Practical pathology and genetics of Hirschsprung's disease

Seminars in Pediatric Surgery (2009) 18, 212-223

Practical pathology and genetics of Hirschsprung’s disease Raj P. Kapur, MD, PhD From the Department of Laboratories, Seattle Children’s Hospital, and Department of Pathology, University of Washington, Seattle, Washington. KEYWORDS Hirschsprung’s disease; Pathology; Rectal biopsy; Transitional zone; Acetylcholinesterase; Calretinin

Diagnosis and management of Hirschsprung’s disease (HSCR) requires understanding of the malformation’s anatomic features and multigenic nature. Rectal biopsies, intraoperative frozen sections, and resection specimens provide invaluable information. Extraction of these data requires thoughtful biopsy technique, adequate histologic sections, histochemistry, and collaboration of surgeon and pathologist. Critical consideration of transition zone anatomy and published studies of “transition zone pull through” indicate that more research is needed to determine how much ganglionic bowel should be resected from HSCR patients. Many HSCR-susceptibility genes have been identified, but mutational analysis has limited practical value unless family history or clinical findings suggest syndromic HSCR. © 2009 Elsevier Inc. All rights reserved.

Basic anatomy Hirschsprung’s disease (HSCR) is a congenital malformation defined as the absence of ganglion cells in the myenteric and submucosal plexuses of the terminal rectum ⫾ more proximal bowel. HSCR is subdivided based on the rostral extent of aganglionosis into ultrashort- (distalmost rectum), short- (rectosigmoid), and long-segment (proximal to splenic flexure), although some refer only to short- and long-segment subtypes.1 The presence or absence of submucosal ganglion cells generally correlates with the status of ganglion cells in the adjacent myenteric plexus.2,3 The interface between ganglionic and aganglionic gut is often irregular, such that ganglion cells may extend 2-3 cm farther distally along a portion of the bowel circumference, usually on the antimesenteric side (Figure 1). In most cases, the junction between aganglionic and euganglionic bowel is notable for a transition zone in which the myenteric plexus is hypoganglionic and the submucosal plexus is hyper-, eu-, or hypoganglionic. The length of this transition zone can be Address reprint requests and correspondence: Raj P. Kapur, MD, PhD, Department of Laboratories, A6901, Seattle Children’s Hospital, and Department of Pathology, 4800 Sand Point Way NE, Seattle, WA 98115. E-mail address: [email protected].

1055-8586/$ -see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1053/j.sempedsurg.2009.07.003

difficult to define precisely, in part because reliable practical standards for establishing a normal density of ganglion cells do not exist.4 However, obvious abnormalities in the density and distribution of myenteric ganglion cells typically extend several centimeters proximal to the aganglionic zone and occasionally are present over a length of more than 10 cm.5-7 Most, if not all, cases of HSCR result from failure of neural crest cells to fully colonize the gut during embryogenesis, usually in the context of one or more genetic susceptibility factors. The molecular and cellular events that are critical to embryology of the enteric nervous system and the pathogenesis of HSCR are reviewed by Burns and coworkers elsewhere in this issue. The present review focuses on practical aspects of the pathology and genetics of HSCR, primarily for the benefit of pediatric surgeons and their colleagues, who diagnose and treat this disorder.

Rectal biopsies Submucosal biopsies Rectal biopsy is the most common procedure used to diagnose HSCR. It is predicated on the observation that submu-

Kapur

Hirschsprung’s Pathology and Genetics

Figure 1 Diagrammatic illustration of the distribution of ganglion cells in the euganglionic, transition, and aganglionic zones in HSCR. The interface between aganglionic and ganglionic bowel (fine dashed line) is usually irregular and frequently differ by as much as 3 cm along aspects of the bowel circumference. The transition zone is characterized by myenteric hypoganglionosis (and possibly submucosal hyperganglionosis) and may extend proximally for relatively long distances (⬎10 cm) before a normal density of ganglion cells is established.

cosal ganglion cells are invariably absent from the distal rectum of a patient with HSCR and that this submucosal phenotype correlates with absence of myenteric ganglion cells. In neonates and young infants, suction rectal biopsy is the favored approach, as contemporary biopsy guns yield diagnostic specimens with minimal morbidity. The suction device relies on vacuum to elevate and excise an adequate amount of submucosa.8 In older patients, the procedure is more prone to produce inadequate specimens, probably because the submucosa is more fibrous. A “deep” biopsy with forceps or another instrument is a reasonable alternative, particularly if suction biopsy fails. To exclude HSCR, clinicians are taught to biopsy 2-3 cm proximal to the pectinate line (transition between rectal and squamous mucosa) because a number of studies have demonstrated that the distal 1-2 cm of rectum is normally hypoganglionic, and a justifiable concern exists that sampling of this area may lead to a false impression of aganglionosis.5,9-11 In practice, although the distal rectum is hypoganglionic, ganglion cells can often be found in a distal biopsy from a patient who does not have HSCR if the biopsy size is adequate and sectioned thoroughly. However, this can require examination of more than a hundred histologic sections to find a single unequivocal ganglion cell. Cognizant of the physiological submucosal aganglionosis that exists in the terminal rectum, Aldridge and Campbell recommended that at least two biopsies should be obtained 2-3 cm proximal to the pectinate line.9 This recommendation is generally considered the minimal requirement in most practices. However, no universal standard exists, and some practitioners advocate more extensive sampling, as discussed below. When operating the biopsy device, the precise location of the biopsy port relative to the dentate line is difficult, and it is not uncommon to find that a biopsy was either at or below

213 the squamocolumnar junction. Presumably, similar deviation may occur in the oral direction, in which case biopsies taken more proximally risk missing a short segment of aganglionosis. For these reasons, the author and others advocate biopsies from multiple levels (eg, 1, 2, and 3 cm proximal to the dentate line) to increase the likelihood that adequate tissue is obtained and reduce the likelihood that very short-segment disease will be overlooked.12 This strategy has effectively eliminated a problem with inadequate biopsies because, even if the “1 cm” biopsy is too low (squamous mucosa), the others are not and the likelihood that all three biopsies will have inadequate submucosa is negligible. From the surgical pathology perspective, HSCR is excluded if one or more ganglion cells is identified in the submucosa of a distal rectal biopsy. Conversely, the diagnosis of HSCR can be established with confidence, when hypertrophic nerves, but no ganglion cells, are identified in an adequately sectioned biopsy. Assessment of specimen adequacy and recognition of ganglion cells and hypertrophic nerves is influenced by experience and regular practice. Even the most experienced pathologist will encounter some biopsies that yield equivocal results, usually because an inadequate amount of submucosa is present. Above all, a pathologist must have the confidence to distinguish diagnostic from equivocal findings and clearly communicate the results to the clinician. In some instances, rebiopsy may be necessary. Two different approaches have evolved for the pathologic evaluation of rectal biopsies. The first, which is based solely on analysis of enzyme histochemistry using frozen sections to identify ganglion cells and acetylcholinesterase (AChE)-positive nerves, is outlined by Feichter, Bruder, and Meier-Ruge in this issue. This strategy was pioneered by Meier-Ruge and colleagues and is employed by a relatively small number of laboratories in Europe and other parts of the world. The more widely used approach relies primarily on paraffin sections stained with hematoxylin and eosin (H&E), although many laboratories complement the latter with AChE histochemistry and/or paraffin-based immunohistochemistry.4 Although well-controlled comparative data are not available, the different approaches appear to be equally accurate and reliable. Adequacy of the biopsy and technical and interpretational experience of the laboratory are probably the most important variables to influence diagnostic accuracy. At a minimum, a diagnostic biopsy should measure ⬃3 mm in diameter and be at least one-third submucosa (Figure 2A). When properly oriented and sectioned adequately (50-75 sections), H&E-stained, paraffin-embedded sections are generally sufficient to exclude the presence of submucosal ganglion cells and suggest the diagnosis of HSCR.12 The presence of multiple hypertrophic nerve fibers (⬎40 ␮m diameter) is observed in many, but not all cases, and helps establish the diagnosis (Figure 2B).13 The hypertrophic nerves that exist in most patients with HSCR arise from

214

Seminars in Pediatric Surgery, Vol 18, No 4, November 2009

Figure 2 Histopathology of HSCR. (A) An adequate suction rectal biopsy should be ⱖ3 mm in greatest dimension, and at least one-third of the biopsy should be submucosa. (B) Hypertrophic nerves (ⱖ40 ␮m) are typically present in the aganglionic submucosa in HSCR. (C) A biopsy with ganglion cells contains sparse thin AChE-positive nerves (not easily resolved in black and white image) in the muscularis mucosae and lamina propria. (D) In contrast, coarse large AChE-positive nerves (arrowheads) are present in a biopsy from a HSCR patient. (E) Calretinin immunoreactivity exists in small nerves in the muscularis mucosae and lamina propria (arrowheads), when ganglion cells are present. (F) No calretinin immunoreactivity is present in the mucosa or superficial submucosa of aganglionic bowel. Scale bars: (A) 1 mm; (B) 40 ␮m; (C-F) 25 ␮m.

extrinsic autonomic and sensory fibers, which enter along with vessels from the perirectal region and project for a finite distance rostrally.14 It is the number and diameter of these fibers that increase in HSCR, giving rise to the “hypertrophic” nerves that are frequently, but not always, observed in the myenteric plexus, submucosa, and mucosa of HSCR patients. Because these fibers project only a finite distance proximal to the rectum, hypertrophic innervation may not be observed in biopsies taken rostrally in longsegment disease.14 Furthermore, extrinsic nerve hypertrophy of the rectum may not be observed in patients with combined deficiency of intrinsic ganglion cells and other peripheral ganglia (more common with long-segment HSCR) or very premature infants with delayed extrinsic innervation.15 Unfortunately, some biopsies are suboptimal due to paucity of submucosa, crush artifact, or both. In several series dating back to the original suction rectal biopsy paper by Dobbins and Bill,16 reported rates of inadequate biopsies have generally hovered between 10% and 20%.17-19 It is in

these cases that AChE histochemistry can be particularly valuable.

Acetylcholinesterase histochemistry Histochemical staining for AChE activity is a useful adjunct for the diagnosis of HSCR. The procedure is only performed with frozen sections and therefore requires an additional suction biopsy if paraffin sections are also going to be evaluated. The traditional protocol for AChE staining requires approximately 90 minutes,20 but rapid procedures have been developed that require 5-10 minutes.21,22 AChE staining in the rectum of normal children includes staining of nerves in the submucosa and small fibers in the muscularis mucosae (Figure 2C). Usually the latter are confined to the inner half of the muscularis mucosae.23 If fibers are positively stained in the lamina propria, they are extremely thin and few. Most, but not all, of the rectal submucosal biopsies from HSCR patients contain more densely packed,

Kapur

Hirschsprung’s Pathology and Genetics

large, AChE-positive fibers through the full thickness of the muscularis mucosae (Figure 2D). In addition, prominent AChE-positive fibers are often present in the lamina propria. The latter finding should not be relied on too heavily because hypertrophic nerves are confined to the muscularis mucosae in biopsies from a significant subset of HSCR patients, particularly young infants.

Immunohistochemistry and the diagnosis of Hirschsprung’s disease A sizable number of papers have been published that essentially describe some neuron-specific marker that the authors suggest could be used to facilitate the diagnosis of HSCR.24 Although most of these antibodies are fairly specific, none is used widely in practice because pathologists who regularly search for ganglion cells in H&E-stained sections are quite good at discriminating neuronal cell bodies without the need for special stains. False-positive diagnoses result either from inadequate sampling or observer inexperience. Among the many antigens expressed by enteric neurons, the calcium-binding protein calretinin is a potentially useful immunohistochemical target for immunohistochemical studies that complement traditional H&E-based diagnosis of HSCR.25 Calretinin immunoreactivity is normally present in small intrinsic nerves located in the muscularis mucosae and lamina propria (Figure 2E). In the mucosa of aganglionic bowel, expression is completely absent (Figure 2F). The results of a recent case-control study suggested that calretinin immunohistochemistry is at least as sensitive and specific as AChE histochemistry for the diagnosis/exclusion of HSCR from suction rectal biopsies.26 In contrast with AChE staining, which requires a second frozen biopsy and specialized methods, calretinin immunostaining can be performed on paraffin sections of formalinfixed biopsies with methods that are available in most pathology laboratories.

215 plished by sequential seromuscular biopsies, from distal to proximal. Seromuscular biopsies should be a minimum of 1 cm in length and extend for a depth of 3-5 mm, to include the longitudinal and most of the circular layers of the muscularis propria. Proper orientation of the biopsy for frozen sections greatly facilitates sampling and identification of ganglion cells. The goal is to cut perpendicular to the serosal surface, thereby visualizing the both muscle layers and their interface in the histologic sections. With a well-oriented biopsy, ⱕ10 sections are generally sufficient to confirm/ exclude aganglionosis. Recognition of ganglion cells is usually not difficult, although inflammation sometimes obscures their cytologic features. In the operating room, the surgeon must be aware of limitations to the seromuscular biopsy.1 The biopsy only examines a portion of the circumference. Because the transition from ganglion cells to aganglionic bowel typically extends up to 3 cm more distally along some part of the circumference (Figure 1),2,3 one should not infer from a small seromuscular biopsy that the entire circumference contains ganglion cells. However, it is likely that bowel ⱖ3 cm proximal to any ganglion cell-containing seromuscular biopsy will have ganglion cells around the entire circumference. Therefore, it may be prudent to routinely resect at least 3 cm of bowel proximal to a “positive” biopsy site for ostomy placement or pull-through.2 It is difficult, if not impossible, to distinguish a seromuscular biopsy of euganglionic bowel from hypoganglionic transitional zone, with the possible exception of severe hypoganglionosis, in which case ganglia consist of individual ganglion cells. Delineation of mild-to-moderate hypoganglionosis requires nearfull circumference tissue samples and rigorous neuronal counts, which are not possible intraoperatively. Nonetheless, intraoperative examination of a frozen section that represents the full circumference of the proximal resection margin can assess the general distribution of ganglion cells and give some insight into whether the distribution of ganglion cells is relatively uniform, and thereby reduce the likelihood of a transition zone pull-through (TZPT).

Transition zone pull-through

Intraoperative frozen sections Many surgical approaches to HSCR are employed with a trend toward one-step procedures that are often transanal. In other instances, diagnosis based on suction rectal biopsy is followed by a two-stage procedure that begins with placement of an ostomy proximal to the aganglionic segment. Intraoperative seromuscular biopsies are important to determine that ganglion cells are present at the level where the ostomy or anastomosis will be placed. During ostomy placement, aganglionic gut may not be biopsied intraoperatively. However, whether the definitive surgery is done in one or two stages, it is helpful if the boundary between aganglionic and ganglionic gut is localized intraoperatively before resection of any bowel. Generally, this is accom-

Resection of aganglionic bowel followed by an anal “pullthrough” procedure is the treatment for most cases of HSCR. However, chronic postoperative problems, including soiling, constipation, diarrhea, and enterocolitis, are reported in many patients in various series and appear equally common with the three major types of surgical anastomosis.27 Obstructive symptoms, in particular, are reported in 10% to 40% of patients.28 Potential causes for persistent constipation are scarring/stenosis at the anastomotic site, “acquired aganglionosis” due to ischemia around the anastomosis, “skip lesions” (coincident aganglionosis of the cecal region), dysfunction of the ganglionic bowel, and TZPT.28 The transition zone is a segment of variable length, which is partially aganglionic and/or hypoganglionic and

216 lies immediately proximal to the aganglionic zone in HSCR. TZPT implies incorporation of transitional zone bowel (partially aganglionic and/or hypoganglionic) into the anorectal anastomosis (or ostomy). TZPT is generally assumed to be a cause of post pullthrough obstruction and is touted as an indication for surgical revision by some authors.28 For example, Boman and coworkers (2007) evaluated the resection specimens from 41 HSCR patients and measured the length of the resection, length of the transition zone, and length of the aganglionic zone.29 They defined the transition zone as “insufficient numbers of ganglionic myenteric nervous plexuses (hypoganglionosis) proximal to the aganglionic segment in HD, sometimes with an excess of ganglion cells in the submucosa (hyperganglionosis).” However, no quantitative or objective criteria were presented to define “hypoganglionosis.” They also “estimated the percentage of the circumference of the muscularis propria at the proximal end of the resection specimens that contained ganglionic nervous plexuses irrespective of the number of neuronal cells associated with the plexuses.” If transitional zone is defined as myenteric ganglion cells present in less than 100% of the bowel circumference, their data indicate that the transition zone extended to the proximal margin in 33 of 39 cases (2 discarded for inadequate data). More importantly, their data suggest that the risk for postoperative complications is greatest for patients whose proximal margins have ganglion cells distributed around less than 80% of the circumference (Figure 3). Interestingly, the distance between the aganglionic zone and proximal margin in all the latter cases was less than 3 cm, which reinforces the notion that resection ⱖ3 cm proximal to a “positive” seromuscular biopsy may be a good surgical guideline.

Figure 3 Potential long-term complication of TZPT. Data extracted from Table 1 in Boman and coworkers (2007) are graphed to show the relationship observed between the distance from the proximal surgical margin to the aganglionic segment (x axis), fraction of the circumference stated to contain ganglion cells (y axis), and either the presence (open symbols) or absence (filled symbols) of long-term postoperative obstructive symptoms. In general, long-term obstructive symptoms were more common in patients with aganglionosis of more than 20% of the proximal surgical margin and ⱕ3 cm between the surgical margin and aganglionic zone (dashed line).

Seminars in Pediatric Surgery, Vol 18, No 4, November 2009 Including the aforementioned study, I am unaware of research on TZPT with appropriate controls, unbiased pathologic review, and objective histopathological criteria to define transition zone. Furthermore, each of the surgical procedures currently used to construct an anastomosis retains a small amount of aganglionic bowel. Therefore, it is unclear why incorporation of some hypoganglionic gut and/or a bit more aganglionic gut is significant. More research is needed to resolve this important question.

Clinical relevance of post-resection pathology Analysis of Hirschsprung’s disease resections The pathologist’s goals in analyzing resected aganglionic gut from an HSCR patient are to confirm the diagnosis of HSCR, map the transitional zone, and assess the integrity of the nervous system at the proximal end of the resection. Confirmation that the distal gut is aganglionic is straightforward. A map of the transition zone can be completed by sampling multiple areas along the length of the resected segment to document the presence/absence of submucosal and myenteric ganglia. Some pathologists prepare rolls from longitudinal strips of the entire length of the resection.20 I prefer to use transverse sections because the circumferential interface between aganglionic and ganglionic gut is often irregular, which will not be apparent in any single longitudinal strip.2,3 Full-circumference sections are easier to interpret and are particularly useful when the aganglionic zone is close to the proximal resection margin. Evaluation of the integrity of the proximal gut is one of the most challenging aspects of HSCR pathology. In principle, the genetic etiologies that produce distal aganglionosis may have more subtle effects on the number, distribution, circuitry, and/or differentiation of proximal neurons. Certainly, most HSCR patients harbor a transitional zone of variable length, in which the density of myenteric ganglia is obviously less than in normal gut. However, mild to moderate changes in neuronal density are extremely difficult to diagnose by routine analysis, but are hypothetical basis for the persistence of HSCR-like symptoms postoperatively. Given the diverse functions of the products of the HSCR susceptibility genes (discussed below), it is also conceivable that functionally significant alterations in neural differentiation or other properties may exist in the “euganglionic” proximal bowel of some HSCR patients. Experimental evidence with some animal models for HSCR suggests that such subtle alterations probably exist, but their recognition requires sophisticated methods and appropriate normative data.30,31 As it is very unlikely that routine histopathology will resolve these alterations, translational research is needed to determine whether molecular genetic, immunohistochemical, or other methods can detect clinically significant alterations in the ganglionic bowel of HSCR patients.

Kapur Table 1

Hirschsprung’s Pathology and Genetics

217

Miscellaneous histopathological findings in HSCR

Histopathological finding

Location

Reference(s)

Eosinophilic neural infiltrates

Aganglionic and transition zones; transmural, including nerve plexuses Transition zone Conflicting data

38

Aganglionic and transition zones Aganglionic zone Aganglionic, transition, and/or euganglionic

43 44-46 47,48

Submucosal arterial fibromuscular dysplasia Loss of c-kit immunoreactive interstitial cells of Cajal Peripheral nerve pattern of laminin expression Loss of nNOS innervation in muscularis propria Altered immunoreactivity of cytoskeletal proteins in muscularis propria

Hirschsprung’s disease-associated intestinal neuronal dysplasia In addition to hypoganglionosis, submucosal hyperganglionosis (intestinal neuronal dysplasia, type B; IND) has been reported in the proximal gut of 20% to 75% of HSCR patients.32 IND is a controversial form of submucosal hyperganglionosis that has been defined entirely based on histochemically stained biopsies.33,34 Similar changes have been observed as isolated “neuropathy” in some children with HSCR-like symptoms, as well as in contexts of other primary disorders of dysmotility. The significance of Hirschsprung-associated IND is hotly debated in the literature; some authors have advocated screening for IND with frozen sections at the time of surgery or as a staged procedure, to extend the resection proximal to the affected area.32,35,36 However, no compelling data exist to suggest that IND-like changes in HSCR either predict a poor outcome or should be managed any differently from isolated HSCR. The study by Estevao-Costa and coworkers (2006) warrants additional comment because a significant effort was made to compare postoperative entercolitis rates in Hirschsprung patients with traditional resections of their aganglionic bowel (group I) versus a cohort who underwent preoperative mapping of their gut for IND before resection of the aganglionic bowel and any proximal gut with IND pathology (group II). Group II patients with IND had a lower rate of enterocolitis than group I patients with IND, leading to the suggestion that resection of proximal bowel involved by IND may be beneficial. The results are interesting but the study has several limitations, including a nonrandomized process for patient selection, lack of detailed information regarding the characteristics of the two groups (eg, lengths of aganglionic segments, genders, associated anomalies, etc), no statement that the pathology results for group I were obtained in an unbiased manner (“blinded” review), and no statistically significant difference in the overall rate of enterocolitis between the two groups. In contrast, Haricharan and coworkers found no relationship between the length of ganglionated bowel in 36 HSCR resections and postoperative enterocolitis, although the prevalence of IND was not examined.37 The role of IND as a marker or cause of postop-

39 40-42

erative enterocolitis should be investigated further before changes to the traditional surgical approach are made. A variety of other poorly understood histopathological findings are observed in aganglionic bowel and or the transition zone. Some of these are summarized in Table 1. Although most of these findings have no established clinical and/or genetic significance, their potential correlations with specific genetic defects and/or postoperative complications have not been adequately studied.

Enterocolitis Hirschsprung’s disease-associated enterocolitis (HAEC) is a serious complication that can occur pre- or postresection of the aganglionic bowel. HAEC manifests clinically as abdominal distension and diarrhea, usually accompanied by fever and often bloody stools.49 The pathogenesis is unclear and may be multifactorial.50 Histopathological features include inflammation and necrosis, which span the full thickness of the bowel wall in severe cases. Pathologic findings are often patchy and may affect the aganglionic or ganglionic segments, or both. Teitelman and coworkers (1989) devised a histopathological grading system for HAEC, which may be applied to suction rectal biopsies and resection specimens.51 Using this system, they were able to identify patients with histopathological HAEC in their suction rectal biopsies before they became symptomatic. However, identification of high-grade enterocolitis in pullthrough specimens did not correlate well with clinical postoperative HAEC.

Unusual anatomic variations of Hirschsprung’s disease Skip lesions and zonal aganglionosis Two rare forms of congenital aganglionosis deviate from the classic pattern in which the distal rectum and uninterrupted contiguous bowel are devoid of nerve cell bodies. In the intestinal tracts of persons with “zonal” (“segmental”)

218 aganglionosis, ganglion cells are present in the distal rectum but are absent from a proximal segment of gut.52 In contrast, “skip areas” (also termed “double-zonal aganglioosis”) are ganglion cell-containing segments of large intestine, flanked proximally and distally by aganglionic gut.53 Zonal aganglionosis is considered an acquired lesion (disruption) that results when ganglion cells (or their precursors) in a fully colonized segment of gut degenerate due to ischemic, viral, immunologic, or other types of injury.52,54 The aganglionic segment can occur in small or large intestine. In some cases, a specific etiology is suggested by history (eg, necrotizing enterocolitis) or other pathologic findings (eg, viral cytopathy). Alternatively, it has been suggested that zonal aganglionosis might result from failure of vagal and sacral crest cells to converge in the gut wall.55 At the time this hypothesis was introduced, it was less certain that sacral crest cells participate in enteric neurodevelopment. Given recent evidence that a subset of colonic neurons derive normally from the sacral crest (Burns and coworkers, this issue), the proposal is more tenable. Some colleagues and I reviewed the subject of skip areas in 1995 and found that only 11 cases had been reported.53 Since then, I have been informed of several other cases, and I suspect the entity may be more common than the literature might suggest. With rare exception, skip areas are located in the large intestine and bracketed by aganglionic areas that invariably include the distal rectum and appendix. Pathologists and surgeons must be cognizant of skip areas and not use biopsies of the appendix as a means to diagnose total colonic aganglionosis since relatively large skip areas can be present in which ganglion cells exist. In at least 1 patient, the skip area was recognized, preserved, and used to establish a functional anastomosis between small intestine and anus.56

Ultrashort-segment Hirschsprung’s disease Ultrashort-segment Hirschsprung’s disease (US-HSCR) is a very confusing topic, plagued by the fact that multiple different definitions have been applied to the same term. Some regard US-HSCR as a manometric diagnosis characterized by absence of a rectosphincteric reflex response to rectal balloon inflation.57,58 Patients who fulfill this diagnostic criteria may have a normal distribution of ganglion cells and normal AChE-staining results.59,60 The latter situation is better termed internal sphincter achalasia to distinguish it from HSCR, given that aganglionosis is not observed. Deficient nitric oxide synthase (NOS)-immunoreactive nerve fibers have been reported in the sphincters of patients with internal sphincter achalasia.61 Alternative definitions of US-HSCR refer to patients with one of the two following sets of findings: 1. Complete absence of ganglion cells in the terminal 1-4 cm of the rectum ⫾ abnormal AChE staining ⫾ an adjacent hypoganglionic transitional zone62

Seminars in Pediatric Surgery, Vol 18, No 4, November 2009 2. An abnormal AChE staining pattern in the terminal 1-4 cm ⫾ aganglionosis of the terminal rectum ⫾ an adjacent hypoganglionic transitional zone63 Clearly, both of these definitions encompass patients with terminal aganglionosis and HSCR-like AChE histochemistry. The difference rests on which of these two findings is the minimal diagnostic requirement. The first variant (complete absence of ganglion cells in the most distal rectum) fits conceptually well into the spectrum of classic HSCR but is challenging to diagnose, particularly from biopsies, because hypoganglionosis (and possibly aganglionosis) appears to be normal in the terminal 1-2 cm of the rectum. A conservative approach to US-HSCR is to merge the two definitions and require both aganglionosis and an abnormal AChE staining pattern, as is typical of most distal rectal biopsies from short-segment HSCR.

Genetics of Hirschsprung’s disease The genetic factors that predispose to HSCR are heterogeneous and exhibit complex interactions that influence penetrance and severity (length of the aganglionic segment, severity of obstructive symptoms) of the malformation. Because of comprehensive efforts by several groups to understand the molecular and embryologic bases for this complexity, HSCR genetics has become a paradigm for the study and understanding of multigenic disorders. However, despite progress that has been made, mutational analysis has a limited role in the management of most HSCR patients and their families. The text that follows summarizes some of the clinically relevant findings, and the reader is referred to Amiel and coworkers (2008) for a detailed review of this subject.64 It is clear that most, if not all, cases of HSCR have a genetic basis, but risk of the malformation is determined by mutations of one of many HSCR-susceptibility genes, or more often coincident alterations of multiple such genetic loci. Mutations in more than 11 different genes have been implicated in the pathogenesis of HSCR (Table 2); many were first recognized in murine models for this condition. As expected, the products of many of these genes influence critical cellular events during embryogenesis of the enteric nervous system (Burns and coworkers, this issue). This complexity makes it difficult to counsel individual patients or families. However, HSCR patients can be subdivided based on severity and/or associated findings into some broad groups with clinically relevant genetic differences.

Isolated versus syndromic Hirschsprung’s disease HSCR most often occurs as an isolated (sporadic, nonsyndromic) malformation. Comprehensive analysis of most or all HSCR susceptibility genes is rarely performed in such cases, but results from studies of individual loci suggest that a mutation that affects the coding sequence of any one of

Kapur

Hirschsprung’s Pathology and Genetics

Table 2

219

Susceptibility genes in isolated and syndromic HSCR

Syndrome

Gene

% with HSCR

Isolated HSCR*

RET, EDNRB,† EDN3† TTF-1

100% 100%

RET‡

⬍1%65

Syndromic HSCR MEN2A

16%66

Smith–Lemli–Opitz

DHCR7

Down

Trisomy 21

2%-9%67

Waardenburg-Shah Mowat-Wilson

SOX10, EDN3,† EDNRB† ZHFX1B

100%§ 62%68

Haddad Goldberg-Shprintzen

PHOX2B¶ KIAA1279

100%储 100%

X-linked hydrocephalus

L1CAM

NE

Cartilage-hair hypoplasia**

RMRP

NE

Bardet–Biedl

BBS1, BBS2, BBS4, BBS6,†† BBS7, BBS8 MKKS††

Kauffman-McKusick

2%69 NE

Common phenotypic features TTF-1 mutations have been associated with hypothyroidism and cleft palate, but the only reported HSCR patient with a TTF-1 mutation had isolated HSCR. Medullary thyroid carcinoma, pheochromocytoma, parathyroid hyperplasia Growth retardation, pedal syndactyly, mental retardation, hypospadius, dysmorphic facies Prominent epicanthal folds, upslanting palpebrae, hypotonia, mental retardation, flat midface, single transverse palmar crease Deafness, piebaldism, other neurological deficits Abnormal facies, cardiac malformations, mental retardation, genitourinary anomalies Congenital central hypoventilation, neuroblastoma Microcephaly, various brain malformations, cleft palate Cerebral aqueductal stenosis, hydrocephalus, absent corpus callosum Skeletal dysplasia, sparse blond hair, immunodeficiency, anemia Obesity, retinal degeneration, polydactyly, gonadal and renal malformation Polydactyly, congenital heart defect, hydrometrocolpos

ⴱ

Because HSCR most often presents shortly after birth, features of some syndromes (eg, MEN2A) may not be obvious at the time of diagnosis. Isolated HSCR is more commonly associated with heterozygous mutations; Waardenburg-Shah syndrome is more commonly associated with mutations in both alleles. ‡ Missense mutations affecting one of three cysteine codons (Cys609, Cys618, or Cys620). § HSCR is a diagnostic feature that differentiates Waardenburg-Shah syndrome from other variants of Waardenburg’s syndrome. ¶ CCHS has also been associated with RET, GDNF, and EDN3 mutations in rare patients, but concurrent HSCR has only been associated with PHOX2B mutations. 储 HSCR is a diagnostic feature of Haddad syndrome, but PHOX2B mutations also occur in patients with isolated congenital central hypoventilation. **Many patients have been presented with a HSCR and a variety of limb defects different from cartilage-hair hypoplasia. †† BBS6 and MKKS are the same gene. Abbreviation: NE, not established. †

these genes will be detected in less than 25% of patients with isolated disease. RET is the gene in which mutations are most frequently detected in patients with nonsyndromic HSCR (approximately 15% of all isolated HSCR patients). In addition, recent studies indicate that noncoding polymorphisms (base-pair differences that do not affect protein structure and which are present in ⬎1% of the normal population) in portions of the RET gene pose a significant risk for HSCR, possibly by reducing RET expression.70-72 The best characterized of these alterations involves a short evolutionarily conserved region in the first intron, which exists as “low”-and “high”-risk polymorphic alleles.72 A contemporary model, based partially on data derived from animal models and selective human populations with high rates of HSCR, suggests that this polymorphism or other alterations in the RET gene coupled with potentially equally subtle alterations at one or more other loci are responsible for most cases of nonsyndromic HSCR.64 However, neither

the loci involved nor their individual genetic variability have been delineated well enough to put this model to use in the prognosis or management of patients.

Associated anomalies and syndromes Associated anomalies are present in 5% to 30% of HSCR patients, many of whom have recognizable syndromes.64 It is important that every HSCR patient is evaluated carefully to exclude other defects. Malformations of other neural crest derivatives (eg, cardiac conotruncal derivatives, craniofacial musculature and skeleton, melanocytes, irides) are particularly common. A detailed family history should be obtained because HSCR and other anomalies may be incompletely penetrant manifestations of specific syndromes. In addition to dysmorphic features, the history and examination should investigate deafness, pigmentation defects, pregnancy losses, and medullary carcinoma of the thyroid and other lesions associated with

220 multiple endocrine neoplasia, type 2. Some of the latter may not be obvious or present in a neonate. More details are provided for the following specific associations because they are particularly relevant to the pediatric surgeon. Trisomy 21 (Down’s syndrome) Trisomy 21 is present in 2% to 15% of patients with HSCR, and conversely HSCR is found in 2% to 15% infants with trisomy 21.73 In most cases, cytogenetic diagnosis is established either prenatally or in a timely manner postnatally, because pediatric surgeons and their colleagues are well acquainted with other phenotypic features of Down’s syndrome. Although a variety of other structural chromosomal defects have been reported in patients with HSCR, it is unclear whether cytogenetic studies should be conducted in patients with isolated HSCR. With the exception of trisomy 21, abnormal karyotypes are rare and most patients with cytogenetic findings have other physical or neurological indications for chromosomal analysis.64 Familial medullary thyroid carcinoma/multiple endocrine neoplasia, type 2 Familial medullary thyroid carcinoma (FMTC) and both types A and B of multiple endocrine neoplasia, type 2 (MEN2) are caused by dominant mutations in RET. FMTC and MEN2A are due to mutations that affect specific cysteine codons encoded by exons 10 or 11, whereas MEN2B results from an M918T mutation in exon 16. These specific missense mutations produce constitutive active RET receptors, which stimulate abnormal cell proliferation and promote neoplastic transformation. In this respect, they contrast with the wide variety of RET mutations found in HSCR patients, which either reduce RET protein expression or impair the function of the RET receptor. However, a subset of FMTC/MEN2A-associated RET mutations have the dual effect of receptor activation and receptor instability.74 As a consequence, net RET activity may be low in enteric neural precursors, which could explain the empiric observation that FMTC/MEN2A-associated RET mutations have been found in 1% to 3% of patients with isolated HSCR.64,75 Although the likelihood that an individual patient with isolated HSCR will have FMTC/MEN2A is low, routine mutational analysis has been advocated in the literature because the implications for the patient and family are significant. Mutational detection is available commercially from multiple laboratories, and the cost is relatively low because only 2 exons need to be analyzed. However, such analysis is still not the standard practice in many medical centers. Congenital central hypoventilation syndrome ⴞ neuroblastoma Congenital central hypoventilation syndrome (CCHS) is a rare birth defect characterized by defective respiratory drive in response to hypercapnia. The overwhelming majority of patients with isolated CCHS have PHOX2B mu-

Seminars in Pediatric Surgery, Vol 18, No 4, November 2009 tations, typically expansions of a polyalanine tract.76 Approximately 20% of CCHS patients also are born with HSCR. Of the latter, the majority carry dominant mutations in PHOX2B other than polyalanine tract expansion, and a subset develop neuroblastoma (Haddad syndrome). Awareness of this association is important because recognition and genetic confirmation may permit early diagnosis of neuroblastoma and more effective treatment of the malignancy. Intestinal atresia Although most cases of intestinal atresia are not associated with HSCR, several patients with both anomalies have been reported.77 Ganglion cells are present in gut proximal to the site of atresia, but entirely missing distally or only present at the blind-ended atresia margin. In this context, HSCR is thought to be a consequence of a vascular event or other primary process that disrupts the gut wall and interrupts the path that vagal neural crest cells take to populate the intestines. Such examples are generally sporadic and may not have the same genetic basis as other forms of HSCR. When flanking segments of intestine are removed to repair atresia, the surgeon should expect the pathology report to comment on the presence/absence of ganglion cells, and aganglionosis should be in the differential diagnosis if obstructive symptoms persist postoperatively. Anal atresia HSCR has also been reported in patients with anal atresia and, if unrecognized, could complicate surgical correction.78,79 The muscular and neural anatomy of atretic distal rectum is usually abnormal, independent of associated HSCR. Hypertrophic nerves and poorly formed ganglia are common.79 Therefore, diagnosis of HSCR in this context may be more challenging and should rely on observations at the proximal end of any resected bowel. Multiple examples of Down’s syndrome and a single case of Pallister–Hall syndrome have been reported with anal atresia and HSCR.80-82

Length of the aganglionic segment Approximately 20% of nonsyndromic patients have longsegment or colonic HSCR, with transition zones located proximal to the sigmoid colon. A nearly equal ratio of males and females fall into this group. They are likely to harbor dominant mutations that affect the coding region of a single gene. An estimated 50% carry mutations in coding regions of RET. In contrast, patients with isolated short-segment disease individuals are nearly six times more likely to be male than female, show only a 5% to 15% rate of mutations in coding regions of RET, and show patterns of inheritance more consistent with multifactorial or autosomal recessive transmission.

Genetic counseling and recurrence risks It is important to realize that intestinal aganglionosis is a malformation, analogous to other birth defects like absent

Kapur

Hirschsprung’s Pathology and Genetics

radii, cleft lip, or congenital heart malformations. Genotype–phenotype correlation in patients with syndromic HSCR is sometimes sufficient to suggest mutation in a particular susceptibility gene, but this must be done cautiously because considerable phenotypic overlap exists among HSCR patients with mutations in different genes. Complete clinical evaluation and family history should be obtained as these details sometimes provide clues to the nature of the genetic defect. A particularly challenging problem in the field of Hirschsprung genetics is the phenomenon of “incomplete penetrance.” Incomplete penetrance refers to the fact that only a subset of individuals who carry a particular mutation in one of the genes listed in Table 2 actually exhibits the HSCR phenotype. Other persons, even immediate family members, with the same mutation are unaffected. Incomplete penetrance is observed in many other human genetic disorders and is usually attributed to one of three variables: genetic modifiers, environmental modifiers, or stochastic events. Substantial evidence has been gathered to suggest that genetic modifiers are responsible for some of the variable penetrance of HSCR mutations.83 In some cases, interactions between alleles of two different genes listed in Table 2 determine whether enteric neurodevelopment is perturbed. For example, subtle polymorphisms that alter the nucleotide sequence of the RET gene, but not the amino acid sequence of the RET protein, have been shown to influence penetrance of a particular missense mutation in the EDNRB gene.84 In addition, genetic loci that influence penetrance of RET mutations, but do not correlate with any of the genes listed in the table, have been mapped.64 Thus, HSCR is a complex genetic disorder that challenges the comprehension of scientists and clinicians, let alone the families with an affected child. At this time, the genetic data listed in Table 2 have limited practical value. Mutational analysis for many of the genes listed is not easily obtained and does not affect clinical management or counseling, except in rare contexts. Possible exceptions are testing for FMTC/MEN2A mutations of the RET gene in patients with sporadic HSCR and PHOX2B mutational analysis in patients with CCHS. Apart from these indications, most counseling is based on recurrence risks that were established empirically based on the length of the aganglionic segment and gender (Table 3).

Conclusions and future directions Fundamental principles for diagnosis and surgical management of HSCR have been in existence for several decades and are predicated on knowledgeable interpretation of rectal biopsies, intraoperative frozen sections, and postoperative resection specimens. The development of ancillary methods (eg, AChE histochemistry, calretinin immunohistochemistry) facilitate diagnosis, but failure to find ganglion cells in

221 Table 3 Empiric risk for recurrence of HSCR in siblings based on genders of proband and sibling and length of aganglionic segment in the proband64

Male sibling Female sibling

Male proband

Female proband

L-HSCR

S-HSCR

L-HSCR

S-HSCR

17% 13%

5% 1%

33% 9%

5% 3%

L-HSCR, long-segment Hirschsprung’s disease (aganglionosis proximal to sigmoid colon); S-HSCR, short-segment Hirschsprung’s disease (aganglionosis distal to sigmoid colon).

adequate histologic sections of an appropriate specimen from the correct site remains the single diagnostic criteria for this condition. Despite dramatic discoveries related to the genetic bases of HSCR, molecular genetic tests have extremely limited value in diagnosis or clinical management. However, additional advances in our genetic understanding may help to explain the multifactorial nature of this malformation and possibly the varied postoperative courses exhibited by patients with similar gross and microscopic pathology. In time, it may be possible to accurately predict the risk for HSCR based on molecular genetic tests, but present risk assessment is largely empiric. Anatomical studies, particularly when coupled with clinical follow-up, may also provide insight into the long-term complications that plague many patients after their aganglionic bowel has been resected. The concept of TZPT (including intestinal neuronal dysplasia) needs to be investigated in a carefully controlled unbiased manner to determine whether and how much resection of ganglionic bowel is necessary to achieve an optimal postoperative outcome. Given the genetic and phenotypic heterogeneity in HSCR, multiinstitutional studies with standardized molecular genetic and pathologic protocols may be required to derive meaningful conclusions in a reasonable time period. Stem cell therapy for HSCR has engendered some enthusiasm as a potential treatment for HSCR.85 The hypothesis has been advanced that function of the aganglionic bowel might be restored by neurons that differentiate from autologous or heterologous neural stem cells. Several groups have already demonstrated that neural stem cells can be isolated from fetal or postnatal gut wall and have the capacity to colonize embryonic gut and give rise to enteric ganglion cells. However, the ability of isolated stem cells to colonize the relatively mature gut wall and establish functionally significant connections may be a serious limitation to this strategy.

References 1. Chakravarti A, Lyonnet S. Hirschsprung’s disease. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The Metabolic and Molecular

222

2. 3.

4.

5. 6. 7.

8. 9.

10. 11.

12.

13.

14.

15.

16. 17.

18.

19.

20. 21. 22.

23. 24. 25.

Bases of Inherited Disease, vol 1. New York, NY: McGraw-Hill, 2001:931-42. Gherardi GJ. Pathology of the ganglionic-aganglionic junction in congenital megacolon. Arch Pathol 1960;69:520-8. White FV, Langer JC. Circumferential distribution of ganglion cells in the transition zone of children with Hirschsprung’s disease. Pediatr Dev Pathol 2000;3:216-22. Knowles CH, De Georgio R, Kapur RP, et al. Gastrointestinal neuromuscular pathology (GINMP): Guidelines for histological techniques and reporting. Acta Neuropathol 2009;118:271-301. Weinberg AG. The anorectal myenteric plexus: Its relation to hypoganglionosis of the colon. Am J Clin Pathol 1970;54:637-42. Whitehouse FR, Kernohan JW. Myenteric plexus in congenital megacolon. Arch Intern Med 1948;82:75-81. Doi T, Kobayashi H, Yamataka A, et al. Complete innervation profile of whole bowel resected at pull-through for Hirschsprung’s disease. Unexpected findings. Pediatr Surg Int 2005;21:889-98. Noblett HR. A rectal suction biopsy tube for use in the diagnosis of Hirschsprung’s disease. J Pediatr Surg 1969;4:406-9. Aldridge RT, Cambell PE. Ganglion cell distribution in the normal rectum and anal canal: A basis for the diagnosis of Hirschsprung’s disease by anorectal biopsy. J Pediatr Surg 1968;3:475-89. Leutenegger F, Untersuchungen des M. sphincter ani internus auf Ganglienzellen. Schweiz Med Wochenschr 1969;99:1431-2. Venugopal S, Mancer K, Shandling B. The validity of rectal biopsy in relation to morphology and distribution of ganglion cells. J Pediatr Surg 1981;16:433-7. Qualman SJ, Jaffe R, Bove KE, et al. Diagnosis of Hirschsprung’s disease using the rectal biopsy: Multi-institutional survey. Pediatr Dev Pathol 1999;2:588-96. Monforte-Munoz H, Gonzales-Gomez I, Rowland JM, et al. Increased submucosal nerve trunk caliber in aganglionosis. A “positive” and objective finding in suction biopsies and segmental resections in Hirschsprung’s disease. Arch Pathol Lab Med 1998;122:721-5. Meier-Ruge W, Hunziger O, Tobler H-J, et al. The pathophysiology of aganglionosis of the entire colon (Zuelzer–Wilson syndrome): Morphometric investigations of the extent of sacral parasympathetic innervation of the circular muscles of the aganglionic colon. Beitr Path Bd 1972;147:228-36. Nakao M, Suita S, Taguchi T, et al. Fourteen-year experience of acetylcholinesterase staining for rectal mucosal biopsy in neonatal Hirschsprung’s disease. J Pediatr Surg 2001;36:1357-63. Dobbins WO, Bill AH. Diagnosis of Hirschsprung’s disease excluded by suction rectal biopsy. N Engl J Med 1965;272:990-6. Yunis EJ, Dibbins AW, Sherman FE. Rectal suction biopsy in the diagnosis of Hirschsprung’s disease. Arch Pathol Lab Med 1976;100: 329-33. Athow AC, Filipe MI, Drake DP. Problems and advantages of acetylcholinesterase histochemistry of rectal suction biopsies in the diagnosis of Hirschsprung’s disease. J Pediatr Surg 1990;25:520-6. Alizai NK, Batcup G, Dixon MF, et al. Rectal biopsy for Hirschsprung’s disease: What is the optimum method? Pediatr Surg Int 1998;13:121-4. Meier-Ruge W, Bruder E. Pathology of chronic constipation in pediatric and adult coloproctology. Pathobiology 2005;72:1-106. Kobayashi H, O’Briain DS, Hirakawa H, et al. A rapid technique of acetylcholinesterase staining. Arch Pathol Lab Med 1994;118:1127-9. Martucciello G, Favre A, Torre M, et al. A new rapid acetylcholinesterase histochemical method for the intraoperative diagnosis of Hirschsprung’s disease and intestinal neuronal dysplasia. Eur J Pediatr Surg 2001;11:300-4. Moore SW, Johnson G. Acetylcholinesterase in Hirschsprung’s disease. Pediatr Surg Int 2005;21:255-63. Kapur RP. Can we stop looking? Immunohistochemistry and the diagnosis of Hirschsprung’s disease. Am J Clin Pathol 2006;126:9-12. Barshack I, Fridman E, Goldberg I, et al. The loss of calretinin expression indicates aganglionosis in Hirschsprung’s disease. J Clin Pathol 2004;57:712-16.

Seminars in Pediatric Surgery, Vol 18, No 4, November 2009 26. Kapur RP, Reed RC, Finn L, et al. Calretinin immunohistochemistry versus acetylcholinesterase histochemistry in the evaluation of suction rectal biopsies for Hirschsprung’s disease. Pediatr Dev Pathol 2009; 12:6-15. 27. Fonkalsrud EW, Snyder CL, Ashcraft KW. Complications of Hirschsprung’s disease and allied disorders. In: Holschneider AM, Puri P, eds. Hirschsprung’s Disease and Allied Disorders (ed 2). Amsterdam: Harwood Academic Publishers, 2000:425-39. 28. Langer JC. Persistent obstructive symptoms after surgery for Hirschsprung’s disease: Development of a diagnostic and therapeutic algorithm. J Pediatr Surg 2004;39:1458-62. 29. Boman F, Sfeir R, Priso R, et al. Advantages of intraoperative semiquantitative evaluation of myenteric nervous plexuses in patients with Hirschsprung’s disease. J Pediatr Surg 2007;42:1089-94. 30. Gianino S, Grider JR, Cresswell J, et al. GDNF: Availability determines enteric neuron number by controlling precursor proliferation. Development 2003;130:2187-98. 31. Breau MA, Pietri T, Eder O, et al. Lack of beta1 integrins in enteric neural crest cells leads to a Hirschsprung-like phenotype. Development 2006;133:1725-34. 32. Estevao-Costa J, Fragoso AC, Campos M, et al. An approach to minimize postoperative enterocolitis in Hirschsprung’s disease. J Pediatr Surg 2006;41:1704-7. 33. Kapur RP. Neuronal dysplasia: A controversial pathological correlate of intestinal pseudo-obstruction. Am J Med Genet 2003;122A:287-93. 34. Meier-Ruge WA, Bruder E, Kapur RP. Intestinal neuronal dysplasia type B: One giant ganglion is not good enough. Pediatr Dev Pathol 2006;9:444-52. 35. Ure BM, Holschneider AM, Meier-Ruge WA. Neuronal intestinal malformations: A retro- and prospective study on 203 patients. Eur J Pediatr Surg 1994;4:279-86. 36. Banani SA, Forootan HR, Kumar PV. Intestinal neuronal dysplasia as a cause of surgical failure in Hirschsprung’s disease: A new modality for surgical management. J Pediatr Surg 1996;31:572-4. 37. Haricharan RN, Seo JM, Kelly DR, et al. Older age at diagnosis of Hirschsprung’s disease decreases risk of postoperative enterocolitis, but resection of additional ganglionated bowel does not. J Pediatr Surg 2008;43:1115-23. 38. Lowichik A, Weinberg AG. Eosinophilic infiltration of the enteric neural plexuses in Hirschsprung’s disease. Ped Pathol Med 1997;17: 885-91. 39. Taguchi T, Suita S, Hirata Y, et al. Abnormally shaped arteries in the intestine of children with Hirschsprung’s disease: Etiological considerations related to ischemic theory. J Pediatr Gastroenterol Nutr 1994; 18:200-4. 40. Vanderwinden J-M, Rumessen JJ, Liu H, et al. Interstitial cells of cajal in human colon and in Hirschsprung’s disease. Gastroenterology 1996; 111:901-10. 41. Taguchi T, Suita S, Masumoto K, et al. An abnormal distribution of C-kit positive cells in the normoganglionic segment can predict a poor clinical outcome in patients with Hirschsprung’s disease. Eur J Pediatr Surg 2005;15:153-8. 42. Newman CJ, Laurini RN, Lesbros Y, et al. Interstitial cells of Cajal are normally distributed in both ganglionated and aganglionic bowel in Hirschsprung’s disease. Pediatr Surg Int 2003;19:662-8. 43. Alpy F, Ritie L, Jaubert F, et al. The expression pattern of laminin isoforms in Hirschsprung’s disease reveals a distal peripheral nerve differentiation. Hum Pathol 2005;36:1055-65. 44. Bealer JF, Natuzzi ES, Buscher C, et al. Nitric oxide synthase is deficient in the aganglionic colon of patients with Hirschsprung’s disease. Pediatrics 1994;93:647-51. 45. Kobayashi H, Hirakawa H, Puri P. NADPH-diaphorase histochemistry: A reliable test for the intraoperative diagnosis of Hirschsprung’s disease. J Pediatr Surg 1996;31:1552-3. 46. O’Kelly TJ, Davies JR, Tam PKH, et al. Abnormalities of nitric-oxideproducing neurons in Hirschsprung’s disease: Morphology and implications. J Pediatr Surg 1994;29:294-300.

Kapur

Hirschsprung’s Pathology and Genetics

47. Fromont-Hankard G, Lafer D, Masood S. Altered expression of alphasmooth muscle isoactin in Hirschsprung’s disease. Arch Pathol Lab Med 1996;120:270-4. 48. Wedel T, Van Eys GJ, Waltregny D, et al. Novel smooth muscle markers reveal abnormalities of the intestinal musculature in severe colorectal motility disorders. Neurogastroenterol Motil 2006;18: 526-38. 49. Teitelbaum DH, Qualman SJ, Caniano DA. Hirschsprung’s disease. Identification of risk factors for enterocolitis. Ann Surg 1988;207: 240-4. 50. Murphy F, Puri P. New insights into the pathogenesis of Hirschsprung’s associated enterocolitis. Pediatr Surg Int 2005;21:773-9. 51. Teitelbaum DH, Caniano DA, Qualman SJ. The pathophysiology of Hirschsprung’s-associated enterocolitis: Importance of histologic correlates. J Pediatr Surg 1989;24:1271-7. 52. Yunis E, Sieber WK, Akers DR. Does zonal aganglionosis really exist? Report of a rare variety of Hirschsprung’s disease and review of the literature. Pediatr Pathol 1983;1:33-49. 53. Kapur RP, deSa DJ, Luquette M, et al. Hypothesis: Pathogenesis of skip areas in long-segment Hirschsprung’s disease. Pediatr Pathol Lab Med 1995;15:23-37. 54. Gonzalez-Vasquez R, Heiss WH. Segmentale Aganglionose des Dunndarms—Mythos oder Wirklichkeit? (Eine Falldarstellung). Z Kinderchir 1988;43:424-6. 55. Vaos GC. Quantitative assessment of the stage of neuronal maturation in the developing human fetal gut—A new dimension in the pathogenesis of developmental anomalies of the myenteric plexus. J Pediatr Surg 1989;24:920-5. 56. Martin LW, Buchino JJ, LeCoultre C, et al. Hirschsprung’s disease with skip area (segmental aganglionosis). J Pediatr Surg 1979;14: 686-7. 57. Davidson M, Bauer CH. Studies of distal colonic motility in children. IV: Achalasia of the distal rectal segment despite presence of ganglia in the myenteric plexuses of this area. Pediatrics 1958;21:746-61. 58. Puri P. Hirschsprung’s disease: Clinical generalities. In: Holschneider AM, Puri P, eds. Hirschsprung’s Disease and Allied Disorders. Amsterdam: Harwood Scientific, 2000:129-35. 59. Lake BD, Puri P, Nixon HH, et al. Hirschsprung’s disease: An appraisal of histochemically demonstrated acetylcholinesterase activity in suction rectal biopsy specimens as an aid to diagnosis. Arch Pathol Lab Med 1978;102:244-7. 60. Neilson IR, Yazbeck S. Ultrashort Hirschsprung’s disease: Myth or reality. J Pediatr Surg 1990;25:1135-8. 61. Hirakawa H, Kobayashi H, O’Briain DS, et al. Absence of NADPHdiaphorase activity in internal anal sphincter (IAS) achalasia. J Pediatr Gastroenterol Nutr 1995;20:54-8. 62. Ballard ET. Ultrashort segment Hirschsprung’s disease: A case report. Pediatr Pathol Lab Med 1996;16:319-25. 63. Meier-Ruge W, Scharli AF. The epidemiology and histotopochemical characterization of ultrashort-segment Hirschsprung’s disease. Pediatr Surg Int 1986;1:37-42. 64. Amiel J, Sproat-Emison E, Garcia-Barcelo M, et al. Hirschsprung’s disease, associated syndromes and genetics: A review. J Med Genet 2008;45:1-14. 65. Yip L, Cote GJ, Shapiro SE, et al. Multiple endocrine neoplasia type 2: Evaluation of the genotype-phenotype relationship. Arch Surg 2003; 138:409-16.

223 66. Jira PE, Waterham HR, Wanders RJ, et al. Smith–Lemli–Opitz syndrome and the DHCR7 gene. Ann Hum Genet 2003;67:269-80. 67. Levy J. Gastrointestinal concerns. In: Pueschel SM, Pueschel JK, eds. Biomedical Concerns in Persons with Down’s Syndrome. Baltimore, MD: Paul H Brookes Publishing, 1992:119-25. 68. Mowat D, Wilson MJ, Goossens M. Mowat-wilson syndrome. J Med Genet 2003;40:305-10. 69. Beales PL, Elcioglu N, Woolf AS, et al. New criteria for improved diagnosis of Bardet–Biedl syndrome: Results of a population survey. J Med Genet 1999;36:437-46. 70. Borrego S, Wright FA, Fernandez RM, et al. A founding locus within the RET proto-oncogene may account for a large proportion of apparently sporadic Hirschsprung’s disease and a subset of cases of sporadic medullary thyroid carcinoma. Am J Hum Genet 2003;72:88-100. 71. Burzynski GM, Nolte IM, Osinga J, et al. Localizing a putative mutation as the major contributor to the development of sporadic Hirschsprung’s disease to the RET genomic sequence between the promoter region and exon 2. Eur J Hum Genet 2004;12:604-12. 72. Emison ES, McCallion AS, Kashuk CS, et al. A common sex-dependent mutation in a RET enhancer underlies Hirschsprung’s disease risk. Nature 2005;434:857-63. 73. Moore SW. Down’s syndrome and the enteric nervous system. Pediatr Surg Int 2008;24:873-83. 74. Iwashita T, Kurokawa K, Qiao S, et al. Functional analysis of RET with Hirschsprung mutations affecting its kinase domain. Gastroenterology 2001;121:24-33. 75. Decker RA, Peacock ML. Occurrence of MEN 2a in familial Hirschsprung’s disease: A new indication for genetic testing of the RET proto-oncogene. J Pediatr Surg 1998;33:207-14. 76. Berry-Kravis EM, Zhou L, Rand CM, et al. Congenital central hypoventilation syndrome: PHOX2B mutations and phenotype. Am J Respir Crit Care Med 2006;174:1139-44. 77. Seo T, Ando H, Watanabe Y, et al. Colonic atresia and Hirschsprung’s disease: Importance of histologic examination of the distal bowel. J Pediatr Surg 2002;37:E19. 78. Kiesewetter WB, Sukarochana K, Sieber WK. The frequency of aganglionosis associated with imperforate anus. Surgery 1965;58:877-80. 79. Meier-Ruge WA, Holschneider AM. Histopathologic observations of anorectal abnormalities in anal atresia. Pediatr Surg Int 2000;16:2-7. 80. Haynes JH, Bagwell CE. Hirschprung’s disease and imperforate anus in Pallister–Hall syndrome: A new association. J Pediatr Surg 2003; 38:1411-12. 81. Clarke SA, Van der Avoirt A. Imperforate anus, Hirschsprung’s disease, and trisomy 21: A rare combination. J Pediatr Surg 1999;34: 1874. 82. Flageole H, Fecteau A, Laberge JM, et al. Hirschsprung’s disease, imperforate anus, and Down’s syndrome: A case report. J Pediatr Surg 1996;31:759-60. 83. de Pontual L, Pelet A, Clement-Ziza M, et al. Epistatic interactions with a common hypomorphic RET allele in syndromic Hirschsprung’s disease. Hum Mutat 2007;28:790-6. 84. Carrasquillo MM, McCallion AS, Puffenberger EG, et al. Genomewide association study and mouse model identify interaction between RET and ENDRB pathways in Hirschsprung’s disease. Nat Genet 2002;32:237-44. 85. Theocharatos S, Kenny SE. Hirschsprung’s disease: Current management and prospects for transplantation of enteric nervous system progenitor cells. Early Hum Dev 2008;84:801-4.