Pneumocystis prophylaxis and survival in patients with advanced human immunodeficiency virus infection treated with zidovudine. The Zidovudine Epidemiology Group.

BACKGROUND: Pneumocystis carinii pneumonia (PCP) is a major cause of morbidity and mortality in persons with advanced human immunodeficiency virus (HIV) infection. We assessed the impact of prophylaxis for PCP on survival in patients with advanced HIV disease who were treated with zidovudine.
METHODS: A cohort of 1048 patients with prior PCP (N = 437), another acquired immunodeficiency syndrome-defining diagnosis (N = 168) or acquired immunodeficiency syndrome-related complex (N = 443) and with less than 0.250 x 10(9)/L CD4 cells initiated zidovudine treatment between April 1987 and April 1988. They were then followed up for 24 months. Morbidity and mortality outcomes were assessed every 2 months. A time-dependent, Cox proportional hazards model was used to identify factors associated with new episodes of PCP and with survival.
RESULTS: Three hundred thirty-six patients (32%) developed PCP after beginning treatment with zidovudine, with a 24-month actuarial rate of 41%. Patients with prior PCP were more likely to develop PCP during follow up (40%) than those without a history of PCP at entry (27% with PCP at follow-up). Other factors associated with developing PCP were baseline acquired immunodeficiency syndrome vs acquired immunodeficiency syndrome-related complex, and dose interruptions of zidovudine. Thirty-six (17%) of 210 patients who received trimethoprim-sulfamethoxazole prophylaxis developed PCP vs 299 (36%) of 838 who never received trimethoprim-sulfamethoxazole (odds ratio, 0.48). One hundred seven (22%) of 483 patients who ever received aerosol pentamidine prophylaxis developed PCP vs 228 (40%) of 565 who did not receive aerosol pentamidine (odds ratio, 0.55). In a time-dependent Cox proportional hazards analysis, trimethoprim-sulfamethoxazole (relative hazard, 0.21; 95% confidence interval [CI], 0.11 to 0.4) and aerosol pentamidine prophylaxis (relative hazard, 0.25; 95% CI, 0.16 to 0.39) were associated with decreased risk of PCP. Pneumocystis carinii pneumonia during follow-up was strongly associated with death when controlling for other factors (odds ratio, 1.8). For all patients, aerosol pentamidine prophylaxis was associated with a reduced risk of death during follow-up (relative hazard, 0.59; 95% CI, 0.44 to 0.78), while trimethoprim-sulfamethoxazole showed a weaker association (relative hazard, 0.74; 95% CI, 0.54 to 1.1). However, there was a significantly reduced risk of death overall for patients who consistently used trimethoprim-sulfamethoxazole (relative hazard, 0.55; 95% CI, 0.35 to 0.88) or aerosol pentamidine (relative hazard, 0.57; 95% CI, 0.42 to 0.77) and this was most pronounced in patients with a baseline history of PCP. DISCUSSION AND
CONCLUSIONS: Pneumocystis carinii pneumonia was common in advanced HIV infection treated with zidovudine. Prophylaxis with trimethoprim-sulfamethoxazole and aerosol pentamidine both were associated with a decreased likelihood of PCP, and consistent use of each was associated with improved survival. Prophylaxis for PCP is associated with prolonged survival for patients with advanced HIV disease.