Growth hormone (GH) replacement therapy regimens in adults using daily subcutaneous (sc) injections may not be optimal with respect to carbohydrate and lipid metabolism. The aim of this study was to compare the efficacy of three times weekly injections with daily sc GH injections in terms of serum IGF-I, IGFBPs, lipoprotein levels, serum bone markers, glucose metabolism, body composition, compliance and well-being. Twenty hypopituitary men, 46-76 years, on a course of stable conventional GH replacement therapy for more than 12 months, were included in a 16-week crossover trial. During the first 8 weeks GH was administered three times per week followed by 8 weeks with daily sc injections with the same weekly dose of GH. Fasting serum samples were collected at baseline and on two consecutive days at the end of each 8-week period. Serum IGF-I and IGFBP-3 concentrations were lower both the first and second morning after the last injection during the period with three injections per week. The second morning after the last GH injection in this period the IGF-I/BP-3 ratio, plasma insulin and FFA were lower whereas IGFBP-1 was increased as compared with values obtained during the period with daily injections. Serum Lp(a) levels, body composition, fat distribution, well-being and compliance were not differently affected by the two treatment regimens. These results suggest that the same weekly dose of GH given as three injections per week reduces serum IGF-I and IGFBP-3 levels without affecting Lp(a) levels. The day-to-day variation in glucose metabolism and FFA serum levels differs considerably between the two modes of GH administration.