Background: Pneumocystis jiroveci pneumonia (PCP) prophylaxis is recommended for patients with CD4 counts of less than 200 cells/mm3. This study examines the proportion of patients in the TREAT Asia HIV Observational Database (TAHOD) receiving PCP prophylaxis, and its effect on PCP and mortality.
Methods: TAHOD patients with prospective follow up had data extracted for prophylaxis using co-trimoxazole, dapsone or pentamidine. The proportion of patients on prophylaxis was calculated for each calendar year since 2003 among patients with CD4 counts of less than 200 cells/mm3. The effect of prophylaxis on PCP and survival were assessed using random-effect Poisson regression models.
Results: There were a total of 4050 patients on prospective follow up, and 90% of them were receiving combination antiretroviral therapy. Of those with CD4 counts of less than 200 cells/mm3, 58% to 72% in any given year received PCP prophylaxis, predominantly co-trimoxazole. During follow up, 62 patients developed PCP (0.5 per 100 person-years) and 169 died from all causes (1.36/100 person-years). After stratifying by site and adjusting for age, CD4 count, CDC stage and antiretroviral treatment, those without prophylaxis had no higher risk of PCP, but had a significantly higher risk of death (incident rate ratio 10.8, p<0.001). PCP prophylaxis had greatest absolute benefit in patients with CD4 counts of less than 50 cells/mm3, lowering mortality rates from 33.5 to 6.3 per 100 person-years.
Conclusions: Approximately two-thirds of TAHOD patients with CD4 counts of less than 200 cells/mm3 received PCP prophylaxis. Patients without prophylaxis had significantly higher mortality, even in the era of combination ART. Although PCP may be under-diagnosed, these data suggest that prophylaxis is associated with important survival benefits.