Background: Insulin resistance is a critical aspect of the pathophysiology of type 2 diabetes mellitus and is also associated with other risk factors for cardiovascular disease (eg, dyslipidemia and hypertension). Accordingly, insulin resistance is a possible target for lowering plasma glucose concentration and preventing diabetic macroangiopathy. Biguanides, such as metformin, and thiazolidinediones (TZDs), such as pioglitazone, improve insulin resistance.
Objectives: The aims of this study were to assess the effects of replacing a biguanide with a TZD on glycemic control in patients with poorly controlled type 2 diabetes mellitus, and also to identify the factors affecting interpatient variation in the effects of treatment change.
Methods: This was a 12-week, open-label, prospective study in which previously prescribed metformin (500 or 750 mg/d) was replaced with pioglitazone (15 or 30 mg/d) in patients with poorly controlled type 2 diabetes mellitus. Patients with a glycosylated hemoglobin (HbA1c) concentration >7% despite treatment with diet, exercise, and hypoglycemic agents other than TZDs were eligible for the study. Patients who never received TZDs were also eligible for inclusion. Vital signs, metabolic parameters, and arterial stiffness were assessed at baseline and after 12 weeks of treatment with pioglitazone. The primary end point was change in HbA1c concentration after replacing metformin with pioglitazone. Tolerability was assessed by medical history, physical examination, and laboratory tests (aspartate aminotransferase, alanine aminotransferase, and γ-glutamyl transpeptidase).
Results: Twenty-one Japanese patients (15 women, 6 men; mean [SD] age, 61.8 [8.4] years; body mass index, 25.5 [3.0] kg/m2) were included in the study. HbA1c concentration was not significantly changed from baseline after 12 weeks of pioglitazone treatment (8.0% [0.7%] vs 8.2% [0.7%]). Fasting plasma glucose (FPG) concentration also was not significantly changed after the replacement of treatment (156  vs 144  mg/dL). In addition, the resistin concentration did not change significantly from baseline after 12 weeks of pioglitazone treatment (6.6 [3.8] vs 6.4 [3.6] ng/mL). In contrast, significant improvement from baseline was observed in triglyceride (TG) concentrations (157  vs 117  mg/dL; P = 0.003), high-density lipoprotein cholesterol (HDL-C) (55  vs 61  mg/dL; P = 0.016), remnant-like particle cholesterol (6.6 [6.0] vs 5.3 [3.5] mg/dL; P = 0.048), and serum adiponectin (8.8 [4.3] vs 23.3 [11.7] μg/mL; P < 0.001). Pulse wave velocity was also significantly improved (1730  vs 1622  m/sec; P = 0.009). Changes in HbA1c were significantly correlated with serum fasting insulin concentration at baseline in the patients not receiving insulin preparations (r = -0.635, P = 0.013). The percentage change in serum adiponectin concentration was correlated with the percentage changes in HbA1c and FPG concentrations (HbA1c, r = -0.518, P = 0.019; FPG, r = -0.594, P = 0.006). Body weight was significantly increased after treatment (62.6 [11.9] vs 65.5 [12.2] kg; P < 0.001). Mild edema was reported in 5 patients. One patient discontinued treatment due to an increase in serum creatine kinase activity to ~6.6 times the upper limit of normal.
Conclusions: Replacement of metformin with pioglitazone did not produce significant differences in HbA1c and FPG concentrations from baseline after 12 weeks of treatment in these patients with poorly controlled type 2 diabetes mellitus. However, the replacement was effective in a subset of patients whose serum insulin concentrations were high or whose serum adiponectin concentrations were sensitive to TZDs. In addition, the replacement was associated with significant improvements in TG, HDL-C, serum adiponectin concentration, pulse wave velocity, and body weight increase from baseline.