OBJECTIVE The bioavailability of GH immunoreactive serum concentrations is reduced following subcutaneous (s.c.) as compared with intravenous (i.v.) administration. Whether this difference also translates into a different biological activity remains to be investigated. The aim of the present study was to evaluate the short‐term metabolic effects of GH following i.v. and s.c. delivery.DESIGN AND MEASUREMENTS In a cross‐over design 10 GH‐deficient patients were randomized to receive GH (0.03 μg (0.1 mU)/kg/min) as a continuous i.v. or s.c. infusion for 39 hours on two different occasions. Preceding each study GH therapy was discontinued for 5 days. Serum profiles of GH, IGF‐I, IGF‐II, IGF binding protein 3 (IGFBP‐3), insulin, glucose and non‐esterified fatty acids (NEFA) were recorded during the studies. Serum GH was measured by a polyclonal radio‐immunoassay (RIA) and by a double monoclonal immunofluorimetric assay (DELFIA).RESULTS Higher mean integrated values (AUC) of serum GH (mU/l) were obtained with i.v. GH delivery [47.4±5.1 (i.v.), 33.3±3.0 (s.c.),P<0.05]. The two GH assays showed qualitatively similar results, but higher mean GH concentrations were measured by RIA following both s.c. (P<0.001) and i.v. infusion (P<0.001). Serum IGF‐I levels displayed different patterns following i.v. and s.c. GH infusion (P<0.05 by ANOVA) and mean IGF‐I levels (μg/l) were lower following s.c. GH infusion [159.5±21.8 (s.c.), 185.2±27.7 (i.v.),P=0.002]. Serum IGF‐II levels were unaffected by short‐term GH treatment and by the route of GH administration. Serum IGFBP‐3 levels increased in response to GH administration (P<0.001), irrespective of route (P=0.76). The IGF‐I/IGFBP‐3 molar ratio increased significantly following GH administration (P<0.001), and a higher ratio was obtained following i.v. infusion (P<0.005). Subcutaneous GH infusion resulted in significantly lower mean levels of serum NEFA (P<0.02), whereas similar mean levels of serum insulin (P=0.54), blood glucose (P=0.24), energy expenditure (P=0.13), and respiratory exchange ratio (P=0.09) were observed on the two occasions.CONCLUSIONS A reduced bioavailability of s.c. as compared with i.v. administered GH has been recorded with two independent GH assays, and this was also accompanied by a significant,