Incomplete Polyp Resection Rate of Colorectal Polyps in the Chinese Population: A Cross-Sectional Study

INTRODUCTION: Colonoscopy effectively reduces the morbidity and mortality of patients with colorectal cancer. However, incomplete resection can result in postcolonoscopy colorectal cancer. Data on the incomplete resection rate (IRR) and associated risk factors are limited; therefore, we aimed to evaluate the IRR by using data from the National Colorectal Polyp Care study. METHODS: We included patients with at least 1 polyp (1-20 mm) removed during colonoscopy. After polyp removal was completed, biopsies were obtained from the resection margins to assess the completeness of the polypectomy. The main outcome was the IRR of neoplastic polyps. Risk factors of incomplete resection were assessed by multivariate logistic regression. RESULTS: A total of 2,041 individuals (62% men) with a mean age of 55 years were included. A total of 2,453 polyps were included in the analysis, of which an IRR of 4.4% was observed for 1,944 neoplastic polyps. The overall IRR was 4.1% for endoscopists with at least 10 resections, but it broadly ranged from 0% to 36.4%, especially among endoscopists with a high level of narrow-band imaging (NBI) experience (>3,000 procedures). Multivariate logistic regression demonstrated that sessile serrated adenomas/polyps (SSA/Ps, absolute risk [AR]SSA/Ps = 0.09, ARadenomas = 0.04, difference in AR [RD] = 0.05), polyps located in the proximal colon (ARproximal colon = 0.06, ARdistal colon = 0.04, RD = 0.02), and low level of NBI experience (1,000-3,000 procedures, ARlow NBI experience = 0.05, ARhigh NBI experience = 0.03, RD = 0.02) were independent risk factors of a high IRR. DISCUSSION: The IRR of polyps varies greatly in clinical practice, even among experienced endoscopists. SSA/Ps and polyps located in the proximal colon tend to be incompletely resected, and a high level of NBI experience might help reduce the IRR.