[关键词]
[摘要]
【目的】基于邱健行教授“毒垢”理论,探讨内镜下结肠息肉切除术后的复发危险因素,建立预测模型并进行复发风 险预测。 【方法】回顾性收集2018年1月~2023年6月在广东省第二中医院脾胃科行结肠息肉切除术的383例患者,所有患者 均在1年以上1年半内再次行肠镜检查,其中254例患者复发,129例患者未再复发;同时,将数据集按8∶2比例随机分为训 练集306例及验证集77例,分析结肠息肉切除后复发相关危险因素,建立Logistic回归预测模型并进行验证。 【结果】(1) 复发 相关因素分析及模型的构建:Logistic回归分析结果显示,年龄 (OR=1.03) 、高血压 (OR=2.91) 、体质量指数 (BMI) 水平 (OR= 2.12) 、饮食不节 (OR=5.01) 、烟酒史 (OR=2.24) 、精神类药物使用史 (OR=2.44) 、湿热 (OR=4.83) 、痰饮 (OR=3.24) 、血瘀 (OR=5.17) 以及息肉数量 (OR=1.38) 、直径 (OR=1.13) 、病理性质 (OR=2.27) 为结直肠息肉复发的相关因素 (P<0.05或P< 0.01) 。 (2) 模型的评价:训练集受试者工作特征 (ROC) 曲线下面积 (AUC) 为0.874 (95%CI:0.834~0.914) ,验证集ROC曲线 下面积为0.883 (95%CI:0.806~0.961) ,训练集及验证集校准曲线显示基于上述因素所建立的模型预测准确性可靠。 (3) 模型 的临床运用:可通过绘制列线图进行风险预测,通过绘制临床决策曲线 (DCA) 及临床影响曲线 (CIC) 提供决策依据。 【结论】 年龄、高血压、BMI水平、饮食不节、烟酒史、精神类药物使用史、湿热、痰饮、血瘀以及息肉数量、直径、病理性质为结 直肠息肉复发的相关危险因素,基于上述因素所建立的预测模型具有良好的区分度、校准度,能对患者的结肠息肉复发概 率作出预测。
[Key word]
[Abstract]
Objective To investigate the risk factors for the recurrence after endoscopic resection of colonic polyps from the toxin-grime theory proposed by Professor QIU Jian-Xing,and to construct the prediction model for predicting the risk of recurrence. Methods A retrospective analysis was conducted in 383 patients who underwent colonic polypectomy in the Department of Spleen and Stomach of Guangdong Second Traditional Chinese Medicine Hospital from January 2018 to June 2023. All of the patients underwent colonoscopy again during the period of one year to one year and a half after the operation. Of the 383 patients,254 patients relapsed and 129 patients did not relapse. Moreover,the data set of the 383 patients was randomly divided into a training set of 306 cases and a validation set of 77 cases at the ratio of 8 to 2. The risk factors related to the recurrence after colonic polypectomy were analyzed,and a logistic regression prediction model was established for verification. Results (1) In the recurrence-related factor analysis and model construction,logistic regression analysis results showed that related factors for the recurrence of colorectal polyps involved age (OR = 1.03) ,hypertension (OR = 2.91) ,body mass index (BMI) (OR = 2.12) ,improper diet (OR = 5.01) ,history of tobacco and alcohol (OR = 2.24) ,history of use of psychotropic drugs (OR = 2.44) ,damp-heat (OR = 4.83) ,phlegm-retention (OR = 3.24) ,blood stasis (OR = 5.17),number of polyps (OR = 1.38),diameter of polyps (OR = 1.13),and pathological nature of polyps (OR = 2.27),the difference being significant (P<0.05 or P<0.01) . (2) The evaluation of the model showed that the area under the receiver operating characteristic (ROC) curve of the training set was 0.874 (95% CI :0.834-0.914) ,and the area under the ROC curve of the validation set was 0.883 (95% CI :0.806-0.961) . The calibration curves of the training set and the validation set showed that the prediction accuracy of the model based on the above factors was reliable. (3) The clinical application of the model showed that the risk prediction can be performed by drawing a nomogram,and decision-making basis can be obtained after drawing a clinical decision curve and a clinical impact curve. Conclusion Age,hypertension,BMI,improper diet,history of tobacco and alcohol,history of use of psychotropic drugs,damp-heat,phlegm-retention,blood stasis,number of polyps, diameter of polyps and pathological nature of polyps are risk factors related to the recurrence of colorectal polyps. The prediction model based on the above factors has good discrimination and calibration,and can predict the probability of recurrence of colon polyps.
[中图分类号]
R259.746
[基金项目]
国家中医药管理局第五批全国中医临床优秀人才研修项目 (国中医药人教函【2022】1号);广东省中医特色重点医院建设工程 项目 (粤财社[2023]139号);广州市科技计划项目 (编号:202201011782)