Diabet. Med. 27, 181–188 (2010)

Aims  Insulin lispro protamine suspension (ILPS) and insulin detemir were compared in insulin‐naive patients with Type 2 diabetes poorly controlled by oral glucose‐lowering agents (OGLAs) to demonstrate non‐inferior overall glycaemic control.

Methods  This was a 24‐week, multinational, open‐label, parallel‐group, treat‐to‐target trial. Adults taking two or more OGLAs were randomized to ILPS (n = 223) or detemir (n = 219) once daily at bedtime. Doses were titrated to target fasting blood glucose (FBG) 5.0–7.2 mmol/l. A pre‐breakfast dose was added up to week 8 per prespecified criteria. The primary objective was comparison of glycated haemoglobin (HbA_1c) change from baseline (non‐inferiority margin 0.4%).

Results  At end‐point, HbA_1c decreased from 8.8 ± 0.7% in both groups to 7.3 ± 0.9% (ILPS) and 7.5 ± 1.1% (detemir). Least‐squares mean difference (95% confidence interval) for HbA_1c [−0.21% (−0.39, −0.03)] and glycaemic variability [0.10 mmol/l (−0.02, 0.23)] demonstrated non‐inferiority. End‐point mean FBG was 7.0 vs. 6.9 mmol/l (P = 0.85), and percentages of patients achieving H < 7.0% were 34.9% vs. 31.2% for ILPS vs. detemir. More ILPS patients used twice‐daily dosing (59% vs. 49%). Mean daily insulin dose was 0.39 vs. 0.46 U/kg (P = 0.005) and weight gain was 1.88 vs. 0.36 kg (P < 0.001) for ILPS vs. detemir. Overall hypoglycaemia (episodes?patient^‐1?year^‐1) (24.2 ± 33.0 vs. 16.2 ± 26.1, P = 0.001) and nocturnal (6.3 ± 12.1 vs. 3.8 ± 13.2, P < 0.001) rates were higher for ILPS.

Conclusions  At end‐point, ILPS was non‐inferior to detemir in HbA_1c change from baseline. Patients using ILPS achieved lower end‐point HbA_1c with lower insulin doses but greater hypoglycaemia and weight gain.