Six-month effectiveness and tolerability of pioglitazone in combination with sulfonylureas or metformin for the treatment of type 2 diabetes mellitus

BACKGROUND AND OBJECTIVE: Pioglitazone has been reported to improve common cardiovascular risk factors in addition to glycemic control in patients with type 2 diabetes mellitus (T2DM). The changes in cardiovascular risk profile were evaluated comparatively in large cohorts either treated g er rea e or not with pioglitazone-containing combinations in the current clinical setting within Spain. PATIENTS AND METHOD: A nationwide prospective, controlled, observational cohort clinical study was performed in 2294 patients with T2DM who started at the criterion of the treating physician, oral antihyperglycemic treatment with either pioglitazone plus a sulfonylurea (Pio+SU; n = 851) pioglitazone, plus metformin (Pio+Met; n = 723) or a sulfonylurea plus metformin (SU+Met; n = 720) due to inadequate control with previous therapy. Serum cholesterol, blood glucose, hemoglobin A(1C), blood pressure and certain anthropometric parameters were measured at baseline and after 6 months of treatment. RESULTS: Serum high density lipoprotein-cholesterol increased in average (mg/dl) 2.08 with Pio+SU 2.06 with Pio+Met and 0.67 with SU+Met; while triglycerides decreased (mg/dl) 26.6 30.6 and 17.6 in the same cohorts. Inter-group differences were significant (p < 0.001 in both parameters). Total cholesterol decreased significantly more with SU+Met than in the pioglitazone cohorts M fasting plasma glucose and hemoglobin A(1C) reductions were significantly greater in the pioglitazone cohorts than in the SU+Met cohort: 27.74, 28.94 and 23.46 mg/dl (p = 0.012); and 0.80 0.87 and, 0.71% (p = 0.016) with Pio+SU, Pio+Met and SU+Met, respectively. Slight, but significant variations of body weight were also registered in the Pio+SU (+1.4 kg) and SU+Met (-0.7 k) groups. CONCLUSIONS: Treatment with pioglitazone was associated with significant improvements of lipid and glycemic parameters that are linked to insulin resistance and cardiovascular risk in patients with T2DM in their routine clinical care. The non-randomised allocation of patients to treatments, inherent to its observational design, is an important limitation of the present study.