Randomized trial of dapsone and aerosolized pentamidine for the prophylaxis of Pneumocystis carinii pneumonia and toxoplasmic encephalitis.

PURPOSE

Pneumocystis carinii pneumonia (PCP) and toxoplasmic encephalitis are the most frequent pulmonary and central nervous system opportunistic infections associated with human immunodeficiency virus (HIV) infection. We designed a prospective study to compare the effects of aerosolized pentamidine and dapsone in the prophylaxis of these infections in HIV-infected persons with CD4+ lymphocyte counts less than 250/mm3.

PATIENTS AND METHODS

Two hundred seventy-eight patients seropositive for HIV who had acquired immunodeficiency syndrome (AIDS) or advanced AIDS-related complex were randomly assigned to receive intermittent dapsone (100 mg twice weekly) or aerosolized pentamidine (100 mg every 2 weeks). The proportion of patients remaining free of PCP or toxoplasmosis was analyzed with the log-rank test as a function of time, as were the effects of zidovudine or prophylaxis on survival.

RESULTS

Dapsone and aerosolized pentamidine demonstrated similar efficacy in the primary and secondary prophylaxis of PCP, with 15 (18%) failures among patients receiving dapsone compared to 15 (14%) among those receiving aerosolized pentamidine (p = 0.4), after a mean length of follow-up of 42 and 44 weeks, respectively. Dapsone was more effective in the primary prophylaxis of toxoplasmic encephalitis, with six toxoplasmic encephalitis events occurring among those receiving aerosolized pentamidine, compared to none among those taking dapsone (p = 0.01). Primary prophylaxis for PCP was more effective than secondary prophylaxis with either therapy. Zidovudine therapy did not prevent PCP yet prolonged the PCP-free interval for those in whom either prophylactic therapy failed. Kaplan-Meier estimates did not show a difference in survival between the patients receiving either therapy, yet zidovudine use was associated with improved survival, independent of race and risk factor (Cox proportional hazards model, p = 0.001). The 1-month survival for patients developing PCP despite prophylaxis was better with those in whom dapsone failed than it was for those in whom aerosolized pentamidine failed (p = 0.08).

CONCLUSION

Dapsone is as effective as aerosolized pentamidine in preventing PCP and has the advantage of a lower cost, easier administration, and possibly an additional preventive effect against toxoplasmosis. Zidovudine prolongs the PCP-free interval for patients receiving prophylaxis, regardless of which prophylactic agent is used.