Higher Corticosteroid Dose Shows No Survival Advantage in Severe HIV-Associated PJP

A retrospective study of hospitalized patients with severe HIV-associated Pneumocystis jirovecii pneumonia found that higher daily corticosteroid doses were not associated with better 28-day or 90-day survival than lower doses after propensity-score matching. The matched cohort included 116 patients, with 58 in each dosing group. Mortality was 25.9% in the lower-dose group and 24.1% in the higher-dose group at 28 days, while 90-day mortality was 39.7% in both groups.

The analysis addressed an unresolved dosing question for patients with severe HIV-associated PJP who receive adjunctive corticosteroids under clinical guidelines. Investigators categorized patients using a prespecified ratio of daily corticosteroid dose to body weight recorded in the study database. Patients receiving less than 1 mg/kg/day were assigned to the lower-dose group, while those receiving at least 1 mg/kg/day were assigned to the higher-dose group.

Because this was a retrospective study, dosing was not assigned randomly, and baseline imbalances were observed between the groups before matching. Researchers used 1:1 propensity-score matching with clinically relevant baseline covariates to reduce measured differences between the dosing groups, then evaluated balance using standardized mean differences. The procedure yielded 58 patients per group.

All-cause mortality did not differ significantly between the regimens, with log-rank p-values of 0.827 for the 28-day comparison and 0.961 for the 90-day comparison. A parsimonious propensity-score sensitivity analysis that excluded the APACHE II score and total Sequential Organ Failure Assessment score produced qualitatively similar conclusions for survival at both time points. These findings show that no survival advantage was detected with the higher-dose regimen in the matched population; they do not establish that the dosing strategies are equivalent, because the retrospective study was not designed as an equivalence or noninferiority study.

The higher-dose regimen also showed no statistically significant overall advantage on the prespecified secondary outcomes. None of these comparisons showed a statistically significant between-group difference. The study likewise identified no statistically significant overall differences in corticosteroid-related adverse events, including hyperglycemia and secondary infection.

That result means no difference was detected in the analyzed sample, rather than demonstrating that the regimens carry identical risks. Several features limit the strength and generalizability of the findings. The study was retrospective, treatment was not randomly assigned, and the matched sample was relatively small.

Propensity-score matching can address measured baseline differences but cannot eliminate all potential residual confounding. The investigators also noted that the conclusions could be sensitive to the specification of the propensity-score model, although the sensitivity analysis excluding APACHE II and total SOFA did not materially alter the survival results. Overall, among patients with severe HIV-associated PJP receiving adjunctive corticosteroids, a daily dose of at least 1 mg/kg was not associated with improved 28-day or 90-day survival compared with a dose below 1 mg/kg after matching.

No statistically significant overall differences were identified in the prespecified secondary outcomes or recorded adverse events, but the evidence warrants cautious interpretation rather than a claim of equivalence.

Subject of Research: Corticosteroid dose and clinical outcomes in severe HIV-associated Pneumocystis jirovecii pneumonia

Article Title: Efficacy and safety of high-dose versus low-dose corticosteroid therapy in severe patients with HIV-associated Pneumocystis jirovecii pneumonia

Article References: Liu, J., Hao, J., Wang, Y., Du, C., Liu, J., & Li, A. (2026). Efficacy and safety of high-dose versus low-dose corticosteroid therapy in severe patients with HIV-associated Pneumocystis jirovecii pneumonia. BMC Infectious Diseases. https://doi.org/10.1186/s12879-026-14545-3

Image Credits: AI Generated

DOI: 10.1186/s12879-026-14545-3

Keywords: HIV, Pneumocystis jirovecii, pneumonia, corticosteroids, dose comparison, propensity-score matching, mortality, retrospective study, Efficacy, safety, high-dose, versus