Effect of 64-hour sleep deprivation on the circadian waveform of thyrotropin (TSH): further evidence of sleep-related inhibition of TSH release.

Half-hourly sampling of plasma TSH was done across 3 days in four normal young men. Sleep was denied for 64 h from 0700 h on awakening from accommodation sleep until polygraphic sleep was resumed at 7100 h of the third day (D3) such that 2 consecutive nights of usual 2300-0700 h sleep were missed. This protocol allowed examination of any modulatory effects on the daily patterns in TSH concentrations during sleep deprivation on D1-2 (1100-3500, 3500-5900 h) or during resumption of usual nightly sleep on D3 (5900-8300) compared to that of a previously studied group of normal young men. The circadian nature of the daily TSH waveform was evidenced by its daily repetition within a subject both basally and during D1-2 sleep deprivation and by its synchronization within the basal, deprived, or resumed sleep days. The peaks in each subject's daily TSH patterns on D1-2 were consistently longer, and the daily maxima and cosine acrophases on D1-2 were consistently later than those on D3 when basal sleep was resumed. About half the daily TSH concentration maxima and daily cosinor amplitudes on D1-2 were greater than those of the respective sleep-resumed TSH patterns of D3. Both the group mean TSH patterns and the cosinor 95% confidence ellipses also indicated the daily peak in the TSH waveform to be significantly longer, later, and larger during D1-2 sleep deprivation than during the basal or D3 periods. These results indicate that significant alteration of the daily TSH waveform can occur in response to absence of sleep and are compatible with the existence of an inhibitory effect in early nightly sleep on TSH release. The TSH patterns during the 1700-2300 h intervals of rising TSH levels were congruent in the basal, deprived, and resumed sleep periods. Prompt reversion to the basal TSH pattern also occurred when sleep was resumed on D3. Both of these observations suggest the alteration in TSH waveform during sleep deprivation to have arisen from an inhibitory effect in sleep rather than from a change in period or phase of a generating oscillator.