Article Text

Rectal biopsy in the investigation of constipation
  1. Department of Pediatrics
  2. Shaare Zedek Medical Center
  3. Hebrew University Medical School
  4. PO Box 3235, Jerusalem 91031, Israel
    1. R BABU,
    2. L CAMADOO,
    3. D P DRAKE,
    4. A PIERRO,
    5. E M KIELY,
    6. L SPITZ
    1. Department of Paediatric Surgery
    2. Great Ormond Street Hospital for Sick Children
    3. London WCIN 3JH, UK

      Statistics from

      Editor,—Ghosh et alconcluded that using rectal biopsy to investigate constipation and, thus, rule out Hirschsprung’s disease, is unnecessary if symptoms begin after the neonatal period.1 We agree with the authors that in most cases there is a history of constipation in the first month of life.

      However, we saw two children recently whose symptoms began at four and five months old, respectively. They had suction rectal biopsy and were subsequently diagnosed with Hirschsprung’s disease, with excellent results after surgery. The parents were unable to provide a neonatal history. A Belgian review supports the view that symptoms can occur first after 1 month of age; they reported that constipation starts, on average, at 38 days.2 Thus, we feel that rectal biopsy may still be useful in atypical cases of infantile constipation.


      Rectal biopsy in the investigation of constipation

      Editor,—Ghosh and Griffiths1-1 reviewed 141 diagnostic rectal biopsy samples to develop criteria for performing the procedure in children with intractable constipation. They concluded that a biopsy sample was unnecessary if the child was not in the neonatal period (older than 28 days) at onset of symptoms. This conclusion was based on the study of 17 children with Hirschsprung’s disease who had developed symptoms within the neonatal period.

      We performed a retrospective study of 122 patients with histologically proven Hirschsprung’s disease who were treated in our department between 1988 and 1994 (7 years). We found that 8% of our patients had no symptoms in the neonatal period; age at onset of symptoms was between 2 and 36 months (mean 7.5 months); this is not a new finding.1-2 1-3

      Failure to perform biopsy samples in children with late onset of symptoms may delay the eventual diagnosis and could increase the risk of developing enterocolitis. We feel that the study from Southampton1-1 is based on a very small cohort of patients with Hirschsprung’s disease and that the authors’ conclusion regarding the indication for rectal biopsy is incorrect.


      1. 1-1.
      2. 1-2.
      3. 1-3.

      Dr Griffiths comments:

      We reviewed three years of rectal biopsies, and Landman2-1 reviewed five years of rectal biopsies in children with constipation; 41 children were studied and there were no cases of Hirschsprung’s disease that started after the neonatal period.

      Babu et al, state that 8% of their patients had no neonatal symptoms and that this is not a new finding. Wheatley and colleagues did not study the onset of symptoms, but found that “each patient had a history of abdominal distension as well as severe, lifelong constipation necessitating regular cathartics or enemas”.2-2 Swenson2-3 conducted a 16 year review of 501 cases of Hirschsprung’s disease, in three hospitals, which ended in 1973. He does not comment on when symptoms began, and reports the passing of meconium in only 268 children; this suggests that retrospective neonatal data collection was difficult. However, study of the birth history of the children (which is of unspecified length), revealed that 93% (453 of 495) were constipated and that 87% (316 of 356) had abdominal distension.

      Dr Wilschanski and colleagues cite the paper by Reding,2-4 a 20 year review which studied the outcome of surgical management, with the biopsy date for completeness. Unfortunately, Redinget al confused age of presentation with age of onset of symptoms (age at presentation in the legend, and age at onset in the table). Thus, we emphasise that the age of presentation is irrelevant, and in our study, it ranged from 1 day to 3 years. However, we found onset of symptoms in children with Hirschsprung’s disease was always within the neonatal period.

      We reiterate that if a detailed history focusing on the neonatal period is taken, many children whose constipation began after the neonatal period could avoid unnecessary biopsy, and the most cases of Hirschsprung’s disease will be found (in our 41 cases, 100%).

      However, if there are other reports of patients with Hirschsprung’s disease who were genuinely asymptomatic in the neonatal period, a greater proportion of babies will need a biopsy.


      1. 2-1.
      2. 2-2.
      3. 2-3.
      4. 2-4.
      View Abstract

      Request Permissions

      If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.