In this study, 60 patients with proven growth hormone deficiency (GHD) of hypothalamic origin were randomized into three equal groups, and received growth hormone-releasing hormone(1-29)-NH2 (GHRH(1-29)-NH2), 30 or 60 micrograms/kg/day, or growth hormone (GH), 0.1 IU/kg/day, for 6 months. There were no significant differences in growth between the two groups given GHRH(1-29)-NH2, but growth in the GH group was significantly better than in the other two groups (p < 0.01). Mean height velocities at 6 months were 9.2, 9.3 and 14.6 cm/year for the three groups, respectively. Plasma GHRH concentrations increased steadily over the 6-month treatment period, with higher levels in the group on the higher dose. During GHRH(1-29)-NH2 treatment, serum concentrations of insulin-like growth factor I rose initially, but then fell to values similar to those before treatment. No GH antibodies were detected, but all 20 patients on high-dose GHRH(1-29)-NH2 and 19 of 20 patients on low-dose GHRH(1-29)-NH2 developed GHRH antibodies. These had almost disappeared by 9 months after stopping treatment. There was no correlation between antibody titres and increase in height. No serious side-effects were seen, but three patients receiving GHRH(1-29)-NH2 reported mild irritation at the injection site. These results from the continuous infusion of GHRH(1-29)-NH2 over 6 months suggest that this treatment, or the related use of a depot preparation, is unlikely to be as effective as GH for the promotion of growth in GHD.