Platelet-derived microparticles (PDMP), selectins, and adiponectin play an important role in the development of atherosclerosis in diabetes. Acarbose has been shown to have a beneficial effect on postprandial hyperglycemia in diabetic patients. However, its influence on PDMP, selectins, and adiponectin in these patients is poorly understood. We investigated the effect of acarbose on circulating levels of PDMP, selectins, and adiponectin in patients with type 2 diabetes. Acarbose (300 mg/day) was administered for 3 months. Levels of PDMP, sP-selectin, sL-selectin, and adiponectin were measured by ELISA at baseline and after 1 and 3 months of treatment. The levels of PDMP, sP-selectin, and sL-selectin were higher in diabetic patients than in hypertensive patients (PDMP; 35.1 +/- 34.2 vs. 53.3 +/- 56.7 U/ml, P < 0.05: sP-selectin; 134 +/- 52 vs. 235 +/- 70 ng/dl, P < 0.01: sL-selectin; 569 +/- 183 vs. 805 +/- 146 ng/ml, P < 0.05), while there were no significant differences between hypertensive and hyperlipidemic patients. Before acarbose treatment, the adiponectin level of diabetic patients was lower than that of hypertensive patients. Acarbose therapy significantly decreased the plasma PDMP level relative to baseline. Acarbose also caused a significant decrease of sP-selectin and sL-selectin. On the other hand, acarbose therapy led to a significant increase of adiponectin after 3 months of administration compared with baseline (adiponectin: diabetes versus hypertension, 3.61 +/- 1.23 vs. 5.87 +/- 1.92 microg/ml, P < 0.05; diabetes versus controls, 2.81 +/- 0.95 vs. 6.13 +/- 1.24 microg/ml, P < 0.01). Twelve of the 30 diabetic patients had a history of thrombotic complications. Furthermore, the reduction of PDMP and selectins during acarbose therapy was significantly greater in the thrombotic group (12 of 30) than in the nonthrombotic group (18 of 30) of diabetic patients. Acarbose may be beneficial for primary prevention of atherothrombosis in patients with type 2 diabetes. However, it requires a large clinical trial to test this hypothesis.