Shorter sleep duration is associated with greater visceral fat mass in US adults

https://www.sciencedirect.com/science/ar...389945723000990

Abstract

Habitual declines in sleep duration and increased rates of obesity are public health concerns worldwide. Accumulating evidence suggests a prominent link between reduced sleep duration and weight gain. Our cross-sectional study investigated the relationship between sleep duration and body fat distribution in US adults. We extracted data for 5151 participants (2575 men and 2576 women) aged 18–59 years from the US National Health and Nutrition Examination Survey 2011–2012 and 2013–2014. Weekday or workday night-time sleep duration was estimated using an in-home interview questionnaire. Dual-energy x-ray absorptiometry scans were used to determine regional body fat mass (arms, legs, trunk [android and gynoid], and abdominal [subcutaneous and visceral]). Multiple linear regression and restricted cubic spline analyses were performed after adjusting for several demographic, anthropometric, and nutritional covariates. There was a significant negative association between sleep duration and visceral fat mass overall (β: −12.139, P < 0.001) and by sex (men: β: −10.096, P < 0.001; women: β: −11.545, P = 0.038), after adjusting for age, ethnicity, body mass index, total body fat mass, daily energy and alcohol intake, sleep quality and sleep disorder status. Sleep duration and visceral fat appeared to plateau at ≥ 8 h of daily sleep. Sleep duration is negatively associated with visceral fat mass accumulation during adulthood with possibly no benefits beyond 8 h of sleep per day. Mechanistic and prospective studies are required to confirm the effect of sleep duration on visceral adiposity and determine its causes.

Previous article in issue
Next article in issue
Keywords
Sleep duration
Visceral fat
Obesity
Fat distribution
Body composition
1. Introduction

Obesity is a global health challenge, affecting approximately 600 million adults worldwide. In the United States alone, more than two-thirds of adults are considered obese (body mass index [BMI] ≥ 30 kg/m2), which contributes to substantial metabolic and financial burden [1]. Increased adiposity throughout the lifespan is linked with several metabolic abnormalities such as insulin resistance, type 2 diabetes, non-alcoholic fatty liver, cardiovascular disease, and cancer [2]. The rapid increase in the prevalence of obesity may be associated with intrinsic (i.e. genetics) and extrinsic factors, including physical inactivity and over-nutrition [3]. Accumulation of adipose tissue is sex-specific and interindividual differences in body fat distribution are linked to sex hormone profiles, genetics, and epigenetic mechanisms [4].

Emerging evidence suggests that sleep deprivation may be an important regulator in metabolic health. In the United States, sleep disorders affect approximately 70 million people, a critical issue that may be manifested due to physiological, psychological, and environmental factors [5]. Sleep is regulatory for metabolic function, including immune and hormonal status [6,7]. Specifically, sleep can modulate appetite and thus plays an important role in reducing obesity [8]. Additionally, research has also demonstrated a pronounced effect of sleep regulation on body fat distribution, which may further exacerbate metabolic health [9]. Indeed, sleep disorders and insufficient sleep can alter neuroendocrine system activity, a major mediator of whole-body metabolism [10]. Interestingly, a bidirectional relationship between sleep and body weight changes has been proposed given that weight loss strategies may promote better sleep quality [11] and decrease wakefulness after sleep onset [12]. However, considering the impact sleep disruption may impose on metabolic health, the majority of research has focused on elucidating its influence on body weight regulation.

At present, a plethora of studies have explored the relationship of sleep duration with regional body fat mass, without accounting for whole-body adiposity and the concurrent effect of other regions of adipose tissue. The purpose of this observational study was to comprehensively investigate the potential association between sleep duration with regional body fat mass in US adults.
2. Methods
2.1. Study design and participants

Data from study participants aged 18–59 years were collected from two consecutive survey cycles in NHANES: 2011–2012 and 2013–2014. The cut-off age was selected based on data availability for sleep duration, sleep quality, sleep disorder status, and body fat distribution.
2.2. Sleep duration and regional body fat mass assessment

Quantity of sleep, in terms of duration during the night on weekdays or workdays, was assessed through an in-home interview questionnaire, using the Computer-Assisted Personal Interviewing (CAPI) system. Responses ranged from 1 to 12, with 1 indicating 1 h of sleep to 12 indicating ≥12 h of sleep. Dual x-ray absorptiometry (DXA) scans were administered to assess regional body fat mass using Hologic Discovery model A densitometers. Regional body fat mass areas included the limbs, trunk (android and gynoid), and abdominal (subcutaneous and visceral). Fat mass was quantified in grams (g). Participants with no information on any of the above measures, were excluded from the study.
2.3. Covariates

Age (years), ethnicity (race), BMI (kg/m2), total fat mass (g), daily energy (kcal), and alcohol intake (g) were classified as covariates alongside sleep quality and sleep disorder status. These variables were considered to be potential confounders in the relationship between body fat mass and sleep duration. With the exception of total fat mass, all covariates were considered as categorical data.

Participants were categorized by age into the following groups: 18–29, 30–39, 40–49, 50–59 years. Ethnic groups consisted of Mexican American, other Hispanic, non-Hispanic White, non-Hispanic Black, non-Hispanic Asian and other (multi) race. Energy and alcohol intake were computed through the average of the 24-h dietary recalls and categorized as low, moderate, or high. Low energy intake in men was considered <2000 kcal/d, moderate as 2000–3000 kcal/d, and high >3000 kcal/d. Low energy intake in women was considered <1600 kcal/d, moderate as 1600–2400 kcal/d, and high as > 2400 kcal/d. Low alcohol intake in men was considered <15 g/d, moderate as 15–30 g/d, and high as > 30 g/d. Low alcohol intake in women was considered <10 g/d, moderate as 10–20 g/d and high as > 20 g/d. A BMI of <18 kg/m2 was classified as low, 18–24.9 kg/m2 as moderate, and ≥25 kg/m2 as high. Sleep quality was categorized as Yes/No responses based on self-reported difficulties in sleeping as reported to a general practitioner or other health professional. Sleep disorder status was classified as a Yes/No response based on diagnosis by a doctor or other health professional.
2.4. Statistical analyses

Multiple linear regression analyses were used to assess the association between daily sleep duration and body fat mass by adipose tissue region upon adjustment of all covariates. Sleep duration was treated both as a continuous (Model 1) and as a categorical (Model 2) variable. For the categorical variable, individuals were categorized into four groups by the amount of reported daily sleep: ≤6, 7–8, 9–10 or ≥11 h. Restricted cubic splines were employed to model the non-linear and dose-response relationship between sleep duration and regional body fat mass with three knots after adjustment. Results from the multiple regression analyses were described by sleep duration as a continuous variable. Additional confirmation of the effect and of the dose-response relationship in terms of significance was established using sleep duration as a categorical variable and by assessing the subgroup effect of sex. Collinearity in terms of linear intercorrelation between the explanatory and covariate variables in the multiple regression model was assessed using the variance inflation factor. Statistical significance was established as p < 0.05. Statistical analysis was ensued using IBM SPSS Statistics v28.
3. Results
3.1. Characteristics of study population

Data for sleep duration and regional body fat mass were available for 5151 participants. Baseline characteristics of socio-demographic, anthropometric, and nutritional relevance amongst all participants are outlined in Table 1 and by sex in Tables S1 and S2. When sleep duration was considered as a continuous variable, the study cohort consisted of 5125 participants and excluded 26 individuals with a reported scores of 12 or more hours of sleep duration (Table S3). The total cohort had a mean age of 37.4 years and was composed equally based on sex (≈50%). Participants were primarily non-Hispanic white (38%) and Black (22%). Participants had an energy intake that was primarily within the recommended range of consumption (44%), with low alcohol intake (81%), and high BMI (73%). The majority of participants reported no trouble sleeping (78%) and were mainly free of any sleep disorders (85%).

Table 1. Socio-demographic, anthropometric and nutritional characteristics of all participants (n = 5151). Values are expressed as count (percentage) unless otherwise specified.
Characteristics
Age
18-29 1658 (32)
30-39 1153 (22)
40-49 1198 (23)
50-59 1142 (22)
Sex
Male 2575 (50)
Female 2576 (50)
Ethnicity
Mexican American 698 (14)
Other Hispanic 489 (10)
Non-Hispanic White 1963 (38)
Non-Hispanic Black 1121 (22)
Non-Hispanic Asian 676 (13)
Other Race - Including Multi-Racial 204 (4)
Body mass index
Low 72 (1)
Normal 1676 (33)
High 3403 (66)
Energy intake
Low 1809 (35)
Moderate 2262 (44)
High 1080 (21)
Alcohol intake
Low 4153 (81)
Moderate 412 (8)
High 586 (11)
Arm fat (g)
Minimum 319.4
Average 1667.1 (12)
Maximum 7248.0
Leg fat (g)
Minimum 574.5
Average 4890.5
Maximum 20863.1
Trunk fat (g)
Minimum 2283.1
Average 12889.8
Maximum 47541.6
Abdominal fat (g)
Minimum 241.4
Average 2070.4
Maximum 5893.1
Visceral fat (g)
Minimum 26.8
Average 469.9
Maximum 1918.7
Subcutaneous fat (g)
Minimum 111.0
Average 1600.5
Maximum 5264.6
Android fat (g)
Minimum 296.6
Average 2352.1
Maximum 10626.5
Gynoid fat (g)
Minimum 659.7
Average 4591.0
Maximum 16843.6
Total fat (g)
Minimum 4902.0
Average 27173.9
Maximum 102288.7
Sleep duration (hr)
Minimum 2.0
Average 6.8
Maximum 12.0
0–6 h 2064 (40)
7–8 h 2751 (53)
9–10 h 304 (6)
11–12+ hours 32 (1)
Sleep quality
Trouble sleeping 1157 (23)
No trouble sleeping 3994 (78)
Sleep disorder status
Yes 792 (15)
No 4359 (85)
3.2. Sleep duration and regional body fat mass

Sleep duration was negatively associated with visceral fat mass when expressed as a continuous (Model 1: β = −12.139, P < 0.001) and categorical (Model 2: β = −26.661, P = 0.002) variable after adjustment for all covariates (Table 2). A difference of 1 h in daily sleep duration corresponded to an increase of 12.1 g in visceral fat mass (Model 1). No associations with limb (arms: P = 0.992; legs: P = 0.074), trunk (P = 0.051) (android [P = 0.157] and gynoid [P = 0.600]), and abdominal (P = 0.166) [subcutaneous (P = 0.471)] fat mass were found following adjustment. Subgroup analysis based on sex after adjustment of covariates revealed a significant negative association between sleep duration and visceral fat mass in men (β: −10.096, P < 0.01) and women (β: −11.545, P = 0.038), even after adjustment for covariates (Table 3, Table S4). Dose-response curves indicated a linear relationship between sleep duration and visceral fat mass, following adjustment for covariates (Fig. 1). A plateau in visceral fat mass changes was observed over 8 h of sleep duration per day in the restricted cubic spline of Model 2. No signs of linear multi-intercorrelation between the effect of sleep duration against other covariates on visceral fat mass were observed (Table S5).

Table 2. Multiple linear regression analysis of the association between sleep duration and regional body fat mass after covariate adjustment of age, ethnicity, BMI, total fat mass, daily energy, and alcohol intake, sleep quality and sleep disorder status. Sleep duration was treated both as a continuous (Model 1) and as a categorical (Model 2) variable.
Regional Body Fat Mass Model 1 Model 2a
β P R2 β P R2
Arms 0.025 0.992 0.917 1.785 0.756 0.917
Legs −14.674 0.074 0.884 −23.537 0.194 0.884
Trunk 30.019 0.051 0.951 46.951 0.166 0.951
Abdominal 4.221 0.166 0.924 9.531 0.155 0.924
Visceral −12.139b <0.001∗∗ 0.624 −26.661b 0.002 0.624
Subcutaneous −1.777 0.471 0.926 −2.997 0.581 0.926
Android 5.888 0.157 0.919 14.172 0.122 0.919
Gynoid −3.138 0.600 0.918 −4.452 0.736 0.918

∗∗P = 0.000828.

a

Participants were categorized in to four groups by the amount of reported daily sleep: ≤6, 7–8, 9–10 or ≥11 h.
b

Unstandardized simple linear regression coefficient of sleep duration against predicted visceral fat mass following multiple linear regression analysis.

Table 3. Multiple linear regression analysis of the association between sleep duration and regional body fat mass after covariate adjustment of age, ethnicity, BMI, total fat mass, daily energy, and alcohol intake, sleep quality and sleep disorder status. Sleep duration was treated both as a continuous (Model 1) and as a categorical (Model 2) variable.
Visceral Fat Mass Model 1 Model 2a
βb P R2 βb P R2
Men −10.096 <0.001∗∗ 0.640 −23.379 0.003 0.639
Women −11.545 0.038 0.617 −24.177 0.033 0.617

∗∗P = 0.000548.

a

Participants were categorized in to four groups by the amount of reported daily sleep: ≤6, 7–8, 9–10 or ≥11 h.
b

Unstandardized simple linear regression coefficient of sleep duration against predicted visceral fat mass following multiple linear regression analysis.

https://pubmed.ncbi.nlm.nih.gov/36966579/

Shorter sleep duration is associated with greater visceral fat mass in US adults: Findings from NHANES, 2011-2014
Panagiotis Giannos 1 , Konstantinos Prokopidis 2 , Darren G Candow 3 , Scott C Forbes 4 , Kamil Celoch 5 , Masoud Isanejad 6 , Vanja Pekovic-Vaughan 6 , Oliver C Witard 7 , Brendan M Gabriel 8 , David Scott 9
Affiliations

PMID: 36966579 DOI: 10.1016/j.sleep.2023.03.013

Free article
Abstract

Habitual declines in sleep duration and increased rates of obesity are public health concerns worldwide. Accumulating evidence suggests a prominent link between reduced sleep duration and weight gain. Our cross-sectional study investigated the relationship between sleep duration and body fat distribution in US adults. We extracted data for 5151 participants (2575 men and 2576 women) aged 18-59 years from the US National Health and Nutrition Examination Survey 2011-2012 and 2013-2014. Weekday or workday night-time sleep duration was estimated using an in-home interview questionnaire. Dual-energy x-ray absorptiometry scans were used to determine regional body fat mass (arms, legs, trunk [android and gynoid], and abdominal [subcutaneous and visceral]). Multiple linear regression and restricted cubic spline analyses were performed after adjusting for several demographic, anthropometric, and nutritional covariates. There was a significant negative association between sleep duration and visceral fat mass overall (β: -12.139, P < 0.001) and by sex (men: β: -10.096, P < 0.001; women: β: -11.545, P = 0.038), after adjusting for age, ethnicity, body mass index, total body fat mass, daily energy and alcohol intake, sleep quality and sleep disorder status. Sleep duration and visceral fat appeared to plateau at ≥ 8 h of daily sleep. Sleep duration is negatively associated with visceral fat mass accumulation during adulthood with possibly no benefits beyond 8 h of sleep per day. Mechanistic and prospective studies are required to confirm the effect of sleep duration on visceral adiposity and determine its causes.

Keywords: Body composition; Fat distribution; Obesity; Sleep duration; Visceral fat.

Copyright © 2023 The Authors. Published by Elsevier B.V. All rights reserved.