Sigmoidoscopy and cytology in the detection of microscopic disease of the rectal mucosa in ulcerative colitis.

EDITORIAL SYNOPSIS This study confirms the value of both the sigmoidoscopic assessment and the cytology of the mucosal secretion in the detection of microscopic disease of the rectal mucosa in ulcerative colitis. On certain occasions, however, microscopic disease is undetected using these methods, so that rectal biopsy is recommended as an aid in the diagnosis of patients with a history suggestive of colitis, but having normal or equivocal sigmoidoscopic appearances. Inspection of the rectal mucous membrane at sigmoidoscopy is generally held to be the most valuable single diagnostic aid in ulcerative colitis. However, a recent study by Baron, Connell, and Lennard-Jones (1964) has demonstrated the difficulties encountered in the precise definition of the mucosal appearances in this disease, considerable discrepancies having been found in the verdicts of several independent observers on the sigmoid-oscopic features of identical areas of mucosa. The relationship between the sigmoidoscopic and microscopic appearances of the rectal mucous membrane in colitis has been studied by Truelove and of these groups of workers has shown that microscopic inflammatory changes in rectal mucosa may be present when sigmoidoscopic appearances are normal, although in general a close correlation was observed between macroscopic and microscopic appearances. In this paper we have attempted to evaluate the sigmoidoscopic appearances and the cytology of the mucosal secretion in ulcerative colitis by comparison with the histological findings in rectal biopsy specimens. As a preliminary step, data are provided on the observer variation in describing the macroscopic mucosal appearances in this disease, the most consistent terms being defined for use in recording sigmoidoscopic appearances. The study comprised 104 patients suffering from idio-pathic ulcerative colitis currently attending the Colitis Clinic, the severity of disease ranging from complete quiescence to severe activity, and the extent of disease varying from localized rectal involvement to complete rectal and colonic involvement at the time of examination; and, as normal controls, 22 patients with simple intemal haemorrhoids and no history of colitis, who were undergoing routine sigmoidoscopy in the Rectal Clinic. The investigations relevant to the present study were: