Adiponectin appears to be an important modulator for metabolic and vascular diseases. A case-controlled study was designed to measure plasma adiponectin levels and investigate the effects of rosiglitazone on adiponectin levels in type 2 diabetic patients with proteinuria. Sixty-four patients (mean age, 46.1+/-4.6 yr; 30 male, 34 female) and 26 healthy volunteers (mean age, 45.3+/-4.8 yr; 14 male, 12 female) were included. Patients with proteinuria were treated with 4-mg/d rosiglitazone (n=21, 10 males, 11 females) for 4 wk. Adiponectin levels in patients were significantly lower than those of controls (p<0.001). There were significant negative correlations between adiponectin concentrations and insulin levels as well as homeostasis model assessment (HOMA) index in patient's group (r=-<FONT "Optima">0.538, p<0.001; r=-<FONT "Optima">0.393, p=0.001, respectively). There was also a significant negative correlation between plasma adiponectin concentrations and the degree of proteinuria (r=-<FONT "Optima">0.526, p=0.002). Plasma adiponectin levels in patients with proteinuria (n=31; 3.91+/-2.57 microg/mL) were significantly lower than those without proteinuria (n=33; 10.15+/-1.97 microg/mL) (p<0.001). After the treatment period, adiponectin levels significantly increased (p<0.001) and proteinuria, plasma insulin, and HOMA indexes significantly decreased in treatment group (p<0.001, p<0.001, p<0.001, respectively). The results suggest that adiponectin is inversely correlated with proteinuria and treatment with peroxisome proliferator-activated receptor-gamma (PPAR-gamma) agonist rosiglitazone both corrects proteinuria and increases the low adiponectin levels in diabetic patients.