广西艾滋病患者免疫重建不良的预测模型建立

Analysis on risk factors for immune reconstitution failure of AIDS and establishmen of prognostic prediction model in Guangxi province

目的探讨广西地区AIDS患者ART后免疫重建的影响因素,建立免疫重建不良的预测模型。方法收集2007年5月至2019年12月就诊于广西医科大学附属武鸣医院及南宁市马山县人民医院艾滋病门诊随访的AIDS患者336例。进行多因素Logistic回归分析建立免疫重建不良预测模型,并建立ROC曲线评价其预测价值。结果共纳入336例研究对象,平均年龄为(47.3±12.9)岁,其中男性有219例,女性有117例。免疫重建不良的发生率为36.3%(122例)。多因素Logistics回归分析显示:男性、年龄≥60岁、合并乙肝或丙肝、持续治疗时间<24个月、基线CD4细胞<100个/μL是免疫重建不良的独立危险因素(P<0.05);ROC曲线下面积为0.784,灵敏度为77.9%,特异度为69.2%,最佳诊断临界值为0.646,诊断价值良好。结论 ART后免疫重建不良可能与多种因素相关,应对HIV感染者尽早治疗干预,以预防免疫重建不良。

Objective To explore the influencing factors of immune reconstitution in AIDS patients after ART treatment in Guangxi, and establish a prediction model of poor immune reconstitution. Methods The information of AIDS patients who were followed up in the Wuming Hospital Affiliated with Guangxi Medical University and the people’s Hospital of Mashan County of Nanning City from May 2007 to December 2019 were collected. Multivariate logistic regression analysis was used to establish the prediction model of poor immune reconstitution, and the ROC curve was established to evaluate its prediction value. Results A total of 336 subjects were included, with an average age of(47.3±12.9) years, including 219 males and 117 females. The incidence of poor immune reconstitution was 36.3%(122 cases).Multivariate Logistics regression analysis showed that males, aged over 60 years, hepatitis B or hepatitis C, continuous treatment time <24 months, baseline CD4 cells < 100. μ L was an independent risk factor for poor immune reconstitution(P<0.05). The area under the ROC curve was 0.784. The sensitivity was 77.9%. The specificity was 69.2%, and the optimal diagnostic threshold was 0.646. Conclusion Poor immune reconstitution after ART may be related to many factors. Early treatment and intervention should be carried out in HIV-infected patients to prevent poor immune reconstitution.