| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Pediatric Radiology, Vanderbilt Children's Hospital, Vanderbilt
University Medical Center-D-1120, Medical Center North, 21st Ave. S.,
Nashville, TN 37232-2675.
2
Department of Pathology, Vanderbilt Children's Hospital, Nashville, TN
37232-2675.
3
Department of Surgery, Vanderbilt Children's Hospital, Nashville, TN
37232-2675.
4
Department of Gastroenterology, Vanderbilt Children's Hospital, Nashville, TN
37232-2675.
Received December 29, 2000;
accepted after revision April 19, 2001.
Address correspondence to M. Hernanz-Schulman.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We identified 102 consecutive infants with surgically confirmed IHPS and determined the thickness of the pyloric mucosa compared with the thickness of the surrounding hypertrophied muscle. Fifty-one infants who did not have pyloric stenosis served as controls.
RESULTS. Mean mucosal thickness in patients with IHPS approximated mean muscle thickness, with a ratio of 0.89. In infants with IHPS, the pyloric mucosa constitutes approximately one third of the cross-sectional diameter of the pyloric mass and fills and obstructs the pyloric canal.
CONCLUSION. Mucosal redundancy is a constant associated finding in IHPS. Although the origin of the redundancy and a cause-and-effect relationship are difficult to establish, our findings support the hypothesis that hypergastrinemia may be implicated in the pathogenesis of IHPS, and suggest that mucosal thickening could be implicated as one of the initiating factors in its development.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
On routine sonographic examinations, we have recognized marked thickening of the pyloric mucosa, which fills the pyloric mucosa, which fills the pyloric lumen and protrudes into the gastric antrum in patients with pyloric stenosis [12, 13]. Our purpose in this study was to show that pyloric mucosal thickening is a constant finding in patients with IHPS and to explore its relationship to the pyloric lumen and pyloric muscle mass. We postulated that mucosal thickening may represent an abnormal or hypertrophied mucosa and in turn could be a related and perhaps an inciting event in the development of obstruction and muscular hypertrophy in IHPS.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because none of the patients in the study group had a biopsy, a specimen from a patient with typical findings, obtained several years earlier [12], was evaluated to show the pathologic correlate to the sonographic findings. This patient had a biopsy for reasons not directly related to her condition, which was otherwise typical in its clinical manifestations, sonographic and surgical findings, and postoperative course.
The sonographic examinations of 83 infants who did not have pyloric stenosis were also reviewed. The mucosal thickness was measured in all those infants in whom it could be distinguished from gastric contents. The resultant control population consisted of 51 infants.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
In all patients, the mucosa could be traced in continuity from the normal portion of the gastric antrum into the pyloric canal. In the pyloric canal, the mucosal and submucosal layers appeared as irregularly alternating hyperechoic and hypoechoic layers, indicating an infolded mucosa protruding into the antral lumen on longitudinal images. On transverse images, the mucosa filled the pyloric canal; its cross-sectional measurement represented the lumen of the pyloric channel (Fig. 2A,2B).
|
|
The biopsy specimen of the patient presenting several years earlier showed mucosal hypertrophy, edema, and eosinophilic infiltration (Fig. 3A,3B). IHPS was confirmed surgically [12].
|
|
One patient merits specific mention. This infant was a 4-month-old girl whose sonographic examination showed that the mucosa was thickened to 8 mm, with a muscle thickness of 4.5 mm (Fig. 4A). Fluoroscopy of the upper gastrointestinal tract, performed because of the unusual sonographic findings, revealed a prominent filling defect within the pyloric channel (Fig. 4B). At pyloromyotomy, the surgeon could detect no difference in the surgical findings: "typical" pyloric stenosis was reported in the operative note, and the girl's postoperative recovery was uneventful.
|
|
Among the control patients, the mean mucosal thickness (two layers) was 2.3 ± 0.5 mm. The difference between the control and pyloric stenosis group was significant (p <0.001).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Imaging
Until the advent of sonography, the gastrointestinal barium meal test had
been the mainstay of diagnosis for many years in patients in whom the pyloric
muscle was not readily palpable. The upper gastrointestinal barium contrast
examination reveals the condition indirectly, by its effect on the pyloric
lumen. The ingested contrast material courses through the pyloric channel,
thus outlining the lumen of the canal. In many cases, the contrast material is
seen to course through more than one channel of infolded mucosa, which is
termed the double track sign, originally described as a differentiating
criterion between pylorospasm and IHPS
[15]
(Fig. 4B). This finding is
recognized to represent contrast material coursing through the interstices of
the pyloric canal. What is not widely acknowledged is that the distance
between the two lines must represent a minimum width of the pyloric channel
and that the intervening area represents a filling defect within the canal. If
this criterion, as expounded by Haran et al.
[15], serves as a method of
differentiating between pylorospasm and IHPS, it suggests that in IHPS the
mucosa is not merely compressed but truly thickened and hypertrophied. Pyloric
mucosal prolapse into the gastric antrum is not typically identified at upper
gastrointestinal barium contrast examination because it is usually small and
obscured by the contrast material within the distended stomach.
Sonography permits identification of the entire pylorus, including the lumen. The submucosa is hypoechoic and readily distinguished from the hyperechoic mucosa (Fig. 2A,2B). Low echogenicity and variable prominence of the submucosal layer may be associated with submucosal edema, which is known to occur in IHPS [2]. Researchers in the earlier period of sonographic examination of the hypertrophied pylorus, limited by equipment constraints and a small base of experience, mistook the edematous submucosa within the canal for fluid traversing the pyloric channel [16]. However, as early as 1984, pathologic interpretation of these sonolucent intrapyloric bands was theorized to represent hypertrophy of the muscularis mucosae [17].
In patients with IHPS, sonographic correlation with the findings at upper gastrointestinal examination confirms that the lumen of the pyloric canal on cross-section measures approximately 4 mm on the average, a diameter that is similar to or larger than that of the normal open pyloric ring, and that this lumen, unlike the normal pyloric ring, is obstructed by the folds of hypertrophied intraluminal mucosa and edematous submucosa. The fact that the obstructed pyloric channel can often be traversed by a stiff enteric tube [7] illustrates the adequate size of the canal and the pliability of the obstructing mucosal folds.
Endoscopic and Surgical Correlation
Recent experience with endoscopy has permitted further evaluation of the
lumen of the pyloric canal in IHPS and provided a visual antral correlate to
the sonographic images. The criteria for diagnosis of IHPS in an endoscopic
series include a cauliflowerlike narrowing
[18] of the gastric entrance
to the pyloric canal (Fig. 5),
identical to that which we have reported and shown at biopsy to represent
hypertrophied and edematous antropyloric mucosa
[12]. After the muscle is
surgically divided, the mucosa "bulges" through the myotomy
incision [19]. These surgical
and endoscopic findings, although indicating the constancy of mucosal
redundancy, do not directly address the question of mucosal redundancy versus
true hypertrophy.
|
Pathologic Correlation
Early pathology reports described an obstructive role played by redundancy
of the mucosa, stressed by Heubner and Fredet in 1906 and 1908
[20]. However, most recent
pathologic data and direct observations of the pyloric anatomy are limited to
external observations during pyloromyotomy and to muscle biopsies obtained
during surgery. Thus, information regarding the pyloric lumen in IHPS is
limited; it is assumed that the thickened pyloric muscle narrows the pyloric
lumen, resulting in the anatomical obstruction that leads to vomiting in these
infants.
However, pediatric pathology texts have described constant pyloric submucosal edema of variable degree and exaggeration of the longitudinal rugae of the pyloric mucosa as contributing significantly to the obstruction [2]. The Ramstedt operation, described in 1912 [20], "proved that transverse closure of the pyloromyotomy incision was unnecessary and probably harmful since it contributed to further obstruction through folding of the redundant mucosa." A mild-to-moderate inflammatory cell infiltration is also described as being a constant finding in the submucosal and muscularis layers of the antropyloric channel [2]. On the other hand, biopsies at the greater curvature of the body of the stomach in these patients show no difference from controls [21].
Because of the success of the Ramstedt operation, few cases in the recent literature describe pathologic correlation of the mucosal abnormalities. One reported case described a prominent filling defect within the lumen of the obstructed pyloric canal on upper gastrointestinal meal test [22]. The article documented and illustrated a large filling defect that was seen protruding into the gastric antrum of that infant. A full-thickness excision and histologic examination of the lesion revealed the hypertrophied and edematous mucosa to be nearly twice the thickness of the muscle (Fig. 6). The exaggerated mucosal thickness described in this patient resembles that in our patient, illustrated in Figure 4A,4B; both patients did well after an otherwise routine pyloromyotomy. Other patients with histologic findings similar to those of our typical patient, illustrated in Figure 3A,3B, with branching glands and submucosal edema, have also been reported and classified as exhibiting idiopathic focal foveolar hyperplasia in association with pyloric stenosis. Those patients also did well after routine pyloromyotomy [23]. Thus, pathologic information suggests that the crowded mucosa, with submucosal edema, inflammatory cellular infiltration, and branching glandular structure is abnormal.
|
Other Associated Findings
Considerable attention has been focused over the past decade on the
muscular layers of the hypertrophied pylorus. The muscular layer has been
found to be deficient in the quantity of nerve terminals
[6], markers for
nerve-supporting cells [24],
peptide-containing nerve fibers
[1,
25], nitric oxide synthase
activity [7], mRNA for nitric
oxide synthase production
[26], and interstitial cells
of Cajal [27,
28]. It also contains
increased expression of insulinlike growth factor ImRNA
[29] when compared with
control specimens. These studies localize the abnormalities largely to the
muscular layer, particularly the circular layer, with fewer or no changes
described in the myenteric plexus. It is postulated that this abnormal
innervation of the muscular layer leads to failure of relaxation of the
pyloric muscle, increased synthesis of growth factors, and subsequent
hypertrophy, hyperplasia [30],
and obstruction.
Prostaglandin E2 generation in the gastric mucosa and its concentration in the gastric secretions of patients with pyloric stenosis have been reported to be significantly greater than in normal controls [31]. Prostaglandin E1 and E2 induce proliferation of the gastric mucosa [32] and are related to muscle contraction in the human gastrointestinal tract [33]. Prostaglandin therapy has been found to result in significant antropyloric mucosal hypertrophy, leading to a reported case of muscular hypertrophy and symptomatic pyloric stenosis requiring surgery [34]. The degree of mucosal hypertrophy illustrated in that patient is very similar to that in all of the consecutive patients in our study.
We have shown the additional associated finding of mucosal thickening filling and obstructing the lumen of the pyloric canal in patients with pyloric stenosis. However, the question of whether this mucosa is merely compressed normal gastric rugae, or whether it is abnormal, needs to be addressed. We believe that some degree of crowding of the mucosa is inevitable, although the obstructed lumen of the hypertrophied pyloric canal is similar to or wider than that of the nonobstructed pyloric ring, suggesting either that the mucosa is indeed thicker than normal or that it is the normally wider, prepyloric antral portion of the stomach that undergoes the changes found in infants with IHPS. However, the filling defect noted on upper gastrointestinal barium contrast examination, distinct from that of pylorospasm, lends credence to the hypothesis that the mucosa is thicker in patients with IHPS. The variable and sometimes greatly exaggerated diameter of the mucosa again points to an inherent change in this component of the lesion. Finally, review of histologic findings, both in older pathologic literature and in the more sparse recent literature, indicates that changes of submucosal edema, inflammatory cellular infiltrate, and gland hypertrophy are present; these are confirmed in our patient who otherwise had a typical course of treatment and findings of IHPS (Fig. 3A,3B).
Cause or Effect?
We have shown the presence of mucosal thickening in the pylorus of patients
with pyloric stenosis, whereas other investigators have identified multiple
abnormalities within the hypertrophied muscle layers that in turn have been
implicated in the causality of the lesion. It is difficult to determine
whether associated findings are the initiation or the result of an abnormality
[8]. It is known that infants
who subsequently develop IHPS are born with a normally functioning pyloric
sphincteric mechanism before oral feeding commences, because neither a
functional obstruction nor an anatomic abnormality is present at birth
[5]. Therefore, it is more
difficult to postulate that the associated muscle abnormalities are
preexistent, or to disregard a stimulus initiated at or soon after birth.
One is thus tempted to ask the obvious question: What is it about the initiation of oral feedings after birth that results in hypertrophy of the pyloric mucosa and muscular layers and in the multiplicity of abnormalities in the sphincteric relaxation mechanism? It seems reasonable to hypothesize that acid secretion induced by the ingested material in contact with the gastric mucosa may in turn be related to the initiation of the antropyloric changes present in infants who develop IHPS.
Maternal administration of pentagastrin in dogs has been implicated in an abnormality in pups that resembles IHPS in humans [9]. Hypergastrinemia in humans has been associated with immaturity of the gut during the first 3 months of life [35]. Furthermore, lactation has resulted in increase in mucosal mass and villus crypt formation in infant animals [36]. Many of these known clinical features of IHPS could be explained by a theory postulating increased antropyloric-acid secretory activity in these infants [11]. It is possible that a genetic propensity towards hyperacidity, augmented by gut immaturity in the first 3 months of life and potentiated by the initiation of oral feedings, may lead to mucosal hypertrophy and the beginning of gastric outlet obstruction. Obstruction, in turn, may lead to a further increase in acidity, increased synthesis of prostaglandins, and antropyloric hypermotility, with subsequent muscular hypertrophy and further obstruction, constituting a vicious circle interrupted by relief of obstruction via pyloromyotomy.
In summary, we have shown that significant mucosal redundancy obstructs the lumen of the pyloric channel in IHPS and that its presence should not lead to an unrelated diagnosis such as gastric polyp. To assess the potential implications of this finding, we have reviewed the existing literature investigating various abnormalities implicated in the pathogenesis of IHPS within the context of our findings. Although redundancy and hypertrophy are not synonymous, data suggest that in IHPS the mucosa is not only crowded within the canal, but may, in fact, be primarily abnormal. It is difficult to tease out a precipitating event from a multiplicity of associated abnormalities, but the uniform presence of pyloric mucosal thickening cannot be ignored and begs further study. The absence of pyloric obstruction and the normal anatomy of the pylorus at birth suggest that the mucosal changes, initiated by oral feedings and mediated by gastric hypersecretion, could be implicated among the initiating events in the development of IHPS in predisposed infants.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Hernanz-Schulman, Y. Zhu, S. M. Stein, R. M. Heller, and L. A. Bethel Hypertrophic Pyloric Stenosis in Infants: US Evaluation of Vascularity of the Pyloric Canal Radiology, November 1, 2003; 229(2): 389 - 393. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Hernanz-Schulman Infantile Hypertrophic Pyloric Stenosis Radiology, May 1, 2003; 227(2): 319 - 331. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
