OBJECTIVE We examined the effects of different doses of GH on insulin-like growth factor 1 (IGF-1), IGF binding protein 3 (IGFBP-3), body composition, energy expenditure, and various metabolites in GH deficient adults, in order to approach a metabolically appropriate GH dosage in young GH deficient adults. DESIGN Ten GH deficient patients (age 21-43) were studied after 4 weeks without GH followed by three consecutive 4-week periods, where the patients received in a fixed order GH 1, 2 and 4 IU/M2 S.C. per day. At the end of each period the patients were hospitalized for a 24-hour examination. RESULTS Mean 24-hour levels of GH (mIU/l) were 2.7+/-0.3 (0 GH), 7.2+/-0.9 (1),10.8+/-1.5 (2) and 18.9+/-2.7 (4 IU/m2) (mean+/-SEM) (P<0.01). Likewise, IGF-I levels increased dose dependently from 61+/-21 to 206+/-65, 260+/-70 and 468+/-171 mug/l (P<0.05); serum IGF-I in an age and sex matched control group was 248+/-25 mug/l. Corresponding serum IGFBP-3 levels also increased from 1860+/-239 to 3261+/-379, 3762+/-434 and 4384+/-652 mg/l (P=0.01) respectively. Significant increases in diurnal serum insulin levels after 4 IU/m2 were recorded, whereas plasma glucose levels remained unchanged. Lipid intermediates increased dose independently during GH administration. GH caused a significant increase in resting energy expenditure, whereas the respiratory exchange ratio was unaltered. Fat mass was increased without GH therapy and decreased during the study. Four patients made complaints during 4 IU/m2 GH administration, probably related to GH induced fluid retention. CONCLUSION Based primarily on IGF-I and IGFBP-3 levels our data suggest that a GH replacement dose in young GH deficient adults in the order of 1-2 IU/m2 per day is adequate. This is a relatively low dose as compared to dose regimens in children and adolescents.