Anorectal disorders are very common in children. It has been estimated that constipation is the chief complaint in 3–5% of all visits to pediatricians and in 25% of pediatric gastroenterology consultations (1). Conventional therapy of childhood constipation is based on dietary manipulation, stool softeners, oral laxatives, and behavioral modification (2). Although these strategies are successful in the majority of patients, there is a small subgroup of patients that fails aggressive treatment and requires further investigation.
An uncommon cause of constipation is the lack of relaxation of the internal anal sphincter (IAS). Although most cases of IAS dysfunction are found in children with Hirschsprung disease, occasionally there are constipated children with ganglion cells in the rectum, but with a non-relaxing IAS. High rectal pressures are required during defecation to overcome the IAS resistance. Hence, these children are prone to fecal retention. In the past, they were thought to have ultra-short segment Hirschsprung disease, a condition clinically indistinguishable from severe idiopathic constipation and often surgically treated with myectomy (3). The lack of IAS relaxation has been also found in patients with intestinal neuronal dysplasia (4). Langer et al. have described the effective use of Clostridium botulinum toxin in children with surgically corrected Hirschsprung disease who had obstructive symptoms due to a hypertonic, non-relaxing IAS (5). This technique could theoretically produce the same effect as internal anal sphincter myectomy, but without the potential long-term problems associated with permanent sphincter destruction.
We have identified a group of patients with chronic constipation refractory to behavioral and pharmacologic treatments whose IAS failed to relax in response to rectal dilation during ano-rectal manometry (IAS achalasia) and who had no evidence of aganglionosis. These patients received a four-quadrant intramuscular injection of Clostridium botulinum toxin into the IAS. We report the characteristics of these children and their outcome after treatment with botulinum toxin injection, based on a retrospective chart review and data collected during telephone interviews. We compare their characteristics with those of other children with functional constipation attending a pediatric gastroenterology clinic.
PATIENTS AND METHODS
We describe 20 patients (8 males, age between 5 months and 17 years, mean: 5.8 ± 4.2 years), who were referred for management of severe constipation. Nine were evaluated at Children's Hospital of Pittsburgh and 11 at Children's Hospital of Boston. At the time of evaluation none of them had more than one spontaneous bowel movement per week (mean interval between stools was 10 ± 3 days). They had all failed pharmacologic and behavioral treatment and 17 had been hospitalized for one or more colon disimpaction procedures. All patients were receiving large doses of laxatives or frequent enemas, and 9 had been treated with daily rectal irrigations. Underlying organic causes of constipation had been ruled out in all patients. One child had developmental delay. In all patients, rectal biopsies demonstrated ganglion cells at least 2 cm above the pectinate line. All children underwent ano-rectal manometry and 9 had colonic manometry, according to published protocols (6,7). On the night before anorectal manometry, unless they had already received colonic lavage in preparation for colonic testing, the children received a phosphate enema. Five children received sedation with chloral hydrate at the time of the anorectal manometry. Anorectal manometry showed absence of the recto-anal inhibitory reflex in all (Fig. 1). The absence of the recto-anal inhibitory reflex in the presence of ganglion cells was the criterion for the diagnosis of IAS achalasia.
The injection of Clostridium botulinum toxin A (Botox, Allergan, Irvine, CA) was performed under general anesthesia after cleaning the rectum with phosphate enemas the preceding night. The clean out was then completed both manually and with irrigations in the operating room before the injections if necessary. The patients received four-quadrant intramuscular injections in the IAS. The injections were performed either by the surgeon or the gastroenterologist, both of whom were present in the operating room. Botox was diluted in saline to a concentration of 100U/ml. Using a small nasal or vaginal speculum with a blade length of 27 mm, the anus was dilated, the dentate line was identified by inspection and the IAS palpated. Using an insulin syringe with a short 27-gauge needle, 15 to 25U of Botox per quadrant were injected into the sphincter. Before each injection the syringe was aspirated to avoid an intravascular injection. All patients were discharged from the hospital on the day of the procedure following a period of brief observation.
A questionnaire administered by telephone was used to monitor the patients 4 weeks-18 months (mean 7.5 ± 4 months) after the injection. The questionnaire asked the parents to rank the patients' improvement as poor, moderate or excellent. Patients were classified by the physicians as “poor response” if they were having bowel movements less than every other day and if laxatives were still used. A “moderate response” was indicated if there was an increase in bowel movement frequency to at least every other day and the patient was still using laxatives. An “excellent response” was indicated if there was an increase in bowel movement frequency to at least every other day and the patient was using no laxatives. Number of bowel movements per week, presence or absence of soiling, use of medications to aid with passage of stools, and side effects related to the injection were also recorded.
The control group consisted of 20 consecutive, newly referred children (14 male, mean age 8.1 + 4.6 y; P > .05 vs. study children) with the chief complaint of constipation, who were seen in the Gastroenterology Clinic at Children's Hospital of Pittsburgh. These 20 patients all were diagnosed as having functional constipation, based on their clinical characteristics, testing and response to treatment.
The patient interviews and the reviews of medical records were approved by the Human Rights Committees of the Children's Hospital of Pittsburgh and Boston. Data are reported as mean + standard deviation. Comparisons between children with IAS achalasia and controls were done with Mann-Whitney U test and χ2 or the two-sided Fisher test. A P value <0.05 was considered as statistically significant.
The children with IAS achalasia had earlier onset of symptoms (2.5 + 4.1 vs 4.7 + 4.4 years, P < .05), less fecal soiling (35% vs. 85% of children, P < .005) and less withholding behavior (5% vs 60% of children, P < .001) than did the control children. Symptoms started at birth in 40% of children with IAS achalasia compared to 10% of control children, (P = .06).
In the children with IAS achalasia, basal sphincter pressures ranged from 45 to 130 mmHg (mean 70.8 ± 20.7 mmHg), (normal value 33–100 mmHg (8). Threshold for rectal sensation ranged from 40 to 300 ml (mean 146 ± 82 ml) (normal value < 20 ml). Colonic manometry was normal in the non-dilated portion of the colon in all 9 patients who had this test. There was no relationship between the results of colonic manometry and the clinical response to Botox injection.
Eight (40%) patients had a bowel movement within 48 hours and 18 (90%) patients within 72 hours of the injection of Botox. The remaining two children did not have a bowel movement for more than 3 days after the injection, remained constipated, and were considered treatment failures. In the patients who responded to therapy, the post-treatment stool frequency varied from twice daily to once every 3 days (mean 5.8 bowel movements/week).
The duration of the response was variable (Fig. 2), ranging from one week to more than 18 months (mean 14 ± 8 weeks). At the time of this report, 8 patients were still using laxatives, but none needed enemas or rectal irrigations to achieve bowel movements.
Four patients, who initially responded to therapy, but relapsed, underwent sphincterotomy. Three of them improved after the procedure. Six children received a second Botulinum toxin injection with sustained benefit for an additional 2 to 6 months (mean 3.5 ± 2 months). Two children had a third injection when the effect of the second wore off and both of them were having normal bowel movements almost 2 months afterwards. Five patients, including the one who did not improve after sphincterotomy, had other surgical interventions after failure of the Botulinum toxin injection or after the effect of the injection had worn off (three had an appendicostomy or a cecostomy, one a sigmoidectomy, and one a colostomy).
According to the parents, the response to treatment was defined as excellent in 60%, moderate in 35%, and poor in 5%. According to the physicians, the response was ranked as excellent in 35%, moderate in 40%, and poor in 25%. There was no correlation between basal sphincter pressure before the injection and improvement of symptoms after the injection (Fig. 3).
There were no systemic or local complications. No patient developed fecal incontinence. Only two patients had transient rectal pain for a few hours after the injection.
Most children with functional or behavioral constipation respond to medical and behavioral treatment. A small subset of children continues to be constipated and requires further diagnostic testing. Anorectal manometry has been used as a diagnostic and investigational technique in the evaluation of childhood constipation. The main indication to perform an anorectal manometry in a constipated child is to demonstrate the presence of the recto-anal reflex and to rule out Hirschsprung disease (9,1). In this report we have described a group of pediatric patients with constipation refractory to aggressive medical and behavioral treatment who were found to have abnormal IAS function and were treated with Clostridium botulinum toxin injection.
Internal anal sphincter dysfunction has typically been attributed to Hirschsprung disease and very little is known about IAS dysfunction in children without classic Hirschsprung disease. This rare entity has been previously described, and has sometimes been classified as ultra-short segment Hirschsprung disease (3) or anal achalasia (10). In the patients we studied there was no evidence of aganglionosis. We therefore suggest classifying them as having IAS achalasia. This condition may be an under-appreciated cause of constipation in children. In 1934, Hurst proposed that one of the possible pathogenic mechanisms of idiopathic constipation might be the failure of the internal sphincter to relax rather than an anal spasm (10). He found similarities between the hypertonicity of this smooth muscle and esophageal achalasia and suggested the term anal achalasia. The pathogenesis and pathophysiology of IAS achalasia is not understood, but intramuscular innervation abnormalities and defective innervation of the neuromuscular junction are believed to be responsible for anal motor dysfunction (11). Nitrergic nerve depletion, cholinergic hyperplasia, and abnormal peptidergic innervation have been reported (12). It has been suggested that there may be an absence of purinergic inhibitory neurons in the distal rectal myenteric plexus of these patients, preventing reflex inhibition of the internal sphincter after cholinergic stimulation in response to rectal distention (13).
By analogy to the treatment of esophageal achalasia, we injected botulinum toxin into the anal sphincter. It has been recently reported that successful treatment of similar children may be achieved with internal sphincter myectomy (14). Botulinum toxin in IAS achalasia may serve as both a diagnostic and therapeutic tool. It has been suggested that if a decrease in anal sphincter pressure after the injection does not improve symptoms, it is unlikely that a myectomy would be curative (15).
The mechanism of action of Clostridium botulinum toxin has been studied extensively. It is a potent bacterial neurotoxin that acts on the neuromuscular junction to block the release of acetylcholine from presynaptic cholinergic nerves (16,17). It weakens the muscle in a focal and transient fashion. This effect is dose-dependent and requires 4 to 6 months to wear off. Botulinum toxin has been used successfully in both adults and children to treat hypertonicity and spasticity of the skeletal and smooth muscles (18,19). In patients with esophageal achalasia, a condition associated with esophageal dysmotility, hypertonicity and impaired relaxation of the lower esophageal sphincter (LES), endoscopically delivered injection of this neurotoxin to the LES results in symptomatic relief in 60% of patients for up to six months (20,21). Botulinum toxin has been used to relax the sphincter of Oddi in patients with biliary dyskinesia (22). Injection of botulinum toxin into the anal sphincter has been also useful in adults for the treatment of non-relaxing puborectalis muscle and chronic anal fissures (23,24).
More recently it has been used in symptomatic children after pull-through surgery for Hirschsprung disease (5).
Most of our patients had a favorable response that lasted for several weeks. We also observed a few children with a much more prolonged response to the injection. It is unclear why such patients continued to do well even after the action of the toxin should theoretically have disappeared. We hypothesize that the long lasting benefit could be secondary to an improved rectal function. Children with chronic constipation often have a dilated rectum, and, after having normal bowel movements for weeks, they may normalize rectal volume and function. Thus, during defecation, they may be able to generate a higher rectal pressure and overcome the obstacle of the non-relaxing sphincter. The very high threshold for the urge to defecate found in most patients and the results of the barium enemas suggest the presence of a dilated rectum in the majority of children. The finding of normal colonic motility in the children who underwent this test, confirms that the patients had an outlet obstruction rather than a generalized motility disorder.
It is unclear whether the lack of response in some patients was secondary to a failure of the treatment, or was due to other problems, such as an injection outside of the sphincter, injection of inactive toxin, use of doses too low, or misdiagnosis. We also noticed that repeated injections were helpful for some patients who failed an initial injection. Thus, we suggest repeating one or two injections before considering a child a treatment failure. In patients with very short-term benefit, we cannot rule out a transient benefit from anal stretching during the injection; however, it is unlikely that such stretching may have led to a benefit lasting more than few days. The use of more aggressive surgical interventions such as sphincterotomy, cecostomy, or partial colectomy was justified in some children by the severity of their symptoms and the lack of benefit from any other therapeutic intervention.
The diagnosis of IAS achalasia can be definitively made only with the use of anorectal manometry. We found that the clinical symptoms that suggested the diagnosis included earlier onset of symptoms, lack of fecal soiling, and no history of withholding behavior. Because the clinical presentation of children with IAS achalasia may overlap with that of children with functional constipation, we suggest that an anorectal manometry be performed in children with intractable symptoms, even if they have rectal biopsies demonstrating ganglion cells. It may be argued that the lack of IAS relaxation on anorectal manometry could be the result of technical problems during the test, failure to stretch the wall of the dilated rectal vault or to simultaneous contraction of the external anal sphincter. We feel it is unlikely in this study that we were dealing with any of these situations. The studies were performed by experienced laboratories, and we inflated the rectal balloon in every child until we triggered either pain or an urge to defecate suggesting that there was contact between the balloon and the rectum. We also routinely demonstrate IAS relaxation in children with functional constipation who often contract the external anal sphincter when feeling the urge of defecate. We acknowledge that only with the use of electromyography could one be absolutely certain of measuring IAS and not external anal sphincter.
In summary, these preliminary, but encouraging data suggest that botulinum toxin injection may be useful for the treatment of children with IAS achalasia. However, a prospective, randomized, double blind study of this intervention is needed to conclusively demonstrate its efficacy.
Dr. Paola Ciamarra was the recipient of a Janssen fellowship in pediatric gastrointestinal motility.
1. Hatch TF. Encopresis and constipation in children. Pediatr Clin North Am 1988; 35:257–80.
2. Baker SS, Liptak GS, Colletti RB, et al. Constipation in infants and children: evaluation and treatment. A medical position statement of the North American Society for Pediatric Gastroenterology and Nutrition. J Pediatr Gastroenterol Nutr 1996; 22:38–44.
3. Neilson IR, Yazbeck S. Ultrashort Hirschsprung's disease: Myth or reality. J Pediatr Surg 1990; 25:1135–8.
4. Koletzko S, Ballauff A, Hadziselimovic F, et al. Is histologic diagnosis of neuronal intestinal dysplasia related to clinical and manometric findings in constipated children? Result of a pilot study. J Pediatr Gastroenterol Nutr 1993; 17:59–65.
5. Minkes RK, Langer JC. A prospective study of botulinum toxin for internal anal sphincter hypertonicity in children with Hirschsprung's disease. J Pediatr Surg 2000; 35:1733–6.
6. Nurko S, Garcia-Aranda JA, Yolena Guerrero V, et al. Treatment of intractable constipation in children: experience with cisapride. J Pediatr Gastroenterol Nutr 1999; 29:612–26.
7. Di Lorenzo C, Flores AF, Reddy SN, et al. Use of colonic manometry to differentiate causes of intractable constipation in children. J Pediatr 1992; 120:690–5.
8. Van der Plas RN, Benninga MA, Redekop WK, et al. Randomised trial of biofeedback training for encopresis. Arch Dis Child 1996; 75:367–74.
9. Molnar D, Taitz LS, Urwin OM, et al. Anorectal manometry results in defecation disorders. Arch Dis Child 1983; 58:257–61.
10. Hurst AF. Anal achalasia and megacolon (Hirschsprung's disease, idiopathic dilation of the colon). Guy's Hosp Rep 1934; 84:317–50.
11. Kobayashi H, Hirakawa H, Puri P. Abnormal internal anal sphincter innervation in patients with Hirschsprung's disease and allied disorders. J Pediatr Surg 1996; 31:794–9.
12. Oue T, Puri P. Altered intramuscular innervation and synapse formation in internal sphincter achalasia. Pediatr Surg Int 1999; 15:192–4.
13. Baumgarten HG, Holstein AG, Stelzner F. Nervous elements in the human colon of Hirschsprung's disease. Vichows Arch Pathol Anat 1973; 358:113–36.
14. De Caluwe D, Yoneda A, Akl U, et al. Internal anal sphincter achalasia: outcome after internal sphincter myectomy. J Pediatr Surg 2001; 36:736–8.
15. Minkes RK, Langer JC, Di Lorenzo C, et al. A prospective study of botulinum toxin for internal anal sphincter hypertonicity in children. Gastroenterology 1999; 116:A1043.
16. Hoogerwerf WA, Pasricha PJ. Botulinum toxin for spastic gastrointestinal disorders. Baillieres Best Pract Res Clin Gastroenterol 1999; 13:131–43.
17. Hallan RI, Williams NS, Melling J, et al. Treatment of anismus in intractable constipation with botulinum A toxin. Lancet 1988; 2:714–7.
18. Friedman A, Diamond M, Johnston MV, et al. Effects of Botulinum toxin A on upper limb spasticity in children with cerebral palsy. Am J Phys Med Rehabil 2000; 79:53–9.
19. Brant CQ, Siqueira ES, Ferrari Jr. AP Botox for oropharyngeal dysphagia: case report of flexible endoscope-guided injection. Dis Esophagus 1999; 12:68–73.
20. Cuilliere C, Ducrotte P, Zerbib F. Achalasia: outcome of patients treated with intrasphincteric injection of botulinum toxin. Gut 1997; 41:87–92.
21. Schiano TD, Parkman HP, Miller LS. Use of botulinum toxin in the treatment of achalasia. Dig Dis 1998; 16:14–22.
22. Wehrmann T, Seifert H, Seipp M. Endoscopic injection of botulinum toxin for biliary sphincter of Oddi dysfunction. Endoscopy 1998; 30:702–7.
23. Jost WH, Schimrigk K. Botulin toxin in therapy of anal fissure. Lancet 1995; 345:188–9.
24. Shafik A, El-Sibai O. Botulinum toxin in the treatment of non-relaxing puborectalis syndrome. Dig Surg 1998; 15:347–51.