Growth response to growth hormone (GH) treatment relates to serum insulin-like growth factor I (IGF-I) and IGF-binding protein-3 in short children with various GH secretion capacities. Swedish Study Group for Growth Hormone Treatment.

Abstract

The purpose of the study was to evaluate the relationship between the 1-yr (n = 193) and 2-yr (n = 128) growth response and the individual serum concentrations of insulin-like growth factor I (IGF-I) and IGF-binding protein 3 (IGFBP-3) before and during GH treatment. Our study group of prepubertal short children had from very low to high GH secretory capacity, estimated during an arginine-insulin tolerance test, and the ages ranged from 3-15 yr at the start of treatment. Their serum levels of IGF-I and IGFBP-3 were low before treatment compared to those in an age-related reference group of prepubertal children and increased significantly from the start to 1 month of GH treatment. The mean increase in height SD score was 0.80 SD score after 1 yr of GH treatment and 1.26 SD score after 2 yr, with a wide range. In univariate analyses the highest correlation coefficients to the 2-yr growth response were found to be vs. the following variables from the start of treatment: IGF-I SD score (r = -0.49), log maximum GH concentration (log GHmax) during the arginine-insulin tolerance test (r = -0.47), difference between the height SD score of the individual child and the midparental height SD score (diffSD score; r = -0.45), IGFBP-3 SD score (r = -0.39), age (r = -0.30), short term change in IGFBP-3 SD score (r = 0.37), and IGF-I SD score (r = 0.34). In multivariate stepwise regression analysis, 41% of the variation in the 2-yr growth response could be explained by IGF-I SD score or log GHmax together with age at the start of treatment, weight SD score at 1 yr of age, and diffSD score. When both IGF-I SD score and GHmax were included and when the short term changes in IGF-I SD score were added, 46% and 58% of the variation, respectively, could be explained. The regression algorithms using different combinations of variables and their corresponding prediction intervals are also presented.