【摘要】 目的 探讨人类免疫缺陷病毒/获得性免疫缺陷综合征(HIV/AIDS)患者心电图异常情况及其影响因素。方法 选取2021年7—12月收治的HIV/AIDS 546例，均行心电图检查，统计心电图异常情况，采用多因素Logistic回归分析探讨HIV/AIDS患者心电图异常影响因素，并建立预测模型，绘制受试者工作特征(ROC)曲线分析模型预测价值。结果 HIV/AIDS患者心电图异常率为67.03%。多因素Logistic回归分析结果显示，有吸烟史、肺部感染、CD4+<200μL、CD8+<500μL、高效抗反转录病毒治疗(HAART)、HAART时间≥2年是HIV/AIDS患者心电图异常的独立危险因素(P<0.01)。HIV/AIDS患者心电图异常影响因素多因素Logistic回归模型评价显示模型建立具有统计学意义(P<0.01),回归方程系数差异有统计学意义(P<0.01),模型拟合效果较好(P>0.05)。多因素Logistic回归模型对HIV/AIDS患者心电图异常预测价值的ROC曲线分析结果显示，曲线下面积为0.851,95%CI(0.817,0.880)。随机抽取1例进行个体预测，结果显示在预测准确率为77.29%的条件下，该患者发生心电图异常。结论 HIV/AIDS患者心电图异常受诸多因素影响，如吸烟史、肺部感染、CD4+<200μL、CD8+<500μL等，根据影响因素构建多因素Logistic回归模型对HIV/AIDS患者心电图异常具有可靠预测价值。更多还原

【Abstract】 Objective To investigate the abnormal electrocardiogram and its influencing factors in human immunodeficiency virus/acquired immunodeficiency syndrome(HIV/AIDS) patients. Methods A total of 546 patients with HIV/AIDS admitted from July to February 2021 were selected, all of whom underwent electrocardiogram examination. The abnormal electrocardiogram was statistically analyzed, and the influencing factors of abnormal electrocardiogram in HIV/AIDS patients were discussed using multivariate Logistic regression analysis, and the prediction model was established. Receiver operating characteristic(ROC) curve was plotted to analyze the predictive value of the model. Results The abnormal rate of electrocardiogram in HIV/AIDS patients was 67.03%. Multivariate Logistic regression analysis showed that smoking history, lung infection, CD4+ < 200 μL, CD8+ < 500 μL, high-potency antiretroviral therapy(HAART), duration of HAART ≥2 years were independent risk factors for ECG abnormalities in HIV/AIDS patients(P<0.01). The multivariate Logistic regression model evaluation of the influencing factors of ECG abnormality in HIV/AIDS patients showed that the model establishment had statistical significance(P<0.01), that the coefficient difference of regression equation had statistical significance(P<0.01), and that the model fitting effect was good(P>0.05). The ROC curve analysis of the predictive value of ECG abnormalities in HIV/AIDS patients by multivariate Logistic regression model showed that the area under the ROC curve(AUC) was 0.851,95%CI(0.817, 0.880). One case was randomly selected for individual prediction, and the results showed that under the condition of prediction accuracy of 77.29%, the patient developed abnormal electrocardiogram. Conclusion ECG abnormalities in HIV/AIDS patients are affected by various factors, such as smoking history, lung infection, CD4+ < 200 μL, and CD8+ < 500 μL. The construction of multivariate Logistic regression model according to the influencing factors has reliable value in predicting ECG abnormalities in HIV/AIDS patients.更多还原