AIDS合并活动性肺结核评分模型的建立与临床诊断价值评估  

Establishment and evaluation of a clinical diagnostic score model for AIDS complicated with active pulmonary tuberculosis

作　　者：莫胜林 黄云 黄小红[1] 覃锦玉 陈涛[1] 胡家光[1] 蒙达礼[1] 张鹏[1] 蒋忠胜[1] 孔晋亮[3] Mo Shenglin;Huang Yun;Huang Xiaohong;Qin Jinyu;Chen Tao;Hu Jiaguang;Meng Dali;Zhang Peng;Jiang Zhongsheng;Kong Jinliang(Department of Infectious Diseases,Liuzhou People's Hospital,Guangxi Medical University,Guangxi Liuzhou 545006,China;Affiliated Health School,Guangxi University of Science and Technology,Guangxi Liuzhou 545006,China;Department of Respiratory and Critical Care Medicine,The First Affiliated Hospital of Guangxi Medical University,Guangxi Nanning 530021,China)

机构地区：[1]广西医科大学附属柳州市人民医院感染性疾病科,广西柳州545006 [2]广西科技大学附属卫生学校,广西柳州545006 [3]广西医科大学一附院呼吸与危重症医学科,广西南宁530021

出　　处：《新发传染病电子杂志》2023年第6期22-26,共5页Electronic Journal of Emerging Infectious Diseases

基　　金：2021年国家临床重点专科建设项目(感染性疾病科);广西医药卫生科研课题(Z-B20221309,Z-B20221305,Z-B20221375);柳州市科技计划项目(2019BJ20601);柳州市科技计划项目(2021CBC0107)。

摘　　要：目的建立AIDS合并活动性肺结核的评分模型,并对AIDS合并活动性肺结核的临床诊断价值进行评估。方法选取2014年10月1日至2021年9月30日在柳州市人民医院感染性疾病科住院的229例患者,其中AIDS合并活动性肺结核组104例,AIDS合并肺部非结核疾病组125例,对患者临床数据进行分析,以AIDS合并活动性肺结核组为结局指标进行单因素分析,单因素分析有意义的指标(P<0.05)作为自变量,进行多因素Logistic回归分析。根据多因素回归模型中的β回归系数进行赋值,建立临床诊断AIDS合并活动性肺结核的评分模型,并评估其对AIDS合并活动性肺结核的临床诊断预测能力。结果本临床诊断评分模型的6个变量分别为咳嗽、咳痰症状(7分),CD4^(+)T淋巴细胞数量(12分),肺上叶病灶(6分),多肺段病灶(7分),肺部病灶并纵隔淋巴结肿大(7分),单侧胸腔积液(6分),总分45分。当分值大于24分时,临床诊断AIDS合并活动性肺结核时的曲线下面积为0.911(95%CI 0.864~0.958)(P<0.05)。敏感度和特异度分别为82.2%、92.0%。阳性预测值和阴性预测值分别为:89.6%、86.9%。验模组验证所得结果正确率达88.2%(60/68)。结论本研究建立的AIDS合并活动性肺结核的临床诊断评分模型在临床上使用操作简单易行,当CD4^(+)T淋巴细胞数量≤200个/μl时,对AIDS合并活动性肺结核的临床诊断具有良好的预测能力。Objective To establish a scoring model for the clinical diagnosis of active pulmonary tuberculosis in patients with AIDS using common clinical indicators and to evaluate the predictive performance of the model.Method A total of 229 patients admitted to the Department of Infectious Diseases,Liuzhou People's Hospital,from October 1,2014,to September 30,2021,were included in the study.Among them,there were 104 cases of AIDS combined with active pulmonary tuberculosis and 125 cases of AIDS combined with other non-tuberculous pulmonary diseases.Univariate analysis and binary logistic regression analysis were used to select variables favorable for diagnosing active pulmonary tuberculosis in AIDS patients.A clinical scoring model for the diagnosis of active pulmonary tuberculosis in AIDS was established,and the predictive ability of the model was assessed.Result In this clinical diagnostic scoring model,six variables were analyzed:cough and sputum symptoms(7 points),CD4^(+)T lymphocyte count(12 points),upper lobe lung lesions(6 points),multi-segment lung lesions(7 points),lung lesions with mediastinal lymphadenopathy(7 points),and unilateral pleural effusion(6 points).When the score was greater than 24 points,the area under the curve for clinical diagnosis of active pulmonary tuberculosis in AIDS patients was 0.911(95%CI 0.864-0.958)(P<0.05).The sensitivity and specificity were 82.2%and 92.0%,respectively.The positive predictive value and negative predictive value were 89.6%and 86.9%,respectively.The results obtained from the validation group achieved an accuracy of 88.2%(60/68).Conclusion The clinical diagnostic scoring model for active pulmonary tuberculosis in AIDS patients established in this study is simple to use in clinical practice and has good predictive and pre-judgment efficacy for evaluating whether HIV-infected individuals with a CD4^(+)T lymphocyte count≤200/μl are combined with active pulmonary tuberculosis.

关 键 词：AIDS 活动性肺结核 机会性感染 临床诊断评分 CD4^(+)T淋巴细胞 

分 类 号：R521[医药卫生—内科学]