OP0152 A longitudinal study of the effects of disease activity on renal function in patients with rheumatoid arthritis utilizing linear mixed effect models - answer cohort study -

Background Patients with RA have higher risk of developing chronic kidney disease compared to the general population. Although some causes of renal impairment in RA include nephrotoxic medications, amyloidosis, and glomerulonephritis, specific causes are not determined among most patients with RA. While experimental researches showed inflammatory process per se might also contribute to renal dysfunction, the results of previous clinical studies that assessed the effects of disease activity or C-reactive protein on renal function were inconsistent. This inconsistency might be attributable to their cross-sectional design, small sample size, and assessment of association between only baseline characteristics and renal outcomes without consideration for the fact that disease activity and medications had changed over time. Objectives To identify the effects of disease activity on renal function in RA in a multi-center cohort study. Methods RA patients with a sampling interval of less than 150 days were enrolled because wide sampling intervals could not take into consideration changes in disease activity and medications during their follow-up. An estimated glomerular filtration rate (eGFR) was calculated using an equation approved by the Japanese Society of Nephrology and used as an outcome variable. Linear mixed effect models were used to evaluate the renal trajectories of patients. Time from baseline (months), disease activity, and their interaction were included as fixed effects while participant identification number and time from baseline were included as random factors. Age, sex, disease duration, RF, ACPA, NSAIDs, and DMARDs that were known as a cause of renal impairment, such as tacrolimus, iguratimod, and tofacitinib, were included as covariates. Results A total of 25661 samples (mean sampling interval: 2.0 months) from 2104 patients was included. Patients with lower DAS28-CRP had worse renal function at inclusion, but a significantly better longitudinal trajectory on eGFR (0.0079 ml/min/1.73m2 per month, P=0.025). Although all RA patients had naturally progressive renal impairment as they got older, patients who achieved remission or low disease activity had slower renal impairment rate of -0.068 ml/min/1.73m2 per month compared to patients with moderate or high disease activity (-0.084 ml/min/1.73m2 per month; P=0.037). These results were also similar using SDAI or CDAI. Conclusions Lower disease activity results in slower renal impairment. Because the effects of disease activity on renal function are mild, additional measures to protect renal function, such as avoiding nephrotoxic medications and treating cardiovascular risk factors are important. Disclosure of Interest None declared