Polyp characteristics at screening colonoscopy and post-colonoscopy colorectal cancer mortality: a retrospective cohort study.

Abstract

Polyp size and high-grade dysplasia in polyps at screening colonoscopy are considered risk factors for post-colonoscopy colorectal cancer (PCCRC) development and death, which might be averted by surveillance colonoscopy. However, robust evidence backing these risk factors is lacking. We aimed to investigate whether polyp size or dysplasia grade is associated with PCCRC mortality. This was a retrospective study including individuals of the Austrian Quality Certificate for Screening Colonoscopy who underwent a colonoscopy between January 2007 and December 2020. We investigated the association of polyp size and dysplasia in polyps with PCCRC mortality according to Cox regression analysis. In addition, whether patients with certain polyp characteristics had similar risk for CRC death compared with the Austrian population was assessed by calculating standardized mortality ratios (SMRs). A total of 316,001 individuals were included. After a median follow-up time of 5.27 years (95% confidence interval [CI], 5.25-5.29), a significant association of polyps 10 to 20mm (hazard ratio, 4.00; 95% CI, 2.46-6.50; P< .001) as well as high-grade dysplasia (hazard ratio, 6.61; 95% CI, 3.31-13.2; P< .001) with PCCRC death was observed. PCCRC mortality was significantly lower than the expected CRC mortality in the general population in patients with polyps<10mm and without high-grade dysplasia (SMR, .27; 95% CI, .21-.33; P< .001), which was not observed for patients with polyps≥10mm or with high-grade dysplasia (SMR, 2.05; 95% CI, 1.64-2.57; P< .001). Polyp size≥10mm and high-grade dysplasia are associated with PCCRC mortality in screening patients. The data suggest that these patients might benefit most from surveillance colonoscopy.