Prediction of Patients with Microscopic Colitis Prior to Colonoscopy: Testing of a Proposed Scoring System in a Community Gastroenterology Practice in the United States: 619

Introduction: A scoring system has been proposed to identify patients with microscopic colitis (MC) prior to endoscopy.1 We sought to validate this scoring system in a community gastroenterology (GI) practice setting. Methods: We performed a retrospective review of 1643 colonoscopies completed for the indication of “chronic diarrhea” at a community GI practice between January 2011 and December 2015 for patients between the ages of 18 and 90. Patients with known IBD or conditions that could explain their chronic diarrhea (e.g., infectious diarrhea, celiac disease) were excluded. Previously proposed risk factors were extracted and weighted according to the proposed scoring system1 that ranged from -8 to +38 and included female gender (+4), age ≥50 (+13), use of PPIs (+6), use of NSAIDs (+11), presence of weight loss (+4), and presence of abdominal pain (-8). Independent sample t-tests and receiver operating curve analyses were performed using SPSS to determine the diagnostic accuracy of this scoring system. Results: In our validation cohort of 182 patients, 91 were diagnosed with MC (74% lymphocytic colitis; 26% collagenous colitis) on the basis of histologic analysis and 91 matched patients were identified as having a macroscopically normal colonoscopy with no evidence of MC on random colonic biopsies. Female gender (p< 0.001), age ≥ 50 (p< 0.001), absence of abdominal pain (p< 0.001), and NSAID use (p=0.026) were significantly associated with the diagnosis of MC. Applying the proposed scoring system with a score cut off of ≥4, patients with MC were identified with 82% sensitivity and 55% specificity [area under receiver operating curve (AUC) = 0.75]. Conclusion: We found comparable performance of the proposed scoring system1 for MC in our population of patients undergoing colonoscopy for the indication of chronic diarrhea at a North American community GI practice. This finding supports the utility of a scoring system to identify patients at low risk for MC prior to colonoscopy who may not need mucosal biopsy. (1) Kane JS, Rotimi O, Everett SM, Samji S, Michelotti F, Ford AC. Development and validation of a scoring system to identify patients with microscopic colitis. Clin Gastroenterol Hepatol. 2015 Jun; 13(6):1125-3.Figure 1Figure 2