Current issues of ACP Journal Club are published in Annals of Internal Medicine


Prednisone decreased mortality and respiratory failure in moderate-to-severe Pneumocystis carinii pneumonia

ACP J Club. 1991 Mar-April;114:41. doi:10.7326/ACPJC-1991-114-2-041

Related Content in this Issue
• Companion Abstract and Commentary: Methylprednisolone decreased mortality and respiratory failure in severe Pneumocystis carinii pneumonia

Source Citation

Bozzette SA, Sattler FR, Chiu J, et al. A controlled trial of early adjunctive treatment with corticosteroids for Pneumocystis carinii pneumonia in the acquired immunodeficiency syndrome. N Engl J Med. 1990;323:1451-7.



To evaluate the clinical efficacy and adverse effects of adjunctive corticosteroids in the treatment of patients with Pneumocystis carinii pneumonia and the acquired immunodeficiency syndrome.


Randomized, unblinded, controlled trial with analyses at 21, 31, and 84 days after entering the study.


Multicenter trial conducted in California.


333 adult patients with AIDS were entered into the trial. Most of the analysis of outcome was done on 251 of these patients who had morphologically confirmed (n = 225) or presumed pneumocystis pneumonia (n = 26). Patients were excluded if they were receiving mechanical ventilation, had a hypoxemia ratio (PaO2/ FIO2) of less than 75, or had started antipneumocystis therapy more than 36 hours before study entry.


123 patients were assigned to receive corticosteroids and 128 were assigned to standard therapy. Standard therapies were oral or parenteral trimethoprimsulfamethoxazole, parenteral pentamidine, or oral dapsone plus trimethoprim. Corticosteroid therapy consisted of oral or parenteral prednisone (40 mg twice daily for 5 days, then 40 mg daily for 5 days, followed by 20 mg daily for the duration of antipneumocystis therapy).

Main outcome measures

Respiratory failure was defined as a hypoxemia ratio of less than 75, institution of mechanical ventilation, or death.

Main results

At 31 days follow-up, 17 patients (14%) receiving corticosteriods had developed respiratory failure compared with 38 patients (30%) receiving standard therapy (P = 0.001) {This absolute risk reduction (ARR) of 16% means that 6 patients would need to be treated (NNT) with corticosteroid therapy for 31 days (rather than standard treatment) to prevent 1 additional patient from developing respiratory failure; the relative risk reduction (RRR) was 53%, CI 23% to 72%}*. Also at 31 days follow-up, 13 patients (11%) receiving corticosteroid therapy died compared with 28 patients (22%) receiving standard treatment {P = 0.02} {ARR 11%, NNT 9, CI 5 to 45; RRR 52%, CI 12% to 74%}.* In patients whose pneumocystis pneumonia was initially rated mild (hypoxemia ratio > 350), the risks for respiratory failure and death were not significantly affected by corticosteroid therapy. Fever (P < 0.01), cough (P < 0.001), and dyspnea (P = 0.01) were observed significantly less often in the corticosteroid treatment group. After 84 days, recurrence of most opportunistic infections was not significantly affected.


The addition of corticosteroid to standard therapy for pneumocystis pneumonia was associated with lower rates of respiratory failure and death for patients with moderate-to-severe illness. No serious side effects could be directly attributed to prednisone.

Source of funding: California University-Wide AIDS Research Program.

Address for article reprint: Dr. S. Bozzette, Division of Infectious Diseases, Box H208, University of California-San Diego Medical Center, 225 Dickinson Street, San Diego, CA 92103, USA.

*Numbers calculated from data in article.


This trial adds weight to the accumulating evidence in support of the benefit of corticosteroid administration as adjunctive therapy for pneumocystis pneumonia. Although it is the largest trial yet published, it is unblinded and should therefore be viewed in the context of other available information. Such an overview was recently performed by an NIH-University of California expert panel (1) that reviewed this trial as well as 3 other North American trials and 1 unpublished European trial. On the basis of the combined evidence, the panel recommended that persons older than 13 years with documented or suspected Pneumocystis carinii pneumonia accompanied by moderate-to-severe pulmonary dysfunction (defined as PaO2 < 70 mm Hg or an A-a gradient > 35 mm Hg) be started on adjunctive corticosteroid therapy as early as possible in the course of the illness. The need for prompt institution of therapy is underscored by the fact that benefit from corticosteroids was seen in all 3 studies in which the drug was instituted within 72 hours of starting antipneumocystis therapy. Benefit was not observed in the 1 study allowing longer periods to elapse. The optimal dose and regimen of corticosteroid therapy have not been established. The role of adjunctive corticosteroid therapy in mild episodes of pneumocystis pneumonia is uncertain. When respiratory failure occurs in the face of adjunctive corticosteroid treatment, the proper course of action remains undetermined.

Martin T. Schechter, MD, PhD, MSc
University of British ColumbiaVancouver, British Columbia, Canada


1. The National Institutes of Health-University of California Expert Panel for Corticosteroids as Adjunctive Therapy for Pneumocystis Pneumonia. Consensus statement on the use of corticosteroids as adjunctive therapy for pneumocystis pneumonia in the acquired immunodeficiency syndrome. N Engl J Med. 1990;323:1500-4.