Case Based Pediatrics For Medical Students and Residents
Department of Pediatrics, University of Hawaii John A. Burns School of Medicine
Chapter IX.9. Hirschsprung's Disease
Walton K.T. Shim, MD
March 2003

Return to Table of Contents

An infant male presents in the second day of life with a large bilious emesis. He had been "spitty" for a day and had yielded 15 ml of greenish gastric aspirate at birth. He had not passed meconium for the first 36 hours of life. He was born at term weighing 3.5 kg.

Exam: VS T 37.2 (rectal), P 125, R 30, BP 75/55, weight 3.4 kg. He is an active hungry infant with a moderately distended abdomen. Bowel sounds are very active but not obstructive in nature. No organs or abdominal masses are appreciated and no herniae are present. His anus is patent.

An abdominal series reveals large dilated loops of bowel but no air in the rectum. A hand injected contrast enema on the third day of life shows no distinct transition zone. A 24-hour delayed film shows retained contrast and a rectal mucosal suction biopsy reveals an absence of ganglion cells and the presence of hypertrophied nerve fibers consistent with a diagnosis of Hirschsprung's disease.

Rectal irrigations are not successful in decompressing the colon leading to the establishment of a descending colonic ostomy, placed under biopsy guidance. When the infant achieves a weight of 7 kg (15 pounds) a definitive resection will be performed.


Hirschsprung’s disease (also known as congenital megacolon or congenital intestinal aganglionosis) is a disease condition most commonly affecting the rectosigmoid portion of the colon. It presents with constipation in older infants and children, but mainly by distention and vomiting in newborn infants. The affected segment lacks ganglion cells which aid in normal peristalsis. Without these ganglion cells, normal peristalsis is lacking, resulting in a functional obstruction. This produces a proximal dilated colon and a distal normal appearing segment. Classically, there is an obvious transition zone where the dilated colon (with normal ganglion cells and peristalsis) meets the non-dilated colon (which is abnormal and aganglionic). The appearance is paradoxical, and in the past, has led surgeons to remove the grossly dilated (normal) portion rather than the normal appearing aganglionic segment of the colon. This of course resulted in recurrence of the obstruction and dilatation. The severity of Hirschsprung’s disease varies with the length of the involved segment and may be very difficult to diagnose especially in the ultra short segment disease because of the variability of the constipation. Total aganglionosis of the colon is quite uncommon but aganglionosis involving the small bowel is rare.

The earliest description of a case of congenital megacolon was by Fredrick Ruysch in 1691, almost two centuries prior to the classic description of the Danish physician Harald Hirschsprung who reported two cases of young boys dying with a hugely dilated proximal colon and a narrowed distal colon and rectum in 1886.

Early in the history of the disease attention focused on the hugely dilated proximal colon as the abnormal portion so that resection of this area was attempted. Of course this failed soon after operation with resumption of a megacolon.

A pediatric surgeon, Orvar Swenson, was the first to devise a procedure based on observations that a colostomy established in the dilated segment functioned normally but again became obstructed when reconnected to the distal narrow portion. He concluded that functional obstruction occurred in the narrower but normally appearing distal segment. Swenson followed up these observations with pressure measurements through colostomies in patients both with and without Hirschsprung's disease and showed normal peristaltic activity in the proximal dilated colon but no peristalsis in the distal narrowed segment (1). His contribution was to resect the distally narrowed area and connect the dilated segment to two or three centimeters of distal rectum. The procedure was successful and described in 1948 (2).

This was followed a few years later by Duhamel who incorporated a portion of the anteriorly placed aganglionic rectum with a posteriorly placed, normally innervated colon to produce a new rectum composed of half aganglionic and half ganglionic musculature.

A still later modification was proposed and used by Soave who stripped the mucosa from the distal aganglionic rectum and passed the normally innervated colon through the sleeve of dysfunctional rectum (an endorectal pull-through) relying on the normal portion to propel through the abnormal cuff.

Each of these procedures has been successful in overcoming the functional obstruction in the great majority of cases, but each has its own complications.

Common to each procedure is post-operative enterocolitis characterized by abdominal distention, loose foul smelling stools, and vomiting. It occurs in a quarter to a third of cases and should be treated early and aggressively with rectal irrigations, anal dilatations and intravenous support as death may occur if it is neglected (3). Although fever and signs of infection may be present, stool cultures are often not helpful. The author routinely has parents or caregivers dilate the anus or irrigate the rectum postoperatively for several months to prevent enterocolitis. Most patients continue to improve bowel control for several years postoperatively (4).

About five percent of Swenson procedures experience anastomotic leaks. Incomplete emptying of the aganglionic portion of the pouch plagues some Duhamel patients. Patients with endorectal Soave procedures suffer from cuff abscesses and may require continued dilatations.

Recently with the introduction of minimally invasive procedures involving laparoscopic dissection and various stapling devices, techniques have changed but the basic concepts for overcoming the non-relaxing, functionally obstructive distal colon are unchanged.

The diagnosis is suggested in a term newborn who has emesis and abdominal distention early in the newborn period. Since a newborn usually passes his/her first meconium on the first day, the most suggestive symptom is the lack of meconium passage during the first day of life. Ninety-nine percent of normal newborn infants pass stool within the first 48 hours of life (5). A digital rectal examination is not helpful and may prevent an accurate contrast enema study, although a temperature probe may be gently inserted to prove anal patency.

In the face of delayed meconium passage, vomiting and abdominal distention, an abdominal series should be obtained. In congenital megacolon, intestinal dilatation is usually present with a gasless rectum. A hand injected contrast enema should be obtained to outline the rectum and sigmoid colon. Particular attention should be directed at not overfilling the intestines, thus obscuring the transition zone.

Infants with Hirschsprung's disease frequently retain contrast material longer than 24 hours and this delayed passage strongly suggests the diagnosis despite the absence of a definite transition area. Absence of ganglion cells as the cause for uncoordinated peristalsis was correctly identified as the cause for Hirschsprung's disease in infancy in1938 by Robertson and Kernahan.

Although the gold standard of diagnosis is the histological absence of ganglion cells and hypertrophied autonomic nerves, the typical radiographic transition zone between the proximal dilated and distally narrowed colon is sufficient evidence for the diagnosis in the face of supportive presence of delayed meconium passage, vomiting, and distention. Histochemical patterns with special staining techniques have also been correlated with ganglion cell absence.

Occasionally an older child presents with a history of long standing constipation requiring enemas and other attentive measures directed at producing defecation. In such cases the diagnosis is made by contrast enema as the transition zone is usually easily demonstrated. Contrast enemas in infants less than two months old may be non-diagnostic in over 20% of cases (6). In these instances when clinical and radiographic findings are unable to make a definitive diagnosis, a rectal biopsy becomes necessary. A full thickness biopsy should have an absence of ganglion cells in Auerbach's plexus located between the circular and longitudinal muscle layers. Although ganglion cells are more plentiful in this area, the full thickness biopsy complicates later surgical dissection so a rectal mucosal suction biopsy of Meissner's plexus located in the muscularis propria (i.e., performed more superficially and less invasively) is the preferred biopsy technique. Although ganglion cells are more sparse, the associated presence of hypertrophied nerve fibers is diagnostic.

Ganglion cells are absent in the most distal two centimeters of the normal rectum which is of importance when performing the biopsy and in the determination of an ultra short segment Hirschsprung's (i.e., the rectal biopsy should ideally be obtained proximal to this region).

The normal physiologic pressure in the anal canal during defecation involves a decrease in internal sphincter pressure (relaxation) with rectal distention, thus allowing passage of the fecal bolus. In a baby with Hirschsprung's disease this relaxation of the involuntary internal sphincter does not occur, thus providing another means of making the diagnosis by ano-rectal manometry.

The most frequently involved areas of aganglionosis are the rectum and sigmoid, with decreasing incidence progressing cephalad. Total aganglionosis of the colon is a rarity, and small bowel involvement is even less common.

The incidence of Hirschsprung's disease is about 1 in 5000 births with a 4:1 predominance in males. There is a familial inheritance factor greatest among siblings but less common among children of parents with the disease. It is one of the most common causes of infant intestinal obstruction and is exceeded only by intestinal atresia, malrotation and meconium ileus (in Caucasians). Hirschsprung's disease mutations have been mapped to RET protooncogenes at 10Q11.2, the recessive EDNRB gene at 13Q22, its ligand endothelin 3 (EDN3) and glial cell line-derived neurotrophic factor (GDNF) in humans. Although the majority of cases are multigenic or multifactorial, there are some conditions associated with Hirschsprung's disease such as Down's syndrome, Waardenburg syndrome, neurofibromatosis, neuroblastoma, pheochromocytoma, the MEN2B syndrome and others. The trypanosome causing Chagas' disease is responsible for an acquired form of aganglionosis which may affect not only the colon but the esophagus and heart as well.

There have been described a limited number of Hirschsprung's patients with neuronal intestinal dysplasia that may explain continued post-operative morbidity. The diagnosis of neuronal intestinal dysplasia is not easily made but is associated with abnormal neural elements and their distribution in both the submucosal (Meissner's) and intermuscular (Auerbach's) plexus. This may explain some cases of continued post-operative constipation. Since ganglion cells are of neural crest origin, other conditions affecting the physiology, distribution and migration of these cells may be related to Hirschsprung's disease (7).


Questions

1. True/False: A digital rectal examination carefully performed is most important in the diagnosis of Hirschsprung's disease in a newborn infant.

2. True/False: Post operative diarrhea from enterocolitis is a common occurrence.

3. In a newborn infant with abdominal distention and/or vomiting, what is the most significant clinical finding to raise the suspicion of Hirschsprung's disease?

4. True/False: In a child over a year of age with a radiographic transition zone, a rectal biopsy is required for a definitive diagnosis?

5. What cell line differentiates into Auerbach's and Meissner's plexus and may be responsible for other associated neurological defects?


References

1. Swenson O. Historical Review of Hirschsprung's Disease. Proceedings of the Plenary Session: Hundred Years of Hirschsprung's Disease (1886-1996), 8th Congress Asian Association of Pediatric Surgeons, Calcutta, India, 1986.

2. Swenson O, Raffensperger JG. In: Raffensperger JG (ed). Swenson's Pediatric Surgery, 5th edition. 1990, McGraw-Hill Professional, pp. 555-579.

3. Marty TL, Takahito S, Sullivan JJ, et al. Rectal Irrigations for the Prevention of Postoperative Enterocolitis in Hirschsprung's disease. J Pediatr Surg 1995;30:652-654.

4. Swenson O, Sherman JO, Fisher JH, et al. The Treatment and Postoperative Complications of Congenital Megacolon: A 25 Year Follow Up. Ann Surg 1975;82:226.

5. Kirschner BS, Black DD. Chapter 11 - The Gastrointestinal Tract. In: Behrman RE, Kliegman RM (eds). Nelson Essentials of Pediatrics, 4th edition. 2002, Philadelphia: W.B. Saunders, p. 482.

6. Shim WKT, Swenson O. Treatment of Congenital Megacolon in 50 Infants. Pediatrics 1966;38:185.

7. Cheng W, Au DK, Knowles CH, Anand P, Tam PK. Hirschsprung's Disease: A More Generalized Neuropathy. J Pediatr Surg 2001;36:296-300.


Answers to questions

1. false

2. true

3. No meconium for the first day of life.

4. false

5. neural crest cells


Return to Table of Contents

University of Hawaii Department of Pediatrics Home Page