RESUMO
BACKGROUND: Corticosteroids (CS) are widely used to treat Crohn's disease (CD) and ulcerative colitis (UC), but are not recommended as maintenance therapy. Biologic drugs are widely used as an alternative to or in conjunction with CS to induce and maintain remission. This meta-analysis tested the hypothesis that CS use is associated with differential response to biologics. METHODS: We identified published placebo-controlled clinical trials of biologic drugs approved for the treatment of CD or UC. Pooled estimates of the risk difference (RD) and 95% confidence intervals were derived from random effects models for induction of response and remission and maintenance of remission comparing biologic with CS versus biologic alone. Heterogeneity of response was estimated using I2. RESULTS: Fifteen studies met the inclusion criteria. Pooled estimates of the RD and I2 comparing biologic plus CS versus biologic alone were as follows: induction of UC response 0.15 (0.05, 0.25), I2 = 57.29% and CD response 0.02 (- 0.03, 0.06), I2 = 0.01%; induction of UC remission 0.03 (- 0.01, 0.08), I2 = 0.00% and CD remission 0.08(0.02, 0.14), I2 = 7.81%; and maintenance of UC remission - 0.06 (- 0.13, 0.01), I2 = 0.00% and CD remission - 0.06 (- 0.14, 0.03), I2 = 11.24%. Patients in the placebo arm of CD trials who were receiving CS were less likely to achieve remission during the induction phase (pooled RD - 0.05 (- 0.09, - 0.00), I2 = 0.00%). CONCLUSIONS: In this meta-analysis of placebo-controlled trials, CS use was associated with higher biologic response rates for UC and remission rates for CD during the induction phase, but were not associated with improved maintenance of remission.