Continuing metformin when starting insulin in patients with Type 2 diabetes: a double‐blind randomized placebo‐controlled trial

Aims  To test the effect of continuing metformin on weight gain and glycaemic control in patients with poorly controlled Type 2 diabetes who need to start insulin. Methods  Patients with Type 2 diabetes on maximum tolerated oral agents referred for insulin conversion were recruited from hospital diabetes clinics into a double‐blind randomized placebo‐controlled trial. The 183 participants received metformin or placebo, titrated up to 2 g a day or maximum tolerated dose, with insulin started according to local practice. The main outcome measures were weight change over 12 months, HbA 1c , insulin dose, frequency of hypoglycaemia, treatment satisfaction, and well‐being. Results  Over 12 months, metformin was associated with less weight gain than placebo [mean 6.1 kg vs. 7.6 kg; adjusted difference 1.5 kg (95% confidence interval 0.2–2.9); P  = 0.02], a greater reduction in HbA 1c [1.5% vs. 1.3%; adjusted difference 0.5% (0.1–0.9); P  = 0.02] and a lower insulin requirement [62 units vs. 86; adjusted difference 25 units (15–34); P  < 0.001], but also more hypoglycaemia [relative risk of any episode 1.24 (1.02–1.1); P  = 0.03]. Treatment satisfaction improved more in patients on metformin than on placebo ( P <  0.001), as did the positive well‐being score ( P =  0.02). Conclusions  Metformin decreases weight gain, lowers insulin requirement, and improves glycaemic control, and should be continued in patients with Type 2 diabetes who transfer to insulin.