570 Improved Outcomes with Earlier Initiation of HAART: Longer Follow-up of an Observational Cohort Study T Sterling*, R Chaisson, J Keruly, R Moore Johns Hopkins Univ Sch of Med, Baltimore, MD
Background: A number of treatment guidelines currently recommend that HAART be initiated in asymptomatic HIV-infected persons when the CD4+ lymphocyte count is = 200 cells/mm3. These recommendations are based on data from observational cohort studies, which have been limited by relatively short follow-up time. We assessed differences in clinical disease progression over a longer follow-up period than prior studies, which could influence the decision regarding when to start HAART.
Methods: We conducted an observational cohort study from 7/1/96 to 6/30/02 among patients (pts) initiating HAART (= 90 days of a regimen including an HIV-1 protease inhibitor, non-nucleoside reverse transcriptase inhibitor, or triple nucleosides) at the Johns Hopkins HIV Clinic. Development of a new opportunistic infection or death (combined endpoint) was assessed. Durable virologic suppression was defined as having a majority of undetectable (< 400 copies/ml) viral loads after the start of therapy. Differences in disease progression were assessed by log-rank comparisons of Kaplan-Meier estimates and Cox proportional hazards models.
Results: There were 1,173 pts who initiated HAART. The median time on therapy was 30 months (mos) (range: 3-72 mos), and the median follow-up was 36 mos (range: 2-72 mos), both longer than in our previous analyses. Among all persons on HAART (e.g., not accounting for adherence or durable response), persons with baseline CD4 < 200 progressed more rapidly than persons with baseline CD4 201-350 (p < 0.001) and CD4 > 350 (p < 0.001). Among persons who achieved durable virologic suppression, those with baseline CD4 < 200 tended to progress faster than those with baseline CD4 201-350 (p = 0.09) and did progress faster than persons with baseline CD4 > 350 (p = 0.01). In addition, disease progression of those with baseline CD4 201-350 was slower than those with CD4 < 50 (p = 0.03) but not CD4 51-200 (p = 0.3).
Conclusions: With longer follow-up in this observational cohort study, differences in clinical disease progression have become apparent, even among persons with durable virologic suppression. HAART should be started before pts achieve low CD4 counts.
