Effectiveness of pulse methylprednisolone in patients with non-human immunodeficiency virus pneumocystis pneumonia: a multicentre, retrospective registry-based cohort study.

Background: A recent database study and meta-analysis reported that adjunctive glucocorticoid therapy reduces mortality in patients with non-human immunodeficiency virus-associated (non-HIV) Pneumocystis jirovecii pneumonia (PCP), having hypoxemia. However, the optimal glucocorticoid dose remains unclear. Our study aimed to evaluate the effectiveness of pulse methylprednisolone compared with mild-to-moderate steroid doses in patients with non-HIV PCP.

Methods: This multicentre retrospective cohort study included adults with non-HIV PCP receiving adjunctive steroids at three Japanese tertiary care hospitals from June 2006 to March 2021. Patients were categorised into pulse methylprednisolone and mild-to-moderate dose groups. Pulse methylprednisolone involved an initial intravenous infusion of 500-1000 mg methylprednisolone daily, while the mild-to-moderate dose was lower. Primary and secondary outcomes were 30-day and 180-day mortality from treatment initiation. Patient characteristics were adjusted using propensity score analysis with overlap weighting. Subgroup analysis focused on patients with respiratory failure.

Results: The study included 139 patients with non-HIV PCP: 55 in the pulse methylprednisolone group and 84 in the mild-to-moderate dose group. After adjusting for patient background, 30-day mortality (14.2% vs. 15.5%, P = 0.850) and 180-day mortality (33.5% vs. 27.3%, P = 0.516) did not differ significantly between groups. Subgroup analysis revealed no significant associations among patients with respiratory failure.

Conclusions: After adjusting for patient characteristics, no difference in prognosis was observed between pulse methylprednisolone and mild-to-moderate dose groups in patients with non-HIV PCP. A mild-to-moderate dose of adjunctive corticosteroid may suffice for treating non-HIV PCP.