Short term studies have demonstrated the acceleration of growth velocity after the administration of GH in short children born with intrauterine growth retardation (IUGR). We report the final heights of 70 IUGR children whose short stature was attributed to idiopathic GH deficiency (peak plasma GH <10 ng/mL at 2 provocative tests) and treated with GH at a mean dosage of 0.4 +/- 0.1 U/kg x week during an average of 4.6 +/- 2.5 yr. They were compared to a control group of 40 untreated short children born with IUGR, without GH deficiency. At the time of evaluation, age, auxological data, and pubertal status were similar in the 2 groups (height, -2.9 +/- 0.8 and -2.8 +/- 0.7 SD score). Final heights were comparable in both groups of children (-2 +/- 0.7 and -2.2 +/- 1.1 SD score). A multivariate analysis identified 4 independent predictors of final height, namely target height, age and body mass index at evaluation, and GH treatment. Treatment was associated with a gain of 0.6 SD score, suggesting a final height gain of about 3.4 cm. Fifty-three of 70 treated children were reevaluated after completion of growth, and 43 of 53 had a peak plasma GH level of 10 ng/mL or more. Auxological characteristics of these 53 patients were not different from those of nonreevaluated patients. We believe that the transient character of the GH deficiency in most patients and the nonstringent initial criteria used for the diagnosis of GH deficiency render the spontaneous growth potentials identical in the 2 groups of patients. Our data, therefore, suggest that GH treatment at this dosage has a limited effect on the final height of short children born with IUGR.