The effect of short-and long-term corticosteroid treatment on sleep-associated growth hormone secretion

Eight healthy medical studients and four renal transplant patients had blood sampled two or three times hourly throughout EEG monitored nocturnal sleep. This was carried out on the healthy subjects for a total of 12 nights without medication (control nights asleep), a total of 12 nights following 40 mg of flucortolone the previous morning, and a total of 6 nights with similar blood sampling when sleep was prevented (control nights awake). Four renal transplant patients who were receiving long-term therapy with prednisolone were similarly studied (total of 7 nights asleep). Circulating corticosteroid and growth hormone (GH) levels were determined. A peak of GH was seen during the first 2 h of sleep on the control nights when slow-wave sleep predominated. The GH peak was absent on the control nights awake. The pattern of plasma corticosteroid levels was identical during control nights asleep and awake. Both single-dose and chronic corticosteroid administration inhibited the GH peak associated with slow-wave sleep. Chronic corticosteroid therapy, but no single-dose administration in the morning, suppressed the circadian rise of plasma corticosteroids which normally occurs late in sleep.