Longitudinal changes in renal function in rheumatoid arthritis patients treated with biologic agents: a retrospective cohort study.

OBJECTIVES: To evaluate longitudinal renal trajectories and predictors of kidney function in patients with rheumatoid arthritis (RA) receiving biologic therapies, using serial estimated glomerular filtration rate (eGFR) measurements during follow-up. As a secondary objective, renal outcomes were compared between anti-tumor necrosis factor (anti-TNF) agents and other biologic therapies. METHOD: This single-center retrospective cohort study included 141 patients with RA fulfilling the 2010 ACR/EULAR classification criteria and receiving the same biologic therapy for at least one year. Renal parameters, including eGFR, creatinine, and urea, were assessed at baseline, every 6 months up to 24 months, and at final follow-up. Chronic kidney disease (CKD) was defined according to the 2024 KDIGO criteria. Propensity score matching (PSM) was performed to balance baseline characteristics between anti-TNF and non-TNF groups. Longitudinal renal changes were evaluated using linear mixed-effects models. RESULTS: Baseline CKD prevalence was 5.0%, and most patients had preserved renal function. During follow-up, significant longitudinal changes were observed in creatinine (p < 0.001) and eGFR (p = 0.002) in unadjusted analyses, whereas urea remained stable. However, in the adjusted linear mixed-effects model, the overall time effect on eGFR was not statistically significant (F = 1.080, p = 0.370). Type III tests demonstrated significant associations of treatment group (p = 0.032), RF group (p = 0.024), ILD (p = 0.008), disease duration (p = 0.028), and age (p < 0.001) with longitudinal eGFR variability. Patients receiving Anti-TNF therapy demonstrated higher eGFR values during follow-up (101.49 vs. 98.68 mL/min/1.73m2). In estimates of fixed effects, age remained the strongest independent predictor of eGFR (β =  - 1.024, 95% CI: - 1.208 to - 0.840; p < 0.001). Propensity score-matched analyses demonstrated a significant time-by-treatment interaction (β =  - 0.574, p = 0.037), suggesting differential renal trajectories according to treatment exposure. CONCLUSION: Differences in longitudinal renal function were observed between treatment groups, but these findings should be interpreted cautiously because of residual confounding and baseline imbalances. Prospective studies are warranted to confirm these observations. Key Points • Renal function remained largely preserved in patients with rheumatoid arthritis receiving biologic therapies in real-world clinical practice. • Patients receiving anti-TNF therapy demonstrated higher eGFR values and distinct longitudinal renal trajectories compared with those receiving non-TNF biologics. • Age emerged as the strongest independent predictor of renal function, supporting the importance of long-term renal monitoring in RA patients receiving biologic therapy.