Background/Aims: In prepubertal short children with idiopathic growth retardation, growth hormone (GH) peak after GH release stimuli classifies patients as growth hormone- deficient (GHD) or non-GHD. This study compared a 2-year growth response to GH therapy in 318 prepubertal short children. Methods: Patients were classified as: severe GHD (GH peaks <5 ng/ml after 2 stimuli; n = 54), mild GHD (GH peaks <10 ng/ml, but one or two between 5 and 10 ng/ml; n = 140), dissociated GH release (GH peak ≧10 ng/ml after 1 stimulus and <10 ng/ml after the other; n = 89), and normal GH release (GH peaks ≧10 ng/ml after 2 stimuli; n = 35). Results: Two-year height gain did not differ statistically among the 4 groups: 1.39 ± 0.51 SD, 16.4 ± 2.3 cm; 1.23 ± 0.56 SD, 15.8 ± 2.1 cm; 1.18 ± 0.53 SD, 15.3 ± 2.0 cm, and 1.14 ± 0.53 SD, 15.4 ± 2.0 cm, respectively, as was also the case for bone age gain: 2.5 ± 0.6, 2.4 ± 0.7, 2.6 ± 0.7 and 2.3 ± 0.5 years, respectively. Conclusions: Our results suggest that GH release stimuli are of little help for deciding on GH therapy in the clinical management of prepubertal short children with idiopathic growth retardation, while well-defined anthropometric and biochemical criteria may be useful.