We evaluated the effects of rosiglitazone (4 mg b.i.d.) and metformin (1 g b.i.d.) monotherapy for 26 weeks on adipose tissue insulin-stimulated glucose uptake in patients (n = 41) with type 2 diabetes. Before and after the treatment, glucose uptake was measured using 2-[¹⁸F]fluoro-2-deoxyglucose and positron emission tomography and adipose tissue masses were quantified using magnetic resonance imaging. Rosiglitazone improved insulin-stimulated whole-body glucose uptake by 44% (P < 0.01 vs. placebo). Mean body weight was unchanged in the rosiglitazone group, while it decreased by 2.0 kg in the metformin group (P < 0.05 vs. placebo). In visceral adipose tissue, glucose uptake increased by 29% (from 17.8 ± 2.0 to 23.0 ± 2.6 μmol · kg⁻¹ · min⁻¹, P < 0.05 vs. placebo) in the rosiglitazone group but to a lesser extent (17%) in the metformin group (from 16.2 ± 1.5 to 18.9 ± 1.7 μmol · kg⁻¹ · min⁻¹, P < 0.05 vs. baseline). Because the visceral adipose tissue mass simultaneously decreased with both treatments (P < 0.05), no change was observed in total visceral glucose uptake per depot. Rosiglitazone significantly enhanced glucose uptake in the femoral subcutaneous area, either when expressed per tissue mass (from 10.8 ± 1.2 to 17.1 ± 1.7 μmol · kg⁻¹ · min⁻¹, P < 0.01 vs. placebo) or per whole-fat depot (P < 0.05 vs. placebo). In conclusion, metformin treatment resulted in improvement of glycemic control without enhancement of peripheral insulin sensitivity. The improved insulin sensitivity of the nonabdominal subcutaneous adipose tissue during treatment with rosiglitazone partly explains the enhanced whole-body insulin sensitivity and underlies the central role of adipose tissue for action of peroxisome proliferator-activated receptor γ agonist in vivo.