Rethinking Post-colonoscopy Colorectal Cancer Risk: Endoscopist Performance Versus Presence of High-risk Polyps

Abstract

Background and Aims: Current post-polypectomy guidelines set intervals based solely on features of resected polyps. Despite the well-known inverse relationship between both adenoma detection rate (ADR) and proximal serrated polyp detection rate (PSPDR) with post-colonoscopy colorectal cancers (PCCRCs), both quality indicators are not considered when determining surveillance intervals. Methods: We used colonoscopy data from 2014 to 2020 performed for a positive fecal immunochemical test in the Dutch colorectal cancer screening program. Individuals were categorized into having high-risk polyps or no/low-risk polyps resected. The association between 3-year PCCRC-risk and presence of high-risk polyps and either ADR or PSPDR was studied with Cox proportional hazard regression. Secondly, endoscopists were categorized into low/medium/high ADR and PSPDR to enable stratified analysis. Results: A total of 239,217 individuals were included; 74,289 had high-risk polyps resected at baseline and 202 had PCCRC within the subsequent 3 years. Presence of high-risk polyps was not associated with PCCRC-risk (hazard ratio [HR], 1.00; 95% confidence interval [CI], 0.75–1.35), whereas ADR and PSPDR showed a strong association with PCCRC (per point increase HR, 0.94; 95% CI, 0.92–0.96; HR, 0.92; 95% CI, 0.89–0.95, respectively). For individuals with no/low-risk polyps but examined by endoscopists with low ADR, the HR of PCCRC was 2.11 (95% CI, 1.21–3.65), as compared with individuals with high-risk polyps but examined by endoscopists with high ADR. Conclusions: An individual's PCCRC risk in the initial years is primarily influenced by endoscopist performance, rather than the presence of high-risk polyps. To reduce PCCRCs, besides ensuring appropriate surveillance intervals, it is crucial to monitor and audit endoscopist quality indicators.