Metformin as firstline treatment for type 2 diabetes: are we sure?

Remy Boussageon and colleagues ask whether metformin is bringing practical benefit to patients and question the focus on surrogate measures Metformin is recommended as the first glucose lowering treatment for people with type 2 diabetes.(1) The recommendation is based on the supposedly conclusive results of the UK Prospective Diabetes Study (UKPDS 34) published in 1998.(2) The study found a reduction in 10 year mortality from any cause (relative risk 0.64, 95% confidence interval 0.45 to 0.91), and myocardial infarction (0.61, 95% CI 0.41 to 0.89). The number needed to treat to avoid one death was 14 and the absolute risk reduction was 0.07. However, these impressive results were obtained in a randomised subgroup of obese patients (342 patients in the metformin group and 411 in the conventional group) and have never been reproduced.(3) From a scientific point of view, the reproducibility of results is an essential validity criterion. Meta-analyses of randomised controlled trials evaluating the effectiveness of metformin in patients with type 2 diabetes found that metformin did not significantly modify clinically relevant outcomes (table 1 double down arrow).(4 5) The analysis of all types of trial shows no efficacy of metformin at all.