Different thresholds of tissue-specific dose-responses to growth hormone in short prepubertal children

Background In addition to stimulating linear growth in children, growth hormone (GH) influences metabolism and body composition. These effects should be considered when individualizing GH treatment as dose-dependent changes in metabolic markers have been reported. Hypothesis: There are different dose-dependent thresholds for metabolic effects in response to GH treatment. Method A randomized, prospective, multicentre trial TRN 98-0198-003 was performed for a 2-year catch-up growth period, with two treatment regimens (a) individualized GH dose including six different dose groups ranging from 17–100 μg/kg/day (n=87) and (b) fixed GH dose of 43 μg/kg/day (n=41). The individualized GH dose group was used for finding dose–response effects, where the effective GH dose (ED 50%) required to achieve 50% Δ effect was calculated with piecewise linear regressions. Results Different thresholds for the GH dose were found for the metabolic effects. The GH dose to achieve half of a given effect (ED 50%, with 90% confidence interval) was calculated as 33(±24.4) μg/kg/day for Δ left ventricular diastolic diameter (cm), 39(±24.5) μg/kg/day for Δ alkaline phosphatase (μkat/L), 47(±43.5) μg/kg/day for Δ lean soft tissue (SDS), 48(±35.7) μg/kg/day for Δ insulin (mU/L), 51(±47.6) μg/kg/day for Δ height (SDS), and 57(±52.7) μg/kg/day for Δ insulin-like growth factor I (IGF-I) SDS. Even though lipolysis was seen in all subjects, there was no dose–response effect for Δ fat mass (SDS) or Δ leptin ng/ml in the dose range studied. None of the metabolic effects presented here were related to the dose selection procedure in the trial. Conclusions Dose-dependent thresholds were observed for different GH effects, with cardiac tissue being the most responsive and level of IGF-I the least responsive. The level of insulin was more responsive than that of IGF-I, with the threshold effect for height in the interval between.