BACKGROUND Because both type 2 diabetes and elevated plasma lipid levels are important independent risk factors for cardiovascular disease and coronary heart disease, the choice of an antihyperglycemic agent for patients with type 2 diabetes--in whom abnormal plasma lipid levels are often seen-should take into account effects on lipids as well as on markers of glycemic control. OBJECTIVE This study assessed the effects on lipid levels of glyburide/metformin tablets in the treatment of type 2 diabetes, particularly in a group of patients who had poor glycemic control and dyslipidemia at baseline. METHODS This 52-week, open-label study was an extension of a 32-week, double-blind, placebo-controlled study. The patient population was drawn from 3 groups: those who completed the double-blind study, those who were discontinued from the double-blind study, and those who were ineligible for the double-blind study based on predefined measures of glycemic control (screening fasting plasma glucose > 240 mg/dL and glycosylated hemoglobin [HbA1c] < or = 12%, or HbA1c 11%-12%) and were directly enrolled in the open-label extension study. Patients with an HbA1c of < 9% received glyburide/ metformin tablets 1.25 mg/250 mg BID; those with an HbA1c > or = 9% received glyburide/ metformin tablets 2.5 mg/500 mg BID. Changes in total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and triglyceride (TG) levels were assessed for 52 weeks. RESULTS The study population included 828 patients: 515 who completed the double-blind study, 138 who were discontinued from the double-blind study, and 175 who were enrolled directly. Direct enrollees had poor glycemic control and dyslipidemia at baseline. Improvements in plasma lipid levels were seen as early as week 13. At week 52, the mean change in TC from baseline was -8.0 mg/dL for the total population (95% CI, -10.9 to -5.2; P < 0.05) and -23.2 mg/dL for direct enrollees (95% CI, -30.1 to -16.4; P < 0.05). The mean decrease in LDL-C from baseline for the total population was 2.86 mg/dL (95% CI, -5.3 to -0.4; P < 0.05), compared with a reduction of 13.3 mg/dL for direct enrollees (95% CI, -18.5 to -8.1; P < 0.05). Mean HDL-C levels were minimally affected. Mean TG levels decreased by 27.8 mg/dL for the entire population (95% CI, -42.9 to -12.8; P < 0.05) and by 99.7 mg/dL for direct enrollees (95% CI, -152.5 to -46.8; P < 0.05). CONCLUSION In this open-label extension study, treatment with glyburide/ metformin tablets for type 2 diabetes had a durable, favorable effect on lipid levels, particularly in those with poor glycemic control and dyslipidemia at baseline.