Polyp detection rates as quality indicator in clinical versus screening colonoscopy

Background Adenoma and Polyp Detection Rates (ADR and PDR) are advocated as general performance measures for screening and clinical colonoscopy, but their evidence is largely derived from screening data. This study compares PDRs in colonoscopy for screening versus clinical indications. Methods Consecutive patients at two Norwegian centers were examined by eight endoscopists either for colonoscopy screening in a randomized colonoscopy screening trial (Nordic-European Initiative on Colorectal Cancer, NordICC) or for clinical indications during the same time period (January 2013 to December 2014). PDR-5mm, defined as the proportion of colonoscopies with detection of at least one polyp with diameter 5mm, was measured prospectively. We fitted multivariable logistic regression models and calculated the adjusted odds ratios (OR) to evaluate factors for differences in PDR-5mm between screening and clinical colonoscopies. Results The study included 2939 clinical and 771 screening colonoscopies. The PDR-5mm was 26% and 31%, respectively (P=0.005). Among sex, age, cecum intubation, bowel cleansing, and endoscopist, only the latter explained the higher PDR-5mm in screening compared to routine colonoscopy. In the fully adjusted logistic regression model, the detection of polyps 5mm was not associated with indication for colonoscopy. The OR for polyp detection in screening vs. routine colonoscopy was 1.04; 95% confidence interval 0.85-1.27. Conclusion In this study, the differences in PDR-5mm between clinical and screening colonoscopies could be explained by the endoscopist. Accordingly, PDR-5mm benchmarks may be similar for clinical and screening colonoscopy.