University Of Tasmania
whole_GriffinSusanJessica1985_thesis.pdf (5.74 MB)

Physical fitness and sleep

Download (5.74 MB)
posted on 2023-05-27, 15:04 authored by Griffin, SJ
Physically fit athletes have been found to have elevated slow wave sleep (SWS) , longer sleep duration and decreased sleep onset latency (SOL) compared with unfit sedentary individuals. It has been hypothesized that the critical variable in these effects is the subject's level of fitness resulting from habitual exercise. Certain negative findings, however, cast doubt on this interpretation. The aim of the current series of experiments was to further investigate the effects of physical fitness on electroencephalographic (EEG) and hormonal aspects of sleep. These investigations are of theoretical interest as they potentially provide information regarding the functional significance of sleep and SWS in particular. Those theories which hypothesize a relationship between peripheral metabolism and sleep, for example, the restorative and energy conservation - theories, would predict that chronic physical exercise would promote SWS and sleep-related anabolic hormones. In contrast, those theories which do not propose a direct relationship between peripheral metabolism and sleep, for example, the cerebral restitution and immobilization theories, would predict no effect of physical fitness on sleep. By using independent group designs, previous studies have potentially confounded aerobic fitness with other characteristics of athletic individuals. Therefore, the first experiment in this thesis assessed the sleep of proficient athletes on two occasions; initially when they were unfit and subsequently when aerobically fit. In addition, the sleep of athletes was compared with that of an unfit, non-athletic, sedentary group. The athletes tended to sleep longer and had elevated SWS and non-rapid eye-movement (NREM) sleep compared with non-athletes. These differences were independent of the aerobic fitness of the athletes. Thus, it was concluded that aerobic fitness was not a critical factor but, rather, had been confounded with a more enduring variable associated with physical fitness. A secondary issue addressed in Experiment 1 was the relationship of daytime exercise, physical fitness and SWS. It has been hypothesized that a facilitative effect on SWS of a single exercise session is dependent on the subjects being physically fit. This hypothesis was tested by assessing the effect of exercise on sleep in the athletes when they were unfit and subsequently when they were fit. The hypothesis was not supported as no effect of exercise on sleep was observed. It is possible, however, that fitness may increase the probability of an exercise effect occurring when one or more other conditions are met. Experiments investigating the effects of physical fitness on sleep have typically concentrated on EEG variables. A number of factors have also suggested that hormonal aspects of sleep may be influenced by physical fitness. Experiment 2, therefore, was designed to examine the effect of physical fitness on the night-time secretion of human growth hormone (hGH), prolactin and cortisol. In addition, the relationship between the secretion of these hormones during the night and body composition was assessed. Two groups of 17 subjects, one of fit athletes and the other of unfit non-athletes, were selected so that the groups were matched for weight, height, lean body mass (LBM) and fat levels. Subjects slept in a sleep laboratory for 3 non-consecutive nights; an adaptation night and two experimental nights. On one experimental night blood samples were collected, while on the other, baseline sleep was assessed and the catheter was not inserted. Weight and height were measured and LBM assessed by 24hr urinary creatinine. The effect of physical fitness was tested by a comparison of the two groups, while the effect of body composition was assessed by correlation analyses. Physical fitness did not have a significant effect on either sleep or hormone levels, though in the latter case the results were marginal and are worthy of further investigation. Body composition was related to hGH level, percentage LBM being positively correlated with hGH levels. These results were significant for all subjects combined and for the fit group, though not the unfit group alone. Consistent with the findings of Experiment 1, it was concluded that physical fitness is not a critical factor influencing sleep variables and that previous studies may have confounded it with other variables. Since body composition is related to physical training it was hypothesized that differences in SWS observed between athletes and non-athletes may be related to differences in body composition. This hypothesis was tested by comparing sleep and anthropometric variables of fit athletes and unfit non-athletes. Two sets of data were analysed. One from the Hobart laboratory included subjects from Experiment 1 plus others run in other experiments at that time (designated Experiment 3a) , while the other consisted of the subjects from Experiment 2 (designated Experiment 3b). Twenty-five fit and 22 unfit subjects were run in Experiment 3a and 17 fit and 17 unfit in Experiment 3b. LBM and fat were estimated using a different method in each experiment. The results showed percentage LBM was negatively correlated to SWS in fit subjects while the amount of LBM and weight were negatively related in the unfit groups. When all subjects were combined within each experiment, significant negative correlations were found between SWS and both LBM and percentage LBM in Experiment 3b. The results, therefore supported the hypothesis that body composition influences SWS levels. Different types of physical training develop different anthropometric characteristics and other physiological attributes. Thus, it was considered possible that different types of training would influence sleep. This hypothesis was tested in Experiment 4. The sleep of four groups of 10 young male subjects who differed with respect to the type of athletic training in which they habitually engaged~was compared on two consecutive, non-exercise nights. The groups were : aerobically trained endurance runners, power trained weight lifters and bodybuilders; athletes with mixed anaerobic, aerobic and power training; and an unfit, non-athletic, sedentary, control group. Pre-planned comparisons showed that the control group did not differ from the combi~ed athletic groups on any sleep variable. However, the aerobic group had more xiii SWS and NREM sleep, slept longer and had shorter SOLs than the power group. The mixed group was intermediate on each of these variables. The data show that the type of physical training in which athletes engage has substantial effects of their sleep. It was not possible to determine from Experiments 3a, 3b and 4 if the effects of type of training and body composition are related. The results, however, demonstrate that variations in peripheral physiological factors are related to sleep architecture; particularly to sws. The results of the experiments reported in this thesis clearly show that aerobic fitness has no direct effect on the sleep variables assessed. They do, however, indicate that peripheral factors associated with physical training are related to aspects of sleep architecture. The results also have implications for theories of sleep. While sleep may serve cerebral restitution and immobilization functions it also appears to be influenced by peripheral factors and thus the findings are inconsistent with present formulations of the cerebral restitution and immobilization theories of sleep. However, despite finding peripheral effects on sleep, the data were not consistent with a general restorative view as the direction of the results was largely incompatible with this theory. Finally, in relation to the energy conservation theory of sleep as applied within species or within individuals, the data indicate that athletes as a broad group do not use the elevation of SWS or TST to compensate for high energy expenditure. A subgroup of athletes (endurance athletes), however, may use this method to balance their energy intake and expenditure.


Publication status

  • Unpublished

Rights statement

Copyright 1985 the Author - The University is continuing to endeavour to trace the copyright owner(s) and in the meantime this item has been reproduced here in good faith. We would be pleased to hear from the copyright owner(s). Thesis (Ph.D.)--University of Tasmania, 1985.

Repository Status

  • Restricted

Usage metrics

    Thesis collection


    No categories selected