Predictors of growth response to growth hormone in otherwise normal short children.

Sixty prepubertal short children (39 boys) with heights less than 2 SD for age and gender were treated daily for 1 year with recombinant human growth hormone (GH), either 0.1 IU/kg (group 0.1, n = 32) or 0.05 IU/kg (group 0.05, n = 28). Reserve of GH was determined by at least one GH provocative stimulus and 24-hour continuous blood withdrawal to determine the integrated concentration of GH (IC-GH). All participants had a GH response to provocative tests greater than 10 micrograms/L. The height velocity (mean +/- SD) of the group as a whole increased from 4.46 +/- 1.02 to 7.59 +/- 1.65 cm/yr (p less than 0.001). The growth velocity of group 0.1 was significantly greater than that of group 0.05 (8.1 +/- 1.5 vs 7.0 +/- 1.65 cm/yr; p less than 0.01). Bone age did not advance more than 1 year during the treatment period. Growth velocity after 1 year of GH therapy was inversely correlated with the IC-GH in both groups, as was the pretreatment height velocity. We found no correlation of growth velocity during GH therapy with other measures such as parental heights, bone age/chronologic age ratio, maximal GH response to provocative tests, chronologic age, or pretreatment insulin-like growth factor I levels. We conclude that the best predictors for the 1-year growth outcome of short children with a normal GH response to provocative tests are the pretreatment growth velocity and the IC-GH. The short-term benefit from GH therapy in children with a normal growth velocity and a normal IC-GH is poor, whereas marked growth acceleration is noted in children with a low growth velocity and a low 24-hour IC-GH.