Establishment of a Predictive Model for the Response to intravenous immunoglobulin in Children with Kawasaki Disease

Background:Kawasaki disease (KD) is an acute and self-limited vasculitis that predominantly affects children.Intravenous immunoglobulin (IVIG) remains the standard treatment.However, a subset of patients shows resistance to IVIG, which increases the risk of coronary artery complications.This study aims to identify reliable clinical biomarkers for predicting IVIG response in pediatric patients with KD. Methods:This retrospective study included 136 pediatric patients diagnosed with KD who received IVIG treatment between January 2021 and December 2023.Patients were categorized into IVIG responders or non-responders.Baseline clinical characteristics and laboratory parameters were compared between groups.Multivariate logistic regression analysis was used to identify independent predictors of IVIG response (responders = 1, non-responders = 0).Receiver operating characteristic (ROC) analysis was applied to evaluate the predictive performance of significant variables. Results:IVIG responders exhibited significantly higher hemoglobin and sodium levels, and significantly lower neutrophil percentage, platelet count, aspartate transaminase (AST), total bilirubin (TBIL), and C-reactive protein (CRP) levels (P < 0.05).Serum sodium concentration, platelet count, AST, and TBIL were independent predictors of IVIG response.Higher sodium levels were associated with increased IVIG response.Elevated platelet count, AST, and TBIL levels were associated with reduced response probability.AST and sodium had the highest AUC values and specificity, indicating strong diagnostic value.Platelet count provided supportive predictive value. Conclusions:J o u r n a l P r e -p r o o f Higher serum sodium levels were associated with an increased likelihood of IVIG response, whereas elevated platelet count, AST, and TBIL were associated with IVIG non-responsiveness in pediatric KD.