The effects of partial sleep deprivation on energy balance: a systematic review and meta-analysis. HK Al Khatib, S V Harding, J Darzi, GK Pot. European Journal of Clinical Nutrition (2016); DOI: 10.1038/ejcn.2016.201

In a nutshell
This well conducted systematic review and meta-analysis suggests that partial sleep deprivation may increase energy intake, whilst having no significant effect on energy expenditure, leading to a net positive energy balance.

Background
It has been suggested that average human sleep duration has declined over the past century. Observational studies suggest that self-reported short sleep duration (typically defined as ranging from <4 to <8 hours) and sleep disorders are associated with weight gain and non-communicable diseases, including obesity, hypertension, type 2 diabetes and cardiovascular disease.
Whether the relationship between short sleep duration and weight gain is causal is still uncertain, but has been examined in a number of short term intervention studies looking at the impact of sleep deficiency on the energy balance equation; energy intake (as well as appetite regulating hormones) and energy expenditure (including resting metabolic rate, basic metabolic rate and physical activity).

Study objective
This systematic review and meta-analysis was aimed at evaluating and quantifying the effects of partial sleep deprivation (restricted but not complete elimination of sleep) on energy intake and energy expenditure. Secondary outcomes included looking at changes in macronutrient intake and resting metabolic rate.

Study design
A systematic search of publications (until mid-November 2014) was performed to find all relevant randomised or non-randomised human intervention studies that examined the effect of partial sleep deprivation on 24 hour total energy intake, ad libitum energy intake and/or 24 hour total energy expenditure in healthy adults (no chronic disease or sleep conditions) of all Body Mass Index (BMI) ranges.

Main findings
The search identified 16 eligible studies (including 496 subjects), 11 of which had enough data to be included in the meta-analyses.
The vast majority of studies were performed in the laboratory setting and most were short in duration (less than a week spent in the partial sleep deprivation and control conditions). Partial sleep deprivation ranged from 3 ½ - 5 ½ hours sleep per night, whilst the control conditions ranged from 7 – 10 or 12 hours per night.

  • Meta-analysis on total 24-hour mean energy intake
    Ten studies were included, 7 showed an increase in total 24h energy intake, 3 showed no effect. Pooled mean increase in total 24-hour energy intake, compared to the control condition, was 385 kcal (95% CI 252, 517; p <0.00001, low heterogeneity).
  • Meta-analysis on total 24-hour total energy expenditure
    Five studies were able to be included in the meta-analysis, 2 studies observed an increase and 3 reported no effect. Pooled data found no significant change in 24-hour energy expenditure after partial sleep deprivation (88 kcal 95% CI -21, 198; p = 0.11, low heterogeneity).
  • Meta-analysis on macronutrient intake
    The meta-analysis, which included 7 studies, found a pooled mean increase in fat intake (1.6% of energy, 95% CI 0.3, 2.9; p = 0.02, low heterogeneity), decrease in protein intake (-0.8% of energy, 95% CI -1.5, -0.1; p= 0.02, low heterogeneity) and no significant difference in carbohydrate intake after partial sleep deprivation, compared to the control condition.
  • Meta-analysis on resting metabolic rate
    Four studies assessed the effects of partial sleep deprivation on resting metabolic rate but no significant differences between conditions were found.

Author conclusions
Partial sleep deprivation may result in an increased energy intake, leading to a net positive energy balance of 385 kcal per day. In the long term, this may implicate weight gain; however, further research conducted in free-living conditions is needed to understand the effect of partial sleep deprivation on body weight.

Limitations
There were limitations with the evidence-base identified, including variation in the methodology used in the studies (e.g. intervention duration and degree of sleep deprivation and sleep time in the control group). Most studies were short in duration and conducted in the laboratory setting. Therefore, it is unclear what the effect of partial sleep deprivation may be in free-living situations and whether there is any adaptation in the long term. The mechanism by which sleep deprivation causes an increase in energy intake also needs to be elucidated.

Not all of the identified studies could be included in the meta-analysis as the reported data was inextractible and unavailable when the study authors were contacted. As some of the studies which couldn’t be included observed no effect on total 24 hour energy intake, this could have resulted in an overestimation of energy intake when the meta-analysis was performed.

Subgroup analysis was also not possible with this relatively small data set. Further research in diverse populations such as ethnic minorities, older and obese populations would be of benefit in increasing our understanding of this area.

Relevance
Sleep is important for our mental and physical health. These findings support the view that short sleep duration may increase the risk of excess weight gain and obesity. This meta-analysis found energy intake to increase by 385 kcal per day after partial sleep deprivation, compared to control conditions. However, further research is needed to see how transferable these findings are to free-living situations and to understand the effect and implications of this in chronically sleep-deprived population groups. Research also needs to be performed to determine whether the opposite effect is true - if increasing sleep in habitual short sleepers can reduce calorie consumption and play a role in obesity prevention.