Objective: Impaired cognitive function has been demonstrated in adults with growth hormone (GH) deficiency (GHD) by using different neuropsychological tests. Despite several studies, present knowledge about the impact of GHD and GH replacement therapy (GHRT) on cognitive function is limited. P300 event-related potential (ERP) application is a well-established neurophysiological approach in the assessment of cognitive functions including the updating of working memory content and the speed of stimulus evaluation. GHD is a well-known feature of Sheehan's syndrome and cognitive changes due to GHD and the effects of GHRT remain to be clarified. The present study was designed to investigate the effects of GHD and 6 months of GHRT on cognitive function in patients with Sheehan's syndrome by using P300 latency. Design and methods: The study comprised 14 patients with Sheehan's syndrome (mean age, 49.5^7.8 years) and 10 age-, education-and sex-matched healthy controls. With hormone replacement therapy, basal hormone levels other than GH were stable before enrollment and throughout the GHRT. The diagnosis of GH deficiency was established by insulin-tolerance test (ITT), and mean peak level of GH in response to insulin hypoglycemia was 0.77^0.35 mIU/l. Treatment with GH was started at a dose of 0.45 IU (0.15 mg)/day in month 1, was increased to 0.9 IU (0.30 mg)/day in month 2 and was maintained at 2 IU (0.66 mg)/day. Initially baseline auditory ERPs in patients and controls were recorded at frontal (Fz), central (Cz), and parietal (P3 and P4) electrode sites. In the patient group, ERPs were re-evaluated after 6 months of GH replacement therapy. During each session P300 amplitude and latency were measured. Results: Mean serum insulin-like growth factor-I (IGF-I) concentration in the patient group before GHRT was 23^13 ng/ml. After 6 months of GH therapy mean IGF-I significantly increased to an acceptable level, 234^71 ng/ml (P , 0.05). The mean latencies (at all electrode sites) of the patients before GHRT were found to be significantly prolonged when compared with those of normal controls (P , 0.05). After 6 months of GHRT mean P300 latencies (at all electrode sites) were decreased significantly when compared with latencies before treatment (P , 0.05). Conclusions: The present study, using P300 ERP latencies, therefore suggests an impairment of cognitive abilities due to severe GHD in patients with Sheehan's syndrome and an improvement of cognitive function after 6 months of physiological GHRT. Moreover, this was a novel application of P300 ERP latencies in cognitive function detection in patients with GHD. Further studies with different patient groups need to be done to assess the clinical use of this electrophysiological method in the diagnosis of cognitive dysfunction due to GHD.
European Journal of Endocrinology 150 153-159