Administration of growth hormone (GH) to patients with growth hormone deficiency (GHD) has beneficial effects, but so far has been employed only empirically. We have, therefore, investigated the dose-dependent effect of GH on target tissue by studying biochemical markers of bone and collagen turnover in GHD. Then patients with GHD (nine males and one female aged 21-43 years, mean age 28 years) participated in the study. Growth hormone deficiency was defined as a peak serum GH response of less than 15 mU/l in two provocation tests. After a 4-week run-in period, the study population received increasing doses of GH at 4-week intervals (1, 2 and 4 U/m2). Blood samples were collected in the fasting state at 7.00 h on the last day of each period and assayed for serum levels of osteocalcin (S-BGP), bone alkaline phosphatase (B-ALP), C-terminal propeptide of type I collagen (S-PICP), carboxy-terminal pyridinoline cross-linked telopeptide of type I collagen (S-ICTP) and N-terminal propeptide of type III collagen (S-PIIINP). Following replacement therapy, serum insulin-like growth factor I and insulin-like growth factor binding protein 3 increased sequentially with time (p < 0.001 and p < 0.001, MANOVA) and the values were elevated significantly over baseline levels after treatment with 1 U/m2. Serum BGP values were below normal at the start of the study and increased gradually following GH treatment to levels in the low-normal range. Baseline values for serum bone alkaline phosphatase (B-ALP), PICP and PIIINP were within the normal range. The collagen parameters increased with GH replacement (p < 0.001, MANOVA) to levels above normal, whereas B-ALP stayed within normal limits. Serum ICTP values were elevated above the normal range at baseline, indicating increased bone resorption in GHD. A linear increase in values was observed with GH treatment (p < 0.001, MANOVA). Serum ICTP did not correlate significantly with the bone formative parameters but was correlated positively to PIIINP. The sensitivity of S-ICTP as a bone resorptive marker is thus questioned. In conclusion, a dose-dependent increase in markers of growth hormone metabolism and in biochemical markers of both bone and non-bone collagen synthesis was seen following incremental doses of GH in GHD.