Poor sleep quality and sleep apnea are associated with higher resting energy expenditure in obese individuals with short sleep duration

Abstract

Context: Epidemiological studies reported an inverse or U-shaped relationship between sleep duration and weight. The relationship between sleep and resting energy expenditure (REE) has not been well characterized.

Objective: The aim of the study was to determine the relationship between sleep, REE, and stress hormones.

Design and setting: We conducted a cross-sectional evaluation of a prospective cohort study at a tertiary referral research clinical center.

Subjects: Subjects included 126 obese individuals (30 males, 96 females; age, 40.5 ± 6.9 yr; body mass index, 38.6 ± 6.5 kg/m(2); sleep duration, 360 ± 50 min/night; and sleep efficiency, 79.5 ± 7.5%).

Main outcome measure(s): REE and respiratory quotient (RQ) were assessed by indirect calorimetry. Sleep duration and sleep efficiency were assessed by actigraphy. Sleep quality was estimated by questionnaires, and sleep apnea was evaluated by respiratory disturbance index (RDI). Morning plasma ACTH, serum cortisol, and 24-h urinary free cortisol and catecholamines were also measured.

Results: RDI was positively correlated with REE adjusted by fat-free mass (r = 0.307; P = 0.003) and RQ (r = 0.377; P < 0.001). Sleep efficiency was inversely correlated with RQ (r = -0.200; P = 0.033). The relationship of RDI score and REE was stronger in men than women (P = 0.03). In women, serum cortisol was positively correlated (r = 0.407; P < 0.001), and Epworth sleepiness score tended to be inversely (r = -0.190; P = 0.086) correlated with adjusted REE. The RQ was positively related to RDI in women, whereas subjective sleep time was related to RQ in men. In a multiple regression model, RDI, serum cortisol, and urinary norepinephrine were directly related to REE, whereas serum cortisol also directly related to adjusted REE.

Conclusion: Poor sleep quality was associated with increased REE, a higher RQ indicating a shift from fat toward carbohydrate oxidation, and activation of the stress system.

Trial registration: ClinicalTrials.gov NCT00261898.

Fig. 1.

Correlation of RDI with REE adjusted (A) and RQ (B). Closed diamonds, women; open circles, men.

Fig. 2.

Correlation of RDI with ACTH (A) and plasma cortisol (B).