Background:  Recent reports have suggested an increased risk of coronary artery disease in HIV-infected patients, particularly those on protease inhibitor (PI) therapy. Our objective was to evaluate the association of various factors with risk of coronary artery disease in HIV-infected patients.

Methods:  A cross-sectional study was performed, using flow-mediated vasodilatation measured by high-resolution brachial artery ultrasound as a surrogate marker of early arteriosclerotic changes. Waist-to-hip ratio, body mass index, ankle brachial index, fasting lipoprotein profile, fasting glucose, insulin levels, HbA1c, and high-sensitivity C-reactive protein were measured. flow-mediated vasodilatation between the different groups was compared using the non-parametric t-test.

Results:  Of the 67 patients studied, 52% were male, 54% were African American, mean age was 44 years, 37% were on PI therapy, and 70% were infected with hepatitis C virus (HCV). We compared values of flow-mediated vasodilatation and lipid profile to a control group of 61 non-diabetic presumably HIV-negative patients matched on a 5-year age group, sex, and gender. Flow-mediated vasodilatation was significantly impaired in the HIV-infected group (7.3±4.7%  vs 12.2±6.2%, p <0.0001). Nitroglycerin-mediated vasodilatation was available for 67% of the matched group. Nitroglycerin-mediated vasodilatation was significantly lower in the HIV-infected groups (17.4±8.3% vs 21.8±9.1%, p = 0.02). Among HIV-infected persons, flow-mediated vasodilatation in the PI group did not differ significantly from the non-PI group (7.8±5.2% vs 6.9±4.5%, p = 0.40). Flow-mediated vasodilatation in HCV⁺ patients did not differ from HCV patients (7.6±5.2% vs 6.8±3.0%, p = 0.38), but non-African Americans were more likely to have flow-mediated vasodilatation impairment than African Americans (5.8±4.0% vs 8.6±5.0%, p = 0.04). Among the HIV-infected patients, no associations were seen between age, gender, smoking, dyslipidemia, obesity, or hyperinsulinemia and flow-mediated vasodilatation.

Conclusions:  HIV infection is associated with a substantial impairment of endothelium-dependent vasodilation and a modest impairment of non-endothelium-dependent vasodilation that is consistent with a reduction in the biological activity of nitric oxide. These abnormalities were not related to PI therapy, and might predispose HIV-infected patients to cardiovascular disease. 

Keywords: Coronary Artery Disease in HIV Infection; Endothelial Dysfunction in HIV Infection