To explore the growth hormone-releasing hormone (GHRH)-GH-somatomedin axis in major depressive disorder, 12 patients and 12 normal controls matched to the patients on age, sex, ovarian status and body weight received synthetic human GHRH-44 amide (1 microgram/kg) as an intravenous bolus. Compared to controls, the depressed patients showed a reduction in baseline plasma GH and a significant attenuation of net plasma GH responses to GHRH. The blunted GH responses occurred along with significantly increased somatomedin C (Sm-C) concentrations. The impairment of GH responses to GHRH and the increased Sm-C concentrations in patients with depression could have resulted from episodic hypersecretion of GH during the daytime, indicating integrity of the negative feedback circuitry. Normal feedback regulation suggests that diurnal episodic hypersecretion of GH reflects an abnormality at or above the level of the hypothalamus, so that the GHRH-GH-somatomedin axis hyperactivity observed in certain patients with major depressive disorder may be due, at least in part, to hypersecretion of hypothalamic GHRH. Our failure to demonstrate a difference in plasma prolactin (PRL) responses to GHRH between controls and depressed patients indicates that GHRH is not a PRL releaser in patients with major depression and that the altered GH secretory dynamics may not be directly related to the altered circadian PRL secretion linked to depression.