Background:  Individual components of the metabolic syndrome have been described among HIV-infected HAART recipients and include abdominal adiposity, hypertriglyceridemia (HTG), low serum HDL, fasting hyperglycemia (HG), and hypertension. However, rates of metabolic syndrome among HIV-infected persons have not been well-described.

Methods:  We evaluated metabolic syndrome among HIV seropositive and seronegative men in the Multicenter AIDS Cohort Study (MACS) seen from 1999 to 2004. Metabolic syndrome was defined as ≥3 of:  waist circumference >102 cm, fasting glucose (>100 mg/dL), serum TG (>150 mg/dL), HDL (<40 mg/dL), blood pressure >130 systolic or >85 diastolic (HTN). We also evaluated, in multivariate regression models, associations with ART received, demographics, and stage of HIV disease.

Results:  Among 645 HIV seropositive and 398 seronegative men, metabolic syndrome prevalence was consistently higher among HIV seropositive (20 to 33%) than seronegative men (19 to 27%) (OR = 1.43, p = 0.002). Likewise, HIV seropositive were more likely to have HTG (OR = 2.43, p <0.001), low HDL (OR = 3.09, p <0.001), and HG (OR = 1.29, p = 0.009) than HIV seronegative. Seropositives were less likely to have elevated waist circumference than seronegatives (OR = 0.38, p <0.001.) There were no significant differences in HTN rates. Those with CD4⁺ lymphocyte counts/cm³ plasma (CD4) of 200 to 350 were less likely than those with CD4 <200 to have metabolic syndrome (OR = 0.39, p = 0.026)). Alcohol use (≥1 drink/day) was associated with decreased likelihood of metabolic syndrome (OR = 0.71, p = 0.016). Advancing age conferred a greater risk for metabolic syndrome (OR = 1.08 per year, p <0.0001). HAART use in general was associated with an increased metabolic syndrome risk (OR = 1.18 per year of use, p = 0.012), as was protease inhibitor (PI) use (OR = 1.17 per year of use, p = 0.039). Drug factors associated with:  HTG, were PI use (OR = 1.30 per year, p <0.001), especially lopinavir/ritonavir (LPV/rtv) (OR = 1.70 per year, p = 0.020) and RTV (OR = 1.17, p = 0.0.28); non-nucleoside reverse transcriptase inhibitor (NNRTI) use (OR = 1.21 per year, p = 0.005), especially efavirenz (EFV) (OR = 1.38, p <0.001); and NRTI use (OR = 1.07 per year, p = 0.044), especially stavudine (d4T) (OR = 1.12 per year, p = 0.010); low HDL, none. NNRTI use was protective (OR = 0.85 per year, p = 0.012); increased waist circumference, none. NRTI use was protective (OR = 0.88 per year, p = 0.007), especially d4T (OR = 0.82, p = 0.001); HG, RTV use (OR = 0.82, p = 0.010.)

Conclusions:  Metabolic syndrome is more common in HIV seropositive than seronegative men. Only increased waist circumference is less likely in HIV seropositives, and this is most associated with d4T use. Lower CD4 cell counts, advancing age, and HAART use (especially PI) are also associated with greater metabolic syndrome risk. The HAART-associated risk is progressive over time. Alcohol use appears to be protective. All HAART drug classes are assoc with HTG.