Validation and modification of a diagnostic scoring system to predict microscopic colitis

Area of Science:

• Gastroenterology
• Internal Medicine
• Diagnostic Medicine

Background:

• Diarrhoea often leads to colonoscopy, with random biopsies taken if macroscopic findings are normal to exclude microscopic colitis (MC).
• Most patients with diarrhoea and normal colonoscopy have functional bowel disorders, not MC.
• Identifying accurate predictors of MC could reduce unnecessary biopsies and healthcare costs.

Purpose of the Study:

• To validate a previously described diagnostic scoring system for microscopic colitis (MC).
• To assess if incorporating additional variables improves the diagnostic performance of the MC scoring system.
• To identify predictors that can help avoid random colonic biopsies in patients with diarrhoea.

Main Methods:

• Prospective data collection of demographic and symptom data from adults with loose stools undergoing colonoscopy.
• Application of a previously established diagnostic scoring system for MC.
• Assessment of additional variables, including medications, stool frequency, nocturnal stools, and symptom duration, for inclusion in the scoring system.

Main Results:

• The original scoring system (cut-off ≥4) showed 92.3% sensitivity and 35.2% specificity for MC in 242 patients (10.7% with MC).
• Nocturnal passage of stools and symptom duration <6 months were significant predictors of MC.
• A new scoring system (cut-off ≥6) incorporating these factors achieved 95.7% sensitivity and 46.0% specificity.

Conclusions:

• Incorporating nocturnal stool passage and shorter symptom duration into a scoring system may enhance the prediction of microscopic colitis (MC).
• This improved prediction could lead to a greater avoidance of random colonic biopsies in patients presenting with loose stools.
• Refined diagnostic tools can optimize patient management and resource allocation in gastroenterology.