Maxime Verhoeven

27 Systematic review of remission-induction strategies in early RA Several of the 29 studies used more than 1 remission definition; in all, 46 remission definitions were used, range 1-4 per study. Most studies used at least a definition of remission where remission had to be present ≥ 1 visit within 6 to 12 months follow-up and according to one of our remission outcome definitions, and we will describe the results based on these outcomes (Table 1). Seventeen studies defined remission as DAS28 <2.6, 12 studies used the Boolean remission definition, 7 studies used CDAI ≤2.8 and 10 studies used SDAI ≤3.3, results are described separately below. Overall, for 32 of the 46 remission definitions (70%), a statistically significant effect in favor of remission-induction strategy was found. DAS28-based remission When DAS28 was used for remission definition, 13/17 (76%) studies showed a statistically significant effect in favor of the remission-induction strategy, Figure 1. The pooled RR of achieving remission for strategies using a bDMARD in the remission- induction strategy compared to the single csDMARD-initiating strategy without GC bridging was 1.73 [95%CI 1.59 - 1.88] versus 1.20 [95%CI 1.03 - 1.40] for studies which used a combination csDMARD based remission-induction strategy compared to the single csDMARD-initiating strategy without GC bridging. For studies using GC bridging in the single csDMARD-initiating strategy, no statistically significant additional effect for the remission-induction strategy was found (pooled RR 1.06 [95%CI 0.83 - 1.35]). One of them used a bDMARD in the remission-induction strategy arm. 29 One cohort study only provided an odds ratio (OR) for achieving remission in patients treated with a remission-induction strategy compared to a single csDMARD-initiating strategy, with or without additional GC use (without sufficient information to calculate an RR). Results were in favor of the remission-induction strategy (OR 1.82 [95%CI 1.01 - 3.29]). 25 2