The absolute glucose disposal of insulin glargine (Lantus) was compared to that of regular human insulin in healthy subjects (n=20) using the euglycaemic clamp technique in a single-dose, double-blind, randomized, two-way crossover design. Subjects received 30-minute intravenous infusions of insulin glargine (0.1 IU/kg) or human insulin (0.1 IU/kg) and a 20% glucose solution infused at a variable rate to maintain euglycaemia at the subject's baseline glucose level. At equal baseline blood glucose levels (4.42 mmol/l [range, 4.00-5.16 mmol/l] and 4.42 mmol/l [range, 4.01-4.94 mmol/l], respectively), the area under the glucose infusion rate (GIR) time curves from 0-6 hours (AUC(0-6h)) was within the bioequivalence range (insulin glargine, 663.92 mg/kg; human insulin, 734.85 mg/kg). Both the time to maximum GIR and the suppression of serum C-peptide were similar with insulin glargine and human insulin. The resulting maximum serum insulin concentrations (Cmax) were 151.16 microIU/ml and 202.23 microIU/ml, and the time to Cmax (Tmax) was 30 minutes (the duration of the infusion). The observed differences in the Cmax (the mean value for insulin glargine was about 25% lower than that of human insulin) could be explained by lower cross-reactivity of insulin glargine in the human insulin radioimmunoassay. The employed intravenous route, though definitely not the intended clinical use of insulin glargine, provided the clinical evidence in healthy subjects that on a molar basis insulin glargine is equipotent to regular human insulin regarding glucose disposal.