Psoas muscle attenuation measurement with computed tomography indicates intramuscular fat accumulation in patients with the HIV-lipodystrophy syndrome

The human immunodeficiency virus (HIV)-lipodystrophy syndrome is characterized by abnormalities of lipid metabolism, glucose homeostasis, and fat distribution. Overaccumulation of intramuscular lipid may contribute to insulin resistance in this population. We examined 63 men: HIV positive with lipodystrophy (n = 22), HIV positive without lipodystrophy (n = 20), and age- and body mass index-matched HIV-negative controls (n = 21). Single-slice computed tomography was used to determine psoas muscle attenuation and visceral fat area. Plasma free fatty acids (FFA), lipid profile, and markers of glucose homeostasis were measured. Muscle attenuation was significantly decreased in subjects with lipodystrophy [median (interquartile range), 55.0 (51.0-58.3)] compared with subjects without lipodystrophy [57.0 ( 55.0-59.0); P = 0.05] and HIV-negative controls [59.5 (57.3-64.8); P < 0.01]. Among HIV-infected subjects, muscle attenuation correlated significantly with FFA (r = -0.38; P = 0.02), visceral fat (r = -0.49; P = 0.002), glucose (r = -0.38; P = 0.02) and insulin (r = -0.60; P = 0.0001) response to a 75-g oral glucose tolerance test. In forward stepwise regression analysis with psoas attenuation as the dependent variable, visceral fat (P = 0.02) and FFA (P < 0.05), but neither body mass index, subcutaneous fat, nor antiretroviral use, were strong independent predictors of muscle attenuation (r(2) = 0.39 for model). Muscle attenuation (P = 0.02) and visceral fat (P = 0.02), but not BMI, subcutaneous fat, FFA, or antiretroviral use, were strong independent predictors of insulin response ( area under the curve) to glucose challenge (r(2) = 0.47 for model). These data demonstrate that decreased psoas muscle attenuation due to intramuscular fat accumulation may contribute significantly to hyperinsulinemia and insulin resistance in HIV-lipodystrophy patients. Further studies are needed to assess the mechanisms and consequences of intramuscular lipid accumulation in HIV-infected patients.