OBJECTIVETo analyze the risk factors of carcinogenesis of large colorectal polyps (diameter ≥ 10 mm) found by colonoscopy.

METHODSClinicopathological and follow-up data of 418 consecutive patients who were diagnosed as colorectal polyps with diameter≥10 mm by colonoscopy at two endoscopy centers of the Affiliated Wuxi Second People's Hospital, Nanjing Medical University (n=207) and Zhongshan Hospital, Fudan University (n=211) from January 2015 to December 2016 were retrospectively collected. High-grade intraepithelial neoplasia and cancer were defined as malignancy in this study. Chi square test was used for univariate analysis, and logistic regression was used for multivariate analysis (in patients with multiple polyps, if the pathological findings were all low grade intraepithelial neoplasia, one polyp with the largest diameter was selected to enter the model; in patients with high grade intraepithelial neoplasia, one polyp of high grade intraepithelial neoplasia with the largest diameter was selected to enter the model). Associated risk factors of malignancy were analyzed.

RESULTSAmong the 418 patients, 278(66.5%) were male and 140(33.5%) were female, with mean age of (58.7±10.2) (range 15-87) years old. Of 398 patients undergoing endoscopic treatment with resected 456 polyps, 142 cases with 150 polyps were malignant, including 134 polyps of high-grade intraepithelial neoplasia and 16 polyps of intra-mucosal cancer. The other 20 patients showed negative elevation signs after endoscopic submucosal injection and were transferred to surgery, of whom 20 polyps were resected. Histological examination of these 20 polyps indicated invasive cancer. Univariate analysis showed that age ≥ 50 years [40.5% (150/370) vs. 25.0% (12/48), χ² =4.323, P=0.041], multiple polyps [77.5%(31/40) vs. 34.7%(131/378), χ² =12.900, P=0.001], polyp locating at rectum [59.0%(36/61) vs. 32.3%(134/415), χ² =22.736, P=0.000], polyp diameter ≥31 mm [74.1%(20/27) vs. 33.4%(150/449), χ² =36.493, P=0.000] and tubular villous adenoma [67.4%(120/178) vs. 16.8%(50/298), χ² =71.810, P=0.000] were associated with malignancy. Multivariate analysis showed that age ≥ 50 years(OR=2.473, 95%CI:1.209-5.058, P=0.013), multiple polyps (OR=2.472, 95%CI: 1.300-4.702, P=0.006), polyp locating at rectum (OR=1.253, 95%CI: 1.091-1.439, P=0.001) and the polyp diameter ≥31 mm (OR=1.500, 95%CI:1.196-1.881, P=0.000) were independent risk factors for malignancy of large colorectal polyps. The mean follow-up time was (9.6±4.2) months. During the follow-up period, 86 patients (20.5%) who received endoscopic resection developed recurrent adenoma which all were successfully removed by colonoscopic polypectomy. Two patients(0.5%) developed colon cancer 6 months after endoscopic resection and both underwent radical surgery and chemotherapy. Their previous pathology from endoscopic resection was tubular villous adenoma and high grade intraepithelial neoplasia. All the patients were alive during the follow-up period.

CONCLUSIONSAge ≥50 years old, multiple polyps, polyps locating at rectum and polyps with diameter ≥ 31 mm are the risk factors of malignancy. Emphasized examination should be recommended for those with the above mentioned risk factors to avoid missed diagnosis and misdiagnosis. The choice of endoscopic treatment must be reasonable for curative resection.