Eight trials (n=1,876) were included in the review: six RCTs, one retrospective cohort study and one case-control study. Overall study quality varied; median Cronin scores for the six RCTs ranged from 11 to 13.
Corticosteroid therapy significantly improved shock reversal at seven days compared to control groups (RR 1.41 95% CI 1.22 to 1.64), but did not differ significantly for all-cause mortality or incidence of superinfection. Adrenal responsiveness did not influence results for any outcomes.
Sensitivity analyses showed that cumulative effects, quality of included trials and single study exclusion did not influence results. No significant heterogeneity was observed for any of the analyses.