Objective: To assess the long-term safety of adjunctive corticosteroids in the treatment of Pneumocystis carinii pneumonia (PCP).
Design: Analysis of data from a large prospective observational database.
Setting: HIV clinic at a large urban teaching hospital.
Patients: One hundred seventy-four patients who developed PCP after being enrolled in the database.
Results: Fifty-three patients (30%) received adjunctive corticosteroids and 121 (70%) did not. Survival did not differ between groups after adjusting for CD4 count (relative risk for adjunctive corticosteroids = 0.74, p = 0.13). There were no differences in the incidence of cytomegalovirus disease (adjunctive corticosteroids: 18.5 cases per 100 person-years vs no adjunctive corticosteroids: 15.7, p = 0.22), Mycobacterium avium complex (23.4 vs 27.0, p = 0.73), cryptococcal meningitis (1.8 vs 4.1, p = 0.58), toxoplasmosis (3.6 vs 11.0, p = 0.28), Kaposi's sarcoma (1.8 vs 2.2, p = 0.92), herpes simplex (27.1 vs 42.7, p = 0.66), herpes zoster (3.8 vs 6.9, p = 0.71), oropharyngeal candidiasis (18.9 vs 10.9, p = 0.09), or non-Hodgkin's lymphoma (3.5 vs 4.2, p = 0.92). Esophageal candidiasis was more common among adjunctive corticosteroid recipients (45.1 vs 26.6, p = 0.01). Results were similar for time to development of opportunistic conditions.
Conclusions: Adjunctive corticosteroids do not increase mortality or the risk of most common HIV-associated complications.