CORTICOSTEROIDS PREVENT EARLY DETERIORATION IN PATIENTS WITH MODERATELY SEVERE PNEUMOCYSTIS-CARINII PNEUMONIA AND THE ACQUIRED-IMMUNODEFICIENCY-SYNDROME (AIDS)

Objective: To determine whether oral corticosteroids can prevent early deterioration in patients with acquired immunodeficiency syndrome (AIDS)-related Pneumocystis carinii pneumonia. Design: Prospective, double-blind, placebo-controlled, randomized trial. Methods: Included patients were having their first P. carinii pneumonia episode, had no other known active pulmonary pathology, had no contraindications for corticosteroids, received no anti-P. carinii pneumonia medications for more than 48 hours, and had oxygen saturation by pulse oximetry of 85% or more and less than 90% at rest or a 5-percentage-point decrease in oxygen saturation with exercise while breathing room air. Consenting subjects were randomly assigned to prednisone, 60 mg/d for 7 days, followed by a progressive tapering over 14 days or to an identical placebo. Early deterioration, the endpoint of the trial, was defined as a 10% decrease in baseline oxygen saturation on day 3 or thereafter. The cases of patients developing early deterioration were considered to be failures of treatment; the code was then broken, and the patient's treatment was left to the judgment of the treating physician. Sequential analysis was done with the primary variable being development of early deterioration. Results: The trial was terminated 5 April 1989 on the basis of the sequential analysis when a total of nine episodes of early deterioration had occurred in the first 37 patients at an overall significance level of P = 0.0136. A total of 8 of 19 placebo-treated patients (42.1%) developed early deterioration compared with only 1 of 18 patients (5.6%) treated with corticosteroids. Baseline characteristics were not statistically different between the two treatment groups. The adjusted odds ratio for the treatment effect was 5.87 (95% CI, 1.7 to 27.4). The adjusted point estimates for the probability of early deterioration in the placebo and corticosteroid groups were 43% and 12%, respectively. All 8 patients in the placebo group developing early deterioration recovered rapidly with addition of corticosteroid treatment. The single patient with early deterioration in the corticosteroid group died on day 6 from overwhelming P. carinii pneumonia, as documented at autopsy. The corticosteroid group had an increased exercise tolerance on day 7 that persisted at day 30. Conclusion: Oral corticosteroids prevent early deterioration and increase exercise tolerance in patients with moderately severe AIDS-related P. carinii pneumonia.