Background: Supplementary observational data in the community setting are required to better assess the predictors of colorectal polyp recurrence and the effectiveness of colonoscopy surveillance under real circumstances.
Aim: The goal of this study was to identify patient characteristics and polyp features at baseline colonoscopy that are associated with the recurrence of colorectal polyps (including hyperplastic polyps) among patients consulting private practice physicians.
Patients and methods: This cohort study was conducted from March 2004 to December 2010 in 26 private gastroenterology practices (France). It included 1023 patients with a first-time diagnosis of histologically confirmed polyp removed during a diagnostic or screening colonoscopy. At enrollment, interviews were conducted to obtain data on socio-demographic variables and risk factors. Pathology reports were reviewed to abstract data on polyp features at baseline colonoscopy. Colorectal polyps diagnosed at the surveillance colonoscopy were considered as end points. The time to event was analyzed with an accelerated failure time model assuming a Weibull distribution.
Results: Among the 1023 patients with colorectal polyp at baseline, 553 underwent a surveillance colonoscopy. The mean time interval from baseline colonoscopy to first surveillance examination was 3.42 (standard deviation, 1.45) years. The recurrence rates were 50.5% and 32.9% for all polyps and adenomas, respectively. In multivariate models, the number of polyps at baseline was the only significant predictor for both polyp recurrence (hazard ratio [HR] 1.19, 95% CI 1.06 to 1.33), and adenoma recurrence (HR 1.17, 95% CI 1.03 to 1.34).
Conclusion: The efficacy of surveillance colonoscopy in community gastroenterology practice compared favorably with academic settings. This study provides further evidence that the number of initial colorectal polyps is useful for predicting the risk of polyp recurrence, even in the community setting.