest in normal development of the urethra,
foreskin, and ventral aspect of the penis.3 This results in a
wide range of abnormalities; the urethral opening can be any-
where along the ventral shaft of the penis, within the scrotum,
or even in the perineum (Fig. 121-1). Hypospadias is also
associated with a ventral curvature of the penis, or chordee.
Left uncorrected, patients with severe hypospadias may need
to sit down to void and tend to shun intimate relationships
because of fears related to abnormal sexuality. Babies born
with severe hypospadias and penile curvature may have
“ambiguous genitalia” in the newborn period, making an
immediate and accurate sex assignment difficult.
Hypospadias is classified by the location of the urethral me-
atus (Fig. 121-2). Mild (distal) hypospadias may be glanular
(meatus on the ventral surface of the glans penis), coronal
(meatus in the balanopenile furrow), or distal (in the distal
third of the penile shaft). Moderate hypospadias is along the
middle third of the penile shaft. Severe (posterior) hypospa-
dias extends through the proximal third of the penile shaft
to the perineum and may be described as posterior penile
(at the base of the shaft), penoscrotal (at the base of the shaft
in front of the scrotum), scrotal (on the scrotum or between
the genital swellings), or perineal (behind the scrotum or
behind the genital swellings). Classifying hypospadias is not
necessarily useful in determining surgical approach, and these
classifications do not take into account the associated penile
curvature (Fig. 121-3). A patient with severe curvature and
an anterior urethral meatus may require more extensive
surgery to correct both anomalies.
Historical Notes
------------------------------------------------------------------------------------------------------------------------------------------------
Throughout Greek culture, there was high appreciation for the
goddess Hermaphrodite, who was half man and half woman.
Many statues reflect hypospadiac genitalia, perhaps indicative
of admiration for this condition. It is, therefore, understand-
able why it was not until the first and second centuries AD that
the Alexandrian surgeons Heliodorus and Antyllus attempted
to correct this anomaly by amputation of the distal curved
portion.4 Sexually, the dystopia of the meatus may cause
impotentia generandi. Henry II of France was known to have
hypospadias, as recorded by his physician Fernal. Henry’s
marriage with Catherine de Medici was infertile until Fernal
“advised his patient that in such cases coitus more ferarum
permitted him to overcome the difficulty.”5 Henry II went on
to sire three kings of France, along with seven other children.
Embryology
------------------------------------------------------------------------------------------------------------------------------------------------
Formation of the external male genitalia is a complex deve-
lopmental process involving genetic programming, cell differ-
entiation, hormonal signaling, enzyme activity, and tissue
remodeling.6,7 By the end of the first month of gestation,
the hindgut and future urogenital system reach the ventral sur-
face of the embryo at the cloacal membrane.8 The urorectal
septum divides the cloacal membrane into a posterior, or anal,
half and an anterior half, the urogenital membrane. Three
protuberances appear around the latter. The most cephalad
is the genital tubercle. The other two, the genital swellings,
flank the urogenital membrane on each side. Up to this point,
the male and female genitalia are essentially indistinguishable.
Under the influence of testosterone and dihydrotestosterone
in response to a surge of luteinizing hormone from the
pituitary, masculinization of the external genitalia takes place
(Fig. 121-4). One of the first signs of masculinization is an
increase in the distance between the anus and the genital
structures, followed by elongation of the phallus, formation
of the penile urethra from the urethral groove, and deve-
lopment of the prepuce.9,10
At 8 weeks’ gestation, the external genitalia remain in the
indifferent stage (see Fig. 121-4). The urethral groove on
the ventral surface of the phallus lies between the paired
urethral folds (Fig. 121-5). The penile urethra forms as a result
of fusion of the medial edges of the endodermal urethral folds.
The ectodermal edges of the urethral groove fuse to form the
median raphe. By 12 weeks, the coronal sulcus separates the
glans from the shaft of the penis. The urethral folds have
completely fused in the midline on the ventrum of the penile
shaft. During the 16th week of gestation, the glanular urethra
appears. Two possible explanations for formation of the
glanular urethra have been proposed (see Fig. 121-5):
(1) endodermal cellular differentiation or (2) primary intru-
sion of ectodermal tissue from the glans (Fig. 121-6).
1531
1532 PART VIII GENITOURINARY DISORDERS

A B C

D E F
FIGURE 121-1 Mild to severe hypospadias. A, Mild hypospadias with the urethral opening on the glans. B, Mild hypospadias with the urethral opening at
the coronal margin. C, Moderate hypospadias with the urethral opening on the distal penile shaft. D, Moderate hypospadias with the urethral opening on
the midpenile shaft. E, Severe hypospadias with the urethral opening at the penoscrotal junction. F, Severe hypospadias with the urethral opening in the
scrotum (the arrows indicate the opening of the hypospadiac urethral meatus). Note that the foreskin is absent on the ventral surface of the penis and
excessive on the dorsal aspect. The more severe forms of hypospadias are associated with penile curvature. (From Baskin LS: Hypospadias and urethral
development. J Urol 2000;163:951-956.)

thereby explaining the cell type of the glans penis.12 There

is no evidence of ectodermal ingrowth, as proposed under
Glanular the second theory.13
Anterior (50%) The future prepuce is forming at the same time as the ure-
Subcoronal thra and is dependent on normal urethral development. At
Distal penile about 8 weeks’ gestation, low preputial folds appear on both
sides of the penile shaft, which join dorsally to form a flat ridge
at the proximal edge of the corona. The ridge does not entirely
Midshaft Middle (30%) encircle the glans because it is blocked on the ventrum by incom-
plete development of the glanular urethra (see Fig. 121-5, A).
Proximal penile Thus the preputial fold is transported distally by active
Penoscrotal growth of the mesenchyma between it and the glanular lamella.
The process continues until the preputial fold (foreskin) covers
all of the glans (see Fig. 121-5, C). The fusion is usually present
at birth, but subsequent desquamation of the epithelial fusion
Scrotal Posterior (20%)
allows the prepuce to retract. If the genital folds fail to fuse,
the preputial tissues do not form ventrally; consequently, in
hypospadias, preputial tissue is absent on the ventrum and is
Perineal excessive dorsally (see Fig. 121-1).
At the molecular level, testosterone must be converted to
FIGURE 121-2 Classification and incidence of hypospadias: distal 5a-dihydrotestosterone (DHT) by the microsomal enzyme
(anterior), moderate (middle), and posterior. type 2 5a-reductase for complete differentiation of the penis
with a male-type urethra and glans.14–16
Anatomic and immunohistochemical studies advocate
the theory of endodermal differentiation on the basis of evi-
dence that the epithelium of the entire urethra is of urogenital Incidence
sinus origin.11 The entire male urethra including the glanular ------------------------------------------------------------------------------------------------------------------------------------------------

urethra is formed by dorsal growth of the urethral plate
into the genital tubercle and ventral growth and fusion of
the urethral folds. Under proper mesenchymal induction,
urothelium has the ability to differentiate into a stratified
squamous phenotype with characteristic keratin staining,
Several European countries including Norway, Sweden,
England, Wales, Hungary, Denmark, Finland, Italy, France,
and the United States published independent reports of
increasing rates of hypospadias during the 1960s, 1970s,
and 1980s.2,17 More recently, data from the International
 CHAPTER 121 HYPOSPADIAS 1533

A B C
FIGURE 121-3 A, Coronal hypospadias with penile curvature. B, The thin ventral skin is not suitable for a parameatal-type procedure. C, After release
of the skin and dartos fascia (straightening the penis), the meatus (arrow) is cut back to remove thin urethral skin in preparation for an onlay island
flap urethroplasty. (From Baskin LS, Duckett JW, Ueoka K, et al: Changing concepts of hypospadias curvature lead to more onlay island flap procedures.
J Urol 1994;151:191.)

Umbilical
cord

Umbilicophallic
groove
Epithelial tag
Genital tubercle
Genital swelling
Urethral Lateral phallic groove
groove
Urethral fold

Perineal body Urogenital membrane

Anal hillock
Anal pit
Gluteal fold
Epithelial tag
Glans

Secondary urethral groove

Urethral plate
A Urethral fold
Ectoderm
Mesoderm
Endoderm
Penile urethra

Perineal raphe

11 weeks

B
FIGURE 121-4 Development of the penis and urethra. A, Indifferent stage at about 8 weeks. Note that the primitive urethral groove forms on the caudal
slope of the genital tubercle. Paired genital (labioscrotal) swellings arise on either side of the urogenital membrane above the anal pit and the perineal
body. B, Enclosure of the urethra at 11 weeks. Beginning near the anus, the adjacent ectodermal urethral folds fuse over the urethral plate to form the penile
urethra, with the distal urethra at the coronal sulcus being the last to close.
Continued
1534 PART VIII GENITOURINARY DISORDERS

Prepuce
Preputial Meatus
fold
Frenulum
Urethral
Groove plate
Urethral
folds

Glanular urethra Prepuce

Lumen
Ectodermal
intrusion
Urethral
plate VS Endoderm
Ectoderm
Urethral
folds

Urethral fold
Urethral Urethral plate Endoderm
folds (fused) Ectodermal
Ectodermal Urethral Ectoderm
C intrusion intrusion
plate
Preputial fold

Glans
Preputial groove

Preputial
fold

Glanular
lamella

FIGURE 121-4—CONT’D C, Formation of the glanular urethra and fossa navicularis occurs late in gestation. A plug of ectoderm from the tip of the glans
invades the mesenchyma as an ectodermal intrusion. The floor of the ectodermal intrusion makes contact with the end of the urethral plate that forms the
roof of the advancing urethra, and the intervening double wall breaks down. D, The prepuce forms by the differentiation of the epithelial cells of the
glanular lamella, which forms a groove between the preputial folds and the glans. (From Hinman F Jr: Surgical Anatomy of the Genitourinary Tract.
Philadelphia, WB Saunders, 1994, pp 418-470.)

Clearinghouse for Birth Defects Monitoring Systems

(ICBDMS) were analyzed to determine whether these in- Associated Anomalies
creases were worldwide and continuing and whether they ------------------------------------------------------------------------------------------------------------------------------------------------

had any geographic pattern.2 The ICBDMS data suggested that
the increased incidence of hypospadias was not a worldwide
trend; it was most notable in the United States, Norway, and
Denmark. Also, it was determined that the incidence was not
increasing in the less affluent and less industrialized nations
(on the basis of gross domestic product) for which data were
available. Increasing trends in England, Canada, and northern
Netherlands appear to have leveled off since 1985.
Data from the Metropolitan Atlanta Congenital Defects
Program showed that the incidence of severe hypospadias
increased between 1968 and 1990.1 The Birth Defects Moni-
toring Program (BDMP), which gathers diagnoses recorded on
newborn discharge summaries from hospitals nationwide,
also reported an increase in hypospadias. However, more
recent studies in the New York area question whether the
incidence is actually changing.18
Undescended testis and inguinal hernia are the most common
anomalies associated with hypospadias. In one series, 9.3%
of hypospadias patients had an undescended testis; the inci-
dence was 32% with posterior hypospadias, 6% for moderate,
and 5% with distal.19 The same investigators found the overall
incidence of inguinal hernia to be 9%, with 17% of those cases
associated with posterior hypospadias. A utriculus masculinus
(utricle) is more often found in cases of severe hypo-
spadias.20,21 Combining two large studies in severe hypospa-
dias, there was an 11% incidence of a utricle. Usually, the only
complications caused by the presence of a utricle are difficulty
passing a catheter and rarely infection.22
It is not surprising that urinary tract anomalies are
infrequent because the external genitalia are formed after
the supravesical portion of the urinary tract. McArdle and
Lebowitz found only 6 genitourinary anomalies among
 CHAPTER 121 HYPOSPADIAS 1535

B C D
FIGURE 121-5 Normal human male genitalia development at 11, 16.5, 20, and 24 weeks’ gestation. A, Note the open urethra and prominent urethral
folds (uf). B, Note the natural phase of penile curvature that occurs during development. C, The foreskin is completely formed, and the curvature has
resolved. D, Continued growth with visualization of the midline skin seam (ms).

Ectodermal ingrowth theory Ectodermal differentiation theory

Ectoderm
Endoderm
Stratified
squamous

FIGURE 121-6 Two theories of urethral development: the older theory of ectodermal intrusion and the newer theory of endodermal
differentiation. (From Kurzrock EL, Baskin LS, Cunha GR: Ontogeny of the male urethra: Theory of endodermal differentiation. Differentiation
1999;64:115-122.)

200 patients with hypospadias (3%).23 Cerasaro and col- Cause

leagues24 found that 1.7% of patients (4 of 233) had signifi- ------------------------------------------------------------------------------------------------------------------------------------------------

cant anomalies. On the basis of a review of 169 patients,
Shelton and Noe25 did not recommend routine urinary tract
evaluation. Khuri and colleagues,19 in a review of 1076 pa-
tients, did not find any significant associated urinary tract
anomalies. They concluded that patients with hypospadias
and an associated inguinal hernia or undescended testis did
not require further urinary tract evaluation; however, patients
with hypospadias and other organ system anomalies found on
physical examination should undergo upper urinary tract
screening with abdominal ultrasonography.
In the majority of patients the etiology of hypospadias remains
unknown.3 For example, 33 patients with severe hypospadias
were evaluated with a range of diagnostic techniques
including clinical assessment, ultrasonography, karyotyping,
endocrine evaluation, and molecular genetic analysis of
the androgen receptor (AR) and 5a-reductase genes to classify
and determine the cause of hypospadias. Diagnoses were
determined in only 12 patients (36%); the remaining 64%
of patients were classified as having hypospadias of unknown
etiology.26
1536 PART VIII GENITOURINARY DISORDERS
HORMONE RECEPTOR IMPAIRMENT
Many investigators have also attempted to link abnormalities in
androgen metabolism or the AR to hypospadias. For example,
Gearhart and colleagues27 found no deficiencies in either AR
levels or 5a-reductase in their study of preputial skin from boys
with hypospadias. Allera and colleagues28 analyzed nine pa-
tients with severe hypospadias and found a defect in the
open-reading frame of the AR in only one patient. Sutherland
and colleagues29 also concluded that mutations in the AR gene
are rarely associated with hypospadias. Molecular biology
techniques have demonstrated that defects in the AR gene are
definitely associated with isolated hypospadias. However, these
genetic defects account for an extremely small subset of cases,
implying that other factors are responsible for hypospadias.
GENETIC IMPAIRMENT
Increasingly, researchers are examining the role of cellular signals
other than testosterone and DHT in the morphogenesis of the
phallus and the cause of hypospadias. Normal embryogenesis
of the urogenital system depends on epithelial-mesenchymal
interactions, and it has been hypothesized that aberrant signal-
ing between the epithelium and mesenchyma could lead to hy-
pospadias.12 For example, prostate development requires
testosterone-dependent Sonic hedgehog (Shh) expression in
the epithelium of the urogenital sinus.30 In mice null for Shh,
penile development is inhibited and cloacal defects exist.31,32
Mice null for fibroblast growth factor 10 have a proximal urethral
opening and a flattened glans, consistent with the hypospadias
defect.32 It is likely that researchers will find similar genetic
signaling molecules involved in epithelial-mesenchymal
interactions in the phallus that play a role in its development.
Another area of investigation with respect to the cause of
hypospadias is the expression and regulation of homeobox
(Hox) genes. These genes are transcriptional regulators that
play an essential role in directing embryonic development.
Genes of the Hox A and Hox D clusters are expressed in
regionalized domains along the axis of the urogenital tract.
Transgenic mice with loss of function of single Hox A or
Hox D genes exhibit homeotic transformations and impaired
morphogenesis of the urogenital tract.30,33–35 Human males
with hand-foot-genital syndrome, an autosomal dominant
disorder characterized by a mutation in Hox A13, exhibit hy-
pospadias of variable severity, suggesting that Hox A13 may be
important in the normal patterning of the penis.36–38 Further-
more, research has shown that the embryonic expression of
certain Hox genes is regulated by hormonal factors.39 Estrogen
and the synthetic estrogen DES, for example, inhibit Hox A9,
Hox A10, Hox A11, and Hox A13 genes in mice. Thus in
addition to defects in Hox genes, it is possible that improper
regulation or expression of hormonal factors during embryo-
genesis could disrupt normal expression of Hox genes and
lead to reproductive tract anomalies. More recently, ATF3
and MAMLD1 (Cxorf6) are new candidate genes that are asso-
ciated with hypospadias.40 ATF3 is an estrogen-responsive
gene whose expression is increased in hypospadiac boys.41,42
MAMLD1, which is expressed in the gonad during sex
differentiation and interacts with steroidogenic factor 1
(SF1), a transcription factor expressed in the mammalian
sex-determining pathway, has also been shown to be mutated
in patients with isolated hypospadias.43
ENZYME IMPAIRMENT
Despite the central role that testosterone plays, attempts to
ascribe all hypospadias to an underlying genetic defect in
this pathway have only rarely been successful. Holmes and
colleagues44 determined the incidence of defects in three
major enzymes in the biosynthetic pathway of testosterone:
3a-hydroxysteroid dehydrogenase, 17a-hydroxylase, and
17,20-lyase. Forty-eight boys with fully descended testes
and various degrees of hypospadias were compared with
age-matched controls who were undergoing circumcision.
Morning fasting serum samples of pregnenolone, proges-
terone, 11-deoxycorticosterone, 17-OH pregnenolone, 17-OH
progesterone, 11-deoxycortisol, cortisol, dehydroepian-
drosterone, androstenedione, androstenediol, testosterone,
and dihydrotestosterone were obtained. To focus on the prox-
imal steps in androgen biosynthesis, 12 individuals with
hypospadias underwent standard adrenocorticotropic hor-
mone (ACTH) stimulation. No significant differences in the
androgen precursors and metabolites were found between
the controls and the individuals with hypospadias. The
response to ACTH was variable, with no significant difference
between patients with different degrees of hypospadias and
controls. These data indicate that enzymatic defects in the
steroidogenic steps from cholesterol to dihydrotestosterone
are not a common cause of hypospadias.
ENVIRONMENTAL FACTORS AND
ENDOCRINE DISRUPTERS
In the past, environmental factors were generally ruled out as
causes of hypospadias.45,46 More recently, multicausality
models have included environmental contaminants to deter-
mine the risk of developing a given phenotype. For example,
familial clustering of hypospadias among first-degree relatives
has been perceived as being under the influence of a strong
genetic and heritable component, but there have been many
exceptions in which genetics was ruled out. It has thus been
suggested that environmental influences should be considered
as well, given that families share similar exposures. In partic-
ular, in cases in which the effects are profound, genetic pre-
disposition exacerbated by environmental exposure should
be considered.47–49
Attempts to determine risk factors for hypospadias have
yielded a number of maternal and paternal risk factors.
Among traditional studies of maternal risk factors for congen-
ital anomalies, maternal age and primiparity are significantly
associated with hypospadias, although some studies have con-
tested the maternal age effect.45 Paternal risk factors associated
with hypospadias include abnormalities of the scrotum or
testes and low spermatozoa motility and abnormal sperm mor-
phology.48,50 It has been suggested that the recent increase in
hypospadias reflects the improvement in fertility treatments,
contributing to more subfertile men fathering children.
As Fritz and Czeizel state, this “relaxed-selection hypothesis,
which states that there is a redistribution in the number of
children born to fertile and infertile (subfertile) couples,
may account for the increasing number of other defects and
cancers of male genitalia observed today and the fall in sperm
counts.”48,51
In addition to parental risk factors, there is strong consen-
sus in the literature that boys with hypospadias have lower
 CHAPTER 121 HYPOSPADIAS 1537

birth weights.52 Fredell and colleagues52 examined hypos- thick, elastic tunica albuginea, with a midline septum
padias in discordant monozygotic twins and found that (Fig. 121-7).9,66–69 The urethral spongiosum lies in a ventral
the twin with hypospadias weighed 78% of the twin without position, intimately engaged between the two corporal bodies.
hypospadias. The birth weight difference was still significant The Buck fascia surrounds the corpora cavernosa and splits to
in healthy monozygotic twins. Another study found that contain the corpus spongiosum in a separate compartment.
boys with hypospadias had a lower placental weight than The neurovascular bundle lies deep to the Buck fascia, and
control boys did.46 where the two crural bodies join to form the corporal bodies,
A 1995 meta-analysis of first-trimester exposure to proges- the neurovascular bundle completely fans out around the
tins and oral contraceptives did not indicate an increased risk corpora cavernosa, all the way to the junction of the corpus
of hypospadias.53 Exposure to DES was excluded in that spongiosum (see Figs. 121-6 and 121-7).68,70 This concept
study. However, a number of other studies did list gestational disagrees with the classic dogma that the neurovascular bun-
exposure to progestins as a causal agent. Another recent study dle lies in the 11 o’clock and 1 o’clock positions. Superficial to
linked a maternal vegetarian diet during pregnancy to an in- the Buck fascia is the dartos fascia, which lies immediately
crease in the incidence of hypospadias.54 This study looked at beneath the skin. This fascia contains the blood supply to
51 boys with hypospadias from a group of 7928 boys born to the prepuce. The prepuce is supplied by two branches of
mothers taking part in the Avon Longitudinal Study of Preg- the inferior external pudendal arteries, the superficial penile
nancy and Childhood. The authors hypothesized that vege- arteries.71 These arteries divide into the anterolateral and
tarians have a greater exposure to phytoestrogens than posterolateral branches. In hypospadias surgery, the island
omnivores do. The phytoestrogens may come in the form of flap is typically based on the anterolateral superficial vessels.
soy, which is high in isoflavones, or may be related to endo- The onlay island flap, tubularized island flap, and de-
crine disrupters in pesticides and fertilizers.55 epithelialized pedicle flap are dependent on careful preser-
Although these risk factors may not be direct causes of hy- vation of these blood vessels. The outer skin survives from
pospadias, they provide additional information that may re- the intrinsic subcutaneous vessels.
veal a common developmental pathway and inform future Compared with the normal penis, the anatomy of the
research. For example, there is growing evidence that andro- hypospadiac penis is no different in terms of neuronal inner-
gens play a central role in the lower birth weight of girls vation, corpora cavernosa and tunica albuginea architecture,
compared with boys.56 Androgens are also crucial to the and blood supply, except at the region of the abnormal ure-
development of the male reproductive tract. Thus exposure thral spongiosum and glans (Figs. 121-8 and 121-9).66,72
to an agent that compromises the weight-gaining advantage The nerves in both the normal and the hypospadiac penis start
of androgen during gestation could play a role in the deve- as two well-defined bundles superior and slightly lateral to the
lopment of hypospadias and low birth weight. urethra. As the two crural bodies converge into the bodies of
Increasing rates of hypospadias have paralleled reports of the corpora cavernosa, the nerves diverge, spreading around
other untoward end points related to male reproductive health the cavernosal bodies up to the junction with the urethral
including increases in testicular cancer, an increased inci- spongiosum, not limiting themselves to the 11 and 1 o’clock
dence of cryptorchidism, and decreased semen and sperm positions (see Figs. 121-7 and 121-8). The 12 o’clock position
quality.57,58 The increasing incidence of such anomalies over in a hypospadiac penis is spared of neuronal structures, just as
the past 50 years concomitant with the increased production in a normal penis. At the hilum of penis where the bodies of
and use of synthetic chemicals has raised concerns that envi- the corpora start to separate, the cavernous nerve sends
ronmental factors may play a role in these problems.17,47 It is nNOS-positive fibers to join the dorsal nerve of the penis,
now well documented from wildlife studies and accompany- thereby changing the functional characteristics of the distal
ing laboratory data that a number of synthetic and natural che- penile dorsal nerve (see Fig. 121-8). Similarly, the nNOS-
micals commonly found in the environment can mimic or negative, ventrally located perineal nerve originating from
antagonize hormones or otherwise interfere with the develop- the pudendal nerve becomes nNOS reactive at the caver-
ment and function of the endocrine and reproductive sys- nosa-spongiosum junction. The redundant wiring in the penis
tems.59,60 The offspring of pregnant mice exposed in utero may be important in the preservation of erectile function.68
to both estrogen and prednisone have been shown to develop The most striking difference between the normal penis
hypospadias.61 Whether endocrine disrupters are having an and the hypospadiac penis is a difference in vascularity (see
impact on human male reproductive health and on hypospa- Fig. 121-9). The hypospadiac penis has large endothelium-
dias in particular is difficult to determine.62 Regardless, public lined vascular channels filled with red blood cells on the
health agencies worldwide are increasingly concerned about abnormal ventral shaft. In contrast, the normal penis has
endocrine disruption and it remains an active area of re- well-defined, small capillaries around the urethra, fanning
search.63–65 into the glans. Anatomic studies of the urethral plate show
no evidence of fibrosis or scarring.73 The urethral plate is well
vascularized, has a rich nerve supply, and has an extensive
Normal and Hypospadiac muscular and connective tissue backing (Fig. 121-10). These
features may explain the success of incorporating the ure-
Penile Anatomy thral plate or abortive spongiosum into hypospadias
------------------------------------------------------------------------------------------------------------------------------------------------
reconstruction.74,75
Surgical repair of hypospadias requires an expert understand- Hypospadias repair requires attention to three major
ing of the normal anatomy of the penis, as well as an under- anatomic defects: (1) abnormal urethral meatus, (2) penile
standing of the anatomy of the hypospadiac penis. The human curvature, and (3) foreskin defect and, in more severe cases,
penis consists of paired corpora cavernosa covered by a scrotal anomalies.
1538 PART VIII GENITOURINARY DISORDERS

A B C D

E F G H
FIGURE 121-7 Normal human fetal penis at 25 weeks’ gestation. A to H, Transverse sections, distal to proximal, immunostained with neuronal marker
S-100 (25). Note localization of the S-100 nerve marker in brown, completely surrounding the cavernous bodies up to the junction with the urethral
spongiosum along the penile shaft, except at the 12 o’clock position (A to D). On the proximal penis at the point where the corporal bodies split into
two (E) and continue in a lateral fashion inferior and adjacent to the pubic rami, the nerves localize to an imaginary triangular area at the 11 and 1 o’clock
positions. At this point (E), the nerves reach their farthest vertical distance from the corporal body (about half the diameter of the corporal body) and
continue (F, G) in a tighter formation at the 11 and 1 o’clock positions, well away from the urethra. (From Baskin LS, Erol A, Li YW, Cunha GR: Anatomical
studies of hypospadias. J Urol 1998;160:1108-1115.)

Left lateral Posterior Right lateral

FIGURE 121-8 Computer-generated three-dimensional reconstruction of the normal human fetal penis at 27 weeks’ gestation. Note that the pathway of
the cavernosal nerves (white arrows) and dorsal nerves (arrowheads) is away from the origin of the crural bodies (white), which is along the ischial tuberosity.
The cavernosal nerves follow a more direct pathway to reach the penile hilum, where they penetrate the corporal bodies under the pubic arch. As the
cavernosal nerves travel close to the penile hilum, they send interconnecting fibers to the dorsal nerves, with subsequent positive immunostaining for
nNOS. nNOS-positive dorsal nerves travel on the corporal bodies (gray) to the glans. As the dorsal nerves of the penis change from nNOS-negative immu-
nostaining character into nNOS positive, the proximal nNOS-negative perineal nerves turn into nNOS-immunoreactive nerves.

it may be so far back in the perineum that it appears as “vaginal

MEATAL ABNORMALITIES
Hypospadias is characterized primarily by location and con-
figuration abnormalities of the urethral meatus. The urethral
meatus may be located only slightly ventrally, just below a
blind dimple at the normal meatal opening on the glans, or
hypospadias.”76 Most patients present with the meatus in one
of the many transitional forms. The meatus is encountered in a
variety of configurations in terms of form, diameter, elasticity,
and rigidity. It can be fissured in both transverse and longitu-
dinal directions, or it can be covered with delicate skin. In the

A B
C D
E F
FIGURE 121-9 Hypospadiac penis at 33 weeks’ gestation. A to F, Trans-
verse sections, distal to proximal, immunostained with neuronal marker
(dark staining) S-100 (20). Note that the anatomy of the hypospadiac pe-
nis is the same as the normal penis, except for the abnormal formation of
the distal urethra and glans (A to C). (From Baskin LS, Erol A, Li YW, Cunha
GR: Anatomical studies of hypospadias. J Urol 1998;160:1108-1115.)
case of the megameatus intact prepuce variant, the distal ure-
thra is enlarged, tapering to a normal caliber in the penile
shaft.77 Often, there is an orifice of a periurethral duct located
distal to the meatus that courses dorsal to the urethral channel
for a short distance. It has a blind ending and does not com-
municate in any way with the urinary stream. The periurethral
duct corresponds with the Guérin sinus or Morgagni lacu-
nae.78 Unless these ducts are inadvertently closed, leading
to a blind-ending epithelial pouch, they are of no clinical
consequence.
PENILE CURVATURE
The curvature of the penis is caused by a deficiency of the nor-
mal structures on the ventral side of the penis. The causes of
penile curvature are diverse: a skin deficiency, a dartos fascial
deficiency, a true fibrous chordee with tethering of the ventral
shaft, or deficiency of the corpora cavernosa on the concave
(ventral) side of the penis.
Occasionally other penile anomalies, which represent
variations of the embryologic defect causing hypospadias,
CHAPTER 121 HYPOSPADIAS 1539
FIGURE 121-10 Urethral plate in a newborn human penis with proximal
hypospadias (smooth muscle alpha actin immunostaining, 25). The ure-
thral plate is well vascularized, without any evidence of fibrosis or scarring.
Epithelial nest or an abortive attempt at the formation of the urethra is
seen within the plate (inset).
are reported. They can be characterized as a defect in the
course of the urethra (congenital urethral fistula) and a group
of defects characterized by curvature of the penis (chordee)
without hypospadias.80,81
SKIN AND SCROTAL ABNORMALITIES
The skin of the penis is radically changed as a result of the dis-
turbance in the formation of the urethra. Distal to the meatus,
there is often a paucity of ventral skin, which may contribute
to penile curvature. The frenulum is always absent in hypo-
spadias. Vestiges of a frenulum are sometimes found inserting
on either side of the open navicular fossa.
The skin proximal to the urethral meatus may be extremely
thin, to the extent that a catheter or probe passed proximally
may be readily apparent through the tissue-paper thickness of
skin. The urethral plate extending from the meatus to the
glanular groove may be well developed. Even with a meatus
located quite proximal on the shaft, this normal urethral plate
is elastic and typically nontethering. Artificial erection demon-
strates no ventral curvature in these situations. A normal
urethral plate may be incorporated into the surgical repair.
However, if the urethral plate is underdeveloped, it will act
as a tethering fibrous band that bends the penis ventrally
during artificial erection. When this fibrous tissue is divided,
the penis frequently straightens.
1540 PART VIII GENITOURINARY DISORDERS

Normally, the genital tubercle should develop in a cranial
position above the two genital swellings. The penis may be
caught between the two scrotal halves and become engulfed
with fusion of the penoscrotal area. The boundary between
the penis and scrotum may be formed by two oblique
raphes that extend from the proximal meatus to the dorsal
side of the penis.
TREATMENT
The object of therapy is to reconstruct a straight penis with a
meatus as close as possible to the normal site (ventrum of the
terminal aspect of the glans) to allow a forward-directed urine
stream and normal coitus. There are five basic phases for a
successful hypospadias outcome: (1) orthoplasty (straighten-
ing), (2) urethroplasty, (3) meatoplasty and glanuloplasty,
(4) scrotoplasty, and (5) skin cover. These elements can be ap-
plied sequentially or in various combinations to achieve surgi-
cal success (Fig. 121-11). They are described in conjunction
with a number of specific surgical techniques in this section.
DISTAL OR ANTERIOR HYPOSPADIAS
The treatment of distal hypospadias is dependent on the cul-
tural preference of the child’s family. Many patients with distal
hypospadias do not have a functional defect, lack significant
penile curvature, and will be able to stand and void with a
straight stream. Therefore the goal of placing the meatus in
its normal position within the glans is essentially cosmetic.
The outcome needs to be as close to perfect as possible.
The present standard is outpatient surgery. The technique
chosen depends on the anatomy of the hypospadiac penis.
Algorithm for Hypospadias Repair
Preservation of urethral plate
Skin and dartos dissection
Distal Proximal
Glans configuration
Assess curvature
Meatal quality and location
Urethral plate width
(finger test)
Dorsal plication if necessary
Rarely resect plate
(dermal graft)
Assess curvature
Dorsal plication if necessary
MAGPI
Onlay
GAP/Pyramid
Two stage (foreskin amount)
Tubularization
Bracka two-stage buccal graft
Snodgrass modification
FIGURE 121-11 Algorithm for hypospadias repair. GAP, Glans approxi-
mation procedure; MAGPI, meatal advancement glansplasty.
The most common procedures performed presently are the
meatal advancement with glansplasty incorporated (MAGPI),
the glans approximation procedure (GAP), primary tubu-
larization, the Mathieu or flip-flap, and the Duplay or primary
tubularization with the incision of the urethral plate
(Snodgrass modification) when the plate is too small for
primary closure.75,82–86
MAGPI Technique
The MAGPI technique was devised by Duckett in 1981. This
technique provides outstanding results in appropriately
selected patients. The hypospadiac penis that is amenable to
the MAGPI technique is characterized by a dorsal web of tissue
within the glans that deflects the urine from either a coronal or
a slightly subcoronal meatus. The urethra itself must have a
normal ventral wall, without any thin or atretic urethral
spongiosum. The urethra must also be mobile so that it can
be advanced into the glans (Fig. 121-12).
A circumferential incision is made around the corona
proximal to the meatus on the ventrum and 5 mm below
the coronal margin to create a Firlit collar on the dorsum
(see Fig. 121-12, A). The meatal advancement of the dorsal
urethral wall is accomplished by a Heineke-Mikulicz vertical
incision and horizontal closure (see Fig. 121-12, B). More
commonly, a wedge of glanular tissue that includes the glanu-
lar meatal wall is removed. The horizontal closure flattens
out the glanular bridge and permits the dorsal urethra to be
advanced out onto the glans tissue to the apex of the glanular
groove, where it is sutured with interrupted 7-0 Vicryl (see
Fig. 121-12, C). This flattens the deflecting ridge of glans
and permits the stream to be directed forward.
The glansplasty is made by reconfiguring the flattened
glans into a conical shape (see Fig. 121-12, D). By rotating
the lateral wings around to the midline proximal to the mea-
tus, a proper conical glans shape can be recreated. The deep
glanular tissue is brought together with interrupted 6-0 Vicryl,
and the superficial epithelial edges are closed with 7-0 chromic
suture (see Fig. 121-12, E). In this way, mesenchymal glans
tissue heals to glans tissue between the epithelial layers of
the urethra and the outer epithelium of the glans; this prevents
meatal retraction. The rotation of the glans wings reconfigures
a nearly normal glanular appearance. When there is ventral
skin deficiency, dorsal preputial skin can be transposed in a
Byar flap fashion to the ventrum (see Fig. 121-12, E). No
stents or catheters are required for diversion in children youn-
ger than 12 months. Preservation of the foreskin is possible
with the MAGPI procedure as long as penile curvature is min-
imal.46,87 Polus and Lotte,122 Gilpin,66 and Snodgrass88 dem-
onstrated foreskin preservation in anterior hypospadias repair
without chordee (Fig. 121-13).88–90
GAP Procedure
The GAP procedure is applicable in a small subset of patients
with distal hypospadias who have wide and deep glanular
grooves.86 These patients do not have a bridge of glanular tis-
sue that typically deflects the urine stream, as seen in patients
who would be more appropriately treated with the MAGPI
procedure. In the GAP procedure the wide-mouthed urethra
is tubularized primarily, over a stent (Fig. 121-14). Ventral gla-
nular tilt, meatal retraction, and splaying of the urine stream
can result from the inappropriate use of the MAGPI technique
in these circumstances.
 A B

C D E
FIGURE 121-12 MAGPI hypospadias technique. A, Initial circumferential subcoronal incision. B, Heineke-Mikulicz closure of the dorsal meatus after ex-
cision of the dorsal web skin bridge. C, Exposure of the glans mesenchyme, the most critical step, is accomplished by trimming the excess skin (dashed lines)
and advancing the mobile urethra with the use of a 6-0 chromic suture or a skin hook. D, Two-layer closure of the glans mesenchyme over the advanced
urethra, allowing for a normal-appearing glans with excellent support of the urethra. E, Skin closure with a sleeve approximation of the penile shaft skin.
If there is ventral skin deficiency, a Byar flap rearrangement with a standard midline seam is appropriate. (From Hinman F Jr: Atlas of Pediatric Urologic
Surgery. Philadelphia, WB Saunders, 1994, pp 575-578.)

A B
FIGURE 121-13 A, Preoperative patient with anterior hypospadias amenable to the MAGPI procedure. Note the dorsal web of tissue. B, Postoperative
result of the MAGPI procedure. (From Duckett JW, Baskin LA: Hypospadias. In Gillenwater J, Grayhack JT, Howards SS, Duckett JW (eds): Adult and Pediatric
Urology, 3rd ed. St Louis, Mosby-Year Book, 1996.)
1542 PART VIII GENITOURINARY DISORDERS

A B

FIGURE 121-14 GAP hypospadias

technique. A, Initial incision. B, Ex-
posure of the glans mesenchyme
by de-epithelialization of tissue. This
is critical for a two-layer glans clo-
sure and provides good support of
the urethroplasty. C, Tubularization
of the neourethra, followed by glans
closure. D, Completed repair. E, An-
terior hypospadias with a patulous
fish-mouth urethra amenable to F
the GAP procedure. F, Postoperative
outcome. C D

Urethral Mobilization
Another method of dealing with a subcoronal meatus is ure-
thral mobilization. Reported as far back as 1917 by Beck,91
urethral mobilization was revived by Koff and de Sy.92–94
This technique offers little advantage over the other distal
techniques and requires a more extensive procedure to mobi-
lize the penile urethra potential compromising the urethra’s
vascular supply. The meatal stenosis rate is significant.
Pyramid Procedure
In 6% of those with distal hypospadias, the prepuce is intact and
a megameatus exists under the normal foreskin, with a wide
glanular defect and no penile curvature (Fig. 121-15).77 These
anomalies may be recognized only after circumcision is per-
formed, potentially making correction more complex. Duckett
and Keating designed the “pyramid” technique to repair the
megameatus variant.77 The enormous distal urethra is carefully
dissected down the shaft of the glans and distal penis by way of
a four-quadrant exposure (hence the pyramid designation).
The urethra is tapered and buried in the glans, similar to an
epispadias repair (see Fig. 121-15, C).
Mathieu or Perimeatal-Based Flap Procedure
When the meatus is too proximal on the shaft to perform a
MAGPI procedure, or when there is no deep glanular groove
appropriate for a GAP or in situ tubularization technique (see
later), the meatus is advanced onto the glans using the tech-
nique described in 1932 by the French surgeon Mathieu.84
This technique involves the use of a perimeatal skin flap,
based on the intrinsic blood supply. To ensure viability, the
length-to-width ratio of the skin flap should not exceed 2:1.
The flap has also been used essentially as a free skin graft, with
A B C
FIGURE 121-15 Megameatus hypospadias repaired by the pyramid pro-
cedure. A, Patients with a megameatus have a wide glanular defect and no
penile curvature. Calibration reveals that the meatus is 22 to 24 Fr in a new-
born baby, compared with the normal caliber of 12 to 14 Fr. Often this
anomaly is recognized only after circumcision. Correction, however, is
the same as if the foreskin were intact. B, The technique involves careful
periurethral dissection, exposing the urethra and glans tissue and then re-
moving a wedge of the abnormally enlarged urethra. C, The exposed glans
tissue is closed over the newly closed neourethra. Technically, this proce-
dure involves careful dissection down the shaft and distal penis by way of a
four-quadrant exposure, hence its designation as the pyramid technique.
Tapering the urethra and burying it into the glans are similar to the tech-
nique used for epispadias.
no attempt made to preserve any of the subcutaneous tissue.
A pedicle of foreskin is brought around, as in a vascularized
flap technique, and used as a recipient bed. When designing
a Mathieu repair (Fig. 121-16), the ventral skin must be able to
be advanced to its new location on the glans. Some modi-
fications of the Mathieu procedure involve a second layer as
a subcutaneous pedicle.

A
C D
FIGURE 121-16 Mathieu procedure. A, Dotted lines outline the skin
flaps. B, The proximal flap is rotated, and the lateral glans flaps are devel-
oped. C, The glans flaps cover the neourethra, and preputial skin is moved,
if necessary. D, Completed repair. (From Duckett JW: Hypospadias. In
Walsh PC, Gittes RF, Perlmutter AD, et al [eds]: Campbell’s Urology,
5th ed. Philadelphia, WB Saunders, 1986, pp 1969-1999.)
Redman95 revived the Barcat96 modification of the Mathieu
procedure by mobilizing a glans flap in addition to the peri-
meatal-based flap and splitting the glans dorsally to bury
the urethral extension farther toward the apex of the glans.
Koff and colleagues93 have extensive experience with this
technique and have reported excellent results.
Tubularized Plate Urethroplasty with and Without
Snodgrass Modification
Historically, if the urethral groove was not wide enough for
tubularization in situ, an alternative approach such as the
Mathieu technique or, for more severe hypospadias, a vascu-
larized pedicle flap was taken. More recently, the concept of
incising the urethral plate, with subsequent tubularization
and secondary healing, was introduced by Snodgrass
(Fig. 121-17).75 Short-term results have been excellent, and
this procedure is enjoying extensive popularity.85 One appeal-
ing aspect is the slitlike meatus, which is created with a dorsal
midline incision. This technique has also been applied to more
posterior forms of hypospadias.97 Theoretically, one concern
is the possibility of meatal stenosis from scarring, as occurs
in patients with urethral stricture disease; in the latter, di-
rect-vision internal urethrotomy often leads to recurrent stric-
ture. However, reports of meatal stenosis have been rare.98
In hypospadias, the native virgin tissue, with its excellent
blood supply and large vascular sinuses, seems to respond
to primary incision and secondary healing without scarring.
The tubularized incised plate urethroplasty is conducive to
preservation of the foreskin (Fig. 121-18).88 To preserve the
foreskin, the incision is made only on the ventrum; therefore
CHAPTER 121 HYPOSPADIAS 1543
A B
FIGURE 121-17 Tubularized plate urethroplasty. A, Deep incision in the
urethral plate down to corporal tissue. B, Tubularization of the neourethra,
with subsequent glansplasty.
patients with significant penile curvature are not candidates
for this procedure. A three-layer closure of the prepuce
prevents foreskin fistula. The fact that the foreskin cannot
be used as a de-epithelialized flap theoretically increases the
possibility of urethral fistula.
POSTERIOR HYPOSPADIAS
Urethral Plate Preservation
Duckett popularized the concept of preserving the urethral
plate, which is now standard practice for the repair of almost
all hypospadias cases.83,99 The urethral plate serves as the dorsal
urethral wall, and the ventral urethra is created by a vascular
onlay flap of tissue from the inner prepuce. Extensive experience
has shown that the urethral plate is rarely the cause of penile
curvature (Fig. 121-19). This knowledge was gained by repeti-
tive resection of the urethral plate and subsequent artificial
erection, which showed no improvement in correction of the
penile curvature (see Fig. 121-19).99 Further efforts and experi-
ence have shown that the urethral plate is typically supple
and pliable and that ancillary penile strengthening procedures
such as midline dorsal plication (see later), with preservation
of the urethral plate, have led to fewer complications such as
fistula and stenosis at the proximal anastomosis.99,100
The concept of preserving the urethral plate but undermin-
ing it108,110 and exposing the corporal bodies—with the goal
of releasing the chordee tissue—has been advocated.101,102
Careful anatomic studies have shown that there is an extensive
network of blood vessels supplying the urethral plate in the
hypospadiac penis, and lifting the urethral plate defeats the
purpose of preservation by violating this intricate blood
supply.66 Historically, posterior hypospadias was approached
by complete resection of the abnormal urethra and all tissue
down to normal corporal bodies. The urethra was replaced
by a tubularized vascular preputial flap from either the inner
or outer prepuce.103–105 Presently, in the majority of posterior
hypospadias cases including perineal hypospadias, the ure-
thral plate can be preserved and a vascularized flap used
in an onlay fashion. In the rare case in which the urethral
plate needs to be resected, a two-stage technique can be used
(see later).
1544 PART VIII GENITOURINARY DISORDERS

A B C D

E F G H
FIGURE 121-18 Foreskin preservation technique for distal hypospadias. A, Distal hypospadias. B, The incision is marked for foreskin preservation.
C, Dissection of ventral skin. Note the thin spongiosum with the distal urethra exposed. D, Glans wings have been mobilized, and the distal urethroplasty
is completed by the Snodgrass technique. E, Two-layer glansplasty. F, Three-layer foreskin reconstruction: (1) inner prepuce, (2) subcutaneous tissue to
prevent foreskin fistulas, and (3) outer foreskin. G, Completed repair with foreskin retracted. H, Completed repair.

Onlay Island Flap

A B
C D
FIGURE 121-19 A, Penile curvature with preservation of the urethral
plate. B, Resection of the urethral plate with continued severe curvature.
C, Penile curvature with preservation of the urethral plate. D, Resection of
the urethral plate with continued severe curvature. Extensive experience
has shown that the urethral plate is rarely the cause of penile curvature.
The blood supply to the hypospadiac preputial tissue is reli-
able and easily delineated (see Fig. 121-18).71,106 The abun-
dance of cutaneous tissue on the dorsum of the penis is
vascularized in a longitudinal fashion.107 This tissue may be
dissected from the penile skin, creating an island flap from
the inner layer of the prepuce. The blood supply to the dorsal
skin of the foreskin and the penile skin comes from its broad
base and is not dependent on the subcutaneous tissues, except
at the remote edges of the dorsal preputial skin. The tips of the
distal portion of the penile skin flaps can be excised and not
used in the repair.
All cases of posterior hypospadias (i.e., with or without
penile curvature) are approached by initially leaving the ure-
thral plate intact. This technique can be applied to the penile
shaft, as well as scrotal and perineal hypospadias. The intact
dorsal plate essentially avoids the complication of proximal
stricture, and the excellent blood supply has decreased the
fistula rate to approximately 15% for all cases of onlay island
flap hypospadias repair (Fig. 121-20).99 For shorter repairs,
the flap may be dissected from half the prepuce, as described
by Rushton and Belman,108 leaving the remaining half of the
foreskin available for a second layer of coverage. Long-term
results with the onlay island flap have been durable.101,109,110
For severe hypospadias, the prepuce can be designed in a
horseshoe style to bridge extensive gaps (see Fig. 121-20, K).
Transverse Tubularized Island Flap
The technique of using the transverse tubularized island
flap was used extensively before the concept of preserving
the urethral plate (Fig. 121-21). It is still successful in severe
 CHAPTER 121 HYPOSPADIAS 1545

A B C

D E
FIGURE 121-20 Onlay island flap hypospadias repair. A, A U-shaped incision is made around the urethral plate, preserving a dorsal urethral strip
approximately 8 mm wide. B, Takedown of the skin and subcutaneous tissue and outlining of the inner prepuce for the onlay island flap. Glans wings
are mobilized along the plane of the corporal body and the glans mesenchyme. C, Preservation of the urethral plate with penile curvature in a case
of penoscrotal hypospadias. D, Suturing of the onlay flap to the urethral plate with running 7-0 sutures. The flap is trimmed to obtain a 12-Fr bougie
in a 1-year-old child to prevent the complication of urethral diverticulum, which results from leaving excess tissue. The glans wings are approximated over
the new urethra after maturing the meatus. The skin is then closed by a classic Byar flap rearrangement. E, The split prepuce in situ technique for dissection
of a short vascularized onlay island flap. This method is useful for shorter onlays, leaving the remaining half of the foreskin available for a second layer of
coverage.
Continued

cases when the urethral plate needs to be resected,
although long-term problems with diverticulum have
resulted in a high reoperation rate. Technical nuances involve
an oblique proximal anastomosis with interrupted sutures
to avoid stenosis; fixation of the neourethra to the corporal
bodies to prevent diverticulum and improve ease of catheter-
ization; and a wide glans channel made underneath the
glans cap, against the corporal bodies to avoid meatal
stenosis.104
Two-Stage Hypospadias Repair
An alternative approach for severe hypospadias is to transfer
the dorsal prepuce to the ventrum after correcting the penile
curvature (Fig. 121-22). In severe cases the urethral plate may
need to be resected to correct chordee. Dermal grafting may be
required, and performing a urethroplasty on top of the healing
graft is not suggested. Instead, Byar flaps can be rotated from
the dorsum, setting up ventral coverage for subsequent ure-
throplasty.111 Occasionally the chordee can be corrected with-
out resection of the urethral plate. In this case the dorsal skin
can be sutured to each side of the preserved urethral plate (see
Fig. 121-22, D). The second stage is performed at least
6 months after the first stage. To assist the urethroplasty within
the glans, dorsal skin can be tucked within the glans wings
during the first stage. Subcutaneous secondary coverage of
the reconstructed urethra is performed to prevent fistula. In
cases where local tissue is not readily available, a tunica vagi-
nalis flap from the testicle can be mobilized and used to cover
the urethroplasty.112–114 Long-term results documented
through puberty have been durable.115
1546 PART VIII GENITOURINARY DISORDERS
F G
J K
FIGURE 121-20—CONT’D F, Penoscrotal hypospadias amenable to the onlay island flap technique. G, Preservation of the urethral plate and flap dis-
section. H, Suturing of the onlay to the urethral plate. I, Completed repair. J, Perineal hypospadias. K, Horseshoe vascularized onlay island flap to bridge a
long urethral deficit. L, Onlay flap rotated to bridge the urethral defect. M, Completed repair.
Bracka Two-Stage Buccal Graft Repair
For patients with prior surgery or with severe hypospadias,
Bracka described a two-stage buccal graft repair.116–118 In
the first stage the penis is straightened, and the scarred urethra
is discarded (Figs. 121-23 and 121-24). Buccal mucosa is
harvested from either the cheek or the lip and grafted to the
prepared bed.119 Extensive quilting of the graft is performed
to prevent hematoma from lifting off the buccal mucosa. Dur-
ing the first stage, glans wings are mobilized in preparation for
the creation of a slitlike meatus during the second stage. The
second-stage urethroplasty is undertaken at least 6 months
after the first stage. In the second stage, excess buccal mucosa
is trimmed off the glans, setting up a two-layer glans closure
(see Figs. 121-23 and 121-24). The buccal mucosa is rolled
into the new urethra, and subcutaneous tissue is used for
secondary coverage.
PENILE CURVATURE
Correction of penile curvature has evolved along with the
concept of preserving the urethral plate. Artificial erection, in-
troduced by Gittes and McLaughlin in 1974, provides a mech-
anism to check for penile curvature and the success of
correction at the time of surgery.120 A tourniquet is placed
at the base of the penis, and a corpus cavernosum is injected
with saline. Both corporal bodies fill, so it is possible to deter-
mine the extent of curvature and the success after correction.
The assurance of complete correction is essential before
proceeding to urethroplasty and one-stage repair. There have
been no reports of damage to the cavernous tissue with this
H I
L M
technique, as long as care is taken to ensure that injectable
saline is used.
In reality, all penises have penile curvature during develop-
ment (Fig. 121-25). The majority of penile curvature resolves
when the penile skin is dissected to the penoscrotal
junction—hence the term “skin chordee.”121 Curvature may
also result from differential growth of the dorsal and ventral
aspects of the corpora body.122 Finally, in rare cases the
urethra itself may be short or atretic, requiring resection to
correct curvature and subsequent augmentation.
Historically, chordee was corrected by a modification of
Nesbit’s dorsal plication, taking out wedges of tunica albugi-
nea in an ellipse and closing this with permanent suture.123
This technique was first described by Syng Physick, the “father
of American surgery,” in the early nineteenth century. Physick
treated chordee by shortening the dorsal tunica albuginea.124
When the arc of maximal curvature is identified during
artificial erection, wedges of tunica albuginea are excised in
a stepwise fashion. These diamond wedges are closed trans-
versely with permanent suture until the penis is straight.
Nesbit’s technique has been modified into dorsal tunica albu-
ginea plication for the correction of penile curvature in the
setting of corporal disproportion, as well as hypospadias.100
On the basis of anatomic studies of the human fetal penis, a
simpler approach—placing dorsal midline plication sutures
in the nerve-free zone at the 12 o’clock position—is now
advocated.66 Midline dorsal plication avoids the need for
mobilization of the neurovascular bundle (Fig. 121-26).72
Midline plication can be applied to mild to moderate degrees
of curvature (Fig. 121-27).125 If more than two rows of
plication sutures or more than four permanent sutures are
 CHAPTER 121 HYPOSPADIAS 1547

B C

D E

FIGURE 121-21 Transverse preputial island flap with glans channel hypospadias repair. A, Correction of penile cur-
vature by surgical release of the skin and dartos fascia and ancillary straightening procedures as needed. B, The trans-
verse island flap is designed. C, Development of the island flap by dissecting subcutaneous tissue from the dorsal
penile skin. D, The transverse preputial island flap is developed and tubularized to 12 Fr, which is monitored by a
bougie à boule. The distal edges of the tube are sewn with interrupted sutures so that the edges can be trimmed
to the appropriate length. E, Rotation to the ventrum must avoid torsion of the shaft by freeing the base of the flap
adequately. A proximal oblique anastomosis is made, fixing the urethroplasty to the tunica albuginea along its pos-
terior anastomosis. A wide glans channel is made underneath the glans cap against the corporal bodies by removing
glans tissue within the channel. F, The neourethra is tacked to the corporal bodies. G, Lateral transposition of Byar flaps
of dorsal penile skin to the midline and excision of the tips. The repair is stented with a 6-Fr catheter. (From Hinman F
G Jr: Atlas of Pediatric Urologic Surgery. Philadelphia, WB Saunders, 1994, pp 590-592.)

necessary, however, an alternative approach such as complete
resection of the urethral plate and dermal grafting should be
considered. Midline dorsal plication has also been effective for
recurrent curvature.126 During artificial erection, if the curva-
ture cannot be corrected with the surgeon’s finger, midline
dorsal plication is not advised. A glans tilt may also be repaired
by permanent sutures on the dorsum, but care must be taken
to avoid the neurovascular structures supplying the glans.18
In rare cases the tunica albuginea is so deficient on the ven-
trum that excision of the tunica is required, with replacement
1548 PART VIII GENITOURINARY DISORDERS

A B C D E

F G H I
FIGURE 121-22 Two-stage hypospadias repair. The first stage includes correction of penile chordee and transfer of dorsal foreskin to the ventral aspect of
the penis. A, Scrotal hypospadias with chordee and penoscrotal transposition. B, Foreskin attached to the scrotal skin. C, Penile straightening with removal
of the ventral tethering urethral plate. D, In select cases, the urethral plate can be preserved and the dorsal skin split and wrapped to the ventrum. E, First
stage complete. F, Second-stage urethroplasty is done 6 months after the first-stage surgery. G, Urethroplasty. H, Completed repair. I, Six-month follow-up.

A B C D

E F G H
FIGURE 121-23 Two-stage Bracka buccal hypospadias repair. First stage: A, Patient with midshaft hypospadias and a paucity of available skin after mul-
tiple previous hypospadias repairs. B, Resection of scar tissue. C, Mobilization of glans wings. D, Buccal free graft quilted into the resected scar. The second
stage is done after 6 months of healing. E, Exposure of the glans mesenchyme, and trimming of the buccal graft for subsequent urethroplasty. F, Meatal
stitch for start of urethroplasty. G, Secondary de-epithelialized pedicle coverage of the urethroplasty. H, Two-layer glansplasty and completed repair.
 B

A D E

F G
FIGURE 121-24 A, Patient with a distal urethral stricture after multiple previous hypospadias repairs. B, Urethrogram. C, Outline of buccal mucosa graft
harvest. D, Free buccal harvest prepared for grafting. E, Quilted buccal mucosa graft applied to the ventral penis after resection of scar tissue. F, Healed
buccal mucosa 6 months after grafting. G, Completed repair.

A B C
FIGURE 121-25 A, Congenital fistula. B and C, Chordee without hypospadias. Abnormal development of the urethral spongiosum necessitates urethro-
plasty and causes penile curvature.
1550 PART VIII GENITOURINARY DISORDERS

FIGURE 121-26 Midline dorsal

plication technique. In this tech-
nique, parallel plication sutures are
placed in the tunica albuginea in
the 12 o’clock position, which is free
of both nerves and vascular struc-
tures. This technique requires a min-
imal amount of manipulation of the
penis. It is not necessary to incise
into the corporal body or exten-
sively mobilize Buck fascia. If the cur-
vature is severe, a maximum of two
rows of parallel plications can be
placed for correction.

A B C

D E F G
FIGURE 121-27 Midline dorsal plication technique. A to C, Correction of mild curvature. D to G, Correction of moderate to severe curvature. Note the two
permanent 5-0 Prolene plication sutures in the midline nerve-free zone.

by an elastic graft. I prefer the use of dermis, as described by
Horton and Devine,80 which has withstood the test of time
(Fig. 121-28).127
About 5% of patients need ancillary penile straightening
procedures after the release of skin curvature by aggressive
dissection to the penoscrotal junction.3,99,128 Of those requir-
ing ancillary procedures, most (>90%) can be corrected by
dorsal midline plication without the need for resection of the
urethral plate. This leaves a small minority of severe cases requi-
ring resection of the urethral plate, dermal grafting, or both.
PATIENTS WITH MULTIPLE FAILURES
Some patients undergo multiple hypospadias repairs that fail;
this unfortunate outcome can occur even in skilled hands.129
In such patients, it is often necessary to discard the previously
 CHAPTER 121 HYPOSPADIAS 1551

A B C
FIGURE 121-28 Dermal graft for severe ventral penile curvature. A, Scrotal hypospadias. B, Severe right-angle curvature after aggressive skin and
subcutaneous dissection. C, Ventral incisions for placement of dermal graft. (From Hinman F Jr: Atlas of Urologic Surgery. Philadelphia, WB Saunders,
1989, pp 91-92.)

created problem-plagued urethra and start anew. The penis of 1:100,000 epinephrine in 1% lidocaine (Xylocaine) using
may be further straightened, if necessary. In some cases it a 29-gauge needle. Infiltration around the urethra and into
may be appropriate to place a meshed split-thickness skin the glans is helpful for developing glans wings. Usually, only
graft and go back later for tubularization130,131 or to place a 1 to 1.5 mL are required. There is a wide margin of safety be-
free graft of buccal mucosa or skin where the scarred urethra fore epinephrine sensitizes the myocardium to arrhythmias
was resected.117 In patients with a paucity of penile skin, a with the use of inhalational anesthetics.144 A safe dose is
full-thickness skin graft can provide adequate skin cover- 1 mL/kg of a 1:100,000 solution. I do not believe that epineph-
age.132 I prefer harvesting the skin from the patient’s buttock rine compromises the vascularity of the flaps. Various methods
so that the scar from the graft site will be well hidden. A wound of coagulation have been recommended; I prefer a low-current
vac for 5 days postoperatively has ensured excellent graft take. electrocautery (Bovie) applied to the 0.5-mm forceps.
Rarely, tissue expanders have been useful to stretch local gen-
ital skin.133 In the past, bladder mucosa was used for the
ANALGESIA
neourethra134,135 because the long-term results with free skin
grafts were not as good as previously reported.136 Unfortu- In anticipation of postoperative discomfort, the anesthesia team
nately, bladder mucosa grafts were complicated by eversion places a long-acting caudal nerve block before the proce-
at the meatus in up to 30% of patients.134 Meatal dilatation dure.145,146 If the repair takes longer than 2 hours, the block
on a daily basis could not avoid this problem. More recently, can be repeated at the completion of the procedure. Evidence
buccal mucosa grafts have been used for urethral replacement suggests that the caudal block also reduces the amount of
in these difficult patients.134,137,138 I presently advocate the bleeding.147 Alternatively, a supplemental local block with 3 mL
two-stage approach of Bracka (see Fig. 121-23).118,139 of 0.5% bupivacaine (Marcaine) is placed just beneath the symp-
In the future it may be possible to treat difficult cases of hy- hysis to infiltrate the dorsal penile nerve bundle.148 Penile blocks
pospadias with autologous grafts of cultured urethral epithe- also reduce the general anesthetic needs and prevent erections,
lium.140,141 Finding the ideal support matrix to scaffold the which can be bothersome and cause extra bleeding.
urothelial cells may make this technique widely applicable.

DRESSING
Technical Considerations
------------------------------------------------------------------------------------------------------------------------------------------------
For distal hypospadias repair performed without urinary
The principles of hypospadiology have been enhanced over diversion, I currently use a Tegaderm wrap.3 The parents
the years by the precise, delicate tissue handling taught by remove the dressing at home in 24 to 48 hours. For hypospa-
plastic surgeons and the use of optical magnification. Sharp- dias repair with urinary diversion, a “sandwich” dressing allows
pointed iris scissors and delicate-toothed forceps are the the application of more diffuse pressure by placing the penis
principal tools. Castroviejo needle holders and 0.5-mm onto the abdominal wall. A layer of Telfa followed by a 4- by
forceps are useful for delicate maneuvers with 7-0 sutures. 4-inch gauze folded three times on the ventrum of the penis
Bougie à boule probes are useful as the procedure progresses is secured with a medium Tegaderm sheet of sticky plastic.
for calibrating the sizes of tubes and the anastomosis, as well as The dressing is typically removed at home after 48 hours.
the meatus. Fine lacrimal duct probes are useful for identify- I encourage bathing twice a day with the stent in place after
ing fistulas and periurethral ducts. 48 hours to allow the dressing to soak off. Warm bathwater
allows the repair to remain clean and assists healing.
HEMOSTASIS
DIVERSION
Control of bleeding from the vascular penis and glans can be
achieved in a number of ways. Placing a tourniquet at the base Bladder spasms caused by irritation of the trigone with
of the penis is simplest.142,143 Recommended tourniquet a catheter can be an aggravating problem. Oral oxybutynin
times vary from 20 minutes to 1 hour. I prefer the injection chloride may be of help. Another problem relates to
1552 PART VIII GENITOURINARY DISORDERS

connections between tubing and a drainage bag. I avoid a and often traumatic for the patient and family. Internal
Foley catheter, except in teenagers and adults, and prefer a urethrotomy also has poor long-term results. Often the glans-
“dripping stent” in infants aged 6 to 18 months. Kendall man- plasty will have to be taken apart and reperformed. In some
ufactures a prepackaged 6-Fr hypospadias catheter; alterna- cases a two-stage approach (Bracka) is necessary.117,118
tively, 5- and 8-Fr feeding tubes are effective. Urethral
diversions are used for 3 to 10 days. If the diversion is used
URETHROCUTANEOUS FISTULA
for 5 days or less, the family is advised to allow the stent to
fall out when the dressing comes off. If used for more than Urethrocutaneous fistulas are the most common late compli-
5 days, the temporary stents can be sutured to the glans with cation of hypospadias repair, and their incidence has been
5-0 polypropylene on a noncutting needle. I have had good used to evaluate the effectiveness of the surgical proce-
success with this technique, which allows the overwhelming dure.160,161 The expected fistula rate is between 10% and
majority of hypospadias repairs to be done on an outpatient 20% for most one-stage hypospadias surgery. Often, distal
basis. Patients are followed at 6 months to 1 year after surgery, obstruction is related to persistent fistulas. The principles of
after toilet training, and after puberty. The urethra is not repairing urethrocutaneous fistulas require wide mobilization,
instrumented during follow-up visits. After toilet training adjacent skin flaps, and multilayered closure.162–164 I recom-
the child urinary stream is observed to assess for occult fistula, mend using magnification and delicate instruments for fistula
diverticulum, and urethral stricture. repair. Dissection is carried down to the urethra, and a
7-0 suture is used to close the urethral edges in an inverting
fashion. Watertight closure is ensured by irrigating the urethra
AGE FOR REPAIR
with proximal compression. Irrigating the urethra will also
Hypospadias repair is being performed at progressively confirm the absence of an additional fistula. In straightforward
younger ages. Schultz and colleagues149 pointed out that an fistulas no urinary diversion is used. The patient is allowed
ideal age might be 6 to 18 months to minimize the emotional to void through his urethra and goes home on the same day.
effect of this traumatic experience. The consensus statement It is extremely important to recognize a concomitant ure-
on the timing of genital surgery from the American Academy thral stricture or diverticulum at the time of fistula closure.
of Pediatrics also supports early surgery before 18 months of These should be repaired to ensure the success of the fistula
age.150 Gender identity does not seem to be defined until after closure. Occasionally, a meatoplasty may be necessary at the
18 months. I now prefer to operate when patients are between time of fistula repair. A fistula near the glans is best repaired
4 and 9 months old; at that age, the penis is of sufficient size to by opening the bridge between the meatus and fistula, mobi-
achieve success comparable with that obtained by waiting until lizing the inner urethral edges, and closing in two layers with
age 2 to 5 years, which was previously popular. For many rea- good glans approximation (Fig. 121-29).
sons (medical and social), however, surgery may be delayed.
Kaplan and colleagues151–154 studied 69 boys aged 6 to 10
years who had had hypospadias repair in infancy and found
STRICTURE
no increase in significant psychopathology during childhood,
although more recent studies suggest that before age 5 is the Strictures result from technical problems during the initial hy-
most ideal for both functional and psychosocial outcomes. pospadias repair. Proximal anastomotic strictures may result
from either a luminal calibration miscalculation, resulting in
a narrow neourethra, or an anastomotic overlap. A meatal
TESTOSTERONE STIMULATION
stricture may result from chronic balanitis xerotica obliterans.
Enlargement of the infant penis is possible by testosterone stim- This reaction is located at the meatus or may extend into the
ulation. The recommended dose is 25 to 50 mg of testosterone more proximal urethroplasty. This may be caused by a poorly
propionate (2 mg/kg) intramuscularly given at 3-week intervals vascularized meatoplasty, especially the Mathieu procedure.
for up to three preoperative doses.14,155 In Europe dihydrotes-
tosterone transdermal gel has also proven to be effective.156
I have not found the routine use of testosterone stimulation
to be beneficial. In severe cases when the penis is small such
as partial androgen insensitivity syndrome or for diagnostic
purposes, testosterone is helpful. A patient with a small glans
may benefit from presurgical testosterone stimulation. There
is a suggestion in animal models that testosterone stimulation
may be detrimental to the later maturation of prostatic tissue
and penile growth.157,158 To date, clinical practice has not
documented any untoward effects in humans.159

Complications
------------------------------------------------------------------------------------------------------------------------------------------------

MEATAL STENOSIS A B
Meatal stenosis is caused by technical errors in operative FIGURE 121-29 Hypospadias fistula. A, Coronal fistula requiring redo
design or poor vascularization of the urethroplasty or glansplasty. B, Preservation of the urethral plate with eccentric vascular-
glansplasty.160 Postoperative dilatation is seldom therapeutic ized onlay flap preparation.
 CHAPTER 121 HYPOSPADIAS 1553

The most likely cause of meatal stenosis is vascular com- problems. Evidence shows that the neourethra grows with
promise of the urethra at the apex of the meatus. It may be the child. Early hypospadias repair with minimal hospitaliza-
secondary to an inadequate glans channel that compresses tion helps avoid separation anxiety and castration fears. We
the vascularity of the pedicle. Once a stricture has developed, can now counsel parents confidently that there is an excellent
it can be repaired by excision and reanastomosis, a vascular- chance of a good cosmetic, functional, and emotional result in
ized pedicle graft, or a two-stage buccal graft using the tech- boys with all degrees of hypospadias.
nique of Bracka.117,165 Of note, there has been little success
treating strictures secondary to hypospadias surgery with
optical internal urethrotomy.165
Summary
------------------------------------------------------------------------------------------------------------------------------------------------

DIVERTICULUM
1. Hypospadias should be repaired within the first year of life,
A fusiform urethral diverticulum may form because the preferably at 4 to 6 months of age. Pain and catheters seem
neourethra was made too wide and meatal stenosis allowed to be better tolerated at this age, and the baby’s lack of mo-
ballooning of the proximal urethra.1 Reduction of a divertic- bility simplifies postoperative care.
ulum may be necessary and should be done in a longitudinal 2. A terminal slitlike meatus should be the goal, with or with-
fashion.166 Most often, a circumcising incision is used and the out preservation of the foreskin in distal hypospadias,
penile skin is dropped to the penoscrotal junction, which depending on parental preference.
allows for longitudinal repair of the diverticulum without 3. Preservation of the urethral plate provides the best possible
overlying suture lines. Care must be taken to evaluate the chance of recreating normal urethral anatomy by incorpo-
neourethra for an associated fistula. A discrete diverticulum rating the abortive spongiosum into the repair.
may develop from urinary extravasation into the tissues 4. Midline dorsal plication is safe and effective for the correc-
adjacent to the anastomosis. tion of penile curvature in the majority of patients. (Placing
more than two rows of sutures is a sign that another tech-
nique such as dermal grafting is indicated.)
Results 5. In the small percentage of patients who require resection of
------------------------------------------------------------------------------------------------------------------------------------------------
the urethral plate, a two-stage approach is generally
Long-term follow-up reflects the older age at which repair was warranted.
undertaken and the difficulties of hypospadias surgery in the 6. Vascularized pedicle onlay flaps are successful in primary
past.167–170 Today’s results are much better both cosmetically and redo hypospadias surgery.
and functionally than those in the past.171–174,175 The use of 7. De-epithelialized vascular flaps should be used as a second
one-stage hypospadias repair at an early age with a low com- layer for all urethroplasties.
plication rate encourages our current positive outlook for pa- 8. Patients with a paucity of skin are best managed with the
tients with this condition. Curvature correction with the aid of Bracka two-stage buccal repair.
an artificial erection is extremely important for ensuring satis- 9. Coronal fistulas require a redo glansplasty.
factory sexual function. With the placement of the urinary me-
atus at the tip of the glans, the fertility potential has been The complete reference list is available online at www.
improved, unless the patient has other coexisting testicular expertconsult.com.
 nonobstructive and are of historical interest only. Type I rep-
resents 95% of PUVs. They are membranes that originate at
the verumontanum and travel distally to insert in the anterior
proximal membranous urethra with an opening present
posteriorly at the verumontanum. The etiology is probably a
result of the mesonephric ducts entering the cloaca more an-
teriorly than normal and fusing in the midline.7 Type III PUVs
represent the other 5% and consist of a ring-type membrane
distal to the verumontanum with a perforation present
centrally. The membrane may occasionally migrate distally,
forming a windsock appearance.8 The cause of these PUVs
is an incomplete dissolution of the urogenital membrane.
Presently, most cases of PUVs are detected prenatally by US
showing hydronephrosis and/or a distended thick-walled
bladder. Postnatally, the US will show a thickened bladder
wall and classically a dilated and elongated posterior urethra.
Hydronephrosis will vary in degree and may be unilateral or
bilateral. Physical examination may exhibit a distended, firm
bladder, weak urinary stream, abdominal distention, and,
possibly, urinary ascites. If the fetus has been subjected to oli-
gohydramnios, there may be respiratory distress and the stig-
mata of Potter syndrome may be present. Older boys may
present with urinary tract infection (UTI) and voiding dys-
CHAPTER 122 function, particularly daytime urge incontinence. The voiding
cystourethrogram (VCUG) is usually diagnostic with a dilated
and often elongated posterior urethra and abrupt transition
to a narrower distal urethra; a thickened bladder wall, trabe-
Abnormalities of culation, bladder diverticulum, and vesicoureteral reflux
(VUR) may also be seen (Fig. 122-1).
Fetal intervention for PUVs is controversial. The diagnostic
the Urethra, Penis, accuracy of prenatal US has improved significantly over the
past 10 years, but nonobstructive disorders such as prune-

and Scrotum belly syndrome and high-grade VUR may be difficult to

differentiate from PUVs. Fetal intervention may involve early
delivery, vesicoamniotic shunting,3,9 amnioinfusion,10 percu-
J. Patrick Murphy and John M. Gatti taneous fetal cystoscopy,11–13 or open fetal surgery.14 All of
these have significant risks to the fetus and mother including
preterm labor, bleeding, and infection. Therefore the consid-
eration of fetal intervention should be limited to centers with
experienced personnel in both the diagnostic and technical
The evaluation of abnormalities of the urethra, penis, and skills involved. Renal dysplasia (RD) may occur early in fetal
scrotum now begins before birth. Prenatal ultrasound (US) development before the time of consideration for fetal manip-
has become standard of care, and with it has come the early ulation. Intervention to relieve obstruction will not reverse
diagnosis of many congenital anomalies of the genitourinary RD. Therefore most clinicians agree that a fetus with signs
tract. Virtually every hydronephrotic or obstructive lesion of severe RD is not a candidate for prenatal treatment. Several
has been documented prenatally, but new controversies have parameters are used for determining the severity of RD.
been generated regarding the efficacy and appropriateness of Increased echodensity of the renal parenchyma, cystic
prenatal therapy. In no area is this more apparent than in the changes of the parenchyma, and early moderate or severe oli-
diagnosis of posterior urethral valves (PUVs). Interventions in gohydramnios are all US signs of severe RD. The character of
the way of vesicoamniotic shunting and fetoscopic urethral percutaneously sampled fetal urine also predicts the degree of
valve ablation have been described and used, but a benefit RD.1,2,15,16 Normal fetal urine is hypotonic and low in so-
in outcome has yet to be confirmed.1–3 dium. Fetal urine sodium greater than 100 mEq/L, osmolality
greater than 210 mOsm, protein greater than 20 mg/dL, and
b2-microglobulin greater than 4 mg/L17 are suggestive of sig-
nificant RD. Improvement in these parameters on serial blad-
Posterior Urethral Valves der taps is encouraging, but a fetus with these radiographic
------------------------------------------------------------------------------------------------------------------------------------------------
and urine parameters probably would not benefit from prena-
PUVs are the most common obstructive anomaly of the tal manipulation. Prenatal intervention has not been shown to
urethra. The incidence is between 1 in 5000 and 1 in 8000 improve the overall long-term renal function when compared
male births.4,5 Hugh Hampton Young is credited with the first with conventional postnatal therapy.3,9 However, the patients
description and classification of PUVs.6 He described three who are treated prenatally likely represent the more severe
types. Type II PUVs are now generally considered to be degree of obstruction, and it could be argued that these
1555
1556 PART VIII GENITOURINARY DISORDERS
FIGURE 122-1 Voiding cystourethrogram showing posterior urethral
valves with dilated posterior urethra (open arrow) and abrupt transition
to narrower anterior urethra (solid arrow). The trabeculated bladder is
above the posterior urethra.
patients should be predicted to have worse outcomes than
those treated postnatally. Therefore prenatal treatment may
have had a beneficial effect in these higher-risk patients.
The ultimate role of fetal intervention for PUVs is still evolv-
ing; newer technology allowing safer and earlier treatment
may improve outcomes.2,18
Postnatal management of PUVs initially involves obtaining
bladder drainage. Usually this can be accomplished with a
small, soft 5- or 8-Fr feeding tube passed per urethra. Care
must be taken to ensure it does not coil in the dilated posterior
urethra. Bladder US can help confirm proper placement. Gen-
erally, Foley catheters should be avoided for initial drainage
because the balloon may cause bladder spasm in the thick-
walled bladder and affect urine drainage, or the balloon
may slip into the dilated posterior urethra (Fig. 122-2).
In the rare case that the urethra cannot be cannulated, percu-
taneous suprapubic access can be used. Fluid and electrolyte
management is critical in the first 24 to 48 hours. Postobstruc-
tive diuresis may occur, and both water and solute may be
rapidly depleted, requiring aggressive fluid and electrolyte
replacement. Acid-base balance is also important and is more
of a problem in severe renal insufficiency. Serum creatinine is
monitored closely, realizing that for the first few days of life
the value reflects the maternal renal function. Historically, when
the creatinine stayed elevated beyond several days, supravesical
diversion was considered. However, a number of studies suggest
this is not necessary and subsequent renal function was not
improved when supravesical diversion was compared with treat-
ment with standard postnatal valve ablation or vesicostomy.19–21
Antibiotic prophylaxis is indicated, especially in those patients
presenting with hydronephrosis or VUR.
When fluid electrolyte status is stable, most patients can be
treated with endoscopic valve ablation. With advanced fiber-
optic technology, cystoscopes of 7-Fr size are now available,
allowing urethral access in all but the smallest premature
infants. A Bugbee or angled wire electrode with cutting
current or a small-caliber laser fiber22 can be used to incise
the valves at the 5, 7, and 12 o’clock positions in a retrograde
transurethral fashion. In the older child, a small resectoscope
FIGURE 122-2 Voiding cystourethrogram with posterior urethral valves
and heavily trabeculated bladder and Foley balloon inflated in dilated
posterior urethra (arrow).
can be used. Some have advocated a percutaneous antegrade
approach to fulgurate PUVs.23 In the rare case in which the
valves cannot be fulgurated endoscopically, cutaneous vesi-
costomy is used. The Blocksom technique is favored, bringing
the bladder dome to the skin to decrease the chance of bladder
prolapse.24,25 When the child has grown and stabilized med-
ically, the valves may then be ablated endoscopically and the
vesicostomy closed. Vesicostomy probably does not decrease
long-term bladder capacity or function.5,26,27
The long-term prognosis for patients with PUVs has
improved over the decades and is affected mostly by three
factors: (1) degree of RD; (2) incidence of UTI with or without
VUR; and (3) bladder function. The overall infant mortality
rate has improved from about 50% to 1% to 3% in the past
3 decades.28,29 Improved neonatal critical care, along with
careful attention to the treatment of these factors, are respon-
sible for this improvement. RD is irreversible, but attention to
the other issues of UTI and bladder dysfunction can decrease
or delay ongoing renal deterioration. Renal failure occurs in
as high as 40% of patients treated for PUVs.19,20 Renal trans-
plantation has improved in this group of patients owing to
improved medical care and modern immunosuppressive ther-
apy. The graft survival is comparable with patients without
significant urologic pathology.30,31 Attention to bladder dys-
function and treating these higher pressure bladders for urge
incontinence with anticholinergic therapy and frequent void-
ing are all important to delay renal deterioration or to protect
the transplanted kidneys. In bladders that have progressed to
myogenic failure and incomplete emptying, clean intermittent
catheterization and/or overnight bladder drainage may be
necessary. Urodynamics are helpful in directing this type of
bladder therapy. In rare cases, bladder augmentation may be
necessary.20,32–34 VUR occurs in up to half of patients with
PUVs.4,5,35 The high incidence is probably related to the
high-pressure bladder, but anatomic studies suggest primary
VUR related to ureteral position is frequent.36 The VUR is more
often bilateral, and resolution after valve fulguration occurs in
one third to one half of patients.4,35 Unilateral VUR is some-
times associated with a dilated, dysplastic, poorly functioning
 CHAPTER 122 ABNORMALITIES OF THE URETHRA, PENIS, AND SCROTUM 1557

kidney. It has been suggested that this vesicoureteral reflux-

renal dysplasia (VURD) syndrome may have a protective Urethral Stricture
effect on the bladder and the opposite kidney, much like other ------------------------------------------------------------------------------------------------------------------------------------------------

“popoff” mechanisms such as a large bladder diverticulum or Urethral strictures are more common in boys than girls. They
urinary ascites secondary to a ruptured renal fornix.37–39 are classified as congenital, inflammatory, iatrogenic, or trau-
Traditionally, it has been advocated that removal of this matic.50 Congenital strictures are rare, but urethral hypoplasia
dilated dysplastic kidney and ureter would improve voiding can occur. Inflammatory strictures are uncommon in children
efficiency and decrease potential for UTI.38 More recent evi- and are more likely associated with gonococcal or chlamydial
dence suggests that retaining this dysplastic unit does not urethritis in sexually active adolescents. Inflammatory
affect UTI or function, and these units may be left in place.40 strictures related to chronic indwelling catheters are rare in
These dilated ureters also have the potential for use as a the modern era. Iatrogenic strictures are encountered after
ureterocystoplasty to augment the bladder in those rare cases urethral surgery or instrumentation.
of high-pressure bladder refractory to standard therapy.41 Traumatic strictures of the anterior urethra are seen after
The presence of VUR should not change the initial overall straddle injuries in which the urethra is compressed against
treatment of PUVs.42 Ureteral reimplantation is generally the pubic bone, such as falling on the crossbar of a bicycle.
indicated only in those patients with recurrent UTI despite Posterior urethral strictures are generally seen associated with
appropriate chemoprophylaxis and after appropriate therapy displaced fractures of the pelvis. Blood at the tip of the meatus,
to treat bladder dysfunction. The presence of VUR probably a high-riding bladder or prostate on abdominal and rectal
does not change long-term prognosis unless recurrent UTI examination, or a suspected pelvic fracture should all merit
is an issue.35,43 Over time, with appropriate bladder therapy, a retrograde urethrogram (RUG) before catheter placement
the dilation of the ureters may decrease and bladder wall to avoid converting a partial urethral disruption into a com-
thickness improves, making surgery technically easier with plete one. A partial injury is generally treated with endoscopic
improved results if reimplantation is ultimately necessary. or fluoroscopic catheter placement of an indwelling catheter
until the injury has healed and no extravasation of contrast
agent is seen on the RUG done alongside the indwelling cath-
Anterior Urethral Valves in Boys eter. Controversy exists on initial management of the complete
------------------------------------------------------------------------------------------------------------------------------------------------
disruption. Suprapubic tube placement or vesicostomy and
Anterior urethral valves occur much less frequently than delayed reconstruction approximately 6 months later is gene-
PUVs, but their overall presentation and impact on the urinary rally accepted, but some advocate primary realignment of
tract are quite similar. They can occur anywhere along the the urethra using retrograde urethroscopy and antegrade
anterior urethra, with a slight predominance in the bulbar cystoscopy to place a catheter across the defect.51
urethra.44 The etiology is most likely to be the development The location and length of the stricture determine therapy.
of a ventral urethral diverticulum. With antegrade flow of Most clinicians use antegrade cystography with RUG to define
urine during voiding, the diverticulum undermines the distal these parameters, but others report more accurate assessments
urethra with the common wall becoming an obstructive flap.45 using intraoperative US.52 Short, filmy strictures can generally
Others postulate that this lesion develops from a primary be incised or dilated with reasonable results.53 Some authors
weakness in the spongiosum or an abortive attempt at urethral have been quite successful with progressive urethral dilation
duplication.46 The diagnosis is made radiographically, with a in the setting of urethral hypoplasia.54 Longer strictures or re-
VCUG revealing a dilated proximal anterior urethra, a narrow current strictures of the bulbar urethra are generally treated by
distal anterior urethra, and often a subtle flap of tissue. These excision of the stricture and spatulated reapproximation in an
valves are treated with endoscopic incision and are best visu- end-to-end fashion. This may require an inferior pubectomy
alized with minimal irrigation to prevent flattening of the valve or corporal rerouting to cover a long distance.55 The navicular
with forceful retrograde flow. Hydronephrosis and VUR are and pendulous urethra are less forgiving because excision and
commonly associated and are generally managed in the same reapproximation can result in ventral chordee. Longer stric-
manner as PUVs.47 tures in these locations are generally treated with patch grafts
A similar lesion is the lacuna magna, or valve of Guérin, a or flaps using prepuce, penile shaft skin, or buccal mucosa.56
dorsal urethral diverticulum located in the fossa navicularis. These patients require long-term follow-up, although most
This is a rare entity and may require incision as described strictures that recur do so during the first year.
previously. The lesion can be difficult to diagnose owing to
its proximity to the tip of the penis, but voiding images of
the VCUG that include the glans reveal the distal lesion.48 Urethral Atresia
------------------------------------------------------------------------------------------------------------------------------------------------

Urethral atresia is incompatible with renal development un-

Urethral Diverticulum in Girls less an alternative communication with the bladder such as
------------------------------------------------------------------------------------------------------------------------------------------------
a patent urachus exists. Prenatal intervention with vesicoam-
Urethral diverticula are rare lesions in children. They can niotic shunting may aid in getting the fetus to delivery, but
present as dysuria, hematuria, or symptoms of obstruction. there is often significant RD. There is a strong association with
Their diagnosis can be extremely difficult because of problems prune-belly syndrome.57 The associated perinatal problems
of imaging the short length of the female urethra. Cystoscopy associated with obstructive infravesical uropathy must be
may be required for diagnosis and treatment by excision, dealt with initially. Further reconstruction is individualized
or, if small, incision of the distal lip endoscopically may to the degree of urethral development, but some will require
be warranted.49 continent diversion ultimately.
1558 PART VIII GENITOURINARY DISORDERS

Imperforate hymen may present as hydrocolpos or hydro-

Urethral Stenosis in Girls metrocolpos as a bulging introital mass often with palpable
------------------------------------------------------------------------------------------------------------------------------------------------
distended abdominal uterus or vagina. The urethra is dis-
It was once believed that irritative or obstructive voiding cretely visualized anterior to the mass. Treatment is incision
symptoms or recurring urinary tract infections in girls could and drainage.
be related to urethral stenosis. The diagnosis was supported Various cysts including epithelial inclusion cysts, Skene’s
by the narrowed urethra at the level of the genitourinary duct cysts, müllerian duct cysts, and wolffian or Gartner duct
diaphragm and sphincter on VCUG. It is now realized that cysts (GDC) can occur in this area, which can be incised and
the radiographic and clinical findings represent voiding dys- drained or excised if they recur. GDCs, which line the vaginal
function and its associated inappropriate sphincter activity wall adjacent to the bladder, are worth elaboration because of
during voiding. Although some short-term benefit of incapa- their embryologic origins from the wolffian or mesonephric
citating the external sphincter muscle with dilation may be duct. As a result, ectopic ureters may end in a GDC, and
seen, urethral dilation has been largely abandoned, owing rupture into the vagina results in chronic drip incontinence.
to its long-term ineffectiveness and potential for creating a true Although GDCs are unusual in infants, they are the most com-
urethral stricture after overly vigorous dilation. mon benign cause of vaginal swelling in children. They are
generally asymptomatic, but when large they can protrude
from the vagina and can be associated with urinary retention
or dyspareunia in sexually active adolescents.63 These are
Urethral Mass in Girls usually treated with marsupialization, but those associated
------------------------------------------------------------------------------------------------------------------------------------------------
with ectopic ureters require more proximal reconstruction
Multiple lesions may present as a urethral mass in young girls. involving ureteral reimplant or nephrectomy.
Careful attention to the lesion can often make the diagnosis by Urethral polyps are rare but present as intermittent bleed-
inspection. Lesions of the urethral meatus include urethral ing or obstruction and are treated with excision.64 Anterior
prolapse, prolapsing ureterocele, urethral cyst, or sarcoma. vaginal wall or retention cysts are commonly seen in newborn
Lesions of the vagina may be mistaken for having a urethral females and lie between the posterior urethral wall and the
origin and include Gartner duct cysts, imperforate hymen, anterior vaginal wall. These most commonly resolve spontane-
and sarcoma. ously with a few days after birth.
Urethral prolapse has classically been described in prepu-
bertal African American girls, but Caucasian girls can also be
affected.58,59 The chief complaint is blood spotting in the
underwear and painful urination. On physical examination, Labial Adhesions
the markedly edematous urethra protrudes circumferentially ------------------------------------------------------------------------------------------------------------------------------------------------

at the level of the meatus and is often seen as a friable rosette Labial adhesions, or fusion of the labia minora, are believed to
that is bright red or cyanotic with a central dimple (the ure- occur as a result of chronic inflammation related to vulvo-
thral meatus). A trial of sitz baths, estrogen cream, or a mild vaginitis or chronic dampness resulting from urinary in-
corticosteroid cream is reasonable, but if this is ineffective, the continence. The labia may fuse near completely, causing
redundant tissue is excised and the urethral mucosa is anas- obstructive-type symptoms or incontinence with trapping of
tomosed to the adjacent introital epithelium in the operating urine. More commonly they present as postvoid drip inconti-
room.60 Complications are rare but can include bleeding, nence as small volumes of urine pool above this shelf of tissue
recurrence, or urethral stricture. while seated to void, and they drip into the underwear when
A prolapsing ureterocele can also be quite edematous, but the child stands. This chronic dampness can also cause irrita-
it can usually be discerned because it is not connected to the tive symptoms of itching or dysuria, and an aseptic urine
surrounding urethral meatus. The presentation may include specimen is difficult to obtain owing to the impediment of
blood spotting in the underwear, painful urination, or urinary preparing this adhesed area for culture acquisition.
retention. The diagnosis is confirmed by findings of hydro- Treatment of labial adhesions is not warranted in the ab-
ureteronephrosis associated with the ectopic moiety on sence of urinary tract infection, dysuria, obstruction, or drip
US. If the ureterocele cannot be manually reduced, it can incontinence because these adhesions commonly resolve with
be reduced and unroofed cystoscopically with definitive ex- the physiologic estrogen surge at puberty.65 If the adhesions
cision/reconstruction performed in staged fashion. warrant treatment, a trial of topical estrogen cream (0.01%)
Rhabdomyosarcoma is the most common primary malig- applied twice daily for 2 to 4 weeks may be attempted but
nant tumor involving the uterus, vagina, or bladder in infants can be unsuccessful in up to half of patients.66 Estrogens also
and children and can present in similar fashion. The botryoid carry the risk of vulvar pigmentation, development of breast
type lesion is usually exophytic and commonly emanates from buds, or breast tenderness in prepubertal patients with pro-
the vagina or urethrovaginal septum. It generally presents longed use. The effects reverse with the withdrawal of treat-
before 2 years of age. The urethral margins are at least partially ment. An effective alternative is the use of betamethasone
discrete, which differentiates it from urethral prolapse. It is cream (0.05%) topically twice daily for 1 month.67
classically a grapelike clustered mass that extrudes through If the adhesions are persistent or thick and well developed,
the introitus and presents as bleeding. Radiographic imaging surgical lysis can generally be performed in the office setting
is warranted to fully evaluate the mass and rule out metastases using eutectic mixture of local anesthetics (EMLA) cream.
to the lungs, liver, or bone marrow. After tissue diagnosis, A cotton-tipped applicator or hemostat can then be used to
multimodal therapy including chemotherapy, radiation, and bluntly push the adhesions apart in an anterior to posterior
surgical excision is usually used.61,62 direction. It is imperative that the excoriated labia minora
 CHAPTER 122 ABNORMALITIES OF THE URETHRA, PENIS, AND SCROTUM 1559

be dressed with a lubricating or antibiotic ointment to prevent

the recurrence of adhesions until healed and an ointment be
applied regularly for a period of time after lysis to ensure
re-epithelialization of the lysed edges.

Cowper Gland Anomalies

------------------------------------------------------------------------------------------------------------------------------------------------

Cowper glands are paired urethral glands that sit in the

urogenital diaphragm and drain into the bulbar urethra.
When the ducts draining these glands become obstructed,
cysts termed syringoceles can form.68 They may be seen on a
urethrogram as a filling defect in the bulbar urethra or found
on cystoscopy for another reason in which they appear as a
blue-domed cyst on the floor of the bulbar urethra. Most cysts
are asymptomatic; some may present with terminal hematuria,
blood spotting from the urethra, urinary tract infection, or
obstructive symptoms. These are generally treated with endo-
scopic unroofing, incising the leading lip to prevent obstruc-
tion. Open resection perineally is reserved for large cysts.69

Urethral Polyps
------------------------------------------------------------------------------------------------------------------------------------------------

Urethral polyps are fibromuscular epithelial structures with

transitional epithelium covering the surface.70 They may
occur either in the posterior or anterior urethra.71,72 In either
position, they may cause hematuria, urgency, or obstructive
FIGURE 122-3 Posterior urethral polyp seen on ultrasound of bladder.
symptoms. Bladder ultrasound may show the polyps in the Top view shows transverse image of bladder with polyp arising from
posterior urethra (Fig. 122-3), but VCUG and cystoscopy urethra (cursors). Bottom view shows longitudinal view of bladder with
are diagnostic and excision through the cystoscope is usually polyp seen at entry to posterior urethra (cursors).
curative.73 Polyps can present as congenital obstructing
lesions with all of the characteristics of bladder outlet
obstruction.

Prostatic Utricle
------------------------------------------------------------------------------------------------------------------------------------------------

Prostatic utricles are müllerian duct remnants, which are

common in patients with ambiguous genitalia or proximal
hypospadias. Enlarged utricles can cause urinary tract infec-
tion, dysuria, urgency, hematuria, or epididymitis.74 Diagno-
sis is usually made by VCUG or RUG. However, US may detect
the larger remnants.75 Surgical treatment is reserved for those
that are associated with recurrent symptoms. The surgical
approach may be transabdominal, transvesical, perineal,
posterior sagittal (transrectal or perirectal), or laparoscopic
depending on the experience of the surgeon.74,76–80

Urethral Duplication
------------------------------------------------------------------------------------------------------------------------------------------------

Urethral duplication is rare but more common in males. FIGURE 122-4 Duplicate urethra with a meatus on each side of glans
and counterclockwise penile torsion.
Although embryologic explanations for the defect have been
proposed, the multiple variants of the anomaly suggest that
there is probably not a common etiology for all forms. Dupli- or duplicate phallus (Fig. 122-4). With the sagittal-type dupli-
cation of the urethra can occur along with bladder and genital cations, the ventral urethra is almost always the more normal
duplication. In males the duplications usually occur in the one and passes through the prostate and sphincteric mecha-
same sagittal plane on a single phallus.81 Less often, the ure- nism. It may end anywhere from the tip of the glans to the
thras may be side by side on the glans with either a widened perianal area.
1560 PART VIII GENITOURINARY DISORDERS
The dorsal urethra may open anywhere on the shaft from
an epispadiac location on the glans to the penopubic area.
Dorsal chordee may be present, and a widened symphysis
pubis may be present, suggesting possible association with
the exstrophy complex.82–85 This dorsal urethra may com-
municate with the bladder or ventral urethra (Fig. 122-5)
but can end blindly beneath the symphysis pubis. If it does
communicate, incontinence can be an issue because it often
does not traverse the sphincter. Treatment in the symptomatic
patient involves excision of the dorsal accessory urethra and
correction of the chordee. Urethroplasty may be necessary
to bring the ventral urethra to the glans. In the rare case of
a dominant dorsal urethra and an accessory ventral one, the
excision of the ventral tract cures the incontinence. The treat-
ment of the even more rare side-by-side type requires an
individualized approach depending on the varied anatomy.
Full urinary tract evaluation radiographically is required in
all forms of urethral duplication.
Female duplication may have complete bladder and genital
duplication and may be associated with colonic atresia and/or
cloacal anomalies.86 A dorsal accessory urethra may occur in
FIGURE 122-5 Urethral duplication with dorsal urethral duplication
(arrow) exiting in suprapubic area.
the pubic area and has been reported to communicate with a
urachal remnant.87 Treatment again is individualized to
the particular anatomy involved, with attempts to preserve
bladder tissue and maintain continence.
Megalourethra
------------------------------------------------------------------------------------------------------------------------------------------------
Megalourethra is a rare syndrome of urethral dilation. There
are two types: scaphoid and fusiform. The scaphoid form is
associated with abnormal development of the corpus spongio-
sum, and anterior bulging of the urethra is seen with voiding.
In the more severe fusiform type (Fig. 122-6), the corpora
cavernosa are also involved, which may have an impact on
long-term potency, and the dilation during voiding is circum-
ferential.88 The lesion is generally discovered at birth but has
also been diagnosed on prenatal US.89
The etiology is unclear, but it is thought to be a defect of
mesodermal development, especially given its association
with prune-belly syndrome.90 A renal-bladder US is indicated
to rule out other congenital anomalies. The dilation is nonob-
structive, and urethroplasty with excision of the redundant
tissue and tapering of the urethra is generally undertaken
for cosmetic reasons.
Congenital Urethral Fistula
------------------------------------------------------------------------------------------------------------------------------------------------
Congenital urethral fistula is rare and usually occurs in the
subcoronal area of the penis. Associated hypospadias and chor-
dee occur, suggesting this may be a form of the hypospadias
anomaly. The fistula generally has a well-formed urethra
distally, but it may be thinned with poor glans formation.
Repair involves the techniques used in hypospadias surgery
and may involve simple multilayer closure of the fistula or more
complex reconstruction of the distal urethra and glans.91,92
FIGURE 122-6 Megalourethra, fusiform type.
 CHAPTER 122 ABNORMALITIES OF THE URETHRA, PENIS, AND SCROTUM 1561

of glans injury, and this technique should be limited to those

Phimosis
------------------------------------------------------------------------------------------------------------------------------------------------
adequately trained in its use.

Phimosis is defined as the inability to retract the foreskin. At TECHNIQUE

birth, physiologic phimosis is present as adhesions between
the prepuce and glans preclude retracting the foreskin. As Newborn circumcisions can be carried out at the bedside or in
the child grows, the two layers begin to separate as sloughed the office safely to 2 or 3 months of age. The authors advocate
epithelial debris, or smegma, accumulates between them, the use of local anesthetic, either topically or subcutaneously
defining this plane. This smegma is commonly referred to before the procedure. With the foreskin in an unstretched
as “foreskin pearls” and can be mistaken for infection or position, the palpable coronal margin is marked on the over-
purulence by the uneducated parent. With spontaneous lying shaft skin to indicate the appropriate position of the
erections and natural manipulation, more than 90% of fore- clamp device. A hemostat is used to develop the plane be-
skin becomes retractable by age 3 to 4 years.93 tween the glans and adherent prepuce. In the dorsal midline,
Forceful retraction is not required for this to occur and may the tissue is clamped with a straight hemostat and then incised
initiate the vicious cycle of tearing and scarring, which can with scissors to allow placement of the device. All adhesions
lead to pathologic phimosis. In children older than 4 years should be taken down and smegma débrided. The appropriate
who are unable to retract the foreskin and are symptomatic size of clamp device should be chosen. The conical portion of
with episodes of posthitis or balanoposthitis or ballooning the clamp should cover the majority of the glans. Once this
of the foreskin with voiding, a trial of betamethasone cream portion is in position, the excess foreskin is reduced back over
(0.05%) two times per day for 1 to 2 months allows the the clamp device and the clamp is engaged (Gomco) or the su-
foreskin to retract in up to 90% of boys.94 For those refractory ture is tied (Plastibell) at the previously marked site. Care
to corticosteroid treatment, a temporizing dorsal slit, pre- should be taken to draw the foreskin over the clamp at its lead-
putioplasty (surgical enlargement of the phimotic ring), or ing edge to avoid the foreskin intussuscepting over the clamp
circumcision is indicated. and giving a mismatch of inner prepuce and proximal shaft
skin engaged in the clamp. The amount of inner prepuce
and shaft skin excised should be equal to avoid the risk of
Circumcision recurrent phimosis creating a hidden or buried penis due to
------------------------------------------------------------------------------------------------------------------------------------------------
an inappropriately large amount of residual inner prepuce.
Circumcision remains one of the most controversial topics in Hemostasis is usually excellent with these techniques. Electro-
urology. The American Academy of Pediatrics issued a policy cautery should never be applied to the metal clamp because
guideline in 1975 stating that there is no absolute medical the penis can be completely devascularized (Fig. 122-7).
indication for routine circumcision of the newborn.95 In older children, the procedure is performed under a gen-
In 1999 the Academy offered the opinion that there are some eral anesthetic when the risk of anesthetic is minimized (after
medical benefits of the procedure but not enough to warrant 60 weeks’ gestation). Because the clamp devices become less
routine circumcision.96 reliable in older children and have a less-tailored fit, a free-
Circumcision advocates argue that circumcised boys have hand surgical technique is used at our institution. The prox-
lower urinary tract infection rates, lower incidence of zipper imal and distal incisions are carefully marked before cutting
injury or paraphimosis, and lower rates of sexually transmit- sharply, with the proximal mark overlying the coronal margin
ted disease and penile cancer as adults.96–100 Circumcision when any prominent prepubic fat pad is reduced. The excess
opponents argue the procedure is nonphysiologic and may shaft skin is excised at the level of the dartos pedicle with
be unnecessary or even harmful. Some more extremist groups electrocautery. Hemostasis is attained by electrocautery as in-
argue that circumcised males have decreased penile sensation, dicated, and larger vessels can be ligated. The edges are
less satisfaction with intercourse, and possibly even higher reapproximated with fine chromic suture in an interrupted
divorce rates.101 Unfortunately, this issue tends to be emotion- subcuticular technique to avoid the formation of epithelialized
ally charged and a large body of “supportive research” is highly suture sinuses. A minimal dressing of antibiotic ointment and
subjective. In a cost utility analysis of circumcision, Ganiats gauze is placed with the expectation of it falling off spontane-
and colleagues determined that financial and medical advan- ously or being removed the following day. The authors also
tages and disadvantages of routine neonatal circumcision advocate early bathing on postoperative day 1 and have the
cancel out one another and that personal cultural or religious family apply the ointment to the area at least three times daily
views rather than cost or health outcomes should be the basis for the next few weeks. A caudal block or dorsal penile nerve
of decision making.102 and ring block is usually given at the end of the procedure to
Despite the controversy, circumcision is still one of the help with postoperative pain management.
most common elective procedures in the United States and
is usually done for cosmetic and cultural reasons.
The risk of circumcision is generally low in larger series
(<1%) and includes bleeding, infection, skin separation, Meatal Stenosis
adhesions, too much or too little skin removed, and meatal ------------------------------------------------------------------------------------------------------------------------------------------------

stenosis.103,104 The two major techniques used in neonates Meatal stenosis occurs in approximately 10% of circumcised
are the Plastibell and the Gomco clamp. Similar complication boys.106 It is virtually nonexistent in uncircumcised boys,
rates are noted with both, the Plastibell having more problems with the exception of those with balanitis xerotica obliterans.
with infection and the Gomco clamp more problems with It is thought to be related to subclinical chronic inflammation
separation.105 The Mogan clamp carries the additional risk of the meatus in the circumcised phallus related to exposure to
1562 PART VIII GENITOURINARY DISORDERS

A B
FIGURE 122-7 Neonatal circumcision injury after applying electrocautery to metal clamp. A, Immediate postoperative appearance. B, End result
3 months after surgery.

urine (ammoniacal dermatitis) or to relative ischemia caused psychiatry be done before full informed consent is obtained
by compromising the vasculature to the meatus when the fre- from the family.112 Discussion of maintaining male gender as-
nular artery is taken during circumcision. The classic history is signment with later construction of a neophallus along with
that of a fine-caliber urinary stream with dorsal deflection, of- the cosmetic and functional limitations of this approach
ten greater than 90 degrees. The meatus is pinpoint and tight should be included in the informed consent.112
with 6- or 8-Fr calibration. It may or may not be associated
with pain or blood spotting in the underwear. The disorder
is commonly overdiagnosed, and those with painful urination Diphallia
who lack these physical findings often have voiding dysfunc- ------------------------------------------------------------------------------------------------------------------------------------------------

tion and its associated inappropriate sphincter activity during Penile duplication has a frequency of 1 in 5 million
voiding. births.113,114 The extent of the duplication varies, and the
The treatment of meatal stenosis is a meatotomy in which etiology is suggested to be a failure of mesodermal banding
the ventral parameatal tissue is engaged in a clamp or hemo- or mesoderm encountering two urethral anlages.115 Presenta-
stat (the authors prefer a nontoothed bowel clamp), then tion may range from a small accessory penis to complete du-
incised roughly half the distance to the coronal margin. The plication of urethra, glans, and corporal bodies (Fig. 122-8).
edges are dressed with antibiotic ointment, and the family is There is often size discrepancy, and orientation is most
encouraged to continue this application three times daily commonly side by side. Associated abnormalities include
for at least a week to avoid recurrence, which is uncommon.
This procedure can be performed safely in the office with local
anesthetic or in the operating room setting.107

Penile Agenesis
------------------------------------------------------------------------------------------------------------------------------------------------

Congenital aphallia is extremely rare, with an incidence of 1 in

10 million to 30 million.108,109 It is a result of partial or com-
plete developmental failure of the genital tubercle. The karyo-
type is almost always 46,XY, and the urethra usually opens
into the anal verge or rectum.108,109 Common associated
anomalies can include cryptorchidism, renal anomalies,
VUR, anal anomalies, and cardiac anomalies.108–110 The more
proximal the urethral communication, the higher the inci-
dence of associated anomalies and of neonatal mortality.109
Female gender reassignment with orchiectomy, along with
urinary tract and genital reconstruction, has traditionally
been recommended for this anomaly.108,111 However, recent
evidence that persistence of male gender identity may occur
despite female gender reassignment makes it imperative that
careful consideration by a team dedicated to disorders of
sexual differentiation including urology, endocrinology, and FIGURE 122-8 Diphallia, side-by-side configuration.
 CHAPTER 122 ABNORMALITIES OF THE URETHRA, PENIS, AND SCROTUM 1563

hypospadias, scrotal anomalies, duplicate bladder, renal scrotal ectopia.119 Association with hypospadias and chordee
anomalies, exstrophy defects, and anal and cardiac anomalies. is common. The more significant the scrotal defect, the more
Complete imaging of the urinary tract should be performed, likely other anomalies are to occur including caudal regres-
and US or magnetic resonance imaging may be helpful in sion, VATER syndrome, cryptorchidism, and other urinary
assessment of penile anatomic development.116 Surgical treat- tract abnormalities.119–121
ment must be individualized to the extent of the defect and Repair of significant forms of penoscrotal transposition
may range from simple resection of the accessory penis to require various advancement flap techniques. When associ-
complex reconstruction.113 ated with severe hypospadias, which requires preputial island
flap type repairs, correction of penoscrotal transposition is
usually done at a second stage to avoid the possibility of devas-
Penile Torsion cularizing the preputial flap.122,123 Others propose combined
------------------------------------------------------------------------------------------------------------------------------------------------
repair of the hypospadias and scrotal transposition in one
Torsion of the penis is a rotational defect of the phallus, which stage, but complications tend to be higher.124–126 In cases
is usually counterclockwise in direction and may be associated of scrotal ectopia associated with cryptorchidism, scrotoplasty
with hypospadias, chordee, and hooded dorsal foreskin. The and orchiopexy can usually be accomplished at the same time.
median raphe generally courses obliquely around the shaft
to the left. Rotation of less than 90 degrees often is not symp- The complete reference list is available online at www.
tomatic and will not require correction. Surgical treatment expertconsult.com.
involves degloving the penis and releasing any fibrous, dys-
genic bands all the way to the base of the penis, allowing
the phallus to reorient to the appropriate position.117 In rare
cases, fixation of the base of the corporal bodies to the SUGGESTED READINGS
symphysis pubis or creation of a dartos flap to rotate the Austin PF. Circumcision. Curr Opin Urol 2010;20:318–322.
corpora may be necessary to maintain proper orientation.118 Hodges SJ, Patel B, McLorie G, et al. Posterior urethral valves. Sci World J
2009;13:1119–1126.
Kajbafzadeh A. Congenital urethral anomalies in boys. Part II. J Urol 2005;2:
125–131.
Penoscrotal Transposition Krishnan A, de Souza A, Konijeti R, et al. The anatomy and embryology of
posterior urethral valves. J Urol 2006;175:1214–1220.
and Scrotal Ectopia
------------------------------------------------------------------------------------------------------------------------------------------------
Rattan KN, Kajal P, Pathak M, et al. J Pediatr Surg 2010;45:E13–E16.
Smith DK, Taylor A, Kilmarx PH, et al. Male circumcision in the United States
The scrotum forms by migration of the labioscrotal folds for the prevention of HIV infection and other adverse health outcomes:
Report from a CDC consultation. Public Health Rep 2010;125:72–82.
inferomedial to the genital tubercle. Failure of migration pos- Xu X, Patel DA, Dalton VK, et al. Can routine neonatal circumcision help
sibly related to a gubernacular defect results in scrotal anom- prevent human immunodeficiency virus transmission in the United States?
alies such as penoscrotal transposition, bifid scrotum, or Am J Mens Health 2009;3:79–84.
 depends on the expression of a number of important genes in-
cluding the Wilms’ tumor gene (WT1) and steroidogenesis
factor (SF-1) genes; in their absence, the gonad and adjacent
structures fail to form. For example, mutations in WT1 in mice
result in absence of the gonad and kidney4 and abnormalities
of the urogenital system, heart, and thorax. Mutations in
Wilms’ tumor 1 (WT1) cause Denys-Drash syndrome, Frasier
syndrome, and Wilms’ tumor associated with aniridia, genito-
urinary (GU) malformations, and mental retardation (WAGR)
in human patients.5 Mutations in SF-1 in mice result in absence
of gonads and adrenal glands6 (see Fig. 123-1) and produce
similar phenotypes in humans.7 Other genes such as LIM1,
LHX9, EMX2,8,9 and GATA-4 that are essential for develop-
ment during this early process are shown in Figure 123-1.
Simultaneously, genes important for wolffian and müllerian
duct development are expressed. The wolffian duct requires
the action of PAX2 and possibly PAX8, which are paired genes
important in Drosophila development. Müllerian duct forma-
tion requires the activity of a series of secreted factors whose
expression is directed by the WNT family of proteins.10 Dur-
ing formation of the urogenital ridge, primordial germ cells,
the progenitors of the oocytes and spermatocytes, must mi-
grate from outside the embryo in the epiblast.11 During this
CHAPTER 123 time they undergo proliferation and imprint erasure imposed
by DNA methylation and complex histone modification to re-
set epigenetic memory in germ cells as elucidated during
mouse development,12–14 where, under the influence of
Disorders of Sexual Blimp-1,15 expression of BMP 2, 4, and 8b and Fragilis ex-
pands this alkaline phosphatase and Stella-expressing popula-
tion16,17 at 3 weeks’ gestation in the human.11 These germ
Development cells migrate through the ectodermal layer of the embryo,
the primitive streak, the base of the allantois, the wall of the
hindgut, and then to the urogenital ridge, where they take
Rafael V. Pieretti and Patricia K. Donahoe up residence in the as yet undifferentiated gonad. Fragilis in-
duces Stella, which in turn induces the pluripotency genes
OCT/4, SOX 2, and Nanog. The migratory phenotype, con-
trolled by integrins18 and cKit signaling,19 is characterized
Nomenclature
------------------------------------------------------------------------------------------------------------------------------------------------
by extensive mitosis (170-fold in mice).20 After entering the
mesonephros, female germ cells enter meiosis in an anterior
“Disorders of Sex Development” (DSD) and “Disorders of to posterior wave under the influence of retinoic acid–
sexual differentiation” (DSD)1 are more inclusive terms than mediated STRA8 (Fig. 123-2). Because the testes degrade
“ambiguous genitalia” or “intersex.” They include a broad retinoic acid with the enzyme CYP26b1, STRA8 is unavailable
clinical spectrum of hormonal, metabolic, and chromosomal to initiate meiosis, so male germ cells remain arrested in
abnormalities resulting in abnormal genital development.2,3 Go.21,22 Further ovarian development of the urogenital ridge
requires the presence of germ cells. Oocyte development also
requires residence in the urogenital ridge,23 as ectopic germ
cells either fail to develop or form tumors outside the confines
Developmental Biology of the urogenital ridge.8,24
of Mammalian Sexual Differentiation of the gonads commences as an indifferent
blastema of mesenchymal cells covered by coelomic epithe-
Differentiation lium. The testes are directed by the expression of small tran-
------------------------------------------------------------------------------------------------------------------------------------------------
scription factors such as SRY, a master switch gene present on
Phenotypic males and females develop as a result of a precise the short arm of the Y chromosome, and an SRY-related
cascade of sequential morphologic, genetic, and molecular autosomal gene called SOX 9 from chromosome 17q. The fact
events, many of which have recently been elucidated. The in- that testicular differentiation requires a signal from the Y chro-
termediate mesoderm situated between the ectoderm and the mosome became evident in 1959, when the 45,X Turner phe-
endoderm is the site that gives rise to the urogenital ridge in notype was first recognized.25 Subsequent cytogenetic studies
the early embryo (Fig. 123-1).2 The urogenital ridge, which progressively localized the male differentiation region to the
develops from a thickening of the ventromedial aspect of short arm of the Y chromosome.26 Much later, Page and
the intermediate mesoderm, contains the undifferentiated colleagues27 defined a critical region for testis determination
gonad and the mesonephros, wherein both the reproductive near the pseudoautosomal region of the short arm of the Y
wolffian and müllerian ducts reside. Formation of the ridge chromosome by comparing chromosomal deletions and
1565
1566 PART VIII GENITOURINARY DISORDERS

Müllerian-duct–derived
Müllerian
female reproductive tract
agenesis

Deficiency of WNT-4,
HOXA9, 10, 11, 13
Gonadal and
adrenal
Ovary
Gastrulation agenesis

Primitive streak Urogenital ridge

Adrenal
Deficiency
of SF-1

Emx2, GATA-4, SF-1

Mixed gonadal
Excess dysgenesis
Lim1, Lhx9, WT1
DAX1
Notochord
Deficiency Gonad
Intermediate
mesoderm of WT1 Deficiency of
SRY, SOX9
Kidney
Renal and
gonadal Pure
Pax2 gonadal Testis
agenesis
agenesis

Wolffian-duct–derived
male reproductive tract
FIGURE 123-1 Mutations in a number of genes can lead to a variety of syndromes of dysgenesis involving the müllerian or wolffian ducts, gonads,
kidneys, and adrenal glands as a result of a deficiency or excess of the proteins shown. DAX1, Tgene duplicated in congenital adrenal hypoplasia on
the X chromosome; Emx2, the empty spiracles homeobox gene; GATA-4, the gene encoding a protein that binds to a GATA DNA sequence; HOXA, homeo-
box protein; Lhx9, a LIM homeobox family member; Lim1, a homeobox gene important for limb development; Pax2, a paired box homeotic gene; SF-1, the
gene for steroidogenic factor 1; SRY, the sex-determining region of the Y chromosome; SOX9, SRY homeobox 9; WNT-4, a protein that induces development
of the müllerian mesenchyma; and WT1, Wilms’ tumor suppressor gene 1. (From MacLaughlin DT, Donahoe PK: Sex determination and differentiation.
N Engl J Med 2004;350:367-378.)

GERM CELL MIGRATION AND PROLIFERATION DURING

EMBRYONIC AND FETAL LIFE

Alkaline
Extraembryonic phosphatase
ectoderm positive
BMP4 cells
Epiblast Urogenital
Primitive Base of the Hindgut
cells ridge and Birth
streak allantois
Blimp 1 gonads

G0 arrest
Mitosis δ meiosis
XY germ cells

Meiosis inhibiting facor (CYP26b1) degrades RA ---// STRA8

Meiosis stimulation (δ CYP26b1) RA → STRA8
Mitosis Meiosis I
XX germ cells

FIGURE 123-2 Migration and pro-

liferation of germ cells during Migration and proliferation of germ cells during embryonic and fetal life
embryonic and fetal life.2,133

translocations in a series of XY phenotypic females. Later,
Sinclair and colleagues28 defined a single-copy gene called
the sex-determining region of the Y chromosome (SRY) that
encodes a DNA binding protein, which is expressed in the go-
nadal ridge immediately before testis differentiation.29 When
the SRY gene, which expresses a protein homologous to the
high mobility group (HMG) class of nuclear proteins, was
transfected into XX female mouse embryos, a substantial pro-
portion of these transgenic animals developed testes and
assumed anatomic and functional male phenotypes.30,29

In addition to the Y chromosome, autosomal and X factors
contribute to sex differentiation. For example, mutations on
the long arm of chromosome 17q, found in cases of campto-
melic dysplasia associated with sex reversal,31–33 caused a de-
fect in the SRY-related gene product, SOX 9, which contributes
to sex differentiation. In addition, analysis of 46,XY sex-
reversed females with intact SRY led to the discovery of the
dosage-sensitive sex (DSS) reversal locus on the short arm
of the X chromosome, duplication of which is required for
female differentiation.34 This region, which contains the gene
DAX-1, was then assumed to have a negative influence on tes-
tis differentiation because a double dose of the X chromosome
is associated with dysgenesis of the testis. However, knockout
of DAX-1 resulted in normal female gonadal development,35
which undermines the hypothesis that DAX-1 is responsible
for ovarian differentiation. The only abnormality seen as a re-
sult of null mutation in DAX-1 occurs, counterintuitively, in
the male, which displays abnormalities in spermatogenesis
and spermatogenic cord formation.36 Defects in the distal
end of the short arm of chromosome 9p37 and the distal
end of the long arm of chromosome 10q38 are also associated
with sex reversal.
The genes responsible for ovarian development are rela-
tively obscure; in fact, the conventional wisdom has been that
the ovaries developed passively as a result of the absence of
testicular determining genes. Of the few genes studied in
the mouse, it is clear that WNT-4 is associated with ovarian
development, as homozygous females have masculinized
gonads and absence of the müllerian ducts.39 Furthermore,
WNT-4 activation was recently shown to be regulated by
Respondin-1, which regulates canonical-b-catenin, cytoplas-
mic stabilization of which in the XY gonad causes male-to-
female sex reversal9,40; FOXL2, a forkhead transcription
factor, was identified as another gene that represses the male
development, allowing for normal ovarian development.41–43
It is important to note, moreover, that prenatal ovarian devel-
opment appears to be independent of steroid hormone ac-
tion.44 Taken together, it now seems reasonable to assume
that ovarian differentiation is not merely a default pathway
resulting from improper testis differentiation.
Thus the embryo must have the appropriate chromosomal
endowment (i.e., 46,XX for females and 46,XY for males).
Germ cells must then accurately migrate to the hindgut and
subsequently take up residence in the retroperitoneum, where
they condense in the urogenital ridge to become either a testis
or an ovary. Germ cell migration into the urogenital ridge co-
incides with and probably induces the morphologic formation
of a sex-specific gonad, which in turn produces steroid hor-
mones and proteins. Receptors in local and more distant sites
subsequently respond to the secreted extracellular hormones
and proteins to activate intracellular signaling pathways
and somatic gene responses, which lead to morphologic
and biochemical changes resulting in the appropriate male
or female phenotype.45
Male and female primordial reproductive ducts coexist for
a short period in all mammalian embryos. Müllerian ducts, the
anlage of the uterus, Fallopian tubes, and upper third of the
vagina, develop autonomously in the female in the absence
of the testis. Wolffian ducts, the anlage of the epididymis,
vas deferens, and seminal vesicles, require testosterone to
develop. The müllerian duct first forms by invagination of
the coelomic epithelium, which tubularizes under the
CHAPTER 123 DISORDERS OF SEXUAL DEVELOPMENT 1567
influence of WNT-4 and then elongates, as the epithelial cells
migrate in close approximation to the wolffian duct.46 The
wolffian duct provides no cells, but its presence is essential
for full elongation of the müllerian duct to the urogenital
sinus.47 Wnt9b47 and a functional PI3K/AKT pathway are
essential for this final tubularization of the müllerian duct.48
The fetal testis, after morphogenesis into seminiferous
tubules with Sertoli cells surrounding germ cells and intersti-
tial differentiation to Leydig cells, produces two products
required for further male differentiation, müllerian inhibiting
substance (MIS), which inhibits differentiation of the
müllerian duct, and testosterone, which stimulates wolffian
structures.49 The external genital primordia develop autono-
mously into clitoris, labia minora, and labia majora. Complete
differentiation of the external genitalia to phallus and scrotum
requires reduction of testosterone to dihydrotestosterone by
5a-reductase.50 The interaction of dihydrotestosterone and
an intracytoplasmic androgen receptor results in lengthening
of the phallus into a penis, the urogenital folds fuse to form
the penile urethra, and the labioscrotal swellings fuse in the
midline to form the scrotum. The influence of testosterone
is discussed in more detail later in the chapter. Autonomous
female development can occur in the absence of ovaries.
The existence of a müllerian inhibitor was proposed by
Jost,49 who showed that testicular implants in female rabbit
embryos stimulated the wolffian duct but also caused regres-
sion of müllerian ducts. This regression is characterized mor-
phologically by programmed cell death. Because of the
importance of these events, MIS was purified51,52 and then
used to clone the MIS gene.53,54 The bioactive C-terminal
domain of MIS is homologous to a group of evolutionary
conserved proteins, referred to as the transforming growth
factor-b (TGF-b) family, which is composed of TGF-b, MIS,
activin, inhibin, bone morphogenesis factor, Drosophila
decapentaplagia, Xenopus Vg1, and a number of growth and
differentiation factors (GDFs).54 The MIS ligand binds to a
heterologous receptor composed of at least two serine-
threonine kinase transmembrane units, the type II receptor,55
which phosphorylates or activates the type I receptor,56 which
signals downstream via SMAD1/5/8 to begin the series of
molecular events that results in regression of the müllerian
ducts. This substance has been developed as an antiprolifera-
tive agent for tumors of müllerian duct origin and ovarian,57,58
endometrial,59 cervical,60 breast,61–63 and prostate61 cancers.
Abnormalities of the MIS gene itself can result in the retained
müllerian duct syndrome, in which otherwise normal males,
usually with undescended testes, have persistent müllerian
structures that have not undergone normal regression.64,65
Pathophysiology of Disorders
of Sexual Differentiation
------------------------------------------------------------------------------------------------------------------------------------------------
Three major categories of developmental aberrations are
responsible for the most common forms of DSD in newborns
(Table 123-1). In the first category, genetic females are mascu-
linized by an overabundance of androgenic steroid produc-
tion, causing a genital abnormality that requires highly
specialized medical or surgical management (or both).
In the second category, DSD occur because of deficient
1568 PART VIII GENITOURINARY DISORDERS

TABLE 123-1
Disorders of Sex Development Diagnosis and Treatment
Physical Examination Gender
Disease Diagnostic Features Phenotype Assignment Medical Therapy Surgical Therapy
46,XX DSD Karyotype 46,XX Symmetric gonads F Hydrocortisone or Perioperative stress
(Overandrogenized Electrolytes: K high, Na Clitoral hypertrophy cortisone steroids
Female) low Sinogram: UG sinus acetate þ Florinef Clitoral reduction
Congenital adrenal Androgen high defect Embryo selection from Vaginal exteriorization
hyperplasia 17- Enlarged labioscrotum the blastocyst Labioscrotal reduction
(adrenogenital hydroxyprogesterone Steroid replacement
syndrome) high in utero from 6 wk
MIS 0 gestation
Sequence CYP21
Chromosomal FISH
46,XY DSD Karyotype 46,XY Symmetric, undescended, If M Presurgical Hypospadias repair
(Underandrogenized Testosterone low small testis testosterone Prepenile scrotal repair
Male) Sequence enzyme Severe hypospadias stimulation Orchiopexy
Testosterone genes: No müllerian structures Testosterone at
deficiency 17-KS OH, P450scc, adolescence
3b-HSD, CYP17
FISH
Predominant female If F Estrogen/ Gonadectomy
phenotype if adolescence progesterone at
adolescence
Androgen receptor Karyotype 46,XY Female phenotype F Estrogen/ Infancy: gonadectomy,
deficiency Testosterone high No müllerian structures progesterone at Adolescence: vaginal
Testicular MIS high Normal testes adolescence replacement
feminization Sequence androgen Narrow male pelvic
(complete androgen receptor gene structures
insensitivity) FISH Sparse axillary and pubic
hair
Reifenstein Karyotype 46,XY As for testosterone if If F Estrogen/ Infancy: gonadectomy,
syndrome Testosterone normal deficiency progesterone at clitoral reduction,
(incomplete MIS slightly elevated adolescence labioscrotal reduction
androgen Adolescence: vaginal
insensitivity) replacement
Sequence androgen If M As for testosterone AS for testosterone
receptor gene deficiency (above) deficiency (above)
FISH
CGH
5a-Reductase Karyotype 46,XY Severe hypospadias M Presurgical Hypospadias repair
deficiency Testosterone high Symmetrical, testosterone Prepenile scrotal repair
DHT low undescended, normal stimulation Prostatic and utrical
MIS high testes DHT replacement opening
Sequence 5a- Prepenile scrotum
reductase type 2 gene
FISH

Chromosomal Abnormalities
46,XY (complete Karyotype 46,XY Phenotypic female F Estrogen/ Gonadectomy
gonadal dysgenesis) Sequence candidate Vagina present progesterone at
genes: SRY, SOX 9, No gonads adolescence
DAX-1 Symmetric
Testosterone absent
MIS absent
FISH
CGH
45,X/46,XY Karyotype 45,X/46,XY Asymmetry of gonads: If F Estrogen/ Clitoral recession
(Ovotesticular or 46,XY streak ovary and progesterone at Vaginal exteriorization
DSD; MGD (mixed Testosterone low dysgenetic testis adolescence Labioscrotal reduction
gonadial dysgenesis) MIS low UG sinus Gonadectomy
Sequence DAX-1 Clitoral hypertrophy
FISH
CGH
If M Presurgical
testosterone
stimulation
Testosterone at
adolescence
 CHAPTER 123 DISORDERS OF SEXUAL DEVELOPMENT 1569

TABLE 123-1
Disorders of Sex Development Diagnosis and Treatment—Cont’d
Physical Examination Gender
Disease Diagnostic Features Phenotype Assignment Medical Therapy Surgical Therapy
46XY (90%) Karyotype 46,XX Asymmetric testis, ovary, If F Estrogen/ Clitoral recession
Ovotesticular DSD Testosterone normal or ovotestes progesterone at Vaginal exteriorization
True hermaphrodite or low UG sinus defect adolescence Labioscrotal reduction
MIS normal or low Clitoral hypertrophy Preserve normal ovary or
CGH polar ovarian tissue
If M Presurgical Staged hypospadias repair
testosterone Prepenile scrotal repair
stimulation Removal of müllerian
Testosterone at structures
adolescence Preserve vas and normal
testis or central testicular
tissue
Orchiopexy?
Prosthesis?

CGH, Comparative genomic hybridization; DHT, dihydrotestosterone; F, female; FISH, fluorescent in situ hybridization; M, male; MIS, müllerian inhibiting substance;
PCR, polymerase chain reaction; UG, urogenital.

androgen production or action in genetic males. The third cat- TABLE 123-2
egory of abnormalities results from mutations leading to ab- Steroid Biosynthetic Enzyme Nomenclature
sent, incomplete, or asymmetric gonadal differentiation.45 It Old New-Protein New-Gene
is essential to understand the pathophysiology underlying
these disorders in order to make a timely and accurate diag- Testis
nosis and to plan optimal treatment strategies. 20,22-desmolase P450scc CYP11A
(side chain
cleavage)
46,XX DSD (OVERANDROGENIZATION 17-hydroxylase P450c17 CYP17
OF 46,XX GENETIC FEMALES) 17-lyase P450c17 CYP17
3b-hydroxysteroid dehydrogenase Same Same
The most common cause of overandrogenization or viriliza- (3b-HSD)
tion is a defect in the P450 (heme pigment 450 oxidases, 17-ketosteroid reductase Same Same
which metabolize multiple substrates) adrenal enzymes re- Adrenal
sponsible for the conversion of progesterones to glucocorti- 21-hydroxylase P450c21 CYP21
coids and mineralocorticoids, resulting in the syndrome of 11-hydroxylase P450c11 CYP11B1
congenital adrenal hyperplasia (CAH) or adrenogenital syn- 18-hydroxylase P450c11 CYP11B2
drome, which is now known as 46,XX DSD (overandrogeniza-
tion).66 This syndrome affects both males and females but
causes ambiguous genitalia only in females. Adrenal diffe- to prevent loss of pregnancy. In case of classical CAH, deficient
rentiation occurs in the human fetus at 11 weeks’ gestational glucocorticoids and mineralocorticoids result in negative
age, after differentiation of the gonads and reproductive feedback on the pituitary, which then leads to excessive pro-
tract. Therefore the excess androgen most severely affects duction of adrenocorticotropic hormone. This, in turn, results
the developing external genitalia, the genital tubercle, and in hyperplasia of the fasciculata and granulosa layers of the ad-
the urogenital folds, which develop after 11 weeks. When renal gland and preferential production of androgenic steroids
exposed to excess androgens, the genital tubercle, instead of from the reticulata layer. 17-hydroxyprogesterone accu-
forming a clitoris, is stimulated to develop as a penile struc- mulates proximal to the P450c21 block, resulting in over-
ture. The urogenital folds, rather than developing as labia, production of pregnenolone and progesterone and their
acquire a bifid infused or fused scrotal appearance. Important 17-hydroxylated derivatives, leading to the overproduction
to the differential diagnosis, ovaries remain in their normal of testosterone and ultimately its reduced derivative, dihydro-
intra-abdominal position above the inguinal canal. testosterone (see Fig. 123-2). Elevations in 17-hydroxyproges-
Mutations in the P450c21, or 21-hydroxylase gene, now terone can be detected from a spot serum sample.73
known as CYP21 (Table 123-2), cause 90% of cases of Polymerase chain reaction (PCR), sequencing of known genes,
CAH. The mutated gene, which resides on chromosome 6p or fluorescent in situ hybridization (FISH) can now be used to
in the HLA locus, results from recombination between the make this diagnosis in utero.45
functional CYP21B and the nonfunctional pseudogene The molecular defect in these 46,XX genetic females with
CYP21A. Incorporation of deletions from this pseudogene adrenogenital syndrome resides in the adrenal glands. The
act to disable the functional gene.67–70 The remainder of condition presents in a wide clinical spectrum. The ovaries,
CAH cases are distributed among deficiencies of P450c11, uterus, and Fallopian tubes are normal. The vagina, however,
3b-hydroxysteroid dehydrogenase, CYP17, or steroid acute is foreshortened as a result of having failed to migrate to the
regulatory protein (StAR), depending on the ethnic origin of perineum. It joins the urethra either at the position of the pros-
the patient.45,71,72 Rarely, 46,XX DSD can result from expo- tate, if masculinization is severe, or at a more distal position, if
sure to exogenous androgens such as those given in the past masculinization is less severe. The external genitalia are
1570 PART VIII GENITOURINARY DISORDERS

characterized by variable clitoral enlargement, ranging from
trivial to severe; some patients form an almost normal phallus.
The labia can be masculinized to form either labioscrotal folds
or, in the most severe cases, complete scrotal fusion. Because
the ovaries are normal, the gonads never descend into these
labioscrotal folds or fused scrotum.74 Therefore an infant with
bilateral nonpalpable testes should have a spot test for 17-
hydroxyprogesterone and a karyotype determination to identify
the presence or absence of the Y chromosome. All the enzymatic
defects result in deficient cortisol biosynthesis. Failure of feed-
back results in increased corticotropin production, which con-
tinues to stimulate the adrenal gland and leads to hyperplasia,
elevated production of products proximal to the enzymatic
defect, and preferential overproduction of androgenic steroids.
Concomitant production of melanocyte-stimulating hormone
can also darken the genitalia and breast areolae.
46,XY DISORDERS OF SEX DEVELOPMENT,
UNDERVIRILIZATION OF THE MALE (46,XY)
Insufficient masculinization of a 46,XY genetic male can occur
due to insufficient testosterone production, an androgen
receptor deficiency, or an inability to convert testosterone to
dihydrotestosterone resulting in 46,XY DSD (formerly known
as male pseudohermaphroditism).75 Deficiency of androgen
production (Fig. 123-3) occurs because of genetic enzymatic
defects including steroid acute regulatory protein (StAR), resul-
ting in lipoid adrenal hyperplasia (formerly thought to represent
defects in cholesterol desmolase, P450scc, or CYP11A1).
CYP11A1 (20,22 desmolase), CYP17 (17-hydroxylase or
lyase), 3b-hydroxysteroid dehydrogenase (3b-HSD), and 17b-
hydroxysteroid dehydrogenase (17b-HSD), formerly known as
17-ketosteroid reductase, together and sequentially are respon-
sible for the cascade of metabolism from cholesterol to
testosterone.71 In these patients, stimulation by chorionic
gonadotropin produces little or no testosterone; in contrast,
MIS levels are normal or high for age. The testes may be unde-
scended, small, or both. Under the influence of normal MIS,
there should be no müllerian structures. Because of testosterone
deficiency, the penis may be small and hypospadiac.70
Androgen insensitivity can occur because of receptor defi-
ciency. Most abnormalities are caused by point mutations in
the androgen receptor gene,76,77 which can now be detected
by PCR and DNA sequencing. Only rarely have large deletions
been found. The phenotype can be mild or severe; the com-
plete androgen insensitivity syndrome (CAIS) manifestation
was often referred to as testicular feminization, in which the
patient has an almost complete female phenotype. Serum
levels of testosterone may be high; MIS may be normal or,

Adrenal and Gonadal Biosynthesis of Steroid Hormones

CH2OH
StAR O
Cholesterol
C O
H C
HO

Aldosterone

HO
Deoxycorticosterone O
CYP11A1 (DOC)
(20,22 Desm) CH3 CH3 CH2 OH CH2OH

C O C O CYP21 C O Corticosterone
C O CYP11B2
Pregnenolone Progesterone
(21-OH) CYP11B1 HO (18-OH)
3β (11-OH)

HO O O O

CYP17 CYP17
CH3 CH3 CH2 OH
(17-OH) (17-OH) CH2OH

C O C O 11-Deoxycortisol C O Cortisol O
17-OH Progesterone C
17-OH Pregnenolone
OH OH OH HO OH
3β
CYP21 CYP11B1
(21-OH) (11-OH)
HO O O O

CYP17
CYP17 (17 Lyase)
(17 Lyase) O O
Androstenedione
O H Dihydrotestosterone
3β (DHT) OH
17-KS R Testosterone
ED
17-OH
Dehydro
HO
O genase
Dehydroepiandrosterone
5αRED
O
(DHEA) O

FIGURE 123-3 Pathways of steroid hormone biosynthesis from cholesterol. New enzyme nomenclature is in bold lettering; old nomenclature is in
parentheses. Enzymatic defects of CYP21, CYP11, or 3b-hydroxysteroid dehydrogenase can lead to congenital adrenal hyperplasia. Deficiencies in testos-
terone production leading to a form of male pseudohermaphroditism can result from defects in CYP11A1, CYP17, 3b-hydroxysteroid dehydrogenase, and
17-ketosteroid reductase.

in some cases, considerably elevated; and müllerian structures
are normally regressed. The gonads are symmetric and may be
intra-abdominal or descended. However, not all patients with
the 46,XY karyotype and the characteristics of 46,XY DSD
have a detectable molecular defect in the coding region of
the androgen receptor. It is possible that the noncoding
promoter, the 30 untranslated region, or other transcription
factors or their cofactors are the cause of 46,XY DSD in the
large subset of patients in which the molecular defect has
not yet been defined.
3b-HSD acts by reducing the 3b-hydroxyl group to a 3-
ketone, and by isomerizing the C5-6 double bond to the C4-
5 position in the conversion of pregnenolone to progesterone,
or 17-hydroxypregnenolone to 17-hydroxyprogesterone. All
these steps are essential for the production of glucocorticoids
and mineralocorticoids. The activity is also important in the
testis for the conversion of dehydroepiandrosterone (DHEA)
to androstenedione. Various isoforms of this enzyme have also
been found in the placenta, but their functions have not been as
carefully detailed. Another non-P450 enzyme in the pathway
is 17b-HSD, which converts androstenedione to testosterone.
Mutations in the 5a-reductase type 2 genes cause another
form of 46,XY DSD. Because of this deficiency, testosterone is
not converted to dihydrotestosterone in peripheral, particu-
larly genital, target tissues.50,78 This non-P450 enzyme
depends on the reduced form of NADPH (nicotinamide-
adenine dinucleotide phosphate). Action in the genital area
results in elongation and ventral closing of the penile raphe,
which encloses the urethra and displaces the urethral orifice
from the perineum to the tip of the penis.75 The labioscrotal
folds are also fused in a posterior direction to create scrotal
sacs. This autosomal recessive form of severe hypospadias is
associated with undescended testis, prepenile scrotum, and
enlarged prostatic utricle. The enzyme has binding sites for
both testosterone and the NADPH cofactor, where point
mutations have been associated with the defective phenotype.
Two known isoforms exist; however, type 2, located on chro-
mosome 19, is the one expressed predominantly in the exter-
nal genitalia. The paradoxical virilization that occurs at
puberty and results in a change of sexual identity is caused
by increased activity of the 5a-reductase type 1 isoform in a
multitude of other tissues in these genetic males.
ABNORMALITIES OF GONADAL
DEVELOPMENT AND DIFFERENTIATION
Abnormalities of the sex chromosomes usually manifest as
failed, incomplete, or asymmetric gonadal differentiation.
Patients with these disorders have either bilateral streak go-
nads, as in 46,XY pure gonadal dysgenesis, or asymmetric
gonadal development, as in mixed gonadal dysgenesis
(MGD) or ovotesticular DSD (formerly termed true hermaph-
roditism). Patients with pure gonadal dysgenesis and a 46,XY
karyotype may have a defective Y chromosome, with either
deletion in the 1A1 pseudoautosomal region27 or a point mu-
tation in the SRY gene.79 Mutations associated with camptome-
lic dysplasia, a severe disorder occurring in patients with a
translocation in the distal arm of chromosome 9p near the
SRY-related SOX 9 gene,31,33 or mutations in WT1 associated
with Frasier syndrome37 also result in pure gonadal dysgenesis.
These streak gonads fail to develop bilaterally, producing little
CHAPTER 123 DISORDERS OF SEXUAL DEVELOPMENT 1571
or no end products of testosterone or MIS; gonadotropin levels
are compensatorily high. Müllerian ducts are preserved, and
the phenotype is female.
MGD (also known as asymmetric gonadal dysgenesis) with a
45,X/46,XY karyotype is by far the most common of the chro-
mosomal abnormalities.45,80 The gonads are asymmetric,
most often with a small dysgenetic testis on one side and a
streak gonad on the other.81 Most patients with this defect
have retained müllerian ducts. The small testis can produce
enough testosterone to cause masculinization and hypertro-
phy of the clitoris. The vagina fails to migrate to the perineum
and enters the urethra as a urogenital sinus defect more distal
to the bladder neck than seen in severe cases of CAH. It is im-
portant to note that 40% of patients with MGD can have a
46,XY karyotype, and some have bilateral testes or streak ova-
ries. The absence of the second X chromosome is in some way
related to the early ovarian dysgenesis. The mosaicism of MGD
results from the presence of at least two gonadal (chimeric)
germ cell lines, and the degree of testicular differentiation is
determined by the percentage of cells expressing the XY geno-
type, which may also influence the degree of asymmetry. Loss
of the Y chromosome can occur because of nondisjunction,
the failure of paired chromosomes to migrate to opposite poles
during cell division.2,25,82,83 Because prenatal genetic testing
has become more common, the 45,X/46,XY karyotype is now
being detected in a small percentage of amniotic cells in
phenotypically normal males.
True ovotesticular DSD (true hermaphrodites) are rare,
except among the Bantu in Southern Africa; the cause of this
anomaly remains elusive.84,85 More than 90% of these patients
have a 46,XX karyotype. Asymmetry characterizes many of
these patients, who have simultaneous ovarian and testicular
differentiation without the dysgenesis characteristic of MGD.
The testicular and ovarian tissue can be separated on both sides
or combined in one or both gonads as an ovotestis. When ovar-
ian tissue and testicular tissue coexist in the same gonad, the
testis is always central and the ovarian tissue is polar.86 Al-
though the molecular cause of this disorder has not been elu-
cidated, translocation of a fragment containing the SRY gene to a
cryptic site on the X chromosome has been observed.87 Other-
wise, SRY has not been detected in these patients. Although
early testicular differentiation occurs, spermatogenesis is not
evident,84 which may reflect the absence of other necessary
Y-directed functions. The müllerian structures are regressed
on the side of the testicular tissue but retained on the side of
the ovarian tissue and in the midline as well, with the vagina
entering the distal urethra as a urogenital sinus defect. The
asymmetry associated with this disorder remains an enigma.
Diagnosis
------------------------------------------------------------------------------------------------------------------------------------------------
A baby born with DSD must have an expeditious and thought-
ful evaluation to determine whether gender assignment can be
made. It is important to minimize the emotional trauma to the
family and later to the child. Referral to a well-organized team
of endocrinologists, geneticists, psychologists, and pediatric
surgeons/urologists experienced with the complexity of these
disorders is absolutely necessary. Although many syndromes
can affect later sexual development, only four result in
DSD at birth: 46,XX DSD (formerly known as female pseudo-
hermaphroditism), ovotesticular DSD (true hermaphroditism);
1572 PART VIII GENITOURINARY DISORDERS

46,XY DSD (previously termed male pseudohermaphroditism,
undervirilization of an XY male, and undermasculinization of
an XY male); and MGD 45,X/46,XY. A child with pure gonadal
dysgenesis, although having a 46,XY karyotype, is phenotyp-
ically female. It is critical that the sex of rearing not be biased
by the skills of the surgeons, who must be equally capable at
reconstructing the infant as male or female. Two screening cri-
teria can be used to diagnose the infant as having one of the
four disorders: symmetry and the presence of a Y chromosome
(Table 123-3). The diagnostic evaluation of patients with am-
biguous genitalia is outlined in Table 123-1.
The first criterion is the physical finding of gonadal symme-
try or asymmetry, and the second is a rapid analysis of the sex
chromosomes. Probes for SRY can now be used to detect the
distal end of Yp, the short arm of the Y chromosome, by PCR
or FISH. Direct sequencing of the SRY gene is now available.
Symmetry of the gonads is determined by the position of one
gonad relative to the other, either above or below the external
inguinal ring. Both gonads lie either above or below the exter-
nal inguinal ring when a diffuse biochemical cause underlies
TABLE 123-3
Rapid Diagnostic Algorithm
Y Chromosome Absent or
Abnormal Y Chromosome Present
Symmetry Asymmetry Symmetry Asymmetry
46,XX DSD Ovotesticular 46,XY 46,XX/46,XY MGD
[congenital DSD DSD [Mixed gonadal
adrenal dysgenesis]
hyperplasia]
These two diagnostic criteria—presence or absence of Y chromosome and
gonadal symmetry or asymmetry—allow the rapid, accurate assignment of
a patient into one of the four diagnostic categories with approximately 90%
accuracy.
the abnormality. For example, in 46,XX DSD or 46,XY DSD, a
biochemical defect influences both gonads equally. Asymme-
try occurs in chromosomal abnormalities such as MGD or
ovotesticular DSD (true hermaphroditism), in which a pre-
dominant testis descends and a predominant ovary remains
above the external ring. In cases of 46,XX DSD (female pseu-
dohermaphroditism with adrenogenital syndrome) and in
most cases of ovotesticular DSD (true hermaphroditism),
the karyotype is 46,XX. Patients with 46,XY DSD (male pseu-
dohermaphroditism) or MGD have a Y chromosome in their
karyotype.
After this initial evaluation, rapid analysis for 17-hydroxy-
progesterone to detect CYP21 disorders should be done emer-
gently in patients with 46,XX with symmetric undescended
gonads. This screening test is done in some states’ laboratories
on all newborns. A detailed history and physical examination
can be coupled with PCR, FISH, or direct sequencing to detect
mutations or deletions known to cause defects in the four ma-
jor categories (Table 123-4) and to define these disorders more
fully and guide sex assignment. Subsequently, ultrasonogra-
phy, magnetic resonance imaging, contrast radiography, and
later panendoscopy, coupled with accurate laboratory analysis
for disorders of enzymes affecting testosterone synthesis,
should provide a definitive diagnosis and permit appropriate
gender assignment with a high degree of accuracy.
The size and location of the penis with respect to the scro-
tum should be carefully noted. Although the size of the phal-
lus varies considerably, some guidelines are helpful.70,88 The
average length of the penis is 3.5  0.4 cm at term, 3.0 
0.4 cm at 35 weeks’ gestation, and 2.5  0.4 cm at 30 weeks’
gestation. At term, a diameter of 1 cm is average. A penis smal-
ler than 1.5  0.7 cm in full-term infants may raise the option
of female gender assignment, except in cases of 17-ketosteroid
reductase or 5a-reductase deficiencies, in which large body
habitus or sex reversal at puberty, respectively, may favor or
dictate male sex assignment. An early rectal examination

TABLE 123-4
Mutations in Genes Involved in Sex Determination and Development and Associated with Disorders of Sex Development
Gene (Locus) Protein and Proposed Function Mutant Phenotype
WT1 (11p13) Transcription factor Frasier syndrome; Denys-Drash syndrome with Wilms’
tumor
SF-1 (9q33) Transcription factor, nuclear receptor Gonadal and adrenal dysgenesis
SOX 9 (17q24) High mobility group (HMG) protein transcription factor Camptomelic dysplasia and male gonadal dysgenesis or
XY sex reversal
DAX-1 (Xp21.3) Transcriptional regulator, nuclear receptor Gonadal dysgenesis and congenital adrenal hypoplasia
SRY (Yp11) HMG protein transcription factor Gonadal dysgenesis, XX sex reversal
MIS (AMH) type II receptor Serine threonine kinase receptor Persistent müllerian duct syndrome
(12q12-13)
MIS (AMH) (19p13) Secreted protein, fetal müllerian duct regressor, Leydig Persistent müllerian duct syndrome
cell inhibitor
AR (Xq11-12) Androgen receptor, ligand transcription factor 46,XY DSD, CAS or PAIS
HSD17B3 (9q22) 17-hydroxysteroid dehydrogenase–17-ketosteroid 46,XY DSD
reductase 3
SRD5A2 (5p15) 5a-steroid reductase type 2 46,XY DSD (may virilize at puberty)
CYP17 (10q24q25.) 17-hydroxylase, 20,22-lyase 46,XY DSD
CYP21 (6q21.3) 21-hydroxylase 46,XX DSD (Congenital adrenal hyperplasia)
HSD3B2 (1p13.1) 3b-hydroxysteroid dehydrogenase type 2 46,XX DSD (Congenital adrenal hyperplasia)
CYP11B1 (8q24) 11b-hydroxylase 46,XX DSD (Congenital adrenal hyperplasia)
StAR (8p11.2) Steroidogenic acute regulatory protein Congenital lipoid adrenal hyperplasia

From MacLaughlin DT, Donahoe PK: Sex determination and differentiation. N Engl J Med 2004;350:367-378.

may allow the detection of the uterus while it is still under the
influence of placental human chorionic gonadotropin, after
which the uterus softens. The presence and severity of hypo-
spadias should be noted; the confluence point of the vagina
with the urogenital sinus (UGS) (urethra) in relation to the
bladder neck and perineum and the length of the vagina
should also be noted. Palpation of the gonads can help differ-
entiate the firm testis from the softer ovotestis, and the wrap-
ping of the epididymis around the testis can help differentiate
it from an ovary in the inguinal position.45 Often the hypos-
padiac phallus is displaced posteriorly behind a prepenile
scrotum.
46,XX DSD (OVERANDROGENIZATION
OF GENETIC FEMALES, 46XX)
Female pseudohermaphroditism is a diagnostic problem if the
patient is the proband for the family. If a sibling already has the
disease, the level of awareness for this autosomal disorder will
be high. Amniocentesis with cytogenetic analysis for the ge-
netic defect allows early prenatal diagnosis. However, in un-
suspected cases, when the child is born at term, the clitoris is
hypertrophied, the gonads are symmetric and intra-abdominal,
and the vagina fails to descend to the perineum and enters
the UGS (urethra) distal to the bladder neck; however,
a significant subset of patients may have a verumontanum,
with the urethrovaginal confluence quite close to the bladder
neck.89 These patients can also have an enlarged, placentally
stimulated, palpable prostate at the level of the verumontanum.
The labioscrotal folds are rugated, enlarged, and, in some cases,
completely scrotalized. In addition to normal ovaries, the
uterus and Fallopian tubes are normal. The vagina, however,
is foreshortened, as a result of having failed to migrate to the
perineum. It is essential to measure the distance from the blad-
der neck to the UGS confluence at the time of panendoscopy to
aid future plans for definitive reconstruction. The external gen-
italia are characterized by variable clitoral enlargement, ranging
from trivial to severe. Some patients have a male-appearing
phallus. The labia may be masculinized to form labioscrotal
folds or, in severe cases, complete scrotal fusion may occur. Be-
cause the ovaries are normal, the gonads never descend into the
labioscrotal folds or fused scrotum.45,74 The karyotype is al-
ways 46,XX in females, MIS is undetectable, and a spot serum
analysis reveals elevated 17-hydroxyprogesterone. Androgen
levels are also elevated. Concomitant production of melano-
cyte-stimulating hormone can darken the genitalia and breast
areolae. PCR with appropriate probes demonstrates point mu-
tations in the P450c21 gene.69 Electrolytes are often normal at
birth, but when maternal steroids wane after 5 to 7 days, potas-
sium levels may become markedly elevated and serum sodium
levels may fall. As a result, electrocardiography reveals inverted
Twaves, which, if unattended, can lead to cardiac arrest. Hence
this disorder must be considered a medical emergency in the
new forum.
In rare cases with 3b-HSD deficiency, 17-hydroxypregnenolone,
pregnenolone, and DHEA levels are elevated and serum
11-deoxycortisol or deoxycorticosterone levels are elevated
in patients with P450c11 deficiency. Gender assignment
in these patients is almost always female, except when the
diagnosis is delayed and the patient has already been raised
as a male.90,91
CHAPTER 123 DISORDERS OF SEXUAL DEVELOPMENT 1573
46,XY DSD (MALE
PSEUDOHERMAPHRODITISM)
Testosterone Deficiency
By definition, 46,XY DSD have deficient androgenization of
the external genitalia with a 46,XY karyotype. If the disorder
is caused by an enzymatic defect in the testosterone-androgen
pathway, patients have low basal serum testosterone and stim-
ulation by chorionic gonadotropin produces little or no
increase in testosterone. Because the genes coding for the en-
zymes in the pathway have been cloned, PCR or direct se-
quencing can be performed to detect deficiencies in StAR,
3b-HSD, CYP17, or 17-ketosteroid reductase genes. The testes
may be small and are often bilaterally and symmetrically
undescended. The penis is small, and hypospadias is usually
severe. Under the influence of normal MIS, there are no de-
tectable müllerian structures. In some cases, the phenotype
is completely female.70,92 Gender assignment depends on
the size of the phallus. It may be better to raise those with a
small phallus as females, whereas it is invariably preferable
to raise those with a reasonably sized phallus as males because
they will respond to exogenous testosterone. Special consider-
ation should be given to those rare children with 17-ketosteroid
reductase deficiency, which may be better raised as males
(P. K. Donahoe and R. Pieretti, oral communication).
Androgen Receptor Deficiency
Cases in which the androgen receptor is severely dysfunc-
tional, as occurs in CAIS (complete androgen insensitivity
syndrome), previously known as testicular feminization, have
been attributed to complete androgen receptor insufficiency.93
The phenotype is completely female, so these children should
be raised in concordance with that phenotype. No müllerian
structures are present because high levels of biologically active
MIS are produced. The testes are usually in the inguinal
region, and their firmness and investing epididymis can be
palpated to distinguish them from ovaries. The karyotype is
46,XY. Although the androgen receptor is deficient, levels of
testosterone and MIS are high. PCR or sequencing of the
androgen receptor gene can often pinpoint the molecular
defect and provide the definitive diagnosis.
PAIS (partial androgen insensitivity syndrome) results in
only partially masculinized 46,XY patients, with a wide varia-
tion from minimal to severe anomalies. As with testosterone-
deficiency syndrome, the penis is small and hypospadiac.
The testes may be small and are often undescended but sym-
metric, and müllerian structures are not present. The testoster-
one levels are normal, and MIS levels are elevated. Most
abnormalities are caused by point mutations in the androgen
receptor gene,76,77 which can be detected by PCR and DNA
sequencing. Only rarely have large deletions been found.
Gender assignment, again, depends on the size of the phallus.
5a-Reductase Deficiency
46,XY DSD can also result from a deficiency of 5a-reductase,
which is responsible for the conversion of testosterone to
dihydrotestosterone. The normal action of 5a-reductase on
the genital area results in elongation and ventral closing of
the penile raphe, displacing the urethral orifice from the per-
ineum to the center of the glans penis.75 This process is
abnormal in these patients, resulting in severe penoscrotal
1574 PART VIII GENITOURINARY DISORDERS
hypospadias. Normal and symmetric testes can be either un-
descended or fully descended. The labioscrotal folds are also
closed posteriorly to create partial or bifid scrotal sacs. The
prostatic utricle is often quite enlarged. Again, the karyotype
is 46,XY. Serum levels of testosterone are high, but levels of
dihydrotestosterone are low. The MIS levels are normal.
PCR and full DNA sequencing can be used to genotype and
detect mutations in the 5a-reductase type 2 gene. Given the
dramatic phenotypic conversion at puberty, it may be more
appropriate to raise these children as males. It should be
noted, however, that in this group, gender assignment in-
volves the greatest dilemma in societies committed to two
sexes, but not in societies that more readily accept assignment
to a third sex.50,94,95
CHROMOSOMAL ABNORMALITIES
46,XY Pure Gonadal Dysgenesis
Patients with pure gonadal dysgenesis are born as phenotypic
females. Amniocentesis with a 46,XY karyotype that does not
produce the expected phenotype on fetal ultrasonography
now allows earlier diagnosis than was previously possible.
Dorsal pedal edema and some Turner characteristics may be
the only obvious somatic manifestations of the defect. Müllerian
structures are present, but gonads are not palpable due to failure
of gonadal differentiation. Testosterone and MIS are un-
detectable, indicating dysgenesis of the gonad. Candidate gene
mutations include SRY, SOX 9, WT1, and SF-1.2
Mixed Gonadal Dysgenesis
Patients with MGD are characterized by asymmetry, with a
streak gonad on one side and a dysgenetic testis on the other.
They also have retained müllerian structures.80,81,96 The clito-
ris is usually hypertrophied. The most common karyotype is
45,X/46,XY, but 40% of patients have the 46,XY karyotype.
Because the testes are dysgenetic, testosterone and MIS levels
may be low. There is a propensity for neoplastic transforma-
tion to gonadoblastoma or seminoma (or both),83,97,98 even
as early as the neonatal period. These tumors can cause torsion
and apparent loss of a gonad that, in rare cases, leads to a uni-
lateral dysgenetic testis or streak ovary. Because gonadoblasto-
mas may occur at any stage, gonadectomy and female
reconstruction and rearing are the options usually chosen.
The X-linked DAX-1, which may suppress testicular differen-
tiation, may have a role in MGD, which is the least understood
of the intersex abnormalities at the molecular level.99 When
we identify the genes responsible for the separation of paired
chromosomes during meiosis, the defects responsible for this
enigmatic abnormality may be discovered. We should men-
tion a subgroup of 45,XO/46,XY who are diagnosed prenatally
with normal external genitalia and raised as males—indicating
a spectrum of the disorder.
Ovotesticular DSD (True Hermaphroditism)
The molecular cause of this disorder remains enigmatic.84,85
Patients with this disease also have asymmetry, with a testis
on one side and an ovotestis on the other; however, various
other gonadal combinations can occur. If an ovotestis is pre-
sent, testis is always central and the ovarian tissue is polar.86
Neoplastic transformation is not characteristic of these unique
gonads. The müllerian structures are regressed on the side of
the testicular tissue but retained on the side of the ovarian tis-
sue and as a midline uterus as well, with the vagina entering
the urethra as a urogenital sinus defect. The clitoris is hyper-
trophied. The karyotype is 46,XX in 90% of cases; the levels of
testosterone and MIS are low or sometimes normal. Although
early testicular differentiation occurs, spermatogenesis is not
evident,84 which may reflect the absence of other necessary
Y-directed functions. Abnormalities in SRY have not been
detected in these patients;100 both the molecular cause and
the reason for the asymmetry remain an enigma. The sex of
rearing is dictated by the phenotype, which is directed by
the predominant gonad.
Medical Management
------------------------------------------------------------------------------------------------------------------------------------------------
46,XX DSD (CAH)
Masculinized females with pseudohermaphroditism can be at
profound risk for life-threatening complications including
cardiac arrest in the prenatal period. Because maternal corti-
sone crosses the placenta and is released slowly from fat stores,
the clinical manifestations of adrenogenital crisis may not be-
come apparent until 5 to 7 days after birth. Heightened clinical
awareness and prompt, appropriate metabolic treatment can
prevent serious complications. Glucocorticoid replacement
as oral hydrocortisone 8 to 10 mg/m2 per day in two or three
doses or cortisone acetate 25 mg/m2 injected every 3 days101
will prevent serious metabolic abnormalities associated with
acute adrenal insufficiency.
Fludrocortisone (9a-fluorocortisol) 0.05 to 0.2 mg/day is
started in severely virilized infants and in those less virilized
with a family history of salt wasting as part of 46,XX DSD
(CAH). Recent evidence that even patients with milder viriliz-
ing hermaphroditism have subclinical aldosterone deficiency
has led to more liberal use of fluorocortisone, although some
clinicians follow electrolyte levels and plasma renin activity
before starting such treatment.
An infant in adrenal crisis should be rehydrated with iso-
tonic saline. After the first hour, half-strength saline in 5% to
10% dextrose should be administered. This regimen corrects
plasma sodium and chloride imbalances rapidly, but hyperka-
lemia and acidosis are corrected more slowly. Dramatic eleva-
tion of plasma 17-hydroxyprogesterone can be used to
monitor the effectiveness of treatment. Infants (average
0.25 m2) in adrenal crisis are treated with 25 mg hydrocortisone
sodium succinate (2 mg/kg), whereas older children receive
50 to 100 mg. This preparation has both glucocorticoid and min-
eralocorticoid activity; the mineralocorticoid activity is equiva-
lent to 0.1 mg of fludrocortisone. A similar regimen is used for
infants exposed to the stress of surgery. Parents are taught to
recognize situations that contribute to adrenal insufficiency such
as high fever, exposure to hot environments, or surgery. Breast
milk and prepared formulas are low in sodium, so table salt
should be given as a supplement. Growth charts must be mon-
itored carefully to avoid inadequate or inconsistent treatment or
oversuppression and chronic adrenal insufficiency.
Prenatal diagnosis and treatment are now available.102,103
Dexamethasone treatment, if used to suppress masculiniza-
tion in utero, is initiated by the fifth or sixth week of gestation
before the start of sexual differentiation. Dexamethasone
crosses the placenta to suppress the fetal adrenal gland and

decrease fetal androgen production, thereby minimizing in
utero virilization. The sex of the fetus is determined by chro-
mosome analysis at amniocentesis or chorionic villus sam-
pling, and PCR and DNA sequencing can be used to
analyze the CYP21 gene. Treatment is discontinued in males
and unaffected females. Affected females can be treated to
term. Currently, embryo selection or genetic manipulation
can be done at the early blastocyst stage. Blastocysts with a
normal CYP21 gene can be preselected for subsequent implan-
tation if a previous pregnancy produced a child with the
CYP21 mutation.
46,XY DSD (MALE
PSEUDOHERMAPHRODITISM)
46,XY DSD patients with poor penile development who have
testosterone biosynthetic defects or androgen resistance may
be considered for female gender assignment. However, for
those with 5a-reductase or 17b-hydroxysteroid reductase de-
ficiency, it may not be appropriate to assign a gender at birth.
Patients assigned to the female gender should receive early
surgical correction of the external genitalia and gonadectomy
but do not receive hormone therapy until puberty, at which
time they will require estrogen and progesterone treatment.
Vaginal replacement should be planned for late puberty.
If it is elected to assign these infants to the male gender, 25
mg of testosterone enanthate or cypionate is given once every
3 or 4 weeks for about 3 doses to confirm that the penis re-
sponds to androgens or to improve the size of the penis before
surgery. Repair should be done before 1 year of age if the size
of the phallus permits. This approach takes advantage of an
early period of presumed enhanced sensitivity to androgens.
After one to three courses of monthly testosterone for
3 months, an interval of at least 1 month without testosterone
must be allowed before undertaking hypospadias repair to
reduce postoperative hyperactivity that can accompany
testosterone therapy. If the child requires a staged procedure,
testosterone therapy, if needed, can precede the second stage.
Testosterone replacement is resumed at adolescence. Patients
with 5 a-reductase deficiency should receive dihydrotestoster-
one replacement, if available. Otherwise, replacement can be
achieved by giving higher doses of testosterone to overcome
the enzyme block.
CHROMOSOMAL ABNORMALITIES
In patients with MGD or ovotesticular disease assigned to the
female gender, no steroidal replacement is required in child-
hood. However, if the child has been assigned to the male
gender, presurgical testosterone stimulation of the penis may
be required before hypospadias repair. Treatment does not
recommence until adolescence. If the patient is assigned to
the female gender, replacement of estrogen and progesterone
is started at adolescence. Patients with 46,XYpure gonadal dys-
genesis require neither surgical nor medical therapy as new-
borns, and estrogen and progesterone replacement begins at
adolescence. Renal function must be followed carefully in pa-
tients with the various WT1 isoform defects. Adrenal replace-
ment therapy may be complex in patients with SF-1 mutations
characteristic of the adrenal hypoplasia congenita syndrome.
Vaginal replacement should be planned for late puberty.
CHAPTER 123 DISORDERS OF SEXUAL DEVELOPMENT 1575
Surgical Treatment of Urogenital
Sinus Anomalies and Disorders
of Sexual Differentiation
------------------------------------------------------------------------------------------------------------------------------------------------
Patients with DSD and UGS abnormalities who have anatomic
problems such as clitoral enlargement and labial fusion are a
source of great concern and challenges for parents, patients,
and the medical team involved in their treatment. Although we
have witnessed significant advancements resulting from studies
on clitoral innervation and new nerve-sparing clitoroplasty tech-
niques,15,104 we lack convincing long-term results regarding
sexual function and acceptance of genital appearance in females
with CAH because most of the published studies correspond to
patients reconstructed, which have been revised.107
Preoperative Evaluation
------------------------------------------------------------------------------------------------------------------------------------------------
IMAGING EVALUATION
A retrograde genitogram is performed by occluding the open-
ing of the urogenital sinus with the balloon of an 8-Fr Foley
catheter placed outside the meatus and secured in place with
tape. Lateral and oblique images are then obtained. The study
should be performed by the surgeon and an experienced ra-
diologist. Next, the catheter should be advanced into the blad-
der for a vesicoureterogram (VCUG). The study allows a
preliminary diagnosis of the level of confluence of the urogen-
ital sinus in relation to the bladder neck, thus facilitating
the planning of the surgical procedure (Figs. 123-4, A and
B, and 123-5).
Ultrasonography gives valuable information about the uri-
nary tract, and in most cases the uterus, vagina, and gonads
can be visualized (Fig. 123-6, A and B). In those cases in which
the anatomy is not well demonstrated, magnetic resonance
imaging (MRI) of the pelvis could clearly outline the anatomy
of the pelvic organs (Fig. 123-7).
Laparoscopy
------------------------------------------------------------------------------------------------------------------------------------------------
Laparoscopy provides excellent visualization of the pelvic
structures and may play an important role in those cases that
need a gonadal biopsy or gonadectomy. It can also be helpful
for identification or removal of müllerian structures.108,109
Also, laparoscopy can be used to perform other procedures
such as a laparoscopic-assisted sigmoid vaginoplasty, or the
excision of an enlarged müllerian duct remnant.
Preoperative Preparation
------------------------------------------------------------------------------------------------------------------------------------------------
It is important to prepare the bowel adequately before repair.
This should be done at home because most children are pref-
erentially admitted on the morning of surgery. For low repairs,
magnesium citrate should be given on each of the 2 days be-
fore repair. For high repairs, Golytely, a polyethylene glycol
isotonic solution, is administered by mouth beginning 3 days
before surgery for 2 consecutive days, followed by magnesium
citrate 1 day before surgery. In some cases, ondansetron may
be indicated to prevent nausea, or Golytely can be
1576 PART VIII GENITOURINARY DISORDERS

C
C
P
P B
B

V R V
R

A Low confluence B High confluence

FIGURE 123-4 A, Low-confluence urogenital sinus. B, High-confluence urogenital sinus. B, Bladder; C, Clitoris; P, pubis; R, rectum; V, vagina.

administered through a small nasogastric tube. It is important

to discontinue Golytely at least 24 hours before surgery to
avoid leakage during the procedure. Magnesium citrate,
which shrinks the bowel, is given on the last day to prevent
leakage. Oral administration of neomycin plus erythromycin
can be used to reduce bacterial concentration.
As outlined earlier, stress doses of steroids, based on the
weight of the child, are given at the time of anesthetic

Uterus

Bladder
Cath in vagina

SAG
A

1 2
RT streak gonad

Urethra
Vagina

Low confluence B
FIGURE 123-6 A, Pelvic ultrasound in a patient with mixed gonadal dysgen-
FIGURE 123-5 Retrograde genitogram showing a low-confluence uro- esis demonstrating the bladder, a catheter inside the vagina, and the uterus.
genital sinus in a patient with 46,XX DSD (CAH). B, Ultrasound of the right inguinal region ultrasound showing a streak gonad.
 CHAPTER 123 DISORDERS OF SEXUAL DEVELOPMENT 1577

probing with a 3-Fr ureteral catheter can be used to find a

small vaginal orifice (Fig. 123-8). For surgical planning one
must precisely define the location of the confluence point be-
tween the vagina and urethra in relation with the bladder
neck. To do so, we use the verumontanum/external sphincter
Vagina
as a landmark. In addition, more accurate measurements can
Bladder be obtained with a 3- or 4-Fr ureteral catheter with 1-cm
markings passed alongside the cystoscope, with its tip placed
at the bladder neck and using the catheter markings as a ruler.
Those anomalies with the confluence point at or above the
verumontanum/external sphincter are considered high, and
Rectum those below are considered low (Fig. 123-4, A and B). Patients
with 46,XX DSD, MGD, and ovotesticular DSD have a cervix
at the most proximal part of the vagina. Patients with 46,XY
DSD have either a small prostatic utricle or a deeper, more
generous cavity that has no proximal cervix. This prostatic
utricle is characteristically found in the center of a flattened
verumontanum but has no surrounding prostatic tissue. In
midlevel and high-confluence cases, a Fogarty catheter with
a stopcock valve is passed into the vagina. The balloon is in-
FIGURE 123-7 Magnetic resonance imaging of the pelvis in a patient flated, a small Foley catheter is placed in the bladder, and both
with a posterior cloaca (persistent urogenital sinus and rectovaginal fistula)
showing an enlarged vagina, the bladder, and the rectum.
are tied together.

induction. Steroids are continued during surgery and for 2 to

3 days after surgery at double the usual oral dose, followed by
Reconstruction for Female
a tapering of the dosage. Children on dexamethasone are Gender Assignment
asked to omit their medications, but children on prednisone ------------------------------------------------------------------------------------------------------------------------------------------------

take the usual morning dose on the day of surgery. Medica- All 46,XX DSD newborns should be assigned to the female
tions after surgery are planned with pediatric endocrinology gender, regardless of the extent of masculinization, and un-
consultation, and dosing is dependent on the length and ex- dergo surgical reconstruction consistent with the female gen-
tent of surgery and the expected length of hospitalization. der assignment. Similar repairs can be used for selected
patients who are not severely masculinized because of 46,XY
DSD, MGD, or ovotesticular DSD. The mainstays of a fem-
Surgical Reconstruction inizing genitoplasty are clitoroplasty, labioplasty, and vagino-
------------------------------------------------------------------------------------------------------------------------------------------------ plasty. Surgical procedures should preserve clitoral sensation
PANENDOSCOPY and provide a female appropriate cosmesis, with a well-
lubricated vagina that will allow pleasant and painless inter-
Each reconstructive procedure is preceded by a panendoscopy course. It should be mentioned that recent trends have raised
using a pediatric cystoscope with 0- and 30-degree optics. concerns regarding the benefits of clitoroplasty. Therefore it
High-flow irrigation of the urogenital sinus assists finding a should be undertaken only after extensive discussions with
small vaginal orifice in the back wall of the UGS; in some cases, the family because another option would be deferral until
there may be only a pinpoint orifice, and in some patients the patient is capable of deciding by herself.

Urethra
Verumontanum

External
Proximal
sphincter
vagina
(catheter)

Vaginal
orifice
External
sphincter

A B
FIGURE 123-8 Cystoscopic examination indicates whether the vagina joins the urethra in a high or low confluence. A, The vaginal orifice is clearly high.
B, Catheter entering the vaginal opening at the verumontanum in a patient with severe masculinization from female pseudohermaphroditism.
1578 PART VIII GENITOURINARY DISORDERS

Planning and Timing the for future orgasms, provide an acceptable cosmesis, and avoid
painful erections. Kogan described a subtunical excision of the
Surgical Reconstruction erectile tissue, which has been used extensively and led to
------------------------------------------------------------------------------------------------------------------------------------------------
newer nerve-sparing techniques.113 Baskin and colleagues
The timing and magnitude of surgical reconstruction is the (1999) found that distribution of the sensory nerves of the cli-
subject of significant controversy. Some groups advocate toris is similar to the sensory nerves of the penis. These nerves
delaying sex assignment to an age in which each patient can are found on the top or dorsal aspect of the clitoris and course
make his or her own decisions110; however, as mentioned ear- under the pubis; circumferential branches from the dorsal
lier, most of these studies analyze the outcomes of older sur- neurovascular bundle encircle the clitoral shaft toward the
gical procedures.111,112 We believe that newer techniques ventrum, thus making a ventral approach to the corpora most
result in an improved cosmetic appearance, achieve a reduced likely to avoid nerve injury (Fig. 123-9).104–106 In Baskin’s
complication rate, and are more likely to preserve sensation; technique corporal tissue proximal to the bifurcation is left in-
however, long-term studies are required to evaluate the out- tact. Hypothetically this preserved erectile tissue may play an
comes of these procedures, which include ventral clitoroplasty important role in sexual function. In most cases we try to
and the total and partial urogenital sinus mobilization (TUM avoid reduction of the glans clitoris, aiming to preserve
and PUM). sensation, but in patients with a large clitoris a wedge of glans
We discuss with the parents all available treatment alterna- tissue may be cautiously excised from its ventral aspect
tives and recommend that the different steps of the surgical (Fig. 123-10, A and B).
reconstruction should be incorporated into a single surgical It is our practice to incorporate the three components into a
procedure and be performed at an early age in order to take unified surgical strategy. A 5-0 Prolene holding suture is
advantage of all available tissues, with the objective of achiev- placed in the glans. Two vertical incisions are outlined with
ing the best possible functional and cosmetic results. This ap- a marking pen on each side of the urethral plate, and the me-
proach is in keeping with the recommendations of an atus is circumscribed as one does for hypospadias surgery, tak-
international group charged with drafting a consensus state- ing care of leaving a redundant segment of dorsal inner
ment on the management of children with disorders of sexual foreskin to allow the fashioning of a hooded prepuce to pre-
differentiation.1 Patients with a low-confluence urogenital si- serve an important source of sensation (Fig. 123-11).114 Prior
nus can be operated once their metabolic management is well injection of incisions with 1% lidocaine/1:200,000 epineph-
controlled; in most cases we undertake an elective reconstruc- rine prevents excess blood loss. The vaginoplasty is incorporated
tion at 3 to 6 months of age. Patients with a midlevel or high into the surgical reconstruction, outlining a wide-base, inverted
confluence can be electively repaired at 9 to 12 months of age. “U”-shaped perineal incision based on the anus; the apex of this
is placed at the estimated final location of the vagina, using the
ischial tuberosities as a landmark (see Fig. 123-14). The clitoris is
degloved down to the peno-scrotal junction; the ventral strip of
Clitoroplasty, Vaginoplasty, the urethral plate should be initially kept intact. Following
Labioplasty, Incorporated into degloving of the dorsal skin, the distal portion of the UGS
(urethral plate) is divided horizontally below the glans clitoris.
a Single Procedure
------------------------------------------------------------------------------------------------------------------------------------------------
The urethral plate (UGS) is mobilized off the ventral aspect of

Planning of the surgical reconstruction should incorporate the

three components of a feminizing genitoplasty, in which the
prepuce is used to create labia minora, the clitoris is reduced
with preservation of sensation, and the labioscrotal swellings
are used to fashion female-appearing labia majora and to en-
hance the vaginoplasty.
To optimize surgical exposure and elevate the perineum,
we place our patients in a hyperextended lithotomy position
with the buttocks lying over and slightly beyond several
folded towels. As previously described, all procedures must
begin with a panendoscopy.

Clitoroplasty
------------------------------------------------------------------------------------------------------------------------------------------------

Surgical procedures to correct the enlarged clitoris have un-

dergone significant advances. Surgeons now recognize that
preservation of clitoral sensation is essential for future or-
gasms. Clitoral resection and recession are only of historical
interest and are no longer recommended. In cases with severe
masculinization, the clitoris is too large, resembling a penis; in
FIGURE 123-9 Circumferential nerves branches coming from the dorsal
such cases, we discuss the anatomic characteristics with the neurovascular encircling the clitoral shaft toward the ventrum. Vertical in-
parents and recommend a clitoroplasty. The goal of current cisions outlined on each side of the urethral plate for the excision of the
techniques is to preserve sensation (neurovascular integrity) erectile tissue.
 CHAPTER 123 DISORDERS OF SEXUAL DEVELOPMENT 1579

impairing the neurovascular supply, and its lateral edges are su-
tured to the adjacent periosteum of the pubis. The reduced cli-
toral shaft is covered with surrounding fat tissue, thus fashioning
a normal-looking mons pubis. In cases with a large glans clitoris,
a wedge of glans tissue can be excised from the ventral aspect of
the glans with careful reapproximation of the glans tissue with
interrupted 6-0 polydioxanone sutures (PDS) (see Fig. 123-10, A
and B). The dorsal mucosal collar should cover the glans partially,
giving it a hooded appearance (see Fig. 123-11). The remainder
of the dorsal prepuce is divided in the midline (Byars technique)
and reattached to the mucosal collar as with hypospadias
repair. The preputial skin wings are rotated inferiorly and
incorporated lateral to the urethral plate to create labia minora
(Fig. 123-12).

Labioplasty
------------------------------------------------------------------------------------------------------------------------------------------------

Most girls with CAH have labioscrotal swellings that are more
anterior than normal labia majora. Significant skin rugation
may be present as well. To move this labioscrotal skin poste-
A Clitoroplasty B Glansplasty riorly, Y- shaped incisions are outlined with an extension pos-
terior to the swellings. The scrotal flaps are cautiously defatted
FIGURE 123-10 A, Excision of erectile tissue. An elliptical incision is out-
lined on the ventral aspect of the glans for glansplasty. B, Completed cli- and moved posteriorly, beside the introitus, as bilateral Y-V ad-
toroplasty and glansplasty (body of the glans sutured to the corporal body vancements. The medial aspects of these skin flaps are then
stumps with absorbable sutures). sutured to the lateral edges of the preputial skin flaps mobi-
lized during clitoroplasty (now labia minora). The end result
is an anatomically correct positioning of the labia minora and
the clitoral shaft downward to below the bifurcation of the cor- majora posteriorly, beside the introitus, rather than anterior-
poral bodies. Next, clitoral reduction is carried out; in our expe- laterally (Fig. 123-13).
rience placement of a tourniquet is unnecessary because we have
observed that bleeding from the erectile tissues is not significant,
particularly in infants, although it can be considerable in the
older child. Longitudinal ventral incisions are made; the erectile
tissue is dissected within the bodies (see Fig. 123-10, A and B).
The body of the glans is sutured to the corporal body stumps P
with absorbable sutures (Fig. 123-10, B). This step should, hy-
pothetically, allow for painless engorgement of the proximally
preserved corpora remnants with sexual activity. The redundant
dorsal tunica albuginea and neurovascular bundle are placed in a
subcutaneous pocket above the pubic bone, folded without

S C

Dorsal
hood A

FIGURE 123-11 Outlined incision to fashion a hooded clitoris from the FIGURE 123-12 Split prepuce used to create labia minora; labia majora
dorsal foreskin. Y-V-plasty; and flap vaginoplasty. C, Clitoris; F, flap; P, split prepuce; S, labio-
scrotal fold.
1580 PART VIII GENITOURINARY DISORDERS

by Fortunoff and colleagues of an inverted perineal skin

U-flap that could be advanced into the opened vagina.115
The posterior flap based on the anus is outlined with a mark-
ing pen, and anteriorly it should reach the edge of the sinus.
A thick, long, flap is mobilized and then the posterior wall of
LM the vagina is dissected with care not to enter the rectum.
Next, the posterior wall of the sinus is opened longitudinally
into a normal-caliber vagina to avoid a vaginal stricture. The
apex of the flap is inserted into the apex of the vagina wall
and secured in place beginning with three interrupted,
C
full-thickness sutures of 4-0 Vicryl, which should be tied si-
multaneously to prevent tearing the fragile vaginal wall; the
rest of the sutures are placed in a sequential manner. In low-
F confluence vaginoplasty, there is no need to insert a finger in
the rectum, but a roll of petroleum jelly (Vaseline) gauze can
be inserted in the rectum to avoid rectal injury (Fig. 123-14,
A to C). In a low vaginoplasty there is no need to do a com-
plete mobilization of the UGS. Once it is dissected off the cor-
pora, the dorsal wall is opened longitudinally down to the
A
vaginal opening to create a more normal-looking introitus.
FIGURE 123-13 Bilateral Y-V advancement of the labia majora. The flaps Later during labioplasty, the medial edge of the prepuce (la-
are cautiously defatted and moved posteriorly, beside the introitus to en- bia minora) will be sutured to the outer edge of the opened,
hance the vaginoplasty. The medial aspects of the skin flaps are sutured to
the lateral edges of the preputial skin mobilized during clitoroplasty (now nonmobilized urethral plate/UGS.
labia minora). C, Clitoris; F, flap; LM, labia majora.
VAGINOPLASTY USING UROGENITAL
Vaginoplasty
------------------------------------------------------------------------------------------------------------------------------------------------
MOBILIZATION
Total urogenital sinus mobilization (TUM) was described in
The type of vaginoplasty required depends on the anatomic
1997 by Alberto Peña as a technique to repair the UGS com-
location of the merging point of the vagina into the urethra
ponent of cloacal malformations.116 Currently, this procedure,
as a UGS. There are six types of vaginoplasty: cut-back, flap,
or a modification thereof, is being used by most surgeons for
pull-through, TUM and PUM, and vaginal replacement.
UGS repair.117 Circumferential, partial mobilization of the
UGS allows the midlevel vaginal confluence to be brought
down to the perineum without tension, avoiding the need
LOW-CONFLUENCE FLAP VAGINOPLASTY
for separation of the vagina from the urethra as in the classical
Flap vaginoplasty is only indicated for cases with a very-low- pull-through vaginoplasty. The Fogarty balloon, placed in the
confluence UGS because otherwise it does not bring the vagina during the panendoscopy, allows the identification
merging point of the vagina with the urethra any closer to of the confluence. Urogenital sinus mobilization has the ad-
the perineum. The technique is based on the description vantage of better visualization of the merging point, and it

UGS
incision Vaginal
incision Incised vagina

Perineal flap

Flap
A B C
FIGURE 123-14 Flap vaginoplasty. A, Wide-based inverted perineal U-flap based on the anus. B, Dissected vagina with outlined back wall incision. C,
Inverted U-flap sutured to back wall of the vagina.

obviates the need for vaginal separation, thus reducing the dif-
ficulty of the procedure. Also, in those cases where vaginal sep-
aration is required, it becomes less extensive. In this technique,
because the confluence is brought closer to the perineum, the
mobilization of skin flaps is minimized. The posterior dissec-
tion is similar to that done for a pull-through or flap procedure,
with careful, midline mobilization of the UGS off the rectum.
The anterior dissection in cases requiring a total urogenital si-
nus mobilization is done, staying close to the urogenital sinus
under the pubis, and dividing the ligament from the pubis to
the urogenital sinus, resulting in significant mobilization of
the sinus, and in most cases the confluence can be brought
more easily to the perineum (Fig. 123-15, A).
In response to concerns for possible complications of uri-
nary incontinence, resulting from the circumferential dissec-
tion of the UGS beyond the pubourethral ligament, Rink
and colleagues (2006) proposed the use of a PUM. In this
technique the anterior dissection stops at the pubourethral
ligament, aiming to avoid compromising the innervation
to the bladder outlet (Fig. 123-15, B).118 This procedure is
adequate in most cases except for patients with a high con-
fluence in whom additional mobilization beyond the pub-
ourethral ligament may be necessary. Alternatively, if the
perineum cannot be reached without tension the vagina must
be sharply dissected from the back wall of the bladder before
the anastomosis to the inverted perineal U-flap can be
attempted. In these cases the use of a prone position can as-
sist the dissection of the vagina off the bladder.
In both total and partial urogenital sinus mobilization, as
previously described for low-confluence vaginoplasty, the
distal segment of the vagina can be quite narrow; hence its
C
P B
B
R V
A High confluence
FIGURE 123-15 A, Total urogenital sinus mobilization: mobilization of the urogenital sinus past the pubourethral ligament. B, Bladder; C, clitoris; U, uterus. B,
Partial urogenital sinus mobilization: circumferential mobilization of the urogenital sinus stops at the pubourethral ligament. B, Bladder; C, clitoris, U, uterus.
CHAPTER 123 DISORDERS OF SEXUAL DEVELOPMENT 1581
posterior wall must be incised up to a normal-caliber vagina
to avoid a vaginal stricture. The apex of the inverted perineal
U-flap is inserted into the apex of the vagina wall and se-
cured with interrupted full-thickness sutures of 4-0 Vicryl as
previously described for a low vaginoplasty (see Fig. 123-14,
A to C).
SPLITTING THE UROGENITAL SINUS
As proposed by Rink, the urogenital sinus can be split to en-
hance the feminizing genitoplasty.118 The common urogenital
sinus can be used in the following manners: (1) In very-low-
confluence cases we do not mobilize the urogenital sinus and
incise it longitudinally on its ventral aspect up to the conflu-
ence point; the lateral aspects of the opened sinus are sutured
to medial aspects of the prepuce wings, thus resulting in a
more normal anatomic configuration; (2) ventrally to fashion
a mucosa-lined vestibule, (3) dorsally to create a flap to en-
hance the anterior half of the vaginoplasty, and (4) laterally
to create a flap that will be rotated to extend the vagina, thus
completing the vaginoplasty (Fig. 123-16).
PULL-THROUGH VAGINOPLASTY FOR MID-
AND HIGH-LEVEL VAGINAL CONFLUENCE
Before the advent of the urogenital sinus mobilization proce-
dures, the majority of girls were reconstructed using a pull-
through vaginoplasty as described by Hendren and Crawford.89
All patients undergo a total lower body preparation from
nipples to toes, the legs are wrapped, and the lower body is
C
P
V
R
B Low confluence
1582 PART VIII GENITOURINARY DISORDERS

C P C P C P

B B B Urethra

V F V V
F

R U U R U UGS flap
R

A B C D
FIGURE 123-16 Use of the redundant sinus. A, Split the sinus ventrally to fashion a mucosa-lined vestibule. B, Split the sinus dorsally to create a flap to
enhance the anterior half of the vagina. C and D, Split the sinus laterally and rotate the flap to extend the vagina, thus completing the vaginoplasty.

passed through the aperture in drapes, which allows the patient postoperative days. Hormonal replacement appropriate to
to be rotated either supine or prone during the procedure. the patient’s diagnosis is given.
The initial aspects of the dissection are similar to the description
previously mentioned for lower confluences. The operation
begins with a panendoscopy, during which the confluence is Reconstruction for Male Gender
localized, and a Fogarty catheter, connected to a stopcock valve,
is maneuvered into the vagina, where the balloon is inflated and Assignment
the valve closed. A Foley catheter is passed into the bladder; both ------------------------------------------------------------------------------------------------------------------------------------------------

catheters are secured together with a silk suture. An inverted per- The treatment strategy is similar for all patients assigned to the
ineal U-flap is made, the urogenital sinus is exposed, and the bal- male gender. Most of these patients have a small penis with a
loon of the Fogarty catheter is palpated. Intermittent insertion of penoscrotal, scrotal, or perineal hypospadias; a severe ventral
a finger inside the rectum, to flatten out its anterior wall, curvature; and a partial or complete prepenile scrotum
reassures the surgeon that the rectum has not been injured. (Fig. 123-18). As expected, the dorsal foreskin is redundant
The vagina is mobilized circumferentially and meticulously and the glans is often split, lacking its normal conical config-
separated off the bladder anteriorly.119 The UGS is closed with uration. Preoperative treatment with testosterone is helpful in
interrupted stitches of 6-0 PDS. A labial flap or an anterior island those cases with a small penis.
flap, which for this procedure must be fashioned in the midline, Our preferred approach now is to perform a one-stage hy-
can now easily reach the anterior vagina. In patients with a very pospadias repair in patients with an adequate urethral plate,
high confluence we have found that rotating to the prone posi- using the extended applications of the tubularized incised
tion improves visualization and exposure, allowing the vagina plate urethroplasty described by Snodgrass.121 In this tech-
to be safely mobilized off the urogenital sinus and bladder.120 nique the urethral plate is preserved; aggressive, meticulous
The placement of a small malleable retractor into the vagina degloving of the foreskin is undertaken beyond the hypospa-
combined with slight upward traction assists the surgical diac meatus into the scrotum, thus achieving in many cases
dissection. The opening in the UGS is then closed with inter- significant correction of the ventral curvature. Next, an
rupted stitches of 6-0 PDS. The vagina is then mobilized cir- artificial erection is induced and, if needed, additional ventral
cumferentially and brought to the perineum, where it is curvature correction is accomplished with a dorsal plication,
sutured to the perineal skin flap as described for vaginoplasty placing one or two 5-0 Prolene stitches at the 12 o’clock
(Fig. 123-17). position dorsally, at the point of maximum angulation. In
some cases dissecting and elevating the urethral plate from
the corpora cavernosa down to normal urethra can be done
Postoperative Care of with excellent results. This maneuver can avoid the need to
divide the urethral plate (Fig. 123-19). We perform the
the Female Patient anastomosis in two layers, the first layer with interrupted sub-
------------------------------------------------------------------------------------------------------------------------------------------------
cuticular stitches of 7-0 PDS, followed by a second running
At the end of each procedure a small Foley catheter is left in- layer of the same material. The suture line should be covered
dwelling for 3 to 4 days. The use of epidural anesthesia has with a well-vascularized dartos flap harvested from the dorsal
improved pain management. Immobilization with a mer- prepuce, or with a tunica vaginalis flap.
maid dressing in which the lower extremities are wrapped If the urethral plate cannot be preserved, a multistage re-
with stockinet prevents tension on the suture lines. A soft pair is indicated. In some cases resection of the ventral teth-
plastic foam or cotton is placed between the knees and ering tissue and division of the urethral plate may not fully
ankles, and the anus is left exposed. The legs are loosely correct the ventral curvature; in these patients a dermal graft
wrapped with an elastic bandage without covering the anus. harvested from a non-hair-bearing donor is defatted and
The mermaid dressing is kept in place for 4 to 5 days. A placed in normal saline solution. A transverse incision is
broad-spectrum antibiotic is used during the first 2 made at the site of maximal concavity, and the inserted graft
 CHAPTER 123 DISORDERS OF SEXUAL DEVELOPMENT 1583

Bladder
catheter
Vagina
catheter

Bladder

I
S F
L L
A A
N P
D
Uterus

Rectum

A B

Clitoral
Labia recession
minora

Preserve
N-V bundle
I
S F
L L
A A
N P
D

Stimulate ISLAND
sphincter rectum
Dissect
up to
peritoneal
Circumscribe reflection
urethral
C fistula D

FIGURE 123-17 Surgical reconstruction for female gender assignment. A, Locations of incisions viewed from the surgeon’s perspective, from the per-
ineum with the patient in the lithotomy position. B, Sagittal view showing the downward and inward direction of movement of the anterior island flap
(right-pointing arrows) and the inward position of the posterior flap. The smaller arrow shows the distance that must be traversed by the vagina, which has
been separated from the urethra. C, Dissection between the vagina and the rectum in a posterior direction and the beginning of the anterior separation of
the vagina from the urethra. D, Completely dissected flaps.
Continued

is sutured to the edges of the defect while preserving the dis- Penoscrotal Transposition
tal urethral plate. The second stage, composite repair, is per- ------------------------------------------------------------------------------------------------------------------------------------------------

formed 6 to 9 months later using an anterior, tubularized,
incised plate urethroplasty combined with a posterior
Thiersch-Duplay procedure. Also, the use of a buccal mu-
cosa graft or an onlay island flap gives satisfactory results
at this stage.
Partial or complete penoscrotal transposition is often found in
cases with penoscrotal and perineal hypospadias (see
Fig. 123-18). The least severe forms are known as bifid scro-
tum, pre-penile scrotum, and shawl scrotum. The scrotoplasty
should be delayed until after the hypospadias repair is
1584 PART VIII GENITOURINARY DISORDERS

External
sphincter

Bladder catheter
seen through Recessed
fistula clitoris

Labia Dissect vagina

minora from
Circumscribing
urethral fistula bladder neck

I
S F
V
a g a
in L L
A A
N P
D
Open for
advancement of Posterior
E posterior flap F flap

Island
Island
flap
flap

Labioscrotal
advancement
Island flap
turned inward
to join anterior
edge of vagina
Posterior flap
Posterior flap
sutured to posterior
G incision in vagina H

FIGURE 123-17—CONT’D E, Separation of the vagina from the urethra. F, Anterior plane of dissection behind the urethra and bladder. G, Advancement of
the labioscrotal flaps to cover defects and inward rotation of the anterior island flap and the posterior U-flap to augment the introitus. H, Completed
reconstruction.

FIGURE 123-18 Patient with scrotal hypospadias, severe chordee, and
scrotal transposition
completed because the base of the flaps needed for the hypo-
spadias repair must be divided and displaced during correction
of the prepenile scrotum. Six months or more should elapse
between the urethroplasty and correction of the prepenile
deformity. A square flap that circumscribes the base of the penis
is displaced in a caudal direction, where it serves to lengthen the
scrotum and further correct the bifid scrotal defect. The base of
the penis is then advanced and held in place in a cephalad di-
rection while the abdominal flaps are rotated in a posterior di-
rection to the ventral base of the penis. During the forward,
upward displacement of the penis, this is done to give it a more
erect appearance. Rotation of the abdominal flaps, posterior
and caudally, is extremely important. Otherwise, the base of
the shaft skin will simply pull away from the apical position
on the mons, creating an unacceptable tented appearance.
The rotated abdominal flap must be of sufficient length to create
half a diameter around the base of the penis to meet the flap
from the other side without tension. Midline and lateral closure
of the now-lengthened scrotum completes the repair and cor-
rects the bifid scrotum. It is important that these patients stay
on bed rest with urine diversion until the ventral base of the
penis has effectively healed (Fig. 123-20, A to F).
 CHAPTER 123 DISORDERS OF SEXUAL DEVELOPMENT 1585

Penile Agenesis
------------------------------------------------------------------------------------------------------------------------------------------------

Penile agenesis is a rare condition occurring in 1 in 30 million

births (Fig. 123-23).123,124 It seems to be the result of a devel-
opment failure of the genital tubercle during the fourth week
of embryogenesis; the scrotum appears normal and contains
normal testicles.125 Patients can present with an imperforate
anus and a recto-urethral fistula (Fig. 123-24), a recto-ure-
thral fistula with a normal anus, or with the urethra located
in the perineum inside a skin tag, resembling a foreskin.123
Patients are otherwise normal 46,XY males.
Parents of affected children face a stressful decision, and
for the surgeons penile agenesis is one of the most challeng-
ing conditions. In the past gender reassignment was the
most frequent choice for these patients. Initial surgical treat-
ment of female-assigned patients requires bilateral orchiec-
tomy. Cases with an associated imperforate anus and
recto-urethral fistula require an urgent colostomy and a vesi-
costomy. Those babies assigned to the female sex require the
creation of a neovagina, which is most frequently performed
with a sleeve of sigmoid colon. However, patients have been
unhappy with the assigned female sex and prefer to be
males.126
The diagnostic evaluation of patients with penile agenesis
FIGURE 123-19 Elevating the preserved urethral plate from the corpora includes a renal ultrasound, pelvic MRI, retrograde urethro-
cavernosa down to normal urethra for additional chordee correction. gram, and in cases associated with an imperforate anus a distal
colonogram through the mucous fistula, using hydro-soluble
contrast material combined with an antegrade VCUG via
Müllerian Duct Remnants
------------------------------------------------------------------------------------------------------------------------------------------------
the cutaneous vesicostomy (see Fig. 123-24). Newborns with
penile agenesis must be evaluated by a multidisciplinary
In patients with severe hypospadias, ovotesticular disease, or team composed of surgeons, radiologists, geneticists, psy-
MGD who have been assigned the male gender, the retained chologists, endocrinologists, and pediatricians. Families must
müllerian ducts can become quite enlarged, leading to recur- be given all available information regarding sex assignment,
rent urinary tract infections, epididymo-orchitis, urinary re- surgical procedures, and immediate and long-term results
tention, secondary incontinence due to urine trapping, and so that they can make a decision in the best interest of their
infertility (Fig. 123-21). Also, malignant transformation has child.
been reported. In such cases the müllerian duct remnant In adolescents and adults the most frequently used phallo-
should be removed. plasty procedure is the microvascular transfer radial forearm
Surgical treatment of müllerian duct remnants is chal- flap, which is a complex and rare operation performed only
lenging because of their close proximity to the ejaculatory in highly specialized centers. The recent description by De
ducts, pelvic nerves, rectum, vas deferens, and ureters. A Castro of a phalloplasty technique and complete urethroplasty
number of surgical approaches have been described including using a quadrangle lower abdominal flap can bridge the inter-
transperitoneal, posterior with rectal retraction, anterior and val between childhood and adolescence until a more defini-
posterior saggital transrectal, transtrigonal, perineal, transure- tive procedure can be undertaken. In this procedure a
thral fulguration, laparoscopic, and robotic. We have success- quadrangle of lower abdominal flap is created, 4  5 cm
fully used the transtrigonal technique; however, the for babies, and slightly larger in an older child to fashion
laparoscopic- or robotic-assisted procedures are less invasive the new penis (Fig. 123-25, A to C). De Castro recommends
and minimize possible damage to adjacent anatomic struc- the use of oral or bladder mucosa for the urethroplasty,127
tures. The case shown in Figure 123-21 was successfully trea- although the single-stage buccal mucosa urethroplasty has
ted by a robotic-assisted laparoscopy. had a high complication rate.
One of the authors (Dr. Patricia Donahoe) has preserved We favor staging the surgical reconstruction. As mentioned
the vas deferens as it courses within the outer wall of the di- earlier, cases with an imperforate anus and a rectourethral fis-
lated vagina.122 Its position must be estimated until it exits at tula require an urgent colostomy with a mucous fistula and a
the site of the verumontanum. Lateral incisions are made on vesicostomy, followed a few months later by a posterior saggi-
both sides of the predicted course of the vas deferens. The re- tal rectourethral pull-through, with subsequent closure of
mainders of the müllerian structures are removed including both the vesicostomy and colostomy. Patients born with a nor-
the sometimes markedly dilated vagina. The strip of vaginal mal anus and a rectourethral fistula can be operated through a
wall containing the vas deferens is tubularized with its mucosa transperineal approach. We have used the quadrangle lower
internalized from the urogenital sinus to the point of entry of abdominal flap to fashion the neopenis in three patients,
the vas deferens. Running absorbable suture can be used; one but we prefer a two-stage buccal mucosa urethroplasty to
layer usually suffices (Fig. 123-22, A to C). avoid or minimize complications.
1586 PART VIII GENITOURINARY DISORDERS

Normal Prepenile scrotum

Dorsal incisions Lateral and ventral incisions

B B
B B

A A A
B′
B′ A′ A′ B′
BB B′ B

B′ B′ Dorsal B

A C
B B

Base of
penis

Normal scrotal length B′ B′ Scrotum to be lengthened

A′ A′ and bifid scrotum
A A to be corrected
B Ventral C

Undermine Completed
closure
B B

B B
B B B′
A A B′ B′ B′ B′
B′ B′ B′
A′ A′
A A′ A′ A A′ A′ A
BB A
B′ B′
C
C

D E F
Prepenile skin to
be shifted caudad
FIGURE 123-20 Surgical correction of prepenile and bifid scrotum. A, Correction is reserved for phenotypic patients in whom abundant scrotal tissue lies
in an anterior direction to the dorsal base of the penis in association with a bifid scrotum. B and C, Posterior transposition of the scrotum and anterior
transposition of the penis are accomplished through a series of dorsal, lateral, and ventral incisions and the movement of anterior tissue. D, Generous
abdominal skin flaps are raised with broad bases to retain blood supply. E, The anterior abdominal wall flaps are mobilized sufficiently to advance to
the ventral base of the penis with no tension. This is necessary to prevent unsatisfactory tenting of the anterior or dorsal base of the penis. F, The remaining
incisions are closed using fine absorbable suture. The transposed scrotalized skin, now in a caudal position, lengthens the scrotum and corrects the bifid
appearance.
 CHAPTER 123 DISORDERS OF SEXUAL DEVELOPMENT 1587

dimple or even an introitus are present because these struc-

tures are derived from the urogenital sinus. The diagnosis
can be confirmed by pelvic ultrasound and MRI.

V. remnant
Surgical Treatment of Vaginal
Agenesis
------------------------------------------------------------------------------------------------------------------------------------------------

Vaginal dilatation is a valid alternative in patients with a vag-

Urethra inal dimple or introitus. A program of daily, graduated dilata-
tions over several months, using Hegar or similar sounds, can
result in a good-sized vagina to allow intercourse. Once a sat-
isfactory vaginal size is obtained, regular sexual intercourse
maintains an adequate vaginal cavity.
Different techniques for vaginal construction in patients
with vaginal agenesis have been used including the construc-
tion of a skin neovagina and the creation of an intestinal neo-
vagina using sigmoid, cecum, and small intestine. Our
preferred technique for vaginal replacement is the use of a
10-cm segment of distal sigmoid colon based on the left colic
or superior hemorrhoidal vessels (Fig. 123-26, A and B). Me-
chanical and antibiotic bowel preparation must be done before
the procedure. A lower transverse abdominal incision gives ad-
equate exposure. A supine position with legs spread and the
knees slightly bent, using Allen stirrups, is recommended.
FIGURE 123-21 Retrograde urethrogram showing an enlarged vaginal The sleeve of sigmoid is selected between noncrushing clamps;
remnant in a patient with mixed gonadal dysgenesis reared as a male.
a short distal sigmoid segment can be discarded to increase the
length on the mesenteric vasculature for the neovagina. The sig-
moid sleeve can be rotated 180 degrees to allow placement in the
Vaginal Agenesis perineum. The proximal end is closed with two layers of absorb-
------------------------------------------------------------------------------------------------------------------------------------------------
able suture material. Either a cruciate or H-shaped large perineal
Vaginal agenesis, known as the Mayer-Rokitansky-Kuster- opening is made; the rectovesical space is dissected to create a
Hauser syndrome, is a rare abnormality affecting approximately large tunnel that will allow the easy introduction of two fingers
1 in 5000 to 10,000 females. In this condition there is absence in the perineum (Fig. 123-27). The sigmoid segment is pulled
of the proximal portion of the vagina, resulting from failure of through the perineal channel and anastomosed directly to the per-
the sinovaginal bulbs to develop and form the vaginal plate. ineum with absorbable sutures. Two-point fixation between the
Two different types of Mayer-Rokitansky-Kuster-Hauser proximal end of the neovagina and the presacral fascia prevents
syndrome have been described; in the typical variety (type prolapse of the bowel. We do not elevate the proximal vaginal
A) patients have symmetric uterine remnants and normal dimple into the cul-de-sac to perform the anastomosis because
Fallopian tubes. Cases with the atypical form (type B) present in our experience, it results in a higher frequency of strictures.
with asymmetric uterine buds or abnormally developed Alternatively, a laparoscopic-assisted procedure can be done with
fallopian tubes. The surgeon must be aware of the differen- a satisfactory outcome.
ces between the two types because the majority of associated The creation of a sigmoid neovagina has several advantages
findings in other organs or systems occur in type B such as natural lubrication without excessive mucous produc-
patients.128,129 tion; long-term use of a stent is avoided, stenosis is infrequent,
Skeletal anomalies are present in up to 12% of females af- and most patients have normal sexual intercourse.
fected with this condition, most commonly in type B cases in-
cluding patients with congenital fusion of the cervical vertebra
known as the Klippel-Feil syndrome.130 Duncan proposed Summary
the term MURCS association to describe the combination of ------------------------------------------------------------------------------------------------------------------------------------------------

müllerian duct aplasia, renal aplasia, and cervicothoracic The medical and surgical management of DSD should be un-
somite dysplasia.131 dertaken by a multidisciplinary specialized team. The treat-
Renal anomalies present most frequently in patients with ment plan must be thoroughly discussed with parents, with
the type B variety129; these anomalies consist of either unilat- the goal of giving the child, thereafter, the most satisfactory
eral renal agenesis, horseshoe kidney, or ectopia of one or both quality of life possible. However, long-term studies are neces-
kidneys, which occurs in 74% of affected patients.132 sary to assess the functional and sensory outcomes of newer
This condition can be discovered at birth when the exam- surgical techniques.
ination does not reveal a vaginal opening, but most cases pre- We recommend that surgeons taking care of these complex
sent with primary amenorrhea. A few patients complain of cases be familiar with all the available techniques so that sat-
dyspareunia or failed intercourse. Physical examination re- isfactory repairs can be performed either to support the male
veals an absent vagina, but the hymen and a distal vaginal of female gender.
 Ovary
Uterus

Vagina Bladder

Penoscrotal
hypospadias

Vas Vas in outer

uterine wall R. testis Future
A prosthesis

To resect remainder
of vagina
Completed
To tubularize vaginal strip neoseminal
as neoseminal vesicle vesicle

FIGURE 123-22 Surgical manage-

ment of retained müllerian ducts.
A, In males with retained müllerian
structures, every attempt is made
to preserve the vas deferens, which
often enters the side wall of the en-
larged vagina and travels between
the mucosa and muscularis or in
the muscularis, as shown. B, A strip
of vaginal wall containing the vas is
preserved and tubularized, and the Vaginal
remaining vagina and uterus are re- mucosa
moved. C, The remaining strip of va-
gina is closed from the entrance of B C
the vas into the wall of the vagina Vas in
to the narrowed neck, where the wall
vas deferens joins the urethra.

FIGURE 123-23 Patient with penile agenesis, imperforate anus, and a FIGURE 123-24 Antegrade cystography and colonogram through the
recto-urethral fistula. mucous fistula demonstrating a high imperforate anus with a rectourethral
fistula in a patient with penile agenesis.
 CHAPTER 123 DISORDERS OF SEXUAL DEVELOPMENT 1589

8–10 cm
Rotate
180°

B 4–5 cm removed
FIGURE 123-26 A, Fashioning of an 8- to 10-cm sigmoid sleeve. B, A
short distal sigmoid segment can be discarded to provide greater length
on the mesenteric vascular pedicle so that the neovagina can reach the
perineum under no tension.

C
FIGURE 123-25 A, Incisions for penile reconstruction using a lower ab-
domen rectangle flap. B, Penile construction in progress. C, 7 days postop-
erative picture of neopenis.

FIGURE 123-27 Cruciate type of incision outlined on the perineum for

the anastomosis with the sigmoid sleeve.
1590 PART VIII GENITOURINARY DISORDERS

Acknowlegments
The authors want to acknowledge the outstanding help of Dr. Anuradha
Shenoy-Bhangle, Fellow in Pediatric Radiology at the Massachusetts General
Hospital, with the preparation of the diagnostic images included in the
chapter. Also, we want to acknowledge the outstanding help of Dr. Rafael
Pieretti-Vanmarcke, Fellow in General Surgery, with the preparation of the fig-
ures included in the chapter.
The complete reference list is available online at www.
expertconsult.com.

CHAPTER 124
Abnormalities of
the Female Genital
Tract
Marc R. Laufer
Abnormalities of the reproductive tract can present and be
identified at birth through adulthood. Most genital anomalies
in female infants present at birth and are congenital. In some
instances, the abnormalities are discovered during evaluation
of coexisting urologic problems or as the result of an associa-
tion with a specific pattern of anomalies. Advances in perina-
tology, obstetrics, neonatology, and pediatric radiology have
improved the recognition and subsequent care of female in-
fants, children, and adolescents with genital tract anomalies.
Many of these conditions are now being diagnosed during fe-
tal life by prenatal maternal ultrasonography. A large number
of these patients may also present at puberty with primary
amenorrhea, pelvic/abdominal pain, and difficulty using a
tampon or having sex.
In addition, females present with reproductive tract ac-
quired conditions such as those related to infection, foreign
bodies, trauma, and neoplasia. A thorough appreciation of
normal anatomy and an understanding of the embryologic
development of the female urogenital tract are necessary to
understand the multiplicity of congenital anomalies and
acquired conditions that may occur.
Embryology
------------------------------------------------------------------------------------------------------------------------------------------------
The female reproductive system is derived primarily from the
müllerian or paramesonephric ducts. The ureteral bud arises
from the mesonephric (wolffian) duct during the third week
of gestation, interacting with the metanephric blastema to ulti-
mately form the kidney and ureter. In the absence of a fetal testis,
there is no production of androgen or müllerian-inhibiting
substance (MIS), and consequently the wolffian duct distal to
the ureteral bud is absorbed into the urogenital sinus. During
the sixth week the paired paramesonephric (müllerian) ducts ap-
pear alongside the wolffian ducts with which they are intimately
related. Between the sixth and eighth weeks of gestation, the
paired müllerian ducts develop lateral to the wolffian ducts
and then cross over the mesonephric ducts medially to fuse in
the midline. Fusion occurs in a caudal to cranial direction, form-
ing a single lumen. These fused müllerian ducts then join the
urogenital sinus at the müllerian tubercle. By the 10th week of
gestation, the ducts form a single midline tubular structure
called the uterovaginal canal. The lateral aspect of the müllerian
ducts, which remain separate (unfused), ultimately become
the fallopian tubes. The caudal fused portion thickens and forms
the uterus (fundus and body) and cervix.
The exact embryologic origin of the vagina is somewhat un-
certain. Koff1 suggested that the upper two thirds to four fifths
of the vagina are probably formed from the fused müllerian
duct, whereas the lower one third to one fifth is formed from
the urogenital sinus. Although evidence supports the dual or-
igin of the vagina, the exact contribution from each source is
still unknown. By the ninth week of gestation, the fused
müllerian ducts push the urogenital sinus caudad, creating
the Müller tubercle. Vaginal development begins at the müller-
ian tubercle, where the uterovaginal canal joins the urogenital
sinus. The tubercle thickens and elongates during the third to
fifth months of gestation, forming bilateral endodermal
invaginations called the sinovaginal bulbs. The bulbs grow,
the müllerian tubercle regresses, and the distance between
the uterovaginal lumen and the urogenital sinus increases.
The sinovaginal bulb completes its growth by the 15th to
26th weeks of gestation, giving rise to a solid cord of tissue
called the primitive vaginal plate. Canalization of this cord of
cells begins at the urogenital sinus and progresses in a
cephalad direction to form the distal third of the vagina. By
the fifth month of gestation, continuity between the proximal
and distal vaginal lumens is complete. Defects can occur
during development, fusion, and canalization.
Failure of development may result in agenesis, failure of fusion
leads to a variety of anomalies including duplication, and failure
of the canalization phase results in formation of vaginal septa.2
The close association and interaction between the müllerian
and wolffian duct structures during embryogenesis may lead to
associated anomalies between the genital and urinary systems.
Differentiation of the female external genitalia occurs
between 12 and 16 weeks of gestation. In the absence of fetal
androgen (particularly, dihydrotestosterone, and androgen
receptor), the genital anlage develops passively into the exter-
nal genitalia of the female. The genital tubercle elongates
slightly and becomes the clitoris. The urethral folds form
the labia minora and do not fuse; the unfused genital swellings
form the labia majora. The urogenital groove remains open
and forms the vestibule of the vagina.
1591
1592 PART VIII GENITOURINARY DISORDERS

Vaginal Agenesis Most patients have distal fallopian tubes12 and ovaries.
------------------------------------------------------------------------------------------------------------------------------------------------
However, unilateral13 and bilateral gonadal agenesis14 have
Complete absence or agenesis of the vagina occurs with differ- occasionally been observed in MRKH syndrome.
ing clinical situations. There can be pure müllerian agenesis The association between congenital absence of the vagina
with normal chromosomal and gonadal development with and anomalies of the urinary tract is common, with approxi-
normal female hormonal production and variable uterine mately one third of these patients having renal abnormali-
structures. There can also be androgen insensitivity with a ties.15,16 The most common abnormality is agenesis of one
defect in the androgen receptor in 46,XY individuals who kidney or ectopia of one or both kidneys. Fusion anomalies
develop female external genitalia, and because they make such as horseshoe kidney and crossed renal ectopia are also
müllerian inhibiting substance, they have no uterus, cervix, commonly observed.17
or upper two thirds of the vagina. Griffin and colleagues16 reported skeletal anomalies in
Müllerian agenesis (also known as Mayer-Rokitansky- approximately 12% of MRKH patients. Two thirds of the pa-
Küster-Hauser [MRKH] syndrome) probably results from a fail- tients with skeletal anomalies have spine, limb, or rib anom-
ure of the müllerian ducts to reach the urogenital sinus due to alies. Six percent of the skeletal malformations were of the
a disorder involving the ureterovaginal canal or the vaginal Klippel-Feil type, which reflects an aberration in cervical tho-
plate. The condition appears to be a sporadic polygenic mul- racic somite development.18 In 1979 Duncan and colleagues19
tifactorial disorder occurring between the 4th and 12th weeks reported on the frequency of these associated events and sug-
of gestation.3 No specific genetic abnormality has been iden- gested that this combination of malformations be called the
tified. Mayer4 first reported vaginal agenesis in stillborn MURCS association. This is an acronym for müllerian duct
infants with multiple birth defects in 1829. Rokitansky5 aplasia, renal aplasia, and cervicothoracic somite association.
(in 1838) and Küster6,7 (in 1910) further described this clin- MRKH syndrome may also occur in association with the TAR
ical entity and recognized the presence of a rudimentary syndrome (thrombocytopenia-absent radius)20,21 and rarely
uterus with normal ovaries and normal external genitalia with anorectal malformations.22 Radiographic abnormalities
(Fig. 124-1). Subsequently, in 1961 Hauser and colleagues8 of the hand and hearing loss also have been observed.23,24
described the frequent association of renal and skeletal anom- The McKusick-Kaufman syndrome (MKS) was described in
alies. Thus the combination of names as MRKH syndrome is 1964 in an Amish population and includes hydrometrocolpos
used to describe this disorder. The incidence of vaginal agen- due to vaginal atresia or aplasia, postaxial poly-dactyly, and
esis varies from 1 in 4000 to 1 in 5000 live female births.9–11 congenital heart disease.25–27 This is inherited in an autoso-
By definition, patients are 46,XX females with normal second- mal recessive pattern, and a mutation in the MKKS gene has
ary sex characteristics that commonly present with amenor- been identified in these cases.28,29 Patients with MKS may
rhea.11 Uterine development in these patients may vary have an overlap with the Bardet-Biedl syndrome (BBS).27
from normal to the more characteristic rudimentary uteri with Children with BBS have genital abnormalities and polydactyly
or without a lumen/endometrial cavity. Because of the lack of a in addition to retinitis pigmentosa, learning disabilities, cen-
vagina, they do not have menstrual bleeding but may present tral obesity, and cardiac abnormalities. Mutations in the MKKS
with cyclic cramping abdominal pain because of rudimentary gene occur in 10% to 15% of BBS patients. BBS can also be
functioning endometrium and retrograde menses. In addition, caused by the BBS6 gene on chromosome 20p12, demonstrat-
there can be a normal midline uterus with cervical and vaginal ing heterogeneity.
agenesis. The vagina is completely absent in 75% of cases, Although some specific syndromes with known genetic
whereas 25% have a short vaginal pouch (often < 2.5 cm). inheritance have MRKH, the genetics/inheritance is not under-
stood. Monochorionic monoamniotic twins have been
reported to be discordant for MRKH.30 In addition, assistant
reproductive technologies have been successful in offering
fertility options to MRKH patients with the assistance of a
gestational carrier and the offspring have not been found to
have MRKH.31

DIAGNOSIS
Diagnosis of vaginal agenesis is most frequently made at the
time of an evaluation of primary amenorrhea. Approximately
15% of girls presenting with primary amenorrhea will have
MRKH syndrome. The diagnosis should be suspected when
commonly associated defects (e.g., skeletal, renal, inguinal
hernias) are identified in a young woman with amenorrhea.
A diagnosis of vaginal agenesis is emotional for the patient
and the family. It is important to have educational information
available.32 Support for patients and their family can be pro-
FIGURE 124-1 External genitalia of vaginal agenesis. The normal ap- vided from www.mrkh.org and www.youngwomenshealth.
pearance indicates how readily this diagnosis can be overlooked. (From org. We have also found it greatly beneficial to have Social
Laufer MR. Structural abnormalities of the female reproductive tract. In
Emans SJ, Laufer MR [eds]: Emans, Laufer, Goldstein’s Pediatric and Adoles- Work Service support for patients and families.
cent Gynecology, 6th ed. Philadelphia, Lippincott Williams & Wilkins, 2011.) It is important to differentiate MRKH and androgen insen-
sitivity syndrome (AIS), both of which present with primary
 CHAPTER 124
amenorrhea and absence of a functional vagina. With AIS
there is absent or scant pubic hair, and in the postpubertal
time period a normal “male range” serum testosterone is
present. In the prepubertal time period, before elevation of
testosterone, a chromosomal analysis is necessary to confirm
the diagnosis. The presence of a Y chromosome increases the
risk of dysgerminoma in the gonadal tissue, and thus removal
of gonadal tissue is indicated. Women with AIS have a lack of
müllerian structures (uterus, cervix, and upper vagina), due to
the presence of müllerian inhibiting substance (MIS). Women
with AIS tend to have the presence of the lower vagina, with
1 to 3 cm of a lower vagina. In contrast, women with MRKH
have absence of müllerian structures with normal functioning
ovaries and hormones. In addition, as noted earlier, these
women have the ability to use their oocytes for assisted repro-
ductive technologies (ART) and pregnancy with the use of a
gestational carrier.
In recent years, instances of hydrocolpos have been noted
on prenatal ultrasound studies. Prenatal magnetic resonance
imaging (MRI) may also document such cases. Occasionally,
the disorder can be diagnosed in the neonatal period in infants
with an associated abdominal mass. In infants and older
patients, a complete physical examination including vagino-
scopy and cystoscopy, pelvic ultrasonography, computed
tomography (CT), MRI,33 and laparoscopy are the best means
of making an accurate diagnosis and identifying associated
urinary tract abnormalities. Ultrasound, MRI, and endoscopy
avoid unnecessary exposure to radiation.
It is extremely important to adequately define the anatomy.
From a gynecologic standpoint, this is best accomplished in
the postpubertal period. For example, it is possible that a
uterus, cervix, and upper vagina are present (agenesis of the
lower vagina), and this is addressed differently than the
presence of a midline uterus with cervical and vaginal
agenesis. In the prepubertal time period, when the uterus,
cervix, and upper vagina are not estrogenized and fully
developed, the small structures may be unappreciated and
thus an incorrect surgical procedure/management may be
undertaken. Thus if there is no surgically pressing issue for
a prepubertal intervention, then evaluation and treatment is
best delayed until after puberty.
TREATMENT
Treatment of vaginal agenesis depends on the anatomy of the
individual patient. Functional, reproductive, and psychologic
issues must be carefully evaluated and addressed, taking into
account both the physical and intellectual maturity of the pa-
tient.34–36 The goal of therapy is to provide adequate sexual
function and deal with the psychologic impact that the patient
has no uterus or vagina. Fertility is possible and is an option
that should be offered to these patients because both ovaries
are usually normal and successful in vitro fertilization with
surrogate pregnancy has been achieved.31,37 Psychologic
studies have shown that infertility rather than expected
difficulty with intercourse was the most problematic issue
for MRKH syndrome patients to deal with.38,39 Proper
counseling of the patient and family members is essential.
Patients and families need to be educated in all options for
creation of a functional vagina, not just the technique pre-
ferred by or most familiar to the treating clinician.
ABNORMALITIES OF THE FEMALE GENITAL TRACT 1593
The timing of creation of a functional vagina is important.
In cases of MRKH and AIS it is best to wait for intervention
until the patient is postpubertal and the young woman is
interested in creating a functional vagina. It can be challenging
to definitively identify the anatomy in an unestrogenized, pre-
pubertal state. The surgeon may believe that the prepubescent
patient has MRKH and create a vagina surgically only to
find out later that there is a uterus, cervix, and upper vagina
present and that the patient could have had a pull-through
vaginoplasty with her native vagina and never needed a
bowel vaginoplasty (see “Agenesis of the Lower Vagina” later).
It is of utmost importance that the affected young woman
decides which treatment plan to pursue and when to move
ahead with it.
If uterine tissue is present, it is important to determine the
size and location of the tissue and the presence or absence of a
cervix. This can best be accomplished with MRI or three-
dimensional ultrasound. If a cervix is absent, with the pres-
ence of a midline uterus, then we do not recommend attempts
to connect the obstructed uterus to a created vagina due to
the risk of ascending infection, sepsis, and death.40 A patient
with a midline uterus without a cervix should be maintained
on hormonal therapy for menstrual suppression to decrease
pelvic pain and endometriosis and maintain the options
for the use of GIFT or ZIFT for conception, pregnancy, and
a cesarean delivery.
In the majority of cases of müllerian agenesis, there is no
uterus or cervix present and creation of a functional vagina
should be undertaken when the young woman desires crea-
tion of a functional vagina.
The goal for creation of a functional vagina is pleasurable
sexual intercourse with acceptable cosmesis of the external
genitalia with minimal short- and long-term morbidity.
Over the years a variety of techniques have been used to
develop a functional vagina. A native vagina is lined with strat-
ified nonkeratinized squamous epithelium, is inclined poster-
osuperiorly, and has an average length of 9 cm on the posterior
wall and 7.5 cm on the anterior wall. Ideally, the neovagina
should be located appropriately (posterosuperiorly), be of
adequate dimensions, be lined by elastile tissue (either mu-
cosa or skin), be neither constantly moist nor malodorous,
be hairless, and be sensate at least at the introitus.41 The
method of vaginoplasty used should be a simple, one-staged,
easily reproducible procedure, with low morbidity and good
outcomes (i.e., achieving painless sexual satisfaction). The
ideal procedure has been elusive, and there are differences
of opinion among various specialists concerning the most
appropriate technique for neovaginal construction in patients
with MRKH.2,10,34,42–44
Nonoperative Management
A policy statement from the American College of Obstetrics
and Gynecology in both 2002 and 2006 suggested that non-
operative management using vaginal dilators is the first line of
treatment.42,45 In 1938 Frank described the use of daily dila-
tation of the vaginal dimple for 20 minutes using Pyrex tubes
in six patients with MRKH syndrome.46 In 1983 Rock and col-
leagues reported that only 43% of patients were successfully
managed by dilatation with identification of lack of patient
compliance affecting the results.47 Our team in Boston has
shown 88% successful creation of a functional vagina. The
length of time to success was found to depend on the number
1594 PART VIII GENITOURINARY DISORDERS
of times per day that the dilator was used, with those using
dilators at least once daily for 20 minutes being able to create
a vagina in 4.3  2.4 months.48 We have found success with
the use of hard dilators regardless of the depth of the native
lower vagina. Because the native vaginal mucosa is stretched,
there is normal lubrication and texture to the created vagina.
The dilation technique is best managed by a team of nurse
and physician. Educational materials are the key to success.
The young woman is instructed on the use of the hard dilator
with constant pressure to the appropriate area twice daily. The
angle of insertion of the dilator is a key to success (Fig. 124-2).
Patients are asked to return monthly for examination to assure
that the vagina is being created in the appropriate location and
at the appropriate angle. The next size dilator is dispensed
when appropriate. If the patient elects to cease or hold off
on the process, she can resume when ready.
The vast advantages of this technique support the rationale
for being the first-line therapy for vaginal agenesis in cases of
MRKH and AIS. The advantages include the fact that this tech-
nique is under the control of the young woman, so if she elects
to proceed at a faster or slower rate depending on circum-
stances in her life, she can easily adapt. In addition, it does
not require anesthesia or a surgical procedure. Creation of a
vagina with the use of dilators is also more cost effective than
other treatment options.49 Critics have concerns regarding
adequate length of the vagina, poor lubrication, and dys-
pareunia during intercourse.
Operative Management
A number of operative procedures have been advocated for
creation of the neovagina including insertion of a skin-covered
vaginal mold (Abbe-McIndoe procedure),50,51 artificial skin
grafts,52 various fasciocutaneous53 and myocutaneous flaps
(gluteal-thigh flap, gracilis, rectus abdominis,54 Malaga flap,55
Singapore flap,56,57 vulvovaginoplasty of Williams,58 vulvova-
ginal flap (lotus petal labia majora flap),59 horseshoe flap
using labia minora,60 labial flaps using tissue expanders,61
Bladder
FIGURE 124-2 Creation of a vagi- Vaginal
nal with the use of progressive peri- Rectum
dimple
neal dilation. (From Laufer MR.
Structural abnormalities of the fe-
male reproductive tract. In Emans
SJ, Laufer MR [eds]: Emans, Laufer,
Goldstein’s Pediatric and Adolescent
Gynecology, 6th ed. Philadelphia,
Lippincott Williams & Wilkins, 2011.)
full-thickness skin grafts with vacuum-assisted wound clo-
sure,62 and the use of amnion7,63 and peritoneum.64 Varying
laparoscopic techniques have been described.65–67 The Vec-
chietti68 procedure consists of intra-abdominal traction on
the perineal membrane causing invagination of the shallow
vaginal dimple. This requires a laparoscopic approach and
long-term dilatation, and there have been reports of modifica-
tions and the development of revised instrumentation.69 The
Wharton procedure70,71 simply places a condom-covered
mold in a created space that is left in place for many months.
Bowel vaginoplasty using cecum, small bowel, and sigmoid
colon has been recommended as an alternative to these other
procedures.2,43,72,73 Some of the more popular procedures are
presented here in more detail.
The most popular tissue for vaginal replacement during the
past 3 decades has been the split-thickness skin graft as de-
scribed by McIndoe.51 The McIndoe procedure uses several
split-thickness skin grafts, which are usually harvested from
the buttocks (Fig. 124-3). The grafts are then refashioned over
a vaginal mold that serves as a stent to form a neovagina. The
neovagina is then transferred to its normal anatomic position
in a space surgically created through an incision in the apex of
the vaginal dimple with dissection superiorly between the ure-
thra and bladder anteriorly and the rectum posteriorly. The
dissection is carried up to the pelvic peritoneum to provide
length and to avoid postoperative contracture. The labia are
sutured over the mold to hold it in position. After 7 days of
bedrest the patient is taken back to the operating room for cut-
ting of the labial sutures, removal of the mold, and sewing of
the edge of the perineal skin to the skin grafts. Postoperative
hematoma and fistula due to pressure necrosis related to the
mold are early complications. Use of an inflatable soft vaginal
mold has been considered to reduce these risks. The reported
complication rate is less than 10%. Follow-up reports have
been encouraging,74,75 but as with most surgical techniques,
use of an inflatable mold is hampered by the need for regular
and long-term home dilatation. Due to the use of dermis,
lubricants are required, and a significant incidence of
Bladder
Rectum
Dilator
creating vagina
 CHAPTER 124 ABNORMALITIES OF THE FEMALE GENITAL TRACT 1595

S.P. tube

Bladder
Dissection
between Peritoneal
urethra reflection
and mold
rectum Vaginal

Graft
over Rectum
mold

A B

C
FIGURE 124-3 The McIndoe procedure. A, Split-thickness skin graft sutured over a vaginal mold. B, The mold is placed in a surgically created space
between the bladder and rectum. C, After the mold is removed, the skin graft can be seen in place. (From Laufer MR. Structural abnormalities of the female
reproductive tract. In Emans SJ, Laufer MR [eds]: Emans, Laufer, Goldstein’s Pediatric and Adolescent Gynecology, 6th ed. Philadelphia, Lippincott Williams &
Wilkins, 2011.)

inadequate vaginal length, vaginal stenosis, and dyspareunia
due to contraction of the split-thickness skin graft are draw-
backs of this procedure.76 Antibiotics and analgesics are nec-
essary, and the mold requires changing under an anesthetic
and replacement with a larger-sized stent.
In order to avoid the need for a skin graft donor site, the
use of an acellular human dermal allograft has been
reported.52 Other substances have been recommended as
variations on the technique. Amnion has been used to avoid
the need for autologous graft sites; however, acquiring am-
nion requires scheduled cesarean section and tissue banking
and donors must be carefully screened for HIV infection and
Creutzfeldt-Jakob disease.77 In order to avoid the issues of
infection transmission, chemically processed and freeze-
dried human amnion has been used.78 The peritoneum
has been used, as described in the Davydov procedure, to
line the neovaginal space, but this requires laparotomy or
laparoscopy.64,67,79 A long-term follow-up study on the
use of the Davydov procedure reports that sexuality ap-
proaches so-called “normal sexuality.”80
As noted earlier, a variety of full-thickness skin flaps have
been used to obviate the problem of contracture noted with
split-thickness grafts. The disadvantage of using hair-bearing
areas as the donor source of the flap for vaginal reconstruction
is that hair growth results in neovaginal discharge and is
associated with dyspareunia. These procedures also result in
additional wounds at the donor sites.
Use of vulvar tissues to create a vagina grossly change the
appearance of the external genitalia, and in the Williams
vulvovaginoplasty (Fig. 124-4), an abnormal angle for
intercourse is problematic, resulting in dyspareunia and
making this operation undesirable.58,77 Modifications to the
Williams vaginoplasty have been proposed with the report
of 200 patients with the Creatsas modification.81,82
Use of tissue expanders to create more tissue for a labia
minora graft may prove useful, but late results are not yet
available to ensure that labia minora flaps will result in ade-
quate vaginal length and will endure.60,61 Autologous buccal
mucosa has been used successfully for vaginoplasty.83,84 Lin
and colleagues85 described eight cases using buccal mucosal
grafts. This technique requires a vaginal stent mold and dila-
tation. Complications included vaginal bleeding presumably
from granulation tissue in one patient and a bladder injury
in another. Baker and colleagues84 reported on 21 patients
1596 PART VIII GENITOURINARY DISORDERS
Bladder
FIGURE 124-4 A, Schematic Vaginal
drawing at the completion of the dimple
Williams procedure. B, Schematic
drawing months after completion
of the Williams procedure with Perineal
the creation of a vagina with the tissue
use of dilators and vaginal inter- forms
course. (From Laufer MR. Structural “vaginal
abnormalities of the female repro- pouch” Rectum
ductive tract. In Emans SJ, Laufer
MR [eds]: Emans, Laufer, Goldstein’s
Pediatric and Adolescent Gynecol-
ogy, 6th ed. Philadelphia, Lippincott
Williams & Wilkins, 2011.) A B
with 5-year follow-up from buccal mucosa graft vaginoplasty.
Artificial dermis and recombinant basic fibroblast growth fac-
tor (bFGF) have also been used to create a neovagina in hopes
of accelerating epithelialization and reducing the incidence
of bloody discharge from granulation tissue.86 Because of
the small numbers of patients involved in these studies, no
conclusions can be made at this time regarding the efficacy
of these techniques.
In the Vecchietti procedure,68 at laparotomy (or laparos-
copy87) and with modifications,69 a suture is passed through
the perineal membrane from above and then through a plastic
olive and back up through the abdomen. The suture is then
attached to a traction device strapped to the patient’s abdomen
for a week to 10 days, with pressure on the olive causing in-
vagination of the vaginal dimple, creating a vaginal vault.
Claims of early vaginal length of 10 to 12 cm and a 100% an-
atomic and functional success rate have been reported.88
Complications include bladder and rectal fistula in a small
percentage of cases. The traction method is associated with
some degree of patient discomfort, is cumbersome, still re-
quires long-term vaginal dilatation, and has not been widely
adopted in the United States. Recent revisions to the technique
and refinements to the surgical equipment have resulted in
U.S. Food and Drug Administration approval and wider avail-
ability of the equipment for this technique.69 It should be re-
alized that this technique must be followed by the patient’s use
of manual dilation.
The Wharton procedure71 involves placing a condom-
covered mold in the neovagina to epithelialize vaginal granu-
lation tissue. The process is lengthy because epithelialization
may take 3 to 6 months to occur and is associated with
prolonged bloody vaginal discharge owing to the presence
of granulation tissue.
Schätz and colleagues89 described initial dilatation of rudi-
mentary müllerian ducts in three patients by gradually in-
creasing the Hegar dilator size. The two cavities formed
were joined into one by diathermy, and the resultant space
is lined by smooth tissue. The dome of the space is composed
of the peritoneum reflected from the bladder to the rectum.
Sheares’ modification of the Wharton technique70 was ap-
plied, and an adjustable vaginal stent was placed in the space.
Antibiotics and analgesics were administered for 8 days, and a
Foley catheter was left in place for 1 week to avoid urinary
retention. The vaginal mold was changed under anesthesia
Bladder
Dilatation
completed
Rectum
after 1 week, and a new mold was inserted. The mold remained
in place 24 hours per day for 3 months and then at night there-
after until regular sexual intercourse was initiated. Periodic
changing of the mold required general anesthesia. Subsequent
neovaginal length was 7 to 10 cm, and the width was described
as two fingerbreadths. Insufficient lubrication and dyspareunia
were noted initially with sexual intercourse but resolved using
lubricating gels. The two girls who experienced sexual relations
expressed satisfaction with the procedure.89 The extended
period of time required to maintain the mold in place, the need
for general anesthesia to repeatedly change the mold, inade-
quate lubrication, and lack of long-term follow-up data are
drawbacks of this procedure.
Bowel Vaginoplasty
The use of isolated bowel segments for vaginal replacement
has historically been popular with pediatric surgeons and
pediatric urologists. The use of rectum for the creation of a
vagina was reported as early as 1892 but was not popular
due to the need for a colostomy.90 The use of isolated segments
of small intestine for vaginal replacement was first described
by Baldwin91 in 1904 and performed in 1907.92 Ruge was
the first to use the sigmoid colon to replace the vagina in
1914.93 Experience with this technique was reported by Fall94
in 1940, but efforts were abandoned because of an unaccept-
ably high mortality rate in the preantibiotic era. Attention was
refocused on the use of small bowel and colon for vaginal
replacement in 1972 by Pratt,41 and there have been scattered
reports in the literature ever since.2,44,72,73,95–108
Bowel vaginoplasty uses a section of bowel approximately
10 cm that is mobilized and retains its vascular pedicle to
reach the perineum without compromise to the graft or the
pedicle. A neovaginal space is dissected below the bladder
and above the rectum from the perineum to the pouch of
Douglas, and the bowel segment is sewn in place in an isoper-
istaltic or antiperistaltic orientation.103–105 Most surgeons
advocate anchoring the proximal bowel segment in order to
decrease the risk of prolapse of the bowel vagina.106–108
Currently, an interposed segment of the sigmoid colon is
preferred for vaginal replacement. When the use of this seg-
ment is unavailable because of previous surgery, the cecum
is an acceptable alternative. Although small intestine has
occasionally been used, results are not as gratifying because
 CHAPTER 124 ABNORMALITIES OF THE FEMALE GENITAL TRACT 1597

of excessive mucus formation and friability of small bowel
mucosa with intercourse.
Patients require mechanical bowel preparation for 24 hours
before surgery. Perioperative systemic antibiotics are adminis-
tered 30 to 45 minutes before the incision. The operation is
accomplished with the patient in the lithotomy position with
the legs positioned in stirrups so that the abdomen and peri-
neum can be prepared in the same field. A Foley catheter is
placed in the bladder, and the abdomen and pelvis can be
approached through a Pfannenstiel or lower midline abdom-
inal incision. Alternatively, the procedure has been carried out
using a minimally invasive laparoscopic approach.99,100,109
The peritoneal cavity is explored, and the genital organs are
evaluated. If small rudimentary bicornuate uteri are noted,
these are excised. Uncommonly, if a relatively normal uterus
is observed, this is maintained (see options for uterus with
cervical and vaginal agenesis later). The ovaries should be left
in place to maintain the potential for in vitro fertilization
with a gestational carrier should the patient desire to pursue
this in the future.37,35,36
A 10- to 15-cm sleeve of sigmoid colon, based on the left
colic or superior hemorrhoidal vessels, is sufficient to provide
a functional neovagina. Keeping the segment short avoids ex-
cess mucus production, which can be associated with longer
segments. The sigmoid segment is divided, and intestinal
continuity is reestablished with either a stapled or hand-sewn
end-to-end colocolostomy (Fig. 124-5). In patients with
Mayer-Rokitansky syndrome or those with a hypoplastic va-
gina, an incision is made in the vertex of the vaginal dimple
and a space is created between the bladder anteriorly and
the rectum posteriorly up to the peritoneal reflection.
Depending on the length of mesentery, the isolated segment
is usually brought down to the perineum in an isoperistaltic
fashion or rotated 180 degrees. The proximal end of the neo-
vagina is then closed in two layers, with absorbable suture to
form the vertex of the neovagina. An unrestricted one-layer
anastomosis of the perineal opening to the distal end of the
interposed colon segment using interrupted absorbable su-
tures (3-0 Vicryl or polydioxanone sutures) is then performed.
A three-point fixation of the interposed segment to the

Descending colon

Inferior
mesenteric
artery

Two-layer
closure

10–15 cm sigmoid colon

Introitus
Rectum
A B

Colo-colostomy
Peritoneum

End-to-end
colo-colostomy

Introital
anastomosis
Loose gauze pack

D
C Introitus
FIGURE 124-5 The patient is placed in a lithotomy position with the feet in stirrups. The procedure is performed through a Pfannenstiel incision or
laparoscopically. A, A 10- to 15-cm sigmoid colon segment is isolated. B and C, The proximal portion of the sigmoid neovagina is closed in two layers,
and bowel continuity is restored by an end-to-end colon anastomosis. D, The isolated neovagina is then brought into the cul-de-sac, and the distal end of
the sigmoid vagina is anastomosed to the vaginal remnant, if present, or to the small introital cuff with one layer of interrupted absorbable sutures. The
interposed colon is sutured to the retroperitoneal tissues to prevent prolapse. A Vaseline gauze pack is placed in the new orifice.
1598 PART VIII GENITOURINARY DISORDERS
retroperitoneum with nonabsorbable suture is essential to
prevent prolapse.
Perioperative systemic antibiotics are discontinued after
24 hours. Analgesics are administered for 48 hours and
discontinued after that time. Oral pain medication is available
on request. The neovagina is stented for 5 days after the pro-
cedure with petrolatum gauze. A Foley catheter drains the
bladder during the period of vaginal stenting. Patients are
discharged when they are ambulating and tolerating a diet,
usually on the fifth postoperative day. Patients are examined
at 3 weeks and 3 months. If a stenosis is evident, dilatation
under anesthesia is performed. Home dilatation is usually
unnecessary. Bicycling should be avoided for a 6-week period.
Other exercises are permitted as tolerated. Figure 124-6 shows
the postoperative view after healing.
RESULTS
Long-term follow-up and sexual satisfaction have been
assessed in a few series of patients who have undergone bowel
vaginoplasty. Communal and colleagues106 used the standard-
ized Female Sexual Function Index in 16 patients, and of the
11 who were sexually active and who responded, 8 (75%)
reported satisfactory sexual function results after sigmoid
vaginoplasty. Hensle and colleagues described their experi-
ence with intestinal-vaginal replacement in 44 patients rang-
ing in age from 1 to 20 years.44,72,110 Long-term results were
assessed by a validated Female Sexual Dysfunction Question-
naire, and 78% of the 36 who responded reported sexual
satisfaction.
Syed and colleagues111 described the occurrence of diver-
sion colitis in 3 of 18 children who had undergone sigmoid
vaginoplasty between 18 months and 8 years of age. On the
basis of the report by Syed and colleagues,111 Edmonds77
has suggested that the procedure not be performed in this
young age group. Hiroi and colleagues112 in Japan described
FIGURE 124-6 A, Postoperative
view of the perineum with the sig-
moid vagina anastomosed to vagi-
nal introitus. B, No. 26 Hegar
dilator (6 inches long) reveals nor-
mal vaginal caliber and depth. Pa-
tients are advised to pass the
dilator periodically to ensure contin-
uation of adequate caliber. This pro-
cedure is not as important as it
would be for a skin graft vagina,
which will contract if not perio-
dically dilated in this fashion or by
coitus. (From Hendren WH, Atala
A: Use of bowel for vaginal recon-
struction. J Urol 1994;152:752.)
the only case of malignancy occurring in a sigmoid vagino-
plasty. A mucinous adenocarcinoma of the colovaginoplasty
was observed 30 years after the procedure.
A 7-year follow-up study of 29 patients who underwent
intestinal vaginoplasty has shown positive results, and the
authors recommend the treatment, especially where there is
a cultural moral taboo to the use of vaginal dilators in unmar-
ried women.104
Pediatric surgeons have generally felt that the sigmoid
colon is an ideal tissue for vaginal replacement. Unlike other
methods of vaginoplasty, this procedure creates a vagina that is
capable of providing natural lubrication during sexual activity.
Many women do complain of the natural mucus production
and state that they need to wear a pad daily. Limiting the
length of the interposed segment to the usual vaginal length
of 9 to 10 cm will reduce the risk of excessive mucus produc-
tion. Recognized complications such as prolapse can be
avoided by retroperitoneal fixation of the sigmoid segment.
Use of the sigmoid colon rather than the small intestine will
obviate some of the other complications including stenosis
and excessive mucus production that were seen in the past.
The fact that the procedure can be performed laparoscopically
will reduce the risk of wound complications and intra-abdom-
inal adhesions and make this procedure a more attractive
alternative in the future.99,109,113 It should be noted that
the intestinal procedures do not eliminate the need for vaginal
dilation because patients commonly complain of stenosis at
the site of the anastomosis.
REPRODUCTIVE ISSUES
Reproductive issues relating to MRKH can create emotional
hardship for the affected young woman and her fam-
ily.35,36,38,114 In cases of MRKH, ovarian function is normal.
Women may experience pain with ovulation or with ovarian
cyst formation. The possibility for the use of Assisted
 CHAPTER 124 ABNORMALITIES OF THE FEMALE GENITAL TRACT 1599

Reproductive Technologies (ART) has shown that there is

IMPERFORATE HYMEN
normal response of the ovaries to controlled ovarian hyper-
stimulation for the retrieval of oocytes.35,36,115,116 Perforation of the hymen usually occurs during the fifth
If there is a midline uterus present with the absence of a month of gestation. The hymen occurs at the junction of
cervix and vagina, the uterus can be maintained but it should the sinovaginal bulbs with the urogenital sinus and is usually
not be attached to the neovagina due to the risk of ascending perforated during fetal life. Failure of perforation to occur re-
infection, sepsis, and death.40 Instead, the vagina can be cre- sults in an imperforate hymen.11,77 Different types of hymens
ated by the selected modality and the uterus can be main- are shown in Figure 124-7. An imperforate hymen can result
tained for ART. In order to maintain the uterus and avoid in gross distention of the vagina, which can be filled with
pain with retrograde menstruation and the risk of the devel- fluid (hydrocolpos), mucus (mucocolpos), pus (pyocolpos),
opment of endometriosis, the patient should be treated with or blood (hematocolpos), or in distention of both the vagina
continuous monophasic combination hormonal therapy. and uterus, which is termed hydrometrocolpos, mucometro-
The patient is asked to take the monophasic progesterone colpos, pyometrocolpos, or hematometrocolpos.
dominant oral contraceptive pill daily without breaks or Congenital vaginal obstruction is most commonly caused by
placebos for the suppression of retrograde menses. This treat- simple obstructive anomalies and can include agenesis of the
ment plan maintains the uterus without risk of ascending lower vagina, segmental agenesis, or transverse vaginal septum.
infection and preserves the option for fertility. In this clinical
scenario, pregnancies have been reported.117
DIAGNOSIS
The etiology of MRKH is unknown, but reports have shown
that a specific genetic abnormality has not been identified. We Imperforate hymen usually presents in the newborn period
recently reported a case of monozygotic (monoamniotic/ when there is increased mucus production from maternal es-
monochorionic) twins in which only one twin exhibited tradiol production or at the time of menarche.100 The mucus
MRKH.30 In addition, in 58 women with MRKH undergoing distention will usually resolve with time because there is ces-
ART with gestational carriers, none of the 17 female offspring sation of the estradiol stimulation. The imperforate hymen
had MRKH.31 Thus women with MRKH who desire fertility that is asymptomatic (not causing compromise of urination
can use their oocytes, fertilized with a partner’s or donor’s or other issue) can be observed and addressed as an adolescent
sperm, and the conceptus is implanted and carried in a gesta- or surgically addressed in the newborn period of time.
tional carrier. An imperforate hymen can be suspected on prenatal
ultrasound.118 In the fetus, the imperforate hymen appears
as a thin membrane associated with a distended vagina (sono-
Congenital Vaginal Obstruction lucent mass) and spread labia majora. Imperforate hymen and
------------------------------------------------------------------------------------------------------------------------------------------------
hydrocolpos can be diagnosed as early as the second trimester.
Congenital vaginal obstruction is probably caused by incom- Neonates with vaginal obstruction can present with a lower
plete canalization of the vagina and can occur at differing midline abdominal mass. Frequently, these infants have asso-
levels. The obstruction can result from an imperforate hymen, ciated urinary tract obstruction. The abdominal mass is the
agenesis of the lower vagina, or a transverse vaginal septum. distended vagina, which results from continued collection
It is important to differentiate these anomalies because the of cervical gland secretions in response to maternal estro-
appropriate surgical management varies greatly. gens. Abdominal ultrasonography reveals a large midline

A B C D E
Normal Imperforate Microperforate Cribriform Septate
FIGURE 124-7 Types of hymens: A, normal, B, imperforate, C, microperforate, D, cribriform or complex, and E, septate. (From Emans SJ. Office eval-
uation of the child and adolescent. In Emans SJ, Laufer MR [eds]: Emans, Laufer, Goldstein’s Pediatric and Adolescent Gynecology, 6th ed. Philadelphia,
Lippincott Williams & Wilkins, 2011.)
1600 PART VIII GENITOURINARY DISORDERS

sonolucent mass. This mass causes forward displacement of Agenesis of the Lower Vagina
the bladder and posterior displacement of the rectum. Per- ------------------------------------------------------------------------------------------------------------------------------------------------

cutaneous needle aspiration and injection of contrast medium Agenesis of the lower vagina (ALV) (Fig. 124-11) occurs with a
may aid in the diagnosis and can be done through the peri- normal upper vagina in patients with a uterus and cervix.
neum or the anterior abdominal wall (Fig. 124-8). If no The presentation is similar to an imperforate hymen, and
abdominal mass is present at birth, the condition is often the diagnosis is determined by physical examination and
not detected until early adolescence. At the time of puberty, imaging studies.
symptoms may include amenorrhea, cyclic abdominal pain,
and an abdominal mass secondary to hematocolpos or hydro-
metrocolpos (Fig. 124-9). Introital examination may show a
bulging membrane with bluish discoloration behind it due
to hematocolpos.

TREATMENT
An imperforate hymen in neonates that bulges at the introitus
can be easily incised, but there is a risk that when the hymenal
tissue is no longer bulging, the edges of incised tissue will
adhere, resulting in a persistent structural issue. That is the
rationale for not performing a cruciate incision type procedure
in either the infant or adolescent (Fig. 124-10). This technique
will result in achieving adequate drainage, but there is high
risk of reobstruction. We thus recommend the use of an ellip-
tical incision and removal of the excess hymenal tissue. It is
necessary to clearly identify the anatomy at the time of the
drainage/corrective procedure to ensure that a common uro-
genital sinus or other urinary tract anomaly is not associated
with the obstructive anomaly.

FIGURE 124-9 Lateral view of a contrast study in a patient with a midline

mass pushing the bladder (B) forward and the rectum (R) in a posterior
direction. V indicates the obstructed vagina.

Bulging
hymen
Incision

Vaginal
orifice

FIGURE 124-8 Percutaneous injection of an obstructed vagina (V) in a FIGURE 124-10 Surgical correction of imperforate hymen. (From Laufer
newborn, with simultaneous cystogram of bladder (B) and retrograde MR. Structural abnormalities of the female reproductive tract. In Emans SJ,
catheterization of a solitary left ureter and kidney. Laufer MR [eds]: Emans, Laufer, Goldstein’s Pediatric and Adolescent Gyne-
cology, 6th ed. Philadelphia, Lippincott Williams & Wilkins, 2011.)
 CHAPTER 124 ABNORMALITIES OF THE FEMALE GENITAL TRACT 1601

examiner may be able to insert a finger in the rectum (with

the patient in the dorsal supine position) and appreciate a
space between the obstructed vagina and the location for
the introitus (see Fig. 124-11). The surgeon may not appreci-
Bladder ate the difference until an operation is in progress. If there is
some amount of areolar tissue until the obstructed vagina is
Vaginal reached, then this is ALV and not an imperforate hymen. Note
dimple the distance that has been dissected before reaching the vaginal
mucosa in Figure 124-12, A. This is an important differential
because if the upper vagina is distended and pointing toward
Blood-filled
uterus and the introitus, the blood acts as a natural tissue expander and in-
Agenesis of upper vagina creases the amount of native vaginal tissue. Once the obstruc-
lower vagina Rectum
tion is relieved and the old blood is evacuated, the native upper
vagina will tend to retract back up toward the uterus. This is the
reason the native vaginal tissue must be sewn to the newly cre-
ated introitus (Fig. 124-12, B and C). A flexible vaginal dilator
should be used 24 hours a day 7 days a week for some number
FIGURE 124-11 Agenesis of the lower vagina with a blood-filled upper va- of months so that stenosis of the vaginal orifice will not occur
gina. (From Laufer MR. Structural abnormalities of the female reproductive
tract. In Emans SJ, Laufer MR [eds]: Emans, Laufer, Goldstein’s Pediatric and Ad-
and a fistulous tract will not form if the upper vagina retracts
olescent Gynecology, 6th ed. Philadelphia, Lippincott Williams & Wilkins, 2011.) after the pull-through vaginoplasty.

DIAGNOSIS
Transverse Vaginal Septum
The external appearance of ALV can easily be confused with an ------------------------------------------------------------------------------------------------------------------------------------------------

imperforate hymen. The key means to a correct diagnosis Transverse vaginal septum (TVS) can occur at varying levels of
include the observation that there are normal fronds of hymen the vagina (Fig. 124-13) and is referred to as low, middle, and
and there is not just a membrane of tissue. In addition, the high transverse vaginal septum. Although highly variable, it

Pull through Bladder

of edges of
vaginal mucosa
to perineum

Rectum

A B

C
FIGURE 124-12 A, Transverse incision is made to create the vaginal orifice. The dissection is carried through the areolar type tissue to reach the obstructed
vagina. B, After drainage, the previously obstructed upper vaginal tissue is pulled through to the perineum to create a patent tract. C, The normal upper vagina has
been pulled down and will now be sewn in place. (From Laufer MR. Structural abnormalities of the female reproductive tract. In Emans SJ, Laufer MR [eds]: Emans,
Laufer, Goldstein’s Pediatric and Adolescent Gynecology, 6th ed. Philadelphia, Lippincott Williams & Wilkins, 2011.)
1602 PART VIII GENITOURINARY DISORDERS

Bladder

4
Levels of 2 3
septum 1

A
Hymen
Rectum

FIGURE 124-13 Locations of transverse vaginal septum. (From Laufer

MR. Structural abnormalities of the female reproductive tract. In Emans
SJ, Laufer MR [eds]: Emans, Laufer, Goldstein’s Pediatric and Adolescent
Gynecology, 6th ed. Philadelphia, Lippincott Williams & Wilkins, 2011.)

has been reported that approximately 46% of vaginal septa oc-

cur in the upper vagina, 40% in the middle vagina, and 14% in
the lower vagina.119 The septum can be obstructing or have a
small perforation either in the midline or laterally. The presen- B
tation will vary depending on whether there is a perforation.
FIGURE 124-14 Fusion anomalies of the uterus (A) and vagina (B).
Pelvic ultrasonography is usually diagnostic of hemato-
colpos. The high transverse vaginal septum is best treated
OHVIRA
by draining the vagina by cutting into the septum, resecting
the septum, and performing a simultaneous transperineal In cases of complete vaginal duplication, one vagina can be
vaginal pull-through procedure attaching the upper and lower obstructed and the other can be patent, which is referred to as
vagina using absorbable sutures. This condition is best obstructed hemivagina with ipsilateral renal anomaly (OHVIRA)121
addressed after menarche, and the patient is asked to wear or Herlyn-Werner-Wunderlich syndrome (Fig. 124-15).122,123
a flexible vaginal dilator to avoid circumferential stricture. Urinary tract anomalies may confound the situation with an ec-
The results with lower septum defect repair are much topic ureter inserting into a duplicated imperforate hemiva-
better than outcomes achieved with high vaginal septum gina, resulting in a mass.13,44,124 The external genitalia will
defects due to difficulty with the use of dilators reaching usually appear normal. The diagnosis of uterus didelphys and
the area of the high septum. a unilateral obstructed vagina should be entertained as part of
the differential diagnosis in a newborn with a sonolucent ab-
dominal mass. The mass usually pushes the bladder forward
and the normal vagina in a posterior direction. Pelvic examina-
Anomalies of Fusion
------------------------------------------------------------------------------------------------------------------------------------------------ tion without possible vaginoscopy is helpful in this situation
and may reveal a bulging mass high in the vaginal side wall.
VAGINAL DUPLICATION
A simple incision of the obstructing septum with resection
Disorders of embryogenesis that produce duplication anoma- of a “window of tissue” often provides adequate drainage; how-
lies of the vagina and uterus occur at approximately 9 weeks ever, further resection of the septum is necessary to prevent
of gestation and involve a complete or partial failure of union possible collection of bacteria on the abnormal side and ab-
of the two lateral müllerian ducts. These anomalies120 are scess formation and is usually performed after menarche.11,121
common and can include two uteri and two cervices with We advocate a single procedure for OHVIRA during adoles-
two separate vaginas (uterus didelphys) or a single vagina cence.121 Rarely a staged procedure is necessary but is useful
(uterus duplex bicollis). They may also consist of two uteri when the procedure must be performed on children (due to
fused with a single cervix and a single vagina (uterus duplex urinary symptoms, pain, or infection) and when there is a
unicollis or bicornuate uterus) (Fig. 124-14). pyocolpos with associated inflammation in the adolescent.
 CHAPTER 124
FIGURE 124-15 Obstructed hemivagina with ipsilateral renal anomaly
[OHVIRA]. (From Laufer MR. Structural abnormalities of the female repro-
ductive tract. In Emans SJ, Laufer MR [eds]: Emans, Laufer, Goldstein’s
Pediatric and Adolescent Gynecology, 6th ed. Philadelphia, Lippincott
Williams & Wilkins, 2011.)
FIGURE 124-16 Longitudinal vaginal septum. (From Laufer MR. Struc-
tural abnormalities of the female reproductive tract. In Emans SJ, Laufer
MR [eds]: Emans, Laufer, Goldstein’s Pediatric and Adolescent Gynecology,
6th ed. Philadelphia, Lippincott Williams & Wilkins, 2011.)
ABNORMALITIES OF THE FEMALE GENITAL TRACT 1603
Longitudinal Vaginal Septum
------------------------------------------------------------------------------------------------------------------------------------------------
For nonobstructed longitudinal vaginal septum (Fig. 124-16),
a surgical excision of the septum is required only if the young
woman desires such a procedure. Some patients complain
that they need to use a pad in addition to a tampon (because
the tampon is absorbing menstrual blood from only one of the
two vaginas). Some maintain the longitudinal vaginal septum
and have no issues with sexual activity or vaginal delivery.
Most have two cervices, so a Papanicolaou smear must be
obtained from both.
The septum should be excised by “wedging” out the fibrous
septal tissue. Great care is necessary to avoid the bladder
above, the rectum below, and the septal tissue between the
two cervixes at the apex. Once the septal tissue is excised,
the normal vaginal mucosa from one vagina is sewn to the
other vagina to create a smooth vaginal mucosa and close
the resulting defect. Excision of the septal tissue removes
the fibrous tissue to avoid dyspareunia.
If there are two systems that are not obstructed, each can
function independently, as two unicornuate systems, from a
fertility standpoint.125
RUDIMENTARY UTERINE HORNS
A young woman may present with pain and regular menses.
The possibility of a patent unicornuate system (Fig. 124-17)
and an obstructed hemi uterus (Fig. 124-18) should be consid-
ered. The obstructed hemi uterus may be appreciated on ultra-
sonography or MRI. In cases of MRKH, if rudimentary uteri are
present, they may be difficult to appreciate even on MRI.126
The obstructed hemi uterus should be removed so that the
patient will no longer have retrograde menses and pain. The
procedure can be performed laparoscopically. The cervix
should be canalized, and blue dye should be injected at the
time of the laparoscopy so that it can be confirmed which
FIGURE 124-17 Unicornuate system. (From Laufer MR. Structural ab-
normalities of the female reproductive tract. In Emans SJ, Laufer MR
[eds]: Emans, Laufer, Goldstein’s Pediatric and Adolescent Gynecology,
6th ed. Philadelphia, Lippincott Williams & Wilkins, 2011.)
1604 PART VIII GENITOURINARY DISORDERS

vagina and urethra into the urogenital sinus are frequently as-
sociated with an anteriorly displaced anus. This indicates poor
formation of the urorectal septum.

DIAGNOSIS
Adequate definition of the patient’s anatomy is essential for
proper evaluation of any urogenital sinus anomaly. A retro-
grade genitogram (Fig. 124-21) can assist in the definition
of the urogenital sinus by delineating the length of the com-
mon channel and identifying the anatomic relationship of
the vagina and urethra.128 This procedure is best done with
a blunt adaptor placed against the perineal opening to instill
the contrast agent. Cystoscopy and vaginoscopy should be
performed in each case.

FIGURE 124-18 Unicornuate system with an obstructed hemiuterus. TREATMENT

(From Laufer MR. Structural abnormalities of the female reproductive
tract. In Emans SJ, Laufer MR [eds]: Emans, Laufer, Goldstein’s Pediatric Once the anatomy of the defect has been clearly identified, var-
and Adolescent Gynecology, 6th ed. Philadelphia, Lippincott Williams & ious treatment options can be considered. A simple U-flap vagi-
Wilkins, 2011.) noplasty129 can be done if the urogenital sinus is low and the
common channel is short. In the cases of a urogenital sinus with
uterus communicates with the cervix (Fig. 124-19). The a high confluence of the bladder and vagina, Donahoe and Gus-
remaining unicornuate system will most likely function nor- tafson130 advocate an inverted perineal U-flap in combination
mally but has an increased risk of breech presentation and with bilateral buttock flaps to augment the vagina. However, the
premature labor.35,125 urogenital sinus is frequently associated with an anteriorly
placed anus, which requires posterior relocation of the anus be-
fore the perineal skin flap vaginoplasty is done.131 If the vagina
Urogenital Sinus Anomalies
------------------------------------------------------------------------------------------------------------------------------------------------
enters the urogenital sinus too far proximally, a division of the
vaginal moiety from the urogenital sinus in conjunction with a
The urogenital sinus is defined as a common channel into pull-through vaginoplasty will be necessary. A confluence of
which both the urinary and genital tracts open. The common both the vagina and the bladder high in the urogenital sinus re-
urogenital sinus is a normal stage of embryonic development quires a pull-through vaginoplasty and often requires the cre-
in both sexes. In normal females, an arrest in development of ation of a neourethra from the anterior vaginal wall.132
the müllerian ducts at 9 weeks’ gestation, after fusion with the
urogenital sinus, manifests as a urovaginal confluence or com-
mon urogenital sinus. An arrest of vaginal differentiation at a Cloacal Malformations
slightly later date can lead to varying degrees of persistence of ------------------------------------------------------------------------------------------------------------------------------------------------

the urogenital sinus seen clinically (Fig. 124-20). A long uro- Cloacal malformations result from the combination of a
genital sinus with a short vagina and high urethral opening urogenital sinus with an anorectal anomaly.133 In the early
will result if the defect occurs at an early stage. Conversely, embryo there is a cloacal stage during which there is a single
a short urogenital sinus with an almost normal vaginal vesti- common tract. At 4 to 6 weeks’ gestation, the urorectal
bule and low urethral orifice will occur if the arrest occurs late septum divides the allantois from the hindgut. If this separa-
in development.127 Early defects with a high insertion of the tion does not take place, a common cloaca will result. Infants

FIGURE 124-19 Unicornuate sys-

tem with a right obstructed hemiu-
terus. Note blue dye coming from
left fallopian tube after injection of
dye into the single cervix. (From
Laufer MR. Structural abnormalities
of the female reproductive tract. In
Emans SJ, Laufer MR [eds]: Emans,
Laufer, Goldstein’s Pediatric and
Adolescent Gynecology, 6th ed.
Philadelphia, Lippincott Williams &
Wilkins, 2011.)
 CHAPTER 124 ABNORMALITIES OF THE FEMALE GENITAL TRACT 1605

A B

FIGURE 124-20 A, Urogenital sinus with a low confluence. B, Entry of

vagina in the midportion of the urogenital sinus. C, High confluence of
C the vagina and urogenital sinus at the level of the urinary sphincter.

with this defect present at birth with abdominal distention DIAGNOSIS

and an abnormal perineum. They usually have no anus or
vagina but possess a single perineal opening. The phallus-like
structure frequently has a hooded appearance, which gives a
masculinized gender appearance (Fig. 124-22). The bladder
may enter the cloaca immediately inside the perineal opening
and may be associated with a relatively normal length ure-
thra. Conversely, the bladder may enter the cloaca very
high and be located in proximity to the bladder neck. The
entrance of the rectum into the cloaca can similarly be either
high or low (Fig. 124-23).134 Vaginal anomalies are common
in patients with a cloaca, and vaginal duplication is seen most
often. There may also be significant perineal obstruction
in these neonates, resulting in both hydrocolpos and an
obstructive uropathy (Fig. 124-24).
Diagnosis of a cloacal anomaly is confirmed by a retrograde
contrast study through the single perineal opening. Fre-
quently, these studies demonstrate a confluence of both the
genitourinary and gastrointestinal systems. Blask and col-
leagues135 reported on the usefulness of ultrasonography in
nine neonates who had either an obstructed urogenital sinus
or a cloacal anomaly. On abdominopelvic ultrasonography, the
bladder was often compressed by the distended vagina and
difficult to visualize. Typically, the vagina was markedly dis-
tended with urine and a vaginal fluid-debris level distin-
guished the vagina from a distended bladder. However, all
components of the cloacal anomaly are often not identified
on ultrasonography or retrograde studies. Therefore it is only
1606 PART VIII GENITOURINARY DISORDERS
FIGURE 124-21 Retrograde genitogram of a urogenital sinus showing
a high confluence of the bladder (B) and vagina (V) into a long urogenital sinus.
FIGURE 124-22 Perineum of a patient with a cloacal anomaly.
at the time of panendoscopy or an open surgical procedure
that the exact anatomy can be defined. Complete definition
of the patient’s anatomy is the first principle of treatment.136
Information concerning cloacal exstrophy can be found in
Chapter 119.
Other Gynecologic Issues
------------------------------------------------------------------------------------------------------------------------------------------------
LABIAL ADHESIONS
Labial fusion is a common finding in infants and young chil-
dren. Labial fusion may present as a primary skin bridge and
may be associated with or identified at the time of an evalua-
tion of a urinary tract infection. The etiology of labial adhe-
sions is unknown. In most cases the adhesions are
asymptomatic and do not require an intervention. They will
usually resolve completely with exogenous estrogen produc-
tion with puberty. The adhesions may be symptomatic, caus-
ing irritation, pulling sensation, or complaints of wetness
due to trapping of urine in the vagina such that the child
wipes after urination and then stands, only to have resulting
wetness in her underwear from spontaneous evacuation of
the trapped vaginal urine. In symptomatic cases the labial ad-
hesions will easily respond to the twice-daily application of
topical estrogen cream or a mild corticosteroid such as beta-
methasone 0.05%. A surgical intervention is rarely necessary
and is reserved for cases of complete obstruction and the in-
ability to urinate. If a surgical intervention is required, great
care should be taken to make sure that the lysis of adhesions
is performed without cutting into the normal labial tissue. If
the normal labial tissue is incised, bleeding may occur and
resulting scar tissue may not be estrogen responsive in the
future. Thus daily topical application of conjugated estrogen
cream should be the first line of therapy in almost all cases of
labial adhesions.
MASSES OF THE INTROITUS
A mass presenting at the introitus between the labia is often a
diagnostic dilemma. The differential diagnosis includes
prolapse of the urethra, prolapsed ectopic ureterocele, pro-
lapsed yolk-sac tumor, and rhabdomyosarcoma (sarcoma
botryoides). It is important to recognize each of these clinical
entities and to understand the diagnostic and treatment
criteria for each.
URETHRAL PROLAPSE
Prolapse of the urethra (Fig. 124-25) occurs most commonly
in the prepubescent years. Affected children frequently
present with blood spotting on the underwear and a mass
at the introitus. The tissue appears inflamed and friable and
represents a prolapsed portion of the anterior or posterior
urethra. The cause remains unclear. An accurate diagnosis de-
pends on recognition of the entity. No special diagnostic stud-
ies are necessary. Medical treatment with sitz baths and topical
estrogen cream is usually successful. Once treated medically
with topical estrogen, an examination is necessary to be cer-
tain that a urethral polyp is not present. Operative treatment
is rarely required, even in cases of distal necrosis. If a surgical
procedure is used, it usually involves simple excision of the
prolapsed segment, with suturing of the normal urethra to
the meatus.137 With a surgical procedure, complications are
uncommon except for mild degrees of meatal stenosis after
excision of the prolapsed segment. Valerie and colleagues138
noted recurrence of prolapse in 1 of 20 patients treated by
excision.
 CHAPTER 124
A B
FIGURE 124-23 A to D, Various forms of the cloacal anomaly demonstrating proximal and distal forms.
FIGURE 124-24 Antegrade study of the bladder (B) and vagina (V) in
a patient with a cloacal anomaly. A dilated urogenital sinus (arrow) and
relative obstruction at the perineal outlet are shown.
PROLAPSED ECTOPIC URETEROCELE
Prolapsed ectopic ureteroceles can present as an intralabial
introital mass (Fig. 124-26).139 They are usually smooth
and white or slightly edematous lesions that extrude from
the urethral meatus. The urethral meatus can often be seen
above the lesion. The diagnosis can usually be made on intra-
venous urography, which demonstrates a lower pole collecting
system that is pushed laterally and inferiorly (i.e., drooping
lily sign). The ureter is displaced away from the midline on
the affected side, and the bladder has a filling defect (see
Fig. 124-26).140 Current management of these lesions is
presented in Chapter 114.
SARCOMA BOTRYOIDES
Sarcoma botryoides, or rhabdomyosarcoma of the bladder or
vagina, is the most common malignant condition of the lower
genitourinary system presenting in infancy.141 The tumor may
be observed protruding from the introitus. It often presents as
ABNORMALITIES OF THE FEMALE GENITAL TRACT 1607
C D
FIGURE 124-25 Prolapsed urethra in a 9-year-old girl. Note the
congested, necrotic appearance of the prolapsed tissue.
blood spotting on the undergarments with the diagnosis
then achieved by vaginoscopy and biopsy.142 The lesion, as
the name denotes, appears as a lobulated, grapelike mass
(Fig. 124-27). The extent of the lesion is best determined
by pelvic ultrasonography and CT. Rhabdomyosarcoma
occurring at this primary site (vulvovaginal) is associated
with an excellent outcome, with more than 90% of patients
surviving with limited fertility-preserving surgery (including
vaginoscopic resection) and multimodal chemotherapy.142–146
Further details concerning management can be found in
Chapter 34.
YOLK SAC TUMOR
The vagina is a usual primary site for yolk sac tumors. These
neoplasms can present protruding from the introitus or with
vaginal spotting or bleeding.147 These lesions are classified as
endodermal sinus tumors (germ cell tumors), and affected
patients frequently have an elevated serum alpha fetoprotein
1608 PART VIII GENITOURINARY DISORDERS

A B
FIGURE 124-26 A, Prolapsed ureterocele in a 3-year-old girl. Note the white, healthy appearance of the cystlike structure. B, Intravenous urogram in a
patient with an ectopic ureterocele demonstrating a right lower pole collecting system being pushed in a lateral and inferior direction (drooping lily sign)
along with a filling defect in the bladder.

In fact, some cases have demonstrated resolution of the tumor

with chemotherapy alone.150 This topic is presented in depth
in Chapter 39.

MISCELLANEOUS INTROITAL LESIONS

FIGURE 124-27 Sarcoma botryoides. (From Emans SJ. Vulvovaginal
problems in the prepubertal child. In Emans SJ, Laufer MR [eds]:
Emans, Laufer, Goldstein’s Pediatric and Adolescent Gynecology, 6th ed.
Philadelphia, Lippincott Williams & Wilkins, 2011.)
level. Germ cell tumors are treated with platinum-based che-
motherapy programs and often have an excellent outcome
(85% survival). Fertility-sparing vaginal surgery and adjuvant
chemotherapy have been successful in these cases.148,149
Paraurethral cysts also present as interlabial masses in fe-
male infants.151 They tend to displace the urethral meatus
and vaginal introitus to one side. A normal urethra and vag-
inal introitus can usually be identified (see Fig. 124-20).
These cysts are probably embryologic remnants of Skene
glands or the Gartner duct. Most of these cysts regress spon-
taneously and rarely need to be incised and drained. The
Gartner duct cysts are remnants of the wolffian duct and
may be associated with unilateral renal dysplasia152 and du-
plicated collecting system with ectopic ureter emptying into
the vagina or rarely the cyst wall.153,154 Goldstein and col-
leagues153 were the first to describe the relationship of the
Gartner duct cyst and ipsilateral renal agenesis in 1973. Both
patients in that report had a bicornuate uterus. Patients often
present with urinary tract infection and incontinence due to
the ectopic ureter emptying into the vagina or the Gartner cyst
wall.155,156 Once the urinary tract abnormalities have
been addressed, the Gartner cyst can either be excised or
marsupialized.
 CHAPTER 124 ABNORMALITIES OF THE FEMALE GENITAL TRACT 1609

Other Vaginal Tumors (21%) with vaginal tumors had metastases at diagnosis. Delay
------------------------------------------------------------------------------------------------------------------------------------------------ in diagnosis was common because of the lack of an appropri-
HEMANGIOMA ate pelvic examination in patients younger than 13 years old.
Nine patients (36%) had not reached menarche. Seventy-six
Hemangiomas can occur in the vagina, as they can in many percent had International Federation of Gynecology and Ob-
other tissues. Vaginal hemangiomas are occasionally extensive stetrics (FIGO) stage I or II disease, whereas 24% had stage III
and coexist with hemangiomas affecting the skin and sub- or IV (advanced-stage) disease. Twenty-two girls (58%) under-
cutaneous tissues of the perineum and perianal areas and went a radical operative procedure (radical hysterectomy or
rectal wall.157 Conservative management is usually recom- pelvic exenteration): 19 had stage I or II disease, and in 3
mended because the natural history of these lesions suggests the stage was undocumented. Seven of these patients received
that spontaneous resolution will occur over a period of radiation therapy. Seven girls (18%) had tumor excision alone
months to years. Ulceration and bleeding are complications combined with radiation therapy in three and chemotherapy
that may require treatment with corticosteroids and inter- and radiation in one. Seven patients with advanced disease
feron-a.157 Failure of conservative treatment may require had no surgery. Three received radiation therapy, three had
excision. Further information regarding hemangiomas is chemotherapy, and one had both. Overall, the 3-year
available in Chapter 124. disease-free survival was 71% (27/38). Ten patients died of
progressive or recurrent disease. Relapse was an ominous
finding. Sites of recurrence were the lungs in six patients,
ADENOCARCINOMA OF THE CERVIX the lymph nodes or soft tissues in two, and the brain in
one. The 3-year survival estimate for those undergoing surgi-
AND VAGINA
cal removal of tumor was 85%. Girls with tumor excision
Adenocarcinoma of the cervix and vagina is rare in children. (conservative surgery) did better than those with radical sur-
Less than 3% of sexually active adolescents may show evi- gery (100% 3-year survival vs. 78%), although there were only
dence of cervical intraepithelial neoplasia, and annual Papani- seven patients with tumor excision and four received adjuvant
colaou smears have been recommended. In 1971 Herbst and therapy. The study did not evaluate tumor size. Two of the
colleagues158 described the occurrence of clear cell adenocar- seven patients with tumor excision had stage III disease,
cinoma of the vagina in young adolescent girls who were prog- whereas all of those undergoing radical surgery had stage I
eny of mothers who had received diethylstilbestrol during or II disease. The only statistically significant predictors of sur-
pregnancy. Offspring of women with this history are recom- vival were surgical resection and the presence of metastases at
mended to have a gynecologic examination by age 14 or after diagnosis. The data suggest that conservative surgical treat-
menarche and then annually thereafter. Vaginal clear cell ad- ment with adjuvant therapy may obviate the need for radical
enocarcinoma has been observed in an infant who presented surgery. A prospective randomized study in a larger number of
with hematuria.159 In 2004 McNall and colleagues160 patients is necessary to confirm these observations. Because of
reviewed 37 patients 18 years of age or younger with vaginal the rarity of the tumor, this may be difficult to achieve. Long-
or cervical mesonephric adenocarcinoma and clear cell adeno- term follow-up is important in these patients because late
carcinoma. The patients’ ages ranged from 7 months to recurrences have been observed.161,162
18 years. The most common presenting symptoms were
vaginal bleeding (89%) and/or discharge (16%). Sixty-two The complete reference list is available online at www.
percent were exposed to diethylstilbestrol. Eight of 27 girls expertconsult.com.
 Virchow is likely the first to have categorized vascular anom-
alies on the basis of histologic features.4
For centuries, the field has been plagued by overlapping
vernacular, clinical, and histopathologic terms resulting in
misdiagnosis, inappropriate treatment, and misdirected
research. A common example is the continued application
of the word “hemangioma” to describe vascular lesions of
any kind irrespective of behavior and biology. The biologic
classification system in use today was formally accepted in
1996 by the International Society for the Study of Vascular
Anomalies (Table 125-1).5 In addition to classifying tumors
and malformations based on cellular and clinical behavior,
they also exhibit their own unique radiographic findings
and immunohistochemical properties.2

Vascular Tumors
------------------------------------------------------------------------------------------------------------------------------------------------

INFANTILE HEMANGIOMA
Infantile hemangiomas (IHs), benign tumors of the endothe-
lium, are the most common tumor of infancy (Fig. 125-1). It
occurs in approximately 4% of white-skinned infants. A fre-
CHAPTER 125 quently quoted study probably overestimated the incidence
of 10% by including other vascular lesions.6 The incidence
is lower in dark-skinned babies. There is a female-to-male pre-
ponderance of 3:1 to 5:1.2 Extremely-low-birth-weight infants

Vascular Anomalies (<1000 g) have the highest incidence of IHs, approaching

23%.7 Additional risk factors include advanced maternal
age, multiple gestations, and placental abnormalities.8
Ann M. Kulungowski and Steven J. Fishman IHs most often occur as a single cutaneous lesion (80%)
with a predilection for the head and neck (60%), trunk
(25%), and extremities (20%) (see Fig. 125-1).9 Multiple tu-
mors are present in up to 20% of patients and, when present,
may signal involvement of extracutaneous organs such as the
A reliable classification system for vascular anomalies was liver or gastrointestinal (GI) tract.2 Median age of onset is 1 to
developed 30 years ago.1 Vascular anomalies are divided into 2 weeks. A premonitory cutaneous mark such as a pale spot or
two categories, tumors and malformations, on the basis of faint macular stain is present at birth in 30% to 50% of cases.2
clinical behavior and cellular kinetics.2 Vascular tumors are The majority (90%) of IHs are small, localized lesions that do
characterized by endothelial hyperplasia; vascular malforma- not involve aesthetically or functionally vital structures.2,10
tions arise due to vascular dysmorphogenesis and exhibit A hallmark of IH is its predictable life cycle (Fig. 125-2).
normal endothelial turnover.2 The classification scheme pro- During the proliferating phase, there is rapid growth of the tu-
vides the framework for understanding the pathophysiology, mor, which typically lasts until 10 to 12 months of age. Super-
diagnosis, and prognosis of these lesions. ficial hemangiomas are red; deeper lesions may be noted later
Pediatric surgeons are often called on to assist with diag- as a blue mass visualized through the skin. Around 12 months
nosis and treatment of vascular anomalies. The majority of of age, growth of the IH plateaus, marking the beginning of the
lesions are noted at birth or during infancy because they involuting phase. Over the next 6 to 7 years, the crimson color
commonly involve the skin. A thorough history and physical fades, the center becomes pale, and the lesion appears to de-
examination leads to the correct diagnosis in more than 90% flate.2 By 5 years of age, 50% of tumors have completed invo-
of patients. Familiarity with the field of vascular anomalies, lution, which increases to 70% at 7 years of age. There is often
its classification scheme, and use of correct nomenclature continued gradual regression of the color and bulk of the
will ultimately improve outcomes for patients affected by tumor until 10 to 12 years of age.11 When the IH has ceased
these lesions. involuting, it enters the involuted phase. At the end of invo-
lution, 50% of patients have nearly normal skin in the area
of the prior lesion. Large tumors can leave lax, redundant skin
Historical Perspective and/or a fibrofatty residuum. Previously ulcerated lesions can
------------------------------------------------------------------------------------------------------------------------------------------------
leave permanently damaged skin, scars, and discoloration.2
Historically, vascular birthmarks were thought to be imprinted Endangering or life-threatening IHs are rare.12 Recognition
on the unborn child due to a mother’s emotions.1,3 The use of of the anatomic distribution of the lesions helps predict pos-
adjectives such as “strawberry” and “port wine” are reflected in sible complications. Cervicofacial and subglottic IHs can be
this doctrine of maternal impressions.4 Physicians preferred life-threatening due to airway obstruction.2 Subglottic IHs
the Latin term naevus maternus. In the nineteenth century, are characterized initially by hoarseness and later biphasic

1613
1614 PART IX SPECIAL AREAS

TABLE 125-1 may be indicated. The liver is the most frequently involved
Classification of Vascular Anomalies location.
Tumors Malformations Etiology and Pathogenesis The exact etiology and patho-
Hemangioma Slow-flow genesis of IH remains to be elucidated. There is emerging ev-
Infantile Hemangioma Capillary idence for an endothelial stem/progenitor cell as the cellular
Congenital Hemangioma Lymphatic origin of hemangioma.13–15 The source of these endothelial
- Rapidly involuting congenital - Microcystic progenitors remains elusive. Some studies suggest a popula-
hemangioma (RICH) - Macrocystic tion of resident angioblasts, arrested in an early stage of vas-
- Noninvoluting congenital - Mixed cular development, as a source. Hemangioma endothelial
hemangioma (NICH) - Lymphedema cells may be of placental origin because they coexpress several
- Gorham-Stout disease markers: glucose transporter-1 (GLUT-1), type III iodothyro-
- Lymphangiectasia (pleural, nine deiodinase, Fcg-RIIb, merosin, and Lewis Y antigen.16–20
pulmonary, intestinal) Disruption of the maternal-fetal barrier may allow an embolic
Venous nidus of placental endothelial cells to reach fetal tissues
Kaposiform Hemangioendothelioma Fast-flow through the permissive right to left shunt of fetal circulation.21
Kaposiform Lymphatic Anomaly Arteriovenous fistula This could explain the increased incidence of IH observed
Tufted Angioma Arteriovenous malformation with chorionic villus sampling and prematurity, sometimes
Cutaneovisceral Angiomatosis with a result of placental disorders.22
Thrombocytopenia The characteristic growth and involution of IH seen on
Angiosarcoma physical examination parallel the cellular activity of the tumor.
Kaposi Sarcoma During the proliferating phase, angiogenesis occurs as endo-
Pyogenic Granuloma Complex-Combined thelial cells rapidly divide, forming a mass of sinusoidal vas-
Capillary-lymphatico-venous cular channels with feeding arteries and draining veins.23–25
(Klippel-Trenaunay) A proangiogenic environment is created by the presence of
Capillary-arteriovenous factors such as basic fibroblast growth factor, vascular endo-
(Parkes Weber)
thelial growth factor, and matrix metalloproteinases. In con-
Lymphatico-venous
trast, angiogenesis is decreased during the involuting phase
Rare syndromes (see text)
as endothelial cells undergo apoptosis.26 This coincides
with increased expression of angiogenesis inhibitors such as
interferon-b and tissue inhibitor of metalloproteinase.23,24
stridor around 6 to 12 weeks of age. Ulceration of the eyelid, Associated Structural Abnormalities Congenital abnor-
nasal tip, lip, and ear can be disfiguring. Periorbital IHs can malities are rarely associated with IH, but larger hemangiomas
block the visual axis and cause deprivation amblyopia; upper and those encountered in the midline merit attention. A sub-
eyelid and supraorbital lesions can distort the cornea, produc- group of patients with IH exhibits associated structural anom-
ing astigmatic amblyopia. Infants with periorbital hemangi- alies of the brain (e.g., posterior fossa abnormalities), cerebral
oma should be examined by a pediatric ophthalmologist. GI vasculature (e.g., hypoplasia or absent carotid and vertebral
IHs can cause GI bleeding. Complications of hepatic heman- vessels, aneurysms), eye (e.g., cataracts and optic nerve hypo-
gioma include high-output cardiac failure, hypothyroidism, plasia), aorta (e.g., coarctation), and chest wall defects (e.g.,
and abdominal compartment syndrome. sternal clefts) in the neurocutaneous disorder called PHACES
The occurrence of multiple IHs is called hemangiomatosis syndrome (Posterior fossa malformations, Hemangioma,
(see Fig. 125-1, left, middle). In this setting, the cutaneous le- Arterial anomalies, Cardiac defects, Eye anomalies, and Sternal
sions are usually tiny (<5 mm) and domelike, but some may defects).27–30 Tumors of the lumbosacral region may occur
have the more typical appearance of a single tumor. Occult vis- along with spinal dysraphism abnormalities such as a tethered
ceral lesions may be present when multiple cutaneous IHs cord or lipomeningocele. Ultrasonography is used to screen
(usually five or more) are found. Screening ultrasonography infants younger than 4 months of age for the presence of

FIGURE 125-1 Morphologic variation of infantile hemangioma (IH). Proliferating phase IH on the trunk (left). Multiple cutaneous dome-shaped IHs, which
are often seen in association with hepatic hemangioma (left, middle). Proliferating IH in the perineum (right, middle). Reticular IH over the midline (right).
 CHAPTER 125 VASCULAR ANOMALIES 1615

FIGURE 125-2 Progression and involution of infantile hemangioma (IH) in the beard distribution. Proliferating IH at 2 weeks of age (top, left). Cortico-
steroid treatment initiated. After 2 weeks of therapy (top, middle). Notice ulcerated area in lower lip. Ulceration has healed after 6 weeks of corticosteroid
(2 months) (top, right). Continued proliferation of IH despite therapy (3 months) (bottom, left). Corticosteroid was weaned with rebound growth (age 11
months) (bottom, middle). Propranolol therapy was initiated. After 6 months of propranolol therapy (bottom, right). IH has decreased in size (18 months).

occult spinal dysraphism; magnetic resonance imaging (MRI)

may be necessary in older children. Anorectal and
genitourinary anomalies can occur with IH of the pelvis or
perineum, sometimes as part of PELVIS syndrome (Pelvic
hemangioma, External genital malformations, Lipomeningo-
cele, Vesicorenal, Imperforate anus, Skin tags).31 A macular,
network-like hemangioma of the lower extremity has been
seen with structural anomalies of the ventral-caudal region,
visceral hemangiomas, and cardiac overload.32

Radiologic Features Radiographic studies assist with

classifying indistinguishable vascular tumors. Ultrasono-
graphic features of proliferating IHs include a focal fast-flow
soft tissue mass with increased vessel density.33,34 MRI is
the second-line study and reserved for further clarification
in the face of diagnostic uncertainty to confirm the extent
and tissue characteristics of the lesion. MRI characteristics
of a proliferative-phase IH reveal a solid mass with dilated
feeding and draining vessels (Fig. 125-3). IHs are of variable
intensity on T1 and are hyperintense on T2.35 Flow voids are
seen around and within the mass.35 As the IHs involute,
vessels decrease in size and number. Involuted IHs are seen
as avascular fatty masses on MRI.36

Treatment
Observation and Local Treatment The majority of IHs
require no specific treatment other than observation. Even if
the proper decision is not to intervene, this does not mean that
FIGURE 125-3 Magnetic resonance imaging (MRI) of infantile hemangi-
oma (IH). Axial T2 weighted fat-saturation MRI sequence of a parotid IH,
which shows a well-circumscribed, uniformly hyperintense mass with
dilated feeding and draining vessels (arrow).
1616 PART IX SPECIAL AREAS
nothing should be done.2 The consultant plays a crucial role
in providing guidance and support.37 Regularly scheduled
follow-up is imperative. Serial photographs are useful to
document progression, response to treatment, and sub-
sequent regression. Disfiguring hemangiomas, particularly
when located on the face, frequently evoke strong parental
emotions of loss and grief. Deformity and severe complica-
tions occur in 10% cases.12
The most frequent complications of proliferating cutane-
ous IHs include epithelial breakdown, ulceration, bleeding,
and pain. The lips, perineum, and parotid area are especially
vulnerable to ulceration.38 Treatment of ulceration includes
daily cleansing; application of petrolatum and viscous lido-
caine may be used for pain control. Superficial ulceration
usually heals within days to weeks, whereas a deep ulceration
may take several weeks.2 Eschars should be debrided and
treated with wet-to-dry dressings to stimulate granulation
tissue. Total excision of an ulcerated IH is usually reserved
for lesions of the chest, scalp, extremity, and rarely for facial
lesions when primary closure is possible.
Pharmacotherapy Systemic pharmacologic intervention
may be necessary for endangering, ulcerating, problematic, or
life-threatening IHs. Corticosteroids inhibit the vasculogenic
potential of hemangioma-derived stem cells, as well as the
expression of vascular endothelial growth factor.39 Prednisone
or prednisolone is administered in the morning at 2 to
3 mg/kg/day for 2 weeks. Initial improvement in color and
tension is usually evident in the first 1 to 2 weeks. If the IH
stabilizes, the dosage is tapered every 2 to 4 weeks with a goal
of discontinuation by 10 to 11 months of age. The overall re-
sponse rate is 80% to 90%. In some cases the tumor may ex-
hibit rebound growth as the corticosteroid is tapered. A return
to the initial dose and slower taper will usually suffice. Possi-
ble side effects of corticosteroid include Cushingoid facies,
irritability, GI reflux, a slowing in the rate of height and
weight gain, steroid-induced cardiomyopathy, and hyperten-
sion.40–42 Nearly all children (88%) return to their pretreat-
ment curves for height and weight within 24 months.41
Live vaccines (e.g., polio, measles, mumps, rubella, varicella)
should not be administered during corticosteroid treatment.
Intralesional injection of corticosteroid for well-localized
tumors of the nasal tip, cheek, lip, or eyelid is used to mini-
mize deformity. There are reports of retinal artery occlusion
and eyelid necrosis, presumably from embolization of colloi-
dal particles.43,44 Compression at the periphery of the lesion
will help minimize this risk. Triamcinolone (25 mg/mL) is
injected slowly at a low pressure with a 3-mL syringe and
25-gauge needle. The dosage is 3 to 5 mg/kg per injection.45
Additional injections can be done at 6- to 8-week intervals;
usually, three to five are necessary.45 Response rates are similar
to that of oral corticosteroid.
Propranolol, a nonselective beta blocker, is being used with
increasing frequency for the treatment of IHs.46 Early reports
suggest that propranolol may be as efficacious as corticoste-
roids for the treatment of problematic IHs.46–48 The mecha-
nism of action for propranolol is unknown; theories include
vasoconstriction of the tumor vasculature or downregulation
of angiogenic proteins.46,48 An ongoing, prospective, random-
ized controlled trial will help elucidate propranolol’s safety,
efficacy, and tolerability for the treatment of IHs.
Recombinant interferon-alfa has fallen out of favor as
second-line therapy due to the risk of spastic diplegia, which
occurs in 5% to 20% of treated infants, particularly those
younger than 6 months of age.49,50 More suitable second-line
agents are chemotherapeutic drugs with antiangiogenic
properties such as vincristine. The drugs are administered
in a low-dose, high-frequency metronomic regimen.
Embolic Therapy Rarely, embolization is indicated for
IHs that result in congestive heart failure and do not respond
to drug therapy. Hepatic hemangiomas are the most common
lesions to require embolization. Its effectiveness is determined
by the ability to occlude a large percentage of the shunts.
Infants are maintained on pharmacotherapy even after an
apparently successful embolization.2
Laser Therapy Flashlamp pulsed-dye laser is not benefi-
cial for nascent or proliferating hemangiomas. It penetrates
only the most superficial portion of the dermis and leaves
the majority of the lesion untreated. Moreover, a superficial
IH is often the tumor that requires no treatment and involutes
without a trace. Laser treatment carries the risks of scarring,
ulceration, and hypopigmentation. In the involuting/invo-
luted phase, remnant telangiectasias can be effectively treated
with pulsed-dye laser. The use of endoscopic continuous-
wave carbon dioxide laser can control proliferating, unilateral
subglottic IHs.51
Surgical Therapy Indications for surgical resection vary
according to patient age and hemangioma stage. During
infancy (proliferating phase), excision may be necessary for
IHs causing ulceration, obstruction, and bleeding. Occasion-
ally, a well-localized or pedunculated lesion may be removed
when the scar formed would be similar to that if the excision
were to occur at a later stage. Sites most amenable to resection
during this phase include the scalp, trunk, and extremity
due to skin laxity. IHs of the forehead, eyelid, cheek, lips, nose,
nasal tip, ear, and neck require surgical expertise.
IHs of the stomach, small intestine, or colon can present
with mild or life-threatening GI bleeding during infancy.
Endoscopy or laparoscopy, or both, are sometimes necessary
to localize lesions of the small bowel.52 Most infants with
GI hemangiomas are medically managed with pharmacother-
apy and intermittent blood transfusions. Focal lesions that are
not responsive to drug therapy can be treated with endoscopic
band ligation or segmental bowel or wedge resections. More
commonly, though, GI hemangiomas present with diffuse,
patchy involvement, making endoscopic and surgical inter-
vention problematic.53 Extensive resections should not be
performed. These infants should be supported medically.
Involution and cessation of bleeding will eventually occur.
Removal of IHs during the involuting phase may be con-
sidered for large, protuberant lesions, which are likely later
to be associated with excess skin or a large fibrofatty residuum.
Indications for resection include inevitable excision, need for
staged excision, ability to hide the scar, and desire to avoid an
altered self-image.2 Waiting until the involuted phase to
determine the need for resection is reasonable barring any
endangering or life-threatening complications. Many IHs
involute, leaving normal or minimally blemished skin. Ulcer-
ation increases the likelihood of scar. Excision may be consid-
ered during late childhood (involuted phase) for damaged
skin and contour deformity.
Surgeons would agree that the scar for resection of IH
should be minimized irrespective of indication and timing.54
Lenticular excision with linear closure is the usual method
for removal of an ovoid mass. The incisional length often

extends beyond the lesional border into uninvolved skin in an
effort to avoid terminal dog-ears. Lenticular excision may be
used when a transverse incision is necessary. Convex areas
such as the cheek and forehead are susceptible to central
flattening when lenticular excision and linear closure are used.
Instead, circular excision and intradermal purse-string closure
are particularly applicable. Radial folds may be evident
initially and usually smooth out rapidly within weeks.54
Staged excisions are possible with this method.
CONGENITAL HEMANGIOMA
Unlike IHs, congenital hemangiomas are present at birth and
do not exhibit postnatal growth. Two different lesions are
recognized: rapidly involuting congenital hemangioma (RICH)
and noninvoluting congenital hemangioma (NICH). Most
lesions are solitary. There is no gender bias. These lesions
are further distinguished from IH by staining negatively for
GLUT-1.
RICHs undergo involution early, often beginning before
birth. Involution is complete between 6 and 14 months of
age.55 RICHs commonly affect the head/neck and extre-
mities.55,56 On physical examination, the lesions are raised,
firm, pink to violaceous with central depression or ulceration
and a surrounding pale rim (Fig. 125-4). Coarse telangiecta-
sias are often present.55 After involution, the residual area
often appears deflated and is not accompanied by the usual
fatty residuum of IH.56 Prenatal detection is possible because
RICHs arise in utero.55,57,58
NICHs present as bossed, round-to-ovoid shape lesions in
shades of pink to purple. The average diameter is 5 cm. There
may be overlying coarse telangiectasia.59 NICHs most com-
monly affect the head/neck (43%) followed by the limbs
(38%) and trunk (19%). As the name implies, NICHs do
not undergo involution and persist essentially unchanged.59
Although they are tumors histologically, the static behavior
of NICHs resembles that of a malformation.
RICHs and NICHs are fast-flow by Doppler evaluation.
RICHs exhibit large flow voids near the surface and have areas
that enhance inhomogenously on MRI. Arterial aneurysms
and direct arteriovenous shunts can be seen on angiography.56
NICHs and IHs are indistinguishable radiographically. The
differential diagnosis of congenital hemangioma includes
infantile fibrosarcoma.60 Biopsy must be performed when
the diagnosis is in question. The periphery of the lesion
FIGURE 125-4 Rapidly involuting congenital hemangioma (RICH). RICH of the extremity at 1 week of age (left). Involution has occurred by 2 months of
age (middle). By 1 year of life the RICH has completely involuted (right).
CHAPTER 125 VASCULAR ANOMALIES 1617
should be included in the biopsy specimen because lobular
architecture is preserved here in congenital hemangioma.56
HEPATIC HEMANGIOMA
Infantile hepatic hemangiomas (HHs) must be differentiated
from so-called “hepatic hemangiomas” that are seen in adult-
hood. Adult “hepatic hemangiomas,” also referred to as “cav-
ernous hemangiomas,” are actually venous malformations.
Conversely, hepatic hemangiomas of infancy are true vascular
tumors. HHs present in one of three recognizable patterns:
focal, multifocal, and diffuse.61 The majority of HHs are not
life-threatening and involute without long-term sequelae.
Focal HHs are likely the hepatic equivalent of the cutaneous
RICHs. Focal HHs are GLUT-1 negative. These tumors are fully
grown at birth and involute faster than multifocal or diffuse le-
sions, which are typical infantile hemangiomas. There is no
gender predilection. Cutaneous IHs are rarely present. They
can be detected on antenatal ultrasound because they develop
antenatally.62 Transient thrombocytopenia and anemia are ob-
served in some infants due to intralesional thrombosis. Con-
trary to previous beliefs, true Kasabach-Merritt phenomenon
is not observed with focal HH. Most lesions are discovered as
an abdominal mass in an otherwise healthy infant. A subset
of these focal lesions, however, results in high-output cardiac
failure due to the presence of macrovascular high-flow shunts
(arteriovenous or portovenous). These shunts can cause steal
accounting for blood-flow demands above and beyond the
hypervascular tumor parenchyma. As the tumor involutes,
the shunts usually close. The benefit of pharmacotherapy is un-
known because these lesions are biologically distinct from cu-
taneous and hepatic infantile hemangiomas and involute
rapidly. Nevertheless, medical therapy is attempted when treat-
ment is indicated because it is less invasive and more widely
available than embolization. Skilled pediatric interventional ra-
diologists can frequently successfully embolize symptomatic
shunts to improve heart failure while protecting major hepatic
vessels. Resection is almost never indicated.
Multifocal HH is a true infantile hemangioma. This lesion
was formerly known as infantile hemangioendothelioma because
of histologic similarities with kaposiform hemangioendo-
thelioma, a cutaneous tumor with a more aggressive behavior
(see later). Females are more affected than males. These
lesions are GLUT-1 positive. Most infants are asymptomatic
and require no treatment. Multifocal HHs often come to
1618 PART IX SPECIAL AREAS
clinical attention while screening for visceral hemangioma on
the basis of the presence of multiple cutaneous infantile hem-
angiomas. Occasionally, an infant can present in high-output
cardiac failure due to macrovascular shunting. Corticosteroid
is used to treat symptomatic infants in an effort to close
shunts. Propranolol use is increasing.63 Asymptomatic infants
without shunts do not require therapy. Embolization is some-
times necessary to improve heart failure when the lesions are
refractory to pharmacotherapy or if time is insufficient to
allow for clinical improvement. Infants with multifocal HHs
should be followed closely with serial abdominal ultraso-
nography until involution of their hemangiomas.
Diffuse HHs are the most feared. These lesions are also true
IHs. There is a female gender bias. The normal liver paren-
chyma is almost completely replaced by innumerable com-
pact, nodular tumors. The massive hepatomegaly can
compress the inferior vena cava and thoracic cavity leading
to abdominal compartment syndrome, respiratory distress,
and multiorgan failure. All IHs express type 3 iodothyronine
deiodinase, which converts thyroid hormone to its inactive
forms, resulting in an acquired hypothyroidism.20 This is
most prevalent in diffuse HHs but is sometimes seen with mul-
tifocal HHs. Levels of type 3 iodothyronine deiodinase are
proportional to tumor burden, placing infants with diffuse
HHs (and occasionally larger multifocal HHs) at risk for
hypothyroidism. Exogenous thyroid hormone replacement,
often in large quantities, is necessary to prevent mental retar-
dation and cardiac failure.20 Serum TSH should be checked in
all patients with larger multifocal and diffuse HHs. Hypothy-
roidism improves with tumor involution. Corticosteroid and
increasingly propranolol are given for diffuse HHs to acceler-
ate tumor involution. Hepatic transplantation may occasion-
ally be indicated in some infants in extremis without time
to wait for a response to pharmacologic therapy.61 Other ex-
traordinary maneuvers in the presence of respiratory distress
and abdominal compartment syndrome include temporary
abdominal enlargement with Silastic sheets or silos and
low-dose vincristine treatment.64,65
HHs exhibit unique radiographic patterns, which aid in
diagnosis. On computed tomography (CT), focal HHs are het-
erogeneous. Enhancement is usually centripetal with central
sparing due to thrombosis or necrosis. Calcifications can
also be evident and increase as the tumor involutes.66
MRI of focal HHs shows a well-defined, solitary, spherical
FIGURE 125-5 Axial magnetic resonance images of hepatic hemangioma. T1 postcontrast appearance of a focal hepatic hemangioma (left), which
demonstrates peripheral enhancement (arrows) with sparing centrally due to thrombosis or scar. T2 images of multifocal hepatic hemangioma (middle).
Lesions are hyperintense relative to liver. T2 sequences of diffuse hepatic hemangioma show near-total hepatic parenchymal replacement (right).
tumor (Fig. 125-5). It is hypointense relative to liver on T1
and hyperintense on T2. Centripetal enhancement is seen
on gadolinium sequences. The solid, nonthrombosed portions
at the periphery demonstrate intense homogeneous enhance-
ment (see Fig. 125-5). Centrally there is heterogeneous
enhancement due to necrosis, thrombosis, or intralesional
hemorrhage.61
Multifocal and diffuse HHs on CT appear as multiple well-
defined, spherical lesions. Multifocal HHs will have interven-
ing normal hepatic parenchyma, whereas diffuse lesions
nearly totally replace the liver. Without contrast the lesions
are hypodense relative to the liver. They enhance centripetally
with contrast. In comparison with focal lesions, there is no
thrombosis, necrosis, or central sparing. The supraceliac aorta
is usually dilated and tapers distally. Enlarged hepatic arteries
and veins may indicate shunting.66 Flow-voids can be seen
in and around the lesions. On MRI, the lesions enhance
homogenously and are hypointense relative to liver on T1
and hyperintense on T2 (see Fig. 125-5).66
Close follow-up is mandatory, especially in the first few
months of life, because clinical deterioration is possible.
The differential diagnosis of HH is broad and includes arterio-
venous malformation, arterioportal fistula, mesenchymal
hamartoma, hepatoblastoma, angiosarcoma, and metastatic
neuroblastoma. Percutaneous needle or open biopsy should
not be delayed in the face of diagnostic uncertainty.
PYOGENIC GRANULOMA
Pyogenic granuloma is a benign, acquired vascular lesion of
the skin and mucous membranes seen in children and preg-
nant women.67 Pyogenic granulomas rarely occur before
6 months of age. The etiology is unknown. Most lesions are
not associated with trauma or dermatologic conditions.67
They develop as small erythematous papules that enlarge to
5 mm to 10 mm, mostly seen in the head and neck area
(Fig. 125-6). The lesions bleed easily and are often peduncu-
lated.68 Treatment options include curettage, full-thickness
excision with linear closure, shave excision and cautery,
cautery alone, and laser phototherapy.69 Silver nitrate can
be used for small pyogenic granulomas. Larger lesions may
require liquid nitrogen. Recurrence is high (45%) when
pyogenic granuloma is not completely excised or ablated.67

FIGURE 125-6 Pyogenic granuloma.
KAPOSIFORM HEMANGIOENDOTHELIOMA
AND KASABACH-MERRITT PHENOMENON
Sixty years after the initial report by Kasabach and Merritt, we
have learned that the findings of an enlarging vascular lesion
and thrombocytopenia are never associated with infantile or
congenital hemangioma.70 Additional features of Kasabach-
Merritt phenomenon include microangiopathic hemolytic ane-
mia and a mild consumptive coagulopathy.71 Kasabach-Merritt
phenomenon occurs with more aggressive and invasive vascu-
lar tumors such as kaposiform hemangioendothelioma (KHE),
kaposiform lymphatic anomaly (KLA), and tufted angioma
(TA).72–74 These tumors are biologically distinct from infantile
hemangioma.75 The term “Kasabach-Merritt phenomenon” is
often used mistakenly to describe transient moderate neonatal
thrombocytopenia seen with focal hepatic hemangioma
(RICH), as well as a localized intravascular coagulopathy seen
with some extensive venous and lymphatic malformations.
KHE and TA are typically present at birth, although
they may also appear postnatally. There is an equal gender
distribution. The tumors are unifocal and mostly located on
the trunk and extremities. KHE can occur in the retroperito-
neum. TA appears as erythematous macules; Kasabach-Merritt
FIGURE 125-7 Kaposiform hemangioendothelioma (KHE) with Kasabach-Merritt phenomenon (KMP). KHE of the chest at 2 years of age (left). Appearance
at 6 years of age after treatment with vincristine and resolution of KMP (middle). Tumor regression has resulted in joint contracture. Further involution at
8 years of age (right).
CHAPTER 125 VASCULAR ANOMALIES 1619
phenomenon may also be evident. KHE is more extensive than
tufted angioma and can rapidly expand. KHE continues to
proliferate into early childhood followed by incomplete
regression in midchildhood. The superficial component of
KHE appears in the skin and causes a red-purple discoloration
with surrounding ecchymosis (Fig. 125-7). Generalized
petechiae may be apparent due to profound thrombocytope-
nia (<10,000 platelets/mL). Infants with Kasabach-Merritt
phenomenon are at risk for intracranial, pleural, pulmo-
nary, peritoneal, and GI hemorrhage. Thrombocytopenia is
unresponsive to platelet transfusion due to intralesional
trapping. The prothrombin time and activated partial throm-
boplastin time are usually normal to mildly elevated. D-dimer
is increased while fibrinogen is low. KHEs less than 8 cm in
size are less likely to have Kasabach-Merritt phenomenon.76
Histopathology of KHE reveals aggressive infiltration of nor-
mal tissues by sheets or lobules of oval or spindled endothelial
cells, dilated lymphatic channels, and slitlike vascular spaces
filled with hemosiderin and compacted erythrocytes suggesting
stasis.74 When lymphatic channels predominate, the condition
is called KLA. The microscopic appearance of TA shows tufts
of capillaries in the middle to deep dermis.77
On MRI, KHE demonstrates enhanced signal on T2. Tumor
margins are not well defined. The tumor does not respect
tissue planes as it passes through skin down to muscle. There
is stranding in the subcutaneous fat due to lymphatic obstruc-
tion or invasion. Feeding and draining vessels are small
relative to tumor size. Osteolysis may be evident when KHE
is adjacent to bone.74 Imaging patterns are similar for TAs.
The treatment for KHE and TA with Kasabach-Merritt phe-
nomenon is primarily medical. Due to its size and permeation
of multiple tissue planes, KHE can rarely be resected safely.
Drug options include corticosteroid, vincristine, or interferon.
No treatment is uniformly effective. Vincristine has been asso-
ciated with improved platelet counts.71 Platelet transfusions
should not be given unless there is active bleeding. Heparin
should be avoided because it can stimulate tumor growth,
exacerbating platelet trapping and thrombocytopenia. For
KHE not associated with Kasabach-Merritt phenomenon,
the decision to treat should be based on size, location, treat-
ment side effects, and long-term complications (such as joint
1620 PART IX SPECIAL AREAS

contracture or myofascial pain syndromes).76,78 All infants vascular channels. Angiogenesis is the formation of new blood
should be followed closely because mortality is high (20% vessels from preexisting vessels by budding or branching.
to 30%). Previously quiescent tumors can be associated with Around the third week of development, embryonic blood ves-
recurrent Kasabach-Merritt phenomenon. sels begin developing when mesodermally derived hemangio-
blasts aggregate to form blood islands. The inner cells of the
OTHER RARE AND MALIGNANT VASCULAR blood islands become hematopoietic stem cells. The outer
TUMORS layer differentiates into angioblasts, endothelial cell precur-
sors. Proliferating angioblasts form a capillary-like network,
Infants with cutaneovisceral angiomatosis with thrombocyto- which constitutes the primary vascular plexus. Angiogenesis
penia present with multiple red to brown cutaneous lesions in occurs as this plexus is reorganized into a functional vascular
one or multiple sites. The skin lesions are present at birth and system and new capillaries begin sprouting.21,84
may be mistaken for infantile hemangioma. Infants may pre- The predisposition of endothelial precursors to differenti-
sent with GI bleeding or hemoptysis. GI endoscopy demon- ate into distinct channel types is imprinted early in embryo-
strates multiple, flat, red lesions. The lung parenchyma may genesis by the detection of unique cell-surface markers.85
also be involved. The lesions proliferate throughout early in- Arterial endothelial cells express ephrin-B2, and venous endo-
fancy. Histology shows variable endothelial hyperplasia. Due thelial cells express Eph-B4.86 The recruitment of perien-
to the rarity of cutaneovisceral angiomatosis, optimal treat- dothelial cells stabilizes the vessel wall and allows for
ment has not been delineated. It is postulated that angiogenic deposition of the extracellular matrix and basement mem-
inhibitors may be of benefit.79 brane. Vascular endothelial growth factor, platelet-derived
Angiosarcoma is a rarely encountered aggressive malig- growth factor-b, angiopoietins and their receptors, and trans-
nancy in children. It is characterized by rapidly proliferating, forming growth factor-b1 all play roles in vessel assembly and
infiltrating anaplastic cells derived from blood vessels. Angio- maturation.87
sarcomas occur in soft tissue, bones, and viscera (liver, spleen, The lymphatic system begins its formation near the end of
and heart). More common locations in the pediatric age group the sixth week, only after the establishment of functional
include the head, neck, and mediastinum.80 Environmental blood vessels.88,89 Lymphatics most likely arise from
exposure to arsenic and vinyl chloride can increase the risk veins.88–91 The earliest indication of the future lymphatic
of angiosarcoma. The most efficacious treatment is resection. system is the appearance of lymphatic vessel endothelial
Curative resections are rare given that many angiosarcomas hyaluronan receptor (LYVE-1) in a number of the endothelial
are not limited to one anatomic area. They have a high rate cells lining the anterior cardinal vein.88,92 A subpopulation of
of local recurrence and metastasis. these LYVE-1 venous endothelial cells begin to express the
Kaposi sarcoma is a low-grade vascular neoplasm mediated transcription factor prospero-related homeobox 1 (PROX-1)
by the human herpesvirus-8. The endemic/African clinical on one side of the anterior cardinal vein.91 As development
subtype is seen in the pediatric population. Unlike adults, continues, the number of PROX-1 cells increases and they ul-
children rarely present with cutaneous findings. Generalized timately migrate from the veins to establish primitive lymph
lymphadenopathy may be their only presenting symptom. sacs.91 PROX-1 acts as a master regulator of lymphatic endothe-
Definitive diagnosis requires histologic examination.81 lial differentiation.93 Lymph sacs are formed close to major
veins throughout the embryo. Lymphatic endothelial cells
sprout from the sacs to form the peripheral lymphatic
Vascular Malformations
------------------------------------------------------------------------------------------------------------------------------------------------
network.88,89 Lymphatic endothelial cells express vascular
endothelial growth factor receptor-3, which via interaction
Vascular malformations are localized or diffuse errors of develop- with its ligand, vascular endothelial growth factor-C, promotes
ment that may affect any segment of the vascular tree including and guides the budding of lymphatic endothelial cells.91,94
arterial, venous, capillary, and lymphatic vessels. They are
named on the basis of the predominant channel type and flow
CAPILLARY MALFORMATIONS
characteristics (Table 125-1). Slow-flow anomalies include cap-
illary, lymphatic, and venous malformations; fast-flow lesions Capillary malformation (CM) is the proper name for “port-
include arteriovenous malformations and arteriovenous fistulae. wine stain.” Nevus flammeus neonatorum, also known as “angel
Vascular malformations, unlike vascular tumors, which expand kiss” (on the forehead) and “stork bite” (on the nuchal area),
postnatally, exhibit growth commensurate with that of the child. are faint macular stains seen in 50% of Caucasian neonates.
The prevalence of congenital vascular malformations is 1.2% They are due to transient dilation of dermal vessels. These le-
to 1.5%.21 Most vascular malformations are sporadic. Some sions predictably fade, whereas a CM does not. CMs affect
inherited forms have been observed, usually in autosomal- 0.3% of infants with an equal sex distribution.95 Most CMs
dominant pattern. Multiple lesions are frequently observed in are sporadic, but some are inherited in an autosomal domi-
the familial forms.82,83 Unfortunately, vascular malformations nant pattern.96 They are present at birth and appear as flat,
tend to enlarge during puberty and recur after treatment. pink-red, cutaneous patches. They can be localized or exten-
sive and can be found anywhere on the body (Fig. 125-8).
EMBRYOLOGY AND DEVELOPMENT OF THE CMs are composed of dilated, ectatic capillary-to-venule-sized
VASCULAR AND LYMPHATIC SYSTEMS vessels in the superficial dermis. Histopathology demonstrates
a paucity of nerves surrounding these vessels.97 With age,
The development of the embryonic vascular system involves the vessels gradually dilate due to lack of innervation, explain-
two separate but intimately connected processes: vasculogen- ing the observed darkening and nodular expansion.98 Asso-
esis and angiogenesis. Vasculogenesis is the formation of new ciated hypertrophy of the subcutaneous tissue, muscle, and

FIGURE 125-8 Diffuse capillary malformation with overgrowth.
Posterior view shows the capillary malformation involving the right arm,
trunk, and left leg. Soft tissue overgrowth of the left leg is apparent.
bone underlying a CM is common, and when located on an
extremity a limb length discrepancy may be apparent.
Occasionally, the presence of a CM signals an underlying
structural abnormality. A CM overlying the cervical or lumbar
spine may be associated with an occult spinal dysraphism or
tethered cord. Encephalocele or ectopic meninges may be found
beneath an occipital CM. Children with CMs in the distribution
of the ophthalmic and maxillary branches of the trigeminal der-
matomes should be evaluated for Sturge-Weber syndrome. This
syndrome consists of a facial CM with ipsilateral ocular and
leptomeningeal vascular anomalies. Extensive leptomeningeal
involvement can manifest as seizures, contralateral hemiplegia,
and variable motor and cognitive delays. CMs are also seen in
complex-combined vascular malformations.
Treatment for CM is indicated primarily for cosmetic pur-
poses. Vascular-selective pulsed dye lasers cause photother-
molysis of the CM and will improve the appearance by
lightening the color of the lesion in 70% of patients.99 Early,
thin CMs such as in the neonate respond well especially when
located on the lateral aspect of the face. Complete eradication
of the CM is difficult, and despite multiple laser sessions some
lesions continue to darken.100 The timing of therapy remains
controversial. Initiating treatment before 6 months of age is an
option and short-term follow-up appears promising.101
Surgical intervention may be required for associated soft tissue
hypertrophy and limb length discrepancy.
Cutis Marmorata Telangiectatica Congenita
Cutis marmorata telangiectatica congenita (CMTC) is a rare
vascular anomaly.102 The involved skin is bluish to deep pur-
ple in color with a characteristic reticular vascular pattern.
CHAPTER 125 VASCULAR ANOMALIES 1621
Cutaneous marbling is apparent at room temperature and
becomes more pronounced with lower temperatures
and crying. The affected skin is depressed and can ulcerate
and bleed. CMTC occurs in a localized, segmental, or gener-
alized distribution. It is more frequent on the trunk and lower
extremities.103,104 Histopathology reveals dilated capillaries in
the papillary dermis and proliferation of blood vessels in the
reticular dermis with occasional dilated veins and venous
lakes.105 It is sporadic and has no gender bias. The etiology
of CMTC is unknown.
The clinical course of CMTC is benign with partial regres-
sion of the capillary stains in the first year of life, which
continues into adolescence. Atrophy, pigmentation, and
prominent veins often persist into adulthood. CMTC has been
“associated” with abnormalities of many kinds including
neurologic, skeletal, and cardiovascular.106 Stenosis of the
common iliac and femoral arteries with resultant claudication
has been observed in children with CMTC of the lower
extremity.106,107
Telangiectasia
Tiny acquired capillary vascular marks, commonly referred to
as spider nevi or spider telangiectasias, can appear on children
in the preschool and school-aged years. There is no gender
bias, and epidemiologic studies suggest that they may be pre-
sent in nearly half of all children.21 Spontaneous resolution is
possible, but pulsed-dye laser can eradicate the lesion.
Hereditary hemorrhagic telangiectasia (Osler-Weber-
Rendu syndrome) is an autosomal dominant disorder
that commonly results from mutations in endoglin, activin
receptor-like kinase 1, or rarely Smad4. These genes encode
proteins that modulate transforming growth factor-b signaling
in vascular endothelial cells.108 Mutations lead to the deve-
lopment of fragile telangiectatic vessels and arteriovenous
malformations (AVMs). There is marked intrafamilial pheno-
typic variation. A definite diagnosis of hereditary hemorrhagic
telangiectasia is made in the presence of at least three separate
manifestations: spontaneous recurrent nosebleeds; mucocu-
taneous telangiectasia (frequently located on the fingertip
pulps, lips, oral mucosa, or tongue); visceral involvement
(GI tract, pulmonary, hepatic, cerebral, or spinal AVMs);
and family history.109 Presentation with spontaneous recur-
rent nosebleeds before school age is most common. Approx-
imately one third of patients have chronic anemia due to lower
GI bleeding, which usually becomes apparent in adult-
hood.108 In infancy and early childhood, hepatic AVMs of
hereditary hemorrhagic telangiectasia can be mistaken for
hepatic hemangiomas.
LYMPHATIC MALFORMATION
Lymphatic malformations (LMs) are frequently called “lym-
phangiomas,” erroneously suggesting a proliferative tendency.
LMs are not tumors. Their pathogenesis is unknown. LMs of
the neck and axilla may be due to failure of lymphatic sacs to
communicate with the central venous system. Sequestration of
lymphatic buds may account for peripheral lesions. Typical
LMs seem to be sporadic. Heritable forms of lymphedema,
also a type of LM, have been identified.82
LMs are classified as microcystic, macrocystic, or combined
macrocystic and microcystic lesions (Table 125-1). The dif-
ference is clinically and therapeutically useful. Size is
1622 PART IX SPECIAL AREAS
distinguished by whether or not the cystic cavity can be suc-
cessfully aspirated, resulting in visible decompression. LMs
may be focal, discrete masses or infiltrate multiple anatomic
locations.
LMs are usually apparent at birth and tend to occur in areas
of major lymphatic channels, especially the cervical and axil-
lary locations (Fig. 125-9). They can be found in all tissues or
organ systems with the exception of the central nervous sys-
tem. LMs most commonly appear as ballottable masses with
normal overlying skin, although a blue hue may result if large
underlying cysts are present. Dermal involvement can appear
as puckering or deep cutaneous dimpling. LMs in the subcutis
or submucosa appear as tiny clear to white vesicles. Intra-
vesicular bleeding is evidenced by dark red, dome-shaped
nodules. Histologically, LMs appear as thin-walled vascular
channels lined by lymphatic endothelial cells, which are
immunopositive for podoplanin (D2-40) and LYVE-1.110
The lumens may be empty or filled with a proteinaceous fluid
containing macrophages and lymphocytes.111
Increasingly, mass lesions and vascular anomalies are
detected on antenatal ultrasound.57,112 The term cystic
hygroma is used in perinatology to describe a lymphatic anom-
aly distinct from LMs seen in postnatal pediatrics. A fetal cystic
hygroma consists of collections of lymphatic fluid due to a
maldevelopment of the cervical lymphatic sacs. Nuchal thick-
ness greater than or equal to 2.5 mm, measured between
10 and 14 weeks gestation, is considered abnormal. Some
authors reserve the term cystic hygroma for only those with
visible septations and refer to the rest as simple increased nuchal
translucency.112 A fetal cystic hygroma ranges from 6.5 to
7.9 mm.112,113 The finding of a first-trimester septated cystic
hygroma has definitively been shown to be associated with a
higher risk of aneuploidy, cardiac malformation, and perinatal
death over simple increased nuchal translucency.112 Chromo-
somal abnormalities such as Down syndrome or Turner syn-
drome are observed in 50% of fetuses with cystic hygroma;
however, 17% to 25% of these pregnancies will result in the
birth of a healthy newborn.112,113 Resolution of a fetal cystic
FIGURE 125-9 Cervical lymphatic malformation (LM). Prenatal magnetic resonance image shows a fetus with a right cervical macrocystic LM (left,
arrowhead). Postnatal appearance of the same infant at 1 week of age (right).
hygroma improves the chance of a normal outcome to as high
as 95%.112
Fetal cystic hygroma and simple increased nuchal translu-
cencies are always situated in the posterior neck and can ex-
tend down the entire dorsum of the fetus.112 This posterior
location is distinct from cervical LMs located in the subman-
dibular, supraclavicular, and anterior/lateral neck regions (see
Fig. 125-9). LMs are usually not detected until the second or
third trimesters, unlike the more concerning fetal cystic
hygroma, which is seen by 14 weeks.57 Unfortunately, because
of continued use of the imprecise term cystic hygroma, some
families receive misinformation regarding their true diagnosis
and conclude incorrectly that their fetus cannot survive or will
have chromosomal abnormalities. Antenatal consultation with
pediatric specialists knowledgeable about vascular malfor-
mations can benefit families. Reassurance can be given that
a lateral or anterior LM is not associated with chromosomal
or developmental disorders. It also provides an educational
opportunity to discuss the concerns facing a child with
a cervicofacial LM.
Large cervicofacial LMs detected in utero should raise con-
cern about potential airway obstruction leading to asphyxia in
postnatal life. Results from prenatal ultrasound and fetal MRI
allow for assessment of potential airway difficulties at birth.
When the risk of airway obstruction is high, consideration
should be given to delivery of the fetus via an ex-utero intra-
partum treatment (EXIT) procedure.114 Although the EXIT
procedure had early popular appeal, experience has shown
that most LMs are compressible, allowing for intubation after
delivery in the face of airway compromise. Differentiating LM
from teratoma, which is usually firm and noncompressible,
is an important distinction in delivery planning.
Fetal pleural, pulmonary, and peritoneal lymphangiectasia
may also present in utero as pleural effusions or ascites. These
may be isolated or may be a manifestation of Turner, Noonan,
or other syndromes. Secondary hydrops fetalis may develop
secondary to large pleural effusions. Diagnosis is confirmed
when a tap shows more than 80% lymphocytes on the

differential cell count. After chromosomal studies, fetal inter-
vention, when indicated, consists of repeated taps or shunting
into the amniotic fluid.115–117
The morbidity of LMs is anticipated by anatomic location
and extent. Proptosis can result from periorbital and orbital
lesions. Facial LMs can be associated with skeletal overgrowth,
macroglossia, and macrocheilia.10 Tongue and oral floor
lesions can cause chronic airway problems, recurrent infec-
tion, and functional issues related to speech, oral hygiene,
and malocclusion.118 Cervical or axillary LMs can signal an
associated mediastinal LM. Chylous pericardial and pleural
effusions can be associated with diffuse thoracic lymphatic
anomalies or rare abnormalities of the thoracic duct or cisterna
chyli. GI LMs can cause chronic protein-losing enteropathy
resulting in profound hypoalbuminemia. Pelvic lesions can
cause recurrent infections, constipation, and bladder outlet
obstruction. LMs in the extremity may result in overgrowth
and limb length discrepancy. Gorham-Stout syndrome is a
rare, potentially fatal form of skeletal and soft tissue LM.
It is characterized by the gradual and often complete resorp-
tion of one or multiple skeletal elements (hence the synonym
“vanishing bone disease”), which may result in pathologic
fractures.111
Lymphedema, also a type of LM, is characterized by local-
ized accumulation of protein-rich fluid in the superficial,
interstitial space causing enlargement of the subcutaneous
tissue; over time, deposition of adipose and fibrous tissue fur-
ther increases limb volume.119 It may be primary (idiopathic)
or secondary (acquired). The vast majority of patients with
primary lymphedema have no family history, but a genetic ba-
sis has been found in some families.120 Milroy disease, a con-
genital form of lymphedema, is an autosomal dominant
condition linked to chromosome 5q35.3, which codes for a
mutation in the vascular endothelial growth factor receptor
3 gene.121,122 Sporadic forms also occur. Greater than 90%
of patients with Milroy disease present with lymphedema
below the knees.83 Intrauterine pleural effusion and hydrops
fetalis have been observed.123,124 Skin biopsy from the swol-
len feet of patients with Milroy disease shows abundant skin
lymphatics suggesting malfunction.125 Meige disease typically
presents in puberty. It is an autosomal dominant disorder with
FIGURE 125-10 Magnetic resonance image of cervicofacial lymphatic malformation (LM). Coronal T1 image of an extensive cervicofacial LM (left), which
shows contrast enhancement of septations and rims of cysts (arrowhead). Coronal T2 image (middle), which demonstrates macrocysts (arrow) and micro-
cysts (arrowhead). Macrocystic component with a fluid-fluid level (right, arrowheads).
CHAPTER 125 VASCULAR ANOMALIES 1623
variable penetrance and phenotype. Families with lymphedema
distichiasis (double row of eyelashes) have been found to have a
mutation in transcription factor FOXC2, which plays a role in
somite development.126–128 Systemic involvement such as
intestinal lymphangiectasia, ascites, pleural or pericardial effu-
sions, and pulmonary lymphangiectasia have been observed
in patients with primary extremity lymphedema.125
Ultrasonography is useful to characterize superficial and
well-localized LMs. MRI allows for determination of extent
and type (microcystic versus macrocystic) (Fig. 125-10).
LMs are hyperintense on T2-weighted and turbo-STIR im-
ages.129 Fluid-fluid levels are often seen in macrocystic LMs
as a result of layering of blood, protein, or both. Cyst walls
and septae enhance with contrast.36 Microcystic LMs show
an intermediate signal on T1 and intermediate to high signal
on T2 spin echo sequences.129 Conventional contrast lymph-
angiography can be useful for characterization of lymphatic
anomalies of the thoracic duct and chylous effusions, allowing
for localization of abnormal channels or sites of leakage.130
It has largely been abandoned in small children due to
technical difficulties and morbidity.131 Lymphoscintigraphy
is minimally invasive and safe in infants and children.132
The indications for treatment of LMs vary with extent and
location. The most common complications requiring treat-
ment include bleeding and infection. Intralesional bleeding,
either spontaneous or due to local trauma, can cause expan-
sion. Previously imperceptible lesions may suddenly appear.
On physical examination, the LM may be firm and ecchy-
motic. Treatment is focused on pain control. Intralesional
bleeding may transform macrocystic lesions into microcysts.
LMs, unlike other vascular anomalies, may swell in re-
sponse to systemic viral or bacterial infection. Usually, there
are no consequences of this passive engorgement. If accompa-
nied by the rapid onset of localized swelling, tenseness,
erythema, and systemic signs of infection, antibiotics (oral
or intravenous depending on severity) should be initiated.
Frank septic shock may be the presenting symptom in patients
with occult retroperitoneal and mesenteric LMs. Bacterial
cellulitis is a serious complication of cervicofacial LM due to
the risk of airway obstruction, possibly requiring emergent
intubation.
1624 PART IX SPECIAL AREAS
LMs may be treated with sclerotherapy, surgical resection,
or both. Sclerotherapy works well for macrocystic LMs and in
many cases may render lesions undetectable. Common scler-
osants include pure ethanol, sodium tetradecyl sulfate, and
doxycycline. Technically, sclerotherapy is usually straightfor-
ward. The cystic cavity is entered by direct puncture, and
the fluid is aspirated. The needle is maintained in situ, and
the sclerosant is injected. Multiple punctures, aspirations,
and injections are often performed. Ultrasound guidance is
useful. Staged sessions may be necessary for larger lesions.
Sclerotherapy preceding resection can shrink LMs, allowing
for less morbid surgical procedures. LMs may reexpand after
treatment, requiring additional procedures. Microcystic LM is
less responsive to sclerotherapy. Superficial lymphatic vesicles
can be treated with local intravesicular injection to improve
leakage. Complications of sclerotherapy to be avoided include
injury to adjacent nerves, necrosis of overlying skin, and
cardiotoxicity related to overdose.
The only potential for a definitive cure of LM is surgical
resection. Operations near vital structures can be long and
tedious. Serial, staged excisions are often necessary, while en-
suring complete resection in a given anatomic area as much as
possible, because repeat excision of the same area is techni-
cally difficult and increases morbidity.2 Dissection in previ-
ously sclerosed areas can be challenging because of fibrosis
and scarring, although some authors have found excision after
sclerotherapy to be easier.133 General principles regarding
surgical procedures include (1) limiting resection to a defined
anatomic region; (2) minimizing blood loss and planning for
replacement; (3) performing closed suction drainage of the
resection cavity. The “recurrence” rate after macroscopically
complete resection ranges from 17% to 40%.134 The perceived
recurrence is thought to be due to growth and expansion of
affected lymphatics in the remaining, seemingly normal, yet
involved tissue. Sometimes cutaneous “warty” lymphatic
vesicles develop in the scar. These can be treated easily with
intravesicular sclerotherapy, cauterization, laser coagulation,
or excision.
FIGURE 125-11 Venous malformation (VM). VM of the perineum (left). Truncal VM (middle). Coronal fast spin-echo inversion recovery magnetic reso-
nance image of a VM of the knee (right). There is diffuse involvement of the quadriceps muscle. A phlebolith is present (arrowhead).
VENOUS MALFORMATION
Venous malformations (VMs), often incorrectly referred to as
cavernous hemangiomas, are slow-flow lesions. They may be
seen at birth or become apparent later depending on location.
VMs are most common in the skin and soft tissues but can be
located anywhere in the body (Fig. 125-11). They present in a
heterogeneous fashion including simple varicosities and ecta-
sias, discrete spongy masses, or a diffuse network of complex
channels that permeate an organ or tissue. VMs grow propor-
tionately with the child and tend to expand slowly with time.
Some VMs will dilate when placed in a dependent position or
during a Valsalva maneuver. Phlebothrombosis is common
and may be painful. Phleboliths can also be easily palpated
in many VMs. Histologically, a VM is composed of thin-walled,
dilated abnormal channels with clumping of vascular smooth
muscle cells. There is often evidence of clot formation and
fibrovascular ingrowth.
Most VMs are sporadic (90%). Somatic mutations in the
tyrosine kinase receptor TIE2 have been discovered in approx-
imately 50% of sporadic VMs.135 An autosomal dominant
inheritance pattern occurs in cutaneomucosal venous mal-
formation, which accounts for only 1% to 2% of lesions. Mu-
tations in the TEK gene, which encodes TIE2, have been
identified.136,137 Cutaneomucosal VMs are dome shaped
and range in size from tiny to several centimeters. Glomuve-
nous malformations account for 5% of lesions and are also
inherited in an autosomal dominant fashion.83 The lesions
are mostly superficial and occur as multiple blue to deep-
purple nodules or confluent, cobblestone-appearing plaques
frequently on the trunk or extremities.45 They are caused
by loss-of-function mutations in glomulin, which derails
vascular smooth muscle cell differentiation.138
Complications of VMs are related to their location. Cervi-
cofacial VMs are often unilateral. They may distort facial fea-
tures and cause exophthalmia, dental malalignment, and
obstructive sleep apnea. Limb length discrepancy can be seen
in an affected limb. VMs involving the synovial lining of the

knee can cause episodic pain due to repeated bloody effu-
sions. Hemarthrosis may lead to degenerative arthritis.2
VMs of the GI tract can be solitary or multifocal. They can
involve any or all layers of the bowel wall and can be small or
massive. The majority of GI VMs occur as transmural lesions
of the left colon and rectum with variable local extension into
pelvic structures.2,53,139 GI bleeding may be the initial symp-
tom. Discoloration in the perineum may indicate an underly-
ing rectal VM. Mesenteric and portal venous anomalies may be
associated with GI VMs. A rectal VM associated with ectasia of
mesenteric veins is a risk factor for portomesenteric venous
thrombosis.140
Blue-rubber bleb nevus syndrome (BRBNS) represents a
rare disorder consisting of multifocal VMs that affect the skin
and GI tract primarily. This should not be confused with be-
nign or malignant blue nevi. The cutaneous lesions of BRBNS
are VMs, not true nevi; they are often quite numerous
(Fig. 125-12). They range in color from blue to purple and
are commonly found on the palms and soles of the feet.141
There is often a large dominant VM. Diagnosis of GI VMs is
generally based on endoscopic findings. As with other GI
VMs, chronic bleeding and anemia can result. Intussusception
may also occur.142
VMs with phleboliths and those that are large (10 cm2)
and/or deep are more likely to cause a localized intravascular
coagulopathy.143 Constant activation of coagulation caused by
stasis and stagnation of blood within the malformation leads
to the consumption of coagulation factors. Thrombin is pro-
duced and converts fibrinogen into fibrin. Fibrinolysis ele-
vates fibrin degradation products and can be measured by
obtaining D-dimer levels.144 Occasionally, fibrinogen levels
are also low.143,145 Coagulation studies (prothrombin time,
activated partial thromboplastin time, and D-dimer and fibrin-
ogen levels) should be reviewed before invasive procedures in
patients with extensive VMs. Low-molecular-weight heparin
can be given during the perioperative period to improve the
hematologic status for patients with elevated D-dimer levels.
Platelet counts can be slightly decreased in the 100,000 to
150,000/mm3 range (unlike the marked thrombocytopenia
seen with Kasabach-Merritt phenomenon).144
Imaging modalities useful for the diagnosis of VM include
ultrasonography, MRI, and venography. MRI is the most
informative. Lesions are hyperintense with T2 sequences
FIGURE 125-12 Blue rubber bleb nevus syndrome. Characteristic dome-shaped plantar lesions (left). Serosal view of venous malformations (VMs) in the
small intestine (middle). Endoluminal appearance of a VM in the colon (right).
CHAPTER 125 VASCULAR ANOMALIES 1625
(see Fig. 125-11). Identifiable spaces separated by septations
can be seen. The vascular spaces of VM enhance with contrast,
whereas LM does not; the exception to this rule is an LM with
intralesional bleeding. Phleboliths are often seen as signal
voids on all sequences. High-flow vessels are not seen within
or around the lesions. Magnetic resonance venography helps
delineate deep venous anatomy, especially of the extremity.36
Doppler ultrasonography reveals soft, compressible masses
with monophasic low velocity.45
The indications for treatment of VMs include pain, func-
tional loss, bleeding, and appearance. No known systemic
pharmacologic agents induce involution of vascular malfor-
mations, and therefore they are not used for VMs. Compres-
sion therapy aims to reduce swelling and pain in diffuse VMs
of the extremities. Sclerotherapy can be effective for symptom-
atic VMs. Sclerosing agents cause direct endothelial damage,
thrombosis, and scarring. Large VMs are accessed by direct
puncture under fluoroscopy. Compression and tourniquets
can be used to limit venous drainage and minimize systemic
delivery of the sclerosant. Staged sessions and occasional em-
bolization of large draining channels are sometimes necessary.
Possible local complications are blistering, full-thickness cuta-
neous necrosis, and nerve injury.146 Systemic complications
include hemolysis, sudden pulmonary hypertension, and car-
diac and renal toxicities. Unfortunately, many VMs recanalize
and expand after treatment, thus requiring additional pro-
cedures.147 Focal VMs can often be successfully excised.
Sclerotherapy before resection may allow removal of larger
lesions.146 Staged subtotal resections may be necessary. Trans-
fusion dependent anemia due to GI VMs should be evaluated
for surgical resection. Multifocal VMs of BRBNS can be
removed via wedge excision and polypectomy by intussuscep-
tion of successive lengths of intestine. Intraoperative entero-
scopy aids to identify location of all lesions. This technique
provides the only chance for cure.141 Diffuse colorectal VMs
causing significant bleeding may be treated with colectomy,
anorectal mucosectomy, and coloanal pull-through.148
ARTERIOVENOUS MALFORMATION
Arteriovenous malformations (AVMs) are fast-flow malfor-
mations characterized by abnormal collections of arteries
and veins that directly communicate (shunts), thus bypassing
1626 PART IX SPECIAL AREAS
the high-resistance capillary bed. The shunts comprise the
epicenter of the AVM, called the nidus. Intracranial AVMs
are more common than extracranial AVMs. Areas affected
by extracranial AVMs in decreasing frequency are the head
and neck, limbs, trunk, and viscera. AVMs may be apparent
at birth but are often misdiagnosed initially as a capillary mal-
formation or infantile hemangioma due to pink staining in the
overlying skin (Fig. 125-13). The diagnosis becomes more ob-
vious later, particularly during adolescence, when they are
more likely to expand.149 On palpation, the area may be warm
with an underlying palpable thrill. Manual compression of
feeding arteries and draining veins results in a slowing of
the heart rate known as Nicoladoni sign.150 The natural history
of AVMs can be documented by a clinical staging system
(Table 125-2).151 As an AVM grows, it can become more mass-
like, causing ulceration of the overlying soft tissue, bleeding,
pain, or heart failure. Lower-extremity AVMs often develop
curious, dry, brown-violaceous-colored plaques.2
The majority of AVMs are sporadic. Some forms are
heritable (e.g., the autosomal dominant disorder capillary
malformation-arteriovenous malformation [CM-AVM], which
is caused by mutations in RASA).152 The hallmark of CM-AVM
is the randomly distributed, pink to red, multifocal CMs
accompanied by a fast-flow vascular lesion such as an intra-
cranial or extracranial AVM, Parkes Weber syndrome (PWS),
or a vein of Galen aneurysmal malformation in one third of
patients.153
Ultrasonography and Doppler imaging can quickly detect
the fast flow of an AVM. Dilated feeding arteries and draining
veins can be seen as areas of contrast enhancement on CT,
signal flow voids (black tubular structures) on spin echo
MR images, and signal enhancement (white tubular struc-
tures) on gradient sequences including MR angiography.
Ability to discern the nidus is variable. Unlike infantile hem-
angioma, there is no obvious parenchymal mass.36 Muscle
enlargement, bony changes, and increased fat may be seen.
FIGURE 125-13 Arteriovenous malformation of the foot. AVM of the foot shows faint pink staining of the overlying skin (left). Angiographic appearance
of the AVM (right), which delineates the nidus (arrow).
TABLE 125-2
Clinical Staging System for Arteriovenous Malformation
Stage Clinical Findings
I (Quiescence) Pink-blue warm stain, shunting on Doppler
examination
II (Expansion) Enlargement, pulsation, thrill, bruit, tense
veins
III (Destruction) Dystrophic skin changes, ulceration,
bleeding, pain, or tissue necrosis
IV (Decompensation) Cardiac failure
Superselective angiography has its role at the time when treat-
ment is planned and often more clearly delineates the nidus
(see Fig. 125-13).2
Many AVMs require treatment as a result of continued
lifelong expansion. Angiographic embolization alone or in
combination with surgical excision is the mainstay of treat-
ment. Controversy exists about when to intervene. In the early
stages, the full extent of the lesion may not be appreciated at
the time of resection, which could result in local recurrence
and complicate future procedures. Meanwhile, a well-
localized stage I AVM is often more amenable to resection
and complete removal. During infancy, treatment is rarely in-
dicated except in rare cases of postnatal heart failure. After a
complete diagnostic evaluation, infants and children should
be followed annually. Extremity AVMs can result in limb
length discrepancy. Stage I and II AVMs are generally ob-
served. Treatment is usually delayed until symptoms indica-
tive of stage III develop: tissue destruction, pain, bleeding,
or ulceration. For treatment initiated at any age or stage, prox-
imal feeding arteries must never be ligated or embolized. The
nidus of the AVM tends to recruit nearby arteries leading to
ongoing growth and continued progression when feeders
are ligated.25 Furthermore, despite temporary improvement
 CHAPTER 125 VASCULAR ANOMALIES 1627

in bleeding and heart failure, occlusion of feeding arteries

precludes future embolization.
All efforts to treat an AVM should be directed toward the
nidus. Superselective arterial or retrograde venous emboliza-
tion can decrease pain and bleeding. Embolic agents include
destructive liquids such as ethanol, glue, onyx, particles, or
coils. Repeated and staged embolizations are usually neces-
sary. Symptomatic improvement is only temporary because
it is nearly impossible to occlude all microscopic arteriove-
nous shunts. The preferred strategy consists of arterial embo-
lization of the nidus followed by surgical resection 2 to 3 days
later. Preoperative embolization decreases intraoperative
blood loss but does not decrease the extent of resection. Com-
plete removal of the nidus and overlying soft tissue and skin is
the goal; this will decrease recurrence. Resection extent is
determined by reviewing the earliest radiologic imaging, pref-
erably before any treatment. Primary closure is sometimes
possible. More frequently, though, skin grafting or tissue
transfer is necessary. The best results are seen with well-
localized AVMs. These patients must be followed closely with
physical examination and MRI or ultrasonography because
the recurrence rate is high. In some cases, AVMs are unresec-
table. Amputation is a last resort for debilitating AVMs of
the extremities.2,25,45 Drugs that inhibit angiogenesis and
extracellular matrix remodeling are being explored given poor
outcomes with other modalities.154
FIGURE 125-14 Capillary-lymphatico-venous malformation of the lower
extremity. Note the soft tissue hypertrophy and lateral position of the
capillary malformation. Acral anomalies are also present.
Complex-Combined Vascular
Malformations
------------------------------------------------------------------------------------------------------------------------------------------------
or hypotrophic in 10% of patients. Lymphatic anomalies
Like single-channel-type vascular malformations, combined include lymphedema or lymphatic cysts (macrocysts or micro-
lesions are also categorized as slow-flow and fast-flow (See cysts). Lymphatic vesicles often erupt through the CM. LMs
Table 125-1). These disorders are usually associated with are common in the buttock, perineum, and pelvis. Persistent
hypertrophy and skeletal overgrowth. embryonic veins are found in patients with lower extremity
CLVM; the most common is the marginal vein of Servelle.
CAPILLARY-LYMPHATICOVENOUS Branches of the marginal vein become prominent due to
MALFORMATION incompetent valves. The deep venous system may be absent.
Pelvic, perineal, and genital involvement is common in
Klippel-Trenaunay syndrome is a well-known eponym for CLVM. The vascular malformations may cause bladder outlet
capillary-lymphatico-venous malformation (CLVM).2 It is obstruction. In some infants and children, extension of an LM
often incorrectly called “Klippel-Trenaunay-Weber syndrome,” through the inguinal canal is mistaken for an inguinal hernia.
suggesting a relation to PWS, a fast-flow malformation consist- LMs in the perineum can serve as an infectious source. The
ing of a capillary-arteriovenous malformation in association rectum is variably involved. Portomesenteric venous thrombo-
with limb hypertrophy. CLVM is usually diagnosed at birth sis and portal hypertension have been observed in some
by the presence of an enlarged lower extremity with lateral patients.156 Venous abnormalities of the lower limbs often
capillary malformations (CMs), lymphatic vesicles, and visible extend into the pelvis, connecting anomalously with the fem-
varicosities. CLVM can be suspected on antenatal imaging oral and iliac veins and inferior vena cava. This aberrant connec-
but often cannot be confirmed until delivery because congen- tion can result in pulmonary embolism in 4% to 25% of cases.
ital lymphedema, PWS, and CLOVES syndrome (Congenital, MRI and MR venography (MRV) provide the foundation for
Lipomatous, Overgrowth, Vascular malformations, Epidermal describing the type, location, and extent of the vascular mal-
nevi and Spinal/skeletal anomalies/scoliosis) are included in formation components of CLVM. Hypertrophic fatty tissue in
the differential diagnosis. areas of overgrowth can be evaluated (Fig. 125-15). MRV
CLVM most frequently involves the lower extremity (88%) delineates the normal state of the veins and reliably demon-
but can involve the upper extremity (29%) and trunk strates the anomalous venous channels present in CLVM of
(23%).155 The deformity can range from barely perceptible the extremity. The lateral marginal vein of Servelle can be seen
capillary staining with mild soft tissue overgrowth to a gro- coursing along the lateral calf and thigh (see Fig. 125-15).
tesquely deformed limb (Fig. 125-14). Soft tissue and skeletal Microcystic LMs can be seen in the abdominal wall, buttock,
hypertrophy of the involved limb predominate. CLVM of the and extremity, whereas macrocysts are seen in the pelvis and
torso and upper extremities may involve the mediastinum perineum. Plain radiographs are used to evaluate and follow
or retropleural space. The affected extremity may be short limb-length discrepancies. Venography may be used to map
1628 PART IX SPECIAL AREAS

FIGURE 125-15 Imaging of lower extremity capillary-lymphatico-venous malformation (CLVM). Axial T2 view of CLVM of the right leg (left) shows pre-
dominantly extrafascial soft tissue overgrowth and microcystic lymphatic malformation (arrow). Magnetic resonance venogram (right) depicts the lateral
marginal vein of Servelle (arrows).

the venous drainage of the extremity when resection or sclero-
therapy is being considered.
In general, treatment for CLVM is conservative and focused
on symptoms. Compression therapy is the mainstay of conser-
vative treatment. It can be delayed until the child is walking.
Lymphatic oozing may improve with compression therapy.
Sclerotherapy can be used to treat macrocystic LMs and
venous components. Intradermal lymphatic vesicles can also
be sclerosed. Persistent embryonic and other anomalous veins
may be amenable to endovenous laser ablation or resection.
Veins with direct connections to the femoral or iliac veins
or inferior vena cava should be considered for pre-emptive
ablation or resection to prevent pulmonary embolism.
Surgical debulking procedures can be of tremendous
physical and psychological benefit (Fig. 125-16). The location
of the soft tissue overgrowth, either extrafascial or intrafascial,
is critical in determining which patients will benefit from
staged contour resection. In general, intrafascial overgrowth
should not be debulked due to risk of injury to major neuro-
vascular structures and immobility. Truncal and thoracic wall
CLVM are also amenable to staged resections. Excision of me-
diastinal and intrathoracic components is performed only
when symptomatic. Postoperative healing can be problematic
regardless of location since flaps are made of abnormal tis-
sue with poor lymphatic drainage and altered circulation.
Prolonged closed-suction drainage is used. Perioperative

FIGURE 125-16 Debulking of capillary-lymphatico-venous malformation. Overgrowth of right buttock and extremity (left). Intraoperative appearance of
buttock flaps and excision of soft tissue overgrowth (middle). Postoperative appearance of leg (right).

anticoagulation and placement of a temporary inferior vena
cava filter to decrease deep venous thrombosis and prevent
pulmonary embolism should be considered for extensive
surgical procedures.25
Gross foot enlargement that impairs ambulation and the
ability to wear shoes requires orthopedic corrective proce-
dures and partial amputations to permit the use of custom
footwear.25 Limb length discrepancy should be monitored
annually by an orthopedic surgeon. Differences less than
0.5 cm require no therapy. Those between 0.5 and 2 cm are
managed non-operatively with internal or external heel lifts.
Discrepancies in the legs greater than 2 cm are often treated
with epiphysiodesis at the distal femoral growth plate around
12 years of age. Correction of upper extremity discrepancies
is not necessary.
PARKES WEBER SYNDROME
Capillary-arteriovenous malformation (CAVM) and capillary-
arteriovenous fistulas (CAVFs) correspond to the eponym Parkes
Weber syndrome. Lymphatic anomalies are often present (capil-
lary-arteriovenous-lymphatico-venous malformation). The syn-
drome is characterized by the presence of a confluent or patchy
capillary malformation with multiple underlying micro-AVFs
(Fig. 125-17). There is soft tissue and skeletal hypertrophy of
the affected limb, usually a lower extremity.153 The bony over-
growth can result in a limb length discrepancy. The stained areas
are usually warm; a thrill may be palpable. Hand-held Doppler
examination often reveals increased flow and low-resistance
run-off when placed over the stained areas.
De novo or inherited mutations in RASA1 have been
identified in those patients with multifocal CMs.153 The
FIGURE 125-17 Parkes Weber syndrome and capillary malformation-arteriovenous malformation (CM-AVM). Parkes Weber syndrome of the left lower
extremity (left). Notice geographic CM with fusiform, symmetric hypertrophy of the entire leg and foot. Characteristic appearance of CMs seen in CM-AVM
(right).
CHAPTER 125 VASCULAR ANOMALIES 1629
genetic mutation in those without multifocal CMs is yet to
be elucidated (see Fig. 125-17). The distinction between the
two presentations is clinically relevant. Some patients with
RASA1 mutations have been found to have vein of Galen
aneurysmal malformations and tumors similar to those
observed in neurofibromatosis.153
On imaging, the involved extremity usually has fusiform
subcutaneous, muscular, and bony overgrowth with diffuse
microfistulas. Angiography and venography can reveal gener-
alized arterial and venous dilation and a soft tissue blush
involving muscles and subcutaneous fat.
Some patients with PWS are incorrectly diagnosed as having
CLVM or diffuse capillary malformation with overgrowth. An
accurate diagnosis is critical. Up to 30% of patients with
PWS exhibit signs of cardiac volume overload, which is gener-
ally well tolerated.153 In rare instances, cardiac failure can de-
velop secondary to shunting through the arteriovenous fistulas.
Infants and children are seen yearly and monitored for axial
overgrowth, signs of cardiac failure, and cutaneous problems
related to ischemia. Treatment is reserved for symptomatic pa-
tients. Flow reduction may be accomplished with repetitive
superselective embolization to improve heart failure.2 Limb
amputation may be required for recalcitrant disease.
CLOVES SYNDROME
Congenital lipomatous overgrowth, vascular malformations,
epidermal nevi, and skeletal anomalies are recognized as
features in a new syndrome.157,158 CLOVES syndrome’s main
features are truncal lipomatous masses, vascular malforma-
tions, and acral/musculoskeletal anomalies (Fig. 125-18).
The key feature is the presence of a truncal lipomatous mass,
1630 PART IX SPECIAL AREAS
FIGURE 125-18 CLOVES syndrome. Truncal lipomatous mass in an infant (left). A common acral skeletal anomaly is depicted showing widened triangular
feet and polydactyly (middle). Coronal fast spin-echo inversion recovery magnetic resonance image of a truncal lipomatous mass (right, arrowheads). Sco-
liosis is also apparent.
which is usually evident at birth. The lesions may be mistaken
for an LM; however, the truncal masses are hypervascular and
exhibit rapid postresection recurrence. On imaging, the le-
sions are also infiltrative and can extend into adjacent areas
such as the retroperitoneum, mediastinum, and pleural cavity.
The lipomatous growths may involve the spinal column and
epidural space. Compression of the cord, thecal sac, and nerve
roots can occur.
Vascular malformations are also present with slow-flow
lesions having been identified in all documented cases.157,158
CMs can be seen on the trunk. LMs are often located adjacent
to or within the truncal lipomatous masses. VMs, in the form
of phlebectasia, can course over or around the truncal lesions
and can be a source of pulmonary embolus. Paraspinal AVMs
can result in paresis and spasticity. MRI with venous and
arterial sequences to determine the presence, location, and
extent of these lesions should be obtained early in life. Acral
deformities include large, wide feet and hands; macrodactyly;
and a wide sandal gap. Scoliosis has been documented in 50%
of patients.157
PTEN HAMARTOMA-TUMOR SYNDROME
Mutations in the PTEN (phosphatase tensin homolog on
chromosome 10) gene cause two autosomal dominant disorders
that have predisposition for cancer, Bannayan-Riley-Ruvalcaba
and Cowden syndromes. These two entities comprise the
PTEN hamartoma-tumor syndrome. Bannayan-Riley-Ruvalcaba
syndrome is characterized by macrosomia at birth, macro-
cephaly, lipomas, hamartomatous intestinal polyposis, and var-
iable degrees of developmental delay. Genital lentiginosis, seen
best in males as freckling on the glans penis, can be present at
birth or develop in early childhood or puberty. Cowden syn-
drome is characterized by multiple hamartomas and neoplasias
of ectodermal, mesodermal, and endodermal origin. Vascular
anomalies are present in about 50% of patients with PTEN
mutations and are typically multifocal intramuscular com-
binations of fast-flow channels and ectopic fat.159 The presence
of a fast-flow lesion in a macrocephalic patient should raise
suspicion of PTEN hamartoma-tumor syndrome.
Conclusions
------------------------------------------------------------------------------------------------------------------------------------------------
The past 30 years have witnessed remarkable gains in under-
standing the pathogenesis of vascular anomalies. Confusing
nomenclature has been supplanted by precise terminology
based on a genetic-anatomic-histologic classification system.
Multidisciplinary vascular anomaly centers combine medical,
surgical, radiologic, and pathologic expertise, assisting proper
diagnosis and treatment of what were previously hopeless
situations.
The complete reference list is available online at www.
expertconsult.com.
 complexity of the aortic reconstruction, with considerable dif-
ferences if the celiac artery (CA) and superior mesenteric
artery (SMA) are involved, compared with the renal arteries
alone. Most aortic coarctations are diminutive vessels, having
an hourglass appearance.
The cause of most abdominal aortic coarctations appears
related to events occurring around day 25 day of fetal devel-
opment when the two embryonic dorsal aortas fuse and lose
their intervening wall to form a single vessel. Overfusion of the
two embryonic dorsal aortas or their failure to fuse with sub-
sequent obliteration of one of these vessels would predictably
result in an aortic narrowing.
Developmental overfusion of the two primitive dorsal
aortas receives support in patients with decreased aortic diam-
eters who have single origins of the lumbar arteries. In addi-
tion, the presence of multiple renal arteries to one or both
kidneys in these patients is nearly twice that observed in
the general population, lending further support to a develop-
mental etiology.5,7,8 Normal aortic development occurs at
approximately the same embryonic time that the multiple
metanephric arteries involute, leaving a single renal artery.
It is likely that if aortic narrowings exist, flow disturbances will

CHAPTER 126
occur in the vicinity of what would usually be the principal
renal artery and diminish its hemodynamic advantage, allow-
ing persistence of adjacent metanephric channels.
Certain well-known genetic diseases appear to be asso-
ciated with these arterial anomalies. The most common is
Pediatric Arterial neurofibromatosis-1 (NF-1), in which patients exhibit an
unusually high frequency of developmental abdominal aortic

Diseases coarctations and renal artery stenoses.5 Because of the protean

nature of NF-1 and infrequent genetic analyses of patients
with abdominal aortic coarctation, the exact frequency of
James C. Stanley and Jonathan L. Eliason this disease is likely to exceed the 29% reported in a recent
University of Michigan series. The primary vascular pathology
in neurofibromatosis appears to be related to an inhibition of
vessel growth, not entrapment or invasion of the arterial wall
by neural elements. Williams’ syndrome is a second genetic
Arterial occlusive and aneurysmal diseases in infancy, child- disease associated with atypical abdominal aortic narrowings.
hood, and adolescence are uncommon. An understanding Viral-mediated events may impede transition of fetal
of their etiology, manifestations, and treatment is often based mesenchymal tissue to vascular smooth muscle or alter its
on anecdotal case reports. Nevertheless, the more frequent use organization and growth in utero. They may also result in de-
of contemporary imaging and advances in vascular surgery velopmental aortic narrowings. Certain viruses including
have defined the management of most arterial diseases affect- rubella are cytocidal and inhibitory to cell replication, with in-
ing pediatric patients. This chapter addresses common arterial timal fibroplasia and aortic hypoplasia occurring as a conse-
problems affecting the aortic, renal, splanchnic, extremity, quence. In this regard, fibroproliferative intimal disorders
and cerebrovascular circulations. have been documented in the aorta and large elastic arteries of
16.5% of patients exhibiting the congenital rubella syndrome.9
Panaortitis with adventitial or periadventitial fibrosis
and associated inflammatory cell infiltrates, suggesting an
Aortic Disease
------------------------------------------------------------------------------------------------------------------------------------------------
active or chronic aortitis, is another well-recognized cause
of abdominal aortic coarctations. The proposition that most
ABDOMINAL AORTIC COARCTATION abdominal aortic coarctations are a variant of an inflam-
matory aortitis like Takayasu disease is quite controversial
AND HYPOPLASIA
yet such may be the most common cause of these aortic
Coarctation and hypoplasia of the abdominal aorta are uncom- narrowings in the subcontinent populations of Asia and
mon and are often associated with coexisting splanchnic and South America.
renal artery occlusive disease.1–6 The anatomic location of the Associated narrowings of the abdominal aortic branches
aortic narrowing is important, with 69% being suprarenal, are common in these patients. Nearly 90% of those with
23% intrarenal, and 8% infrarenal (Figs. 126-1, 126-2 and abdominal aortic coarctation have renal artery stenoses.5
126-3).5 The contemporary classification of aortic coarctation Splanchnic arterial occlusive disease occurs in 62% of these
is based on the most superior level of the narrowing. It is patients, with both CA and SMA stenoses and occlusions
the most cephalad extent of the disease that defines the involved in 82% of the affected patients.5
1631
1632 PART IX SPECIAL AREAS

A B
FIGURE 126-1 A, Suptarenal abdominal aortic coarctation (bracket) with superior mesenteric artery stenosis (arrow) and B, bilateral renal artery stenoses
(arrows). Note common trunk of lower lumbar artery (arrow) on posterior projection. (From Stanley JC, Criado E, Eliason JL, et al: Abdominal aortic coarc-
tation: Surgical treatment of 53 patients with a thoracoabdominal bypass, patch aortoplasty, or interposition aortoaortic graft. J Vasc Surg 2008;48:1073-
1082.)

FIGURE 126-2 Intrarenal abdominal aortic coarctation with bilateral

artery stenosis (arrows). (From Stanley JC, Criado E, Eliason JL, et al:
Abdominal aortic coarctation: Surgical treatment of 53 patients with a
thoracoabdominal bypass, patch aortoplasty, or interposition aortoaortic
graft. J Vasc Surg 2008;48:1073-1082).

Clinical manifestations of this disease usually reflect

uncontrolled hypertension due to suprarenal or intrarenal
aortic narrowings. Two thirds of developmental coarctations
and a quarter of those having inflammatory-related aortic nar-
rowings have associated renal artery stenoses. The secondary
hypertension in these instances is usually resistant to sim-
ple pharmacologic control. Occasionally, a patient reports
exercise-related lower extremity fatigue, but true claudication
is rare.5 Symptomatic intestinal ischemia is uncommon,
affecting only 6% of those having known CA and SMA disease
accompanying their aortic narrowings.10
FIGURE 126-3 Infrarenal abdominal aortic coarctation manifest by
tubular stenosis extending from dilated inferior mesenteric artery to the
aortic bifurcation.
Abdominal aortic coarctations usually cause signs or symp-
toms during the first or second decade of life in contemporary
times. An earlier review noted that patients had reached a
mean age of 22 years before the diagnosis was actually con-
firmed.3 Untreated, this entity has been associated with
 CHAPTER 126 PEDIATRIC ARTERIAL DISEASES 1633

stroke, progressive left ventricular hypertrophy, congestive
heart failure, flash pulmonary edema, and less often with renal
insufficiency.11,12 In one review, 55% of untreated patients
died at a mean age of 34 years.3
Treatment of abdominal aortic narrowings in children has
most often entailed open surgery.5 Historically, thoracoab-
dominal bypass was the conventional means of treating these
patients (Figs. 126-4 and 126-5). In more recent years patch
aortoplasty has become preferred when it can be performed
safely (Figs. 126-6 and 126-7). Both procedures can be
technically challenging, especially in the face of coexisting
splanchnic or renal disease requiring simultaneous treatment.
Thoracoabdominal bypass grafts originate from the distal
thoracic aorta above the diaphragm or from the supraceliac
aorta at the diaphragmatic hiatus and are passed behind the
left kidney to the distal aorta. Expanded Teflon grafts are
favored because of their greater stability regarding postim-
plantation dilatation. Graft diameter should be chosen to be
as big as possible, short of being so large that excessive luminal
thrombus would accumulate. The intent is to oversize grafts
compared with the aorta, with anticipated growth otherwise
resulting in a graft too small to maintain normal distal pres-
sures and flow. In the ideal circumstance, one should use a
graft whose size would not represent an energy-consuming
constriction as the child grows into maturity. This means
having a conduit at least 60% or 70% the size of the adult
aorta. In early childhood the use of large conduits may not
be possible. Graft length is a nonissue in older children and

FIGURE 126-4 Left, Suprarenal coarctation (bracket) with superior
mesenteric artery stenosis (arrow). Right, Subsequent thoracoabdominal
bypass (broad arrow) with aortic implantation of superior mesenteric artery
(arrow). (From Stanley JC, Criado E, Eliason JL, et al: Abdominal aortic
coarctation: Surgical treatment of 53 patients with a thoracoabdominal
bypass, patch aortoplasty, or interposition aortoaortic graft. J Vasc Surg
2008;48:1073-1082.)
FIGURE 126-6 Left, Suprarenal coarctation (bracket) with bilateral renal
artery ostial stenoses. Preoperative magnetic resonance angiography.
Right, Subsequent patch aortoplasty (broad arrow) with aortic diameter ex-
ceeding that of uninvolved proximal and distal aorta. Reimplantation of
the renal arteries accompanied the aortic reconstruction. (From Stanley JC,
Criado E, Eliason JL, et al: Abdominal aortic coarctation: Surgical treatment
of 53 patients with a thoracoabdominal bypass, patch aortoplasty, or
interposition aortoaortic graft. J Vasc Surg 2008;48:1073-1082.)

FIGURE 126-5 Complex aortic, splanchnic, and renal arterial reconstruction: 1, Celiac artery (CA) stenosis; 2, Superior mesenteric artery (SMA) stenosis;
3, suprarenal midabdominal aortic coarctation; 4, right renal artery ostial stenosis; 5, left segmental renal artery stenosis; 6, CA implanted onto aorto-
SMA bypass (with autogenous internal iliac artery graft); 7, reconstructed SMA; 8, thoracoabdominal aortic bypass; 9, right renal artery reimplantation onto
aorta; 10, left segmental renal artery implantation onto adjacent segmental renal artery. (From Upchurch GR Jr, Henke PK, Eagleton MJ, et al: Pediatric
splanchnic arterial occlusive disease: Clinical relevance and operative treatment. J Vasc Surg 2002;35:860-867.)
1634 PART IX SPECIAL AREAS
FIGURE 126-7 Complex aortic, splanchnic, and renal arterial reconstruction: 1, Celiac artery (CA) stenosis; 2, superior mesenteric artery (SMA) stenosis,
3, interrenal midabdominal aortic coarctation; 4 and 5, right and left renal artery stenosis; 6, CA implanted onto stenotic SMA origin; 7, widely patent SMA;
8, polytetrafluoroethylene patch aortoplasty; 9 and 10, bilateral implantation of renal arteries onto aorta. (From Upchurch GR Jr, Henke PK, Eagleton MJ,
et al: Pediatric splanchnic arterial occlusive disease: Clinical relevance and operative treatment. J Vasc Surg 2002;35:860-867.)
adolescents, with axial growth from the diaphragm to pelvis
being minimal after reaching an age of 9 or 10 years.
Patch aortoplasty is usually undertaken when the coarcta-
tion segment has a large enough diameter to allow completion
of an anastomosis without an overlap of sutures from the
opposing sides of the patch. Whenever possible, patches in
children should be large enough, similar to thoracoabdominal
graft sizing, so as to not be constrictive with growth into adult-
hood, yet not so generous as to risk development of an exten-
sive lining of unstable thrombus. Teflon graft material is again
favored over Dacron graft because of the latter’s propensity for
dilatation years after implantation.
Reoperations are infrequent but may be required for anas-
tomotic narrowings or if a patient outgrows the adequacy
of the primary aortic reconstructive procedure. The fact that
nearly 10% of the cases in a recent report from the University
of Michigan required late secondary operations supports
the importance of long-term follow-up of these patients.5
Simultaneous visceral artery reconstructions depend on
the clinical relevance of the nonaortic disease, as well as the
proximity of the aortic reconstruction to the affected aortic
branches. Renal artery stenoses causing secondary renovascu-
lar hypertension justify their reconstruction. A mandate to
reconstruct the CA or SMA applies only to symptomatic cases.
Nevertheless, a relative indication to prophylactically recon-
struct these vessels exists when performance of an aortoplasty
or renal revascularization would make a subsequent CA or
SMA revascularization exceedingly difficult. When the aortic
reconstruction is distant from the CA or SMA, such as with
thoracoabdominal bypass, a concomitant splanchnic revas-
cularization is less likely to be warranted.
Treatment of select focal abdominal aortic coarctations
distant from the renal and splanchnic branches may involve
endoluminal interventions. When such is undertaken, stent
placement is necessary to overcome the significant recoil of
these hypoplastic and highly fibrotic aortic narrowings.13–15
However, one should remain cautious about accepting the
long-term benefits of endoluminal treatment of abdominal
aortic coarctation in children. Given the high frequency with
which the visceral arteries are involved in abdominal aortic
narrowings, the number of lesions amenable to endovascular
repair is likely to be limited.
More than 90% of patients with abdominal aortic coarcta-
tion benefit when treated with a patch aortoplasty or
a thoracoabdominal bypass.5 Accompanying postoperative
morbidity is low, and mortality should approach zero. Never-
theless, long-term follow-up of these young patients with
annual noninvasive assessments of lower extremity blood
flow is recommended. Imaging with magnetic resonance an-
giogram (MRA) studies or computed tomography angiogram
(CTA) should be obtained if any evidence of diminished blood
flow exists.
ABDOMINAL AORTIC ANEURYSMS
Abdominal aortic aneurysms (AAAs) are rare in children. All
are considered life threatening. The etiology is often difficult
to ascertain, but in general these AAAs have been categorized
as (1) infectious, (2) related to known vascular disease entities,
or (3) due to unknown congenital developmental misadven-
tures. The clinical manifestations vary widely among these
different categories. The diagnosis typically follows imaging
with ultrasonography, MRA, or CTA.
Infectious AAAs associated with umbilical artery catheter-
izations (UACs) account for approximately one third of aortic
aneurysms reported in children. These aneurysms may pre-
sent anytime during the first 2 years of life, but most are
encountered in the newborn period. Repeated UAC insertion
and prolonged UAC use have been implicated as risk factors
for this type of AAA. Whether these aneurysms are the result

of a bacteremic invasion of a catheter-injured aorta or due to an
infected thrombus is problematic, but the outcome is destruc-
tion of the involved vessel and aneurysm formation. More than
75% of these aneurysms are associated with a Staphylococcus
aureus or Staphylococcus albus infection.16 The resulting AAAs
are usually saccular, and more than 50% involve the abdominal
aorta (Fig. 126-8). Aortic infection related to UAC represents
a life-threatening illness,16,17 with AAA rupture being the
most serious complication. These infected AAAs, with rare ex-
ception, require early operation. Interventions range from
simple aortic ligation to aneurysmectomy with an autologous
FIGURE 126-8 Aortic and iliac artery aneurysms due to infection second-
ary to umbilical artery catheterization.
A B
FIGURE 126-9 A, Preoperative tuberous sclerosis-related abdominal aortic aneurysm with renal and splanchnic arterial involvement. B, Postoperative
appearance following aneurysmectomy, aortic reconstruction, and revascularization of the celiac and superior mesenteric arteries.
CHAPTER 126 PEDIATRIC ARTERIAL DISEASES 1635
or synthetic graft aortic reconstruction. Long-term antibiotic
administration must accompany these procedures.
Aortic aneurysms in childhood may also be related to
known vascular diseases, the more common of which deserve
mention. Tuberous sclerosis (TS) is an autosomal dominant
disease with hamartomas in multiple organ systems, with cen-
tral nervous system involvement manifest by learning diffi-
culties and seizures, as well as vascular lesions.18,19 Most
TS-related aortic aneurysms are abdominal in location, with
less frequent thoracic aortic involvement. Fatal rupture of
these AAAs is well recognized.19,20 Treatment strategies are
often complex and tailored to anatomic involvement, with a
frequent need to reconstruct the splanchnic and renal vessels
(Fig. 126-9).
Ehlers Danlos syndrome (EDS) includes a spectrum of dis-
orders with vascular involvement. This is especially the case in
type IV EDS, an autosomal dominant disease resulting in
defective type III collagen production. It is often manifest by
spontaneous rupture of a hollow viscus or large arteries.
Among 132 vascular complications in 24 patients with EDS,
17 had thoracic or abdominal aortic involvement including
dissections, aneurysms, and ruptures.21 Unfortunately, the
true prevalence of this disease is unknown, in that cata-
strophic complications of EDS resulting in death may occur
before a definitive diagnosis is established. Vascular surgery
in these patients is often required, but because of vessel fragil-
ity and bleeding, operative treatment must be approached
with extreme care.22
Marfan syndrome is an autosomal dominant disorder,
affecting collagen cross-linking due to mutations in the
fibrillin-1 gene. Infantile arterial involvement in Marfan syn-
drome is rare. The vast majority of these children have cardiac
problems including dilation of the aortic root, mitral valve
prolapse, and valvular regurgitation.23 Nevertheless, 20% of
these children develop aortic aneurysmal changes as they
grow older.24
Takayasu aortoarteritis is also a recognized cause of pedia-
tric aortic aneurysm formation, pseudoaneurysm formation,
aortic stenoses, and aortic dissections.25,26 The clinical pre-
sentation may be marked by systemic symptoms of
inflammation or may be related to aortic involvement. Aortic
1636 PART IX SPECIAL AREAS
reconstructions are less urgent in this group of children and
are best undertaken when the aortoarteritis is quiescent.
Certain pediatric AAAs are not infectious or related to
an underlying connective tissue disease. Their etiology is
related to developmental events evolving during fetal growth
(Fig. 126-10). Aneurysms in this group are commonly asso-
ciated with stenotic lesions and multiple aneurysms in other
locations. Concomitant renal, upper extremity, or iliac artery
aneurysms occur most often in these children, with cerebro-
vascular, mesenteric, and lower extremity arterial involvement
being less common.27
Seven children, ranging in age from 2 weeks to 8 years,
have undergone surgical repair of their AAAs at University
of Michigan from 2002 to 2010, with six survivors. The AAAs
involved 10 renal arteries in five patients, all of whom were
hypertensive. No two aneurysms were of the same clear etiol-
ogy, and aortic reconstructions varied in each case including
concomitant renal or splanchnic arterial reconstructions in
four children.
ABDOMINAL AORTIC THROMBOSIS
Acute thrombotic occlusions of the aorta and its branches have
multiple causes including prothrombotic coagulation disor-
ders, sepsis, and certain cardiopulmonary diseases. However,
iatrogenic injury is the most common inciting event, usually a
complication of a neonatal UAC.17,28,29 Some degree of aortic
thrombosis accompanies 20% to 30% of UACs,30 and 89% of
neonates experiencing aortic thrombosis had indwelling
arterial catheters.31
Clinical presentations vary widely from silent thromboses
to profound tissue ischemia manifest by lower extremity
discoloration and multiple systemic organ failure. Overall
mortality with neonatal aortic occlusion approaches 15%
to 20%.31,32 Diagnoses are most often made with Doppler
FIGURE 126-10 Idiopathic multiple aortic aneurysms in a newborn.
evidence of absent extremity blood flow in the face of no pal-
pable peripheral pulses and CTA or MRA imaging confirming
the aortic involvement. Intravenous digital subtraction arteri-
ography has a role in select patients. Conventional catheter-
based diagnostic arteriography in the neonate is ill-advised
but may provide a means for delivering lytic agents in select
cases. In that setting, access may be best through an umbilical
artery catheter.29
Treatment of acute aortic thromboses in neonates is
initiated with heparin anticoagulation. Fibrinolytic therapy,
a challenge because of low plasminogen levels in the newborn,
is an important therapy.33–35 Heparin administration should
accompany tissue plasminogen activator (tPA) infusions in
these cases. Hemorrhagic complications may accompany
tPA therapy and such risks must be weighed when considering
the benefits of lytic treatment. Mechanical thrombectomy
using small balloon catheters or polyethylene tubing advanced
through an aortotomy allows disobliteration of the aorta and
extraction of branch thrombus.17
Aortic thromboses in older children and adolescents are
frequently associated with intimal flaps and dissections
caused by blunt trauma.36,37 Abdominal and chest injuries
occurring in motor vehicle accidents and falls from great
heights are common contributors to this trauma. Other life-
threatening injuries often accompany the aortic thrombosis
in these cases. Obvious pulse deficits and tissue ischemia
are manifestations of aortic occlusion in these older children
with the diagnosis being confirmed by CTA, MRA, or conven-
tional arteriography.38 Aortic interposition or bypass grafts are
usually undertaken in these instances with nonanatomic
axillofemoral or thoracoabdominal bypasses required if intes-
tinal perforation has contaminated the immediate area of
aortic injury. Endovascular repairs offer certain benefits in
the adolescents having severe coexisting injuries in addition
to the aortic injury.39,40
Renal Artery Disease
------------------------------------------------------------------------------------------------------------------------------------------------
RENAL ARTERY STENOSIS
Occlusive disease affecting the renal arteries is the third most
common cause of hypertension in children following coarcta-
tion of the thoracic aorta and parenchymal renal disease. The
precise incidence of this type of secondary hypertension in
children is unknown but is likely to account for 5% to 10%
of those exhibiting marked blood pressure elevations. Pedi-
atric renovascular hypertension, when unrecognized and
untreated, has been associated with hemorrhagic stroke, hy-
pertensive encephalopathy with impaired mental develop-
ment, and failure to thrive. Poorly controlled hypertension
may result in left ventricular hypertrophy and severe diastolic
dysfunction. When the entire renal mass is involved, flash pul-
monary edema associated with renal insufficiency may occur.
Pediatric renal artery stenoses represent a spectrum of dis-
eases. Most developmental renal artery stenoses encountered
in North American Caucasian children11,12 are different from
the inflammatory aortoarteritis related stenoses that dominate
reports from the Asian subcontinent.41–43 The cause of most
developmental renal occlusive lesions are the same as those
causing abdominal aortic coarctations.5,11 Many of these
patients have NF-1.7,11,44,45 The vast majority of renal artery

narrowings are ostial in location, representing true hypoplas-
tic stenoses (Fig. 126-11).
Recognition of renal artery stenotic disease in children
requires detailed vascular imaging.11,46 Conventional
catheter-based digital-subtraction arteriography has been a
standard, but at many institutions it is being replaced with
newer CTA technology.5 Nevertheless, digital-subtraction
arteriography should be pursued when a strong suspicion
of renovascular disease exists.5 Screening modalities including
MRA, nuclear renal scans with ACE inhibitors, and deep
abdominal ultrasonography all have value but are currently
too inconsistent to be used as confirmatory diagnostic or
prognostic tests.
Optimal surgical therapy depends on the specific character
of the renal artery disease being treated. The most common
surgical procedure has been renal artery implantation into
the aorta, as well as into the main or adjacent segmental renal
arteries (Figs. 126-12 and 126-13).11 Aortorenal bypasses
with vein grafts were frequently used in the past but are not
favored because in follow-up more than half of these conduits
undergo aneurysmal deterioration.47 The internal iliac artery
has become the preferred conduit for aortorenal bypasses
(Fig. 126-14). Splenorenal reconstructions with a direct
anastomosis of the splenic and renal vessels are not favored
because of coexistent or later development of a celiac artery
stenosis that would perpetuate the hypertensive state. Resec-
tions of the stenotic renal artery with primary reanastomosis,
focal arterioplasty, or open operative dilation are less common
reconstructive procedures. Nephrectomy may be appropriate
for irreparable renal disease including multiple intrarenal
stenoses not amenable to open in situ or ex vivo repairs, as
well as diminutive kidneys 2 to 3 cm in size. In these cases
it is assumed that the contralateral kidney will suffice in keep-
ing the child from going into renal failure. There is a cer-
tain logic in undertaking single-stage vascular reconstructive
FIGURE 126-11 Ostial main renal artery stenoses.
CHAPTER 126 PEDIATRIC ARTERIAL DISEASES 1637
FIGURE 126-12 Bilateral renal artery-aortic implantations. Note the
residual origins of the native renal arteries. (From Stanley JC, Zelenock
GB, Messina LM, Wakefield TW: Pediatric renovascular hypertension:
A thirty-year experience of operative treatment. J Vasc Surg 1995;
21:212-227.)
procedures given that a later operation in the same anatomic
area entails significant hazards of reoperations.
Undertaking surgical therapy in small infants is controver-
sial. Technical challenges exist in reconstructing renal arteries
less than 2mm in diameter and renal revascularization in these
circumstances should be pursued only when uncontrolled se-
vere hypertension or renal failure threaten the patient. Renal
revascularizations are most likely to be successful after age 3
years, and deferring reconstructive procedures in younger
children when possible is reasonable. Drug treatment of hy-
pertension in the young child may be difficult and requires
frequent and fastidious monitoring.
Large clinical experiences with pediatric-aged renovascular
hypertension are uncommon.11,48–51 The University of Michigan
series included 39 girls and 58 boys, aged from 3 months
to 17 years, who underwent operation from 1963 to 2006.11
All but one patient had refractory hypertension not responsive
to contemporary medical therapy. Developmental renal
artery stenoses accounted for 80% of the renal artery disease.
Splanchnic arterial occlusive lesions affected 24%, and
abdominal coarctations affected 33%. Primary renal artery
operations were undertaken 132 times. Procedures included
resection beyond the stenosis and implantation into the aorta
in 49, renal artery in 7, or SMA in 3; aortorenal and iliorenal by-
passes with vein or iliac artery grafts in 40; focal arterioplasty in
10; resection with reanastomosis in 4; operative dilation in 4;
splenorenal bypass in 2; and primary nephrectomy in 13 when
arterial reconstructions proved impossible. Bilateral renal artery
operations were done in 32 children. The fact that 17 underwent
CA or SMA reconstruction, 19 underwent patch aortoplasty,
and 11 underwent thoracoabdominal bypass underscored the
complexity of disease being treated. Thirty secondary renal artery
procedures were done in 19 patients including nine nephrecto-
mies. Hypertension was cured in 68 children (70%), improved
in 26 (27%), and unchanged in 3 (3%). Follow-up averaged
4.2 years. No patients required dialysis, and there were no
operative deaths.
1638 PART IX SPECIAL AREAS
A B
FIGURE 126-13 A, Focal stenosis of intraparenchymal segmental renal artery with poststenotic dilation (arrow). B, Segmental renal artery implantation
into the distal main renal artery (arrow). (From Stanley JC, Zelenock GB, Messina LM, Wakefield TW: Pediatric renovascular hypertension: A thirty-year ex-
perience of operative treatment. J Vasc Surg 1995;21:212-227.)
A second large series was reported from the Hospital
Beaujon in France.48 That series encompassed 78 children,
35 boys and 43 girls, ranging in age from 1.4 to 18 years.
There were 91 primary revascularization procedures and
15 nephrectomies. These numbers were quite similar to the
FIGURE 126-14 Aortorenal bypass with an internal iliac artery
graft. (From Stanley JC, Zelenock GB, Messina LM, Wakefield TW: Pediatric
renovascular hypertension: A thirty-year experience of operative
treatment. J Vasc Surg 1995;21:212-227.)
University of Michigan series. A third series emanated
from the Cleveland Clinic involving 56 children including
23 boys and 33 girls ranging in age from 0.7 to 21 years.49
They reported 46 primary operations and 10 primary
nephrectomies. A fourth series was from Vanderbilt University
and the Children’s Hospital of Philadelphia.50 This experi-
ence included 50 children, 24 boys and 26 girls, ranging
in age from 0.4 to 16 years. These children were subjected
to 28 primary reconstructive procedures and an addi-
tional 12 primary nephrectomies. The latter two series,
unlike the University of Michigan experience, did not
contain many patients treated for aortic or splanchnic arterial
disease.
Outcomes regarding blood pressure control were remarkably
similar at the University of Michigan, Vanderbilt-University-
Children’s Hospital of Philadelphia, and Cleveland Clinic,
with respective cure rates of 70%, 70%, and 66%; reported
improvement rates of 27%, 26%, and 23%; and failure rates of
3%, 4%, and 11%. Surgical therapy in all three of the former
experiences consisted of conventional open operative
interventions.
The role of percutaneous transluminal angioplasty (PTA) in
treating pediatric renovascular hypertension remains contro-
versial.52,53 Failure following PTA for developmental disease
might be anticipated, given the excessive elastic tissue in many
stenoses that would predictably contribute to early postdila-
tion recoil, as well as the minute caliber of developmentally
narrowed vessels that might lead to their disruption. Never-
theless, a small number of recent reports suggest success with
catheter-based interventions.54,55 It is of note that if the dis-
ease being treated is a quiescent inflammatory aortoarteritis,
then more salutary outcomes would follow PTA than might
be the case if treating developmentally hypoplastic renal
arteries. However, even in the former setting recurrent stenoses
are frequent.42,43
 CHAPTER 126 PEDIATRIC ARTERIAL DISEASES 1639

ACUTE RENAL ARTERY DISSECTIONS

AND THROMBOSES
Acute renal artery dissections and thromboses are most often a
result of UAC in the neonate or iatrogenic events accompany-
ing diagnostic or therapeutic catheterizations in older
children. Among adolescents, blunt or deceleration injury
with stretching and dissections of the renal artery is a common
cause of an acute occlusion. Flank pain, hematuria, and
hypertension are frequent manifestations of acute renal artery
occlusions with evolving renal infarction in these children.
Acute thrombosis of an otherwise nondiseased renal artery
resulting in interruption of blood flow for more than 90 minutes
invariably results in an irreversible loss of renal function.
Nevertheless, an attempted operative repair is justified, espe-
cially when treating a vascular injury to a solitary kidney or
bilateral injuries.56 Endovascular interventions offer a less
invasive means of treating select patients in this setting.10
Surgical therapy in the chronic setting and in cases of sub-
total occlusions may require an aortorenal bypass. Revascular-
ization attempts in these children are justified if sufficient
parenchyma exists and nuclide studies suggest the affected
kidney to be functioning at 10% to 20% of the contralateral
unaffected kidney. Nephrectomy is undertaken when recovery
of renal function is impossible.

FIGURE 126-15 Renal artery aneurysm involving second order segmen-

RENAL ARTERY ANEURYSMS tal renal artery. Note the relative hypoperfusion of the upper pole.
The clinical relevance of renal artery aneurysms in children
is poorly understood including an undefined risk of rup-
ture.11,57 Aneurysms affecting the renal artery are generally residual opening at the base of the aneurysm or a closed aneur-
of two categories. ysmorrhaphy in the case of small aneurysms.
The most common aneurysms are those associated with an
underlying congenital vasculopathy or a connective tissue
disorder. More often than not the congenital nature of this
remains unknown, although in some patients genetic diseases
like Ehlers-Danlos are obvious. These aneurysms usually
Splanchnic Artery Disease
------------------------------------------------------------------------------------------------------------------------------------------------

occur at branch points as saccular lesions (Fig. 126-15). Their
discovery usually occurs as an incidental finding during
imaging for other suspected vascular or renal diseases. An oc-
casional aneurysm may have been the source of embolic ma-
terial responsible for distal arterial occlusion, renal ischemia,
and secondary renovascular hypertension.
Treatment is recommended for extraparenchymal aneu-
rysms in hypertensive children or when they exceed 1 cm
in diameter regardless of blood pressure levels. Larger aneu-
rysms may be excised with a primary angioplastic closure,
and smaller ones plicated with fine cardiovascular suture in
the form of a closed aneurysmorrhaphy. Aneurysms arising
within the cortical substance may be treated with catheter-
delivered occlusive material or carefully delivered absolute
alcohol. In that setting there is a predictable loss of kidney
substance beyond the aneurysm.
The second group of renal aneurysms are due to poststeno-
tic turbulent blood flow. Renal artery aneurysms in this cate-
gory were encountered in 8 children among 97 treated for
renovascular hypertension at the University of Michigan.11
These aneurysms tend to be more globular than saccular.
Treatment in these cases necessitates eliminating the stenosis
with a renal artery reconstruction similar to that in managing
renovascular hypertension. Accompanying such a revascular-
ization is aneurysm excision and angioplastic closure of the
SPLANCHNIC ARTERY STENOSES
AND ANEURYSMS
Occlusive disease of the splanchnic arteries is rare in pediatric-
aged patients and is an even rarer cause of clinically relevant
intestinal ischemia.58 These stenotic narrowings usually affect
the origins of the celiac and superior mesenteric arteries
(Fig. 126-16). They are recognized most frequently during
arteriographic studies for suspected aortic or renal artery
pathology.
Developmental lesions appear related to embryonic events
similar to those causing aortic and renal artery narrowings.
These events likely affect the reorganization of the ventral seg-
mental vessels associated with the cephalic roots of the vitel-
line arteries that form the CA and SMA. This results in an ostial
stenosis or occlusion. Minimal CA and SMA narrowings dem-
onstrated at a young age may not become intrinsically nar-
rower as the child grows, but the vessel origin may simply
remain the same size as everything else becomes larger. Thus
the artery’s origin becomes proportionately narrower until it
becomes a critical stenosis.58 This entire process suggests a
major growth arrest of the aortic origins of these arteries.
Symptomatic splanchnic artery disease presents as post-
prandial intestinal angina with periumbilical discomfort that
lasts for the duration of small bowel transit of the ingested
1640 PART IX SPECIAL AREAS

to reconstruct the CA or SMA exists when performance of a

FIGURE 126-16 Severe celiac artery and superior mesenteric artery
ostial stenoses evident on a lateral aortogram.
food. These children develop a food aversion and exhibit a
failure to gain weight. In extreme cases they may experience
weight loss.
Operative treatment of pediatric splanchnic arterial occlu-
sive disease is complex.58,59 A mandate to reconstruct these
vessels applies only to symptomatic cases. A relative indication
primary procedure such as patch aortoplasty or renal revascu-
larization would potentially place these splanchnic arteries at
risk of occlusion. In these circumstances the proximity of the
origins of the CA and SMA to the other reconstruction may
justify a prophylactic splanchnic revascularization.
In certain children internal iliac artery grafts are used to treat
lengthy CA or SMA stenoses. Anastomoses are fashioned with
interrupted sutures when reconstructing small arteries 2 to 3 mm
in diameter. In older adolescents, a continuous suture may be
used to complete anastomoses of larger splanchnic arteries
(see Fig. 126-5). In other children, aortic implantation of the
CA or SMA after spatulation of the transected vessel beyond
its stenotic origin is favored over an aortosplanchnic bypass
(see Figs. 126-7 and 126-17).58 Children treated by both
methods have invariably gained weight and are free of abdom-
inal discomfort.58 Percutaneous transluminal balloon angio-
plasty with or without stenting for these developmental ostial
lesions is not appropriate and ill-advised.
Splanchnic arterial occlusive disease when encountered in
childhood has an uncertain natural history. Long-term follow-
up of these children with periodic imaging studies in those op-
erated on seems appropriate, and regular clinical assessment
of those not subjected to surgical therapy is recommended.
In regard to the latter, it is important to remember that most
CA and SMA stenoses are asymptomatic because of the IMA
serving effectively as a source of collateral circulation. Unop-
erated patients should be aware of their splanchnic arterial
anatomy and be able to pass this information on to anyone un-
dertaking a later abdominal operation. In these individuals,
inadvertently interrupting the collateral circulation might
cause catastrophic intestinal ischemia.
The median arcuate ligament syndrome is a poorly defined
disorder that has been sporadically described in children.60,61

FIGURE 126-17 Left, celiac artery

occlusion and severe ostial stenosis
of the superior mesenteric artery
(SMA). Right, Postoperative docu-
mentation of widely patent SMA
implanted onto anterior aorta.
(From Upchurch GR Jr, Henke PK,
Eagleton MJ, et al: Pediatric splanch-
nic arterial occlusive disease: Clinical
relevance and operative treatment.
J Vasc Surg 2002;35:860-867.)
 CHAPTER 126 PEDIATRIC ARTERIAL DISEASES 1641

Compression or entrapment of the celiac artery by unyielding often suffices (Fig. 126-18). In cases of arterial injury due
fibers of the diaphragmatic crura the aortic hiatus is a recog- to blunt trauma or orthopedic trauma with long-bone frac-
nized anatomic finding. Intermittent upper abdominal pain tures or posterior knee dislocations, the exact level of the
has been reported to be the most common clinical manifesta- injury may be difficult to discern and imaging in this setting
tion alleged to occur in symptomatic cases. Weight loss in will be essential to the diagnosis and proper treatment.
children has not been a constant accompanying finding in In situations of limited lower extremity arterial injury,
these individuals. Treatment usually includes transaction of a local resection and primary repair may be possible. Inter-
the compressing tissue and skeletonization of the celiac artery. rupted sutures in young children with spatulation of the
Celiac artery narrowing may persist if secondary mural fibrosis involved vessels, so as to create an ovoid anastomosis, will
has occurred, but such is less likely to occur in children lessen the risk of later stenosis of the repair. In older patients
compared with adults. Early relief of symptoms is expected, a patch angioplasty or bypass in the lower extremity with
but long-term outcomes remain to be defined. reversed saphenous vein is appropriate (Fig. 126-19).63,71
Splanchnic artery aneurysms in children are uncommon.57 Late aneurysmal deterioration of these grafts is unlikely, in
The etiology that underlies these lesions becomes an impor- contrast to their use in renal artery reconstruction.
tant determinant of therapy. Blunt or deceleration injury to
the liver and spleen may result in small pseudoaneurysms CHRONIC LOWER EXTREMITY ARTERIAL
recognized on a computed tomography (CT) study, but such OCCLUSIONS
may often be safely followed without a specific intervention.
Most small pseudoaneurysms thrombose without sequelae. Chronic lower extremity ischemia in children has the potential
Endovascular embolic obliteration of large or unstable to lead to growth retardation and later limb length discrepan-
aneurysms in these solid organs is usually possible in older cies.63,72 Arterial catheterization-related thromboses are the
children and adolescents. most common cause of delayed lower extremity ischemia.63
Splanchnic artery aneurysms associated with childhood In this regard, limb length discrepancy rates of 8% have been
Ehlers-Danlos syndrome or Kawasaki disease, as well as some reported after femoral catheterizations.73
of the necrotizing arteritides such as polyarteritis nodosa, are Chronic limb ischemia in children is often initially un-
unlike those traumatic lesions affecting the liver and spleen. recognized because of the child’s ability to develop extensive
The former aneurysms carry a potential risk of rupture or collaterals, with palpable pulses in the affected extremity
thrombosis. Most are diagnosed by CTA or MRA studies. similar to those in the uninvolved opposite extremity. Thus
Treatment includes endoluminal catheter-directed throm-
bosis, simple ligation, or aneurysmectomy with an arterial
repair if necessary. Explicit clinical guidelines have yet to be
established in the management of these rare aneurysms.

Extremity Arterial Disease

------------------------------------------------------------------------------------------------------------------------------------------------

ACUTE LOWER EXTREMITY ARTERIAL

OCCLUSION
Catastrophic lower extremity arterial occlusion in pediatric
patients is uncommon.62 It is often a consequence of UAC
thromboembolic events in the neonate and femoral artery
thromboses following arterial catheterizations or cannulations
in the management of congenital heart disease.63–66 In older
children and adolescents, both penetrating and nonpene-
trating arterial injuries can cause profound lower extremity
ischemia.67,68 On rare occasions less common etiologies of
acute arterial occlusion exist including thrombosed aneu-
rysms, paradoxical emboli, and thromboembolism associated
with bacterial endocarditis.69
In the very young an isolated femoral artery occlusion
without initial evidence of severe ischemia may be managed
conservatively with heparin anticoagulation. Some have
reported acceptable results of nonoperative therapy even in
patients with critical ischemia.70 The relative effectiveness of
the collateral circulation in these circumstances usually pre- FIGURE 126-18 Intravenous digital subtraction arteriogram document-
serves limb viability. If impending tissue loss or ischemic nerve ing occluded right external iliac and common femoral arteries. Iodinated
dysfunction exists, then an open revascularization is appro- contrast was injected centrally through a catheter advanced through
priate and preoperative imaging is important by CTA or the right brachial vein under fluoroscopic guidance to the junction of
the vena cava and right atrium. (From Cardenau JD, Henke PK, Upchurch
conventional arteriography by way of umbilical artery cathe- GR Jr, et al: Efficacy and durability of autogenous saphenous vein conduits
terization. In instances where further catheterizations should for lower extremity arterial reconstructions in preadolescent children.
be avoided, an intravenous digital subtraction arteriography J Vasc Surg 2001;34:34-40.)
1642 PART IX SPECIAL AREAS
A B
FIGURE 126-19 A, Occlusion of common femoral and external iliac arteries due to thrombosis after cardiopulmonary bypass cannulation of the vessel in
a child with a limb length discrepancy. The deep femoral and superficial femoral arteries are reconstituted through abundant collaterals. B, Subsequent
iliofemoral bypass graft with saphenous vein.
it is often not until a discrepancy in limb length develops with
gait abnormalities or the child complains of exercise-related
symptoms that the diagnosis is even considered.
Simple Doppler ankle-brachial indices (ABIs) may suggest
limb ischemia. However, detection may be evident only if a
substantial decline in ABI occurs following treadmill exercise.
Conventional arteriographic studies will best define the sus-
pected occlusion in these children. Intravenous digital sub-
traction arteriography with newer computer-enhanced
imaging has proven useful in defining these lesions while
avoiding further vascular complications. Both MRA and
CTA are evolving as commonly performed imaging tech-
niques, but MRA may not provide small vessel detail and cau-
tion is appropriate regarding excessive radiation exposure if
serial CTA studies are necessary.
Interestingly, 93% of patients in one study whose acute is-
chemia was managed without operation had their ABI return
to normal after being treated with heparin anticoagulation
alone.74 Thus a normal resting ABI in the presence of an oc-
cluded iliac or femoral artery does not necessarily eliminate
the risk for later growth problems. It is advised that arterial
reconstruction using vein or synthetic grafts is appropriate
for occluded iliofemoral or femoral arteries before a child en-
ters rapid growth phases. Revascularization before adoles-
cence when long bone growth is the most active is
important to lessen limb length discrepancies. In the past, it
had been generally accepted that late operation would not re-
verse an existing limb length discrepancy, but this has not al-
ways proven to be the case.
UPPER EXTREMITY ARTERY OCCLUSIONS
Upper extremity ischemia is most commonly associated with
blunt or penetrating trauma, shoulder or elbow dislocations,
or supracondylar humerus fractures. Concomitant nerve inju-
ries accompany many of these vascular injuries. Diagnosis is
usually suspected from the clinical findings of pallor, reduced
or absent pulses, and weakness with repetitive motions.
Formal imaging is not necessary if the site of injury is
obvious by physical examination. In such a setting a standard
vascular repair should be pursued if the ischemia is severe with
impending tissue loss or pain with minimal activity. Lesser
degrees of acute ischemia may abate with development of
collateral vessels. In those children, especially the very young,
conservative nonoperative therapy may be appropriate.
If exercise-related claudication occurs later, a formal revascu-
larization may be undertaken. In many early cases with a focal
injury, the involved arterial segment may be excised, the vessel
mobilized so as to have no tension on it, and a primary reanas-
tomosis undertaken. In other cases a bypass with autologous
saphenous or basilic vein may be required.
Subclavian-axillary artery occlusion is an unusual complica-
tion of the birthing process (Fig. 126-20). Mechanisms causing
this complication are believed to be similar to those cited for
brachial plexus birth injuries. Arterial occlusions may result
from intimal fracture secondary to stretching of the vessel or from
direct compression due to thoracic outlet closure or shoulder
dislocation. Acute ischemia placing the limb at jeopardy in these
neonates is rare. Development of more severe symptoms is re-
lated to damage of collateral arterial flow around the obstructed
vessel. Subclavian artery sacrifice for Blalock-Taussig procedures
rarely causes acute ischemia because the collaterals are carefully
preserved.75 Exercise fatigue may accompany later growth
abnormalities. Diminutions in longitudinal bone growth and
muscle mass are common with subclavian artery occlusion.
Growth disturbance alone is not an indication for surgical inter-
vention, inasmuch as upper extremity limb length discrepancies
are well tolerated and do not carry the same clinical significance
as they do in the lower extremities. Carotid-subclavian or
carotid-axillobrachial bypasses in late childhood are often
required to relieve symptomatic forearm and hand ischemia.
EXTREMITY ARTERIAL ANEURYSMS
Upper and lower extremity arterial aneurysms are rare, being
associated with penetrating and blunt trauma, iatrogenic pseu-
doaneurysms following arterial punctures and catheterizations,
 CHAPTER 126 PEDIATRIC ARTERIAL DISEASES 1643

FIGURE 126-21 Brachial artery aneurysm associated with Kawasaki

disease.

cell disease; structural cardiac defects accompanying congenital

FIGURE 126-20 High-grade stenosis of proximal subclavian artery with heart disease; a remote history of varicella zoster infection; a re-
poststenotic dilation and an axillary artery occlusion associated with birth cent history of head or neck trauma; and various arteritides such
trauma.
as Takayasu disease (Fig. 126-22). A comprehensive investiga-
tion of pediatric patients experiencing a stroke should include
and nontraumatic aneurysms in cases of Kawasaki disease, characterization of the extracranial and intracranial vessels by
Ehlers-Danlos syndrome, and other infrequently encountered
vasculopathies. Aneurysms, regardless of etiology, affecting
the larger arteries of both upper and lower extremities carry a
greater risk of thromboembolism than rupture. Depending on
the presence of collateral vessels, most of these aneurysms
should be treated operatively with either ligation or vascular re-
construction.76
Peripheral aneurysms in Kawasaki disease are of particular
note (Fig. 126-21). The syndrome affects children younger
than 4 years of age in 80% of cases,77 and the most lethal
aspect of this disease relates to its cardiac complications.
Nearly 15% of these children develop coronary aneurysms,
which are prone to thrombose, causing myocardial infarction
and sudden death. Operative interventions for peripheral
aneurysms in children with Ehlers-Danlos syndrome often in-
volve simple ligation with large suture or umbilical tape, with
formal arterial reconstructions undertaken when the ligation
may cause critical limb ischemia. Ultrasound-guided percuta-
neous thrombin injection as a means of occluding traumatic
and iatrogenic pseudoaneurysms has merit, but such must
not result in propagation of thrombus and subsequent severe
limb ischemia.78

Cerebrovascular Disease
------------------------------------------------------------------------------------------------------------------------------------------------

Ischemic strokes in childhood and adolescence are rare and often

multifactorial.79 Contributing to these strokes are numerous FIGURE 126-22 Magnetic resonance angiography of Takayasu disease
prothrombotic states including protein C and S deficiencies, evident by stenoses of left carotid and subclavian (arrowhead) arterial
antithrombin III deficiency, hyperhomocysteinemia, and sickle origins. Irregular narrowing of distal aortic arch (arrow).
1644 PART IX SPECIAL AREAS

MRA, CTA, or a conventional digital subtraction arteriogram.

Such imaging will lessen the chance of overlooking a lesion
amenable to surgical correction.

CAROTID ARTERY OCCLUSION

Carotid artery stenoses or occlusions were present in approx-
imately 50% of ischemic stroke patients studied at a major
children’s hospital.80 Moyamoya disease accounted for nearly
a third of these cases. Moyamoya disease is characterized by
occlusions of the terminal branches of the internal carotid ar-
tery and proximal middle cerebral artery, with a “plume-
of-smoke” appearance of the resulting collaterals that typify
this entity. Moyamoya disease is a noninflammatory vasculitis
of unknown etiology. It is seen most often in children from
Southwest Asia, although all racial groups have been reported
affected. The use of anticoagulants and calcium channel
blockers in the acute setting have a place, although the exact
benefits are ill defined. Most relevant to surgeons is recogni-
tion of the importance of maintaining normotension, euvole-
mia, and normocapnia during anesthesia to lessen the risk of
further cerebral ischemic injury.81–83 The operative revascu-
larization procedures including superficial temporal-middle
cerebral artery bypass or indirect revascularization procedures
are complex and generally lessen the risk of future stroke.84
The role of thrombolysis in children experiencing a stroke
is still undetermined.85 FIGURE 126-23 Internal carotid artery coiling.
Dissections affect the upper cervical region of the internal
carotid artery and are known to follow excessive hyperexten-
sion or forceful rotation of the head and neck.86–88 Spontane- Most children with internal carotid artery coiling are
ous dissections of this type are uncommon in children. asymptomatic. However, neurologic symptoms including
Carotid artery dissections are usually accompanied by severe transient ischemic attacks, hemiparesis, and seizures have
facial and neck pain (hemicrania) and are often associated been alleged to be associated with coiling in certain individ-
with syncope, near syncope, a seizure, or stroke. The majority uals.90 Although the subject of considerable controversy, there
of patients will exhibit Horner syndrome. Treatment is usually may be a small number of children having lateralizing neuro-
conservative, with the use of anticoagulation. The mural he- logic symptoms who may benefit from surgical intervention.91
matomas associated with the dissection frequently resolve, Reimplantation of the internal carotid artery more proximally
leaving a near normal artery. When such does not occur onto the common carotid artery or excision of the redundant
and the child is experiencing cerebral ischemic symptoms, artery with reanastomosis is the most frequent reconstruction
operative therapy is recommended. Resection of the affected undertaken in symptomatic children.
carotid artery and reconstruction with an interposition graft
is appropriate in that setting.
CAROTID ARTERY ANEURYSMS
CAROTID ARTERY COILING
Aneurysms of the proximal extracranial carotid artery are rare
Developmental coiling and tortuosity of the extracranial inter- in children.92 True aneurysms are usually manifestations of a
nal carotid artery have been reported to affect a quarter to generalized vascular disease like Ehlers-Danlos syndrome.
nearly half of young children undergoing arteriographic stud- Traumatic pseudoaneurysms are uncommon and are usually
ies (Fig. 126-23).65,89 The true incidence of this abnormality is the consequence of blunt trauma (Fig. 126-24). Other trau-
likely less because those being studied may represent a se- matic aneurysms represent a complication of tonsillectomy
lected group, but nevertheless these findings are not rare in or, occasionally, errant placement of central venous catheters.
the very young. Most carotid artery aneurysms deserve early operative treat-
Coiling and tortuosity of the internal carotid artery are nor- ment to lessen the risk of cerebral embolization. Simple arte-
mal findings in the fetus. As the heart and great vessels migrate rial suture of the entrance into a pseudoaneurysm or an
caudally, the vessel usually straightens. Occasionally the em- interposition graft, usually with an internal iliac artery seg-
bryonic tortuosity persists. Frequent demonstration of bilat- ment, in the case of extensive aneurysmal involvement is
eral abnormalities lends credence to arrested development appropriate. There may be a role for endoluminal therapy,
as the basis of these lesions. The patterns of internal carotid with placement of covered stents in certain children.93
artery coiling and tortuosity range from gentle S-shaped tortu- Small mural aneurysms affecting the internal carotid artery
osities to complete coiling.89 These changes characteristically adjacent to the upper cervical vertebrae are often due to vio-
begin a few centimeters beyond the common carotid artery lent head trauma with stretching and fracture of the vessel
bifurcation. wall. These may be the source of emboli or thromboses acutely
 CHAPTER 126 PEDIATRIC ARTERIAL DISEASES 1645

causing a massive stroke. Anticoagulant and antiplatelet

therapy lessen these events. Lining the involved artery with
a covered stent or replacement with an interposition graft
may be advised if recurrent cerebral ischemic symptoms occur
despite medical therapy.

VERTEBRAL ARTERY DISSECTIONS

AND OCCLUSION
Injury of the vertebral artery due to forceful stretching or blunt
trauma may result in a dissection causing an acute stroke.88,94
However, many dissections are initially asymptomatic. CTA
imaging in children with severe head and cervical spine
trauma may allow earlier recognition and treatment of these
lesions. Simple administration of heparin and antiplatelet
agents has been successful in many patients.95 Operative
revascularization is usually deferred to symptomatic patients
due to chronic occlusions in whom rotational nystagmus
and other posterior circulatory phenomena can be demon-
strated.

The complete reference list is available online at www.

expertconsult.com.
FIGURE 126-24 Internal carotid artery aneurysm following blunt
trauma.

CHAPTER 127
Congenital Heart
Disease and
Anomalies of the
Great Vessels
Richard G. Ohye and Jennifer C. Hirsch
Congenital cardiac malformations are the most common con-
genital defects, with an incidence of 0.5% to 1% of live births
(excluding trivial lesions and bicuspid aortic valves).
Although the technical details of the repair of complex malfor-
mations are beyond the scope of this textbook, pediatric sur-
geons should have a basic understanding of the anomalies and
their repair because they are often involved in the manage-
ment of these patients. All patients undergoing open cardiac
surgical procedures receive standard gram-positive antibiotic
prophylaxis at the time of surgery. Prophylaxis for future non-
cardiac surgical procedures is based on the current American
Heart Association guidelines recommending prophylaxis for
any patients with significant residual defects or significant
valvar dysfunction. Prophylaxis is not required for patients
who have patch material in place for greater than 6 months.
The pediatric surgeons’ “encounters” with the pediatric
cardiac population may occur in many different contexts:
(1) A cardiac malformation may be present or suspected
in a fetus for whom the pediatric surgeon is consulted for a
general surgical malformation, the prognosis of which may
be affected by the cardiac defect (i.e., giant omphalocele);
(2) intestinal ischemia is often suspected in premature infants
with a patent ductus arteriosus or term babies with cardiac
malformations leading to decreased splanchnic blood flow;
(3) cardiac defects are common in patients with esophageal
atresia, with aortic arch anomalies presenting specific chal-
lenges to pediatric surgeons; (4) vascular anomalies such as
a persistent left superior vena cava may be associated with car-
diac malformations and pose problems during attempts at
central venous access; (5) polysplenia/asplenia and intestinal
malrotation are frequent in patients with situs inversus; and
(6) finally, pediatric surgeons may be involved in postcardiac
surgery ECMO or may be called urgently for insertion of
a peritoneal drainage catheter after an infant has undergone
a cardiac surgical procedure. For all these reasons, pediatric
surgeons should understand the anatomy and pathophysiol-
ogy of common malformations of the heart and great vessels.
Patent Ductus Arteriosus
------------------------------------------------------------------------------------------------------------------------------------------------
The ductus arteriosus is a normal fetal structure that commu-
nicates between the pulmonary trunk or proximal left pul-
monary artery (PA) and the proximal descending thoracic
aorta (Fig. 127-1). During fetal development, this vessel pro-
vides a pathway for blood leaving the right ventricle to bypass
the high-resistance pulmonary vascular bed and traverse the
systemic circulation instead. Histologically, the media of the
aorta and PA contain circumferentially arranged elastic fibers,
whereas the media of the ductus is composed mostly of
smooth muscle cells. Normally after birth, the rise in oxygen
tension that accompanies ventilation of the lungs signals clo-
sure of the ductus. In a full-term infant, closure of the ductus
is usually complete by 12 to 24 hours. Closure of the ductus
creates a fibrous cord called the ligamentum arteriosum. Failure
of closure of the ductus leads to the condition known as patent
ductus arteriosus (PDA). The incidence of isolated PDA is
approximately 1 in 1200 live births and accounts for about
7% of all congenital heart defects.1 The incidence is higher
in preterm infants (>20%).2 This defect may occur in isolation
or in association with a number of other anomalies. In some
heart defects in which there is inadequate pulmonary blood
flow (such as pulmonary atresia) or inadequate systemic blood
flow (e.g., severe coarctation of the aorta), persistent patency
of the ductus is desirable. The discovery of prostaglandins
to maintain ductal patency has played a significant role in
improving the survival of such patients.3
NATURAL HISTORY AND DIAGNOSIS
The pathophysiology of PDA involves shunting of blood
across the ductus. The shunt volume is determined by the size
of the ductus, as well as the ratio of pulmonary to systemic
vascular resistance. Pulmonary vascular resistance drops dra-
matically at birth and continues to decrease over the first
weeks of life. This leads to left-to-right flow across the ductus.
Excessive pulmonary blood flow typically leads to congestive
heart failure, and in extreme cases, hypotension and systemic
malperfusion may also occur. For patients with a large PDA
who survive infancy, there is a risk of developing pulmonary
1647
1648 PART IX SPECIAL AREAS

IA LCCA inhibitors, are useful for stimulating PDA closure in premature

LSA infants.8 Indomethacin and ibuprofen are known to have a
number of possible side effects including hypotension,
decreased gastrointestinal blood flow (which may lead to nec-
rotizing enterocolitis or spontaneous intestinal perforation),
Isthmus
decreased renal blood flow, and interference with platelet
Ao function. Due to an improved side effect profile regarding
gastrointestinal bleeding and renal dysfunction, ibuprofen is
the current drug of choice.9 It is rarely effective in full-term
PDA babies. The dosing regimen is 10 mg/kg intravenously,
PA followed by 5 mg/kg intravenously at 24-hour intervals for
a total of three doses. This approach is successful in nearly
80% of premature infants.9,10 Contraindications to ibuprofen
LPA therapy include sepsis, renal insufficiency, and bleeding dis-
orders. Failure of ibuprofen after two complete courses results
in referral for surgical closure.
The surgical approach to the PDA is usually via a left pos-
terolateral thoracotomy through the third or fourth intercos-
tal space (Fig. 127-2). The pleura is divided longitudinally
over the proximal descending thoracic aorta. The vagus nerve
FIGURE 127-1 Anatomy of patent ductus arteriosus (PDA) as seen from
a left thoracotomy. The ductus extends from the main pulmonary artery is thereby lifted medially. Dissection is carried out to demon-
(PA) and enters the proximal descending thoracic aorta distal to the left strate unequivocally the distal transverse aortic arch and duc-
subclavian artery (LSA). Ao, aorta; IA, innominate artery; LCCA, left common tus. In some cases the ductus may be the largest vascular
carotid artery. (From Hillman ND, Mavroudis C, Backer CL: Patent ductus structure present, so it is critical that it not be confused with
arteriosus. In Mavroudis C, Backer CL [eds]: Pediatric Cardiac Surgery,
3rd ed. Philadelphia, Mosby, 2003.)
the aorta. Once the anatomy is confirmed, the ductus is gently
dissected. Ductal tissue is extremely friable, especially in pre-
mature infants, so direct manipulation is not recommended.
vascular obstructive disease. In these cases, Eisenmenger’s
physiology may develop when pulmonary vascular resistance
exceeds systemic vascular resistance, leading to a reversal of
shunting across the ductus (from left to right to right to left),
causing cyanosis and, eventually, right ventricular failure.
Some patients with a small PDA may remain asymptomatic
until adulthood. Endocarditis and endarteritis have been
reported as long-term complications of PDA.4
The clinical manifestations of PDA are determined by the
shunt volume and the presence of associated cardiac defects.
Left-to-right shunt flow leads to volume overload of the left
heart. Signs of congestive heart failure in infants commonly Ao
PA
include tachypnea, tachycardia, and poor feeding. Older Vagus
children may present with recurrent respiratory infections, Recurrent nerve
fatigue, and failure to thrive. Physical findings include a wid- laryngeal
ened pulse pressure and an active precordium. Auscultation nerve
reveals a continuous “machinery” murmur heard best along
the left upper sternal border. Radiographic findings include
increased pulmonary vascular markings and left heart enlar- PDA
gement generally in proportion to the degree of shunting.
Electrocardiography may demonstrate left ventricular hyper-
trophy and left atrial enlargement. Echocardiography is
currently the diagnostic method of choice and can addition-
ally rule out the presence of associated defects. Cardiac cath-
eterization is reserved for two principal indications. First, in
older patients with suspected pulmonary hypertension, it can
be used to evaluate for pulmonary vascular obstructive dis-
ease. Second, transcatheter techniques have been developed
to occlude the ductus in selected cases.5–7

FIGURE 127-2 Exposure of patent ductus arteriosus (PDA) by left

MANAGEMENT
Three management schemes for the closure of a PDA exist:
pharmacologic therapy, surgical closure, and endovascular
device closure. Indomethacin and ibuprofen, cyclooxygenase
thoracotomy. The mediastinal pleura has been divided and reflected with
preservation of the vagus and recurrent laryngeal nerves. Ao, aorta; PA,
pulmonary artery. (From Hillman ND, Mavroudis C, Backer CL: Patent
ductus arteriosus. In Mavroudis C, Backer CL [eds]: Pediatric Cardiac
Surgery, 3rd ed. Philadelphia, Mosby, 2003.)
 CHAPTER 127 CONGENITAL HEART DISEASE AND ANOMALIES OF THE GREAT VESSELS 1649

The recurrent laryngeal nerve curves behind the ductus and
must be preserved during the dissection. The ductus
may be occluded with a surgical clip or silk ligature, or it
may be divided between ligatures (Fig. 127-3). The latter is
preferred in most cases; however, in premature infants, the
safest approach is clip occlusion. The surgical procedure is
commonly performed in the neonatal intensive care unit,
which avoids problems associated with patient transfer.11,12
Recently, PDA ligation has been performed using video-
assisted thoracoscopy.13 This approach has the potential
benefits of decreased pain and shorter hospital stay. Dis-
advantages include a substantial learning curve and increased
operating time.
Several endovascular devices have been developed for
the purpose of transcatheter occlusion of the PDA.5–7 These
devices have proved to be so successful that at most centers,
transcatheter occlusion has become the treatment of choice
for older infants and children with small- to moderate-sized
PDAs. Surgical therapy is reserved for patients with larger-
sized ducts.
RESULTS
Closure of the PDA by surgical or transcatheter approach
can be performed with a mortality close to zero.11 Potential
morbidity from surgical therapy may include pneumothorax,
recurrent laryngeal nerve injury, phrenic nerve injury, and
chylothorax. Most patients should experience a normal life
expectancy following PDA ligation. Long-term survival in pre-
mature infants depends primarily on the extent of prematurity
and the presence of associated anomalies.

Ao Vagus
nerve
MPA

Recurrent
laryngeal
nerve

PDA

C D
FIGURE 127-3 Various techniques for control of the patent ductus arteriosus (PDA). A, Simple ligation. B, Ligation and hemoclip application. C and D,
Ligation and adventitial pursestring. Ao, aorta; MPA, main pulmonary artery. (From Hillman ND, Mavroudis C, Backer CL: Patent ductus arteriosus.
In Mavroudis C, Backer CL [eds]: Pediatric Cardiac Surgery, 3rd ed. Philadelphia, Mosby, 2003.)
1650 PART IX SPECIAL AREAS

Coarctation of the Aorta

------------------------------------------------------------------------------------------------------------------------------------------------
Coarctation can usually be diagnosed clinically. The new-
born infant with significant coarctation may appear normal
Coarctation of the aorta is a congenital narrowing of the at birth but following ductal closure develops signs of heart
proximal descending thoracic aorta near the insertion of the failure such as irritability, tachypnea, and poor feeding.
ductus arteriosus (or ligamentum arteriosum). This creates Lower-extremity pulses are absent, and upper-extremity pulses
obstruction to blood flow and results in a pressure load on may be weak. Chest radiography may show cardiomegaly
the left ventricle. The incidence of coarctation is about 0.5 and pulmonary venous congestion. There is a left ventricular
per 1000 live births, and its prevalence is 5% of congenital strain pattern on the electrocardiogram. Echocardiography is
heart defects.1 Coarctation is commonly associated with other usually diagnostic, demonstrating anatomic narrowing at
heart defects including bicuspid aortic valve (in > 50% of the coarctation site with a loss of pulsatility in the descending
cases), patent ductus arteriosus, and ventricular septal defect. aorta; additional cardiac defects are also demonstrable.
Other left-sided obstructive lesions may be present including In older children and adults, there is usually a pressure
aortic arch hypoplasia, aortic stenosis, mitral stenosis, and left gradient between the arms and legs, which can be identified
ventricular hypoplasia. by measuring cuff pressures in all four extremities. Radio-
graphy may demonstrate rib notching, which is secondary
to the development of large intercostal collaterals that erode
NATURAL HISTORY AND DIAGNOSIS into the inferior aspects of the ribs. Echocardiography is
usually sufficient to confirm the diagnosis in older patients;
Two patterns of presentation are based on anatomy, age, and however, computed tomography (CT) and magnetic reso-
symptoms (Fig. 127-4). In infantile coarctation, patients de- nance imaging (MRI) provide excellent anatomic detail, which
velop symptoms in the first week of life. These infants have may aid in therapeutic planning. Cardiac catheterization is
such severe aortic obstruction that perfusion of the lower body usually not necessary.
depends on flow from the ductus arteriosus. Spontaneous
ductal closure (which also tends to worsen the aortic obstruc-
MANAGEMENT
tion) then precipitates ischemia to tissues beyond the coarc-
tation. The resultant pressure load on the left ventricle may The diagnosis of coarctation usually mandates surgical cor-
lead to congestive heart failure. Patients may exhibit shock rection; however, the acute medical management of the sick
with severe acidosis, oliguria, and diminished distal pulses. neonate demands careful attention. Prostaglandin E1 is usually
Survival for these patients is unlikely without intervention. effective for reopening the ductus when administered within
The second pattern of presentation is sometimes called 7 to 10 days after birth; its effectiveness wanes when initiated
adult coarctation. Patients with adult coarctation are usually after 2 weeks of life. Intravenous fluids, inotropic agents,
asymptomatic during infancy and present later in life with and correction of anemia may also be important for the resus-
hypertension. These patients generally develop extensive col- citation of the patient with coarctation.
laterals, which serve to bypass the obstruction. Life expec- Surgical exposure of coarctation is performed by left
tancy for these patients is limited due to the development of posterolateral thoracotomy through the third or fourth inter-
heart failure later in life and the cumulative risk of endocardi- costal space. The mediastinal pleura overlying the left sub-
tis (frequently involving a bicuspid aortic valve) or endarteritis clavian artery and proximal descending thoracic aorta is
(in the poststenotic aorta at the site of the turbulent jet), aortic incised, and the descending aorta, transverse arch, brachioce-
rupture, or intracranial hemorrhage (from Berry aneurysms, phalic vessels, and ductus (or ligamentum) are mobilized.
which are more common in patients with coarctation).14 Care is taken to avoid injury to the vagus nerve and its

Coarctation Coarctation
Ao
Ao

PDA
PA

PA

Ligamentum
A B arteriosum

FIGURE 127-4 A, Infantile coarctation of the aorta (Ao). The isthmus (segment of the aorta between the left subclavian artery and ductal insertion)
is hypoplastic, and the descending aorta receives most of its flow from the ductus. B, Adult coarctation with juxtaductal narrowing of the aorta and a
prominent posterior shelf. The ductus has contracted to form the ligamentum arteriosum. PA, pulmonary artery; PDA, patent ductus arteriosus. (From
Backer CL, Mavroudis C: Coarctation of the aorta. In Mavroudis C, Backer CL [eds]: Pediatric Cardiac Surgery, 3rd ed. Philadelphia, Mosby, 2003.)
 CHAPTER 127 CONGENITAL HEART DISEASE AND ANOMALIES OF THE GREAT VESSELS 1651

recurrent branch. The coarctation is usually evident externally
by narrowing or posterior indentation; however, the degree of
internal narrowing is usually much more severe. A dose of
heparin (100 units/kg) may be given intravenously for patients
younger than 2 years. Proximal and distal control of the aorta
is achieved using clamps. Usually the proximal clamp is posi-
tioned on the transverse arch between the innominate and left
carotid vessels with concomitant occlusion of the left carotid
and left subclavian. In infants and children, the preferred
surgical approach to coarctation is resection with extended
end-to-end repair (Fig. 127-5).15 A generous resection of
the coarctation segment is performed. The proximal aorta
is then spatulated along the lesser curvature and the distal
aorta along the greater curvature. An extended end-to-end
anastomosis is then performed.
In older children and adults, it may not be possible to per-
form a resection with primary repair without creating exces-
sive tension on the anastomosis, which might lead to
hemorrhage or scarring with recurrent coarctation. In these
cases, an alternative strategy is indicated. In children, in
whom further growth is expected, the coarctation may be
repaired by patch aortoplasty.16,17 Less initial dissection is re-
quired. After achieving proximal and distal control, a longitu-
dinal aortotomy is made across the area of narrowing. A patch
of polytetrafluoroethylene or the left subclavian artery can be
used to patch augment the coarctation. By avoiding circumfer-
ential prosthetic material, growth potential of the native aortic
tissue is preserved. In adults, growth is no longer an issue and
resection of the coarctation may be performed with sub-
sequent placement of an interposition graft (either polytetra-
fluoroethylene or Dacron).
During the operative repair of coarctation, injury to the spinal
cord is a major concern. In patients who do not have well-formed
collaterals, ischemia of the spinal cord may be precipitated by
aortic cross-clamping and paraplegia may result. Standard
protective measures include induction of mild hypothermia
(35 C), maintenance of a high proximal aortic pressure, and
minimization of cross-clamp time. In older children and adults

PA Ao

Ao
PA Vagus
nerve
PDA
LCCA

LSA B
Recurrent
laryngeal
nerve

Ao
PA

Ao

Divided
intercostal
collaterals D

FIGURE 127-5 Resection with extended end-to-end anastomosis. A, Exposure by left posterolateral thoracotomy. Vessels are mobilized and
intercostals divided when necessary. B, The ductus is ligated and divided. Proximal and distal control is achieved, and the coarctation segment is resected.
The proximal and distal aortic segments are spatulated along the lesser and greater curvatures, respectively. C, The anastomosis is performed using a
running technique. D, The completed repair. Ao, aorta; LCCA, left common carotid artery; LSA, left subclavian artery; PA, pulmonary artery; PDA, patent
ductus arteriosus. (From Backer CL, Mavroudis C: Coarctation of the aorta. In Mavroudis C, Backer CL [eds]: Pediatric Cardiac Surgery, 3rd ed. Philadelphia,
Mosby, 2003.)
1652 PART IX SPECIAL AREAS
with inadequate collateral flow, distal aortic perfusion may be
maintained by the technique of left heart bypass (in which
the left atrium is cannulated for drainage of oxygenated blood,
which is then delivered to the femoral artery or distal aorta using
a centrifugal pump).18 Overall, the incidence of paraplegia
following coarctation repair is less than 1%.19
Following repair, patients may develop severe hyper-
tension. This can be managed using intravenous beta blockers
(such as esmolol). Uncontrolled hypertension may lead to the
complication of mesenteric arteritis. Hypertension usually
resolves within days to weeks after repair, although older
children and adults may require permanent antihypertensive
therapy. Repair of coarctation during infancy is thought to
minimize the risk of late hypertension.20
Transcatheter therapy has been attempted as primary
therapy for coarctation, but this approach is controver-
sial.21–24 Limitations of this approach include the incidence
of recurrent coarctation, the potential need for multiple
interventions, injury to the femoral vasculature (for access),
and incidence of aneurysm formation. Technical success rates
range from 80% to 98% with a reintervention rate of 10% to
20%.25,26 Balloon angioplasty is generally recommended for
the treatment of recurrent coarctation following surgery, in
which its efficacy is on the order of 90%.27–29
RESULTS
The early mortality following repair of coarctation in neonates
is 2% to 10%, while the risk in older children and adults
is about 1%.15,30 The incidence of recurrent coarctation
Heavily trabeculated
portion of RA
RA appendage
Ao
SVC
Crista terminalis
FIGURE 127-6 Normal intraatrial anatomy as viewed by the surgeon through a right atriotomy. Ao, aorta; IVC, inferior vena cava; SVC, superior vena cava;
RA, right atrium. (From Backer CL, Mavroudis C: Atrial septal defect, partial anomalous pulmonary venous connection, and scimitar syndrome. In Mavroudis C,
Backer CL [eds]: Pediatric Cardiac Surgery, 3rd ed. Philadelphia, Mosby, 2003.)
following resection and end-to-end repair is about 4% to
8%.15,31 The long-term survival following coarctation repair
is determined by the presence of associated defects and the
persistence of hypertension.
Atrial Septal Defects
------------------------------------------------------------------------------------------------------------------------------------------------
Atrial septal defects (ASDs) are the third most common
congenital heart defect, occurring in 1 out of 1000 live births
and representing 10% of congenital heart defects.1 In order to
understand the terminology of ASDs, a review of normal atrial
septation is useful. During embryologic development, the
septum primum initially divides the common atrium. The os-
tium primum, which exists at the inferior edge of the septum
primum, is obliterated as the septum primum fuses with the
endocardial cushions. The ostium secundum forms in the
midportion of the septum primum and is then covered by
the septum secundum, which descends from the roof of the
atrium along the right side of the septum primum. This ar-
rangement creates a flap valve whereby blood from the inferior
vena cava may preferentially stream beneath the edge of the
septum secundum and through the ostium secundum into
the left atrium. Following birth, the septum secundum fuses
with the septum primum, thereby sealing the interatrial
communication (Fig. 127-6).
Atrial septal defects may be classified on the basis of this
embryology. The most common defect is the secundum atrial
septal defect (80%), which occurs when the ostium secundum
is too large for complete coverage by the septum secundum
Tricuspid
valve
Anterior
Posterior
Coronary sinus
Septal
Thebesian valve
Eustachian valve
IVC
Septum primum;
fossa ovalis
Septum Location of patent foramen ovale
secundum
 CHAPTER 127 CONGENITAL HEART DISEASE AND ANOMALIES OF THE GREAT VESSELS
Septum
Septum primum
secundum
Ostium
secundum
FIGURE 127-7 Ostium secundum atrial septal defect. Note the deficiency
of the septum primum in the region of the normal fossa ovalis. (From Backer
CL, Mavroudis C: Atrial septal defect, partial anomalous pulmonary venous
connection, and scimitar syndrome. In Mavroudis C, Backer CL [eds]:
Pediatric Cardiac Surgery, 3rd ed. Philadelphia, Mosby, 2003.)
(Fig. 127-7). Failure of postnatal fusion of the septum secun-
dum to the septum primum results in a persistent slitlike
communication known as a patent foramen ovale (PFO).
A primum atrial septal defect (10%) represents persistence
of the ostium primum, which occurs as a result of failure of
fusion of the septum primum with the endocardial cushions.
This defect is discussed in the section on atrioventricular sep-
tal defects. Sinus venosus ASDs (10%) are caused by abnormal
fusion of the venous pathways with the atrium; these defects
can be located high in the atrial septum near the orifice of the
Sinus venosus ASD
B RSPVs RIPV
SVC
Azygos vein
Anomalous RSPVs
A
FIGURE 127-8 Sinus venosus atrial septal defect. A, Defect near the superior cavoatrial junction cephalad to the superior limbic band. Note the
anomalous drainage of right superior pulmonary veins (RSPVs) to the superior vena cava (SVC) and normal drainage of the right inferior pulmonary
veins (RIPVs). B, Another type of sinus venosus ASD with anomalous RSPVs draining adjacent to the defect. (From Backer CL, Mavroudis C: Atrial septal
defect, partial anomalous pulmonary venous connection, and scimitar syndrome. In Mavroudis C, Backer CL [eds]: Pediatric Cardiac Surgery, 3rd ed.
Philadelphia, Mosby, 2003.)
1653
superior vena cava, or, less commonly, they may be low in the
atrial septum near the orifice of the inferior vena cava
(Fig. 127-8). Sinus venosus ASDs occur commonly with par-
tial anomalous pulmonary venous return, usually with the
right upper pulmonary vein draining into the superior vena
cava near the cavoatrial junction. An uncommon type of
ASD is the unroofed coronary sinus septal defect. This occurs
when there is loss of the common wall between the coronary
sinus and left atrium adjacent to the atrial septum. This
unroofing of the coronary sinus results in a communication
between the right and left atria at the site of the coronary sinus.
NATURAL HISTORY AND DIAGNOSIS
Shunting at the atrial level occurs primarily during diastole
and is determined in part by the size of the atrial defect and
more importantly by the relative ventricular compliance.
In other words, blood preferentially fills the more compliant
chamber. At birth, both chambers are equally compliant,
but as pulmonary vascular resistance falls, the right ventricle
remodels and becomes more compliant. As a result, shunting
across an ASD in postnatal life is from left to right. This results
in a volume load on the right heart. A volume load is a burden
created by additional venous return to a chamber during
diastole.
The volume overload created by an ASD is usually well
tolerated. As a result, patients are frequently asymptomatic.
Congestive heart failure is rare in infants. Children may
develop symptoms of exertional dyspnea, exercise intolerance,
or recurrent respiratory infections. Older patients with un-
treated ASDs tend to develop atrial dysrhythmias, and adults
may develop congestive heart failure and right ventricular
dysfunction. Pulmonary vascular obstructive disease may de-
velop as a late complication of untreated ASD; severe pulmo-
nary hypertension has been shown to develop in 14% of adults
Sinus venosus ASD
Fossa
ovalis
Superior
limbic
band RIPV
1654 PART IX SPECIAL AREAS
with untreated large ASDs between the ages of 20 and
40 years.32 Paradoxical embolization (deep venous thrombus
passing to the systemic circulation via the ASD) is also an
important potential complication of ASD.
On physical examination, a patient with an ASD has fixed
splitting of the second heart sound and a systolic ejection
murmur at the left upper sternal border due to relative pulmo-
nary stenosis (increased flow across a normal pulmonary
valve). A chest radiograph shows cardiomegaly, and electro-
cardiography frequently demonstrates incomplete right
bundle branch block. Echocardiography confirms the clinical
diagnosis of ASD and clarifies the anatomy. Cardiac catheter-
ization is rarely necessary for diagnostic purposes but may be
necessary to evaluate pulmonary vascular resistance in older
patients. Cardiac catheterization is most widely used with
therapeutic intent (device closure).
MANAGEMENT
Due to the long-term complications associated with ASD,
repair is recommended for all patients with symptomatic
defects and in asymptomatic patients in whom the ratio of
pulmonary to systemic blood flow (Qp/Qs) is greater than
1.5. Repair is usually performed in children before school
age. Closure of ASDs may be accomplished surgically or using
a device deployed in the cardiac catheterization laboratory.
Surgical repair of ASDs is relatively straightforward
(Fig. 127-9). Surgical closure is recommended for large
secundum defects and most other types of ASDs. The heart
is exposed by median sternotomy. Cardiopulmonary bypass
is necessary using bicaval cannulation and mild hypothermia.
Following aortic clamping and arrest of the heart with cardio-
plegia, the atrial septum is exposed through a right atriotomy.
The defect is then closed using a patch (polytetrafluoroethylene
or autologous pericardium) with a running polypropylene
Coronary
sinus
SVC
IVC
Patch: Pericardium or PTFE
FIGURE 127-9 Patch closure of an ostium secundum atrial septal defect.
Bicaval cannulation is used, and cardioplegia is infused through the aortic
root to arrest the heart following application of the aortic crossclamp. IVC,
inferior vena cava; SVC, superior vena cava. (From Backer CL, Mavroudis C:
Atrial septal defect, partial anomalous pulmonary venous connection,
and scimitar syndrome. In Mavroudis C, Backer CL [eds]: Pediatric Cardiac
Surgery, 3rd ed. Philadelphia, Mosby, 2003.)
suture. If there is anomalous pulmonary venous return,
a baffle may be created to redirect the flow across the atrial sep-
tal defect into the left atrium. Small defects and PFOs can be
closed primarily without a patch. In all cases, care is taken to
de-air the left atrium to avoid the complication of air
embolization.
Small-to-moderate secundum ASDs and PFOs can now be
closed using transcatheter techniques. Device closure of these
defects has now become the standard of care at most centers
with a device deployment rate of 96% with complete closure at
24 hours of 99% or greater.33,34
RESULTS
ASD closure by either approach is successful.35 Mortality
is less than 1%. Morbidity is infrequent. The long-term
survival for patients undergoing ASD repair in childhood is
normal.36,37
Ventricular Septal Defects
------------------------------------------------------------------------------------------------------------------------------------------------
Ventricular septal defects (VSDs) are the most common con-
genital heart anomalies (with the exception of bicuspid aortic
valve), occurring in 4 of 1000 live births and representing
about 40% of congenital heart defects.1 There is heterogeneity
with respect to the location, size, and number of defects pre-
sent. Associated cardiac defects are common. VSDs occur as a
result of failure of ventricular septation. They are most com-
monly classified on the basis of their location: perimembra-
nous (80%), inlet (5%), outlet (10%), or trabecular (5%)
(Fig. 127-10). The most common, perimembranous VSDs,
are located in the area of the membranous septum, near the
point of contact of the tricuspid, mitral, and aortic valve an-
nuli. Inlet defects are located beneath the septal leaflet of
the tricuspid valve. Outlet defects are also known as supracris-
tal or doubly committed subarterial. These defects are bordered
superiorly by both semilunar valves. Outlet defects are more
common in the Asian population. Trabecular (or muscular)
VSDs are completely bordered by muscle. They are frequently
multiple and may be associated with perimembranous or
outlet defects. The size of VSDs varies. A VSD is defined as
nonrestrictive when its size (or the cumulative size of multiple
defects) approximates that of the aortic annulus.
NATURAL HISTORY AND DIAGNOSIS
The physiology of a VSD involves left-to-right shunting pri-
marily during systole. This creates a volume load on the left
heart (the left atrium and ventricle receive the increased ve-
nous return during diastole). The right ventricle is not volume
loaded (blood is ejected from the left ventricle through the
VSD and directly into the pulmonary circulation); however,
it does experience a pressure load. Shunt volume across a
VSD is determined by the size of the defect and by the ratio
of pulmonary-to-systemic vascular resistance. As pulmonary
vascular resistance decreases during the first few weeks of
life, the shunting across a VSD tends to increase. Therefore
a VSD that was asymptomatic at birth may eventually cause
severe congestive heart failure.
The natural history of patients with VSD is variable. Most
VSDs are restrictive and tend to close spontaneously during
 CHAPTER 127 CONGENITAL HEART DISEASE AND ANOMALIES OF THE GREAT VESSELS 1655

A B

C D
FIGURE 127-10 Types of ventricular septal defects (VSDs) based on location. A, Outlet (supracristal, doubly committed subarterial). B, Perimembranous.
C, Inlet. D, Trabecular (muscular). (From Mavroudis C, Backer CL, Jacobs JP: Ventricular septal defect. In Mavroudis C, Backer CL [eds]: Pediatric Cardiac
Surgery, 3rd ed. Philadelphia, Mosby, 2003.)

the first year of life.38 Patients with large VSDs develop symptoms
caused by congestive heart failure. Untreated, excessive pulmo-
nary blood flow leads to pulmonary vascular obstructive disease
by 1 to 2 years of age. Smaller VSDs may remain asymptomatic.
In patients with outlet VSDs, proximity of the aortic valve may
lead to prolapse of a valve cusp and aortic insufficiency.39
Signs of heart failure in infants with large VSDs include
tachypnea, hepatomegaly, poor feeding, and failure to thrive.
On examination, there is a holosystolic murmur at the left ster-
nal border. The precordium is active. Usually, the murmur is
louder with smaller defects. In patients with pulmonary vas-
cular obstructive disease, there may be no murmur present
but the pulmonary component of the second heart sound is
prominent. Chest radiography shows increased pulmonary
vascular markings and cardiomegaly. Electrocardiography is
significant for right ventricular hypertrophy.
Patients with small VSDs have little shunting and are
usually asymptomatic, having only a pansystolic murmur.
Patients with moderate VSDs manifest symptoms and signs
that are proportional to the degree of shunting.
Echocardiography is diagnostic in most cases. The anatomy
can be accurately defined, and the presence of associated
abnormalities can be excluded. Cardiac catheterization is
used selectively in older children and adults with VSDs to
determine pulmonary vascular resistance and assess reactivity
to pulmonary vasodilators.
MANAGEMENT
Treatment of a patient with a VSD depends on the size of the
defect, type of defect, shunt volume, and pulmonary vascular
resistance. In general, patients with large defects who
1656 PART IX SPECIAL AREAS
demonstrate intractable congestive heart failure (CHF) or
failure to thrive should undergo early surgical repair. If the
congestive symptoms can be moderated by medical therapy,
surgery can be delayed to 6 months of age. Patients with mod-
erate defects may be safely followed. If spontaneous closure
has not occurred by school age, then surgery may be consid-
ered. Small VSDs with Qp/Qs less than 1.5 do not require
closure. In these patients there is a small long-term risk of
endocarditis, but this is minimized by appropriate antibiotic
prophylaxis.40 As noted previously, patients with outlet VSDs
have a significant risk of developing aortic insufficiency
secondary to leaflet prolapse; as a result, all outlet defects
should be referred for closure.41 Older children and adults with
large VSDs who present late must undergo catheterization
to evaluate the pulmonary vasculature. A fixed pulmonary
vascular resistance greater than 8 to 10 Woods units/m2
represents a contraindication to surgical closure of the VSD.
Surgical closure is performed through a median sternot-
omy (Fig. 127-11). Cardiopulmonary bypass is required,
using bicaval cannulation. Following delivery of cardioplegia,
a right atriotomy is performed. Exposure of the ventricular
septum is achieved through the tricuspid valve. This provides
access to perimembranous, inlet, and most outlet defects.
Most trabecular defects may also be exposed in this fashion.
Some outlet VSDs are best exposed via a pulmonary arteriot-
omy because the defect sits just beneath the valve. Trabecular
VSDs located near the ventricular apex are notoriously
difficult to expose, and an apical ventriculotomy (left or right)
may be necessary. Once the defect is exposed, it is generally
closed using a polytetrafluoroethylene patch and a running
polypropylene suture, although other centers may prefer other
Muscle of
Lancisi
Right
bundle
branch
VSD
AV
node
FIGURE 127-11 Transatrial exposure of a perimembranous ventricular septal defect (VSD) as viewed by the surgeon. The septal and anterior leaflets of the
tricuspid valve are retracted to expose the VSD. The conduction tissue is demonstrated by dashed lines leading from the atrioventricular (AV) node along the
posterior and inferior margin of the VSD. When closing these defects, sutures are placed a few millimeters away from the edge of the defect on the right
ventricular side of the septum to avoid conduction injury. (From Mavroudis C, Backer CL, Jacobs JP: Ventricular septal defect. In Mavroudis C, Backer CL [eds]:
Pediatric Cardiac Surgery, 3rd ed. Philadelphia, Mosby, 2003.)
patch material or an interrupted suture technique. Knowledge
of the anatomy of the conduction tissue is critical when closing
VSDs. The atrioventricular node is an atrial structure that lies
in the apex of the triangle of Koch (formed by the coronary
sinus, the tendon of Todaro, and the annular attachment of
the septal leaflet of the tricuspid valve). The node then gives
rise to the bundle of His, which penetrates the atrioventricular
junction beneath the membranous septum. The bundle then
bifurcates into right and left bundle branches, which descend
along either side of the muscular ventricular septum. When a
perimembranous VSD is present, the bundle of His traverses
the posterior and inferior rim of the defect, generally on the
left ventricular side. In this danger area, sutures must be
placed superficially on the right ventricular side of the defect
and a few millimeters away from the edge of the defect. The
His bundle also tends to run along the posterior and inferior
margin of inlet VSDs, whereas in outlet and trabecular defects,
the conduction tissue is remote.
PA banding is a palliative maneuver that is designed to pro-
tect the pulmonary circulation from excessive flow in circum-
stances when the patient is not a candidate for a surgical
closure, either due to associated illness or anatomic complex-
ity (e.g., multiple trabecular VSDs). Placement of a PA band
may be accomplished via sternotomy or thoracotomy. The
band is tightened, and pressures are measured proximally
and distally with a goal of achieving a distal pulmonary pres-
sure of approximately one-half systemic. The band should
then be secured to the adventitia of the main PA to prevent
its migration. Distal migration may result in narrowing (and
poor growth) of one or both branch pulmonary arteries,
whereas proximal migration may lead to deformity of the
Tricuspid
valve:
Anterior
Posterior
Septal
Fossa
ovalis
Coronary
sinus
 CHAPTER 127 CONGENITAL HEART DISEASE AND ANOMALIES OF THE GREAT VESSELS 1657

pulmonary valve. Ultimately, when the patient is a candidate results in variable deficiency of the atrial and ventricular septa
for VSD closure, the band must be removed. In most cases, a and malformation of the AV valves. Atrioventricular septal
repair (scar resection and primary closure or patch repair) of defects are generally categorized into incomplete and com-
the main PA will be necessary. plete on the basis of the AV valve morphology. Incomplete,
Recently, transcatheter technology has evolved and closure or partial, defects have two separate AV valve orifices
of some VSDs may be accomplished in the cardiac catheteri- (Fig. 127-12). The mitral valve in an incomplete AVSD is in-
zation laboratory or by periventricular deployment without variably associated with a cleft in the anterior leaflet. Although
the use of cardiopulmonary bypass.42,43 Device closure is most incomplete AVSDs have no ventricular level shunting,
optimal for large muscular defects with sufficient rims for the classification of AVSDs as complete and incomplete
securing the device. Periventricular closure allows the use of depends only on the valve anatomy, not on the presence or
VSD device closure in smaller infants with symptomatic mus- absence of a VSD. Incomplete defects without associated ven-
cular VSDs with excellent results without the challenges of tricular level shunting have also been termed ostium primum
transvenous access in this population.43 The use of devices ASDs, whereas those with a VSD have been described as inter-
for perimembranous VSDs remains limited by the risk of dam- mediate or transitional AVSDs. Complete AVSDs have a single
age to the conduction system or impingement on the function common AV valve orifice resulting in a single five-leaflet valve
of the tricuspid or aortic valves. A recent series demonstrated overlying both the right and left ventricles (Fig. 127-13).
a 22% rate of complete heart block, which is prohibitively Rastelli and colleagues47 further subclassified complete
high in comparison with the surgical rate of complete heart AVSDs into types A, B, and C on the basis of the morphology
block of less than 1%.44 of the superior bridging leaflet (SBL) of the common valve.
If both left and right AV valves equally share the common
AV valve orifice, the AVSD is termed a balanced defect. Occa-
RESULTS
sionally the orifice of the valve may be unbalanced, with
Repair of a VSD is associated with a mortality of approximately one side being dominant. In marked right dominance, the left
1%.45 The main complications include injury to the con- AV valve and left ventricle are hypoplastic and frequently
duction tissue and injury to the tricuspid or aortic valves. coexist with other left-sided abnormalities including aortic
Transient heart block may develop as a result of tissue swell- stenosis, hypoplasia of the aorta, and coarctation. Conversely,
ing, but permanent heart block occurs in only 1% to 2% of marked left dominance is associated hypoplasia of the
cases.45 Patients who develop heart block after surgery are right ventricle, pulmonary stenosis or atresia, and tetralogy
observed for a period of 7 to 10 days. If normal conduction of Fallot.
has not returned, then pacemaker implantation is indicated. The location of the conduction tissue is of importance in
Tricuspid insufficiency may be precipitated by annular distor- the surgical treatment of AVSDs because it is at risk during
tion or chordal restriction by the VSD patch or sutures. The the repair. The AV node is displaced posteriorly and inferiorly
aortic valve may also be injured by inaccurate suturing toward the coronary sinus. The bundle of His courses anteri-
(especially in perimembranous and outlet defects). A residual orly and superiorly to run along the leftward aspect of the
VSD may be observed in 5% of cases, and reoperation is crest of the VSD, giving off the left bundle branch before
indicated when significant shunting persists (Qp/Qs > 1.5). continuing as the right bundle branch.
Intraoperative echocardiography is used routinely to identify
valvar dysfunction and residual shunting.

ASD Cleft
mitral
Atrioventricular Septal Defect
------------------------------------------------------------------------------------------------------------------------------------------------
LA valve

Atrioventricular septal defects (AVSD) represent a group of

congenital abnormalities bound by a variable deficiency of RA
the atrioventricular (AV) septum immediately above and
below the AV valves. Other terms commonly applied to an LV
AVSD include atrioventricular canal defects, endocardial cushion
defects, and atrioventricular communis. These resulting septal
defects are invariably associated with AV valve abnormalities.
Atrioventricular septal defects include incomplete AVSDs, also
termed ostium primum atrial septal defects, which have only a
deficiency of the atrial septum immediately superior to the RV
AV valves and two separate valve orifices. The far end of the
spectrum encompasses complete AVSDs, with both an ASD
and a VSD and a single common AV valve. AVSDs represent
approximately 4% of congenital cardiac anomalies and are
infrequently associated with other cardiac malformations.
AVSDs comprise 30% to 40% of the cardiac abnormalities FIGURE 127-12 Incomplete atrioventricular septal defect. ASD, primum
atrial septal defect; LA, left atrium; LV, left ventricle; RA, right atrium; RV, right
seen in patients with Down syndrome.46 ventricle. (From Backer CL, Mavroudis C: Atrioventricular canal defects. In
The embryologic abnormality in AVSDs is the failure of Mavroudis C, Backer CL [eds]: Pediatric Cardiac Surgery, Philadelphia,
the proper development of the endocardial cushions, which Mosby, 2003.)
1658 PART IX SPECIAL AREAS
ASD Common
AV valve
LA
RA
LV
RV
VSD
FIGURE 127-13 Complete atrioventricular septal defect. ASD, primum
atrial septal defect; AV, atrioventricular LA, left atrium; LV, left
ventricle; RA, right atrium; RV, right ventricle; VSD, inlet ventricular septal
defect. (From Backer CL, Mavroudis C. Atrioventricular canal defects. In
Mavroudis C, Backer CL [eds]: Pediatric Cardiac Surgery, Philadelphia,
Mosby, 2003.)
Cardiac anomalies associated with AVSDs include a patent
ductus arteriosus (10%) and tetralogy of Fallot (10%).48
Important abnormalities of the left AV valve include single
papillary muscle (parachute mitral valve) (2% to 6%) and
double orifice mitral valve (8% to 14%).49 A persistent left su-
perior vena cava with or without an unroofed coronary sinus is
encountered in 3% of patients with an AVSD. A double-outlet
right ventricle (2%) significantly complicates or may even
preclude complete surgical correction.48 Left ventricular out-
flow tract obstruction from subaortic stenosis or redundant
AV valve tissue occurs in 4% to 7%.50,51 Associated trans-
position of the great arteries and left ventricular inflow
obstruction have been reported rarely.48,50
NATURAL HISTORY AND DIAGNOSIS
The predominant hemodynamic features of an AVSD are the
result of left-to-right shunting at the atrial and ventricular
levels. In the absence of ventricular level shunting, the hemo-
dynamics and clinical presentation resemble that of a typical
secundum ASD with right atrial (RA) and right ventricular
(RV) volume overload. On physical examination, there is an
active precordium, a pulmonary outflow murmur, and a fixed,
widely split second heart sound. Electrocardiogram reveals a
leftward axis, prominent P waves associated with atrial
enlargement, and a prolonged PR interval. Chest radiograph
generally shows mild cardiomegaly and increased pulmonary
vascular markings. Echocardiography is diagnostic of the
atrial septal defect, the absence of ventricular level shunting,
and the presence of any AV valve abnormalities. Cardiac cath-
eterization is only indicated in older patients or those mani-
festing physical or radiologic signs of decreased pulmonary
blood flow. The decreased PA blood flow may be a result
of pulmonary vascular disease or concurrent right-sided
obstructive lesions. A fixed pulmonary vascular resistance of
greater than 10 units/m2 is a contraindication to surgical
closure.
As with an uncomplicated ASD, the natural history of
decades of chronic volume overload results in atrial dilatation
and arrhythmias, ventricular dysfunction, and potentially
pulmonary vascular disease. Therefore repair is indicated
and generally undertaken by age 2 to 4 years.
Patients with a complete AVSD often have both atrial and
ventricular level shunting. These patients generally present
early in infancy with signs and symptoms of CHF. Moderate
or greater left AV valve regurgitation may occur with a
complete AVSD, worsening the clinical picture. On physical
examination, the precordium is hyperactive, with auscultatory
findings of a systolic murmur along the left sternal border, a
high-pitched murmur at the apex from left AV valve regurgi-
tation, and a mid-diastolic flow murmur across the common
AV valve. Chest radiograph demonstrates significant cardio-
megaly and pulmonary overcirculation. Electrocardiogram
reveals biventricular hypertrophy, atrial enlargement, pro-
longed PR interval, leftward axis, and counterclockwise
frontal plane loop. Echocardiography is diagnostic, defining
the atrial and ventricular level shunting, valvular anatomy,
and any associated anomalies. Cardiac catheterization should
be performed for patients older than the age of 1 year, patients
with signs or symptoms of increased pulmonary vascular
resistance, or in select cases to further evaluate other asso-
ciated major cardiac anomalies.
Up to 90% of untreated individuals with a complete
AVSD develop pulmonary vascular disease by 1 year of age
due to the large left-to-right shunt, potentially exacerbated
by the associated AV valve regurgitation.52 Patients with
trisomy 21 tend to develop pulmonary vascular obstructive
disease earlier than chromosomally normal infants due to
small airway disease, chronic hypoventilation, and elevated
pCO2. Initial aggressive medical management is undertaken
to relieve the symptoms of CHF. Elective surgical correction
should be performed by age 3 to 6 months. Earlier inter-
vention is indicated for failure of medical management.
MANAGEMENT
The treatment of choice for an incomplete or complete AVSD
is surgical repair. Currently, PA banding for palliation has a
limited role in the management of these lesions. Indications
for PA banding may include those patients with associated
complex cardiac anomalies, functional single ventricle anat-
omy necessitating ultimate Fontan procedure, and poor
clinical condition precluding major cardiac surgery.
A median sternotomy approach is employed, with routine
conduct of cardiopulmonary bypass for the majority of patients.
Repair of an incomplete AVSD with only atrial level shunting is
similar to the approach employed for a secundum-type ASD.
There are two techniques widely employed for the repair
of complete AVSDs, a one-patch technique and a two-patch
technique. Regardless of which approach is selected, the goals
are to close the ASD and VSD and to separate the common
AV valve into two nonstenotic, competent valves. The cleft
in the anterior leaflet of the mitral valve is generally closed
to lessen the risk of long-term mitral regurgitation.53–55
For the two-patch technique, separate patches are used
for the ASD and VSD (Fig. 127-14, A). For the one-patch
 CHAPTER 127 CONGENITAL HEART DISEASE AND ANOMALIES OF THE GREAT VESSELS 1659

Single-patch Two-patch Modified

A B C single-patch
FIGURE 127-14 Surgical repair of an atrioventricular septal defect using a single-patch (A), two-patch (B), and modified single-patch
(C) technique. (From Backer CL, Mavroudis C. Atrioventricular canal defects. In Mavroudis C, Backer CL [eds]: Pediatric Cardiac Surgery, Philadelphia,
Mosby, 2003.)

technique, the superior and inferior bridging leaflets are

divided along a line separating them into right and left
components. A single patch is used to close both the ventric- MPT
ular and atrial septal defects. The cut edges of the leaflets are
then resuspended to the patch (see Fig. 127-14, B). For defects
with a small VSD component, a modified single-patch tech-
nique may be employed. For this method, a single patch
is sewn directly to the rim of the VSD, sandwiching the brid- Ao
ging leaflets between the patch and the crest of the VSD (see Parietal extension
Fig. 127-14, C). of infundibulum

Posterior limb
RESULTS septal bond

Operative mortality is largely related to any associated cardiac

anomalies and left AV valve regurgitation. Mortality for repair
of uncomplicated incomplete AVSDs is 0% to 0.6%. The ad-
dition of left AV valve regurgitation increases mortality to TV
4% to 6%.51,56 For complete AVSDs, the mortality without left
AV valve regurgitation is approximately 5%, compared with
13% when significant degrees of regurgitation are present.51
The difference in operative mortality between patients
with and without regurgitation underscores the importance
of careful management of the left AV valve. In addition,
the majority of reoperations after repair of AVSD are due to VSD
Papillary muscle
left AV valve regurgitation. Significant postoperative AV valve of conus
regurgitation occurs in 10% to 15% of patients, necessitating FIGURE 127-15 Pathologic anatomy of tetralogy of Fallot. The ventri-
reoperation for valve repair or replacement in 7% to 12%.53–55 cular septal defect (VSD) is nonrestrictive. The infundibular septum is mala-
The incidence of permanent complete heart block is ligned, leading to aortic override and crowding of the right ventricular
approximately 1%.51,57 Heart block encountered in the imme- outflow tract. Note the hypoplastic main pulmonary trunk (MPT). Ao,
diate postoperative period may be transient due to edema aorta; TV, tricuspid valve. (From Hirsch JC, Bove EL: Tetralogy of Fallot.
In Mavroudis C, Backer CL [eds]: Pediatric Cardiac Surgery, 3rd ed.
or trauma to the AV node or bundle of His. However, right Philadelphia, Mosby, 2003.)
bundle branch block is common (22%).57

The infundibular septum normally separates the primitive

Tetralogy of Fallot outflow tracts and fuses with the ventricular septum. Anterior
------------------------------------------------------------------------------------------------------------------------------------------------
malalignment of the infundibular septum creates a VSD due to
Tetralogy of Fallot (TOF) was described by Etienne Fallot in failure of fusion with the ventricular septum and also displaces
1888. The pathologic anatomy is frequently described as the aorta over the VSD and right ventricle. The malaligned
having four components: a malalignment ventricular septal infundibular septum crowds the right ventricular outflow
defect, overriding aorta, pulmonary stenosis, and right ven- tract, producing pulmonary stenosis and, secondarily, right
tricular hypertrophy (Fig. 127-15). The anatomy of tetralogy ventricular hypertrophy. Prominent muscle bands also extend
of Fallot has also been related to a single embryologic from the septal insertion of the infundibular septum to the
defect: anterior malalignment of the infundibular septum.58 right ventricular free wall and contribute to the obstruction
1660 PART IX SPECIAL AREAS
of the right ventricular outflow tract. The pulmonary valve is
usually stenotic with thickened leaflets, which are bicuspid in
58% of cases.59 Abnormalities of coronary artery anatomy
may be present, with origin of the left anterior descending
from the right coronary artery in 5% of cases. In 25% of cases,
there is a right aortic arch. Associated defects may include
ASD, complete AVSD, PDA, or multiple VSDs.
Tetralogy of Fallot is the most common cyanotic con-
genital heart defect. It occurs in 0.6 per 1000 live births
and has a prevalence of 5% among all patients with congenital
heart disease.1
NATURAL HISTORY AND DIAGNOSIS
Patients with TOF develop cyanosis as a result of shunting
from right to left across the ventricular septal defect. The
degree of cyanosis is proportional to the degree of obstruction
of the right ventricular outflow tract. There is a spectrum of
severity, with some patients developing symptoms as neonates
and others remaining asymptomatic. The occurrence of
intermittent cyanotic spells is a well-known clinical feature
of TOF. The etiology of spelling is still controversial but is
clearly related to a transient imbalance between pulmonary
and systemic blood flow. A spell may be triggered by hypo-
volemia, peripheral vasodilation (e.g., after a bath), or infun-
dibular spasm. Spells have been reported in neonates but tend
to occur most frequently between 3 and 18 months of age.
Older children have been observed to spontaneously squat
to terminate spells. The squatting position is thought to
increase systemic vascular resistance, which thereby favors
pulmonary blood flow.
Long-term complications of untreated TOF include club-
bing of the fingers and toes, severe dyspnea on exertion, brain
abscesses (secondary to right-to-left shunting), paradoxical
embolization, and polycythemia (which may lead to cerebral
thrombosis). Long-term survival is unlikely for most patients
with untreated TOF.
Cyanosis is the most frequent physical examination
finding. There is usually a normal first heart sound and a
single second heart sound. A systolic pulmonary ejection
murmur is present at the left upper sternal border. Older
children and adults may exhibit clubbing. Chest radiogra-
phy characteristically demonstrates a boot-shaped heart be-
cause of elevation of the cardiac apex from right ventricular
hypertrophy. A right aortic arch may be evident. An echo-
cardiogram shows right ventricular hypertrophy. Echocardi-
ography is definitive, and catheterization is unnecessary in
most cases.
MANAGEMENT
The medical management of TOF is directed at the treatment
and prevention of cyanotic spells. The spelling patient should
be given oxygen and sedation, and acidosis, if present, should
be corrected. Transfusion may be indicated in anemic infants.
Alpha-agonists can be administered to increase systemic
vascular resistance, which favors pulmonary blood flow. Some
centers have used long-term therapy with beta blockers to
reduce the incidence of cyanotic spells.
All patients with TOF should undergo surgical repair.
In general, asymptomatic patients should be repaired elec-
tively between 4 and 6 months of age. Early repair is indicated
for neonates with severe cyanosis, as well as for infants who
have had a documented spell.60
Classically, the repair of TOF was performed in two stages.
The first stage involved creation of a systemic-to-pulmonary
shunt to relieve cyanosis. The second stage was a complete re-
pair. Currently, one-stage complete repair is preferred by most
centers. In some patients (such as those with multiple congen-
ital anomalies, severe concurrent illness, or an anomalous cor-
onary artery crossing a hypoplastic infundibulum), initial
palliation with a shunt may still be indicated. The modified
Blalock-Taussig shunt is the most common type of shunt
used today and consists of an interposition graft (polytetra-
fluoroethylene) between the innominate or subclavian
artery and the ipsilateral pulmonary artery. Creation of a shunt
may be performed with or without the use of cardiopulmo-
nary bypass.
Complete repair of TOF is performed using a median ster-
notomy and cardiopulmonary bypass with bicaval cannula-
tion (Fig. 127-16). Via a transatrial approach, the muscle
bundles obstructing the right ventricular outflow tract are di-
vided; resection is rarely necessary. The VSD is closed using a
patch. Depending on the status of the pulmonary valve, pul-
monary valvotomy may be appropriate. In cases of severe hy-
poplasia of the pulmonary annulus or infundibulum, a right
ventricular outflow tract transannular patch may be required
to relieve obstruction (Fig. 127-17). Our philosophy is to min-
imize right ventricular incisions whenever possible in order to
preserve right ventricular function. When an anomalous cor-
onary artery crosses the right ventricular infundibulum, a
transannular incision may be contraindicated; in these cases,
placement of a conduit (cryopreserved homograft or biopros-
thetic heterograft) between the right ventricle (via a separate
ventriculotomy) and main PA may be necessary. Patients re-
quiring a transannular patch will develop free pulmonary
insufficiency. This is well tolerated in most infants, as long
as the tricuspid valve is competent. As adults, these patients
will develop right ventricular failure due to chronic pul-
monary insufficiency and pulmonary valve implantation will
be necessary.61,62
RESULTS
The early mortality following repair of TOF is between 1% and
5%.59,63 Long-term complications include recurrent obstruc-
tion of the right ventricular outflow tract, conduit failure, and
development of right ventricular dysfunction due to chronic
pulmonary insufficiency. Actuarial survival is 86% at 20 years
with excellent functional status.64
Transposition of the Great
Arteries
------------------------------------------------------------------------------------------------------------------------------------------------
Transposition of the great arteries (TGA) is a common congen-
ital cardiovascular malformation in which there is ventricu-
loarterial discordance. This discordant anatomy is a result
of the aorta arising from the morphologic right ventricle
and the PA arising from the morphologic left ventricle (LV).
Transposition of the great arteries is divided into dextro
looped- or d-TGA, and levo looped- or l-TGA. The looping
refers to the right or left looping of the heart during fetal
 CHAPTER 127 CONGENITAL HEART DISEASE AND ANOMALIES OF THE GREAT VESSELS 1661

Parietal extension
of infundibulum

RVOT
VSD RV

VSD

AL
Papillary muscle
PA
of conus;
SL septal band
Ao
RA Ao Posterior limb
septal band
RA

B
A
RVOT
Parietal extension Muscle
of infundibulum RVOT bundles
VSD
Muscle
bundles RVOT

Dilator

Ao

VSD
C
E patch

FIGURE 127-16 Transatrial repair of tetralogy of Fallot. A, The location of the ventricular septal defect (VSD) is denoted by the dashed line. B, Stay sutures
on the septal and anterior leaflets of the tricuspid valve allow for exposure of the VSD. C and D, Using a dilator to demonstrate the course of the right
ventricular outflow tract (RVOT), hypertrophied muscle bundles may be divided or resected. E, The VSD is closed with a patch. AL, anterior leaflet
of tricuspid valve; Ao, aorta; PA, pulmonary artery; RA, right atrium; SL, septal leaflet of tricuspid valve. (From Hirsch JC, Bove EL: Tetralogy of Fallot.
In Mavroudis C, Backer CL [eds]: Pediatric Cardiac Surgery, 3rd ed. Philadelphia, Mosby, 2003.)

development, which determines whether the atria and ventri-

cles are concordant (right atrium attaches to right ventricle
and left atrium attaches to left ventricle) or discordant.
L-transposition of the great arteries is associated with both
atrioventricular discordance (right atrium attaches to left ven-
tricle and left atrium attaches to right ventricle) and ventricu-
loarterial discordance and is also termed congenitally corrected
TGA. L-transposition of the great arteries is a rare variant of
TGA and is beyond the scope of this chapter, which focuses
on d-TGA.
d-Transposition of the great arteries is the most common
cause of cyanosis in the infant and accounts for approximately
10% of all congenital cardiovascular malformations.65 The
defect can be subdivided into d-TGA with intact ventricular
septum (IVS) (55% to 60%) and d-TGA with ventricular
septal defect (VSD) (40% to 45%), one third of which are he-
modynamically insignificant. Pulmonic stenosis (PS), causing
significant left ventricular outflow tract obstruction, occurs
rarely with an IVS and in approximately 10% of d-TGA/VSD.66
NATURAL HISTORY AND DIAGNOSIS
Without intervention, d-TGA is universally fatal; 30% of
neonates will die in the first week of life, 50% by the first
month, 70% within 6 months, and 90% by 1 year.67 Clinical
characteristics depend on the degree of mixing and the
amount of pulmonary blood flow (PBF). These factors relate
to the specific anatomic subtype of d-TGA.
Neonates with d-TGA and IVS (or small VSD) have mixing
limited to the atrial level and PDA. The ASD may be restrictive,
and the PDA generally will close over the first days to week of
life. As the degree of mixing decreases, the patient becomes
increasingly cyanotic and will eventually suffer cardiovascular
collapse. Fortunately, the majority of these neonates will
manifest cyanosis early in life, which is recognized by a nurse
or physician within the first hour in 56% and in the first
day in 92%.68
In d-TGA with a large VSD, there is additional opportunity
for mixing and increased PBF. The neonate with d-TGA/VSD
1662 PART IX SPECIAL AREAS
Transannular
patch
FIGURE 127-17 A transannular patch is used to enlarge a hypoplastic
pulmonary annulus and main pulmonary trunk. (From Hirsch JC, Bove
EL: Tetralogy of Fallot. In Mavroudis C, Backer CL [eds]: Pediatric Cardiac
Surgery, 3rd ed. Philadelphia, Mosby, 2003.)
may only manifest mild cyanosis, which may be initially
overlooked. Generally within 2 to 6 weeks, signs and symp-
toms of congestive heart failure will emerge. Tachypnea and
tachycardia become prominent, while cyanosis may remain
mild. Auscultatory findings are consistent with congestive
heart failure with increased PBF including a pansystolic
murmur, third heart sound, mid-diastolic rumble, gallop,
and narrowly split second heart sound with increased pul-
monary component. Neonates with d-TGA and significant
PS present with severe cyanosis at birth. Lesser degrees of
PS will result in varying levels of cyanosis.
The diagnosis of d-TGA is generally suspected by the
clinical presentation, or it may be seen on prenatal ultrasonog-
raphy and fetal echocardiography. Chest radiograph reveals an
egg-shaped heart with a narrow superior mediastinal shadow,
mild cardiomegaly, and increased pulmonary vascular mark-
ings. Echocardiography is the diagnostic modality of choice.
Diagnostic cardiac catheterization is rarely necessary in the
current era.
MANAGEMENT
With all forms of d-TGA, there is the need to maintain ade-
quate mixing of oxygenated and deoxygenated blood between
the right and left sides of the heart. The circulation in d-TGA is
of two separate circuits, pulmonary and systemic, in parallel
(Fig. 127-18). Without communication between the two sides
of the heart, deoxygenated blood does not reach the lungs and
oxygenated blood does not reach the systemic circulation. The
majority of patients will require two levels of mixing until the
time of repair to maintain adequate saturations. In d-TGA/IVS,
this is achieved by maintaining a patent ductus arteriosus
(PDA) with prostaglandin E1 (PGE1) infusion and ensuring
Pulmonary
artery
Aorta
Left atrium
Right
atrium
Left
Right ventricle
ventricle
FIGURE 127-18 The anatomy of d-transposition of great arteries with
intact ventricular septum. (From Bove EL, Lupinetti FM: Congenital heart
disease and cardiac tumors. In Greenfield LJ, Mulholland MW, Oldham KT,
Zelenock GB [eds]: Surgery: Scientific Principles and Practice. Philadelphia,
JB Lippincott, 1993.)
an adequate atrial septal defect (ASD), which may require a
balloon atrial septostomy. Occasionally, with d-TGA/VSD with
mild or no PS, there will be adequate mixing at the atrial and
ventricular levels to allow for discontinuation of PGE1. In the
presence of significant PS, PGE1 will be necessary to provide
sufficient pulmonary blood flow. These management strategies
are used to palliate the patient until definitive surgical
correction can be undertaken. In the absence of mitigating
factors such as poor clinical condition preventing major
open heart surgery, the current treatment of all forms of
d-TGA is neonatal complete repair.
d-Transposition of the Great Arteries Without
Pulmonic Stenosis
The current treatment for d-TGA is an arterial switch opera-
tion (ASO). The repair is performed via a median sternotomy
with standard techniques of cardiopulmonary bypass. The
aorta and main PA are divided. The PA is translocated anterior
to the aorta using the Lecompte maneuver, and the distal aorta
is anastomosed to the proximal PA. The coronary arteries
are removed from the aorta with buttons of adjacent artery
and transferred to the proximal PA. The proximal aorta is
reconstructed using a pantaloon-shaped patch of autologous
pericardium and anastomosed to the distal PA. The ASD
and, if present, the VSD are closed.
d-Transposition of the Great Arteries with
Ventricular Septal Defect and Pulmonic Stenosis
In the presence of significant PS, a Rastelli procedure is per-
formed because an ASO will result in systemic (LV) outflow
tract obstruction. The Rastelli operation uses the VSD to allow
the LV to eject via the aorta. The PA is ligated and divided at the
level of the pulmonary annulus. The VSD patch is constructed
to incorporate both the VSD and the aortic valve, creating a
tunnel from the LV to the aorta (Fig. 127-19). Right ventricle-
to-PA continuity is established using a conduit from the RV
infundibulum to the distal PA.
 CHAPTER 127 CONGENITAL HEART DISEASE AND ANOMALIES OF THE GREAT VESSELS 1663

significant atrioventricular or semilunar valve regurgitation.

Another option to staged palliation is cardiac transplantation.

NATURAL HISTORY AND DIAGNOSIS

Subaortic obstruction
The natural history of these lesions depends on the amount of
pulmonary blood flow and the degree of systemic outflow
tract obstruction. Patients with severely limited pulmonary
blood flow or significant systemic outflow tract obstruction
Dacron patch closure of will succumb rapidly as the ductus arteriosus closes. Excessive
ventricular septal defect
pulmonary blood flow without systemic outflow tract obstruc-
Allograft reconstruction tion will result in death due to pulmonary vascular obstructive
of right ventricular disease, which develops at months to decades of age, depending
outflow tract
on the degree of pulmonary overcirculation. Occasionally,
patients may be “balanced” with an appropriate ratio of systemic
FIGURE 127-19 The Rastelli procedure. (From Bove EL, Lupinetti FM: and pulmonary blood flow, which in rare cases may allow sur-
Congenital heart disease and cardiac tumors. In Greenfield LJ, Mulholland vival into adulthood. The treatment of choice for all of these
MW, Oldham KT, Zelenock GB [eds]: Surgery: Scientific Principles and patients is still staged reconstruction to a Fontan procedure
Practice. Philadelphia, JB Lippincott, 1993.) because even a perfectly balanced circulation leaves the heart
volume overloaded and subject to ultimate failure. More
commonly, this scenario simply allows the patient to avoid the
RESULTS
need for an initial stage to control pulmonary blood flow in
Current hospital survival for the ASO ranges from approxi- the neonatal period.
mately 90% to 98%.66,69–72 In earlier eras, d-TGA/IVS had a
lower mortality than d-TGA/VSD or d-TGA/VSD/PS; however, MANAGEMENT
recent studies have neutralized this difference.70,72 Long-term
survival at 5 to 10 years and 15 years range from 88% to 93% Due to the limited number of suitable donors, the uncertain
and 86% to 88%, respectively.69–71 The most common cause effects of long-term immunosuppression, and the limited life
for reintervention is supravalvar pulmonic stenosis, occurring span of transplanted hearts, the majority of centers have pur-
in 3.9% to 16%.69,71 Late follow-up of arterial switch operation sued a policy of staged reconstruction as the primary therapy
patients has led to increased concerns regarding coronary for single ventricle lesions. Staged reconstruction involves a
artery patency and neoaortic root dilation.73 series of two or three procedures. An initial neonatal operation
A study of 101 patients undergoing a Rastelli operation may or may not be necessary depending on the presence of
over a 25-year period revealed a hospital mortality of 7%, with systemic outflow tract obstruction or the need to limit or aug-
no deaths in the past 7 years of the study.74 Actuarial survival ment pulmonary blood flow. This is followed at 4 to 6 months
at 5, 10, 15, and 20 years was 82%, 80%, 68%, and 52%. of age by a bidirectional Glenn or hemi-Fontan procedure and
the Fontan procedure at 18 to 24 months of age.
Because of the frequent prenatal detection of HLHS, its
high morbidity and mortality, and the thought that it may
Single Ventricle and the be a consequence of critical aortic stenosis during develop-
Hypoplastic Left Heart ment, there have been attempts at fetal ultrasound-guided
balloon valvuloplasty in a few centers, with limited success
Syndrome
------------------------------------------------------------------------------------------------------------------------------------------------
thus far.

A variety of congenital cardiovascular malformations may re- Initial Palliation

sult in functional single ventricle anatomy, most commonly Patients who do not have balanced systemic and PA flow ratios
tricuspid atresia, pulmonary atresia, unbalanced atrioven- require an initial operation in the neonatal period. The goal of
tricular septal defect, and hypoplastic left heart syndrome. this first stage of reconstruction is to provide unobstructed
The most common lesion is hypoplastic left heart syndrome blood flow to the systemic and coronary circulations, nonre-
(HLHS). There are approximately 1000 infants with HLHS strictive pulmonary venous return, and a controlled amount of
born in the United States each year. It is the most common flow to the lungs. For patients with unobstructed systemic
severe congenital heart defect, comprising 7% to 9% of all blood flow and excessive pulmonary blood flow, the PA
anomalies diagnosed within the first year of life.75 All single may be banded or disconnected with the placement of a sys-
ventricle lesions share the common physiology of only a single temic to PA shunt to control the pulmonary overcirculation.
ventricle capable of supporting cardiac output. They also For patients with unobstructed systemic blood flow and inad-
share the need for a two- or three-stage approach to recon- equate pulmonary blood flow, a systemic to PA shunt is
struction, ultimately resulting in a Fontan procedure. The goal placed. The most common single ventricle lesion, HLHS, re-
of the initial procedure is to maintain adequate cardiac output sults in systemic outflow tract obstruction and excessive pul-
and oxygen saturation while optimizing the patient for the monary blood flow and requires an initial Norwood operation.
ultimate Fontan procedure. Characteristics for an optimal During the Norwood procedure, the aorta is reconstructed, in-
Fontan candidate include good ventricular function, low corporating the proximal PA to allow the single right ventricle
PA vascular resistance, no outflow tract obstruction, and no to eject via the aorta. A systemic to PA shunt or right ventricle
1664 PART IX SPECIAL AREAS
AP shunt
RV
FIGURE 127-20 The completed Norwood procedure. AP, aortopulmon-
ary; RV, right ventricle. (From Ohye RG, Mosca RM, Bove EL, et al: Hypoplas-
tic left heart syndrome. In Mavroudis C, Backer CL [eds]: Pediatric Cardiac
Surgery, Philadelphia, Mosby, 2003.)
FIGURE 127-21 The completed hemi-Fontan procedure. (From Ohye
RG, Mosca RM, Bove EL, et al: Hypoplastic left heart syndrome. In Mavroudis
C, Backer CL [eds]: Pediatric Cardiac Surgery, Philadelphia, Mosby, 2003.)
to PA conduit is then placed to provide PA blood flow
(Fig. 127-20). In response to the high mortality associated
with the Norwood procedure at lower volume centers, a
hybrid procedure has been introduced. It involves placement
of a patent ductus arteriosus stent and bilateral PA bands
via a median sternotomy in the catheterization laboratory.
FIGURE 127-22 The completed Fontan procedure. (From Ohye RG,
Mosca RM, Bove EL, et al. Hypoplastic left heart syndrome. In Mavroudis C,
Backer CL [eds]: Pediatric Cardiac Surgery. Philadelphia, Mosby, 2003.)
This approach has produced improved early results with
long-term results still pending.76 This approach delays the
complex aortic arch reconstruction to the second stage.
Bidirectional Glenn or Hemi-Fontan Procedure
In order to minimize the period of time during which the right
ventricle is subject to volume overload, the hemi-Fontan
operation or a bidirectional Glenn anastomosis (BDG) is typ-
ically performed between 4 to 6 months of age. The BDG
entails removing the systemic to PA shunt and connecting
the superior vena cava (SVC) directly to the right PA by an
end-to-side anastomosis. The pulmonary blood flow after this
second stage is provided solely by the SVC. The hemi-Fontan
is a modification of the bidirectional Glenn procedure. The
hemi-Fontan involves a side-to-side connection between the
superior vena cava/right atrial junction and the pulmonary
arteries, routine augmentation of the branch pulmonary arter-
ies, and temporary patch closure between the pulmonary
arteries and the right atrium (Fig. 127-21).
Fontan Procedure
The completion of the Fontan procedure is usually performed
at 18 to 24 months of age. The Fontan technique that we have
employed for HLHS anatomy is the total cavopulmonary con-
nection with a lateral tunnel (Fig. 127-22). During the Fontan
procedure, the inferior vena caval venous return is directed to
the pulmonary artery. Other centers use an extracardiac
 CHAPTER 127 CONGENITAL HEART DISEASE AND ANOMALIES OF THE GREAT VESSELS 1665

conduit from the IVC to the right pulmonary artery. Of signif- subclavian artery and right ligamentum. The incidence of
icance to pediatric surgeons, a Fontan procedure may lead to clinically significant vascular rings is 1% to 2% of all con-
an increased venous pressure in the IVC, which may in turn genital heart defects.
cause ascites in the early postoperative period, leading to res- Vascular rings and pulmonary slings have been described
piratory compromise. Although this has become less frequent in conjunction with other cardiac defects including tetralogy
with fenestrated Fontans, peritoneal drainage catheters may of Fallot, atrial septal defect, branch PA stenosis, coarctation,
be required for several days postoperatively in some patients. atrioventricular septal defect, ventricular septal defect, inter-
rupted aortic arch, and aortopulmonary window. Significant
associated cardiac anomalies occur in 11% to 20% of patients
RESULTS
with a vascular ring.83,85,86 A right aortic arch is generally
Recently, tremendous strides have been made in the outcomes associated with a greater incidence of coexisting anomalies.
for patients with single ventricle anatomy. Most notably, Pediatric surgeons should be aware that arch anomalies
survival for patients with HLHS, universally fatal only 2 and aberrant subclavian arteries are more common in chil-
decades ago, has drastically improved. The highest risk stage dren with esophageal atresia, even in the absence of cardiac
of the repair remains the Norwood operation. During the malformations.
1990s, the hospital survival for the Norwood procedure across By the end of the fourth week of embryonic development,
the United States was approximately 40%.77 Currently, several the six aortic or branchial arches have formed between the
select centers are reporting hospital survivals of approximately dorsal aortae and ventral roots. Subsequent involution and
90% or greater.78–81 Reported survival for the hemi-Fontan migration of the arches results in the anatomically normal
and Fontan procedures has been excellent at 98% for both or abnormal development of the aorta and its branches. The
operations.82–84 majority of the first, second, and fifth arches regress. The third
arch forms the common carotid artery and proximal internal
carotid artery. The right fourth arch forms the proximal right
Vascular Rings and Slings subclavian artery. The left fourth arch contributes to the por-
------------------------------------------------------------------------------------------------------------------------------------------------
tion of the aortic arch from left carotid to left subclavian
Vascular rings comprise a spectrum of vascular anomalies of arteries. The proximal portion of the right sixth arch becomes
the aortic arch, pulmonary artery, and brachiocephalic vessels. the proximal portion of the right pulmonary artery, while the
The clinically significant manifestation of these lesions is a distal segment involutes. Similarly, the proximal left sixth
varying degree of tracheoesophageal compression. These arch contributes to the proximal left pulmonary artery, and the
vascular anomalies can be divided into complete vascular distal sixth arch becomes the ductus arteriosus (Fig. 127-23).
rings and partial vascular rings. Complete vascular rings can The PA is formed from two vascular precursors, as well as
be divided into double aortic arch and right aortic arch with through a combination of angiogenesis, the de novo develop-
retroesophageal left ligamentum arteriosum. These two cate- ment of new blood vessels, and vasculogenesis, the budding
gories can be further subdivided on the basis of the specific and migration of existing vessels. As stated earlier, the proxi-
anatomy (Table 127-1). Incomplete vascular rings include ab- mal pulmonary arteries are based on the sixth arches, whereas
errant right subclavian artery, innominate artery compression, the primitive lung buds initially derive their blood supply
and PA sling. Other rare variations, which have been de- from the splanchnic plexus. Ultimately, these two segments
scribed, include left aortic arch with right descending aorta of the PA join to form the vascular network of the lung
and right ligamentum and left aortic arch with aberrant right parenchyma (Fig. 127-24).

NATURAL HISTORY AND DIAGNOSIS

TABLE 127-1 Children with a complete vascular ring generally present
Anatomic Classification and Distribution of Vascular Anomalies within the first weeks to months of life. Typically, children
Anomaly Prevalence %
with a double aortic arch present earlier in life than those with
a right arch and retroesophageal left ligamentum. In the youn-
Complete vascular rings 65-82 ger age group, respiratory symptoms predominate, as liquids
Double aortic arch 38-55 are generally well tolerated. Respiratory symptoms may in-
Right dominant 73-81 clude stridor, nonproductive cough, apnea, or frequent respi-
Left dominant 15-20 ratory infections. The cough is classically described as “seal
Co-dominant 3-11 bark” or “brassy.” These symptoms may mimic asthma, respi-
Right arch, left ligamentum/ductus 24-55 ratory infection, or reflux, and children with vascular rings are
Aberrant left subclavian 65-92 often initially misdiagnosed. With the transition to solid food,
Mirror-image branching 8-35 dysphagia becomes more apparent.
Right ligamentum 0-3 The presentation of a patient with an incomplete vascular
Incomplete vascular rings 14-29 ring is variable. Children with innominate artery compression
Innominate artery compression 80-87 usually present within the first 1 to 2 years of life with res-
Aberrant right subclavian 13-20 piratory symptoms. Although aberrant right subclavian artery
Pulmonary sling 2-9 is the most common arch abnormality, occurring in approxi-
From Ohye RG, Wild LC, Mutabagani K, Bove EL: Vascular rings and slings. In
mately 0.5% to 1% of the population, it rarely causes symp-
Franco KL, Putman JB (eds): Advanced Therapy in Thoracic Surgery. toms. Classically, when symptoms do occur, they present in
Hamilton, Ontario, BC Decker, 2004. the seventh and eighth decade, as the aberrant vessel becomes
1666 PART IX SPECIAL AREAS

Ventral
roots
LECA
LICA RCCA
LCCA
Aortic arch I RSCA LSCA

II

III
IV

AO
V
PA
VI
AO
R. dorsal PA L. dorsal
aorta aorta
LSCA

FIGURE 127-23 Normal aortic arch development. AO, aorta; DA, ductus arteriosum; LCCA, left common carotid artery; LECA, left external carotid artery;
LICA, left internal carotid artery; LSCA, left subclavian artery; PA, pulmonary artery; RCCA, right common carotid artery; RSCA, right subclavian artery. (From
Ohye RG, Wild LC, Mutabagani K, Bove EL: Vascular rings and slings. In Franco KL, Putman JB [eds]: Advanced Therapy in Thoracic Surgery. Hamilton, Ontario,
BC Decker, 2004.)

3-mm embryo

Pulm. art.
bud from VI
Pulm. art. bud from 5-mm embryo
lung primordium

R. dorsal aorta

RPA Ligamentum
or ductus
Involutes
LPA

R. dorsal aorta

L. dorsal aorta
FIGURE 127-24 Normal pulmonary artery development. LPA, left pulmonary artery; RPA, right pulmonary artery. (From Ohye RG, Wild LC, Mutabagani K,
Bove EL: Vascular rings and slings. In Franco KL, Putman JB [eds]: Advanced Therapy in Thoracic Surgery. Hamilton, Ontario, BC Decker, 2004.)

ectatic and calcified, causing dysphagia lusoria due to impinge-
ment of the artery on the posterior esophagus. An aberrant right
subclavian rarely causes symptoms, except when it is of an ab-
normally large caliber or associated with tracheomalacia.
Children with PA slings generally present with respiratory
symptoms within the first few weeks to months of life. As with
complete rings, respiratory symptoms may include stridor,
nonproductive cough, apnea, or frequent respiratory infections
and may mimic other conditions leading to misdiagnosis. PA
slings are associated with complete tracheal rings in 30% to
40% of patients, leading to focal or diffuse tracheal stenosis.87
The methods for diagnosing a vascular ring are variable due to
the variability in presentation and the spectrum of diagnostic
tests available. A child with a presumptive diagnosis of asthma
or tracheomalacia may be referred to a pulmonologist, and a
diagnosis of vascular ring made or suspected initially by chest
radiograph and bronchoscopy. In some situations, the diagnosis
is made by echocardiography during evaluation for concurrent
cardiac defects. Regardless, the diagnosis generally begins with a
chest radiograph. Complementary studies may include barium
esophagogram, CT, MRI, and bronchoscopy. CT, MRI, and bron-
choscopy are important modalities to define the tracheal
anatomy in a patient with a PA sling. Echocardiography may
be diagnostic and is used to rule out other cardiac anomalies. Tra-
cheograms and cardiac catheterizations, which have been used
extensively in the past, are rarely indicated in the current era.
 CHAPTER 127
MANAGEMENT
Complete Vascular Rings
Double Aortic Arch A double aortic arch occurs when the
distal portion of the right dorsal aorta fails to regress
(Fig. 127-25). The two arches form a complete ring, encircling
the trachea and esophagus. The right arch is dominant in
the majority of cases, followed by left dominant, with co-
dominant arches being the least common (Table 127-1).
The left and right carotid and subclavian arteries generally
arise from their respective arches. The ligamentum arteriosum
and descending aorta usually remain on the left.
The approach to repair of a double aortic arch is via a left
posterolateral thoracotomy. The procedure can easily be
accomplished through a limited, muscle-sparing incision
through the third or fourth intercostal space. The lung is retracted
anteriorly and inferiorly, exposing the posterior mediastinum.
The pleura is incised, after identifying the vagus and phrenic
nerves. The ligamentum or ductus arteriosum is divided while
preserving the recurrent laryngeal nerve. The nondominant arch
is then divided between two vascular clamps at the point where
brachiocephalic flow is optimally preserved. If there is concern
regarding the location for division, the arches can be temporarily
occluded at various points while monitoring pulse and blood
pressure in each limb. If there is an atretic segment, the division
is done at the point of the atresia. Dissection around the esoph-
agus and trachea in the regions of the ligamentum/ductus and
nondominant arch allows for retraction of the vascular structures
and lysis of any residual obstructing adhesions.
Right Aortic Arch with Left Ligamentum Arteriosum There
are three anatomic variations for a right arch with a left liga-
mentum, which cause a complete vascular ring. If the left
fourth arch regresses between the aorta and left subclavian,
a right aortic arch with aberrant left subclavian artery results.
The ligamentum arteriosum is retroesophageal, bridging the
left PA and aberrant left subclavian, forming a complete vas-
cular ring (Fig. 127-26). If the left fourth arch regresses after
RCCA
RSCA
AO
PA
Fail to
regress
FIGURE 127-25 Formation of a double aortic arch. AO, aorta; DA, ductus arteriosum; LCCA, left common carotid artery; Lig., ligamentum arteriosum; LSCA,
left subclavian artery; PA, pulmonary artery; RCCA, right common carotid artery; RSCA, right subclavian artery. (From Ohye RG, Wild LC, Mutabagani K,
Bove EL: Vascular rings and slings. In Franco KL, Putman JB [eds]: Advanced Therapy in Thoracic Surgery. Hamilton, Ontario, BC Decker, 2004.)
CONGENITAL HEART DISEASE AND ANOMALIES OF THE GREAT VESSELS 1667
the origin of the left subclavian artery, but before the arch
reaches the dorsal aorta to communicate with the left sixth
arch (which becomes the ductus arteriosum), there is mirror-
image branching. The retroesophageal ligamentum arterio-
sum arises directly from the descending aorta, or from a
Kommerell diverticulum off of the descending aorta, forming
the complete ring (Fig. 127-27). If communication is main-
tained between the left fourth and sixth arches, there is mirror
image branching with the ligamentum arising from the ante-
rior and mirror image left subclavian and a ring is not formed
(Fig. 127-28).
The surgical approach for a right aortic arch with retroeso-
phageal left ligamentum arteriosum is the same as for a double
arch. The ligamentum is divided, and any adhesions around
the esophagus and trachea are lysed. Rarely, the Kommerell di-
verticulum has been reported to cause compression even after
division of the ligamentum. As such, it may be prudent to re-
sect or suspend the diverticulum posteriorly to the preverte-
bral fascia, if it is particularly prominent.
Incomplete Vascular Rings
Innominate Artery Compression In innominate artery
compression syndrome, the aortic arch and ligamentum are
in their normal leftward position; however, the innominate
artery arises partially or totally to the left of midline
(Fig. 127-29). As the artery courses from left to right anterior
to the trachea, it causes tracheal compression. The symptoms
of innominate artery compression may be mild to severe.
With mild symptoms and minimal tracheal compression on
bronchoscopy, children can be observed expectantly because
the symptoms may resolve with growth. Indications for
surgery include apnea, severe respiratory distress, significant
stridor, or recurrent respiratory tract infection.
Several approaches for the correction of innominate artery
compression syndrome have been described. These include
simple division, division with reimplantation into the right
side of the ascending aorta, and suspension to the overlying
sternum. Suspension is currently the most widely used
LCCA
RCCA
LSCA
RSCA
III
LCCA
IV Lig.
VI (DA)
AO
LSCA PA
1668 PART IX SPECIAL AREAS

RCCA
LCCA
RSCA
Aberrant LSCA
III
LCCA Lig.
RCCA
IV Involutes
AO
RSCA VI (DA) PA
AO
PA
LSCA

FIGURE 127-26 Formation of a right aortic arch with aberrant left subclavian and retroesophageal left ligamentum arteriosum. AO, aorta; DA, ductus
arteriosum; LCCA, left common carotid artery; Lig., ligamentum arteriosum; LSCA, left subclavian artery; PA, pulmonary artery; RCCA, right common carotid
artery; RSCA, right subclavian artery. (From Ohye RG, Wild LC, Mutabagani K, Bove EL: Vascular rings and slings. In Franco KL, Putman JB [eds]: Advanced
Therapy in Thoracic Surgery. Hamilton, Ontario, BC Decker, 2004.)

LCCA
RCCA

RSCA LSCA
III
Kommerell’s
RCCA LCCA diverticulum
IV DA Lig.
VI Involutes AO
AO
RSCA LSCA PA
PA

Kommerell’s
diverticulum

FIGURE 127-27 Formation of a right aortic arch with mirror image branching and retroesophageal left ligamentum arteriosum. AO, aorta; DA, ductus
arteriosum; LCCA, left common carotid artery; Lig., ligamentum arteriosum; LSCA, left subclavian artery; PA, pulmonary artery; RCCA, right common carotid
artery; RSCA, right subclavian artery. (From Ohye RG, Wild LC, Mutabagani K, Bove EL: Vascular rings and slings. In Franco KL, Putman JB [eds]: Advanced
Therapy in Thoracic Surgery. Hamilton, Ontario, BC Decker, 2004.)

technique. Exposure is obtained through a limited left
anterior, right anterior, or right inframammary anterolateral
thoracotomy. Once the innominate artery is exposed, no dis-
section of the artery is undertaken. By not performing a cir-
cumferential dissection of the innominate artery, the
suspension of the vessel will also pull up on the anterior tra-
chea. Pledgeted polypropylene sutures are passed partial
thickness through both the innominate artery and the aorta
at the origin of the innominate. Temporary distraction on
the sutures under bronchoscopic guidance aids in the optimal
placement of the sutures in the vessels and overlying sternum.
Once satisfactory establishment of tracheal patency is con-
firmed by bronchoscopy, the sutures are brought through
the sternum and secured. The operation is essentially the
same as an aortopexy done to relieve severe localized tracheo-
malacia associated with esophageal atresia (see Chapter 69).
Left Aortic Arch with Aberrant Right Subclavian Artery
An aberrant right subclavian artery occurs when there is re-
gression of the right fourth arch between the right common
carotid and right subclavian arteries (Fig. 127-30). The right
subclavian then arises from the leftward descending aorta, lay-
ing posterior to the esophagus as it crosses from left to right.
Although the artery can compress the esophagus posteriorly, it
is rarely the cause of symptoms in children. Surgical treatment
involves simple division via a left posterolateral thoracotomy.
Rarely, reimplantation or grafting from the right carotid or
aortic arch may be necessary. More significantly, any vascular
ring or sling that goes around the esophagus should be a
contraindication to long-term nasogastric tubes or any esoph-
ageal stent (as may be used in the postoperative management
of patients with esophageal atresia), because of the risk of
arterio-esophageal fistula.
 CHAPTER 127 CONGENITAL HEART DISEASE AND ANOMALIES OF THE GREAT VESSELS 1669

LCCA
RCCA
LSCA
III RSCA
RCCA LCCA Lig.
IV
VI (DA)
RSCA
AO LSCA AO
PA
PA

Involutes

FIGURE 127-28 Formation of a right aortic arch with mirror imaging branching and left ligamentum arteriosum, which does not form a vascular ring. AO,
aorta; DA, ductus arteriosum; LCCA, left common carotid artery; Lig., ligamentum arteriosum; LSCA, left subclavian artery; PA, pulmonary artery; RCCA, right
common carotid artery; RSCA, right subclavian artery. (From Ohye RG, Wild LC, Mutabagani K, Bove EL: Vascular rings and slings. In Franco KL, Putman JB
[eds]: Advanced Therapy in Thoracic Surgery. Hamilton, Ontario, BC Decker, 2004.)

LCCA
Innominate a.
Innominate a. LSCA

AO AO

PA PA

Involutes

FIGURE 127-29 Embryologic origin of innominate artery compression syndrome. AO, aorta; LCCA, left common carotid artery; LSCA, left subclavian
artery; PA, pulmonary artery. (From Ohye RG, Wild LC, Mutabagani K, Bove EL: Vascular rings and slings. In Franco KL, Putman JB [eds]: Advanced Therapy
in Thoracic Surgery, Hamilton, Ontario, BC Decker, 2004.)

Pulmonary Artery Sling

Normally, the right and left sixth aortic arches contribute to
the proximal portions of their respective pulmonary arteries.
If the proximal left sixth arch involutes and the bud from the
left lung migrates rightward to meet the right pulmonary
artery, a PA sling is formed (Fig. 127-31). PA slings are asso-
ciated with complete tracheal rings and tracheal stenosis in
30% to 40% of patients.87 Origin of the right upper lobe
bronchus from the trachea has been reported in frequent
association with PA sling.88
Initial attempts at the repair of a PA sling involved reim-
plantation after division of the left pulmonary artery (LPA)
and translocation of the trachea without cardiopulmonary
bypass. These early reports had a high incidence of LPA
thrombosis. This has led some authors to advocate division
of the trachea and translocation of the LPA. This approach
would seem sensible if the trachea were being divided in
the course of tracheal reconstruction. However, currently
most authors advocate the reimplantation of the LPA, which
has resulted in excellent results.89,90 The procedure is done
via a median sternotomy on cardiopulmonary bypass to
ensure optimal visualization of the repair. Aortic cross clam-
ping is not necessary. The LPA is divided from the RPA, trans-
located anterior to the trachea, and reimplanted into the main
pulmonary artery.
Any necessary reconstruction of the trachea is done
concurrently with bronchoscopic assistance. Many techniques
for tracheal reconstruction have been described, the most
1670 PART IX SPECIAL AREAS

LCCA

RCCA
Aberrant LSCA
III RSCA
LCCA RCCA
Involutes
IV
AO
RSCA AO LSCA
PA PA

FIGURE 127-30 Formation of a left aortic arch with aberrant right subclavian artery. LCCA, left common carotid artery; LSCA, left subclavian artery; RCCA,
right common carotid artery; RSCA, right subclavian artery. (From Ohye RG, Wild LC, Mutabagani K, Bove EL: Vascular rings and slings. In Franco KL, Putman
JB [eds]: Advanced Therapy in Thoracic Surgery, Hamilton, Ontario, BC Decker, 2004.)

3-mm embryo

No bud develops
on left side

Bud from I. lung Pulmonary

primordium migrates sling
to right side

5-mm embryo
Anom.
attachment
to RPA
Ductus
RPA
LPA

FIGURE 127-31 Formation of a pulmonary artery sling. LPA, left pulmonary artery; RPA, right pulmonary artery. (From Ohye RG, Wild LC, Mutabagani K,
Bove EL: Vascular rings and slings. In Franco KL, Putman JB [eds]: Advanced Therapy in Thoracic Surgery, Hamilton, Ontario, BC Decker, 2004.)

common of which are resection with primary reanastomosis
and sliding tracheoplasty for short segment stenosis and rib
cartilage or pericardial patch for long areas of narrowing.
Vascular Rings Requiring a Right Thoracotomy
More than 95% of vascular rings without concurrent cardiac
defects can be performed through a left thoracotomy. A right
thoracotomy is indicated for the rare cases where there is a
right ligamentum arteriosum. A right ligamentum occurs in
the setting of a left aortic arch with right descending aorta,
where the ligamentum bridges from the descending aorta to
the right PA, forming a complete ring. Right ligamentum
arteriosum has also been described with a left aortic arch with
aberrant right subclavian artery. In this case, the ligamentum
may arise from the aberrant subclavian artery, a diverticulum
off of the arch, or directly from the left arch to the right
pulmonary artery. In addition, a double aortic arch with an
atretic segment proximal to the right carotid artery is more eas-
ily divided through a right thoracotomy. The approach to
these anomalies is the same as for a left-sided ring division,
with the caveat that the right recurrent laryngeal nerve will
loop around the right ligamentum.
Video-Assisted Thoracoscopic Surgery
We and Burke and colleagues have repaired vascular rings
using video-assisted thoracoscopic surgery (VATS) both with91
and without92 robotic assistance. Candidates for thoracoscopic
division in both series were limited to those patients requiring
 CHAPTER 127 CONGENITAL HEART DISEASE AND ANOMALIES OF THE GREAT VESSELS
only the division of nonpatent vascular structures. There were
no operative deaths, and all procedures were completed by
VATS. In general, VATS is used for patients greater than
15 kg due to current size limitations of the instruments.93
RESULTS
Mortality for the repair of a vascular ring is 0.5% to 7.6%, with
improved survival occurring in more recent series.83,85,88,94
The majority of deaths are related to other cardiac defects
1671
or respiratory infection and failure. Backer and colleagues89
reported a series of 16 patients repaired using LPA division
and reimplantation for PA sling, all of whom also required
tracheal reconstruction. There were no operative mortali-
ties and one late death due to respiratory complications.
The major source of morbidity, as well as mortality, in this
and other series is related to the tracheal reconstruction.89,90
The complete reference list is available online at www.
expertconsult.com.
 that transform it into an NT. Secondary neurulation involves the
formation of an epithelial cord (i.e., the medullary cord) and its
subsequent cavitation to form an NT. In humans, the brain and
SC as far caudally as the S2 level form by primary neurulation,
whereas the SC caudal to the S2 level and the filum terminale form
by secondary neurulation. For more than a century, investigators
have used many different animal models and experimental
paradigms to ascertain the mechanisms underlying primary
and secondary neurulation. When these processes go awry in
humans, neural tube defects (NTDs) result. NTDs are among
the most common of all human birth defects, with incidence rates
varying from less than 1 to more than 6 per 1000 pregnancies
around the world and affecting about 3000 pregnancies
per year in the United States.1,2,2a These complex congenital mal-
formations occur when the NT, which ultimately forms the brain
and SC, fails to close during the first few weeks of embryonic de-
velopment. NTDs are commonly classified as open or closed on
the basis of the presence or absence of exposed neural tissue. In
open NTDs, which occur as a consequence of failed primary neu-
rulation, the NTand its membranous coverings are abnormal and
are exposed at birth through a defect in the skull or spine. In
closed NTDs, which occur because of faulty secondary neurula-
tion, the NTand its membranous coverings are abnormal but the
CHAPTER 128 overlying skin is intact. Open NTDs, which occur more
frequently than closed NTDs, include anencephaly, spinal
rachischisis or spina bifida aperta/cystica (i.e., myeloschisis, mye-
lomeningocele, and meningocele), and encephalocele. Closed
Management of NTDs include spina bifida occulta; lipomatous malformations
(e.g., lipomas and lipomyelomeningoceles); split cord malforma-
tions (diastematomyelia, diplomyelia); neurenteric cysts; dermal
Neural Tube sinuses; tethered SC; and sacral agenesis (caudal regression). An-
encephaly, or absence of the brain, is invariably fatal and results

Defects, when the cephalic part of the NT fails to close. Infants born with
this condition typically die at birth or within the first few days af-
ter birth. Spina bifida cystica (i.e., myeloschisis, myelomeningo-
Hydrocephalus, cele, and meningocele) results from incomplete development of
the NT more caudally, with protrusion of the malformed neural
tissue, meninges, or both through an opening in the vertebral
Refractory Epilepsy, arches, muscle, and skin (Fig. 128-1). Open NTDs range in sever-
ity from craniorachischisis, in which the entire NTremains open,
to a lesion limited to a single vertebral level. Of the open NTDs,
and Central myelomeningocele is the most common and the most severe
birth defect compatible with survival.3 Recent studies have sug-

Nervous System gested that periconceptional folic acid supplementation reduces

the occurrence and recurrence risk of NTDs by 50% to 70%4,5;
however, these serious birth defects continue to affect approxi-
Infections mately 1 per 1000 live-born infants. In other words, including
pregnancies that are electively terminated, approximately 4000
pregnancies per year, or 11 to 12 pregnancies per day, in the
Jodi L. Smith United States are affected by an NTD. These devastating birth de-
fects are associated with fetal wastage, infant mortality, lifelong
disability, and substantial health care costs. To eradicate NTDs,
we must continue to advance our understanding regarding
the cellular and molecular mechanisms responsible for normal
Neural Tube Defects
------------------------------------------------------------------------------------------------------------------------------------------------
NT formation through dedicated research efforts.

Neurulation, one of the earliest and most crucial events in human

EMBRYOLOGY
development, generates the neural tube (NT), the rudiment of the
entire adult central nervous system (CNS) (i.e., the brain and spi- Most of what we already know about the mechanisms that
nal cord [SC]). The NT forms during primary and secondary comprise normal human NT formation comes from extensive
neurulation. Primary neurulation involves the formation of a descriptive studies on both human and nonhuman embryos
neural plate (NP) and the subsequent morphogenetic movements and from experimental studies on nonhuman animal models
1673
1674 PART IX SPECIAL AREAS
FIGURE 128-1 Three-hour-old male infant with an L5-S1 myelomenin-
gocele diagnosed prenatally by ultrasound.
such as the chick and mouse, from which much of our current
understanding of neurulation is derived. Such studies, which
provide insight into the mechanisms underlying early human
neural development, have revealed that primary neurulation
occurs in four spatially and temporally overlapping stages:
(1) formation of the NP, (2) shaping of the NP, (3) bending
of the NP, and (4) fusion of the neural folds (Fig. 128-2,
A to E).6,7 The important stages in embryologic development
are further summarized in Figures 128-3 and 128-4.6–9
For many years, neurulation was viewed as a simple, intrin-
sic force-driven, all-or-none process in which the NP either
rolled up into a tube or did not. Moreover, a change in neu-
roendocrine (NE) cell shape from column-like to wedgelike
by contraction of apical microfilaments was the intrinsic force
C E
FIGURE 128-2 A-E, Scanning electron micrographs of transverse sections through the chick blastoderm demonstrating the four stages of primary
neurulation—formation, shaping, and bending of the neural plate and fusion of the neural folds. During primary neurulation, the neural tube, the precursor
of the entire adult central nervous system, develops from the ectodermal flat neural plate. DHP, dorsolateral hinge point; MHP, median hinge
point. (Modified from Smith JL, Schoenwolf GC: Neurulation: Coming to closure. Trends Neurosci 1997;20:510.)
proposed to drive this process. However, through more recent
advances in research we now know that this “traditional view
of neurulation” is incorrect. Instead, neurulation is a highly
complex, multifactorial process that requires not only intrinsic
forces generated by NE cells but also extrinsic forces generated
by epidermal ectoderm cells, with the intrinsic and extrinsic
forces acting in concert.6,7
A “multifactorial model of neurulation” has been proposed
to explain how intrinsic and extrinsic morphogenetic forces
generated by fundamental cell behaviors such as changes in
cell shape, position, and number interact during neurulation
to transform a flat pseudostratified sheet of ectodermal cells,
the NP, into an elongated NT, the precursor of the entire adult
CNS.7 According to this model, intrinsic forces generated by
elongation, rearrangement, and nonrandomly oriented rostro-
caudal division of NE cells drive shaping of the NP and intrin-
sic forces generated by NE cell wedging drive furrowing of the
NP. Extrinsic forces are generated by changes in epidermal ec-
toderm cell shape from low cuboidal to squamous, rearrange-
ment of epidermal ectoderm cells with caudal-to-medial
convergent extension, and nonrandomly oriented rostrocau-
dal and mediolateral cell division. Such changes in epidermal
ectoderm cell behavior drive rostrocaudal lengthening of the
epidermal ectoderm, as well as medial epidermal ectoderm
expansion, which in turn drives neural fold elevation, conver-
gence, and fusion. Failure of this process at any step along the
way presumably will generate a neural tube defect.
EPIDEMIOLOGY
The incidence of spina bifida is estimated at one to two cases
per 1000 population, with certain populations having a
significantly higher incidence based on genetic predilection.
There is also a marked geographic variation in incidence.
 CHAPTER 128 MANAGEMENT OF NEURAL TUBE DEFECTS, HYDROCEPHALUS, REFRACTORY EPILEPSY, AND CNS INFECTIONS 1675

fb

mb

hb

sc

A B C D E
FIGURE 128-3 Light micrographs of dorsal views of chick blastoderms at Hamburger and Hamilton stages 4 to 10þ. A, Flat neural plate (stage 4)
183

shortly after its formation. B, Flat neural plate with a notochord (head process; stage 5). C, Neural groove stage (stage 6) showing rostrocaudal lengthening
and mediolateral narrowing characteristic of shaping of the neural plate. The broken line in A to C indicates approximate rostrolateral boundaries of the
neural plate. D, Incipient neural tube (stage 8þ). E, Definitive neural tube (stage 10þ). fb, forebrain; hb, hindbrain; mb, midbrain; sc, spinal cord. (Modified
from Smith JL, Schoenwolf GC: Neurulation: Coming to closure. Trends Neurosci 1997;20:510.)

main reasons for the decline of NTDs in the United States

are pregnancy termination and periconceptional use of folic
Shaping
acid. Finally, the risk of giving birth to a second child with
an NTD in the United States is 1% to 3%.

MHP
Folding
Elevation
DLHP
Convergence
SC SR
FIGURE 128-4 Schematic diagram illustrating the major morphogenetic
events of primary neurulation at the future midbrain/hindbrain level and
differing morphologies at the spinal cord (SC) and sinus rhomboidalis (SR)
levels. DLHP, dorsolateral hinge point; MHP, median hinge point. (Modi-
fied from Smith JL, Schoenwolf GC: Neurulation: Coming to closure. Trends
Neurosci 1997;20:510.)
For instance, the highest rates occur in parts of the British
Isles, mainly Ireland and Wales, where the incidence of mye-
lomeningocele is as high as three to four cases per 1000 pop-
ulation and the incidence of anencephaly is more than six
cases per 1000 population for both live births and stillbirths.
In the United States, the prevalence is declining, with African
Americans having a rate of 0.1 to 0.4 per 1000 live births ver-
sus 1 per 1000 live births in the white population. The two
ETIOLOGY
The precise etiology and specific genes responsible for the gen-
eration of NTDs such as myelomeningocele have not yet been
elucidated. The cause of such defects is probably multifactorial,
with both genetic and environmental factors playing a role. Be-
sides problems occurring during neurulation that prevent NT
closure, problems in the postneurulation period (e.g., as a re-
sult of exposure of the unprotected nervous system to amniotic
fluid) may also have a deleterious effect on NT development
and subsequent neural function. In addition, numerous terato-
gens have been implicated in the etiology of NTDs. Of all the
teratogens evaluated thus far, carbamazepine, valproic acid,
and folic acid deficiency have been most strongly tied to the de-
velopment of NTDs. In support of this association, periconcep-
tional folic acid supplementation has been shown to reduce the
occurrence and recurrence risks of NTDs,4,5 suggesting that
NTDs result, at least in part, from folic acid deficiency. How-
ever, the precise role of folic acid in preventing NTDs remains
unclear. Nevertheless, the U.S. Public Health Service made a
strong recommendation in September 1992 that all women
of childbearing age in the United States who are capable of be-
coming pregnant should consume 0.4 mg of folic acid per day
for the purpose of reducing the risk of having a pregnancy
affected with an NTD. Furthermore, valproic acid (a known
folate antagonist), if taken during pregnancy, results in a 1%
to 2% estimated risk of having a child with an NTD.
PATHOLOGY
The two most common types of NTDs seen with spina bifida
aperta or cystica are myelomeningocele and meningocele.
Myelomeningocele, the most common NTD compatible with
1676 PART IX SPECIAL AREAS
life, consists of an open neural placode surrounded by an in-
termediate zone of thin epithelium, which in turn is sur-
rounded by normal skin (see Fig. 128-1). The dorsal (i.e.,
exposed) surface of the neural placode constitutes the everted
interior of the NT and is continuous with the central canal of
the SC rostrally. In contrast, the ventral surface constitutes the
outside of what should have been a closed NT. Ventral or mo-
tor roots exit from the ventral surface of the placode just lateral
to the midline on both sides; dorsal or sensory roots enter the
periphery of the placode lateral to the motor roots. An arach-
noid sac and subarachnoid space underlie the neural placode,
and the junction between the skin and dura underlies the skin
within a few millimeters from the skin edge. Subjacent to the
dura is an epidural space that contains fat. The spinous pro-
cesses are absent, and the paraspinous muscle masses, hypo-
plastic laminae, and pedicles are everted. In most cases, the SC
rostral to the neural placode is normal; however, additional
anomalies such as a split cord malformation (i.e., diastemato-
myelia), arteriovenous malformation, epidermoid, or lipoma
may be present. In addition, most neonates with myelomenin-
gocele have associated neurologic malformations such as
hydrocephalus and Chiari II malformation (deformity of the
hindbrain, cervical SC, and craniovertebral junction), as well
as orthopedic anomalies of their lower extremities and uro-
genital anomalies caused by involvement of the sacral nerve
roots. A meningocele is a skin-covered anomaly characterized
by herniation of just the meninges through a dorsal bony
defect (spina bifida). However, unlike myelomeningocele,
the SC and nerve roots do not herniate into the skin-covered
dural sac and neonates with these lesions typically do not have
associated neurologic malformations.
DIAGNOSIS OF NEURAL TUBE DEFECTS
Open NTDs such as anencephaly and myelomeningocele can
be diagnosed prenatally by measuring alpha-fetoprotein (AFP)
in the amniotic fluid or maternal bloodstream. AFP, the major
serum protein in early embryonic life, can leak into the
amniotic fluid from an open NTD. The first step in prenatal
screening for open NTDs is drawing blood to measure mater-
nal serum AFP between 15 and 20 weeks of gestation.
A patient-specific risk is then calculated on the basis of gesta-
tional age and AFP level. For example, at an estimated gesta-
tional age of 20 weeks, a maternal serum AFP concentration
higher than 1000 ng/mL would be consistent with the diagno-
sis of an open NTD. Measurement of maternal serum AFP is
more than 75% accurate in detecting an open NTD when ges-
tational age is greater than 15 weeks. If the diagnosis is uncer-
tain with maternal serum AFP, an amniotic AFP level can be
obtained, which will detect approximately 98% of all open
NTDs. Moreover, fetal ultrasound can detect open and some-
times closed NTDs prenatally, especially in the hands of a
skilled ultrasonographer. If the diagnosis of myelomeningo-
cele is made prenatally, the parent or parents undergo exten-
sive prenatal counseling and cesarean delivery is planned.
EVALUATION AND TREATMENT
Neonates with myelomeningocele have a saclike protrusion
containing a neural placode bathed in cerebrospinal fluid
(CSF) (see Fig. 128-1). The size of the sac on the child’s back
at the time of birth depends on the amount of CSF that has
collected ventral to the neural placode. These children may
also have other associated CNS abnormalities including Chiari
II malformation, hydrocephalus, syringomyelia, brainstem
malformations, agenesis of the corpus callosum, and polymi-
crogyria. Shortly after birth, a thorough physical examination
is performed that includes measurement of head circumfer-
ence and assessment of general vigor (especially cry and suck);
anal sphincter function (anocutaneous reflex or anal wink);
upper and lower extremity motor and sensory function; the
site, level, and size of the myelomeningocele defect; and
whether or not the defect is leaking CSF. In addition, the
neonate should be evaluated for signs and symptoms of
hydrocephalus and brainstem compression (i.e., Chiari II mal-
formation), as well as for associated orthopedic deformities
such as clubfeet and kyphoscoliosis.
Before repair, the neonate is kept prone to prevent rupture
of the sac and avoid trauma to the neural placode. In addition,
the myelomeningocele defect should be covered with sterile
saline-soaked gauze to prevent desiccation of the exposed
neural tissue. The gauze, in turn, should be covered with a
plastic wrap to prevent heat loss.3 An intravenous catheter
is placed, and broad-spectrum antibiotics are administered
to reduce the risk of CNS infection. Head ultrasonography
or computed tomography (CT) is performed to evaluate the
extent of ventricular enlargement and determine the need
for shunt placement. Initially, the ventricles may be normal
or only slightly enlarged. However, after the NTD is closed,
the ventricles often enlarge. The incidence of hydrocephalus
associated with myelomeningocele ranges from 80% to 95%.
A myelomeningocele is typically closed within 24 to 72
hours following birth unless the infant has associated medical
conditions that prevent administration of a general anesthetic
or surgery.10 The goal of surgery is to close the neural placode
into an NT to establish a microenvironment conducive to neu-
ronal function.11 Closure involves (1) separation of the neural
placode from the intermediate zone of epithelium and recon-
struction of the placode into a tube with preservation of all
neural tissue; (2) separation of the dura from the epidural
space at the lateral margins of the defect and closure of the
dura in a watertight but patulous fashion around the newly
created NT; (3) surgical correction of any significant kyphotic
deformity; (4) mobilization and midline approximation of the
paraspinal muscles and fascia; and (5) tension-free closure of
the skin in the midline, which often requires mobilization of
the skin and subcutaneous tissue from the underlying fascia
rostrally, caudally, and bilaterally.
Common postoperative complications include CSF leak,
wound-healing problems, and tethered SC. The incidence of
CSF leak may be reduced by careful closure of the dura and
by placement of a shunt to treat associated hydrocephalus.
Wound-healing problems may occur, especially in patients
with large myelomeningoceles, and can be managed by keeping
the patient prone and performing moist-to-dry dressing
changes. Currently, no techniques are available to reduce the
incidence of SC tethering after myelomeningocele repair.
Fetal Surgery for the Treatment of Neural
Tube Defects
Intrauterine repair of myelomeningocele has been advocated
as a means of improving the neurologic outcome and reducing
hindbrain herniation in infants with myelomeningocele. This
is based on the idea that secondary damage and resultant
 CHAPTER 128 MANAGEMENT OF NEURAL TUBE DEFECTS, HYDROCEPHALUS, REFRACTORY EPILEPSY, AND CNS INFECTIONS
disability, which may occur when exposed neural tissue is in
contact with amniotic fluid, may be reduced or even
completely eliminated by closing the defect as early as possible
in utero. Moreover, halting CSF loss by in utero closure of the
neural placode may reverse some of the potentially devastating
neurologic sequelae of NTDs such as shunt-dependent hydro-
cephalus and Chiari II malformation, by reducing hindbrain
herniation.
A single-institution, nonrandomized observational study
conducted between 1990 and 1999 compared outcomes
including the requirement for ventriculoperitoneal shunt
placement, obstetric complications, gestational age at delivery,
and birth weight in 29 patients who underwent intrauterine
myelomeningocele repair between 24 and 30 weeks of gesta-
tion with 23 lesion-matched controls who underwent stan-
dard postnatal repair.12 The results of this study suggested
that intrauterine repair decreased the incidence of hindbrain
herniation and shunt-dependent hydrocephalus. However,
this study also demonstrated an increased risk of oligohy-
dramnios and admission to the hospital for preterm uterine
contractions, an earlier estimated gestational age at delivery
(33.2 vs. 37.0 weeks), and a smaller birth weight (2171 vs.
3075 g) in patients who underwent intrauterine repair. Other
problems with this study included relatively short follow-up
times, small sample size, varied selection criteria, and lack
of a comparable control group of children with myelomenin-
gocele who did not undergo intrauterine repair.
Thus, the study by Bruner and others12 provided prelim-
inary evidence that suggested that intrauterine myelomenin-
gocele repair may decrease the incidence of hydrocephalus
and reduce the severity of hindbrain herniation (including
the incidence of Chiari II malformation). However, this study
also showed that intrauterine repair involved substantial risks
not associated with standard postnatal repair. Moreover, an-
other study demonstrated that intrauterine repair between 20
and 28 weeks of gestation did not statistically improve lower
extremity function when compared with lower extremity
function in patients undergoing postnatal closure.13 For the
risks and benefits of intrauterine repair to be more clearly
established, a multicenter, prospective, randomized con-
trolled trial of intrauterine and conventional therapies, the
management of myelomeningocele study (MOMs), was
funded by the National Institutes of Health.14–16 The MOMs
trial began in 2003 and was carried out at three hospitals in
the United States. In this trial, fetuses were randomized to
surgery at 19–25 weeks of gestation or to standard postnatal
repair. The primary objective of the trial was to ascertain
whether intrauterine repair of myelomeningocele at 19 to
25 weeks of gestation improved outcome as measured by
death or the need for a shunt by one year of life.15 Other out-
comes included a composite of mental development and
motor function at 30 months, the degree of Chiari II malfor-
mation, and maternal morbidity. The trial was stopped for ef-
ficacy of prenatal surgery after 183 of a planned 200 patients
were recruited, and the results of the study were based on 158
patients whose children were evaluated at 12 months. This
study provided evidence that prenatal repair of myelomenin-
gocele reduced the need for shunting and improved motor
outcomes at 30 months when compared with standard post-
natal repair. However, in contrast to postnatal repair, prenatal
surgery was associated with an increased risk of preterm de-
livery, as well as uterine dehiscence at delivery.
1677
Associated Anomalies
Children with myelomeningocele often have other anomalies
of the CNS that require attention. For example, more than
90% of children with myelomeningocele have a Chiari II mal-
formation and hydrocephalus. Chiari II malformation is a
complex deformity involving the calvarium, dura, and
hindbrain. The spectrum of abnormalities observed in Chiari
II malformation includes a lacunar skull (i.e., thinning and
scalloping of the calvaria producing a “copper-beaten” appear-
ance), small posterior fossa, low-lying transverse sinuses and
torcular Herophili, fenestrated falx, heart-shaped tentorial
incisura with upward herniation of the cerebellum, medullary
kinking, beaking of the tectal plate, prominence of the massa
intermedia, enlargement of the suprapineal recess, elongation
of the fourth ventricle, syringomyelia, and downward dis-
placement of the cerebellar vermis, fourth ventricle, medulla,
and pons through the foramen magnum (Fig. 128-5, A and B).
Neonates with symptomatic Chiari II malformation commonly
have inspiratory stridor, apnea, dysphagia or nasal regurgita-
tion, aspiration, weak or absent cry, weakness/spasticity in
the upper or lower extremities (or both), and opisthotonic pos-
turing. Older children and adolescents have a more insidious
manifestation consisting of syncopal episodes, nystagmus,
oscillopsia, lower cranial nerve palsies, hyperreflexia, and spas-
tic quadriparesis. Radiologic diagnosis of Chiari II malforma-
tion is best made by brain magnetic resonance imaging
(MRI).Treatment of patients with symptomatic Chiari II malfor-
mation (i.e., those with signs and symptoms of brainstem com-
pression) involves decompression of the posterior fossa or
cervical SC, or both, which can be surgically challenging be-
cause the torcular Herophili and transverse sinuses are low ly-
ing (i.e., near the foramen magnum), the cerebellum is often
adherent to the medulla, and the fourth ventricle is displaced
caudally and dorsally. Before subjecting the patient to a Chiari
decompression, one must make certain that the patient has a
functioning shunt because shunt malfunction can produce
signs and symptoms of brainstem compression similar to those
observed in a Chiari II malformation. When evaluating shunt
function, it is important to remember that CT findings can
be misleading because the ventricles may remain small or fail
to show an increase in size despite a high-grade shunt obstruc-
tion. Therefore a shunt tap, radionuclide shunt injection, or
shunt exploration should be performed to evaluate shunt func-
tion before decompressing a Chiari II malformation in a patient
with myelomeningocele. Unfortunately, many patients fail to re-
spond to surgical decompression of their Chiari II malforma-
tion. This is thought to be related at least in part to the
possibility that their symptoms arise from incomplete formation
of the brainstem nuclei rather than to brainstem compression
from the Chiari II malformation.17
Hydrocephalus, which is discussed in this chapter in
greater detail later, is present in 80% to 95% of children with
myelomeningocele. Factors contributing to the development
of hydrocephalus in this patient population include the Chiari
type II malformation, aqueductal stenosis, fourth ventricle
outlet obstruction, anomalous venous drainage in the poste-
rior fossa caused by compression of the dural venous sinuses,
open myelomeningocele with associated CSF leak, and the
presence of other CNS malformations. Hydrocephalus is usu-
ally treated with a ventriculoperitoneal shunt, which diverts
CSF from the brain to the peritoneal cavity for reabsorption.
1678 PART IX SPECIAL AREAS

A B
FIGURE 128-5 Five-year-old girl with myelomeningocele repaired in utero. A, Sagittal brain magnetic resonance image (MRI) showing a Chiari II mal-
formation with upward herniation of the cerebellum and tectal plate beaking (asterisk), as well as downward displacement of the cerebellar tonsils through
the foramen magnum (arrow). B, Sagittal spine MRI showing syringomyelia extending from C5 to L2 (upper arrow). The spinal cord and conus end at L4
(lower arrow), consistent with a tethered spinal cord.

A shunt may be placed at the time of myelomeningocele
closure in patients with severe symptomatic hydrocephalus
with associated macrocrania and massive ventriculomegaly
apparent at birth or several days later under separate anesthe-
sia to reduce the risk of shunt infection. CSF diversion is in-
dicated in patients with rapidly progressive hydrocephalus
and in those manifesting acute neurologic change such as stri-
dor, swallowing dysfunction, or central apnea, with or without
significant change in the size of the ventricles.3
OCCULT SPINAL DYSRAPHISM-CLOSED
NEURAL TUBE DEFECTS
During secondary neurulation, the caudal part of the NT de-
velops from the tail bud, a mesenchymal mass of cells derived
from remnants of the cranial part of the primitive streak. Sec-
ondary neurulation occurs in three stages: (1) formation of
the medullary cord from the tail bud; (2) cavitation of the
central aspect of the medullary cord, which results in the for-
mation of multiple lumina; and (3) coalescence of the lumina
into a single, central cavity surrounded by a peripheral layer
of radially oriented medullary cord cells.18 When secondary
neurulation goes awry, closed NTDs result in which the over-
lying skin is intact but the caudal NT and its membranous
coverings develop abnormally. Developmental anomalies of
the caudal SC include spina bifida occulta; tight or thickened
filum terminale; split cord malformations (also known as
diastematomyelia, diplomyelia); neurenteric cysts; dermal si-
nuses; sacral agenesis/caudal regression; and lipomatous
malformations such as lumbosacral lipoma, leptomyeloli-
poma, lipomyelomeningocele, and fatty filum. Such anoma-
lies may lead to SC tethering, which, in turn, can cause
progressive neurologic deterioration secondary to traction
on the conus medullaris and resultant ischemic injury.19
Cutaneous stigmata are often found in association with oc-
cult spinal dysraphism and thus serve as markers for SC teth-
ering, especially when they occur above the gluteal cleft. Such
skin lesions include lumbosacral hypertrichosis (i.e., hairy
patch), dermal sinus tract, sacral dimple, capillary hemangi-
oma, caudal appendage, and subcutaneous lipoma. Patients
with one or more of these lesions above the gluteal cleft should
undergo further evaluation including a full spine MRI to look
for SC tethering. In contrast, a skin lesion such as a dimple or
pit below the gluteal cleft most likely represents a benign, non-
neurologic finding that requires no further workup.
Besides cutaneous stigmata, children with tethered cord
syndrome and occult spinal dysraphism frequently have
(1) dysraphic posterior spinal elements and vertebral abnor-
malities; (2) orthopedic abnormalities such as scoliosis, lower
extremity atrophy or asymmetry, and foot deformities;
(3) progressive neurologic dysfunction such as lower extrem-
ity weakness, loss of sensation, radiculopathy, spasticity,
hyperreflexia, and abnormal gait; (4) urologic dysfunction
such as neurogenic bladder and urinary incontinence; and
(5) bowel dysfunction. In addition, there is a high incidence
of caudal SC abnormalities and associated SC tethering in pa-
tients with anorectal and urogenital malformations including
cloacal exstrophy, persistent cloaca, penoscrotal transposition,
 CHAPTER 128 MANAGEMENT OF NEURAL TUBE DEFECTS, HYDROCEPHALUS, REFRACTORY EPILEPSY, AND CNS INFECTIONS
imperforate anus, the Currarino syndrome, and the VATER/
VACTERL association (vertebral defects, imperforate anus,
tracheoesophageal fistula with esophageal atresia, and radial
and renal dysplasia with or without cardiac and limb abnor-
malities).20 Although the explanation for the simultaneous
development of cloacal and caudal SC malformations remains
unknown, it is likely related at least in part to the intimate tem-
porospatial relationship of urogenital, anorectal, and caudal
NT development. Because of the high incidence of coexistent
malformations and SC tethering, patients with anorectal and
urogenital malformations should undergo a routine screening
lumbosacral spine ultrasound or MRI to evaluate for asymp-
tomatic tethering of the SC.
Lipomatous SC malformations such as fatty filum, lipomye-
lomeningocele, and leptomyelolipoma cause tethering of the
SC and are the most common types of closed NTDs requiring
neurosurgical treatment.19,21,22 A fatty filum consists of a short,
thickened filum terminale in which there is partial or complete
fatty infiltration. A lipomyelomeningocele consists of a skin-
covered subcutaneous lipoma that extends through a defect
in the lumbosacral fascia, lamina, dura, and pia into a low-lying
SC. Leptomyelolipoma, a type of lipomyelomeningocele,
consists of a conus lipoma. Most children with lipomatous mal-
formations have intact neurologic function at birth and are
brought to medical attention for evaluation of a subcutaneous
lipoma or other cutaneous stigmata as described earlier.
The natural history of tethered SC is that of progressive
neurologic deterioration with deficits in sensory, motor,
bowel, and bladder function arising during periods of rapid
growth and weight gain.21 As a consequence, early prophylac-
tic untethering is recommended for all patients to stabilize
neurologic function and prevent irreversible injury. Surgery
involves microsurgical technique with monitoring of anal
sphincter electromyography and somatosensory evoked po-
tentials. Intraoperative stimulation of nerve roots helps distin-
guish functional from nonfunctional roots. In the case of a
fatty filum, surgical treatment consists of a partial laminect-
omy at L4-L5 or L5-S1 through a short posterior midline
incision and opening of the dura in the midline to expose
the filum. After stimulating the filum to ensure that no neural
function is present, the filum is sectioned at the rostral and
caudal ends of the exposure and the intervening segment is
sent to pathology for evaluation. In the case of lipomyelome-
ningocele or leptomyelolipoma, surgical treatment involves
detaching the subcutaneous lipoma from the SC or conus li-
poma at the point where it emerges through the defect in
the lumbosacral fascia. It is unnecessary to resect the subcu-
taneous fatty mass completely because doing so can result
in devascularization of the overlying skin, which in turn
can compromise wound healing.22 A laminectomy is per-
formed rostral to the dural defect to identify normal anatomy,
followed by careful dissection progressively more caudally to
the region where the lipoma traverses the dural defect and en-
ters the SC or conus. Subsequent steps include (1) releasing
the conus from its attachments to the lipoma, leptomeninges,
and dura; (2) decompressing the SC and conus by debulking
the intramedullary lipomatous mass with the laser or ultra-
sonic aspirator; (3) sectioning the filum; (4) closing the dura
in a watertight but patulous fashion; and (5) closing the para-
spinal muscles, lumbosacral fascia, and skin using meticulous
technique. Surgery-related complications include anesthesia-
related risks, worsening of neurologic or urologic function
(or both), CSF leak, and infection.
1679
Other types of occult spinal dysraphism include split cord
malformations (a.k.a. diastematomyelia), dermal sinus tracts,
and neurenteric cysts. In split cord malformations, the SC is
split into two halves in the sagittal plane, with each hemicord
residing either together in a single dural tube (type I malfor-
mation) or in separate dural tubes (type II malformation).23
A hairy patch usually overlies the region of the split cord mal-
formation, and tethering of the SC may occur as a conse-
quence of a thickened filum or a midline bony spur, or
both, in the case of type II malformations or dorsal tethering
bands/fibrous median septum between the dura and hemi-
cords in the case of type I malformations. To prevent neuro-
logic and urologic deterioration, the SC is untethered by
first resecting the midline bony spur or sectioning the dorsal
bands/fibrous septum between the dura and hemicords
followed by sectioning of the filum.
Dermal sinus tracts often present in childhood with cuta-
neous findings, neurologic deficits, or infectious or chemical
meningitis, or any combination of these findings.24,25 Dermal
sinus tracts represent remnants of incomplete NT closure, de-
veloping as a consequence of localized failure of dysjunction,
the process whereby the surface ectoderm and dermal ele-
ments separate from the neuroectoderm. Failure of dysjunc-
tion generates an epithelial-lined tract that extends from an
opening on the surface of the skin to the fascia, dura, or
SC. Dermal sinus tracts can be found anywhere along the mid-
line of the neuraxis and may be associated with drainage of
CSF, intradural dermoid or epidermoid cysts, split cord mal-
formations,23 and tethering of the SC. Skin lesions associated
with dermal sinus tracts are typically found in the lumbodor-
sal midline rostral to the gluteal cleft and include a dimple, pit,
capillary hemangioma, hairy patch, skin tag, and subcutane-
ous lipoma. In the case of a dimple or pit, there may be drain-
age of cyst contents or CSF or local signs of infection such as
erythema or induration. MRI is the neurodiagnostic tool of
choice because it enables visualization of the dermal sinus
tract and associated pathology such as a tethered SC, inclusion
tumor, syrinx, or split cord malformation. However, some der-
mal sinus tracts may not be well visualized on MRI if they are
small or out of the plane of imaging.24 Hence if a patient has a
cutaneous finding overlying the midline neuraxis that is above
the gluteal cleft, surgical exploration may be warranted even if
the MRI appearance is normal. Treatment consists of complete
excision of the dimple and tract, intradural exploration with
resection of any intradural connections or masses, and
untethering of the SC. Timely diagnosis and appropriate sur-
gical intervention with intradural exploration are essential in
preventing infection including meningitis and preserving or
improving neurologic function.
Spinal neurenteric cysts are rare congenital malformations
lined by alimentary tract mucosa.19 They form as a conse-
quence of a persistent abnormal connection between the prim-
itive ectoderm and endoderm26 and are often accompanied by
vertebral abnormalities and other forms of occult spinal dys-
raphism such as split cord malformation, lipoma, dermal sinus
tract, and tethered SC. Patients with spinal neurenteric cysts
may have cutaneous stigmata or signs of SC compression such
as pain and neurologic deficits. Treatment involves complete
excision of the neurenteric cyst, as well as appropriate surgical
management of other associated congenital abnormalities
including untethering of the SC. Gross total resection is
essential because there is a high incidence of cyst recurrence
associated with subtotal resection of neurenteric cysts.26
1680 PART IX SPECIAL AREAS
In patients with occult spinal dysraphism and tethered SC
who have undergone an untethering procedure, progressive
neurologic or urologic deterioration may signal recurrent tether-
ing of the SC, which can occur in up to 15% of patients.27 There-
fore long-term neurologic and urologic follow-up is warranted to
determine patients who may benefit from reoperation.
OUTCOME AND PROGNOSIS
Treatment of NTDs has evolved over the past 50 years. Previ-
ously, neonates with NTDs either were left untreated or were
treated selectively; however, most of them died of meningitis,
hydrocephalus, and/or sepsis. In contrast, during the past 3
decades, neonates with NTDs have received prompt, aggres-
sive treatment in almost all pediatric centers in the North
America. Such treatment including early closure of open
NTDs and aggressive shunting of hydrocephalus leads to sur-
vival with nearly normal intelligence in many patients.
Initial closure of an open NTD is just the beginning of the
medical care typically required. In general, such patients re-
quire frequent and consistent follow-up in a comprehensive
multidisciplinary clinic where they can be evaluated by a team
of pediatric specialists. The team would include a develop-
mental pediatrician, neurosurgeon, urologist, orthopedic sur-
geon, physiatrist, physical therapist, occupational therapist,
nurse, and nutritionist. The multidisciplinary team approach
is critical to the ultimate success and long-term management
of these patients. With proper medical care, children with
open NTDs can lead active and productive lives. For example,
in a 20- to 25-year follow-up study of children with open
spina bifida who were treated aggressively in a nonselective,
prospective manner, these children entered college in the
same proportion as the general population and many were ac-
tively employed.28 With aggressive treatment and a multidis-
ciplinary clinical approach, long-term survival into adulthood
and advanced age is now common.
The goal of treating children with myelomeningocele is
to maintain stable neurologic functioning throughout their life-
time.3 Neurosurgeons must always be on the alert for neurologic
deterioration in children and adults with myelomeningocele.
A B
FIGURE 128-6 One-day-old girl with severe obstructive hydrocephalus secondary to congenital aqueductal stenosis. A and B, Head computed tomog-
raphy (CT) scan showing severe ventriculomegaly. C, Head CT scan 1 day after shunt placement.
Besides shunt malfunction, the most common cause of neuro-
logic deterioration is symptomatic SC tethering in which there
is tension on the SC resulting from fixation of the SC due to ad-
hesions between the previously exposed neural tissue and the
surrounding tissues. In almost all patients with myelomeningo-
cele, the SC is low lying and ends in the lumbar or sacral region.
This may be observed in patients without any new neurologic
complaints. However, symptomatic tethering of the SC develops
in at least 20% of all patients with myelomeningocele despite
careful surgical closure of the original neural placode. Such pa-
tients often have one or more of the following signs/symptoms:
gait difficulty, back pain, leg weakness, sensory loss, a new foot
deformity, a change in urodynamic data, or urinary inconti-
nence. When one or more of these signs/symptoms occur in pa-
tients with shunted hydrocephalus, the shunt must be evaluated
first to confirm that it is functioning appropriately. After this has
been established, surgery is performed to free the neural placode
and nerve roots from the dorsal surface of the dura and thereby
untether the SC. In contrast, asymptomatic patients who demon-
strate SC tethering on routine MRI do not require reoperation.
Hydrocephalus
------------------------------------------------------------------------------------------------------------------------------------------------
Hydrocephalus is a common pediatric disorder in which there
is an increase in CSF volume, which in turn causes enlarge-
ment of the ventricles, thinning of the cortical mantle
(Fig. 128-6), and elevation of intracranial pressure (ICP).
The incidence of congenital hydrocephalus is approximately
0.9 to 1.8 per 1000 births,29 and the mortality rate is approx-
imately 1% per year. To prevent neurologic deterioration
associated with increased ICP, CSF diversion is required.
Treatment with CSF diversion techniques such as valve-
regulated CSF shunt systems and endoscopic third ventricu-
lostomy increases life expectancy in pediatric patients with
hydrocephalus and improves their intellectual outcome.
Hydrocephalus results from a disparity between CSF pro-
duction and absorption. Most cases of hydrocephalus occur
as a consequence of impaired CSF absorption. The exception
to this rule is hydrocephalus in patients with a choroid plexus
papilloma, which results, at least in part, from overproduction
C
 CHAPTER 128 MANAGEMENT OF NEURAL TUBE DEFECTS, HYDROCEPHALUS, REFRACTORY EPILEPSY, AND CNS INFECTIONS
of CSF. Increased CSF volume secondary to impaired absorp-
tion or increased production (or both) leads to progressive
ventricular dilatation. In some children, CSF absorption can
occur through alternative pathways, thereby resulting in stabi-
lization of their ventricular enlargement. This, in turn, leads to
arrested or compensated hydrocephalus. In support of this, a
detailed analysis of the natural history of unoperated hydro-
cephalus has shown that arrest of the progressive ventricular
enlargement eventually occurs in up to 45% of children with
hydrocephalus.30 Children with compensated hydrocephalus
do not require CSF diversion; however, many factors such as
fever and infection can cause sudden decompensation of the
hydrocephalus and elevated ICP. Therefore children with
suspected shunt-independent “compensated” or “arrested”
hydrocephalus must be monitored closely for signs and symp-
toms suggesting decompensation such as headaches, vomiting,
ataxia, and visual symptoms.
Formation of CSF is an energy-dependent, ICP-independent
process that requires carbonic anhydrase and occurs at a rate of
approximately 450 mL/day, or 0.3 mL/min. Choroid plexus in
the lateral, third, and fourth ventricles produces 50% to 80%
of the CSF; the remaining 20% to 50% of CSF is produced by
the ventricular ependyma and brain parenchyma as a byproduct
of cerebral metabolism. CSF flows in a caudal direction through
the ventricular system and exits by way of the foramina of
Luschka and Magendie into the cortical and spinal subarachnoid
space. CSF then travels through the tentorial incisura, passes
over the hemispheric convexity, and is absorbed into the venous
system at the level of the arachnoid villi.
CSF absorption is an energy-independent process that oc-
curs predominantly by bulk flow through the arachnoid villi,
which represent the interface between the cortical subarach-
noid space and the intradural venous sinuses. The rate of
CSF absorption depends on a pressure gradient from the sub-
arachnoid space across the arachnoid villi to the venous space
in the dural venous sinuses. When ICP is normal (i.e., 7 cm
H2O or 5 mm Hg), CSF is produced at a rate of 0.3 mL/min
but no CSF absorption occurs. When ICP increases, CSF ab-
sorption occurs in direct proportion to the increase in ICP.
Other pathways for CSF absorption have been proposed. Such
pathways including the lymphatic system, nasal mucosa, para-
nasal sinuses, and the nerve root sleeves of cranial and spinal
nerves are probably recruited for CSF absorption when ICP is
elevated; however, definitive evidence for these alternative
pathways is currently lacking.
Obstruction of CSF flow at any point along the ventricular
pathway or impaired absorption of CSF generates increased
intraventricular pressure, which in turn results in progressive
dilatation of the ventricular system proximal to the block. The
pathologic effects of ventricular obstruction or impaired CSF
absorption include ventricular dilatation, which is often more
pronounced initially in the occipital horns, thinning of the
cortical mantle, disruption of the ependymal membrane,
transependymal absorption of CSF into the periventricular
white matter, white matter injury and scarring, elevation of
ICP, brain herniation, coma, and ultimately, death.
ETIOLOGY
The etiology of hydrocephalus depends on the site of the ob-
struction of CSF flow. If the obstruction is proximal to the
arachnoid villi, there is selective enlargement of the ventricles
1681
proximal to the obstruction and the resultant hydrocephalus
is termed noncommunicating or obstructive hydrocephalus. For
example, obstruction of the aqueduct of Sylvius results in en-
largement of the lateral and third ventricles out of proportion
to the fourth ventricle. In contrast, when the block is at the
level of the arachnoid villi and CSF absorption is impaired,
the lateral, third, and fourth ventricles become dilated and
the volume of CSF in the subarachnoid space is increased.
The resultant hydrocephalus is referred to as communicating
or nonobstructive hydrocephalus. Of note, hydrocephalus
ex vacuo, a condition in which the ventricles become
enlarged as a consequence of cerebral atrophy, is not true
hydrocephalus.
Obstructive hydrocephalus can be caused by blockage at
the foramen of Monro. Common causes include congenital
atresia or stenosis; intracranial cysts such as arachnoid cysts
within the subarachnoid space or ventricle, porencephalic
cysts within the brain adjacent to the ventricle, and colloid
cysts; tumors such as hypothalamic gliomas, craniopharyngio-
mas, and subependymal giant cell astrocytomas; and cavern-
ous malformations. Obstruction at the level of the aqueduct of
Sylvius is another common cause of obstructive hydrocepha-
lus (see Fig. 128-6). Causes of aqueductal obstruction include
upward herniation of the cerebellum through the tentorial
incisura in patients with myelomeningocele and associated
Chiari II malformation, vein of Galen vascular malformations,
gliosis of the aqueduct from infection or hemorrhage, tumors
of the pineal region, and tectal plate gliomas. In addition, ob-
structive hydrocephalus can occur as a consequence of fourth
ventricle outflow obstruction from posterior fossa and fourth
ventricle tumors such as medulloblastomas, ependymomas,
and pilocytic astrocytomas; a Dandy-Walker malformation,
in which there is a large posterior fossa cyst that communi-
cates with an enlarged fourth ventricle, as well as atresia of
the outlet foramina (i.e., lateral foramina of Luschka and
midline foramen of Magendie); and a Chiari II malformation,
in which there is herniation of the fourth ventricle through the
foramen magnum because of a small posterior fossa.
In communicating or nonobstructive hydrocephalus, the
block can occur at the level of the basal cisterns. When this
occurs, CSF is blocked between the spinal and cortical sub-
arachnoid spaces and cannot reach the arachnoid villi for ab-
sorption. As a consequence, the lateral, third, and fourth
ventricles become dilated. Common causes include congenital
infections; meningitis from acquired pyogenic, tuberculous,
and fungal infections; subarachnoid hemorrhage from aneu-
rysm rupture, a vascular malformation, or trauma; intraven-
tricular hemorrhage associated with a germinal matrix
hemorrhage in a preterm infant; basal arachnoiditis; leptome-
ningeal carcinomatosis; neurosarcoidosis; and tumors pro-
ducing high protein levels in the CSF. The block can also
occur at the level of the arachnoid villi from occlusion or atre-
sia of the arachnoid villi, which results in dilatation of the sub-
arachnoid space and ventricles. Furthermore, the block can
occur at the level of the dural venous sinus from venous out-
flow obstruction. Such a block can be seen in patients with
achondroplasia or multisutural craniosynostosis who have
stenosis of the jugular foramina, in patients with high right
atrial pressure from congenital heart disease, and in those with
thrombosis of the dural venous sinuses or superior vena cava.
Venous outflow obstruction causes increased venous pressure,
which in turn results in decreased drainage of cortical veins,
1682 PART IX SPECIAL AREAS
increased cerebral blood volume, elevated ICP, increased brain
stiffness, and decreased CSF absorption. In infants with an
open anterior fontanelle and cranial sutures, increased venous
pressure and decreased CSF absorption result in macroce-
phaly, with splitting of the cranial sutures and ventricular en-
largement. In contrast, when the cranial sutures and anterior
fontanelle are closed, the elevated venous pressure produces
venous engorgement, which in turn raises ICP. In such pa-
tients the ventricular system is compressed despite decreased
CSF absorption, and the clinical entity known as pseudotumor
cerebri is produced.
Maximal hydrocephalus is characterized by extreme ven-
tricular enlargement with only a thin rim of cortical mantle,
the presence of background cortical activity throughout the
brain on electroencephalography (EEG), and at least some
restitution of the cortical mantle with CSF diversion (see
Fig. 128-6). In contrast, hydranencephaly is a condition in
which the intracranial cavity is filled with CSF instead of
brain because of a total or nearly total loss of brain tissue sup-
plied by the anterior and middle cerebral arteries bilaterally
(Fig. 128-7). The cranial vault and meninges remain intact
along with the thalamus, brainstem, and a small amount of
occipital lobe supplied by the posterior cerebral artery. The
most common causes of hydranencephaly are bilateral internal
carotid artery infarcts and infection. Cortical activity is absent
on EEG. Such infants cry, suck, and feed; they are typically
hyperirritable; they retain primitive reflexes; and rarely pro-
gress beyond spontaneous vowel production or a social smile.
CSF shunting may be performed to control head size in the
face of progressive head enlargement; however, shunting does
not improve neurologic function or reduce hyperirritability.
CLINICAL FEATURES
In infants, hydrocephalus results in irritability, lethargy, failure
to thrive (with or without vomiting), delayed development,
apnea, bradycardia, hyperreflexia, hypertonia, increasing
head circumference, bulging anterior fontanelle, splaying of
the cranial sutures, thin scalp with distended scalp veins, fron-
tal bossing, impaired upward gaze with eyelid retraction (i.e.,
“setting sun” sign from pressure on the tectal plate), decreased
A B
FIGURE 128-7 Two-day-old girl with hydranencephaly. A, Transcranial illumination showing the superior sagittal sinus (arrow) and a scant amount of
right inferior temporal and occipital cortex (arrowheads). B, Head computed tomographic scan shortly after birth revealing a scant amount of right cerebral
cortex posteriorly (arrow). The cerebral hemispheres are replaced almost entirely by cerebrospinal fluid.
levels of consciousness, and papilledema, as well as third,
fourth, and sixth nerve palsies. In older children with a rigid
cranial vault, hydrocephalus can cause headache (especially in
the morning), nausea, vomiting, lethargy, papilledema, deteri-
oration in vision, decline in cognitive function or behavior,
memory problems, decreased attention span, worsening of
school performance, gait changes, upgaze palsy, diplopia
(especially from sixth nerve palsy), and seizures.
RADIOLOGIC FEATURES
OF HYDROCEPHALUS
Plain skull films from patients with hydrocephalus often dem-
onstrate a beaten-copper appearance and splitting of the cor-
onal sutures. CT and MRI frequently reveal dilated temporal
horns; obliteration of the sylvian and interhemispheric fis-
sures, sulci, and basilar cisterns; ballooning of the frontal
horns and third ventricle; upward bowing or atrophy of the
corpus callosum, or both; and periventricular edema related
to transependymal absorption of CSF.
Benign external hydrocephalus (also known as benign
extra-axial fluid of infancy), a condition that rarely requires
CSF diversion, is characterized by enlargement of the anterior
cortical subarachnoid spaces (i.e., sulci and cisterns) bilater-
ally; normal or mildly dilated ventricles; a prominent pulsatile
anterior fontanelle; and progressive enlargement of head cir-
cumference, with crossing of percentile lines when head cir-
cumference is plotted with respect to age. Infants with
benign external hydrocephalus have large heads and may
demonstrate slight delays in motor development related to
their large head size. The etiology of this clinical entity is
unclear; however, defective CSF absorption has been pro-
posed. Frequently, the family history is significant for large
heads. At approximately 12 to 18 months of age, the enlarging
head circumference tends to plateau, which allows the child’s
body growth to catch up with head growth, and the hydro-
cephalus typically resolves spontaneously by 2 years of age
without a need for shunting. Although shunting is not usually
required, these children should be monitored closely with
serial head circumference measurements and head CT or
ultrasound to evaluate for abnormal ventricular enlargement.
 CHAPTER 128 MANAGEMENT OF NEURAL TUBE DEFECTS, HYDROCEPHALUS, REFRACTORY EPILEPSY, AND CNS INFECTIONS
MANAGEMENT
The goal of treatment is to achieve optimum neurologic func-
tion and prevent or reverse neurologic damage caused by dis-
tortion of the brain from ventricular enlargement. The best
predictor of a good outcome is postoperative reconstitution
of the cerebral mantle to at least 2.8 cm.31 This is more likely
to occur if a symptomatic hydrocephalic infant undergoes
shunting by 5 months of age. The treatment of hydrocephalus
is surgical, and the type of surgery depends on the cause of the
hydrocephalus. For example, in patients with obstructive hy-
drocephalus, surgery is performed to remove the obstruction
(e.g., tumor). In patients with communicating hydrocephalus
or obstructive hydrocephalus in which the obstruction cannot
be removed, CSF is shunted from the site of its secretion
within the ventricle proximal to the obstruction to a distal site
capable of reabsorption. The most commonly used distal sites
include the peritoneal cavity, right atrium, pleural cavity, gall-
bladder, bladder/ureter, and basilar cisterns (i.e., third ventri-
culostomy). The preferred site for the distal catheter is the
peritoneal cavity (i.e., ventriculoperitoneal shunt), unless
there is a problem with absorption or abdominal sepsis. When
abdominal abnormalities such as extensive postsurgical
abdominal adhesions, peritonitis, morbid obesity, and necro-
tizing enterocolitis are present in preterm infants, a ventricu-
loatrial shunt is the treatment of choice. Alternatively, if the
peritoneum and atrium are not available, a ventriculopleural
shunt can be placed in children who are 7 years of age or older.
CSF shunt systems consist of at least three components: a
ventricular catheter, which is usually placed in the occipital
or frontal horn of the lateral ventricle (see Fig. 128-6, C); distal
shunt tubing to drain CSF into a distal site for reabsorption; and
a valve. Conventional differential pressure shunt valves control
unidirectional CSF flow by opening at a fixed pressure differ-
ential across the valve. This pressure differential is determined
by the characteristics of the valve and is designated as low, me-
dium, or high (typically 5, 10, and 15 cm H2O, respectively).
Once the valve is open, it provides little resistance to CSF flow.
Consequently, the gravitational effects of an upright posture
can lead to high CSF flow rates, which in turn can generate a
large negative ICP, a process referred to as “siphoning.”
Because siphoning can result in slit ventricles, which in turn
are associated with a higher incidence of proximal shunt ob-
struction, some shunt valves have been designed to limit excess
CSF flow, particularly in the upright position.32 For example,
the Delta valve (Medtronic PS Medical, Goleta, CA) is a stan-
dard diaphragm-type, differential pressure shunt valve with
an integral siphon control device to diminish overdrainage in
upright positions. The Orbis-Sigma valve (Cordis, Miami) con-
tains a variable-resistance, flow-limiting component that limits
excess CSF flow by progressively narrowing the flow orifice
with increasing pressure. Although use of these valves has been
advocated as a means of preventing overdrainage and improv-
ing treatment results, a randomized trial comparing three
different shunt valve designs (i.e., standard differential pres-
sure valve, Delta valve, and Orbis-Sigma valve) in the treatment
of children with newly diagnosed hydrocephalus showed no
significant difference in shunt failure rates among the valves
tested.33 Currently, there are also three types of programmable
differential pressure shunt valves available for use. Such valves,
which allow the pressure setting to be altered after implanta-
tion, are commonly used in the treatment of pediatric
1683
hydrocephalus. Despite the vast array of shunt systems avail-
able to treat hydrocephalus, no prospective, randomized, con-
trolled, double-blind, multicenter trial has ever proved a
particular shunt system to be more effective than any other.
Ventriculoperitoneal shunting is commonly used to treat
childhood hydrocephalus. Placement of a ventriculoperitoneal
shunt is performed under general anesthesia after prophy-
lactic intravenous antibiotics are administered. The role of anti-
biotic prophylaxis has been studied by meta-analysis, and
antibiotic coverage is recommended.34,35 The head of the patient
is positioned on a gel donut at the top edge of the operating table
in close proximity to the surgeon and turned to the side opposite
the shunt insertion. The neck and trunk are extended with a gel
roll under the shoulders to assist subcutaneous tunneling of the
shunt. Skin incisions are delineated before skin preparation,
depending on the desired location of the burr hole for placement
of the ventricular catheter (e.g., frontal or occipital). A meticu-
lous skin preparation (e.g., with povidone-iodine scrub and
paint solution or chlorhexidine 2% in 70% alcohol) is per-
formed, and the incisions are marked again. The abdominal in-
cision is typically marked in the right upper quadrant or in the
midline approximately two to three fingerbreadths below the xi-
phoid process. Surgical drapes including an iodine-impregnated
sterile adhesive drape are placed over the incisions and tunneling
site. A curvilinear incision is made in the scalp overlying the pro-
posed burr hole site, and the scalp is retracted. The pericranium
is opened at the base of the incision, and a burr hole is drilled.
The dura, arachnoid, and pia are then coagulated and opened.
Next, the abdominal incision is opened and entrance into
the peritoneal cavity is achieved and confirmed by gently prob-
ing the opening into the peritoneal cavity. A shunt tunneling de-
vice is then used to create a subcutaneous tunnel extending
between the two incisions. The shunt components are removed
from their sterile packages just before shunt placement and
placed into antibiotic solution. The distal shunt system includ-
ing the valve is passed within the subcutaneous tunnel, and the
proximal and distal ends of the shunt are covered with sponges
soaked in antibiotic solution. Care is taken to prevent contact of
the shunt tubing with the skin during this portion of the pro-
cedure. After this, the ventricular catheter is passed on a wire
stylet into the ventricle. Intraoperative ultrasound or ventricu-
lar endoscopy, or both, can be used to accurately place the cath-
eter within the ventricle. Ventriculoscopy can also be used to
fenestrate intraventricular cysts and provide communication
between loculated ventricles.
Once the catheter is in the ventricle and CSF flow has been
confirmed, the catheter is cut to the desired length, guided by
measurements obtained from the preoperative head CT scan,
and then connected to the distal shunt system, usually by way
of a reservoir. Silk ties (2-0 or 3-0) are used to secure the catheter
to the reservoir. After this, the distal shunt tubing is evaluated for
CSF flow. Once confirmed, the distal shunt tubing is passed into
the peritoneal cavity; it should pass freely without resistance. The
incisions are then irrigated with antibiotic solution, closed in a
meticulous, two-layered fashion with good approximation of
the skin edges, and covered with sterile dressings.
COMPLICATIONS OF SHUNT MANAGEMENT
Most children are shunt dependent after shunt implantation.
Consequently, children with shunts require lifelong follow-up
to make certain that their shunt is functioning adequately,
1684 PART IX SPECIAL AREAS
especially in light of the high incidence of failure over the life
of a shunt. For example, in a previous retrospective study in
which 1719 hydrocephalic children were evaluated in an at-
tempt to understand why shunts fail,36 the actuarial probabil-
ity of the occurrence of shunt failure was 70% at 10 years after
the initial shunt insertion, with the highest risk for shunt fail-
ure occurring in the immediate postoperative period (i.e., a
30% risk of shunt failure during the first year and a 2% to
5% risk per year thereafter). In addition, the actuarial proba-
bility of the occurrence of a shunt infection was 9% over the
same 10-year period and the time of shunt failure varied sig-
nificantly according to the cause of the shunt failure. Other
studies have reported shunt failure rates of approximately
40% within the first year for standard CSF shunts implanted
for the treatment of pediatric hydrocephalus.37,38 Moreover,
in a randomized trial of CSF shunt valve design,33 the shunt
failure rate was 40% at 1 year and 60% at 2 years and the in-
fection rate was 8% in the hands of experienced pediatric neu-
rosurgeons working in centers where many shunts are
implanted each year.
CSF shunt systems malfunction due to three main factors:
(1) mechanical failure, (2) infection, and (3) overdrainage or
underdrainage (i.e., “functional” failure). Proximal obstruc-
tion (i.e., obstruction of the ventricular catheter) is the most
frequent cause of mechanical shunt failure.36 Other causes
include obstruction of the distal catheter or any accessory
part of the shunt (e.g., antisiphon device); formation of a
peritoneal pseudocyst, which is usually associated with in-
fection and causes obstruction of the peritoneal catheter; as-
cites resulting from failure of intraperitoneal reabsorption of
CSF; fracture of the shunt tubing; disconnection of the shunt
at a junction point of shunt components (e.g., proximal or
distal to a straight connector); migration of the shunt; inad-
equate length of the distal catheter; improper proximal or
distal catheter placement; infection; and overdrainage, which
can lead to the formation of subdural fluid collections,
slit ventricles, craniosynostosis, or intracranial hypotension.
In a study by Sainte-Rose and colleagues,36 two thirds of the
shunt failures were due to obstruction or fracture, with ob-
struction responsible for 56.1% of shunt failures and frac-
tures occurring almost exclusively at sites of connectors.
Moreover, they found that a significantly higher percentage
of shunt failure was due to proximal obstruction in patients
with slit ventricles than in those with normal or enlarged
ventricles and to distal obstruction in patients with slit-
ended peritoneal catheters than in those with open-ended
catheters.
Repeated shunt failures are common in pediatric patients
and are associated with significant morbidity and occasion-
ally death. A prospective, single-institution, observational
study designed to identify risk factors predisposing pediatric
patients to repeated shunt failures39 revealed that the timing
of the previous shunt procedure is significant, with shunt re-
visions performed within less than 6 months from the time of
the preceding implantation having a significantly increased
risk for failure. Furthermore, in this study the age of the
patient at the time of first shunt insertion was a significant
risk factor for subsequent shunt failure, with patients youn-
ger than 40 weeks and those between 40 weeks and 1 year of
age at the time of shunt insertion having a higher risk for
subsequent shunt failure than those older than 1 year.
With respect to the effect of the cause of hydrocephalus on
repeated shunt failure, this study showed a higher risk for
recurrent shunt failure in patients with intraventricular hem-
orrhage, meningitis, and tumor. Finally, concurrent surgical
procedures also increased the risk for shunt failure in this
study.
In most cases of shunt malfunction, the diagnosis is obvi-
ous because the child has acute and overt signs and symptoms
of elevated ICP such as irritability, headache, vomiting, and
lethargy and will progress to stupor, coma, and death if the
shunt is not revised promptly. In the study by Sainte-Rose
and colleagues,36 there was a 1.05% mortality rate directly re-
lated to shunt failure. Alternatively, some children with shunt
failure will have more subtle signs of deterioration such as a
decline in school performance, change in attention span, or
change in behavior. These “subtle deteriorators” often demon-
strate a gradual increase in ventricular size over time and may
have a marked increase in ventricular size at the time of pre-
sentation. In contrast, acute deteriorators often have only a
slight increase in ventricular size. When a child has a sus-
pected shunt malfunction, a head CT is performed to evaluate
for interval enlargement of the ventricles. This must be com-
pared with a previous scan obtained when the patient was do-
ing well. A shunt series is also performed to look for continuity
of the shunt, especially at sites of connectors, the position of
the ventricular catheter or catheters, and the length of the
distal shunt tubing.
When a shunt malfunction is suspected, the site of shunt
failure must be determined. The most common cause of shunt
failure is obstruction of the ventricular catheter by choroid
plexus, glial tissue, connective tissue, leptomeninges, epen-
dyma, or brain tissue (or any combination of these tissues),
especially in patients with slitlike ventricles.36 Proximal ob-
struction is typically associated with precipitous development
of intracranial hypertension, which necessitates immediate
shunt revision. To reduce the incidence of proximal catheter
obstruction, surgeons attempt to place the ventricular catheter
with the use of external anatomic landmarks either within the
frontal horn of the lateral ventricle anterior to the foramen of
Monro or within the occipital horn to keep the catheter away
from choroid plexus.36 With the introduction of fiberoptic en-
doscopes in the early 1990s, surgeons began using ventricular
endoscopy to insert the ventricular catheter away from cho-
roid plexus under direct visualization. Although evidence
from uncontrolled series suggested that endoscopic insertion
of the ventricular catheter decreased the incidence of shunt
failure,40 a multicenter randomized trial evaluating time to
first shunt failure after endoscopic versus nonendoscopic
placement of the ventricular catheter in children with hydro-
cephalus showed no reduction in the incidence of shunt fail-
ure with endoscopically assisted placement of the ventricular
catheter.41 However, in this trial, ventricular catheters that
were thought by surgeons to be placed away from choroid
plexus at the time of shunt insertion were actually found to
be located away from choroid plexus on postoperative imag-
ing studies only two thirds of the time in both endoscopic and
nonendoscopic insertions. This suggests that the endoscope
did not help achieve the goal of placing ventricular catheters
away from choroid plexus in this trial. However, a secondary
analysis of catheter position in this trial did demonstrate a
reduced incidence of shunt failure when the ventricular cath-
eter was positioned away from choroid plexus on the basis of
postoperative imaging studies.
 CHAPTER 128 MANAGEMENT OF NEURAL TUBE DEFECTS, HYDROCEPHALUS, REFRACTORY EPILEPSY, AND CNS INFECTIONS
Distal shunt malfunctions can occur as a consequence of
shunt fracture, inadequate length of the distal shunt tubing,
shunt infection, and impaired absorption of CSF at the distal
site. Shunt tubing is made of silicone elastomers, which can
calcify and break over time. Breakage often occurs above
the clavicle, where there is increased motion, or in the region
of a connector. Consequently, a shunt series (i.e., anteropos-
terior and lateral skull, chest, and abdominal radiographs)
should be evaluated carefully to look for discontinuity in
the shunt system, particularly at sites of connectors. If a shunt
fracture is found, the shunt should be revised. The shunt se-
ries can also detect whether the distal shunt tubing is too short
and a distal lengthening procedure is necessary.
Shunt infection can also cause a distal shunt malfunction
and is the second leading cause of shunt failure after mechan-
ical obstruction. With each shunt operation, there is a 2% to
8% incidence of shunt infection,33 with 5% to 15% of shunts
becoming infected over the life of a shunt. The consequences
of shunt infection are devastating and can include intellectual
and focal neurologic deficits, enormous health care costs, and
death. Most shunt infections occur within the first 6 months
after a shunt operation,34 with approximately 70% of shunt
infections diagnosed within the first month after surgery
and 90% by 6 months. The most common pathogens are
staphylococci (Staphylococcus epidermidis, 40%; Staphylococcus
aureus, 20%). Others include coryneforms, streptococci, en-
terococci, aerobic gram-negative rods, and yeasts. Although
shunt infections can occur after 6 months, they are rare and
almost always result from indolent bacteria that are a normal
part of skin flora, such as S. epidermidis or Propionibacter acnes.
The fact that most shunt infections occur as a consequence of
contamination with the patient’s own skin flora underscores
the need for meticulous attention to surgical technique at
the time of shunt insertion.
Shunt infections can range from isolated wound infections
and colonization of the shunt tubing to ventriculitis, peritoni-
tis, and an infected pseudocyst. Risk factors implicated in
shunt infection include young age of patient, poor skin con-
dition, prolonged surgery, an open NTD, CSF leakage from
an incision or wound breakdown postoperatively, increased
number of shunt revisions, and concomitant infection. Pa-
tients with a shunt infection typically have a low-grade fever
(i.e., between 100 F and 102 F) or evidence of shunt mal-
function. However, they can also manifest signs of meningitis,
ventriculitis, peritonitis, or cellulitis, or any combination of
such signs. Symptoms include irritability, headache, nausea
and vomiting, lethargy, decreased appetite, abdominal pain,
erythema and tenderness along the shunt tract, photophobia,
and neck stiffness. A head CT may or may not show a change
in ventricular size. It is important for pediatric surgeons to be
aware that shunt infection may present only as an acute
surgical abdomen, often mimicking appendicitis.40a
If shunt infection is suspected, the patient should undergo
a radiologic evaluation that includes a shunt series and a head
CT scan. The shunt series provides information regarding the
continuity of the shunt tubing and location of the shunt res-
ervoir for tapping, possible sources of shunt-related problems,
and other potential sources of infection such as pneumonia.
The head CT reveals ventricular catheter location, size, and
contents including loculations and purulent fluid collections,
which can be seen in cases of severe gram-negative ventricu-
litis. If the patient complains of abdominal pain or abdominal
1685
distention is present, an abdominal CTor ultrasound should be
performed to evaluate for a CSF pseudocyst. In addition, a
peripheral white blood cell count and blood cultures should
be obtained because patients with shunt infections often have
leukocytosis and positive blood cultures. Finally, a multisystem
fever workup should be performed to evaluate for concurrent
infections.
If the initial evaluation suggests possible shunt infection, a
shunt tap is performed to obtain CSF for study. To do this, the
area over the shunt reservoir is shaved and prepared with
povidone-iodine or chlorhexidine solution. A 25-gauge but-
terfly needle is then passed percutaneously into the reservoir
using sterile technique. Proximal and distal shunt flow dy-
namics are evaluated, and CSF is obtained and sent for cell
count and differential, protein, glucose, Gram stain, and cul-
ture and sensitivity studies. Broad-spectrum intravenous
antibiotics are then started.
If CSF studies demonstrate a shunt infection (i.e., isolation
of bacteria in CSF obtained from the shunt tap, elevated neu-
trophils and protein in CSF, or decreased CSF glucose), the en-
tire shunt is removed and an external ventricular drain and
central venous line are placed. The patient is treated with
the appropriate systemic antibiotics (i.e., antibiotics that have
good activity against the isolated organism and adequate CSF
penetration) until the shunt infection has cleared, at which
time a new shunt may be placed. Monitoring of CSF during
treatment is essential. At present, significant variation exists
in the duration of antibiotic therapy, with treatment regimens
ranging from 2 days to 3 weeks.42 Studies are currently under
way to determine the most effective duration of antibiotic ther-
apy in an effort to shorten hospitalization and minimize com-
plications without sacrificing efficacy. It is the practice at
the author’s institution to treat with antibiotics until at least
two consecutive CSF cultures, at least 48 hours apart, are
negative and the inflammatory component of the infection
has resolved. Aside from the practical problems associated
with their treatment, shunt infections have been linked to
an increase in the development of loculated CSF compart-
ments, impaired intellectual outcome (with as high as an 8-
to 10-point decrease in IQ), an increased risk for seizures,
and an increased mortality rate.34
Shunt overdrainage, or “overshunting,” can also be associ-
ated with shunt failure. Most shunts with differential pressure
valves will overdrain regardless of whether the valve pressure
is high or low.32 Overdrainage can lead to the formation of
subdural hematomas, a low-ICP syndrome, or slit ventricle
syndrome. Subdural hematomas result from the tearing of
bridging veins as the ventricles collapse and the cortical sur-
face pulls away from the dura. Although they often resolve
without treatment, symptomatic or progressive subdural he-
matomas from overshunting may require a reduction in the
degree of shunting to allow the ventricles to re-expand and
the subdural space to become obliterated. Alternatively, drain-
age of the subdural space either by placing burr holes or by
shunting the subdural space with a low-pressure valve may
be required. Overshunting can also lead to an intracranial hy-
potension syndrome with symptoms such as headache, nau-
sea, vomiting, tachycardia, and lethargy that are sensitive to
changes in position. In such patients, overdrainage occurs
when the patient assumes an upright position. This produces
a negative ICP, which in turn leads to intense postural head-
aches that are relieved by recumbency.43,44 If symptoms
1686 PART IX SPECIAL AREAS
persist and are frequent or recurrent enough to interfere with
daily activities, especially school, the shunt should be revised
by placing a higher-resistance valve or adding an antisiphon
device, or both.
Overshunting can also lead to the formation of slitlike ven-
tricles. Ventricles commonly become small or slitlike after
placement of a shunt. In a previous retrospective study, slit
ventricles developed in 80% of the total population of shunted
patients.32 Interestingly, 88.5% of the patients in whom slit-
like ventricles developed were completely asymptomatic.
Moreover, of the 11.5% of patients with slit ventricles who
were symptomatic, only 6.5% required surgical intervention.
Symptomatic slit ventricle syndrome occurs infrequently and
is associated with intermittent episodes of vomiting, head-
ache, and lethargy. Patients with symptomatic slit ventricle
syndrome tend to have small ventricles, diminished extraven-
tricular CSF spaces, a thick skull, and no room for intracranial
accumulation of CSF. In this syndrome the ventricular walls
collapse around the ventricular catheter, which becomes
obstructed, and the shunt fails to drain.32 Eventually, intra-
ventricular pressure builds up, the ventricles dilate slightly,
and the shunt begins to drain. These patients often experience
acute neurologic deterioration related to waves of elevated ICP
that occur during periods of intermittent shunt obstruction.
Moreover, because of the tight intracranial space in these pa-
tients, there is no room for any increase in intracranial volume
including cerebral blood volume. Consequently, any event
that leads to cerebral vasodilatation such as migraine head-
aches or to an increase in cerebral blood flow such as exercise
or playing outside in the hot summer sun will cause symptoms
of elevated ICP.32 Patients with slit ventricle syndrome become
symptomatic early in the phase of shunt obstruction, and head
CT scans obtained in symptomatic patients frequently reveal
no change in ventricular size when compared with their usual
slitlike state.
Patients with acute neurologic deterioration in the face of a
functioning shunt may benefit from treatment with medica-
tions such as furosemide (Lasix) or acetazolamide (Diamox)
to lower ICP until the elevated pressure waves subside. Some
patients may also benefit from taking antimigraine medica-
tions such as cyproheptadine (Periactin) or propranolol
(Inderal). If symptoms persist despite conservative treatment,
surgical intervention may be necessary. Such interventions
include ventricular catheter revision, shunt valve upgrade to
increase resistance, addition of an antisiphon or flow control
device, decompressive subtemporal craniectomy ipsilateral to
the ventricular catheter, or any combination of these interven-
tions. ICP monitoring is useful to differentiate between high-
pressure headaches caused by waves of elevated ICP in the
face of a functioning shunt and low-pressure headaches
caused by negative ICP in patients with true overdrainage.45
ICP monitoring can also be helpful in identifying children
with normal ICP physiology who are having headaches
unrelated to shunt function.
Trapping of the fourth ventricle can occur in children with
chronic shunting of the lateral ventricles, especially in those
with a history of posthemorrhagic hydrocephalus from intra-
ventricular hemorrhage related to prematurity, postinfectious
hydrocephalus, or repeated shunt infections/ventriculitis. In
this condition, overdrainage of CSF leads to slit ventricles,
aqueductal narrowing, or both. When aqueductal stenosis
occurs, the fourth ventricle does not communicate with the
third ventricle. Likewise, the fourth ventricle does not com-
municate with the basilar cisterns because of occlusion of
the outlets (i.e., foramina of Luschka and Magendie). Produc-
tion of CSF by choroid plexus in the trapped fourth ventricle
results in progressive ventricular enlargement, which may
lead to headache, swallowing difficulty, lower cranial nerve
palsies, ataxia, lethargy, spastic quadriparesis, and nausea/
vomiting. Infants may have apneic spells and bradycardia.
In symptomatic patients, the isolated fourth ventricle can
be shunted with a separate shunt system or by adding a
fourth ventricular catheter into an existing supratentorial
shunt system above the valve. However, as the fourth ventri-
cle is decompressed with CSF drainage, the brainstem moves
posteriorly into a more normal position and the catheter can
injure the brainstem.46 In addition, approximately 40% of
patients require a shunt revision within 1 year.47 Alterna-
tively, a suboccipital craniotomy with open fenestration of
the fourth ventricle to the subarachnoid space and basal
cisterns may be performed with or without placement of a
shunt from the fourth ventricle to the spinal subarachnoid
space.48,49 Endoscopic aqueductoplasty and interventricu-
lostomy with stent placement have also been proposed as
treatment options to reestablish communication between
the supratentorial ventricular system and the isolated fourth
ventricle.50
ENDOSCOPIC THIRD VENTRICULOSTOMY—
AN ALTERNATIVE TO SHUNTING
Endoscopic third ventriculostomy (ETV) can be performed to
treat certain types of obstructive hydrocephalus such as pri-
mary untreated aqueductal stenosis from tectal plate and pi-
neal region tumors. This technique eliminates the need for a
shunt and thereby avoids the complications associated with
shunting. It is contraindicated in patients with communicat-
ing hydrocephalus, and the success rate is poorer in patients
younger than 1 year of age. The highest success rate occurs in
older children and adults with acquired obstructive hydro-
cephalus, with approximately 70% of patients achieving
shunt independence.51,52 In contrast, in children younger
than 3 years of age, the success rate has traditionally been
poorer, ranging from 40% to 50%. However, in a more recent
study, the outcome of ETV was analyzed in patients younger 2
years of age.53 ETV was successful in 71.4% of procedures in
children younger than 2 years of age and in 75% of proce-
dures in infants. The results of this study suggested that suc-
cess of ETV in infants and young children depends primarily
on the thickness of the floor of the third ventricle and the
patient’s age at the time they initially manifested their hydro-
cephalus. The success rate is also lower in patients with pre-
existing pathology such as a tumor, previous shunt, prior
subarachnoid hemorrhage, previous whole-brain irradiation,
and significant scarring of the third ventricle floor, with as
little as 20% of third ventriculostomies remaining patent in
such patients.53
ETV is performed through a frontal burr hole that is made
approximately 2.5 to 3 cm lateral to the midline, just ante-
rior to the coronal suture.52 A peel-away sheath is passed
through the foramen of Monro to protect bordering struc-
tures from damage caused by repeated passage of the endo-
scope. A rigid endoscope is passed by way of the peel-away
 CHAPTER 128 MANAGEMENT OF NEURAL TUBE DEFECTS, HYDROCEPHALUS, REFRACTORY EPILEPSY, AND CNS INFECTIONS
sheath through the foramen of Monro into the third ventri-
cle, and the floor of the third ventricle is fenestrated in the
midline just anterior to the cleavage plane of the mammillary
bodies, which delineate the posterior border of the third
ventricular floor. The fenestration is then dilated with a 2-
Fr Fogarty balloon catheter by means of repeated inflation
and deflation of the balloon. The endoscope is then passed
through the fenestration into the interpeduncular cistern to
evaluate the arachnoid membranes and make certain there
are no redundant arachnoid membranes impeding CSF flow
into the subarachnoid cistern.
Complications related to ETV include failure to complete
the procedure for technical reasons, which has been reported
in up to 26% of patients,54 hemorrhage secondary to vascular
injury,55 cardiac arrest, diabetes insipidus, the syndrome of in-
appropriate antidiuretic hormone secretion, subdural hema-
toma, meningitis, cerebral infarction, transient third and sixth
nerve palsies, and disturbances in short-term memory.54,55
Risks of ETV include infection; injury to adjacent neural struc-
tures (e.g., hypothalamus, pituitary gland, optic chiasm); and
arterial injury with intraoperative rupture or delayed hemor-
rhage related to traumatic aneurysm formation. In a small num-
ber of patients, delayed closure of the ETV has been described
related to scar formation or metastatic tumor seeding.56 In pa-
tients with recurrent symptoms of intracranial hypertension,
cine-MRI can be performed to assess the patency of the third
ventriculostomy by evaluating CSF flow across the floor of
the third ventricle. If the floor of the third ventricle is no longer
patent, endoscopic exploration with an attempt at repeat fenes-
tration is reasonable. If postoperative scarring precludes safe
repeat fenestration of the third ventricular floor, a ventricular
shunt system should be placed.
Late rapid deterioration is an infrequent but deadly compli-
cation of ETV in which deterioration can occur long after the
ETV becomes occluded.57 It is imperative that patients and
caregivers be counseled about this potential complication.
As a means of avoiding this type of complication, the author
places a permanent ventricular access device in patients
undergoing ETV.
OUTCOME AND PROGNOSIS
The natural history of untreated hydrocephalus is poor. In one
study, 50% of children with untreated hydrocephalus died be-
fore 3 years of age58 and only 20% to 23% reached adult life.
Of the survivors, only 38% had normal intelligence.30 The
development of CSF diversion techniques has improved the
prognosis for children with hydrocephalus. Many patients
with shunted hydrocephalus have normal intelligence and
participate in all aspects of social life, with cognitive develop-
ment and life expectancy determined primarily by the nature
of the underlying condition. Overall, about 50% to 55% of
shunt-dependent children will achieve an IQ greater than
80, with verbal cognitive skills being superior to nonverbal
ones.59–61 Not surprisingly, epilepsy appears to be an impor-
tant predictor of poor intellectual outcome in children with
shunted hydrocephalus, with an IQ higher than 90 occurring
in 66% of children without epilepsy versus in 24% of children
with epilepsy.62 Unfortunately, repeated shunt complications
in pediatric patients with hydrocephalus carry significant
morbidity.
1687
Epilepsy Surgery in Children
------------------------------------------------------------------------------------------------------------------------------------------------
Epilepsy is a common childhood disorder.63 It affects approx-
imately 0.5% to 1% of the population in the United States
and Canada,63–65 and up to 50% of epilepsy cases begin
before 5 years of age.63,66 Children with epilepsy are usually
managed with antiepileptic medications initially. However,
20% to 25% of children with epilepsy have inadequate control
of their seizures with pharmacotherapy67,68 and it is now rec-
ognized that seizures starting in childhood that are caused by a
known structural lesion and are not controlled by medication
rarely remit spontaneously as the child matures.69 Repeated
seizures and the side effects of seizure medications have a neg-
ative impact on the developing brain and can lead to severe
impairment of neurocognitive function.69–75 Moreover, recur-
ring seizures can have a negative impact on a child’s education
as a consequence of learning difficulties and the associated
psychosocial effects such as poor peer relationships, behav-
ioral difficulties, poor school performance, depression, anxi-
ety, and poor self-esteem.73,76,77 Because refractory epilepsy
in childhood is a progressive, debilitating condition associated
with eventual deterioration in both intellectual and behav-
ioral function, aggressive treatment is warranted. Seizure
surgery provides an effective alternative therapeutic option
for children who suffer from intractable seizures. The goal
of surgery is to remove the seizure focus as completely as pos-
sible without creating a new neurologic deficit or worsening
an existing one.
Epilepsy surgery is indicated when (1) a child’s seizures
persist despite an adequate trial of at least two appropriate an-
ticonvulsants, alone or in combination, with each drug
pushed to the maximum tolerated dosage; (2) the seizures sig-
nificantly reduce the child’s quality of life; (3) there is a uni-
lateral seizure focus that can be reliably and reproducibly
identified; and (4) resection of the seizure focus does not cause
unacceptable neurologic deficits.69,78–80 A preoperative
workup is initiated as soon as a child fails medical therapy be-
cause early operative intervention is paramount to a good
functional outcome.69,80,81
PREOPERATIVE EVALUATION
The primary objective of the preoperative evaluation is to
identify the area of seizure onset (i.e., the seizure focus) and
its relationship to eloquent cortex.69,74 This begins with non-
invasive characterization of the seizure syndrome and optimi-
zation of medical management. Patients undergo a thorough
neurologic history and physical examination, baseline EEG,
and high-resolution brain MRI with thin coronal cuts through
the temporal lobe (e.g., three-dimensional spoiled gradient
echo [3D-SPGR] scan), fluid-attenuated inversion recovery
(FLAIR) sequences, and MR spectroscopy. EEG can provide
important information by confirming the presence of an epi-
leptogenic focus, by helping to characterize the seizures and
epilepsy syndrome, and by localizing the seizure focus.82
Helpful localizing features on EEG include focal slowing, focal
attenuation of fast activity, and focal epileptiform discharges
(e.g., spikes or sharp waves). A good prognostic indicator
for success with focal resective surgery is the presence of a sin-
gle focus of epileptiform activity.82 The MRI is evaluated for
focal structural abnormalities (e.g., cytoarchitectural) or
1688 PART IX SPECIAL AREAS

abnormal lesions (e.g., neoplastic) that could be causing the
child’s seizures. The presence of a focal abnormality on
MRI, especially when it corresponds to the location of seizure
onset on EEG, increases confidence that the epileptogenic re-
gion has been identified and also increases the likelihood of
achieving seizure control with surgery.80,82,83, Structural ab-
normalities that can be visualized on MRI and that are associ-
ated with medically refractory seizures include low-grade
tumors such as a ganglioglioma (Fig. 128-8), dysembryoplas-
tic neuroepithelial tumor (DNET), astrocytoma, oligodendro-
glioma, and pleomorphic xanthoastrocytoma; developmental
abnormalities such as cortical dysplasia and focal neuronal
migration abnormalities (e.g., gray matter heterotopia); vascu-
lar malformations such as cavernomas and arteriovenous
malformations; secondary post-traumatic or postischemic
encephalomalacia lesions; and mesial temporal sclerosis.82
If the initial evaluation suggests a possible seizure focus,
additional studies are performed to define the seizure focus
further including prolonged video-EEG monitoring with sur-
face (i.e., scalp) electrodes, functional neuroimaging studies
(positron emission tomography [PET] or single-photon com-
puted emission tomography [SPECT], or both), and neuro-
psychological testing, with or without intracarotid injection
of sodium amobarbital (i.e., Wada testing). Prolonged
video-EEG monitoring involves the simultaneous recording
of brain activity and clinical behavior and is the cornerstone
of the presurgical evaluation.82,84 It combines simultaneous
EEG recording via scalp electrodes with a synchronized, con-
tinuous real-time audio/video recording of a child’s typical sei-
zures and thereby enables lateralization and localization of the
child’s seizure focus. When a temporal focus is suspected,
sphenoidal electrodes can be placed to record activity from
the mesial temporal lobes. In addition, antiepileptic medica-
tions can be withdrawn gradually while continuously moni-
toring for seizures until several of the child’s “habitual
events” are recorded. In the majority of cases, the video-
EEG study enables one to determine whether a patient has ep-
ilepsy, whether the patient has partial or generalized seizures,
and where the partial seizures arise.78,80 When considering a
child for epilepsy surgery, the ideal circumstance is to find that
the seizures are highly stereotyped on EEG and that all
seizures arise from the same region.80
Occasionally, video-EEG recording will demonstrate the
presence of independent, bilateral seizure onset, especially
in children with epilepsy secondary to multifocal abnormali-
ties, as seen in tuberous sclerosis.85 In such cases a child may
still be considered a candidate for epilepsy surgery if a major-
ity of the seizures come from one side; however, complete sei-
zure control may not be possible.80 Additionally, in children
with tuberous sclerosis complex who have multiple and often-
times bilateral epileptogenic tubers, a novel surgical approach
has been proposed using multistaged and bilateral invasive in-
tracranial monitoring to identify both primary and secondary
epileptogenic zones.86,87 In contrast to the typical epilepsy
surgery approach where electrodes are placed to map the
location of seizure onset during the first operative stage and
surgical resection follows during the second stage, the multi-
staged approach involves resection of the primary seizure
focus during the second operative stage followed by replace-
ment of the electrodes during that same stage to determine
whether a second epileptogenic region requires resection at
the third and final stage. With such an approach, multiple

A B C

D E F
FIGURE 128-8 Three-month-old girl with 30 to 60 seizures per day on three anticonvulsants. A to F, Brain magnetic resonance image (MRI) showing a
2.5  2.5  2-cm right posterior frontal mass. Preoperative video electroencephalographic studies localized her seizures to the region of the abnormality on
MRI. She underwent complete resection of the mass and adjacent epileptogenic zone, which was localized via intraoperative electrocorticography. The final
pathology was a ganglioglioma. She is now 33 months of age with no evidence of tumor recurrence, has mild residual spastic left hemiparesis, and is
completely seizure free and off all anticonvulsants.
 CHAPTER 128 MANAGEMENT OF NEURAL TUBE DEFECTS, HYDROCEPHALUS, REFRACTORY EPILEPSY, AND CNS INFECTIONS
or bilateral seizure foci have been successfully resected to
achieve good seizure control, suggesting that multiple or bilat-
eral seizure foci are not necessarily a contraindication to sur-
gery in selected patients. However, long-term follow-up is
necessary to determine the durability of the effects of this
novel surgical approach.
Functional neuroimaging with PET88–90 and SPECT91 also
provides valuable localizing information by detecting general-
ized and focal functional brain abnormalities. This informa-
tion can then be compared with the electrophysiologic
dysfunction observed on EEG and the anatomic abnormalities
on MRI. PET is a noninvasive functional imaging study that
measures regional cerebral glucose utilization or metabolism
by determining the extent of uptake of the radionuclide
[18F]deoxyglucose (FDG). Interictally, an epileptic focus is
seen on PET as an area of glucose hypometabolism. This study
is especially helpful in localizing focal areas of cortical dyspla-
sia, mesial temporal sclerosis, and other structural abnor-
malities that correspond to surface EEG localization of
epileptogenic regions. SPECT is another noninvasive func-
tional imaging study that measures regional cerebral blood
flow by using perfusion radiotracers such as technetium Tc
99m hexamethylpropyleneamine oxine (99mTc-HMPAO). It
provides a snapshot of local cerebral blood flow at a specific
point in time.70 Between seizures, a seizure focus is hypome-
tabolic and receives less blood flow. Consequently, relative
hypoperfusion is observed in this region on a SPECT scan.92
In contrast, hyperperfusion is observed in the region of the sei-
zure focus on an ictal SPECT study if the tracer is injected at
the time of seizure onset because local blood flow to the sei-
zure focus increases significantly during a seizure as a conse-
quence of increased metabolic demands. Ictal SPECT is more
reliable than interictal SPECT in localizing the epileptogenic
zone,93–95 and ictal SPECT is more feasible than ictal PET be-
cause, unlike the ictal PET scan, which must be performed
shortly after the tracer is given, the ictal SPECT scan can be
performed at a convenient time after the tracer is injected.105
Functional imaging studies such as PET and SPECT have
reduced the need for invasive intracranial monitoring in
children by up to 90%.88,93,94
Another important component of the presurgical evalua-
tion is the identification of areas of brain function relative
to the location of the seizure focus. This helps predict the risks
associated with removal of the seizure focus. Neuropsycholog-
ical testing plays an important role in this part of the presur-
gical evaluation.82 Such testing can provide lateralizing and
localizing information by identifying areas of neurologic dys-
function that correlate with the patient’s underlying lesion.96
For example, patients with seizures arising from their domi-
nant temporal lobe may have deficits in verbal memory or lan-
guage acquisition, whereas deficits in visuospatial memory
suggest seizure onset in the nondominant temporal lobe. Sig-
nificant deficits in both verbal and visuospatial memory sug-
gest bilateral temporal lobe damage and probable bilateral
seizure foci. In addition, preoperative neuropsychological
testing can help predict the prognosis for cognitive function-
ing after surgery and can be used as a baseline to monitor dis-
ease progression, successful therapy, and the adverse effects of
treatment.70,82 Moreover, postoperative neuropsychological
testing can be performed to identify residual or resultant
cognitive deficits and to assist in planning psychoeducational
interventions to address any impairments.
1689
Intracarotid amobarbital injection (i.e., Wada testing) can
also be used, especially in older children, to determine which
cerebral hemisphere is dominant for language and to ascertain
the relative contributions of each cerebral hemisphere to
memory function.70,78,97,98 During this study, a cerebral an-
giogram is performed and amobarbital is injected into each in-
ternal carotid artery individually. Intracarotid injection of
amobarbital results in arrest of ipsilateral cerebral function,
which mimics the effect of surgical removal of the ipsilateral
cerebral hemisphere. A successful injection is demonstrated
by contralateral hemiparesis and ipsilateral slowing of the
EEG. Speech and memory are then tested to determine
whether the hemisphere contralateral to a planned resection
can adequately support language and memory function. Wada
testing is the most reliable means of ascertaining lateralization
of language dominance and assessing residual memory capac-
ity after temporal lobectomy; however, it is often difficult to
perform in young children because it is necessary for them
to be awake and cooperative.
Results from the preoperative investigation determine
whether a patient is a candidate for epilepsy surgery or
whether additional studies are required to more accurately
pinpoint the location of the seizure focus and areas of eloquent
cortex. The results of the presurgical studies are reviewed in a
multidisciplinary clinical conference attended by epileptolo-
gists, neurosurgeons, neuroradiologists, neuropsychologists,
and other members of the epilepsy surgery team to determine
whether the patient is a candidate for surgery. The pediatric
neurosurgeon discusses the risks and benefits of surgery, de-
termines the surgical approach, and makes the final decision
on whether or not to proceed.82 Results obtained from video-
EEG, MRI, and PET are typically given the strongest consider-
ation,78 and surgery is recommended when there is a clearly
defined focal area of seizure onset that is consistent with EEG,
radiographic, neuropsychological, and clinical evidence.79
The likelihood of a successful outcome from seizure surgery
increases with the number of independent elements that
converge toward a single localization.78,95,99,100
In some patients surgery appears to be a good option, but
the data are inconclusive or discordant. Such patients include
those with normal or nonlocalizing imaging studies and a
video-EEG that regionalizes the seizure onset without suffi-
ciently localizing it, patients with a seizure focus that is more
extensive than the structural lesion observed on imaging stud-
ies, and those with a seizure focus in the vicinity of eloquent
cortex. For these patients, prolonged, invasive intracranial
EEG monitoring may be required to obtain more precise local-
izing information.78,79,101–104 Intracranial EEG monitoring
has the following advantages: (1) it helps define the bound-
aries of the epileptogenic zone around the lesion, which in
turn guides the extent of resection; (2) it helps determine
whether children with multiple structural lesions and multi-
focal interictal spike discharges are surgical candidates by
ascertaining whether their seizures arise from a single opera-
ble epileptogenic zone; and (3) it enables mapping of areas of
eloquent cortex that are contiguous to the seizure focus to de-
termine whether resection of the seizure focus can be per-
formed safely without creating new deficits or worsening
existing ones.
This technique involves placing surface electrodes such as
grid and strip electrodes embedded in a thin pliable sheet of
plastic, directly onto the cortical surface through a craniotomy
1690 PART IX SPECIAL AREAS
and dural opening in the case of subdural grids or through
burr holes in the case of subdural strips.105 Before closing
the dura, an ICP monitor can be placed in patients with sub-
dural grids to assist postoperative monitoring and manage-
ment of elevated ICP related to grid placement. ICP
monitoring is particularly beneficial in the case of a prolonged
postictal state because it can help differentiate postictal leth-
argy from neurologic deterioration as a result of hemorrhage
or edema.79 In addition, the strips and grids are sutured to
the dura to minimize the risk of postoperative movement
and the electrode leads are tunneled to a distant exit site to
minimize the risk of infection. Duraplasty is performed in
cases in which the subdural space must be enlarged to accom-
modate the electrodes, the bone flap is replaced loosely, and
the scalp incision is closed in a two-layered fashion. The skin
around the exiting leads is closed tightly to avoid a CSF leak,
and the leads are secured to the scalp with 4-0 nylon suture to
minimize movement of the electrodes. If necessary, multicon-
tact depth electrodes can be placed stereotactically with a
frame-based system to achieve accurate EEG recordings from
structures located deep to the cortical surface. Depth elec-
trodes survey restricted areas of cortex and are not helpful
in delineating large epileptogenic zones. However, monitoring
with depth electrodes can be advantageous in children when
the seizures are thought to arise from mesial temporal lobe
structures such as the hippocampus and amygdala, but the
side of origin is uncertain.102
After implanting the surface electrodes, a postoperative
skull radiograph is obtained to document electrode posi-
tion, prophylactic intravenous antibiotics and low-dose
corticosteroids are administered, and continuous video-
EEG monitoring is performed. The child is monitored until
several typical seizures are recorded. Once an adequate
number of the child’s typical seizures have been recorded,
cortical stimulation studies are carried out to map func-
tional cortex and induce auras or seizures that can be of
further help in localizing the seizure focus.70,78,79 A surgi-
cal plan is then formulated on the basis of the electrical,
structural, and functional data, and the child is taken back
to the operating room, where the craniotomy is reopened
and the planned resection is performed. A repeat skull ra-
diograph is performed before the second operation and
compared with the initial skull film to document electrode
position and evaluate for electrode movement. Occasion-
ally, an epileptogenic focus cannot be found or resection
of the epileptogenic focus may lead to an unacceptable neu-
rologic deficit. In such cases, the subdural electrodes are
removed and the craniotomy is closed without performing
resective surgery.79
Although intracranial EEG monitoring can help determine
the feasibility and safety of epilepsy surgery, it is an invasive
procedure associated with a number of possible risks includ-
ing hemorrhage, infection, CSF leak, electrode migration, in-
creased ICP, and even death.104,106 The complication rate for
strips is 1% to 3%,107 whereas that for depth electrodes is 3%
to 10%.108 Complications occur more commonly with grids
than with strips. The risk of infection increases with the
duration of invasive monitoring and may be decreased by
administering prophylactic intravenous antibiotics. In many
cases, extraoperative electrode grid placement is unneces-
sary.102,103,109 Instead, intraoperative electrocorticography
can be carried out at the time of surgery to define the precise
limits of the cortical resection on the basis of the location of
interictal spike discharges, differences in background EEG
patterns, and identification of critical functional cortex (e.g.,
primary sensorimotor cortex) by intraoperative somatosensory
evoked potentials.102,110
TYPES OF EPILEPSY SURGERY
Three main types of surgery are used to treat intractable sei-
zures in children: (1) resections, in which the epileptogenic
cortex is removed; (2) disconnections, in which critical path-
ways involved in the propagation of epileptiform discharges
are interrupted; and (3) implantations, in which electrical
stimulation is used.74,82,111 The type of surgery performed
is determined by the results of the presurgical workup.
Resection Surgery
Specific types of resection surgery include anterior temporal
lobectomy with amygdalohippocampectomy for intractable
complex partial seizures; extratemporal resections for nonle-
sional epilepsy79; multilobar, lobar, and focal cortical resec-
tions for intractable seizures caused by structural lesions
such as cortical malformations, intracerebral hamartomas, tu-
mors (see Fig. 128-8), and vascular malformations74; and
hemispherectomy, the extreme of resection surgery, which in-
volves either complete removal (anatomic) or isolation (func-
tional) of the abnormal cerebral hemisphere and is reported to
be one of the most successful of all operations for the relief of
epilepsy.111–113 Hemispherectomy is reserved for patients
who have severe damage to one hemisphere and a relatively
intact contralateral hemisphere. It is particularly effective in
children with severe unilateral motor seizures who already
have contralateral hemiparesis and hemianopsia. Candidates
for hemispherectomy include patients with Rasmussen
syndrome,114 Sturge-Weber syndrome,115 hemimegalence-
phaly,116 diffuse hemispheric cortical dysplasia, infantile
hemiplegia, and cerebral infarction. When the presurgical
workup reveals that a child’s seizures are arising from the en-
tire hemisphere and that hemisphere is removed with sparing
of the ipsilateral basal ganglia (anatomic hemispherectomy) or
disconnected with removal of less cortical tissue (functional
hemispherectomy or hemispherotomy), 75% to 80% of pa-
tients have a marked reduction (80% or more) in seizure fre-
quency with improvement in cognitive function.74,114,117,118
Temporal lobe epilepsy (TLE) is the most common
localization-related epilepsy in adult and adolescent surgical
candidates but is less common than extratemporal epilepsy
in infants and children.119,120 In TLE, patients generally expe-
rience complex partial seizures with oral and gestural autom-
atisms, which may be preceded by an aura.78 Although the
most common cause of TLE in adults and adolescents is mesial
temporal sclerosis, such is not the case for infants and chil-
dren.121,122 Instead, TLE in this age group is more commonly
caused by congenital abnormalities in cerebral development
such as cortical dysplasia, low-grade neoplasms, and vascular
malformations.81,123 In patients with lesion-related TLE,
which occurs as a result of a temporal lobe mass, the source
of the seizures is not generally the lesion itself but the margins
around the lesion124 and a successful outcome in such pa-
tients depends on adequately defining the extent of the epilep-
togenic zone around the lesion and resecting this region
along with the lesion.
 CHAPTER 128 MANAGEMENT OF NEURAL TUBE DEFECTS, HYDROCEPHALUS, REFRACTORY EPILEPSY, AND CNS INFECTIONS
When the results of diagnostic studies indicate that the sei-
zures are arising from a single temporal lobe, surgery is recom-
mended. Both the lateral neocortex and the medial basal
portion of the temporal lobe can cause temporal lobe seizures
in children.69,70 The medial basal portion of the temporal lobe
consists of the fusiform and parahippocampal gyri and the
amygdalohippocampal complex. Several surgical approaches
have been described for resecting epileptic foci arising from
the medial temporal lobe in children with medically refractory
mesial TLE of unilateral origin.74 For example, an anterior
temporal lobectomy with amygdalohippocampectomy can
be performed.70,111,125 This two-step approach involves the
initial removal of approximately 3.5 cm of anterolateral
temporal lobe (measured with a ruler from the temporal tip
along the middle temporal gyrus) below the superior temporal
gyrus with a subpial dissection technique. This results in
exposure of the temporal horn and mesial temporal lobe struc-
tures including the hippocampus, amygdala, and parahippo-
campal gyrus. The mesial structures are then removed by
subpial dissection under the operating microscope, with the
posterior limit of the dissection extending to the level of the
tectal plate. In children with medically refractory mesial
TLE of unilateral origin, removal of the affected mesial tempo-
ral lobe, with complete removal of the medial temporal
structures, has a greater than 75% chance of significantly re-
ducing a child’s seizures or rendering a child seizure free off
medication.69,70,74,81,126
Extratemporal epilepsy is found more frequently than TLE
in pediatric epilepsy surgery candidates.119 The epileptiform
abnormalities can be frontal, parietal, or occipital and are of-
ten poorly localized.127–129 Moreover, it is not uncommon for
more than one lobe to be involved in children with intractable
epilepsy.79 When the seizure focus is outside the temporal
lobe and is not associated with a lesion on imaging studies
(nonlesional, extratemporal epilepsy), patients often require
invasive intracranial monitoring and stimulation studies to de-
fine and accurately localize the seizure focus, as well as to map
areas of eloquent brain function relative to the epileptogenic
cortex.78,79 Such seizures may be clinically silent and suggest
a parietal or lateral frontal focus, complex partial and suggest a
frontal lobe focus, or associated with amaurosis and suggest an
occipital lobe focus. Resection margins are defined on the ba-
sis of the epileptogenic gyri identified during preoperative in-
tracranial monitoring and stimulation studies with implanted
subdural electrodes, as well as by intraoperative electrocorti-
cography and functional cortical mapping. Corticectomy is
then carried out under the operating microscope, and the en-
tire epileptogenic focus is resected en bloc using a subpial dis-
section technique. Roughly two thirds of children with
extratemporal (both lesional and nonlesional) epilepsy derive
significant improvement from surgical intervention.130,131
The risks associated with resection procedures are gener-
ally related to the area or areas of brain resected and include
visual field cuts (e.g., contralateral superior quadrantanopsia
or “pie-in-the-sky” defect caused by an injury to Meyer
loop/optic radiation during posterior temporal lobe resection),
manipulation hemiplegia, speech and memory deficits, stroke
from injury to nearby blood vessels (e.g., sylvian branches of
the middle cerebral artery during temporal lobectomy), and
cranial nerve palsies (e.g., injury to the third nerve during re-
section of the hippocampus). The incidence of major compli-
cations such as stroke or paresis ranges from 2% to 5%.131,132
1691
Disconnection Surgery
Resection of an epileptogenic focus is the mainstay of epilepsy
surgery. However, in some children, the epileptogenic area is
in a functionally critical cortical region (e.g., sensorimotor,
language, or visual cortex, or more than one of these regions)
and cannot be resected safely. Such children may be candi-
dates for disconnection surgery—surgery in which the seizure
focus is isolated from the rest of the brain.111 For example, if
the presurgical workup reveals that a child has a seizure focus
in an area of eloquent brain such as the motor cortex, resection
of the focus might cause unacceptable deficits such as contra-
lateral paralysis or weakness. To prevent this, children with
seizures arising in eloquent cortex can undergo a type of
“disconnection” procedure called multiple subpial transec-
tion.79,111,133 The efficacy of this type of surgery is based on
the observation that normal cortical function is transmitted ver-
tically, whereas seizures are transmitted through horizontal
pathways. A small ball-tipped probe is used to transect the hor-
izontal fibers in regions of eloquent cortex, thereby interrupting
the horizontal spread of the ictal discharge without damaging
the vertical columns of functional cortex. The cuts are made
within the superficial convexity of the cortex under direct vi-
sion through the operating microscope, with each cut occur-
ring at a right angle to the long axis of the gyrus at 5-mm
intervals in a parallel fashion until the entire epileptogenic zone
has been covered. With this technique, selected patients can
achieve a substantial reduction in seizure frequency, without
sustaining any persistent neurologic deficits. In extratemporal
epilepsy, seizure control with multiple subpial transection is
comparable with that achieved with resective surgery.134
Corpus callosotomy is another type of disconnection sur-
gery that can be performed in the absence of an identifiable
seizure focus or in the presence of multiple foci. During
this procedure, the anterior two thirds or the entire corpus
callosum is sectioned through a bifrontal craniotomy ap-
proach.23,74,135 It is typically reserved for patients who have
symptomatic generalized epilepsy with multiple seizure types
and foci of epileptogenicity, as well as some degree of devel-
opmental delay or mental retardation.78,82 Sectioning the cor-
pus callosum results in separation of the cerebral hemispheres
and prevents rapid bilateral generalization of epileptic dis-
charges. Candidates for operative intervention include pa-
tients with atonic seizures (“drop attacks”) who are prone to
violent falls and closed head injuries and patients with
Lennox-Gastaut syndrome. In appropriate patients, this type
of procedure can be effective. For example, in patients with
atonic seizures, up to 80% may experience complete or nearly
complete cessation of attacks.82 However, rarely do seizures
totally remit, and patients with absence or myoclonic seizures
either derive no benefit or experience an inconsistent response
with corpus callosotomy. In addition, corpus callosotomy is
associated with a number of possible adverse effects, some
of which can be quite devastating, including language dis-
turbances (e.g., complete mutism vs. slowness in initiating
speech), weakness, increased frequency or intensity of partial
seizures, and a disconnection syndrome. This syndrome,
which can occur in patients with a dominant left hemisphere,
consists of left tactile anomia, left-sided dyspraxia, pseudohe-
mianopsia, right-sided anomia for smell, impaired spatial syn-
thesis of the right hand resulting in difficulty copying complex
figures, decreased spontaneity of speech, and incontinence.74
1692 PART IX SPECIAL AREAS
Outcome Successful outcome from epilepsy surgery de-
pends on (1) proper and timely patient selection guided
by a thorough presurgical evaluation; (2) good correlation
among clinical, electrophysiologic, and neuroradiologic
data; (3) complete or nearly complete resection of the sei-
zure focus; and (4) lack of surgery-related injury. Patients
undergo routine EEG, MRI, and neuropsychological testing
postoperatively and are graded in terms of percentage of
seizure reduction and length of time since surgery. Results
from preoperative neuropsychological testing can also be
compared with those from postoperative testing to ascer-
tain the effects of surgery on the child’s behavior and cog-
nition. In this manner, postoperative neuropsychological
testing can help identify residual or resultant cognitive
deficits and assist in planning psychoeducational interven-
tions to address these deficits. Patients continue taking an-
ticonvulsants for 1 to 2 years postoperatively. Those who
are seizure free at that time, with a relatively benign-
appearing EEG, are gradually weaned from their seizure
medications.
Some studies have shown that outcome after epilepsy sur-
gery (especially in the case of multilobar, lobar, and focal
cortical resections for intractable seizures caused by struc-
tural lesions) is related to the pathology and location of
the lesion.74,82 Structural lesions associated with intractable
epilepsy in children include developmental lesions such as
focal cortical dysplasia and hamartomas; vascular lesions
such as cavernomas and arteriovenous malformations; low-
grade tumors such as DNET, ganglioglioma, ganglioneur-
oma, astrocytoma, pleomorphic xanthoastrocytoma, and
oligodendroglioma; ischemic or hypoxic lesions; traumatic
lesions; and mesial temporal sclerosis.74,82 With focal corti-
cal dysplasia, an important cause of intractable epilepsy in
children, the outcome after surgical intervention is good
within the first 2 years after surgery; however, the percentage
of patients who remain seizure free declines over time. For
example, in one series,136 although 65% of patients with
cortical dysplasia were seizure free at 2 years, only 40%
remained seizure free at 5 years and just 33% were seizure
free at 10 years. In another series with a mean follow-up
period of 3.6 years, children with cortical dysplasia had
seizure-free rates of 52%.137 With mesial temporal sclerosis,
a relatively rare pathologic entity in children, seizure-free
rates are excellent for the vast majority of children and ad-
olescents, with surgical results comparable with those ob-
served in adults.136,137 A DNET is a benign cortical tumor
that has cystic or microcystic components and is frequently
associated with cortical dysplasia.82 In contrast to cortical
dysplasia, the epileptogenic area associated with a DNET re-
sides within the cortex immediately adjacent to the lesion
rather than within the DNET itself.138 Total resection of
these lesions, aided by preresection electrocorticography to
identify adjacent epileptogenic cortex, can result in complete
seizure freedom in about 75% of patients.139 In patients
with ganglioglioma who have seizures (see Fig. 128-8), com-
plete surgical resection is the preferred management and the
most important factor predicting a seizure-free outcome,140
with about 78% of patients becoming seizure free off
anticonvulsants.141
The goal of surgical management is complete elimina-
tion of seizures or excellent seizure control with seizure
medications and an improvement in the child’s long-term
cognitive, behavioral, and social development. Surgical
treatment of epilepsy in children can reduce seizure fre-
quency and limit the cognitive and psychologic impairment
that often occurs as a consequence of chronic uncontrolled
seizures. However, surgical intervention must be timed ap-
propriately to enable children to develop their maximal po-
tential and reach their educational, vocational, and social
goals.82
As with any surgical procedure, epilepsy ablative proce-
dures are not risk free and the risk of creating a new per-
manent neurologic deficit or worsening an existing deficit
must be balanced against the potential benefit of
completely eliminating a patient’s seizures or significantly
reducing the frequency of seizures. Risks of epilepsy sur-
gery include removal of or injury to eloquent cortex; injury
to projection fibers, association fibers, or commissural fi-
bers underlying the cortical resection; injury to the Meyer
loop causing a contralateral “pie-in-the-sky” visual field de-
fect; injury to vessels in the area of the resection causing
ischemic damage or stroke in the corresponding areas of
vascular distribution; and injury to nearby cranial nerves.
Despite the risks of the procedure, early surgical interven-
tion in appropriate cases can arrest progressive deleterious
changes in the developing brain including progressive psy-
chologic and intellectual impairment, allow for optimum
brain development, reduce the undesirable side effects of
medical therapy, and allow many children with seizures
to lead normal, productive lives. Currently, surgery is the
only therapy that offers the chance of a cure and early iden-
tification of appropriate surgical candidates by means of a
comprehensive preoperative evaluation is paramount to
achieving a successful outcome.82
Vagus Nerve Stimulation
Vagus nerve stimulation by means of the implantable Neuro-
Cybernetic Prosthesis (NCP; Cyberonics, Houston) is a rela-
tively recent development in the surgical treatment of
childhood epilepsy and is gaining increasing popularity as
an effective treatment option for children with drug-resistant
epilepsy who have failed medical therapy and are not candi-
dates for resection or disconnection procedures. Such chil-
dren often have many generalized or focal seizures (or both)
every day that arise from multiple areas on both sides of the
brain, and they are not candidates for or have already failed
previous cortical resection. The FDA limits its use to patients
older than 12 years of age; however, it is also commonly used
in children younger than 12 years of age because earlier con-
trol of seizures generally results in improved long-term
outcomes.
Placement of a vagal nerve stimulator involves wrapping
helical, platinum, bifurcated ribbon electrodes around the left
vagus nerve in the cervical region and connecting the bifur-
cated helical lead to a pulse generator, which in turn is
inserted into a subcutaneous or submuscular pocket in the left
pectoral region.111,142 Together, they deliver intermittent elec-
trical stimulation to the left cervical vagus nerve—usually for
30 seconds every 5 minutes. Electrical impulses, which travel
to the cerebral cortex by way of ascending sensory nuclei (e.g.,
nucleus tractus solitarius), exert widespread effects on neuro-
nal excitability throughout the CNS.143 Although it has been
postulated that these rostral impulses stop seizure activity by
disrupting the abnormal electrical activity caused by abnormal
 CHAPTER 128 MANAGEMENT OF NEURAL TUBE DEFECTS, HYDROCEPHALUS, REFRACTORY EPILEPSY, AND CNS INFECTIONS
groups of neurons firing in an uncontrolled synchronous fash-
ion, the exact mechanism of seizure modulation remains
unknown.64,144,145
The pulse generator has eight programmable parameters
that are adjusted noninvasively through the skin by a teleme-
try programming wand controlled by a hand-held computer.
The stimulation parameters are adjusted according to patient
tolerance and seizure frequency, and seizure medications are
weaned as tolerated. An additional benefit of vagus nerve stim-
ulation is the ability to control seizure activity with the use of a
hand-held magnet. For example, at the beginning of an aura or
seizure, the patient or a family member may pass a hand-held
magnet across the chest pocket where the generator resides.
This triggers a train of stimulation superimposed on the base-
line output that can attenuate or even abort an impending sei-
zure.143 Other advantages of vagus nerve stimulation include
guaranteed treatment compliance, sustained efficacy over
time,146 and global improvement in quality of life and cogni-
tive function.143 Adverse effects of vagus nerve stimulation
can occur and include hoarseness, throat pain, cough, dys-
pnea, and paresthesias. Such effects tend to occur intermit-
tently, concomitantly with stimulus delivery, and are
typically transient.146,147 Surgical complications of vagus
nerve stimulation are rare and include left vocal cord injury,
lower facial paresis, bradycardia, and infection requiring ex-
plantation in 1.1%.143 Although the vagus nerve is the prin-
cipal efferent component of the parasympathetic nervous
system, vagus nerve stimulation by the NCP system has not
been shown to adversely influence pulmonary function,
gastrointestinal motility, or secretion.147 Nevertheless, there
is a case report in the literature of bradyarrhythmia, perfectly
correlated with VNS stimulation periods, which suddenly
occurred 2 years and 4 months after VNS implantation in
a pediatric patient who presented with syncope-like
episodes.148
Previous studies have shown that children and adoles-
cents with medically and surgically refractory epilepsy derive
substantial benefit from vagus nerve stimulation, with many
patients achieving a significant reduction in seizure fre-
quency and some being able to reduce the number of anti-
convulsants.149,150 Patients with idiopathic epilepsy, as well
as those with seizures arising from a structural etiology, are
considered appropriate candidates.143 In particular, vagal
nerve stimulation has been shown to be highly effective in
patients with Lennox-Gastaut syndrome, with five of six
children in one series achieving a 90% reduction in seizure
frequency.151 Moreover, use of the magnet in pediatric pa-
tients has been shown to reduce seizure duration in some
patients and improve postictal lethargy in others.150 In a pre-
vious retrospective analysis of seizure frequency and quality
of life in pediatric patients with medically refractory epi-
lepsy,150 29% of children achieved a greater than 90% reduc-
tion in seizure frequency, 39% had a 50% to 90% reduction,
and 13% had less than a 50% reduction with vagus nerve
stimulation. When compared with the efficacy in adult pa-
tients, multiple studies have shown an equivalent or slightly
higher response rate in children with about 45% to 55% of
patients deriving a greater than 50% reduction in seizure fre-
quency. Moreover, there is a suggestion that these patients
also showed significant improvement in quality of life mea-
sures.152–156 Although complete seizure control is possible
with vagus nerve stimulation, the likelihood of complete
control is low.143
1693
Suppurative Central Nervous
System Infections
------------------------------------------------------------------------------------------------------------------------------------------------
Suppurative infections of the CNS include intracranial infec-
tions such as brain abscess, subdural empyema, epidural ab-
scess, osteomyelitis of the skull, Pott puffy tumor, and acute
pyogenic meningitis,157 as well as intraspinal infections such
as SC abscess, spinal epidural abscess, diskitis, and vertebral
osteomyelitis. Intracranial infections can be life threatening
and must be considered in the differential diagnosis in
children with fever, headache, and cranial wounds or sinus
disease. Similarly, intraspinal infections can result in devastat-
ing neurologic impairment, especially if the diagnosis is
delayed, and must be considered in the differential diagnosis
in children with back pain, spinal tenderness, and fever. The
remainder of this chapter discusses suppurative CNS
infections.
INTRACRANIAL INFECTIONS
Suppurative processes in the epidural and subdural spaces
and brain parenchyma are classic neurosurgical complications
of otorhinologic, middle ear, and post-traumatic infections, as
well as immunosuppression, and they commonly manifest as
neurosurgical emergencies.158 An epidural abscess refers to
suppuration in the epidural space between the dura and cal-
varia or skull base, a subdural empyema refers to suppuration
in the subdural space between the dura and arachnoid, and a
brain abscess refers to a circumscribed focus of suppurative
tissue necrosis within the brain parenchyma.
Incidence
The reported incidence of intracranial infections in children
remains low. In busy pediatric neurosurgical centers, the inci-
dence of an isolated epidural abscess is about one case per year
and that of a subdural empyema is one to two cases per
year.159,160 The incidence of brain abscess is less than one
to three cases per year.158 Although intracranial infections
are relatively uncommon in children, they can be rapidly fatal
if not recognized promptly and managed appropriately. Suc-
cessful treatment requires early diagnosis, guided by CT or
MRI (or both), aggressive antibiotic therapy, and timely and
adequate surgical intervention targeted to maximal areas of
suppuration. Response to treatment must also be monitored
neuroradiographically to ensure that the infection is resolving.
Etiology
Intracranial infections occur as a consequence of (1) direct ex-
tension from contiguous infection, such as paranasal sinusitis,
otitis media, or mastoiditis, or propagation through venous
channels; (2) direct inoculation of the brain after penetrating
cranial trauma or a neurosurgical procedure; and (3) hematog-
enous spread from a distant focus of infection. Contiguous in-
fection in the paranasal sinuses, mastoid air cells, orbit, or
skull is the most commonly reported cause of suppuration
in the epidural space.161 Epidural abscess can also occur as
a complication of a congenital dermal sinus, trauma (espe-
cially penetrating wounds), craniotomy, application of skull
pins or tongs, and paranasal sinus and skull base operative
procedures.162 Subdural empyema commonly occurs as a re-
sult of infection in the frontal and ethmoidal sinuses, middle
ear, and mastoid air cells.163,164 Other causes of subdural
1694 PART IX SPECIAL AREAS

empyema include spread from contiguous infections of the
scalp, subgaleal space, calvaria, or epidural space (or any com-
bination of these regions); meningitis; hematogenous infec-
tion of a preexisting subdural effusion or hematoma165;
rupture of a parenchymal brain abscess into the subdural
space; and subdural puncture or subdural shunt placement.
In the pediatric population, the most common cause of
brain abscess is direct or indirect spread from infection in
the paranasal sinuses, middle ear, and teeth.166 In adolescent
boys, frontal lobe abscess is a relatively common complication
of acute frontal sinusitis.158 In children with congenital heart
defects or pulmonary right-to-left shunt disease, hematoge-
nous spread from remote sources of infection such as pulmo-
nary infections, cutaneous infections, osteomyelitis, and
bacterial endocarditis remains an important pathogenic mech-
anism of brain abscess formation.167 In neonates, bacterial
meningitis caused by Citrobacter, Proteus, Serratia, or Entero-
bacter is complicated by the formation of brain abscesses in
a high percentage of patients.168 Moreover, direct inoculation
of the brain after penetrating cranial trauma, penetrating
wounds of the orbit, compound skull fractures, scalp wounds,
facial sepsis, CSF fistula, and post-traumatic meningitis can
lead to the development of brain abscesses in children. The
location of the abscess is dependent on the portal of entry.
For example, otogenic abscesses or those resulting from
mastoiditis often develop in the temporal lobe or cerebellar
hemisphere adjacent to the infected ear or mastoid air cells,
paranasal sinusitis commonly gives rise to abscess formation
in the frontal lobe, and brain abscesses arising from hematog-
enous spread tend to occur in the distribution of the middle
cerebral artery.
Aerobic organisms commonly isolated from suppurative
intracranial infections include Staphylococcus, Streptococcus,
Enterobacteriaceae, and Haemophilus. Anaerobic organisms in-
clude Bacteroides, Peptostreptococcus, Fusobacterium, Veillonella,
Propionibacterium, and Actinomyces. In many cases, culture of
pus obtained from intracranial infections reveals multiple or-
ganisms including both anaerobic and aerobic organisms.158
However, in some cases, no organisms are isolated despite
proper handling of specimens. With regard to the causative
organisms, the bacteriologic diagnosis often points to the
source of the intracranial infection. For example, aerobic or
anaerobic streptococci (or both) are often cultured from intra-
cranial infections complicating paranasal sinus infection,
gram-negative organisms such as Bacteroides and Haemophilus
are cultured from intracranial infections complicating oto-
genic infections, and S. aureus is cultured from intracranial in-
fections complicating compound skull fractures and
neurosurgical procedures. Fungal organisms such as Candida,
Aspergillus, Nocardia, and Cryptococcus commonly cause intra-
cranial infections in children who are immunologically sup-
pressed after organ transplantation or chemotherapy or who
have impaired host defenses.169
Pathogenesis
Spread of infection from the paranasal sinuses, mastoid air
cells, or middle ear to the epidural and subdural spaces or
to the brain parenchyma occurs as a consequence of infective
thrombophlebitis of the diploic vessels or nutrient vessels that
supply the outer layer of the dura.158 Once infection enters the
diploic space, it spreads rapidly through this space to involve
widespread areas of the skull. Occasionally, a subperiosteal
abscess (Pott puffy tumor) forms as a result of the osteomye-
litis and leads to swelling of the overlying scalp (Fig. 128-9).
Once the inflammatory process reaches the subdural space, it
can spread rapidly throughout the subdural space because this
space is in continuity over the surface of the brain, on each
side of the falx and tentorium, and through the foramen mag-
num with the spinal subdural space. Direct exposure of the
brain to subdural infection can lead to occlusion of cortical
veins and septic thrombophlebitis. This, in turn, can cause

A B
FIGURE 128-9 A and B, Axial magnetic resonance images from a 5-year-old boy with a history of strep throat, treated with 2 weeks of oral antibiotics,
followed by fever, malaise, headaches, progressive swelling in the midline of his forehead, vomiting, and lethargy. A frontal subperiosteal abscess (i.e., Pott
puffy tumor; A, arrow) and an epidural abscess (B, arrow) are evident. He was managed by surgical evacuation and aggressive antibiotic therapy; the bone
flap was left in place. The infection resolved completely with no neurologic sequelae.
 CHAPTER 128 MANAGEMENT OF NEURAL TUBE DEFECTS, HYDROCEPHALUS, REFRACTORY EPILEPSY, AND CNS INFECTIONS 1695

cerebral infarction and edema, and the edema can produce
significant mass effect and associated brain compression
and herniation.
A brain abscess develops as a consequence of seeding of
bacteria in areas of preexisting necrosis or direct traumatic
injury, hemorrhage, or infarction from septic thrombophlebi-
tis. Once a nidus of bacteria is established within the brain
parenchyma, the organisms produce an acute inflammatory
response that results in polymorphonuclear infiltration and
edema. The central area eventually undergoes necrosis and
liquefaction and becomes surrounded by a peripheral zone
of inflammatory cells, neovascularization, and fibroblasts.158
The fibroblasts, in turn, create a dense collagenous capsule
that is surrounded by edema and reactive gliosis. An abscess
with its surrounding edema can lead to significantly elevated
ICP and cause brain herniation and death. Moreover, sudden
deterioration and death can occur from rupture of a brain ab-
scess into the ventricular system167 or through the cerebral
cortex into the subarachnoid space.170
Clinical Features
Children with intracranial infection often have a history of ce-
rebral trauma, purulent nasal or aural discharge, fever, or
headache. If there is suppuration within the epidural space,
the child typically appears acutely ill with fever, headache,
or earache at the time of the initial evaluation. If the epidural
abscess is associated with a subperiosteal abscess, patients also
commonly have a tender, fluctuant swelling in the scalp over-
lying the involved frontal sinus, mastoid air cells, or affected
area of the diploic space (see Fig. 128-9). As infection spreads
within the epidural space, the mass effect on the subjacent
dura and brain increases and causes escalating headaches.
Eventually, focal neurologic signs and a decreased level of con-
sciousness develop as the abscess increases in size and brain
herniation occurs.160
With spread of infection into the subdural space, rapid
neurologic deterioration occurs as a result of cerebral edema,
infarction, and herniation. Patients typically have signs and
symptoms of a systemic febrile illness and raised ICP includ-
ing malaise, fever, chills, meningismus, vomiting, headache,
focal neurologic deficits, and altered mental status.161 In ad-
dition, children with subdural empyema can have localized
tenderness and swelling overlying the affected sinus or mas-
toid secondary to subperiosteal abscess formation (Pott puffy
tumor; see Fig. 128-9), and focal seizures may be observed
in 25% to 50% of patients.158 Depending on the underlying
cause, suppuration within the subdural space may be supraten-
torial or infratentorial, or both (Fig. 128-10). With subdural
empyema in the posterior fossa, children typically have a sys-
temic febrile illness, headaches, and a stiff neck, but focal neu-
rologic signs and seizures are generally lacking.171
The clinical manifestations of a brain abscess are influenced
by the size and location of the abscess, as well as the number of
abscesses present, the virulence of the organism or organisms,
and the patient’s age and host defenses.158 Children and ado-
lescents with a brain abscess exhibit signs and symptoms of
(1) systemic infection including malaise, fever, chills, and
neck stiffness; (2) raised ICP as a result of the mass effect from
the abscess and surrounding cerebral edema including head-
ache, vomiting, and a declining level of consciousness; and
(3) focal neurologic deficits reflecting the location of the ab-
scess such as hemiparesis with an abscess in the posterior
frontal region or as a consequence of uncal herniation, speech
difficulty with an abscess in the dominant temporal lobe, a
visual field deficit with a posterior temporal or occipital lobe
abscess, and nystagmus, defective conjugate eye movements,
ataxia, and hypotonia with a cerebellar abscess.158 Infants
with a brain abscess have signs and symptoms of increased
ICP including irritability, lethargy, vomiting, a bulging anterior
fontanelle, frontal bossing, and increasing head circumfer-
ence. At least 25% to 50% of pediatric patients with a supra-
tentorial brain abscess also manifest focal or generalized
seizures during the course of their disease.172
Diagnostic Studies
If intracranial infection is suspected, the patient should un-
dergo a contrast-enhanced head CT or brain MRI study (or
both). A head CT scan with contrast enhancement enables
rapid diagnosis, whereas MRI provides accurate preoperative
localization and is beneficial for monitoring resolution of

A B C
FIGURE 128-10 Axial (A and B) and sagittal (C) brain magnetic resonance imaging (MRI) in a 35-week preterm male infant with a history of meningitis,
organism unknown, who was on multiple antibiotics, but not metronidazole. He began having seizures and respiratory difficulty and required intubation.
MRI showed extensive subdural empyema extending bilaterally, both above and below the tentorium (arrows). Cultures at the time of drainage grew
Bacteroides fragilis. The infection eventually cleared with a prolonged course of antibiotics including metronidazole. He is now almost 4 years old, is
no longer taking anticonvulsants, and has no neurologic sequelae.
1696 PART IX SPECIAL AREAS
infection. Depending on the type, extent, and stage of the intra-
cranial infection, these studies may reveal dural enhancement,
an empyema wall, an abscess capsule, and hypodense areas of
cerebral edema, venous infarction, and early cerebritis. Sagittal
and coronal views are also helpful for visualizing affected
sinuses and suppurative collections at the vertex and skull
base. In patients with a brain abscess, contrast-enhanced CT
and MRI demonstrate a characteristic ring-enhancing lesion
with surrounding edema. In particular, on MRI a brain abscess
appears as an area of central hypointensity surrounded by a ring
of enhancement after gadolinium administration on T1 weight-
ing and as a hyperintense area of suppuration surrounded
by a hypointense capsule and an outer layer of edema on T2
weighting.173
Laboratory studies helpful in the diagnosis of intracranial
infection include a peripheral white blood cell count, erythro-
cyte sedimentation rate, and C-reactive protein, all of which
may be mildly elevated.172 Blood cultures may reveal the caus-
ative organism, especially in cases of epidural abscess and sub-
dural empyema.172 Because of the risk of herniation and the
low diagnostic yield of CSF studies, a lumbar puncture should
not be performed before neuroimaging, especially if a mass-
occupying lesion is suspected. Identification of the causative
organism(s) is accomplished by culturing purulent material
obtained at the time of surgical drainage.158 Such cultures de-
termine the antibiotic sensitivity of the causative organism(s),
which in turn guides antimicrobial therapy. Proper handling
of specimens and appropriate aerobic and anaerobic culture
techniques are required to achieve positive culture results.
Treatment
A high index of suspicion is essential to ensure prompt diag-
nosis. Successful treatment requires surgical drainage, appro-
priate antimicrobial therapy, and close coordination of care
between pediatric neurosurgeons and infectious disease spe-
cialists. Systemic antibiotics should be initiated as soon as
the diagnosis is considered.158 Although preoperative antibi-
otics may achieve some bone penetration, it is unlikely that
they will interfere with culture results.
The initial choice of antibiotics is contingent on the micro-
bial spectrum, which varies according to the primary site of in-
fection.158 Hence knowledge of the primary infection site can
be used to guide antimicrobial therapy. For example, if frontal
sinusitis is the primary site, the most likely infective agent is a
streptococcal species and antibiotic coverage should include
metronidazole and penicillin or a third-generation cephalo-
sporin. If the primary site is unknown, the initial choice of an-
tibiotics should include a penicillinase-resistant synthetic
penicillin, a third-generation cephalosporin, and metronida-
zole to provide coverage of aerobic and anaerobic streptococci,
other anaerobes, and staphylococci. Although the optimal du-
ration of therapy remains unknown, intravenous antibiotics
should be continued until systemic evidence of the infection
has resolved and the intracranial infection appears to be
diminishing in size.160,163 An additional 2 to 3 months of oral
antibiotics is recommended by some to prevent recurrence.174
Besides systemic antibiotics, children with intracranial in-
fections may also require treatment with corticosteroids and
anticonvulsants. Corticosteroids such as dexamethasone are
effective in reducing the cerebral edema associated with intra-
cranial infection and have not been shown to exacerbate the
inflammatory response. Consequently, corticosteroids can be
administered during the acute phase of the illness to reduce in-
tracranial hypertension and prevent herniation. Because 25% to
50% of pediatric patients with intracranial infections exhibit
seizures during the course of their illness, prophylactic anticon-
vulsants should also be started as soon as the diagnosis is made.
Besides enabling bacteriologic diagnosis and guiding anti-
microbial therapy, surgical treatment permits evacuation of
sizable collections of liquefied pus and reduces ICP.158 If
the primary site of infection is known and the patient’s clinical
condition permits, surgical intervention should also address
the primary infection source. Surgical drainage of intracranial
pus can be performed through a limited craniotomy, craniect-
omy, or burr hole. The bone removed should be just large
enough to provide access to the abscess or empyema. If an epi-
dural abscess occurs in the face of a previously devascularized
bone flap, the devitalized bone is frequently left out and a
delayed cranioplasty performed. However, if epidural infec-
tion complicates an extensive craniofacial reconstruction in
which bone removal would probably generate irreparable de-
fects, the bone is replaced after extensive irrigation of the epi-
dural space with antibiotic solution and debridement, and
systemic antibiotics are administered until the infection
clears.158 Follow-up neuroimaging studies are performed on
all patients to monitor the progress of treatment.
For a subdural empyema, the subdural space is exposed by
craniotomy, craniectomy, or a burr hole, often using stereotac-
tic CTor MRI guidance to accurately localize the purulent fluid
collection and keep the bone removal and dural opening to a
minimum. Once the dura is opened, specimens are obtained
for aerobic and anaerobic culture, the empyema is evacuated,
and the subdural space is irrigated thoroughly with antibiotic
solution. If a craniotomy has been performed and neither os-
teomyelitis nor epidural abscess is observed, the bone flap
may be replaced. However, healing of the bone flap should
be monitored closely because failure of healing can occur
and result in reabsorption or sequestration of the bone flap.175
Repeated drainage may also be necessary if the subdural em-
pyema recurs. Infants in whom subdural fluid collections de-
velop as a complication of Haemophilus influenzae meningitis
typically do not require surgical intervention unless the fluid
collection causes a mass effect and the patient becomes symp-
tomatic from elevated ICP. If surgery is warranted, needle
drainage through the lateral aspect of the anterior fontanelle
or through a burr hole can be performed to remove the
subdural fluid, which is typically sterile.158
Conservative management of brain abscesses with intrave-
nous antibiotics, with or without neurosurgical intervention,
has been reported. However, antibiotic treatment alone must
be accompanied by frequent neuroradiographic follow-up to
determine the effectiveness of the treatment. Careful monitor-
ing of the patient’s neurologic findings and level of conscious-
ness is also essential. Immediate neurosurgical intervention is
required if (1) the patient demonstrates any evidence of neu-
rologic decline, (2) the abscess shows significant mass effect
on CT or MRI, or (3) the abscess fails to respond to antibiotics
within 2 weeks of the initiation of treatment.
Two surgical options are available to treat brain abscesses
including needle aspiration, with or without catheter drain-
age, and excision. Ultrasound guidance or guidance with
frameless or framed stereotaxy can be used to accurately local-
ize an abscess and place a drainage needle or catheter within
the abscess cavity. Ultrasound can also be used to visualize
 CHAPTER 128 MANAGEMENT OF NEURAL TUBE DEFECTS, HYDROCEPHALUS, REFRACTORY EPILEPSY, AND CNS INFECTIONS
decompression of the abscess cavity as the pus is aspirated.
After gentle aspiration with a syringe and irrigation with nor-
mal saline, a catheter may be left in the abscess cavity for sev-
eral days until drainage of purulent fluid ceases. If aspiration
fails, as frequently occurs in the case of multiloculated ab-
scesses, surgical excision may be required. To accomplish this,
a limited craniotomy bone flap is turned, the dura is opened, a
small corticectomy is generated in the closest noneloquent pa-
renchymal area, and the abscess is excised en bloc while leav-
ing the surrounding white matter intact. The craniotomy is
closed in the usual fashion after thorough irrigation with sa-
line, and the patient is treated with long-term antibiotics.
Outcome
Outcome is determined by the rapidity of the diagnosis and
initiation of treatment. Moreover, the prognosis for survival
and neurologic recovery is contingent on the patient’s level
of consciousness at the time of diagnosis and institution of
treatment. Consequently, early diagnosis and prompt aggres-
sive therapy, with appropriate systemic antibiotics and surgi-
cal removal of suppurative collections, are paramount to
minimizing the morbidity and mortality associated with sup-
purative intracranial infections. In support of this, the mortal-
ity associated with a brain abscess in older series approached
32% but has decreased to 10% in recent years as a conse-
quence of more rapid diagnosis and treatment with appropri-
ate antibiotics.158 In contrast, the mortality rates associated
with intraventricular rupture of a brain abscess remain high,
ranging from 23.8% to 80%.167,170 Long-term neurologic def-
icits occur in about 44% of patients who survive surgical treat-
ment of brain abscess.176 Moreover, children who survive are
commonly left with learning disabilities168 and seizures.174
INTRASPINAL INFECTIONS
Spinal epidural abscess in children is a rare clinical entity177
that often escapes diagnosis until significant neurologic deficits
develop.178 It is a rapidly progressive, compressive lesion of
the SC that often requires rapid decompression to prevent per-
manent paraplegia. Hematogenous spread of bacteria from cu-
taneous or mucosal sources of infection such as furuncles,
pharyngitis, and dental abscesses is the most commonly
reported cause of spinal epidural abscess.177,179,180 Other
potential causes include direct extension from vertebral osteo-
myelitis, a paraspinal abscess, or a septic focus in the pelvis, ret-
roperitoneum, or posterior mediastinum. Such extension can
occur by way of veins passing through the intervertebral foram-
ina or bacterial seeding of an epidural hematoma that forms as a
result of blunt trauma.177 The most common pathogen of a spi-
nal epidural abscess in children is S. aureus, although anaerobic
and gram-negative bacteria have also been isolated.
Spinal epidural abscesses occur posterior to the SC in about
80% of cases. Posteriorly, the dura is adherent to the posterior
longitudinal ligament from C1 to S2. Consequently, pus that
accumulates posterior or posterolateral to the dura can spread
extensively rostrocaudally but is restricted dorsally by the lam-
inae and ligamenta flava. Although epidural infection can oc-
cur anywhere along the spinal axis, lower thoracic and lumbar
abscesses are most commonly reported in the literature.180
Once infection is established in the epidural space, pus
collects deep to the laminae and ligamenta flava and is sur-
rounded by thrombosed veins and granulation tissue. The
1697
dura provides a formidable barrier to infection. Consequently,
meningitis is uncommon unless a spinal tap is performed.
However, the SC is defenseless against compression and
may be injured as a result of vascular compromise from arterial
or venous thrombosis.180
Children with spinal epidural abscess may have fever and
malaise, and they typically complain of deep-seated back pain,
spinal tenderness, or radicular pain. The back pain is usually
midthoracic and radiates around the chest wall. The pain is con-
stant and can be exacerbated by coughing or movement. Patients
often exhibit nuchal rigidity and exquisite tenderness to percus-
sion over the involved spinal segments. A primary source of
infection may be evident. Symptomatic SC compression is man-
ifested by rapidly progressive weakness and sensory loss in the
legs and evidence of bladder or bowel dysfunction, or both.
If a child is suspected of having a spinal epidural abscess, a
peripheral white blood cell count, erythrocyte sedimentation
rate, and C-reactive protein level are obtained and are usually
elevated. Moreover, blood cultures are performed because
they may reveal the offending organism. A total spine MRI
with and without gadolinium enhancement enables visualiza-
tion of the extent and location of a spinal epidural abscess and
is the diagnostic procedure of choice. A spinal tap should not
be performed. If a myelogram must be performed as part of the
diagnostic workup (e.g., if MRI is not available), the spinal
puncture site should be placed as far away from the suspected
position of the abscess as possible.
After the diagnosis of spinal epidural abscess is made, treat-
ment must commence immediately to prevent long-term neu-
rologic disability. Aggressive antimicrobial therapy is initiated
with intravenous antibiotics that have good coverage against
staphylococci, anaerobes, and gram-negative organisms. Par-
enteral antibiotics are continued for at least 4 to 8 weeks.
Although there are a few reports in the literature of neurolog-
ically intact children being treated successfully with antibi-
otics alone,179 in the majority of cases, successful treatment
requires appropriate antibiotic therapy and emergency evacu-
ation of the abscess with decompression of the SC and nerve
roots, especially if the abscess is widespread. This is typically
performed by means of a posterior decompressive laminect-
omy with drainage of the purulent material. A one- or two-
level laminectomy at one or both ends of the abscess with
catheter irrigation and drainage is usually sufficient. If a multi-
level laminectomy is required for surgical decompression, the
child must be monitored carefully postoperatively for kypho-
sis and spinal instability, which can occur as a complication of
multiple-level laminectomies in children.179
The prognosis is related to the patient’s preoperative neu-
rologic condition.181 If the infection is recognized early and
the patient receives appropriate treatment, recovery is
expected to be excellent. If mild motor weakness has been pre-
sent for less than 36 hours, a good outcome is also possible.182
However, once a child develops paraplegia, recovery is highly
unlikely. Thus early diagnosis and appropriate treatment are
paramount to achieving a good outcome in children with a
spinal epidural abscess. Mortality from spinal epidural ab-
scess, which is unquestionably related to a delay in instituting
appropriate therapy, has remained stable over the past several
decades at about 14%.182
The complete reference list is available online at www.
expertconsult.com.
 such as myelomeningocele, arthrogryposis, or other type of
syndrome. This type of dislocation is a rigid, fixed dislocation
that occurs prenatally and will not respond to nonoperative
methods of treatment as the more common types of hip dis-
location usually will. Most cases of hip dysplasia are evident
at birth by physical examination or ultrasound, but, rarely, in-
stability may not develop until later in the first year of life and
may have different findings. Still other types will not be iden-
tifiable until adolescence, when they are often manifested as
pain without other physical examination findings. A change
in terminology has occurred in the past 15 years from “con-
genital dislocation” to “developmental dysplasia” of the hip.
This change reflects more accurate information about the less
apparent types of dysplasia not identifiable at birth and the
variations that are not always complete dislocations of the
joint but rather abnormal development or “dysplasia.”3,4
Hip dysplasia is the most common hip disorder in children.
One in 1000 newborns has a dislocated hip, and 10 in 1000
will have some form of subluxation or dysplasia. The majority
of hips will tighten up on their own, usually within the first
2 weeks of life. Approximately 4 in 10,000 will eventually
have problems of dislocation, subluxation, or acetabular dys-
plasia. Females have a higher incidence than males by a ratio
CHAPTER 129 of 4:1, which is thought to be due to sensitivity to the relaxin
hormone produced by the mother during pregnancy; relaxin
causes transient ligamentous laxity in females and not in males
(because males are not sensitive to the hormone). The left hip
Major Congenital is affected 60% of the time, probably because of pressure on
the hip from the mother’s sacrum because most birth presen-
tations are left occiput anterior.
Orthopedic The cause of hip dysplasia is multifactorial. Genetics, phys-
iologic factors, and mechanical factors such as prenatal and

Deformities postnatal positioning all play a role. Genetics is a factor be-

cause babies with a positive family history have a higher inci-
dence of DDH, as do siblings and twins. Mechanical factors
Kosmas Kayes and William Didelot are also involved because any form of intrauterine crowding
or abnormal positioning such as a first pregnancy, breech pre-
sentation, oligohydramnios, or multiple births puts the infant
at higher risk. This is also true postnatally because cultures
such as Native Americans that swaddle, strap, or papoose their
Developmental Dysplasia babies have a higher incidence of DDH as well.2,5–8

of the Hip
------------------------------------------------------------------------------------------------------------------------------------------------ DIAGNOSIS
TERMINOLOGY/INCIDENCE/ETIOLOGY History and Screening
Developmental dysplasia of the hip (DDH) is a term used to de- Early detection of DDH is crucial; the sooner treatment is ini-
scribe a wide spectrum of pathology related to the hip joint. tiated, the better the expected outcome. Classically, the perti-
All cases involve some type of incongruity between the femo- nent history of the newborn consists of whether the newborn
ral head and acetabulum. This can vary from minimal acetab- is the first child, the presentation in utero (breech), whether
ular dysplasia, in which the acetabulum has not developed there is a family history of DDH, the presence of any associated
fully (with a femoral head that may be normal in position orthopedic or genitourinary abnormalities, multiple births,
or slightly subluxated), to complete dislocation of the femoral and the presence of oligohydramnios. However, more evi-
head, in which there is no contact with the acetabulum. Dif- dence is showing that previous risk factors may no longer
ferent findings will be present at different ages of the individ- be completely accurate and the important risk factors are first
ual. In a newborn the changes vary from an unstable hip that is born, family history, and oligohydramnios. Some studies show
fully reduced within the acetabulum but can be gently sub- that breech position may not be a significant risk factor and
luxated or fully dislocated to a hip that is subluxated partially that risk factors are different for early and late diagnosis.9–11
out of the acetabulum and may or may not dislocate to a hip Most cases are detected early as a result of the screening
that is dislocated completely and may be reducible or not.1,2 protocol for pediatricians developed by the American Acad-
Another variation of dislocation, called a teratologic hip disloca- emy of Pediatrics; the protocol consists of serial physical
tion, usually occurs with some type of neuromuscular problem examinations at birth, at 2 weeks, and then at every well-baby

1699
1700 PART IX SPECIAL AREAS

follow-up visit at 1, 2, 4, 6, 9, and 12 months of age. New ev- angle of the posterior wall in relation to the ilium (Fig. 129-
idence suggests that risk factors alone are not enough to warrant 1). It can also provide visualization of the labrum, capsule,
ultrasound screening, so only a child with an abnormal physical iliopsoas, and femoral head ossific nucleus. An alpha angle
examination should be imaged by ultrasound. Physical exami- greater than 60 degrees is considered normal.
nation remains the gold standard for screening newborns, and Plain radiographs of the hip are not reliable before 3 to 4
ultrasound can be used after 4 weeks of life. Many hips are over- months of age when the nucleus of the acetabulum is starting
read on ultrasound as positive findings of dysplasia before 4 to ossify and becomes apparent radiographically. After this
weeks in children who will otherwise spontaneously resolve age, it is the standard means of diagnosis and the best way
their instability. Also, the accuracy of ultrasound is still operator to monitor progression of a dysplastic hip.23 An anteroposter-
dependent and variable.12 A child at risk should have either an ior (AP) view and a “frog-leg” view are both required when or-
ultrasound before 4 months or radiographs of the hip taken af- dering any hip radiographs on a child (referred to as “AP and
ter 4 months of age. Screening of the entire newborn population frog pelvis”). Several important parameters should be mea-
has not been necessary, although some cases of hip dislocation sured on the AP film (Fig. 129-2). The Shenton line is the con-
and dysplasia are not detected until later in life.13–19 There is tinuation of a semicircle drawn on the superior portion of the
some thought that screening may even potentially cause harm
because well over 90% of newborn hip instability will resolve
spontaneously and many children may be overtreated. One
meta-analysis showed that there was no clear evidence for or
against ultrasound screening in DDH.20–22
Physical Examination
Findings on physical examination vary with the age of the
child. From birth to about 3 months of age, the most reliable
findings are a positive Barlow or Ortolani maneuver. In the
Barlow test the hip is held flexed 90 degrees with neutral ad-
duction, and the knee is flexed 90 degrees. The first web space
and the palm of the hand should cradle the knee and be out- A
side the thigh with the long finger down the length of the
thigh. The hand should gently push the thigh posteriorly
while adducting the hip in an attempt to dislocate a reduced
hip out the back of the socket. The Ortolani maneuver is the
same, but instead of pushing posteriorly, the thigh is abducted
with pressure on the femur toward the middle of the socket
while pulling it up anteriorly and trying to reduce a dislocated
hip. Symmetric abduction is also important during the evalu-
ation. Any asymmetry of abduction on examination should be
evaluated further. Asymmetric skinfolds have been shown to
be a less reliable measure of hip dysplasia. B
After 3 to 4 months of age, the hip is usually dislocated and
no longer reducible. It may occasionally be reduced but is sub-
luxable. The Barlow and Ortolani maneuvers will no longer be
useful. Asymmetric abduction is the best indicator, as is leg-
length inequality, or the Galeazzi sign, which is positive when
both hips and both knees are flexed to 90 degrees and one
knee is higher than the other, giving an apparent limb-length
inequality. More accurately, the dislocated hip is more poste-
rior in position, thus making it look like the leg is shorter. At
walking age the most obvious findings are apparent leg-length
inequality with pelvic obliquity and increased lumbar lordo-
sis. The child will have a Trendelenburg gait from shortened
(weak) abductors and hip flexion contracture. If the disloca-
tion is bilateral, the child will have increased lumbar lordosis.
Imaging
Ultrasound and plain radiographs are both useful tools for C
assessment of hip dysplasia, but their timing is different.
Ultrasound is highly dependent on the technician performing FIGURE 129-1 A, Standard coronal plane ultrasonogram of a normal in-
the scan and the radiologist interpreting it. General screening fant’s hip. The alpha angle (A) corresponds to the bony acetabular roof
of all newborns is not recommended in the United States. (normal, >60 degrees). The beta angle (B) represents the cartilaginous roof
(normal, <50 degrees). B, Method for determining percent coverage of the
Ultrasound can assess the amount of subluxation by identify- femoral head: d/D  100 (normal,  50%). C, Coronal plane ultrasonogram
ing the position of the cartilaginous femoral head, as well as of a dysplastic hip. The alpha angle was 40 degrees; the femoral head was
evaluating acetabular development by measuring the alpha subluxated with only 25% coverage.
 CHAPTER 129
Normal DDH
Acetabular Index
CE Angle
Hilgenreiner Line
Disrupted
Shenton Line Shenton Line
Perkins Line Perkins Line
FIGURE 129-2 Diagram of radiographic measurements. CE, center edge
angle; DDH, developmental dysplasia of the hip.
obturator foramen and should be at the same level as the fem-
oral neck. If the line is disrupted, the femur is too superior in
the acetabulum. The Hilgenreiner line is a horizontal line
drawn through the level of the triradiate cartilage on each side.
A perpendicular line is then drawn at the lateral edge of the
acetabulum and intersects the Hilgenreiner line to form four
quadrants around the hip. The medial metaphyseal beak of
the femur should be in the inferior/medial quadrant, or it is
not in a reduced position. The acetabular index is the angle
formed by the Hilgenreiner line and a line drawn from the tri-
radiate cartilage to the lateral edge of the acetabulum. This an-
gle should be less than 25 degrees by 1 year of age to indicate
normal acetabular development. The higher the acetabular
index is, the more dysplastic (more vertical) the acetabulum.
This measurement has been shown to correlate with future hip
osteoarthritis.
The most definitive way to assess a dysplastic hip is by
injection of contrast material into the hip joint (hip arthrogra-
phy); because this requires general anesthesia, it is usually per-
formed at the time of operative reduction. Contrast outlines the
cartilaginous femoral head and shows the exact relationship to
the acetabulum. It can also show the position of the labrum and
ligamentum teres and whether there is a pooling of dye in the
medial joint space, which is indicative of laxity of the hip and a
less than ideal reduction. Although ultrasound is helpful in
assessing the position of the hip, arthrography can be per-
formed if there is any question regarding stability and position
of the hip. It is also extremely useful in assessing a closed reduc-
tion and in preoperative planning for any reconstructive sur-
gery. Computed tomography (CT) is useful postoperatively
to check the position of the hip in the cast after closed or open
reduction with immediate casting. CTwith 3-D reconstructions
can also be used preoperatively to assess the acetabulum.
Magnetic resonance imaging (MRI) has had limited use in
DDH, but recent studies show that it may be as useful as CT in
postoperative assessment and without the radiation exposure.
ANATOMY/PATHOANATOMY
Hip development starts with the femoral head and acetabulum
being formed as a block of cartilaginous cells. At 7 to 8 weeks a
cleft develops between the head and acetabulum and forms
the joint. So the hip starts “reduced” in the socket, and other
factors, previously discussed, contribute to the later instability.
MAJOR CONGENITAL ORTHOPEDIC DEFORMITIES 1701
The hip joint is completely developed by 11 weeks of gesta-
tion. The femoral head remains totally cartilaginous until
4 months of age, when the ossific nucleus starts to form bone.
The hip is dependent on the femoral head and acetabulum,
maintaining concentric reduction for continued normal devel-
opment until growth ceases. A socket without a reduced fem-
oral head will not form normally. If the hip stabilizes early, the
joint will most likely remodel any dysplasia and form a normal
joint. The longer the femoral head is not concentrically re-
duced, the less chance of the hip having normal anatomy at
maturity. The acetabulum undergoes the most remodeling
and development in the first year of life, but some remodeling
continues until year 4 or 5, after which time there is minimal
remodeling potential.
The anatomic structures involved in hip dysplasia
(Table 129-1) contribute differently in each case. The adduc-
tor muscle becomes contracted. The iliopsoas muscle can be a
major block to reduction of a completely dislocated hip. It be-
comes contracted and can exert a powerful obstacle to reduc-
tion by forming an hourglass-shaped constriction of the
anterior medial capsule. The ligamentum teres connecting
the fovea of the femoral head to deep inside the acetabulum
can hypertrophy and pose a block to a good concentric reduc-
tion as well. The pulvinar is an accumulation of fat and fibrous
tissue in the space next to the medial wall of the acetabulum,
which can keep the femoral head lateralized during a reduc-
tion attempt. The labrum or cartilaginous rim of the acetabu-
lum can become infolded and blunted and also block the
femoral head. The anterior-medial joint capsule can become
constricted. The transverse acetabular ligament spanning
the inferior portion of the hip between the medial and poste-
rior walls can also limit reduction inferiorly. All these prob-
lems may become more marked as the child gets older and
pose more difficulty in achieving an adequate reduction.
TREATMENT
There are two goals of treatment. The first is to reduce the hip
and allow it to stabilize within the acetabulum. The second goal
is to ensure remodeling of any acetabular dysplasia. If the first
goal is met at a young age and concentric reduction is achieved,
the acetabulum will be able to remodel with further growth.
Because the remodeling potential of the acetabulum is greatest
in the first year of life, it is important to begin treatment as soon
as possible. Treatment protocols depend on the age of the child
and the amount of dysplasia of the hip joint. Table 129-2 lists all
possible methods to help achieve reduction and stability.
Treatment is discussed as it relates to the age of the child. Some
of the methods are useful only in certain age groups, whereas
others span the entire age range.
TABLE 129-1
Potential Blocks to Hip Reduction
Adductor tendon
Iliopsoas tendon
Medial capsule
Infolded labrum
Ligamentum teres
Pulvinar
Transverse acetabular ligament
1702 PART IX SPECIAL AREAS
TABLE 129-2
Available Treatments for Hip Dysplasia
Pavlik harness
Closed reduction and casting
Open reduction and capsulorrhaphy
Femoral shortening osteotomy
Acetabular procedures
0 to 6 Months
If a positive examination is recorded at birth (positive Barlow
or Ortolani maneuver), immediate orthopedic referral is
recommended. Evidence suggests that most hips will tighten
up and stabilize within the first 2 weeks of life; therefore ini-
tiation of treatment before 2 weeks is controversial, and even
how and when to screen is uncertain. Either way, close follow-
up is mandatory. A Pavlik harness is the gold standard of treat-
ment at this age and can often be used in children up to
6 months of age or less than 20 lb/9 kg (Fig. 129-3). When
the hip is still dislocatable or reducible, the harness is the best
choice. It is an active method of treatment that allows the baby
to still have motion while encouraging the hip to remain in a
position of stability. The harness maintains a hip flexion of 90
to 100 degrees (but should not be tightened any more than at
90 degrees to avoid hyperflexion) and abduction to keep the
hip reduced in the most stable position. It is worn 24 hours a
day, and correct placement and adjustments are crucial to help
lower the incidence of complications. The harness can cause
avascular necrosis (AVN; discussed later), which is thought to
be due to hyperflexion of the hip or overtightening of the ab-
duction straps. Once treatment is initiated, the harness should
FIGURE 129-3 Infant being treated with a Pavlik harness. Hip flexion of
approximately 100 degrees is maintained by anterior straps. Posterior
straps (not seen) limit hip adduction.
be checked 1 week later to ensure proper fit and use by the
family. The length of treatment is generally twice as many
weeks as the child’s age at diagnosis, with some authors
recommending a minimum of 3 months’ treatment regardless
of age at initiation of treatment. It can be weaned when the
examination and imaging studies have normalized. It is effec-
tive in 90% or more children with Barlow-positive hips (re-
duced but dislocatable).24–26 If the hip is Ortolani positive
(dislocated but reducible), the harness may not be effective
and should be discontinued after 3 weeks of treatment.
Closed reduction may also be necessary in this age group if
the hip will not easily reduce with gentle manipulation or does
not respond to treatment in the Pavlik harness within 2 weeks.
Arthrography is usually performed under general anesthesia,
and the stability of the hip is checked. The child can then be
placed in a hip spica cast if certain criteria are met during
arthrography. The hip must be stable within the safe zone of
30 to 60 degrees of abduction and 90 degrees of hip flexion.
If excessive abduction is required for stability, open reduction
should be performed. The hip is at risk for AVN if extreme ab-
duction is necessary to keep the hip reduced in the cast. There
must also be only minimal (<3 to 5 mm) pooling of contrast
in the medial aspect of the joint or the closed reduction will be
less successful. Prereduction traction is controversial, with
studies both supporting and refuting its use in reducing the
incidence of AVN.26–28
After closed reduction, the child is placed in a hip spica cast
in the human position for 6 to 8 weeks. A CT (or MRI) scan is
usually checked postoperatively to ensure successful reduc-
tion. Another arthrogram is often obtained at the 6- to 8-week
mark to assess progress of the hip. Casting may last 6 to
12 weeks, and then abduction bracing is used all day or night-
time only to encourage remodeling of the acetabulum. Some
authors recommend some type of abduction bracing until
radiographs have normalized.29
6 to 12 Months
As the child gets older, if the hip has not been reduced
(Fig. 129-4), the compensatory changes become more severe
and more invasive treatment is often necessary to overcome
FIGURE 129-4 Right hip dislocation. Left hip is normal.
 CHAPTER 129
the blocks to reduction. At this age a Pavlik harness is not in-
dicated. Closed reduction is still the preferred method if a gen-
tle, concentric, stable reduction can be obtained. If too
aggressive abduction or pressure is necessary to keep the
hip reduced easily in the cast, the chance of complications,
mainly AVN, increases. At this point surgical open reduction
should be considered. Some authors think that unless a closed
reduction gives nearly perfect concentricity and is stable, care
should proceed immediately to open reduction, which they
believe is safer than a forced, incongruous closed reduction.
This view may be changing somewhat as closed reduction is
becoming more widely used. The goal of open reduction is
the same as that of closed reduction, to achieve concentric
stable reduction of the femoral head in the acetabulum.
During open reduction, as many of the blocks to reduction
as possible are corrected.
If the femoral head is superior to the acetabulum on radio-
graphs and the soft tissue is significantly contracted, a femoral
shortening osteotomy may be helpful to achieve a more stable
reduction with less pressure on the femoral head. It helps
decrease the possibility of AVN and improves the stability of
the hip. If excessive coxa valga or femoral anteversion is pre-
sent, the osteotomy can be readily modified to correct these de-
formities as well. The incision is made at the subtrochanteric
level, and the bone is then fixed with a plate and screws to hold
the alignment while bony union is achieved. Femoral shorten-
ing is almost always necessary in an older child of walking age.
Walking Age, Older Than 12 Months
Once children have been bearing weight on the hip, adaptive
changes occur and closed reduction becomes less successful.
Closed reduction may be useful up to 18 months of age or
more, but a larger percentage of these older hips will require
open reduction and many will need femoral shortening osteot-
omy (Fig. 129-5).30,31 The remodeling potential of the acetab-
ulum decreases as the child gets older, and some children may
need to undergo pelvic osteotomy to help gain coverage of the
femoral head and provide a more stable reduction (see
Fig. 129-5). A pelvic osteotomy can be performed in several
ways, with most aimed at redirecting the socket to provide
more anterior and lateral coverage of the femoral head. If more
coverage is obtained and better congruency of the hip is
FIGURE 129-5 Femoral osteotomy fixed with plate and screws. Pelvic
osteotomy (arrow).
MAJOR CONGENITAL ORTHOPEDIC DEFORMITIES 1703
achieved, there is more potential for the hip to form a more
normal joint with time.32,33 Evidence suggests that remodel-
ing will occur until 2 years after surgery; then if it is not
adequate, a secondary procedure may be necessary.
Some cases of hip dysplasia will go undetected until later in
life. After 4 years of age, the risks of reducing a hip surgically
increase. Some authors recommend attempting reduction up
to the age of 8 years. If both hips are dislocated in older
children, one must consider leaving them dislocated because
they may do well for a long time before osteoarthritis develops.
Other variations of DDH will be manifested in the early to late
teens as hip pain, and acetabular dysplasia will be detected on
radiographs. These hips are not dislocated, but the acetabu-
lum is deficient in coverage of the femoral head and the
femoral head may be subluxated to varying degrees. These
children will benefit from redirectional acetabuloplasty of
some type to provide a more congruous joint and better cov-
erage. This is done in the hope of delaying further arthritic
changes in the hip. Evidence supports the natural history of
DDH in that any residual dysplasia of the joint will lead to
osteoarthritis, which will eventually require total hip replace-
ment. All treatment of DDH is directed at delaying this process
for as long as possible.29,32,33,34
COMPLICATIONS
Avascular Necrosis
AVN of the femoral head can occur in up to 60% of treated
hips if the long-term results are looked at critically. All forms
of treatment put the blood supply at risk. The exact cause is
unknown but is suspected to be a pressure phenomenon on
the femoral head during closed reduction and injury to the
medial circumflex vessels surgically. In younger children the
cartilaginous femoral head is thought to be protective, and
younger kids with AVN have a good chance of healing and
doing well. AVN can cause growth arrest of the physis as well
and give rise to shorter femoral necks or varus or valgus
position of the neck at maturity. The final outcome, however,
is unpredictable at any age.34
Recurrent Dislocation
Some hips will not stabilize with conventional treatment whether
it is the Pavlik harness or surgical intervention. Even after open
reduction and bony procedures, the hip may again subluxate
over time or even dislocate again. This will necessitate another
surgical procedure, which increases the risk for AVN even
more and also bodes for a poorer result in the long term.34,35
Congenital Abnormalities
of the Feet
------------------------------------------------------------------------------------------------------------------------------------------------
METATARSUS ADDUCTUS
Terminology/Incidence/Etiology
Metatarsus adductus is a deformity in which an adducted fore-
foot gives the medial border of the foot a concave appearance
and the lateral border looks convex (Fig. 129-6). The hindfoot
is normal and has normal range of motion. It occurs in about 1
in 1000 births and has no gender predisposition.36 The most
1704 PART IX SPECIAL AREAS
FIGURE 129-6 Metatarsus adductus.
likely cause is intrauterine positioning. Although there was
evidence of a positive association with hip dysplasia, this is
no longer thought to be true.
Diagnosis/imaging
The deformity is best identified with the child prone, and on
examination the lateral border of the foot will deviate inward
and look convex. Its severity is measured by a line called the
heel bisector, which passes through the heel and is drawn out to
the toes. Normally, this line bisects the second and third toe. In
mild metatarsus adductus it bisects the third toe, and in severe
forms it will bisect the fourth or fifth. The rest of the foot is
normal. The hindfoot is normal, and there is normal ankle
dorsiflexion. The abductor hallucis muscles may be overactive
and pull the big toe into adduction as well (atavistic great toe).
Radiographs may be obtained to assess the foot structurally
but are not necessary unless there are other concerns.
Treatment
The natural history is that most feet with metatarsus adductus
will correct spontaneously if they are supple and correctable past
neutral.37 If the deformity is more severe and left untreated, it
may cause problems with shoe wear later from a prominent base
of the fifth metatarsal and medial deviation of the foot. It can also
be a source of in-toeing in a child. Most deformities resolve
without any treatment or with the parents manipulating the
feet. If the deformity persists or the feet are stiff and not passively
correctable, stretching casts may fix the deformity. If the problem
persists past 5 years of age, consideration may be given to surgery.
Correction at a young age by joint release may leave the foot
stiff. In the past, metatarsal osteotomies were performed for
correction and are still used at times. Waiting until after 5 years
of age and then performing opening wedge cuneiform and
closing wedge cuboid osteotomies is more reliable.38–40
FLEXIBLE FLATFOOT
Terminology/Incidence/Etiology
A flatfoot deformity occurs when the medial longitudinal arch
is absent. This finding is common in younger children, espe-
cially when they first begin walking. Many of these children
have ligamentous laxity that contributes to the arch being
depressed because the ligaments that support the arch are also
lax and unable to support the foot in the weight-bearing
position. The posterior tibial tendon may be incompetent
and contribute to the deformity.
Diagnosis/Imaging
Diagnosis is by clinical examination. The medial arch is absent
in the weight-bearing position and gives it the “flat-foot” ap-
pearance. The foot is flexible, and an arch will be reestablished
when the foot is hanging free or the patient stands on tiptoes.
As the patient goes up on toes, the heel will normally rotate
into varus and the arch will become apparent. Radiographs
will show a plantar-flexed talus that is not lined up with the
first metatarsal on the lateral projection; however, radiographs
are normally unnecessary unless other foot pathology is
suspected.
Treatment
No treatment is necessary if the foot is flexible. The parents
need to be reassured that it does not need to be treated and
an orthotic device will not help develop an arch—it will only
support the arch. Many patients will tighten up over time as
the ligamentous laxity improves and reestablish a medial arch
as they get older. Arthroereisis or stabilization of the subtalar
joint by an implant of some type (usually a Silastic or metal
screw or staple) may be considered if the patient is symptom-
atic. An opening wedge osteotomy of the calcaneus or cuboid
can be considered as well.41–43 Fusion of the subtalar joint is
reserved for patients with the most severe cases who continue
to have pain despite other treatment options.
If the foot is not flexible and the arch does not reconstitute
when the patient goes up on tiptoes (Root sign), it is consid-
ered a rigid flatfoot. This deformity may cause problems later
in life, and surgery may be indicated in those with pain or sig-
nificant deformity. Results are variable, and some patients may
eventually undergo fusion of the subtalar joint for pain relief.
CLUBFOOT
Terminology/Incidence/Etiology
Clubfoot is a condition isolated to the foot and leg, in which
the foot has varying degrees of the following deformities:
metatarsus adductus, cavus, heel or hindfoot varus, and equi-
nus (Fig. 129-7). Hence it is also called talipes equinovarus. It
occurs in 1 in 1000 live births. The severity of the deformity is
variable and ranges from a positional clubfoot that resolves
with minimal stretching and casting to rigid deformities that
are difficult to correct without surgery. The exact cause of
clubfoot is unknown but multifactorial, and it is becoming
more obvious that genetics plays a role.44 Positioning prob-
lems can result from anything that decreases the in-utero
space for the fetus (e.g., oligohydramnios, being the first baby,
multiple births). Many of the rigid, resistant feet are associated
with neuromuscular problems or syndromes such as spina
bifida, diastrophic dysplasia, sacral agenesis, and arthrogrypo-
sis. Those that are not associated with one of these conditions
are thought to result from a defect in formation. Classification
was difficult, but recently methods by Dimeglio and Pirani
have shown good correlation between the two systems.45–49
 CHAPTER 129
FIGURE 129-7 An infant with bilateral clubfeet demonstrates the
deformities that are part of clubfeet. The infant is prone, a position that
reveals the varus deformity of the hindfeet. The crease in the arch shows
that these feet have a cavus deformity. The feet are internally rotated and
are in equinus.
Diagnosis/Imaging
The diagnosis is usually apparent from the findings on phys-
ical examination discussed earlier. The forefoot is medially de-
viated as a result of the metatarsus adductus and supinated,
which causes the forefoot to be almost touching the tibia.
The calcaneus is rotated into varus and is high in the calf
and not palpable in the heel pad. There is a medial crease
of varying depth in the area of the medial arch. Finally, the
Achilles tendon is tight, which causes the high-riding calca-
neus. Occasionally, the diagnosis will be subtle if most of these
deformities are mild, but such cases are rare. If the foot does
not lack dorsiflexion and the heel is not in varus, it is probably
not a true clubfoot.
Radiographs are not necessary at birth because most of the
midtarsal bones are not yet ossified. Radiographs are helpful
as the child gets older to monitor the progress of treatment.
Anatomy/Pathoanatomy
The calcaneus and talus are normally divergent both in the
axial and sagittal planes. In a clubfoot they are more parallel
because of the varus deformity of the heel. This, along with
the metatarsus adductus, is the major bony deformity. The first
ray is also plantar-flexed, and the navicular is subluxated me-
dially and dorsally on the talar head. The talus is often smaller
and misshapen with a short neck. The soft tissue deformities
consist of a severe contracture of the Achilles tendon, as well
as contractures of the posterior tibial tendon, flexor hallucis
longus, and flexor digitorum longus. Where these tendons
cross in the medial aspect of the foot, they are occasionally
confluent with abnormal slips of other tendons. These ten-
dons are often entrapped in scar tissue and form what is called
the knot of Henry. The plantar fascia is also tight.50
Treatment
Treatment of a clubfoot usually requires some type of manip-
ulation and casting and may require surgery. A technique of
manipulation and casting advocated by Ponseti has been
around for more than 40 years, but only in the past 10 to
12 years has it been gaining widespread popularity among
pediatric orthopedists. There have been good short-term
MAJOR CONGENITAL ORTHOPEDIC DEFORMITIES 1705
results, and now older patients are showing maintenance of
the correction.51–54 The process consists of the application
of a series of four to five long leg casts with manipulation be-
fore each casting. The first step is to correct the plantar-flexed
first ray and line it up with the other metatarsals in the foot.
The next several casts bring the foot out of adduction and over
into neutral alignment with the tibia and finally mild overcor-
rection into valgus. The last cast tries to improve the Achilles
contracture by dorsiflexing the foot. Care should be taken to
not dorsiflex the foot too quickly or before complete correc-
tion of the varus or a rocker-bottom foot deformity may be
created. A tenotomy of the Achilles tendon is required in
the majority of patients (>90%) to release the heel cord
contracture, allow the calcaneus to come down into the heel
pad, and permit the foot to dorsiflex. This can be done in
the office under local anesthesia with or without sedation,
and then a final long leg cast is applied with the foot slightly
dorsiflexed and abducted 70 degrees. The cast is worn for
3 weeks, and then the patient goes into abduction bracing
to keep the deformities from recurring. These shoes are worn
until 4 years of age. Recurrence usually occurs after failure to
fully correct the foot or inability to wear the abduction braces.
If supination persists or the heel cord contracture recurs,
the patient may need to undergo posterior release with open
heel cord lengthening and a transfer of all or half of the
anterior tibialis tendon to the lateral side of the foot. This tech-
nique is usually successful in keeping the foot in a functional
position. However, many centers favor manipulation for any
deformity and any recurrence of deformity and avoid surgery
if at all possible.
Correction can be achieved surgically, but overcorrection
and undercorrection are both possible and over time recur-
rence or overcorrection of the foot may develop.55–57 Many
studies show deterioration of surgical results leading to arthri-
tis and pain in the foot as the patient gets older.58 The Ponseti
method appears to produce a better foot clinically, both
cosmetically and functionally.
In France, a technique developed by Dimeglio has
produced good results. The feet are manipulated daily by
the therapist and placed in a machine to give passive range
of motion to the foot. It is labor intensive and has not been
extremely practical in the United States. The Ponseti and
Dimeglio techniques are considered to be the gold standards,
with surgery having dropped out of favor.
Congenital Dislocation
of the Knee
------------------------------------------------------------------------------------------------------------------------------------------------
Congenital knee dislocation covers a spectrum of deformity
ranging from simple positional contractures that correct spon-
taneously to rigid dislocation of the knee joint requiring sur-
gical intervention (Fig. 129-8). The problem often occurs in
children with myelomeningocele, arthrogryposis, or other
syndromes such as Larsen’s. This deformity is frequently asso-
ciated with DDH, clubfoot, and metatarsus adductus. It is ex-
tremely important to look for associated hip dysplasia when a
knee dislocation is present. The deformity is recognized at
birth in an infant with a knee that is hyperextended and a foot
that is up by the infant’s head because of a flexed hip. The pa-
thology consists of one or all of the following: contracture of
1706 PART IX SPECIAL AREAS

Congenital Deformities
of the Spine
------------------------------------------------------------------------------------------------------------------------------------------------

Congenital scoliosis is relatively uncommon. It is a malforma-

tion in vertebral column development, typically a failure of
formation or a failure of segmentation.59 Congenital implies
that the deformity is present at birth. Even though the verte-
bral deformity is present at birth, the clinical deformity may
not be recognized until years later, depending on the amount
of growth imbalance that occurs. Isolated anomalies are usu-
ally sporadic, but multiple anomalies can be hereditary.60 The
true incidence is not known because some vertebral anomalies
produce little deformity and go unrecognized.
Embryologic development of the spine occurs during the
fourth to sixth weeks of gestation. By the sixth week the mes-
enchymal spinal anlage is complete. Defects in vertebral devel-
FIGURE 129-8 Congenital knee dislocation.
opment occur at this time. They typically stem from failure of
segmentation, failure of formation, or a mixed lesion. The re-
sultant clinical deformity is determined by the type and loca-
tion of vertebral deformity and the potential for asymmetric
the quadriceps tendon, absent suprapatellar pouch, tight spinal growth.
collateral ligaments, and anterior subluxation of the hamstring Failure of segmentation most commonly occurs in the tho-
tendons. racic or thoracolumbar region. It can be unilateral or bilateral.
If the problem has resulted from positioning in utero, the Unilateral failure of segmentation, or a “bar,” results in asym-
deformity is often mild and the knee is still fairly flexible. The metric growth of the spine (Fig. 129-9). It may not be evident
knee will correct spontaneously within a few weeks or can be on initial radiographs; however, there may be associated rib
assisted by gentle range-of-motion exercises to accelerate cor- fusions that should raise suspicion. It is often rapidly progres-
rection. Serial casting may also be helpful, and if hip dysplasia sive. The resultant deformity depends on the location of the
is present, a Pavlik harness can be worn once knee range of bony tether. A lateral bar results in scoliosis, an anterior bar
motion increases enough to allow the harness to fit properly. results in kyphosis, and a posterior bar results in lordosis.
A more severe contracture or true dislocation will be appar- It can also occur in combination and result in kyphoscoliosis
ent fairly early in the course of treatment because there will be or lordoscoliosis. By far the most common is scoliosis,
minimal response to stretching and casting. It is usually evi- followed by kyphosis. Lordosis is quite rare. Bilateral involve-
dent within the first 3 months of treatment. If 30 degrees of ment creates a block vertebra with minimal deformity.
knee flexion has not been gained by this time, surgery is indi- Failure of formation can range from mild wedging to
cated. Surgery may consist of one or more of the following complete absence of half of the vertebra (hemivertebra)
procedures, depending on the severity of the condition. The (Fig. 129-10). It can occur anywhere along the spinal axis.
first step is V-Y quadricepsplasty or Z-lengthening of the ten- A hemivertebra is the most common. If present in the thoracic
don, followed as needed by release of the anterior joint spine, there will be an associated extra rib. The extent of
capsule, posterior transposition of the hamstring tendons, the clinical deformity depends on the location of the
and mobilization of the collateral ligaments. Ninety degrees
of knee flexion should be achieved, and then the knee is casted
in 45 to 60 degrees of flexion for 3 to 4 weeks.

Congenital Dislocation
of the Patella
------------------------------------------------------------------------------------------------------------------------------------------------

True congenital dislocation of the patella is rare. This entity

must be distinguished from a recurrent dislocation that occurs
later in life. In congenital dislocation of the patella, the patella
is hypoplastic or absent and the femoral trochlea is often
flattened. The lateral retinaculum is tight, and the patella is
completely dislocated laterally. The patella is frequently adher-
ent to the iliotibial band and is often irreducible over the A B C D
condyle. Genu valgus is frequently present, as is a flexion FIGURE 129-9 Defects of segmentation. A, Unilateral unsegmented bar
contracture of the knee. Surgical intervention is required causing scoliosis, which is usually progressive. B, Block vertebra caused by
bilateral failure of segmentation. C, Anterior segmentation defect causing
and involves extensive lateral retinacular release, often to kyphosis. D, Failure of segmentation posteriorly causing lordosis. (Modi-
the greater trochanter, medial plication, and either hamstring fied with permission from Winter RB: Congenital Deformities of the Spine.
tenodesis or transfer of half of the patellar tendon. New York, Thieme-Stratton, 1983.)
 CHAPTER 129
A B
C D E A thorough neurologic examination must be performed,
FIGURE 129-10 Defects of formation. A, Anterior central defect (“butter-
fly vertebra”), usually nonprogressive. B, Incarcerated hemivertebra. Note
the accommodation by adjacent vertebrae and minimal deformity. C to E,
Wedge, free, and multiple hemivertebrae, respectively, demonstrating var-
iation in the degree of vertebral wedging and increasing tendency for pro-
gression. (Modified with permission from Bradford DS, Hensinger RM: The
Pediatric Spine. New York, Thieme, 1985.)
hemivertebra and on the potential for growth of the hemiver-
tebra. A lateral hemivertebra will result in scoliosis, whereas a
posterior hemivertebra will produce kyphosis.
The remaining deformities are either mixed or cannot be
classified because of the extent of their pathology
(Fig. 129-11). The amount of clinical deformity is variable,
depending on the defect. The most progressive deformity,
however, is a combination of a hemivertebra with a contralate-
ral unsegmented bar.61 Such deformities rapidly progress and
will require intervention, usually at the time of diagnosis.
The neural axis forms in conjunction with the bony canal,
and the frequency of intraspinal anomalies has been reported
to be as high as 40%.62,63 Abnormality of the overlying skin
such as a dimple, lipoma, hairy patch, or nevus may be present
70% of the time. Diastematomyelia is believed to be the most
common associated defect.64 It is a sagittal split in the spinal
cord or cauda equina by an osseous or fibrocartilaginous
spur protruding posteriorly from the vertebral body. It should
be removed if symptomatic or before any corrective spinal
surgery. Other intraspinal anomalies include tethered cord,
syringomyelia, epidermoid cyst, lipoma, and teratoma.
MAJOR CONGENITAL ORTHOPEDIC DEFORMITIES 1707
FIGURE 129-11 Multiple anomalies with combined failure of formation
and failure of segmentation. Note the fused ribs in a concavity suggestive
of a “bar.”
and any clinical suspicion of neurologic abnormality requires
imaging with MRI.
Associated anomalies occur in other systems up to 60%
of the time65 including genitourinary, 25%; Klippel-Feil
syndrome (cervical spine fusions), 25%; and cardiac, 10%.
Abnormalities of the gastrointestinal system can also be seen,
as well as esophageal atresia and imperforate anus in connec-
tion with the VATER association (in addition to vertebral
defects and radial and renal dysplasia).66 Other musculoskel-
etal deformities such as Sprengel deformity (congenital
elevated scapula), pectus abnormalities, and appendicular
skeletal deficiencies are less common. All patients should
undergo renal ultrasound screening to exclude underlying
abnormalities because the genitourinary system is difficult
to assess by physical examination alone.
An infant with congenital scoliosis may or may not have a
spinal deformity at birth. If a deformity is present, it is typi-
cally a lateral deviation of the spine. It can be confused with
congenital muscular torticollis (see Chapter 60) or infantile id-
iopathic scoliosis (thoracic spinal curvature). Both these con-
ditions can cause clinical deformity at birth, but radiographs
will fail to show vertebral anomalies. Most often, vertebral
anomalies are incidental findings during the workup for other
conditions (Fig. 129-12), and initial detection can occur at
any age. The diagnosis is made with radiographs, but MRI
or CT may be necessary to better delineate the pathology.
There is little nonoperative treatment to offer other than
observation. Unlike other types of scoliosis that may respond
to bracing, it is not routinely indicated for congenital scoliosis.
1708 PART IX SPECIAL AREAS

performed to balance the remaining growth. Posterior fusion

FIGURE 129-12 Incidental finding on a kidney-ureter-bladder radio-
graphic workup for imperforate anus. Note the S1 hemivertebra.
alone is adequate for older children with little remaining spi-
nal growth. Hemiepiphysiodesis is a variation of fusion in situ.
It involves fusion of only the convex side of the curve, thus
permitting the concave side to continue unopposed growth
and allow correction to occur. Because the correction is grad-
ual, the risk for neurologic impairment is decreased. Hemiver-
tebra excision is typically reserved for deformities in the
lumbar spine where spinal imbalance is more pronounced
and the cauda equina is safer to retract (Fig. 129-13). Excision
has been performed in the thoracic and thoracolumbar spine
less often, but with favorable results.69 An anterior and poste-
rior approach is usually necessary. Spinal osteotomy or verteb-
rectomy is reserved for the most severe curves with imbalance
that are not amenable to other more standard techniques.
It carries the highest risk for neurologic injury because of
the acute correction and is considered a “salvage” procedure
by most.
Anterior approaches to the spine require the traditional
thoracotomy or thoracoabdominal approaches. Postoperative
morbidity can be significant in this population because of
other coexisting abnormalities, and pulmonary function may
be further diminished after these approaches. Video-assisted
thoracoscopic surgery can be useful in some instances.70 Visu-
alization of the spine is adequate, and the instrumentation
continues to improve to allow for more complete diskectomy
and anterior fusion. Postoperative pulmonary morbidity
should be lessened.
Congenital kyphosis is a variant of congenital scoliosis and
usually occurs secondary to failure of formation of the anterior

The ideal therapeutic modality is surgical stabilization before

spinal deterioration.
The surgical treatment of congenital scoliosis depends on
the type of deformity present, the location of the deformity,
and the amount of growth remaining. The goal of treatment
is to halt progression and achieve spinal balance at maturity.
The most rapid vertebral growth occurs up to the age of 2 years
and again during adolescence. Thus clinical deformities can
change rapidly during these times and require close monitor-
ing. Surgical complications are increased in this population.
This type of scoliosis carries the highest risk for neurologic
complications with attempts at intraoperative correction. Ide-
ally, patients who have deformities with a high probability of
progression should undergo stabilization when their curves
are small.67 Other patients with a lower probability or an un-
known probability of progression are managed by close fol-
low-up with serial radiographs. Surgical intervention is
delayed as long as the curve is not progressing and the spine
remains in balance. Overall, 50% of curves will be stable, 25%
will progress slowly, and 25% will have rapid progression.
Surgery is the definitive treatment of congenital scoliosis. It
is indicated for rapidly progressive curves, deformities with a
high propensity to progress, and deformities that cause spinal
imbalance. Numerous surgical techniques are available,
depending on the type and location of the deformity. Attempts
at correction carry a high risk of neurologic impairment.
Fusion in situ is considered the “gold standard”68 and is indi-
cated for curves that are progressive or prone to progress but
still within a range where spinal balance can be maintained. FIGURE 129-13 Left L5 hemivertebra causing spinal imbalance. This is
In younger children, anterior and posterior fusion must be the ideal location for hemivertebra excision.
 CHAPTER 129
FIGURE 129-14 Posterior hemivertebra with anterior failure of forma-
tion resulting in kyphosis.
body (Fig. 129-14).71 It can also result from failure of segmen-
tation with an anterior or, more commonly, an anterolateral
bar. This type of deformity has a poor prognosis for progres-
sion and neurologic involvement. If unrecognized, it can re-
sult in paraplegia. Congenital kyphosis requires surgical
stabilization at the time of diagnosis to prevent further deteri-
oration. Bracing is of no value. Because of the variability in
deformity, no single surgical approach can be applied to all.
Typically, if the patient is younger than 5 years and the
kyphosis is less than 50 degrees, posterior fusion will suffice.
If older than 5 years or more severe kyphosis exists, anterior
and posterior fusion is required.72
Sacral agenesis, or complete or partial absence of the sa-
crum, is another type of congenital spinal anomaly. It has been
associated with maternal insulin-dependent diabetes,73 al-
though this relationship has more recently been questioned.74
It is often associated with gastrointestinal and genitourinary
abnormalities. Neurologic involvement is variable and often
results in hip, knee, and foot deformities. The prognosis
and treatment depend on the degree of malformation, the sta-
bility of the spine, the level of neurologic involvement, and
any other associated anomalies. Typically, only patients with
stability of the spinal attachment to the pelvis are able to am-
bulate.75 They may require surgical procedures to correct their
lower extremity deformities. Most others with spinal instabil-
ity at the spinal-pelvic junction will not ambulate functionally
and require treatments aimed at improved sitting balance and
positioning.
The complete reference list is available online at www.
expertconsult.com.
MAJOR CONGENITAL ORTHOPEDIC DEFORMITIES 1709
SUGGESTED READINGS
Bialik V, Fishman J, Katzir J, Zeltzer M. Clinical assessment of hip instability in
the newborn by an orthopaedic surgeon and a pediatrician. J Pediatr
Orthop 1986;6:703.
Cheng JC, Chan YL, Hui PW, et al. Ultrasonographic hip morphometry in
infants. J Pediatr Orthop 1994;14:24.
Curtis BH, Fisher RL. Congenital hyperextension with anterior subluxation of
the knee. J Bone Joint Surg Am 1969;51:255.
Drummond DS, O’Donnell J, Breed A, et al. Arthrography in the evaluation of
congenital dislocation of the hip. Clin Orthop 1989;243:148.
Eggli KD, King SH, Boal DK, Quiogue T. Low-dose CT of developmental
dysplasia of the hip after reduction: Diagnostic accuracy and dosimetry.
AJR Am J Roentgenol 1994;163:1441.
Ferris B, Aichroth P. The treatment of congenital knee dislocation. Clin Orthop
1987;216:135.
Fish DN, Herzenberg JE, Hensinger RN. Current practice in use of
pre-reduction traction for congenital dislocation of the hip. J Pediatr
Orthop 1991;11:149.
Fleissner Jr PR, Ciccarelli CJ, Eilert RE, et al. The success of closed reduction in
the treatment of complex developmental dislocation of the hip. J Pediatr
Orthop 1994;14:631.
Grill F, Benshael H, Canadell J, et al. The Pavlik harness in the treatment of
congenital dislocating hip: Report of a multicenter study of the European
Paediatric Orthopaedic Society. J Pediatr Orthop 1988;8:1.
Hensinger RN. Congenital dislocation of the hip. CIBA Clin Symp 1979;31:1.
Herring JA, Cummings DR. The limb-deficient child. In: Morrissy RT,
Weinstein SL, eds. Lovell and Winter’s Pediatric Orthopaedics. 4th ed. Phil-
adelphia: Lippincott-Raven; 1996.
Heyman CH, Herndon CH, Strong JM. Mobilization of the tarsometatarsal and
intermetatarsal joints for the correction of resistant adduction of the fore
part of the foot in congenital clubfoot or congenital metatarsus varus. J Bone
Joint Surg Am 1958;40:299.
Hinderaker T, Daltveit AK, Irgens LM, et al. The impact of intra-uterine factors
on neonatal hip instability: An analysis of 1,059,479 children in Norway.
Acta Orthop Scand 1994;65:239.
Hummer CD, MacEwen GD. The coexistence of torticollis and congenital
dysplasia of the hip. J Bone Joint Surg Am 1972;54:1255.
Ilfeld FW, Westin GW, Makin M. Missed or developmental dislocation of the
hip. Clin Orthop 1986;203:276.
Ishii Y, Weinstein SL, Ponsetti IV. Correlation between arthrograms and
operative findings in congenital dislocation of the hip. Clin Orthop
1980;153:138.
Jacobs JE. Metatarsus varus and hip dysplasia. Clin Orthop 1960;16:203.
Jones GT, Schoenecker PL, Dias LS. Developmental hip dysplasia potenti-
ated by inappropriate use of the Pavlik harness. J Pediatr Orthop
1992;12:722.
Kahle WK, Anderson MB, Alpert J, et al. The value of preliminary traction in
the treatment of congenital dislocation of the hip. J Bone Joint Surg Am
1990;72:1043.
Katz K, David R, Soudry M. Below-knee plaster cast for the treatment of meta-
tarsus adductus. J Pediatr Orthop 1999;19:49.
Kershaw CJ, Ware HE, Pattison R, Fixsen JA. Revision of failed open
reduction of congenital dislocation of the hip. J Bone Joint Surg Br
1993;75:744.
Kite JH. Congenital metatarsus varus. J Bone Joint Surg Am 1967;49:388.
Lourenco AF, Morcuende JA. Correction of neglected idiopathic clubfoot by
the Ponseti method. J Bone Joint Surg Br 2007;89:378–381.
Malvitz TA, Weinstein SL. Closed reduction for congenital dysplasia of the hip.
J Bone Joint Surg Am 1994;76:1777.
McKay DW. New concept of and approach to clubfoot treatment, I: Principles
and morbid anatomy. J Pediatr Orthop 1982;2:347.
McKay DW. New concept of and approach to clubfoot treatment, Section III:
Evaluation and results. Correction of the clubfoot. J Pediatr Orthop 1983;
3:141.
Morcuende JA, Dolan LA, Dietz FR, Ponseti IV. Radical reduction in the rate
of extensive corrective surgery for clubfoot using the Ponseti method.
Pediatrics 2004;113:376–380.
Nogi J, MacEwen GD. Congenital dislocation of the knee. J Pediatr Orthop
1982;2:509.
Ooishi T, Sugioka Y, Matsumoto S, Fujii T. Congenital dislocation of the knee.
Clin Orthop 1993;287:189.
Pavlik A. Stirrups as an aid in the treatment of congenital dysplasias of the
hip in children. J Pediatr Orthop 1989;9:157.
Ponseti IV, Becker JR. Congenital metatarsus adductus: The results of treat-
ment. J Bone Joint Surg Am 1966;48:702.
1710 PART IX SPECIAL AREAS
Quinn RH, Renshaw TS, DeLuca PA. Preliminary traction in the treatment of
developmental dislocation of the hip. J Pediatr Orthop 1994;14:636.
Race C, Herring JA. Congenital dislocation of the hip: An evaluation of closed
reduction. J Pediatr Orthop 1983;3:166.
Ramsey PL, Lasser S, MacEwen GD. Congenital dislocation of the hip: Use of
the Pavlik harness in the child during the first six months of life. J Bone Joint
Surg Am 1976;58:1000.
Richards BS, Faulks S, Rathjen KE, et al. A comparison of two nonoperative
methods of idiopathic clubfoot correction: The Ponseti method and the
French functional method. J Bone Joint Surg Am 2008;90:2313–2321.
Rosendahl K, Markestad T, Lie RT. Ultrasound screening for developmental
dysplasia of the hip in the neonate: The effect on treatment rate and prev-
alence of late cases. Pediatrics 1995;96:982.
Smith JT, Bleck EE, Gamble JG, et al. Simple method of documenting meta-
tarsus adductus. J Pediatr Orthop 1991;11:679.
Stanton RP, Capecci R. Computed tomography for early evaluation of devel-
opmental dysplasia of the hip. J Pediatr Orthop 1992;12:727.
Tavares JO, Gottwald DH, Rochelle JR. Guided abduction traction
in the treatment of congenital hip dislocation. J Pediatr Orthop 1994;14:643.
Viere RG, Birch JG, Herring JA, et al. Use of the Pavlik harness in congenital
dislocation of the hip. J Bone Joint Surg Am 1990;72:238.
Weinstein SL. Developmental hip dysplasia and dislocation. In: Morissy RT,
Weinstein SL, eds. Lovell and Winter’s Pediatric Orthopedics. Vol 2. 5th ed.
Philadelphia: Lippincott Williams and Wilkins; 2001:905.
Weinstein SL, Mubarak SJ, Wenger DR. Developmental hip dysplasia and dis-
location: Part 1. J Bone Joint Surg Am 2003;85:1824.
Weinstein SL, Mubarak SJ, Wenger DR. Developmental hip dysplasia and dis-
location: Part 2. J Bone Joint Surg Am 2003;85:2024.
Weintroub S, Grill F. Ultrasonography in developmental dysplasia of the hip. J
Bone Joint Surg Am 2000;82:1004.
Wenger DR, Maudlin D, Speck G, et al. Corrective shoes and inserts as treat-
ment for flexible flatfoot in infants and children. J Bone Joint Surg Am
1989;71:800.
 burden. At least one malformation is identifiable in about 15%
of newborns. The distinction between major and minor is ar-
bitrary because all human deformities exist on a continuum.
However, even the most minor physical aberration can predis-
pose the afflicted person to psychosocial isolation, justifying
surgical therapy.

SKIN AND SOFT TISSUES

Embryology
The developmental origins of the skin, muscle, connective tis-
sue, and tendons commence during the third week of embry-
onic life. During this period, a primitive streak appears and
facilitates the formation of the trilaminar embryo (ectoderm,
mesoderm, and endoderm), notochord, and somites.
Development of the Epidermis
The skin is the largest organ in the body and is composed of
two layers: the dermis and the epidermis. It serves a variety of
functions beyond simply providing an interface between the
body and the environment. The skin provides temperature
regulation, sensory perception, and a barrier against water loss
CHAPTER 130 and microbial invasion and is a major neurally regulated
immune organ.6,7 The cutaneous system is also a medium
for nonverbal communication, and the appearance of the skin
contributes to interpersonal recognition and one’s sense
Congenital Defects of self.
The epidermis originates from a monolayer of ectodermal
cells forming the periderm at about 4 weeks of gestation. The
of the Skin early epidermis is a monolayer exposed to the amniotic fluid,
with which it participates in significant salt and water ex-

and Hands change. By the third month the epidermis becomes three cell
layers thick (basal, intermediate, and superficial layers). The
epidermis continues to thicken, and by the sixth month differ-
Edward P. Miranda entiation has progressed far enough to achieve effective barrier
function. Upon microscopy the skin is similar to newborn epi-
dermis with several observable layers: the strata basale (inner
layer), spinosum, granulosum, and corneum (outer layer).
The stratum basale retains the keratinocyte stem cells that will
The sequence of cellular proliferation, differentiation, and re- repopulate the other layers over the lifetime of an individual.
gression that is characteristic of human embryogenesis and Keratinocytes over their lifetime progress from the stratum
subsequent development is complex and possesses little toler- basale to ultimate sloughing from the stratum corneum. This
ance for error. Errors and injury during gestation do occur migration is characterized by progressive differentiation and
quite often as evidenced by spontaneous abortion rates of accumulation of keratin granules (in the stratum granulosum)
up to 26.9% when implantation is detected by measurement and the intermediate filament filaggrin in the stratum corneum
of human chorionic gonadotropin (HCG).1 The “failure” of to maintain an effective barrier to the environment.
these gestations can also be viewed as “success” because em- Several other cell types migrate into the epidermis to assist
bryos with errors incompatible with life (e.g., major chromo- in its function. Melanocytes are neural crest–derived cells that
somal abnormalities) are culled out. This selection for major first migrate into the dermis and then into the epidermis.
errors is not precise because many gestations continue with These cells produce melanin pigment for defense against ul-
sublethal abnormalities. traviolet photoinjury. At approximately 3 months of gestation,
Langerhans cells (LCs) are found in the epidermis. LCs are
bone marrow derived, professional antigen-presenting cells
that arrange themselves in the epidermis to form an immuno-
General Incidence and Risk
------------------------------------------------------------------------------------------------------------------------------------------------
logic net to detect and present foreign antigen to T cells.

A major malformation has been defined as a structural anom- Development of the Dermis and Other
aly with medical and social consequences. The incidence of Mesodermal Tissues
major malformations has been estimated to be 2% to 3% of During the third week of embryogenesis a primitive knot ap-
live births.2–5 Minor malformations have been defined as pears at the cranial end of the primitive streak. Mesenchymal
structural alterations that pose no significant medical or social cells migrate cranially from the knot forming a midline cellular
1711
1712 PART IX SPECIAL AREAS

cord known as the notochordal process. The process grows until maintain the potential for growth and wound healing, and
it reaches the oropharyngeal membrane; at this point it can ex- resist infection after healing. Nevertheless, alloplastic surgical
tend no farther because of the densely adherent ectoderm and implants, judiciously placed during adolescence, remain an
endoderm. The notochord gives rise to the vertebral column excellent reconstructive standard in special clinical circum-
and induces the overlying ectoderm to form the neural plate, stances such as congenital hypoplastic breast deformity.
the primordium of the nervous system. A wide spectrum of surgical procedures has evolved to op-
As the notochord forms, the intraembryonic mesoderm timize the use of autologous tissues. Use of nonvascularized
thickens on each side of the notochord to form paired longi- tissue grafts is a simple and effective way to replace skin, mus-
tudinal columns of paraxial mesoderm. At the end of the third cle, connective tissue, and tendons. This technique is justifi-
week, columns of paraxial mesoderm divide into paired able only when the gain at the recipient site will compensate
cuboidal bodies called somites. Eventually, 42 to 44 pairs of so- for the loss sustained at the donor site. Furthermore, a well-
mites develop in a craniocaudal sequence. Somites are the seg- vascularized recipient bed is necessary to nourish the graft un-
mental precursors to most of the axial skeleton, musculature, til neovascularization occurs. When the bed is not suitable for
and its associated dermis of the skin. a graft or when specialized tissue is desired for the reconstruc-
The emergence of mesenchyme at 3 weeks of embryonic life is tion, then en bloc (composite) flap transfer of specialized
essential to organogenesis. By 8 weeks of life, the face and hands tissue with its own intrinsic vascular supply should be used.
are recognizably human. At 12 weeks of life, the palate has The method of tissue transfer depends on the proximity of the
formed and the sex can be determined by visual inspection of donor and recipient sites. Free microvascular flap transfer is
the external genitalia. Interruptions in the development and an elegant method of moving tissue when the distance from
differentiation of mesodermal events (mesenchymal migrations, donor to recipient site is too great for local rearrangement.
notochord, somites) accurately reflect the wider diversity of More recently, off-the-shelf biologic or combined biologic-
congenital malformations found in the skin, connective tissues, synthetic materials have been used to reconstruct congenital
muscles, and tendons. With such an expansive array of defor- and acquired defects of normal tissue. Recent improvements
mities, systematic analysis and anatomic regionalization of the in acellular dermal matrices (ADMs) (e.g., Alloderm or Strat-
defect are necessary to formulate an effective treatment plan. tice) and bilayered dermal regeneration templates (combining
an outer silicone film to recreate barrier function and a deeper
synthetic layer of cross-linked collagen and cartilage, analo-
Analysis of Defects and gous to a synthetic extracellular matrix [ECM]) have bridged
the divide between pure autologous tissue and alloplast-only
Technical Aspects of Treatment reconstructions. ADMs have been used for reconstructions of
------------------------------------------------------------------------------------------------------------------------------------------------
the abdominal wall, diaphragmatic hernia, and other de-
The structural aspect of any defect can be subdivided into fects.12,13 Postimplantation histologic analyses reveal that
malposition of otherwise normal structures, regional absence ADMs are incorporated by the body and used as a scaffold
of normal tissue, and aberrations of tissue composition. for the production of new fascia-like tissue with revasculariza-
tion and ingrowth of fibroblasts.14
MALPOSITIONING OF OTHERWISE NORMAL Bilayered dermal regeneration templates have been used as
STRUCTURES an intermediate step for reconstruction of full-thickness cuta-
neous defects followed by thin split-thickness skin grafting in-
Congenital malpositioning of normal structures can occur cluding those with limited exposure of tissues that have
from persistence of embryonic lines of fusion, displacement traditionally fared poorly with skin grafts alone (e.g., bone
of anatomic landmarks from abnormal growth of adjacent and tendon). These constructs limit the inflammatory re-
tissues, or developmental malrotation. Visceral malposition- sponse, decrease scar contracture, and allow for the formation
ing may be entirely inconsequential to the affected individual. of a neo-dermis via vascular and fibroblast ingrowth without
Dextrocardia with situs inversus is an incidental finding in the need for potentially morbid flap harvest (Fig. 130-1).15
asymptomatic patients. More commonly, malpositioning has Tissue expansion is an excellent way to create composite
more deleterious effects. The magnitude of the mishap soft tissue for use with fewer donor-site demands. Hair-
depends on the cascade of events resulting from the derail- bearing replacement of deficient scalp in conditions such as
ment of normal development. Spina bifida and facial clefts aplasia cutis (see later) best illustrates the potential dividends
are examples of defects resulting in severe functional and gained from tissue expansion. Distraction osteogenesis is the
social consequences.8–11 analogous procedure for bone. The use of this powerful tech-
nique for bone lengthening, first described by Ilizarov for the
REGIONAL ABSENCE OF NORMAL TISSUE lengthening of long bones, is now being used in the facial
skeleton. Gradual facial-bone lengthening without donor de-
Structural anomalies resulting from a deficiency or absence of formity has been effective in treating mandibular hypoplasia,
tissue have taxed the ingenuity of plastic surgeons. Tradition- which is found in disorders such as hemifacial microsomia.
ally, alloplastic or autologous materials have been used in the
reconstruction of congenital defects that are caused by
ABERRATIONS OF TISSUE COMPOSITION
deficient or absent tissues. Benefits of reconstruction with
alloplastic materials include unlimited supply, no donor de- An imbalance in the relative cellular composition of a tissue
fect, and ease of use. However, reconstruction with autologous may lead to significant malformation without a deficiency in
tissues is generally superior for congenital malformations tissue mass. Skin lesions such as congenital giant hairy nevus
because such tissues can fully integrate with living structures, (see later) and port wine stains are anomalies that express an
 CHAPTER 130 CONGENITAL DEFECTS OF THE SKIN AND HANDS 1713

A B

C D
FIGURE 130-1 Bilayered dermal regeneration matrix (Integra) for reconstruction of soft tissue defects. A, Preoperative appearance of a soft tissue defect
in a 7-year-old girl with small amounts of exposed bone and tendon visible. B, Postoperative (2 weeks) after placement of bilayered dermal regeneration
matrix (Integra). C, Postoperative (2 weeks) after removal of the silicone outer layer; note the vascularity of the neodermis. The tissue is ready for grafting.
D, Postoperative appearance after skin grafting over the neodermis. An additional 2-mm dermis is present with decreased contracture and increased sup-
pleness compared with traditional skin grafting. (Courtesy Edward P Miranda.)

excess of a specific cell or tissue type. In congenital giant hairy

nevus, the melanocyte-related nevus cell has excessively pro- Specific Anatomic Regions
liferated in the dermis. After birth, these cells have an in- ------------------------------------------------------------------------------------------------------------------------------------------------

creased risk of degenerating into melanomas, leading to the SKIN

rationale for surgical ablation.16 Cutaneous capillary vascular
malformations (port wine stains) result from an excessive de-
velopmental accumulation of abnormal microvascular tissue
in the dermis. These lesions are not mitotically active and have
no neoplastic potential. However, they are disfiguring and so-
cially disturbing. Direct excision is generally undesirable due
to subsequent scarring and deformity. Selective ablation of
vascular tissue with laser light is an effective method for treat-
ing this disorder.17
Many localized congenital malformations are inconsequen-
tial and can be considered normal variants. For example, the
congenital absence of the palmaris longus tendon has a prev-
alence of 15% to 20%.18 Such anomalies require no treatment.
However, when function or body image is threatened, a care-
ful analysis of the magnitude of malpositioning of normal
structure, degree of tissue deficiency, and composition of
the aberrant tissue is essential in planning the reconstruction.
Judicious timing of the operation can often interrupt a cascade
of predictable secondary deformities.
Aplasia Cutis Congenita
Aplasia cutis congenita (ACC) is a rare congenital disorder
characterized by full-thickness absence of the skin and under-
lying tissues over a section of the body. The scalp is involved in
approximately 80% to 90% of cases with occasional loss of
all cranial tissues including calvarium and dura (15% to
30%). The true incidence is unknown, and approximately
500 cases have been reported in the medical literature. The
diagnosis of ACC is clinical and is generally discovered at
birth, although diagnosis by prenatal ultrasound has been
reported. The lesion generally presents as a thin, transparent
membrane or as a dried cicatrix in place of normal skin; no
hair follicles are present. The vertex of the scalp is the most
common area involved. For small defects, the natural history
is that of epithelialization from the lateral edges, but larger
defects may not close and generally warrant treatment. The
mortality for extensive scalp defects may be as high as 20%
to 50%.19
1714 PART IX SPECIAL AREAS
The etiology of ACC is not well established. It has been
reported to be associated with other defects including limb
deformities, omphalocele, and clefts. Surgical treatment is
generally indicated immediately for the larger defects and ones
in which there is a bony defect. Bony involvement at the vertex
of the scalp places the patient at risk for hemorrhage from the
sagittal sinus and also mandates operative treatment. The de-
fect is generally repaired with flaps and/or skin grafts. Recently
ADM has been used for regeneration/reconstruction of the
tissue defect.20 Bony skull reconstruction is generally delayed
until 2 years of age; occasionally the osteogenic dura may
spontaneously generate new bone replacing the defect.19
Reconstruction of a normal hair pattern is accomplished via
tissue expansion of the scalp and subsequent coverage of
the affected area with local flaps at about 5 to 7 years of age.21
Congenital Melanocytic Nevus and Giant
Hairy Nevus
Congenital nevi are composed of large clusters of pigment
producing cells of melanocyte lineage that are clustered in
the dermis and produce pigment at birth. When compared
with the more common acquired nevi, congenital nevi are
larger, have increased cellularity, and lie deeper in the dermis.
The incidence of congenital nevi has been estimated to be 1 in
100 births for small lesions and 1 in 20,000 for large lesions.
After birth it is difficult to distinguish congenital from
acquired nevi even by histology. Congenital melanocytic nevi
are classified according to size: small, less than 1.5 cm in
diameter; medium, 1.5 to 19.9 cm; and large (giant), greater
than 20 cm.22,23
The potential for congenital nevi to transform into mela-
noma is the prime reason that these lesions come to surgical
attention. However, considerable uncertainty surrounds esti-
mation of the true risk for malignant degeneration. Rates of
1% to 31% have been reported for giant congenital nevi
(GCN), although these data may be even less reliable than
the range suggests because the total number of patients with
GCN is unknown. A metareview calculates an overall mela-
noma risk of 8.2% in GCN, which is high enough to make sur-
gical ablation a priority. The risk for malignant melanoma is
also immediate; 60% of observed GCN-related melanomas
occurred in the first decade of life, with several studies sug-
gesting that many malignancies occur by 3 years of age. Mel-
anoma has also been reported in small congenital nevi, and
melanoma rates of 2.6% to 4.9% have been estimated.24
Giant hairy nevi (GHN) are a special subset of congenital
nevi. These melanocytic cutaneous lesions are extensive and
encompass entire anatomic regions that may exceed 20% of
the body surface area (Fig. 130-2). These nevi are often hairy
with benign surface nodularity consisting of focal growths of
neuroectodermal tissue.25 When the head or upper part of
the trunk is involved, neurologic abnormalities have been ob-
served in up to 20% of patients. Neurologic findings of lepto-
meningeal melanosis, hydrocephalus, seizures, and myelopathy
in association with GHN are consistent with the developmental
origin of the neuroectoderm.25
GHN are disfiguring and easily identifiable at birth.26 Par-
ents of patients with GHN often seek treatment early. Some ne-
onates with GHN have been found to have malignant
melanoma shortly after birth, thus justifying early surgical ab-
lation. Serial resection of GHN lesions with the use of subcu-
taneous tissue expanders is indicated to improve aesthetic
appearance and diminish the risk for malignancy.27 The in-
creased compliance and elasticity of infant skin provide the
first opportunity to debulk a large portion of the giant nevus
in the first 6 months of life.28 At this time areas of nodularity or
increased pigment are resected. The timing of subsequent
resections is based on changes in appearance of the nevus
and the parents’ decision to proceed with complete resection.
Repeated excisions may expedite lesion removal with comple-
tion in early childhood. Complete excision of the nevus with
skin grafts results in scar and contracture deformities both at
the site of nevus resection and the skin graft donor site. The
use of sequential excision with expansion of normal adjacent
skin allows coverage of areas of nevus excision with normal
skin and does not cause additional scarring.
Nevus Sebaceous of Jadassohn
Sebaceous nevus (SN) is a solitary, well-circumscribed, alope-
cic plaque of epidermal origin. SN is commonly found on the
scalp, temple, or preauricular region and is sometimes diag-
nosed at birth. Histologic analysis in children reveals epider-
mal hyperplasia and hypoplastic hair follicles and sebaceous
glands. At puberty the sebaceous glands mature, and progres-
sion of the epithelial hyperplasia occurs. Progression to basal
cell carcinoma (BCC) has been reported in up to 6% to 22% of
adults.29
Controversy exists in the management of SN in children.
Early reports of BCC formation in 10% to 15% of adolescents
have been refuted by large retrospective studies that have
shown a low incidence of benign epithelioid lesions and very
low (0% to 0.8%) incidence of BCC in excised lesions from
children up to 16 to 18 years of age.29,30 However, there is still
a well-documented incidence of NS-related BCC in adults.
The management of choice remains excision, but delay until
later in adolescence is reasonable.29
Congenital Webs
Congenital webs (pterygia) are most commonly found in the
neck (pterygium colli). In the neck, thick folds of skin and
subcutaneous tissue extend from the mastoid process laterally
to the acromion. The lesions are generally bilateral and can
distort local structures including the posterior hairline, the
border of which is often displaced laterally on the neck, caus-
ing a short-neck appearance. Webs have been identified sim-
ilarly in the axillary folds and crural regions.
Congenital webbing may occur in isolation or in conjunc-
tion with other congenital deformities, however the etiology is
unknown. Certain regional deformities such as Klippel-Feil
syndrome can display this developmental lesion. Neck web-
bing is often a manifestation of Turner’s syndrome. The webs
can cause functional and/or aesthetic disabilities. Correction is
by lengthening with multiple Z-plasties generally between
ages 1 and 6 years.31
CONGENITAL DEFORMITIES
OF THE BREASTS
Embryology
The breast develops in the fetus from ectodermal tissue during
the fifth week of gestation. Bilateral mammary ridges extend
from the axillae to the groins. During the seventh to eighth
week of gestation, the mammary ridges undergo involution
 CHAPTER 130 CONGENITAL DEFECTS OF THE SKIN AND HANDS 1715

A B

FIGURE 130-2 A, Congenital giant hairy nevus of the trunk. The large
midline mass in this infant proved to be a malignant neural crest cell tumor
already present at birth. B, The malignant tumor was completely resected.
Multiple implanted tissue expanders have been inflated to facilitate wound
C coverage after excision of the nevus. C, Most of the nevus has been
excised.
1716 PART IX SPECIAL AREAS
in all areas except on the anterior thoracic wall and axillae.
After 10 to 14 weeks of gestation, the primordial breast tissue
penetrates the thoracic mesenchymal tissue and begins to
form glands. The mesenchyme envelops the breast and forms
layers in continuity with Camper fascia (outermost layer) and
a second layer separating the breast from the underlying tho-
racic musculature. Camper fascia surrounds the anterior
breast in all areas except just deep to the areola. A network
of connective tissue runs through the mammary gland forming
Cooper ligaments, which suspend the breast. Milk ducts begin
to form at about 20 weeks of gestation, and the glandular
tissue continues to develop until the end of puberty.32,33
Polythelia and Polymastia
Supernumerary nipples (polythelia) and supernumerary
breasts (polymastia) are relatively common congenital abnor-
malities with an incidence of approximately 0.2% to 2.5%
(polythelia) and 0.1% to 1.0% (polymastia) (see also
Chapter 61). Both of these deformities are observed to form
along the milk lines that track from the axillae to the groins.
Both supernumerary nipples and breasts form from a failure
of normal apoptotic regression of the mammary ridges during
gestation. Diagnosis of supernumerary nipples can often be
made at birth. The most common site for polythelia is just
inferior to a normal breast; the most common site for an acces-
sory breast is in the axilla.33
Supernumerary nipples and breasts typically present as an
aesthetic complaint but can also respond to hormonal changes
similar to a normal breast including enlargement during pu-
berty and milk production after pregnancy.34 Treatment of
these lesions is generally simple excision, but it is important
to consider that diseases that afflict the normal breast can also
affect supernumerary ones.35 Breast masses must be excluded.
Furthermore, several other anomalies have been associated
with polythelia and polymastia including a higher rate of
testicular tumors in boys.36 Other more controversial as-
sociations include genitourinary tract abnormalities and renal
cell carcinoma. 37,38
Congenital Absence of the Breast
Complete absence of one or both breasts is an extremely
rare anomaly that can occur in isolation or with a variety of
other developmental syndromes including Poland syndrome,
Ullrich-Turner syndrome, and AREDYLD syndrome (acrore-
nal field defect, ectodermal dysplasia, lipatrophic diabetes)
(Fig. 130-3). Bilateral congenital absence without other asso-
ciated ectodermal defects is extremely rare, with few cases
reported in the literature. Mendelian inheritance is possible
but has not been definitively established; teratogen exposure
has also been implicated. Due to the rarity of this condition,
optimal treatment has not been established. Successful use
of autogenous tissues including the use of abdominal or
buttock free flaps for construction of a breast de novo has been
reported.38a Tissue expansion and implant placement may
also be reasonable options.33,39
Tuberous Breast Disease
The breast anomaly known as the “tuberous breast” was ini-
tially described in 1976 by Rees and Aston for its likeness
to a tuber root (Fig. 130-4).40 Although there has been much
confusion in the literature about both the nomenclature and
the description of the tuberous breast and breast asymmetries
in general, the following facts are known: 88% of women pre-
senting for breast augmentation mammaplasty have breast
asymmetry and 29% have breast base constriction.41 How-
ever, the true tuberous breast is much less common and is de-
fined not simply by asymmetry or constriction but also by the
herniation of mammary tissue through a constricting fascial
ring deep to the areola. The true incidence is uncertain, and
there seems to be no heritable genetic component to the
disorder.33
The presence of the tight fascial ring causes the breast to
become asymmetric during pubertal development and ulti-
mately herniate through it, causing areolar enlargement. This
may be associated with hypoplasia of one or more quadrants,
with the deficiency being most common in the inferior pole.
The inferior mammary fold is often variably elevated.42
The tuberous breast often presents as an aesthetic concern
in the early teen years. Differentiation of this disorder from
more common asymmetries is essential because the treatment
is generally different. Nearly 90% of cases are bilateral, and
significant hypoplasia of the breast tissue is present in approx-
imately one quarter of patients.33
Treatment for the tuberous breast is somewhat controver-
sial, although several principles can be relied on. The patient
should not be operated on until late in the second decade when
breast development has finished. In nearly all cases, the herni-
ated breast tissue causes enlargement of the areola, mandating a
circum-areolar incision to reduce its size. This incision also al-
lows for access to the fascial ring, which has constricting bands
that must be released. Operative plans subsequent to these two
steps are less definitive, although local tissue rearrangement
is generally recommended with the establishment of a normal
position of the inframammary fold and a natural inferior pole
contour.43 For hypoplastic or asymmetric breasts, tissue ex-
pansion and/or implant placement can be a useful adjunct to
tissue rearrangement, but the use of an implant alone has been
shown to be unsatisfactory with abnormalities of the inframam-
mary crease being common.33,44,45
Gynecomastia (Infantile and Adolescent)
Gynecomastia, or the abnormal presence of breast tissue in
males, is a common but complex disorder with both congen-
ital and acquired features (Fig. 130-5). The overall incidence
has been reported to be 32% to 40% with a reported incidence
of palpable breast tissue of 65% in adolescent boys.46 Other
studies have disputed these data; a cross-sectional study of
3082 healthy boys (aged 10 to 19 years) in Eastern Europe
found a 3.93% prevalence of gynecomastia.47 Variations in es-
timates are likely to be related to varying thresholds for diag-
nosis of overt gynecomastia (grade IIb/III gynecomastia)
versus the presence of palpable breast tissue on the male chest
(grade I/IIa gynecomastia).
Although numerous individual cases of gynecomastia have
been identified, breast development in males generally occurs
secondary to an excess of estrogens relative to testosterone be-
cause of an absolute estrogen excess, a testosterone deficiency,
or receptor resistance. A full discussion of the causes of gy-
necomastia is beyond the scope of this section and is reviewed
elsewhere.48 Gynecomastia can be classified as congenital or
acquired, as well as physiologic or pathologic. In boys, most
cases are congenital and physiologic and occur in either the
neonatal period or during adolescence. In utero, exposure
to high maternal estrogen levels can cause the development
 CHAPTER 130 CONGENITAL DEFECTS OF THE SKIN AND HANDS 1717

A B

C D

FIGURE 130-3 Congenital absence of the breast. A, Anterior preopera-

tive view of a patient with absence of natural breast volume on the right
hemithorax. B, Lateral view. C, An expander has been inflated to establish
an adequate skin envelope for breast mound reconstruction. D, Postoper-
ative anterior view. The patient has a permanent saline implant in place to
establish the right breast mound. E, Lateral view. Note that preexpansion
of skin allows adequate projection and shape after the permanent implant
E is inserted.
1718 PART IX SPECIAL AREAS

A B

C D
FIGURE 130-4 Tuberous left breast with mild pectus excavatum. A, Preoperative anterior view of a congenital breast and central chest deformity. B,
Lateral view. Note the tuberous breast deformity. C, Postoperative release of periareolar congenital constricting bands and the use of an implant to es-
tablish the breast mound. D, Postoperative lateral view.

of mammary tissue (neonatal gynecomastia), which regresses
within a few weeks after birth. In adolescence, elevation of the
estradiol-testosterone ratio causes detectable gynecomastia in
up to 65% of boys. However, in most it is not noticeable with-
out a focused examination and most noticeable cases occur
during pubic Tanner stages 3 and 4.47 Nearly all cases regress
by the end of puberty.33 Pathologic gynecomastia in children
and adolescents is relatively rare but requires special attention,
especially in the case of Klinefelter syndrome, which is asso-
ciated with a 19.2- to 57.8-fold risk for breast cancer (unlike
gynecomastia in general, which has no elevated risk).49
Although gynecomastia in adolescents is typically self-
limited and resolves by the end of puberty, persistent gyneco-
mastia (present for more than 1 year and beyond puberty)
generally does not resolve without surgical treatment because
the abnormal breast inevitably becomes fibrotic.50 The gy-
necomastia for which a plastic surgeon is initially contacted
is usually an aesthetic complaint; however, this does not ob-
viate the need for a detailed history and physical examination.
Although gynecomastia in the general population and even
cases examined by a pediatrician are generally uncomplicated
and benign, there is good evidence that gynecomastia needing
 CHAPTER 130 CONGENITAL DEFECTS OF THE SKIN AND HANDS 1719

A B

C D
FIGURE 130-5 Gynecomastia. A, Preoperative anterior view of an adolescent with abnormal breast development. B, Lateral view. Note the excessive
breast tissue with the development of early breast ptosis. C, Postoperative view after the use of liposuction. D, Lateral view. A contour deformity caused by
overresection was avoided by combining direct excision and liposuction.

surgical treatment is a more complicated subset of the disor-
der. In particular, several large studies have shown an in-
creased incidence of testicular and nonmelanoma skin
cancer in males with gynecomastia requiring operative treat-
ment. The risk for testicular cancer is particularly elevated
in young men and has a standardized incidence ratio reported
at 5.82; 7% to 11% of patients with testicular cancer present to
surgical evaluation for gynecomastia.51,52 Examination of the
testicles and skin, as well as education regarding skin cancer,
is mandatory.
Poland Syndrome
In 1841 Alfred Poland, a medical student, reported some un-
usual anatomic findings on a cadaver, describing unilateral hy-
poplasia of the pectoralis musculature, upper limb, and hand.
The expression of the various components of this trait varies
1720 PART IX SPECIAL AREAS
widely. The incidence has been estimated to be 1:30,000 with
a predilection in females of approximately 2-3:1.53,54
The etiology of Poland syndrome (PS) has been theorized
to be due to obstruction of the subclavian blood supply during
the sixth to seventh weeks of gestation. During this time, the
ribs grow anterior and medially, depressing the subclavian ar-
tery into a U-shaped configuration. Tension on the arterial
wall may cause kinking and obstruction. The site of dimin-
ished blood flow may determine the tissues involved in the de-
fect, which may involve the thoracic wall, the pectoralis
muscle, breast, upper extremity, and hand. Vascular studies
have shown diminished blood flow in the subclavian artery
on the affected side. More proximal vascular involvement of
the subclavian or vertebral arterial systems has been shown
to give rise to the bony cervical anomalies found in Klippel-
Feil syndrome and the congenital palsies of cranial nerves
VI and VII in Moebius syndrome. The synchronous presence
of Poland, Klippel-Feil, and Moebius syndromes support a
common embryologic event (i.e., vascular accident) in their
genesis.55–58
The essential feature of PS is the absence of the sternal head
of the pectoralis major muscle. Other features are hypoplasia
or aplasia of the breast or nipple, deficiency of subcutaneous
fat and hair across the chest and axilla, and abnormalities of
the ipsilateral limb including shortening and brachysyndac-
tyly. Other muscles may also be involved to variable degrees
including absence of the pectoralis muscle (Fig. 130-6) and
hypoplasia of the serratus anterior, supraspinatus, latissimus
dorsi, and external oblique muscles. The thoracic cage itself
may be involved with absence of the anterolateral ribs with
possible lung herniation or symptomatic depressions in the
chest wall. The hand deformities can be more involved with
symphalangism occurring with syndactyly and hypoplasia
or absence of the middle phalanges. Detection of syndactyly
in a patient should make PS a consideration because 10%
of all cases of syndactyly have been reported to have features
of PS.53,59,60
Despite the hand deformities, the functional disability is
generally mild and related to the presence of syndactyly. In
milder forms of PS, treatment is primarily for aesthetic rea-
sons. For more severe presentations, particularly in which
the thoracic cage is involved, stabilization of the protective
chest wall is warranted first because all other chest reconstruc-
tions will be placed on top of this foundation. Minor chest wall
deformities are generally reconstructed with an ipsilateral
latissimus dorsi muscle flap. Major chest wall deformities re-
quire better stabilization with either bone or prosthetic mesh.
Split rib homografts and Marlex mesh have been used for
moderate and severe defects and are covered with a latissimus
dorsi muscle transfer. The transferred latissimus dorsi muscle
replicates the missing anterior axillary fold of the pectoralis
muscle and improves soft tissue coverage. Remaining major
contour irregularities may be treated using a custom alloplas-
tic implant.53,61
The reconstruction of the female breast in PS often pre-
sents a challenging problem. In severe cases the breast recon-
struction generally occurs as a second-stage procedure late
in adolescence, years after stabilization of the chest wall.
The placement of an implant behind a latissimus dorsi mus-
cle transposition has been effective. The successful use of
free-tissue transfer has also been reported. In cases where
the cutaneous envelope is hypoplastic, tissue expansion early
in puberty will augment the soft tissues for later final repair
and may guard against the psychologic distress of an
absent breast.62–65
HAND
Embryology
Arm buds appear at 31 days, the fourteenth stage in embry-
onic development, coincident with the formation of the mar-
ginal vein beneath the surface ectoderm. The paddle-like
hand segment forms on approximately day 33. Finger rays
are evident at 41 days but do not begin to separate until
day 47. Interdigital zones destined for apoptosis determine
the separation of the fingers. The median artery supplying
the embryonic hand is replaced by the radial and ulnar arter-
ies between the fifth and sixth weeks. The reduced median
artery maintains only the median nerve and interosseous
blood supply. Dorsal and ventral blastemas differentiate into
dorsal and volar forearm musculature, respectively, from the
fifth to the eighth week, and are formed superficially and
deeply. Intrinsic hand muscles arise from five embryonic
layers; the deep hypothenar and thenar muscles in the sev-
enth week are the last to separate. Condensation and then
chondrification of the mesenchyme begins in the fifth week,
ending with chondrification of the phalanges at week 6 and
sesamoids at week 7. The eighth week marks the end of em-
bryogenesis; much of the final form is in place. During weeks
4 to 8 of gestation, the embryo is at the highest risk of hand
anomalies. Growth dominates the rest of fetal development,
during which the fingernail fields develop and cartilaginous
bones ossify.66
Incidence
The estimated incidence of upper limb deformities has been
studied in a longitudinal study of 261,914 live births in Swe-
den with a recorded incidence of 21.5 per 10,000 live births.67
This result confirmed an earlier study from Western Australia,
where an incidence of 19.8 per 10,000 was reported without
an observed difference between Caucasian and non-Caucasian
individuals.68 The Centers for Disease Control and Prevention
reported the following rates per 1000 live births: 0.8 for trans-
verse deficiencies, 0.9 for syndactyly, 0.2 for polydactyly in
Caucasians, and 0.12 for polydactyly in black persons.69 In
40% to 50% of the cases, no clear cause can be determined.
It is estimated that environmental factors account for 10%
of limb anomalies. Rubella, cytomegalovirus (CMV), toxo-
plasma, and varicella in the first trimester have been linked
to longitudinal deficiencies. Chemicals affecting limb develop-
ment include thalidomide, ethanol, phenytoin, and warfarin.
Many genetic syndromes (of Mendelian inheritance patterns
or otherwise) have been associated with skeletal anomalies.
Brachydactyly, clinodactyly, syndactyly, and polydactyly have
been noted in trisomy 21; camptodactyly in trisomy 18; and
polydactyly in trisomy 13.
Congenital Hand Deformities: Classification
and Treatment
The International Federation for Surgery of the Hand (IFSH)
has established a seven-category classification system for hand
deformities in 1976 (Table 130-1).70
 CHAPTER 130 CONGENITAL DEFECTS OF THE SKIN AND HANDS 1721

A B

C D

E F
FIGURE 130-6 Poland syndrome on the left side of the chest with an associated asymmetric pectus carinatum deformity. A, Preoperative anterior view.
Note the absence of the left pectoralis muscle. B, Lateral view. C, Posterior view of the chest with the patient testing for the presence of the left latissimus
dorsi muscle. Note the lateral edge of the muscle is visible with isotonic contraction. D, Postoperative anterior view. Note that functional transfer of the
latissimus dorsi muscle to the anterior hemithorax provides contour correction and maintains latissimus dorsi function. E, Postoperative lateral view. The
incision is concealed beneath the upper extremity. Note that the contour of the latissimus dorsi on the anterior aspect of the chest simulates the absent
pectoralis muscle without contour deformity at the posterior chest donor site. F, A postoperative lateral view with the upper extremity abducted reveals the
incision used for muscle harvest.
1722 PART IX SPECIAL AREAS
TABLE 130-1
International Federation for Surgery of the Hand Classification
System for Congenital Hand Anomalies
1. Failure of formation
2. Failure of differentiation
3. Duplication
4. Overgrowth
5. Undergrowth
6. Congenital constriction band syndrome
7. Generalized skeletal abnormalities
Failure of Formation Failure of formation is further divided
into transverse and longitudinal arrest of development. Fail-
ure of formation has an incidence of 3.9 per 10,000 live births
and contributes to 17.6% of all congenital hand anomalies.67
Transverse deficiencies are congenital amputations, the most
severe of which occur at the shoulder level and result in ame-
lia. The short below-elbow defect, a common transverse defi-
ciency, ends at the upper forearm. Prosthetic fitting should
begin in infancy. Wrist disarticulation defects occur more of-
ten in girls than in boys. In this condition, the skeleton is ab-
sent beyond the distal radial and ulnar epiphyses, but
pronation and supination abilities are generally maintained.
In bilateral defects the Krukenberg procedure may be indi-
cated on the dominant arm with a prosthesis worn only on
the nondominant arm. The Krukenberg procedure separates
the forearm bones into opposing prehensile forceps. Progres-
sive length reduction defines longitudinal deficiencies, which
may be preaxial (radial), postaxial (ulnar), central, or complete
(phocomelia).71
Radial Deficiencies Radial deficiencies variably affect
the radius, scaphoid, trapezius, trapezoid, and thumb, as well
as associated musculotendinous, neural, and vascular struc-
tures. In the classic radial clubhand, the forearm is short,
the hand is radially deviated, and the thumb is absent or float-
ing. The defect begins in the first weeks of fetal life. Radial
shaft development coincides with the development of the car-
diac septum, which may explain the association of radial ray
defects with the atrial septal defect found in Holt-Oram syn-
drome. Radial defects are also associated with vertebral de-
fects, imperforate anus, esophageal atresia, radial and renal
dysplasia syndrome (VATER or VACTERL), and blood dyscra-
sia such as Fanconi anemia and thrombocytopenia with ab-
sent radii (TAR) syndrome. Treatment begins shortly after
birth; traditional methods stretch the hand (centralization)
over the ulna by serial castings or distraction lengthening. Free
fibula transfers have also been investigated.71
Ulnar Defects Ulnar ray deficiencies are rare and can be
associated with defects of the fourth or fifth finger, elbow, hu-
merus, or shoulder girdle (PS). Unlike the radius, which func-
tions as a buttress for the wrist, the ulna may be deficient
without any accompanying ulnar wrist deviation. Severity varies
from hypoplasia to partial or total defects with or without humer-
oradial synostosis to ulnar deficiency with amputation at the
level of the wrist. Prostheses such as an elbow disarticulation
device can improve limb use. In selected patients, fusion of
the ulnar remnant to the radius provides stability at the elbow.71
Thumb Defects The management of thumb defects de-
pends on their severity. In total aplasia of the thumb, pollici-
zation of the index finger is the most successful method of
achieving functional opposition. Toe-to-thumb transfer is
not successful because recipient nerve and tendon structures
are absent and the brain lacks the motor and sensory homun-
culus for the thumb. Short thumbs can be lengthened by dis-
traction osteogenesis. Widening of the first web space and
tendon transfer to establish opposition may improve function.
Floating thumbs are generally amputated and opposition is ac-
complished by pollicization of the index finger.71
Central Longitudinal Deficiency Central longitudinal
deficiency, also known as cleft hand or split hand, applies to de-
fects of the central (second, third, or fourth) rays. Typical
deficiencies involve partial or complete loss of the central
digit(s). Atypical deficiencies are syndactylous (the central fin-
ger remnants fuse to the second or fourth ray) or polydacty-
lous. The polydactylous type contains supernumerary bony
elements. Function determines treatment; appearance is a sec-
ondary consideration. Resection of the remnant third metacar-
pal assists closure of a deep cleft. Osteotomies correct rotatory
deformities of the adjacent fingers. Thumb web space recon-
struction is critical to overall hand function.71
Failure of Differentiation Failure of differentiation in-
cludes soft tissue and skeletal failure of the separation of parts,
as well as congenital tumors. Failure of differentiation is the
most common congenital hand anomaly and has an incidence
of 10.5 per 10,000 live births. It contributes to 47.2% of all
congenital hand anomalies.67
Syndactyly Synostosis of the phalanges (symphalan-
gism) is uncommon and usually occurs with syndactyly. Syn-
dactyly itself is common and most often occurs between the
middle and ring fingers. It has an estimated incidence of
5 per 10,000 live births.72 True failure of differentiation is
distinguished from the brachysyndactyly (short, webbed fin-
gers) of PS and from acrosyndactyly with fusion of the distal
digits secondary to a constriction band. Syndromes featuring
syndactyly include Apert (acrocephalosyndactyly), Chotzen
(cephalodactyly), Pfeiffer, and Golz; oculodentodigital; and
trisomy 13.
Digits with acrosyndactyly should be separated before 1 year
of age if the digits are of discrepant lengths or there is fusion of
the phalangeal bones. In the absence of these features, separa-
tion can be performed at 2 to 3 years. The complexity of an
Apert “mitten” hand presents a challenge to provide a func-
tional hand with both prehensile function and fingertip sensa-
tion (Fig. 130-7). To this end, a three-finger hand is often
preferable to four fingers and a thumb. The thumb and small
fingers are released before 1 year of age. The remaining digits
are separated within 6 to 9 months. Another feature of Apert
syndrome is radioclinodactyly of the thumb (radial curvature
of the thumb). Thumb osteotomy and bone graft are per-
formed at 4 to 7 years of age, along with Z-plasty of the web
space between the thumb and the index finger.73
Congenital Flexion Deformities (Arthrogryposis) Failure
of soft tissue differentiation can lead to congenital flexion de-
formities. In the clasped thumb or thumb-in-palm contrac-
ture, the extensor pollicis longus, extensor pollicis brevis,
and abductor pollicis longus muscle and tendon units are
absent or attenuated. Splinting and manipulation begin in
infancy, and surgical correction is delayed until 3 years of
age. The clasped thumb must be differentiated from the
trigger thumb, wherein a flexor pollicis longus nodule proxi-
mal to the A1 pulley interferes with interphalangeal and
 CHAPTER 130
A and radially deviated, and the thumb and fingers flexed and
B Splinting and manipulation begins in infancy. If the defor-
C more often than preaxial thumb duplication. Central duplica-
D consists of rotational osteotomy, tendon transfer (e.g., Huber
FIGURE 130-7 A, Infant with Apert syndrome afflicted with the typical
acrosyndactyly deformity of the hand. B, Complex bilateral acrosyndactyly.
C and D, Postoperative appearance after release of syndactyly.
metacarpophalangeal extension. Because of a 30% spontane-
ous resolution rate, surgical trigger finger release is recom-
mended for contractures that persist beyond 3 years of age.74
Clinodactyly most often affects the middle phalanx of the
little finger and is seen in many syndromes (e.g., Treacher-
Collins and Silver-Russell syndromes, orofaciodigital and ocu-
lodentodigital dysplasia). Camptodactyly refers to congenital
proximal intraphalangeal flexion contracture, usually of the
little finger, and is associated with such syndromes as Marfan,
popliteal pterygium, and orofaciodigital and oculodentodigital
CONGENITAL DEFECTS OF THE SKIN AND HANDS 1723
dysplasia. Mild variants of camptodactyly can be found in up
to 1% of the population.75 If disabling or deforming, the con-
tracture is released by dividing the flexor digitorum superficia-
lis tendon and lengthening the volar skin with a full-thickness
skin graft. In more severe cases arthroplasty or arthrodesis is
performed. Arthrogryposis multiplex congenita features mul-
tiple congenital nonprogressive joint contractures and affects
all extremities and the spine. In the upper extremity the shoul-
der may be flexed and internally rotated, the elbow flexed or
extended, the wrist flexed and ulnarly deviated or extended
ulnarly deviated.74
Arthrogryposis isolated to the hand manifests as the “wind-
blown hand,” with metacarpophalangeal flexion and ulnar de-
viation caused by malformation of the palmar fascia and
retaining ligaments, with or without involvement of tendons
and capsules. Thick, fibrous subcutaneous bands and short-
ened digitopalmar skin contribute to this malformation. Sec-
ondary metacarpal head deformation occurs with time. The
thumb is flexed, and abduction is limited by a retracted volar
carpal ligament and carpal skin.
mity progresses, as often happens during the adolescent
growth, operative intervention is indicated. Corrective proce-
dures include release of the metacarpophalangeal joint capsule,
ligament, and intrinsic tendon; relocation of the extrinsic ten-
don; metacarpal osteotomy; and division of palmar skin and
fibrous bands followed by skin grafting for coverage.74
Duplication of Parts Duplication of parts has an incidence
of 5.9 per 10,000 live births and contributes to 26.5% of all
congenital hand anomalies.67 Duplication of parts occurs
early in embryonic development. Polydactyly is a common
hand malformation. Postaxial little finger duplication occurs
tions are rare. A flail, a poorly attached appendage, should be
amputated early. Operative correction of preaxial polydactyly
is sometimes delayed until 12 months of age, when functional
evaluation is possible to identify the dominant (usually ulnar)
thumb to be preserved. The collateral ligaments, even ten-
dons and bony elements, of the ablated digit are salvaged
for reconstruction of the retained digit. Alternatively, the
Bilhaut-Cloquet procedure consisting of excising a central
wedge from adjacent parts of both thumbs and then joining
the remaining lateral portions can be performed. If the thumb
resembles an index finger and cannot be opposed, treatment
transfer of the abductor digiti minimi across the palm), and
widening of the first web space.76
The mirror hand possesses seven or eight digits, one or two
index with two ring and middle fingers, two ulnae, absent ra-
dius, absent thumb, and weak or absent extensor tendons.
The hand loses its ability to pinch or grasp. An opposing digit
needs to be constructed, and excess fingers should be dis-
carded; the filleted skin is used to create a wide web space.
Wrist arthrodesis may be required in early adolescence. Elbow
motion may be improved by excising the olecranon off of one
of the ulnae.76
Overgrowth Congenital overgrowth anomalies are rare with
an incidence of 0.4 per 10,000 live births and contribute to
1.7% of all congenital hand anomalies.67 Digital gigantism
1724 PART IX SPECIAL AREAS
involves the overgrowth of bone and soft tissues. Classification
has been difficult due to the wide range of etiologies for the
overgrowth phenotypes and is beyond the scope of this chap-
ter.77 Classic type I macrodactyly is associated with lipofibro-
matous hamartoma of the median or (less frequently) the
ulnar nerve. Phalangeal involvement is constant, and metacar-
pal involvement varies. Soft tissue overgrowth can affect skin
and lymphatic or nerve structures, although flexor tendons
and blood vessels may be spared. Gigantism of several adjacent
digits is more prevalent than single-digit macrodactyly. Median
nerve hamartoma can compress structures in the carpal tunnel.
Type II macrodactyly is part of von Recklinghausen disease
(neurofibromatosis). The distribution follows the path of a
major peripheral nerve, most frequently the median nerve.
Contrary to types I and II, type III macrodactyly (hyperostotic)
is not associated with nerve enlargement but does follow the
course of the sensory branches of the median nerve. Treatment
has included epiphysiodesis, epiphyseal arrest, epiphyseal
plate excision, multiple defatting procedures, longitudinal
phalangeal osteotomies, soft tissue debulking, excision of distal
nerves, and partial or ray amputation. Skin flap necrosis is not
an uncommon complication after repair of macrodactyly.78
Undergrowth Undergrowth malformations vary in the
structures affected and the degree of the hypoplasia. Under-
growth anomalies have an incidence of 0.7 per 10,000 live
births and contribute to 3.1% of all congenital hand anoma-
lies.67 Brachydactyly (shortened digits) is commonly seen in
association with syndromic disorders, generally of autosomal
dominant transmission. Ectrodactyly is the complete absence
of phalanges or metacarpals. Short metacarpals are uncom-
mon and often go unrecognized until the adolescent growth
spurt; associated syndromes include pseudohyperparathy-
roidism, Turner, and cri-du-chat (chromosome 5p deletion).
Treatment is usually unnecessary, but lengthening the affected
ulnar metacarpals does increase palm size and improve grip.
Lengthening procedures use either bone grafting or distrac-
tion osteogenesis.79 Short phalanges are the major cause of
brachydactyly. The middle phalanx of the little finger is the
hand bone, the length of which varies the most, particularly
in girls. Other than osteotomies to correct deviations, there
is little role for operative intervention. Phalangeal lengthening
procedures rarely improve function.80
Constriction Band Deformities Constriction band defor-
mities are rare (incidence of 0.3 per 10,000 live births,
1.7% of congenital hand anomalies) and occur during fetal
rather than embryonic development.67 The predominant the-
ory is that amniotic bands wrapped around a limb cause local
ischemia and necrosis. An alternate explanation proposes an
intrinsic defect wherein hemorrhage or a local defect in the
limb leads to necrosis of the superficial tissues. The necrotic
area heals as a circumferential scar, and amniotic bands appear
secondary to the limb injury. The distal regions of the extrem-
ities are more often affected than are proximal regions. Com-
pression neuropathy distal to the constriction band scar has
been reported. In the digits, acrosyndactyly results from tissue
necrosis and fusion. Separations between the digits proximal
to the fusion and distal transverse grooves or amputations
characterize acrosyndactyly. For most constriction rings, sim-
ple or staged excision and Z-plasty suffice. Broad rings may
require local flap coverage of large defects. Treatment of fused
digits is individualized; separation may include cross-finger
flaps, completion amputation, distraction lengthening, or
free-toe transfer.81
Summary
------------------------------------------------------------------------------------------------------------------------------------------------
A wide spectrum of anomalies of varying severity can
result from the congenital absence or alteration of tissues.
Any anatomic region of the body can be affected. In determin-
ing the indications and timing for surgical correction, the sever-
ity of functional impairment, the degree of disfigurement, and
the social pressures associated with the malformation must be
considered. A sound plan of management relies on accurate di-
agnosis and analysis of the defect. Almost all defects can be sys-
tematically evaluated by the degree of malposition of normal
structures, absence of tissues, and aberrations in tissue compo-
sition. Surgical correction can then be formulated to specifically
address each component of the deformity. Using a systematic
approach simplifies and optimizes a safe treatment plan for
even the most complex malformations.
The complete reference list is available online at www.
expertconsult.com.
 the shoulders”2 (Fig. 131-2). They lived together for 34 years.
When Mary became ill and died, Eliza was advised to be
separated but absolutely refused, saying, “as we came together
we will also go together.” She died 6 hours later. They amassed
a vast fortune and bequeathed to the parish of Biddenden
20 acres of land with the expressed wish that proceeds from
rent should be spent on the distribution of cakes, bearing
an impression of their images, to be given to the poor each
Easter Sunday—a practice that endures to the present day.
It is impossible that they were joined as depicted in numerous
illustrations at the hips and shoulders. The more plausible
explanation was that they were ischiopagus or parapagus
twins.
The Scottish Brothers were born near Glasgow in 1490.3
They were two complete individuals above the waist but the
lower half of their bodies was fused—one set of genitalia
and two legs. They were taken to the court of King James
IV, who ordered that they be carefully brought up and edu-
cated at court. They learned to sing, one tenor and the other
treble bass; played various musical instruments; and became
fluent in several languages. They often differed in opinions
and sometimes quarreled. They died in 1518, aged 28 years.
Geoffrey-Saint-Hilaire cited an example of twin girls, born
CHAPTER 131 in 1495, joined at the forehead, causing them to stand face to
face: “When one walked forward, the other was compelled to
walk backward.” They lived to the age of 10 years, and when
one died an unsuccessful attempt was made at separation.4,5
Conjoined Twins In the sixteenth century, Ambroise Paré collected examples
of six sets of conjoined twins and was one of the first to classify
the different varieties of conjoined twins.6
Lewis Spitz, Edward M. Kiely, and Agostino Pierro The Isle-Brewer xiphopagus twins, Priscilla and Aquila,
were born in 1680 joined “from the navel up to a point just
below the nipples.”7 The vicar at their christening believed
that the monstrous birth was a sign of impending evil. They
were visited daily by hundreds of people keen to view “the
Conjoined twins can be subdivided into (a) symmetric monstrous work of Nature and admire so great a piece of
conjoined twins, which form the main bulk of this chapter, curiosity.” They were abducted by Henry Walrond and
and (b) heteropagus or parasitic twins. displayed but died in 1683.
The first successful separation of conjoined twins took place
Symmetrical Conjoined Twins in 1689. The surgeon, Johannes Fatio, separated omphalopagus
------------------------------------------------------------------------------------------------------------------------------------------------ twins in Basel, Switzerland, by “tracing the umbilical vessels to
HISTORY the navel, where he tied them separately. He then transfixed
and tied the bridge between the two infants with a silken cord
Although there are numerous examples of figurines depicting and cut the isthmus.” The ligature fell off on the ninth postoper-
conjoined twins dating back to classical times, particularly ative day, and both children survived. Koenig,8 an observer at
among Aztec art and ancient Turkish culture, it is doubtful the procedure, published the case as his own and to him is
whether these were factual representations. credited the first successful separation.9,10
The earliest example is a 17-cm marble statuette portraying Table 131-1 lists the early history of conjoined twins dating
parapagus twins, “the double goddess,” dating from the sixth back from the mid nineteenth century.
millennium BC. The statue of sisters of Catalhoyuk is housed The most celebrated pair of conjoined twins was Chang
in the Museum of Anatolian Civilization in Ankara, Turkey.1 and Eng Bunker, born on a riverboat in Siam (Thailand) in
Another early example that appears more authentic is a stone 1911. They were joined at the xiphisternum by a short band
carving of pygopagus twins dating back to 80 BC, discovered that stretched so that they were eventually able to stand side by
in Fiesole and kept in the San Marco Museum in Florence, side (Fig. 131-3). Their father was a fisherman, and the family
Italy (Fig. 131-1). lived on a floating houseboat on the Mekong River. They be-
The earliest attempt at separation of conjoined twins took came proficient swimmers, and their peculiar movement
place in Kapadokia, Armenia, around 945 AD. When one of the attracted the attention of Robert Hunter, a traveling Scottish
male ischiopagus twins died at the age of 30 years, an attempt merchant, who eventually persuaded their mother to permit
was made to save the surviving twin by separating him from him to take them to the United States, where they were exhib-
his dead brother, but he died 3 days later.1 ited by Captain Coffin and later by the showman Phineas
The first well-documented case is that of the Biddenden Barnum. They married sisters, lived in separate houses in
maids born in Kent in 1100 AD and “joined at the hips and North Carolina spending 3 days in each house alternately,
1725
1726 PART IX SPECIAL AREAS

and had 21 children between them. Physiologically and psy-

FIGURE 131-1 Stone carving of pygopagus conjoined twins dating
back to 80 BC.
chologically they were totally different—Chang was smaller
and more feeble; what Chang liked to eat, Eng detested;
Eng was good-natured, Chang cross and irritable; Chang
drank heavily, but Eng appeared unaffected by the alcohol.
They wished to be separated but could not persuade surgeons
to carry out the operation, which was considered to be too
hazardous at the time. In retrospect, according to the autopsy
findings, the procedure would have been quite straightfor-
ward because the connecting bridge between them contained
only a small tongue of liver.11–13 They lived together for 63
years and died in 1874.14,15
In 1840 an African village was raided, and many of the res-
idents were sold to a slave trader and transported by ship to
America. During the voyage, a woman gave birth to xiphopa-
gus twins by cesarean section by the naval surgeon Brito y
Boin, who surgically separated them. The mother died the
next day, and one of the twins died 2 days later. Another slave
woman took the surviving twin and raised him. He had keloid
scars removed at age 8 years and was still reported as being
alive at the age of 66 years at that time.16
In 1860 a physician separated his own omphalopagus twin
daughters. The incision was closed with three through-and-
through sutures. One twin died, but the other was well 5 years
later.17
The “first successful separation of conjoined twins” was
performed at The Military Families Hospital in Portsmouth,
England, in 1912. One twin survived, but the other died of
pneumonia at the age of 4 months.18
The Tocci Brothers, born in 1877, were extensively united
parapagus conjoined twins. They also wished to be separated,
but the extent of their union dictated that surgery at the time
was clearly not feasible. They lived together for 63 years and
died in 1940.19,20 Other conjoined twins have declined any
suggestion of separation. Lori and Reba Schappell, cranio-
pagus twins, born in 1961, never wanted to be separated,
stating “God made us this way and He had a purpose for us
and you do not ruin what God has made.”15
In 1990 Hoyle21 reviewed all attempts at surgical sepa-
ration of conjoined twins, performed successfully and unsuc-
cessfully, through 1987. Of a total of more than 600
publications, separation was attempted in 167 cases. The sur-
vival rate increased dramatically in the most recent decade
before 1987. He concluded that “the current excellent out-
comes (for separation) suggest that separation should always
be considered with rare exceptions.”

ETIOLOGY AND EMBRYOLOGY22,23

In the absence of an experimental model, the exact etiology of
conjoined twins is unknown. The uncertainty between fission
and fusion theories remains. Because it seems likely that fis-
A sion of a single fertilized ovum occurs in the same manner
each time, the wide variety of types of conjoined twinning
make fusion the more likely mechanism. Unlike identical
monovular twins, conjoined twins often have remarkably dif-
ferent personalities. This difference in personality is clear from
the early months of life and remains unexplained. The infer-
ence may be drawn that these twins may not always result
B
from monovular pregnancies. Regardless of the mechanism,
FIGURE 131-2 A and B, Broadsheet of Biddenden maids born in Kent twins are always joined homologously—chest to chest, pelvis
in 1100 AD. to pelvis, etc.—and are always the same sex. Spencer has
TABLE 131-1
History of Conjoined Twins to 1851
Yr Type Sex Details Outcome Origin
6500 BC Dicephalic Female Marble statuette “Double goddess” Unknown Turkey1
500-800 BC Dicephalic ? Various clay figures Unknown Mexico5
Figure 131-1
80 BC Pygopagus ? Stone carving Unknown Fiesole,
Florence, Italy
375 Dicephalus Male One twin ate and slept, but the other did Died at age 2 Castle of Emaus,
neither England19
945 Thoracopagus Male Attempt at separation after one twin died Died at age 30 Kappadokia,
aged 30 years Armenia1
961 Dicephalus Female When one laughed, ate and talked, the Unknown—both died in the space of Gascony,
other wept, fasted, and was silent 2 days France19
1100 ? Ischiopagus Female Joined at hips and shoulders Lived for 34 years—when one died, Biddenden, Kent,
the survivor refused separation England2
1490 Dicephalus Male Scottish brothers—brought up in the Musical and linguistic—lived to age Glasgow,
court of King James IV of Scotland 28 Scotland19
1495 Craniopagus Female Joined at the foreheads—when one Lived to age 10—when one died, Worms,
walked forward, the other walked attempt at separation failed Germany5
backward
1533 Omphalopagus Female Umbilicus to stomach Died after 8 days Hispaniola57
Liver
1552 Pygopagus ? Male Three lower limbs ? Middleton
Stony, Oxon.
England58,59
1573 Parapagus Male Described by ? Paris
(1546) Ambroise Paré in Verona, Italy
Pygopagus Female 1573 Paris
(1475) Angers, France
Ischiopagus ? Tours, France6
(1570)
Thoracopagus Female
(1572)
Craniopagus Female
(1569)
1576 Ischiopagus Male Born in the Jewish ghetto of Venice Impossible to circumcise—threat of Venice, Italy5,60
anti-Semitism
Died soon after birth
1664 Pygopagus Female Waterman—Martha and Mary Lived for 2 days Normal female triplet Somerset,
England61,15
1668 Dicephalus Female “Fair maidens of Foscott” Apparently lived to a “mature age” Somerset,
England19
1670 Thoracopagus Female One twin had esophageal Died at birth Plymouth,
atresia England62
1680 Pygopagus Female “The Isle-Brewers Twins” Died at age 3 Somerset,
England7
1680 Craniopagus Female Voyoen-Pieternella and One slept while the other was awake, Bruges,
Barbara crying, and Belgium5,63
eating
1689 Omphalopagus Female Connecting bridge ligated at 2 days old— Both survived Basel,
both survived Surgeon—Fatio Switzerland9
Reported by Koenig
1701 Pygopagus Female “Hungarian Sisters” Lived to age 22 Szony,
Helen—active and intelligent Hungary5
Judith—hemiplegic
1706 Parapagus ? Single trunk, 4 arms 30 minutes Hitchin,
and legs England58
1783 Craniopagus Male “Two-Headed Boy of Died after being bitten India64
Bengal”—separate brains by a cobra at age 4 yr
1811 Xiphi- Male Connecting bridge contained liver only Lived for 62 yr Mekong, Siam
omphalopagus Married sisters—had 21 children (Thailand)15,14
1829 Parapagus Female Parodi—Ritta (weak, cyanotic) and Died at 8 months old Ritta—severely Sardinia,
Christina (healthy, voracious appetite) deformed heart—2 superior vena Italy15
cavas
1848 Ischiopagus Female “Martha and Maria” Lived 10 days Copenhagen,
Denmark15
1851 Pygopagus Female “Millie and Christine” McKoy. Lived to age 61—died 1912 North Carolina,
Walked at 12 mo USA15
1728 PART IX SPECIAL AREAS

With the advent of routine prenatal ultrasound, the early

diagnosis of the nature of the problem will become evident
early in gestation and elective termination will be an option.
Female twins predominate in the ratio of 3:1.

CLASSIFICATION23–25
Conjoined twins are classified on the basis of the union’s site,
with the suffix pagus meaning fixed or fastened. The twins can
have four (tetrapus), three (tripus), or two (bipus) legs. The
classification of conjoined twins, limited to eight types with
approximate percentages of frequency, is summarized in
Table 131-2.
1. Thoracopagus: The twins lie face to face and share the ster-
num, diaphragm, upper abdomen wall, and liver and have
an exomphalos (Fig. 131-4). In the majority of cases they
share the pericardium (90%) and heart (85%). They may
have a common small intestine (50%), which joins at the
duodenum and separates at the ileum; the biliary tree
can be joined in 25% of patients. There may be associated

TABLE 131-2
FIGURE 131-3 Portrait of Chang and Eng from the Royal College of
Characteristic Features of Different Conjoined Twins
Surgeons of England, London.
Type of Fusion Incidence Extent of Shared Structures
Union
proposed that union occurs at sites where ectoderm is absent
or programmed to disrupt or fuse. The fusion theory suggests Ventral (87%)
that two separate embryonic discs from a monovular preg- Cephalopagus 11% Top of head
to umbilicus
nancy lie on the surface of a single yolk sac. At around the
Thoracopagus 19% Thorax Liver 100%
third week of pregnancy, fusion occurs at sites where the
Upper Pericardium 90%
ectoderm is not present or where it is disrupted. Ectoderm abdomen
is absent over the precursors of the septum transversum Conjoined Cardiac defects 75%
and heart. Fusion at this level results in thoracopagus and hearts 85%
omphalopagus twins. Ectoderm disrupts at the sites of the Upper intestine 60%
oropharyngeal and cloacal membranes and fuses at the edge Biliary tree 17%
of the embryonic disc. Union at these sites results in cepha- Omphalopagus 18% Upper Liver 90%
lopagus, ischiopagus, and parapagus twins. Dorsal union— abdomen
craniopagus, rachipagus, pygopagus—results in each twin Separate Upper foregut 16%
having its own umbilicus and separate abdomen. The other hearts Cardiac 25%
types of union are considered ventral and usually share a sin- Ischiopagus 11% Cloacal Pelvic bones 100%
gle cord and peritoneal cavity. These cords will frequently membrane Lower
gastrointestinal tract
have more than three vessels. Associated anomalies are com- 70%
mon and predominantly affect the structures that are joined. Genitourinary 50%
The patterns of abnormality encountered depend on the type Parapagus 28% Cloacal Cardiac 75%
of union, and each of the eight types of twinning has a range of membrane Intestine 100%
associated anatomic defects. Liver 100%
These abnormalities may preclude extrauterine survival Genitourinary 100%
and may render separation impossible. According to Spencer,
Dorsal (13%)
thoracopagus twins always have a single heart with multiple
Craniopagus 5% Cranial Skull, venous sinus,
chambers and separation is rarely an option. Others have been neuropore and meninges 100%
less rigid in their definition of thoracopagus, and there Cerebral cortex 37%
have been well-documented cases in which separation of Rachipagus 2% Neural tube Vertebral column
thoracopagus with separate hearts has been successful. Pygopagus 6% Caudal Sacrum and coccyx
neuropore 100%
Lower
INCIDENCE gastrointestinal tract
25%
The frequency of conjoined twins has been estimated at Genitourinary tract
1:50,000 pregnancies, but because up to 60% of these twins 15%
succumb in utero the true incidence is 1:250,000 live births.
 CHAPTER 131 CONJOINED TWINS 1729

FIGURE 131-4 Thoracopagus conjoined twins sharing a heart.

FIGURE 131-5 Omphalopagus.

cardiac anomalies such as ventricular septal defect, atrial
septal defect, and tetralogy of Fallot.
2. Omphalopagus: The heart is never fused; the liver is joined
in 80% of cases, and there is an exomphalos (Fig. 131-5).
The stomach and proximal small bowel are usually
separate, and each twin has a rectum. In up to one third
of omphalopagus twins, the intestine usually joins at the
Meckel diverticulum, the terminal ileum and colon are
shared, and a dual blood supply may exist. There is usually
no union of the genitourinary tract.
3. Pygopagus: The twins are joined dorsally, sharing the sacro-
coccygeal and perineal regions (Fig. 131-6). They face
away from each other—share the sacrum, coccyx, and part
of the pelvic bones. The spinal cords are usually separate.26
Twenty-five percent share the lower gastrointestinal tract
and have a single anus and one or two rectums. In 15%
of cases there is a single bladder. There is an increased in-
cidence of vertebral anomalies including hemivertebrae,
hemisacral agenesis, and thoracic anomalies.27 Although
the pelvic conjunction is fundamentally different than in
ischiopagus twinning, the types are similar insofar as nu-
merous other associated orthopaedic anomalies have been
reported in association with pelvic conjunction such as hip
subluxation or dislocation, congenital vertical talus, talipes
equinovarus, Sprengel shoulder, and scoliosis. There can
also be a variable degree of spinal and cord fusion.
Although there may be only one anus and rectum, the re-
mainder of the intestines are usually separate. The upper
bodies are not fused, and there are four arms and four legs.
4. Ischiopagus: The twins may lie face to face or end to end
(Fig. 131-7). They have two sacra or two symphysis pubis.
They may share the lower gastrointestinal tract (70%) and/
or the genitourinary tract (50%) and may have crossing FIGURE 131-6 Pygopagus.
1730 PART IX SPECIAL AREAS

faces (janiceps) facing away from each other; one face may
be rudimentary. These twins are terminated or die in utero.
They are nonviable.
7. Rachipagus: The twins generally have vertebral anomalies
and neural tube defects.
8. Parapagus: This is a relatively new term denoting extensive
side-to-side fusion (Fig. 131-9). The twins share the umbi-
licus, lower abdomen, pelvis (single symphysis pubis), and
the genitourinary tract. They can have anorectal anomaly
and colovesical fistula and may be at risk of anencephaly.

DIAGNOSIS
Antenatal Diagnosis and Imaging
Fetal ultrasound can detect the presence of conjoined twins in
almost all cases. Accurate antenatal assessment allows the
parents to be counseled about the probable outcome of the
pregnancy and the likelihood of successful postnatal separa-
tion. Prenatal diagnosis of conjoined twins is important for
FIGURE 131-7 Ischiopagus.
optimum obstetric management including decisions regarding
termination of pregnancy and the timing and method of
ureters. The twins can be tetrapus, tripus, or bipus, al- delivery to minimize maternal and fetal mortality.
though the most common arrangement is the presence
of four legs. Pelvic conjunction leads to complex urogenital Fetal Ultrasound Prenatal ultrasonography (US) is capable
and orthopedic anatomy. The kidneys usually function of diagnosing conjoined twin pregnancies as early as 12 weeks’
normally but are often malrotated or ectopic in location. gestation.28–30 Transvaginal US may also aid early diagnosis.31
When two bladders are present, they lie side by side in a Diagnosis of conjoined twins may be straightforward when
collateral position or they may lie in a sagittal midline fusion of fetal parts is obvious (Fig. 131-10). In addition, the
location with one bladder draining into the other. The ure- possibility of conjoined twins should be suspected in a twin
ters frequently cross over and insert into a contralateral
bladder such that they will need to be rerouted during
separation. Partial urethral duplication is possible, but a
single urethral orifice is typical. The distal gastrointestinal
tract is often shared, with anorectal agenesis and rectove-
sical fistula. Contrast studies are necessary to delineate
distal bowel anatomy. Urogenital sinus or cloaca may
be present. In boys there is an increased incidence of
undescended testes.
5. Craniopagus: The conjoined twins share the skull,
meninges, and venous sinuses (Fig. 131-8). The brains
are usually separate, although some cortical fusion can
occur in 33% of cases.
6. Cephalopagus: The twins often have a fused thorax in addi-
tion to a fused head. The single fused head may have two

FIGURE 131-9 Parapagus. (From Spitz L, Stringer MD, Kiely EM, et al:
Separation of brachio-thoraco-omphalo-ischiopagus bipus conjoined
twins. J Pediatr Surg 1994;29:477-481.)
FIGURE 131-8 Craniopagus.

FIGURE 131-10 Prenatal ultrasound of conjoined twins.
pregnancy with a single placenta and no visible separating am-
niotic membrane. Conversely, observation of two placentas or
an amniotic membrane excludes conjoining.32 The sono-
graphic findings in conjoined twins include inseparable fetal
bodies and skin contours, an unchanged relative position of
the fetuses, both fetal heads persistently at the same level,
bibreech or bicephalic presentations, fewer limbs than
expected, and a single umbilical cord with more than three
vessels.32 Most conjoined twins face each other and are fused
ventrally, resulting in hyperextension of their cervical spines.
Polyhydramnios occurs in up to 50% of conjoined twin preg-
nancies compared with 10% of normal twins and 2% of sin-
gleton pregnancies. Detailed US assessment at around 20
weeks’ gestation should be able to define the site and extent
of the conjoined area and provide a reasonable evaluation
of which viscera are, and are not, shared.
Fetal echocardiographic assessment needs to be detailed
because there is an increased incidence of congenital heart
disease in conjoined twins overall, particularly thoracopagus
twins.33 A shared heart is seldom compatible with life and
usually an indication for termination of the pregnancy. Hearts
can be confirmed as separate when they are seen to be anatom-
ically separate or when the heart rates are different. It has been
the experience of many authors that fetal echocardiography
underestimates the severity of cardiac anomalies.32
The thoracopagus heart typically has six chambers. Due to
the abnormal cardiac anatomy and function, an increased
nuchal translucency and subcutaneous edema have been
noted in thoracopagus twins in particular.29 Communications
at the atrial and ventricular levels can often be well defined
with US. However, the great vessel relationships and atrial
morphology are often difficult to determine. Prenatal US does,
nevertheless, provide an opportunity to accurately define the
morphologic connections among the various chambers of the
hearts. That is important because the state of the cardiac
conjunction is the fundamental pointer to eventual outcome.
Prenatal echo is made easier by amniotic fluid, particularly
polyhydramnios in later pregnancy, acting as an acoustic
window, and the lack of lung aeration permits US evaluation
from different approaches. Prenatal diagnosis can be techni-
cally easier for the operator because it avoids the anatomic
constraints of scanning fused chests postnatally, and fetal
fluid-filled lungs may allow better images.35
Recently three-dimensional (3D) US imaging has been
advocated as a new tool to demonstrate the extent of fusion
CHAPTER 131 CONJOINED TWINS 1731
in conjoined twins. 3D US may add anatomic information
and thus improve the accuracy of classification of individual
twins.33,35 In some cases the category of conjoined twins is
suspected by two-dimensional scanning but can be better
appreciated on 3D US. It would seem sensible, however, to
postpone attempts at detailed 3D US until after 14 weeks’ ges-
tation, when better views of the anatomy can be obtained.30
Although 3D US may add more information regarding the
extent of the union in utero, it is questionable whether these
findings affect management prenatally.33
Magnetic Resonance Imaging Magnetic resonance imaging
(MRI) can be superior to US for overall fetal assessment. MRI,
with its ability to differentiate soft tissues, provides an excel-
lent alternative technique.33 Ultrafast T2-weighted (T2-W)
sequences of short duration such as the single-shot fast
spin-echo sequence allow minimal image degradation by fetal
motion and high-quality images of fetal organs without the
need for fetal or maternal sedation.36 The larger field of view
of MRI permits better evaluation of the spatial relation-
ships of anatomic anomalies or between normal structures
(Fig. 131-11). Additional 3D MRI models may be useful in
specific circumstances.37
Postnatal Imaging
The choice of imaging study will depend to some degree on
the site of fusion. All conjoined twins should have chest
and abdominal radiography for an overall general assessment,
as well as to help health professionals understand the extent
of the conjoined area. Unexpected diaphragmatic hernia or
vertebral anomalies can thus be detected early.
Ultrasound All neonates should have routine cerebral US
and, when indicated, a spinal US as baseline investigations.33
In addition, abdominal US to assess the liver and to document
the presence of two spleens, gallbladders, biliary systems,
bladders, and four kidneys is necessary. Detailed Doppler
studies to evaluate the great vessels in the abdomen and he-
patic venous drainage should also be performed, but midline
abdominal conjunction may make accurate Doppler assess-
ment unreliable. Meticulous labeling of the images ensures
the correct twin is consistently noted to be on the same side.
Echocardiography Echocardiography (ECHO) is manda-
tory for every twin due to the high frequency of congenital
heart disease. 3D ECHO has been advocated postnatally to
make it easier to understand the cardiac connections and help
plan treatment.35 Andrews and colleagues35 analyzed 23 sets
of conjoined twins and classified twins according to the degree
of cardiac fusion as follows: separate hearts and pericardium
(group A, n ¼ 5), separate hearts and common pericardium
(group B, n ¼ 7), fused atria and separate ventricles (group
C, n ¼ 2), and fused atria and ventricles (group D, n ¼ 9).
The degree of cardiac fusion was correctly diagnosed in all
but one set. The intracardiac anatomy was correctly diagnosed
in all cases, although the antenatal diagnosis was revised post-
natally in three cases. Abnormal intracardiac anatomy was
found in one twin only in two group A pairs, one group B pair,
and both group C pairs. All group D twins had abnormal anat-
omy. Ventricular function was good in all twins scanned pre-
natally, and postnatally function correlated well with clinical
condition. None of the twins from groups C or D survived
1732 PART IX SPECIAL AREAS
B
FIGURE 131-11 A, Prenatal magnetic resonance imaging of conjoined
twins. B, Scan of twins showing abdominal conjunction.
demonstrating that the outcome in twins with fused hearts
remains dismal.
Computed Tomography Radiologists avoid CTwhere possi-
ble in infancy due to the high radiation burden. Conjoined
twins, even when stable and asymptomatic, are a well-justified
exception. Due to the high spatial resolution and speed in
particular of multidetector CT (MDCT), it is the best overall
modality for evaluating conjoined twins in the postnatal
setting (Fig. 131-12).
Contrast enhancement is mandatory to assess the vascular
anatomy and so prior cannulation to minimize the stress
of the procedure is advisable. Intravenous contrast agents
(Fig. 131-13) for CT are given33; this can demonstrate the
FIGURE 131-12 Computed tomography scan of thoracopagus
conjoined twins.
FIGURE 131-13 Contrast-enhanced computed tomography showing
conjoined liver and large arterial connection.
shared liver and vascular connections. Delayed images give
very useful information on excretion by both twins’ kidneys,
the ureteric anatomy, and the location and number of blad-
ders. Separate studies examining each twin’s vascular anatomy
on different days are recommended.33
MRI MRI has an increasing role in the postnatal evaluation
of conjoined twins, particularly those joined at the head
or thorax (Fig. 131-14). MRI has the capability of producing
3D-reconstructed images in any direction with much better
resolution and tissue characterization than 3D US. MRI is
the optimum examination to assess for any cortical fusion
in craniopagus twins. Although the ultrafast sequences such

FIGURE 131-14 Postnatal magnetic resonance imaging: coronal T1
image after contrast administration to the right twin showing enhance-
ment of that twin’s liver parenchyma.
as single-shot T2-W images have advantages in the constantly
moving fetus, more conventional T1-W, T2-W, and T1-W im-
ages after IV gadolinium administration are used in the post-
natal imaging of conjoined twins. Intracardiac anatomy, great
vessel anatomy and blood flow, and ventricular wall motion
can all be accurately assessed in thoracopagus cases. Many
complex fusions will inevitably get both MRI and MDCT.
These modalities are often complementary with MRI showing
soft tissue anatomy to best effect and CT detailing complex
bony anatomy in pelvic conjunction. MRI with magnetic
resonance cholangiopancreatography (MRCP) is likely to
offer the best hope of assessing the biliary anatomy. Liver
anatomy can be best appreciated after intravenous contrast
enhancement, either via CT or MRI. Additional coronal or
3D reconstructions will provide further information. Dem-
onstration of separate hepatic venous drainage into the infe-
rior vena cava and right atrium of each twin is essential to
plan separation as absent or severely anomalous hepatic
venous drainage in one twin is incompatible with survival
after surgery.
It is not unusual to find large-caliber vessels crossing the
area of fusion. It is important to visualize these vessels by
CT or MRI before separation to minimize blood loss and mor-
tality during separation. It is rarely necessary to perform an
angiogram before separation to clarify the vascular anatomy.
Contrast Studies of the Gastrointestinal and Genitourinary
Tracts Abdominal conjunction often involves fusion of
parts of the intestine. In addition, there is always a common
peritoneal cavity. Even in the absence of fused bowel, when
contrast medium is given to one twin the bowel loops may
be freely mobile across the “midline” and be seen to project into
the peritoneal cavity of the other twin during fluoroscopy.33 Pel-
vic conjunction leads to complex fusion and anomalies in the
anorectal region. These require an individual approach with
contrast studies (enemas, loopograms, and cystograms) of the
CHAPTER 131 CONJOINED TWINS 1733
distal bowel and bladder. Although performed in many cases,
in practice, contrast studies provide limited information and of-
ten, particularly due to overlapping bowel loops, the anatomy
can only be revealed at the time of separation.38
Urologic abnormalities are confined to those in whom the
pelvis is joined: ischiopagus, parapagus, or pygopagus twins.
Most twins share four kidneys and two bladders, occasionally
with one ureter crossing from the contralateral twin to the
other. The bladders are usually side by side, but they may
be sagittally placed, which presents a reconstructive
challenge.39 CT scanning usually provides sufficient infor-
mation regarding the upper renal tracts and bladders. Detailed
urethral anatomy and possible fistulas require retrograde
contrast medium examinations.
Nuclear Medicine The kidneys, albeit ectopic or too few in
number, usually function normally in conjoined twins.33
Cross-sectional imaging can usually clarify the genitourinary
anatomy without the need for isotope renography studies.
There is an increased frequency of pelviureteric or ureterove-
sical obstruction in these twins, but this is seldom of major
importance before separation.
Although hepatobiliary agents have been used to demon-
strate separate biliary drainage, these are seldom necessary
in practice. It is important to demonstrate the presence of
two separate gallbladders. When two gallbladders are defi-
nitely seen in thoracopagus or omphalopagus twins, the like-
lihood of separate biliary systems is high.40 When twins share
the liver or have shared hepatic veins or other large shunts,
then mixing of blood precludes the use of a hepatobiliary
radiopharmaceutical. Despite detailed preoperative imaging,
complex biliary anatomy may only become apparent at the
time of surgery. MRCP may have a particular role to play in
this area in the future. Individual twins raise unique individ-
ual dilemmas, and nuclear medicine studies may nevertheless
be useful in specific circumstances.
Obstetric Management
Delivery should take place at, or close to, the surgical unit
where separation will be performed. Delivery must always
be by classical cesarean section at 36 to 38 weeks’ gestation.
Anesthetic Management
Anesthesia involves two completely separate teams with all
members being clearly aware for which twin they are respon-
sible.41 All drugs and intravenous fluids administered are
calculated on a total weight basis with half being delivered
to each twin. Because of the cross-circulation, drugs given in-
travenously may have an unpredictable effect and particular
care must be taken to administer such drugs incrementally.
Endotracheal intubation should be via the nasal passage for
added security during repositioning, especially once separa-
tion has been achieved and the infants are moved into separate
operating rooms or onto separate operating tables, as well as
for continuing respiratory support postoperatively. Full
arterial and central venous monitoring is essential and, in ad-
dition, ECG, pulse oximetry, capnography, and urine output
must be carefully monitored. All lines and monitoring cables
must be color coded for the individual twin to avoid confusion
during repositioning. Adequate venous access is essential
because brisk hemorrhage may be encountered, necessitating
1734 PART IX SPECIAL AREAS
rapid large-volume transfusion. Cardiovascular monitoring is
particularly important because, with relative changes in posi-
tion of the two infants, significant shifts in blood volume
may occur.
Separation Procedure
The key to a successful outcome is thorough preoperative
planning. From the imaging, the extent of major organ union
will be apparent, and the details of each phase of the proce-
dure are planned in advance. The order in which these phases
of separation will be performed is best decided during the op-
eration itself because operative findings will often dictate a dif-
ferent sequence of events to what was originally envisaged.
Many anatomic variants exist for each of the different types
of union. For instance, although the liver is joined in 100%
of thoracopagus twins, in 50% there is intestinal union in
addition. The surgeon will need to have planned for each
possibility—such as what to do if there is a single duodenum
or a single common bile duct.
In the most extensive forms of ischiopagus or parapagus
union, the terminal ileum and the large bowel are single.
Allocation of these structures cannot be planned in detail
because the mesenteric blood supply is abnormal and unpre-
dictable. As a general rule, these children have two bladders,
which may lie side by side or anteriorly and posteriorly. Plans
must be in place to deal with either eventuality, together with
plans for dealing with crossed or uncrossed ureters. Finally,
there may be a single urethra, although more commonly there
are two urethras. Finally, with extensive union, there will be a
substantial body wall defect after separation. The management
of this problem also needs to be planned in advance. We have
used tissue expanders in the past and found them to be trou-
blesome and unhelpful in providing substantial extra tissue.
With the sole exception of craniopagus twins, we would not
recommend their use. We use polypropylene mesh over a
plastic liner to close the defect and plicate this, as one would
with any other abdominal wall defect. Even with the largest of
defects, the body wall can be closed within 2 weeks.
Initially, one surgical team comprising both lead surgeons is
sufficient to commence the procedure. Our preference is to
start the operation posteriorly. The skin and soft tissue are
divided down to muscle and fascia. The wound is then su-
tured closed, and the babies are turned around. We have done
this with the more extensive unions; otherwise, it is difficult to
allocate body wall equally and one may easily leave one with
significantly less. Although this sounds cumbersome, we have
never regretted taking the time for this maneuver. Once this
incision has been made, it is easy to complete the separation
at the final stage of the operation.
If a vestigial limb is present posteriorly, it may be bivalved
and divided and the bones filleted out. This provides vascular-
ized healthy tissue to assist in closure. Allocation of this extra
tissue is decided at the time of surgery.
On opening the chest and abdomen anteriorly, it is usually
best to divide the conjoined sternum to improve access. The
lobar anatomy of conjoined livers is not readily discerned, and
in any case it is not necessary to be precise about this. A point
is chosen, midway between the gallbladders and the porta
hepatis on either side, and the liver is divided in the usual
manner. The use of an ultrasonic dissector combined with
one of the high-energy sealing devices results in virtually
bloodless conditions.
Once the liver is divided, the remainder of the gastrointes-
tinal anatomy is easier to manage. The urologic aspects are
dealt with before the decision about allocating the rectum.39
Usually each twin has a bladder and at least a posterior
urethra. Occasionally, two bladders lying fore and after are
divided and the two halves are united on either side. If the
bladders are side by side, the division takes place between
them. In ischiopagus and parapagus twins the ureters may
cross into the opposite twin’s bladder. Our experience suggests
that in almost all cases, it is possible to preserve at least one
corpus of the penis for each twin. Occasionally, there are three
or more corpora cavernosa in which case a more normal penis
will result.
Once the urologic aspect has been completed, a decision is
made on allocation of the rectum and anus. This will depend
on the blood supply that is available. It has been our practice
to leave the terminal ileum and colon with one twin while
leaving the rectum with the other when conditions permit.
Once the pelvic viscera have been allocated, the posterior
part of the pelvic ring is divided. Preoperative imaging will
have revealed if there are substantial vessels crossing at this
level. Finally, the previously sutured skin wound is opened
and separation is complete.
Closure of a large body wall defect is straightforward. If the
sternum is short and the heart or pericardium is exposed, a
polypropylene patch is sutured to the chest wall laterally
and superiorly and to the diaphragm inferiorly. This patch
is left permanently in situ. The abdomen is closed by suturing
two flaps of polypropylene mesh to the muscle on either
side and covering the viscera with a plastic liner such as an
intestinal bag.
Usually there is sufficient skin to cover the chest wall de-
fect. The skin over the abdomen is tacked to the mesh to cover
the area where the mesh is sutured to muscle (Fig. 131-15).
This allows tissue to grow from the fascia through the mesh
to the overlying subcutaneous fat and ensures a strong attach-
ment of mesh to muscle.42–46
Postoperative Management
Almost all these patients are electively ventilated after surgery
with the usual monitoring that is used after major surgery.
There are, however, two problems that are not common in
other children—fluid loss and cardiac performance.
FIGURE 131-15 Large anterior abdominal wall defect covered with
polypropylene (Prolene) mesh after separation.

If the abdomen has been closed with a mesh, there will be
substantial fluid losses for the first few days.
We estimate the losses by use of absorbent gauze dressings,
which are weighed on an hourly basis. The fluid is then
replaced milliliter for milliliter with human albumin solution.
The mesh is tightened, commencing on the day after opera-
tion, although for the first few days, the degree of tension
applied is limited. However, after the first 4 days, increasing
tension can be applied to the mesh because it is tightened
and the abdominal wall stretches and grows in an impressive
fashion. It is usual to achieve abdominal wall closure within
2 weeks.
The other aspect in which conjoined twins differ is in car-
diac performance. It is a feature of those who have extensive
union that one circulation supports the other. In general, the
thinner and more active twin has a heart that is more robust
and supports the other twin’s circulation. This difference in
cardiac performance is not evident on echocardiography.
However, once the circulations are separated, the more
dependent twin may have a compromised circulation. It is
of critical importance that the lowest filling pressure possible
is used and that the afterload is also reduced. We try to main-
tain as low a central venous pressure as possible, together
with as low a blood pressure as possible, commensurate with
an adequate urine output. This phase of compromised cardiac
performance lasts a number of days but is usually resolved
by 7 days postoperation.
The timing of enteral feeding is variable but should not
necessarily await body wall closure. In the majority, complete
removal of the mesh is achieved when the abdomen is closed.
On occasion, it is necessary to leave some mesh, and this has
not been a particular problem over the following years.
The management after birth can take one of three courses
(Table 131-3):
1. Nonoperative management is indicated when there is
complex cardiac union without the possibility of recon-
structing even a single functioning heart or where there is
extensive cerebral fusion. Parents faced with the prospect
of a surviving twin having severe deformities may refuse
consent for separation. In eight of nine cases in our series
there was cardiac fusion, while in one with extensive para-
pagus fusion (a normal triplet survived) parents refused
surgery. All nine sets died within a short period.
2. Emergency separation is required when one twin is dead
or dying and threatening the survival of its sibling, where
there is a correctable congenital anomaly incompatible with
survival if left untreated such as esophageal or intestinal atre-
sia, or where serious damage has occurred to the connecting
bridge or area of fusion (ruptured omphalocele).
Emergency separation was necessary in eight of our
cases. One had a rupture of the shared omphalocele result-
ing in tearing of the united liver with exsanguinating hem-
orrhage resulting in the death of one twin. Emergency
separation was undertaken in an attempt to salvage the
surviving twin who unfortunately died at the end of the
procedure. In three sets of thoracopagus twins, cardiac in-
stability prompted emergency separation. Only one of
these infants survived. Prenatal volvulus with necrosis
and perforation occurred in a set of omphalopagus twins,
necessitating emergency separation. Both twins had intes-
tinal atresia that was repaired, but both had short bowel
syndrome that required many months for adaptation to
CHAPTER 131 CONJOINED TWINS 1735
occur. They survived and eventually thrived. Another set
of omphalopagus twins was transferred soon after birth
because one had severe pulmonary hypoplasia needing
mechanical ventilation. This twin died in the ambulance
during transfer. The transport team was instructed to dig-
itally compress the area of union to prevent exsanguination
into the surviving twin and avoid applying a clamp because
prior experience showed that much of the surviving twin’s
intestine may reside in the peritoneal cavity of its sibling.
This twin did survive.
The survival rate for emergency separation was 4 out
of a possible 16 infants—25%. In retrospect, survival
was not possible in six of these cases.
3. Planned separation—if the twins are stable and healthy at
birth, they should be allowed to feed normally and thrive
with the intention to separate at around 2 to 3 months.
This allows time to carry out investigations to define as
accurately as possible the nature and extent of shared
organs. The operative procedure can then be planned with
involvement of all the relevant staff.
Fourteen sets underwent planned separation, of whom
three died (two at home 6 months after successful separation)
and 25 survived (89%) (Table 131-4).
Follow-up
The extent of the union will determine whether the infants
will have minor or major residual problems. One or both
twins may have congenital cardiac abnormalities that may re-
quire surgical correction. Residual scars or contractures may
require plastic surgical correction. Short bowel syndrome oc-
curs when each twin is allocated 50% of the shared intestine.
There will be a need for parenteral nutrition and meticulous
dietary support for a period of time pending intestinal adap-
tation, which generally occurs. Intestinal stomas may be tem-
porary or permanent and will require the input of a stoma
therapist. Imperforate anus will need reconstruction if pos-
sible. Prolonged urinary follow-up is essential, particularly
in the cases of ischiopagus and pygopagus twins. Shared
bladder, cross ureters, vesicoureteric reflux, and renal abnor-
malities will require regular monitoring to prevent further
damage and to diagnose early and treat urinary calculi. Life-
long orthopedic follow-up will be necessary for limb defor-
mities, deletion of lower limbs, and scoliosis and chest wall
deformities.
Psychologic support should commence preoperatively
when separation is undertaken at a later age and is essential
following separation, especially when one twin dies. When
the separation has involved prolonged periods of hospitaliza-
tion (some of our twins have been discharged within 2 weeks
of separation), there may be developmental and behavioral
problems that require attention.
Ethics
If separation is possible with the expected survival of both
twins, we strongly recommend that the surgical procedure
be performed. If the hearts of the twins are fused, the chances
of survival of one or both twins is so remote that surgery
should be declined and nature allowed to take its course.
If one twin has a lethal abnormality and cannot survive
independently from his or her “normal” twin, and without
separation both twins would die, then separation should pro-
ceed, even though it is at the expense of the abnormal twin.
1736 PART IX SPECIAL AREAS

TABLE 131-3
Outcome of Cases
Separation Not Attempted
Type (-pagus) Sex Age at Operation Shared Organs Associated Anomalies Outcome
Thoraco F Cardiac fusion Pulmonary atresia, esophageal atresia Died
Thoraco F Cardiac fusion Exomphalos Died
Thoraco F Cardiac fusion Died
Para F Extensive union (Surviving triplet) Died
Thoraco F Cardiac fusion Died
Thoraco F Cardiac fusion Died
Thoraco F Cardiac fusion Died
Thoraco F Cardiac fusion Died
Thoraco F Cardiac fusion Died

Treated by Emergency Separation

Type Sex Age at Shared Organs Associated Anomalies Outcome
(-pagus) Operation
Thoraco F 3d Pericardium, diaphragm, liver, Pulmonary atresia 1 died of pulmonary atresia
CBD, small bowel 1 alive
Omphalo F 1d Liver, small bowel Exomphalos (ruptured), lacerated liver, 1 died before admission
cloacal anomaly 1 died postoperation
Thoraco M 8d Cardiac-atrial communication, Unilocular heart 1 early death,
liver One absent hepatic veins 1 died 6 after wk of SIDS
Thoraco F 1d heart, liver Unilocular heart, CDH Both died
Omphalo F 2d Liver, midgut Exomphalos, volvulus, bowel atresia Both alive
Omphalo M 1d Bowel, bladder Imperforate anus, hypoplastic lungs 1 died during transfer—
pulmonary hypoplasia
1 alive
Para F 1d Lungs, diaphragm, aorta, liver, AV canal defect, PPH Both died
bowel
Thoraco F 5d Pericardium, liver Cardiac, Porencephalic cyst One died immediately (PPH)
The other died after 2 wk

Treated by Planned Separation

Type Sex Age at Shared Organs Associated Outcome
(-pagus) Operation Anomalies
Cranio F 2m Sagittal sinus 1 alive
1 died 6m
Cranio Sagittal sinus Both alive
Ischio M 8m Liver, distal gut, genitalia ARA Both alive
Tripus
Para F 3y Pericardium, diaphragm, liver, distal gut, genito-urinary Renal agenesis 1 alive
1 died 6m
Para M 10m Pericardium, liver, distal gut, genito-urinary ARA 1 alive
1 died 6m
Thoraco F 3m Pericardium, liver, biliary atresia, small bowel atresia Exomphalos Both alive
Omphalo F 3m Liver Both alive
Pygo F 2m Spinal cord ARA Both alive
Ischio F 2m Pelvis Bladder exstrophy Both alive
ARA
Omphalo F 6w Liver Both alive
Omphalo F 8m Liver Both alive
Pygo F 3m Sacrum Both alive
Spinal cord
Ischio M 9m Liver, colon Tripus Both alive
Ischio M 5m Sternum, pericardium, liver, distal ileum and colon, anorectum, Both alive
penis

d, days; w, weeks; m, months; y, years; AV, atrioventricular; CBD, common bile duct; CDH, congenital diaphragmatic hernia; SIDS, sudden infant death syndrome;
PPH, persistent pulmonary hypertension; ARA, anorectal anomaly.
 CHAPTER 131 CONJOINED TWINS 1737

TABLE 131-4
Outcome of Management of Conjoined Twins in the Major Series
Authors Yr No. No Operation Emergency Separation Planned Separation
Survivors (%) Survived (%)
O’Neill (USA)65 1988 18 5 5 sets 8 sets
1 (10%) survived 13 survived (81%)
Cywes/Rode (South 2006 33 16 None 17 sets
Africa)66,67 22 survived (65%)
Al Rabeeah (Saudi Arabia)68 2006 29 19 (1 abandoned None 10 sets
operation) 19 survived (95%)
Saguil (Philippines)69 2009 22 6 6 sets 9 sets
1 (8%) survived 15 survived (83%)
Spitz/Kiely/Pierro (UK) 2010 31 9 8 sets 14 sets
4 (25%) survived 25 survived (89%)

Yet twins can survive joined together for many years. Alice D. surgical separation. Others have disputed this theory because
Dreger, PhD, a professor in medical humanities and bioethics, it cannot account for all the abnormalities found in these twins
contends that “when it comes to cases in which one twin must and particularly cannot explain the acephalic occurrence. It is
be ‘sacrificed,’ it is ethically wrong to take one life so another generally accepted that the heteropagus twin is genetically
may live.”46a She further states that “not in a single case has identical to the autosite, and this has been substantiated by
the twin chosen to survive ever actually survived to go home.” DNA analyses.
This is clearly incorrect as borne out in our and other series.
ANATOMIC TYPES
Reports of 157 cases of heteropagus twins have appeared in
Heteropagus (Parasitic) Twins the literature until 2003. Recently, Sharma and colleagues48
------------------------------------------------------------------------------------------------------------------------------------------------
reviewed 39 cases reported in 23 publications since 2003.
Heteropagus or parasitic twin is a grossly defective fetus, or fetal In six cases the specific variety was not defined.
parts, attached externally, with or without internal connections, Table 131-5 provides a breakdown of these cases:
to a relatively normal twin (the autosite) in one of the same eight Dipygus or caudal duplication (Fig. 131-16)49 is a rare abnormal-
areas in which symmetrical twins are united.23 They are usually ity comprising two sets of accessory lower limbs sited between
composed of externally attached supernumerary limbs but may the autosite’s limbs and including accessory external genitalia.
also contain viscera or visceral parts and only rarely a beating Omphalopagus (Fig. 131-17) or epigastric heteropagus ac-
heart or intact brain. counts for the majority of cases—more prevalent in recent
Fetus in fetu is a fetiform mass enclosed within the body of series, in which almost 60% of cases were in this region.50–52
the autosite, usually the abdominal cavity, rarely within the Fetus in fetu53–55 occurs most commonly within the abdom-
brain, with grossly recognizable fetal parts including an axial inal cavity with 57 reported cases with only 9 having a brain
skeleton, attached to the autosite by a pedicle containing a and 6 a heart, both always small and rudimentary. In most
few large blood vessels. Its growth rate is similar to the host cases a pedicle of vessels arises from the retroperitoneal region
within which it is discovered. and attaches to the umbilical area of the parasite. In the cranial
region the majority of fetuses in fetu are epignathi attaching to
the posterior pharynx near the Rathke pouch. Only seven
INCIDENCE
cases of intracranial fetuses in fetu have been reported.
The estimated incidence of heteropagus twins is much less Sacral parasite is a rare variety of heteropagus twin and needs
than for symmetrical twins—1 per 1 million births or less. to be differentiated from sacrococcygeal teratoma.56 Some
The female preponderance seen in symmetrical twins is not authors consider fetus in fetu and fetiform teratomas within
evident in heteropagus twins where there is an equal distribu- the spectrum of teratomas, but this remains controversial.
tion of the sexes.

EMBRYOLOGIC CONSIDERATIONS TABLE 131-5

The most plausible theory for the occurrence of heteropagus Types of Heteropagus Twins Reported
twins was proposed by Dönitz in 1866.47 He postulated that Thoracopagus 1
heteropagus twins originated from symmetrical twins, one of Cephalopagus 6
which suffered secondary damage as a consequence of vascu- Parapagus 3
lar compromise. The affected “twin” would then have to rely Omphalopagus 71
on collateral blood supply from the autosite while ischemic Ischiopagus 32
damage occurred in various parts of the affected “twin.” In Pygopagus 46
support of this theory, hypoplastic umbilical vessels have been Craniopagus 11
found in the heteropagus twin and vascular connections from Rachipagus 28
the autosite to the heteropagus twin may be found during
1738 PART IX SPECIAL AREAS

INVESTIGATION
Sharing of organs and vascular connections are infrequent in
heteropagus twins. As a result, preoperative imaging has usu-
ally been restricted to CT, ultrasound, and MRI scans, which
reveal bony and soft tissue structures. Occasionally angiogra-
phy has been undertaken to show vascular communications,
but in recent years magnetic resonance angiography, which is
safer and noninvasive, has replaced the invasive conventional
angiography. Echocardiography of the autosite is mandatory
in thoracopagus parasitic twins because more than 25% have
congenital cardiac defects.

SURGERY
FIGURE 131-16 Dipygus with prolapsed intestine.
Separation of the parasite from the autosite is generally under-
taken in early infancy. As mentioned earlier, sharing of organs
and major vascular connections is unusual in heteropagus
twins. Therefore the surgical procedure is generally straight-
forward and uncomplicated. The initial skin incision should
be carefully planned to take sufficient skin and subcutaneous
tissue from the parasite to ensure tension-free closure of the
wound following separation. One of the common complica-
tions following separation is wound breakdown or infection,
and this can be avoided by careful planning and meticulous
closure of the incision. A thorough search of the autosite at
the site of attachment must be carried out to exclude any
retained abnormal or extraneous tissue and to correct possible
anomalies in the autosite such as associated congenital intes-
tinal abnormalities.

FIGURE 131-17 Parasitic omphalopagus twins. OUTCOME

The prognosis for the autosite is generally excellent except
for cases with severe cardiorespiratory problems. Wound
DIAGNOSIS complications can be avoided by careful preoperative
planning and meticulous attention to detail in the operative
Prenatal identification of heteropagus twins has been docu-
procedure.
mented on at least seven occasions at gestational ages ranging
from 9 to 28 weeks. The importance of recognition of the
nature of the abnormality is evident in that the condition is
entirely benign and compatible with normal development. Acknowledgments
The diagnosis in utero should not be considered an indication
Special thanks to Dr. Simon Eaton for collating the references and organizing
for termination of the pregnancy provided that it is not the illustrations.
causing gross deformity of adjacent structures.
The complete reference list is available online at www.
expertconsult.com.
OBSTETRIC CARE
Depending on the size and extent of the parasitic twin, deliv-
ery can be by normal vaginal route. If there is a suspicion of
expected obstructive delivery, then cesarean section at around
38 weeks of gestation would be indicated.
 Index

Page numbers followed by f indicate figures and t indicate tables.

A Abdominal trauma (Continued) Abuse (Continued)

Aarskog syndrome, 967 to duodenum. See Duodenum, trauma to. drug, abdominal wall defects and, 976
ABCA 3 mutation, 674 to external genitalia, 308 sexual
ABCDE sequence, 263 gastrointestinal, 305–308, 307f. anorectal pathology secondary to, 1320
Abdomen See also Gastrointestinal tract, trauma to. genital injuries in, 308
acute surgical, after bladder augmentation or historical perspective on, 289 Accident prevention. See Injury, prevention of.
replacement, 1483–1484 to kidney. See Kidney, trauma to. Acetaminophen, 216, 216t
bruising of, from seat-belt restraints, 307, 307f to liver. See Liver, trauma to. for burns, 382
distention of to pancreas. See Pancreas, trauma to. Acetate, in parenteral nutrition, 190–191
in adhesive bowel obstruction, 1127–1128 to perineum, 308 Acetylcholinesterase
in ascites, 1171 solid organ, 291–299 in abdominal wall defects, 978
in colonic atresia, 1248 to spleen. See Spleen, trauma to. in Hirschsprung disease, 1265, 1267
in intestinal neuronal dysplasia, 1280 Abdominal wall N-Acetylcysteine, for meconium ileus, 1079–1080,
in jejunoileal atresia and stenosis, 1061, 1061t closure of 1081–1082
in meconium ileus, 1075, 1075f, 1081 after intestinal transplantation, 655, 656f Achalasia
in megacystis-microcolon-intestinal temporary, 298 cricopharyngeal, 942
hypoperistalsis syndrome, 1285 in trauma patient, 270 esophageal, 944–946, 945f, 946f
in necrotizing enterocolitis, 1195–1196 defects of, 973–988 internal anal sphincter, 1276, 1278, 1283–1287,
in portal hypertension, 1360 antenatal considerations in, 977–978 1284f
germ cell tumors of, 516 associated conditions with, 979–982, 979t Acid-base balance, 94–95
perforation of, in meconium ileus, 1075 complications of, 983 Acid-base regulation, 115
Abdominal compartment syndrome, 298–299, 299f, cryptorchidism in, 1004–1005 Acid burns, 383
481, 481f embryogenesis of, 975–976, 976f Acid ingestion. See Esophagus, caustic injury to.
Abdominal mass gastroesophageal reflux disease and, 958 Acidosis. See also Metabolic acidosis.
in choledochal cyst, 1334 genetics and familial occurrence of, lactic, with bacterial overgrowth, 1140
differential diagnosis for, 427 977, 977t in neonate, 94
in intussusception, 1099 historical perspective on, 973 Acinic cell carcinoma, salivary gland,
in mesenteric and omental cysts, 1168, 1168f obstetric delivery with, 978–979 733, 733f
in neuroblastoma, 442 outcome of, 983–984 Acinus(i), 111
palpable, management of, 1262 spectrum of, 973–975, 974f, 974t pancreatic, 1371
Abdominal packing, 297–298 umbilicoplasty for, 971–972, 972f pulmonary, 811
Abdominal pain embryology of, 975, 975f Acoustic neuroma, 601
in adhesive bowel obstruction, 1127–1128 expansion of, 298–299, 299f Acquired immunodeficiency syndrome (AIDS).
in appendicitis, 1256, 1259 in prune-belly syndrome, 1506, 1507f, See HIV/AIDS.
in choledochal cyst, 1334 1508f Acrocephalopolydactylous dysplasia, 977t
differential diagnosis of, 1258–1259, 1258t rhabdomyosarcoma of, 497 Acrosyndactyly, 1722, 1723f
in intussusception, 1099 Abdominoplasty ACTH. See Adrenocorticotropic hormone (ACTH).
in midgut volvulus, 1116 patch, 298–299, 299f Actinomycin, 412
multidetector computed tomography in, 41, 42f for prune-belly syndrome, 1506, 1508f Actinomycosis, cervicofacial, 743
in ovarian tumors, 530 ABI (ankle-brachial index), 365 Activated partial thromboplastin time, in coagulation
in pancreatitis, 1372 ABO-incompatible heart transplantation, 663 disorders, 171
in peptic ulcer disease, 1031–1032 Abscess Actuators, microelectromechanical, 61
in peritonitis, 1231 anal, 1214–1215, 1318–1319, 1318f Acute chest syndrome, in sickle cell disease, 168,
in ureteropelvic junction obstruction, 1414 in appendicitis, 1257, 1262 1342, 1387
Abdominal paracentesis, in chylous ascites, 1174 brain, 1693, 1694, 1695–1696, 1697 Acute respiratory distress syndrome (ARDS).
Abdominal pressure breast, 773 See also Respiratory failure.
measurement of, 298 in Crohn disease, 1213, 1214–1215 lung transplantation in, 675
in trauma patient, 298–299, 299f epidural pharmacologic adjuncts in, 120
Abdominal surgery intracranial, 1693–1694, 1694f, 1695, 1696 Acute tubular necrosis, in renal transplantation,
neonatal energy expenditure after, 103–104, 104f intraspinal, 1697 626–627
robotic, 60 hepatic. See Liver, abscess of. Adalimumab, for Crohn disease,
single incision laparoscopic, 55–56, 55f kidney, in pyelonephritis, 1431 1212
Abdominal torso vascular injuries, 364 lung, 867–870, 869f Adenitis, cervical, 727. See also Lymphadenitis.
Abdominal trauma, 289–313 perinephric, after renal trauma, 318 Adeno-associated viral vectors, for gene transfer,
to anus, 308 peritonsillar, 717–718 24–25, 24t
to bile duct, 299, 300f retropharyngeal, 718, 718f Adenocarcinoma
compartment syndrome in, 298–299, 299f salivary gland, 731 bladder, 1523–1524
damage-control strategies for, 294–298, 296f, spleen, 1387–1388 cervical, 1609
297f, 297t subperiosteal, 1694–1695, 1694f colonic, 1250
diagnostic modalities in, 289–291, 290f Abuse esophageal, 483
to diaphragm, 308–309 child. See Child abuse. gastric, 483

Volume 1, pages 1–768; Volume 2, pages 769–1738 I-1

I-2 INDEX

Adenocarcinoma (Continued) Adrenal rests, in inguinal hernia repair, 1001 Allergic fungal sinusitis, 713
pancreatic, 1382, 1384 Adrenalectomy, 564–566 Allergy
vaginal, 1609 anterior (transabdominal), 564–566, 565f latex, bladder augmentation or replacement and,
Adenoid cystic carcinoma cortical-sparing, 566 1491
bronchial, 568 laparoscopic, 566 metal, in pectus excavatum repair, 784, 789–790,
salivary gland, 733 posterior, 565–566 792
Adenoidectomy, 720 thoracoabdominal, 565–566 milk, hematemesis from, 1151
Adenoma Adrenocortical rests, 557 Allgrove syndrome, 945
adrenal, 563–564 Adrenocorticotropic hormone (ACTH), 558 Alloderm, in burn care, 378
bronchial, 567–568 in Cushing syndrome, 561, 562, 562f Allograft, for bone tumors, 587–588, 587f, 589f
hepatocellular, 461 ectopic production of, 562, 563 Alloplastic material, 1712
nipple, 774, 777 Adrenogenital crisis, 1574 Allotransplantations, islet cell, 638–641, 639f
pancreatic, 1382 Adriamycin, 407t Aloe vera, for burns, 372
parathyroid gland, 751–752 ADVANCE program, 250 Alpha-adrenergic blocking agents
parotid gland, 732–733, 733f AESOP, 58 for dysfunctional elimination syndromes,
pleomorphic, 721, 732–733, 733f Africa, pediatric surgery in, 17, 17f 1463–1464
salivary gland, 733 Afterload, 134 for pheochromocytoma, 560
Adenoma sebaceum, in tuberous sclerosis, 1399 Afterload agents, for congestive heart failure, 135 Alpha fetoprotein
Adenomatoid malformation, congenital cystic, 825, Aganglionosis in abdominal wall defects, 978
826, 826f, 827 colonic. See Hirschsprung disease. in hepatoblastoma, 464–465
Adenomatosis, erosive, 777 intestinal, near-total, 1272–1274 in neural tube defects, 1676
Adenomatous polyposis syndromes, 1179. persistent or acquired, after pull-through, in ovarian tumors, 530, 530t
See also Familial adenomatous polyposis. 1274–1276 screening for, 77
Adenosine, for supraventricular tachycardia, 138, 139t AIDS. See HIV/AIDS. serum, 460t
Adenotonsillar hypertrophy, 719, 719f Air embolus, in pulmonary laceration, 277 in testicular tumors, 550
Adenoviral vectors, for gene transfer, 24–25, 24t Air enema Alum-precipitated toxoid (APT) test, 1148
Adenovirus infection, intussusception and, 1097 intestinal perforation with, 1108 Aluminum toxicity, 193
Adhesion barriers, 1129 in intussusception, 1103, 1103f Alveolar-arterial oxygen gradient, in congenital
Adhesions Air trapping, in congenital lobar emphysema, 828 diaphragmatic hernia, 816, 821
anomalous, in alimentary tract duplications, 1156 Airway. See also Larynx; Trachea. Alveolar dead space, 114–115
intestinal obstruction from assessment of, 722–723, 837 Alveolar development, 111, 111f
inflammatory, 1130 development of, 109, 110f Alveolar rhabdomyosarcoma, 400–401, 491–492,
postoperative, 1127–1129, 1128f, 1129f dilation of, for laryngotracheal stenosis, 846, 846f 494
labial, 1558–1559, 1606 inflammatory disease of, 725–726 Alveolar ridge, 716
Adipogenesis inhibitory factor, in necrotizing surgical, in trauma patient, 265 Amastia, 771, 772f
enterocolitis, 1190t, 1192 trauma to, 277–279, 279f Amebic abscess, hepatic, 1352
Adipose tissue, brown, 98–99 vascular compression of, 853–854 Amenorrhea, in vaginal agenesis, 1592
Adjuvant chemotherapy, 406 Airway management American Burn Association, major burn criteria of,
Adolescents in burn injury, 372 375, 375t
bariatric surgery in. See Bariatric surgery in in sepsis, 155 American College of Critical Care Medicine, sepsis
adolescents. in trauma patient, 263–265, 264f guidelines of, 154–162
cognitive development in, 1044–1045 in upper airway obstruction, 723 American Joint Committee on Cancer (AJCC) staging
postoperative compliance in, 627, 1044–1045, Airway obstruction system, 582
1045t acute, 722–723 American Pediatric Surgical Association (APSA), 7,
Adrenal adenoma, 563–564 in cervicofacial lymphatic malformation, 1622 235
Adrenal cortex chronic, 726 Amino acids
anatomy of, 557 clinical presentation in, 837 formula based on, in short bowel syndrome, 1137
functions of, 558 evaluation of, 837 metabolism of, in neonate, 102–103
insufficiency of, Addison disease and, 564 fetal interventions for, 83 surgery and, 107
lesions of, 561–563 in laryngomalacia, 840 in parenteral nutrition, 103, 189
Cushing syndrome as, 561–563, 562f management of, 723 requirements for, 181–182
sex hormone–producing, 563 pathophysiology of, 837–838 Aminoglycosides, for necrotizing enterocolitis, 1206
treatment of, 563 sleep-disordered breathing and, 718–720, 719f 5-Aminosalicylic acids, for ulcerative colitis, 1221
Adrenal gland tracheotomy for. See Tracheotomy. Amiodarone, for supraventricular tachycardia, 138,
anatomy of, 557 in vocal cord immobility, 842 139t
embryology of, 557 Airway resistance, 114 Amniocentesis, 77
hemorrhage of ALADIN syndrome, 945 Amnion, human, in burn care, 378–379
fetal, 564 Alarm therapy, for nocturnal enuresis, 1465 Amniotic fluid tests, in abdominal wall defects, 978
neonatal, 564 Albendazole, for hepatic hydatid disease, 1353 Amobarbital, intracarotid injection of, 1689
injury to, in birth trauma, 393 Albumin Amoxicillin
physiology of, 558 serum, nutritional status and, 180 for otitis media, 710
steroid biosynthetic enzyme nomenclature related supplementation of, in burn injury, 374 for peptic ulcer disease, 1033, 1033t
to, 1569t Aldosterone Amoxicillin-clavulanate, for otitis media, 710
steroid hormone synthesis in, 1570f overproduction of, 563–564 Ampicillin, for urinary tract infection, 1431–1432
tumors of, 557–567 regulation of, 558 Amplification, DNA, 401
cortical, 561–563 Algorithms, 234 Amputation
incidental, 564 Alimentary tract duplications, 834–835, 834f, 835f, for bone tumors, 586, 586f
medullary, 558–561 1133, 1155–1165 partial, in capillary-lymphaticovenous
Adrenal hyperplasia anomalies associated with, 1155 malformation, 1629
congenital, 564, 1568t. See also Disorders of sex clinical manifestations of, 1156–1157 penile, 324, 324f
development (DSD). diagnosis of, 1157–1158, 1157f, 1158f through diaphysis, 334
diagnosis of, 1573 embryology of, 1155–1156 traumatic, 340
genitogram in, 1576f incidence of, 1155 Amygdalohippocampectomy, 1691
medical management of, 1574–1575 intestinal obstruction in, 1133 Anabolic steroids, for aplastic anemia, 166
pathophysiology of, 1569–1570 locations of, 1156t Anal canal, 1291, 1311, 1312f
reconstruction for. See Female gender Alkali burns, 383 Anal continence, 1311–1312. See also Incontinence,
assignment surgery. Alkali ingestion. See Esophagus, caustic injury to. fecal.
hyperaldosteronism and, 563–564 Alkaline phosphatase, in bone tumors, 581 Anal fissure, 1317–1318
lipoid, 1570 Alkalosis rectal bleeding in, 1151, 1317
nodular, 563 after gastrocystoplasty, 1484 sexual abuse and, 1320
Adrenal medulla in neonate, 94 treatment of, 1317–1318
anatomy of, 557 Alkylating agents, 406, 407t Anal sphincter
functions of, 558 Allantois, 961–963, 962f in anal continence, 1312
lesions of, 558–561. See also Pheochromocytoma. Allen test, 337 anatomy of, 1291, 1311, 1312f

Volume 1, pages 1–768; Volume 2, pages 769–1738


Anal sphincter (Continued)
external, 1311, 1312f
internal, 1311, 1312f
achalasia of, 1276, 1278, 1283–1287, 1284f
myectomy of, 1282, 1284–1285
Analgesia. See Pain management.
Analgesic ladder, 215f
Anaplastic large cell lymphoma, 525, 526–527
Anaplastic lymphoma kinase (ALK) oncogene, 405
in neuroblastoma, 441
Anaplastic Wilms’ tumor, 428, 434, 435
Anastomosis
leakage around, after colonic interposition, 931
simulated, 73
Anastomotic stricture
after hypospadias repair, 1552–1553
after portal hypertension surgery, 1366, 1367f
after pull-through for Hirschsprung disease, 1274,
1275f
Anastomotic ulceration, in necrotizing enterocolitis,
1204
Anatomic barriers, in host defense, 145–146, 145f
Anatomic dead space, 114–115
Anderson-Hynes pyeloplasty, 1421, 1422f, 1423
Androgen. See also Testosterone.
adrenal, 558
deficiency of
cryptorchidism and, 1005, 1007
in disorders of sex development, 1568t, 1570,
1570f, 1573
malignancy risk in, 508
excess of. See Adrenal hyperplasia, congenital.
hypospadias and, 1536
Androgen insensitivity syndrome, 1568t,
1570–1571, 1573
inguinal hernia and, 1001
versus vaginal agenesis, 1592–1593
Androgen receptor deficiency, 1568t, 1570–1571,
1573
Anemia, 165–169
aplastic, 166
from blood loss, 167
from bone marrow failure, 165–167
Diamond-Blackfan, 166–167
Fanconi, 166, 169
hemolytic, 168–169
iron-deficiency, 167–168
postoperative apnea and, 203–204
sickle cell, 168
in ulcerative colitis, 1220
workup for, 165, 166f
Anencephaly, 1673, 1674–1675, 1676
Anesthesia, 201–232. See also Local anesthetics; Pain
management.
apnea after, 203
cardiac arrest associated with, 203
complications of, 201–204
in conjoined twins, 1733–1734
emergence delirium with, 209
fluid management with, 205–207
induction of, parental presence during, 250–251
inhalation
agents for, 201, 202f, 202t, 207–209, 207t
laryngospasm associated with, 203
malignant hyperthermia with, 210–211, 211t
intravenous agents in, 201, 202f, 211–212, 212f
monitoring of
invasive, 214
noninvasive, 212–213, 213f
mortality associated with, 202
neuromuscular blocking agents with, 209–210, 210t
physiologic considerations in, 201
premedications for, 201, 204–205
preoperative evaluation for, 202, 204–205
preoperative fluid restrictions for, 203t, 204
regional. See Regional anesthesia.
risk of, 201–204
Aneuploidy, screening for, 77
Aneurysm
abdominal aortic, 1631–1636, 1635f, 1636f
arterial, lower extremity, 1642–1643
brachial artery, 1643, 1643f
carotid artery, 1644–1645, 1645f
false, after splenic injury, 294, 295f
Aneurysm (Continued)
intracranial, traumatic, 353
renal artery, 1639, 1639f
splanchnic artery, 1641
Aneurysmal bone cyst
chest wall, 573
location of, in relation to physis, 579f
resection of, 583f
Angiofibroma, juvenile nasopharyngeal, 715–716
Angiogenesis, 1620
inhibition of, 410–411
in neuroblastoma, 449
versus preformed vascular networks, 33, 34f
Angiographic embolization, in abdominal trauma,
294–296, 296f
Angiography. See also Cholangiography.
in arteriovenous malformation, 1626, 1626f
computed tomography. See Computed
tomography angiography.
magnetic resonance, in portal hypertension, 1361
in Meckel diverticulum, 1089
in musculoskeletal trauma, 331–332
in portal hypertension, 1361
radionuclide, in pectus excavatum, 783–784
in renal injury, 313
in vascular trauma, 362
Angioma. See also Hemangioma; Lymphangioma.
tufted, 1619–1620
Angiomatosis, cutaneovisceral, with
thrombocytopenia, 1620
Angiomyolipoma, renal cysts in, 1399
Angioplasty, percutaneous transluminal, for
renovascular hypertension, 1638
Angiosarcoma, 1620
breast, 777
hepatic, 480
Angiotensin, for septic shock, 161
Angiotensin-converting enzyme (ACE) inhibitors, for
congestive heart failure, 135, 137t
Ankle-brachial index, 365
Ankyloglossia, 720, 720f
Ann Arbor staging system for Hodgkin lymphoma,
519, 519t
Annular pancreas, 1051, 1053, 1053f, 1054, 1056
Anoplasty, 1298, 1300f
Anorectal angle, 1311
Anorectal malformations, 1289–1312.
See also specific disorders, e.g., Anus,
imperforate.
associated anomalies with, 1289–1290, 1290f
classification of, 1289, 1290t
clinical findings and initial management of,
1291–1296
in females, 1291–1296, 1294f, 1295f
in males, 1291–1296, 1291f, 1292f, 1293f
colostography of, 1296, 1296f
colostomy for, 1293–1294, 1295–1296, 1296f
closure of, 1306
management after, 1296, 1296f
embryogenesis of, 1289
historical perspective on, 1289
incidence of, 1289
pathophysiology of, 1291
reconstruction for, 1296–1307.
See also Anorectoplasty.
in females, 1301–1305
limited, 1294–1295
in males, 1297–1301
outcome of, 1307–1309, 1308t, 1309f
postoperative care in, 1306–1307, 1307t
principles of, 1296–1297, 1297f
with spinal cord tethering, 1459–1460
vascular, 1319
Anorectal manometry
in constipation, 1314
in Hirschsprung disease, 1267
in internal anal sphincter achalasia, 1283–1284
Anorectal pain, in proctalgia fugax, 1320
Anorectal trauma, 308, 1153, 1153f
Anorectoplasty, posterior sagittal
for cloaca, 1301–1305
in females, 1301–1305
for imperforate anus without fistula, 1300
limited, 1294–1295
Volume 1, pages 1–768; Volume 2, pages 769–1738
INDEX I-3
Anorectoplasty, posterior sagittal (Continued)
in males, 1297–1301
outcome of, 1307–1309, 1308t, 1309f
postoperative care after, 1306–1307, 1307t
for rectal atresia and stenosis, 1301
for rectobladder neck fistula, 1298–1300, 1300f,
1301f
for rectourethral fistula, 1297–1298, 1298f,
1299f, 1300f
for vestibular fistula, 1301
Anosmia, 715
Anoxic brain damage, in birth trauma, 392
Antacids
for peptic ulcer disease, 1033
for stress ulcers, 1034
Anthracyclines, 407t
Antiangiogenic therapy, 410–411
for hepatocellular carcinoma, 479
Antiarrhythmic agents, for supraventricular
tachycardia, 138, 139t
Antibiotic-lock technique, 193–194
Antibiotics
for airway obstruction, 723
antitumor, 406, 407t
for appendicitis, 1259
for atypical mycobacterial lymphadenitis, 742
for bacterial overgrowth, 1140
for burns
intravenous, 383
topical, 376–377, 377t
for cat-scratch disease, 1351
for catheter-related infections, 1139–1140
for cervical adenitis, 727
for dialysis-related peritonitis, 1233
for intracranial infections, 1696
for liver abscess, 1351
for lung abscess, 869
for lymphadenitis, 740
for meconium ileus, 1082
for necrotizing enterocolitis, 1199–1200,
1206–1207
for otitis media, 709–710, 709t
for peptic ulcer disease, 1033, 1033t
for pouchitis, 1228
prophylactic
cardiac surgery and, 1647
in heart transplantation, 666–667
for sepsis, 154
after splenectomy, 1391
in ureteropelvic junction obstruction, 1421
for urinary tract infection, 1432, 1433
for sepsis, 158–159
for shunt infection, 1685
for spinal epidural abscess, 1697
for umbilical cord cleansing, 963
for urinary tract infection, 1431–1432, 1432t
Antibody(ies). See also Immunoglobulin(s).
radiolabeled, 54
Antibody-mediated rejection, in lung
transplantation, 678–679
Anticholinergic agents
for fecal incontinence, 1315
for neuropathic bladder, 1459, 1459f, 1460, 1461
for overactive bladder syndrome, 1464
for posterior urethral valves, 1462
Anticoagulants
for abdominal aortic thrombosis, 1636
lupus, 174
naturally occurring, disorders of, 174–175
for renal vein thrombosis, 1439–1440
for venous thromboembolism, 175
Anticonvulsants
orofacial clefting and, 699
prophylactic, in intracranial infections, 1696
Antidepressants, tricyclic, for nocturnal enuresis,
1464–1465
Antidiarrheal agents, in ulcerative colitis, 1222
Antifungal agents, for necrotizing enterocolitis,
1199–1200
Antigens
crossmatching of, for transplantation, 615
matching of, for transplantation, 614–615, 615f
migration and localization of, 610–611,
611f, 612f
I-4 INDEX
Antilymphocyte antibodies
in renal transplantation, 624
in transplantation, 606–607
Antilymphoid antibodies, in transplantation,
612–613, 613f
Antimetabolites, 406, 407t
Antireflux procedure. See Fundoplication.
Antithrombin III deficiency, 174
Antithymocyte globulin. See Thymoglobulin.
Antitumor antibiotics, 406, 407t
Antivenin, 340–341
Antrectomy, for stress ulcers, 1034–1035
Anus. See also Anorectal entries.
abscess in, 1214–1215, 1318–1319, 1318f
anatomy of, 1311, 1312f
dilatation of
after anorectoplasty, 1306, 1307t
sexual abuse and, 1320
fistula in, 1318–1319, 1318f
in Crohn disease, 1210–1211, 1212, 1215,
1215t
imperforate. See also Anorectal malformations.
associated anomalies with, 1290, 1290f
colostomy for, 1239f, 1240, 1241f, 1242
genitourinary anomalies associated with, 1470,
1471f
penile agenesis with, 1585, 1588f
surgical management of, 1470, 1472f
without fistula, 1289, 1294, 1300
normal position of, 1313–1314
rhabdomyosarcoma of, 497
sexual abuse and, 1320
streptococcal dermatitis in, 1318, 1318f
warts in, sexual abuse and, 1320
Anxiolytics
for burns, 382–383, 382t
parental presence during induction of anesthesia
and, 250, 251
Aorta
aneurysm of, abdominal, 1631–1636, 1635f,
1636f
coarctation of
abdominal, 1631–1634, 1632f, 1633f, 1634f
congenital, 1650–1652, 1650f, 1651f
traumatic, 283
renal artery implantation into, 1637, 1637f,
1638f
splanchnic artery implantation into, 1640, 1640f
thrombosis of, abdominal, 1636
trauma to, 282–286, 283t, 284f, 285f
Aortic arch
development of, 1665, 1666f
double, 853, 854, 1665, 1667, 1667f
left, with aberrant right subclavian artery,
1665–1666, 1668, 1670f
right, with left ligamentum arteriosum, 1665,
1667, 1668f, 1669f
Aortography, in trauma, 274, 283
Aortopexy, for tracheomalacia, 851, 851f, 914
Aortoplasty, patch, for aortic coarctation, 1631,
1633f, 1634f, 1651
Aortorenal bypass, 1637, 1638f
APC gene, in familial adenomatous polyposis, 488,
1180, 1181
Apert syndrome, 693, 1722, 1723f
Aphthous stomatitis, in ulcerative colitis, 1219
Aphthous ulcers, in Crohn disease, 1210
Aplasia cutis congenita, 1713–1714
Aplastic anemia, 166
Apnea, 718–719
obstructive sleep, 203–204, 719, 1043
postoperative, 203
reflex, in tracheobronchial vascular compression,
853
reflux with, 950–951, 951t
tracheomalacia with spells of, 914
Apnea index, 719
Apoptosis, 399
Appendectomy, 1259–1262, 1260f, 1261f
complications of, 1262
in Ladd procedure, 1122
laparoscopic, 1259–1262, 1261f
with meconium evacuation or irrigation,
1079–1080
small bowel obstruction after, 1127
Appendicitis, 1255–1267
chronic, 1255
clinical presentation in, 1256
complicated, 1262
complications of, 1262
diagnosis of, 1256–1259
differential diagnosis of, 1258–1259, 1258t
imaging studies in, 1257–1258
laboratory studies in, 1257
in meconium ileus, 1082
multidetector computed tomography in, 41, 42f
outcomes of, 1262–1263
perforated, 1256, 1257
physical examination in, 1256–1257
spectrum of, 1255
treatment of, 1259–1262, 1260f, 1261f
Appendicostomy
choices for, 1240
continent, 1309, 1309f
indications for, 1237
Appendix
anatomy of, 1255
carcinoid tumors of, 485–486, 1259
duplication of, 1255
embryology of, 1255
for Mitrofanoff neourethra, 1480, 1481f, 1493,
1494f
in sliding hernia sac, 1000
Apple-peel deformity, 1064–1065, 1066f
Applicability of study, 234
Appropriate for gestational age, 89, 91f
Aqueductal stenosis, 1680f, 1681
Arginine
for necrotizing enterocolitis, 1207
requirements for, 182
Argon beam coagulator, 49
Arrhythmias, in neonate, 138–139, 139t
Arterial anastomosis
in liver transplantation, 648
in renal transplantation, 622
Arterial blood gas analyzer, microelectromechanical,
61
Arterial blood gases, in congenital diaphragmatic
hernia, 816
Arterial catheterization, 116–117
for intraoperative monitoring, 214
Arterial disease, 1631–1648
aortic, 1631–1636
cerebrovascular, 1643–1645
extremity, 1641–1643
renal artery, 1636–1639
splanchnic artery, 1639–1641
Arterial switch operation, 1662
Arteriography. See Angiography.
Arteriovenous-capillary fistula, 1629
Arteriovenous fistula, 1358
Arteriovenous malformation, 1625–1627, 1626f
capillary malformation with, 1626, 1629, 1629f
cerebral, 53
hepatic, 460–461
nidus of, 1626–1627
staging of, 1625–1626, 1626t
treatment of, 1626–1627
Arthralgia, in ulcerative colitis, 1219
Arthrography, hip, 1701
Arthrogryposis, 1722–1723
Arthropathy, in Crohn disease, 1211
Arthrotomy, traumatic, 334
Arytenoidopexy, for vocal cord immobility, 843
Ascariasis, intestinal obstruction in, 1133
Ascites, 1171–1178. See also Intraperitoneal fluid.
anatomy and pathophysiology of, 1171
bacterial, 1171
biliary, 1173–1174
causes of, 1171, 1172t
chylous, 1174–1175
clinical features of, 1171
diagnosis of, 1172, 1173f, 1173t
hepatocellular, 1172–1173
laboratory evaluation of, 1172, 1173t
in necrotizing enterocolitis, 1198, 1201
in portal hypertension, 1359
after portoenterostomy, 1329
after shunt operations, 1366
urinary, 1175
Volume 1, pages 1–768; Volume 2, pages 769–1738
Ascorbic acid, during burn fluid resuscitation, 374
Asia, pediatric surgery in, 15–16, 16f
Asparaginase, 407t
Aspergillus infection
pulmonary
in cancer patient, 860–861, 860f
in HIV-infected patient, 864
in renal transplant patient, 651t
Asphyxia, traumatic, 286, 286f
Asphyxiating thoracic dystrophy, 805–807, 807f,
808f
Aspiration
with enteral nutrition, 187–188
fine-needle. See Fine-needle aspiration.
gastric, lung abscess from, 868
Aspirin, platelet abnormalities from, 170
Asplenia, 1386–1387
Assent, pediatric, 238–239, 239t
Assist-control mode, in mechanical ventilation, 118
Astrocytoma
cerebellar, 594, 595f
cervicomedullary, 597
genetics of, 601
hypothalamic/chiasmatic, 597–598, 598f
pilocytic, 53
supratentorial
low-grade, 599–600, 599f
malignant, 600, 600f
ASVS technique, in hyperinsulinism, 1380
Ataxia
in brain tumors, 591–592
cerebellar, in neuroblastoma, 443
ATF3 gene, in hypospadias, 1536
ATG15L1 gene, in Crohn disease, 1209
Athelia, 771
Atlanto-occipital dislocation, 359
Atlanto-occipital subluxation, 765f
Atlantoaxial subluxation, 359, 765
Atracurium, 210t
Atrial flutter, after lung transplantation, 678
Atrial septal defect, 1652–1654, 1652f, 1653f,
1654f
Atrioventricular septal defect, 1657–1659
cardiac anomalies associated with, 1658
cardiovascular management in, 140
classification of, 1657, 1657f, 1658f
management of, 1658–1659, 1659f
natural history and diagnosis of, 1658
results of, 1659
Atrioventricular valves, injury to, 281, 281f
Atrium
anatomy of, 1652, 1652f
Wilms’ tumor extension to, 431
Atropine sulfate, during endotracheal intubation,
265
Audiometry, 708
Auditory canal, external, 707, 708
absence of, 708
Aural atresia, 708
Auricle, 707
laceration of, 711
Australia, pediatric surgery in, 15
Austria, pediatric surgery in, 15
Autograft, in burn care
cultured epithelial, 380
mesh, 379–380, 379f
Autologous tissue, 1712
Autonomic nerve tumor, gastrointestinal, 484
Autonomy principle, 237
bariatric surgery and, 242
Autotransplantations, islet cell, 638, 638f
Avalon cannula, for extracorporeal life
support, 127
Avascular necrosis, femoral head, 1703
AVPU pneumonic, 268
Axonal injury, diffuse, 345, 347–348, 347f, 348f
Azathioprine
for Crohn disease, 1212
in transplant patient, 606–607, 608, 609f
heart, 665, 667t
lung, 676–677, 676t
for ulcerative colitis, 1221
Azithromycin
for intestinal dysmotility, 1140
for Pseudomonas infection, 865

B Beckwith-Wiedemann syndrome, 405
B cell(s), in host defense, 147
B-cell lymphoma, 485, 523, 523t, 524–525, 524f
Baby Doe law, 240
Back pain, in spinal epidural abscess, 1697
Backboard, 335f, 357
Baclofen, for gastroesophageal reflux disease, 953
Bacterial infection
in ascites, 1171
in community-acquired pneumonia, 855–858,
856f
in necrotizing enterocolitis, 1194–1195, 1197
in renal transplant patient, 651t
Bacterial overgrowth
methods to decrease, 1206–1207
in short bowel syndrome, 1140
Bacterial toxins, 150
Bacterial virulence, 149–150
Ballard score for gestational age, 89, 90f
Balloon dilatation
for Crohn strictures, 1214
esophageal. See Esophagus, dilatation of.
for laryngotracheal stenosis, 846, 846f
Bannayan-Riley-Ruvalcaba syndrome, 1630
Bardach two-flap palatoplasty, 703, 704f
Bardet-Biedl syndrome, 1592
Bariatric surgery in adolescents, 1041–1054
clinical pathway for, 1049–1050
cognitive developmental concepts related to,
1044–1045, 1045t
compliance following, 1044–1045, 1045t
ethical considerations in, 241–242
guidelines for, 1048–1049
historical perspective on, 1041–1042
nutritional and metabolic consequences of,
1046–1048
obesity science related to, 1042
outcome of, 1041
patient selection for, 1048, 1048t
procedures for, 1045–1048, 1045t, 1047f
psychological factors related to, 1043–1044,
1049
regionalization of, 1041
team approach to, 1048–1049
timing of, 1049
training for, 1045
Barium enema. See also Enema, contrast.
in alimentary tract duplications, 1158f
in appendicitis, 1257
in constipation, 1314
hydrostatic, in intussusception, 1104, 1104f
intestinal perforation with, 1108, 1108f
in intussusception, 1101
in megacystis-microcolon-intestinal
hypoperistalsis syndrome, 1286, 1286f
in ulcerative colitis, 1221, 1221f
Barium meal, in achalasia, 945, 945f
Barium swallow
in motility disorders, 941
in pyloric atresia, 1035, 1035f
in thoracic enteric duplications, 1158–1159,
1159f
Barlow test, 1700
Barotrauma
with emphysema, 828
with mechanical ventilation, 122
Barrett esophagus, in gastroesophageal reflux
disease, 483, 956, 956f
after caustic injury, 924
after esophageal atresia repair, 913
Bartonella henselae infection, 727–728
Bartsocas-Papas syndrome, 977t
Basal cell carcinoma, sebaceous nevus and, 1714
Basal cell nevus syndrome, 529
Basal ganglia, tumors of, 593–594
Basal metabolic rate, 97
Basement membrane, 370
Basiliximab
in liver transplantation, 650t
in lung transplantation, 676–677, 676t
in renal transplantation, 624
BB gun injuries, 348
Bcl-2 gene, in neuroblastoma, 449
Beardmore, H., 7, 7f
Becker nevus, of breast, 773
INDEX I-5
Biobrane, in burn care, 377–378, 378f, 384
abdominal wall defects in, 977, 977t Biochemical markers
adrenocortical tumors in, 561 of sepsis, 153–154
hepatoblastoma in, 466–467 of ureteropelvic junction obstruction,
umbilical defects in, 969 1419–1420
Wilms’ tumor in, 424–425, 427 Biochemical screening, in prenatal diagnosis, 77
Bell-clapper anomaly, 1014 Bioelectrical impedance analysis, 180
Bell necrotizing enterocolitis staging criteria, 1187, Bioethics. See Ethics.
1188t, 1199 Biofeedback program, for dysfunctional elimination
Beneficence principle, 237 syndromes, 1463–1464
bariatric surgery and, 242 Biologic materials, 1712
Bentec bag, in gastroschisis reduction, 982 Bioluminescent imaging, 47–48
Benzodiazepines, for burns, 382–383 Biopsy, 417–423
Bernard-Soulier syndrome, 171 of bone tumors, 582, 583f
Best Evidence, 233 of brain tumors, 593
Best interests standard, 241 of burn wound, 372
Beta blockers of chest wall tumors, 573
in burn injury, 380–381 core needle, 418–420, 419f, 419t
for congestive heart failure, 135, 137t fine-needle aspiration, 418. See also Fine-needle
for pheochromocytoma, 560 aspiration.
for variceal hemorrhage, 1362–1363 laparoscopy with, 420
Beta-catenin mutation, in hepatoblastoma, 467 liver
Betamethasone in biliary atresia, 1324
for labial adhesions, 1558 in portal hypertension, 1361–1362
maternal, for cystic lung lesions, 826 lung, 875–876, 875f, 876f
Bias open incisional, 422
in case-control studies, 229 percutaneous needle, 418
in case reports, 227–228 rectal
identification of, 233–234 in Hirschsprung disease, 1267–1268, 1268f
Biatrial anastomosis, in heart transplantation, 664, in intestinal neuronal dysplasia, 1280, 1281f
664f, 665 in isolated hypoganglionosis, 1282, 1283f
Bicaval anastomosis, in heart transplantation, of rhabdomyosarcoma, 493–494
664–665, 666f salivary gland, 730
Bicycle injury, prevention of, 259 specimen handling for, 417–418
Bier block, 221–222 stereotactic, of brain tumors, 593
Bifid nipples, 772 thoracoscopy with, 420–422, 421f
Bifid scrotum, 1583–1584, 1586f Birth, transitional circulation at, 112, 135
Bile cysts, post-traumatic, 462 Birth injuries, 391–393
Bile duct fractures in, 391–392
common, 1371 neurologic, 392
anomalies of, 1372 soft tissue, 391
cystic dilatation of. See Choledochal cyst. thoracoabdominal, 392–393, 392f
development of, 1332 Birth weight
trauma to, 299, 300f gestational age and, 89, 91f
Bile lake, 464, 465f hypospadias and, 1536–1537
Bile reflux, 882 necrotizing enterocolitis and, 1135–1136,
Bilhaut-Cloquet procedure, 1723 1188–1189, 1203
Biliary ascites, 1173–1174 subgroups for, 89
Biliary atresia, 1321–1332 Bishop-Koop enterostomy, 1080–1081, 1080f
choledochal cyst with, 1333–1334 Bite injuries, 340–341
classification of, 1321–1322, 1322f BK virus infection, in renal transplant patient, 628
clinical presentation in, 1323–1324, 1323f Black widow spider bite, 341
embryogenesis of, 1322 Bladder
epidemiology of, 1321–1322 adenocarcinoma of, after bladder exstrophy repair,
etiology of, 1322, 1322t 1523–1524
historical perspective on, 1321 anatomic relationships of, 320
liver biopsy in, 1324 capacity of, 1454
nutritional complications of, 1328 compliance of, 1457, 1458f, 1467
nutritional support in, 197, 197t drainage of, for posterior urethral valves, 1556
outcomes with, 1327–1328 neuropathic, 1467, 1469–1470, 1470f
pathology of, 1322–1323, 1323f in cerebral palsy, 1460–1461
scintigraphy in, 1324, 1334–1335 detrusor-sphincter dyssynergy in, 1455–1456,
treatment of, 1324–1327 1458f
complications of, 1328–1329 megaureter with, 1497, 1498
liver transplantation in, 644, 644f, 1326, 1327, in myelodysplasia, 1457–1459, 1458f, 1459f
1329 in sacral agenesis, 1460, 1460f
postoperative care in, 1327, 1327t in spinal cord tethering, 1459–1460
preoperative care in, 1324 treatment of, 1459, 1459f
steroid controversy in, 1327, 1327t urinary tract infection and, 1428
surgical controversies in, 1326 voiding cystourethrography in, 1454, 1455f
surgical technique for, 1324–1326, 1325f, overactive, 1464
1326f pressure in, 1454–1455, 1456f, 1458f
Biliary dyskinesia, 1343 rhabdomyosarcoma of, 498
Biliary stent, in bile duct injury, 299, 300f stones in, 1438. See also Urolithiasis.
Biliary tract suspensory ligament of, 1303
carcinoma of, 1333, 1339 trauma to, 312, 313, 320–322, 321f
embryology of, 1332 causes of, 320
rhabdomyosarcoma of, 480, 497 classification and definitions of, 320–321
Biliopancreatic diversion, 1046 diagnosis of, 321
Bilirubin grading of, 314t
in choledochal cyst, 1334 management of, 321–322
in intestinal failure–associated liver disease, pelvic fracture with, 321, 321f
1139 valve, 1468, 1468f
Billroth, endothelial cords of, 1385 wall thickening in, imaging of, 1429–1430,
Bioartificial liver device (BAL), 33 1429f
Volume 1, pages 1–768; Volume 2, pages 769–1738
I-6 INDEX
Bladder augmentation or replacement, 1467–1489.
See also Urinary diversion.
in cloacal exstrophy, 1528
complications of, 1483–1485, 1495
continent channels for use with, 1479–1480,
1481f, 1493–1494, 1493f, 1494f
donor site considerations in, 1472–1473
fecal incontinence and, 1480–1482, 1482f
gastric segment for, 1475, 1475f, 1484, 1492
ileocecal segment for, 1473, 1492
indications for, 1471–1472
large bowel for, 1473–1474, 1492
patient evaluation and selection for, 1491
philosophy of, 1471, 1491
physiologic considerations in, 1471–1473, 1473f,
1492, 1492f
before renal transplantation, 1482–1483
seromuscular segments for use with, 1493
small bowel for, 1473, 1474f, 1475f, 1492
ureter for, 1475, 1476f, 1477f
Bladder base reconstruction, for ureterocele,
1450–1451
Bladder dysfunction, 1453–1470
anatomic, 1461–1462, 1461f, 1468–1469, 1469f
combined anatomic and neurogenic, 1470, 1471f,
1472f
in dysfunctional elimination syndromes,
1462–1464, 1462f, 1463f
history and physical examination in, 1453, 1454f,
1455f
neurogenic. See Bladder, neuropathic.
in nocturnal enuresis, 1464–1466
in posterior urethral valves, 1461–1462, 1461f
in prune-belly syndrome, 1507, 1510f
renal disease contributing to, 1467
renal transplantation and, 617–618, 619
ultrasonography in, 1454, 1455f
urinary tract infection and, 1428
urodynamic evaluation of, 1454–1457, 1456f,
1458f
voiding cystourethrography in, 1454, 1455f
Bladder exstrophy, 1515–1524
clinical presentation in, 1516, 1516f, 1517f
embryogenesis of, 1515–1516, 1516f
epidemiology of, 1515
genital defects in, 1516–1517, 1517f, 1518f
historical perspective on, 1515
initial management of, 1519
pelvic defects in, 1517
prenatal diagnosis of, 1517–1518
surgical reconstruction of, 1518
approaches in, 1518
bladder neck reconstruction in, 1522–1523,
1524f
complications of, 1523–1524
single-stage, 1523, 1525f
staged, 1519–1523
epispadias repair in, 1521–1522, 1522f,
1523f
functional closure in, 1519–1521, 1519f,
1520f, 1521f, 1522f
urinary diversion in, 1523
umbilical abnormalities in, 967
Bladder neck reconstruction, in bladder exstrophy
repair, 1522–1523, 1524f
Bladder outlet obstruction, 1468
in posterior urethral valves, 1461–1462, 1461f
Bladder outlet resistance
correction procedures for, 1475–1479, 1478f,
1479f, 1480f
factors influencing, 1467
pathologic, 1467
Blalock-Taussig shunt, 1660
Blastoma, pulmonary, 569–570, 570f
Bleeding. See Coagulation, disorders of; Hemorrhage.
Bleomycin, 407t
Blind loop syndrome, after jejunoileal atresia and
stenosis repair, 1067
Blisters, after burn injury, 384
Blocksom vesicostomy, 1488, 1488f, 1556
Blood loss, allowable, estimation of, 206
Blood pressure. See Hypertension; Hypotension.
Blood pressure sensor, microelectromechanical, 61
Blood transfusion. See Transfusion therapy.
Blood volume, in neonates, 92
Blue cell tumors, small, round, 418
Blue-rubber bleb nevus syndrome, 1625, 1625f
Body composition, direct measurement of, 180
Body fluids. See Fluid(s).
Body image, in pectus excavatum, 784
Body length
gestational age and, 89, 91f
normal changes in, 179
nutritional status and, 179–180
Body mass index (BMI)
nutritional status and, 179–180
obesity and, 1041, 1042
Body surface area, 374, 374t
Body temperature. See Hyperthermia; Hypothermia.
Body weight
normal changes in, 179
nutritional status and, 179–180
Boerhaave syndrome, 889–893. See also Esophagus,
rupture of.
BOLD imaging, 44–45
Bone
congenital anomalies of, 1699–1712
demineralization of, after bladder augmentation or
replacement, 1484
immature, 327–329, 328f, 329f, 330f, 331f
remodeling of, 329, 330f
tissue engineering of, 29–31, 30f
trauma to. See Fracture; Musculoskeletal trauma.
Bone criteria for sepsis, 141–142, 142f
Bone cyst
aneurysmal
chest wall, 573
location of, in relation to physis, 579f
resection of, 583f
unicameral, 578–579, 579f
injection therapy for, 584
location of, in relation to physis, 579f
Bone density, in short bowel syndrome, 1137
Bone disease
metabolic, with parenteral nutrition, 193
in renal transplant patient, 629
Bone marrow failure, anemia from, 165–167
Bone marrow injection, for unicameral bone cysts, 584
Bone marrow transplantation
bronchiolitis obliterans after, 673–674
for CAMT, 169
historical perspective on, 607, 608f, 610, 610f
for Wilms’ tumor, 435
Bone scintigraphy, in thoracic trauma, 274
Bone transport, 588–590, 589f, 590f
Bone tumors, 577–592
age of child and, 578t, 581
benign, 577–580, 578t
fracture through, 578–579, 579f
metastatic potential of, 580
multiplicity of, 579
reconstruction of, 587
resection of, 585, 585f
site of involvement of, 579–580
size of, 578, 578f
biopsy of, 582, 583f
chemotherapy for, 583
diagnosis of, 581–582
general considerations in, 577–590
incidence of, 578t
injection therapy for, 584
local recurrence of, 587
location of, in relation to physis, 579, 579f
malignant, 578t, 580–581
epidemiology of, 580–581
fracture through, 578
genetics of, 580–581, 580f
reconstruction of, 587–590, 587f, 588f, 589f,
590f
resection of, 585–587, 586f
minimally invasive surgery for, 584
pathophysiology of, 577–580
radiation therapy for, 583
radiofrequency ablation of, 584
reconstruction of
for benign lesions, 587
for malignant lesions, 587–590, 587f, 588f,
589f, 590f
Volume 1, pages 1–768; Volume 2, pages 769–1738
Bone tumors (Continued)
resection of, 584–585
adjuvants in, 585, 585f
for benign lesions, 585, 585f
compartments and, 584, 584f
growth after, 588
for malignant lesions, 585–587, 586f
staging of, 582
surgery for, 583–590
Botulinum toxin
for anal fissure, 1317
for internal anal sphincter achalasia, 1285
Bougienage
for congenital esophageal stenosis,
915–916
for esophageal anastomotic stricture, 912
Bovie, 49
Bowel. See Colon; Duodenum; Intestine; Small
intestine.
Brachial artery, aneurysm of, 1643, 1643f
Brachial plexus injury, in birth trauma, 392
Brachydactyly, 1724
Brachytherapy, 413
Bracing, for pectus carinatum, 795
Bracka two-stage buccal graft hypospadias repair,
1546, 1548f, 1549f
BRAF gene, in thyroid cancer, 749
Brain. See also Central nervous system.
abscess of, 1693, 1694, 1695–1696, 1697
arteriovenous malformations of, 53
positron emission tomography of, 46
Brain injury, traumatic, 344–354
birth-related, 392
in child abuse, 349, 353, 387–388, 387f, 388f
coagulopathy after, 269
contusion as, 344, 345–346, 345f
from crush injuries, 348–349, 349f
diffuse, 345, 347–348, 347f, 348f
early complications of, 352–353
early management of
in minor injury, 351–352
in severe injury, 350–352, 351f, 351t, 352f
emergency management of, 268–269
epidemiology of, 344–345
focal, 345–347, 345f, 346f
from gunshot wounds, 348
initial assessment of, 349–350
management of
basic concepts for, 343–344
medical, 351t
resuscitation and transport in, 344
outcomes with, 353
penetrating, 346, 346f, 351
primary versus secondary, 343–344
spectrum of, 345–349
vascular, 346–347, 353
Brain tumors, 591–604
age at diagnosis of, 591, 592t
clinical features of, 591–592
dural-based, 601
genetics of, 601
location of, relative to tentorium, 591, 592
metastatic, 601
radiologic evaluation of, 592–593
surgical management of, 593–594
types of, 591, 592t, 594–601
Brainstem
decompression of, for Chiari II malformation,
1677
glioma of, 597, 597f
tumors of, 593–594
Branchial anomalies, 757–760
first, 758, 758f
fourth, 758–760, 759f
piriform sinus, 759, 759f
second, 757–758, 757f, 758f
third, 758–760, 759f
Branchial (pharyngeal) apparatus, 753–755, 754f,
754t, 755f
Branchial cleft cyst, 721, 731–732
Branchial cleft sinus, 708
Branchio-oculo-facial syndrome, 757
Branchio-oto-renal syndrome, 757
BRCA genes, in ovarian tumors, 529–530
 INDEX I-7

Breast Bupivacaine, 220–221, 221t Cancer (Continued)

absence of, congenital, 1716, 1717f caudal, 224–225 chemotherapy for, 405–410, 407t.
amastia of, 771, 772f after inguinal hernia repair, 988–989 See also Chemotherapy.
asymmetry of, 773 Burkholderia cenocepacia, 671–672 clinical trials on, 415
Becker nevus of, 773 Burkholderia cepacia, 865 communication in, 249
cancer of, 777 Burkitt-like lymphoma, 524–525, 526 after conduit diversion, 1490
congenital anomalies of, 771–772, 772f, Burkitt lymphoma, 524–525, 524f, 526 epidemiology of, 397–398, 398t
1714–1720, 1717f, 1718f, 1719f, 1721f Burn center, 375, 375t genetic screening for, 405
development of Burns, 369–386 heredity and, 404–405
normal, 771, 772t acute management of, 371–376 historical perspective on, 9, 397
premature, 771 analgesia for, 382, 382t immunotherapy for, 411
disorders of, 769–779 antibiotics for lung infections in, 860–862, 860f, 862f
ectopic, 771–772 intravenous, 383 metastasis of, 402–403
embryology of, 1714–1716 topical, 376–377, 377t molecular biology of, 398–403, 400t, 401t
enlargement of, 773, 773t body surface area calculation for, 374, 374t molecular diagnostics in, 403–404, 404t
in boys, 777–778, 778f chemical, 383 radiation therapy for, 411–414. See also Radiation
juvenile or virginal, 773 in child abuse, 369, 370f, 389–390, 390f therapy.
neonatal, 774 degree of, 370, 370f in renal transplant patient, 628–629
fibrocystic changes in, 776 depths of, 370, 370f, 372 stem cell transplantation for, 415
gynecomastia of, 777–778, 778f, 1716–1719, dressings for, 377–379, 377t, 378f survival rate for, 398, 518f
1719f electrical, 383–384 transfusion therapy in, 176
hypomastia of, 771, 772f, 773 epidemiology of, 369 Cancer predisposition syndromes, 404–405
hypoplasia of, in Poland syndrome, 1719–1720, escharotomy for, 372, 373f, 379, 379f Candida infection
1721f excision and grafting for, 379–380, 379f in lung cancer patient, 861
infection of, 773 fluid resuscitation for, 374–375, 374t, 375t in renal transplant patient, 651t
masses of historical perspective on, 369 Cantrell, pentalogy of, 973, 974t, 975–976, 981, 983
in adolescent girls, 774–777, 775f, 776t hypermetabolic response to, 380–381 Cantwell-Ransley epispadias repair, 1521–1522,
prepubertal, 774, 774f, 775t inhalation injury in, 375–376, 376t 1522f, 1523f
polymastia and polythelia of, 771–772, 772f, 1716 initial evaluation of, 371–374, 373f, 373t Capillary-arteriovenous fistula, 1629
reconstruction of, in Poland syndrome, 798–799 nutritional support for, 381–382, 381t Capillary-arteriovenous malformation, 1626, 1629,
trauma to, 777 outpatient therapy for, 384 1629f
tuberous, 1716, 1718f pathophysiology of, 369–371 Capillary-lymphaticovenous malformation,
Breast feeding, in biliary atresia, 197 pharmacotherapy for, 382–383, 382t 1627–1629, 1627f, 1628f
Breast milk, 187–188 prevention of, 258–259 Capillary malformation, 1620–1621, 1621f,
fortifiers for, 187–188 rehabilitation for, 384–385 1712–1713
for necrotizing enterocolitis prophylaxis, 1205 sedatives and anxiolytics for, 382–383, 382t Capillary refill time, 337
Breath phases, in mechanical ventilation, 117 simulated treatment of, 74 Capnography, intraoperative, 213, 213f
Breath testing, in Helicobacter pylori infection, 1032 size of, estimation of, 372, 373f, 373t Capnometry, 116
Breathing, sleep-disordered, 718–720, 719f transfer to burn center for, 375, 375t Capsule endoscopy
bariatric surgery and, 1043 wound care for, 376–380, 377t in Crohn disease, 1211, 1211f
Breathing support, in trauma patient, 265–266, 266f zones of injury in, 370–371, 371f in gastrointestinal bleeding, 1154
Breech delivery Busulfan, 407t Carbaminohemoglobins, 115
developmental dysplasia of hip and, 1699 Buttocks, flattened, in sacral agenesis, 1460, 1460f Carbohydrate
with torticollis, 763 malabsorption of
Bronchial adenoma, 567–568 in necrotizing enterocolitis, 1196–1197
Bronchial artery embolization, for hemoptysis, 867 C in short bowel syndrome, 198
Bronchial stenosis, after lung transplantation, 677 C-reactive protein, in necrotizing enterocolitis, 1197 metabolism of
Bronchial trauma, 277–279 C syndrome, 977t in neonate, 99–102, 105–106
Bronchiectasis, 865–866, 866f C3, 148 postoperative, 105–106
Bronchioalveolar carcinoma, 568–569, 568t C3b, 151–152 in parenteral nutrition, 100–101, 105–106, 189
cystic malformations with, 568–569, 568t CA 125, in ovarian tumors, 530t, 531 overfeeding from, 194
Bronchiolitis, as pneumonia, 858–859, 858f Calcification, in pancreatitis, 1375 requirements for, 182
Bronchiolitis obliterans Calcineurin inhibitors, in transplant patient, 608, Carbon dioxide
after bone marrow transplantation, 673–674 609f diffusion of, 114
after lung transplantation, 674, 679, 680–681, 680f heart, 665, 667t end-tidal, 116
lung transplantation for, 673–674 liver, 649, 650t equilibrium of, 115
Bronchiolitis obliterans with organizing pneumonia pancreas, 636, 636f extracorporeal removal of, 119
(BOOP), 875, 875f renal, 624 hemoglobin binding to, 115
Bronchogenic carcinoma, 568–569, 568t, 569f Calcitonin, 745 partial pressure of, 114
Bronchogenic cysts, 832–833, 833f, 834f in Hirschsprung disease, 1265, 1267 arterial, 117
Bronchopulmonary dysplasia, with mechanical Calcium in congenital diaphragmatic hernia, 816
ventilation, 122 after bariatric surgery, 1046, 1048 mechanical ventilation and, 121
Bronchopulmonary sequestration, 825, 826, imbalance of. See Hypercalcemia; Hypocalcemia. transcutaneous monitoring of, 116
827–828, 827f in parenteral nutrition, 190t, 191 Carboplatin, 407t
Bronchoscopy regulation of, 93–94 Carcinoembryonic antigen, in colorectal cancer,
in airway trauma, 278 release of, calcium-induced, 134 489
in esophageal atresia with upper pouch fistula, Calcium oxalate stones, 1437–1438 Carcinogenesis, 399–400, 400t, 401f
910–911, 911f Caliceal diverticulum, 1403 Carcinoid tumors
in inhalation injury, 375 Calprotectin, in necrotizing enterocolitis, 1197 bronchial, 567–568
after lung transplantation, 677 Calretinin, in ovarian tumors, 531 gastrointestinal, 485–486
in pulmonary tumors, 570 Camey enterocystoplasty, 1473, 1474f in appendix, 1259
simulated, 73 Camptodactyly, 1723 in Meckel diverticulum, 1091
in tracheobronchial vascular compression, Camptothecins, 407t, 412 Carcinoma. See Cancer; Tumor(s).
853–854 CAMT, 169 Cardiac. See also Heart.
in tracheomalacia, 914, 914f Canada, pediatric surgery in, 9–10 Cardiac anomalies. See also Heart disease, congenital.
BronchSim device, 73 Canalicular testis, 1005, 1005f with anorectal malformations, 1290
Browne, D., 11, 11f, 12f Canalization defects, in alimentary tract duplications, with atrioventricular septal defect, 1658
Buccal graft hypospadias repair, 1546, 1548f, 1549f 1156 with congenital diaphragmatic hernia, 810
Buck fascia, 1537 Cancer, 395–419. See also specific organ or tumor type. with esophageal atresia, 896–897, 898, 898t
Bucket-handle fracture, 388 angiogenesis in, 403 with omphalocele, 979
Buckle fracture, 327, 328f antiangiogenic therapy for, 410–411 Cardiac arrest
Budd-Chiari syndrome, 1357 biopsy techniques for, 417–423 with anesthesia, 203
Budesonide, for Crohn disease, 1211–1212 after bladder augmentation or replacement, 1485, extracorporeal life support after, 124
Buffer systems, 94 1495 in thoracic trauma, 280

Volume 1, pages 1–768; Volume 2, pages 769–1738

I-8 INDEX
Cardiac arrhythmias, in neonate, 138–139, 139t
Cardiac catheterization
in heart transplantation, 662
in patent ductus arteriosus, 1648
Cardiac evaluation, in heart transplantation, 662
Cardiac failure. See Heart failure.
Cardiac index, in pectus excavatum, 783
Cardiac output
in burn injury, 371
left ventricular end-diastolic pressure and,
133–134, 134f
Cardiac surgery, antibiotic prophylaxis and, 1647
Cardiac tissue engineering, 30–31
Cardiac tissue viability, positron emission
tomography of, 46
Cardiac transplantation. See Heart transplantation.
Cardiac valves
injury to, 281, 281f
tissue-engineered, 30–31
Cardiomyopathy, heart transplantation for, 660–661,
660f, 661f
Cardiopulmonary bypass, for aortic injury, 283
Cardiopulmonary dysfunction, after radiation
therapy for Hodgkin lymphoma, 522
Cardiopulmonary indices, in congenital
diaphragmatic hernia, 816
Cardiopulmonary resuscitation
extracorporeal life support for, 123–136.
See also Extracorporeal life support.
in lung transplantation, 676
Cardiovascular death, in renal transplant patient, 629
Cardiovascular disorders. See also Heart disease,
congenital.
neonatal, 135–140
arrhythmias as, 138–139, 139t
congenital heart disease as, 139–140
congestive heart failure as, 135–138, 137t
obesity and, 1042–1043
in renal transplant patient, 629
Cardiovascular physiology, neonatal, 133–134, 134f
Cardioversion, for supraventricular tachycardia, 138
Carney-Stratakis syndrome, 484
Carney triad, 484, 567
Carnitine, in parenteral nutrition, 191–192
Caroli disease, 1331–1332, 1335, 1336
Carotid artery
aneurysm of, 1644–1645, 1645f
coiling of, 1644, 1644f
dissection of, 1644
in extracorporeal life support, 127, 127f
injury to, 717
occlusion of, 1644
Carpenter syndrome, 693
Cartilage, tissue engineering of, 29–31, 30f
Case-control studies, 228–229
Case reports, 227–228
Caspase 8, 410
Casting
for clubfoot, 1705
for developmental dysplasia of hip, 1702
for knee dislocation, 1706
Castleman disease, cervical lymphadenopathy in,
743
Cat-scratch disease, 727–728
hepatic, 1351
lymphadenitis in, 742–743
Catecholamine-resistant shock, 159t, 160–161
persistent, 161
Catecholamines
adrenal regulation of, 558
in burn injury, 380
in neuroblastoma, 444–445
in pheochromocytoma, 559
postoperative elevation of, 105
Catheterizable channels, continent, 1462,
1479–1480, 1481f, 1493–1494, 1493f, 1494f
Catheterization
arterial, 116–117, 214
cardiac
in heart transplantation, 662
in patent ductus arteriosus, 1648
central venous
for intraoperative monitoring, 214
venous thromboembolism with, 175
Catheterization (Continued)
epidural, 225
infections related to
with parenteral nutrition, 193–194
in short bowel syndrome, 1139–1140
pulmonary artery, 117
radial artery, 116–117
umbilical, 116–117, 1634–1635, 1635f
urinary. See Urinary catheterization.
vascular injuries during, 366
venous, for parenteral nutrition, 188–189
Caudal block, 224–225, 224f
in hypospadias repair, 1551
in inguinal hernia repair, 988–989
Caustic injury. See Esophagus, caustic injury to.
Cavitation devices, 50
CCAM volume ratio (CVR), 85
CD11b/CD18, in neutrophil adhesion, 146
CD30, in Hodgkin lymphoma, 519, 519f
CD44, in neuroblastoma, 449
Cecal volvulus, 1117, 1124, 1252
Cecocolic loop, 1113, 1113f, 1114f
Cecostomy
continent, 1482, 1482f
tube, 1237, 1240
Celecoxib, for familial adenomatous polyposis, 488
Celiac artery, stenosis of, 1639–1641, 1640f
Cell cycle, 398
Cell death, programmed, 399
Cell differentiation
extracellular matrix in, 29
multipotent, 28–29, 28f
Cell-mediated immunity, 146–148
Cell physiology, normal, 398–402
Cellulitis
orbital, 713
peritonsillar, 717–718
Central nervous system. See also Brain; Spinal cord.
anomalies of, with cloacal exstrophy, 1527
formation of, 1673–1674, 1674f, 1675f
injuries to, 343–364
in birth trauma, 392
suppurative infections of, 1693–1697
Central venous catheter
for intraoperative monitoring, 214
venous thromboembolism with, 175
Central venous parenteral nutrition, 189, 193–194
Cephalosporins, for urinary tract infection,
1431–1432
Cerebellar ataxia, in neuroblastoma, 443
Cerebellum
astrocytoma of, 594, 595f
tumors of, 594
Cerebral contusion, 344, 345–346, 345f
Cerebral palsy, 392
neuropathic bladder in, 1460–1461
nutritional support in, 199, 199t
Cerebral perfusion pressure, in trauma patient,
268–269
Cerebrocostomandibular syndrome, 807–808, 977t
Cerebrospinal fluid
absorption of, 1681
formation of, 1681
leakage of
in basilar skull fracture, 352–353
after myelomeningocele repair, 1676
in temporal bone fracture, 712
obstruction of, hydrocephalus from, 1681
shunting of
complications of, 1683–1686
for hydrocephalus, 1683
Cerebrovascular disease, 1643–1645, 1643f, 1644f,
1645f
Cerebrovascular injuries, 346–347, 353
Cerumen, removal of, 708
Cervical adenitis, 727. See also Lymphadenitis.
Cervical clefts, midline, 760, 760f
Cervical dermoid cyst, 760
Cervical ectopia cordis, 803
Cervical esophagus, duplications of, 1158
Cervical lymphadenopathy, 737–746
anatomy of, 737, 738f
differential diagnosis of, 737, 738t
evaluation of, 740, 740t
Volume 1, pages 1–768; Volume 2, pages 769–1738
Cervical lymphadenopathy (Continued)
infectious, 743
in inflammatory disorders, 743
in malignant disorders, 743
management of, 740–743, 741f
Cervical spine
control of, in trauma patient, 263–265, 264f
hemivertebrae involving, 765
injury to, 335, 335f, 354, 356–357, 356t,
358–359, 358f
in birth trauma, 392
Cervical thymic cysts, 760–761
Cervical torso vascular injuries, 363
Cervical tumors, in Peutz-Jeghers
syndrome, 1184
Cervicofacial teratoma, 516
Cervicomedullary astrocytoma, 597
Cervix, adenocarcinoma of, 1609
Cesarean delivery
for conjoined twins, 1733
defects managed by, 78t
CFTR gene. See Cystic fibrosis transmembrane
regulator (CFTR) gene.
Cheatle slit, 1067, 1068f
Chédiak-Higashi syndrome, 529
Chemical burns, 383
Chemical exposure, hypospadias and, 1537
Chemoattractants, 149
Chemoembolization, hepatic arterial (transarterial)
for hepatoblastoma, 475
for hepatocellular carcinoma, 479–480
Chemotaxins, in neutrophil diapedesis, 146
Chemotherapy
for bone tumors, 583
for colorectal cancer, 490
common agents for, 406, 407t
for Ewing sarcoma family/primitive
neuroectodermal tumors of chest wall, 575,
575f
for fibrosarcoma of chest wall, 575–576
for hepatoblastoma, 470, 471–472
for hepatocellular carcinoma, 477–478
for Hodgkin lymphoma, 520
hyperthermic intraperitoneal, 503, 504f
for hypothalamic/chiasmatic astrocytoma, 598
for neuroblastoma, 456
for non-Hodgkin lymphoma, 525–526
for primitive neuroectodermal tumors,
594–596
principles of, 405–410, 407t
for pulmonary blastoma, 569–570
radiation therapy with, 412
for rhabdomyosarcoma, 495–496
risk stratification in, 406
for sacrococcygeal teratoma, 512–513, 515f
side effects of, 406, 407t
targeted, 406–410
terminology in, 406
for testicular tumors, 556, 556t
for tuberculosis, 742, 857
for Wilms’ tumor, 434–435, 435t
Chest
examination of, in thoracic trauma, 273
flail, 275
funnel. See Pectus excavatum.
trauma to. See Thoracic trauma.
Chest radiography. See Radiography, chest.
Chest tube
breast deformity from, 771, 772f
for chylothorax, 878
complications related to, 874
for empyema, 872
for hemothorax, 277
for pneumothorax, 275, 276f, 872–873
care and removal of, 874
in neonate, 873–874, 874f
in older child, 875
size of, guide for, 875, 875t
in trauma patient, 265–266, 266f
Chest wall
congenital deformities of, 779–812
in diffuse skeletal disorders, 805–808, 807f, 808f
involving depression. See Pectus excavatum.
involving protrusion. See Pectus carinatum.

Chest wall (Continued)
in Poland syndrome. See Poland syndrome.
sternal, 799–804, 803f, 804f, 805f
rhabdomyosarcoma of, 497
trauma to, 275
tumors of, 572–576, 574t
benign, 573–574, 574f
clinical presentation in, 572–573
diagnosis of, 573
epidemiology of, 572
imaging of, 573
malignant, 497, 575–576, 575f
treatment of, 573
types of, 573–576, 574t
Chiari malformation, 842–843, 1677, 1678f
Child abuse, 385–386
burns in, 369, 370f, 389–390, 390f
cycle of, 386
epidemiology of, 385–386
fractures in, 275, 336, 388–389, 388f, 389f
presentation of, 386–391, 386t
reporting of, 385
thoracoabdominal injury in, 390–391, 390f, 391f
traumatic brain injury in, 349, 353, 387–388,
387f, 388f
Child-Pugh score, 1362
Child safety seats, 258, 258f
Children’s Oncology Group (COG), 235
staging system of
for germ cell tumors, 509, 509f
for liver tumors, 469, 469t
for ovarian cancer, 510, 511f
for testicular cancer, 510, 510f
for testicular tumors, 550, 551t
for Wilms’ tumor, 423, 424t, 429–430
China, pediatric surgery in, 15–16
Chlamydia pneumoniae, 857
Chlamydia trachomatis, perihepatitis and, 1352
Chlorhexidine, for umbilical cord cleansing, 963
Chloride
CFTR gene regulation of, 1074
in parenteral nutrition, 190–191, 190t
Choanal atresia, 713–714, 714f
Cholangiography
in choledochal cyst, 1335
intraoperative
with cholecystectomy, 1344–1345, 1345f, 1345t
in choledochal cyst, 1337, 1338f
with portoenterostomy, 1325, 1325f
Cholangiopancreatography
endoscopic retrograde
in bile duct injury, 299, 300f
in biliary atresia, 1324
in choledochal cyst, 1335
in choledocholithiasis, 1344, 1344f
in pancreas divisum, 1375–1376, 1377f
in pancreatic injury, 303
in pancreatitis, 1373, 1375
magnetic resonance
in choledochal cyst, 1335, 1335f
in pancreas divisum, 1375–1376, 1376f
in pancreatitis, 1373, 1375
Cholangitis
in choledochal cyst, 1334
in cholelithiasis, 1342
after portoenterostomy, 1328–1329
primary sclerosing, in ulcerative colitis, 1219
Cholecystectomy
for biliary dyskinesia, 1343
for choledocholithiasis, 1344, 1344f
for cholelithiasis, 1344
laparoscopic, 1345–1348
Children’s Mercy Hospital experience with,
1348–1349, 1349t
four-port technique in, 1345–1347, 1346f, 1347f
single-site umbilical, 1347–1348, 1347f, 1348f,
1349
stab incision technique in, 1345, 1346f
in sickle cell anemia, 1341–1342, 1344
Cholecystitis
acalculous, 1343
in cholelithiasis, 1342
cholescintigraphy in, 1343
Cholecystolithotomy, 1343–1344
Choledochal cyst, 1331–1342
adult form of, 1333–1334
anatomic classification of, 1332, 1332f, 1333f
carcinoma arising in, 1333, 1339
clinical presentation in, 1334
diagnosis of, 1334–1336
embryology of, 1332
epidemiology and etiology of, 1331–1332
imaging of, 1334–1336, 1335f
infantile form of, 1333–1334
laboratory studies in, 1334
pancreatitis in, 1373
pathology of, 1332–1333
prenatal diagnosis of, 1333–1334
surgical management of, 1336–1338, 1336f,
1337f, 1338f
outcome and complications of, 1338–1339
timing of, 1334
Choledochocele, 1332, 1333f
etiology of, 1331–1332
imaging of, 1335
pathology of, 1333
surgical management of, 1336
Choledocholithiasis, 1344, 1344f, 1345t
Cholelithiasis
after bladder augmentation or replacement, 1485
clinical presentation in, 1342, 1342f
complications of, 1341–1342
hemolytic, 1341
in hereditary spherocytosis, 1342
nonhemolytic, 1341
nonsurgical treatment of, 1343
in pancreatitis, 1373
with parenteral nutrition, 193, 1341
radiographic evaluation of, 1343
in sickle cell anemia, 1341–1342
surgical treatment of, 1343–1345, 1344f, 1345f,
1345t, 1346f
in thalassemia, 1342
Cholescintigraphy, 1343
Cholestasis
intrahepatic, progressive familial, pruritus in,
1343
in necrotizing enterocolitis, 1204
nutritional support in, 197, 197t
with parenteral nutrition, 193
Cholesteatoma, of ear, 711
Cholesterol, steroid hormone synthesis from, 1570f
Choline magnesium trisalicylate, 216, 216t
Chondroblastoma
location of, in relation to physis, 579f
site of involvement of, 579–580
Chondrogladiolar pectus carinatum, 794, 794f
Chondroma, chest wall, 573
Chondromanubrial pectus carinatum, 794, 795f
Chondromyxoid fibroma, location of, in relation to
physis, 579f
Chondrosarcoma
chest wall, 575
cryosurgery for, 585f
epidemiology of, 580
resection and reconstruction of, 578f
Chorda tympani nerve, 707
Chordee
in hypospadias, 1531, 1533f, 1539
repair of, 1546–1550, 1549f, 1550f, 1551f,
1582–1583, 1585f
urethral plate preservation and, 1543, 1544f,
1545, 1548f
without hypospadias, 1546, 1549f
Choriocarcinoma, 508
ovarian, 543
testicular, 552
Chorionic villus sampling, 77
Choristoma, 721
Choroid plexus papilloma, 600–601, 601f,
1680–1681
Choroid plexus tumors, 600–601, 601f
Christmas-tree deformity, jejunoileal atresia and
stenosis with, 1064–1065, 1066f
Chromaffin cells, 558
Chromium, requirements for, 184, 184t
Chromosomal abnormalities
in congenital diaphragmatic hernia, 810
Volume 1, pages 1–768; Volume 2, pages 769–1738
INDEX I-9
Chromosomal abnormalities (Continued)
in esophageal atresia with tracheoesophageal
fistula, 895–896
malignant transformation associated with,
399–400, 401f
Chromosomal translocations, 400–401, 401t
Chromosome 9p deletion syndrome, 977t
Chyle, fat content of, 877
Chylothorax, 286, 876–879
anatomy and pathophysiology of, 877, 877f
clinical manifestations of, 877–878
etiology of, 876–877, 876f
historical perspective on, 876
treatment of, 878–879
Chylous ascites, 1174–1175
Chymotrypsin, stool, in meconium ileus, 1077
Cidofovir, for recurrent respiratory papillomatosis,
844
Cigarette lighters, safety standard for, 258
Cigarette smoking, ulcerative colitis and, 1218
Cimetidine, for peptic ulcer disease, 1033
Ciprofloxacin, for Crohn disease, 1212
Circulation
fetal, 134–135, 136f
persistence of, 135
pulmonary, 114
after separation of conjoined twins, 1735
transitional, 112, 135
Circulatory support, for trauma patient, 266–268,
267f
Circumcision, 1561, 1562f
Circumvallate papillae, 716
Cirrhosis
in choledochal cyst, 1334
liver transplantation for, 644–645, 644f
Cisapride, for intestinal dysmotility, 1140
Cisatracurium, 210t
Cisplatin, 407t
for hepatoblastoma, 470, 471–472
for hepatocellular carcinoma, 477–478
ototoxicity of, 475
for ovarian germ cell tumors, 546–547
Citrate toxicity, 176
Citrulline, serum, in short bowel syndrome, 1135
Clam ileocystoplasty, 1472, 1473, 1474f
Clarithromycin, for peptic ulcer disease, 1033, 1033t
Clatworthy, H. W., 5–6, 6f
Clavicular fracture
in birth trauma, 391
in child abuse, 389
Clear cell renal cell carcinoma, 438
Clear cell sarcoma, 503
renal, 437
Cleft(s)
craniofacial, 695–697. See also Craniofacial clefts.
laryngeal, 850–851, 850f
laryngotracheoesophageal, 916–918, 917f, 918f
sternal, 804, 805f, 805t
Cleft hand, 1722
Cleft lip and palate, 699–707
anatomy of, 699–701, 700f, 701f
embryology of, 699, 700f
epidemiology of, 699
etiology of, 699
multidisciplinary care in, 704–705
omphalocele in, 977t
surgical correction of, 701–704
for bilateral cleft lip, 703, 703f
for cleft palate, 703–704, 704f, 705f
orthopedics prior to, 701–702, 702f, 703, 703f
secondary, 705–706
timing of, 701–702
for unilateral cleft lip, 702, 702f
Clinical guidelines and pathways, 234
Clinical Risk Index for Babies, 90–91
Clinical trials, cancer, 415
Clinodactyly, 1723
Clitoroplasty, in female gender assignment surgery,
1578–1579, 1578f, 1579f
Cloaca, 1289, 1302f, 1604–1606, 1606f, 1607f
with common channel longer than 3 cm, 1304
with common channel shorter than 3 cm,
1302–1303, 1302f, 1303f
posterior, 1577f
I-10 INDEX

Cloaca (Continued) Colon (Continued) Compliance

reconstruction for, 1301–1305 acquired, 1249–1250, 1250f of bladder, 1457, 1458f, 1467
vaginal replacement for, 1304–1305, 1305f, causes of, 1132 of lung, 113, 113f
1306f, 1307f congenital, 1247, 1248f postoperative, in adolescents, 627, 1044–1045,
vaginal switch maneuver for, 1304, 1305f functional, 1250–1252 1045t
Cloacal exstrophy, 973, 974t, 975–976, 982, 983, in meconium ileus, 1252, 1252f Compression therapy, for capillary-
1524–1529 in meconium plug syndrome, 1250–1251, 1251f lymphaticovenous malformation, 1628
anomalies associated with, 1526, 1526t miscellaneous causes of, 1252–1253, 1252f Computational fluid dynamics, 29
embryogenesis of, 1525–1526, 1526f in reversed intestinal malrotation, 1117, 1117f, Computed tomography, 40–43
genetics of, 1526 1124 in abdominal trauma, 289–290, 290f
postoperative care in, 1528–1529 in small left colon syndrome, 1251–1252, 1251f in adhesive bowel obstruction, 1128, 1129f
prenatal diagnosis of, 1527 perforation of, 1248–1249, 1248f in alimentary tract duplications, 834–835, 835f,
surgical repair of, 1527–1528, 1528f polyps of. See Polyp(s), gastrointestinal. 1157–1158, 1157f, 1158f
umbilical abnormalities in, 967 segmental dilatation (ectasia) of, 1252 in aortic trauma, 283, 285f
Cloacal membrane, 1289 stenosis of, congenital, 1247, 1248f in appendicitis, 1258
development of, 1515–1516, 1516f strictures of, 1132 of brain tumors, 592
premature rupture of, 1516, 1525 infectious and inflammatory causes of, in cerebellar astrocytoma, 594
Clonidine 1249–1250 in cervical lymphadenopathy, 739
caudal, 224–225 in meconium ileus, 1083 of chest wall tumors, 573
epidural infusion of, 225 in necrotizing enterocolitis, 1203, 1249, 1250f in choledochal cyst, 1335
Closing capacity, 113 trauma to, 305–308 colorectal cancer after, 489
Clostridium difficile infection, ulcerative colitis and, vaginal replacement with, 1304, 1306f in congenital lobar emphysema, 828–829,
1220 volvulus of, 1132, 1252, 1252f 828f
Cloves syndrome, 1629–1630, 1630f Colon conduit diversion, 1489–1490 in conjoined twins, 1732, 1732f
Clubfoot, 1704–1705, 1705f Colon patch procedure, for long-segment in cystic mediastinal masses, 830, 831f
Coagulation Hirschsprung disease, 1272 in diffuse axonal brain injury, 347–348,
disorders of, 171–175 Colonic interposition 347f, 348f
acquired, 173–174 esophageal, 907, 929–932, 929t, 930f, 931t electron-beam, 42–43, 43f
assays for, 171, 173t for caustic stricture, 925–926, 926f in gastrointestinal trauma, 307–308
genetic, 171–173, 173t left/transverse technique of, 931 in hepatic abscess, 1349f, 1350–1351, 1353,
thrombotic, 174–175 right/retrosternal technique of, 930–931, 930f 1353f
in trauma patient, 269–270, 296, 296f Colonic neobladder, 1473–1474 in hepatic hemangioma, 460, 460f, 1618
disseminated intravascular, 173–174 Colonoscopy in Hodgkin lymphoma, 518–519, 518f
zone of, in burns, 370–371, 371f in gastrointestinal bleeding, 1153 in intestinal rotation and fixation disorders, 1118,
Coagulation cascade, 172f in ulcerative colitis, 1220, 1220f 1120f
Coagulation factor deficiencies Colorectal cancer, 486 in intracranial infections, 1695–1696
genetic, 171–173, 173t diagnosis of, 489 in intussusception, 1101
in liver disease, 174 hereditary associations with, 487–489 in liver and spleen trauma, 290f, 291, 291t
Coagulation proteins, inhibitors of, 174 hereditary nonpolyposis, 488, 489 in Meckel diverticulum, 1089
Coarctation of aorta in inflammatory bowel disease, 1215 in mesenteric and omental cysts, 1168, 1169f
abdominal, 1631–1634, 1632f, 1633f, 1634f intestinal obstruction with, 1132 multidetector, 41–42, 42f
congenital, 1650–1652, 1650f, 1651f nonhereditary associations with, 489 in musculoskeletal trauma, 331–332
traumatic, 283 polypoid disease and, 486–487, 488 in neck mass, 727
Coating, nanoelectromechanical systems for, 62 sporadic, 489–490 in neuroblastoma, 444, 444f, 445f
Coccidioidomycosis spp. infection, pulmonary, 864 treatment of, 490 of ovarian tumors, 533
Cochlea, 707 in ulcerative colitis, 1219 in pancreatic pseudocysts, 1377, 1378f
Cochlear implant, 708–709 Colostography, of anorectal malformations, 1296, in pancreatic trauma, 303, 304t, 305f, 305t
Cochrane Collaboration and Best Evidence, 233 1296f in pancreatitis, 1373, 1375, 1375f
Codeine, 218, 218t Colostomy. See also Enterostoma. percutaneous needle biopsy guided by, 418
Codman triangle, 581 for anorectal malformations, 1293–1294, in pheochromocytoma, 559–560, 559f
Coe, H., 5, 5f 1295–1296, 1296f in pneumothorax, 274
Cognitive development closure of, 1306 in Poland syndrome, 797, 800f
in adolescents, 1044–1045 management after, 1296, 1296f positron emission tomography with, 46, 47f
in renal transplant patient, 629 choices for, 1239f, 1240, 1241f in hyperinsulinism, 1380, 1380f
Cohort studies complications of, 1244–1245, 1245f, 1245t of primitive neuroectodermal tumors, 594
prospective, 230 decompressing, for Hirschsprung disease, 1269, in renal injury, 313
retrospective, 229–230 1269f, 1270 of salivary glands, 730
Cold, neonatal response to, 98–99 descending, 1295–1296, 1296f single photon emission
Cold shock, 159t, 160, 161 indications for, 1237 in epilepsy surgery, 1689
Colectomy. See also Proctocolectomy. loop, 1296 molecular imaging using, 48
for Crohn disease, 1214, 1214t, 1215 parastomal hernia after, 1132 of thyroid gland, 746, 746f
for familial adenomatous polyposis, 488, 1181 stoma care in, 1244 in spine and spinal cord injury, 358, 358f, 359
for juvenile polyposis syndrome, 1183 takedown of, 1242–1244 in thoracic trauma, 274
for ulcerative colitis, 1222 technical aspects of, 1242–1244 three-dimensional, 42, 42f, 43f
Colic transverse, 1296 of thyroid gland, 746, 746f
in mesocolic hernia, 1117 Coma, diffuse axonal injury and, 348 in tracheobronchial vascular compression,
renal, 1434–1437 Combination chemotherapy, 406 853–854
Colitis. See also Enterocolitis; Necrotizing Common cavity phenomenon, manometric, in traumatic brain injury, 349–350, 352
enterocolitis. 947–948, 949f, 950f, 950t in ureteral injury, 319
in Crohn disease, 1213 Common cold, 713 in Wilms’ tumor, 427
ulcerative. See Ulcerative colitis. Commotio cordis, 280 Computed tomography angiography
Collaboration, in patient- and family-centered care, Communication, in patient- and family-centered in portal hypertension, 1361
251–252 care, 248–250 in portal vein thrombosis, 1357, 1357f
Collateral circulation, in portal hypertension, 1356 Community-acquired bacterial pneumonia, in vascular trauma, 362–363, 363f
Collecting system, duplication of. See Duplex 855–858, 856f Computed tomography cystography, in bladder
collecting system. Compartment syndrome, 334 injury, 313
Colloids, for sepsis, 155–158 abdominal, 298–299, 299f, 481, 481f Computer-assisted surgery, 50–52. See also Image-
Colon. See also Sigmoid entries. Compartments, bone tumor resection and, 584, 584f guided therapy.
adenocarcinoma of, 1250 Complement system Concussion, 348, 352
atresia of, 1247, 1248f in host defense, 148 neurologic dysfunction after, 353
duplications of, 1157, 1161–1163 in neonate, 151–152 outcomes from, 354
fibrosarcoma of, congenital infantile, 1250 Complete blood count Conduit diversions, 1489–1490, 1489f
motility of, 1291 in anemia, 165 Confidence interval, 232
obstruction of, 1247–1256. See also Intestinal in coagulation disorders, 171 Conflicts of interest, 242–243
obstruction. in liver tumors, 464 Conformal radiation therapy, 413

Volume 1, pages 1–768; Volume 2, pages 769–1738


Congenital anomalies
of bone, 1699–1712
branchial. See Branchial anomalies.
of breast, 771–772, 772f, 1714–1720, 1717f,
1718f, 1719f, 1721f
communication in, 249, 249t
craniofacial, 689–699
of ear, 708
of esophagus, 893–924
endoscopy of, 883, 884f
laryngotracheoesophageal cleft as, 916–918,
917f, 918f
true congenital stenosis as, 915–916, 916f
of foot, 1703–1705, 1704f, 1705f
of hand, 1720–1724, 1722t, 1723f
incidence of, 1711–1712
of kidney
related to abnormal ascent, 1405–1406, 1407f,
1408f
related to abnormal fusion, 1406–1409, 1408f,
1409f, 1410f
of larynx, 723–725, 724f, 725f
major versus minor, 1711
of nipple, 772, 772f
of nose, 713–715, 714f
of oral cavity and pharynx, 721
of pancreas, 1371, 1372t, 1374, 1375f
parental reactions to, 249, 249t
prenatal diagnosis of, ultrasonography in, 38–39,
39f
prenatal surgical consultation on, 249
of scrotum, 1563
of skin and soft tissue, 1713–1714
of spine, 807, 808f, 1706–1709, 1706f, 1707f,
1708f, 1709f
of umbilicus, 961, 964–968, 965f, 966f
of urogenital sinus, 1470, 1575, 1576f, 1604,
1605f, 1606f
vascular, 1611–1633
Congestive heart failure. See Heart failure.
Conjoined twins, 1725–1738
anesthetic management for, 1733–1734
classification of, 1728–1730, 1728t, 1729f,
1730f
diagnosis of
postnatal, 1731–1733, 1732f, 1733f
prenatal, 1730–1731, 1731f, 1732f
ethical considerations in, 1735–1737
etiology and embryology of, 1726–1728
follow-up for, 1735
heteropagus (parasitic), 1737–1738, 1737t,
1738f
historical perspective on, 1725–1726, 1726f,
1727t, 1728f
incidence of, 1728
nonoperative management of, 1735
obstetric management of, 1733
outcome of, 1736t
separation of
care after, 1734–1735
emergency, 1735
planned, 1735
procedure for, 1734, 1734f
Connective tissue disorders, inguinal hernia in,
1000
Consciousness, evaluation of, 349
Consent, informed, 238–239, 239t
Constipation, 1312–1315
acute, 1313
in anorectal malformations, 1291
after anorectoplasty, 1307–1308, 1308t
in cerebral palsy, 1461
chronic, 1313–1314, 1313t, 1314t
in Hirschsprung disease, 1266
management of, 1314–1315
definitions of, 1312–1313
in dysfunctional elimination syndromes, 1463,
1463f
functional, 1312–1313
in internal anal sphincter achalasia, 1284
in intestinal neuronal dysplasia, 1280
in isolated hypoganglionosis, 1282
in overactive bladder syndrome, 1464
slow-transit, 1314
Constriction band deformities, 1724
INDEX I-11
Continence. See also Incontinence. Craniopharyngioma, 598–599, 599f
anal, 1311–1312 Craniosynostosis, 691–692
after anorectoplasty, 1308–1309, 1308t, 1309f diagnosis of, 693–694
urinary, 1467 etiology and pathologic anatomy of, 691–692
Continent catheterizable channels, 1462, single suture, 692, 692f
1479–1480, 1481f, 1493–1494, 1493f, syndromic, 692–694, 693f
1494f treatment of, 694
Continent cecostomy, 1482, 1482f Craniosynostosis–mental retardation syndrome of
Continent urinary diversions, 1490–1492, 1492f, Lin and Gettig, 977t
1493f, 1494f, 1495f Craniotomy, for brain tumors, 593
Continent urinary reservoirs, 1494, 1495f Cricoid cartilage, 722
Continuous positive airway pressure (CPAP), 118 Cricoid split, for laryngotracheal stenosis, 846–847,
Contractility, cardiac, 134 847t
Contrast studies Cricopharyngeus, disorders of, 942
BOLD magnetic resonance imaging, 44–45 Cricothyroid muscles, 722
in conjoined twins, 1733 Cricothyrotomy, 723
in necrotizing enterocolitis, 1199 needle, 265
ultrasound, 40, 40f surgical, 265
Control mode, in mechanical ventilation, 118 Cricotracheal resection, for laryngotracheal stenosis,
Contusion 847, 847f, 848–849
bladder, 321 Critical illness
in central nervous system injury, 344 fluid and electrolyte balance in, 205
cerebral, 344, 345–346, 345f nutritional support in, 197
pulmonary, 277, 278f Croatia, pediatric surgery in, 15
Conus elasticus, 722 Crohn disease, 1209–1218
Copper, requirements for, 184, 184t cancer and, 489, 1215
Core needle biopsy, 418–420, 419f, 419t diagnosis of, 1211, 1211f
Corner fracture, 388 epidemiology of, 1209
Coronal suture, premature fusion of, 692, 692f etiology of, 1209
Coronary artery disease, after heart transplantation, after ileoanal pouch procedure, 1228–1229
668 Meckel diverticulum in, 1088
Coronary sinus septal defect, 1652–1653 medical treatment of, 1211–1213
Corpora cavernosa, 1537 pathology and clinical features of, 1210–1211,
Corpus callosotomy, 1691 1210f
Corpus luteum cysts, 536 surgical treatment of, 1213–1215, 1213f, 1214f,
Cortical dysplasia, seizures in, 1692 1214t, 1215t
Corticectomy, 1691 Cronkhite-Canada syndrome, 487
Corticosteroids, 407t Cross-sectional studies, 228
for airway obstruction, 723 Croup
for Crohn disease, 1211–1212 membranous, 726
for Diamond-Blackfan anemia, 166–167 viral, 725
for esophageal caustic injury, 922–923 Crouzon syndrome, 693, 693f
for hepatic hemangioma, 1617–1618 Crural sling, 940
for immune thrombocytopenic purpura, 1387 Cryoprecipitate, for von Willebrand disease, 170–171
for infantile hemangioma, 1616 Cryotherapy, 50
for intracranial infections, 1696 for bone tumors, 585, 585f
for intussusception, 1102 for hepatocellular carcinoma, 480
peptic ulcer disease and, 1031 Cryptococcus neoformans infection, 864
prenatal, for congenital diaphragmatic hernia, 817 Cryptorchidism, 1003–1014
for sepsis, 160 acquired, 1006, 1009
in spinal cord injury, 359 associated abnormalities with, 1004–1005
for subglottic hemangioma, 849–850 classification of, 1005–1006, 1005f
in transplantation, 606–607 complications of, 1006–1008, 1007f
liver, 650t diagnosis of, 1008–1009
lung, 674–675, 676–677, 676t embryology of, 1003–1005, 1004f
renal, 625 after hernia repair, 997
for ulcerative colitis, 1221, 1222 historical perspective on, 1003
for unicameral bone cysts, 584 with hypospadias, 1534
Corticotropin. See Adrenocorticotropic hormone incidence of, 1006
(ACTH). inguinal testis trauma in, 1008
Corticotropin-releasing hormone, 558 malignancy risk in, 508
Cortisol in meconium ileus, 1083
adrenocortical production of, 558 in prune-belly syndrome, 1509
in burn injury, 380 testicular cancer and, 549–550, 1008, 1014
elevation of torsion in patient with, 1008
in Cushing syndrome, 561–562, 562f treatment of, 1009–1014
postoperative, 105 cancer after, 1014
functions of, 558 complications of, 1013, 1013t
insufficiency of, 564 fertility after, 1013–1014, 1013t
Cough, 722–723 hormonal, 1009
Couinaud-8 segments, for liver tumors, 469, 469f laparoscopic, 1013
Cowden syndrome, 1630 surgical, 1009–1013, 1011f
Cowper gland anomalies, 1559 Crystalloids
Coxsackie virus infection, oropharyngeal, 716–717 for sepsis, 155–158
Cranial burst fracture, 352 in trauma patient, 267
Cranial neuropathy, in basilar skull fracture, 353 Cultural barriers, 243–244
Cranial vault remodeling, surgical, 692, 694 Curettage, for bone tumors, 585
Craniectomy, for brain tumors, 593 Currarino-Silverman deformity, 780
Craniofacial anomalies, 689–699 Currarino triad, 1162, 1290
Craniofacial clefts, 695–697 Cushing disease, 561
number 7, 696–697, 696f, 697f Cushing syndrome, 561–563, 562f
Tessier classification of, 695–696, 696f Cutaneous. See Skin.
in Treacher Collins syndrome, 697, 697f Cutaneovisceral angiomatosis with
Craniofacial reconstruction, 691 thrombocytopenia, 1620
virtual reality in, 72–73, 72f Cutis marmorata telangiectatica congenita, 1621
Volume 1, pages 1–768; Volume 2, pages 769–1738
I-12 INDEX

Cyanide toxicity, in inhalation injury, 375 Cystolitholapaxy, for bladder stones, 1438 Developing countries, pediatric surgery in, 17, 17f
Cyanosis, in tetralogy of Fallot, 1660 Cystoplasty. See Bladder augmentation or Developmental disability, nutritional support in, 199,
CyberKnife, 47, 52, 52f, 58 replacement. 199t
CyberWare scanner, 72, 72f Cystostomy Dexamethasone
Cyclooxygenase-2, in necrotizing enterocolitis, for echinococcal cyst, 1353 for esophageal caustic injury, 922–923
1190t, 1193 suprapubic, for urethral injury, 323 for intracranial infections, 1696
Cyclophosphamide, 407t Cystourethrography. See Voiding cystourethrography. prenatal, for congenital adrenal hyperplasia,
for neuroblastoma stage IV-S disease, 450 Cytarabine, 407t 1574–1575
Cyclosporine Cytogenetics, 404t Dexamethasone suppression test, in Cushing
in transplant patient, 608, 609f Cytokines syndrome, 561–562
heart, 665, 667t in host defense, 148–149 Dexmedetomidine
liver, 649, 650t in necrotizing enterocolitis, 1190t, 1191–1193 for burns, 382–383
lung, 676–677, 676t in neonate, 151 for emergence delirium, 209
renal, 624 in stress response, 104–105 Dextrocardia, in congenital diaphragmatic hernia, 814
in ulcerative colitis, 1222 Cytomegalovirus infection Dextrose. See Glucose (dextrose).
CYP21 gene, in congenital adrenal hyperplasia, 1569 pulmonary Diabetes mellitus
Cystadenoma and cystadenocarcinoma, pancreatic, in cancer patient, 861 after lung transplantation, 680
1382 in HIV-infected patient, 861, 864 maternal, small left colon syndrome and,
Cystadenoma lymphomatosum, papillary, 733 transfusion-related, 177 1251–1252, 1251f
Cyst(s) and cystic lesions in transplant patient in neonate, 101–102
bile duct. See Choledochal cyst. heart, 667 obesity and, 1043
bone. See Bone cyst. renal, 628, 650–651, 651t sacral agenesis and, 1460
dermoid Czech Republic, pediatric surgery in, 15 type 1, pancreas and islet cell transplantation for,
cervical, 760 631–647
nasal, 714, 714f Diagnostic technological innovations, 38–48
echinococcal, 859, 859f D Diagnostic tests, rating of, 227, 228t
epidermoid Da Vinci surgical system, 58–59, 58f, 59f Dialysis
hepatic, 462 Dacarbazine, 407t peritoneal
testicular, 550–551, 553, 554f Daclizumab inguinal hernia and, 999–1000
Gartner duct, 1558, 1608 in liver transplantation, 650t peritonitis with, 1232–1233, 1233t
hepatic in renal transplantation, 624 renal transplantation and, 619
congenital, 464, 465f Dactinomycin, 407t renal cystic disease and, 1403
nonparasitic, 462 Damage-control strategies renal transplantation and, 619
mediastinal, 825–839. See also Mediastinal mass for abdominal trauma, 294–298, 296f, 297f, 297t Diamond anastomosis, duodenoduodenostomy
(es), cystic. for thoracic trauma, 274–275 with, 1055–1056, 1055f
mesenteric. See Mesenteric and omental cysts. Dancing eye syndrome, 443 Diamond-Blackfan anemia, 166–167
Montgomery, 776 Dantrolene, for malignant hyperthermia, 211, 211t Diaphragm
neck. See Neck, cysts and sinuses of. Dartos fascia, 1537 crural, 947–948, 948f
neurenteric, 835, 1679 Data, subjective versus objective, 233 development of, 811
omental. See Mesenteric and omental cysts. Data Knife, 61, 61f and esophagus, functional relationship between,
ovarian. See Ovarian cysts. Data mining, 230, 232 939
pancreatic, 1376–1378, 1382–1383 Databases for quality improvement and outcomes eventration of, 815, 824
pericardial, 832 research, 235–236, 235f trauma to, 279, 280f, 280t, 308–309
pharyngeal, 758 Dataglove, 70–71, 70f Diaphragmatic crural sling, 940
pulmonary, 825–839. See also Lung, cystic Daunomycin, 407t Diaphragmatic hernia
lesions of. DAX-1 gene, in ovarian differentiation, 1565–1567 congenital, 809–832
renal. See Kidney, cystic disease of. Dead space, pulmonary, 114–115 associated anomalies with, 810
salivary gland, 731–732, 732f Death of child, communication in, 249 diagnosis of, 813–815
splenic, 1386 Debridement postnatal, 810f, 814–815
thymic in burn care, 379 prenatal, 79f, 813–814, 814f, 815
cervical, 760–761 for open fractures, 334 differential diagnosis of, 815
mediastinal, 830–832, 831f, 832f Debulking procedure, for capillary- embryology of, 811–812
thyroglossal duct, 721, 755–756, 755f, 756f lymphaticovenous malformation, epidemiology of, 810
urethral, 1558 1628–1629, 1628f extracorporeal life support for, 124, 131, 821
Cystic fibrosis Decannulation, accidental, after tracheotomy, 839 fetal interventions for, 85–86, 817, 823
bronchiectasis in, 865, 866f Decision making, phases of, 244 genetics of, 810
diagnosis of, 1076–1077 Defecation. See also Constipation; Fecal entries; Stool. historical perspective on, 809
genetics of, 1073–1074 physiology of, 1312 left-sided, 812, 812f, 813f, 815
hemoptysis in, 867, 867f Denmark, pediatric surgery in, 13 outcome of, 821–823, 822f
inguinal hernia in, 1000–1001 Denys-Drash syndrome, Wilms’ tumor in, 424–425 pathology of, 812–815, 812f, 813f
lung transplantation for, 671–672 Dermal graft pectus excavatum in, 779–780, 780t
meconium ileus in. See Meconium ileus. for chordee repair, 1547–1550, 1551f prognostic factors in, 815–817
meconium plug syndrome and, 1250–1251 for vaginoplasty, 1595 anatomic, 815–816
molecular genetics of, 20, 20f Dermal regeneration templates, bilayered, 1712, physiologic, 816
peritonitis in, 1233 1713f pulmonary function, 816–817
pneumothorax in, 873 Dermal sinus tracts, 1679 recurrent, 823
prenatal diagnosis of, 1075 Dermatitis, streptococcal, perianal, 1318, 1318f repair of, 819–821, 820f
pulmonary disease in, 864–865 Dermatofibrosarcoma protuberans, 505 extracorporeal membrane oxygenation
rectal prolapse in, 1316 Dermis, 370, 1711–1712 during, 128
Cystic fibrosis transmembrane regulator (CFTR) Dermoepidermal junction, 370 gastroesophageal reflux disease after, 822,
gene, 864 Dermoid cyst 822f, 958
in cystic fibrosis, 20, 20f cervical, 760 timing of, 819
in meconium ileus, 1073–1074 nasal, 714, 714f right-sided, 812, 815, 823
in pancreatitis, 1373 Desflurane, 202t, 207t, 208 treatment of, 817–824
testing for, 1076–1077 Desmoid tumors, 503–504, 504f fetal, 817
Cystic hygroma, fetal, 1622 chest wall, 573–574 future, 823–824, 823f
Cystic nephroma, 439–440, 439f, 1401–1402, 1402f in Gardner syndrome, 1182 postoperative, 820–821
Cystic neuroblastoma, 450f, 451 Desmoplastic small round cell tumor, 503, 503f, 504f preoperative, 817–819
Cystine stones, 1437–1438 Desmopressin surgical, 819–821, 820f
Cystoduodenostomy, for pancreatic pseudocysts, for nocturnal enuresis, 1464–1465 of Morgagni, 815, 824
1378 for von Willebrand disease, 170–171 Diaphysis, 327, 328f
Cystography, computed tomography, in bladder Desmosis coli, 1278 fracture of, 332, 388
injury, 313 Detrusor-sphincter dyssynergy, 1455–1456, 1458f, Diarrhea
Cystojejunostomy, for pancreatic pseudocysts, 1378 1459, 1467, 1468, 1469, 1470f intractable, in neuroblastoma, 443

Volume 1, pages 1–768; Volume 2, pages 769–1738


Diarrhea (Continued)
in short bowel syndrome, 1140
in ulcerative colitis, 1219–1220
Diastematomyelia, 1707
Diazepam, for burns, 382–383
Diet
intussusception and, 1097
ulcerative colitis and, 1222, 1227
Diethylenetriamine pentaacetic acid, 1431
Diffusion, pulmonary, 114
Digit flexor mechanism, disruption of, 338–339,
338f
Digital gigantism, 1723–1724
Digital ischemia, 339, 367
Dignity, in patient- and family-centered care, 248
Digoxin (digitalis)
for congestive heart failure, 135–138, 137t
for supraventricular tachycardia, 138
Dihydrotestosterone, in sexual differentiation,
1567
Dimeglio clubfoot procedure, 1705
Dimercaptosuccinic acid (DMSA)
in multicystic dysplastic kidney, 1400–1401
in pyelonephritis, 1427, 1431
Dimercaptosuccinic acid (DMSA) renal scan, in
ureterocele, 1449–1450, 1449f
Diphenoxylate hydrochloride, in ulcerative colitis,
1222
Disability
developmental, nutritional support in,
199, 199t
in emergency management, 268
Disc batteries, 715
Disconnection surgery, for epilepsy, 1691–1692
Disimpaction, for constipation, 1314–1315
Dismembered pyeloplasty, for ureteropelvic junction
obstruction, 1421, 1422f, 1423
Disorders of sex development (DSD)
46,XX (overandrogenization), 1569–1570, 1573,
1574–1575
46,XY (male pseudohermaphroditism),
1570–1571, 1573–1574, 1575
classification of, 1567–1569
diagnosis of, 1568t, 1571–1574, 1572t
genetic mutations in, 1572, 1572t
inguinal hernia in, 1001
medical management of, 1574–1575
ovotesticular, 1568t, 1571, 1574, 1575
pathophysiology of, 1567–1571
sex chromosome, 1568t, 1571, 1574, 1575
surgical treatment of, 1568t, 1575
for gender assignment. See Gender assignment
surgery.
imaging evaluation for, 1575, 1576f, 1577f
laparoscopy in, 1575
panendoscopy in, 1577, 1577f
preparation for, 1575–1577
Distraction osteogenesis, 695, 1712
Diuretic renography, in ureteropelvic junction
obstruction, 1416–1417, 1416f, 1417f,
1418f
Diuretics, for congestive heart failure, 135, 137t
Diverticular defects, in alimentary tract duplications,
1156
Diverticulectomy
incidental, 1088
for Meckel diverticulum, 1092
Diverticulum
caliceal, 1403
Meckel. See Meckel diverticulum.
urethral
in boys, 1557
in girls, 1557
after hypospadias repair, 1553
DNA
abnormal content of, 400
amplification of, 401
histone modification of, 402
methylation of, 402
mutations of, 399–400, 400t, 401f
replication of, 398
DNA microarrays, 48
DNET (dysembryoplastic neuroepithelial tumor),
591, 599–600, 599f, 1692
INDEX I-13
Dobutamine Duplex collecting system, 1429–1430, 1430f.
for congestive heart failure, 135–138, 137t See also Ureteral duplication.
for fluid-refractory shock, 159–160 complete, 1444–1447
Docetaxel, 407t definition of, 1441
Domectomy, in prune-belly syndrome, 1507 incomplete, 1444–1447
Domperidone, for intestinal dysmotility, 1140 renal dysplasia with, 1442f, 1443
Donnai-Barrow syndrome, 977t upper pole ureter ectopia with, 1445–1447, 1446f
Dopamine, 558 ureteropelvic obstruction with, 1447
for congestive heart failure, 135–138, 137t vesicoureteral reflux with, 1441, 1444–1445,
for fluid-refractory shock, 159–160 1444f, 1445f
Dorsal lumbotomy approach, 1423, 1424f Dura, in craniosynostosis, 691–692
Double aortic arch, 853, 854, 1665, 1667, 1667f Dysgerminoma. See Seminoma (dysgerminoma).
Double bubble sign, in duodenal obstruction, 1053, Dysmorphic syndromes, ovarian cysts in, 529
1053f, 1117–1118, 1119f Dysphagia
Doughnut sign, in intussusception, 1100, 1100f after esophageal atresia repair, 914–915
Down syndrome esophagoscopy in, 882
anorectal malformations with, 1289 after fundoplication, 955
atrioventricular septal defect with, 1658
duodenal atresia and stenosis with, 1051
jejunoileal atresia and stenosis with, 1059–1060 E
nutritional support in, 199, 199t Ear, 707–712. See also Hearing loss.
Doxazosin, for dysfunctional elimination syndromes, anatomy of, 707
1463–1464 anomalies of, 708
Doxorubicin, 412 cholesteatoma of, 711
Dressings embryology of, 708
for burns, 377–379, 377t, 378f examination of, 708–709
after hypospadias repair, 1551 fetal, ultrasonography of, 39, 39f
mermaid, 1582 middle, infection of. See Otitis media.
Drug(s) trauma to, 711–712, 711f
abuse of, abdominal wall defects and, 976 tumors of, 712
overdose of, prevention of, 259–260 Ear tags, preauricular, in craniofacial cleft number 7,
pancreatitis from, 1373 696–697, 696f
Drug delivery systems, microelectromechanical, Ecchymosis, orbital, in neuroblastoma, 442–443,
61–62 442f
Drug eluting stent, 62 ECE1 gene, in Hirschsprung disease, 21t
Dual energy x-ray absorptiometry, 180 Echinococcal infection
Ductus arteriosus, patent, 1647–1649 hepatic, 1352–1353, 1353f
anatomy of, 1647, 1648f pulmonary, 859, 859f
management of, 1648–1649, 1648f, 1649f Echocardiography
natural history and diagnosis of, 1647–1648 in aortic trauma, 283
results of, 1649 in conjoined twins, 1731–1732
Duhamel, B., 12–13, 13f in heart transplantation, 667
Duhamel procedure in pectus excavatum, 784
complications of, 1274, 1275f in prenatal diagnosis, 78
for Hirschsprung disease, 1269f, 1270 in thoracic trauma, 274
Duke Abdominal Assessment Scale (DAAS), 1200 Ectopia cordis
Duodenal atresia and stenosis, 1051–1060 cervical, 803
clinical presentation in, 1053–1054, 1053f, 1054f thoracic, 800–803, 803f, 973, 974f, 974t,
complications of, 1056–1057 975–976, 978, 981, 983
embryology of, 1051, 1052t thoracoabdominal, 803–804, 804f
genetics of, 1051 Ectopia vesicae, rectal prolapse in, 1316
historical perspective on, 1051 Ectopic testis, 1005–1006, 1005f, 1009
incidence of, 1051 Ectrodactyly, 1724
outcomes of, 1057 EDN3 gene, in Hirschsprung disease, 21t
spectrum of, 1052–1053, 1052f, 1053f EDNRB gene
treatment of, 1054–1056, 1054f, 1055f, 1056f in Hirschsprung disease, 21t
Duodenal switch, 1046 in intestinal neuronal dysplasia, 1280
Duodenoduodenostomy Education, in pediatric surgery, 6–9, 73–74
for duodenal atresia and stenosis, 1055 EEA stapler, in pancreas transplantation, 633, 633f
laparoscopic, 1056 EGlass II, 70f
Duodenojejunal loop, 1111, 1112f Ehlers-Danlos syndrome
Duodenojejunostomy, for duodenal atresia and abdominal aortic aneurysm in, 1635
stenosis, 1055, 1055f inguinal hernia in, 1000
Duodenostomy, choledochal cyst excision with, peripheral aneurysms in, 1643
1336, 1336f Eisenmenger syndrome
Duodenum lung transplantation for, 672
diverticularization of, 304f in patent ductus arteriosus, 1647–1648
duplications of, 1156–1157, 1159–1160, ELA-Max, 221, 221t
1159f Elbow, injury to, 331–332, 331f
obstruction of. See also Duodenal atresia and Electrical burns, 383–384
stenosis. Electrocardiography, in trauma patient, 268
double bubble sign of, 1053, 1053f, Electrocautery, 49
1117–1118, 1119f eschar excision with, 379
by Ladd bands, 1116–1117, 1122, 1123f Electroencephalography, in epilepsy surgery,
perforation of, with duodenal atresia and stenosis, 1687–1688
1052, 1052f intracranial, 1689–1690
polyps of, 1181 video-assisted, 1688, 1690
transplantation of, with pancreas transplantation, Electrolytes. See also Fluid management or
631–632, 633f, 634f resuscitation.
trauma to, 299–302, 301t in critically ill infant, 205
computed tomography in, 300, 301f, 302f, in hypertrophic pyloric stenosis, 1024
302t after jejunoileal atresia and stenosis repair,
surgical treatment of, 300–301, 302, 303f, 304f, 1069–1070
304t in neonate, 91–95
Volume 1, pages 1–768; Volume 2, pages 769–1738
I-14 INDEX

Electrolytes (Continued)
with parenteral nutrition
disturbances of, 192–193
requirements for, 190–191, 190t
in premature infant, 205
after renal transplantation, 623
Elejalde syndrome, 977t
Elliptocytosis, hereditary, 1387
Embolization. See also Chemoembolization.
in abdominal trauma, 294–296, 296f
of arteriovenous malformation, 1626–1627
bronchial artery, for hemoptysis, 867
of hepatic hemangioma, 1617–1618
of infantile hemangioma, 1616
portal venous, for hepatocellular carcinoma, 480
Embolus, air, in pulmonary laceration, 277
Embryonal carcinoma, 507, 508f
ovarian, 543
testicular, 552
Embryonal rhabdomyosarcoma, 491–492, 494
Embryonal sarcoma, hepatic, 480
Embryonal tumors, 591
Emergence delirium, 209
Emergency management
airway and cervical spine control in, 263–265,
264f
breathing support in, 265–266, 266f
circulatory support and vascular access in,
266–268, 267f
coagulopathy and, 269–270
damage control in, 270
disability/neurologic evaluation in, 268
exposure in, 268
impact of, 262–263
neuroresuscitation in, 268–269
of pain, 270
prehospital care in, 263
primary survey and treatment of life-threatening
injuries in, 263–268
resuscitation phase in, 268
resuscitation principles in, 263–270
of spinal cord injury, 344
of thoracic trauma, 272–274, 273t
of traumatic brain injury, 344
Emergency personnel, objectives of, 263
Emesis
bilious
in adhesive bowel obstruction, 1127–1128
in jejunoileal atresia and stenosis, 1061, 1061t
in Ladd bands, 1117
in midgut volvulus, 1116
bloody-appearing, evaluation of, 1147–1148
in brain tumors, 591–592
in hypertrophic pyloric stenosis, 1022
in intussusception, 1099
nonbilious, causes of, 1022
EMLA (eutectic mixture of local anesthetics), 221,
221t
Emphysema, lobar, congenital, 825, 828–829, 828f
Empyema, 870–872, 871f
subdural, 1693–1694, 1695, 1695f, 1696
Encephalocele, 715
Encephalopathy, in portal hypertension,
1359, 1360
Enchondroma
location of, in relation to physis, 579f
multiple, 579
Encopresis, 1313, 1315
End-of-life care
communication in, 249
ethics in, 240–241
End-tidal carbon dioxide, 116
Endocardial cushion defect, 1657–1659
EndoCinch system, 57, 957
Endocrine disrupters, in hypospadias, 1536–1537
Endocrine response to surgery, 105
Endodermal sinus tumors. See Yolk sac tumors.
Endolumenal therapies, 57
Endometrial stromal sarcoma, ovarian, 547
Endometriosis, 537
Endophthalmitis, in liver abscess, 1351
Endoprostheses
for bone tumors, 587–588, 588f, 589f
extensible, 588f, 590
Endorectal pull-through. See Pull-through, endorectal. Enteral nutrition (Continued)
Endoscopic injection, in megaureter repair, 1502, for intestinal failure–associated liver disease,
1502f 1138–1139
Endoscopic retrograde cholangiopancreatography. after intestinal transplantation, 655
See Cholangiopancreatography, endoscopic after jejunoileal atresia and stenosis repair, 1070
retrograde. for meconium ileus, 1081–1082
Endoscopy. See also specific types, e.g., Esophagoscopy. for short bowel syndrome, 198, 1138
airway, 837 Enteric duplications. See Alimentary tract
in airway obstruction, 723 duplications.
in bleeding ulcer, 1034 Enteric infections, hematochezia associated with,
in esophageal caustic injury, 921–922, 922t 1152–1153, 1153t
in gastroesophageal reflux disease, 952–953, 953f Enterocolitis
in gastrointestinal bleeding, 1153 in Hirschsprung disease, 1266, 1277, 1277t
in laryngotracheoesophageal cleft, 913–914, 917f in isolated hypoganglionosis, 1282
in motility disorders, 941 necrotizing. See Necrotizing enterocolitis.
in peptic ulcer disease, 1032 Enterocystoplasty. See Bladder augmentation or
in portal hypertension, 1361 replacement.
subcutaneous, 54–55, 55f Enteroliths, in Meckel diverticulum, 1092
transaxillary, 54–55, 55f Enteroplasty
virtual, 42, 43f serial transverse, 1142–1144, 1144f
in Crohn disease, 1211, 1211f tapering, for jejunoileal atresia and stenosis, 1067,
in gastrointestinal bleeding, 1154 1068, 1069f
Endosurgery, natural orifice translumenal, 56–57 Enterostoma, 1235–1249
Endothelial cells care of, 1244
development of, 1620 child with, 1235–1236
lymphatic, 1620 choices for, 1238–1240, 1240f, 1241f
Endothelial cords of Billroth, 1385 closure of, 1244
Endothelin gene, in Hirschsprung disease, 1266 in necrotizing enterocolitis, 1203
Endotoxins, bacterial, 150 complications of, 1203, 1244–1245, 1245f, 1245t
Endotracheal intubation historical perspective on, 1235
in central nervous system injury, 344 increased output with, in short bowel syndrome,
in congenital diaphragmatic hernia, 817–818 1140
in fluid-refractory shock, 160 indications for, 1236–1238
laryngotracheal stenosis with, 845 in jejunoileal atresia and stenosis, 1068–1069,
in respiratory failure, 120 1069f
in spinal cord injury, 359 laparoscopic, 1244
in thoracic trauma, 273 in meconium ileus, 1080–1081, 1080f
in trauma patient, 264, 264f in necrotizing enterocolitis, 1202–1203
in upper airway obstruction, 723 sites for, umbilical, 971
Endotracheal tube technical aspects of, 1240–1244, 1241f, 1242f,
cuffed versus uncuffed, 120 1243f, 1245f
selection of, 263 types of, 1236, 1236t, 1237f, 1238f, 1238t, 1239f
Endovascular procedures, vascular injuries during, Enterotomy, for meconium ileus, 1079–1080, 1080f
365–366 Enteryx systems, 57, 957
Enema Enuresis. See also Incontinence.
air nocturnal, bladder dysfunction in, 1464–1466
intestinal perforation with, 1108 Envenomation injuries, 340–341
in intussusception, 1103, 1103f Environmental factors
antegrade continence, continent cecostomy for, in alimentary tract duplications, 1156
1482, 1482f in hypospadias, 1536–1537
contrast. See also Barium enema. in ulcerative colitis, 1218
in Hirschsprung disease, 1267, 1267f Eosinophilic esophagitis, 944, 944f, 952–953
in intestinal atresia, 1249 Ependymomas, 596–597, 596f
in long-segment Hirschsprung disease, 1272, Epidermal growth factor, in necrotizing enterocolitis,
1272f 1189–1190, 1190t, 1207
in meconium ileus, 1075–1076, 1076f Epidermis, 369–370
delayed repeat, for intussusception, 1105 coagulation necrosis of, 370
Energy expenditure development of, 1711
estimation of, 180 Epidermoid cysts
in neonate hepatic, 462
activity-based, 97 testicular, 550–551, 553, 554f
resting, 97–98, 98f Epididymis
surgery and, 103–104, 104f deficiency of, cryptorchidism and, 1005
resting, 180 testis connection with, in cryptorchidism, 1008
Energy metabolism, in neonate, 95–98 Epididymo-orchitis, 1015
intake and, 95–96 Epidural abscess
losses and, 97 intracranial, 1693–1694, 1694f, 1695, 1696
storage and, 96, 96f intraspinal, 1697
Energy requirements, 180–181, 181t Epidural anesthesia, 225–226, 225t
after burn injury, 381–382, 381t Epidural catheter, 225
Energy stores, by gestational age, 96, 96f Epidural hematoma, 351
Engineering, tissue. See Tissue engineering. Epigastric hernia, 970
Enneking staging system, 582 Epigenetic alterations, 402
Enoxaparin, for renal vein thrombosis, 1439–1440 Epiglottitis, 725, 726
Enoximone, for septic shock, 161 Epilepsy surgery, 1687–1693
Enteral nutrition, 184–188. See also Formula(s). disconnection surgery in, 1691–1692
administration of, 187, 187t preoperative evaluation in, 1687–1690, 1688f
breast milk in, 187–188 resection surgery in, 1690–1691
after burn injury, 381–382, 381t vagus nerve stimulation in, 1692–1693
complications of, 188 Epinephrine, 558
for Crohn disease, 1212 for fluid-refractory shock, 159–160
delivery modalities for, 186 Epiphysiodeses, contralateral, for bone tumors, 588
formulas for, 184–187, 185t, 186t Epiphysis, 327, 328f
indications for, 184–186 fracture of, 328, 329f, 388

Volume 1, pages 1–768; Volume 2, pages 769–1738

 INDEX I-15

Epipodophyllotoxins, 407t Esophageal atresia (Continued) Esophagogastric junction, 947–948, 948f

Episcleritis, in Crohn disease, 1211 with long gap, 902f, 905–907, 906f, 907f Esophagography, 881
Epispadias repair, 1521–1522, 1522f, 1523f by replacement, 907, 908f, 927. of congenital esophageal stenosis, 915, 916f
Epistaxis, 715 See also Esophageal replacement. of esophageal perforation, 890, 891f
Epithelial disorders, intestinal transplantation for, with upper pouch fistula, 910–911, 911f of H-type tracheoesophageal fistula, 909
653, 654f without fistula, 907–908, 909f of tracheoesophageal fistula, 898, 900f
Epithelial-stromal ovarian tumors outcomes of, 911 Esophagomyotomy, for esophageal atresia
laboratory tests in, 530–531, 530t preoperative treatment of, 899 with distal fistula, 899–901, 902f
of low malignant potential, 538–539, 538f tracheomalacia from, 851–852 with long gap, 905, 906, 906f
staging of, 534, 535t Waterston risk groups for, 895, 895t, 897–898 Esophagoscopy, 881–889
surface, 537–538 Esophageal construct, tissue-engineered, 32 complications of, 887
Epithelial testicular tumors, 550–551 Esophageal dysmotility, 939–951 in congenital upper gastrointestinal anomalies,
Epithelioid sarcoma, 503 anatomic basis of, 940 883
Epstein-Barr virus infection distal, 944–946, 945f, 946f in dysphagia, 882
in lymphoma, 522–523 after esophageal atresia repair, 914–915, 944 in esophageal caustic injury, 882–883, 921–922,
oropharyngeal, 716–717 evaluation of, 941–942, 941f, 942f, 943f 922f
post-transplant lymphoproliferative disorders historical perspective on, 939 in gastroesophageal reflux, 882, 882f, 883f
and, 628–629 physiologic basis of, 940 historical perspective on, 881
in transplant patient primary, 942–943, 943f indications and applications of, 882–885
kidney, 650–651, 651t secondary, 943–944, 944f diagnostic, 882–883, 882f, 883f, 884f
lung, 679 Esophageal impedance monitoring, 882 therapeutic, 883–885, 884f
Epulis, 720–721, 720f Esophageal manometry, 881, 947, 948f, 949t instrumentation for, 885–886
Equipoise, clinical, 245 in achalasia, 945–946, 946f patient preparation for, 886
ERBB2 gene, amplification of, 401 common cavity phenomenon and, 947–948, 949f, rigid, 885–887
Errors 950f, 950t technical considerations in, 886–887
surgical, ethics in, 244–245 in cricopharyngeal disorders, 942 in upper gastrointestinal bleeding, 883
type I and II, 233 in gastroesophageal reflux disease, 952 Esophagus
Erythema nodosum in motility disorders, 941–942, 941f, 942f, 943f anatomy of, 940
in Crohn disease, 1211 Esophageal overdrive pacing, for supraventricular Barrett. See Barrett esophagus.
in ulcerative colitis, 1219 tachycardia, 138 body of
Erythroblastopenia of childhood, transient, 166–167 Esophageal pH monitoring, 881–882 anatomy of, 940
Erythrocyte(s) Esophageal replacement, 927–941 disorders of, 942–944, 943f, 944f
enzyme deficiencies in, 169 for atresia, 907, 908f, 927 caustic injury to, 919–929
splenic maintenance of, 1385 Barrett esophagus after, 483 Barrett esophagus after, 483
transfusion of, 175–177 after caustic injury, 924, 925–926, 926f, 927 causes of, 919–920, 920t
in cancer or immunodeficient patient, 176 colonic interposition for, 907, 929–932, 929t, clinical presentation in, 920–921
choice of product for, 176 930f, 931t complications of, 923–924, 923f, 924f, 925t
complications of, 176–177 for caustic stricture, 925–926, 926f diagnosis of, 882–883
fresh whole blood for, 176 gastric transposition for, 907, 908f, 929t, dysmotility after, 944
frozen deglycerolized, 176 934–938, 936f, 937t epidemiology of, 919
intraoperative, 206, 206t gastric tube for, 907, 929t, 932–934, 933f, 934f, historical perspective on, 919
leukocyte-reduced, 176 934t initial management and diagnosis of, 921–922,
packed, 175, 176 ideal, characteristics of, 928 921f, 922f, 922t
hematocrit and, 206, 206t indications for, 927–928 long-term outcome of, 924, 925f
in trauma patient, 267 jejunal interposition for, 907, 929t, 934, 934t pathophysiology of, 920
washed, 176 for peptic strictures, 927 results of, 924–926, 926f
underproduction of, 165–168 positioning routes for, 928, 929t treatment of, 922–923, 927
Erythromycin timing of, 929 congenital anomalies of, 893–924.
for intestinal dysmotility, 1140 types of, 928, 928f, 929t See also Esophageal atresia;
for intestinal pseudoobstruction, 1134 Esophageal sphincter Tracheoesophageal fistula.
Erythropoietin, in necrotizing enterocolitis, lower endoscopy of, 883, 884f
1190–1191, 1190t anatomy of, 940 laryngotracheoesophageal cleft as, 916–918,
Escharotomy, 372, 373f, 379, 379f length of, 947, 949t 917f, 918f
Esmolol, for supraventricular tachycardia, 138, 139t manometry of, 941, 941f, 942f, 947, 948f, 949t true congenital stenosis as, 915–916, 916f
Esophageal anastomosis physiology of, 940 dilatation of, 883
for esophageal atresia transient relaxations of, 947–948, 949f, 950f, for achalasia, 946
with distal fistula, 902–903, 903f 950t for anastomotic stricture, 912
with long gap, 906–907, 907f pathologic, 948–949. for caustic stricture, 923, 923f, 924f
without fistula, 908 See also Gastroesophageal reflux disease. for congenital esophageal stenosis, 915–916
leaks of, 911–912, 911f upper with direct endoscopic visualization, 884, 884f
stricture with, 912, 912f anatomy of, 940 with fluoroscopic control, 883–884
Esophageal atresia. See also Tracheoesophageal fistula. manometry of, 941, 941f with guidewire left in situ, 884
associated anomalies with, 896–897, 897t physiology of, 940 distal, disorders of, 944–946, 945f, 946f
classification of, 894, 895, 895f, 895t, 897–898, Esophageal stricture embryology of, 939
897t, 898t anastomotic, 912, 912f evaluation of, 881–882. See also Esophagoscopy.
complications after repair of, 911–915 caustic, 919–929 foreign body in
early, 911–913, 911f, 912f complications of, 923–924, 923f, 924f, 925t endoscopic removal of, 885
late, 913–915, 914f esophageal replacement for, 925–926, 926f, 927 esophageal replacement for, 928
diagnosis of, 898–899, 899f, 900f prevention of, 922–923 perforation by, 890
embryology of, 895–896 dilatation of. See Esophagus, dilatation of. innervation of, 939
epidemiology of, 896 peptic, esophageal replacement for, 927 laser ablation of, 885
gastroesophageal reflux disease in after portal hypertension surgery, 1367 motility of, 940
preoperative, 912 Esophageal varices. See also Varices. disorders of. See Esophageal dysmotility.
after repair, 913, 957–958 banding of, 1363 evaluation of, 881, 941–942, 941f, 942f, 943f
historical background on, 893–895, 894f bleeding from, 1150 nutcracker, 942–943, 943f
laryngeal cleft with, 850 esophageal replacement for, 928 perforation of, 889–893
long-gap injection sclerotherapy for, 885 caustic, 921–922, 924, 925–926, 925f
surgical management of, 927, 928t in portal hypertension, 1150 classification and incidence of, 889
tissue-engineered esophageal construct for, 32 after portoenterostomy, 1329 clinical findings in, 889
natural orifice translumenal endosurgery for, Esophagitis diagnosis of, 889–890, 890f, 891f
40–41 classification of, 952–953 results of therapy for, 891–892
operative repair of, 899–911 reflux treatment of, 891
with distal fistula, 899–905, 901f, 902f, 903f, bile reflux in, 882 physiology of, 940
904f esophagoscopy in, 882, 882f, 883f rupture of, 889–893

Volume 1, pages 1–768; Volume 2, pages 769–1738

I-16 INDEX
Esophagus (Continued)
classification and incidence of, 889
clinical findings in, 889
diagnosis of, 889–890, 890f, 891f
results of therapy for, 891–892
treatment of, 891
spasm of, diffuse, 942
stenosis of. See also Esophageal stricture.
congenital, 915–916, 916f
trauma to, 279, 883
tumors of, 483
Barrett esophagus and, 956
after caustic injury, 924
esophageal replacement for, 928
smooth muscle, 483
upper, disorders of, 942
EsophyX, 57, 957
Estradiol, ovarian tumors and, 530, 531t
Estrogen
for labial adhesions, 1558
secretion of, ovarian tumors associated with, 530
Ethanol, percutaneous injection of, for hepatocellular
carcinoma, 480
Ethanol-lock therapy, 193–194, 1140
Ethics, 237–248
in bariatric surgery, 241–242
in conflicts of interest, 242–243
in end-of-life care, 240–241
in informed consent and assent, 238–239, 239t
in innovation and research, 245
in multiculturalism, 243–244
in prenatal surgical consultation, 239–240
principles of, 237
resolution of dilemmas in, 237–238
in separation of conjoined twins, 1735–1737
in surgical error, 244–245
virtue, 237
Etomidate, 212
Etoposide, 407t
Europe, pediatric surgery in, 12–15, 13f, 14f, 15f
Eutectic mixture of local anesthetics (EMLA), 221, 221t
Everolimus, in renal transplantation, 625
Evidence
rating of, 227, 228t
sources of, 227–233
summaries of, 232–233, 232f
Evidence-based medicine, 227–237
clinical application of, 233–234
definition of, 227
quality improvement and, 234–236, 235f
study design in, 227–233, 228t
Ewing family tumors, 575, 575f
Ewing sarcoma
chromosomal translocations in, 400–401
Enneking staging system of, 582
epidemiology of, 580
genetics of, 580–581
location of, in relation to physis, 579f
pulmonary metastasis in, 571–572
resection and reconstruction of, 586f, 587f, 588f,
589f
EWS-FLI-1 fusion, 400–401
in Ewing sarcoma family/primitive
neuroectodermal tumors, 575
Excisional biopsy, in cervical lymphadenopathy, 740,
740t
Excretory urogram, in ureterocele, 1449, 1449f
Exercise program, after pectus excavatum repair, 790
EXIT procedure
for airway obstruction, 83
for cervicofacial lymphatic malformation, 1622
for cystic lung lesions, 826
indications for, 78t
Expiratory reserve volume, 112, 113f
Exposure, in emergency management, 268
Exstrophy-epispadias complex, 1515
Extracellular fluid, in neonates, 91–92
Extracellular matrix, 29
Extracorporeal carbon dioxide removal, 119
Extracorporeal life support, 119, 123–136
anticoagulation during, 128
background on, 123
as bridge to heart transplantation, 662
cannulation for cardiac support in, 127–129
Extracorporeal life support (Continued) Fanconi anemia, 166, 169
circuit for, 125–126, 126f, 126t Fascia iliaca nerve block, 223, 223f
complications of, 129–130 Fascial sling, for bladder outlet competence, 1478,
for congenital diaphragmatic hernia, 821 1480f
contraindications to, 124 FAST (focused abdominal sonography for trauma),
cost of, 131 290, 290f, 308, 313
development of, 8 Fasting, preoperative, 203t, 204
discontinuation of, 128–129 Fats (lipids)
future of, 132 carbohydrate conversion to, postoperative,
hemofiltration during, 128 105–106
indications for, 124 emulsions of, 189
methods of, 125, 125f intravenous, 182–183
operative procedures performed during, 128 metabolism of, in neonate, 102–103
patient management for, 126–127, 127f surgery and, 106
for refractory shock, 161–162 in parenteral nutrition, 106, 189
results of, 130–131, 130t, 131t requirements for, 182–183
vascular injuries associated with, 366 restriction of, in intestinal failure–associated liver
venoarterial, 125, 125f, 127f disease, 1139
venoarterial-venous, 127 soybean, 193
venovenous, 125, 125f, 126t Fatty acid–binding protein, intestinal, in necrotizing
Extracorporeal Life Support Organization, 123 enterocolitis, 1197
Extracorporeal membrane oxygenation. Fatty acids
See Extracorporeal life support. deficiency of, 182–183
Extracorporeal shock wave lithotripsy, for in intestinal failure–associated liver disease,
urolithiasis, 1438 1139
Extracranial lesions, stereotactic radiosurgery for, 54 oxidation of, in neonate, 102
Extradural hematoma, 351, 352f saturated versus unsaturated, 183
Extremity(ies). See also Limb entries. Fecaliths, in appendicitis, 1256, 1257
arterial aneurysm of, 1642–1643, 1643f Feces. See Constipation; Defecation; Incontinence,
arterial occlusion of, 1641–1642, 1641f, 1642f, fecal.
1643f Feeding intolerance, after duodenal obstruction
mangled, 335, 365, 365t surgery, 1057
rhabdomyosarcoma of, 497–498 Feet, congenital anomalies of, 1703–1705, 1704f,
vascular trauma to, 361, 364–365, 365t 1705f
Extubation, 121 Female gender assignment surgery, 1577
failure of, 121–122 clitoroplasty in, 1578–1579, 1578f, 1579f
Eye(s) labioplasty in, 1579, 1579f, 1580f
in brain tumors, 592 planning and timing of, 1578
dancing, 443 postoperative care in, 1582
endophthalmitis of, in liver abscess, 1351 single-stage, 1578
panda or raccoon, in neuroblastoma, 442–443, vaginoplasty in, 1580–1582, 1580f, 1581f, 1582f,
442f 1583f
uveitis of, in ulcerative colitis, 1219 Female genital tract. See also Ovary(ies); Uterus;
Vagina.
abnormalities of, 1591–1613
F Feminization, in adrenocortical lesions, 563
Face Femoral artery
burns to, 384 in extracorporeal life support, 127
hemangioma of, 849 injury to, 366
hypoplasia of, in torticollis, 766, 766f occlusion of, 1641, 1641f, 1642f
Facial clefting syndrome, 977t Femoral fracture
Facial nerve, 707 in birth trauma, 392
paralysis of, 710, 712 in child abuse, 389, 389f
Faciooculoacousticorenal syndrome, 977t fixation of, 333f
Factor IX, deficiency of, 171–172 Femoral head, avascular necrosis of, in
Factor VII replacement, for variceal hemorrhage, developmental dysplasia of hip, 1703
1362 Femoral hernia, 987, 1000
Factor VIII Femoral neck fracture, 334–335, 335f
deficiency of, 171–172 Femoral shortening osteotomy, 1703, 1703f
spleen as reservoir for, 1386 Femoral vein, in extracorporeal life support, 126–127
Factor VIII/von Willebrand factor, recombinant, for Fentanyl, 218–219, 219t
von Willebrand disease, 170–171 for burns, 382
Failure to thrive, nutritional support in, 198–199 caudal, 224–225
Falciform ligament, nonfixation of, 1131, 1131f in patient-controlled analgesia, 220, 220t
Fallopian tube, in sliding hernia sac, 991, 991f, 998, Ferrous sulfate, for iron deficiency anemia, 168
1000 Fertility. See Infertility.
Familial adenomatous polyposis, 405, 1179–1182, Fetal access, 79–80, 79t, 80f, 81f
1179t Fetal circulation, 134–135, 136f
clinical presentation in, 1181 persistence of, 135
colorectal cancer and, 488 Fetal diagnosis. See Prenatal diagnosis.
etiology of, 1181 Fetal interventions
follow-up for, 1179t, 1182 for airway obstruction, 83
Gardner syndrome and, 1182 for anomalies of monochorionic twins, 87
hepatoblastoma in, 466–467 for congenital diaphragmatic hernia, 85–86, 817,
historical perspective on, 1180 823
pathology of, 1180–1181, 1180f for congenital pulmonary airway malformation,
treatment of 83–85
medical, 1182 for cystic lung lesions, 826
surgical, 1181 future of, 88
in Turcot syndrome, 1182 for gastroschisis, 87–88
Familial Mediterranean fever, 1234 historical perspective on, 8, 83t, 88
Familial polyposis coli, 487 for intestinal abnormalities, 87
Family-centered care. See Patient- and family- maternal/fetal management during, 80–82
centered care. milestones in, 83t
Volume 1, pages 1–768; Volume 2, pages 769–1738
 INDEX I-17

Fetal interventions (Continued) Fixation Forehead lesions, subcutaneous endoscopy for, 55,
for myelomeningocele, 86, 1676–1677 methods for, 332, 333f 55f
open, 79–80, 80f, 81–82 of open fractures, 334 Foreign body
percutaneous, 82 FLACC pain scoring system, 215 esophageal
for posterior urethral valves, 1555–1556 Flail chest, 275 endoscopic removal of, 885
problems amenable to, 78t, 79, 82–86, 84t Flap procedure, for ureteropelvic junction esophageal replacement for, 928
for pulmonary hyperplasia, 817 obstruction, 1421, 1422f perforation by, 890
risks of, 82 Flap valve, for continent urine drainage, 1479–1480, intestinal obstruction from, 1133
for sacrococcygeal teratoma, 86, 512 1481f in Meckel diverticulum, 1092
for TRAP sequence, 88 Flatfoot, flexible, 1704 nasal, 715
for twin-twin transfusion syndrome, 87 Flexor digitorum superficialis tendon, 338–339, 339f Foreskin
for urinary tract obstruction, 82–83 Flexor mechanism, digital, disruption of, 338–339, inability to retract, 1561
videoendoscopic, 79–80, 81f 338f preservation of
Fetal lung fluid, 112 Fludrocortisone, for adrenogenital crisis, 1574 in meatal advancement glansplasty, 1540, 1541f
Fetal membranes, 1086f Fluid(s) in tubularized plate urethroplasty,
Fetal sampling, 77–78 body 1543, 1544f
Fetus composition of, in neonate, 91–92 Formic acid burns, 383
alpha fetoprotein in, 77 gestational age by, 96, 96f Formula(s), 184–187, 185t, 186t
growth of, 89 measurement of, 180 in biliary atresia, 197
imaging of, 78 cerebrospinal. See Cerebrospinal fluid. after burn injury, 381–382
echocardiography in, 78 intraperitoneal. See Ascites; Intraperitoneal fluid. in gastroesophageal reflux disease, 953
magnetic resonance imaging in, 45, 78, 79f lymph, abdominal, sources of, 1171 low-fat, 1174–1175
ultrasonography in, 78, 79f requirements for, 181, 181t in necrotizing enterocolitis, 1189, 1193–1194,
lymphangiectasia in, 1622–1623 Fluid balance 1206
stem cell transplantation in, 88 after jejunoileal atresia and stenosis repair, after pyloromyotomy, 1027–1028, 1028t
surfactant production by, 111 1069–1070 in short bowel syndrome, 1137
urine production by, 1413 in neonate, 91–95 supplementation of, 187
Fetus in fetu, 1737 in premature or critically ill infant, 205 Fowler-Stephens procedure
Fgf10, in congenital cystic adenomatoid Fluid dynamics, computational, 29 in cryptorchidism, 1010–1013
malformation, 827 Fluid management or resuscitation in prune-belly syndrome, 1509
FGFR2 gene mutation, in syndromic in abdominal wall defects, 983 Fracture(s). See also Musculoskeletal trauma.
craniosynostosis, 693 for adrenogenital crisis, 1574 in birth injury, 391–392
Fibrinogen anesthesia and, 205–207 bone tumor, 578–579, 579f
deficiency of, 172–173 for burns, 374–375, 374t, 375t in child abuse, 336, 388–389, 388f, 389f
during extracorporeal life support, 128 in central nervous system injury, 344 femoral neck, 334–335, 335f
Fibroadenoma, of breast, 774–775 for gastrointestinal bleeding, 1147 fixation of, 332, 333f
giant, 775, 775f for hypertrophic pyloric stenosis, 1024 forearm, 330f
Fibrocystic breast abnormalities, 776 for inhalation injury, 376 hand, 338, 339–340
Fibroma, ovarian, 540 intraoperative, 205–206 management of
Fibromatosis colli, 763 for jejunoileal atresia and stenosis, 1066 definitive, 332, 333f
Fibrosarcoma in neonate, 95, 95t, 163 immediate, 332
breast, 777 parenteral nutrition and, 190–191, 190t nasal, 715
chest wall, 575–576 after renal transplantation, 623 open, 334, 338
colon, congenital infantile, 1250 restriction in periosteum and, 327, 329f, 332
infantile, 501–502, 503 in hepatocellular ascites, 1172 physeal. See Physis, fracture of.
ovarian, 547 preoperative, 203t, 204 rib
Fimbriae, bacterial, 150 after separation of conjoined twins, 1735 in child abuse, 275, 389, 389f
Fine-needle aspiration, 418 for sepsis, 155–158 traumatic, 272, 275
in cervical lymphadenopathy, 739–740 in spinal cord injury, 359 skull. See Skull fracture.
in intracranial infections, 1696–1697 in trauma patient, 267 spinal, 335, 335f, 336f, 354, 359
in pneumothorax, 872–873 Fluid-refractory shock, 159–160, 159t sternal, 275
in thyroid nodules, 748, 748t Fluoro-2-deoxyglucose (FDG) temporal bone, 711–712, 711f
Finland, pediatric surgery in, 13 in molecular imaging, 47 types of, 327, 328f
Finney strictureplasty, 1213–1214, 1213f in positron emission tomography, 45. vascular disruption associated with, 365
Fire safety, 258–259 See also Positron emission tomography. France, pediatric surgery in, 12–13, 13f
Firearm injury, 348 5-Fluorouracil, 407t Frank-Starling relation, 133–134, 134f
prevention of, 259 Foley Y-V-plasty, for ureteropelvic junction Fresh frozen plasma (FFP)
Fish oil, parenteral, in intestinal failure–associated obstruction, 1422f, 1423 for disseminated intravascular coagulation, 174
liver disease, 1139 Folic acid intraoperative, 206–207
Fissurectomy, for anal fissure, 1317–1318 neural tube defects and, 1673, 1675 for sepsis, 158
Fistula supplementation of Frontofacial advancement, monobloc, 695
anal, 1318–1319, 1318f after bariatric surgery, 1046, 1048 Fryns syndrome, 977t
in Crohn disease, 1210–1211, 1212, 1215, 1215t myelodysplasia and, 1458 Functional residual capacity (FRC), 112, 113, 113f
arteriovenous, 1358 Follicular cysts, ovarian, 536, 536f in congenital diaphragmatic hernia, 817
arteriovenous-capillary, 1629 Follow-up, in patient- and family-centered care, 251 Fundoplication
in Crohn disease, 1210–1211, 1210f, 1212, 1213, Fontan procedure, 1664–1665, 1664f complications of, 955–956
1215, 1215t hemi-, 1664, 1664f endoluminal, 57
after ileoanal pouch procedure, 1228 Foot, congenital anomalies of, 1703–1705, 1704f, 1705f endoscopic, 957
perineal, 1293, 1293f, 1294–1295 Foramen ovale, patent, 1652–1653 esophageal repair with
pyriform sinus, 747, 747t closure of, in heart transplantation, 664–665 for achalasia, 946
rectobladder neck, 1296f, 1297 Forced expiratory volume in 1 second (FEV1) for atresia, 913
reconstruction for, 1298–1300, 1300f, 1301f in cystic fibrosis, 671–672 for stricture, 923, 924f
rectoprostatic, 1297 in pectus excavatum, 782 laparoscopic, 954, 954f
rectourethral, 1296f, 1297 Forced vital capacity, in pectus excavatum, 782 mechanisms of, 955
penile agenesis with, 1585, 1588f Forearm in neurologically impaired children, 956
reconstruction for, 1297–1298, 1298f, 1299f, compression of, 338–339, 338f Nissen, 954, 954f, 955
1300f fracture of, 330f results of, 955
rectovaginal, 1162, 1162f Foregut, 825 Toupet, 954, 955
tracheoesophageal. See Tracheoesophageal fistula. duplications of, 721, 832 transoral incisionless, 57
urethral, congenital, 1560 abdominal, 1156–1157, 1159–1160, 1159f Fungal infection
urethrocutaneous bronchogenic, 832–833, 833f, 834f in necrotizing enterocolitis, 1197
after bladder exstrophy repair, 1523 enteric, 834–835, 834f, 835f in sinusitis, 713
after hypospadias repair, 1552, 1552f neurenteric, 835 Funnel chest. See Pectus excavatum.
vestibular, 1294, 1301 embryopathology of, 895–896 Furlow palatoplasty, 703, 704f

Volume 1, pages 1–768; Volume 2, pages 769–1738

I-18 INDEX

G Gastroesophageal reflux disease, 947–961 Gastrointestinal tract (Continued)

Galactorrhea, 774, 774f
Galeazzi sign, 1700
Gallbladder. See also Chole- entries.
carcinoma of, 1339
disorders of, 1341–1349
hydrops of, 1342
polyps of, 1343
Gallbladder-ventriculo shunt, 1343
Gallstones. See Cholelithiasis.
Ganglioglioma, 599–600, 599f
seizures in, 1687–1688, 1688f, 1692
Ganglion cell(s). See also Aganglionosis.
in Hirschsprung disease, 1265, 1267, 1274–1276
in hypertrophic pyloric stenosis, 1021
in intestinal neuronal dysplasia, 1280, 1281f
in isolated hypoganglionosis, 1282, 1283f
in megacystis-microcolon-intestinal
hypoperistalsis syndrome, 1286
Ganglion cell tumors, 591
Ganglioneuroblastoma, 448
Ganglioneuroma, 447–448
Gangliosides, in neuroblastoma, 449
Gangrene
intestinal
in necrotizing enterocolitis, 1195, 1195f, 1200
predictors of, 1200
umbilical, 964
vasospasm and, 366–367
GANT (gastrointestinal autonomic nerve tumor), 484
GAP (glans approximation procedure), 1540–1541,
1542f
Gardner syndrome, 487, 1182
Gartner duct, 1441–1443
cysts of, 1558, 1608
Gas exchange
extreme modes of, 119–120
pulmonary, 114–115, 115f
structural development related to, 109–110, 110f
Gastrectomy
laparoscopic sleeve, 1046
for stress ulcers, 1034–1035
Gastric. See also Stomach.
Gastric acid
secretion of, 1030
ulcers and, 1030
Gastric aspiration, lung abscess from, 868
Gastric banding, laparoscopic adjustable,
1041–1042, 1046
Gastric bypass surgery, 1041–1042, 1046, 1047f
Gastric decompression
for gastric volvulus, 1037–1038
in trauma patient, 268
Gastric duplications, 1036, 1156–1157,
1159–1160
Gastric emptying, delayed, gastroesophageal reflux
disease with, 957
Gastric feedings, 186
Gastric lymphoma, MALT, 522–523
Gastric mucosa
heterotopic, in Meckel diverticulum, 1086–1087,
1087f
ischemia of, stress ulcers and, 1031
Gastric outlet obstruction, congenital, 1035–1036,
1035f, 1036f
Gastric reflux, into colonic interposition, 932
Gastric teratoma, 516
Gastric transposition, for esophageal replacement,
907, 908f, 929t, 934–938, 936f, 937t
Gastric tube, for esophageal replacement, 907, 929t,
932–934, 933f, 934f, 934t
Gastric ulcer. See Peptic ulcer disease.
Gastric varices. See also Varices.
injection therapy for, 1363
Gastrin, in Zollinger-Ellison syndrome, 1034
Gastrinoma, 1383
Gastritis. See also Peptic ulcer disease.
bleeding in, 1149–1150
causes of, 1030t
clinical findings in, 1031t
stress, 1149
Gastrocystoplasty, 1475, 1475f, 1492
complications of, 1484–1485
Gastroduodenostomy, for pyloric atresia, 1036
apneic spells and, 950–951, 951t
Barrett esophagus in, 483, 956, 956f
in caustic esophageal stricture, 923, 924f
in congenital anomalies and diseases, 957–958
in congenital diaphragmatic hernia, 822, 822f
with delayed gastric emptying, 957
diagnostic studies in, 952–953, 952f, 953f
epidemiology of, 947
in esophageal atresia
preoperative, 912
after repair, 913, 957–958
esophagoscopy in, 882, 882f, 883f
hiatal hernia and, 957
with laryngomalacia, 723–724
with laryngopharyngeal reflux, 840
laryngotracheal stenosis and, 846
after lung transplantation, 678
in neurologically impaired children, 956–957
pathophysiology of, 948–949
primary, 944–945
recurrent, 955–956
symptoms of, 949–951, 951f, 951t
treatment of
conservative, 953
endoluminal, 57, 957
surgical, 954, 954f. See also Fundoplication.
Gastrografin enema, for meconium ileus, 1078–1079
Gastroileal pouch, 1494–1495
Gastrointestinal anomalies
with anorectal malformations, 1290
with cloacal exstrophy, 1526
with esophageal atresia, 897
Meckel diverticulum as. See Meckel diverticulum.
Gastrointestinal atresia, familial, 1060–1061, 1061t,
1065, 1066f
Gastrointestinal autonomic nerve tumor, 484
Gastrointestinal bleeding, 1147–1155
evaluation of, 1147–1148, 1148t, 1149f, 1150f
in gastrointestinal vascular malformations, 1154
lower
in anal fissure, 1151
in anorectal trauma, 1153, 1153f
diagnostic algorithm for, 1150f
evaluation of, 1153, 1153f
in juvenile polyps, 1152, 1152f
in Meckel diverticulum, 1089–1091,
1151–1152, 1151f, 1152f
sources of, 1148t, 1151–1152
novel techniques for identification of, 1154
in peptic ulcer disease, 1032, 1033–1034
in portal hypertension, 1358–1360. See also Varices.
resuscitation for, 1147
versus swallowed maternal blood, 1148
upper
diagnostic algorithm for, 1149f
in esophageal varices, 1150
esophagoscopy in, 883
in gastritis, 1149–1150
in hemorrhagic disease of newborn, 1148–1149
nonvariceal, 1150–1151
sources of, 1148–1151, 1148t
Gastrointestinal peptides, in hypertrophic pyloric
stenosis, 1022
Gastrointestinal tissue engineering, 32
Gastrointestinal tract
as barrier to infection, 145–146, 145f
in burn injury, 371
contrast studies of, in conjoined twins, 1733
duplications of. See Alimentary tract duplications.
functional abnormalities of, after bladder
augmentation or replacement, 1484–1485,
1495–1496
hemangioma of, 1616
polypoid disease of, 486–487
trauma to, 305–308, 307f
imaging of, 307–308
intestinal stricture after, 1133
seat-belt sign in, 307, 307f
tumors of, 483–493
carcinoid, 485–486
colorectal cancer as, 486. See also Colorectal
cancer.
esophageal, 483
gastric, 483
intestinal, 485
stromal, 484–485
venous malformation of, 1625
Gastrointestinal vascular malformations, bleeding,
1154
Gastrojejunal feeding tube
in neurologically impaired children with reflux,
956–957
in short bowel syndrome, 1138
Gastrojejunostomy
intussusception around tube in, 1098
pyloric exclusion with, 300–301, 303f
Gastropathy, in portal hypertension, 1359, 1368
Gastroschisis
antenatal considerations in, 977–978
associated conditions with, 979, 979t
complicated, 973–974, 982, 984
complications of, 983
cryptorchidism in, 1004–1005
at delivery, 975–976, 976f, 978–979
description of, 973–974, 974f, 974t
embryogenesis of, 975–976, 976f
fetal interventions for, 87–88
historical perspective on, 973
incidence of, 979
jejunoileal atresia and stenosis with, 1060,
1068–1069, 1241, 1241f
outcome of, 983–984
treatment of, 982, 982f
umbilical hernia versus, 974–975
Gastrostomy, 186. See also Enterostoma.
in cricopharyngeal disorders, 942
endoscopic placement of, 884–885
for esophageal atresia, 893, 907–908, 909f
esophageal dilatations through, 884
intussusception around, 1098
percutaneous endoscopic, in neurologically
impaired children with reflux, 956
Gaucher disease, splenectomy for, 1387
GD2
antibodies against, 411
for neuroblastoma, 457–458
in neuroblastoma, 449
GDNF gene, in Hirschsprung disease, 21t
Gefitinib, 410
Gender assignment
in 46,XX DSD, 1573
in 46,XY DSD, 1573–1574
in cloacal exstrophy, 1528
in penile agenesis, 1585
Gender assignment surgery
female, 1577
clitoroplasty in, 1578–1579, 1578f, 1579f
labioplasty in, 1579, 1579f, 1580f
planning and timing of, 1578
postoperative care in, 1582
single-stage, 1578
vaginoplasty in, 1580–1582, 1580f, 1581f,
1582f, 1583f
male, 1582–1583, 1584f
hypospadias repair in, 1582–1583, 1585f
müllerian duct remnants and, 1585, 1587f,
1588f
penile agenesis and, 1585–1586, 1588f, 1589f
penoscrotal transposition and, 1583–1584,
1586f
Gene chips, 48
Gene therapy, 23–26
Gene transfer
challenges in, 25–26
current status of, 26
viral vectors for, 23–25, 23f, 24t
Genetic counseling, molecular genetics and, 22
Genetic disease
monogenic, 20, 20f
oligogenic, 20–21, 21t
polygenic or complex, 21–22
reconceptualization of, 19–20, 20f
Genetic screening, for cancer, 405
Genetics, molecular. See Molecular genetics.
Genital defects, in bladder exstrophy, 1516–1517,
1517f, 1518f

Volume 1, pages 1–768; Volume 2, pages 769–1738

 INDEX I-19

Genital germ cell tumors, 516 Glial cell–derived neurotrophic factor, in renal Granuloma (Continued)
Genital tract, female, abnormalities of, 1591–1613 development, 1395 pyogenic, 1618, 1619f
Genitalia Glial cells, 591 after tracheotomy, 839–840
ambiguous. See also Disorders of sex development Glioblastoma multiforme, 600, 600f umbilical, 964
(DSD). Glioma, 591, 601 Granulosa-theca cell tumors, ovarian, 539–540, 539f
hypospadias with, 1531 brainstem, 597, 597f Graves disease, 747–748, 748t
external nasal, 715 Great arteries, transposition of, 1660–1663, 1662f,
in congenital adrenal hyperplasia, 1569–1570, pontine, 593, 597, 597f 1663f
1573 tectal, 597 Great saphenous vein, cannulation of, 266, 267f
female, differentiation of, 1591 Glomerular filtration rate, in neonate, 93 Greece, pediatric surgery in, 15
trauma to, 308, 324–325 Glomerular sclerosis, focal segmental, after renal Greenstick fracture, 327, 328f, 338
in females, 324 transplantation, 627–628 Gross, R. E., 4, 5f
in males, 324–325, 324f Glossopexy, for tracheomalacia, 914 “Ground-glass” sign, in meconium ileus, 1075–1076,
of vaginal agenesis, 1592, 1592f Glossoptosis, 720 1076f
Genitofemoral nerve defect, in cryptorchidism, 1004 Glucagon Group A beta-hemolytic streptococcus (GABHS)
Genitogram, 1575, 1576f in burn injury, 380 infection, oropharyngeal, 717
Genitourinary anomalies in intussusception reduction, 1104 Growth
with anorectal malformations, 1290 in perinatal period, 100 of fetus, 89
with cloacal exstrophy, 1526 Glucagon-like peptide 2, for intestinal adaptation intrauterine restriction of, 89–90
with hypospadias, 1534–1535 promotion, 1141 of neonate, 89, 97, 179
Genitourinary system Glucocorticoids. See also Corticosteroids; Cortisol. of premature infant, 179
contrast studies of, in conjoined twins, 1733 adrenocortical production of, 558 Growth disturbances
embryology of, 1531–1532, 1533f, 1535f, 1538f insufficiency of, 564 after bladder augmentation or replacement, 1484
imaging of, 1430t maternal, for necrotizing enterocolitis, 1205 after bone tumor resection, 588
rhabdomyosarcoma of, 498 Gluconeogenesis, in neonate, 100 after physeal fracture, 328, 329, 331f
Genitourinary trauma, 311–329 Glucose (dextrose). See also Hyperglycemia; after radiation therapy for Hodgkin lymphoma, 522
anatomic considerations in, 311–312 Hypoglycemia. after renal transplantation, 629
to bladder, 320–322, 321f in burn injury, 374, 380 Growth factors
clinical features of, 312 in fluid therapy in hypertrophic pyloric stenosis, 1022
diagnostic studies in, 312–314 for burn injury, 374 in necrotizing enterocolitis, 1189–1191, 1190t
epidemiology of, 311 for hyperinsulinism, 1379, 1382 Growth hormone
to external genitalia, 324–325, 324f intraoperative, 205–206 for burn patient, 380, 381
grading of, 314, 314t, 315f metabolism of, in neonate, 99–102, 101t for intestinal adaptation promotion, 1141
to kidney, 315–318, 316f, 318f. See also Kidney, surgery and, 105–106 Growth plate. See Physis.
trauma to. in parenteral nutrition, 189 Grumbach syndrome, 530
management of, 315–318 overfeeding from, 194 Guaiac test, 1147–1148
mechanisms of injury in, 311 requirements for, 182 Gubernaculum, 1003, 1004f
to ureter, 319–320 Glutamine, requirements for, 182 Guidelines, 234
to urethra, 322–324, 322f Gluteal cleft, shortened, in sacral agenesis, 1460, Gunshot wounds, traumatic brain injury from, 348
Genome, human, alteration of, in gene transfer, 1460f Gustilo classification of open fractures, 334
25–26 Gluteal crease, asymmetric, 1453, 1454f Gynecologic anomalies, with anorectal
Genomics, 48 Glycogen, in perinatal period, 100 malformations, 1290
Gentamicin, for urinary tract infection, 1431–1432 Glycogen storage disease, type I, 461 Gynecomastia, 777–778, 778f, 1716–1719, 1719f
Georgeson procedure, for Hirschsprung disease, Glycogenolysis, in perinatal period, 100
1270, 1270f Goiter, 746–747
Germ cell(s) diffuse toxic, 747–748, 748t H
deficiency of, in cryptorchidism, 1007 in hypothyroidism, 747 H-probe, 61
migration and proliferation of, 1565, 1566f Goldstein sepsis criteria, 152, 152t, 153t Haddon approach to injury prevention, 255, 256t
Germ cell tumors, 507–518. See also Teratoma. Goldstein test, in inguinal hernia, 994 Haemophilus influenzae infection
abdominal, 516 Golytely, preoperative, in gender assignment surgery, in cystic fibrosis, 865
classification of, 507–508, 508f 1575–1576 as pneumonia, 856
embryology of, 507–508, 508f Gomco clamp, 1561 Haight, C., 894, 894f
extragonadal, staging system for, 513, 515f Gonad(s) Halothane, 202t, 207t, 208
genetics of, 508 differentiation of, 1565–1567, 1566f Hamartoma
genital (vaginal), 516 dysgenesis of breast, 776
mediastinal, 514–516, 515f malignancy risk in, 508 cystic, pancreatic, 1383
mixed, ovarian, 546 mixed (asymmetric), 1568t, 1571, 1574, 1575, gastrointestinal, 486
ovarian. See Ovarian tumors, germ cell. 1576f mesenchymal
retroperitoneal, 516 XY, 1568t, 1571, 1574, 1575 chest wall, 574, 574f
risk-based therapy for, 508–509, 509f streak, 1571, 1574 hepatic, 461, 461f, 465, 466f
risk factors for, 508 symmetry of, 1572, 1572t pulmonary, 567
testicular, 509–510, 510f, 510t, 551–552, 556, Gonadoblastoma, 508 Hamartomatous polyposis syndrome, 1182–1185
556t in mixed gonadal dysgenesis, 1574 in Cowden syndrome, 1184–1185
Germany, pediatric surgery in, 14, 14f ovarian, 545–546 in juvenile polyposis syndrome, 1177,
Germinoma, ovarian, 541–543, 542f testicular, 552 1182–1183, 1183f
Germline transmission, in gene transfer, 25–26 Gonadosplenic fusion, 1387 in Peutz-Jeghers syndrome, 1183–1184, 1184f
Gershoni-Baruch syndrome, 977t Gonadotropin Hand
Gestational age, 89, 90f cryptorchidism and, 1007 cleft, 1722
birth weight and, 89, 91f human chorionic congenital anomalies of, 1720–1724
body length/head circumference and, 89, 91f in cryptorchidism, 1009 classification of, 1716, 1722t
body water and energy stores by, 96, 96f in ovarian tumors, 530–531, 530t incidence of, 1720
mortality by, 90–91, 92f in testicular tumors, 550 in Poland syndrome, 797
resting energy expenditure and, 97 ovarian tumors and, 531t treatment of, 1716, 1723f
GFRA1 gene, in Hirschsprung disease, 21t Gorham-Stout syndrome, 1623 embryology of, 1720
Giant fibroadenoma, of breast, 775, 775f Graft-versus-host disease mirror, 1723
Gigantism, digital, 1723–1724 immunologic basis of, 609–613, 609f, 610f, 613f trauma to, 337–340
Gila monster bite, 341 tissue typing and, 614–615, 615f early treatment of, 339–340
Gingivostomatitis, herpetic, 716–717 transfusion-related, 177 evaluation of, 337–339, 338f, 339f
Glans approximation procedure (GAP), 1540–1541, Granulocyte colony-stimulating factor, for aplastic windblown, 1723
1542f anemia, 166 Haptic feedback
Glansplasty, meatal advancement, 1540, 1541f Granulocytic sarcoma, ovarian, 548 in surgical simulation, 66
Glanzmann thrombasthenia, 171 Granuloma in virtual reality, 71, 71f, 72
Glasgow Coma Scale (GCS), 349 noncaseating, in Crohn disease, 1210, 1210f Hard signs, in vascular trauma, 362
Glenn procedure, bidirectional, 1664, 1664f plasma cell, pulmonary, 567 Harmonic scalpel, 49

Volume 1, pages 1–768; Volume 2, pages 769–1738

I-20 INDEX
Hashimoto thyroiditis, 747, 747t
Head and neck mass. See also Neck.
fine-needle aspiration biopsy of, 418
rhabdomyosarcoma as, 496
Head circumference
gestational age and, 89, 91f
nutritional status and, 179–180
Head injury. See also Brain injury, traumatic; Skull
fracture.
anosmia from, 715
ear disturbances from, 711–712, 711f
early complications of, 352–353
epidemiology of, 344–345
outcomes with, 353
Head tilt, 764–765, 764f
Health care
patient- and family-centered. See Patient- and
family-centered care.
traditional, 248t
Hearing aid, 708
Hearing loss
assessment of, 708
after congenital diaphragmatic hernia repair, 822
in otitis media, 710–711
in temporal bone fracture, 712
Heart. See also Cardiac entries.
deformity of, in pectus excavatum, 783
ectopic. See Ectopia cordis.
orthotopic, 804, 805f, 805t
trauma to, 280–282, 281f, 282f
penetrating, 286
Heart block
after atrioventricular septal defect repair, 1659
in neonate, 138
after ventricular septal defect repair, 1657
Heart disease. See also Cardiovascular disorders.
congenital, 1647–1674
congenital diaphragmatic hernia and, 822
heart transplantation for, 659–660, 660f, 661f,
661t
in neonate, 139–140
pulmonary hypertension with, lung
transplantation for, 672–673
vascular tissue engineering for, 31–32, 31f
Heart failure
in atrioventricular septal defect, 1658
causes of, 137t
in coarctation of the aorta, 1650
extracorporeal life support for, 123–136.
See also Extracorporeal life support.
in hepatic hemangioma, 1617–1618
in neonate, 135–138, 137t
in patent ductus arteriosus, 1648
pharmacologic therapy for, 135–138, 137t
in transposition of the great arteries, 1661–1662
in ventricular septal defect, 140, 1655–1656
Wilms’ tumor and, 437
Heart rate
in neonate, 133
threshold, by age group, 159, 159t
Heart transplantation, 659–671
ABO-incompatible, 663
age distribution of, 660f
complications of
early, 667, 667t
late, 668
contraindications to, 662, 662t
donor evaluation for, 663–664
historical perspective on, 605–613, 606t, 659
immunosuppressive therapy for, 665–667, 667t
indications for, 659–661, 660f, 661f, 661t
lung transplantation with, 672–673
mechanical support as bridge to, 662
operative procedures in, 664–665, 664f, 666f
organ procurement for, 663–664
postoperative care in, 665
preoperative evaluation for, 661–663, 661t
recipient preparation for, 664–665
results of, 668–670, 668f, 669f
Heineke-Mikulicz pyloroplasty, 1035f, 1036
Heineke-Mikulicz strictureplasty, 1213–1214, 1213f
Helicobacter pylori infection
diagnosis of, 1032
gastritis and, 1149–1150
Helicobacter pylori infection (Continued) Hemofiltration, for fluid-refractory shock, 159, 159t
in MALT gastric lymphoma, 522–523 Hemoglobin
in Meckel diverticulum, 1090 carbon dioxide binding to, 115
in peptic ulcer disease, 1029, 1030–1032 oxygen binding to, 115, 115f
treatment of saturation of, noninvasive monitoring of, 116
medical, 1033, 1033t Hemoglobin S, 168
surgical, 1033–1034 Hemolysis, with extracorporeal life support, 129
Heliox, for airway obstruction, 723 Hemolytic anemia, 168–169
Helium dilution test, 113 Hemolytic cholelithiasis, 1341
Heller myotomy, 946 Hemolytic uremic syndrome, after renal
Helmet use, 259 transplantation, 628
Hemagglutinin, 150 Hemoperitoneum, in neuroblastoma, 442
Hemangioendothelioma, Kaposiform, 464, Hemophagocytic lymphohistiocytosis, of liver, 482
1619–1620, 1619f Hemophilia, 171–172
Hemangioma Hemoptysis, 867, 867f
breast, 774, 774f Hemorrhage. See also Coagulation, disorders of;
cavernous. See Venous malformation. Hematoma; Vascular trauma.
congenital, 1617, 1617f anemia from, 167
cutaneous, 1616 delayed, after splenic injury, 294
gastrointestinal, 1616 with extracorporeal life support, 129
head and neck, 721 gastrointestinal. See Gastrointestinal bleeding.
hepatic, 460–462, 460f, 464, 465, 466f, 480–481, massive
481f, 1613–1614 after liver injury, 296, 296f
infantile, 1617–1618, 1618f in peptic ulcer disease, 1034
infantile, 1613–1617 pulmonary, 867, 867f
associated anomalies with, 1614–1615 after lung transplantation, 678
etiology and pathogenesis of, 1614 stomal, 1245
life cycle of, 1613, 1615f Hemorrhagic disease of newborn, 1148–1149
morphologic variants of, 1613, 1614f Hemorrhoids, 1319, 1319f
radiologic findings in, 1615, 1615f Hemostasis
treatment of, 1615–1617, 1615f in abdominal trauma, 294–296, 296f, 297f
ovarian, 548 in hypospadias repair, 1551
parotid gland, 732 in open incisional biopsy, 422
periorbital, 1613–1614 Hemostatic instruments, 48–50
subglottic, 725, 725f, 849–850, 849f, 1613–1614 Hemothorax, in thoracic trauma, 276–277
tracheal, 849–850 Henoch-Schönlein purpura
vaginal, 1609 intussusception in, 1102
Hemangiomatosis, 1614 after renal transplantation, 628
Hematemesis Heparin
in infant, 1151 for disseminated intravascular coagulation, 174
in neonate, 1148 during extracorporeal life support, 128
in portal hypertension, 1358–1359 low-molecular-weight
Hematochezia, infectious agents associated with, for renal vein thrombosis, 1439–1440
1152–1153, 1153t for venous thromboembolism, 175
Hematocrit, during extracorporeal life in parenteral nutrition, 191
support, 128 for renal vein thrombosis, 1439–1440
Hematologic disease, 165–181 thrombocytopenia from, 170
Hematoma for venous thromboembolism, 175
in birth injury, 391 Heparin-binding epidermal-like growth factor, in
epidural, 351 necrotizing enterocolitis, 1190, 1190t
extradural, 351, 352f Heparin-bonded shunt, for aortic injury, 283
intracerebral, traumatic, 346 Hepatic artery, thrombosis of, in liver
intracranial transplantation, 649
in birth trauma, 392 Hepatic portoenterostomy. See Portoenterostomy.
removal of, 351, 351f, 352f Hepatic venules, occlusion of, 1357–1358
intrahepatic, 464, 465f Hepatitis B
in birth trauma, 392, 392f hepatocellular carcinoma and, 476
pulmonary, 277 transfusion-related, 177
septal, 715 Hepatitis C
subdural, 351, 351f hepatocellular carcinoma and, 476
in child abuse, 387, 388f transfusion-related, 177
from overshunting, 1685–1686 Hepatobiliary anatomy, 646, 646f
vulvar, 324 Hepatobiliary scintigraphy, in biliary atresia, 1324
Hematoma block, for hand fracture, 339–340 Hepatoblastoma, 466–469. See also Liver tumors,
Hematopoietic stem cell transplantation. See Stem malignant.
cell transplantation. alpha fetoprotein in, 464–465
Hematopoietic stem cells, 1620 clinical presentation in, 463–464
Hematuria epidemiology, biology, and genetics of, 466–467,
in bladder injury, 321 468t
in genitourinary trauma, 312–313 with major venous involvement, 474
Hematuria-dysuria syndrome, after gastrocystoplasty, multifocal, transplantation for, 470–472, 474f
1484, 1496 new agents and treatment modalities for, 475–476
Hemihyperplasia, hepatoblastoma and, 466–467 pathology of, 467–469, 468t
Hemisacrum, 1294 PLUTO database for, 475
Hemispherectomy, 1690 with pulmonary metastasis at diagnosis, 474–475,
Hemivagina 475t
hydrocolpos with, 1304, 1305f radiologic evaluation of, 465–466, 466f
obstructed, with ipsilateral renal anomaly, 1602, relapsed, 475, 476
1603f staging and risk stratification of, 467f, 469–476,
Hemivertebrae, 1706–1707, 1707f 469f, 469t, 470f
cervical spine, 765 transplantation for, 472–475, 474f, 474t, 475t, 645
excision of, 1708, 1708f treatment of, 470–472, 471f, 472f, 473t
in Jarcho-Levin syndrome, 807, 808f Hepatocellular adenoma, 461
Hemodialysis, renal transplantation and, 619 Hepatocellular ascites, 1172–1173
Volume 1, pages 1–768; Volume 2, pages 769–1738

Hepatocellular carcinoma, 476–482. See also Liver
tumors, malignant.
clinical presentation in, 463–464
epidemiology, biology, and genetics of, 468t, 476,
476t
fibrolamellar, 477, 478
liver transplantation for, 478–479, 479t, 645
metastatic, 479
new agents and treatment modalities for, 479–480
pathology of, 476–477
radiologic evaluation of, 465–466, 466f
staging of, 477
treatment of, 477–478, 478t
tumor markers in, 465
Hepatomegaly
in neuroblastoma stage IV-S disease, 450
in portal hypertension, 1360
Hepatopulmonary syndrome, 1360
Hereditary hemorrhagic telangiectasia, 487, 1621
Hereditary nonpolyposis colon cancer, 488, 489
Heredity, cancer and, 404–405
Herlyn-Werner-Wunderlich syndrome, 1602, 1603f
Hernia
diaphragmatic. See Diaphragmatic hernia.
epigastric, 970
femoral, 987, 1000
hiatal, 957
inguinal. See Inguinal hernia.
with intestinal obstruction, 1130–1131, 1130f,
1131f
mesenteric, 1132
mesocolic (paraduodenal)
in intestinal malrotation, 1117, 1118f, 1120,
1124
intestinal obstruction in, 1130–1131, 1130f
paraesophageal, 952–953, 953f, 957
parastomal, 1132
umbilical. See Umbilical hernia.
Herniography, 987
Herpangina, oropharyngeal, 716–717
Herpes simplex virus infection, in renal transplant
patient, 651t
Herpes simplex virus (HSV-1) vectors, for gene
transfer, 24t, 25
Herpesvirus infections, in lung cancer patient, 861
Heteropagus twins, 1737–1738, 1737t, 1738f
Heterotaxia, in intestinal rotation and fixation
disorders, 1115, 1120, 1122f
Hiatal hernia, 957
axial, 957
paraesophageal, 957
HIDA scan, in choledochal cyst, 1334–1335
High-frequency ventilation, 119
Hilar twist, in thoracic trauma, 274–275
Hilgenreiner line, 1700–1701
Hindgut duplications, 1157, 1161–1163, 1161f, 1162f
Hip
developmental dysplasia of, 1699–1703
complications of, 1703
diagnosis of, 1699–1701, 1700f, 1701f
etiology of, 1699
incidence of, 1699
pathoanatomy of, 1701, 1701t
recurrent dislocation in, 1703
terminology of, 1699
treatment of, 1701–1703, 1702f, 1702t, 1703f
teratologic dislocation of, 1699
Hirschsprung-associated enterocolitis (HAEC) score,
1277, 1277t
Hirschsprung disease, 1265–1281
clinical presentation in, 1266, 1266t
colonic atresia in, 1247
conditions associated with, 1266, 1266t
diagnosis of, 1266–1268, 1267f, 1268f
enterocolitis in, 1266, 1277, 1277t
etiology of, 1265–1266
historical perspective on, 1265
long-segment, surgical approach to, 1272–1274,
1272f, 1273f
long-term outcomes of, 1274–1277
meconium ileus and, 1077
meconium plug syndrome in, 1250–1251
molecular genetics of, 20–21, 21t, 1266
postoperative care in, 1274
Hirschsprung disease (Continued)
preoperative management of, 1268–1269, 1269f
surgical management of
colostomy in, 1240, 1242, 1269, 1269f, 1270
enterocolitis after, 1277, 1277t
fecal soiling after, 1276, 1276t
for long-segment disease, 1272–1274, 1272f,
1273f
for near-total intestinal aganglionosis,
1272–1274
obstructive symptoms after, 1274–1277, 1274t,
1275f
pull-through for
endorectal, 1269–1270, 1269f, 1272
laparoscopic, 1270, 1270f
obstructive symptoms after, 1274–1277,
1274t, 1275f
transanal (perineal), 1271–1272, 1271f
ultrashort-segment, 1278
variant, 1277–1278, 1279–1289, 1280t
desmosis coli as, 1278
hypoganglionosis as, 1277–1278, 1282–1283,
1283f
internal anal sphincter achalasia as, 1278,
1283–1287, 1284f
intestinal neuronal dysplasia as, 1277–1278,
1279–1282, 1281f
megacystis-microcolon-intestinal hypoperistalsis
syndrome as, 1285, 1286f
ultrashort-segment Hirschsprung disease as,
1278
His angle, 947–948, 948f
Histamine H2 receptor antagonists
for gastroesophageal reflux disease, 953
in parenteral nutrition, 191
for peptic ulcer disease, 1033
for stress ulcers, 1034
Histiocytoma, malignant fibrous
breast, 777
pulmonary, 567
Histiocytosis, Langerhans cell, 482
Histone deacetylase inhibitors, 410
Histone modification of DNA, 402
Histoplasmosis, pulmonary, 864
History of pediatric surgery, 1–20
in 19th century, 3
in 20th century, 4–17
in Asia, 15–16, 16f
in Australia and New Zealand, 15
in Canada, 9–10
clinical advances and, 8–9
in developing countries, 17, 17f
education/training programs and, 6–9
in Europe, 12–15, 13f, 14f, 15f
in Ireland, 11–12
research and, 7–8
in United Kingdom, 10–11, 11f, 12f
in United States, 4–6, 4f, 5f, 6f
HIV/AIDS
lung infections in, 862–864, 863f
transfusion-related, 177
HLA (human leukocyte antigen) system, in
transplantation, 614–615, 615f
heart, 663
kidney, 620
Hoarseness, after lung transplantation, 678
Hodgkin lymphoma, 517–522
clinical presentation in, 518, 518f
diagnosis of, 518–519, 518f
epidemiology of, 517–518
histopathology of, 518f, 519, 519f, 519t
lymphocyte-predominant, 519, 520f, 521
Reed-Sternberg cells in, 517, 518f, 519
staging of, 519–520, 519t
survival rate for, 517, 518f
treatment of, 520–522
chemotherapy for, 520
complications of, 522
novel therapy for, 521–522
radiation therapy for, 520–521, 522
breast cancer after, 777
risk classification and, 520, 521
surgical, 520
Homeobox (Hox) genes, in hypospadias, 1536
Volume 1, pages 1–768; Volume 2, pages 769–1738
INDEX I-21
Hormones, ovarian tumors and, 531t
Horner syndrome, in neuroblastoma, 442–443
Horseshoe kidney, 1406–1409, 1408f, 1409f
Wilms’ tumor in, 1408
Host defense
anatomic barriers in, 145–146, 145f
augmentation of, for necrotizing enterocolitis,
1205–1206
bacterial virulence and, 149–150
cell-mediated immunity in, 146–148
humoral immunity in, 148–149
in necrotizing enterocolitis, 1194
neonatal, 150–152
in sepsis, 145–152, 145f
Host-versus-graft response
immunologic basis of, 609–613, 609f, 610f, 613f
tissue typing and, 614–615, 615f
Hox11L1 gene, in intestinal neuronal dysplasia,
1279–1280
Human bite wounds, 341
Human chorionic gonadotropin
in cryptorchidism, 1009
in ovarian tumors, 530–531, 530t
in testicular tumors, 550
Human genome, alteration of, in gene transfer, 25–26
Human growth hormone
for burn patient, 380, 381
for intestinal adaptation promotion, 1141
Human herpesvirus 8 infection, in non-Hodgkin
lymphoma, 522–523
Human immunodeficiency virus. See HIV/AIDS.
Human leukocyte antigen. See HLA (human
leukocyte antigen) system.
Human metapneumovirus infection, as pneumonia,
859, 861
Human papillomavirus infection
in gingivostomatitis, 716–717
in recurrent respiratory papillomatosis, 726
Human papillomavirus vaccine, for recurrent
respiratory papillomatosis, 844
Human patient simulator, 74
Human T-cell leukemia virus type 1, in non-Hodgkin
lymphoma, 522–523
Humerus, fracture of, 332f
in birth trauma, 391–392
in child abuse, 389
Humoral immunity, 148–149
Hungary, pediatric surgery in, 15
Hunter-Hurler syndrome, 1000
Hydatid disease
hepatic, 1352–1353, 1353f
pulmonary, 859, 859f
Hydatid of Morgagni, 1014–1015
Hydranencephaly, 1682, 1682f
Hydrocele, 986f, 987, 997, 1001, 1083
Hydrocephalus, 1680–1687
benign external, 1682
in brain tumors, 591
in choroid plexus papilloma, 1680–1681
clinical features of, 1682
compensated, 1680–1681
etiology of, 1680f, 1681–1682, 1682f
management of
cerebrospinal fluid shunting for, 1683
complications of, 1683–1686
endoscopic third ventriculostomy for,
1686–1687
with myelomeningocele, 1676, 1677–1678
outcome and prognosis in, 1687
radiologic findings in, 1682
Hydrocodone, 218, 218t
Hydrocolpos, 1290, 1293–1294
with two hemivaginas, 1304, 1305f
Hydrocortisone
for adrenogenital crisis, 1574
for sepsis, 160
Hydrofluoric acid burns, 383
Hydromorphone, 218, 219t
caudal, 224–225
in patient-controlled analgesia, 220, 220t
Hydronephrosis, 1400–1401. See also Ureteropelvic
junction obstruction.
definition of, 1411
diagnosis of, 1414–1420
I-22 INDEX
Hydronephrosis (Continued)
differential diagnosis of, 1414, 1414t
embryogenesis of, 1411–1412
etiology of, 1411, 1412f
imaging of, 1429–1430, 1429f
management of, 1420–1425, 1421f
natural history of, 1411
prenatal
counseling for, 1414
manifestations of, 1413–1414, 1413f, 1413t
spontaneous resolution of, 1420
Hydrops
in congenital pulmonary airway malformation, 85
in cystic lung lesions, 825–826
of gallbladder, 1342
Hydroxyapatite-coated implant, 62
17-Hydroxysteroid, in Cushing syndrome, 561–562
Hymen
imperforate, 1558, 1599–1600, 1599f, 1600f
types of, 1599f
Hyoid bone, 722
Hyperaldosteronism, 563–564
Hyperammonemia, after bladder augmentation or
replacement, 1484
Hyperbilirubinemia, in intestinal failure–associated
liver disease, 1139
Hypercalcemia
differential diagnosis of, 751, 751t
in hyperparathyroidism, 751–752
in neonate, 94
in neuroblastoma, 442
in parathyroid carcinoma, 752
Hypercapnia, permissive, in congenital
diaphragmatic hernia, 818
Hyperemia, zone of, in burns, 371, 371f
Hyperganglionosis, in intestinal neuronal dysplasia,
1280, 1281f
Hypergastrinemia
in short bowel syndrome, 1140–1141
in Zollinger-Ellison syndrome, 1034
Hyperglycemia
after central nervous system injury, 344
in neonate, 101–102, 105–106
with parenteral nutrition, 192
postoperative, 105–106
Hyperinsulinism, 1379–1382, 1380f, 1381f
Hyperkalemia
in neonate, 93
with parenteral nutrition, 192–193
Hypermagnesemia, with parenteral nutrition,
192–193
Hypermetabolic response, to burns, 380–381
Hypernatremia, in neonate, 93
Hyperparathyroidism
causes of, 751–752, 751t
incidence of, 745
neonatal severe, 751
Hyperphosphatemia, with parenteral nutrition,
192–193
Hypersensitization, to pain, 215
Hypersplenism, 1385
in portal hypertension, 1359, 1365, 1368
after portoenterostomy, 1329
Hypertension
after coarctation of the aorta repair, 1652
induced, in trauma patient, 269
after kidney transplantation, 623, 629
after lung transplantation, 680
in neuroblastoma, 442
in pheochromocytoma, 559
portal. See Portal hypertension.
pulmonary. See Pulmonary hypertension.
renovascular. See Renovascular hypertension.
Hyperthermia
after central nervous system injury, 344
malignant, 210–211, 211t
in trauma patient, 269
Hyperthermic intraperitoneal chemotherapy, 503,
504f
Hyperthyroidism
causes of, 747–748, 748t
congenital, 745
Hypertonic saline, for burn fluid resuscitation, 374
Hypertriglyceridemia, with parenteral nutrition, 192
Hypertrophic scarring, after burn injury, 384
Hyperventilation, controlled, in trauma patient, 269
Hypocalcemia
after bladder augmentation or replacement, 1484
in neonate, 94
in sepsis, 155
Hypocalciuric hypercalcemia, familial, 751
Hypoganglionosis, 1277–1278
isolated, 1282–1283, 1283f
Hypoglycemia
in hyperinsulinism, 1379
islet allotransplantations for, 638
in neonate, 100–101, 101t
with parenteral nutrition, 192
in sepsis, 155
Hypokalemia
after bladder augmentation or replacement, 1484
in neonate, 93
with parenteral nutrition, 192–193
Hypomagnesemia
after bladder augmentation or replacement, 1484
with parenteral nutrition, 192–193
Hypomastia, 771, 772f, 773
Hyponatremia
in neonate, 93
in short bowel syndrome, 198, 1137
Hypoparathyroidism, after thyroidectomy, 749
Hypopharynx, anatomy of, 716
Hypophosphatemia, with parenteral nutrition,
192–193
Hypoplastic left heart syndrome, 1663–1665, 1664f
cardiovascular management in, 139–140
heart transplantation for, 659, 662–663, 664, 666f
Hypopnea, 718–719
Hypopnea index, 719
Hypospadias, 1531–1557
in 46,XY DSD, 1571
anatomy of, 1537–1551, 1539f
associated anomalies with, 1534–1535
chordee in, 1531, 1533f, 1539
repair of, 1546–1550, 1549f, 1550f, 1551f
urethral plate preservation and, 1543, 1544f,
1545, 1548f
classification of, 1531, 1532f
embryogenesis of, 1531–1532, 1533f, 1535f, 1538f
etiology of, 1535–1537
historical perspective on, 1531
incidence of, 1532–1534, 1532f
meatal abnormalities in, 1538–1539
penoscrotal transposition with, 1563
skin and scrotal abnormalities in, 1539–1540,
1539f
Hypospadias repair, 1540
age for, 1552
algorithm for, 1540f
Bracka two-stage buccal graft repair in, 1546,
1548f, 1549f
complications of, 1552–1553, 1552f
for curvature, 1546–1550, 1549f, 1550f, 1551f
distal (anterior), 1540–1543, 1541f, 1542f, 1543f
GAP procedure in, 1540–1541, 1542f
MAGPI technique in, 1540, 1541f
in male gender assignment surgery, 1582–1583,
1585f
Mathieu or perimeatal-based flap procedure in,
1542–1543, 1543f
multiple failures with, 1550–1551
onlay island flap in, 1544, 1545f
posterior, 1543–1546, 1544f, 1545f, 1547f,
1548f, 1549f
pyramid procedure in, 1542, 1542f
results of, 1553
technical considerations in, 1551–1552
transverse tubularized island flap in, 1544–1545,
1547f
tubularized plate urethroplasty in, 1543, 1543f,
1544f
two-stage, 1545, 1548f
urethral mobilization in, 1542
urethral plate preservation in, 1543, 1544f
Hypotension
after central nervous system injury, 343
intracranial, from overshunting, 1685–1686
sepsis-induced, 141
Volume 1, pages 1–768; Volume 2, pages 769–1738
Hypothalamic/chiasmatic astrocytoma, 597–598,
598f
Hypothalamic-pituitary-adrenal (HPA) axis, 558
Hypothalamic tumors, 593–594
Hypothermia
in neonate, 99
in trauma patient, 268, 269
Hypothesis testing, 233
Hypothyroidism
causes of, 747, 748t
congenital, 745
in hepatic hemangioma, 1618
ovarian cysts and, 536, 548
Hypoventilation, in burn injury, 374–375
Hypovolemia, permissive, during burn fluid
resuscitation, 374
Hypovolemic shock, in cervical spine injury,
356–357
Hypoxemia, 114
Hypoxia, after central nervous system injury, 343
Hysterotomy, 79–80, 80f
I
Ibuprofen, 216, 216t
for patent ductus arteriosus, 1648
Ice or ice-water bag, for supraventricular tachycardia,
138
Ifosfamide, 407t
Ileal atresia. See Jejunoileal atresia and stenosis.
Ileal conduit diversion, 1489–1490, 1489f
Ileal pull-through, straight endorectal, 1223,
1225–1227, 1225f, 1227f
laparoscopic, 1225–1226, 1226f
Ileoanal pouch procedure, 1223–1224
anastomosis in, 1214
complications and outcomes of, 1227–1229
Crohn disease after, 1228–1229
failure of, 1228–1229
J pouch construction in, 1225
long-term follow-up for, 1229
open operative approach in, 1224–1225,
1224f
pouch configurations for, 1223–1224, 1223f
pouch construction in, 1224–1225, 1224f, 1225f,
1226f
preoperative medical therapy and, 1229
quality of life after, 1229
straight pull-through technique in, 1223,
1225–1226, 1225f
Ileoanal pull-through
approaches to, 1223
laparoscopic, 1225–1226, 1226f
stooling patterns after, 1226, 1227f
Ileocecal cystoplasty, 1473, 1485, 1492
Ileocecal exstrophy, 1526
Ileocecal valve, in short bowel syndrome, 1135
Ileocecectomy, in Crohn disease, 1214
Ileocolic intussusception, 1098, 1098f, 1107
Ileocystoplasty, 1473, 1474f, 1475f, 1492
Ileorectal anastomosis
colectomy with, 1181
for familial adenomatous polyposis, 488
Ileostomy. See also Enterostoma.
choices for, 1238–1240, 1239f, 1240f
complications of, 1244–1245, 1245t
indications for, 1236
permanent, proctocolectomy with, 1223
protective, in ulcerative colitis, 1226–1227
stoma care in, 1244
technical aspects of, 1241–1242, 1241f, 1242f,
1243f, 1245f
Ileovesicostomy, incontinent, 1490
Ileum
duplications of, 1160–1161, 1161f
inflammation of, in ulcerative colitis, 1218
for Mitrofanoff neourethra, 1480, 1493, 1494f
vaginal replacement with, 1304, 1306f
Ileus. See also Meconium ileus.
in isolated hypoganglionosis, 1282
postoperative, 1129
Iliac ectopic kidney, 1405
Ilioinguinal-iliohypogastric nerve block, 222–223,
222f

Image-guided therapy, 50–52
with computed tomography, 51
general requirements for, 50–51
with magnetic resonance imaging, 51
significance of, 50
with ultrasonography, 51
Imatinib, 410
for dermatofibrosarcoma protuberans, 505
for gastrointestinal stromal tumors, 485
Imbrication, in megaureter repair, 1501–1502,
1501f, 1502f
Imipramine, for nocturnal enuresis, 1464–1465
Immobilization, after bladder exstrophy repair,
1519–1521, 1522f
Immune response
in burn injury, 371
in gene transfer, 25
postoperative, 105
spleen in, 1385–1386
in transplantation, 610–611, 611f, 612f
Immunity
cell-mediated, 146–148
humoral, 148–149
Immunization, before splenectomy, 1388
Immunocompromised patient, lung infections in,
860–862
with cancer, 860–862, 860f, 862f
with HIV/AIDS, 862–864, 863f
Immunodeficiency, transfusion therapy in, 176
Immunoglobulin(s)
in host defense, 148
intravenous
for immune thrombocytopenic purpura, 170,
1387
for necrotizing enterocolitis, 1205
for sepsis, 162
in neonate, 151
radiolabeled, 54
Immunoglobulin A, 148, 151
prophylactic, for necrotizing enterocolitis,
1205
Immunoglobulin D, anti-, for immune
thrombocytopenic purpura, 170
Immunoglobulin G, 148, 151
prophylactic, for necrotizing enterocolitis, 1205
Immunoglobulin M, 148, 151
Immunohistochemistry, of ovarian tumors, 531
Immunosuppressive therapy
for aplastic anemia, 166
for transplantation
heart, 665–667, 667t
intestinal, 655–656
islet cell, 638–640
liver, 649–650, 650t
in hepatoblastoma, 475
in hepatocellular carcinoma, 479
lung, 676–677, 676t
pancreas, 634–635
principles of, 605–606, 607, 607f, 608–609,
611–613, 612f
renal, 624–625
for ulcerative colitis, 1222, 1229
Immunotherapy, 411
for neuroblastoma, 457–458
Implant, hydroxyapatite-coated, 62
In situ hybridization, 404t
Incisional biopsy, open, 422
Incontinence
in anorectal malformations, 1291
after bladder exstrophy repair, 1523
in cerebral palsy, 1460–1461
ectopic ureter and, 1445
fecal
after anorectoplasty, 1308–1309, 1308t, 1309f
bladder augmentation or replacement and,
1480–1482, 1482f
with constipation, 1313, 1314
after pull-through for Hirschsprung disease,
1276, 1276t
without constipation, 1315
after ileoanal pouch procedure, 1228
in posterior urethral valves, 1462
pseudo-, after pull-through, 1276
structural consequences of, 1467
Incontinent urinary diversions, 1487–1490, 1488f,
1489f
India, pediatric surgery in, 17
Indiana pouch, 1473, 1495, 1495f
Indomethacin
necrotizing enterocolitis and, 1189
for patent ductus arteriosus, 1648
Induction chemotherapy, 406
Industry, surgeons and, 242–243
Infant. See Neonate; Premature infant.
Infection
host defense against, 145–152, 145f. See also Host
defense.
intracranial, 1693–1697, 1694f, 1695f
intraspinal, 1697
with open fractures, 334
shunt-related, 1685
after tracheotomy, 839
transfusion-related, 177
in transplant patient
heart, 667
intestinal, 656
lung, 680
renal, 628, 650–651, 651t
Infection control measures, for necrotizing
enterocolitis, 1204–1205
Infertility
in bladder exstrophy, 1524
in cryptorchidism, 1007–1008
after inguinal hernia repair, 998, 999
after orchidopexy, 1013–1014, 1013t
after radiation therapy for Hodgkin lymphoma,
522
tubal, appendicitis and, 1262
in varicocele, 1017
Infiltration anesthesia, 221
Inflammation. See also Host defense; Systemic
inflammatory response syndrome (SIRS).
in burn injury, 371
in Meckel diverticulum, 1091, 1091f
in necrotizing enterocolitis, 1195, 1196f
Inflammatory bowel disease. See also Crohn disease;
Ulcerative colitis.
colorectal cancer in, 489
cost of, 1209
Crohn disease and, 1215
multidetector computed tomography in, 41–42
Inflammatory cascade, 1189, 1190t
Inflammatory markers, 153–154
Inflammatory mediator antagonists, for necrotizing
enterocolitis, 1207
Inflammatory polyps, intestinal obstruction with,
1132–1133
Inflammatory pseudotumor
hepatic, 462
intestinal obstruction in, 1133
pulmonary, 567
Infliximab
for Crohn disease, 1212
for ulcerative colitis, 1221–1222, 1229
Information bias, 233
Information-guided therapy, 50. See also Image-
guided therapy.
Information sharing, communication versus, 249
Informed consent, 238–239, 239t
Infundibular septum, in tetralogy of Fallot,
1659–1660, 1659f
Inguinal canal, 1003, 1004f
Inguinal hernia, 985–1004
bilateral, 993–994
clinical features of, 986–987
in connective tissue disorders, 1000
cryptorchidism and, 1008
in cystic fibrosis, 1000–1001
direct, 1000
embryology of, 986, 986f, 986t
examination for, 987, 987f
historical perspective on, 985
versus hydrocele, 986f, 1001
hypospadias with, 1534
incarcerated, 987, 994–997, 995f
diagnosis of, 995
intestinal injury with, 998
nonoperative management of, 995–996, 995f
Volume 1, pages 1–768; Volume 2, pages 769–1738
INDEX I-23
Inguinal hernia (Continued)
operative management of, 996–997
in premature infant, 994
testicular atrophy with, 998
incidence of, 985
in intersex patient, 1001
in meconium ileus, 1083
mortality in, 999
peritoneal dialysis and, 999–1000
in premature infant, 989, 994, 999
radiologic findings in, 987–988, 988f
recurrent, 997
repair of, 988–993
adrenal rests found during, 1001
anesthesia for, 988–989
complications of, 997–999
contralateral exploration with, 993–994
laparoscopic, 994
laparoscopic, 991–993, 992f
with mesh, 999
open technique of
in female, 991, 991f
in male, 989–991, 990f
pain control after, 988–989
same-day, 989
timing of, 989
sliding, 1000
fallopian tube in, 991, 991f, 998, 1000
ventriculoperitoneal shunts and, 999
Inguinal hydrocele, in meconium ileus, 1083
Inguinal pouch, superficial, testis in, 1005–1006,
1005f, 1008–1009
Inguinodynia, after inguinal hernia repair, 999
Inhalation anesthesia
agents for, 201, 202f, 202t, 207–209, 207t
laryngospasm associated with, 203
malignant hyperthermia with, 210–211, 211t
Inhalation injury, 375–376, 376t
Inhibin, in ovarian tumors, 530t, 531
Injection therapy, for unicameral bone cysts, 584
Injury. See also Trauma.
bicycle/motorcycle, prevention of, 259
epidemiology of, 261–262, 262f
fire, prevention of, 258–259
firearm, prevention of, 259
intentional, 262
mortality from, 261–262, 262f
motor vehicle, prevention of, 258, 258f
pedestrian, 259
poisoning, unintentional, prevention of, 259–260
prevention of, 253–261
design strategies for, 257–258
evaluation of, 260
Haddon approach to, 255, 256t
initiatives related to, 258–260, 258f
Internet resources for, 260t
priorities of, 255–260, 257t
savings from, 255, 256t
pyramid characterization for, 256f
resuscitation after, 262–263. See also Emergency
management.
Innominate artery
compression of, 1665–1666, 1667–1668, 1669f
erosion of, after tracheotomy, 839–840
tracheal compression by, 851, 851f, 853, 854
Innominate vessel, injury to, 285–286
Inotropic agents
for congestive heart failure, 135–138, 137t
for fluid-refractory shock, 159–160
after heart transplantation, 665
Insertional mutagenesis, in gene transfer, 25
Inspiratory capacity, 112, 113f
Inspiratory reserve volume, 113, 113f
Institute for Patient- and Family-Centered Care
(IPFCC), 247
Institutional review boards, 63, 245
Insulin
for burn injury, 380, 381
micropump delivery of, 62
parenteral nutrition and, 191, 192
in perinatal period, 100
Insulin/glucose ratio, postoperative, 105
Insulin-like growth factor 2, in renal development,
1395
I-24 INDEX

Insulin-like growth factor 3, in testicular descent, Intestinal neuronal dysplasia (Continued) Intestinal transplantation (Continued)
1003 outcome of, 1282 indications for, 653–654, 654f
Insulin resistance syndrome, 1042–1043 treatment of, 1282 operative procedures in, 654–655, 655f
Insulinoma, 1383 Intestinal obstruction. See also Meconium ileus; postoperative care in, 655–656
Integra, in burn care, 378 Volvulus. results of, 656–658, 657f
Intensity-modulated radiation therapy, 413 adhesions with in short bowel syndrome, 1145
Intensive care unit, neonatal, 8 inflammatory, 1130 timing of, 653–654
Intercellular adhesion molecule-1, in neutrophil postoperative, 1127–1129, 1128f, 1129f Intestinal tumors, 485
adhesion, 146 in ascariasis, 1133 in Peutz-Jeghers syndrome, 1184
Intercostal drainage. See Chest tube. causes of, 1127–1135 Intestinal vaginoplasty, 1596–1598, 1597f
Interferon-a, 411 distal, 1082 Intestine. See also Colon; Gastrointestinal entries;
for infantile hemangioma, 1616 in duplication cysts, 1133 Small intestine.
for subglottic hemangioma, 850 embryology of, 1127 bacterial overgrowth in
Interferon-g, 149 after foreign body ingestion, 1133 methods to decrease, 1206–1207
in atypical mycobacterial lymphadenitis, 741–742 gastrointestinal lesions with, 1132–1133 in short bowel syndrome, 1140
Interleukin(s), in stress response, 104 hernias with, 1130–1131, 1130f, 1131f distention of, in necrotizing enterocolitis, 1188f,
Interleukin-1, 149 in Hirschsprung disease, 1266 1198
in necrotizing enterocolitis, 1190t, 1191 after ileoanal pouch procedure, 1228 echogenic, fetal, 87
Interleukin-2, 149 in inflammatory pseudotumor, 1133 invagination of. See Intussusception.
Interleukin-2 receptor antibodies, in transplantation in intussusception, 1093–1094 mucosal gland abnormalities of, in meconium
heart, 665–666 in Meckel diverticulum, 1090–1091, 1090f, 1091f ileus, 1074
liver, 650t in mesenteric and omental cysts, 1133, stretching of, 1144
renal, 624 1166–1167, 1167f tissue-engineered, 32, 1144
Interleukin-4, in necrotizing enterocolitis, 1190t, postoperative Intra-abdominal pressure
1191 adhesive, 1127–1129, 1128f, 1129f measurement of, 298
Interleukin-6, 149 after appendectomy, 1262 in trauma patient, 298–299, 299f
in necrotizing enterocolitis, 1190t, 1191 ileus and, 1129 Intra-abdominal testis, 1005, 1005f
Interleukin-8, 149 intussusception and, 1130 Intracellular fluid, in neonates, 91–92
in necrotizing enterocolitis, 1190t, 1191 after pull-through for Hirschsprung disease, Intracerebral hematoma, traumatic, 346
Interleukin-10, in necrotizing enterocolitis, 1190t, 1274–1277, 1274t, 1275f Intracranial aneurysm, traumatic, 353
1191–1192 spectrum of disorders causing, 1127 Intracranial hematoma
Interleukin-11, in necrotizing enterocolitis, 1190t, Intestinal perforation in birth trauma, 392
1192 diagnosis of, 290 removal of, 351, 351f, 352f
Interleukin-12, 149 with intussusception reduction, 1107–1108, 1108f Intracranial hemorrhage, with extracorporeal life
in necrotizing enterocolitis, 1190t, 1191 primary peritoneal drainage for, 1201 support, 129
Interleukin-18, in necrotizing enterocolitis, 1190t, in utero, 87 Intracranial hypotension, from overshunting,
1191 Intestinal pseudoobstruction, 1133–1134, 1134t 1685–1686
Intermittent mandatory ventilation, 118 chronic, 1250 Intracranial infections, 1693–1697, 1694f,
synchronized, 118 esophageal dysmotility in, 944 1695f
Intermittent positive-pressure ventilation, 117 Intestinal rotation and fixation, 1111–1127 Intracranial lesions, stereotactic radiosurgery for,
International Neuroblastoma Pathology disorders of. See also Volvulus. 53–54
Classification (INPC), 446–447, 447f anomalies associated with, 1115 Intracranial pressure. See also Hydrocephalus.
International Society of Pediatric Oncology (SIOP) asymptomatic, 1116 increased
staging system for Wilms’ tumor, 423, 424t, atypical, 1114–1115, 1120, 1120f, 1124 in brain tumors, 591
429–430 classification of, 1114–1115, 1120, 1120f in craniosynostosis, 692
Intersex. See Disorders of sex development (DSD). clinical manifestations of, 1115–1116, 1116f management of, 350–351, 351t
Interstitial cells of Cajal, deficiency of, in internal anal complications of, 1124–1125 monitoring of
sphincter achalasia, 1284 growth disorders in, 1114 in epilepsy surgery, 1689–1690
Intestinal adaptation, promotion of, in short bowel with heterotaxia, 1115, 1120, 1122f after shunt implantation, 1686
syndrome, 1141 historical perspective on, 1111 in traumatic brain injury, 350
Intestinal aganglionosis, near-total, 1272–1274 management of in trauma patient, 268–269
Intestinal atresia and stenosis laparoscopic versus open reduction in, Intrahepatic duct cysts, dilatation of, 462
colonic, 1247, 1248f 1122–1123 Intrahepatic hematoma, 464, 465f
duodenal, 1051–1060. See also Duodenal atresia operative, 1120–1122, 1123f in birth trauma, 392, 392f
and stenosis. postoperative, 1124–1125 Intraosseous line, in trauma patient, 266, 267
familial, 1060–1061, 1061t, 1065, 1066f preoperative, 1120 Intraperitoneal fluid. See also Ascites.
genetics of, 1248 resection and second-look procedures in, free, 1171
jejunoileal. See Jejunoileal atresia and stenosis. 1124 computed tomography of, 308
Intestinal conservation, in short bowel syndrome, with mesocolic hernia, 1117, 1118f, 1120, 1124 in necrotizing enterocolitis, 1198
1141 radiologic findings in, 1117–1120, 1119f, Intraspinal infections, 1697
Intestinal dysmotility. See also Constipation; 1120f, 1121f, 1122f Intrathoracic access and procedures, 873–876
Hirschsprung disease. reversed, 1115 Intrauterine growth restriction, 89–90
in intestinal rotation and fixation, 1124 with colonic obstruction, 1117, 1117f, 1124 Intravenous anesthesia, 201, 202f, 211–212, 212f
intestinal transplantation for, 653, 654f terminology in, 1114–1115, 1115f Intravenous immunoglobulin
after pull-through, 1275–1276 normal, 1111–1117, 1115f, 1119f for immune thrombocytopenic purpura, 170,
in short bowel syndrome, 1140 cecocolic loop in, 1113, 1113f, 1114f 1387
Intestinal failure. See also Short bowel syndrome. duodenojejunal loop in, 1111, 1112f for necrotizing enterocolitis, 1205
causes of, 653, 654f fixation in, 1114–1115, 1115f for sepsis, 162
definition of, 1135 side and direction of, 1112 Intraventricular hemorrhage, 347–348, 347f
liver disease with, 1138–1139 simultaneous rotation of both ends and entire Introital cysts, 1608
management of, 653 intestinal tract in, 1113–1114, 1114f Introital masses, 1606
nutritional support for, 1137–1138 Intestinal stoma. See Enterostoma. Intubation, endotracheal. See Endotracheal
transplantation for, 653–659 Intestinal stricture intubation.
Intestinal ischemia, in necrotizing enterocolitis, 1194 in Crohn disease, 1210, 1210f Intussusception, 1093–1114
Intestinal lengthening procedures, in short bowel balloon dilatation for, 1214 anatomic, 1098
syndrome, 1141–1144, 1142f, 1143f, 1144f surgery for, 1213–1214, 1213f, 1214f definition of, 1093
Intestinal loops, persistent dilated, in necrotizing after ileoanal pouch procedure, 1228 diagnosis of, 1095f, 1099
enterocolitis, 1198–1199, 1201 in necrotizing enterocolitis, 1203, 1249, 1250f clinical, 1095f, 1099
Intestinal neuronal dysplasia, 1250, 1277–1278, posttraumatic, 1133 radiologic, 1099–1101, 1100f, 1101f
1279–1282 Intestinal transplantation, 653–659 epidemiology of, 1094
clinical presentation in, 1280 abdominal wall closure after, 655, 656f future expectations for, 1109–1110
diagnosis of, 1280–1282, 1281f assessment and preparation for, 654 historical perspective on, 1093
history and pathogenesis of, 1279–1280 complications of, 656 idiopathic, 1095f, 1096–1097
incidence of, 1280 immunosuppressive therapy for, 655–656 ileocolic, 1098, 1098f, 1107

Volume 1, pages 1–768; Volume 2, pages 769–1738


Intussusception (Continued)
in jejunoileal atresia and stenosis, 1060, 1062,
1065f
in Meckel diverticulum, 1090–1091, 1090f
in meconium ileus, 1082
mortality in, 1109
in neonate, 1099
outcome of, 1108–1109
overview of, 1093–1094, 1094f, 1095f
from pathologic lead point, 1097–1098, 1097f
pathophysiology of, 1095–1099
physical examination in, 1099
postoperative, 1098
after Ladd procedure, 1125
recurrent, 1098–1099, 1108, 1109
red currant jelly stool in, 1147, 1149f
spontaneous reduction of, 1096
treatment of, 1101–1108
medical, 1102
operative, 1106–1108, 1106f
complications of, 1108
laparoscopic, 1106
laparotomy for, 1106–1108, 1107f
radiologic, 1102–1106, 1102f
care after, 1105–1106
complications of, 1108
with delayed repeat enema, 1105
history of, 1093
with hydrostatic barium enema, 1104,
1104f
methods to improve, 1104–1105
with pneumatic air enema, 1103, 1103f
tube-related, 1098
types of, 1095–1099
Invasins, 150
Inverse ratio ventilation, 118–119
Iodine, radioactive
for Graves disease, 747–748
for thyroid cancer, 749–750
Iowa I operation, 1142, 1143f
Ireland, pediatric surgery in, 11–12
Irinotecan, 407t
Iron
after bariatric surgery, 1046
dietary, requirements for, 167
in parenteral nutrition, 191, 191t
Iron deficiency anemia, 167–168, 1149–1150
Iron dextran, in parenteral nutrition, 191, 191t
Irrigation, enterotomy with, for meconium ileus,
1079–1080, 1080f
Ischemia
in central nervous system injury, 344
prevention of, 344
digital, 339, 367
gastric mucosal, stress ulcers and, 1031
hand, 339
intestinal, in necrotizing enterocolitis, 1194
limb, chronic, 1641–1642
upper extremity, 1642, 1643f
in vascular trauma, 362
Ischemic stroke, 1643–1645, 1643f, 1644f,
1645f
Island flap
onlay, in hypospadias repair, 1544, 1545f
transverse tubularized, in hypospadias repair,
1544–1545, 1547f
Islet cell carcinoma, 1383–1384
Islet transplantation, 637–641
allotransplantations in, 638–641, 639f
autotransplantations in, 638, 638f
Isoflurane, 202f, 202t, 207t, 208
Isoproterenol, after heart transplantation, 665
Isotretinoin, for neuroblastoma, 457
Italy, pediatric surgery in, 14, 15f
J Ketamine, 212, 216t, 217
J pouch
for long-segment Hirschsprung disease, 1272
for ulcerative colitis, 1224, 1225, 1226–1227,
1227f
Jadassohn sebaceous nevus, 1714
Japan, pediatric surgery in, 16, 16f
Jarcho-Levin syndrome, 807, 808f
Jaundice
in biliary atresia, 1321, 1323
in choledochal cyst, 1334
in cholelithiasis, 1342
in jejunoileal atresia and stenosis, 1061,
1061t
in portal hypertension, 1360
Jejunal conduits, metabolic acidosis with, 1484
Jejunal feedings
after gastric transposition esophagoplasty, 937
supplemental, for congenital microgastria,
1039
Jejunal interposition, for esophageal replacement,
907, 929t, 934, 934t
Jejunal tubes, 186
Jejunoileal atresia and stenosis, 87, 1059–1075
with apple-peel (Christmas-tree) deformity,
1064–1065, 1066f
classification of, 1063–1064, 1063f, 1065,
1066f
with colonic atresia, 1247
diagnosis of, 1061–1065, 1061t
postnatal, 1061–1062, 1063f, 1064f, 1065f
prenatal, 1061, 1062f
differential diagnosis of, 1062–1063
etiology of, 1060–1061, 1061t
historical perspective on, 1059
meconium ileus and, 1077, 1077f
morbidity and mortality of, 1070–1071
multiple, 1064, 1065, 1066f
pathologic findings in, 1063–1065, 1063f, 1065f,
1066f
prevalence of, 1059–1060
treatment of, 1066–1069
operative techniques for, 1066–1069, 1068f,
1069f
postoperative care in, 1069–1070
Jejunoileal bypass, 1041–1042
Jejunoplasty, tapering, for jejunoileal atresia and
stenosis, 1068, 1069f
Jejunostomy. See also Enterostoma.
choices for, 1237f, 1238
complications of, 1244–1245, 1245t
indications for, 1236
in necrotizing enterocolitis, 1202, 1203
stoma care in, 1244
technical aspects of, 1240–1241
Jeune syndrome, 805–807, 807f, 808f
Journal of Pediatric Surgery, 7
Jugular vein
injury to, 717
internal, in extracorporeal life support, 125f,
126–127
Justice principle, 237
bariatric surgery and, 242
Juvenile nasopharyngeal angiofibroma,
715–716
Juvenile polyps. See Polyp(s), juvenile.
Juvenile secretory carcinoma, 777
K risk stratification in, 428t, 429
Kaposi sarcoma, 1620
Kaposiform hemangioendothelioma, 464,
1619–1620, 1619f
Kaposiform lymphatic anomaly, 1619
Kasabach-Merritt syndrome, 459, 1619–1620, 1619f
Kasai, M., 16, 16f
Kasai hepatic portoenterostomy.
See Portoenterostomy.
Kawasaki disease
brachial artery aneurysm in, 1643, 1643f
cervical lymphadenopathy in, 743
hydrops of gallbladder in, 1342
Keratinocytes, 1711
cultured, in burn care, 380
for burns, 382–383
caudal, 224–225
Ketogenesis, in neonate, 102
Ketone body use, in neonate, 102
Ketorolac, 216–217, 216t
17-Ketosteroid, urinary, ovarian tumors and, 531t
KIAA1549-BRAF fusion protein, in astrocytoma, 601
Volume 1, pages 1–768; Volume 2, pages 769–1738
INDEX I-25
Kidney. See also Renal entries.
congenital anomalies of
injury risk with, 312
related to abnormal ascent, 1405–1406, 1407f,
1408f
related to abnormal fusion, 1406–1409, 1408f,
1409f, 1410f
crossed ectopia of, 1409, 1410f
cystic disease of, 1396
acquired, 1403
in angiomyolipoma, 1399
benign multilocular, 1401–1402, 1402f
caliceal diverticulum as, 1403
classification of, 1396, 1396t
multicystic dysplastic kidney as, 1395, 1396,
1399–1401, 1400f, 1401f
polycystic, 1396
autosomal dominant (adult), 1396–1397,
1397f
autosomal recessive (infantile), 1397–1398,
1398f
simple, 1402–1403, 1403f
solitary multilocular, 439
in tuberous sclerosis, 1399
in von Hippel-Lindau disease, 1399
duplex collecting system of. See Duplex collecting
system.
ectopic, 1405–1406, 1407f, 1408f
embryology of, 1395, 1405, 1406f, 1411–1412
horseshoe, 1406–1409, 1408f, 1409f
Wilms’ tumor in, 432, 1408
hydronephrotic. See Hydronephrosis.
hypoplasia/hypodysplasia of, 1395–1396
pelvic, 1405, 1407f
in prune-belly syndrome, 1506–1507, 1509f,
1510f
single, Wilms’ tumor in, 432
thoracic, 1406, 1407f
transplantation of. See Renal transplantation.
trauma to, 315–318
anatomic considerations in, 311–312
blunt, 315, 316f
clinical features of, 312
complications of, 317–318
diagnostic studies in, 312–314
epidemiology of, 311
follow-up and outcomes of, 318
grading of, 314, 314t, 315f
management of, 315–318
mechanisms of injury in, 311
penetrating, 315–316
surgical management of, 319
vascular, 311, 313, 316–317, 318f
tumors of, 423–446. See also Wilms’ tumor.
anaplastic histology in, 428
clear cell sarcoma as, 437
congenital mesoblastic nephroma as, 438–439
cystic, 439, 439f
in neonate, 432
renal cell carcinoma as, 438, 438f
rhabdoid, 437–438
treatment strategy for, 423
Kidney stones. See Urolithiasis.
Kiesselbach plexus, 712
Kikuchi disease, cervical lymphadenopathy in, 743
Kimura procedure, for long-segment Hirschsprung
disease, 1272, 1273f
KIT mutations, in gastrointestinal stromal tumors,
485
Klinefelter syndrome, gynecomastia in, 1716–1718
Klippel-Feil syndrome, 1587
Klippel-Trenaunay syndrome, 427, 1627–1629,
1627f, 1628f
Knee, dislocation of, congenital, 1705–1706
Kocher maneuver, in adrenalectomy, 565–566,
565f
Kock pouch, 1472, 1473, 1473f, 1475f, 1494
Koop, C. E., 5, 5f
Kropp procedure, 1477–1478, 1479f
Krukenberg procedure, 1722
Kumar clamp technique, 1344–1345, 1345f
Kyphosis, congenital, 1706, 1706f, 1708–1709,
1709f
I-26 INDEX

L Laryngeal webs, 841–842, 841f, 842f Limb ischemia, chronic, 1641–1642

Labia minora, fusion of, 1558–1559 Laryngocele, 725 Limb length discrepancy
Labial adhesions, 1558–1559, 1606 Laryngomalacia, 723–724, 724f, 840–841, 841f in capillary-lymphaticovenous malformation,
Labioplasty, in female gender assignment surgery, Laryngoscopy, 723, 837 1629
1579, 1579f, 1580f Laryngospasm, with inhalation anesthesia, 203 in developmental dysplasia of hip, 1700
Laboratory tests Laryngotracheal stenosis, 844–849, 845f, 845t. Limb lengthening procedures
in appendicitis, 1257 See also Subglottic stenosis; Tracheal stenosis. for bone tumors, 588–590, 589f, 590f
in ascites, 1172, 1173t endoscopic surgery for, 846–847, 846f extensible prostheses as, 588f, 590
in choledochal cyst, 1334 open surgery for, 847–849, 847f, 847t, 848f Limb malformations, with cloacal exstrophy, 1527
in liver cancer, 464–465 Laryngotracheobronchitis, 725 Limb-sparing surgery, for bone tumors, 586
in meconium ileus, 1076–1077 Laryngotracheoesophageal cleft, 850–851, 916–918, Linear accelerator radiosurgery, 52
in necrotizing enterocolitis, 1196 917f, 918f Lingual thyroid, 721, 745, 746f
in ovarian tumors, 530–531, 530t Laryngotracheoplasty Linoleic acid, 182–183
in portal hypertension, 1360–1361 for laryngeal webs, 841–842, 841f Lip, cleft. See Cleft lip and palate.
Laceration for laryngotracheal stenosis, 847–848, 847f, 848f Lipids. See Fats (lipids).
auricular, 711 for vocal cord immobility, 843 Lipogenesis, postoperative, 105–106
in birth injury, 391 Larynx, 722–726. See also Airway; Vocal cords. Lipoid adrenal hyperplasia, 1570
nasal, 715 anatomy of, 722, 837–838 Lipomatous mass, truncal, in Cloves syndrome,
pericardial, 282 atresia of, 841–842 1629–1630, 1630f
pulmonary, 277–279 congenital anomalies of, 723–725, 724f, 725f Lipomyelomeningocele, 1679
Lactate, postoperative elevation of, 105 functions of, 722, 837–838 Lipopolysaccharide, 144, 144f, 146, 150
Lactate dehydrogenase inflammatory disease of, 725–726 in necrotizing enterocolitis, 1190t, 1193
in bone tumors, 581 lesions of, 840–851 Liposarcoma, breast, 777
in ovarian tumors, 531 papilloma of, 843–844, 843f Liquid ventilation, 119–120
Lactated Ringer solution tumors of, 726, 726f in congenital diaphragmatic hernia, 823, 823f
for burns, 374, 374t Laser ablation Lithopexy, transurethral, for bladder stones,
for intraoperative fluid losses, 206 for capillary malformation, 1621 1438
Lactation, neurogenic, 774 esophageal, 885 Lithotripsy
Lactic acidosis, with bacterial overgrowth, 1140 for infantile hemangioma, 1616 for cholelithiasis, 1343
Lactose, 182, 186–187 for laryngotracheal stenosis, 846 extracorporeal shock wave, for urolithiasis,
Lacuna magna, 1557 for recurrent respiratory papillomatosis, 844 1438
Ladd, W. E., 4, 4f for subglottic hemangioma, 850 Litigations, communication and, 249–250
Ladd bands Lasers, surgical, 49 Littre hernia, Meckel diverticulum with, 1088
division of, 1122, 1123f Latex allergy, bladder augmentation or replacement Liver
intestinal obstruction secondary to, 1116–1117, and, 1491 abscess of, 464, 465f
1131 Laxatives amebic, 1352
Ladd procedure for anal fissure, 1317 diagnosis of, 1349, 1349f, 1350–1351, 1350f
for intestinal rotation and fixation disorders, for constipation, 1314–1315 echinococcal, 1352–1353, 1353f
1120–1122, 1123f Le Fort I osteotomy, 695 pathophysiology of, 1349
intussusception after, 1125 Le Fort III osteotomy, 694, 695 pyogenic, 1349–1350
Lambdoid suture, premature fusion of, 692 LeapFrog Group, 235 treatment of, 1351
Laminectomy Leg length. See Limb length discrepancy. angiosarcoma of, 480
for spinal cord tethering, 1679 Leiomyoma, ovarian, 548 arteriovenous malformation of, 460–461
for spinal epidural abscess, 1697 Leiomyosarcoma biopsy of
Langerhans cell histiocytosis, 482 intestinal obstruction with, 1132 in biliary atresia, 1324
Langerhans cells, 1711 ovarian, 547 in portal hypertension, 1361–1362
Language barriers, 243–244 Lembert sutures, in megaureter repair, 1501–1502, cysts of
Lansoprazole, for peptic ulcer disease, 1501f in autosomal dominant polycystic kidney
1033, 1033t Lennox-Gastaut syndrome, seizures in, 1693 disease, 1396–1397
Lap-belt injuries, 335, 336f Lentiviral vectors, for gene transfer, 24, 24t congenital, 464, 465f
Laparoscopy. See also Minimal access surgery. Leptomyelolipoma, 1679 nonparasitic, 462
in abdominal trauma, 291 Leukemia embryonal sarcoma of, 480
with biopsy, 420 lymphoblastic, testicular tumors in, 552–553 fibrosis of, in choledochal cyst, 1334
contraindications to, 420 megakaryoblastic, hepatic, 482 focal nodular hyperplasia of, 461–462, 462f, 465,
in cryptorchidism, 1013 testicular, 552–553 466f, 482
in disorders of sex development, 1575 Leukocyte(s) hemangioma of, 460–462, 460f, 1613–1614
in inguinal hernia repair, 991–993, 992f count of infantile, 464, 465, 466f, 480–481, 481f,
in intestinal rotation and fixation disorders, abnormal, age group–specific definitions for, 1617–1618, 1618f
1122–1123 152, 152t hemophagocytic lymphohistiocytosis of, 482
in intussusception, 1106 in appendicitis, 1257 herniated, in congenital diaphragmatic hernia, 85,
in Meckel diverticulum, 1152, 1152f in necrotizing enterocolitis, 1196 815
in ovarian tumors, 535 donor infection of, 1349–1353. See also Liver, abscess of.
in rectobladder neck fistula, 1298–1300, 1300f, microchimerism of, 610, 610f cat-scratch disease and, 1351
1301f migration and localization of, 610–611, 611f, hydatid disease and, 1352–1353, 1353f
single incision, 55–56, 55f 612f perihepatitis and, 1351–1352
in ulcerative colitis, 1225–1226, 1226f Leukopenia, in portal hypertension, 1360 inflammatory pseudotumor of, 462
umbilicus as entry site in, 970–971 Levator ani, 1311, 1312f Langerhans cell histiocytosis of, 482
Laparotomy paralysis of, rectal prolapse in, 1316 lymphatic drainage of, 1171
after bladder augmentation or replacement, Levosimendan, for septic shock, 161 megakaryoblastic leukemia of, 482
1483–1484 Leydig cell tumors mesenchymal hamartoma of, 461, 461f, 465, 466f
for bladder injury, 322 ovarian, 541 replacement and tissue engineering of, 33
for cloaca, 1304 Sertoli-, ovarian, 540–541 resection of, 470–471
complications of, 1202 Li-Fraumeni syndrome, 405 rhabdoid tumor of, 480
for intussusception, 1106–1108, 1107f osteogenic sarcoma in, 580, 580f sarcoma of, 480
for necrotizing enterocolitis, 1202 rhabdomyosarcoma in, 492 segmental anatomy of, 646, 646f
for rectobladder neck fistula, 1298, 1300f Lidocaine, 220–221, 221t surface-rendered view of, 69f
Large cell lymphoma Life-saving measures, ethics of, 240–241 teratoma of, 462
anaplastic, 525, 526–527 Life-threatening injuries, treatment of, tissue engineered, 33
diffuse B cell, 525, 526 263–268 trauma to
Large for gestational age, 89, 91f Ligament injury, hand, 339 birth-related, 392, 392f
Large intestine. See Colon. Ligament of Treitz, malrotation and, 1120, 1120f damage-control strategies for, 294–298, 296f,
Laryngeal arteries, 722 Ligamentum arteriosum, left, right aortic arch with, 297f, 297t
Laryngeal cartilages, 722 1665, 1667, 1668f, 1669f imaging of, 290f, 291, 291t
Laryngeal cleft, 850–851, 850f LIL (laparoscopic inversion ligation) technique, treatment of, 291–299
Laryngeal nerve, superior, 722 992–993, 992f guidelines on, 291–292, 292t

Volume 1, pages 1–768; Volume 2, pages 769–1738

 INDEX I-27

Liver disease Lung (Continued) Lymphadenitis

cholestatic. See Cholestasis.
coagulation factor deficiencies in, 174
with intestinal failure, 1138–1139
liver transplantation for, 643–644, 644f
metabolic, liver transplantation for, 645
with parenteral nutrition, 193
peritonitis in, 1233
portal hypertension from, 1356–1357, 1357t
Liver failure, acute, liver transplantation for, 645
Liver transplantation, 643–653
age distribution of, 643, 644f
anatomic considerations in, 646, 646f
for biliary atresia, 1326, 1327, 1329
complications of
infectious, 650–651, 651t
technical, 649
donor operation for, 646–647, 646f
future of, 651–652
for hepatoblastoma, 472–475, 474f, 474t, 475t
for hepatocellular carcinoma, 478–479, 479t
historical perspective on, 605–613, 606f, 606t,
608f, 609f
immunosuppressive therapy for, 649–650, 650t
indications for, 643–644, 644f
with intestinal transplantation, 655, 1145
organ allocation for, 645–646
outcome of, 651–652
pancreatitis after, 1373
PLUTO database for, 475
postoperative care in, 648–649
rejection in, 649–650
segmental, 647, 647f, 648, 648f
transplantation operation in, 647–648, 648f
Liver tumors, 463–487
age at presentation of, 464, 465t
benign, 459–463
clinical presentation in, 459
diagnosis of, 459–460, 460f, 460t
differential diagnosis of, 464, 464t, 465f, 465t
radiologic evaluation of, 465–466, 466f, 467f
malignant, 466–482. See also Hepatoblastoma;
Hepatocellular carcinoma.
clinical presentation in, 463–464
diagnosis of, 463–466
differential diagnosis of, 464, 464t, 465f, 465t
historical perspective on, 463
laboratory evaluation of, 464–465
radiologic evaluation of, 465–466, 466f
metastatic, 435, 465, 466f, 481–482, 482f
as secondary malignancies, 482
syndromes associated with, 468t
transitional, 477
transplantation for, 645
Lobar emphysema, congenital, 825, 828–829, 828f
Lobectomy, temporal, 1691
Local anesthetics, 220–221, 221t
caudal block with, 224–225, 224f
epidural infusion of, 225–226, 225t
infiltration with, 221
neuraxial block with, 224–225, 224f
peripheral nerve and plexus blocks with,
221–222. See also Nerve block.
topical, 221, 221t
Long bone, fracture of, growth stimulation after, 329
Longitudinal intestinal lengthening and tailoring
operation (LLIT), 1141–1142, 1142f
Loperamide
in short bowel syndrome, 1140
in ulcerative colitis, 1222, 1227
Lorazepam, for burns, 382–383
Lordosis, congenital, 1706, 1706f
Loss of heterozygosity (LOH), in Wilms’ tumor,
425–426, 426f
Louw, J., 17, 17f
Lumbar ectopic kidney, 1405
Lumbar spine, injury to, 335, 354, 356t, 358
Lumbosacral spine, dimple over, 1453, 1454f
Lung. See also Pulmonary entries; Respiratory entries.
abscess of, 867–870, 869f
artificial, 125–126
biopsy of, 875–876, 875f, 876f
blastoma of, 569–570, 570f
blood flow in, 114
compliance of, 113, 113f
cystic lesions of, 825–829
diagnosis and treatment of, 825–826, 826f
embryology of, 825
malignancies with, 568–569, 568t, 569t, 570–571
development of, 109–112, 110f, 811–812
alveolar stage of, 111, 111f
arterial growth in, 111–112
canalicular stage of, 109–110
embryonic period of, 109
mediators of, 112
pseudoglandular stage of, 109
terminal saccular stage of, 111
dropped, 277
fibrous histiocytoma of, malignant, 567
gas exchange in, 114–115, 115f
hamartoma of, 567
hemorrhage and hemoptysis in, 867, 867f
infections of, 855–872. See also Pneumonia.
bronchiectasis and, 865–866, 866f
in cystic fibrosis, 864–865
epidemiology of, 855
in immunocompromised patient, 860–862
with cancer, 860–862, 860f, 862f
with HIV/AIDS, 862–864, 863f
after lung transplantation, 680
inflammatory pseudotumor of, 567
physiology of, 112–115
rhabdomyosarcoma of, 569t, 570, 571f
trauma to, 277, 278f
transfusion-related, 177
tumors of
benign, 567, 568t
bronchoscopy of, 570
cystic malformations with, 568–569, 568t,
569t, 570–571
malignant, 567–571, 568t
metastatic
from osteosarcoma, 571, 572t
from soft tissue sarcoma, 503
treatment of, 571–572
from Wilms’ tumor, 572
treatment of, 570
volumes of, 112–113, 113f
in pectus excavatum, 781
Lung buds, 109, 110f
Lung-head ratio (LHR), in congenital diaphragmatic
hernia, 85, 815
Lung transplantation, 671–684
for bronchiolitis obliterans, 673–674
bronchiolitis obliterans after, 674, 679, 680–681,
680f
complications of, 677–680
for congenital diaphragmatic hernia,
823–824
contraindications to, 674–675, 674t
for cystic fibrosis, 671–672, 865
donor evaluation in, 675–676
future of, 681
graft complications in, 679f, 680f
heart transplantation with, 672–673
historical perspective on, 605–613, 606t, 671
immunosuppressive therapy for, 676–677, 676t
indications for, 671–674, 679f
operative procedures in, 676
organ allocation for, 671
organ procurement for, 675–676
preservation solutions in, 676
for pulmonary fibrosis, 673
pulmonary function and growth after, 681
for pulmonary vascular disease, 672–673
retransplantation for failure of, 674
surveillance after, 677
survival rate for, 680–681, 681f
timing of, 671–673
Lupus anticoagulants, 174
Lupus erythematosus, systemic, 1234
Luteinizing hormone–releasing hormone (LHRH),
for cryptorchidism, 1009
Lymph fluid, abdominal, sources of, 1171
Lymph node(s), cervical, 727, 737, 738f
Lymph node dissection
retroperitoneal, radical inguinal orchiectomy and,
554–556, 555f
in rhabdomyosarcoma, 494
acute, 740–741
in cat-scratch disease, 742–743
mycobacterial
atypical (nontuberculous), 741–742
tuberculous, 742
persistent, 741–742
Lymphadenopathy
cervical. See Cervical lymphadenopathy.
differential diagnosis of, 737, 738t
Lymphangiectasia, 1622–1623
Lymphangioma
with chylothorax, 876–877, 876f
cystic, 1165
mediastinal, 835
Lymphatic malformations, 1621–1624,
1622f, 1623f
abdominal cystic, 1133. See also Mesenteric and
omental cysts.
breast, 774
cervicofacial, 1622
oral cavity and pharyngeal, 721
salivary gland, 731–732, 732f, 733f
Lymphatic system, development of, 1620
Lymphaticovenous malformation, capillary-,
1627–1629, 1627f, 1628f
Lymphedema, 1623
Lymphoblastic leukemia, acute, testicular tumors in,
552–553
Lymphoblastic lymphoma, 525, 526
Lymphoceles, after renal transplantation, 624
Lymphocyte-predominant Hodgkin lymphoma, 519,
520f, 521
Lymphocytes, in host defense, 147–148
Lymphoid appendiceal follicles, hyperplasia of,
1256
Lymphoid interstitial pneumonitis, 862–863, 863f
Lymphoid polyps, 486, 1185, 1185f
Lymphoid tissue, as lead point in intussusception,
1095f, 1096–1097
Lymphoma. See also Hodgkin lymphoma;
Non-Hodgkin lymphoma.
anaplastic large cell, 525, 526–527
B-cell, 485
cervical lymphadenopathy in, 743
intestinal, 485
intestinal obstruction with, 1132
oral cavity and pharyngeal, 721–722
Lymphomatosum, papillary cystadenoma, 733
Lymphoproliferative disorders, post-transplant, 525,
526–527
intestinal, 656
lung, 679
renal, 628–629, 650–651
Lymphoscintigraphy, in chylothorax, 878
Lynch syndrome, 488
M
MACE procedure, for constipation, 1315
Macrodactyly, 1723–1724
Macroglossia, 720
Macromastia, 773, 773t
Macrophage inflammatory protein, 149
Macrophages
in host defense, 146
in neonate, 151
Macrostomia, in craniofacial cleft number 7,
696–697, 696f
Mafenide acetate, in burn care, 376–377
Maffucci syndrome, 529, 579
MAG3 (mercapto acetyltriglycine) diuretic
renography, 1431
Magnesium
imbalance of. See Hypermagnesemia;
Hypomagnesemia.
serum, in neonate, 94
Magnesium citrate, in gender assignment surgery,
1575–1576
Magnetic resonance angiography, in portal
hypertension, 1361
Magnetic resonance cholangiopancreatography
in choledochal cyst, 1335, 1335f
in pancreas divisum, 1375–1376, 1376f
in pancreatitis, 1373, 1375

Volume 1, pages 1–768; Volume 2, pages 769–1738

I-28 INDEX
Magnetic resonance imaging, 43–45
of bone tumors, 581
of brain tumors, 592–593
of cerebellar astrocytoma, 594, 595f
of chest wall tumors, 573
in conjoined twins, 1732–1733, 1733f
in constipation, 1314
of craniopharyngioma, 598–599, 599f
in disorders of sex development, 1575, 1577f
in ectopic kidney, 1406, 1408f
in ectopic ureter, 1446, 1446f
of ependymomas, 596, 596f
in epilepsy surgery, 1687–1688, 1688f
functional, 44–45
in gastrointestinal bleeding, 1154
gradient-echo pulse sequence technique in, 44
in hepatic hemangioma, 1618, 1618f
higher field strength, 43–44, 44f
of hypothalamic/chiasmatic astrocytoma,
597–598, 598f
in infantile hemangioma, 1615, 1615f
in intracranial infections, 1695–1696
in intussusception, 1101
in lymphatic malformation, 1623, 1623f
motion artifact reduction techniques in, 44
in musculoskeletal trauma, 331–332
of neck mass, 727
in necrotizing enterocolitis, 1199
in neuroblastoma, 444
of ovarian tumors, 533
parallel, 44
in pheochromocytoma, 559–560
of pontine glioma, 597, 597f
prenatal, 45
in choledochal cyst, 1333
in congenital diaphragmatic hernia, 814, 816
in conjoined twins, 1731, 1732f
in jejunoileal atresia and stenosis, 1061,
1062f
in prenatal diagnosis, 78, 79f
of primitive neuroectodermal tumors, 594, 595f,
596f
of salivary glands, 730
in spine and spinal cord injury, 359
in thoracic trauma, 274
in tracheobronchial vascular compression,
853–854
in traumatic brain injury, 350
ultrafast, 44
in venous malformation, 1624f, 1625
of Wilms’ tumor, 427
Magnetic resonance urography, 1417–1418, 1418f
Magnetic resonance venography, in capillary-
lymphaticovenous malformation,
1627–1628, 1628f
MAGPI (meatal advancement glansplasty), 1540,
1541f
Major histocompatibility complex, in host defense,
147
Malabsorption
after bariatric surgery, 1044–1045, 1045t,
1046–1048
after bladder augmentation or replacement, 1485
in meconium ileus, 1082
in necrotizing enterocolitis, 1196–1197, 1204
in short bowel syndrome, 1071
Male gender assignment surgery, 1582–1583, 1584f
hypospadias repair in, 1582–1583, 1585f
müllerian duct remnants and, 1585, 1587f, 1588f
penile agenesis and, 1585–1586, 1588f, 1589f
penoscrotal transposition and, 1583–1584, 1586f
Malignant hyperthermia, 210–211, 211t
Malignant transformation, 399–400, 400t, 401f
Malnutrition
in burn injury, 381
in meconium ileus, 1082
in neuroblastoma, 443
nutritional support for, 198–199
Malocclusion
in craniofacial cleft number 7, 697, 697f
orthognathic surgery for, 694–695
Malone appendicostomy procedure, 1309, 1309f
Malpositioning, congenital, 1712
Malpuech facial clefting syndrome, 977t
Malrotation. See also Intestinal rotation and fixation,
disorders of; Volvulus.
asymptomatic, 1116
atypical, 1114–1115, 1120, 1120f, 1124
definition of, 1114–1115
with duodenal atresia and stenosis, 1053, 1054
MALT (mucosa-associated lymphoid tissue)
lymphoma, 522–523
MAMLD1 gene, in hypospadias, 1536
Mammary duct ectasia, 773–774
Mandibular defects, in craniofacial cleft number 7,
696–697, 697f
Mandibulofacial dysostosis, 697, 697f
Manganese, requirements for, 184
Manipulation, transabdominal, in intussusception
reduction, 1105
Manitoba oculotrichoanal syndrome, 977t
Mannitol, in trauma patient, 269
Manometry. See Anorectal manometry; Esophageal
manometry.
Marfan syndrome
abdominal aortic aneurysm in, 1635
inguinal hernia in, 1000
pectus excavatum in, 779–780, 780t
Marles syndrome, 977t
Martin procedure, 1272, 1273f
Masculinization
in ovarian tumors, 530
in Sertoli-Leydig cell tumors, 540–541
Mass screening, for neuroblastoma, 442
Mastectomy, for Phyllodes tumors, 776
Mastitis, neonatal, 773
Mastoiditis, 710, 710f
Maternal blood sampling, for fetal disease, 77–78
Mathieu hypospadias repair, 1542–1543, 1543f
Maxillary distraction, 695
Mayer-Rokitansky-Küster-Hauser (MRKH) syndrome,
1587, 1592–1599, 1592f, 1594f, 1595f,
1596f, 1597f. See also Vagina, agenesis of.
McCune-Albright syndrome, 529
McGoon index, 816
McIndoe vaginoplasty procedure, 1594–1595, 1595f
McKusick-Kaufman syndrome, 1592
MCP-1, 149
MDR gene, in neuroblastoma, 449
Mean arterial pressure, in trauma patient, 268–269
Mean corpuscular hemoglobin concentration, in
anemia, 165
Mean corpuscular volume, in anemia, 165
Meatal abnormalities, in hypospadias, 1538–1539
Meatal advancement glansplasty, 1540, 1541f
Mechanical support, as bridge to heart
transplantation, 662
Mechanical ventilation, 117–120
assist-control mode in, 118
breath phases in, 117
complications of, 122
in congenital diaphragmatic hernia, 818
control mode in, 118
cycling mechanisms in, 117
extreme modes of, 119–120
in fluid-refractory shock, 160
high-frequency, 119
intermittent mandatory, 118
inverse ratio, 118–119
investigational adjuncts to, 120
liquid, 119–120
modes of, 118–120
pressure-controlled, 117
pressure support, 118
prolonged, as contraindication to extracorporeal
life support, 124
in respiratory failure, 120–122
synchronized intermittent mandatory, 118
in thoracic trauma, 273
time constants and, 114
transtracheal, for upper airway obstruction, 723
ventilator types in, 117–118
volume-controlled, 118
weaning from, 121
failure of, 121–122
Meckel diverticulum, 964–965, 965f, 1085–1094
bleeding in, 1089–1091, 1151–1152, 1151f, 1152f
clinical findings in, 1088–1092
Volume 1, pages 1–768; Volume 2, pages 769–1738
Meckel diverticulum (Continued)
conditions associated with, 1087–1088
embryology of, 1085, 1086f, 1087f
epidemiology of, 1085
historical perspective on, 1085, 1086f
inflammation in, 1091, 1091f
intestinal obstruction in, 1090–1091, 1090f, 1091f
intussusception in, 1097
neoplasia in, 1091
outcome of, 1092
pathology of, 1086–1087, 1087f
radiologic findings in, 1088–1089, 1089f
treatment of, 1092
Meckel scan, 1088–1089, 1089f, 1151, 1151f
Meconium
albumin concentration in, 1077
composition of, 1074
failure to pass, in jejunoileal atresia and stenosis,
1061, 1061t
Meconium ileus, 1073–1086
clinical features of, 1075–1078, 1075f
colonic obstruction in, 1252, 1252f
complicated, 1075, 1081, 1081f
complications of
gastrointestinal, 1082–1083
inguinoscrotal, 1083
pulmonary, 1083
differential diagnosis of, 1077–1078, 1077f, 1078f
genetics of, 1073–1074
historical perspective on, 1073
laboratory tests in, 1076–1077
management of
nonoperative, 1078–1079
operative, 1079–1081, 1080f, 1081f
postoperative, 1081–1082
results of, 1083, 1083f
pathogenesis of, 1074–1075
pathophysiology of, 1073–1075
radiologic findings in, 1062–1063, 1075–1077,
1076f
uncomplicated, 1075, 1080f, 1081
without cystic fibrosis, 1078
Meconium ileus equivalent, 1082
Meconium peritonitis, 1075, 1081, 1081f
in jejunoileal atresia and stenosis, 1061–1062,
1064f
Meconium plug syndrome, 1078, 1078f
colonic obstruction in, 1250–1251, 1251f
Meconium pseudocyst, 1081, 1081f
Median arcuate ligament syndrome, 1640–1641
Median nerve, injury to, 337–338, 338f
Median sternotomy
mediastinal infection after, 879–880
for ventricular septal defect, 1656, 1656f
Mediastinal infections, 879–880
Mediastinal mass(es)
cystic, 829–830
anatomic considerations in, 830
anterior and superior, 830–832, 831f, 832f
clinical features of, 829–830
diagnosis and treatment of, 830
embryology of, 825
middle, 832
posterior, 832–835, 833f, 834f, 835f
germ cell tumors as, 514–516, 515f
Hodgkin lymphoma as, 518, 518f
incidence of, 829–830, 829t
thoracoscopic biopsy of, 421
Mediastinitis
acute, 879
granulomatous sclerosing, 880
Medical devices, surgeon-developed
innovative, 63
pediatric, 63–65
Medical ethics. See Ethics.
Mediterranean fever, familial, 1234
Medullary thyroid carcinoma, 405, 750
Medulloblastoma, 594–596, 595f, 596f, 601
Megacolon
congenital. See Hirschsprung disease.
functional, after pull-through, 1276
toxic
in Crohn disease, 1213
in ulcerative colitis, 1218

Megacystis-microcolon-intestinal hypoperistalsis
syndrome, 1250, 1285, 1286f
Megakaryoblastic leukemia, hepatic, 482
Megalourethra, 1507, 1560, 1560f
Megameatus hypospadias repair, 1542, 1542f
Megaureter, 1497–1505
in prune-belly syndrome, 1497, 1505–1507, 1510f
repair of
complications of, 1503–1505, 1505f
endoscopic injection in, 1502, 1502f
imbrication in, 1501–1502, 1501f, 1502f
indications for, 1498
lower, 1499–1501, 1500f
peristalsis and, 1503, 1504f
technique of, 1498–1503
upper, 1502–1503, 1503f, 1504f
types of, 1497, 1498f
Meige disease, 1623
Melanocytes, 1711
Melanoma, congenital, 1714
Melena, in portal hypertension, 1358–1359
Melphalan, 407t
for neuroblastoma, 457
Meningioma, 601
Meningitis
bacterial, in basilar skull fracture, 352–353
in otitis media, 710
Meningocele, 1458, 1675–1676.
See also Myelomeningocele.
anterior thoracic, 835
Meperidine, 218, 219t
for pancreatitis, 1374
in patient-controlled analgesia, 220, 220t
6-Mercaptopurine, 407t
for Crohn disease, 1212
in transplantation, 606–607
Mesalamine
for Crohn disease, 1212
for ulcerative colitis, 1222
Mesenchymal hamartoma
chest wall, 574, 574f
hepatic, 461, 461f, 465, 466f
Mesenchymal tumors, mediastinal, 835
Mesenteric and omental cysts, 1165–1171
classification of, 1165–1166, 1169, 1170f
clinical presentation in, 1166–1168, 1167f, 1167t,
1168f
diagnosis of, 1168–1169, 1168f, 1169f
differential diagnosis of, 1166, 1166t
embryology of, 1165
historical perspective on, 1165
incidence of, 1165
intestinal obstruction in, 1133
outcome of, 1169
spectrum of, 1166, 1166t
treatment of, 1167t, 1169, 1170f
Mesenteric artery, superior, stenosis of, 1639–1641,
1640f
Mesenteric hernia, 1132
Mesenteric lymph, 1171
Mesenteric-to-left portal vein bypass (Rex shunt)
for hypersplenism, 1368
for portal hypertension, 1365–1366, 1366f,
1367–1368, 1368f
Mesenteroaxial volvulus, 1037, 1037f
Mesh
in burn care, 379–380, 379f
in inguinal hernia repair, 999
after separation of conjoined twins, 1734, 1734f
Mesial temporal sclerosis, 1692
Mesocaval shunt, 1364
Mesocolic (paraduodenal) hernia
in intestinal malrotation, 1117, 1118f, 1120, 1124
intestinal obstruction in, 1130–1131, 1130f
Mesonephric duct, 1441, 1442f
Mesonephros, 1405, 1411–1412
Mesosalpinx, in sliding hernia sac, 991, 1000
Meta-analysis, 232, 232f
Metabolic acidosis
after bladder augmentation or replacement, 1484,
1496
in necrotizing enterocolitis, 1196–1197
in neonate, 94
with parenteral nutrition, 192
INDEX I-29
Metabolic alkalosis Milk
after gastrocystoplasty, 1484 allergy to, hematemesis from, 1151
in neonate, 94 breast, 187–188
Metabolic bone disease, with parenteral nutrition, 193 fortifiers for, 187–188
Metabolic disorders, after bladder augmentation or for necrotizing enterocolitis prophylaxis, 1205
replacement, 1484 Millard rotation-advancement technique, 702, 702f
Metabolic response to burns, 380–381 Milrinone
Metabolic syndrome, 1042–1043 for congestive heart failure, 135–138, 137t
Metaiodobenzylguanidine (MIBG) radiation therapy, after heart transplantation, 665
for neuroblastoma, 457, 458 Milroy disease, 1623
Metaiodobenzylguanidine (MIBG) scintigraphy Mineralocorticoids, insufficiency of, 564
in neuroblastoma, 444, 444f Minerals
in pheochromocytoma, 560 deficiency of, in short bowel syndrome, 1137
Metal allergy, in pectus excavatum repair, 784, in parenteral nutrition, 184t, 190
789–790, 792 requirements for, 184, 184t
Metanephrine, in pheochromocytoma, 559 supplementation of, after bariatric surgery,
Metanephros, 1405 1044–1045, 1045t
Metaphysis, 327, 328f Minimal access surgery. See also Laparoscopy.
fracture of, 328, 329f, 332 biopsy with, 420
in child abuse, 388 for bone tumors, 584
Metastasis, 402–403 historical perspective on, 9
fine-needle aspiration biopsy of, 418 next-generation, 54–56, 55f
Metatarsus adductus, 1703–1705, 1704f robotic, 57–60. See also Robotic surgery.
Methadone, 218, 219t training in, 74
in patient-controlled analgesia, 220, 220t Minimum alveolar concentration of inhaled
Methicillin-resistant Staphylococcus aureus, 740, 856 anesthetic, 202t, 208
Methimazole, for Graves disease, 747–748 Minute ventilation, 114–115
Methohexital, 202f Mitomycin C, for esophageal caustic injury, 923
Methotrexate, 407t Mitotane, for adrenocortical lesions, 563
for Crohn disease, 1212 Mitotic karyorrhexis index (MKI), in neuroblastoma,
Methylation, DNA, 402 445–447, 446t, 447f
Methylprednisolone Mitral valve prolapse, in pectus excavatum, 784
for immune thrombocytopenic purpura, 170 Mitrofanoff neourethra, 1480, 1481f, 1493–1494,
in spinal cord injury, 359 1493f, 1494f
in transplantation Young-Dees-Leadbetter bladder neck
heart, 667 reconstruction and, 1477
liver, 650t Mitrofanoff principle, 1491, 1493, 1493f
lung, 676–677, 676t Mivacurium, 210t
renal, 625 Mixed venous oxygen saturation, 116
Metoclopramide, for intestinal dysmotility, 1140 MLH1 gene, in hereditary nonpolyposis colon cancer,
Metopic suture, premature fusion of, 692 489
Metronidazole Möbius syndrome, with Poland syndrome, 797
for amebic abscess, 1352 Molecular biology
for Crohn disease, 1212 of cancer, 398–403, 400t, 401t
for peptic ulcer disease, 1033, 1033t of neuroblastoma, 441, 448–449, 448t
for pouchitis, 1228 of Wilms’ tumor, 425–426, 426f, 426t
Metyrosine, for pheochromocytoma, 560 Molecular diagnostics, in cancer, 403–404, 404t
Michelassi strictureplasty, 1213–1214, 1214f Molecular genetics, 19–22
Micro-RNA, 402 changing concepts in, 20–22
Microchimerism, 610, 610f of cystic fibrosis, 20, 20f
Microcolon, in megacystis-microcolon-intestinal of Hirschsprung disease, 20–21, 21t, 1266
hypoperistalsis syndrome, 1285, 1286f pediatric surgical disease and, 19–20, 20f
Microdebrider utility of, 22
for laryngotracheal stenosis, 846 Molecular imaging, 46–48, 48f
for recurrent respiratory papillomatosis, 844 Molybdenum, requirements for, 184
Microelectromechanical systems (MEMS) Monfort abdominoplasty, 1506, 1508f
as actuators, 61 Monobloc frontofacial advancement, 695
for drug delivery, 61–62 Monochorionic twins, anomalies of, fetal
next steps for, 62 interventions for, 87
as sensors, 61, 61f Monocytes
Microform cleft, 699–700, 700f in host defense, 146
Microgastria, congenital, 1039 in neonate, 151
Micrognathia, laryngomalacia with, 840–841 Monogenic disorders, 20, 20f
Microhematuria, in genitourinary trauma, 312–313 Mononucleosis, cervical lymphadenopathy in, 743
Micropumps, 61–62 Montgomery cysts, 776
Microsurgery, for extremity vascular injuries, 364 Moral domain of patient-physician relationship, 238
Microwave energy, 50 Moral problems, resolution of, 237–238
Midazolam Morgagni hydatid, 1014–1015
for burns, 382–383 Morphine, 218, 219t
preoperative, 205 for burns, 382
Midface advancement, 695 caudal, 224–225
Midgut volvulus in patient-controlled analgesia, 220, 220t
acute, 1116, 1116f postoperative ileus and, 1129
chronic, 1116 Mortality, neonatal, 90–91, 91t, 92f
versus duodenal atresia and stenosis, 1054, 1054f Motility disorders. See Esophageal dysmotility;
preoperative management of, 1120 Intestinal dysmotility.
radiography in, 1120, 1121f Motor vehicle injury, prevention of, 258, 258f
recurrent, 1125 Motorcycle injury, prevention of, 259
reduction of, 1122, 1123f Moyamoya disease, 1644
Midline cervical clefts, 760, 760f MSH2 gene, in hereditary nonpolyposis colon cancer,
Midline dorsal plication technique, for chordee 489
repair, 1546–1547, 1550f Mucoepidermoid carcinoma
Mikulicz enterostomy, 1080–1081, 1080f bronchial, 568–569, 569f
Milan criteria for liver transplantation, 478–479 salivary gland, 733
Volume 1, pages 1–768; Volume 2, pages 769–1738
I-30 INDEX
Mucosa-associated lymphoid tissue (MALT)
lymphoma, 522–523
Mucus plugging, after tracheotomy, 839
Müllerian agenesis, 1592–1599, 1592f, 1594f, 1595f,
1596f, 1597f. See also Vagina, agenesis of.
Müllerian anomalies, 1290
Müllerian duct remnants, surgical treatment of,
1585, 1587f, 1588f
Müllerian ducts, 1441–1443, 1565, 1566f, 1567,
1591
Müllerian inhibiting substance, 1567
ovarian tumors and, 531t
Müllerian inhibiting substance (MIS) gene, 1567
Multichannel intraluminal impedance
in motility disorders, 941–942, 943f
pH monitoring combined with, in
gastroesophageal reflux disease, 952, 952f
Multiculturalism
ethics in, 243–244
patient- and family-centered care and, 248
Multifactorial (complex) inheritance disorders, 21–22
Multiple endocrine neoplasia, 405, 751–752
Multiple endocrine neoplasia I, 1383
Multiple endocrine neoplasia II, 561, 750
Multiple subpial transection, 1691
Multivisceral transplantation, 654–655, 655f
Multivitamins, in parenteral nutrition, 189–190, 190t
MURCS association, 1587, 1592
Muscle relaxants, 209–210, 210t
for trauma patient, 265
Musculoskeletal abnormalities, with pectus
excavatum, 779–780, 780t
Musculoskeletal trauma, 327–337
in children versus adults, 327–329, 328f, 329f,
330f, 331f
evaluation of, 329–332, 331f
high-priority, 332–336
child abuse as, 336
compartment syndrome as, 334
femoral neck fracture as, 334–335, 335f
mangled extremities as, 335
open fracture as, 334
spine trauma as, 335, 335f, 336f
management of, 332, 333f
Mutagenesis, in gene transfer, 25
Mutations, 399–400, 400t, 401f
MYCN amplification, 401, 403–404
in neuroblastoma, 445–446, 448, 452–453, 453t
Mycobacterial infection, in neck, 728
Mycobacterial lymphadenitis
atypical (nontuberculous), 741–742
tuberculous, 742
Mycobacterial pneumonia
atypical (nontuberculous), 858, 863–864
tuberculous, 857, 863
Mycophenolate, in transplantation
heart, 665, 667t
liver, 649, 650t
lung, 676–677, 676t
pancreas, 636, 636f
renal, 624–625
Mycoplasma pneumoniae, 857
Mycoses, pulmonary, 864
Myectomy, for near-total intestinal aganglionosis,
1273–1274
Myeloablative therapy, for neuroblastoma, 457
Myelodysplasia, 1469
neuropathic bladder in, 1457–1459, 1458f, 1459f
Myelomeningocele, 1458, 1469, 1673, 1674f
associated anomalies with, 1676, 1677–1678, 1678f
Chiari II malformation with, 1677, 1678f
closure of, 1676
cryptorchidism with, 1005
diagnosis of, 1676
fetal interventions for, 86, 1676–1677
hydrocephalus with, 1676, 1677–1678
incidence of, 1674–1675
nutritional support in, 199, 199t
outcome and prognosis in, 1680
pathology of, 1675–1676
voiding cystourethrography in, 1455f
Myer-Cotton grading of subglottic stenosis,
845, 845f
Myocardial contusion, 281
Myocardial rupture, 281
Myocardial stun, during extracorporeal life support,
127–128
Myofibromatosis, intestinal, 485
Myotomy. See also Esophagomyotomy; Pyloromyotomy.
Heller, 946
for near-total intestinal aganglionosis, 1273–1274
Myringotomy, 709, 710
Myxoma, ovarian, 548
N for multisegmental disease, 1202
N-PASS pain scale, 215
Naloxone, for opioid side effects, 217–218, 217t
Nanoelectromechanical systems (NEMS), 62–63, 62f
Nasal. See also Nose.
Nasal dermoid, 714, 714f
Nasal glioma, 715
Nasal polyp, 713
Nasoalveolar molding, for cleft lip, 701–702, 702f,
703, 703f
Nasogastric feedings, 186, 923
Nasogastric tube
decompression with, for adhesive bowel
obstruction, 1128
in gastrointestinal bleeding, 1147
in thoracic trauma, 272–273
Nasojejunal feedings, 186
Nasopharyngeal angiofibroma, juvenile, 715–716
Nasopharyngoscopy, 837
Nasopharynx
anatomy of, 716
carcinoma of, 716
Natural killer (NK) cell(s)
in host defense, 147–148
in neonate, 151
Natural killer (NK) cell lymphoma, 523t
Natural orifice translumenal endosurgery (NOTES),
56–57
Navigational systems, in image-guided therapy, 51
Neck, 726–728. See also Cervical entries; Head and
neck mass.
clinical evaluation of, 726–727
cysts and sinuses of, 753–763. See also Branchial
anomalies.
dermoid, 760
embryogenesis of, 753–755, 754f, 754t, 755f
thymic, 760–761
thyroglossal duct. See Thyroglossal duct cyst.
inflammatory and infectious masses of, 727–728
lesions of, subcutaneous endoscopy for, 55, 55f
malignant neoplasms in, 728
midline clefts of, 760, 760f
webbing of, 1714
wry. See Torticollis.
Neck space infection, deep, 718, 718f
Necrosis
avascular, femoral head, 1703
coagulation, 370–371, 371f
fat, of scrotum, 1015
Necrotizing enterocolitis, 1187–1216
birth weight and, 1135–1136, 1188–1189, 1203
classification of, 1187, 1188t, 1199
clinical features of, 1195–1196
complications of, 1203–1204
gastrointestinal, 1203–1204
neurodevelopmental, 1204
cytokines in, 1190t, 1191–1193
diagnosis of, 1195–1197
epidemic clusters of, 1197–1198
epidemiology of, 1188–1193
formula feeding and, 1189, 1193–1194, 1206
growth factors in, 1189–1191, 1190t
historical perspective on, 1187
imaging of, 1188f, 1198–1199
incidence of, 1187–1188
indicators of, 1196–1197
indomethacin and, 1189
intestinal strictures in, 1203, 1249, 1250f
laboratory findings in, 1196
management of, 1199–1203
microbiology of, 1197–1198
nonoperative management of, 1199–1200
pathogenesis of, 1193–1195, 1193f
impaired gut barrier in, 1194
infectious agents in, 1194–1195
Volume 1, pages 1–768; Volume 2, pages 769–1738
Necrotizing enterocolitis (Continued)
pathology of, 1195, 1195f, 1196f
prematurity and, 1188–1189, 1193–1194
prevention of, 1204–1207
recurrent, 1204
surgical treatment of, 1201–1203
complications of, 1203
for focal disease, 1202
indications for, 1200–1201
laparotomy for, 1202
for pan involvement, 1202–1203
primary peritoneal drainage for, 1201
stoma closure in, 1203
survival rate for, 1203
unifying hypothesis for, 1195
Necrotizing fasciitis, umbilical, 964
Needle cricothyrotomy, 265
Neisseria gonorrhoeae, perihepatitis and, 1352
Neoadjuvant chemotherapy, 406
Neonatal alloimmune thrombocytopenia,
169–170
Neonatal pain, agitation, and sedation scale
(N-PASS), 215
Neonate
acid-base balance in, 94–95
basal metabolic rate in, 97
body water composition in, 91–92
calcium balance in, 93–94
cardiovascular management in, 135–140
of arrhythmias, 138–139, 139t
of congenital heart disease, 139–140
of congestive heart failure, 135–138, 137t
cardiovascular physiology of, 133–134, 134f
classification of, 89
energy expenditure in
activity-based, 97
resting, 97–98, 98f
surgery and, 103–104, 104f
energy metabolism in, 95–98
intake and, 95–96
losses and, 97
storage and, 96, 96f
enteral nutrition in, 184–188, 185t
extracorporeal life support in, 123, 124, 126, 128,
130, 130t, 131
fluid administration in, 95, 95t
fluid and electrolyte balance in, 91–95
fluid shifts in, 92–93
galactorrhea in, 774, 774f
gastric perforation in, 1038–1039, 1038f
gestational age of. See Gestational age.
glucose metabolism in, 99–102, 101t
surgery and, 105–106
growth of, 89, 97, 179
heart transplantation in, 662–663, 664, 666f
hematemesis in, 1148
hemorrhagic disease in, 1148–1149
host defenses in, 150–152
intussusception in, 1099
juvenile polyposis syndrome in, 1182–1183
larynx of, 722
lipid and fat metabolism in, 102–103
surgery and, 106
liver abscess in, 1350
magnesium balance in, 94
mortality of, 90–91, 91t, 92f
nutrient metabolism in, 99–103
organ failure score for, 90–91, 92t
ovarian cysts in, 536, 536f
pain in, 215, 218
parenteral nutrition in, 188–196, 195f
potassium balance in, 93
protein and amino acid metabolism in, 102–103
surgery and, 107
renal function in, 93
renal transplantation in, 621
renal tumors in, 432
respiratory failure in, mortality risk in, 124
sacrococcygeal teratoma in, 511–512, 512f
sepsis in, 153, 157f, 162–163
sodium balance in, 93
stress response in, 103–107
surgery and, 103–107, 104f, 106f
thermoregulation in, 98–99

Neostigmine, for Ogilvie syndrome, 1250
Neovascularization, tumor-associated, 403
Nephrectomy
partial, with ureterectomy, 1444, 1445f, 1447
renal transplantation and, 618, 619
Nephroblastoma. See Wilms’ tumor.
Nephroblastomatosis, 429, 429f
Nephrogenesis, 1411–1412
Nephrogenic cord, 1405, 1406f
Nephrogenic rests, 429, 429f
diffuse hyperplastic perilobar, 429, 429f
Nephrolithiasis, in ulcerative colitis, 1219
Nephrolithotomy, percutaneous, 1438
Nephroma
congenital mesoblastic, 438–439
cystic, 439–440, 439f, 1401–1402, 1402f
Nephromegaly, in autosomal recessive polycystic
kidney disease, 1397
Nephropathy, chronic allograft, 627
Nephrostomy, percutaneous
for megaureter, 1498
after pyeloplasty, 1425
for urolithiasis, 1437
Nephrotic syndrome, peritonitis in, 1232
Nephroureterectomy, partial, 1444, 1445f, 1447
Nerve block, 221–222
fascia iliaca, 223, 223f
ilioinguinal-iliohypogastric, 222–223, 222f
penile, 223–224, 224f
rectus sheath, 222, 222f
Netherlands, pediatric surgery in, 14–15
Networks for quality improvement and outcomes
research, 235–236, 235f
Neural tube defects, 1673–1680.
See also Myelomeningocele.
classification of, 1673
closed, 1673, 1678–1680
cryptorchidism in, 1005
diagnosis of, 1676
embryogenesis of, 1673–1674, 1674f, 1675f
epidemiology of, 1674–1675
etiology of, 1675
folic acid and, 1673, 1675
incidence of, 1673
pathology of, 1675–1676
prognosis in, 1680
treatment of, 1676–1678
for associated anomalies, 1677–1678
fetal surgery for, 1676–1677
outcome of, 1680
Neuraxial anesthesia, 224–225, 224f
Neurenteric cysts, 835, 1679
foregut, 835
thoracic, 1158
Neurobehavioral outcomes, from traumatic brain
injury, 353
Neuroblastoma, 441–463
cervical, 455–456
cervical lymphadenopathy in, 743
chemotherapy for, 456
clinical presentation in, 442–444, 442f, 443f
cystic, 450f, 451
diagnosis of, 444–445, 444f, 445f
distribution of, 441–442, 442f
epidemiology of, 441
future directions in, 458
genetics and molecular biology of, 441, 448–449,
448t
histopathology of, 445–448, 446t, 447f, 448f,
452–453
immunotherapy for, 411, 457–458
in infancy, 449–450, 454f
with cystic disease, 450f, 451
with stage IV-S disease, 450–451, 455f
mass screening for, 442
metastatic, 443–444
multifocal and bilateral, 452
MYCN amplification in, 403–404, 445–446, 448,
452–453, 453t
myeloablative therapy for, 457
novel therapy for, 458
ploidy in, 400–401, 448
radiation therapy for, 450, 456–457
targeted, 457, 458
risk-based management of, 452–453
INDEX I-31
Neuroblastoma (Continued) Nitrogen mustard, 407t
risk stratification in, 406, 452–453, 453t Nitrogen washout test, 113
staging of, 445, 446t Nitroindazole, for inhalation injury, 375
surgical management of, 451f, 452f, 453–456 Nitrous oxide, 207t, 208
targeted therapy for, 410 Nociception, 214, 215
Neurodevelopmental abnormalities Nocturnal enuresis, bladder dysfunction in,
after congenital diaphragmatic hernia repair, 822 1464–1466
necrotizing enterocolitis and, 1204 NOD2/CARD15 overexpression, in Crohn disease,
Neuroepithelial tumors, dysembryoplastic, 591, 1209
599–600, 599f Noggin, in craniosynostosis, 691–692
Neurofibromatosis, in neck, 728 Non-Hodgkin lymphoma, 522–527
Neurofibromatosis type 1, 405 B-cell, 523, 523t, 524–525, 524f
abdominal aortic coarctation with, 1631 breast, 777
brain tumors with, 597, 601 classification of, 522, 523t
rhabdomyosarcoma with, 492 clinical presentation in, 523–524
Neurofibromatosis type 2, brain tumors with, 601 epidemiology of, 522–523
Neuroimaging gastric, 522–523
in spine and spinal cord injury, 358–359, 358f NK-cell, 523t
in traumatic brain injury, 349–350, 352 ovarian, 547–548
Neurologic evaluation staging of, 523–524, 524t
in trauma patient, 268 subtypes of, 524–525, 524f
in traumatic brain injury, 349–350 T-cell, 485, 523t, 524f, 525
Neurologic injury testicular, 552–553
in birth trauma, 392 treatment and outcomes of, 525–527
with extracorporeal life support, 129 Nonadherence, after renal transplantation, 627
traumatic, 343–364 Nonalcoholic steatohepatitis, bariatric surgery and,
Neurologically impaired children, gastroesophageal 1043
reflux disease in, 956–957 Nonmaleficence principle, 237
Neuroma, acoustic, 601 bariatric surgery and, 242
Neuromuscular blocking agents, 209–210, 210t Nonossifying fibroma
Neuromuscular function, intraoperative monitoring location of, in relation to physis, 579f
of, 213 multiple, 579
Neuronal dysplasia. See Intestinal neuronal Nonsteroidal antiinflammatory drugs (NSAIDs),
dysplasia. 216–217, 216t
Neuronavigational systems, 51 peptic ulcer disease and, 1031
Neuropathic bladder. See Bladder, neuropathic. Norepinephrine, 558
Neuropsychologial testing, epilepsy surgery and, 1689 for fluid-refractory shock, 160
Neuroresuscitation, in trauma patient, 268–269 for septic shock, 161
Neurotrophins, in hypertrophic pyloric stenosis, Norway, pediatric surgery in, 13
1022 Norwood procedure, 1663–1664, 1664f
Neurulation, 1673–1674, 1674f, 1675f Nose, 712–716. See also Nasal entries.
Neutrophils anatomy of, 712
count of, in necrotizing enterocolitis, 1196 congenital malformations of, 713–715, 714f
in host defense, 146 embryology of, 712–713
inflammation of, in Crohn disease, 1210, 1210f foreign bodies in, 715
in neonate, 150–151 fracture of, 715
Nevus(i) inflammatory conditions of, 713
Becker, of breast, 773 trauma to, 715
blue-rubber bleb, 1625, 1625f tumors of, 715–716
giant hairy, 1712–1713, 1714, 1715f NOTCH activation, in hepatoblastoma, 467
melanocytic, congenital, 1714 Notochord, 1711–1712
sebaceous, 1714 split, in alimentary tract duplications, 1156
spider, 1621 Novalis Tx, 52
New Zealand, pediatric surgery in, 15 NP-59 scintigraphy, in adrenocortical lesions, 563
Nicoladoni sign, 1625–1626 NRTN gene, in Hirschsprung disease, 21t
Nipple NSQIP, 235–236, 235f
absence of, 771 Number needed to treat, 233
accessory, 771–772 Numby Stuff, 221, 221t
adenoma of, 774, 777 Nursery, infection control measures in,
congenital anomalies of, 772, 772f 1204–1205
discharge from Nuss procedure, for pectus excavatum, 789–790.
bloody, 773–774 See also Pectus excavatum, minimally invasive
milky, 774, 774f surgery for.
supernumerary, 1716 Nutrient metabolism, in neonate, 99–103
Nipple valve, for continent urine drainage, Nutrition. See also Breast feeding; Diet; Formula(s);
1479–1480, 1481f Malnutrition.
Nitric oxide after bariatric surgery, 1046–1048
for congenital diaphragmatic hernia, 818–819 Nutritional anemia, 167–168
in host defense, 147 Nutritional assessment, 179–180
in hypertrophic pyloric stenosis, 1022 Nutritional requirements, 180–184
inhaled for carbohydrate, 182
as adjunct to mechanical ventilation, 120 for energy, 180–181, 181t
after heart transplantation, 665 for fat, 182–183
for persistent pulmonary hypertension of the for fluids, 181, 181t
newborn, 162–163 for protein, 181–182, 181t
in necrotizing enterocolitis, 1190t, 1192, 1194 for trace elements, 184, 184t
as tocolytic, 80–81 Nutritional status, biochemical measurements of,
Nitric oxide synthase, 147 180
in achalasia, 945 Nutritional support, 179–206
in hypertrophic pyloric stenosis, 21–22 in biliary atresia, 197, 197t
in necrotizing enterocolitis, 1190t, 1192, 1194 in burns, 381–382, 381t
Nitrofurantoin, for urinary tract infection, 1432 in critically ill or septic patient, 197
Nitrogen, body, 181–182 enteral. See Enteral nutrition.
in neonate, 102–103 evolution of, 179
surgery and, 107 in failure to thrive, 198–199
Volume 1, pages 1–768; Volume 2, pages 769–1738
I-32 INDEX

Nutritional support (Continued) Opioids (Continued) Osteotomy(ies) (Continued)

in obesity, 198 postoperative ileus and, 1129 sternal
parenteral. See Parenteral nutrition. side effects of, 217–218, 217t in pectus carinatum repair, 795, 796f
postoperative, 196–197 vicious circle of therapy with, 219, 219f in pectus excavatum repair, 786f
preoperative, 196 Opitz trigonocephaly syndrome, 977t in Poland syndrome repair, 797–798, 801f
in short bowel syndrome, 198, 1137–1138 Opsite, in burn care, 378 Ostium primum atrial septal defect, 1652–1653,
in special care need patient, 199, 199t Optic chiasma, astrocytoma of, 597–598, 598f 1653f, 1657, 1657f, 1658f
Nystatin Optic nerve, injury to, in basilar skull fracture, 353 Ostium secundum atrial septal defect, 1652–1653,
in burn care, 376 Oral cavity 1653f, 1654, 1654f
prophylactic, in heart transplantation, 666–667 anatomy of, 716 Ostling fetal folds, in hydronephrosis, 1411, 1412f
disorders of, 716–722 Otitis media
lesions of acute, 709–710, 709t
O benign, 720–721, 720f, 721f chronic, 710–711
Obesity malignant, 721–722 with effusion, 709
definition of, 1041 trauma to, 717 Otolaryngologic disorders, 707–729
epidemiology of, 1041, 1042–1044 Oral contraceptives, hypospadias and, 1537 Outcomes research, 235
genetics of, 1042 Orbit databases and networks for, 235–236, 235f
health consequences of, 1042–1044, 1043t cellulitis of, 713 Ovarian cysts
nutritional support in, 198 ecchymosis of, in neuroblastoma, 442–443, 442f classification of, 533–534, 534t
risk factors for, 1042–1044 hemangioma of, 1613–1614 clinical presentation in, 530
science of, 1042 rhabdomyosarcoma of, 496 corpus luteum, 536
Obstructive sleep apnea, 203–204, 719, 1043 Orchidopexy endometrioid, 537
Octreotide complications of, 1013, 1013t follicular, 536, 536f
for chylothorax, 878 for cryptorchidism, 1009–1013, 1011f hypothyroidism and, 536, 548
for gastrointestinal bleeding in portal in prune-belly syndrome, 1509 imaging of, 532, 532f
hypertension, 1150 Orchiectomy parovarian, 537
for intestinal dysmotility, 1140 radical inguinal, 554–556, 555f syndromes associated with, 529
for intestinal pseudoobstruction, 1134 for testicular tumors, 509–510 treatment of, 535–537
for variceal hemorrhage, 1362 Orchiopexy, testicular cancer and, 549–550 Ovarian tumors, 529–552
OEIS complex, 977t Orchitis, 1015 classification and staging of, 533–535,
Ogilvie syndrome, 1250 Organ allocation, 645–646, 671 534t, 535t
OHVIRA (obstructed hemivagina with ipsilateral Organ dysfunction criteria, 152, 153t clinical presentation in, 530
renal anomaly), 1602, 1603f Organ failure score, modified, 90–91, 92t diagnosis of, 530–533
OK432, for mesenteric and omental cysts, 1169 Organ preservation, 613–614, 614f epidemiology of, 529–530
OKT3 Organ procurement, 613, 663–664, 675–676 epithelial-stromal
in heart transplant patient, 665–666, 667 Organ transplantation. See Transplantation. laboratory tests in, 530–531, 530t
in lung transplant patient, 676t Organoaxial volvulus, 1037, 1037f of low malignant potential, 538–539, 538f
Olfactory bulb, 712 Organogenesis, 1712 staging of, 534, 535t
Olfactory nerve, injury to, in basilar skull fracture, 353 Orofacial clefting. See Cleft lip and palate. surface, 537–538
Oligogenic disorders, 20–21, 21t Orogastric feedings, supplemental, for congenital frozen section intraoperative diagnosis of, 531
Oligonucleotides, radiolabeled, 46–47 microgastria, 1039 genetics of, 529–530, 531–532
Ollier disease, 529, 579 Oropharynx germ cell, 510–511, 511f, 511t
Omental cysts. See Mesenteric and omental cysts. anatomy of, 716 chemotherapy for, 546–547
Omeprazole, for peptic ulcer disease, 1033, 1033t caustic injury to, 920–921 genetics of, 531–532
Omphalitis, 964 Orthognathic surgery, 694–695, 697 laboratory tests in, 530–531, 530t, 531t
Omphalocele Orthopedic congenital anomalies, 1699–1712 mixed, 546
antenatal considerations in, 977–978 Orthopedics, presurgical, for cleft lip and palate, staging of, 534, 535t
associated conditions with, 979, 979t 701–702, 702f, 703, 703f surgical guidelines for, 546
cloacal exstrophy with, 1526, 1527–1528, 1528f Ortolani maneuver, 1700 treatment of, 541–544, 542f
complications of, 983 Osler-Weber-Rendu syndrome, 487, 1621 imaging of, 532–533, 532f
cryptorchidism with, 1004–1005 Osmotherapy, for trauma patient, 269 immunohistochemistry of, 531
embryogenesis of, 975–976 Ossicles, 707 incidence of, 529
forms of, 973, 974f, 974t Osteoblastoma, 579f laboratory tests in, 530–531, 530t, 531t
genetics and familial occurrence of, 977, 977t Osteochondrodystrophy, Jeune syndrome as, laparoscopy of, 535
giant, 980–981 805–807, 807f, 808f miscellaneous, 547
historical perspective on, 973 Osteochondroma neoplastic
incidence of, 979 chest wall, 574 classification of, 533–534, 534t
Meckel diverticulum with, 1087–1088 location of, in relation to physis, 579f treatment of, 537–546
obstetric delivery with, 978–979 multiple, 579 nonneoplastic
outcome of, 983–984 Osteodystrophy, renal, in transplant patient, 629 classification of, 533–534, 534t
treatment of, 979–981, 980f, 981f Osteogenesis, distraction, 695, 1712 treatment of, 535–537, 536f
umbilical hernia versus, 974–975 Osteogenic sarcoma (osteosarcoma) pelvic washings for, 534–535
Omphalomesenteric duct, 962f, 963t chest wall, 576 in Peutz-Jeghers syndrome, 1184
failed involution of, 1085 epidemiology of, 580 risk factors for, 529
Omphalomesenteric remnants, 964–966, 965f, 966f, fracture through, 582 secondary (metastatic), 547–548, 548t
1085, 1087f, 1091, 1131 genetics of, 580, 580f sex cord-stromal, 530–531, 530t, 531t, 539–541,
Omphalomesenteric vein, 1355, 1356f location of, in relation to physis, 579f 539f
Omphalomesenteric vessels, 962f, 963t pulmonary metastasis of, 571, 572t syndromes associated with, 529
Oncology, nanoelectromechanical systems in, 62–63, resection and reconstruction of, 586f, 588f, 589f, treatment of, 535–546
62f 590f unclassified, 548
Oncoretroviral vectors, for gene transfer, 24, 24t Osteoid osteoma Ovary(ies)
Onion skinning, 581 location of, in relation to physis, 579f differentiation of, 1567
Onlay island flap, in hypospadias repair, 1544, 1545f radiofrequency ablation of, 584 in sliding hernia sac, 991, 998
Open fracture, 334, 338 Osteomalacia, with parenteral nutrition, 193 Overdrive pacing, for supraventricular tachycardia,
Open incisional biopsy, 422 Osteopenia 138
Opioids, 217–220 with parenteral nutrition, 193 Overfeeding, from parenteral nutrition, 194
continuous infusion of, 219 in ulcerative colitis, 1219 Overweight, 198, 1041. See also Obesity.
epidural infusion of, 225, 225t Osteosarcoma. See Osteogenic sarcoma (osteosarcoma). Ovotesticular DSD, 1568t, 1571, 1574, 1575
intravenous, 218–219, 219t Osteotomy(ies) Oxalosis, after renal transplantation, 628
in neonate, 218 for craniosynostosis, 694 Oxandrolone, in burn injury, 381
operative stress response and, 103–104 femoral shortening, 1703, 1703f Oximetry, pulse, 116, 213
oral, 218, 218t, 219t in orthognathic surgery, 695 Oxybutynin
for pancreatitis, 1374 pelvic, in bladder exstrophy repair, 1519–1521, for fecal incontinence, 1315
in patient-controlled analgesia, 219–220, 220t 1521f for overactive bladder syndrome, 1464

Volume 1, pages 1–768; Volume 2, pages 769–1738

 INDEX I-33

Oxycodone, 218, 218t Pancreas (Continued) Parathyroidectomy, 751–752, 752f

Oxygen immunosuppressive therapy for, 634–635 Parental presence during induction of anesthesia
diffusion of, 114 after kidney transplant, 634, 634f (PPIA), 250–251
partial pressure of kidney transplant with, 632, 632f, 634, 634f Parenteral nutrition, 188–196
arterial, 117 outcomes of, 635–637, 635f, 636f additives to, 191–192, 191t
mechanical ventilation and, 121 segmental, 633–634, 634f administration of, 194–196, 195f
transcutaneous monitoring of, 116 surgical techniques of, 632–634, 632f, 633f, amino acids in, 103, 189
toxicity of, with mechanical ventilation, 122 634f carbohydrate in, 100–101, 105–106, 189
transport of, 115, 115f trauma to, 302–305 cholelithiasis with, 1341
Oxygen free radicals, 146 computed tomography in, 303, 304t, 305f, 305t for chylothorax, 878
Oxygen index pancreatic pseudocysts in, 1377, 1378f for chylous ascites, 1174–1175
in congenital diaphragmatic hernia, 816, 821 pancreatitis in, 1373 complications of, 192–194
extracorporeal life support and, 124 treatment of, 304, 306f hepatobiliary, 193
Oxygen saturation Pancreas divisum, 1371, 1375–1376, 1376f, 1377f infectious, 193–194
arterial, 116 Pancreatectomy metabolic, 192–193
mixed venous, 116 distal, 303, 304, 306f from overfeeding, 194
Oxygen therapy, for inhalation injury, 375 for hyperinsulinism, 1380–1381 technical, 194
Oxygenation for pancreatitis, 1375 composition of, 189–191
extracorporeal membrane. See Extracorporeal life pylorus-sparing total, islet autotransplantations fat (lipids) in, 106, 189
support. after, 638, 638f fluids and electrolytes in, 190–191, 190t
intravascular, 119 Pancreatic duct indications for, 188
Oxyhemoglobin dissociation curve, 115, 115f cystic dilatation of. See Choledochal cyst. for intestinal failure, 653
development of, 1332 after jejunoileal atresia and stenosis repair,
main, 1371 1070–1071
P Pancreatic enzymes, supplementation of, for for meconium ileus, 1081–1082
P value, 232 meconium ileus, 1081–1082 monitoring during, 196, 196t
Pacing, overdrive, for supraventricular tachycardia, Pancreatic insufficiency, in Shwachman-Diamond requirements in, 189–191
138 syndrome, 1373 for short bowel syndrome, 1138
Paclitaxel, 407t Pancreaticobiliary maljunction, 1331, 1333, 1339 trace elements in, 184t, 190
PAGOD syndrome, 977t pancreatitis in, 1373, 1374 venous access for, 188–189
Pain Pancreaticoduodenectomy, for adenocarcinoma, vitamins in, 189–190, 190t
abdominal. See Abdominal pain. 1384 Parinaud syndrome, 600
anorectal, in proctalgia fugax, 1320 Pancreaticojejunostomy Parkes Weber syndrome, 1629, 1629f
assessment of, 215 for pancreas divisum, 1376 Parotid ducts, 716
back, in spinal epidural abscess, 1697 for pancreatitis, 1375 Parotid gland
chronic, after inguinal hernia repair, 999 Pancreatitis, 1372–1376 acinic cell carcinoma of, 733, 733f
hypersensitization to, 215 acute, 1372–1374, 1372t anatomy and physiology of, 729
perception of, 201, 214–215 in cholelithiasis, 1342 hemangioma of, 732
undertreatment of, 214 chronic relapsing, 1374–1375, 1374t, 1375f inflammation of, 731, 731f
visceral, in appendicitis, 1256 in enteric duplication cysts, 1377 lymphatic malformations of, 731–732, 732f, 733f
Pain management, 214–220 in pancreas divisum, 1376 papillary cystadenoma lymphomatosum of, 733
for burns, 382, 382t Pancreatoblastoma, 481–482, 482f, 1384 rhabdomyosarcoma of, 733, 733f
for chest tube insertion, 875 Pancuronium, 210t Parotidectomy, 734–735, 734f
developmental considerations in, 214–215 Papillary cancer, thyroid, 749 Parovarian cysts, 537
in hypospadias repair, 1551 Papillary cystadenoma lymphomatosum, 733 Partnership, in patient- and family-centered care,
in inguinal hernia repair, 988–989 Papillary-cystic epithelial neoplasm, pancreatic, 251–252
nonopioid analgesics for, 215–217, 216t 1382–1383 Paraurethral cysts, 1608
opioid analgesics for, 217–220. See also Opioids. Papillary dermis, 370 Passavant ridge, 716
for pancreatitis, 1374 Papillary renal cell carcinoma, 438 Patella, dislocation of, congenital, 1706
patient-controlled, 219–220, 220t Papilloma Pathways, 234
perioperative planning and general approach to, choroid plexus, 600–601, 601f, 1680–1681 Patient- and family-centered care, 247–254
214, 215f intraductal, 774 background on, 247–248
preemptive, 215 laryngeal, 843–844, 843f best practices related to, 252, 252t
regional anesthesia for, 220–226. See also Regional squamous, of oral cavity, 721 collaboration in, 251–252
anesthesia. Papillomatosis commitment to, 252, 252t
in trauma patient, 270 juvenile, 777 communication in, 248–250
Palate recurrent respiratory, 726, 726f, 843–844, 843f core concepts in, 248–252
anatomy of, 716 Paracentesis definition of, 247
cleft. See Cleft lip and palate. abdominal, in chylous ascites, 1174 participation in, 250–251
Palatoplasty, 703–704, 704f, 705f in necrotizing enterocolitis, 1200 respect and dignity in, 248
Palliative radiation therapy, 413–414 Paraduodenal hernia versus traditional care, 248t
Pancreas, 1371–1387 intestinal malrotation in, 1117, 1118f, 1120, 1124 Patient-controlled analgesia, 219–220, 220t
anatomy of, surgical, 1372 intestinal obstruction in, 1130–1131, 1130f Patient-physician relationship, 238
annular, 1051, 1053, 1053f, 1054, 1056, 1371 Parameningeal rhabdomyosarcoma, 496 Pavlik harness
pancreatitis in, 1374, 1375f Paranasal sinuses, 712–713 for developmental dysplasia of hip, 1702
carcinoma of, 1383–1384 Parapharyngeal space infection, 718 for knee dislocation, 1706
congenital anomalies of, 1371, 1372t Paraplegia, after aortic injury repair, 284 PAX7-FKHR fusion, 400–401
pancreatitis in, 1374, 1375f Parasite, sacral, 1737 in rhabdomyosarcoma, 491–492
cysts and cystic neoplasms of, 1376–1378, Parasitic infection Pectoral muscles, anomalies of. See Poland syndrome.
1382–1383 in Meckel diverticulum, 1092 Pectus carinatum, 793–796
duplications of, 1377 pneumonia as, 859, 859f clinical presentation in, 794, 794f, 795f
embryology of, 1371–1372 Parasitic twins, 1737–1738, 1737t, 1738f etiology of, 793–794
hormonally active tumors of, 1383 Paraspinal rhabdomyosarcoma, 497 with pectus excavatum, 780
hyperinsulinism and, 1379–1382, 1380f, 1381f Parastomal hernia, 1132 treatment of, 794–795
neoplasms of, 1382–1384 Paratesticular rhabdomyosarcoma, 498 with bracing, 795
pseudocysts of, 303, 304, 306f, 1377–1378, 1378f Parathyroid glands minimally invasive surgery for, 796
short, congenital, 1371–1372 adenoma of, 751–752 open surgical repair for, 795–796, 796f
transplantation of, 631–637. See also Islet carcinoma of, 752 Pectus excavatum, 779–784
transplantation. disorders of, 751–752, 751t body image in, 784
categories of, 634, 634f embryology and physiology of, 750–751, 751f, 755f cardiovascular function in, 783–784
duodenal transplant with, 631–632, 633f, 634f Parathyroid hormone clinical presentation in, 780–781, 780f
general information on, 634–635 deficiency of, 749 echocardiography in, 784
historical perspective on, 605–613, 606t, elevation in, 745, 751–752, 751t epidemiology of, 779
631–632, 632f secretion of, 751 etiology of, 779–780, 780t

Volume 1, pages 1–768; Volume 2, pages 769–1738

I-34 INDEX
Pectus excavatum (Continued)
minimally invasive surgery for, 789–790
complications of, 790–792
postoperative management of, 790
preoperative considerations in, 789–790, 790f
pulmonary function and, 782, 783
results of, 792–793
technique of, 790, 791f, 792f, 793f
timing of bar removal after, 790, 793f
open surgical repair of
complications of, 785–789, 789f
history of, 784–789
technique for, 785, 786f
preoperative evaluation in, 784
pulmonary function in, 781–783
Pectus support bar. See Steel pectus support bar.
Pedestrian injury, 259
Pediatric Advanced Life Support, sepsis management
guidelines of, 154–162
Pediatric device consortia, 63–65
Pediatric End-Stage Liver Disease (PELD) score,
645–646, 1362
Pediatric surgery. See also Surgical entries.
history of, 1–20. See also History of pediatric
surgery.
stress response to, 103–107, 104f, 106f
Pediatric Ulcerative Colitis Activity Index (PUCAI),
1221, 1221t
Pellet rifle injuries, 348
Pelvic inflammatory disease, perihepatitis and, 1351
Pelvic osteotomy, in bladder exstrophy repair,
1519–1521, 1521f
Pelvic washings, for ovarian tumors, 534–535
Pelvis
defects of, in bladder exstrophy, 1517
fracture of
bladder trauma with, 321, 321f
genitourinary trauma with, 312
rhabdomyosarcoma of, 497
PELVIS syndrome, 1614–1615
Penile disassembly technique, in epispadias repair,
1521–1522, 1523f
Penile nerve block, 223–224, 224f
Penis
agenesis of, 1562, 1585–1586, 1588f, 1589f
amputation of, 324, 324f
curvature of. See Chordee.
development of, 1531–1532, 1533f, 1538f
duplication of, 1562–1563, 1562f
injury to, 324
size of, 1572–1573
torsion of, 1563
Penoscrotal transposition, 1563, 1583–1584,
1586f
Penrose drains, in necrotizing enterocolitis, 1202
Pentalogy of Cantrell, 973, 974t, 975–976, 981, 983
Pentamidine, prophylactic, in heart transplantation,
666–667
Pentoxifylline, for septic shock, 161
Pepsin, secretion of, 1030
Peptic ulcer disease, 1029
bleeding in, 1032, 1033–1034
classification of, 1030t
clinical findings in, 1031–1032, 1031t
diagnosis of, 1032
drug-induced, 1031
epidemiology of, 1029–1035
historical perspective on, 1029
pathophysiology of, 1030–1034, 1030t
primary, 1029
stress ulcers in, 1029–1030, 1030t, 1031, 1032,
1034–1035
treatment of, 1033–1035, 1033t
in Zollinger-Ellison syndrome, 1034
Perfluorocarbons, for liquid ventilation, 120, 823,
823f
Performance analysis, 234–236, 235f
Performance improvement, 235
Perfusion pressure, threshold, by age group, 159,
159t
Pericardial tamponade, 276, 282, 282f
Pericardiocentesis
for pericardial tamponade, 282, 282f
in thoracic trauma, 274
Pericardium Pharyngotonsillitis
cysts of, 832 acute, 716–717
trauma to, 280–282, 282f chronic, 717
Perihepatitis, 1351–1352 localized extension of, 717–718
Perimeatal skin flap, in hypospadias repair, recurrent, 717
1542–1543, 1543f Pharynx, 716–722
Perineal fistula, 1293, 1293f, 1294–1295 anatomy of, 716
Perinephric abscess, after renal trauma, 318 lesions of
Perineum benign, 720–721, 720f, 721f
injury to, in child abuse, 390f, 391, 391f malignant, 721–722
inspection of, in anorectal malformations, 1292, Phenylalanine mustard, for neuroblastoma, 457
1292f, 1293, 1295f Phenytoin, orofacial clefting and, 699
rhabdomyosarcoma of, 497 Pheochromocytoma, 558–561
trauma to, 308 adrenalectomy for, 564–565, 565f
Periosteum in children versus adults, 558–559, 559t
anatomy of, 327 diagnosis of, 559–560, 559f
fracture and, 327, 329f, 332 disorders associated with, 560–561
injury to, 327, 329f malignant, 561
Peripheral nerve and plexus blocks, 221–222 symptoms of, 559
Peripheral nerve injury, hand, 337–338, 338f, 339 treatment of, 560
Peripheral veins, for parenteral nutrition, 188–189 Phimosis, 1561
Peristalsis Phosphate, in parenteral nutrition, 190t, 191
disorders of. See Esophageal dysmotility; Intestinal Phosphodiesterase inhibitors
dysmotility. after heart transplantation, 665
lower megaureter repair and, 1503, 1504f for septic shock, 161
Peritoneal bands. See Ladd bands. Photodynamic therapy, 49
Peritoneal dialysis PHOX2B gene, in neuroblastoma, 441
inguinal hernia and, 999–1000 Phrenic nerve injury
peritonitis with, 1232–1233, 1233t in birth trauma, 392
renal transplantation and, 619 after lung transplantation, 678
Peritoneal drainage, primary, for necrotizing Phyllodes tumors, 775–776
enterocolitis, 1201 Physeal-sparing procedures, for bone tumors, 585,
Peritoneal lavage, diagnostic, in abdominal trauma, 586f
291 Physician-patient relationship, 238
Peritoneovenous shunting, for hepatocellular ascites, Physiologic dead space, 114–115
1173 Physis, 327, 328f
Peritoneum bone tumor location in relation to, 579, 579f
fluid accumulation in. See Ascites; Intraperitoneal fracture of, 327–328
fluid. growth disturbances after, 328, 329, 331f
free air in, in necrotizing enterocolitis, 1198 imaging of, 331–332, 331f
Peritonitis Salter-Harris classification of, 328, 329f
in appendicitis, 1257 Piercing, umbilical, 967–968
ascites and, 1171 Pili, bacterial, 150
conditions associated with, 1231, 1232t Pilocytic astrocytoma, stereotactic radiosurgery for, 53
in familial Mediterranean fever, 1234 Pineal region tumors, 593–594, 600
in healthy children, 1231–1232 Piriform sinus tracts, 759, 759f
in inguinal hernia, 995, 996 PIRO system, 154
in liver disease, 1233 Pit viper bites, 340
meconium, 1075, 1081, 1081f Pituitary tumors, Cushing disease in, 561
in jejunoileal atresia and stenosis, 1061–1062, PKD gene, in polycystic kidney disease, 1396
1064f PKHD1 gene, in polycystic kidney disease, 1397
microorganisms associated with, 1231, 1232t PLA1 antigen, in neonatal alloimmune
in nephrotic syndrome, 1232 thrombocytopenia, 169–170
with peritoneal dialysis, 1232–1233, 1233t Placental-umbilical circulation, 134–135, 136f
primary, 1231–1235 Placental vascular anomalies, jejunoileal atresia and
sterile, 1233–1234 stenosis with, 1060
in systemic lupus erythematosus, 1234 Plagiocephaly
with ventriculoperitoneal shunts, 1233 positional, 693
Perlman syndrome, Wilms’ tumor in, 427 in torticollis, 766, 766f
Permissive hypercapnia, in congenital diaphragmatic Plain radiography. See Radiography.
hernia, 818 Plant alkaloids, 406, 407t
Peutz-Jeghers syndrome, 487 Plasma. See Fresh frozen plasma (FFP).
ovarian tumors in, 529 Plasma cell granuloma, pulmonary, 567
sex cord tumors with annular tubules in, 541 Plasma substitutes, 176
Peyer patches, as lead point in intussusception, Plastibell clamp, 1561
1095f, 1096–1097 Platelet-activating factor, in necrotizing enterocolitis,
Pfannenstiel approach to incarcerated hernia, 1190t, 1192–1193
996–997 Platelet-activating factor antagonists or degrading
PFAPA syndrome, 717 enzymes, for necrotizing enterocolitis, 1207
Pfeiffer syndrome, 693 Platelet-endothelial cell adhesion molecule-1, in
pH, in neonate, 94 neutrophil diapedesis, 146
pH monitoring Platelets
24-hour, in gastroesophageal reflux disease, 952 count of
combined with multiple intraluminal impedance, during extracorporeal life support, 128
in gastroesophageal reflux disease, 952, 952f in necrotizing enterocolitis, 1196
esophageal, 881–882 normal, 177
in motility disorders, 941 functional disorders of, 170–171
PHACES syndrome, 849, 1614–1615 spleen as reservoir for, 1386
Phagocytosis, neutrophil, 146 transfusion of, 177–178
Phalloplasty, 1585, 1589f for immune thrombocytopenic purpura, 170
PhanTOM interfaces, 71, 71f intraoperative, 206
Pharyngeal (branchial) apparatus, 753–755, 754f, for thrombocytopenia, 169–170, 178
754t, 755f Pleomorphic adenoma, 721
Pharyngeal cyst, 758 parotid gland, 732–733, 733f
Volume 1, pages 1–768; Volume 2, pages 769–1738

Plethysmography, total-body, 113
Pleural effusion, fetal lymphangiectasia as,
1622–1623
Pleuroperitoneal membrane, formation of, 811
Pleuroperitoneal shunt, for chylothorax, 879
Plexus block, 221–222
Ploidy, 400
in neuroblastoma, 400–401, 448
screening for, 77
Wilms’ tumor and, 425–426
PNET (primitive neural ectodermal tumor), 575,
575f, 594–596, 595f, 596f
Pneumatocele, 867, 868f
post-traumatic, 277
Pneumatosis intestinalis, in necrotizing enterocolitis,
1187, 1188f, 1195, 1196f, 1198
Pneumococcal pneumonia, 855–856, 856f
Pneumocystis, in renal transplant patient, 651t
Pneumocystis jirovecii pneumonia
in cancer patient, 861–862, 862f
in HIV-infected patient, 862
Pneumomediastinum
in airway trauma, 279
in esophageal perforation or rupture, 889–890,
891f
Pneumonectomy, prior, lung transplantation and,
675
Pneumonia. See also Lung, infections of.
community-acquired bacterial, 855–858, 856f
complications of, 867–872, 868f, 869f, 871f
after congenital diaphragmatic hernia repair, 822
epidemiology of, 855
in immunocompromised patient, 860–862
with cancer, 860–862, 860f, 862f
with HIV/AIDS, 862–864, 863f
parasitic, 859, 859f
tuberculous, 857
viral, 858–859, 858f
Pneumonitis
lymphoid interstitial, 862–863, 863f
in tracheoesophageal fistula, preoperative
treatment of, 899
Pneumoperitoneum
diagnostic, in inguinal hernia, 994
in necrotizing enterocolitis, 1198, 1200
Pneumothorax
chest tube for, 275, 276f, 872–873
care and removal of, 874
in neonate, 873–874, 874f
in older child, 875
in esophageal perforation or rupture, 889–890,
890f, 891f
open, 275
quantification of, 872
simple, 275, 276f
spontaneous, 872–873
tension, 276, 276f
in thoracic trauma, 275–277, 276f
after tracheotomy, 839
Poisoning, unintentional, prevention of, 259–260
Poland, pediatric surgery in, 15
Poland syndrome, 796–799, 1719–1720, 1721f
clinical presentation in, 797, 798f, 799f, 799t
complications and outcome of, 799
embryology of, 797
surgical management of, 797–799, 801f
treatment of, 797
Polyalveolosis, in congenital lobar emphysema,
828
Polycystic kidney disease, 1396
autosomal dominant (adult), 1396–1397, 1397f
autosomal recessive (infantile), 1397–1398,
1398f
Polydactyly, 1720, 1723
Polyembryoma, 543
Polygenic disorders, 21–22
Polyhydramnios
in congenital diaphragmatic hernia, 813–814, 815
in cystic lung lesions, 825–826
in jejunoileal atresia and stenosis, 1061, 1061t
in meconium ileus, 1075
pyloric atresia and, 1035
Polymastia, 771–772, 772f, 1716
Polymerase chain reaction, 404t
Polyp(s)
gallbladder, 1343
gastric, 1151, 1151f, 1181
gastrointestinal, 486–487, 1177–1190
juvenile, 1152, 1152f, 1177–1179
isolated, 1178–1179, 1178f
lymphoid, 1185, 1185f
nonepithelial, 1185
inflammatory, intestinal obstruction with,
1132–1133
juvenile
bleeding, 1152, 1152f
diffuse, 487
intestinal obstruction with, 1132
isolated, 1178–1179, 1178f
in polyposis syndromes, 486–487, 1177,
1182–1183, 1183f
lymphoid, 486, 1185, 1185f
nasal, 713
rectal, 1152, 1152f, 1180, 1183
small intestinal, 1097, 1097f, 1180
urethral, 1558, 1559, 1559f
Polypoid excrescence, bladder exstrophy with, 1516,
1517f
Polyposis syndromes. See also Polyp(s).
adenomatous, 1179. See also Familial
adenomatous polyposis.
classification of, 1177, 1178t
hamartomatous, 1182–1185
in Cowden syndrome, 1184–1185
in juvenile polyposis syndrome, 1177,
1182–1183, 1183f
in Peutz-Jeghers syndrome, 1183–1184, 1184f
Polysplenia, 1386–1387
Polythelia, 771–772, 1716
intra-areolar, 772, 772f
Pons, glioma of, 593, 597, 597f
Ponseti procedure, 1705
Popliteal pterygium syndrome, 977t
Popliteal vascular injuries, 364
Porcine islet cells, for transplantation, 640
Port-site recurrence, in laparoscopy, 420
Port-wine stain, 1620–1621, 1621f, 1712–1713
Portacaval shunt
end-to-side, 1364
for portal hypertension–related bleeding, 1364
side -to-side, 1364
Portal hypertension, 1355–1374
abdominal distention in, 1360
from arteriovenous fistulas, 1358
ascites in, 1359
bleeding in
gastrointestinal, 1358–1360. See also Varices.
from nongut sites, 1359
causes of, 1356–1358, 1357t
hepatocellular, 1356–1357, 1357t
vascular, 1357–1358, 1357f
clinical presentation in, 1358–1360
collateral circulation in, 1356
definition of, 1356
diagnosis of, 1360–1362, 1361f
embryology of, 1355–1356, 1356f
encephalopathy in, 1359, 1360
endoscopy in, 1361
gastropathy in, 1359, 1368
hemorrhoids in, 1319
hepatomegaly in, 1360
historical perspective on, 1355
hypersplenism in, 1359, 1365, 1368
imaging of, 1361, 1361f
jaundice in, 1360
laboratory tests in, 1360–1361
liver biopsy in, 1361–1362
after portoenterostomy, 1329
posthepatic, 1357–1358
prehepatic, 1357, 1357f
pulmonary disorders in, 1360
Rex shunt for, 1365–1366, 1366f, 1367–1368,
1368f
splenomegaly in, 1360
treatment of, 1362–1366, 1365f, 1366f
complications of, 1366–1367, 1367f
historical perspective on, 1355
nonshunt operations for, 1366
Volume 1, pages 1–768; Volume 2, pages 769–1738
INDEX I-35
Portal hypertension (Continued)
options for, 1368, 1369f
outcome of, 1367–1368, 1368f
shunts for, 1364–1365. See also Varices, shunts
for.
varices in. See Varices.
Portal vein
anatomy of, 1356
development of, 1355–1356, 1356f
embolization of, for hepatocellular carcinoma, 480
gas in, in necrotizing enterocolitis, 1188f, 1198,
1200–1201
thrombosis of
in liver transplantation, 649
portal hypertension from, 1357, 1357f, 1359,
1361
Rex shunt for, 1365–1366, 1366f, 1367–1368,
1368f
Portal vein bypass, mesenteric-to-left, 1365–1366,
1366f, 1367–1368, 1368f
Portoenterostomy
for biliary atresia, 644, 644f
complications of, 1328–1329
controversies related to, 1326
historical perspective on, 1321
laparoscopic, 1326
outcomes of, 1327–1328
postoperative care in, 1327, 1327t
technique of, 1324–1326, 1325f, 1326f
Portosystemic shunts
history of, 1355
transjugular intrahepatic, 1363–1364
Positive airway pressure, continuous, 118
Positive end-expiratory pressure (PEEP), 118
Positive inotropic effect, 134
Positive-pressure ventilation, intermittent, 117
Positron emission tomography, 45–46
in chest wall tumors, 573
with computed tomography, 46, 47f
in hyperinsulinism, 1380, 1380f
in epilepsy surgery, 1689
fluoro-2-deoxyglucose in, 45
molecular imaging using, 48
in ovarian tumors, 533
in pheochromocytoma, 560
in rhabdomyosarcoma, 492–493
of Wilms’ tumor, 427
Post-term infant, 89
POST-TEXTstaging of hepatoblastoma, 465–466, 467f
Post-transplant lymphoproliferative disorders.
See Lymphoproliferative disorders, post-
transplant.
Postbiotics, for necrotizing enterocolitis, 1206
Postconcussion syndrome, 353
Posterior fat pad sign, 331–332, 331f
Posterior fossa tumors, 765
Posterior urethral valves. See Urethral valves,
posterior.
Postural compensation, in torticollis, 765f, 766
Postvoid residuals (PVRs), in myelodysplasia, 1459
Potassium
in aldosterone regulation, 558
imbalance of. See Hyperkalemia; Hypokalemia.
in parenteral nutrition, 190, 190t
serum, in neonate, 93
Pott puffy tumor, 1694–1695, 1694f
Pouch
gastroileal, 1494–1495
ileoanal. See Ileoanal pouch procedure.
Indiana, 1473, 1495, 1495f
J
for long-segment Hirschsprung disease, 1272
for ulcerative colitis, 1224, 1225, 1226–1227,
1227f
Kock, 1472, 1473, 1473f, 1475f, 1494
tracheal, 852
Pouchitis, after ileoanal pouch procedure,
1227–1228
Prader-Willi syndrome, nutritional support in, 199t
Preacinar region, 111–112
Prealbumin, serum, nutritional status and, 180
Preauricular pits, with branchial anomalies, 757
Prebiotics, for necrotizing enterocolitis, 1206
Prednisolone, for infantile hemangioma, 1616
I-36 INDEX
Prednisone. See also Corticosteroids.
for Crohn disease, 1211–1212
for immune thrombocytopenic purpura, 170
for infantile hemangioma, 1616
in transplant patient, 606–607
heart, 665
liver, 650t
lung, 676–677, 676t
renal, 625
Pregnancy
after portoenterostomy, 1327–1328
termination of, defects managed by, 78t
Wilms’ tumor and, 436
Prehospital care of trauma patient, 263
Preload, in neonate, 133–134, 134f
Preload agents, for congestive heart failure, 135, 137t
Premature infant
birth weight subgroups for, 89
body water composition in, 92
characteristics of, 89
cryptorchidism in, 1006
definition of, 89
enteral nutrition in, 184–188, 185t, 187t
fluid and electrolyte balance in, 205
growth of, 179
hepatoblastoma in, 466
Hirschsprung disease in, 1267–1268
inguinal hernia in, 989, 994, 999
lung transplantation in, 675
mortality of, 91t
necrotizing enterocolitis in, 1188–1189, 1193–1194
parenteral nutrition in, 188
Premature thelarche, 771
Premaxilla, 700–701
Prenatal counseling, for hydronephrosis, 1414
Prenatal diagnosis, 77–78
biochemical screening in, 77
of bladder exstrophy, 1517–1518
of choledochal cyst, 1333–1334
of cloacal exstrophy, 1527
of conjoined twins, 1730–1731, 1731f, 1732f
of cystic fibrosis, 1075
of cystic lung lesions, 825–826, 826f, 827–829,
827f
of cystic mediastinal masses, 830
echocardiography in, 78
fetal imaging in, 78
fetal sampling in, 77–78
of jejunoileal atresia and stenosis, 1061, 1062f
magnetic resonance imaging in, 45. See also Magnetic
resonance imaging, prenatal.
molecular genetics and, 22
and perinatal management, 78t, 82, 84t
ultrasonography in, 38–39, 39f, 78, 79f.
See also Ultrasonography, prenatal.
Prenatal interventions. See Fetal interventions.
Prenatal surgical consultation
ethics in, 239–240
in patient- and family-centered care, 249
Prentiss maneuver, for cryptorchidism, 1010
Prepubertal Testis Tumor Registry, 549
Prepuce
development of, 1532, 1535f
hooded, fashioning of, 1578–1579, 1579f
Preservation solutions, in lung transplantation, 676
Pressure-controlled ventilation, 117
Pressure-flow study, in ureteropelvic junction
obstruction, 1419, 1419f
Pressure support ventilation, 118
Pressure-volume loop, 113, 113f
Preterm delivery
with abdominal wall defects, 979
defects managed by, 78t, 82
PRETEXT staging
of hepatoblastoma, 465–466, 467f, 469–470,
469f, 470f
of hepatocellular carcinoma, 477, 478t
Primary survey, 263–268
Primitive neural ectodermal tumor (PNET), 575,
575f, 594–596, 595f, 596f
Probiotics
for necrotizing enterocolitis, 1206
for pouchitis, 1228
in short bowel syndrome, 1138
Procainamide, for supraventricular tachycardia, 138,
139t
Process measures, in performance analysis,
234–235
Processus vaginalis, 1003, 1004f
persistence of, 986, 986f, 1006. See also Inguinal
hernia.
Proctalgia fugax, 1320
Proctocolectomy. See also Colectomy.
for Crohn disease, 1214, 1214t
for familial adenomatous polyposis, 488, 1181
laparoscopic, 1225–1226, 1226f
with permanent ileostomy, 1223
for ulcerative colitis, 1217, 1222, 1229
Proctocolitis, infantile, 1320
Progestins, hypospadias and, 1537
Proliferation signal inhibitors, in renal
transplantation, 625
Prone positioning, as adjunct to mechanical
ventilation, 120
Pronephros, 1405, 1411–1412
Propofol, 202f, 211–212
for burns, 382
Propranolol
in burn injury, 380–381
for hepatic hemangioma, 1617–1618
for infantile hemangioma, 1616
for subglottic hemangioma, 849f, 850
for supraventricular tachycardia, 138
Propylthiouracil, for Graves disease, 747–748
Prospective cohort studies, 230
Prospective randomized controlled trials,
230–232
Prostaglandin E, for peptic ulcer disease, 1033
Prostaglandin E1
for coarctation of the aorta, 1650
for patent ductus maintenance, 139
for transposition of the great arteries, 1662
Prostaglandins, stress ulcers and, 1031
Prostate, rhabdomyosarcoma of, 498
Prostatic utricle, 1559
Prosthetic materials, for chest wall defects, 573
Prosthetic patch, for congenital diaphragmatic hernia
repair, 819, 820f
Protein. See also Amino acids.
metabolism of, in neonate, 102–103
surgery and, 107
requirements for, 181–182, 181t
Protein C
activated
in burn injury, 371
recombinant, for sepsis, 162
deficiency of, 174–175
Protein S, deficiency of, 175
Proteomics, 404
Prothrombin time, in coagulation disorders, 171
Proto-oncogenes, 399, 400t
activation of, 401
Proton beam radiation therapy, 413
Proton pump inhibitors
for gastroesophageal reflux disease, 953
for peptic ulcer disease, 1033, 1033t
for stress ulcers, 1034
Prune-belly syndrome, 975, 1505–1514
abdominal wall in, 1506, 1507f, 1508f
associated anomalies with, 1510
bladder in, 1507, 1510f
cryptorchidism in, 1004
kidney in, 1506–1507, 1509f, 1510f
management of, 1510–1514, 1512f, 1513f
abdominoplasty in, 1506, 1508f
megaureter in, 1497, 1505–1507, 1510f
testes in, 1509
urethra in, 1507–1508, 1510f, 1511f
Pruritus, in progressive familial intrahepatic
cholestasis, 1343
Pseudoaneurysm, after splenic injury, 294, 295f
Pseudocysts
meconium, 1081, 1081f
pancreatic, 1377–1378, 1378f
splenic, 294, 295f, 1386
Pseudodiverticulum, after esophageal atresia repair,
906
Pseudomonas infection, in cystic fibrosis, 865
Volume 1, pages 1–768; Volume 2, pages 769–1738
Pseudopolyps
in ulcerative colitis, 1218, 1218f
on vocal cords, 952–953, 953f
Pseudotumor, inflammatory
hepatic, 462
intestinal obstruction in, 1133
pulmonary, 567
Pseudotumor cerebri, 1681–1682
bariatric surgery and, 1043
Psychological factors
in bariatric surgery, 1043–1044, 1049
in cryptorchidism, 1008
Psychosocial development, in renal transplant
patient, 629
PTEN hamartoma-tumor syndrome, 1630
Puberty
precocious
in adrenocortical lesions, 563
in ovarian choriocarcinoma, 543
in ovarian tumors, 530
with premature thelarche, 771
in testicular tumors, 550
pseudoprecocious, in ovarian granulosa-theca cell
tumors, 539–540, 539f
Pubic tubercle, in inguinal hernia repair, 989, 990f
Puborectalis, 1311, 1312f
Pull-through
endorectal
for familial adenomatous polyposis, 488
for Hirschsprung disease, 1269–1270, 1269f,
1272
rectal prolapse after, 1316–1317
for ulcerative colitis, 1223, 1225–1227, 1225f,
1227f
laparoscopic, 1225–1226, 1226f, 1270, 1270f
obstructive symptoms after, 1274–1277, 1274t,
1275f
transanal (perineal), 1271–1272, 1271f
vaginoplasty with, 1581–1582, 1583f
Pulmonary. See also Lung; Respiratory entries.
Pulmonary airway malformation, congenital, fetal
interventions for, 83–85
Pulmonary arterial anastomotic stenosis, after lung
transplantation, 677–678
Pulmonary arteries
catheterization of, 117
development of, 1665, 1666f
growth of, 111–112
Pulmonary artery banding
for atrioventricular septal defect, 1658
for ventricular septal defect, 1656–1657
Pulmonary artery index, in congenital diaphragmatic
hernia, 816
Pulmonary artery sling, 853, 854, 1666, 1669–1670,
1670f
Pulmonary atresia, cardiovascular management in,
139
Pulmonary circulation, 114
Pulmonary compliance, 113, 113f
Pulmonary edema, during extracorporeal life
support, 128
Pulmonary enteral formula, after burn injury,
381–382
Pulmonary fibrosis, lung transplantation for, 673
Pulmonary function
after congenital diaphragmatic hernia repair, 822
after lung transplantation, 681
in meconium ileus, 1083
in pectus excavatum, 781–783
Pulmonary function testing, in congenital
diaphragmatic hernia, 816–817
Pulmonary gas exchange, 114–115, 115f
Pulmonary hematoma, 277
Pulmonary hypertension
congenital diaphragmatic hernia and, 813
fixed, as contraindication to heart transplantation,
662
lung transplantation for, 672–673
persistent, treatment of, 162–163, 818
portal hypertension with, 1360
serum-ascites albumin gradient in, 1172
Pulmonary hypoplasia. See also Diaphragmatic
hernia, congenital.
annular, 1660, 1662f
 INDEX I-37

Pulmonary hypoplasia (Continued) Pyloroplasty Radiography (Continued)

in congenital diaphragmatic hernia, 85 for peptic ulcer disease, 1033 in developmental dysplasia of hip, 1700–1701,
embryology of, 811–812 for pyloric atresia, 1035f, 1036 1701f
fetal interventions for, 817 for stress ulcers, 1034–1035 in gastric volvulus, 1037
model systems for, 812 Pyoderma gangrenosum in intestinal rotation and fixation disorders,
treatment of, 823, 823f in Crohn disease, 1211 1117–1118, 1119f
Pulmonary monitoring, 115–117 in ulcerative colitis, 1219 in intussusception, 1099–1100, 1100f
invasive, 116–117 Pyogenic granuloma, 1618, 1619f in jejunoileal atresia and stenosis, 1061–1062,
noninvasive, 116 Pyramid procedure, in hypospadias repair, 1542, 1063f, 1064f, 1065f
Pulmonary nodules 1542f in meconium ileus, 1075–1077, 1076f
core needle biopsy of, 419–420 Pyriform aperture stenosis, 713 in musculoskeletal trauma, 331–332, 331f
thoracoscopy with biopsy of, 421, 421f Pyriform sinus fistula, 747, 747t in necrotizing enterocolitis, 1188f, 1198
in Wilms’ tumor, 436 in neuroblastoma, 444
Pulmonary parenchymal disease, lung in pancreatitis, 1373
transplantation for, 674 Q in perineal fistula, 1293, 1293f
Pulmonary physiology, 112–115 Quad screen, 77 in pyloric atresia, 1035, 1035f
Pulmonary physiotherapy, for meconium ileus, Quadrangular membrane, 722 of salivary glands, 730, 730f
1082 Quality of care, 234–236, 235f in short bowel syndrome, 1136, 1136f
Pulmonary toilet, for inhalation injury, 376 in spine and spinal cord injury, 358
Pulmonary vascular disease, lung transplantation for, Radioimmunoguided surgery, 54
672–673 R Radioisotope scanning, technetium-labeled sucralfate,
Pulmonary vascular resistance, in congenital Radial artery, catheterization of, 116–117 after caustic ingestion, 921–922, 921f
diaphragmatic hernia, 813, 818 Radial deficiencies, 1722 Radioisotopes, for molecular imaging, 46–47
Pulmonary venous anastomotic stenosis, after lung Radial nerve, injury to, 337–338, 338f Radionuclide imaging. See Scintigraphy.
transplantation, 677–678 Radiation therapy, 51 Radiosurgery
Pulmonary venous return, partial anomalous, atrial adverse effects of, 414, 414t frameless image-guided, 52, 52f, 53f, 54
septal defect with, 1652–1653, 1654 for bone tumors, 583 linear accelerator, 52
Pulmonic stenosis, transposition of the great arteries chemotherapy with, 412 stereotactic, 51–52
with, 1661, 1662, 1663f clinical considerations in, 411–412 in children, 53–54
Pulse oximetry, 116, 213 for colorectal cancer, 490 platforms for, 52
Purpura complications of, 522 Ramstedt pyloromyotomy. See Pyloromyotomy.
Henoch-Schönlein definitive, 411 Randomized controlled trials, 230–232, 245
intussusception in, 1102 for ependymomas, 597 Ranitidine, for peptic ulcer disease, 1033
after renal transplantation, 628 focal, 413 RANTES, 149
immune thrombocytopenic, 170, 1386, 1387 fractionation in, 51, 412 Ranula, 721, 721f, 732, 732f
Push endoscopy, in gastrointestinal bleeding, general principles of, 411–414 Rapamycin, for neuroblastoma, 458
1154 head and neck, thyroid cancer after, 748–749 RAS protein, activation of, 401
Putty sign, in meconium ileus, 1075 hepatic, for neuroblastoma stage IV-S disease, 450 Rastelli procedure, 1662, 1663f
Pyelogenic cyst, 1403 for Hodgkin lymphoma, 520–521, 522 RB gene, inactivation of, 401–402
Pyelography breast cancer after, 777 Recall bias, 229, 233
intravenous intraoperative, 413 Receptors, 399
in renal injury, 313 for bone tumors, 585, 585f Rectal biopsy
in ureteral injury, 319–320 for neuroblastoma, 450, 456–457 in Hirschsprung disease, 1267–1268, 1268f
retrograde targeted, 457, 458 in intestinal neuronal dysplasia, 1280, 1281f
in ureteral trauma, 313 for non-Hodgkin lymphoma, 526 in isolated hypoganglionosis, 1282, 1283f
in ureteropelvic junction obstruction, 1418, palliative, 413–414 Rectal bleeding
1419f, 1422 postoperative, 412 in anal fissure, 1317
Pyelonephritis preoperative, 411–412 differential diagnosis of, 1179
imaging of, 1431 pulmonary, late effects of, 437 in intussusception, 1099
versus lower urinary tract infection, 1427 for rhabdomyosarcoma, 496 in portal hypertension, 1358–1359
after pyeloplasty, 1425 with systemic agents unrelated to efficacy, 412 in short bowel syndrome, 1137
renal abscess in, 1431 treatment techniques in, 412–413 Rectal mucosectomy, colectomy with, 1181
Pyeloplasty for Wilms’ tumor, 435–436, 436t Rectal prolapse, 1316–1317, 1316f
complications of, 1425 Radii, absent, thrombocytopenia with, 169 after anorectoplasty, 1307
dorsal lumbotomy approach to, 1423, 1424f Radioclinodactyly, 1722 in meconium ileus, 1082
minimally invasive, 1424–1425 Radiofrequency ablation postoperative, after pull-through operation,
open, 1421–1423, 1422f, 1424f of bone tumors, 584 1316–1317
outcome of, 1425 of hepatocellular carcinoma, 480 treatment of, 1316–1317
Pyelostomy Radiofrequency energy, 50 Rectal washouts, 1314–1315
cutaneous, 1489 Radiography Recto-anal inhibitory reflex (RAIR), in Hirschsprung
for posterior urethral valves, 1469 in abdominal trauma, 289 disease, 1267
Pyloric atresia, 1033–1034 in adhesive bowel obstruction, 1128, 1128f Rectobladder neck fistula, 1296f, 1297
Pyloric duplications, 1036 in appendicitis, 1257 reconstruction for, 1298–1300, 1300f, 1301f
Pyloric stenosis, hypertrophic, 1021–1030 in ascites, 1172, 1173f Rectoprostatic fistula, 1297
anatomy and histology of, 1021, 1022f in bone tumors, 579f, 581 Rectosigmoid motility, 1291
clinical features of, 1022 in cervical spine injury, 335 Rectosphincteric inhibitory reflex, 1283
diagnosis of, 1023–1024, 1023f chest Rectourethral fistula, 1296f, 1297
differential diagnosis of, 1022 in airway obstruction, 723 penile agenesis with, 1585, 1588f
etiology of, 1021–1022 in aortic trauma, 283, 283t, 284f reconstruction for, 1297–1298, 1298f, 1299f,
historical perspective on, 1021 in aspergilloma, 860, 860f 1300f
molecular genetics of, 21–22 in cervical lymphadenopathy, 739 Rectovaginal fistula, 1162, 1162f
postoperative care in, 1027–1028 in chest tube insertion, 873–874 Rectum. See also Anorectal entries.
treatment of, 1024–1028 in congenital diaphragmatic hernia, 810f, 814 anatomy of, 1311, 1312f
complications after, 1028 in enteric duplications, 834–835, 834f atresia and stenosis of, 1301
nonoperative, 1028 in esophageal perforation or rupture, 889–890, cancer of. See Colorectal cancer.
operative procedures in, 1024–1025, 1025f, 890f, 891f duplications of, 1158f, 1161–1163, 1161f, 1162f
1026f in Hodgkin lymphoma, 518–519, 518f polyps of, 1152, 1152f, 1180, 1183. See also Polyp
outcome of, 1025–1027, 1027t, 1028 of lung abscess, 869, 869f (s), gastrointestinal.
preoperative preparation for, 1024 in pneumococcal infection, 855, 856f sexual abuse and, 1320
Pyloromyotomy in pulmonary contusion, 277, 278f solitary ulcer of, 1319
for hypertrophic pyloric stenosis, 1024–1025 in thoracic trauma, 274 stricture of, after anorectoplasty, 1307
laparoscopic, 1025f, 1026f in cholelithiasis, 1343 trauma to, 308
open, 1024–1025, 1025f in colonic atresia, 1248–1249, 1248f vaginal replacement with, 1304, 1305f
outcome of, 1025–1027, 1027t, 1028 in constipation, 1314 Rectus sheath nerve block, 222, 222f

Volume 1, pages 1–768; Volume 2, pages 769–1738

I-38 INDEX

Recurrent laryngeal nerve, 722 Renal transplantation (Continued) Respiratory failure (Continued)
injury to donor selection for, 619–621, 620t in trauma patient, 265, 266f
in lung transplantation, 678 early graft dysfunction in, 623–624 Respiratory papillomatosis, recurrent, 726, 726f,
in thyroidectomy, 749 electron-beam computed tomography after, 43, 43f 843–844, 843f
Recurrent respiratory papillomatosis, 726, 726f, graft and patient survival in, 625–626, 626f Respiratory quotient, 115
843–844, 843f graft loss in Respiratory syncytial virus infection, as pneumonia,
Red blood cell distribution width index, in anemia, acute rejection and, 627 858–859, 858f
165 risk factors for, 626 in cancer patient, 861
Red blood cells. See Erythrocyte(s). historical perspective on, 605–613, 606t, 607f Resting energy expenditure
Red currant jelly stool, in intussusception, 1095, immunosuppressive therapy for, 624–625 gestational age and, 97
1096f, 1099, 1147, 1149f indications for, 617 in neonate, 97–98, 98f
Red pulp, 1385 medical complications of, 628–629 Resuscitation
5a-Reductase deficiency, 1568t, 1571, 1573–1574 nephrectomy and, 618, 619 cardiopulmonary
Reduction, fracture, 332 nonadherence after, 627 extracorporeal life support for, 123–136.
Reed-Sternberg cells, in Hodgkin lymphoma, 517, organ preservation for, 613–614, 614f See also Extracorporeal life support.
518f, 519 outcomes of, 625–629 in lung transplantation, 676
Reflux pancreas transplantation with, 632, 632f, 634, 634f in congenital diaphragmatic hernia, 817–818
gastroesophageal. See Gastroesophageal reflux for posterior urethral valves, 1556–1557 fluid. See Fluid management or resuscitation.
disease. postoperative care in, 623 in gastrointestinal bleeding, 1147
laryngeal, 952–953, 953f preoperative preparation for, 621 in Hirschsprung disease, 1268
normal, 940 recipient evaluation for, 617–619, 618t of injured patient, 262–263. See also Emergency
uretero-ureteral (yo-yo), 1444 recurrent disease after, 627–628 management.
vesicoureteral. See Vesicoureteral reflux. rejection in principles of, 263–270
Regional anesthesia, 220–226 acute, 627 for sepsis
caudal, 224–225, 224f chronic, 627 continued, 160–162
epidural, 225–226, 225t treatment of, 625 goals of, 159, 159t
infiltration, 221 surgery for, 621–623 initial, 155–160
local anesthetics used in, 220–221, 221t anesthesia with, 621 in variceal hemorrhage, 1362
neuraxial, 224–225, 224f in infants and children, 621–622 Resuscitation phase, 268
peripheral nerve and plexus blocks in, 221–222. in larger children, 622 RET gene
See also Nerve block. in patients with previous urologic procedures, in Hirschsprung disease, 21, 21t, 1266
topical, 221, 221t 622–623 in medullary thyroid carcinoma, 750
Rehabilitation, for burns, 384–385 procedure in, 621 in thyroid cancer, 749
Rehbein, F., 14, 14f timing of, 619 Retention cysts, pancreatic, 1376
Rejection. See Host-versus-graft response. urologic issues in, 617–618, 619, 622–624 Reticular dermis, 370
Renal. See also Kidney. vascular thrombosis in, 627 Retinal hemorrhage, in child abuse, 388
Renal abscess, in pyelonephritis, 1431 Renal vein Retinoblastoma, 405
Renal agenesis, 1395–1396 thrombosis of, 624, 1439–1440 13-cis-Retinoic acid, 410
Renal anomaly, ipsilateral, obstructed hemivagina trauma to, 317 Retinoid-regulated target genes, in congenital
with, 1602, 1603f Wilms’ tumor extension to, 431 diaphragmatic hernia, 810
Renal artery Renin-angiotensin-aldosterone system, 558 Retinoids, for neuroblastoma, 452–453, 457
aneurysm of, 1639, 1639f Renovascular hypertension, 1636 Retinol-binding protein, nutritional status and, 180
dissection and thrombosis of, 1639 abdominal aortic coarctation with, 1632, 1634 Retroperitoneal germ cell tumors, 516
stenosis of, 1636–1638, 1637f, 1638f after renal trauma, 318 Retroperitoneal lymph node dissection, radical
abdominal aortic coarctation with, 1632, 1634 treatment of, 1637, 1638 inguinal orchiectomy and, 554–556, 555f
trauma to, 311, 317 Renovascular trauma, 311, 313, 316–317, 318f Retroperitoneal rhabdomyosarcoma, 497
Renal blood flow, in burn injury, 371 Reperfusion, in liver transplantation, 648 Retropharyngeal space infection, 718, 718f
Renal cell carcinoma, 438, 438f Reperfusion injury, in lung transplantation, 678 Retrospective cohort studies, 229–230
in horseshoe kidney, 1408 Reporter genes, in molecular imaging, 47–48 Retroviral vectors, for gene transfer, 24, 24t
in von Hippel-Lindau disease, 1399 Reproductive issues, in vaginal agenesis, 1598–1599 Revascularization, renal artery, 317
Renal colic, 1434–1437 Reproductive system, female, abnormalities of, Reverse transcription polymerase chain reaction (RT-
Renal compensation, in neonate, 94–95 1591–1613 PCR), 404t
Renal disease Research. See also Technological innovation. Review articles, 232
cystic. See Kidney, cystic disease of. fetal surgical, 88 Reviews, systematic, 232–233
end-stage pediatric Rex shunt
bladder augmentation or replacement in, on innovative devices, 63 for hypersplenism, 1368
1482–1483 on innovative procedures, 63, 64t for varices, 1365–1366, 1366f, 1367–1368, 1368f
epidemiology of, 617 in pediatric surgery, 7–8 Rhabdoid tumors
etiology of, 617, 618t Residual volume, 112, 113f atypical, 600
hyperparathyroidism in, 752 Respect, in patient- and family-centered care, 248 hepatic, 480
renal transplantation for, 617–633. Respiration renal, 437–438
See also Renal transplantation. monitoring of, 115–117 Rhabdomyosarcoma, 491–503
preexisting, injury risk with, 312 invasive, 116–117 abdominal wall, 497
Renal dysgenesis, 1395 noninvasive, 116 alveolar, 491–492, 494
Renal dysplasia physiology of, 112–115 assessment of response to treatment in, 496–498
with duplex collecting system, 1442f, 1443 Respiratory acidosis, 94 biliary, 480, 497
in posterior urethral valves, 1555–1557 Respiratory alkalosis, 94 biopsy of, 493–494
in prune-belly syndrome, 1506, 1509f, 1511 Respiratory compensation, 94 breast, 777
Renal failure Respiratory distress. See also Airway obstruction; chemotherapy for, 495–496
in autosomal recessive polycystic kidney disease, Apnea. chest wall, 497, 576
1397 in congenital diaphragmatic hernia, 814 clinical grouping for, 494–495, 494f, 495f,
in Wilms’ tumor, 433, 433f after esophageal atresia repair, 915 498–499
Renal function, in neonate, 93 in esophageal atresia with distal fistula, 903–905 clinical presentation in, 492
Renal insufficiency, after lung transplantation, 680 Respiratory failure. See also Acute respiratory distress demographics in, 491
Renal scintigraphy syndrome (ARDS). embryonal, 491–492, 494
in ureterocele, 1449–1450, 1449f in congenital diaphragmatic hernia, 821 extremity, 497–498
in ureteropelvic junction obstruction, 1416–1417, after inguinal hernia repair, 998 genitourinary, 498
1416f, 1417f, 1418f management of head and neck, 496, 728
Renal transplantation, 610, 610f, 617–633 extracorporeal life support in, 123–136. lung, 569t, 570, 571f
bladder augmentation or replacement before, See also Extracorporeal life support. lymph node sampling/dissection in, 494
1482–1483 mechanical ventilation in, 120–122 metastatic, 498, 499
contraindications to, 617 pharmacologic agents in, 120 oral cavity and pharyngeal, 721
delayed graft function in, 626–627 mortality risk in, 124 parameningeal, 496
dialysis access and, 619 in spine trauma, 359 paraspinal, 497

Volume 1, pages 1–768; Volume 2, pages 769–1738


Rhabdomyosarcoma (Continued)
paratesticular, 498, 552
perineal/perianal, 497
preoperative workup for, 492–493
prognosis in, 498–499
pulmonary metastasis in, 571–572
radiation therapy for, 496
retroperitoneal/pelvic, 497
salivary gland, 733, 733f
sites of, 496–498
staging of
postsurgical, 494–495, 494f, 495f
pretreatment, 493, 493f
superficial nonparameningeal, 496
surgery for, 493–495
survival rate for, 491, 492f, 494–495, 495f
truncal, 496–497
tumor biology in, 491–492
urethral, 1558
vaginal, 1607, 1608f
Rhinosinusitis, 713
Rib(s)
aplasia of, in Poland syndrome, 797, 798f, 799f
defects of
in cerebrocostomandibular syndrome, 807–808
in Jarcho-Levin syndrome, 807, 808f
in Jeune syndrome, 805, 807f
after pectus excavatum repair, 785–789, 789f
first, fracture of, 275
fracture of
in child abuse, 275, 389, 389f
traumatic, 272, 275
titanium, for Jeune syndrome, 806–807, 808f
Rib graft, for Poland syndrome, 797–798, 801f
Rickets, with parenteral nutrition, 193
Rieger syndrome, 967
Risk, absolute versus relative, 233
Risk stratification, in chemotherapy, 406
Rituximab, for immune thrombocytopenic purpura,
1387
RNA microarrays, 402, 404
Robinow syndrome, 967
ROBODOC system, 58
Robotic surgery, 57–60
advantages and limitations of, 59–60
in children, 60
classification of, 58, 58t
current status of, 58–59, 58f, 59f
Rocuronium, 210t
Ropivacaine, 220–221, 221t
Rosai-Dorfman disease, cervical lymphadenopathy
in, 743
Rosettes, in neuroblastoma, 447–448, 448f
Rotation, intestinal. See Intestinal rotation and
fixation; Volvulus.
Rotationplasty, for bone tumors, 585–586, 586f
Rotavirus infection, intussusception and, 1097
Rotavirus vaccine, intussusception and, 1097
Roux en Y cystojejunostomy, 1378
Roux en Y gastric bypass surgery, 1041–1042, 1046,
1047f
Roux en Y hepaticojejunostomy, 1336–1338, 1336f,
1337f, 1338f
Roux en Y jejunostomy, 1237f, 1238, 1336–1338,
1336f, 1337f
Rule of nines for burns, 372, 373f
S in pectus excavatum, 783–784
Saccules, terminal, 111
Sacral agenesis, 1470, 1709
neuropathic bladder in, 1460, 1460f
Sacral anomalies, with anorectal malformations,
1290, 1290f
Sacral parasite, 1737
Sacral ratio, 1290, 1290f
Sacrococcygeal teratoma, 511–514
classification of, 512, 513f
clinical presentation and initial evaluation of,
511–512, 512f
fetal interventions for, 86, 512
malignancy rate of, 512
postoperative management of, 513–514, 515f
surgical management of, 512–513, 514f, 515f
INDEX I-39
Saethre-Chotzen syndrome, 693 Scoliosis
Safety, fire, 258–259 congenital, 1706, 1706f, 1707
Safety seats, child, 258, 258f treatment of, 1707–1708, 1708f
Sagittal split osteotomy, bilateral, 695 pectus excavatum in, 779–780, 780t
Sagittal suture, premature fusion of, 692, 692f Scorpion sting, 341
St. Jude’s Murphy staging system for non-Hodgkin Scotland, pediatric surgery in, 10–11
lymphoma, 524, 524t Scrofula, 742
Salivary glands, 729–738 Scrotal ectopia, 1563
anatomy and physiology of, 729 Scrotoplasty, for penoscrotal transposition,
biopsy of, 730 1583–1584, 1586f
classification of, 729 Scrotum
cystic disease of, 731–732, 732f abnormalities of, in hypospadias, 1539–1540,
diagnostic evaluation of, 729–730, 730f 1539f
embryology of, 729 acute, 1014–1016, 1014f, 1015t
inflammatory disease of, 731, 731f anomalies of, 1563
neoplasms of, 732–733 bifid, 1583–1584, 1586f
benign, 732–733, 732f, 733f fat necrosis of, 1015
malignant, 722, 733, 733f idiopathic edema of, 1015
pathology of, 729 inflammation of, 1015
surgical considerations for, 734–735, 734f injury to, 324–325
Salter-Harris classification of fractures, 328, 329f, pre-penile, 1583–1584, 1586f
338 swelling of, after hernia repair, 997
Sandifer syndrome, 951 SEAL (subcutaneous endoscopically assisted ligation
Santorini, duct of, 1371 of the hernia sac) technique, 991–992
drainage of, in pancreas divisum, 1376 Seat-belt sign, 307, 307f
Santulli enterostomy, 1080f, 1081 Seat belts, 258, 258f, 335, 336f
Saphenous vein graft, for extremity vascular injuries, Seats, child safety, 258, 258f
364 Sebaceous nevus, 1714
Sarcoma. See also specific sarcoma types. Secretin, before magnetic resonance
breast, 777 cholangiopancreatography, 1335
clear cell, 503 Sedation
Ewing. See Ewing sarcoma. for burns, 382–383, 382t
fine-needle aspiration biopsy of, 418 for chest tube insertion, 875
hepatic, 480 for esophagoscopy, 886
Kaposi, 1620 for intussusception reduction, 1104
osteogenic. See Osteogenic sarcoma parental presence during induction of anesthesia
(osteosarcoma). and, 250, 251
ovarian, primary, 547 preoperative, 204–205
soft tissue. See also Rhabdomyosarcoma. of trauma patient, for intubation, 265
nonrhabdomyosarcoma, 501–503, 502f Seizures
synovial, 502, 502f, 503 arising from temporal lobe, 1690, 1691
Sarcoma botryoides, 1607, 1608f atonic, 1691
Sarfeh shunt, 1364 with extracorporeal life support, 129
Scald burns, 369, 370f refractory, surgery for, 1687–1693.
Scalp, aplasia cutis congenita of, 1713–1714 See also Epilepsy surgery.
Scalpel, harmonic, 49 in trauma patient, 269, 353
Scandinavia, pediatric surgery in, 13 Seldinger technique, 874
Scaphocephaly, 692, 692f Selectins, in neutrophil adhesion, 146
Scardino-Prince vertical flaps, 1422f, 1423 Selection bias, 233
Scarpa fascia, in inguinal hernia repair, 989–991, 990f Selenium, requirements for, 184, 184t
Scarring Self assembly, in nanoelectromechanical systems, 62
hypertrophic, after burn injury, 384 Semicircular canals, 707
natural orifice translumenal endosurgery (NOTES) Seminoma (dysgerminoma), 507, 508, 508f,
and, 56–57 541–543, 542f
stealth surgery and, 54–56, 55f Senescence, cellular, 399
Scene generation software, 67, 67f Sensors, microelectromechanical, 61, 61f
Scintigraphy Sentinel node mapping, in rhabdomyosarcoma, 494
in biliary atresia, 1324, 1334–1335 Sepsis, 141–169
after caustic ingestion, 921–922, 921f biochemical markers of, 153–154
in cholecystitis, 1343 burn wound, 376
in chylothorax, 878 with central venous parenteral nutrition, 193–194
in conjoined twins, 1733 continuum of, 152, 153t
in Meckel diverticulum, 1088–1089, 1089f diagnosis of, 152–154
MIBG diagnostic criteria for, 142t, 152, 152t, 153t
in neuroblastoma, 444, 444f epidemiology of, 142–144
in pheochromocytoma, 560 host defense mechanisms in, 145–152, 145f
in motility disorders, 941 management of, 154–163
NP-59, in adrenocortical lesions, 563 ACCM/PALS guidelines for, 154–162
airway, breathing, and circulation (ABCs) in, 155
renal algorithms for, 155, 156f, 157f
in ureterocele, 1449–1450, 1449f antibiotics in, 158–159
in ureteropelvic junction obstruction, blood transfusion in, 158
1416–1417, 1416f, 1417f, 1418f early goal-directed therapy for, 154–155
in thoracic trauma, 274 fluid resuscitation in, 155–158
of thyroid gland, 746 intravenous immunoglobulin in, 162
ventilation-perfusion, in inhalation injury, 375 recombinant human activated protein C in, 162
Sclerosing stromal tumors, ovarian, 540 resuscitation in
Sclerotherapy continued, 160–162
for capillary-lymphaticovenous malformation, goals of, 159, 159t
1628 initial, 155–160
for esophageal varices, 885 source control in, 159
for lymphatic malformation, 1624 stabilization in, 160–161
for variceal bleeding, 1363 steroids in, 160
for venous malformation, 1625 surviving sepsis campaign in, 154, 155–162
Volume 1, pages 1–768; Volume 2, pages 769–1738
I-40 INDEX
Sepsis (Continued)
mortality in, 143–144
neonatal, 153, 157f, 162–163
nutritional support in, 197
pathogenesis of, 144–152, 144f
PIRO system for, 154
prevention of, 154
severe, 141, 143, 153t
after splenectomy, 1391–1392
terminology of, 141–142, 142f
Septal deviation, 715
Septal hematoma, 715
Septic shock
blood transfusion for, 158
catecholamine-resistant, 159t, 160–161
persistent, 161
cold versus warm, 159t, 160, 161
definition of, 141
diagnosis of, 152, 153t
fluid-refractory, 159–160, 159t
neonatal, 157f, 162–163
refractory, 159t, 161–162
steroids for, 160
Sertoli-Leydig cell tumors, ovarian, 540–541
Serum-ascites albumin gradient (SAAG), 1172
Sevoflurane, 202t, 207t, 208–209
Sex chromosomes, disorders of, 1568t, 1571, 1574,
1575
Sex cord-stromal tumors
ovarian, 530–531, 530t, 531t, 539–541, 539f
testicular, 551
Sex cord tumors with annular tubules, 541
Sex-determining region of the Y chromosome (SRY)
gene, 1565–1567
Sex hormones, adrenocortical lesions producing, 563
Sexual abuse
anorectal pathology secondary to, 1320
genital injuries in, 308
Sexual differentiation
developmental biology of, 1565–1567, 1566f
disorders of. See Disorders of sex development
(DSD).
Shaken baby syndrome, 387, 387f
Shenton line, 1700–1701
Shimada neuroblastoma classification, 445–446,
446t, 452–453
Shock
hypovolemic, in cervical spine injury,
356–357
in inguinal hernia, 995, 996
neurogenic, in cervical spine injury, 356–357
refractory, 159t, 161–162
septic. See Septic shock.
in trauma patient, damage control for, 270
Short bowel syndrome, 1135–1148
clinical assessment of, 1137
clinical presentation in, 1136–1137, 1136f
definitions related to, 1135
etiology of, 1135–1136
incidence of, 1136
medical management of, 1138–1141
for bacterial overgrowth, 1140
for bowel adaptation promotion, 1141
for catheter-related infections, 1139–1140
for decreased intestinal motility, 1140
for hypergastrinemia, 1140–1141
for increased stoma output or diarrhea, 1140
for intestinal failure–associated liver disease,
1138–1139
morbidity and mortality of, 1136
multidisciplinary program for, 1145
in necrotizing enterocolitis, 1204
nutritional support for, 198, 1137–1138
prognostic factors in, 1135, 1136f
after resection for intestinal rotation and fixation
disorders, 1124
surgical treatment of, 1141–1145
bowel conservation in, 1141
intestinal lengthening procedures in,
1141–1144, 1142f, 1143f, 1144f
intestinal transplantation in, 653, 654f, 1145
tissue-engineered intestinal construct in, 32
treatment of, 1070, 1071
Short gut syndrome, cloacal exstrophy with, 1526
Shoulder
dislocation of, in birth trauma, 391–392
elevation of, in torticollis, 764, 765f, 766
Shprintzen omphalocele syndrome, 977t
Shunt(s)
Blalock-Taussig, 1660
cerebrospinal fluid
complications of, 1683–1686
for hydrocephalus, 1683
heparin-bonded, for aortic injury, 283
high-flow, in hepatic hemangioma, 1617
infection associated with, 1685
peritoneovenous, for hepatocellular ascites, 1173
pleuroperitoneal, for chylothorax, 879
portacaval
end-to-side, 1364
for portal hypertension–related bleeding, 1364
side -to-side, 1364
portosystemic
history of, 1355
transjugular intrahepatic, 1363–1364
Rex
for hypersplenism, 1368
for varices, 1365–1366, 1366f, 1367–1368,
1368f
Sarfeh, 1364
splenorenal
distal, 1364–1365, 1365f, 1368
proximal, 1364
thrombosis associated with, 1363–1364, 1366
for variceal bleeding, 1364–1365. See also Varices,
shunts for.
ventriculo-gallbladder, 1343
ventriculoatrial, for hydrocephalus, 1683
ventriculoperitoneal
complications of, 1683–1686
for hydrocephalus, 1677–1678, 1683
inguinal hernia and, 999
peritonitis with, 1233
Warren, 1364–1365, 1365f
Shwachman-Diamond syndrome, 1373
Sialadenitis
bacterial suppurative, 731
chronic, 731, 731f
viral, 731
Sialendoscopy, 730
Sialolithiasis, 731
Sickle cell disease, 168
acute chest syndrome in, 168, 1387
cholecystectomy in, 1341–1342, 1344
cholelithiasis in, 1341–1342
splenectomy for, 1387
Sigmoid cystoplasty, 1473–1474, 1492
Sigmoid vaginoplasty, 1587, 1589f, 1596–1598,
1597f
Sigmoidostomy, tube, 1237, 1240
Signal transduction, 398–399
Silastic silo
in gastroschisis reduction, 980f, 982
in omphalocele reduction, 980
Sildenafil, for esophageal dysmotility, 943
Silicone, for hypertrophic scarring, 384
Silk glove sign, 987, 987f
Silver nitrate, in burn care, 377
Silver sulfadiazine, in burn care, 376
Simonart band, 699–701, 700f, 701f
Singapore, pediatric surgery in, 16
Single photon emission computed tomography
in epilepsy surgery, 1689
molecular imaging using, 48
of thyroid gland, 746, 746f
Single ventricle, 1663–1665, 1664f
Sinus(es)
branchial cleft, 708
neck. See Neck, cysts and sinuses of.
paranasal, 712–713
urogenital. See Urogenital sinus.
Sinus tracts
dermal, 1679
piriform, 759, 759f
Sinus venosus atrial septal defect, 1652–1653,
1653f
Sinusitis, 713
SIOPEL staging system, for liver tumors, 469, 469t
Volume 1, pages 1–768; Volume 2, pages 769–1738
Sirolimus, in transplantation
liver, 649, 650t
lung, 676–677, 676t
pancreas, 636, 636f
renal, 625
Sistrunk procedure
for cervical dermoid cyst, 760
for thyroglossal duct cyst, 756, 756f
Skeletal. See also Bone; Musculoskeletal.
Skeletal anomalies, with vaginal agenesis, 1592
Skeletal disorders
in congenital diaphragmatic hernia, 810,
822–823
diffuse, thoracic deformities in, 805–808, 807f,
808f
Skin
anatomy of, 369–370
as barrier to infection, 145–146
cancer of, in renal transplantation, 629
congenital anomalies of, 1713–1714
embryology of, 1711–1712
functions of, 369–370
hemangioma of, 1616
lesions of
dermal sinus tracts with, 1679
spinal cord tethering with, 1678
Skin expanders. See Tissue expansion.
Skin flaps, for soft tissue trauma, 340
Skin grafts
for burns, 379–380, 379f
for soft tissue trauma, 340
for vaginoplasty
full-thickness, 1595
split-thickness, 1594–1595, 1595f
Skin staples, for cardiac wounds, 280
Skull
crush injuries to, 348–349, 349f
growth pattern of, 691
Skull fracture, 345, 350, 351, 711–712, 711f
basilar, 352–353
in child abuse, 387, 388f
complications of, 352
Sleep apnea, obstructive, 203–204, 719, 1043
Sleep-disordered breathing, 718–720, 719f
bariatric surgery and, 1043
Slit ventricle syndrome, 1686
Small cell carcinoma, ovarian, 547
Small for gestational age, 89, 91f
Small intestine. See also Duodenum; Ileum; Jejunal
entries.
atresia and stenosis of. See Jejunoileal atresia and
stenosis.
cancer of, in inflammatory bowel disease, 1215
duplications of, 1160–1161, 1161f
length of, parenteral nutrition dependence and,
1135, 1136f
obstruction of. See also Intestinal obstruction.
hernias as, 1130–1131, 1130f, 1131f
inflammatory adhesions as, 1130
mesenteric and omental cysts as, 1166–1167,
1167f
postoperative adhesions as, 1127–1129, 1128f,
1129f
postoperative ileus as, 1129
postoperative intussusception as, 1098, 1130
polyps in, 1097, 1097f, 1180. See also Polyp(s),
gastrointestinal.
tissue-engineered, 1144
transplantation of, in short bowel syndrome, 1145
trauma to, 305–308, 307f
Small left colon syndrome
colonic obstruction in, 1251–1252, 1251f
versus meconium ileus, 1077–1078
Small round cell tumor, desmoplastic, 503, 503f,
504f
Smoke alarm, 258
Smoke inhalation injury, 375–376, 376t
Smoking, ulcerative colitis and, 1218
Smooth muscle tumors, esophageal, 483
Snakebites, 340–341
Soak solutions, in burn care, 377
“Soap bubble” sign, in meconium ileus, 1075–1076,
1076f
Soave, F., 14, 15f
 INDEX I-41

Soave procedure Spinal cord injury (Continued) Squamous papilloma, of oral cavity, 721
complications of, 1274, 1275f initial, 357–359 Squint, 765
for Hirschsprung disease, 1269f, 1270 resuscitation and transport in, 344 Stab incision technique, in laparoscopic
transanal, 1271–1272, 1271f outcomes from, 360 cholecystectomy, 1345, 1346f
Sodium primary versus secondary, 343–344 Stapedius muscle, 707
abnormalities of. See Hypernatremia; spectrum of, 355–357, 356t Staphylococcal infection
Hyponatremia. without radiographic abnormality, 357 in liver abscess, 1350
in parenteral nutrition, 190, 190t Spinal dysraphism. See also Myelodysplasia; Neural in lymphadenitis, 740
restriction of, in hepatocellular ascites, 1172 tube defects; Spina bifida. Staphylococcus aureus infection
serum, in neonate, 93 occult, 1678–1680 in cystic fibrosis, 865
Sodium hypochlorite, in burn care, 377 Spine. See also Cervical spine; Lumbar spine; methicillin-resistant, 740, 856
Soft tissue Thoracic spine. as pneumonia, 856
congenital anomalies of, 1713–1714 anomalies of, 807, 808f, 1706–1709, 1706f, Stasis, zone of, in burns, 370–371, 371f
structural analysis of, 1712–1713 1707f, 1708f, 1709f Statin therapy, in burn injury, 371
treatment of, 1712–1713 with anorectal malformations, 1290 Statistics, 232
embryology of, 1711–1712 with cloacal exstrophy, 1527 Statutes, minor consent, 238–239, 239t
trauma to, 339, 340 epidural abscess in, 1697 Stealth surgery, 54–56, 55f
as birth injury, 391 fracture of, in child abuse, 389 Steatohepatitis, nonalcoholic, bariatric surgery and,
chest wall, 275 neurenteric cysts in, 1679 1043
tumors of stabilization of, 344, 357 Steatosis, with parenteral nutrition, 193
chromosomal translocations in, 400–401, 401t trauma to, 335, 335f, 336f, 354–360 Steel pectus support bar
nonrhabdomyosarcoma, 501–508 complications of, 359–360 allergy to, 784, 789–790, 792
rhabdomyosarcoma. See Rhabdomyosarcoma. epidemiology of, 354 displacement of, 792
Soft tissue sarcoma, nonrhabdomyosarcoma evaluation of, 354–355, 355f, 356f in pectus carinatum repair, 796
background and overview of, 501–502, 502f initial management of, 357–359 in pectus excavatum repair, 789–790, 790f
surgical approach and presentation of, 502–503 spectrum of, 355–357, 356t removal of, 790, 793f
Soiling, fecal. See also Incontinence, fecal. Spiral flap, for ureteropelvic junction obstruction, Stem cell transplantation
after pull-through for Hirschsprung disease, 1276, 1422f, 1423 for aplastic anemia, 166
1276t Spirometry, after lung transplantation, 677 for Crohn disease, 1212–1213
Solubilizing agents, for meconium ileus, 1078–1080, Spironolactone, for congestive heart failure, 135 fetal, 88
1080f Splanchnic artery general principles of, 415
Solumedrol, in heart transplant patient, 667t aneurysm of, 1641 for neuroblastoma, 457
Somatostatin stenosis of, 1639–1641, 1640f Stensen ducts, 716, 729
for chylothorax, 878 Spleen, 1385–1395. See also Hypersplenism. Stent(s)
for variceal hemorrhage, 1362 abscess of, 1387–1388 for aortic injury repair, 284
Somites, 1712 accessory, 1386 for bile duct injury, 299, 300f
Sonic hedgehog (Shh) gene, in hypospadias, 1536 anatomic abnormalities of, 1386–1387, 1386f drug eluting, 62
Sorafenib, for hepatocellular carcinoma, 479 anatomy of, 1385 for esophageal caustic injury, 923–924
Source control, in sepsis, 159 asplenia and polysplenia syndromes of, for extremity vascular injuries, 364
South Africa, pediatric surgery in, 17, 17f 1386–1387 for tracheomalacia, 914
SOX10 gene, in Hirschsprung disease, 21t cysts of, 1386 ureteral
Soy, in formulas, 186–187 embryology of, 1385 after pyeloplasty, 1425
Soybean lipid emulsions, 193 function of, 1385–1386 for urolithiasis, 1437
Space of Disse, 1171 historical perspective on, 1385 Stereotactic radiosurgery, 51–52
Spain, pediatric surgery in, 15 pseudocysts of, 1386 in children, 53–54
Special care need patient, nutritional support in, 199, sequestration of, in sickle cell disease, 168 platforms for, 52
199t trauma to Sternomastoid tumor, 763–764, 764f.
Specimen handling, 417–418 associated abdominal injuries with, 294 See also Torticollis.
Spherocytosis, hereditary, 169 birth-related, 392–393 Sternotomy, median, mediastinal infection after,
cholelithiasis in, 1342 damage-control strategies for, 294–298, 296f, 297t 879–880
recurrent, 1386 imaging of, 290f, 291, 291t Sternum
splenectomy in, 1344, 1387 nonoperative treatment of bifid, 804, 805f, 805t
Sphincteroplasty, for pancreatitis, 1375 complications of, 294, 295f congenital defects of, 799–804. See also Chest wall,
Sphincterotomy failure of, 294 congenital deformities of.
for anal fissure, 1317–1318 guidelines on, 291–292, 292t cleft or bifid sternum as, 804, 805f, 805t
for choledocholithiasis, 1344, 1344f treatment of, 291–299 ectopia cordis as
for pancreas divisum, 1376 operative, 292–293, 293t cervical, 803
Spider bites, 341 in trauma centers versus nontrauma centers, thoracic, 800–803, 803f
Spider nevi, 1621 293–294 thoracoabdominal, 803–804, 804f
Spina bifida, 1673, 1674–1675. wandering, 1386, 1386f fracture of, 275
See also Myelomeningocele. Splenectomy, 1388–1391 Steroid biosynthetic enzyme nomenclature, 1569t
Spinal cord. See also Central nervous system. complications of, 1390 Steroid cell tumors, ovarian, 541
compression of, in neuroblastoma, 455 in hereditary spherocytosis, 1344 Steroid hormones, synthesis of, 1570f
malformations of indications for, 1387–1388 Steroidogenesis factor (SF-1) gene, in sexual
lipomatous, 1679 laparoscopic, 1388–1390 differentiation, 1565, 1566f
occult, 1678–1680 conversion of, 1390 Steroids. See also Corticosteroids.
split, 1679 technique of, 1388–1390, 1388f, 1389f anabolic, for aplastic anemia, 166
tethering of, 1678–1680 open, 1388 in gender assignment surgery, 1576–1577
after myelomeningocele repair, 1676, 1680 partial, 1390–1391, 1391f Stertor, 722–723
occult, 1469–1470 in portal hypertension, 1365 Stickler syndrome, 1247
assessment of, 1453, 1454f, 1455f postoperative considerations in, 1391–1392 Stillborns, congenital diaphragmatic hernia in, 810
with cloacal exstrophy, 1527 preoperative immunization in, 1388 Stocking-glove distribution of burns, 369, 370f
with neuropathic bladder, 1459–1460 sepsis after, 1391–1392 Stoma. See Enterostoma; Gastrostomy; Urinary
Spinal cord injury, 354–360 Splenogonadal fusion, 1001, 1387 diversion.
anatomic considerations in, 354 Splenomegaly, in portal hypertension, 1360 Stomach. See also Gastric entries.
in coarctation of the aorta repair, Splenopexy, for wandering spleen, 1386 perforation of, in neonate, 1038–1039, 1038f
1651–1652 Splenoportography, in portal hypertension, 1361 polyps of, 1151, 1151f, 1181. See also Polyp(s),
complications of, 359–360 Splenorenal shunt gastrointestinal.
contusion as, 344 distal, 1364–1365, 1365f, 1368 position of, congenital diaphragmatic hernia
epidemiology of, 354 proximal, 1364 outcome and, 815
evaluation of, 354–355, 355f, 356f Splinting small, congenital, 1039
management of after burn injury, 384 trauma to, 305–308
basic concepts for, 343–344 fracture, 331, 332 tumors of, 483
early, 355f, 359 Spondylothoracic dysplasia, 807, 808f volvulus of, 1036–1038, 1037f, 1037t

Volume 1, pages 1–768; Volume 2, pages 769–1738

I-42 INDEX

Stomatitis, aphthous, in ulcerative colitis, 1219 Sublingual gland, 729 T

Stool Submandibular ducts, 716 T cell(s)
acholic, in biliary atresia, 1323 Submandibular gland cytotoxic, 147
bloody-appearing, evaluation of, 1147–1148 anatomy and physiology of, 729 helper, 147–148
color of, after portoenterostomy, 1327 resection of, 735 in host defense, 147–148
red currant jelly, in intussusception, 1095, 1096f, Submucous cleft, 701 in neonate, 151
1099, 1147, 1149f Subperiosteal abscess, 1694–1695, 1694f T-cell lymphoma, 485, 523t, 524f, 525
Stool antigen test, in Helicobacter pylori infection, Succinylcholine, 209–210, 210t TAC (tetracaine, adrenaline, cocaine), 221, 221t
1032 Sucralfate, for peptic ulcer disease, 1033 Tachyarrhythmias, in neonate, 138–139, 139t
Stool softening agents, for anal fissure, 1317 Sugiura procedure, 1366 Tacrolimus, in transplant patient, 608, 609f
Stooling patterns, after ileoanal pull-through, 1226, Sulfisoxazole, for urinary tract infection, 1432 heart, 665, 667t
1227f Sulindac, for familial adenomatous polyposis, 488, liver, 649, 650t
Straddle injury, genital trauma from, 324 1182 lung, 676–677, 676t
Stratum basale, 1711 Supracondylar fracture, 364–365 pancreas, 636, 636f
Streak gonads, 1571, 1574 Supraglottic stenosis, in laryngomalacia, 841 renal, 624
Streptococcal infection Supraglottitis, 725, 726 Takayasu disease
lymphadenitis in, 740 Supraglottoplasty, for laryngomalacia, 840, abdominal aortic aneurysm in, 1635–1636
oropharyngeal, 717 841f cerebrovascular disease in, 1643–1644, 1643f
perianal dermatitis in, 1318, 1318f Suprapubic catheter Talipes equinovarus, 1704–1705, 1705f
Streptococcus pneumoniae, 855–856, 856f in urethral injury, 323 Tamsulosin, for dysfunctional elimination
Streptokinase, for renal vein thrombosis, 1439–1440 in urinary tract infection, 1428–1429, 1429t syndromes, 1463–1464
Stress gastritis, 1149 Suprapubic cystostomy, for urethral injury, 323 Tanner stages of breast development, 771, 772t
Stress response Supraventricular tachycardia, in neonate, 138–139, Target sign, in intussusception, 1100, 1100f
in neonate, 103–107 139t Taurine, requirements for, 181–182
surgery and, 103–107, 104f, 106f SUR1/Kir6.2 complex, in hyperinsulinism, 1379 Taxanes, 407t
in trauma patient, 270 Surface manipulation, nanoelectromechanical Technetium 99m renal scintigraphy, in ureteropelvic
Stress ulcers, 1029–1030, 1030t, 1031, 1032, systems for, 62 junction obstruction, 1416–1417, 1416f,
1034–1035 Surface rendering, in virtual reality, 68–69, 69f 1417f, 1418f
Stretta procedure, 57, 957 Surfactant Technetium-labeled sucralfate radioisotope
Strictureplasty, in Crohn disease, 1213–1214, 1213f, for congenital diaphragmatic hernia, 818 scanning, after caustic ingestion, 921–922,
1214f, 1215 fetal production of, 111 921f
Stridor, 722–723 Surfactant protein B or C deficiency, lung Technological innovation, 37–80
in airway obstruction, 837 transplantation for, 674 cycle in, 37
expiratory, 837 Surgery in diagnosis, 38–48
inspiratory, 837 minimal access. See Minimal access surgery. in microtechnology and nanotechnology, 60–63
in laryngomalacia, 840 pediatric surgeon-developed, 63–65, 64t
in vocal cord immobility, 842 history of, 1–20. See also History of pediatric in surgical training, 65–75
Stroke, ischemic, 1643–1645, 1643f, 1644f, 1645f surgery. in therapy, 48–60
Stroke volume, in pectus excavatum, 783 stress response to, 103–107, 104f, 106f Tectal glioma, 597
Stromal luteoma, 541 robotic. See Robotic surgery. Tegaderm, in burn care, 378
Stromal sarcoma, ovarian, 547 Surgical-assist devices, 58 Telangiectasia, 1621
Stromal tumors Surgical error, ethics in, 244–245 Teleoperators, 58
gastrointestinal, 484–485 Surgical innovation, 63–65 Telomerase, 399
ovarian device-related, 63 Telomeres, 399
epithelial in pediatric devices, 63–65 Temozolomide, 407t
laboratory tests in, 530–531, 530t procedure-related, 63, 64t Temperature, intraoperative monitoring of, 213
of low malignant potential, 538–539, 538f training in, 74–75 Temporal bone, fracture of, 711–712, 711f
staging of, 534, 535t Surgical lasers, 49 Temporal lobe epilepsy, 1690, 1691
surface, 537–538 Surgical simulation, 65–67 Temporal lobe tumors, 599
sex cord, 530–531, 530t, 531t, 539–541, 539f benefits of, 66–67 Tendon injury, hand, 338–339, 338f, 339f
testicular, 551, 553 future directions in, 67 Teniposide, 407t
Structural measures, in performance analysis, 234 image generation in, 66 Tennison repair, for cleft lip and palate, 702
Struma ovarii, 548 interface in, 66 Tensor tympani muscle, 707
Strut fixation, in pectus excavatum repair, 785, 786f in surgical education, 73–74 Teratogens
Struvite stones, 1437–1438 virtual reality–based, 73, 73f. See also Virtual abdominal wall defects and, 976
Study design, 227–233, 228t reality. neural tube defects and, 1675
case-control studies in, 228–229 visual display systems in, 65–66 Teratoid/rhabdoid tumors, atypical, 600
case reports in, 227–228 Surgical training Teratoma, 507, 508, 508f
cross-sectional studies in, 228 innovative, 65–75 cervicofacial, 516
prospective cohort studies in, 230 in minimal access surgery, 74 cystic
prospective randomized controlled trials in, simulation in, 65–67 benign, 543, 545f
230–232 virtual reality in, 67–73. See also Virtual reality. pancreatic, 1383
retrospective cohort studies in, 229–230 Suruga, K., 16, 16f gastric, 516
Sturge-Weber syndrome, 1621 Surviving sepsis campaign, 154, 155–162 hepatic, 462
Subarachnoid hemorrhage, 347–348, 347f Survivorship bias, 229 immature, 507–508, 544, 545f
Subclavian artery, right, aberrant, 853, 1665–1666, Suture, premature fusion of, 691. mature, 507–508, 543–544, 545f
1668, 1670f See also Craniosynostosis. mediastinal, 830, 831f, 832
Subclavian artery–axillary artery occlusion, 1642, Sweat test, for cystic fibrosis, 1076 monodermal, 544
1643f Swenson, O., 5, 6f ovarian, 543–544, 545f
Subclavian vein, cannulation of, 266–267 Swenson procedure sacrococcygeal, 511–514. See also Sacrococcygeal
Subclavian vessel, injury to, 285–286 complications of, 1274, 1275f teratoma.
Subdural empyema, 1693–1694, 1695, 1695f, 1696 for Hirschsprung disease, 1269–1270, 1269f testicular, 510, 550, 551, 553, 556t
Subdural hematoma, 351, 351f Switzerland, pediatric surgery in, 13–14 Terlipressin, for septic shock, 161
in child abuse, 387, 388f Synchronized intermittent mandatory ventilation, Term infant, definition of, 89
from overshunting, 1685–1686 118 Terminal saccules, 111
Subdural hemorrhage, 347–348 Syndactyly, 1720, 1722, 1723f Tessier classification of craniofacial clefts, 695–696,
Subglottic hemangioma, 725, 725f, 849–850, 849f, in Poland syndrome, 1720 696f
1613–1614 Synovial sarcoma, 502, 502f, 503 Testicular tumors, 549–557
Subglottic space, 837–838 Syringoceles, 1559 chemotherapy for, 556, 556t
Subglottic stenosis, 844–849, 845f, 845t Systematic reviews, 232–233 clinical presentation in, 550
congenital, 724–725, 724f, 844–845, 845f, 845t Systemic inflammatory response syndrome (SIRS) cryptorchidism and, 1008, 1014
endoscopic surgery for, 846–847, 846f diagnosis of, 152, 153t diagnosis of, 550
Myer-Cotton grading of, 845, 845f pathogenesis of, 144–152, 144f epidemiology of, 549
open surgery for, 847–849, 847f, 847t, 848f in sepsis definition, 141, 142f epithelial, 550–551

Volume 1, pages 1–768; Volume 2, pages 769–1738

 INDEX I-43

Testicular tumors (Continued) Thoracic duct Thrombosis (Continued)

germ cell, 509–510, 510f, 510t, 551–552, 556, anatomy of, 877, 877f renal vein, 624, 1439–1440
556t trauma to, 286 shunt, 1363–1364, 1366
gynecomastia and, 1718–1719 Thoracic ectopia cordis, 800–803, 803f venous, after ileoanal pouch procedure,
histologic classification of, 550, 551t Thoracic kidney, 1406, 1407f 1228–1229
markers of, 550 Thoracic spine, injury to, 335, 354, 356t, 357f, 358 Thrombotic disorders, 174–175
in Peutz-Jeghers syndrome, 1184 Thoracic trauma, 271–290 Thumb
primary, 550–552 to aorta, 282–286, 283t, 284f, 285f clasped, 1722–1723
rhabdomyosarcomas as, 552 asphyxia in, 286, 286f defects of, 1722
risk factors for, 549–550 birth-related, 392 radioclinodactyly of, 1722
secondary, 552–553 to chest wall, 275 trigger, 1722–1723
staging of, 550, 551t classification of, 271 Thymic cysts
stromal, 551, 553 clinical presentation in, 272 cervical, 760–761
surgical management of, 553–556, 553f complications of, 287 mediastinal, 830–832, 831f, 832f
radical inguinal orchiectomy and damage control in, 274–275 Thymoglobulin, in transplant patient, 612–613
retroperitoneal lymph node dissection in, diagnosis and initial resuscitation in, 272–274, heart, 665–666, 667, 667t
554–556, 555f 273t liver, 650t
testis-sparing surgery in, 553–554, 554f to diaphragm, 279, 280f, 280t lung, 676–677, 676t
teratoma as, 510 epidemiology of, 271–272, 272t renal, 624, 625
Testicular vessels, ligation of, for varicocele, 1018, to esophagus, 279 Thymoma, 831
1018f to heart, 280–282, 281f, 282f, 286 Thymus
Testis(es). See also Scrotum. hemothorax in, 276–277 development of, 825
appendix, 1014–1015 life-threatening, 271 ectopic tissue of, 831–832
ascending, 1006 to lung, 277, 278f embryology of, 755f
atrophy of mortality in, 271, 287 hyperplasia of, 831–832
after inguinal hernia repair, 998 outcome of, 287 Thyroglossal duct cyst, 721, 755–756, 755f, 756f
after orchidopexy, 1013 penetrating, 271–272, 286 Thyroid carcinoma
from varicocele, 1017 pneumothorax in, 275–277, 276f medullary, 750
descent of, 986, 1003–1005, 1004f prevention of, 272 metastatic, cervical lymphadenopathy in, 743
differentiation of, 1565–1567 with rib fracture, 272, 275 well-differentiated, 748–750
epididymis connection with, in cryptorchidism, thoracoabdominal, 286–287 Thyroid cartilage, 722
1008 to trachea and bronchi, 277–279, 279f Thyroid disorders
fetal, 1567 transmediastinal, 287 neoplastic, 748–750
metastasis to, 553 treatment of, 274–287 non-neoplastic, 746–748
in prune-belly syndrome, 1509 vascular, to torso, 363–364 Thyroid gland
retractile, 1006 Thoracoabdominal bypass, for abdominal aortic abnormalities of, after radiation therapy for
steroid biosynthetic enzyme nomenclature related coarctation, 1633–1634, 1633f Hodgkin lymphoma, 522
to, 1569t Thoracoabdominal ectopia cordis, 803–804, 804f ectopic, 721, 745, 746f, 756
steroid hormone synthesis in, 1570f Thoracoabdominal injury embryology of, 745, 746f, 753, 755f
temperature of, 1006–1008, 1007f in birth trauma, 392–393, 392f enlargement of, 746–747. See also Goiter.
torsion of, 1014–1016, 1014f, 1015t in child abuse, 390–391, 390f, 391f evaluation of, 745–746, 746f
in cryptorchidism, 1008 traumatic, 286–287 physiology of, 745
undescended. See Cryptorchidism. Thoracoscopy Thyroid hormones
vanishing, 1009 with biopsy, 420–422, 421f abnormal levels of. See Hyperthyroidism;
Testosterone. See also Androgen. for chylothorax, 879 Hypothyroidism.
for 46,XY DSD, 1575 for esophageal atresia with distal fistula, 903, 904f synthesis of, 745
in burn injury, 381 Thoracostomy, tube. See Chest tube. Thyroid nodules
deficiency of Thoracotomy cold, 759
cryptorchidism and, 1007 for chylothorax, 878–879 fine-needle aspiration biopsy of, 418
in disorders of sex development, 1568t, 1570, emergency, 273, 273t incidence of, 745
1570f, 1573 for hemothorax, 277 management of, 748, 748t
malignancy risk in, 508 for lung biopsy, 875–876, 876f after radiation therapy for Hodgkin lymphoma, 522
hypospadias and, 1536 for patent ductus arteriosus, 1648–1649, 1648f Thyroidectomy
ovarian tumors and, 531t prior, lung transplantation and, 673 complications of, 749
in Sertoli-Leydig cell tumors, 540–541 right, in vascular ring repair, 1670 for Graves disease, 747–748
in sexual differentiation, 1567 transaxillary, for pneumothorax, 873 prophylactic, 750, 750t
stimulation with, hypospadias repair and, 1552 THPVS technique, in hyperinsulinism, 1380 for thyroid cancer, 749, 750
Tetralogy of Fallot, 139, 1659–1660, 1659f, 1661f, Three-dimensional visualization, in virtual reality, Thyroiditis, 747, 747t
1662f 70, 70f Thyrotropin, 745
TH1 cytokines, 149 Thrombasthenia, Glanzmann, 171 Thyrotropin-releasing hormone, 745
TH2 cytokines, 149 Thrombocytopenia, 169–171 Thyroxine
Thalamus, tumors of, 593–594 acquired, 169–170 deficiency of, 747
Thalassemia cutaneovisceral angiomatosis with, 1620 synthesis of, 745
b-, 168 dilutional, in trauma patient, 269 Tidal volume, 113, 113f
cholelithiasis in, 1342 genetic, 169 Time constants, pulmonary, 114
splenectomy for, 1387 heparin-induced, 170 Tissue. See also Soft tissue.
Thecoma, ovarian, 540 in Kaposiform hemangioendothelioma, 1619 composition of, aberrations of, 1712–1713
Thelarche in necrotizing enterocolitis, 1196 regional absence of, 1712, 1713f
normal, 771, 772t platelet transfusion for, 169–170, 178 synthesis of, energy cost of, 97
premature, 771 in portal hypertension, 1360 typing of, for transplantation, 614–615, 615f
Therapeutic technological innovations, 48–60 Thrombocytopenic purpura, immune, 170, 1386, viability of, positron emission tomography
Thermogenesis, nonshivering, 98–99 1387 of, 46
Thermoregulation, in neonate, 98–99 Thromboembolism, venous, 175, 359–360 Tissue ablative instruments, 48–50
Thiamine, 183 Thrombolysis Tissue engineering, 27–39
in parenteral nutrition, 189–190, 190t for empyema, 872 cardiac, 30–31
Thiopental, 202f, 212, 212f for renal vein thrombosis, 1439–1440 cartilage and bone, 29–31, 30f
Thiopurines, for Crohn disease, 1212 for vasospasm, 366–367 existing products of, 34t
Third-space loss, 206 Thrombosis future directions in, 33–35, 34f
Thoracentesis, for chylothorax, 878 abdominal aortic, 1636 gastrointestinal, 32
Thoracic and thoracoabdominal duplications, 1156, hepatic artery, 649 hepatic, 33
1158–1159, 1159f in liver transplantation, 649 interdisciplinary approach to, 27–29, 28f
Thoracic deformities, in diffuse skeletal disorders, portal vein. See Portal vein, thrombosis of. vascular, 31–32, 31f
805–808, 807f, 808f. See also Chest wall, renal artery, 1639 Tissue expansion, 1712
congenital deformities of. in renal transplantation, 627 for vaginal agenesis, 1595–1596

Volume 1, pages 1–768; Volume 2, pages 769–1738

I-44 INDEX

Tissue plasminogen activator Tracheoesophageal fistula (Continued) Trauma (Continued)

for empyema, 872 laryngeal cleft with, 850 asphyxia in, 286, 286f
for vasospasm, 366–367 operative repair of birth. See Birth injuries.
TNM staging, of rhabdomyosarcoma, 493, 493f with distal fistula, 899–905, 901f, 902f, 903f, bite-related, 340–341
Tobacco exposure, ulcerative colitis and, 1218 904f blunt, extracorporeal life support after, 131
Tobramycin, for Pseudomonas infection, 865 H-type, 909–910, 910f breast, 777
Tocolytics, 80–81 with upper pouch fistula, 910–911, 911f cardiac. See Heart, trauma to; Pericardium, trauma to.
Toll-like receptors, in necrotizing enterocolitis, 1194 preoperative treatment of, 899 central nervous system, 343–364
Tolterodine, for overactive bladder syndrome, 1464 tracheomalacia from, 851–852 diaphragmatic, 279, 280f, 280t, 308–309
Tongue, enlarged, 720 Tracheomalacia, 851–852, 851f, 913–914, 914f ear, 711–712, 711f
Tongue-tie, 720, 720f Tracheoplasty genitourinary, 311–329. See also Genitourinary
Tonsillar hypertrophy, 719, 719f anterior, 852 trauma.
Tonsillectomy, 720 slide, 852–853, 853f hand, 337–340
torticollis after, 765, 765f Tracheotomy, 838–840 head. See Brain injury, traumatic; Head injury;
Tonsillitis complications of, 839–840 Skull fracture.
acute, 716–717 emergent, 723 historical perspective on, 9
chronic, 717 for hemangioma, 849 musculoskeletal, 327–337.
localized extension of, 717–718 indications for, 838 See also Musculoskeletal trauma.
recurrent, 717 for laryngomalacia, 840 nasal, 715
Topical anesthesia, 221, 221t for laryngotracheal stenosis, 846 oral, 717
Topoisomerases, 406, 407t for recurrent respiratory papillomatosis, 844 perineal, 308
Topotecan, 407t technique of, 837–838, 838f, 839f resuscitation after, 262–263. See also Emergency
Tornwaldt cyst, 721 for vocal cord paralysis, 842–843 management.
Torso, vascular injuries to, 361 Training soft tissue, 339, 340
Torticollis, 391, 763–769 in pediatric surgery, 6–9 spinal. See Spinal cord injury; Spine, trauma to.
clinical features of, 763–765, 764f, 765f in surgical innovation, 74–75 stress response to, 103–107, 104f, 106f, 270
conservative management of, 766–767 Transabdominal manipulation, in intussusception thoracic, 271–290. See also Thoracic trauma.
differential diagnosis of, 764–765, 764t, 765f reduction, 1105 vascular. See Vascular trauma.
etiology of, 763, 764t Transannular patch, for tetralogy of Fallot, 1660, Treacher Collins syndrome, 697, 697f
historical perspective on, 763 1662f Treatment
pathology of, 763 Transatrial repair of tetralogy of Fallot, 1660, 1661f informed consent and assent for, 238–239, 239t
secondary effects of, 765–766, 765f, 765t, 766f Transcutaneous monitoring of gas tension, 116 rating of, 227, 228t
surgical management of, 767, 767f TransCyte, in burn care, 378 Triamcinolone
Torus fracture, 327, 328f Transfusion therapy, 175–177 for esophageal caustic injury, 923
Total body water (TBW), in neonates, 91–92 for acute hemorrhage, 167 for infantile hemangioma, 1616
Total lung capacity (TLC), 112, 113f in cancer or immunodeficient patient, 176 Triangular flap technique, 702
in pectus excavatum, 781–782 complications of, 176–177 Triglycerides
Total parenteral nutrition. See Parenteral nutrition. intraoperative, 206–207, 206t elevation in, with parenteral nutrition, 192
Tourniquet, in hypospadias repair, 1551 massive long-chain, in neonate undergoing surgery, 106
Toxic goiter, 747–748, 748t bleeding after, 174 medium-chain
Toxic megacolon intraoperative, 207 for chylous ascites, 1174–1175
in Crohn disease, 1213 risks of, 175–176 in neonate undergoing surgery, 106
in ulcerative colitis, 1218 in trauma patient, 269 in premature infant formulas, 187
Toxins, bacterial, 150 platelet. See Platelets, transfusion of. in neonate undergoing surgery, 106
Toxoplasmosis, cervical lymphadenopathy in, 743 reactions to, 176–177 Trigonocephaly, 692
TP53 gene red blood cell. See Erythrocyte(s), transfusion of. Triiodothyronine, synthesis of, 745
inactivating mutations of, 401–402 for sepsis, 158 Trilogy system, 52, 53f
in Wilms’ tumor, 426 in sickle cell disease, 168 Trimethoprim-sulfamethoxazole
Trace elements toxicity of, 176 prophylactic, in heart transplantation, 666–667
in parenteral nutrition, 184t, 190 Transitional circulation, 112, 135 for urinary tract infection, 1432
requirements for, 184, 184t Transjugular hepatic venography, in portal Trisomy 21. See Down syndrome.
Trachea. See also Airway. hypertension, 1361 TRK gene, in neuroblastoma, 449
anatomy of, 837–838 Transjugular intrahepatic portosystemic shunts TRKA receptors, 410
compression of, by innominate artery, 851, 851f, (TIPS), 1363–1364 Trunk
853, 854 Transjugular portal venography, retrograde, in portal lipomatous mass on, in Cloves syndrome,
hemangioma of, 849–850 hypertension, 1361, 1361f 1629–1630, 1630f
lesions of, 851–853 Transmediastinal injuries, 287 rhabdomyosarcoma of, 496–497
trauma to, 277–279 Transplantation. See also specific cells and organs. Trypsin, stool, in meconium ileus, 1077
Tracheal occlusion future prospects for, 615 TSC gene, in tuberous sclerosis, 1399
fetal historical perspective on, 8, 605–613 Tube feeding. See Enteral nutrition.
for congenital diaphragmatic hernia, 817, 823 from 1953 to 1968, 605–607, 606f, 606t, 607f, Tube thoracostomy. See Chest tube.
endoscopic, 85 608f Tuberculin (PPD) skin testing, 857, 863
lung growth and, 112 from 1969 to 1979, 607–608, 608f Tuberculosis, 728
percutaneous fetoscopic, 86 from 1980 to 1991, 608–609, 609f adenitis in, 742
Tracheal pouch, 852 from 1992 to present, 609–613, 609f, 610f, pulmonary, 857, 863
Tracheal rings, 852–853, 852f, 853f 611f, 612f Tuberous sclerosis
Tracheal stenosis, congenital, 852–853, 852f, 853f therapeutic implications of, 611–613, 612f, 613f abdominal aortic aneurysm with, 1635, 1635f
Tracheitis, bacterial, 726 lymphoproliferative disorders after, 525, 526–527 epilepsy surgery in, 1688–1689
Tracheobronchial anomalies, in congenital organ preservation for, 613–614, 614f renal cysts in, 1399
diaphragmatic hernia, 810 organ procurement for, 613, 663–664, 675–676 Tularemia, cervical lymphadenopathy in, 743
Tracheobronchial remnants, congenital esophageal principles of, 603–619 Tumor(s). See also Cancer; specific organ or tumor
stenosis from, 915 tissue typing for, 614–615, 615f type.
Tracheobronchial vascular compression, 853–854 Transport in horseshoe kidney, 1408
Tracheobronchomalacia, 724 in spine trauma, 344 in Meckel diverticulum, 1091
Tracheoesophageal fistula. See also Esophageal atresia. in traumatic brain injury, 344 nanoelectromechanical identification of, 62–63, 62f
associated anomalies with, 896–897, 897t Transtracheal ventilation, for upper airway positron emission tomography of, 46
classification of, 894, 895, 895f, 895t obstruction, 723 Tumor markers
diagnosis of, 898–899, 899f, 900f TRAP sequence (twin reversed arterial perfusion in liver tumors, 464–465
embryology of, 895–896 sequence), 88 in ovarian tumors, 530, 530t, 532
epidemiology of, 896 Trastuzumab, 410 Tumor necrosis factor, 148–149
H-type, 895f Trauma in necrotizing enterocolitis, 1190t, 1192
diagnosis of, 898, 900f abdominal. See Abdominal trauma. in stress response, 104
operative repair of, 909–910, 910f anorectal, 308, 1153, 1153f Tumor suppressor genes, 399, 400t
historical background on, 893–895, 894f aortic, 282–286, 283t, 284f, 285f inactivation of, 401–402

Volume 1, pages 1–768; Volume 2, pages 769–1738


Tumorigenesis, 399–400, 400t, 401f
Tunica albuginea, 1537
Turbinates, 712
Turcot syndrome, 487, 1182
Turkey, pediatric surgery in, 15
Twin reversed arterial perfusion sequence (TRAP
sequence), 88
Twin-twin transfusion syndrome, fetal interventions
for, 87
Twinning, partial or abortive, in alimentary tract
duplications, 1155
Twins, conjoined. See Conjoined twins.
Two-hit mechanism of carcinogenesis, 399
Tympanic membrane, 707
perforation of, 711
Tympanometry, 708
Tympanostomy tube, 709
Tyrosinemia, hepatocellular carcinoma and, 476
U in congenital diaphragmatic hernia, 813–814,
Ulcer(s). See also Peptic ulcer disease.
acute upper gastrointestinal bleeding from,
1150–1151
aphthous, in Crohn disease, 1210
cutaneous, hemangioma with, 1616
rectal, solitary, 1319
Ulcerative colitis, 1217–1234
clinical examination in, 1219–1221, 1220f, 1221f,
1221t
clinical manifestations of, 1219, 1220f
colorectal cancer in, 489, 1219
epidemiology of, 1218
etiology of, 1218
exacerbations of, 1222
medical management of, 1221–1222
pathology of, 1218–1219, 1218f
postoperative care in, 1226
surgical management of, 1222–1224
complications and outcomes of, 1227–1229
historical perspective on, 1217
ileoanal pouch procedure in, 1223–1224,
1223f, 1224f, 1225f, 1226f
ileoanal pull-through in, 1223
J pouch in, 1224, 1225, 1226–1227, 1227f
laparoscopic, 1225–1226, 1226f
protective ileostomy in, 1226–1227
stooling after, 1226–1227, 1227f
straight pull-through in, 1223, 1225–1227,
1225f, 1227f
Ulnar defects, 1722
Ulnar nerve, injury to, 337–338, 338f
Ultimobranchial body, 755f
Ultrasonography, 38–40
in alimentary tract duplications, 1157, 1157f
in anorectal malformations, 1294
in appendicitis, 1257–1258
in ascites, 1172, 1173f
in biliary atresia, 1323–1324
in bladder dysfunction, 1454, 1455f
in cervical lymphadenopathy, 739
before cholecystectomy, 1344–1345, 1345t
in choledochal cyst, 1334
in conjoined twins, 1731
contrast-enhanced, 40, 40f
in developmental dysplasia of hip, 1700, 1700f
in disorders of sex development, 1575, 1576f
Doppler, 38
of burns, 372
in cervical lymphadenopathy, 739
after renal transplantation, 623
of salivary glands, 730, 730f
in ectopic ureter, 1446
FAST (focused abdominal sonography for trauma),
290, 290f, 308, 313
fetal surgery and, 40
in gallbladder disease, 1343
harmonic, 40, 41f
in hepatic abscess, 1350–1351, 1350f
in hypertrophic pyloric stenosis, 1023, 1023f
in inguinal hernia, 987–988, 988f
in intestinal rotation and fixation disorders, 1118,
1119f
Ultrasonography (Continued)
in intussusception, 1100–1101, 1100f, 1101f,
1103, 1106
in mesenteric and omental cysts, 1168, 1168f
in multicystic dysplastic kidney, 1400–1401,
1400f, 1401f
in musculoskeletal trauma, 331–332
of neck mass, 727
in necrotizing enterocolitis, 1199
of ovarian tumors, 532–533, 532f
in pancreatitis, 1373, 1375
percutaneous needle biopsy guided by, 418
in pheochromocytoma, 559–560
in portal hypertension, 1361
prenatal
of abdominal wall defects, 977–978
in alimentary tract duplications, 1157
in bladder exstrophy, 1517–1518
in bronchopulmonary sequestration, 827–828,
827f
in choledochal cyst, 1333
814f
in congenital lobar emphysema, 828–829
in conjoined twins, 1730–1731, 1731f
diagnostic, 78, 79f
in duodenal atresia and stenosis, 1053, 1053f
in jejunoileal atresia and stenosis, 1061
in megacystis-microcolon-intestinal
hypoperistalsis syndrome, 1285
of ovarian tumors, 532, 532f
in ureteropelvic junction obstruction,
1413–1414, 1413f
in renal injury, 313
in renal vein thrombosis, 1439
of simple renal cyst, 1403, 1403f
of testicular tumors, 550
therapeutic use of, 49
in thoracic trauma, 273–274
three-dimensional, 38–39, 39f
of thyroid gland, 746
of thyroid nodules, 748
in ureterocele, 1448, 1448f
in ureteropelvic junction obstruction, 1415–1416,
1415f, 1420–1421
in urinary tract infection, 1429–1430, 1429f, 1430f
in vascular trauma, 363
of Wilms’ tumor, 427
Umbilical artery
cannulation of, 970–971
catheterization of, 1634–1635, 1635f
embryology of, 962f, 963t
single, 967
Umbilical hernia, 963, 968–970
anatomy of, 968–969
bladder exstrophy with, 1516, 1516f
description of, 974–975, 974f
embryogenesis of, 976
giant, 971
incidence of, 969
natural history of, 969
surgical management of, 969–970, 970f
treatment of, 982
Umbilical-placental circulation, 134–135, 136f
Umbilical vein, 1355, 1356f
cannulation of, 970–971
embryology of, 962f, 963t
Umbilicoplasty
for abdominal wall defects, 971–972, 972f
for giant umbilical hernia with redundant skin,
971
Umbilicus, 959–974
acquired abnormalities of, 964, 968t
appearance of, aesthetics of, 961
at birth, 963–964
catheterization of, 116–117
congenital malformations of, 961, 964–968, 965f,
966f
dysmorphology of, 967
embryology of, 961–963, 962f, 963t
granuloma of, 964
infection of, 964
lint in, 968
new, creation of, 971–972, 972f
Volume 1, pages 1–768; Volume 2, pages 769–1738
INDEX I-45
Umbilicus (Continued)
piercing of, 967–968
protrusions at, 967
reconstruction and preservation of, 971–972
use of, 970–971
Undiversion, 1487
Unicameral bone cyst, 578–579, 579f
injection therapy for, 584
location of, in relation to physis, 579f
Unicornuate system, 1603–1604, 1603f, 1604f
United Kingdom, pediatric surgery in, 10–11, 11f, 12f
United Network for Organ Sharing, 620
United States, pediatric surgery in, 4–6, 4f,
5f, 6f
Upper gastrointestinal series
in duodenal atresia and stenosis, 1054, 1054f
in gastroesophageal reflux disease, 952
in hypertrophic pyloric stenosis, 1023, 1023f
in intestinal rotation and fixation disorders, 1120,
1120f
in necrotizing enterocolitis, 1199
Urachal remnants, 966–967, 966f
Urachus
embryology of, 961–963, 962f, 963t
patent, 966, 966f
tumors arising from, 967, 967t
Ureter. See also Megaureter.
blind-ending, 1443
ectopic, 1445–1447, 1446f
megaureter with, 1497
inverted Y, 1443
for Mitrofanoff neourethra, 1480
reimplantation of, for megaureter, 1499–1501,
1500f, 1502–1503, 1504f
stones in, 1438. See also Urolithiasis.
trauma to, 313, 314t, 319–320
Wilms’ tumor extension to, 431–432
Ureteral anastomosis, in renal transplantation,
622–624
Ureteral buds, 1405
Ureteral duplication, 1441–1454
embryogenesis of, 1441–1444, 1442f
types of, 1441
Ureteral stent
after pyeloplasty, 1425
for urolithiasis, 1437
Ureterectomy, partial nephrectomy with, 1444,
1445f, 1447
Ureteric bud, 1395, 1411–1412
Uretero-ureteral reflux (yo-yo reflux), 1444
Uretero-ureterostomy, 1444, 1445, 1447
Ureterocele, 1447–1451
classification of, 1447–1448, 1447t
clinical presentation in, 1448
definition of, 1441
diagnosis of, 1448–1450, 1448f, 1449f
imaging of, 1429–1430, 1430f
megaureter with, 1497
prolapse of, 1448, 1449f, 1558
prolapsed ectopic, 1607, 1608f
treatment of, 1450
vesicoureteral reflux and, 1450, 1451
Ureterocystoplasty, 1475, 1476f, 1477f
Ureteroneocystostomy, 1445
in bladder exstrophy repair, 1522–1523, 1524f
Ureteropelvic junction, injury to, 319
Ureteropelvic junction obstruction, 1411–1429.
See also Hydronephrosis.
clinical features of, 1412–1414
antenatal, 1413–1414, 1413f, 1413t
postnatal, 1414, 1414t
crossing vessels in, 1420
diagnosis of, 1414–1420
biochemical markers in, 1419–1420
intravenous urography in, 1414
magnetic resonance urography in, 1417–1418,
1418f
pressure-flow study (Whitaker test) in, 1419,
1419f
retrograde pyelography in, 1418, 1419f
scintigraphy in, 1416–1417, 1416f,
1417f, 1418f
ultrasonography in, 1415–1416, 1415f
voiding cystourethrography in, 1417
I-46 INDEX
Ureteropelvic junction obstruction (Continued)
differential diagnosis of, 1414, 1414t
in duplex collecting system, 1447
embryogenesis of, 1411–1412
etiology of, 1411, 1412f
incidence of, 1411
management of, 1420–1425, 1421f
conservative, 1420–1421
minimally invasive surgery for, 1424–1425
open surgery for, 1421–1423, 1422f, 1424f
outcome of, 1425
in prune-belly syndrome, 1506–1507
Ureteropyelostomy, 1447
Ureteroscopic stone extraction, 1438
Ureterosigmoidostomy, colorectal cancer after, 489
Ureterostomy, cutaneous, 1488–1489, 1488f, 1489f,
1498
Urethra
atresia of, 1557
cysts of, 1558
development of, 1531–1532, 1533f, 1535f,
1538f
diverticulum of
in boys, 1557
in girls, 1557
after hypospadias repair, 1553
duplication of, 1469, 1469f, 1559–1560, 1559f,
1560f
fistula of, congenital, 1560
lengthening of, for bladder neck reconstruction,
1477–1478, 1478f, 1479f
mass in, in girls, 1558
obstruction of. See also Urethral valves.
fetal interventions for, 82–83
polyps of, 1558, 1559, 1559f
prolapse of, 1558, 1606, 1607f
in prune-belly syndrome, 1507–1508, 1510f,
1511f
rhabdomyosarcoma of, 1558
stenosis of, 1561–1562
in girls, 1558
after hypospadias repair, 1552
stricture of, 1557
surgical closure of, 1478–1479
suspensory ligament of, 1303
trauma to, 312, 313–314, 322–324, 322f, 1557
in females, 323, 324
grading of, 314t
Urethral catheter, for bladder injury, 321–322
Urethral disorders
anatomic, 1468–1469, 1469f
combined anatomic and neurogenic, 1470, 1471f,
1472f
neurogenic, 1469–1470, 1470f
Urethral diversion, after hypospadias repair,
1551–1552
Urethral mobilization, in hypospadias repair, 1542
Urethral plate
in hypospadias, 1539, 1539f
preservation of, in hypospadias repair, 1543,
1544f, 1545
Urethral spongiosum, 1537
Urethral valves
anterior, 1557
posterior, 1430, 1431f, 1468–1469, 1469f,
1555–1557, 1556f
megaureter with, 1497, 1498
ultrasonography in, 1455f
voiding cystourethrography in, 1454
in prune-belly syndrome, 1507
Urethrocutaneous fistula
after bladder exstrophy repair, 1523
after hypospadias repair, 1552, 1552f
Urethrography, retrograde, in trauma, 313–314
Urethroplasty, tubularized plate, 1543, 1543f, 1544f
Uric acid stones, 1437–1438
Urinalysis
in bladder dysfunction, 1453
in genitourinary trauma, 312
in urinary tract infection, 1428–1429, 1429t
Urinary ascites, 1175
Urinary catheterization
for bladder injury, 321
clean intermittent
Urinary catheterization (Continued) Urostomy, 1237–1238
for neuropathic bladder, 1459, 1459f, 1460, 1461 Ursodeoxycholic acid, in intestinal failure–associated
for posterior urethral valves, 1462 liver disease, 1139
for posterior urethral valves, 1469 Uterine horns, rudimentary, 1603–1604, 1603f,
in trauma patient, 268 1604f
for urine culture, 1428–1429, 1429t Uterus
Urinary diversion, 1487–1499. See also Bladder duplication of, 1602, 1602f, 1603f
augmentation or replacement. hemi-, obstructed, 1603–1604, 1604f
in bladder exstrophy, 1523 rhabdomyosarcoma of, 498
complications of, 1495 Utricle, prostatic, 1559
continent, 1490–1492, 1492f, 1493f, 1494f, 1495f Uveitis, in ulcerative colitis, 1219
after hypospadias repair, 1551–1552
incontinent, 1487–1490, 1488f, 1489f
indications for, 1487 V
intestinal conduits for, 1489–1490, 1489f VAC (vacuum-assisted closure), in trauma patient,
umbilicus as exit site for, 971 270
Urinary incontinence. See Incontinence. Vaccine
Urinary reservoirs, continent, 1494, 1495f Haemophilus, 856
Urinary sphincter, artificial, 1478, 1480f human papilloma virus, 844
Urinary tract. See also Genitourinary entries. pneumococcal, 855–856
anomalies of, with vaginal agenesis, 1592 rotavirus, intussusception and, 1097
development of, 1405, 1406f Vagina
obstruction of. See also specific types, e.g., Bladder adenocarcinoma of, 1609
outlet obstruction. agenesis of, 1587, 1592–1599
fetal interventions for, 82–83 diagnosis of, 1592–1593
after renal transplantation, 624 external genitalia of, 1592, 1592f
reconstruction of. See Bladder augmentation or lower, 1600–1601, 1601f
replacement; Urinary diversion. reproductive issues in, 1598–1599
Urinary tract infection, 1427–1433. treatment of, 1593–1598
See also Pyelonephritis. bowel vaginoplasty for, 1596–1598, 1597f
after bladder augmentation or replacement, 1496 nonoperative, 1593–1594, 1594f
clinical presentation in, 1428 operative, 1587, 1589f, 1594–1596, 1595f,
diagnosis of, 1428–1429, 1429t 1596f
ectopic ureter and, 1445 in congenital adrenal hyperplasia, 1569–1570,
imaging of, 1429–1431, 1429f, 1430f, 1431f 1573
pathogenesis of, 1427, 1428t development of, 1591
posterior urethral valves and, 1430, 1431f, duplication of, 1602, 1602f
1556–1557 germ cell tumors of, 516
recurrent, 1433 hemangioma of, 1609
risk factors for, 1427–1428, 1428t introital cysts of, 1608
treatment of, 1431–1433, 1432f, 1432t introital masses of, 1606
ureterocele and, 1448 obstruction of, congenital, 1599–1600, 1599f,
in ureteropelvic junction obstruction, 1414 1600f
vesicoureteral reflux and, 1428, 1429–1430, rhabdomyosarcoma of, 498, 1607, 1608f
1431f, 1433 stricture of, after anorectoplasty, 1307
Urine trauma to, 308
fetal production of, 1413 yolk sac tumors of, 1607–1608
osmolality of, in neonate, 93 Vaginal dilators, for vaginal agenesis, 1593–1594,
Urine culture, 1428–1429, 1429t 1594f
Urinoma, after renal trauma, 318 Vaginal replacement, for cloaca, 1304–1305, 1305f,
Urodynamic evaluation 1306f, 1307f
in cerebral palsy, 1461 Vaginal septum
in dysfunctional elimination syndromes, 1462 longitudinal, 1603–1604, 1603f
in myelodysplasia, 1459 transverse, 1601–1602, 1602f
in posterior urethral valves, 1462 Vaginal switch maneuver, 1304, 1305f
in spinal cord tethering, 1460 Vaginoplasty, 1594–1596, 1595f, 1596f
Urogenital. See also Genitourinary entries. in female gender assignment surgery, 1578–1579,
Urogenital mobilization, total, for cloaca, 1580–1582, 1580f, 1581f, 1582f, 1583f
1302–1303, 1302f, 1303f flap, low-confluence, 1580, 1580f
Urogenital sinus ideal procedure for, 1593
anomalies of, 1470, 1575, 1576f, 1604, 1605f, intestinal, 1596–1598, 1597f
1606f. See also Disorders of sex development pull-through, for mid- and high-level vaginal
(DSD). confluence, 1581–1582, 1583f
with anorectal anomaly. See Cloaca. sigmoid, 1587, 1589f
mobilization of, vaginoplasty using, 1580–1581, timing of, 1593
1581f using urogenital sinus mobilization, 1580–1581,
splitting of, in female gender assignment surgery, 1581f
1581 Vagotomy
Urography for peptic ulcer disease, 1033
intravenous, 1414 for stress ulcers, 1034–1035
magnetic resonance, 1417–1418, 1418f Vagus nerve stimulation, for epilepsy, 1692–1693
Urokinase, for vasospasm, 366–367 Validity of study, 234
Urolithiasis, 1433–1438 Valproic acid
after bladder augmentation or replacement, 1496 coagulopathy from, 173
classification of, 1434, 1436t neural tube defects and, 1675
clinical presentation in, 1434–1437, 1437f Valve bladder, 1468, 1468f
diagnosis of, 1437 Valves
historical perspective on, 1433–1434 cardiac
in horseshoe kidney, 1408–1409 injury to, 281, 281f
incidence of, 1433–1434 tissue-engineered, 30–31
multidetector computed tomography in, 41–42 urethral. See Urethral valves.
recurrent, 1438 Van Nes rotationplasty, 585–586, 586f
spectrum of, 1434 Van Wyk and Grumbach syndrome, 548
treatment of, 1437–1438, 1437t Vancomycin, for necrotizing enterocolitis, 1206
Volume 1, pages 1–768; Volume 2, pages 769–1738

Vanillylmandelic acid, in pheochromocytoma, 559
Varicella-zoster virus infection
in lung cancer patient, 861
in renal transplant patient, 628
Varices
bleeding, 1358–1360
emergency, 1367
intermittent, 1367–1368, 1368f
medical management of, 1359
prophylactic treatment of, 1359
risk of, 1359
sclerotherapy for, 1363
transjugular intrahepatic portosystemic shunts
(TIPS) for, 1363–1364
endoscopy in, 1361
esophageal. See Esophageal varices.
gastric, injection therapy for, 1363
overview of, 1356
shunts for, 1364–1365
ascites after, 1366
emergency, 1367
nonselective, 1364
outcome of, 1367–1368, 1368f
Rex, 1365–1366, 1366f, 1367–1368, 1368f
selective, 1364–1365, 1365f
thrombosis of, 1363–1364, 1366
transjugular intrahepatic portosystemic,
1363–1364
Varicocele, 1016–1019
clinical presentation in, 1016–1017, 1016f
effects of, 1017
etiology of, 1017, 1017f
treatment of
indications for, 1017–1018
operation for, 1018, 1018f
results of, 1019, 1019t
Vas deferens
abnormalities of
cryptorchidism and, 1005
in cystic fibrosis, 1000–1001
injury to, after hernia repair, 997–998
Vascular access
in burn injury, 372–374
for parenteral nutrition, 188–189
in trauma patient, 266–268, 267f
Vascular anomalies, 1611–1633. See also Vascular
malformations; Vascular tumor(s).
classification of, 1613, 1614t
historical perspective on, 1613
placental, jejunoileal atresia and stenosis with,
1060
Vascular endothelial growth factor, in
neuroblastoma, 449
Vascular malformations, 1614t, 1620–1627
anorectal, 1319
arteriovenous, 1625–1627, 1626f, 1626t
capillary, 1620–1621, 1621f
complex-combined, 1627–1630, 1627f, 1628f,
1629f, 1630f
embryogenesis of, 1620
gastrointestinal, 1154
lymphatic, 1621–1624, 1622f, 1623f
oral cavity and pharyngeal, 721
venous, 1624–1625, 1624f, 1625f
Vascular networks, 33–35, 34f
Vascular rings, 853–854, 1665–1671
classification of, 1665, 1665t
management of, 1667–1671
for complete rings, 1667, 1667f, 1668f, 1669f
for incomplete rings, 1667–1668, 1669f, 1670f
results of, 1671
right thoracotomy in, 1670
video-assisted thoracic surgery in, 1670–1671
natural history and diagnosis of, 1665–1666
Vascular system, development of, 1620
Vascular tissue engineering, 31–32, 31f
Vascular trauma, 361–370
in central nervous system injury, 346–347, 353
digital ischemia syndrome in, 367
epidemiology of, 361, 362t
evaluation of, 362–363, 363f
extremity, 361, 364–365, 365t
fractures associated with, 365
hand, 337, 339
Vascular trauma (Continued)
iatrogenic, 365–366
renal, 311, 313, 316–317, 318f
torso, 361, 363–364
vasospasm in, 366–367
Vascular tumor(s), 1613–1620, 1614t
cutaneovisceral angiomatosis with
thrombocytopenia as, 1620
hemangioma as. See Hemangioma.
Kaposiform hemangioendothelioma and
Kasabach-Merritt syndrome as, 1619–1620,
1619f
pyogenic granuloma as, 1618, 1619f
sarcomatous, 1620
Vasculogenesis, 1620
Vasoactive intestinal peptide (VIP), in
neuroblastoma, 443
Vasodilators
for congestive heart failure, 135, 137t
for septic shock, 161
Vasopressin, for septic shock, 161
Vasospasm, traumatic, 366–367
VATER association, 897
VATERL association, 897
Veau-Ward-Kilner pushback technique, 703, 704f
Vechietti vaginoplasty procedure, 1594, 1596
Vecuronium, 210t
Velocardiofacial syndrome, 841, 841f
Velopharyngeal insufficiency, 705–706
Veloplasty, intravelar, 704
Vena cava, inferior, Wilms’ tumor extension to, 431
Veno-occlusive disease, portal hypertension in,
1357–1358
Venoarterial extracorporeal life support, 125, 125f,
127f
Venoarterial-venous extracorporeal life support, 127
Venography, magnetic resonance, in capillary-
lymphaticovenous malformation,
1627–1628, 1628f
Venous access
in burn injury, 372–374
for parenteral nutrition, 188–189
in trauma patient, 266–267, 267f
Venous anastomosis, in renal transplantation, 621,
622
Venous catheter
central
for intraoperative monitoring, 214
venous thromboembolism with, 175
for parenteral nutrition, 188–189
Venous malformation, 1624–1625, 1624f, 1625f
Venous sampling techniques, in hyperinsulinism,
1380
Venous thromboembolism, 175, 359–360
Venous thrombosis, after ileoanal pouch procedure,
1228–1229
Venovenous extracorporeal life support, 125, 125f,
126t
Ventilation
mechanical. See Mechanical ventilation.
minute, 114–115
Ventilation-perfusion matching, 115
Ventilation-perfusion scintigraphy, in inhalation
injury, 375
Ventilatory index, in congenital diaphragmatic
hernia, 816
Ventricle(s)
enlargement of, 1680, 1680f.
See also Hydrocephalus.
fourth, trapping of, after shunt implantation, 1686
hemorrhage within, 347–348, 347f
left
end-diastolic pressure of, cardiac output and,
133–134, 134f
hypertrophy of, in renal transplant patient, 629
ultrasonography of, 40, 40f
single, 1663–1665, 1664f
slit, 1686
Ventricular assist device, as bridge to heart
transplantation, 662
Ventricular septal defect, 1654–1657
cardiovascular management in, 140
classification of, 1654, 1655f
management of, 1655–1657, 1656f
Volume 1, pages 1–768; Volume 2, pages 769–1738
INDEX I-47
Ventricular septal defect (Continued)
natural history and diagnosis of, 1654–1655
results of, 1657
tetralogy of Fallot with, 1659–1660, 1659f
transposition of the great arteries with,
1661–1662, 1663f
Ventricular tachycardia, in neonate, 139
Ventriculo-gallbladder shunt, 1343
Ventriculoatrial shunts, for hydrocephalus, 1683
Ventriculoperitoneal shunts
complications of, 1683–1686
for hydrocephalus, 1677–1678, 1683
inguinal hernia and, 999
peritonitis with, 1233
Ventriculostomy
for brain tumors, 594
endoscopic third, for hydrocephalus, 1686–1687
Verapamil, for supraventricular tachycardia, 138
Versajet hydrosurgery, eschar excision with, 379
Vertebrae. See Spine.
Vertebral artery, dissection and occlusion of, 1645
Vertebral column injuries, 354, 359
Very low birth weight infant
hepatoblastoma in, 466
parenteral nutrition in, 188
Vesicostomy, cutaneous, 1469, 1487–1488, 1488f,
1556
Vesicoureteral reflux
causes of, 1428, 1433, 1433f, 1433t
duplex collecting system and, 1441, 1444–1445,
1444f, 1445f
imaging of, 1429–1430, 1431f
megaureter and, 1497, 1498, 1498f, 1502, 1502f,
1503
in myelodysplasia, 1459
in posterior urethral valves, 1461–1462, 1556–1557
prune-belly syndrome and, 1506–1507, 1511,
1512f
spontaneous resolution of, 1428
treatment of, 1433
endoscopic injection in, 1433, 1435f, 1436f
surgical, 1433, 1434f, 1435f
ureterocele and, 1450, 1451
ureteropelvic junction obstruction with, 1417
urinary tract infection and, 1428, 1429–1430,
1431f, 1433, 1444
VESPA (virtual environment for surgical planning
and analysis), 72–73
VEST endoscopic surgical trainer, 73, 73f
Vestibular fistula, 1294, 1301
VHL gene, in von Hippel-Lindau disease, 1399
Video-assisted thoracic surgery (VATS)
for empyema, 870–872
for patent ductus arteriosus, 1649
for pneumothorax, 873
for thoracic trauma, 274
for vascular ring repair, 1670–1671
Videoendoscopy, for esophagoscopy, 885, 886
Vinblastine, 407t
Vinca alkaloids, 407t
Vincristine, 407t
for Kasabach-Merritt syndrome, 1619–1620
VIPoma, 1383
Viral infection
intussusception and, 1094, 1097
pneumonia as, 858–859, 858f
Viral vectors
for gene transfer, 23–25, 23f, 24t
targeting of, 25
Virilization
in adrenocortical lesions, 563
in ovarian tumors, 530
in Sertoli-Leydig cell tumors, 540–541
Virtual endoscopy
in Crohn disease, 1211, 1211f
in gastrointestinal bleeding, 1154
Virtual reality, 67–73
challenges of, 72
components of, 68
finite elements in, 69–70
force and tactile feedback in, 71, 71f
historical background on, 67–68, 67f
input devices in, 70–71, 70f
patient-specific, 68, 69f
I-48 INDEX
Virtual reality (Continued)
for preoperative planning, 72–73, 72f
surface rendering in, 68–69, 69f
surgical simulation based on, 73, 73f
tracking in, 71–72
visual displays in, 70, 70f
volume rendering in, 69
Virtue ethics, 237
Virulence, bacterial, 149–150
Visceral pain, in appendicitis, 1256
Visible Human project, 68, 69f
Visual displays
in surgical simulation, 65–66
in virtual reality, 70, 70f
Visual disturbances, in brain tumors, 592
Vital capacity, 112, 113f
in pectus excavatum, 782
Vital signs, abnormal, age group–specific definitions
for, 152, 152t
Vitamin(s)
deficiency of, in short bowel syndrome,
1137
fat-soluble, 183
in parenteral nutrition, 189–190, 190t
requirements for, 183–184
supplementation of
after bariatric surgery, 1044–1045, 1045t,
1046–1048
in cholestasis, 197, 197t
water-soluble, 183
Vitamin A, 183
in parenteral nutrition, 189–190, 190t
Vitamin B, 183
Vitamin B12
with bacterial overgrowth, 1140
after bariatric surgery, 1046
deficiency of, after bladder augmentation or
replacement, 1495–1496
Vitamin C, 183
Vitamin D, 183
deficiency of, in short bowel syndrome,
1137
metabolic bone disease and, 193
supplementation of, 187–188
Vitamin E, 183
Vitamin K, 183
deficiency of, 174, 1148–1149
Vitelline. See Omphalomesenteric entries.
Vocal cords
anatomy of, 837–838
fixation of, 842–843
paralysis of, 724, 842–843
after lung transplantation, 678
pseudopolyps on, 952–953, 953f
Voiding
dysfunctional, 1462–1464, 1462f, 1463f
symptoms related to, 1453, 1454f. See also Bladder
dysfunction.
Voiding cystourethrography, 1429–1430, 1431f
in bladder dysfunction, 1454, 1455f
in disorders of sex development, 1575
in dysfunctional elimination syndromes, 1462,
1462f, 1463, 1463f
in ectopic ureter, 1446
in myelodysplasia, 1459
in penile agenesis, 1585, 1588f
in posterior urethral valves, 1461–1462, 1461f,
1555, 1556f
in ureterocele, 1448–1449, 1449f
in ureteropelvic junction obstruction, 1417
Volume-controlled ventilation, 118
Volume rendering, in virtual reality, 69
Volvulus Wilms’ tumor (Continued)
cecal, 1117, 1124 pathology of, 427–429
colonic, 1132, 1252, 1252f nephrogenic rests and, 429, 429f
gastric, 1036–1038, 1037f, 1037t pretreated tumors and, 428–429, 428t
jejunoileal atresia and stenosis with, 1067 prognostic factors in, 430
Meckel diverticulum with, 1090f, 1091 radiation therapy for, 435–436, 436t
midgut, 1116, 1116f. See also Midgut volvulus. recurrent, 435
Vomiting, in brain tumors, 591–592 renal failure in, 433, 433f
Von Hippel-Lindau disease screening for, 427
pheochromocytoma in, 561 special considerations in, 431–433
renal cysts in, 1399 staging of, 423, 424t, 429–430
Von Langenbeck cleft palate technique, 703, 705f survival rate for, 423, 424f
Von Willebrand disease, 170–171, 432–433 treatment of, 430–437
Vulva unilateral, 424, 425f, 430–431, 434, 435t
hematoma of, 324 unresectable, 431
rhabdomyosarcoma of, 498 Wilms’ tumor 1 (WT-1) gene
Vulvar tissue, for vaginoplasty, 1595, 1596f in renal development, 1395
in sexual differentiation, 1565, 1566f
Winkelmann rotationplasty, 585–586, 586f
Wireless capsule endoscopy, in Meckel diverticulum,
W 1089
Wada testing, 1689 Wirsung, duct of, 1371
WAGR syndrome, 404–405, 424 Wiskott-Aldrich syndrome, 169
Waldeyer ring, 716 WNT-4 gene, in ovarian differentiation, 1567
Warfarin, coagulopathy from, 173 WNT activation, in hepatoblastoma, 467
Warm shock, 159t, 160, 161 Wolffian duct, 1565, 1566f, 1567
Warren shunt, 1364–1365, 1365f Work capacity, in pectus excavatum, 781
Warthin tumor, 733 World Federation of Associations of Pediatric
Water. See Fluid(s). Surgeons (WOFAPS), 17
Waterston risk groups, for esophageal atresia, 895, World Health Organization (WHO) classification of
895t, 897–898 ovarian tumors, 533–534, 534t
Webs Wound care, for burns, 376–380, 377t
congenital, 1714 Wound healing
laryngeal, 841–842, 841f, 842f in children versus adults, 328–329
Weigert-Meyer rule, 1443 after myelomeningocele repair, 1676
Wharton ducts, 716, 729 Wound infections
Wharton vaginoplasty procedure, 1594, 1596 after anorectoplasty, 1307
Wheatstone bridge, 61 after appendectomy, 1262
Whitaker test, in ureteropelvic junction obstruction, after orchidopexy, 1013
1419, 1419f Wyatt, O., 5
White pulp, 1385
Williams syndrome, 1631
Williams vulvovaginoplasty, 1595, 1596f X
Wilms’ tumor, 423 Xenografts, in burn care, 378
acquired von Willebrand disease in patients with,
432–433
anaplastic, 428, 434, 435 Y
anomalies associated with, 424–425
Y chromosome
bilateral, 424, 425f, 433–434, 433f
analysis of, 1572, 1572t
bone marrow transplantation for, 435
sex-determining region of, 1565–1567
chemotherapy for, 434–435, 435t
Yolk sac, 961, 962f, 963t, 1085, 1086f, 1087f
clinical presentation in, 426–427
Yolk sac tumors, 508, 513, 515f
cystic partially differentiated, 439, 440
ovarian, 533, 543
diagnosis of, 427
testicular, 510, 510f, 551–552
epidemiology of, 424–425, 425f
vaginal, 1607–1608
extension of
Young-Dees-Leadbetter bladder neck reconstruction,
in renal vein, inferior vena cava, and atrium, 431
1477, 1478f, 1522–1523, 1524f
in ureter, 431–432
historical perspective on, 423
in horseshoe kidney, single kidney, or
nonfunctioning kidney, 432, 1408 Z
late effects of, 436–437 Z-plasty, for midline cervical clefts, 760, 760f
metastasis of Z sign, in malrotation with midgut volvulus, 1120,
hepatic, 435 1121f
pulmonary, 436, 572 Zinc
molecular biology and genetics of, 425–426, 426f, deficiency of, in short bowel syndrome, 1137
426t requirements for, 184, 184t
in multicystic dysplastic kidney, 1400–1401 Zollinger-Ellison syndrome, 1383
in neonate, 432 peptic ulcer disease in, 1034
operative treatment of, 430–431 Zona fasciculata, 557
spill during, 431 Zona glomerulosa, 557
without chemotherapy, 432 Zona reticularis, 557
Volume 1, pages 1–768; Volume 2, pages 769–1738