Background:  Although incidence and mortality rates due to cryptococcal meningitis (CM) in patients with AIDS, have declined in the West, it remains a leading cause of AIDS related deaths in much of Sub-Saharan Africa.

Methods:  We conducted a randomized clinical trial at Parirenyatwa Hospital in Harare, Zimbabwe to determine whether early vs delayed initiation of ART in patients treated for CM with fluconazole would improve survival rates. We recruited and randomized 54 patients with AIDS-related cryptococcal meningitis to initiate ART (stavudine, lamivudine, or nevirapine) within 72 hours of diagnosis or to delay initiation until 10 weeks after fluconazole monotherapy. Patients were followed for 2 years and the primary end-point was mortality.  

Results:  Mortality due to CM was high with an overall mortality rate of 62%. Mortality rate in the early treatment group was 82% compared with 37% in the delayed treatment group. The early treatment group was associated with early mortality with a median survival time of 35 days (5 weeks) compared with 274 days (39 weeks) in the delayed treatment group (p = 0.028). The results of the adjusted Cox model show that CM patients in the early treatment group have twice the risk of mortality in comparison to those patients who delay ART initiation by 10 weeks (hazard ratio 2.36, 95%CI 1.12 to 4.97).

Conclusion:  This is the first prospective randomized clinical trial showing that timing of initiation of ART is critical in the management of opportunistic infections. Early initiation of ART in patients with cryptococcal meningitis results in increased mortality with most mortality occurring early after initiation of ART.