Propensity score matching analysis of risk factors of colorectal adenoma
10.3760/cma.j.cn311367-20200924-00569
- VernacularTitle:结直肠腺瘤危险因素的倾向评分匹配分析
- Author:
Tiantian LEI
1
;
Bo LIN
;
Hongyu HUANG
;
Hongsheng MA
;
Jinlin YANG
Author Information
1. 四川大学华西医院日间手术中心，成都 610041
- Keywords:
Colorectal adenoma;
Ambulatory surgery;
Propensity score matching;
Risk factors
- From:
Chinese Journal of Digestion
2021;41(6):410-415
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the risk factors of occurrence of colorectal adenoma after endoscopic polypectomy.Methods:From January 2014 to December 2019, at the Department of Day Surgery Centre in West China Hospital, Sichuan University, 6 430 patients with 20 351 polyps who underwent endoscopic colorectal polypectomy were retrospectively analyzed. Patients were divided into adenomas group (4 573 patients) and non-adenomas group (1 857 patients) according to whether they had at least one adenomatous polyp. According to the results of postoperative histopathology, colorectal polyps were divided into adenomatous polyp group (10 656 polyps) and non-adenomatous polyp group (9 695 polyps). The propensity score matching (PSM) method was applied, with 1∶1 matching, in the patients with adenoma group and the patients with non-adenoma group, as well as in adenomatous polyps group and non-adenomatous polyps group. A total of 1 824 pairs of patients and 7 362 pairs of colorectal polyps were successfully matched. After PSM, patients-related factors as gender (male), age (<40 and 40 to 60 years old), number of polyps (>2), obesity (body mass index ≥28 kg/m 2), melanosis, family history of colorectal cancer in first-degree relatives, polyps-related factors as the maximum diameter (6 to 10 and >10 mm), distribution (distal colon), and morphological classification (sessile and flat polyps) were included in the analysis of risk factors of colorectal adenoma. Univariate analysis and multivariate logistic regression were used for statistical analysis. Results:Among 6 430 patients with colorectal polyps, the detection rate of adenoma was 71.12% (4 573/6 430). After PSM, the results of univariate analysis showed that obesity, family history of colorectal cancer in first-degree relatives, the maximum diameter of polyps >10 mm were all correlated with the occurrence of adenoma (odds ratio ( OR)=1.483, 1.426 and 1.503, 95% confidence interval ( CI)1.063 to 2.067, 1.015 to 2.004, 1.198 to 1.887, all P<0.05). The results of multivariate logistic regression analysis indicated that obesity, family history of colorectal cancer in first-degree relatives, the maximum diameter of polyps >10 mm, sessile or flat polyps in morphological classification were independent risk factors of the occurence of colorectal adenomas ( OR=1.425, 1.411, 1.629, 1.165 and 1.151, 95% CI1.019 to 1.994, 1.001 to 1.988, 1.290 to 2.058, 1.030 to 1.316 and 1.012 to 1.310, all P < 0.05). Conclusions:Obesity, family history of colorectal cancer in first-degree relatives, maximum diameter of polyps >10 mm, sessile polyps or flat polyps were the independent risk factors of the occurrence of colorectal adenomas.
