【摘要】 目的:运用临床指标建立狼疮肾炎(LN)肾小球新月体形成的列线图诊断模型。方法:连续收集2010年1月—2023年10月在深圳市第二人民医院肾内科行肾穿刺活检诊断为LN患者。根据纳排标准筛选患者，收集人口学资料及临床指标。采用Lasso回归和Logistic回归进行变量筛选和列线图构建。受试者工作特征(ROC)曲线、校准曲线和临床决策曲线评估模型区分度、校准度和临床实用性。Bootstrap法进行内部验证。结果:符合入组标准患者共144例，筛选出5个预测变量，血红蛋白、肾小球滤过率(EGFR)、三酰甘油、凝血酶原时间(PT)、抗dsDNA抗体。基于上述指标构建列线图诊断模型。模型ROC曲线下面积AUC=0.869(95%CI 0.808,0.931),区分度较好，校准曲线拟合度良好，临床决策曲线和临床影响曲线反映模型临床效用较高。Bootstrap法重复采样500次，得出平均AUC为0.86(95%CI 0.798～0.926),与原模型AUC基本一致。结论:基于临床指标，通过Lasso回归方法建立LN新月体形成的联合诊断模型，具有良好的诊断价值和临床效用性。更多还原

【Abstract】 Objective:To establish a nomogram diagnostic model for crescent formation in lupus nephritis(LN) based on clinical indicators. Methods:Patients diagnosed with LN through renal biopsy in the Nephrology Department of Shenzhen Second People’s Hospital between January 2010 and October 2023 were consecutively enrolled. Patients were screened according to inclusion and exclusion criteria, and demographic data and clinical indicators were collected. Variable selection and nomogram construction were performed using Lasso regression and Logistic regression. The model’s discrimination, calibration, and clinical utility were evaluated using the receiver operating characteristic(ROC) curve, calibration curve, and clinical decision curve. Internal validation was conducted using the Bootstrap method. Results:A total of 144 patients met the inclusion criteria. Five predictive variables were selected: hemoglobin, estimated glomerular filtration rate(eGFR), triglycerides, prothrombin time(PT), and anti-dsDNA antibodies. Based on these indicators, a nomogram diagnostic model was constructed. The area under the ROC curve(AUC) of the model was 0.869(95% CI 0.808, 0.931), indicating good discrimination. The calibration curve showed good fit, and the clinical decision and impact curves demonstrated high clinical utility. Bootstrap resampling was performed 500 times, yielding an average AUC of 0.86(95% CI 0.798～0.926), consistent with the original model’s AUC. Conclusion:A combined diagnostic model for crescent formation in LN was established using clinical indicators and the Lasso regression method, demonstrating good diagnostic value and clinical utility.更多还原