Effect of hydrocortisone on mortality in patients with severe community-acquired pneumonia

In an international adaptive randomized controlled platform trial testing multiple interventions, adults admitted to the intensive care unit (ICU) with severe CAP were randomized to a 7-day course of intravenous hydrocortisone (50 mg every 6 h) or control (no corticosteroid). The primary end point was 90-day all-cause mortality, analyzed iteratively by a Bayesian hierarchical model estimating distinct treatment effects for patients presenting with influenza (Y/N) and shock (Y/N).

Fixed 7-day course hydrocortisone enrollment was stopped for futility ( 20% relative improvement). Of 658 patients enrolled, 536 were randomized to hydrocortisone and 122 to control. Vital status at day 90 was missing for 15 patients. Day 90 mortality was 15% (78/521) and 9.8% (12/122) for the hydrocortisone and control groups. The adjusted odds ratio ranged from 1.52 to 1.63 (with all 95% CrI crossing 1), while the probability of > 20% relative reduction of day 90 mortality ranged from 7.1 to 3.3% across influenza and shock strata. Results were consistent in sensitivity and pre-specified secondary outcomes. In exploratory analyses, the duration of shock appeared lower in the hydrocortisone group compared with control (median (IQR) of 2 (2-5) days compared to control 3 (2-6.75) days, p value = 0.05).