儿童紫癜性肾炎病理分级的列线图预测模型的建立

Establishment of a nomogram prediction model for pathological grade of children with henoch-schonlein purpura nephritis

目的:探讨儿童紫癜性肾炎临床与病理特征的相关性,建立可预测病理分级的列线图模型。方法:选取2019-12-01~2020-12-30于我院住院并行肾脏活检术的HSPN患儿32例。按照国际小儿肾脏病研究组(ISKDC)制定的病理分级标准描述肾脏病理分级及摘录24 h尿蛋白定量、尿系列蛋白如尿N-乙酰-B-氨基葡萄糖苷酶(N-acetyl-B-glucosaminosidase,NAG)、尿微量白蛋白(Albumin,Alba)、尿免疫球蛋白(immunoglobulin G,IgG)等结果,分析ISKDC分级与临床特征关系,建立并评价病理分级列线图模型。结果:32例患儿ISKDC病理Ⅱ级14例(43.8%),Ⅲ级18例(56.2%)。ISKDC病理分级与24 h尿蛋白定量、尿NAG、IgG、Alb和胱抑素C水平有关,联合指标AUC为0.790(95%CI:0.628~0.951,P=0.006),模型预测灵敏度为0.722,特异度为0.714。结论:行肾脏活检术的HSPN患儿病理分级以Ⅱ、Ⅲ级常见。根据临床指标预测病理分级结果,为及时合理治疗及行肾穿检查提供证据。

Objective:To investigate the association between clinical and pathological characteristics of children with henoch-schonlein purpura nephritis(HSPN),and to establish a nomogram model for predicting HSPN pathological grade.Methods:Thirty-two children with HSPN who hospitalized in our hospital and underwent the renal biopsy from December 2019 to December 2020 were included.To describe the ISKDC pathological grade,24 h urine protein,random urine routine test,liver and kidney function test results.To analyze the relationship between ISKDC pathological grade and clinical variables,and to establish and evaluate the nomogram prediction model for the pathological grade.Results:Fourteen HSPN patients wereISKDC gradeⅡ(43.8%),and 18 patients were gradeⅢ(56.2%).The pathological grade of ISKDC was correlated with 24h urine protein,urinary concentrations of NAG,IgG,Alb and cystatin C.AUC of their combined indicator was 0.790(95%CI:0.628-0.951),the sensitivity and specificity were 0.722 and 0.714,respectively.Conclusion:The pathological grades of children with HSPN undergoing renal biopsy are almost at ISKDC gradeⅡandⅢ.According to the results of clinical indicators to predict pathological grade,it could provide evidence for reasonable treatment,timely renal biopsy.