预测结直肠息肉发生风险列线图的建立及评价

Establishment and evaluation of nomogram for predicting the risks of colorectal polyps

目的通过分析结直肠息肉的发生危险因素,建立个体化预测结直肠息肉发生风险的列线图模型并评价其效能。方法收集2018年10月至2020年4月于该院行结肠镜检查患者的临床资料,通过单因素及多因素logistic回归分析结直肠息肉发生的独立危险因素;应用R软件建立预测结直肠息肉发生风险的列线图模型,用Bootstrap法进行模型内部验证,用验证集进行外部验证,采用C-指数、Calibration校正曲线及受试者工作特征(ROC)曲线评价列线图的预测性能。结果年龄、家族肿瘤史、吸烟、脂肪肝、胆囊疾病史、色素痣、高脂血症是结直肠息肉发生的独立危险因素,列线图的C-指数在训练集和验证集中分别为0.803(95%CI:0.761~0.845)、0.787(95%CI:0.716~0.858),Calibration校正曲线拟合良好,通过约登指数计算出列线图的最佳临界值为180.7分,临界值下的灵敏度、特异度、阳性预测值和阴性预测值在训练集中分别为70.4%、77.8%、62.9%和83.1%,在验证集中分别为67.2%、80.3%、63.4%和82.8%。结论该研究所建立的列线图模型具有良好的区分度和准确度,可为直观、个体化地分析结直肠息肉发生风险,甄别高风险人群,制订筛查方案提供参考依据。

Objective To establish a nomogram model to predict the risks of colorectal polyps by analyzing the risk factors of colorectal polyps and to evaluate its effectiveness.Methods Clinical data of patients who underwent colonoscopy in the hospital from October 2018 to April 2020 were collected.The independent risk factors of colorectal polyps were analyzed by univariate and multivariate logistic regression.R software was used to establish a nomogram model for predicting the risk of colorectal polyps.Bootstrap method was used for internal validation of the nomogram,and internal validation set was used for external validation.C-index,Calibration curve and receiver operating characteristic(ROC)curve were used to evaluate the prediction performance of the nomogram.Results Age,family tumor history,smoking,fatty liver,gallbladder disease,pigmented nevus and hyperlipidemia were independent risk factors for the occurrence of colorectal polyps.The C-index of the nomogram in the training sets and validation sets was 0.803(95%CI:0.761-0.845)and 0.787(95%CI:0.716-0.858),respectively,Calibration curves fit well.Based on the Yoden index,the optimal critical value of the nomogram was 180.7.The sensitivity,specificity,positive predictive value and negative predictive value was 70.4%,77.8%,62.9%and 83.1%,respectively in the training set,and 67.2%,80.3%,63.4%and 82.8%,respectively in the validation sets.Conclusion The nomogram model established has a good degree of differentiation and accuracy,which is of certain reference value for medical workers to intuitively and individually analyze the risks of colorectal polyps,to screen high-risk groups,and to formulate screening programs.