HIV infection and certain antiretroviral medications appear to contribute to cardiovascular disease (CVD) risk. CVD events were similar between the early (at CD4 >500cells/µL) and deferred (to CD4 <350cells/µL) antiretroviral therapy (ART) arms in START. We studied individual risk factor changes over time in START to understand the net influence on CVD risk.
Clinical and laboratory measures were ascertained annually among START participants. Mean change from baseline in risk factors between the early and deferred ART arms were compared over follow-up. Framingham and D:A:D 10yr CVD risk scores were calculated using published equations. Incident dyslipidemia (LDL >160mg/dL or use of lipid lowering drug), diabetes (diagnosis or fasting blood glucose ≥126 mg/dL) and hypertension (diagnosis or use of blood pressure medication) during follow-up were assessed using unadjusted Cox proportional hazards regression.
Characteristic among 4,685 START participants across 35 countries at entry were: median age was 36 years, CD4 count 651cells/mm3, HIV viral load 12,759 copies/mL, 73% male, 32% smokers; median CVD risk variables were SBP/DBP 120/76 mmHg, Total Cholesterol (C) 168 mg/dL, LDL-C 102 mg/dL, HDL-C 43 mg/dL, fasting glucose 85 mg/dL. At entry median 10 year risk for CVD was 2.3% and 1.8% (Framingham and D:A:D, respectively). Mean follow-up was 3.0 years. The early and deferred ART groups spent 94% and 28% of follow-up time on ART, respectively. Differences in CVD risk factors between early and deferred ART over time are shown in the table. Early ART increased fasting glucose and all lipid parameters, with a minimal decline in Total:HDL-C. Incident dyslipidemia was greater among early vs. deferred groups (hazard ratio 1.62 [95% CI: 1.33-1.94]), though incident diabetes or hypertension did not differ between groups. The use of lipid-lowering therapy increased overall during follow-up, but use did not differ between ART groups over time.
Among a diverse global population of HIV+ persons with high CD4 counts, early ART initiation had increased LDL-C and the prevalence of dyslipidemia. However, concurrent increases in HDL-C and other mixed effects resulted in no consistent differences in CVD risk scores over time. These randomized data suggest that early ART has both positive and negative influences on CVD risk among HIV+ individuals with preserved immunity.