Objective  To investigate the predictive value of the baseline serum immunoglobulin G (IgG)/immunoglobulin E (IgE) ratio for the risk of renal injury in children with IgA vasculitis (IgAV) and to evaluate its incremental value in the development of a prediction model.

Methods This retrospective cohort study included 198 children with IgAV treated at Beijing Anzhen Nanchong Hospital of Capital Medical University ＆ Nanchong Central Hospital, from January 2022 to August 2024. According to whether renal injury occurred within one year, the patients were divided into a renal injury group and a non-renal injury group. Baseline characteristics, serum IgG and IgE levels, and the IgG/IgE ratio were compared between the two groups. Least absolute shrinkage and selection operator (LASSO) and Logistic regression was used to analyze factors associated with the risk of renal injury in children with IgAV. Receiver operating characteristic (ROC) curve analysis was performed to assess the predictive value of the baseline serum IgG/IgE ratio for renal injury risk in children with IgAV.

Results The incidence of renal injury within one year among children with IgAV was 27.27% (54/198). The proportion of patients with gastrointestinal bleeding, rash duration, white blood cell count (WBC), platelet count (PLT), C-reactive protein (CRP) level, and D-dimer (D-D) level were higher in the renal injury group than in the non-renal injury group (all P<0.05). Baseline serum IgG levels and the IgG/IgE ratio were lower, whereas IgE levels were higher, in the renal injury group than in the non-renal injury group (all P<0.05). LASSO regression identified five core variables with nonzero coefficients: gastrointestinal bleeding, rash duration, CRP, D-D, and the IgG/IgE ratio. Logistic regression model 1 showed that the serum IgG/IgE ratio was a factor influencing the risk of renal injury in children with IgAV (P<0.05). Logistic regression model 2 showed that gastrointestinal bleeding, rash duration, CRP, and D-D were factors influencing the risk of renal injury in children with IgAV (all P<0.05). Logistic regression model 3 showed that gastrointestinal bleeding, rash duration, CRP, D-D, and the IgG/IgE ratio were factors influencing the risk of renal injury in children with IgAV (all P<0.05). ROC curve analysis showed that the areas under the curve (AUCs) of models 1 and 2 for predicting the risk of renal injury in children with IgAV were similar (0.809 vs. 0.833, Z=0.534, P=0.594). The AUC of model 3 was 0.907, which was higher than those of models 1 and 2 (Z=2.305 and 2.018; P=0.021 and 0.044, respectively). Calibration curves showed a high degree of agreement between the predicted risks of renal injury in children with IgAV and the actual clinical outcomes for all three models, indicating good calibration (Hosmer-Lemeshow H-L test: χ²=6.235, 7.891, and 5.124; P=0.620, 0.485, and 0.745, respectively). Internal validation showed that the optimism-corrected AUCs of the three models were 0.792 (95% CI 0.724-0.855), 0.815 (95% CI 0.751-0.878), and 0.886 (95% CI 0.835-0.937), respectively. These values were close to the original AUCs, indicating a low degree of model overfitting and stable discriminative performance.

Conclusions The baseline serum IgG/IgE ratio is a factor influencing the risk of renal injury in children with IgAV and has moderate-to-good independent predictive ability. When combined with gastrointestinal bleeding, rash duration, CRP, and D-D, it significantly improves predictive value and may serve as a candidate indicator for early identification.