Fifty nondiabetic patients who met a modified National Cholesterol Education Program definition for the metabolic syndrome were randomized to receive either rosiglitazone (4 mg/day; n = 25) or placebo (n = 25) for 8 weeks. Compared with those receiving placebo, patients in the rosiglitazone group achieved significant reductions in fasting plasma insulin levels (−40%), homeostasis model assessment indexes (−45%), systolic and diastolic blood pressures, and high-sensitivity C-reactive protein levels (−31%). There were no changes in fasting plasma glucose with either treatment. Although rosiglitazone treatment greatly increased plasma levels of low-density lipoprotein cholesterol (18%) and apolipoprotein B (16%), it significantly improved both endothelium-dependent flow-mediated vasodilation (p <0.001) and endothelium-independent nitroglycerin-induced vasodilation (p = 0.01) of the right brachial artery.