Addition of Thiazolidinedione or Exenatide to Oral Agents in Type 2 Diabetes: A Meta-Analysis

Background: The introduction of several new therapeutic agents for the treatment of type 2 diabetes mellitus has led to significant challenges for providers in deciding which agent to select during the disease course. Objective: To provide a relative comparison of the efficacy and safety of adding thiazolidinediones (TZDs) or exenatide to oral agents for the management of type 2 diabetes mellitus by performing meta-analyses of relevant published studies. Methods: We systematically searched PubMed, MEDLINE, CINHAL, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, EMBASE (inception to March 2008 for all databases), and abstracts presented at the 2006 and 2007 American Diabetes Association conferences to identify all relevant publications. Studies were included in the analysis if they (1) were published in English, (2) were prospective, randomized, and controlled with placebo or comparator, (3) were at least 24 weeks' duration, (4) included nonpregnant adults with type 2 diabetes, (5) were full-text, peer-reviewed articles examining the efficacy of either TZDs (rosiglitazone or pioglitazone) or exenatide in combination with other oral drugs, and (6) included hemoglobin A1C (AIC) outcomes in a manner that allowed data analysis. We evaluated mean change in A1C levels, proportion of subjects reaching A1C goals of less than 7%, mean change in fasting plasma glucose (FPG) and body weight, and the occurrence of nonsevere hypoglycemia and gastrointestinal adverse events. Results: A total of 5212 TZD and 3562 exenatide publications were identified. After critical evaluation, 22 publications met all of the inclusion criteria for the meta-analysis. A1C was reduced from baseline for TZDs (weighted mean difference –0.80%; 95% CI –1.10 to –0.50) and exenaiide (weighted mean difference –0.60%; 95% CI –1.04 to –0.16), Compared with controls, TZD- and exenatide-based therapies had odds ratios greater than 1 for reaching A1C targets of less than 7% (TZD OR 2.27; 95% CI 1.22 to 4.24 and exenatide OR 2.90; 95% CI 1.28 to 6.55). FPG concentrations were reduced significantly from baseline in the TZD-based regimens (weighted mean difference –29.58 mg/dL; 95% CI –39.27 to –19.89), but did not achieve significance in the exenatide trials (weighted mean difference –8.77 mg/dL; 95% CI –28.85 to 11.31). Body weight was reduced with exenatide (weighted mean difference –2.74 kg; 95% CI –4.85 to –0.64) and increased in subgroup analyses for TZDs (weighted mean difference 2.19 kg; 95% CI 1.24 to 3.14). There was no significant association between TZD or exenatide therapy and the risk of nonsevere hypoglycemia. The odds ratios for nausea, vomiting, and diarrhea with exenatide relative to controls were 9.02 (95% CI 3.66 to 22.23), 4.56 (95% CI 3.13 to 6.65), and 2.96 (95% CI 2.05 to 4.26), respectively. Conclusions: TZDs and exenatide have modest but beneficial effects on glycemic control and are relatively safe in regard to the adverse events studied. TZDs produce greater improvement in glycemic control, while exenatide Is associated with reduction in body weight.