Methylprednisolone intravenously — 30 mg twice daily on days 1 to 5, 30 mg once daily on days 6 to 10, then 20 mg once daily until day 21
Control
Placebo
Population
226 HIV-negative immunocompromised patients with severe Pneumocystis jirovecii pneumonia and acute respiratory failure
Follow-up
28 days
Primary endpoint
All-cause mortality at 28 days
Result
28-day mortality was 32.4% in the placebo group versus 21.5% with corticosteroids — mean difference 10.9 percentage points (95% CI -0.9 to 22.5, p=0.069). No significant differences in safety outcomes, including secondary infections (34.2% versus 23.4%) or insulin requirement (22.5% versus 30.8%).
In immunocompromised HIV-negative patients with P jirovecii pneumonia, adjunctive corticosteroid treatment did not significantly decrease 28-day mortality.
The trial authors, in the published abstract
How it has aged
An 11 percentage point absolute difference with p=0.069 — the shape of a trial that may simply be too small rather than one that is truly negative. It leaves practice where it was: steroids are established in HIV-associated disease (Bozzette 1990) and remain unproven outside it.
Publications
Adjunctive corticosteroids in non-AIDS patients with severe Pneumocystis jirovecii pneumonia (PIC): a multicentre, double-blind, randomised controlled trial
Lemiale V, Resche-Rigon M, Zerbib Y, et al. Adjunctive corticosteroids in non-AIDS patients with severe Pneumocystis jirovecii pneumonia (PIC): a multicentre, double-blind, randomised controlled trial. Lancet Respir Med 2025 Sep;13(9):800-808.
28-day mortality 21.5% with steroids versus 32.4% with placebo, p=0.069
No excess of secondary infection or insulin requirement
Confidence interval compatible with meaningful benefit or none