Overtreatment in nonmalignant lesions detected in a colorectal cancer screening program: a retrospective cohort study

Although colorectal cancer (CRC) screening programs reduce CRC incidence and mortality, they are associated with risks in healthy subjects. However, the risk of overtreatment and overdiagnosis has not been determined yet. The aim of this study was to report the surgery rates in patients with nonmalignant lesions detected within the first round of a fecal immunochemical test (FIT) based CRC screening program and the factors associated with it.

We included in this analysis all patients with nonmalignant lesions detected between May 2013 and June 2019 in the Galician (Spain) CRC screening program. We calculated surgery rate according to demographic variables, the risk classification according to the colonoscopy findings (European guidelines for quality assurance), the endoscopist's adenoma detection rate (ADR) classified into quartiles and the hospital's complexity level. We determined which variables were independently associated with surgery rate and expressed the association as Odds Ratio and its 95% confidence interval (CI).

We included 15,707 patients in the analysis with high (19.9%), intermediate (26.9%) low risk (23.3%) adenomas and normal colonoscopy (29.9%) detected in the analyzed period. Colorectal surgery was performed in 162 patients (1.03, 95% CI 0.87-1.19), due to colonoscopy complications (0.02, 95% CI 0.00-0.05) and resection of colorectal benign lesions (1.00, 95% CI 0.85-1.16). Median hospital stay was 6 days with 17.3% patients developing minor complications, 7.4% major complications and one death. After discharge, complications developed in 18.4% patients. In benign lesions, an endoscopic resection was performed in 25.4% and a residual premalignant lesion was detected in 89.9%. The variables independently associated with surgery in the multivariable analysis were age (≥60 years = 1.57, 95% CI 1.11-2.23), sex (female = 2.10, 95% CI 1.52-2.91), the European guidelines classification (high risk = 67.94, 95% CI 24.87-185.59; intermediate risk = 5.63, 95% CI 1.89-16.80; low risk = 1.43; 95% CI 0.36-5.75), the endoscopist's ADR (Q4 = 0.44, 95% CI 0.28-0.68; Q3 = 0.44, 95% CI 0.27-0.71; Q2 = 0.71, 95% CI 0.44-1.14) and the hospital (tertiary = 0.54, 95% CI 0.38-0.79).

In a CRC screening program, the surgery rate and the associated complications in patients with nonmalignant lesions are low, and related to age, sex, endoscopic findings, endoscopist's ADR and the hospital's complexity.