Diagnostic Yield of Colonoscopy in Patients With Symptoms Compatible With Rome IV Functional Bowel Disorders

BACKGROUND & AIMS: There is little data on the diagnostic yield of colonoscopy in patients with symptoms compatible with functional bowel disorders (FBDs). Previous studies have only focused on diagnostic outcomes of colonoscopy in those with suspected irritable bowel syndrome using historic Rome I-III criteria, whilst having partially assessed for alarm features and shown markedly conflicting results. There is also no colonoscopy outcome data for other FBDs, such as functional constipation or functional diarrhea. Using the contemporaneous Rome IV criteria we determined the diagnostic yield of colonoscopy in patients with symptoms compatible with a FBD, stratified diligently according to the presence or absence of alarm features. METHODS: Basic demographics, alarm features, and bowel symptoms using the Rome IV diagnostic questionnaire were collected prospectively from adults attending out-patient colonoscopy in 2019. Endoscopists were blinded to the questionnaire data. Organic disease was defined as the presence of inflammatory bowel disease, colorectal cancer, or microscopic colitis. RESULTS: 646 patients fulfilled symptom-based criteria for the following Rome IV FBDs: IBS (56%), functional diarrhea (27%) and functional constipation (17%). Almost all had alarm features (98%). The combined prevalence of organic disease was 12%, being lowest for functional constipation and IBS-constipation (similar to 6% each), followed by IBS-mixed (similar to 9%), and highest amongst functional diarrhea and IBS-diarrhea ( similar to 17% each); p = .005. The increased prevalence of organic disease in diarrheal versus constipation disorders was accounted for by microscopic colitis (5.7% vs. 0%, p < .001) but not inflammatory bowel disease (7.2% vs. 4.0%, p = .2) or colorectal cancer (4.2% vs. 2.3%, p = .2). However, 1-in-4 chronic diarrhea patients - conceivably at risk for microscopic colitis - did not have colonic biopsies taken. Finally, only 11 of 646 (2%) patients were without alarm features, in whom colonoscopy was normal. CONCLUSIONS: Most patients with symptoms of FBDs who are referred for colonoscopy have alarm features. The presence of organic disease is significantly higher in diarrheal versus constipation disorders, with microscopic colitis largely accounting for the difference whilst also being a missed diagnostic opportunity. In those patients without alarm features, the diagnostic yield of colonoscopy was nil.

Automated summary

There is limited data on the diagnostic yield of colonoscopy in patients with symptoms compatible with functional bowel disorders (FBDs). This study aimed to determine the diagnostic outcomes of colonoscopy in FBD patients using Rome IV criteria and categorizing them based on the presence or absence of alarm features. The presence of organic disease was found to be significantly higher in patients with diarrheal disorders compared to constipation disorders, mainly due to microscopic colitis. Colonoscopy had no diagnostic value in patients without alarm features.