Chronic, unexplained diarrhea: are biopsies necessary if colonoscopy is normal?

• Published in: The American journal of gastroenterology Vol. 90; no. 3; p. 372
• Main Authors: Marshall, J B, Singh, R, Diaz-Arias, A A
• Format: Journal Article
• Language: English
• Published: United States 01.03.1995
• Subjects: Adult; Biopsy - economics; Biopsy - utilization; Chronic Disease; Colitis - epidemiology; Colitis - pathology; Colon - pathology; Colonic Diseases - epidemiology; Colonic Diseases - pathology; Colonoscopy; Costs and Cost Analysis; Diarrhea - etiology; Diarrhea - pathology; Female; Humans; Intestinal Mucosa - pathology; Male; Predictive Value of Tests; Rectal Diseases - epidemiology; Rectal Diseases - pathology; Rectum - pathology; Sigmoidoscopy
• ISSN: 0002-9270
• DOI: 10.1111/j.1572-0241.1995.tb08229.x

We sought to determine the frequency of clinically important histological abnormalities in patients with chronic, unexplained diarrhea who had macroscopically normal colonic endoscopies. Of 855 consecutive patients undergoing colonoscopy (595 cases) or flexible proctosigmoidoscopy (260 cases) by one endoscopist, biopsies were taken in 111 cases of unexplained diarrhea of at least 4-6 weeks duration in which the colorectal mucosa appeared grossly normal. All biopsies were blindly reviewed by one pathologist. In this group of patients with macroscopically normal colons, we identified no cases of Crohn's disease or ulcerative colitis or any definite cases of collagenous colitis (CC) or lymphocytic colitis (LC). There was one case classified as "possible CC" and 13 cases classified as "some features of LC". There were five cases of melanosis coli, one case of cytomegalovirus colitis (in an immunosuppressed patient), and one case of radiation injury. Ninety-one cases were classified as no pathological diagnosis or minimal histological change. Patients with abnormal histology were contacted to see if they had persistence or resolution of diarrhea. For the cases of "possible CC" and "some features of LC," diarrhea had resolved spontaneously in the majority. Interesting to note, only one of the five melanosis coli patients admitted to laxative use, raising the question of surreptitious abuse. We conclude that the yield of biopsies in diarrhea patients with macroscopically normal colons at endoscopy is low. It may be reasonable to obtain biopsies in patients with relatively severe or debilitating symptoms, with diarrhea that sounds "organic" (e.g., nocturnal stools, frequent watery stools, weight loss, elevated sedimentation rate), or in patients who are immunosuppressed. When biopsies are taken at colonoscopy, we suggest taking about six from throughout the colon and placing them into just one specimen container to help minimize costs.