Dyslipidemia is common in patients with HIV infection. In this study, a two-stage euglycemic hyperinsulinemic clamp, with infusion of stable isotopically labeled tracers, was used to evaluate insulin action in skeletal muscle, liver, and adipose tissue in HIV-infected men with dyslipidemia (HIV-DL; plasma triglyceride >250 mg/dl and HDL <45 mg/dl; n=12), HIV-infected men without dyslipidemia (HIV w/o DL; n=12), and healthy men (n=6). Basal rates of glucose production (glucose R(a)), glucose disposal (glucose R(d)), and lipolysis (palmitate R(a)) were similar between groups. The relative suppression of glucose R(a) (63+/- 4, 77+/- 2, and 78+/- 3%, P=0.008) and palmitate R(a) (49+/-4, 63+/-3, and 68+/-3%, P=0.005) during ow-dose insulin infusion (plasma insulin approximately 30 microU/ml), and the relative stimulation of glucose R(d) (214+/-21, 390+/-25, and 393+/-46%, P=0.001) during high-dose insulin infusion (plasma insulin approximately 75 microU/ml) were lower in HIV-DL than in HIV w/o DL and healthy volunteers, respectively. Suppression of basal glucose R(a) correlated with plasma adiponectin (r=0.44, P=0.02) and inversely with plasma IL-6 (r=-0.49, P<0.001). Stimulation of glucose R(d) correlated directly with adiponectin (r=0.48, P<0.01) and inversely with IL-6 (r=-0.49, P=0.02). We conclude that dyslipidemia in HIV-infected men is indicative of multiorgan insulin resistance, and circulating adipokines may be important in the pathogenesis of impaired insulin action.