Differences between long-acting insulins for the treatment of type 2 diabetes.

IMPORTANCE OF THE FIELD Most guidelines suggest that failure of oral antidiabetic drugs should be followed by the addition of a basal insulin with aggressive titration of the dose. In most countries, neutral protamine Hagedorn (NPH)-insulin, glargine and detemir are the only choices. Clinical trials show that the metabolism and metabolic outcomes after treatment with intermediate- or long-acting insulins differ little. Despite this, the hypoglycaemic potential, effect on body weight and adherence to insulin treatment may affect the choice of basal insulin. Adherence seems to be negatively correlated to the prescribed dose and the number of injections. Furthermore, the choice of basal insulin might be influenced by the number of units necessary to achieve the goal for HbA1c. AREAS COVERED IN THIS REVIEW By searching the literature systematically, we identified all randomised clinical trials comparing long-acting insulins (human NPH-insulin and the analogues glargine and detemir) for the treatment of type 2 diabetes conducted over the last 10 years. We continued by reviewing only studies in which similar antihyperglycaemic potential of the treatments was achieved. WHAT THE READER WILL GAIN According to the inclusion criteria for this review, all drugs were efficacious regarding the main purpose of decreasing glycaemia. For an equal efficacy, we were able to detect other differences between the treatments and, furthermore, an estimate on the number of units of insulin needed to achieve comparable glycaemic control. TAKE HOME MESSAGE The analysis confirmed a favourable profile of both analogues regarding hypoglycaemia. For detemir, we additionally identified a favourable profile regarding weight gain and need for an increased number of units of insulin to achieve comparable glycosylated haemoglobin (HbA1c) responses. We conclude that the efficacy of insulin treatment seems to vary little between the available products, however doses needed to achieve similar effects vary; units used per HbA1c reduction could be a relevant parameter for the choice of insulin.