The aim of the present study was to evaluate the characteristics of GH deficiency (GHD) in adult patients after neurosurgery for pituitary adenomas and craniopharingiomas. One hundred and one GHD patients, (42 F/59 M), aged 47.58+/-14.4 yr (mean+/-SD; range 21-78), body mass index (BMI) 28.6+/-0.6, with a history of adult-onset hypothalamic-pituitary disease, were recruited for the study. The whole group included: 45 non-functioning pituitary adenomas, 23 craniopharyngiomas, 16 PRLomas, 8 GHomas, 7 ACTHomas and 2 FSHomas; in particular 51 were macroadenomas and 27 microadenomas. At study entry, GHD diagnosis was carried out by assessing GH secretion after GHRH+arginine. All patients were submitted to the study at least 12 months after neurosurgery and, where needed, subjects were replaced with an appropriate treatment. GHD was mild in 3/101 (3%) and severe in 98/101 patients (97%). Other hormone deficiencies associated with GHD were considered: TSH, ACTH, FSH/LH, ADH. The distribution of peak GH among all patients, according to the type of disease before neurosurgery, showed that patients with Cushing disease were characterized by the presence of higher peak GH. According to the number of additional hormone deficits, the distribution of peak GH among all patients was as follows: GHD was isolated in 4/101 subjects (4%; group A), while it was associated with 1 (14/101, 14%; group B), 2 (22/101, 22%; group C), 3 (44/101, 43%; group D) and 4 hormone deficits (17/101, 16%; group E). GHD was severe in all patients in the panhypopituitaric group. Total IGF-I plasma levels in the whole group of GHD patients were 95.2+/-4.2 microg/l. In all groups of patients IGF-I was lower in subjects with severe GHD than in those with mild GHD (93.6+/-4.1 vs 148.6+/-33.6 microg/l, p<0.03). In particular, according to the type of disease presented before neurosurgery, patients with Cushing disease were characterized by the presence of higher IGF-I plasma levels compared to the other. According to the number of additional deficits, the distribution of IGF-I plasma levels was characterized by higher values when GHD was isolated than when it was associated with multiple hormone deficiencies. IGF-I plasma levels were positively associated to peak GH during GHRH+arginine (r=0.4, p<0.0005). We conclude that patients after neurosurgery approach for sellar and parasellar neoplasia, within an appropriate clinical context, and both the presence of additional pituitary hormone deficiency and low levels of IGF-I can be considered a clear GHD condition, and therefore do not require provocative tests evaluating GH secretion.