There is high risk for H in TBI and SAH patients. Early diagnosis of PH is always confirmed in the long term. Pituitary function in brain-injured patients may improve over time but, although rarely, may also worsen. Thus, brain-injured patients must undergo neuroendocrine follow-up over time.

In a retrospective study of patients with pNETs, we found a significant association between metformin use and longer PFS.

IGF-I is the best marker of GH secretory status but it also depends on the nutritional status and peripheral hormones such as insulin, glucocorticoids, thyroid hormones and gonadal steroids. Though monitoring IGF-I levels is the best way for evaluating appropriate GH replacement, the usefulness of IGF-I assay in the diagnosis of adult GH deficiency (GHD) is still matter of debate. To clarify this point in a large population of GHD adults (no. = 135, 61 women and 74 men; age, mean +/- SE: 43.8 +/- 1.4 yr, range 20-80 yr) we studied IGF-I levels, their reproducibility and association to peak GH response to GHRH + arginine (GHRH + ARG) test and insulin tolerance test (ITT). The results in GHD were compared with those in a large population of normal subjects (no. = 336, 233 women and 103 men, aged 20-80 yr). Mean IGF-I levels in GHD (77.8 +/- 4.9 micrograms/l) were clearly lower (p < 0.001) than those in normal subjects (170.2 +/- 4.7 micrograms/l). In Childhood Onset GHD (CO-GHD; no. = 40; age, mean +/- SE: 27.8 +/- 1.5 yr) IGF-I levels were lower than those in Adult Onset GHD (AO-GHD; no. = 95, age, mean +/- SE: 50.7 +/- 1.4 yr) (56.6 +/- 9.7 vs 87.1 +/- 5.4 micrograms/l, p < 0.0003). In both GHD and normal subjects IGF-I levels showed good, reproducibility (r = 0.92, p < 0.00001 and r = 0.62, p < 0.00001, respectively). In GHD, but not in normal subjects, IGF-I levels were positively associated to peak GH responses to GHRH + ARG (r = 0.57, p < 0.00001); on the other hand, the GH peak after ITT was not associated to IGF-I in GHD. In normal subjects, but not in GHD, IGF-I levels were negatively associated to age (r = -0.60, p < 0.00001). Considering individual IGF-I levels there was a clear overlap between GHD and normal subjects. However, this overlap was strongly dependent on age. In fact, in the third and fourth decade of life 83.6% of GHD had IGF-I levels below the 3rd centile of normal values; on the other hand, in the fifth-sixth decade and in ageing 47% and only 12% of GHD, respectively, had IGF-I levels low for age. In conclusion, our results demonstrate that IGF-I levels represent a reproducible marker of GH status and are reduced more in CO-GHD than in AO-GHD adults. An overlap exists between GHD and normal subjects, however this is small up to the 4th decade of life. Thus, though normal IGF-I levels do not rule out the existence of GHD, up to 40 yr low IGF-I levels strongly point to GHD if malnutrition and liver disease have been ruled out.

Cabergoline (CAB), a new long-acting ergoline derivative, was shown to be very effective in reducing PRL levels in normal volunteers and in hyperprolactinemic patients. We evaluated the hormonal changes after discontinuation of long-term therapy with CAB as well as the safety of drug exposure during pregnancy both for mothers and babies. We therefore studied 48 patients (47 females and one male) with pathological hyperprolactinaemia (mean +/- SE, 117.2 +/- 15.2: median 73.2 micrograms/l), treated for 1-82 months (mean +/- SE, 28.3 +/- 3; median 18). After long-term treatment, CAB was withdrawn in 11 patients and PRL levels were persistently normal for almost 15 days and significantly lower (p < 0.05) than basal at 30, 45, 60, 90, 120 days. Three patients had normal PRL levels still at 45 days after treatment discontinuation. Nine patients became pregnant after 1-37 months (mean 12.4) of therapy. In two patients the pregnancy was interrupted spontaneously in one case and voluntarily in the other. In all but one patients after delivery or three-month breast feeding, PRL levels trended towards reduction. In two cases (one with microadenoma and one with idiopathic hyperprolactinaemia) PRL remained in the normal levels for 1-3 years after delivery. In conclusion CAB is able to inhibit plasma PRL levels for long time (up to 120 days) after withdrawal in patients with pathological hyperprolactinaemia treated with long-term therapy.

Funding: The study was not funded. Acknowledgements: We would like to thank all the members of the Fluid and Electrolytes Disorder Club of the Italian Society of Endocrinology, for the scientific support during the writing of the manuscript.