Glucocorticoid discontinuation in patients with early rheumatoid and undifferentiated arthritis: a post-hoc analysis of the BeSt and IMPROVED studies

Objectives To evaluate the success rate of glucocorticoid discontinuation and to study which factors are associated with successful discontinuation. Methods Data from two treat-to-target studies, BeSt (target Disease Activity Score (DAS) <= 2.4) and IMPROVED (target DAS <1.6), were evaluated for all patients initially treated with a tapered high dose of prednisone with conventional synthetic disease-modifying antirheumatic drugs. Prednisone was discontinued when DAS <= 2.4 was maintained for 28 weeks in BeSt and as soon as DAS was <1.6 in IMPROVED. Discontinuation was considered successful if the target was maintained at the next visit. Logistic regression analyses were performed to identify predictors of successful discontinuation. A mixed effects logistic regression model was used to assess whether primary versus secondary discontinuation was as successful. Results In the BeSt study, 40% (47 of 93) of patients flared after primary prednisone discontinuation, and of the other 60% (56 of 93), 38% had to restart later. Of those who restarted (secondary discontinuation), 47% (17 of 35) again flared. In IMPROVED, after primary discontinuation 39% (158 of 400) flared, and of the other 61% (242 of 400), 40% had to restart later. After secondary discontinuation 49% (68 of 139) flared. Only in IMPROVED a secondary attempt was less successful (BeSt OR 0.71, p=0.45; IMPROVED OR 0.60, p=0.01). A lower DAS both at baseline and stop visit and male gender (in IMPROVED) were associated with successful primary discontinuation. Conclusion Primary glucocorticoid discontinuation resulted in direct loss of disease control in approximately 40% and secondary in 50% of patients. 'Standard' baseline characteristics seem insufficient to personalise the duration of temporary glucocorticoid bridging, but the DAS at the time of discontinuation might provide guidance.