Continuation of metformin after introduction of insulin in type 2 diabetes.

Metformin is the cornerstone of treatment early in the course of type 2 diabetes. Recent studies also provide evidence for the benefits of metformin when given late in the course of the disease, after the introduction of insulin.1 2 However, because metformin targets insulin resistance in overweight patients, can it be as beneficial in non-obese patients? In the linked randomised controlled trial, Lund and colleagues compare the effects of metformin versus repaglinide in non-obese patients with type 2 diabetes treated with insulin.3 They randomised 102 patients (body mass index ≤27 and glycated haemoglobin ≥6.5), after a run-in period with combined repaglinide and metformin, to receive either repaglinide 6 mg plus insulin or metformin 2000 mg plus insulin. Patients had been known to have diabetes for about 10 years. After 12 months, they found no significant difference between the groups in the primary outcome of glycaemic control—glycated haemoglobin decreased from 8.15 to 6.71 in the metformin plus insulin group and from 8.07 to 6.90 in the repaglinide plus insulin group (P=0.125). However, after excluding patients (n=7) with a positive glutamic acid decarboxylase-65 antibody status (a sign of autoimmunity), they found a small but significant difference in mean glycated haemoglobin in favour of the metformin plus insulin group (−0.29, 95% confidence interval −0.56 to –0.02; P=0.034). The daily dose of insulin was similar in both groups, as well as the overall risk of hypoglycaemia. However, metformin (versus repaglinide) significantly prevented weight gain (a secondary outcome) during insulin therapy (weight difference of −2.51 kg, −4.10 to −0.95; P=0.002). This finding is particularly important because repaglinide, the comparator in this study, is a short acting insulin secretagogue associated with a favourable metabolic profile and a slight to moderate weight gain during monotherapy (compared with other sulfonylureas), as shown in a Cochrane review.4 Both regimens seemed to achieve tight glycaemic control in non-obese patients, but what about cardiovascular outcomes? No placebo controlled intervention trials exist with (conclusive) data on the long term effects of repaglinide on cardiovascular outcomes in patients with type 2 diabetes.4 Prevention of weight gain by metformin in overweight patients with type 2 diabetes who are taking insulin may reduce the risk of cardiovascular disease. In a long term placebo controlled trial, about 40% of the favourable effects of metformin on macrovascular disease was explained by reduction in weight, and about 35% was probably caused by an intrinsic effect of metformin on endothelial functions, partly independent of glycaemic control.2 5 Other studies support this last mechanism of action.6 7 In overweight patients with type 2 diabetes and insulin resistance, metformin affects glucose metabolism by improving the responsiveness of the liver to insulin.8 Interestingly, in non-obese patients with type 2 diabetes, in whom insulin resistance is not pronounced, metformin lowers glucose as effectively as insulin secretagogues.3 9 10 Could metformin affect glycaemic control independently of the extent of insulin resistance? New evidence shows that metformin might act as a secretagogue of glucagon-like peptide type 1 (GLP-1), a gut hormone that acts on insulin release from the β cell and on glucagon release from the α cell to normalise blood glucose, in a glucose dependent manner.11 This additional effect of metformin may contribute to better glycaemic control in non-obese patients with type 2 diabetes, but this needs further investigation. What are the implications of these findings for general practice? The prevention of weight gain by metformin (versus repaglinide) in non-obese patients with type 2 diabetes who are taking insulin is promising3 because an increase in body mass index from 25 to 27 is known to increase the risk of cardiovascular disease in the presence of type 2 diabetes.12 Continuation of metformin after the introduction of insulin in nonobese patients with type 2 diabetes may therefore not only reduce weight but have beneficial cardiovascular effects in the longer term. If metformin is contraindicated, repaglinide might be a reasonable alternative, at least for one year.