The physiological pulsatile secretion of GH in humans might be important for the metabolic effects of GH. In the treatment of GH-deficient (GHD) patients, the most common regimen is a single sc injection at bedtime. It has not been completely established if this is the optimal mode of GH administration during long-term GH treatment. The aim of the present study was to evaluate the metabolic effects of two different GH replacement regimens comparing one to two daily injections. Eight men and six women, 42-78 yr old, with verified severe GHD, participated. Patients were matched for gender, age and body mass index (BMI) and were randomized to GH therapy (one or two injections daily) for 12 months. GH doses were individually titrated to IGF-I levels of age-matched controls. IGF-I, glucose, insulin, oral glucose tolerance test (OGTT), cholesterol, triglycerides, lipoproteins, including size fractionation with fast performance liquid chromatography, BMI and body composition were analyzed. After 12 months the median GH dose was 0.45 mg (range 0.25-0.50 mg) in both groups. Body fat had decreased by 20% (p<0.05) in the group receiving one daily GH injection. There were no differences between the two treatment groups in indices of carbohydrate or lipid metabolism. The administration of GH divided into two daily doses offered no major advantage as compared to the more convenient single injection in the evening. The GH-induced reduction in body fat occurred independently from changes in serum lipids.