Abstract: Objective    To investigate the risk factors for renal damage in children with Henoch-Schönlein purpura（HSP），and to develop a nomogram for predicting decreased glomerular filtration rate（GFR）in children with Henoch-Schonlein purpura nephritis（HSPN）. Methods    116 children with HSPN confirmed by renal biopsy from July 2021 to June 2024 were enrolled.The demographic characteristics，clinical and renal pathological characteristics of children with HSPN were utilized to develop a nomogram model for predicting the progression of HSPN. Ten-fold cross-validation was used to select the optimal harmonic coefficient λ in the model.The discrimination and accuracy of the prediction model were evaluated based on the receiver operating characteristic curve（ROC）and calibration curve，respectively，and a nomogram was drawn. Results    The results showed that serum creatinine and renal pathology were independent risk factors affecting the progression of HSPN （P<0.05）. Two variables affecting the prognosis of children with HSPN were selected through the LASSO regression method，and a nomogram was built based on these variables. The predictive model yielded an AUC of 0.957（95%CI 0.925-0.990），The calibration curve showed that the risk factors for HSPN progression predicted by the nomogram were in good agreement with the actual results.The decision curve analysis showed that the nomogram model could obtain clinical benefits within a large threshold range（0-90%）. It has good practicability. Conclusion    In this study，Based on the two independent risk factors of serum creatinine and renal pathology，a nomogram model was established for the first time in predicting the risk of decreased GFR in children with HSPN，which has great value in predicting the decreased renal function in children with HSPN.

Key words: child, Henoch-Sch?nlein purpura nephritis, decreased glomerular filtration rate, nomogram

摘要： 目的    探讨过敏性紫癜（HSP）患儿发生肾脏损害的相关危险因素，初步构建紫癜性肾炎（HSPN）患儿肾小球滤过率（GFR）下降的风险预测列线图。方法    选取2021年7月1日至2024年6月30日徐州医科大学附属徐州儿童医院收治的HSPN行肾穿刺的患儿116例，对HSPN患儿的人口学特征和临床、肾脏病理特征进行分析，开发预测HSPN进展的nomogram模型。构建LASSO Logistic回归模型筛选自变量，通过十倍交叉验证法选择模型中的最优调和系数λ，基于ROC曲线和校准曲线分别评价预测模型的区分度和准确度，并绘制列线图。结果    通过LASSO Logistic回归模型筛选影响HSPN患儿预后变量，结果显示血肌酐、肾脏病理是影响HSPN进展的独立危险因素（P<0.05），并基于变量构建得到列线图。构建的列线图模型的曲线下面积为0.957（95% CI，0.925~0.990），校准曲线显示列线图预测的HSPN进展的危险因素与实际结果具有良好的一致性，决策曲线分析显示列线图模型在较大的阈值范围内（0～90%）可获得临床收益，具有较好的实用性。结论    基于血肌酐、肾脏病理这2项独立危险因素，首次建立了预测 HSPN患儿GFR下降发生风险的列线图模型，该图对HSPN患儿肾功能下降预测具有重要价值。

关键词: 儿童, 紫癜性肾炎, 肾小球滤过率下降, 列线图