The aim of our study was to elucidate the relationship between the level of circulating cortisol and the GH responsiveness to GHRH in six hypoadrenal patients (one male and five females; age range, 35-67 yr; body mass index range, 18-31 kg/m2). Twenty-four hours after taking the last dose of replacement therapy, each patient underwent the following experimental trials on nonconsecutive days: 1) saline, and 2) 12.5 mg, or 3) 25 mg, or 4) 250 mg hydrocortisone hemisuccinate in 250 mL saline constant iv infusion from 0-180 min. On each occasion, 1 micrograms/kg human GHRH-(1-29)NH2 was injected as an iv bolus at 60 min. During GHRH and saline infusion, serum cortisol levels were always less than the detection limit of the assay (55 nmol/L). During 12.5-, 25-, and 250-mg hydrocortisone infusions (from 15-180 min), serum cortisol averaged 413.8 +/- 19.3, 772.5 +/- 46.9, and 1520.2 +/- 110.4 nmol/L, respectively. The GH peaks after GHRH treatment during the various infusions of hydrocortisone were compared to the GH peaks observed after saline, which were normalized to 100% in each subject. GH peaks after GHRH and 25 mg hydrocortisone (70 +/- 11%) and GHRH and 250 mg hydrocortisone (69 +/- 7%) were significantly (P < 0.05) lower than the GH peaks after GHRH and saline or GHRH and 12.5 mg hydrocortisone (83 +/- 15%). No significant differences were observed between the GH peaks after GHRH and 12.5 mg hydrocortisone or GHRH and saline. Our data demonstrate that in hypoadrenal patients, the acute absence of circulating cortisol does not impair the GH secretory response to GHRH with respect to the eucortisolemic state. Moreover, our data suggest that 700 nmol/L is the approximate threshold serum cortisol concentration above which a decrease in the GH responsiveness to GHRH is observed in humans. Further increases in serum cortisol levels above this threshold value do not cause a proportional decrease in the GH responsiveness to GHRH.