Determination of Benchmark Risk of Colorectal Cancer (CRC) at Subsequent Surveillance Colonoscopies

Background: Little is known about the benchmark risk of CRC between surveillance colonoscopies, which is obstructing us to examine surveillance intervals of colonoscopy based on CRC incidence. We aim to determine the benchmark risk through a retrospective cohort. Methods:

Methods: A population-based, retrospective cohort study was identified in a database of colonoscopy screening and surveillance. Patients had colonoscopy (January 2010- March 2014) were selected as study subjects and their examination history was reviewed. Multivariable Weibull regression model was used to estimate the incidence of CRC at 10-year follow-up among patients with low risk level (neither CRC history nor polyp), which was defined as the benchmark risk because the national guidelines recommend a 10-year surveillance interval for them. Results:

Results: Out of total 27,325 events, 5774 (21%) didn’t find any polyp at baseline colonoscopies. Half of them were male, average age was 61±9 years. The incidence at 10-year follow-up (benchmark risk) was 164 and 79 per 100,000 person-years for male and female, respectively. Both male and female exceeded their benchmarks in 3-5 years if they had incomplete polyp removal, ≥3 adenomas during last colonoscopy or personal CRC history, while in 5-10 years if only family CRC history was present. Coexisting risk factors resulted in a sharp increase in the incidence exceeding the benchmarks in less than 2-3 years Conclusions:

Conclusions: The benchmark risk of CRC at subsequent colonoscopy could be used as a threshold for examining surveillance intervals. Gender disparity and coexisting risk should be considered in the future adjustment of intervals.