Aim: The aim of our double-blind, placebo-controlled study was to compare the effect of acarbose and glibenclamide on the insulin sensitivity in type 2 diabetes.
Methods: We investigated 77 patients (mean age 58.7 years, mean BMI 27.3 kg/m2), treated by diet alone for at least 4 weeks. The subjects were randomized into three treatment groups for 16 weeks: 100 mg t.i.d. acarbose (n = 25) or 1 mg t.i.d. glibenclamide (n = 27) or one t.i.d. placebo (n = 25). Before and after therapy, the levels of fasting plasma glucose, glycosylated haemoglobin, fasting insulin, plasma glucose and insulin 1 h after a standardized breakfast were measured and insulin sensitivity determined by euglycaemic hyperinsulinaemic clamp test.
Results: After the treatment period, BMI in the acarbose and placebo group decreased significantly, whereas in the glibenclamide group a significant increase was observed. Fasting plasma glucose was only significant reduced under glibenclamide. The postprandial glucose decreased significantly after acarbose (13.8 vs. 11.4 mmol/l, p < 0.05) and glibenclamide treatment (14.6 vs. 11.4 mmol/l, p < 0.05) and was unchanged under placebo (13.8 vs. 13.7 mmol/l). The fasting insulin levels remained unchanged in all three groups, whereas postprandial insulin values increased significantly under glibenclamide. Neither acarbose nor glibenclamide significantly changed insulin sensitivity [acarbose: glucose disposal rate before treatment 2.3 mg/kg body weight/min/insulin, after treatment 3.2; glibenclamide 2.2 vs. 2.1; placebo 2.6 vs. 3.0].
Conclusions: Our results show a more substantial improvement of glucose control under glibenclamide than under acarbose which, however, was not associated with an increase of insulin sensitivity.