Anatomic Fat Depots and Coronary Plaque Among Human Immunodeficiency Virus-Infected and Uninfected Men in the Multicenter AIDS Cohort Study

Abstract

Methods. In a cross-sectional substudy of the Multicenter AIDS Cohort Study, noncontrast cardiac computed tomography (CT) scanning for coronary artery calcium (CAC) scoring was performed on all men, and, for men with normal renal function, coronary CT angiography (CTA) was performed. Associations between fat depots (visceral adipose tissue [VAT], abdominal subcutaneous adipose tissue [aSAT], and thigh subcutaneous adipose tissue [tSAT]) with coronary plaque presence and extent were assessed with logistic and linear regression adjusted for age, race, cardiovascular disease (CVD) risk factors, body mass index (BMI), and human immunodeficiency virus (HIV) parameters. Results. Among HIV-infected men (n = 597) but not HIV-uninfected men (n = 343), having greater VAT was positively associated with noncalcified plaque presence (odds ratio [OR] = 1.04, P < .05), with a significant interaction (P < .05) by HIV serostatus. Human immunodeficiency virus-infected men had lower median aSAT and tSAT and greater median VAT among men with BMI <25 and 25-29.9 kg/m(2). Among HIV-infected men, VAT was positively associated with presence of coronary plaque on CTA after adjustment for CVD risk factors (OR = 1.04, P < .05), but not after additional adjustment for BMI. There was an inverse association between aSAT and extent of total plaque among HIV-infected men, but not among HIV-uninfected men. Lower tSAT was associated with greater CAC and total plaque score extent regardless of HIV serostatus. Conclusions. The presence of greater amounts of VAT and lower SAT may contribute to increased risk for coronary artery disease among HIV-infected persons.

Keywords: adiposity; coronary plaque.

Figure 1.

Adjusted odds ratios of presence of noncalcified plaque score (NCP) >0 per 10 units increase in abdomen visceral fat volume by human immunodeficiency virus (HIV) serostatus from multiple logistic regression. Factors adjusted for in regression models included age, race, body mass index, cumulative pack years of cigarette smoking, use of antihypertension medications, systolic blood pressure among persons who did not use antihypertensives, use of antidiabetic medications, fasting glucose level among persons who did not use antidiabetic medications, use of lipid-lowering medications, high-density lipoprotein, and low-density lipoprotein levels among persons not receiving lipid-lowering medications.