Abstract 2129: Increased Carotid Intima-Media Thickness in HIV-Infected Patients Exposed to Protease Inhibitor- compared to Non-Nucleoside Reverse Transcriptase Inhibitor-containing Combination Antiretroviral Therapy
Background: Increasing exposure to protease inhibitor (PI)-, but not non-nucleoside reverse transcriptase inhibitor (NNRTI)-containing combination antiretroviral therapy (cART) has been associated with an increased cardiovascular risk. This difference was at least partly explained by different effects of these two antiretroviral drug classes on plasma lipids. These include the propensity of currently licensed NNRTIs to be associated with increases in high density lipoprotein cholesterol (HDL-C) levels which may be atheroprotective.
Methods and Results: In a cross-sectional study we investigated the impact of PI-based versus NNRTI-based cART in 132 HIV-1-infected patients with plasma virus suppressed to below the limit of detection. Patients had been continuously exposed for at least 2 years to either PI or NNRTI containing regimens, but not both. Carotid intima-media thickness (C-IMT), an established surrogate marker for cardiovascular disease (CVD), and fasting metabolic parameters were measured. Mean (± SD) C-IMT in patients treated with PI-based cART was 0.81 (± 0.17) mm as compared to 0.71 (± 0.14) mm in NNRTI treated patients (p=0.0003). Mean HDL-C and apolipoprotein A-I levels were significantly higher in the NNRTI group as compared to the PI group (1.39 mmol/L versus 1.03 mmol/L, p<0.0001, and 1.44 mmol/L versus 1.33 mmol/L, p=0.0008, respectively) and were inversely associated with C-IMT on univariate analysis. In the final multivariate linear regression analysis increased age, body mass index, diastolic blood pressure, as well as use of PI were each independently associated with greater C-IMT.
Conclusions: Treatment of HIV-1-infected patients for two years or more with PI-based compared to NNRTI-based cART is associated with greater C-IMT, consistent with the reported higher risk of CVD in patients using PI. However, this difference seems not fully explained by a more favorable impact of NNRTI-based cART on HDL-C and apo-A1 levels.