The effects of body temperature and oxygen consumption on sleep architecture

Two related theories have had a substantial impact on our understanding of the nature and function of sleep. These are the energy conservation and restorative theories of sleep. Both theories predict an increase in both slow wave sleep (sws) and sleep duration following increased energy expenditure. These predictions form the background to the empirical work of this thesis, and it is the effect of wake-period metabolism on both sleep metabolism and architecture that holds the main research focus. Three studies were designed to evaluate the effects of either metabolic rate (MR), body temperature (T\\(_b\\)), or both, on sleep architecture, in particular sws. The first and second studies imposed variations in waking activity in order to assess the effects of MR and T\\(_b\\) on sleep architecture. The first study actively increased MR by physical exercise, while the second study used passive heating in a warm bath. The third study compared T\\(_b\\), MR and sleep in endurance athletes and sedentary individuals. In the first study, the first sleep cycle of 10 young fit subjects (mean age = 21.8 years) was assessed after a 19km run, either immediately before bedtime or a few hours before retirement. There were four conditions: a no exercise condition; a late afternoon exercise session with evening meal; a late afternoon exercise session without evening meal; and a late evening exercise session with an evening meal. The results showed no evidence of an exercise-induced sws effect, and found that exercise transiently increased wake-period MR which returned to control levels by bedtime. Furthermore, there was some evidence to suggest that a rise in wake-period energy expenditure may have a negative effect upon sleep properties. The second study investigated the effect of passive heating on sleep architecture by using a method similar to that described by Horne and Staff (1983), Sewitch (1987) and Berger and Phillips (1988a,b). It also tested the hypotheses that sws levels increase following heating due to either a compensatory drop in T\\(_b\\)at sleep onset (Sewitch, 1987), or a sustained elevation in absolute T\\(_b\\) at sleep onset, and during sleep (Berger & Phillips, 1988; Berger, Palca, Walker & Phillips, 1988). Five healthy young male subjects (mean age = 20.4 years) were passively heated in a 42-43°c warm bath to induce elevated T\\(_b\\) and MR. A repeated measures design with two conditions was employed. These conditions included a control (no passive heating), and a passive heating condition in the late afternoon. Rectal temperature (T\\(_{re}\\)) was monitored from the early afternoon until the awakening period on the following morning. Metabolic rate was recorded for 20 minutes prior to, and immediately after the passive heating, and then across the sleep period. Sleep recordings also were monitored over the night. Results showed that passive heating significantly increased T\\(_{re}\\), MR and sws levels. Rectal temperature increases were sustained into, and across the sleep period, whereas MR increases were only transient and did not continue into the sleep period. sws levels were significantly elevated in the first 150 minutes of sleep. The direct relationship of T\\(_b\\) to sws supported the theory proposed by Berger and Phillips (1988a,b). The third and final study determined whether the characteristically higher sws levels and longer sleep durations of endurance athletes (Trinder, Paxton, Montgomery & Fraser, 1985) may be attributable to the effect of T\\(_b\\) on sleep. It was designed to assess the role of T\\(_b\\) on sleep by comparing the laboratory sleep of endurance athletes and sedentary individuals with T\\(_b\\) at sleep onset held constant between the two groups based upon evidence from another study (Hedges, 1989) that showed higher average Tres and earlier sleep onset times in athletes compared to sedentary individuals. It was thus considered that the higher T\\(_{re}\\)S reported for athletes in this study may have been a consequence of their earlier sleep onset times. In the final study eight male endurance athletes (mean age = 21.5 years) and eight male non-athletes (mean age = 22.6 years) were compared under conditions of no-exercise. The results showed sws levels to be higher and sleep duration longer in the endurance athletes as compared to sedentary subjects, despite T\\(_b\\) at sleep onset being the same for the two groups, as a result of sleep onset being held constant. The results suggest elevated T\\(_b\\) at sleep onset may not be the mechanism causing particular sleep characteristics of endurance athletes. Rather it is proposed that the sleep properties of endurance athletes are due to a phase delay of the circadian oscillator, which in this group is achieved by an advance of the sleep-wake cycle (earlier usual sleep onset time). In conclusion, it is argued that there is a relationship between metabolism and sleep architecture where sws can be facilitated by either (a) high metabolism during sleep onset and the early part of sleep; or (b) the phase angle in the circadian temperature rhythm at sleep onset, or both.