Sustained effects of pioglitazone vs. glibenclamide on insulin sensitivity, glycaemic control, and lipid profiles in patients with Type 2 diabetes

Aims This study compared the effects of 52 weeks' treatment with pioglitazone, a thiazolidinedione that reduces insulin resistance, and glibenclamide, on insulin sensitivity, glycaemic control, and lipids in patients with Type 2 diabetes. Methods Patients with Type 2 diabetes were randomized to receive either pioglitazone (initially 30 mg QD, n = 91) or micronized glibenclamide (initially 1.75 mg QD, n = 109) as monotherapy. Doses were titrated (to 45 mg for pioglitazone and 10.5 mg for glibenclamide) to achieve glycaemic targets during the next 12 weeks: fasting blood glucose of less than or equal to 7 mmol/l and 1-h postprandial blood glucose of less than or equal to 10 mmol/l. Patients were maintained on the titrated dose for 40 weeks. Results Pioglitazone significantly increased insulin sensitivity compared with glibenclamide, as assessed by homeostasis model assessment (17.0% vs. -13.0%; P < 0.001), quantitative insulin sensitivity check index (0.011 vs. -0.007; P < 0.001) and fasting serum insulin (-1.3 pmol/l vs. 23.8 pmol/l; P = 0.007). The glibenclamide group had significantly lower HbA(1c) than the pioglitazone group after 12 weeks of therapy (7.8% vs. 8.3%, P = 0.015), but significantly higher HbA(1c) after 52 weeks of therapy (7.8% vs. 7.2%, P = 0.001). Pioglitazone significantly (vs. glibenclamide) increased mean HDL-C (P < 0.001), decreased mean triglycerides (P = 0.019), and decreased mean atherogenic index of plasma (AIP; P = 0.001) and mean total cholesterol/HDL-C (P = 0.004), without significantly elevating mean total cholesterol or mean LDL-C compared with glibenclamide. Conclusions These data suggest that the effects of pioglitazone are more sustained than those of glibenclamide for improving insulin sensitivity in patients with Type 2 diabetes, and that 52 weeks' treatment with pioglitazone has favourable effects on glycaemic control and lipoprotein profile.