Comparative effectiveness and harms of long-acting insulins for Type 1 and Type 2 diabetes: a systematic review and meta-analysis.

Aim To review evidence comparing benefits and harms of long-acting insulins in patients with type 1 and 2 diabetes. Methods MEDLINE and two Cochrane databases were searched during February 2018. Two authors selected studies meeting inclusion criteria and assessed their quality. Comparative studies of adult or paediatric patients with diabetes treated with insulin degludec, detemir or glargine were included. Meta-analysis was used to combine results of similar studies, and the I-2 statistic calculated to assess statistical heterogeneity. Results Of 2534 citations reviewed, 70 studies met the inclusion criteria. No statistically significant differences in HbA1c were seen between any two insulins or formulations. Hypoglycaemia was less probable with degludec than with glargine, including nocturnal hypoglycaemia in type 1 (rate ratio 0.68, 95% CI 0.56-0.81) and type 2 diabetes (rate ratio 0.73, 95% CI 0.65-0.82), and severe hypoglycaemia in type 2 diabetes (relative risk 0.72, 95% CI 0.54-0.96). Patients with type 2 diabetes had higher rates of withdrawal because of adverse events when treated with detemir compared with glargine (relative risk 2.1, 95% CI 1.4-3.3). Adults taking detemir gained about 1 kg less body weight than those taking degludec (type 1) or glargine (type 2). Conclusions No differences in glycaemic control were seen between insulin degludec, detemir and glargine. Hypoglycaemia was less probable with degludec than glargine, and patients taking detemir gained less body weight than those given degludec or glargine. In type 2 diabetes, withdrawals as a result of adverse events were more probable with detemir than glargine.