Patients treated with cranial radiation are at risk of GH deficiency (GHD). We evaluated somatotroph responsiveness to maximal provocative tests exploring the GH releasable pool in relation to the impact of radiation damage to the hypothalamic-pituitary unit. The GH-releasing effect of GHRH plus GH secretagogue [GH-releasing peptide (GHRP)-6] (GHRH+GHRP-6) was studied in 22 adult patients (age, 23.2 +/- 1.4 yr; 8 female and 14 male; mean body mass index, 22.6 +/- 0.7 kg/m(2)) who received cranial radiation for primary brain tumor distant from hypothalamic-pituitary region 7.6 +/- 0.7 yr before GH testing. Two stimulatory tests for GH secretion were employed: insulin tolerance test (ITT, 0.15 IU/kg regular insulin i.v. bolus); and GHRH+GHRP-6 test: GHRH (Geref Serono, Madrid, Spain; l microg/kg) plus GHRP-6 (CLINALFA, Laufelingen, Switzerland; 1 microg/kg) as i.v. bolus. Serum GH was measured (Delphia; Perkin Elmer, Wallac, Turku, Finland) at -30, -15, 0, 15, 30, 45, 60, 90, and 120 min. Anterior pituitary function was normal in all except in 1 female with hyperprolactinemia. Twelve out of 22 irradiated patients were GH-deficient (GHD) with both tests. Eleven out of 22 (50%) irradiated patients were severely GHD, according to the ITT (GH < 3 microg/liter; mean GH peak, 1.5 +/- 0.5 microg/liter). In 9 irradiated patients, in whom ITT was performed as well, mean peak GH after the GHRH+GHRP-6 test was 6.2 +/- 0.8 microg/liter, which is considered as severe GHD, according to our own cut-off for the test (peak GH < 10 microg/liter). GH responses to both tests were highly concordant, but the differential in the GH peak concentrations between GHD and non-GHD irradiated patients was significantly larger for the GHRH+GHRP-6 test than that for the ITT. The 2 discordant responses, i.e. poor response to the ITT and good response to the GHRH+GHRP-6 test, were found in 1 hyperprolactinemic female patient and in 1 other female. One irradiated patient was diagnosed as GHD only with the combined test, because ITT was contraindicated because of epilepsy. PRL and cortisol responses to ITT were normal in all irradiated patients and did not depend on the GH status. IGF-I levels were not informative or discriminative between the GHD and non-GHD irradiated adult patients. In conclusion, the use of GH secretagogues plus GHRH is an easy, reliable and accurate way of assessing GH secretion in cranially irradiated patients. Impairment of the GH releasable pool in the irradiated patients, with a maximal provocative test, reflects alterations in the hypothalamic-pituitary unit caused by radiotherapy.