901: CORTICOSTEROIDS FOR PNEUMOCYSTIS JIROVECII PNEUMONIA IN HIV-NEGATIVE CRITICALLY ILL PATIENTS

Introduction: Corticosteroid (CS) treatment reduces the risk of mortality and respiratory failure in patients with human immunodeficiency virus (HIV) and moderate to severe Pneumocystis jirovecii pneumonia (PJP). The role of CS in HIV-negative immunocompromised hosts with PJP remains unclear. Our study aimed to evaluate HIV-negative critically ill patients with PJP and respiratory failure to determine if CS were associated with reduced risk of mortality. Methods: This was a retrospective cohort of HIV-negative critically ill patients with confirmed PJP requiring respiratory support. Patients were identified via the Premier Incorporated database. Inclusion criteria were adult patients requiring respiratory support (supplemental oxygen or non-invasive ventilation [NIV] and mechanical ventilation [MV]) within 48 hours of PJP diagnosis and intensive care unit admission. Exclusion criteria consisted of new or existing HIV diagnosis, death or palliative care within 72 hours, and lack of active PJP treatment. Patients who did not receive CS were compared to those who received prednisone equivalents of at least 40 mg daily (CS). The primary outcome was in-hospital mortality. Results: A total of 807 patients were included (n=176 no CS, n=631 CS) with an average age of 62 years and were 54% male. In the first 48 hours of PJP treatment, 432 patients (54%) required MV and 375 (46%) required NIV. In-hospital mortality occurred in 75 patients (42.6%) who did not receive CS and 237 (37.6%) who received CS (Odds Ratio [OR] 0.81; 95% Confidence Interval [CI] 0.58 to 1.14). There was no difference in ICU free days (11 days [no CS] vs 10.8 days [CS]; p=0.72) or 28-day ventilator free days (13.6 days [no CS] vs 13.3 days [CS]; p=0.55). Furthermore, subgroup analysis of MV patients revealed no significant mortality difference with regard to CS treatment (68% no CS vs 69% CS; OR 0.76; 95% CI 0.51 to 1.14). Conclusions: Corticosteroids are often used for critically ill HIV-negative patients with PJP and respiratory failure. However, CS administration to these patients was not clearly associated with significant reduction in mortality or other clinical outcomes. Further determination of patient selection and risk-benefit analysis of CS in this population is warranted.