Abstract
Background When complex nonmalignant polyps are detected in colorectal cancer (CRC) screening
programs, patients may be referred directly to surgery or may first undergo additional
endoscopy for attempted endoscopic removal by an expert. We compared the impact of
both strategies on screening effectiveness and costs.
Methods We used MISCAN-Colon to simulate the Dutch screening program, and projected CRC deaths
prevented, quality-adjusted life-years (QALYs) gained, and costs for two scenarios:
1) surgery for all complex nonmalignant polyps; 2) attempted removal by an expert
endoscopist first. We made the following assumptions: 3.9 % of screen-detected large
nonmalignant polyps were complex; associated surgery mortality was 0.7 %; the rate
of successful removal by an expert was 87 %, with 0.11 % mortality.
Results The screening program was estimated to prevent 11.2 CRC cases (–16.7 %) and 10.1
CRC deaths (–27.1 %), resulting in 32.9 QALYs gained (+ 17.2 %) per 1000 simulated
individuals over their lifetimes compared with no screening. The program would also
result in 2.1 surgeries for complex nonmalignant polyps with 0.015 associated deaths
per 1000 individuals. If, instead, these patients were referred to an expert endoscopist
first, only 0.2 patients required surgery, reducing associated deaths by 0.013 at
the expense of 0.003 extra colonoscopy deaths. Compared with direct referral to surgery,
referral to an expert endoscopist gained 0.2 QALYs and saved €12 500 per 1000 individuals
in the target population.
Conclusion Referring patients with complex polyps to an expert endoscopist first reduced some
surgery-related deaths while substantially improving cost-effectiveness of the screening
program.
