结肠镜下结直肠腺瘤切除术后再发风险预测模型的建立和验证  

Establishment and evaluation of a nomogram model for predicting the recurrence of colorectal adenoma after colonoscopic resection

作　　者：张君 孙百合 方雅丽[1] 张瑜[1] 刘家铭 庄坤[1] ZHANG Jun;SUN Baihe;FANG Yali;ZHANG Yu;LIU Jiaming;ZHUANG Kun(Department of Gastroenterology and Hepatology,Xi'an Central Hospital,Shaanxi Xi'an 710004,China)

机构地区：[1]西安市中心医院消化病院,陕西西安710004

出　　处：《现代肿瘤医学》2024年第3期496-502,共7页Journal of Modern Oncology

摘　　要：目的:分析结肠镜下结直肠腺瘤切除术后再发的危险因素,构建并验证预测结直肠腺瘤切除术后再发风险的列线图模型。方法:回顾性收集西安市中心医院消化科2012年01月至2021年12月于结肠镜下行结直肠腺瘤切除术且在术后有结肠镜随访结果的患者968例,其中496例组成建模组,472例组成验证组。应用logistic回归模型分析结直肠腺瘤切除术后再发的独立危险因素,并构建预测结直肠腺瘤切除术后再发风险的列线图模型,然后进行验证。结果:多因素logistic回归分析显示,年龄(OR=1.027,95%CI:1.007~1.048)、腺瘤数目(OR=1.232,95%CI:1.123~1.353)、病理分型(OR=2.442,95%CI:1.342~4.441)为影响腺瘤术后再发的独立危险因素(P<0.01)。纳入上述指标构建列线图预测模型,并在建模组和验证组中进行验证。建模组与验证组ROC曲线下面积分别为0.670 (95%CI:0.622~0.718)和0.735(95%CI:0.691~0.780),模型区分度良好。两组校准曲线显示列线图模型的预测值和实际观察结果一致性良好。决策曲线显示该模型的临床净收益也较好。结论:我们的研究基于年龄、腺瘤数目、腺瘤病理分型这3项独立危险因素,建立了预测结直肠腺瘤切除术后再发风险的列线图模型,该模型纳入的指标均是内镜医师非常容易获取的,便于内镜医师快速简便地预测腺瘤术后再发风险,临床实用性高。Objective:To identify the risk factors of colorectal adenoma recurrence after colonoscopic resection,and develop a nomogram model for predicting the recurrence of adenoma after colonoscopic resection.Methods:A total of 968 patients who underwent colonoscopic resection of colorectal adenoma and were followed up by colonoscopy were retrospectively analyzed from January,2012 to December,2021 at digestive department in Xi'an Central Hospital.They were divided into training cohort(n=496)and validation cohort(n=472).Multivariate logistic regression was used to screen out the independent risk factors of adenoma recurrence after colonoscopic resection to construct a nomogram.And the model was verified in both training cohort and validation cohort.Results:Multivariate logistic regression analysis revealed that age(OR=1.027,95%CI:1.007~1.048),the number of adenomas(OR=1.232,95%CI:1.123~1.353),and the pathological type of adenomas(OR=2.442,95%CI:1.342~4.441)were independent risk factors of adenoma recurrence after colonoscopic resection(P<0.01).The nomogram model was constructed based on the multivariate analysis results of the training cohort and verified in both training cohort and validation cohort.The areas under the receiver characteristic curve(ROC)of the training cohort and the validation cohort were 0.670(95%CI:0.622~0.718)and 0.735(95%CI:0.691~0.780),respectively,with a good discrimination.The calibration curve of two groups showed that the predictive probability of the model was in high consistency with the actual incidence.The decision curve analysis(DCA)suggested that the patient's clinical net benefit level was acceptable.Conclusion:This study constructed and validated a nomogram that accurately predicts the recurrence of colorectal adenoma after colonoscopic resection,based on age,the number of adenomas,and the pathological type of adenomas,which were identified the independent risk factors of adenoma recurrence.The indicators included in this model were very easy to obtain,so that endoscopists can quickly and

关 键 词：结直肠腺瘤 术后再发风险 危险因素 列线图 

分 类 号：R735.35[医药卫生—肿瘤]