Cardiac Steatosis and Left Ventricular Dysfunction in HIV-Infected Patients Treated With Highly Active Antiretroviral Therapy

Heart disease is a major contributor to morbidity and mortality in persons infected with human immunodeficiency virus (HIV), and both HIV and highly active antiretroviral therapy (HAART) may be associated with abnormalities in cardiac function and metabolism (1). Ectopic fat deposition in cardiomyocytes constitutes one possible mechanism, especially considering the influence of HIV itself and HAART on the human metabolic milieu. To test this hypothesis, we studied 27 HIV-seropositive (+) cases on stable HAART (ge;3 years) and 22 HIV-negative control subjects; neither group had history of personal or family history of cardiovascular disease.

Myocardial triglyceride content was measured by magnetic resonance spectroscopy (3T MAGNETOM Verio, Siemens, Germany) (2) and revealed a 3-fold elevation in cases compared with control subjects (Figure 1A). To assess left ventricular (LV) function, magnetic resonance tissue tagging was performed on a mid-ventricular short-axis image (3). Both peak circumferential strain and the peak rate of circumferential strain in diastole were impaired in cases compared with control subjects (Figures 1B and 1C, respectively). Clinical evidence of lipo-dystrophy was also assessed by physical exam of the dorsal back, neck, extremities, and abdomen.

FIGURE 1. Ectopic Fat Deposition and LV Dysfunction in HIV-Infected Patients Receiving HAART.

(A) Representative illustration of a typical cardiac magnetic resonance (CMR) image, showing all 4 chambers of the heart. The white box over the intraventricular septum illustrates the voxel position during spectroscopy. The green line illustrates a representative cardiac spectra, showing a large area of resonance for myocardial water and a small area of resonance for myocardial triglyceride content; magnified directly above for a representative human immunodeficiency virus (HIV)-positive (+) subject (pink line) and a representative HIV-negative (−) subject (green line). The summary data demonstrate a 3-fold elevation in myocardial triglyceride content in HIV-positive subjects (pink bar) versus well-matched HIV-negative control subjects (green bar). (B, i to iv) Representative illustration of a mid-short-axis image of the LV, with tissue tagging applied at end diastole (i) and progressing to end systole (iv). The graph shows the circumferential strain from a representative HIV-positive subject (pink line) and control subject (green line). The summary data demonstrate impaired circumferential strain in cases versus control subjects. (C, i to iv) Representative illustration of a mid-short-axis image of the left ventricle, with tissue tagging applied at end systole (i), progressing to diastasis (iv). The graph represents the diastolic circumferential strain rate from a representative HIV-positive subject (pink line) and a control subject (green line). Time 0 represents end systole. The dashed line indicates the rate of ventricular relaxation. The summary data show impaired diastolic strain rate in cases versus control subjects. Data are presented as mean ± SE. HAART = highly active antiretroviral therapy; ¹H MRS = hydrogen ion magnetic resonance spectroscopy.

Myocardial triglyceride content was significantly associated with circumferential strain (r = 0.41, p = 0.009) and diastolic strain rate (r = −0.47, p = 0.004). Moreover, HAART exposure was also significantly associated with myocardial triglyceride content (r = 0.42, p = 0.043) and diastolic strain rate (r = −0.43, p = 0.029), both of which were significantly higher with clinical presence of lipodystrophy (p = 0.02 and 0.001, respectively).

The data herein support the hypothesis that ectopic fat deposition in the myocardium may contribute to cardiac disease in patients with HIV infection treated with HAART. In this study, clinically visible lipodystrophy predicted myocardial triglyceride content and LV dysfunction in HIV(+) patients; this may provide an important mechanistic insight into the pathogenesis of heart disease in patients treated for HIV infection. Lipodystrophy can indeed lead to ectopic spillover of triglyceride into non-adipose tissues—including the myocardium—and has previously been associated with adverse LV remodeling (2). Furthermore, that HAART exposure also predicted myocardial triglyceride content and LV dysfunction provides further insight into the pathogenesis of heart disease in HIV(+) patients. For example, HAART has been associated with lipodystrophy, hyperlipidemia, and hyperglycemia (4) and may be directly cardiotoxic (5). Thus, whereas HAART has significantly reduced HIV-related morbidity and mortality, we speculate that HAART may also contribute to the cardiovascular decline in HIV(+) patients. Future studies are warranted to further investigate these findings.