Association Between Relapse and Long-Term Kidney Outcomes in IgA Nephropathy.

Rationale & Objective: Relapse in IgA nephropathy (IgAN) remains under-researched. This study investigates the incidence of IgAN relapse, its association with adverse outcomes, and its risk factors within a prospective cohort. Study Design: Prospective cohort study. Setting & Participants: The study included 1,069 patients with IgAN who achieved remission following treatment. Predictor: Relapse or not after remission of IgAN. Outcomes: The first occurrence of either end-stage kidney disease (estimated glomerular filtration rate [eGFR] <15 mL/min/1.73 m2, dialysis, or kidney transplantation) or a 50% decline in eGFR. Analytical Approach: We assessed relapse incidence, its correlation with kidney endpoint events, and identified clinicopathological predictors of relapse using Cox regression analysis. Least Absolute Shrinkage and Selection Operator Cox regression was performed to screen predictive variables to predict relapse. Results: Among the 1,069 patients, 348 (33%) experienced 411 relapses. The overall relapse-free rates at 1, 5, and 10 years were 93%, 65%, and 48%, respectively. Relapse patients had a significantly higher incidence kidney endpoint events compared with nonrelapse patients. The relapse group also exhibited a more rapid decline in eGFR compared with the nonrelapse group (-2.3 vs -1.0 mL/min/1.73 m2/year). Patients with corticosteroid-induced remission had a higher likelihood of relapse within 3 years. Controlling 24-hour urinary protein at remission to below 0.3 g/d was associated with a significantly improved relapse-free rate. Limitations: Single-center study; no standardized definition for IgAN remission and relapse. Conclusions: Relapse is prevalent in IgAN and is associated with an increased rate of kidney endpoint events and accelerated eGFR decline. Predictors of relapse encompass various clinicopathological indicators and medications, with stricter control of proteinuria at remission being crucial for reducing relapse rates.