Objective To investigate changes of laboratory indicators and pathological features of Henoch-Schonlein purpura nephritis(HSPN)children after crescent formation,and to explore non-invasive biomarkers for predicting crescent formation.Methods A total of 278 children with HSPN who were hospitalized from January 2018 to July 2023 were selected and divided into the crescent formation group(196 cases)and the non crescent formation group(82 cases)based on their crescent formation status.Patients in the crescent formation group were sub-divided into the cellular crescent formation group(52 cases)and the cellular fibrous crescent formation group(144 cases)based on the type of crescent formation.Laboratory indicators and pathological characteristics were compared between different groups.The correlation between each indicator and the proportion of crescent formation was analyzed.The influencing factors of crescent formation were analyzed by Logistic regression.Receiver operating characteristic(ROC)curves were plotted,and the effectiveness of laboratory indicators in predicting crescent formation was evaluated.Results Compared with the non crescent formation group,24-hour urine protein quantification(24 hUP),urine immunoglobulin G/creatinine(UGCR),urine red blood cell count(URBC),neutrophil/lymphocyte ratio(NLR),blood urea nitrogen(BUN)and triglycerides(TG)were significantly increased in the crescent formation group.There were increased proportion of diffuse mesangial hyperplasia(Mb),renal tubular atrophy or interstitial fibrosis(T1)(P＜0.05).Compared with the cellular crescent group,the proportion of glomerular segmental sclerosis or adhesion(S1)and T1 were increased,and the proportion of crescent formation was higher in the cellular fibrous crescent group.The proportion of capillary endothelial cell proliferation(E1)was decreased in the cellular crescent group(P<0.05).Spearman correlation analysis showed that 24 hUP,UGCR,URBC,NLR,BUN,TG,Mb and T1 were positively correlated with the proportion of crescent formation(all P＜0.05).The results of multivariate Logistic regression analysis showed that elevated UGCR and T1 were risk factors for crescent formation.The area under the curve(AUC)predicted by UGCR for crescent formation was 0.731(95%CI:0.667-0.795,P＜0.05),with an optimal cutoff value of 5.00 mg/mmol,sensitivity of 0.744 and specificity of 0.610.Conclusion UGCR can be used as a non-invasive biomarker to assist in evaluating crescent formation in children with HSPN.