Plasma growth hormone (GH), insulin, cortisol, and glucose were measured during sleep on 38 nights in eight young adults. Blood was drawn from an indwelling catheter at 30-min intervals; EEG and electrooculogram were recorded throughout the night. In seven subjects, a plasma GH peak (13-72 mug/ml) lasting 1.5-3.5 hr appeared with the onset of deep sleep. Smaller GH peaks (6-14 mug/ml) occasionally appeared during subsequent deep sleep phases. Peak GH secretion was delayed if the onset of sleep was delayed. Subjects who were awakened for 2-3 hr and allowed to return to sleep exhibited another peak of GH secretion (14-46 mjug/ml). Peak GH secretion was not correlated with changes in plasma glucose, insulin, and cortisol. The effects of 6-CNS-active drugs on sleep-related GH secretion were investigated. Imipramine (50 mg) completely abolished GH peaks in two of four subjects, whereas chlorpromazine (30 mg), phenobarbital (97 mg), diphenylhydantoin (90 mg), chlordiazepoxide (20 mg), and isocarboxazid (30 mg) .did not inhibit GH peaks. Altered hypothalamic activity associated with initiation of sleep results in a major peak of growth hormone secretion unrelated to hypoglycemia or changes in cortisol and insulin secretion.

No abstract

Spectral luminous efficiency function for a 2-deg field at a photopic level (100 Td) was measured for 91 observers by flicker photometry (FP) and for 97 observers by direct brightness matching (DBM), to find age-related change in the efficiency function as well as to obtain a reliable data set to be used in photometry. Observers ranged in age from 11 to 78 years. A gradual reduction of luminous efficiency with age was observed for both functions by FP and by DBM in the short-wave region, which was expected because of the age-related increase of optical density of eye lens. A similar age-related reduction of efficiency was observed in the long-wave region for the function obtained by DBM; this reduction was regarded as being due to reduced chromatic contribution to brightness with age. Principal components analysis on the spectral efficiency data and an analysis of the efficiency difference between the data obtained by DBM and FP confirmed this conclusion. Assuming a log-linear change in efficiency with age for any wavelength throughout the life span, spectral luminous efficiency function at any age was derived for photometric use.

The range of entrainment of the circadian behavioral rhythm was compared between two groups of Sprague-Dawley rats (each n = 10) exposed to daily cycles of rectangular light-dark alternation (LD) and sinusoidal fluctuations of light intensity (SINE), respectively. The maximum illuminance (20 lx), the minimum illuminance (0.01 lx), and the total amount of light exposure per cycle were the same under the two lighting conditions. The periods (Ts) of both lighting cycles were lengthened stepwise from 24 through 25, 26, 26.5, 27, 27. 5, and 28 h to 28.5 h in experiment 1 and were shortened stepwise from 24 through 23.5, 23, and 22.5 h to 22 h in experiment 2. Each T cycle lasted for 30 cycles. In experiment 1, 60% of rats under the LD condition entrained up to T = 28.5 h, whereas 50% of rats under the SINE condition entrained up to T = 28.5 h. In experiment 2, no animal under the LD condition entrained to T < 23.5 h, whereas 40% of rats under the SINE condition entrained down to T = 23 h and 20% of rats remained to entrain down to T = 22 h cycles. The phase angle of entrainment was systematically changed, depending on T under both conditions. These results suggest that the lower limit of entrainment is expanded under the SINE condition compared with the LD condition.

Twenty-four canine GH (cGH) and cortisol secretion patterns associated with sleep stages were studied in 10 male adult dogs. Plasma samples were obtained at 30- or 15-min intervals via an indwelling catheter. Under baseline conditions, all dogs showed irregular polyphasic sleep, and the episodic cGH secretion had no apparent relationship with sleep or the light-dark cycle. Five dogs were subjected to regular sleep-wake cycles; 3, 6, and 12 h of forced wakefulness (FW) were repeated at 3-, 6-, and 12-h intervals (recovery sleep periods), respectively. Peak cGH secretion (mean +/- SD, 6.4 ng/ml +/- 2.4) occurred soon after recovery sleep onset in 25 of 40 total recovery periods. The incidence of sleep-onset cGH peaks and cGH secretion during the first hour of recovery sleep significantly increased with the length of the preceding FW, but were not affected by the time of day. Delta wave sleep increased during this hour, suggesting a possible correlation with the sleep-onset cGH peak. During the first 3 h of recovery after 6 and 12 h of FW, cGH secretion was significantly enhanced, but cortisol was not. Considering the characteristics of human sleep-related GH secretion, we suggest that this peak cGH secretion represents a model of human GH secretion. Possibly, a close association of cGH secretion with sleep is concealed under the baseline condition and uncovered by inducing longer sleep-wake cycles in dogs. No circadian cortisol variation was detected under the baseline or the experimental conditions.