Background: The optimal time to initiate antiretroviral treatment is unknown.  Survival rates observed among patients who start therapy at high or low CD4 cell counts cannot be attributed solely to initiating treatment early or late in disease because of confounding with patients’ underlying disease stage, unless matched controls are available for comparison.

Methods: We examined disease progression and death among a cohort of patients receiving care at a university-affiliated HIV clinic after July 1, 1995. Patients were required to have at least 12 months of follow up and no HIV-related clinical events prior to or during this period (asymptomatic). We used Cox proportional hazards survival analysis to determine the adjusted hazard ratio (HR) of disease progression or death according to whether or not patients received PI or NNRTI-based antiretroviral treatment within 12 months of their initial clinic visit, controlling for CD4 cell count, HIV-1 RNA level, race, risk factor for HIV transmission, gender, and age. Clinical endpoints included symptomatic conditions, opportunistic infections, and death.

Results: 238 patients were observed for a median of 26.6 months (range 12.3-70.5 months). Patients initiating PI or NNRTI-based antiretroviral treatment at all CD4 cell count levels had less than half the rate of disease progression (HR=0.40; p<0.001) compared with patients who did not receive treatment within 12 months of their initial clinic visit. After controlling for antiretroviral treatment, patients with a CD4 cell count of 200-350 cells/mm³ or >350 cells/mm³ had lower rates of disease progression (HR=0.52; p<0.05 and HR=0.38; p<0.01 respectively) compared with patients with a CD4 cell count <200 cells/mm³. Patients 18-29 years of age also had slower disease progression (HR=0.25; p<0.01) than patients aged 30-39 years.

Conclusions: Our results suggest clinical benefit of antiretroviral therapy initiated at all CD4 cell count levels that must be balanced with the long-term risks of treatment.