系统性红斑狼疮并发狼疮肾炎列线图预测模型的建立

Nomogram model for prediction of systemic lupus erythematosus combined with lupus nephritis

目的:利用实验室生物标志物建立一个简便的模型,以预测系统性红斑狼疮(systemic lupus erythematosus,SLE)患者并发肾脏损害的风险。方法:纳入2021年1月—2023年8月期间在南通大学附属医院诊断为SLE的患者210例进行病例对照研究,根据有无肾脏损害分为狼疮肾炎(lupus nephritis,LN)组(LN组)和非狼疮肾炎组(非LN组)。研究通过单因素和多因素Logistic回归分析构建列线图模型。暴露变量包括中性粒细胞与淋巴细胞比值(neutrophil to lymphocyte ratio,NLR)、D-二聚体(D-dimer,D-D)和24 h尿蛋白(24-hour urine total protein,UTP)。结局变量为LN的发生。模型的预测性能通过ROC曲线和校准图进行评估,临床决策曲线(decision curve analysis,DCA)用于评估模型的临床价值。结果:多因素Logistic回归分析显示,NLR、D-D和UTP是区分LN患者的特征参数。预测公式为:Logit(P)=-3.546+0.997×24UTP+0.481×NLR+0.578×D-D。AUC为0.953,灵敏度为90.1%,特异度为89.9%。Hosmer-Lemeshow拟合优度检验结果为P=0.518,Brier得分为0.085。Bootstrap内部验证后的校正C指数为0.858,校准曲线显示出良好的一致性。结论:构建的列线图模型能够有效预测SLE患者并发肾脏损害的风险,为临床早期干预提供重要参考。

Objective:To establish a simple model using existing laboratory biomarkers to predict the risk of renal damage in patients with systemic lupus erythematosus(SLE).Methods:This case-control study was conducted at the Affiliated Hospital of Nantong University.A total of 210 SLE patients diagnosed between January 2021 and August 2023 were included.The patients were divided into two groups based on the presence or absence of renal damage:the lupus nephritis group(LN group)and the non-lupus nephritis group(non-LN group).A nomogram was constructed using univariate and multivariate Logistic regression analyses.Exposure variables included neutrophil to lymphocyte ratio(NLR),D-dimer(D-D),and 24-hour urine total protein(UTP).The outcome variable was the occurrence of LN.The model's predictive performance was assessed using ROC curves and calibration plots,while decision curve analysis(DCA)was used to evaluate clinical utility.Results:Multivariate Logistic regression identified NLR,D-D,and UTP as key parameters for distinguishing LN patients.The prediction formula was:Logit(P)=-3.546+0.997×24UTP+0.481×NLR+0.578×D-D.The AUC was 0.953,with a sensitivity of 90.1%and specificity of 89.9%.The Hosmer-Lemeshow test result was P=0.518,and the Brier score was 0.085.The corrected C index after Bootstrap validation was 0.858,and the calibration curve showed good consistency.Conclusion:The nomogram model constructed can effectively predict the risk of renal damage in SLE patients,providing an important reference for early clinical intervention.