Printable Version

Home > Abstracts of Current Literature > Pioglitazone vs. Metformin in Type 2 Diabetes

Efficacy and Safety of Pioglitazone Versus Metformin in Patients with Type 2 Diabetes Mellitus: A Double-Blind, Randomized Trial

Background: Pioglitazone’s mechanism of action is to decrease insulin resistance in peripheral tissues while metformin decreases hepatic glucose production with a secondary effect on insulin resistance for patients with type 2 diabetes mellitus. Currently metformin is used as an initial treatment while pioglitazone is usually reserved for adjunctive treatment.

Objective: The purpose of this study was to determine if pioglitazone was as effective as metformin for the initial treatment of type 2 diabetes mellitus.

Methods: This was a prospective, randomized, double blind study. It was a one-year study that included patients from 167 centers in 12 European countries. Patients with type 2 diabetes were included in the study if they were men and women between the ages of 35 and 75, with a hemaglobin A1c (HbA1c) between 7.5 and 11 percent with stable or worsening glycemic control over the past three months. The patients also had to be considered to be inadequately controlled with diet alone. Patients were excluded from the study if they previously had taken any oral diabetes medication or had contraindications to the study medications. Medications known to effect glucose control were permitted provided the patients had been maintained on those medications for at least four weeks prior to the study. The eleven-hundred ninety-four patients enrolled were either given pioglitazone 30 mg daily (597 patients) with placebo or metformin 850 mg (597 patients) up to three times daily with placebo. The doses were adjusted every 4 weeks for the first 12 weeks of the study. At the end of the 12-week titration period, the dosage was kept the same for the remainder of the study (40 weeks). The primary outcome measure for this study was change in HbA1c. This study used a 0.2% difference in HbA1c as the limit of non-inferiority between treatment groups. The secondary outcomes measured included fasting plasma glucose (FPG), insulin levels, lipid panel (HDL, LDL, TC, TG, FFA), adverse events, lab tests, blood pressure and weight gain. This study had a power of 90% and followed the intent to treat approach; however it also reported results as per protocol to give a true comparison of efficacy between the treatment regimens.

Results: A total of one-thousand patients completed this study leaving one-hundred ninety-four patients having dropped out due to various factors including adverse events, lack of efficacy, and protocol violation. Ninety-eight of the dropouts were from the pioglitazone treatment group compared to ninety-six from the metformin group. The HbA1c endpoint showed a decrease of –1.41% (± 0.4) from baseline in the pioglitazone group compared to –1.50% (± 0.4) in the metformin group (90% confidence interval; -0.01 to –0.19). The upper limit confidence interval (0.19) falls within the predetermined 0.2% limit of noninferiority. This test proves pioglitazone’s noninferiority compared to metformin in HbA1C reduction. The FPG decreased by 45 mg/dl (± 16.2) for the pioglitazone group compared to 39.6 mg/dl (± 16.2) for the metformin group. The mean estimated treatment difference in FPG favored pioglitazone (-5.4 mg/dl; P = 0.016). Fasting insulin levels decreased by 0.3 :U/ml in the metformin group compared to –2.4 :U/ml for the pioglitazone group which was statistically significant in favor of pioglitazone (P <0.0001). The OGTT results showed a decrease in AUC0-3h in plasma glucose of 90.1 mg/h×dl for the pioglitazone group compared to 41.4 mg/h×dl in the metformin group (P<0.001). This test also showed the AUC0-3h for insulin increased by 3.1:IU/h×ml compared to 33.2 :IU/h×ml for the metformin group (P= 0.049). Lipid profiles showed a decrease in HDL and TG but an increase in LDL for the pioglitazone group compared to metformin (p = 0.001). The urinary albumin to creatinine ratio decreased by 19% with pioglitazone compared to 1% for metformin (P = 0.002). The authors discuss how the adverse effects seen in the study were different among the treatment groups with edema being most common with pioglitazone versus gastrointestinal effects found with metformin. The authors conclusion of this study was that pioglitazone is not inferior to metformin for diabetes control and that both medications have beneficial effects on FPG and varying effects on lipids.

Strengths: Strengths in this study include the use of a large number of patients, comparing two relevant medications for type 2 diabetes, and adequate measures of comparison between the groups (HbA1c, FPG, lipid levels, adverse events and laboratory tests). Another strength of the study was that it used one central laboratory for the HbA1c measurements. The use of dietary counseling in this study was also a strength to the study design.

Weaknesses: Weaknesses of this study include the use of 167 centers in 12 European countries due to the possibility of intersite variability. Another weakness of the study was that some patients were included in the study even though they didn’t meet the inclusion criteria (based on HbA1c). This study also didn’t thoroughly discuss the dosing adjustments in the study, the definition of “tolerating” and who made the decision for the adjustments. Another weakness of this study was that compliance to the study medications was not defined adequately. The study did mention that 80% of the patients took the medication for the duration of the study; however, they did not explain how this was accounted for such as with pill count, etc.

Conclusion: This study showed that pioglitazone is as effective as metformin in controlling hemaglobin A1c in patients naive to pharmacologic treatment for their type 2 diabetes. This is important from a clinical standpoint because it gives clinicians another alternative from metformin to start the patient on. Since both drugs have different side effect profiles, it will also allow a patient to switch medications if not tolerated. The findings regarding weight gain and lipid effects should be taken into account when deciding which medication to initiate. For example: If weight gain is contraindicated for a patient, it may be advised to start the patient on metformin rather than pioglitazone due to associated weight gain. Due to pioglitazone not being available as a generic, the cost of the medication compared to generic metformin may be significant for some patients.

Schernthaner G., Matthews DR, Charbonnel B, Hanefeld M, and Brunetti P. Efficacy and safety of pioglitazone versus metformin in patients with type 2 diabetes mellitus: a double blind randomized trial. J Clin Endocrinol Metab. 2004 Dec;89(12):6068-76.

Doug Donovall, Pharm.D. Candidate