variables including both environmental and familial factors.

IHPS is now thought to be caused by a mechanism other than

a developmental defect. Based on considerable evidence, in-
cluding a study by Wallgren,8 in which 1000 male infants
had a barium swallow immediately after birth, there were
no anatomic abnormalities of the pylorus identified. Thus it
is generally agreed that IHPS is not a congenital abnormality.
Of interest is that subsequently, IHPS developed in five of
these infants. In a similar but more recent study, 1400 consec-
utive newborn infants underwent ultrasound (US) measure-
ments of the pylorus, with no abnormalities seen.9 In nine
of these infants, however, IHPS subsequently developed.

Anatomy and Histology

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

The gross appearance of the pylorus in IHPS is that of an en-

larged, pale muscle mass usually measuring 2 to 2.5 cm in
length and 1 to 1.5 cm in diameter (Fig. 78-1). Histologically,
there is marked muscle hypertrophy and hyperplasia10 prima-
rily involving the circular layer and hypertrophy of the
underlying mucosa.11 Immunohistochemical analysis of
CHAPTER 78 the hypertrophic muscle reveals an increase in fibroblasts,
fibronectin, proteoglycan chondroitin sulfate,12 desmin,13
elastin,14 and collagen.15 Confocal microscopy identifies ab-
normally contorted and thickened nerve fibers.16 The result
Hypertrophic of these gross and microscopic changes is either partial or
complete obstruction of the pyloric channel.

Pyloric Stenosis
Marshall Z. Schwartz Etiology
------------------------------------------------------------------------------------------------------------------------------------------------

The etiology of IHPS has eluded investigators for several de-

cades. Despite considerable research attempting to elucidate
the etiology of IHPS, no definitive causative factors have been
identified. Both genetic and environmental factors seem to
History
------------------------------------------------------------------------------------------------------------------------------------------------
play a role in the pathophysiology. Evidence for a genetic pre-
disposition includes variability among races, a clear male pre-
The history of what we now refer to as infantile hypertro- ponderance, an increased risk to first-born infants with a
phic pyloric stenosis dates back to the early 1700s. Blair1 positive family history, and certain ABO blood types. Mitchell
described an infant with postmortem findings consistent and Risch studied the familial occurrence patterns of IHPS and
with hypertrophic pyloric stenosis in 1717. Hirschsprung, noted that it was compatible with either multifactorial thresh-
in 1888,2 provided the first complete description of hypertro- old inheritance or the effects of multiple interacting loci.5
phic pyloric stenosis. He believed that this entity was congen- Although pyloric stenosis is more common in boys, the risk
ital and represented failure of involution of the fetal pylorus for IHPS in the offspring of mothers who had pyloric stenosis
and named it angeborener pylorusstenose (congenital pyloric as infants is greater than if the father had IHPS. Environmental
stenosis). In a landmark paper in 1908, Dufour and Fredet3 factors associated with IHPS include the method of feeding
suggested that surgical correction could be accomplished by (breast-feeding versus formula feeding), seasonal variability, ery-
splitting the hypertrophied pyloric muscle to the submucosa thromycin exposure, and transpyloric feeding of premature
and closing the muscle transversely. However, in 1912, infants.5,17–19
Ramstedt4 suggested that closure of the muscle was not nec- Another focus has been on alterations in relaxation of the
essary, and the current standard operation was established. pyloric muscle. One study noted a decrease in ganglion cells in
Infantile hypertrophic pyloric stenosis (IHPS) is the most the circular muscle of the pylorus,20 but a subsequent study
common cause of gastric outlet obstruction in infants. The found that the ganglion cells were normal in number but
prevalence of IHPS ranges from 1.5 to 4.0 per 1000 live births appeared to be immature.21 Zuelzer22 and Jona23 did not find
in Caucasian infants but is less prevalent in African-American any change in number or ultrastructural appearance of the
and Asian children.5 Reports have suggested that the inci- ganglion cells. As new technology and concepts have evolved,
dence is increasing.6,7 It is well known that it is more common additional associations that involve IHPS and gastrointestinal
in boys than girls, with a ratio of approximately 2:1 to 5:1.5 peptides, growth factors, neurotrophins, changes in neural
The occurrence of IHPS has been associated with several development, and nitric oxide have been described.
1021
1022 PART VII ABDOMEN
FIGURE 78-1 Laparoscopic view of a hypertrophied pylorus.
GROWTH FACTORS AND
GASTROINTESTINAL PEPTIDES
Over the past several decades numerous gastrointestinal pep-
tides or growth factors have been implicated by direct or
indirect methodologies as having a causal relationship inclu-
ding gastrin, substance P, epidermal growth factor (EGF),
transforming growth factor-a (TGF-a), insulin-like growth
factor-I, somatostatin, secretin, enteroglucagon, and neuro-
tensin.24,25 Other peptides have been evaluated for their po-
tential relationship to IHPS. Substance P, a neurotransmitter
responsible for enteric muscle contraction, could produce
chronic pylorospasm leading to muscle hypertrophy.26 This
peptide is present in higher concentration in the pyloric
muscle of patients with IHPS. Shima and colleagues27–29
found an increase in the gene expression of EGF, TGF-a, and
insulin-like growth factor-I and an increase in immunostaining
for EGF and TGF-a. Somatostatin, secretin, enteroglucagon,
and neurotensin have also been implicated.24,30,31 However,
their role has not been substantiated, and no clear etiologic
relationship has been identified.
Neurotrophins, Neural Development,
and Nerve Function
Neurotrophins, which are important for nerve differentiation
and survival, have been noted to be decreased in IHPS. Further
studies have shown that one of the specific receptors for these
neurotrophins, the tyrosine kinase A receptor c-kit, is not
present in IHPS tissue.32–34 Investigators also identified that
the pylorus in IHPS is deficient in glial-derived growth factors,
which may be indicative of immature development of the
enteric nervous system.35
Other techniques have been applied to the evaluation of
pyloric innervation. Okazaki and colleagues36 used monoclo-
nal antibodies targeted to nerve terminals and neurofilaments.
They found a reduced density of both neural elements in
patients with IHPS. Kobayashi and colleagues37 demonstrated
a decrease in nerve-supporting cells in both the circular and
longitudinal muscle layers in infants with IHPS. Piotrowska
and colleagues38 found that muscle biopsy specimens from
IHPS patients have decreased numbers of interstitial cells of
Cajal, the pacemaker cells for the gastrointestinal tract, and
a lack of heme oxygenase-2, which may play a role in com-
municating between the interstitial cells of Cajal and smooth
muscle cells.
Nitric Oxide
Nitric oxide can induce smooth muscle relaxation in the
esophagus, stomach, and intestine. It was reasoned that a
deficiency of nitric oxide in the pyloric muscle might result
in failure of relaxation.39–41 Nicotinamide-adenine dinucleo-
tide phosphate (NADPH) diaphorase (essentially identical to
nitric oxide synthase) was measured in biopsy specimens of
hypertrophied pyloric muscle wall from IHPS patients. The
hypertrophied circular muscle did not demonstrate any
NADPH diaphorase activity, but the activity in the longitudinal
muscle was normal. The authors concluded that the lack of
nitric oxide synthase in pyloric tissue might be responsible
for pylorospasm and lead to IHPS.42
Thus at present no definitive cause of IHPS has been elu-
cidated. Whatever the mechanism, it must take into account
that the process of IHPS generally occurs several weeks after
birth and the circular muscle hypertrophy is transient even
without myotomy. It is also apparent that whatever triggers
IHPS causes a dramatic change in cellular architecture and
function through what may be a broad array of pathways.
Clinical Features and Differential
Diagnosis
------------------------------------------------------------------------------------------------------------------------------------------------
The typical clinical finding in an infant with IHPS is the onset
of nonbilious vomiting at 2 to 8 weeks of age with a peak oc-
currence at 3 to 5 weeks. Two reports are at variance with tra-
ditional concepts: IHPS has been diagnosed at birth43 and even
in utero.44 Initially, the emesis may not be frequent or forceful,
but over a period of several days it progresses to nearly every
feeding and becomes forceful (projectile). On occasion, there
may be blood in the emesis that gives it a brownish discoloration
or a coffee-ground appearance as a result of gastritis or esopha-
gitis. Infants with IHPS remain hungry after emesis and are oth-
erwise not ill appearing or febrile. A significant delay in diagnosis
leading to severe dehydration, however, results in a lethargic
infant. Some infants have diarrhea (starvation stools) and are
thought to have gastroenteritis. Approximately 2% to 5% of
infants have jaundice from indirect hyperbilirubinemia, which
can reach levels as high as 15 to 20 mg/dL. This is believed to
be secondary to glucuronyl transferase deficiency.45 In prema-
ture infants, IHPS is generally diagnosed 2 weeks later than in
term infants.19 The emesis may not be projectile and evolves
more slowly, which frequently leads to a delay in diagnosis.
Pylorospasm and gastroesophageal reflux may be difficult
to differentiate from IHPS without further imaging evaluation.
Other medical causes of nonbilious vomiting include gastro-
enteritis, increased intracranial pressure, and metabolic disor-
ders. Other surgical causes of nonbilious emesis include antral
webs, pyloric atresia, duplication cyst of the antropyloric
region, and ectopic pancreatic tissue within the pyloric
muscle, all far less common than IHPS.
 CHAPTER 78 HYPERTROPHIC PYLORIC STENOSIS 1023

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

Nonbilious projectile vomiting, visible peristaltic waves in the

left upper part of the abdomen, and hypochloremic, hypokale-
mic metabolic alkalosis are cardinal features of IHPS. A definitive
diagnosis can be made in 75% of infants with IHPS by careful
physical examination of the upper part of the abdomen. Unfor-
tunately, this is becoming a lost skill. Frequently, imaging proce-
dures are requested by primary care physicians in lieu of a careful
physical examination. To be successful in palpating an enlarged
pylorus, the infant must be calm, warm, and cooperative. The
use of a pacifier or a small feeding (5% dextrose in water) may
be helpful. If the stomach is distended, aspiration with a naso-
gastric tube assists successful palpation of an enlarged pylorus,
often referred to as “the olive.” With the infant supine and the legs A
bent to relax the abdominal muscles, the examining hand should
be placed on the epigastrium. After the edge of the liver has been
identified with the fingertips, gentle pressure deep to the liver
and progressing caudally in the midline a third of the distance
between the umbilicus and xiphoid should reveal a palpable
pylorus if IHPS is present. The examiner should be willing to
commit 5 to 15 minutes of uninterrupted time if necessary to
obtain an adequate examination. One should be able to roll
the hypertrophied pylorus under the fingertips to be convinced
of the diagnosis. Failure to palpate the pylorus requires further
workup to clarify the cause of vomiting. The differential diag-
nosis includes gastroenteritis, food allergy, gastroesophageal
reflux, pylorospasm, antral web, pyloric duplication, ectopic
pancreatic tissue in the wall of the pylorus, and vomiting occa-
sionally accompanying adrenogenital syndrome, metabolic
B
disorders, and increased intracranial pressure.
US has become not only the most common initial imaging FIGURE 78-2 Ultrasound images of the hypertrophied pylorus in a
technique for the diagnosis of IHPS but also the standard for 5-week-old infant. A, A longitudinal view is shown with pyloric length
diagnosing IHPS. Under optimal circumstances, this tech- (X-X) measuring 18 mm and pyloric muscle thickness (þ-þ) measuring
nique is reliable. However, it is dependent on the level of 5.2 mm. B, Transverse view illustrating muscle thickness (þ-þ).
(A, Courtesy of Dorothy Bulas, MD.)
experience and expertise of the ultrasonographer. The gener-
ally accepted criteria for a positive US study are a pyloric mus-
cle thickness of 3.5 (in premature infants) to 4 mm or
more and a pyloric channel length of 16 mm or greater
(Fig. 78-2).19 Some centers also determine pyloric diameter
and consider more than 14 mm to be abnormal.46 Lamki
and colleagues47 reviewed their experience with US and con-
cluded that a muscle thickness of 3 mm should be considered
a positive finding for IHPS in infants younger than 30 days.
US also obviates the need for radiographic contrast studies
and the associated radiation exposure.
If US is not available or diagnostic, an upper gastrointesti-
nal (UGI) contrast examination is highly effective in making
the diagnosis of IHPS. Barium is generally preferred compared
with water-soluble contrast to avoid the chemical pneumonitis
should aspiration occur. This study should demonstrate an
elongated pyloric channel and indentation on the antral
outline, which are indirect findings of pyloric muscle enlarge-
ment (Fig. 78-3). If barium does not leave the stomach, it is
not possible to confirm the diagnosis of IHPS because pylor-
ospasm can also produce transient complete gastric outlet
obstruction. However, if sufficient time is taken with intermit-
tent fluoroscopy, it should be possible to differentiate between
FIGURE 78-3 Upper gastrointestinal study illustrating an elongated py-
the two entities. It is advisable for the radiologist to aspirate loric channel with shoulders proximally as indirect evidence of pyloric mus-
the residual barium from the stomach after termination of cle hypertrophy. Note the small amount of contrast in the duodenal bulb
the radiographic procedure to avoid perioperative aspiration. defining the length of the hypertrophied pylorus.
1024 PART VII ABDOMEN

In 1998 our group conducted a survey of practice patterns be based on the level of hypokalemia and the rate of infusion
for the diagnosis of IHPS among pediatric surgeons in while keeping in mind that a potassium chloride concen-
community, university, and children’s hospitals.48 The ques- tration exceeding 30 mEq/L is rarely indicated.
tionnaire was distributed to 487 board-certified pediatric It is necessary to monitor urine output and serum elec-
surgeons, with a 69% response rate. Ninety-one percent of trolytes. Normalizing the serum bicarbonate level (with the
the respondents stated that they would not request a further goal of decreasing the level below 30 mEq/dL) usually lags
diagnostic test if a pyloric olive was palpable. US was identi- behind normalization of fluid volume and serum potassium
fied as the diagnostic study of choice for an inconclusive phys- and chloride. Administration of ammonium chloride or
ical examination (86% of the time). Finally, 82% of pediatric dilute hydrochloric acid is rarely necessary. As noted previ-
surgeons would proceed with an operation with a positive US ously, indirect hyperbilirubinemia with clinical signs of
study even if the results of physical examination under anes- jaundice occurs in a small percentage of infants with IHPS.
thesia were negative. The response to this question indicates It is not usually necessary to evaluate the hyperbiliru-
that US has essentially supplanted physical examination as binemia further. The hyperbilirubinemia invariably resolves
the diagnostic test of choice for IHPS.48 postoperatively.

OPERATIVE PROCEDURE
Treatment
------------------------------------------------------------------------------------------------------------------------------------------------ Again, it is important to emphasize that fluid and electrolyte
abnormalities must be corrected preoperatively including
PREOPERATIVE PREPARATION
having a serum bicarbonate below 30 mEq/L to avoid respira-
It is most important to prepare the infant adequately for tory depression and prolonged postoperative intubation.
anesthesia and surgical correction of IHPS. The length of prep- In the operating room, before the induction of anesthesia, it
aration depends on the severity of the fluid and electrolyte is important to aspirate the stomach. If a barium UGI study has
abnormalities. Benson and Alpern49 defined three levels of been performed, it is advisable to remove the residual contrast
severity primarily on the basis of the serum carbon dioxide material by gastric irrigation and suctioning to avoid the risk
content (slight, <25 mEq/L; moderate, 26 to 35 mEq/L; and for perioperative aspiration.
severe, >35 mEq/L). In addition to the elevated bicarbonate, The operative procedure of choice remains the Ramstedt
hypokalemia, hypochloremia, dehydration, and possibly pyloromyotomy. This procedure has stood the test of time
malnutrition may be present. because it is straightforward, curative, and associated with
Most infants with IHPS do not have complete gastric outlet remarkably low morbidity. Regardless of abdominal access
obstruction and can therefore tolerate their gastric secretions. techniques, the myotomy created is identical.
Oral feedings should be discontinued. A nasogastric tube
Minimal Laparotomy (“Open”) Technique
should not be placed routinely because it removes additional
fluid and hydrochloric acid from the stomach, which per- The standard open approach is a right upper quadrant trans-
petuates the electrolyte and acid-base imbalance. verse incision of approximately 2.5 to 3 cm made either over
Fluid resuscitation should be based on the degree of dehy- or lateral to the right rectus muscle and just superior to the
dration and the extent of electrolyte abnormalities. Most in- edge of the liver. The fascial layers are divided transversely
fants with IHPS should be able to be resuscitated within a over the rectus muscle, and it can be either retracted laterally
24-hour period. With severe metabolic and fluid abnormali- or split longitudinally in the middle. Several other incisions
ties, however, aggressive resuscitation should be avoided be- have been described.3 A commonly used alternative is a
cause it can produce rapid fluid and electrolyte shifts, possibly supraumbilical curved skin incision followed by division of
leading to seizures and other complications. Intravenous ad- the midline fascia superiorly for 1 to 2 cm. The edge of the
ministration of 5% dextrose in 0.45 normal saline containing liver is retracted superiorly to expose the greater curvature
20 mEq/L of potassium chloride is the optimal resuscitation of the stomach (near the pylorus), which is grasped with a
regimen for fluid and electrolyte replacement. Under circum- noncrushing forceps or clamp and brought out through the
stances of extreme hypokalemia, the concentration of potas- incision. A damp gauze sponge can then be used to grasp
sium chloride can be increased to 30 mEq/L, but because the stomach, and with traction inferiorly and laterally to the
the intravenous fluid rate likely will be above maintenance patient’s left, the pylorus can be delivered through the inci-
rates, the serum potassium level should be carefully moni- sion. Grasping the duodenum or pylorus directly with forceps
tored. Withholding potassium chloride in the intravenous can result in injury or perforation of these structures and
fluid while awaiting urine output only delays appropriate re- should therefore be avoided. The pylorus can be stabilized
placement. The exception (rare) is knowledge of preexisting by the index finger of the operating surgeon standing to the
renal impairment or evidence of acute renal compromise. right of the patient. The serosa on the anterior wall of the
Hyponatremia is rarely a problem. Nonetheless, it is common hypertrophied pylorus is incised with a scalpel from the area
to see normal saline given as an initial bolus. However, there is just proximal to the hypertrophied muscle extending just
little rationale for the use of normal saline because it enhances proximal to the pyloric vein (Fig. 78-4). I and many others
the hypokalemia by dilution and provides an excess amount of prefer to use the back of a scalpel handle to complete the
sodium. Intravenous therapy should be correlated with the myotomy by bluntly splitting the hypertrophied muscle down
level of dehydration. An initial rate for fluid resuscitation is to the submucosa (see Fig. 78-4). Other surgeons prefer the
1.25 to 2 times the normal maintenance rate until adequate use of a pyloric spreading clamp such as that described by
fluid resuscitation and urine output are achieved. The concen- Benson.50 It is appropriate to leave a few pyloric muscle fibers
tration of potassium chloride in the intravenous fluid should intact at the duodenal end to reduce the risk for duodenal

Serosa Stomach
A Pyloric vein
Duodenum
Submucosa
Serosa
C
B
Serosa
Muscle
FIGURE 78-4 Pyloric stenosis, operative technique. A, Pylorus delivered
from the peritoneal cavity and stabilized between the surgeon’s index
finger at the duodenal end with the stomach grasped proximally by the
assistant. The serosal incision is outlined and begun. B, Back of a scalpel
handle being used to split the hypertrophied muscle down to the sub-
mucosa. C, Completed myotomy showing submucosa bulging through
the divided muscle.
perforation. Most incomplete myotomies are a result of failure
to extend it far enough proximally onto the antrum. A careful
check for a leak from the stomach or duodenum should be
made before returning the pylorus to the peritoneal cavity.
Bleeding from the pyloric muscle edges or submucosal surface
is generally venous and almost always stops after returning the
stomach and pylorus to the peritoneal cavity.
Entrance into the lumen during pyloromyotomy should be a
rare event. If it occurs, the submucosa should be approximated
with interrupted fine absorbable suture and a portion of omen-
tum placed over this site. An infrequently needed alternative is
to reapproximate the myotomy site, rotate the pylorus 180 de-
grees, and perform a myotomy on the posterior wall of the py-
lorus. The fascial layers of the abdominal wall are closed with
running absorbable suture. The skin is closed with subcuticular
suture and Steri-Strips and covered with a dressing.
Laparoscopic Procedure
The first description of a laparoscopic pyloromyotomy was by
Alain and colleagues51 in 1991. Since that time, numerous
publications have supported this approach. As the procedure
is currently performed in most centers, the infant is placed su-
pine at the end of the operating table (Fig. 78-5). A 5-mm port
is placed through a small vertical incision in the umbilicus,
and the abdomen is insufflated with CO2 to a pressure of 6
to 8 cm H2O. Two additional access sites are chosen (under
direct vision with the camera) in the left and right upper
quadrants—ports are not usually required (Fig. 78-6). After
CHAPTER 78 HYPERTROPHIC PYLORIC STENOSIS 1025
FIGURE 78-5 Position of the infant on the operating room table for
laparoscopic pyloromyotomy.
infiltration with a local anesthetic, a No. 11 scalpel blade is
used to make a small (2- to 3-mm) skin incision and extended
through the abdominal wall. Until recently an arthrotomy
knife was used to begin the myotomy: This instrument is
no longer available. The extended cautery blade has served
as an adequate substitute. A grasper is placed directly through
the left upper quadrant incision and used to grasp the antrum
just proximal to the pylorus. After the right upper quadrant
2- to 3-mm incision is accomplished, the extended cautery
blade is passed directly into the peritoneal cavity. The grasper
stabilizes the pylorus, and the cautery blade is used to score
the serosa of the hypertrophied pylorus with a low electrical
current at the site of the myotomy in a fashion similar to what
is done during the “open” technique (Fig. 78-7). The cautery
blade (without electrical current) is used to initiate splitting
of the hypertrophied muscle and removed. A pyloric
spreader designed for laparoscopic procedures is placed
through the right upper quadrant incision. This instrument
is used to complete the myotomy (Fig. 78-8). At the comple-
tion of the myotomy, each side of the divided muscle is
grasped and moved in opposite directions to ensure that a
complete myotomy has been achieved. Air is then insufflated
through a nasogastric tube (typically 60 mL) to check for
a leak. If none is seen, the instruments are removed, the
CO2 is evacuated, and the umbilical port is removed. The
umbilical fascial defect is closed, and the skin at all three sites
is closed with subcuticular suture.
COMPARISON OF THE OPEN VERSUS
LAPAROSCOPIC APPROACH
Aesthetics were the initial impetus for developing this pro-
cedure; however, proponents of laparoscopic pyloromyotomy
now cite many other benefits including faster recovery time
with quicker return to full feeding, earlier postoperative dis-
charge, a decrease in postoperative emesis, and a decrease
in pain.52 However, advocates of open pyloromyotomy state
that recovery time is similar and that the laparoscopic
approach has a greater complication rate including perfora-
tion,8,52–55 missed perforation,54 incomplete myotomy,54,56 an
increase in operative time, and increased expense.53 All
1026 PART VII ABDOMEN
A B
FIGURE 78-7 View through the laparoscope with the arthrotomy knife
being used to incise the serosa over the hypertrophied pylorus.
the published studies to date have been retrospective and
demonstrate a marked degree of variability.
The range for the average time to discharge in the pub-
lished literature for laparoscopy is similar to that for open pro-
cedures. Of the four publications that compare time to
discharge between the open procedure and laparoscopy, only
one group found statistical significance and the difference in
discharge time was only 4 hours (Table 78-1).53,55,57–61
There is wide variation in the incidence of postoperative
emesis. The range for laparoscopy was 1% to 68%, whereas
for the open procedure it was 25% to 65%, with no statistical
significance observed between the open and laparoscopic
groups.52,53,57,58,62
The average time to establish full feeding in the laparoscopic
group ranged from 16 to 32 hours versus 20 to 61 hours for the
open procedure (Table 78-2).52–54,57,62 The results are no dif-
ferent if the Fujimoto outlier study is eliminated.52
FIGURE 78-6 Location of “ports”
(A) and after the instruments and
telescope are in place (B).
FIGURE 78-8 Hypertrophied pyloric muscle after being split with the
spreader.
Over the past decade, considerable additional experience
has occurred and laparoscopic pyloromyotomy has become
the standard at the training sites and children’s hospitals in
the United States with low morbidity and short hospital stays.
The wound infection rate for both groups was similar
and ranged from 0% to 6% in the laparoscopic groups and
from 0% to 7% in the open groups. None of these studies
identified a significant difference between the groups (Ta-
ble 78-3).52,54,55,57–59,63
The operative time was similar for both groups, with
laparoscopy taking from 24.3 to 41 minutes and the open
procedure taking from 18.9 to 32.5 minutes. Yagmurlu and col-
leagues56 (the largest study) showed that laparoscopy was faster
by 4.7 minutes on average. Sitsen and colleagues55 and Scorpio
and colleagues60 found that laparoscopy was slower by 14.3
 CHAPTER 78 HYPERTROPHIC PYLORIC STENOSIS 1027

TABLE 78-1 TABLE 78-4

Average Length of Stay (Hours) Average Operative Time (Minutes)
Author N Laparoscopic Open Author N Laparoscopic Open
Kramer WL, et al. 133 61.6 Yagmurlu A, et al. 457 24.3* 29
Caceres M, Liu D 56 60 61.5 Kramer WL, et al. 133 29
Campbell BT, et al. 117 31 28 Caceres M, Liu D 56 36.1 32.5
Shankar KR, et al.* 86 58 Campbell BT, et al. 117 38 33
Sitsen E, Bax NM, van der Zee DC 72 70 74{ Shankar KR, et al.{ 86 30
Najmaldin A, et al. 37 28 Fujimoto T, et al.{ 60 27.4 31.9
Scorpio RJ, et al. 63 44 63 Sitsen E, Bax NM, van der Zee DC 72 33.2 18.9*
Range 28-70 28-74 Ford WD, Crameri JA, Holland AJ 84 41 28
Greason KL, et al. 25 25.4 26.1
*Umbilical approach only. Najmaldin A, et al. 37 29
{
P < .05.
Scorpio RJ, et al. 63 40.2{ 27.3
Range 24-41 19-33
TABLE 78-2
Average Time to Full Feeding (Hours) *P < .01.
{
Umbilical approach only.
{
Author N Laparoscopic Open P < .05.
Caceres M, Liu D 56 24.1 27
Campbell BT, et al. 117 19 20
Fujimoto T, et al.* 60 34.8 61.2{
TABLE 78-5
Ford WD, Crameri JA, Holland AJ 84 32 41
Average Perforation Rate
Greason KL, et al. 25 19 23
Range 19-35 20-61 Author N Laparoscopic Open

*Umbilical approach only. Yagmurlu A, et al. 457 0.40% 3.6%*

{ Kramer WL, et al. 133 3%
P < .01.
Campbell BT, et al. 117 8% 4%
Fujimoto T, et al.{ 60 3% 0%
TABLE 78-3 Sitsen E, Bax NM, van der Zee DC 72 9% 6%
Average Wound Infection Rate Ford WD, Crameri JA, Holland AJ 84 10% 6%
Author N Laparoscopic Open Range 0.4%-10% 0%-6%

Kramer WL, et al. 133 3% *P < .05.

{
Caceres M, Liu D 56 0% 3% Umbilical approach only.
Fujimoto T, et al.* 60 0% 7%
Sitsen E, Bax NM, van der Zee DC 72 6% 3%
Ford WD, Crameri JA, Holland AJ 84 3% 2%
Najmaldin A, et al. 37 0%
colleagues63 found that laparoscopy tended to decrease the
Scorpio RJ, et al. 63 0% 5% temperature of the infant to a greater degree than the open
Range 0%-6% 0%-7% procedure did, which may have detrimental physiologic
effects. These biologic markers indicate subtle differences
*Umbilical approach only. between the open and laparoscopic procedures, but they
may not be clinically significant.
As experience with laparoscopy has increased, the learning
minutes and 13 minutes, respectively, but in each series the smal- curve for the surgeon has become shorter, and technology
ler number of patients probably indicates less operative experi- has continued to improve, application of this technique has
ence at the time of the study. The other five comparative studies become commonplace. The larger laparoscopic studies dem-
failed to reach statistical significance (Table 78-4).53–62 onstrate a comparable complication rate, and there may be
The perforation rate, which includes both duodenal and a decrease in subjective factors such as pain and earlier feed-
pyloric injuries, had ranges of 0.4% to 10% for laparoscopic ing. Ultimately, the main difference may be the increased cost
pyloromyotomy and 0% to 6% for the open procedure. Four of laparoscopy. However, for surgeons who have limited expo-
comparative studies found that laparoscopy had a slightly sure or access to laparoscopic techniques and equipment, the
higher perforation rate, but the differences were not statis- “open” Ramstedt pyloromyotomy remains the gold standard
tically significant (Table 78-5).52–56,58 with which all other methods should be compared.
Other factors that have been analyzed include the abdom-
inal wall hernia rate, which in some laparoscopic groups
ranged from 4% to 7%,55,60 interleukin-6, and heat loss.
POSTOPERATIVE MANAGEMENT
Fujimoto and colleagues52 found that interleukin-6 peak
levels were significantly lower in the laparoscopy group In the majority of infants, feeding can be started within 4
24 hours after surgery. They hypothesized that this could in- hours after the surgical procedure. Infants with hematemesis
dicate a decreased level of stress on the patient. Holland and from gastritis may benefit by delaying feeding for an additional
1028 PART VII ABDOMEN
TABLE 78-6
Postpyloromyotomy Feeding Schedule*
Pedialyte, 30 mL orally every 3 hr " 1
Full-strength formula, 30 mL orally every 3 hr " 1
Full-strength formula, 45 mL orally every 3 hr " 2
Full-strength formula, 60 mL orally every 3 hr " 1
Full-strength formula, 75 mL orally every 3 hr " 1
Full-strength formula as desired
*Begin 4 to 6 hours after surgery. For very small infants, the starting feeding
volume may be reduced to 15 mL and the schedule stopped at volumes of
60 to 75 mL, which provide an adequate calorie supply. Breast milk can be
substituted for formula if appropriate.
6 to 12 hours after the procedure. A typical feeding schedule is
shown in Table 78-6. It is clear that the more aggressive
the feeding schedule, the greater the incidence of emesis in
the initial postoperative period, but this approach is usually
successful and generally allows for earlier hospital dis-
charge.64 The feeding schedule shown in Table 78-6 is mod-
erately aggressive and allows for hospital discharge
approximately 24 hours after the feedings are initiated.
In the future, infants may be discharged within 3 or 4 hours
after pyloromyotomy, in which case resumption of feeding will
be undertaken by the family at home.
NONOPERATIVE TREATMENT
This approach never gained acceptance in North America but
was practiced in the past in some European countries.
Although infants with IHPS can be managed with frequent
small feedings, this practice necessitates either a prolonged
hospital stay or an attentive caregiver at home and may lead
to aspiration and malnutrition. It may take months for the
hypertrophied muscle to resolve. The occasional mortality
and the prolonged interval from diagnosis to resolution led
to abandonment of this type of management.
Complications
------------------------------------------------------------------------------------------------------------------------------------------------
Complications after pyloromyotomy should be minimal if per-
formed by experienced surgeons. Vomiting, frequent in the
early postoperative period, is thought to be secondary to gastro-
esophageal reflux, discoordination of gastric peristalsis, or gas-
tric atony and should not be considered a complication.
Frequent vomiting persisting beyond 3 to 4 days may suggest
an incomplete myotomy or an unsuspected perforation. A post-
operative contrast study may demonstrate a leak but is not
helpful in evaluating the completeness of the myotomy. It takes
several weeks for the radiographic appearance of the pylorus to
improve.65 Persistent and frequent vomiting 1 week beyond
the pyloromyotomy may require reexploration.
Outcome
------------------------------------------------------------------------------------------------------------------------------------------------
After pyloromyotomy, mortality should be a rare occurrence
and morbidity should be low. In a large series of infants
undergoing open pyloromyotomy, the incidence of perfora-
tion was 2.3%, wound-related complications occurred in
1%, and there was one death.50 More recent series have
reported even lower morbidity and mortality.66
As experience with laparoscopic pyloromyotomy has
increased, outcomes have become similar to those after the
open approach. With a somewhat aggressive postoperative
feeding schedule, infants should be able to be discharged
within 24 hours of pyloromyotomy.
The long-term sequelae from pyloromyotomy are minimal.
Ludtke and colleagues67 evaluated the presence of gastrointes-
tinal symptoms, measured scintigraphic gastric emptying, and
determined pyloric measurements by US in 36 adults 17 to 27
years after pyloromyotomy. Their results were compared with
those of age-matched and gender-matched controls. They
identified no differences between the postpyloromyotomy
group and the control group for the parameters measured.
The complete reference list is available online at www.
expertconsult.com.