PWE-020 Diagnosing Microscopic Colitis – is Colonoscopy Necessary?

    loading  Checking for direct PDF access through Ovid



Microscopic colitis is a common cause of chronic diarrhoea, particularly in older people, and the incidence is increasing. As the endoscopic appearance is typically normal, diagnosis of the two subtypes, collagenous and lymphocytic colitis, relies upon specific histology findings. When suspected, guidelines advise colonoscopy with full biopsy series due to reports of a patchy disease distribution, with false negative rates of up to 40% reported with flexible sigmoidoscopy.1 However, more recent data has challenged this assumption, leaving considerable uncertainty.2 We report one of the largest consecutive case series to date, examining whether flexible sigmoidoscopy alone is sufficient.


A retrospective review of all cases of microscopic colitis diagnosed at colonoscopy over a 12-year period (2001–2013) at our hospital was performed. Only colonoscopies with both right (proximal to splenic flexure) and left sided colonic biopsies were included. The diagnostic criteria for microscopic colitis were lymphocytic infiltration in the lamina propria and either >20 intraepithelial lymphocytes per 100 epithelial cells (lymphocytic colitis) or a collagenous layer >10 mm (collagenous colitis). The primary aim was to assess the proportion of patients in which microscopic colitis could be diagnosed on left sided biopsies alone.


84 patients were included in the study. 58 (69.0%) were female with a median age of 62 years. 44 (52.4%) had collagenous colitis and 40 (47.6%) lymphocytic colitis. 76 (90.5%) had features of microscopic colitis on both right and left sided biopsies, 7 (8.3%) right side only and 1 (1.2%) left side only. Hence a diagnosis of microscopic colitis could be made in 77 (91.7%) on left sided biopsies alone. Age, sex and histopathological subtype did not significantly alter the sensitivity of left sided biopsies.


Flexible sigmoidoscopy would have correctly diagnosed microscopic colitis in a very high proportion of patients (92%). Given that flexible sigmoidoscopy is less expensive, better tolerated, and can be combined with CT scanning to exclude a proximal malignancy, this may have important implications for the investigation of non-bloody diarrhoea.

Disclosure of Interest

None Declared.

Related Topics

    loading  Loading Related Articles