Clinical characteristics and renal risk stratification of mixed-type IgA vasculitis: a large-volume study

Background Mixed-type IgA vasculitis (IgAV) is common in clinical practice, but its internal heterogeneity and prognostic significance, especially in renal involvement, remain unclear. This study aimed to characterize mixed-type IgAV, compare renal risk across mixed subtypes, and evaluate the additional predictive value of mixed subtype classification for clinically significant renal involvement. Methods We performed a retrospective cohort study of patients with clinically diagnosed IgAV treated between December 2018 and October 2025. Among 2,818 admissions/records screened, 2,416 unique patients were included after excluding repeated admissions, including 739 non-mixed and 1,677 mixed-type cases. Mixed type and mixed subtypes were defined according to organ involvement at first admission. Clinically significant renal involvement was defined from discharge diagnoses and renal-related clinical documentation, with available renal laboratory indicators used as supporting evidence. Laboratory-based component and sensitivity analyses were additionally performed using predefined proteinuria, eGFR, and serum creatinine criteria. Subtype-specific logistic regression, receiver operating characteristic (ROC) analysis, and a visit-order KM-like proteinuria analysis were performed. Results Compared with non-mixed disease, mixed-type IgAV was associated with lower BMI and albumin, higher white blood cell and platelet counts, longer hospital stay, and markedly higher frequencies of digestive, joint, and renal involvement. Within the mixed-type cohort, clinically significant renal involvement varied substantially across subgroups, ranging from 26.7% to 76.9%. Using Skin–Joint as the reference group, significantly higher odds of clinically significant renal involvement were observed in the Other/rare group, Skin–Renal, Skin–Joint–Renal, and Skin–Abdominal–Renal (all p < 0.001). In complete-case analysis (n=2,286), adding mixed subtype to a model based on routine demographic and laboratory variables improved the AUC from 0.596 to 0.745 (ΔAUC 0.148, DeLong p < 0.001). Mixed-type patients also showed a lower proteinuria-free probability across successive admissions. Conclusions Mixed-type IgAV is a heterogeneous phenotype with marked subtype-dependent differences in renal risk. Mixed subtype classification improves identification of clinically significant renal involvement and may support earlier and more individualized renal risk stratification.