ObjectiveThe randomised placebo-controlled GLORIA (Glucocorticoid LOw-dose in RheumatoId Arthritis) trial evaluated the benefits and harms of prednisolone 5 mg/day added to standard care for 2 years in patients aged 65+ years with rheumatoid arthritis (RA). Here, we studied disease activity, flares and possible adrenal insufficiency after blinded withdrawal of study medication.MethodsPer protocol, patients successfully completing the 2-year trial period linearly tapered and stopped blinded study medication in 3 months. We compared changes in disease activity after taper between treatment groups (one-sided testing). Secondary outcomes (two-sided tests) comprised disease flares (DAS28 (Disease Activity Score 28 joints) increase >0.6, open-label glucocorticoids or disease-modifying antirheumatic drug (DMARD) increase/switch after week 4 of tapering) and symptoms/signs of adrenal insufficiency. In a subset of patients from 3 Dutch centres, cortisol and ACTH were measured in spot serum samples after tapering.Results191 patients were eligible; 36 met treatment-related flare criteria and were only included in the flare analysis. Mean (SD) DAS28 change at follow-up: 0.2 (1.0) in the prednisolone group (n=76) vs 0.0 (1.2) in placebo (n=79). Adjusted for baseline, the between-group difference in DAS28 increase was 0.16 (95% confidence limit –0.06, p=0.12). Flares occurred in 45% of prednisolone patients compared with 33% in placebo, relative risk (RR) 1.37 (95% CI 0.95 to 1.98; p=0.12). We found no evidence for adrenal insufficiency.ConclusionsTapering prednisolone moderately increases disease activity to the levels of the placebo group (mean still at low disease activity levels) and numerically increases the risk of flare without evidence for adrenal insufficiency. This suggests that withdrawal of low-dose prednisolone is feasible and safe after 2 years of administration.