To assess whether short-term growth hormone (GH) treatment can improve the linear growth in children who were born small for gestational age (SGA), we started a randomized multicenter trial in 26 age- and sex-matched prepubertal children born SGA. During the 1st year of GH therapy, all children received GH 0.23 mg/kg/week, then during the 2nd year, 13 children received the same dose (group A), and in the other 13 children, the dose of GH was doubled, i.e., 0.46 mg/kg/week (group B). During the 1st year of therapy, the growth velocity significantly (p < 0.0001) increased in all patients. During the 2nd year, group A showed a significant decrease of the growth velocity (p < 0.015), whereas group B maintained the growth rate. The height in group A children significantly increased during the 1st and the 2nd year of GH therapy (p < 0.000002 and p < 0.000001, respectively), reaching the normal range in 8 out of 13 children at the end of 2 years of GH therapy. The height in group B children significantly increased during the 1st and the 2nd year of GH therapy (p < 0.000001 and p < 0.000001, respectively), reaching the normal range in all 11 children who completed the GH therapy. The height gain was similar in groups A and B treated with the same GH dosage during the 1st year of therapy. A greater increase in height gain was found in children of group B treated with the higher GH dosage during the 2nd year of therapy as compared with group A (p < 0.02). Significant increases in insulin-like growth factor I (p < 0.0001), acid-labile subunit (p < 0.0002), and bone/chronological age ratio (p < 0.0001) were found after the 1st year of GH therapy, but no significant changes were observed during the 2nd year, independently of the GH dose. In conclusion, the height velocity of children born SGA significantly increases during the 1st year of GH therapy, diminishes, but can decrease during the 2nd year, if the GH dosage is not raised.