E103 Determinants of long-term prednisolone use in rheumatoid arthritis: real-world experience

Background/Aims Glucocorticoids provide rapid disease control when initiating or changing csDMARDs in rheumatoid arthritis (RA). Guidelines advise ‘tapering and discontinuing as rapidly as clinically feasible’. Aims To assess demographic, clinical characteristics, co-morbidities, adverse events and determinants of long-term prednisolone (LTP) use in RA patients in a real-world setting. Methods This cross-sectional study included RA patients receiving routine care at St. George’s Hospital, London, identified through departmental databases. Categorical variables were compared using Chi-square or Fisher’s exact tests, and continuous variables using t-tests or Mann-Whitney U tests. Univariable and multivariable logistic regression identified factors associated with LTP use. Variables with p < 0.10 entered the multivariable model. Analyses were conducted using SPSS. Results 557 RA patients were included; 82% female, mean 61.7 years, RF positive 76%, anti-CCP positive 66%, disease duration mean 13.2 years. 79 (14.2%) patients were prescribed LTP, mean dose 6 mg (range 1 - 20 mg), 28 (5%) >5mg daily. Patients on LTP were older (mean 66.7 vs 60.9 years, p = 0.02), had longer disease duration (mean 18.1 vs 12.4 years, p < 0.001), less methotrexate use (36.7% vs 51%, p = 0.018) and higher DAS28 scores (mean 3.5 vs 3.0, p = 0.004). More patients on LTP had received ≥3 bDMARDs (11.4% vs 3.3%, p = 0.007). Of 50/79 patients on LTP and no methotrexate, 73.5% had previously failed methotrexate. There was no difference in gender, smoking status, serology, co-prescription of ≥ 1 csDMARD or any bDMARD. Regression analysis identified longer disease duration, less methotrexate use, and higher DAS28 as independent determinants of LTP (Table). Co-morbidities significantly more prevalent in LTP users were cerebrovascular accident (7.6% vs 2.2% p = 0.023), venous thromboembolism (11.4% vs 2.5% p = 0.001), osteoporosis (40.5% vs 14.4% p < 0.001), fragility fracture (17.7% vs 9.6% p = 0.032) and serious infections (27.8% vs 11.3% p < 0.001). Conclusion LTP in RA was associated with characteristics of difficult-to-treat disease, including higher disease activity, methotrexate failure and more bDMARD use, with significantly more adverse events, notably VTE, fragility fractures, and serious infections. Whilst guidelines advise discontinuation, LTP remains necessary for some RA patients with difficult-to-treat disease. These findings highlight the challenges of managing RA patients, underscoring the need for steroid-sparing strategies to optimise disease control and minimise harm.