Emedicine.medscape

Introduction

Background

In 1886, Harold Hirschsprung first described Hirschsprung disease as a cause of constipation in early infancy. Early recognition and surgical correction of Hirschsprung disease protects affected infants from enterocolitis and debilitating constipation.

Pathophysiology

Hirschsprung disease results from the absence of enteric neurons within the myenteric and submucosal plexus of the rectum and/or colon. Enteric neurons are derived from the neural crest and migrate caudally with the vagal nerve fibers along the intestine. These ganglion cells arrive in the proximal colon by 8 weeks' gestation and in the rectum by 12 weeks' gestation. Arrest in migration leads to an aganglionic segment. This results in clinical Hirschsprung disease.

Frequency

United States

Hirschsprung disease occurs in approximately 1 per 5000 live births.

International

Prevalence may vary by region and has been shown to be as high as 1 per 3000 live births in the Federated States of Micronesia.1

Mortality/Morbidity

The overall mortality of Hirschsprung enterocolitis is 25-30%, which accounts for almost all of the mortality from Hirschsprung disease.

Sex

Hirschsprung disease is approximately 4 times more common in males than females.

Age

Nearly all children with Hirschsprung disease are diagnosed during the first 2 years of life. Approximately one half of children affected with this disease are diagnosed before

they are aged 1 year. A small number of children with Hirschsprung disease are not recognized until much later in childhood or adulthood.

Clinical

History

During the newborn period, infants affected with Hirschsprung disease may present with abdominal distention, failure of passage of meconium within the first 48 hours of life, and repeated vomiting. A family history of a similar condition is present in about 30% of cases.

Nearly one half of all infants with Hirschsprung disease have a history of delayed first passage of meconium (beyond age 36 h), and nearly one half of infants with delayed first passage of meconium have Hirschsprung disease.

Unlike children experiencing functional constipation, children with Hirschsprung disease rarely experience soiling and overflow incontinence.

Children with Hirschsprung disease may be malnourished. Poor nutrition results from the early satiety, abdominal discomfort, and distention associated with chronic constipation.

Older infants and children typically present with chronic constipation. This constipation often is refractory to usual treatment protocols and may require daily enema therapy.

Hirschsprung enterocolitis can be a fatal complication of Hirschsprung disease. Enterocolitis typically presents with abdominal pain, fever, foul-smelling and/or bloody diarrhea, as well as vomiting. If not recognized early, enterocolitis may progress to sepsis, transmural intestinal necrosis, and perforation.

Physical

Examination of infants affected with Hirschsprung disease reveals tympanitic abdominal distention and symptoms of intestinal obstruction. Individuals in this age group may also present with acute enterocolitis or with neonatal meconium plug syndrome.

Children with Hirschsprung disease are usually diagnosed by age 2 years.o Older infants and children with Hirschsprung disease usually present with

chronic constipation. Upon abdominal examination, these children may demonstrate marked abdominal distention with palpable dilated loops of colon. Rectal examination commonly reveals an empty rectal vault and may result in the forceful expulsion of fecal material upon completion of examination.

o Less commonly, older children with Hirschsprung disease may be chronically malnourished and/or present with Hirschsprung enterocolitis.

Causes

Genetic causeso The disease is generally sporadic, although incidence of familial disease

has been increasing.o Multiple loci appear to be involved, including chromosomes 13q22,

21q22, and 10q.o Mutations in the Ret proto-oncogene have been associated with multiple

endocrine neoplasia (MEN) 2A or MEN 2B and familial Hirschsprung disease.2,3

o Other genes associated with Hirschsprung disease include the glial cell-derived neurotrophic factor gene, the endothelin-B receptor gene, and the endothelin-3 gene.

Associated conditionso Hirschsprung disease is strongly associated with Down syndrome; 5-15%

of patients with Hirschsprung disease also have trisomy 21.o Other associations include Waardenburg syndrome, congenital deafness,

malrotation, gastric diverticulum, and intestinal atresia.

Differential Diagnoses

Constipation

Other Problems to Be Considered

Intestinal neuronal dysplasiaMeconium plug syndromeNeonatal small left colon syndromeHypoganglionosis

Workup

Laboratory Studies

CBC count: Order this test if enterocolitis is suspected. Elevation of WBC count or a bandemia should raise concern for enterocolitis.

Imaging Studies

Plain abdominal radiography: Perform this test with any signs or symptoms of abdominal obstruction.

Unprepared single-contrast barium enema: If perforation and enterocolitis are not suspected, an unprepared single-contrast barium enema may help establish the diagnosis by identifying a transition zone between a narrowed aganglionic segment and a dilated and normally innervated segment. The study may also reveal a nondistensible rectum, which is a classic sign of Hirschsprung disease. A transition zone may not be apparent in neonates, because of insufficient time to

develop colonic dilation, or in infants who have undergone rectal washouts, examinations, or enemas.

Other Tests

Rectal manometry: In older children who present with chronic constipation and an atypical history for either Hirschsprung disease or functional constipation, anorectal manometry can be helpful in making or excluding the diagnosis.4

Children with Hirschsprung disease fail to demonstrate reflex relaxation of the internal anal sphincter in response to inflation of a rectal balloon.

Procedures

The definitive diagnosis of Hirschsprung disease rests on histological review of rectal tissue.

o Obtain tissue either by suction rectal biopsy or transanal wedge resection. If a suction biopsy is performed, take the biopsy 2-2.5 cm above the dentate line on the posterior wall to minimize the risk of perforation.

o Carefully examine biopsy specimens for the presence or absence of ganglion cells in the submucous plexus (suction rectal biopsy) or myenteric plexuses (transanal wedge resection).

In the hands of an experienced pathologist, the resulting biopsy and absence of ganglion cells confirm the diagnosis and allow the initiation of treatment. Skip lesions of aganglionosis have been reported in cases of Hirschsprung disease.

Acetylcholinesterase staining of the tissue can be performed to assist with the pathologic assessment. Acetylcholinesterase staining identifies the hypertrophy of extrinsic nerves trunks. In short-segment Hirschsprung disease, the diagnosis can be made with a properly placed rectal suction biopsy alone or in combination with anorectal manometry.

Histologic Findings

Histologic findings include the absence of ganglion cells in the myenteric plexus and hypertrophic extrinsic nerve fibers.

Treatment

Medical Care

If a child with Hirschsprung disease has symptoms and signs of a high-grade intestinal obstruction, initial therapy should include intravenous hydration, withholding of enteral intake, and intestinal and gastric decompression.

Decompression can be accomplished through placement of a nasogastric tube and either digital rectal examination or normal saline rectal irrigations 3-4 times daily.

Administer broad-spectrum antibiotics to patients with enterocolitis.

Immediately request surgical consultation for biopsy confirmation and treatment plan.

While awaiting surgical intervention in the event of a planned single-stage pull-through procedure, the baby should receive scheduled vaccinations.

Surgical Care

The surgical options vary according to the patient's age, mental status, ability to perform activities of daily living, length of the aganglionic segment, degree of colonic dilation, and presence of enterocolitis.

Surgical options include colostomy at the level of normal bowel, rectal irrigations followed by rectal resection with a pull-through procedure once bowel caliber is restored to normal, and a staged procedure with placement of a diverting colostomy followed by a pull-through procedure. The single-stage pull-through procedure may be performed with laparoscopic, open, or transanal techniques. This procedure is generally performed after the newborn has had rectal irrigations at home and has passed the physiologic nadir.

The ability to perform a single-stage pull-through procedure largely depends on the availability, experience, and capabilities of the staff pathologist because aganglionic intestine must not be in the pull-through segment.

Recurrent postoperative enterocolitis may require treatment. Current therapeutic options include rectal dilations, application of topical nitric oxide, posterior myotomy/myectomy,5 or injection of botulinum toxin.6

Diet

A special diet is not required. However, preoperatively and in the early postoperative period, infants on a nonconstipated regimen, such as breast milk, are more easily managed.

Activity

Postoperatively, patients may return to their normal physical activities.

Medication

Drug therapy currently is not a component of the standard of care for this disease itself; however, some medications may be used to treat complications of Hirschsprung disease. See Treatment. Medications may include antibiotics for the treatment of enterocolitis or the use of botulinum toxin injection at the anal sphincter for the treatment of recurrent enterocolitis due to anal hypertonicity.

Medicinet.co.uk

Hirschsprung Disease

What is Hirschsprung's disease?

Hirschsprung's (HURSH-sprungz) disease, or HD, is a disease of the large intestine.

The large intestine is also sometimes called the colon. The word bowel can refer to the large and small intestines. Hirschsprung's disease usually occurs in children. It causes constipation, which means that bowel movements are difficult. Some children with Hirschsprung's disease can't have bowel movements at all. The stool creates a blockage in the intestine.

If Hirschsprung's disease is not treated, stool can fill up the large intestine. This can cause serious problems like infection, bursting of the colon, and even death.

Most parents feel frightened when they learn that their child has a serious disease. This information will help you understand Hirschsprung's disease and how you and the doctor can help your child.

Why does Hirschsprung's disease cause constipation?

Normally, muscles in the intestine push stool to the anus, where stool leaves the body. Special nerve cells in the intestine, called ganglion cells, make the muscles push. A person with Hirschsprung's disease does not have these nerve cells in the last part of the large intestine.

Healthy large intestine: Nerve cells are found throughout the intestine.

Hirschsprung's disease large intestine: Nerve cells are missing from the last part of the intestine.

In a person with Hirschsprung's disease, the healthy muscles of the intestine push the stool until it reaches the part without the nerve cells. At this point, the stool stops moving. New stool then begins to stack up behind it.

Sometimes the ganglion cells are missing from the whole large intestine and even parts of the small intestine before it. When the diseased section reaches to or includes the small intestine, it is called long-segment disease. When the diseased section includes only part of the large intestine, it is called short-segment disease.

What is the treatment for Hirschsprung's disease??

Pull-through Surgery

Hirschsprung's disease is treated with surgery. The surgery is called a pull-through operation. There are three common ways to do a pull-through, and they are called the Swenson, the Soave, and the Duhamel procedures. Each is done a little differently, but all involve taking out the part of the intestine that doesn't work and connecting the healthy part that's left to the anus. After pull-through surgery, the child has a working intestine.

Before surgery: The diseased section is the part of the intestine that doesn't work.

Step 1: The doctor removes the diseased section.

Step 2: The healthy section is attached to the rectum or anus.

Colostomy and Ileostomy

Often, the pull-through can be done right after the diagnosis. However, children who have been very sick may first need surgery called an ostomy. This surgery helps the child

get healthy before having the pull-through. Some doctors do an ostomy in every child before doing the pull-through.

In an ostomy, the doctor takes out the diseased part of the intestine. Then the doctor cuts a small hole in the baby's abdomen. The hole is called a stoma. The doctor connects the top part of the intestine to the stoma. Stool leaves the body through the stoma while the bottom part of the intestine heals. Stool goes into a bag attached to the skin around the stoma. You will need to empty this bag several times a day.

Step 1: The doctor takes out most of the diseased part of the intestine.

Step 2: The doctor attaches the healthy part of the intestine to the stoma (a hole in the abdomen).

If the doctor removes the entire large intestine and connects the small intestine to the stoma, the surgery is called an ileostomy. If the doctor leaves part of the large intestine and connects that to the stoma, the surgery is called a colostomy.

Later, the doctor will do the pull-through. The doctor disconnects the intestine from the stoma and attaches it just above the anus. The stoma isn't needed any more, so the doctor either sews it up during surgery or waits about 6 weeks to make sure that the pull-through worked.

What will my child's life be like after surgery?

Ostomy

Most babies are more comfortable after having an ostomy because they can pass gas more easily and aren't constipated anymore.

Older children will be more comfortable, too, but they may have some trouble getting used to an ostomy. They will need to learn how to take care of the stoma and how to change the bag that collects stool. They may be worried about being different from their friends. Most children can lead a normal life after surgery.

Nurses at the hospital can teach you and your child how to care for a stoma and can talk to you about your worries.

Adjusting After Pull-through

After a pull-through, 9 out of 10 children pass stool normally. Some children may have diarrhea for a while, and babies may develop a nasty diaper rash. Eventually the stool will become more solid and the child will need to go to the bathroom less often. Toilet training may be delayed, as the child learns how to use the bottom muscles only after pull-through surgery. Older children might stain their underwear for a while after the surgery. It is not their fault. They can't control this problem, but it improves with time.

Some children become constipated because 1 in 10 children with Hirschsprung's disease has difficulty moving stool through the part of the colon without nerve cells. A mild laxative may also be helpful. Ask your doctor for suggestions.

Diet and Nutrition

Drinking plenty of liquids is important after surgery for Hirschsprung's disease.

One job of the large intestine is to collect the water and salts the body needs. Since your child's intestine is shorter now, it absorbs less. Your child will need to drink more to make sure his body gets enough fluids.

An infant who has long-segment disease requiring an ileostomy may need special tube feedings. The shortened intestine does not allow the bloodstream enough time to absorb nutrients from food before it is pushed out of the body as stool. Tube feedings that deliver nutrients can make up for what is lost.

Eating high-fiber foods like cereal and bran muffins can help reduce constipation and diarrhea.

Infection

Infections can be very dangerous for a child with Hirschsprung's disease. Infection of the large and small intestines is called enterocolitis. It can happen before or after surgery to treat Hirschsprung's disease. Here are some of the signs to look for:

fever

swollen abdomen vomiting diarrhea bleeding from the rectum sluggishness

Call your doctor immediately if your child shows any of these signs. If the problem is enterocolitis, your child may be admitted to the hospital. In the hospital, an intravenous (I.V.) line may be needed to keep body fluids up and to deliver antibiotics to fight the infection. The large intestine will be rinsed regularly with a mild salt water solution until all remaining stool has been removed. The rinse may also contain antibiotics to kill bacteria.

When the child has recovered from the infection, the doctor may advise surgery. If the child has not had the pull-through surgery yet, the doctor may prepare for it by doing a colostomy or ileostomy before the child leaves the hospital. If the child has already had a pull-through operation, the doctor may correct the obstruction with surgery.

Enterocolitis can be life threatening, so watch for the signs and call your doctor immediately if they occur.

Long-segment Hirschsprung's disease

Sometimes Hirschsprung's disease affects most or all of the large intestine, plus some of the small intestine. Children with long-segment Hirschsprung's disease can be treated with pull-through surgery, but there is a risk of complications such as infection, diarrhea, and diaper rash afterward. Parents need to pay close attention to their child's health. Also, since some, most, or all of the intestine is removed, drinking a lot of fluid is important.

http://surgery.med.umich.edu/pediatric/clinical/physician_content/a-m/hirschsprung.shtml

Enterocolitis Associated with Hirschsprung's Disease

Despite major improvements in the understanding of the pathophysiology and genetics of Hirschsprung's disease, little advancement has occurred in either elucidating the etiology of Hirschsprung's-associated enterocolitis (HAEC) or its prevention. Despite the recognition of this process in Harald Hirschsprung's first description of the disease in 1886, little attention was given to this disease process for the next 70 years (1). Swenson and Fisher, in 1956, were the first to recognize the association of enterocolitis and Hirschsprung's disease (2). Over the past four decades, HAEC has been a major cause of morbidity and mortality in infants and children with Hirschsprung's disease. With an increased understanding of the mucosal defense mechanisms, a host of potential causative factors have been advanced to explain this disease process. This review will attempt to sort out what is known clinically about this disorder and will review potential etiologies and therapy of HAEC.

Clinical Presentation and Diagnosis

Bill and Chapman were the first to accurately characterize the clinical aspects of HAEC (3). They speculated that the cause of this disorder was a partial mechanical obstruction similar to the colitis associated with other forms of bowel obstruction. Their description of the natural history of the process helped to alert physicians to the high risk of HAEC in Hirschsprung's disease patients. The classic clinical manifestations that they described in HAEC include abdominal distention, fever and foul smelling stool (3). There is, however, a wide range of clinical presentations of HAEC. A compilation of symptoms from patients with HAEC, treated at our Children's Hospital, noted the following in decreasing frequency: abdominal distension, explosive diarrhea, vomiting, fever, lethargy, rectal bleeding and shock (4). Many cases of diarrhea or abdominal distention may be mistakenly diagnosed as a gastroenteritis or the obstructing sphincter syndrome; however, most of these are cases of mild HAEC. To facilitate the diagnosis of HAEC, Elhalaby, et al developed a clinical grading system (Table 1) based on several clinical criteria (5). An occasional case of HAEC may present as a perforation of the bowel proximal to the aganglionic segment (6, 7).

The diagnosis of HAEC is typically based on the classic presentation of a neonate with a history of constipation starting in the newborn period, followed by abdominal distention and liquid, foul-smelling stool. Examination will show a markedly distended abdomen which is hyperesonant to percussion. Rectal examination often results in an explosive discharge of gas and stool. Post-pullthrough HAEC may present in a very similar fashion and is typically seen within the first two years following the child's pullthrough (8). Abdominal radiographs may be quite helpful. The proximal colon is distended with an almost toxic megacolon appearance (Lellie, 1997 Holschnier). An extremely useful finding is what we term a 'cut-off sign' in the recto-sigmoid region (Figure 1A) with an absence of air distally (4). This sign can be seen in all forms of HAEC and was noted in 74% of patients during an HAEC episode compared to only 14% of the time in-between

episodes of HAEC. Other common findings were small bowel dilatation in 74% and multiple air-fluid levels in 79%. Occasionally, pneumatosis intestinalis may be seen (4, 9). Because of the risk of perforation, a contrast enema should not be done in the presence of clinical HAEC. However, it is not uncommon to see findings of subclinical HAEC during such radiologic studies (Figure 1B). Typically, an irregular mucosal lining, with a resultant 'sawtooth' appearance is seen.

The timing of HAEC and the clinical course of Hirschsprung's disease shows that the two times an infant is at highest risk for the development of HAEC is either before the diagnosis of Hirschsprung's disease has been made, or following the definitive pullthrough. Although occasionally described, HAEC is distinctly uncommon in those patients with a decompressing colostomy (10). The diagnosis of Hirschsprung's disease after the first week of life places the neonate at a substantially higher risk for the development of HAEC (11). In this report, the mean age at diagnosis of neonates with Hirschsprung's disease was 16.6 days in those who developed HAEC and 4.6 days in those neonates without HAEC. Post-pullthrough HAEC may be due to associated internal sphincter spasm which is commonly associated with Hirschsprung's disease and may act to functionally obstruct the passage of stool (12). Although the course of enterocolitis is usually most severe in those infants who have not yet had the diagnosis of enterocolitis made, a recent report describes the death of 5 infants due to HAEC anywhere from 3 weeks to 20 months after their pullthrough procedure (13).

Incidence and Associated Risk Factors

The incidence of HAEC varies widely among reported series. Table 3 lists the incidence of HAEC in several large series. The mean incidence was 25%, but the range was quite wide (from 17% to50%) and may represent differences in the manner in which HAEC is diagnosed. This is perhaps most evident by the large variation in mortality rates in two of the largest series of Hirschsprung's disease patients (14, 15). In a review of the Surgical Section of the American Academy of Pediatrics, Kleinhaus reported a low rate of HAEC, but a high mortality rate. Whereas in a review of Japanese cases of Hirschsprung's disease, Ikeda noted a high rate of HAEC, but a low mortality rate (14, 15). Additionally, a clear decline in the incidence has occurred over the past 40 years with improved and more prompt diagnosis of the disease as demonstrated by the 50% incidence of HAEC in Bill and Chapman's series, and the much lower incidence (Table 3) in more recent series (3).

Several factors have been associated with an increased incidence of enterocolitis. As stated above, one fairly well substantiated risk factor is a delay in the diagnosis of Hirschsprung's disease (10, 11). Others have claimed that once an infant develops HAEC, they are at increased risk to develop other enterocolitic episodes in the future (16). Although some have speculated that early development of HAEC may somehow alter the defense mechanisms of the intestine and predispose the patient to recurrent episodes of HAEC, many other investigators have not found such a predisposition (11, 17, 18).

Increased length of the aganglionic segment has also been associated with the development of HAEC (8, 14, 15, 19). Intuitively, if the disease process is complicated by the degree of obstruction, longer lengths of aganglionosis should be associated with a higher incidence of HAEC. However, others have failed to find such an association (10, 11, 17, 18).

Infants with Trisomy 21 appear to be at increased risk of developing HAEC (20). In one series on HAEC, almost 45% of infants with Trisomy 21 and Hirschsprung's disease developed HAEC (11). This association of Trisomy 21 and HAEC has been confirmed in subsequent series (10, 18, 21). More than likely, this association is due to an immune deficiency, both humoral and cellular, which probably predisposes these infants to HAEC (22-24).

Other associated anomalies may place the infant at risk for the development of enterocolitis. Caneiro found that 53% of infants with associated anomalies developed HAEC compared to 26% with Hirschsprung's disease alone (18). Elhalaby noted that 47% of infants with anomalies developed HAEC compared to 29% without (4). It is not clear if this association is due to the large number of patients with Down's Syndrome; and, unfortunately, neither author adequately analyzed this data by separating out patients with Trisomy 21.

Post-Pullthrough Enterocolitis

Rates of post pullthrough enterocolitis vary widely among series (table 4), ranging from 2% to 27%. In the two largest series, a significantly higher incidence of enterocolitis was noted in those patients undergoing the Swenson pull-through. This higher incidence was noted in Swenson's own review from 1975, and this may be due to the inclusion of several patients who underwent this pull-through in the earlier years of the Swenson procedure, prior to it being modified to a more distal anastomosis. Post-operative enterocolitis has been associated with a fairly high rate of mortality in several series. In fact, when examining those deaths due to Hirschsprung's disease, several series noted that approximately 50% of deaths resulted from complications directly related to an enterocolitic episode.

Pathology

The gross pathologic description of HAEC (Figure 2) is probably best given by Harald Hirschsprung's first report of the disease itself (25). The following is a quote from his description of the lesion:

'...in the second case separate larger and deeper ulcerations that penetrate to the serosa, and indications of peritonitis can be seen on the serosal surface. Near the larger ulceration, we find an abscess under the mucosa that measures 2 cm...Mottled spaces can be seen in the submucosa containing pus.'

Like many other inflammatory disease processes of the intestine, HAEC is manifested by the appearance of neutrophils within the crypts of the intestine. A careful analysis of the natural progression of this disease process shows several discrete phases which can be used to grade the pathologic severity of the enterocolitic process. A grading system is shown in table 2. The system goes from a grade 0 with no pathologic abnormality to grade I which shows a marked amount of mucus streaming from the crypts of the intestine (Figure 3A). This mucin retention is a histopathologic process unique to only two diseases, Hirschsprung's and cystic fibrosis. Though not absolute, the diagnosis of Hirschsprung's disease is suggested based on this finding alone from a suction rectal biopsy, even without sufficient submucosa. Subsequent grades of HAEC show a progressive increase in crypt abscesses (Figure 3B), followed by the destruction of the intestinal epithelium and eventual perforation of the bowel. The latter 3 grades may look quite similar to ulcerative colitis. Our own group, and others, have had occasional difficulty in differentiating between HAEC and ulcerative colitis (26, 27). Although there are reports of an ischemic enterocolitis as a complication of Hirschsprung's disease, this finding is unusual (28). In a report by Teich, et al, all four of the cases described involved extremely ill neonates who were in septic shock, and this low perfusion state may have been the actual cause of the ischemic findings. Others have reported the association of necrotizing enterocolitis and HAEC in the same patients based on the radiologic finding of pneumatosis intestinalis (18). This radiologic finding, however, can be non-specific and does not necessarily indicate the child has necrotizing enterocolitis.

Morbidity and Mortality

The morbidity and mortality associated with the development of HAEC is quite high. Hospitalizations for many of these infants can be complicated and lengthy. Caneiro noted that hospitalizations ranged from 6 to 29 days (mean 13 days) (18). The cost of caring for an infant with HAEC is over two and one-half times as high as that for an infant with Hirschsprung's disease and no enterocolitis (11). Certainly in this era of cost containment, the ability to avoid such a complication is particularly critical. In addition to morbidity, mortality may also occur. Although death is generally an uncommon complication of HAEC, many series include reports of infants dying of the disease, and in many they comprise the majority of all deaths due to Hirschsprung's disease (Table 4). Mortality rates range from 0% to 33%; again most likely reflecting differences in the way in which HAEC is diagnosed (Table 3). Mortality rates also appear to be due to associated factors such as Trisomy 21.

Potential Causes

An appreciation of the pathologic changes in HAEC allows for consideration of potential pathophysiologic processes which could cause this disease. Historically, Swenson and Fisher postulated that the disorder was due to a defect in water and electrolyte metabolism (2). Subsequently, Swenson revised this concept and stated that improper fluid absorption was actually the result of chronic constipation (26). Several more recent theories on the etiology of HAEC have since been proposed including: infectious,

ischemic, obstructive, as well as hypersentivity reactive causes (29). Below potential contributing factors for the development of HAEC are discussed.

ObstructionPerhaps one of the earliest proposed etiologies of HAEC was primary obstruction. This concept was originally proposed by Glotzer, et al and was further supported by Bill and Chapman (3). In Glotzer's model, he created an artificial obstruction in a dog's colon (30). Although the pathologic description showed multiple punctate ulcerations and some focal inflammation, the pathologic description of HAEC mentioned above, including the profuse amount of mucus and the large number of crypt abscesses, was not seen. Thus, although obstruction may account for some part of the pathogenesis, the etiology of HAEC is probably more complicated.

InfectiousInfectious etiologies, including bacterial and viral, of HAEC have been proposed by several investigators. Hardy, et al found an increased incidence of Clostridium difficile (C difficile) in their patients with HAEC (31). In their report, they noted that children with Hirschsprung's disease had persistent retention of C difficile beyond 12 months of age, despite the usual loss of this organism in gastrointestinal tracts of normal children. This suggested that patients with Hirschsprung's disease may be more susceptible to infectious agents such as C difficile, and that this may predispose them to a high incidence of infections. This higher incidence of C difficile was also noted in Caneiro's series, and interestingly, both reports were from London, England (18). In other reports on HAEC, C difficile was identified in very fewer infants with HAEC (29, 32), suggesting that the specific type of organism may be less important that the actual predisposition to infections with pathologic organisms. With a lack of normal motility, bacterial overgrowth may occur which could allow such pathogens to grow in larger than normal amounts. The entero-invasive property of an organism, allowing it to penetrate the epithelial barrier, may have greater importance than the specific strain of organism (see below). Wilson-Storey has suggested that HAEC may be associated with a rotavirus infection (33).

MucusThe profound amount of mucus seen in the early stages of HAEC has prompted investigators to examine alterations in the intestinal mucus of Hirschsprung's disease patients. Mucus is a comprised of both several glycoproteins as well as secretory immunoglobulin A (IgA), and serves to protect the intestine by directly binding and inactivating a number of different organism (34). Akary first noted the abnormal composition of mucins from patients with Hirschsprung's disease (35). In his work, no changes were noted in the secretion of mucus; however, an increase in the sulphated mucins was noted in Hirschsprung's disease patients, particularly those with enterocolitis. These changes were further confirmed by Teitelbaum, et al, who also noted a loss of acidic (sulphated) mucins and an increase in neutral mucins in colonic sections obtained from patients with HAEC (29). More recently, a study of the dynamic turnover of mucins produced by epithelium from patients with Hirschsprung's disease has been done (36). In this

report, the investigators studied colonic mucosal cultures and demonstrated a significant decrease in mucin turnover in both aganglionic and ganglionic bowel of patients with Hirschsprung's disease compared to age-matched controls. This correlates well with the observed mucin retention seen in both aganglionic and ganglionic bowel, and suggests that an abnormal mucus defense barrier may contribute to HAEC.

Enterocyte AdherenceEnterocyte adherence appears to be a major way in which organisms can gain access to the epithelial barriers (37). A study of HAEC demonstrated that 39% of patients with clinicial , pre-pullthrough HAEC had enterocyte adherence on histologic examination of tissue specimens. The organisms in this study were either Esherichia coli, Clostridia difficile or Cryptosporidium. This suggests that the adherent nature of the organism may be more important that the actual organism itself.

Alteration in Intestinal ImmunocytesOnce an organism adheres to the epithelium of the intestine, it still must get past the complex immunologic system of the intestinal wall. The immunologic defense mechanisms of the intestine comprise both a humoral branch via the production of secretory immunoglobulin A (S-IgA) and a cellular branch with a rich source of T-lymphocytes within the epithelium (intraepithelial lymphocytes), lamina propria, discrete lymphoid follicles (Peyer's Patches) and in mesenteric lymph nodes [Lundqvist, 1996 #838]. Investigation into the alterations noted in these lymphoid populations must be tempered with the fact that changes may be either a primary cause of HAEC or a consequence of its development. Several investigators have examined changes in S-IgA formation and secretion. An analysis of white blood cell (WBC) counts of infants with HAEC has demonstrated lower counts and decreased WBC function compared to other patients (38). This may reflect an overall defect in the immunologic system of patients with HAEC; however, more insight has been found from examination of the intestinal immune system.

Wilson-Storey and Scobie noted in human patients that S-IgA levels were apparently normal in the intestinal lamina propria, however, it failed to be released into the intestinal lumen (39). Imamura, et al examined IgA, IgM and J chain containing plasma cells in patients with HAEC and noted an increased amount of these immunoglobulins (40). However, lower levels of luminal secretory components were found in the aganglionic colons of HAEC patients, again suggesting that there may be a defect in the release of these immunoglobulins. In another study of lamina propria IgA levels, however, lower levels were noted in Hirschsprung's disease patients and even lower levels were found in those patients with HAEC (41).

Using a mouse model of Hirschsprung's disease, the piebald lethal strain which generally dies of enterocolitis within the first 90 days of life, has allowed further immunologic investigation of the pathogenesis of HAEC. Compared to heterozygote controls, an initial rise in S-IgA serum levels was seen early in the

life of these mice, followed by a decline as HAEC developed (42). Similarly, Fujimoto noted that although immunocytes in the lamina propria that were positive for S-IgA were higher in piebald lethal mice, they were markedly lower during acute illness (43). This suggests that declines in S-IgA levels may be a result of HAEC not a cause.

An assessment of T-lymphocyte and natural killer immune systems has also been performed (40, 41). In both of these studies, it was noted that increased levels of T-lymphocytes were found in the bowel wall of patients with HAEC compared to other patients with Hirschsprung's disease. Examination of natural killer cells showed a significant increase in these cells in the lamina propria of ganglionic bowel of patients with HAEC compared to patients with Hirschsprung's disease without HAEC patients and control patients (40). Finally, the aganglionic portion of Hirschsprung's disease has a decreased production of nitric oxide, which may also contribute to the production of HAEC. Nitric oxide has several non-specific actions for defense against bacterial, mycobacterial and viral agents (44). Although this has not been investigated, a decline in its production may result in a deficient immunologic barrier.

Importantly, no study has been done which has examined the gastrointestinal tract in a serial fashion to document whether the observed changes in the intestine occur as a result of HAEC or are the primary cause of HAEC. Figure 5 summarizes the several potential etiologies for the development of HAEC discussed above.

Intestinal Neuronal DysplasiaThe association of Hirschsprung's disease with Intestinal Neuronal Dysplasia (IND)has been described in the literature (45). In a recent report by this same group, several patients with significant stooling problems, including HAEC, after definitive pull-through were shown to have evidence of IND, despite the complete removal of the aganglionic segment (46). In an examination of 10 patients with persistent bowel problems, 5 of whom had enterocolitis, all showed some evidence of IND. It is, however, uncertain what percent of all patients with post-pullthrough HAEC patients may have such abnormalities. Although we have not personally seen IND in association with Hirschsprung's disease in our experience, investigation on the cause of post-pullthrough HAEC should include a rectal biopsy which should be examined for aganglionosis as well as for the presence of IND.

Treatment

In 1964, Swenson suggested that the treatment of HAEC should be rectal tube decompression (2). Rectal washouts should be the initial approach in the care of a child, regardless of age, who presents with HAEC. The technique consists of the use of a soft red rubber catheter (at least 16 Fr) with multiple side holes cut to facilitate drainage. The catheter should be gently advanced (often as saline is flowing) above the aganglionic level. After catheter advancement, gas and stool should be aspirated, followed by

repetitive irrigations of 20 to 30 ml of saline, followed by removal of the tube. The procedure should be repeated every 4 to 6 hours until the child is decompressed. Importantly, enemas without a decompressing tube, should be avoided as they may worsen the enterocolitis. Not uncommonly, this treatment alone can often alleviate an even a fairly high grade enterocolitis.

Along with washouts, intravenous antibiotics or, in mild cases, oral metronidazole should also be used. Should the disease process fail to improve or the infant deteriorate, consideration should be given, in a neonate, to the performance of a leveling colostomy. This typically occurs in infants with long segment disease where rectal washouts cannot reach the dilated proximal bowel (47).

Recently, rectal irrigations have been used in a prophylactic fashion after the definitive pullthrough procedure (48). In this report, a significant reduction in the number of patients who developed enterocolitis was noted. It is possible that the washouts served to decompress the intestine for the first few months after the pullthrough procedure or may have served to either prevent colonic distention or washout enteropathogenic organism in the colonic lumen.

If repeated bouts of enterocolitis persist after the definitive pullthrough procedure, then an investigation into a mechanical cause should be undertaken. This should start with a contrast enema to insure that there is no obstruction in the neo-rectum, a potentially correctable cause (11). If this is normal, a suction rectal biopsy should be performed to rule out retention of an excess amount of aganglionic bowel as well as other possible etiologies such as IND (46, 49). Should these biopsies be normal, as is usually the case, consideration should be given to the performance of a posterior anal myotomy or myectomy. Published experience with this technique by several groups has demonstrated it to be safe and have a very low to absent incidence of fecal incontinence (50, 51). This more recent experience, however, is in contrast to a relatively poor experience with the procedure by Swenson who felt it did not benefit many of his patients (8). Polley, et al performed 3 internal sphincterotomies in 8 patients with persistent enterocolitis and Marty, et al performed 8 in 37 patients with post-pullthrough enterocolitis (52, 53). Overall, results of these internal sphincterotomies are quite good; however, both of these authors advocate a significant period of conservative therapy, since most patients with post-pullthrough enterocolitis will improve over time.

Conclusions

The development of enterocolitis in patients with Hirschsprung's disease remains a perplexing problem. Clearly recognition of this problem is essential to either its prevention or to early intervention in its treatment.

Table 1. Clinical grading system for Hirschsprung's-associated enterocolitis.

Grade Clinical symptoms

IMild explosive diarrhea, mild or moderate abdominal distension, and no systemic manifestations

IIModerate explosive diarrhea, moderate to severe abdominal distention, mild systemic symptoms

IIISevere explosive diarrhea, marked abdominal distention, shock or impending shock

Go Back to Text

Table 2. Pathologic grading of Hirschsprung's-associated enterocolitis.

Grade Pathologic findings

0 Normal mucosa

I Crypt dilatation, mucin retention

II Cryptitis or < 2 crypt abscesses/HPF

III Multiple crypt abscesses/HPF

IV Fibrinopurulent debris and mucosal ulceration

V Transluminal necrosis or perforation

Go Back to Text

Table 3. Incidence of Enterocolitis in several reported series of Hirschsprung's disease (3, 8, 10, 11, 14, 15, 17-19, 54).

Report OverallTrisomy

21Incidence of Long

segmentPre-

pullthroughMX*

Kleinhaus 18% ns 25% 15% 30%

Ikeda ns ns 44.3%29.2%(24.3 to

44.3)1.8-2.4%

Teitelbaum 24% 46% 29% 16% 16%

Elhalaby 33.9% 37.5% 55% not stated 0%

Rescorla 18% 26% 32% 6% 9%

Caneiro 32% 50% not different 16% 4%

Bill 50% ns 66% 45% 33%

Foster 17% ns 5% 10% 0%

Surana 30% 47% 38% 13% 10%

*Mortality (Mx) due to enterocolitis is based on the total number of infants with enterocolitis. Go Back to Text

Table 4. Enterocolitis (HAEC) post-pullthrough from several large series of Hirschsprung's disease (11, 12, 14, 15, 52, 53, 55-59).

Report Type of Incidence of HAEC Percent of deaths

Pullthrough Enterocolitisneeding Surgery

due to post pullthrough enterocolitis

Kleinhaus Mix1RPT 2%Duhamel 6%Swenson 16%

none 75%

Rehbein Rehbein 0% none

Holschneider Mix ERPT

Chapter

13.2%Duhamel 4.7%Swenson 3.7%Rehbein 6.3%

none

Teitelbaum Duhamel 6.3% 60% 0%

Ihezue ERPT2 7.4% none

Swenson Swenson 21% none 46%

Marty Mix 27% 22% 71%

Harrison Mix 12% none 10%

Polley ERPT 16% 38% 0%

Ikeda MixERPT 12% Duhamel 14%Swenson 34%

none

Elhallaby ERPT 21.4% 16% 0%

1 - Many different types of pullthroughs2 - Endorectal pullthrough

Hirschsprung's Disease

Although Hirschsprung's disease has been recognized for over 100 years, it is only within the past 3 decades that improved surgical management has been achieved and only within the past few years that an appreciation for the origin of this disease has developed. Over the past few years, significant insight into the etiology of Hirschsprung's disease has taken place. Recent refinements in the surgical techniques for Hirschsprung's disease have occurred. Included in these advances is the performance of the endorectal pull-through in the newborn period. Others advances in our understanding of Hirschsprung's disease have occurred through detailed studies of patients with Hirschsprung's disease including those with Trisomy 21 syndrome, persistent stooling abnormalities after the pull-through and associated neuronal intestinal dysplasia (NID).

Genetic Deletion in Hirschsprung's Disease

Perhaps one of the most exciting developments in Hirschsprung's disease is the recent attempt at identifying its etiology. Several advances have recently been made in determining the genes associated with Hirschsprung's disease. It has long been appreciated that Hirschsprung's disease may affect more than one family member in 4 to 8% of cases [1] . Martucciello, et al found a deletion in the long arm of chromosome 10 that was associated with Hirschsprung's disease [2]. This child had long segment (total colonic Hirschsprung's disease) and no family history. Last year, further investigation by this group and others narrowed the location of this mutation between 10q11.2 and q21.2 ( Figure 1) [3, 4] . In the work by Luo, et al the location of the Hirschsrpung's genetic abnormality overlaped the region of the RET proto-oncogene. The RET proto-oncogene appears to play a major role in the development of the intestinal nervous system and its deletion is also found in patients with the multiple endocrine neoplasia type 2A (MEN 2A) [5] . This may explain why a few patients with Hirschsprung's disease also have MEN 2A [6].

Etiology of Hirschsprung's disease

Several investigators have suggested that the development of Hirschsprung's disease is due to a lack of migrating nerve cells to develop. One study has recently examined neural cell adhesion molecules (NCAM) in Hirschsprung's disease. Bowel containing ganglion cells (both in control patients and in those with Hirschsprung's disease) had a large amount of NCAM, whereas there was an absence of NCAM in the aganglionic segments [9]. NCAM is believed to be important in nerve cell migration to specific locations during embryogenesis [10] . One might speculate that a loss of NCAM may explain the developmental absence of ganglion cells in Hirschsprung's disease (Figure 2). Another intense area of investigation is the relation of Hirschsprung's disease to nitric oxide

production. It has been well recognized for several years that the loss of ganglion cells in Hirschsprung's disease results in a loss of nerves [11].

Surgical Therapy

With increasing awareness of Hirschsprung's disease by the pediatric community and easier accessibility to obtain a diagnosis with suction rectal biopsy, the diagnosis of Hirschsprung's disease is being made at a much earlier age [21] . As a result, one of the most significant changes in the management of Hirschsprung's disease is the performance of the pull-through procedure in the newborn period. This approach is in contrast to the previous approach of a colostomy during the newborn period with a subsequent pull-through at 9-12 months of life once the child had attained a weight of approximately 20 pounds [15] . Initial success with an initial endorectal pull-through was first noted by So, et al in 1980 [17]. Cilley, et al reported on 15 neonates who underwent a one-stage endorectal pull-through [18*]. Ages at the time of surgery ranged from 2 to 52 days (mean 13 days) and no operative complications occurred. The technique is virtually identical to that performed in older children (Figure 3). All operations were performed within 24 hours of diagnosis and as early as 48 hours of age.

The Duhamel procedure is another approach now used in the neonatal period [19] . In this report, a primary Duhamel pull-through was performed in 22 infants whose ages ranged from 14 to 90 days (mean 44 days). Postoperative complications occurred in 18%, including 3 cases of enterocolitis and one case of a retained rectal spur. Kücükaydin, et al reported on their experience with the Swenson procedure in 10 neonates [20]. A primary pull-through was performed in 6 of these 10 infants with no postoperative complications. Unlike the previous two series, however, these operations included performance of a colostomy.

Limited surgery for Hirschsprung's disease

Anal myectomy has been performed as definitive surgery for low-segment Hirschsprung's disease. A fair number of these patients, however, had persistent constipation requiring enemas and laxatives on a regular basis. Although the authors were

satisfied with these results, most surgeons are reluctant to perform an anal myectomy because of the risk of leaving a significant amount of abnormal aganglionic bowel.

Stooling abnormalities following surgery

Another difficult area in Hirschsprung's disease is the care of the patient with persistent stooling abnormalities after a pull-through. Occasionally, the surgeon is faced with a child with persistent constipation or recurrent enterocolitis after their pull-through procedure. Because the frequency of these problems is fairly low, little has been written about the management of these children. Rectal biopsy shall be performed in the evaluation of the patient with persistent stooling problems after a pull-through. A diagnostic work-up for such patient is shown above.

Trisomy 21

Another group of patients with Hirschsprung's disease that is particularly challenging to manage are those with Trisomy 21 (Down's Syndrome). The association of Trisomy 21 and Hirschsprung's disease has only recently been appreciated [27] . In a study by Quinn, et al, 13% of their patients with Hirschsprung's disease had Trisomy 21 [28] . Interestingly, the outcome of these patients was quite poor. Although a definitive pull-through was performed on 13 of these 17 patients, only one of 13 surviving patients had normal bowel function. Two of the patients had to revert back to a colostomy. These authors suggest that a definitive pull-through in a child with Trisomy 21 requires careful consideration. These authors' results are somewhat in conflict with another previously published series on patients with Trisomy 21 and Hirschsprung's disease [27] . In this latter series, 6 of 13 patients with Trisomy 21 were deemed to be candidates for a

definitive pull-through. Each of these children had a good result with daytime continence although nighttime incontinence did occur.

Enterocolitis of Hirschsprung's disease

Another difficult group of patients are those with Hirschsprung's associated enterocolitis (HEC). This entity can be difficult to differentiate from gastroenteritis or a child with an excessively spastic internal anal sphincter. An intestinal 'cut-off' sign (loss of rectal air approximately at the top of the pelvis on X-ray).

Neuronal Intestinal Dysplasia

Neuronal intestinal dysplasia (NID) is an unusual disorder of the gastrointestinal tract that may be related to Hirschsprung's disease. Three basic forms of the disease are noted. These are: hyperganglionosis (giant ganglia) in Auerbach's (myenteric muscle) plexus; abnormal ganglia; or a combination of the two. All three of these forms may be seen with Hirschsprung's disease; and each of these may be localized or involve all or most of the intestinal tract. Investigators have noted the occurrence of Hirschsprung's disease in 20-75% of NID cases [26- 27,30] . The significance of NID as seen on pathologic examination As such, some surgeons rely predominately on the clinical presentation of constipation for planning any subsequent surgical intervention.