Abstract
AIM OF THE STUDY: Metformin is recommended as the first-line treatment of type 2 diabetes mellitus. Despite its common use, few studies have been conducted to precisely measure the efficacy of metformin versus placebo as a first-line treatment. This study aims to assess the precise effects of metformin monotherapy on glycemic control and weight in drug-naive patients with type 2 diabetes mellitus. METHODS: Medline\textregistered and Cochrane databases were searched until March 19, 2016 to perform a systematic review and meta-analysis of placebo-controlled randomized trials evaluating metformin monotherapy in drug-naive patients with type 2 diabetes mellitus. Assessed outcomes include glycemic control (fasting plasma glucose, glycosated hemoglobin) and weight. RESULTS: Overall, 16~studies (1140~patients) were selected. Compared to placebo, metformin monotherapy was associated with decreased glycosated hemoglobin by 0.95% at 3~months (95% CI: 0.50 to 1.39, I2=87%) and 1.32% at 6~months (95% CI: 1.01 to 1.62, I2=71%), and decreased fasting plasma glucose by 1.92mmol/L at 1 month (95% CI: 0.11 to 3.74, I2=88%), 1.79mmol/L at 3~months (95% CI: 0.92 to 2.66, I2=88%) and 2.14mmol/L at 6~months (95% CI: 1.17 to 3.12, I2=82%). No significant difference was demonstrated for the comparisons of weight due to relatively small number of studies retrieved from the literature resulting in insufficient statistical power. CONCLUSION: This study provides the precise effects of metformin monotherapy regarding the decreases in fasting plasma glucose and glycosated hemoglobin that physician can expected in drug-naive patients with type 2~diabetes mellitus. No evidence was found for the effects on weight.