AAV患者感染的预测列线图模型构建  

Construction of predictive nomogram model for infection in AAV patients

作　　者：周欣 李辛 冯诚怿 范紫芳 狄佳[1,2] ZHOU Xin;LI Xin;FENG Chengyi;FAN Zifang;DI Jia(Nephrology Department,the First People's Hospital of Changzhou,Changzhou 213003,China;Infection Management Department,the First People's Hospital of Changzhou,Changzhou 213003,China)

机构地区：[1]常州市第一人民医院肾内科,江苏常州213003 [2]常州市第一人民医院感染管理科,江苏常州213003

出　　处：《临床医学研究与实践》2025年第5期5-8,共4页Clinical Research and Practice

基　　金：江苏现代医院管理研究中心2019年度课题(No.JSY-3-2019-098)。

摘　　要：目的构建抗中性粒细胞胞浆抗体相关性血管炎(AAV)患者感染的预测列线图模型。方法回顾性分析2010年1月至2023年6月于常州市第一人民医院就诊的193例AAV住院患者的临床资料,分别使用单因素和LASSO回归分析患者发生感染的相关因素,建立列线图模型,绘制校准曲线、受试者工作特征(ROC)曲线进行评价。结果193例AAV患者中,78例发生感染,感染率为40.41%,肺部感染为AAV患者常见感染类型;年龄≥60岁、C反应蛋白(CRP)、中性粒细胞/淋巴细胞比值(NLR)、血小板/淋巴细胞比值(PLR)和伯明翰系统性血管炎活动评分(BVAS)对患者感染有预测价值,基于以上5项变量建立列线图模型,训练集曲线下面积(AUC)为0.874,测试集AUC为0.928,校准曲线与理想曲线拟合较好,内部验证结果未见过拟合或欠拟合现象。结论年龄、BVAS、CRP、NLR、PLR可作为AAV患者发生感染的预测因子;列线图模型预测能力和区分度较好,有助于临床早期筛查感染患者并采取相应对策。Objective To construct a predictive nomogram model for infection in patients with antineutrophil cytoplasmic antibody-associated vasculitis(AAV).Methods The clinical data of 193 patients with AAV who were admitted in the First People's Hospital of Changzhou from January 2010 to June 2023 were retrospectively analyzed.Single factor and LASSO regression were used to analyze the related factors of infection in patients.The nomogram model was established,and the calibration curve and receiver operating characteristic(ROC)curve were drawn for evaluation.Results Among 193 AAV patients,78 were infected,with an infection rate of 40.41%,pulmonary infection was a common type of infection in AAV patients;age≥60 years,C-reactive protein(CRP),neutrophil to lymphocyte ratio(NLR),platelet to lymphocyte ratio(PLR)and Birmingham Vasculitis Activity Score(BVAS)had predictive value for patient infection,a nomogram model was established based on the above five variables,the areas under curve(AUC)of the training set was 0.874,and the AUC of the test set was 0.928,the calibration curve fitted well with the ideal curve,and there was no overfitting or underfitting in the internal validation results.Conclusion Age,BVAS,CRP,NLR and PLR can be used as predictors of infection in AAV patients;the nomogram model has good predictive ability and discrimination,which is helpful for early clinical screening of infected patients and taking corresponding countermeasures.

关 键 词：抗中性粒细胞胞浆抗体相关性血管炎 感染 列线图模型 

分 类 号：R543[医药卫生—心血管疾病]