经抗病毒治疗的成年HIV感染/AIDS患者并发结核病预测模型的建立与验证  

Establishment and validation of a predictive model for tuberculosis in adult HIV infection/AIDS patients treated with antiviral therapy

作　　者：王秋婷 陈燕 陈赛娟 刘华[1] 张燕[1] WANG Qiuting;CHEN Yan;CHEN Saijuan;LIU Hua;ZHANG Yan(School of Public Health/Research Center for Medicine and Social Development/Collaborative Innovation Center for Social Risk Prediction and Governance in Health Field,Chongqing Medical University,Chongqing 400016,China;Department of Information,Chongqing Public Health Medical Treatment Center,Chongqing 400036,China)

机构地区：[1]重庆医科大学公共卫生学院/医学与社会发展研究中心/健康领域社会风险预测治理协同创新中心,重庆400016 [2]重庆市公共卫生医疗救治中心信息科,重庆400036

出　　处：《检验医学与临床》2024年第11期1520-1527,共8页Laboratory Medicine and Clinic

基　　金：重庆市技术创新与应用发展专项重点资助项目(cstc2019jscx-fxydX0037)。

摘　　要：目的探讨经抗病毒治疗(ART)的成年人类免疫缺陷病毒(HIV)感染/获得性免疫缺陷综合征(AIDS)患者并发结核病(TB)的影响因素,构建列线图模型预测HIV感染/AIDS患者并发TB风险。方法对2017年2月至2021年2月于重庆市公共卫生医疗救治中心接受ART的1486例HIV感染/AIDS患者进行回顾性研究。采用单因素、多因素Logistic回归分析HIV感染/AIDS患者并发TB的独立危险因素并构建列线图预测模型。绘制受试者工作特征曲线、校准曲线和决策曲线进行内、外部验证,评价模型预测效能和临床实用性。结果男性、开始ART年龄为18~<45岁、世界卫生组织临床分期为Ⅲ期或Ⅳ期、基线CD4+T淋巴细胞计数为200~<350个/μL或<200个/μL、基线病毒载量为10000~100000 copy/mL或>100000 copy/mL和贫血是HIV感染/AIDS患者并发TB的独立危险因素(P<0.05)。训练集及测试集列线图预测模型内外部验证的曲线下面积分别为0.727(95%CI:0.696~0.758)、0.780(95%CI:0.722~0.838)。校准曲线拟合效果良好,决策曲线分析结果显示该模型具有良好的临床收益。结论基于上述独立危险因素建立的HIV感染/AIDS患者并发TB风险的列线图预测模型具有较强的预测效能,能够为早期降低TB发病风险提供科学依据。Objective To explore the influencing factors of tuberculosis(TB)in adult human immunodeficiency virus(HIV)infection/acquired immunodeficiency syndrome(AIDS)patients treated with antiviral therapy(ART)and to establish a nomogram model to predict the risk of TB in HIV infection/AIDS patients.Methods A retrospective study was conducted on 1486 HIV infection/AIDS patients who treated with ART in Chongqing Public Health Medical Treatment Center from February 2017 to February 2021.Univariate and multivariate Logistic regression analysis were used to screen the independent risk factors of TB in HIV infection/AIDS patients and establish a nomogram prediction model.The receiver operating characteristic curve,calibration curve and decision curve were drawn for internal and external verification to evaluate the prediction efficiency and clinical practicability of the model.Results Male,18-<45 years old at the initiation of ART,World Health Organization clinical stageⅢorⅣ,baseline CD4+T cell count of 200-<350/μL or<200/μL,baseline viral load of 10000-100000 copy/mL or>100000 copy/mL and anemia were independent risk factors for TB in HIV infection/AIDS patients(P<0.05).The area under the curves of internal and external validation of the nomogram prediction model in the training set and test set were 0.727(95%CI:0.696-0.758)and 0.780(95%CI:0.722-0.838),respectively.The calibration curve fitted well,and the decision curve analysis results showed that the model had good clinical benefits.Conclusion The nomogram prediction model based on the above independent risk factors has a strong predictive efficacy and can provide scientific basis for early reduction of the risk of TB in HIV infection/AIDS patients.

关 键 词：人类免疫缺陷病毒感染 获得性免疫缺陷综合征 结核病 抗病毒治疗 危险因素 列线图模型 

分 类 号：R512.91[医药卫生—内科学] R446.11[医药卫生—临床医学]