Effects of three weeks of mild sleep restriction implemented in the home environment on multiple metabolic and endocrine markers in healthy young men

Abstract

Objectives

Evidence for a causal relationship between sleep-loss and metabolism is derived primarily from short-term sleep deprivation studies in the laboratory. The objective of this study was to investigate whether small changes in sleep duration over a three week period while participants are living in their normal environment lead to changes in insulin sensitivity and other metabolic parameters.

Methods

Nineteen healthy, young, normal-weight men were randomised to either sleep restriction (habitual bedtime minus 1.5 h) or a control condition (habitual bedtime) for three weeks. Weekly assessments of insulin sensitivity by hyperinsulinaemic–euglycaemic clamp, anthropometry, vascular function, leptin and adiponectin were made. Sleep was assessed continuously using actigraphy and diaries.

Results

Assessment of sleep by actigraphy confirmed that the intervention reduced daily sleep duration by 01:19 ± 00:15 (SE; p < 0.001). Sleep restriction led to changes in insulin sensitivity, body weight and plasma concentrations of leptin which varied during the three week period. There was no effect on plasma adiponectin or vascular function.

Conclusions

Even minor reductions in sleep duration lead to changes in insulin sensitivity, body weight and other metabolic parameters which vary during the exposure period. Larger and longer longitudinal studies of sleep restriction and sleep extension are warranted.

Introduction

Development of obesity and type 2 diabetes (T2DM) in humans is multifactorial and there is accumulating evidence from epidemiological studies that besides the traditional risk factors such as excessive fat intake and a sedentary lifestyle, novel risk factors, such as short-sleep duration [1], [2] and disruption of circadian rhythms [3] should be considered. Investigating sleep duration as a risk factor is also supported by recent findings in basic circadian rhythm research which have underscored the interrelatedness of sleep–wake cycles, circadian rhythmicity and metabolism [4], [5].

The current evidence that sleep duration is a causal factor in the development of obesity and/or T2DM is limited. Although prospective cohort studies have repeatedly linked short-sleep to a variety of cardiometabolic risk factors, such as increased body weight, glucose intolerance and high blood pressure (reviewed in [6], [7]) neither causality nor the direction of the causality can be derived from these studies [8]. Furthermore, cohort studies typically rely upon subjective responses to a single sleep question, which usually does not distinguish between “sleep duration” and “time in bed”; two measures which may not always have a linear relationship to each other [9].

Laboratory studies in which sleep duration has been reduced under controlled conditions have provided evidence that acute sleep restriction leads to changes in appetite regulating hormones such as leptin and changes in insulin sensitivity and other metabolic parameters. Although these acute studies provide a potential mechanism whereby sleep duration and obesity may be linked, the relevance of these findings has been questioned. The sleep restriction in most laboratory studies has been extreme (0 to 4 h of sleep duration) [10], [11], [12] often compared to a longer than normal sleep duration (10 h) [13] and only imposed for a short period (1 to 14 days) [13]. Therefore the findings of these protocols may not be transferrable to the link between sleep and obesity in society. The lack of adequate interventional data has been highlighted repeatedly [14], [15], [16] and in this study we conducted a controlled trial in healthy lean individuals using a moderate level of sleep-loss (1.5 h per night) over a period of 3 weeks while participants were living in their normal environment, using a similar sleep protocol to that previously described by Zielinski and colleagues in older long-sleepers [17], [18]. We also used a parallel group design in which the sleep restriction was compared to a control group with no change in sleep duration, rather than a cross-over design in which the effects of sleep restriction are compared to a period of recovery sleep-extension or recovery [19]. The aims in this study were, (i) To assess whether sleep duration assessed using sleep diaries and actigraphy – a reliable and validated measure of sleep duration [20] – could be manipulated adequately at home rather than in a controlled sleep-laboratory environment in healthy young men, (ii) to assess whether a smaller amount of sleep loss (1.5 h), over a period of three weeks, has a an impact on insulin sensitivity as assessed through hyperinsulinaemic–euglycaemic clamp, the gold-standard approach and (iii) to assess the time-course of the changes in sleep and metabolic parameters over the three week period.