OBJECTIVE Hyperthyroidism is associated with altered GH secretion. Whether this is due to changes of somatotroph responsiveness or reflects an alteration in negative feedback signals at the hypothalamic level is unknown. We therefore performed a series of studies to shed some light onto this issue. METHODS Study 1: GHRH (1 microgram/kg b.w.) was injected i.v. in 38 hyperthyroid patients and in 30 normal subjects; in 11 of the patients the GHRH test was repeated following methimazole-induced remission of hyperthyroidism. Study 2: hGH (2 U i.v.) or saline were administered 3 hours prior to GHRH; six hyperthyroid patients and six normal subjects were studied. Study 3: ten normal subjects and ten hyperthyroid patients were given 75 g oral glucose or water 30 minutes before GHRH. Study 4: 11 normal subjects and eight hyperthyroid patients were studied. TRH or vehicle were dissolved in 250 ml of saline solution and infused at a rate of 400 micrograms/h for 150 minutes. Thirty minutes after the beginning of the infusions, L-arginine (30 g infused over 45 min i.v.) was administered. METHODS Hyperthyroid patients were compared to normal subjects. METHODS Growth hormone was measured by RIA at 15-minute intervals. RESULTS GH responses to GHRH were subnormal in hyperthyroid patients. Following antithyroid drug treatment with methimazole, GH responses to GHRH increased in these patients in comparison to pretreatment values. Serum IGF-I levels, which were elevated before treatment, decreased after methimazole administration. Exogenous GH administration induced a clear decrease of GH responses to GHRH in both control and hyperthyroid subjects. On the other hand, oral glucose load decreased the GH responses to GHRH in normal but not in hyperthyroid subjects. TRH administration did not modify the GH responses to arginine in either normal subjects or hyperthyroid patients. CONCLUSIONS Hyperthyroidism is associated with increased serum IGF-I levels and marked alterations in the neuroregulation of GH secretion. These changes involve decreased GH responsiveness to GHRH at the pituitary level and, at the hypothalamic level, a lack of suppressive effect of an oral glucose load. The normal inhibitory effect of exogenous GH administration but not of an oral glucose load in hyperthyroid patients suggests that these two feedback signals act through different mechanisms. The lack of effect of a TRH infusion on GH responses to L-arginine in normal and hyperthyroid patients makes an inhibitory role for TRH in GH secretion unlikely, at least in Caucasian subjects.