Quality Sleep Is Associated With Overnight Metabolic Rate in Healthy Older Adults

Abstract

Background:

Increasing age is associated with an increase in overnight metabolic rate.

Objective:

To determine the relationship between quality sleep, sleep efficiency, and overnight metabolic rate as measured in a respiration chamber in older participants.

Methods:

The study design was cross sectional. Forty participants, aged 50 to 83 years (17 males, age 63±7 years, body mass index 25.7±2.3kg/m²) spent one night in a respiration chamber to measure sleep stages by polysomnography and overnight metabolic rate (OMR). Data were collected between 23:00 and 07:00. Subsequently basal metabolic rate (BMR) was measured under a ventilated hood. Quality sleep was calculated as time spent in rapid-eye movement (REM) sleep and slow wave sleep divided by total sleep time, and sleep efficiency was calculated as total sleep time divided by the sleep period time. Body movement was measured between 23:00 and 07:00 with an accelerometer on the wrist. Overnight metabolic rate was adjusted for body size by dividing by basal metabolic rate (OMR/BMR).

Results:

OMR/BMR was positively associated with age (r = 0.48, p < .001), and quality sleep was negatively associated with age (r = −0.51, p < .001). The variance of OMR/BMR was significantly explained by quality sleep (r = −0.58, p < 0.001). Body movement was negatively related to sleep efficiency (r = -0.38, p<0.01) with no effect on OMR/BMR. When OMR/BMR was adjusted for quality sleep, the effect of age was non significant.

Conclusion:

Quality sleep is inversely associated with the age-related rise in overnight metabolic rate, suggesting that increased overnight metabolic rate is a biological sign of ageing as a consequence of diminished quality sleep.

Sleep is a temporary state of altered consciousness that occurs for approximately one third of human life. Sleep promotes growth, recovery, and cognitive well-being. Reduced total sleep time (TST) affects endocrine and metabolic functions and sleep-deprived participants can show confusion, depression, hallucinations, and in extreme cases death (1–9) During sleep, slow wave sleep (SWS) is the stage associated with growth and recovery (10). Rapid-eye movement (REM) sleep is associated with cognitive functions and its selective deprivation is associated with executive functions, pain threshold, and emotional memory consolidation (11–13). Studies that selectively deprived participants of SWS or REM showed a rebound effect during the period following the deprivation, underlying the importance of both SWS and REM.

Aging is associated with a reduction of TST, REM, and SWS. After 30 years of age, wake time after sleep onset increases 10min per decade, reducing the sleep time between sleep onset and final awakening, known as sleep efficiency, from 95% in young adults to less than 80% in older adults (14,15). Similarly, quality sleep, defined as the sum of rapid REM and SWS, progressively decreases from 60% of TST in children to 20% in old adults (14,15). Older individuals also show a reduction of basal metabolic rate (BMR) with age, while overnight metabolic rate (OMR) is preserved or increased. The reduction in BMR is a sign of decreased fat-free mass, and it had been assumed that OMR would proxy this trend and decrease with age. On the contrary, previous studies showed that OMR is lower than BMR in 11-year-old children, similar to BMR in 25-year-old adults and higher than BMR in 61 year participants (16). Changes in quality sleep or sleep efficiency have possible consequences for OMR that would indicate that OMR is a biological sign of aging.

Possible explanations for an increased OMR in older individuals include less quality sleep and decreased sleep efficiency. Some studies observed a lower metabolic rate during quality sleep, particularly SWS (17–19). Older adults spend less time in quality sleep and this could contribute to higher OMR. Additionally, body movement is reduced during sleep time (20). Older adults with lower sleep efficiency spend more time awake and are more likely to move during the night with a consequent increase in OMR (21,22).

This study aimed at determining the relationship between quality sleep, sleep efficiency, and metabolic rate in older participants staying overnight in a respiration chamber. Sleep was assessed with polysomnography to determine sleep and wakefulness time as well as to allow the distinction between different sleep stages. Body movement was assessed simultaneously with actigraphy. The respiration chamber allowed the assessment of OMR, followed by measurement of BMR under a ventilated hood (16).

Participants and Methods

Population

Participants were 17 men and 23 women, aged 50 to 83, 63±7 years, recruited by advertisements in local newspapers (Table 1). They underwent a medical screening including measurement of body weight and height and the completion of a questionnaire related to health, use of medication, physical activity, alcohol consumption, smoking, and sleeping behavior. Good health was required for inclusion, while exclusion criteria were medications, sleep complaints, smoking, or heavy drinking. All participants included in the study were in good health, not taking medication, nonsmokers, at most moderate alcohol consumers, and with no complaint of sleep disorders such as sleep apnea. The study was conducted according to the Declaration of Helsinki, and the Ethics Committee of the Maastricht University Medical Center approved the study. The trial was registered at www.clinicaltrials.gov as NCT01609764.

Study design

On the day of the experiment, participants arrived at the university at 19:00. Electrodes for polysomnographic measurements were applied and a triaxial accelerometer was placed on the right wrist. At 21:00, participants entered a respiration chamber and were instructed to sleep between 23:00 and 07:00. BMR was measured directly after leaving the respiration chamber between 07:00 and 08:00 and before any consumption of any food or beverage. At the end of the visit, participants were asked to evaluate how good they had slept using a visual analog scale consisting of a 10-cm line marked from 0 (very bad) to 100% (very good).

Polysomnography

Polysomnographic recordings were obtained between 23:00 and 07:00 with a digital electroencephalogram system (BrainRT, OSG BVBA, Rumst, Belgium). Electrodes for electroencephalogram, bilateral electrooculograms, and submental electromyograms were applied according to standardized criteria (23–27). Sleep records were visually scored in 30s epochs by an experienced researcher following the standard guidelines (23–27). Stages included awake, REM sleep, non-REM sleep stage 1, non-REM sleep stage 2, and SWS. Sleep period time was defined as the time between sleep onset and last awaking, including wake time after sleep onset. TST was defined as the time between sleep onset and last awaking, excluding wake time after sleep onset. Body movement during the night was monitored with a triaxial accelerometer (GT3X+, ActiGraph LLC, Pensacola, FL) on the right wrist.

Energy Expenditure

OMR was measured in a respiration chamber. The respiration chamber is an airtight room sized 18 m³ and equipped with a bed, table, chair, freeze toilet, washing bowl, and television. During the stay in the respiration chamber, oxygen consumption and carbon dioxide production were measured according to the protocol used in previous studies (26–28). BMR was measured for 30min under a ventilated hood (29) in the supine position under standard conditions of wake, rest, fasting, immobility, thermoneutrality, and mental relaxation.

Data Analysis

Quality sleep was calculated as time spent in REM sleep and SWS divided by TST, and sleep efficiency was calculated as TST divided by the sleep period time. Body movement was calculated from the accelerometer as activity counts averaged between 23:00 and 07:00. Calculation of metabolic rate from measured oxygen consumption and carbon dioxide production was based on Brouwer’s formula (30). OMR was calculated as the average energy expenditure between 23:00 and 07:00. BMR was calculated as the average energy expenditure during the last 20min under the ventilated hood. To adjust for body size, OMR was divided by BMR (OMR/BMR) (16,31).

Statistical Analysis

Pearson product–moment correlation coefficient (r) was used to quantify associations between variables. All variables were expressed as mean ± standard deviation. The significance threshold was set to p < .05.

Results

A typical example of chronological display of the sleep stages (hypnogram) is presented in Figure 1, aligned with energy expenditure and body movement. Sleep period time was on average 434±67min, wake time after sleep onset was 64±50min, and TST was 370±84min. Sleep efficiency was 85±13% (Table 2). Quality sleep was 41±10% of the TST; 69±26min was spent in REM (18±6% of TST) and 84±37min in SWS (23±9% of TST). OMR was on average 4.6±0.7 kJ/min and BMR was 4.4±0.6 kJ/min resulting in an OMR/BMR of 1.06±0.06 (Tables 1 and 2).

OMR/BMR was positively associated with age (r = 0.48, p < .001) and quality sleep was negatively associated with age (r = −0.51, p < .001; Table 3). The variance of OMR/BMR was partly explained by quality sleep (r = −0.58, p < .001; Figure 2). Body movement was negatively related to sleep efficiency (r = −0.38, p < .01) with no effect on OMR/BMR (Table 3).

When OMR/BMR was adjusted for quality sleep, the association with age was nonsignificant (r = 0.24). Sleep scores from the visual analog scale were negatively associated with age (r = −0.40, p < .01) and OMR/BMR (r = −0.35, p < .05) and were related to both quality sleep (r = 0.44, p < .01) and sleep efficiency (r = 0.41, p < .01).

Discussion

This study demonstrated for the first time that quality sleep was inversely associated with the age-related rise in OMR/BMR and when OMR/BMR was adjusted for quality sleep, the effect of age was nonsignificant. Body movement increased with reduced sleep efficiency with no consequence for OMR/BMR.

On average, quality sleep was 41±10% of TST. Quality sleep was negatively associated with age. This is in line with previous observations. In a study on 2,685 participants aged 62±11 years, quality sleep was about 38% of TST (32). The same study reported that quality sleep was significantly reduced in participants older than 61 years, compared with younger adults. Similar results were shown previously for smaller groups (33–35). In addition, OMR was on average 4.7 kJ/min, which is comparable with the previous measures in older adults where OMR was on average 4.6 kJ/min (16).

Overnight metabolic rate was on average 6% higher than BMR and increased with age; participants in their sixth decade showed similar OMR and BMR, while in participants in their eighth decade OMR was around 10% higher than BMR. Previous studies showed that overnight metabolic rate is similar to BMR in young adults (16,18,21,36). Some studies also showed that OMR/BMR is lower in children and higher in old adults (16,37). Our results suggest that this is partly determined by diminished quality sleep. When OMR/BMR was adjusted for quality sleep, the effect of age was nonsignificant, indicating that the increase of OMR/BMR with age is mediated by quality sleep and that there is no direct effect of age on OMR/BMR.

This study showed that lower OMR/BMR is associated with more quality sleep. This result is consistent with previous studies where quality sleep, particularly SWS, was associated with a reduction of metabolic rate of 2%–3% (17–19). While some studies report that this reduction is nonsignificant, SWS is consistently reported as the stage with the lowest metabolic rate, often followed by REM (27,38,39), especially in comparison with wakeful moments in between periods of time sleeping (27). The association between OMR/BMR and quality sleep supports the hypothesis that metabolic rate decreases during quality sleep relative to increases during wakeful moments in between periods of time sleeping. One possible mechanism to explain increased OMR/BMR and decreased quality sleep with increasing age is altered autonomic regulation. Autonomic regulation is responsible for regulating physiological parameters such as blood pressure, which is typically elevated in older adults. High blood pressure overnight has been associated with diminished sleep quality and altered sleep duration (40) and could result in increased OMR. Thus increased OMR/BMR might be one of the signs of age-related alterations of autonomic regulation.

Reduced sleep efficiency has been previously suggested as an explanation for the increased OMR/BMR in older individuals (16). Aging reduces sleep efficiency and increases time spent awake during the night (15). Time spent awake is associated with increased body movement and thus increased metabolic rate (21,22). In the population of this study, increased body movement was associated with lower sleep efficiency but with no effect on OMR/BMR. Overnight body movement intervals were short and the increase in metabolic rate was possibly too short to significantly increase OMR. Furthermore, sleep efficiency in this population was higher than previously reported in older adults (14,32,41). Increased sleep efficiency could have reduced the effect of aging, hence the correlation between sleep efficiency and age was not significant. Similar studies in populations of less efficient sleepers could reveal whether sleep efficiency and body movement are mediators between age and OMR/BMR, together with quality sleep.

Changes in sleep architecture have been related to aging, but this is the first study to associate these changes with OMR (14,15). Results show not only that both quality sleep and OMR/BMR increase with age but also that quality sleep and OMR/BMR are related to each other. This suggests that increased OMR/BMR is a biological sign of aging as a consequence of diminished quality sleep. The association of increased OMR/BMR with adverse outcomes needs further studies.

One limitation of this study is that all participants were healthy and the effect of age-related conditions was not explored. Among these conditions, breathing disorders, such as sleep apnea, high blood pressure, and dementia are associated with decreased sleep and quality sleep and might further increase OMR/BMR in older individuals. Additionally, the high sleep efficiency measured in this population might have reduced the effect of aging on OMR/BMR, thus limiting the conclusion drawn from the results.

In conclusion, quality sleep is inversely associated with the age-related rise in OMR, suggesting that increased OMR is a biological sign of aging as a consequence of diminished quality sleep.

Funding

The research was funded by Maastricht University.

Conflict of Interest

The authors have no conflict of interest to declare.

Acknowledgments

K.R. Westerterp and G. Valenti designed the study; G. Valenti collected the data, analyzed the data, and wrote the manuscript; K.R. Westerterp and A.G. Bonomi contributed to the interpretation of the data and reviewed the manuscript; and the study was executed under the supervision of K.R. Westerterp. Everyone who contributed significantly to the work has been listed; all authors read and approved the final manuscript.

References

Author notes