Associations Between Lifestyle Characteristics and the Presence of Nonalcoholic Fatty Liver Disease: A Case

Abstract

Background:

Dietary and physical activity (PA) habits have been suggested as important factors for nonalcoholic fatty liver disease (NAFLD). Published data are mainly focused on the effect of either diet or exercise, whereas data on other aspects like sleep remain sparse. The aim of this study was to explore potential associations between dietary intake, PA, and sleeping habits, and the presence of NAFLD.

Methods:

One hundred patients with ultrasound-proven NAFLD and 55 healthy controls matched for age, sex, and body mass index were included. Dietary habits were assessed through a semiquantitative validated food frequency questionnaire. PA level was assessed with a validated questionnaire. Total night sleep hours and duration of midday rest were also recorded. Optimal sleep duration was defined as sleep hours ≥7 and ≤9 hr/day.

Results:

Patients compared to controls consumed less vegetables and nuts, more sweets, drank less coffee and alcohol (all P < 0.05), and exhibited a lower level of PA (P = 0.006). PA level [odds ratio (OR) per 100 metabolic equivalent of task-min/day = 0.74, 95% confidence interval (CI) 0.61–0.89, P = 0.002] was associated with lower probability of having NAFLD, whereas sweets consumption (OR = 2.13, 95% CI 1.22–3.71, P = 0.008) was associated with increased probability, after adjusting for several confounders, including body weight status. Optimal sleep duration was marginally and inversely associated with NAFLD presence (OR = 0.38, 95% CI 0.14–1.01, P = 0.05).

Conclusion:

Higher PA level and optimal sleep duration are associated with lower likelihood, whereas sweets consumption is associated with higher likelihood of having NAFLD. These associations are independent of body weight status and energy intake.

Background

Nonalcoholic fatty liver disease (NAFLD) is a pathological condition of liver fat accumulation, not caused by alcohol abuse or other steatosis causes, ranging from simple steatosis to necroinflammatory steatohepatitis (NASH).^1,2 NAFLD represents the most common cause for chronic liver diseases in Western countries,³ following the rising obesity prevalence.⁴ It is considered the hepatic manifestation of metabolic syndrome,⁵ having insulin resistance as the key underlying pathophysiological mechanism. NAFLD patients are at increased risk for cardiovascular diseases⁶ and type 2 diabetes mellitus,⁷ appearing to have higher mortality from both liver- and nonliver-related causes, compared to the general population.⁸

Positive energy balance, unhealthy food choices, and physical inactivity have been linked to increased NAFLD risk.⁹ Regarding food groups, high consumption of sugar-sweetened beverages and soft drinks combined with high energy intake,^10,11 refined carbohydrates,¹² as well as red and processed meats^13,14 consumption have been associated with increased NAFLD likelihood. On the contrary, high consumption of whole-grain foods,¹² fruits, vegetables, and coffee drinking¹⁵ have been linked to favorable effects on NAFLD. Possible underlying mechanisms seem to be the effect of foods on insulin resistance, inflammation, lipid profile, and weight control.¹⁶

In addition, NAFLD patients seem to spend more time on sedentary activities^17,18 and have lower physical activity (PA) levels compared to healthy controls.¹⁹ Higher PA level has been associated with lower serum triglycerides (TG), weight control, and higher high-density lipoprotein cholesterol (HDL-C) levels in healthy subjects, making them important factors mediating NAFLD development.¹⁵ Recently, sleep disturbances have been linked to increased obesity,²⁰ insulin resistance,²¹ and cardiometabolic risk,²² whereas studies exploring the effect of sleep habits on NALFD risk are limited.^23
–25

Published data focus on associations between one aspect of lifestyle, for example, dietary intake or PA, and NAFLD. Other aspects of lifestyle such as sleep remain fragmentary. Thus, the aim of this study was to explore potential associations between lifestyle characteristics, namely dietary intake, PA, and sleep habits, and the presence of NAFLD, by considering also body weight status, an important NAFLD risk factor.

Materials and Methods

One hundred consecutive adult patients (18–67 years old) with newly diagnosed NAFLD (within 6 months), who visited two outpatient liver clinics (Department of Gastroenterology, Athens University Medical School, Laiko Hospital of Athens and Second Department of Internal Medicine, Athens University Medical School, Hippokration Hospital of Athens) between May 2013 and December 2015, and fulfilled the inclusion criteria were included in this analysis.

NAFLD diagnosis was based on the following criteria: elevated alanine aminotransferase (ALT) and/or gamma-glutamyl transpeptidase (GGT) levels, evidence of hepatic steatosis on ultrasound and/or compatible liver histology, and no other causes of liver injury or steatosis. Patients were excluded in case of positive serological markers for hepatitis B (HBsAg), hepatitis C (anti-HCV), and human immunodeficiency virus (anti-HIV), weekly alcohol consumption more than 210 g for men or 140 g for women or use of potentially hepatotoxic agents, evidence of metabolic or autoimmune liver disease, and presence of known systemic disease with potential liver involvement. Moreover, patients with BMI > 40 kg/m² were also excluded. Equally, patients who had changed their eating habits since NAFLD diagnosis, were on a weight loss diet, and had diabetes mellitus or other diagnosed malignancy, were not enrolled.

Information regarding HBsAg, anti-HBc, anti-HBs, anti-HCV, anti-HIV, and liver autoantibodies (anti-nuclear, anti-smooth muscles, anti-microsomal, anti-mitochondrial) was retrieved from medical records and recent laboratory data. The history of alcohol use was taken and confirmed by patients' relatives or friends. Blood pressure was measured in a sitting position, and hypertension was defined as systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg, or antihypertensive medication treatment. After baseline evaluation, NAFLD patients were encouraged to participate in an ongoing randomized clinical trial, aiming at evaluating the potential benefit of interventions based either on the Mediterranean diet or Mediterranean lifestyle compared to usual care (ClinicalTrials.gov ID: NCT01894438).

A group of 55 control subjects was also enrolled. Controls were subjects who visited other outpatient clinics of the same hospitals for routine examinations, or hospital or university employees, during the aforementioned study period. Controls had normal glucose metabolism and liver enzymes, no evidence of hepatic steatosis at ultrasonography, and stable weight, dietary, and exercise habits during the last year. The study aimed at enrolling a control group, not differing from the patients' in age (±3 years), sex distribution, and body mass index (BMI) (±1 kg/m²). Matching controls for BMI were considered very important because body fat has been proven as an important NAFLD risk factor, however, finding obese people without steatosis was really hard and that is why a 1:1 matching was not accomplished.

The study was approved by the Ethics Committees of Laiko General Hospital of Athens and Harokopio University, and was carried out in accordance with the Declaration of Helsinki.²⁶

Lifestyle assessment

Participants' (both cases and controls) habitual dietary intake over the last 6 months was assessed through a semiquantitative 69-item food frequency questionnaire (FFQ) validated for the Greek population.²⁷ Consumption of main food groups, namely total, low and full fat dairy products; total, refined, and nonrefined cereals; red meat, poultry, fish, legumes, vegetables, fruits, sweets, soft drinks, and nuts were calculated in servings per day. Servings were determined based on the dietary guidelines for Greek adults published in 1999.²⁸ Alcohol consumption was expressed in grams of ethanol/day, coffee intake in times consuming coffee per day, sweets consumption in servings per day (1 serving = 1 scoop/piece ice cream, 60 g chocolate, 3–4 pieces of biscuits, etc.), and soft drinks as servings per day [(1 serving = 1 can (330 mL)]. Dietary intake was also assessed through two nonconsecutive 24-hr dietary recalls, which were analyzed in terms of energy, macronutrient, and micronutrient intakes, using Nutritionist Pro version 2.2 (Axxya Systems).

The PA level was assessed through a validated for the Greek population questionnaire [Athens Physical Activity Questionnaire—APAQ²⁹]. Total volume of PA was determined by calculating total metabolic equivalent of task (MET) per minutes per day (MET-min/day). MET-min/day was also calculated for moderate intensity PA, such as playing tennis or doing yoga. Duration of leisure time sedentary activities, like TV viewing, computer games, and reading was also estimated.

Habitual total night sleep hours, as well as duration of midday naps (siesta) or other ways of midday resting (e.g., watching television or lying on the couch), were recorded both for weekdays and weekends. Short siesta was defined as siesta or midday resting ≤1 hr/day and optimal sleep duration was defined as sleep hours ≥7 and ≤9 hr/day.³⁰

Information regarding smoking habits and routine medication use was also recorded for both patients and healthy controls.

Anthropometric assessment

Body weight was measured with digital scale (Seca Robusta 813, Hamburg, Germany) to the nearest 100 g and height with stadiometer to the nearest 0.5 cm. BMI was calculated as weight (kg) divided by height squared (m²). Waist circumference (WC)³¹ was tape measured to the nearest 0.1 cm. Increased WC was defined as >102 cm for men and >88 cm for women.

Biochemical and inflammatory markers

Blood collection was performed in both patients and controls after a 12-hr fasting period. Serum and plasma samples were immediately frozen at −80°C. Glucose, total cholesterol, and HDL-C were measured using enzymatic colorimetric method (Cobas^® 8000 analyzer; Roche), and low-density lipoprotein cholesterol (LDL-C) was calculated using the Friedewald formula.³² TG were measured with chromatometric enzymatic method (Cobas 8000 analyzer; Roche) and insulin with chemiluminescence (E170 modular analyzer; Roche). The Homeostatic Model Assessment of Insulin Resistance (HOMA-IR) was calculated using the formula by Matthews et al.³³ Metabolic syndrome presence was defined according to the criteria published by Alberti et al.³⁴

ALT, aspartate aminotransferase (AST), and GGT were measured with chromatometric enzymatic method (Cobas 8000 analyzer; Roche). High-sensitivity C-reactive protein (hsCRP) was measured using a nephelometric assay (BN II nephelometer; Siemens). Tumor necrosis factor-a (TNF-a) and adiponectin were measured by sensitivity enzyme-linked immunosorbent assay (ELISA, Quantikine/immunoassay kit; R&D Systems, Minneapolis). The intra-assay coefficient of variation was <7% for TNF-a and <5% for adiponectin. The interassay coefficient of variation was <8% for both TNF-a and adiponectin measurements.

Shear wave elastography

Eighty five of the 100 patients had reliable liver stiffness measurements (LSM) (in kPa) by two-dimension liver shear wave elastography, performed using the Aixplorer^® MultiWave™ ultrasound system (Supersonic Imagine S.A., Aix-en-Provence, France) with the single crystal curved probe at frequencies 1–6 MHz. The examination was considered reliable, if at least five valid measurements were obtained based on good sonographic signal combined with <25% ratio of SD to the mean LSM (mean value of all valid measurements per examination).

NAFLD patients were divided according to elastography results to those of LSM ≤6.6 and LSM >6.6, suggesting the presence of no fibrosis (F ≥ 2) or significant fibrosis (F ≥ 3 or F ≥ 4), based on the cutoff values published by Abenavoli and Beaugrand.³⁵

Liver histology

Liver biopsies were available only for 32 patients. Liver specimens were evaluated by a single hepatopathologist, who was unaware of the clinical data. A liver biopsy was considered adequate, if at least six portal tracts were identified and the specimen length was ≥1.5 cm. NASH diagnosis was based on the criteria of Brunt et al.³⁶ modified by Kleiner et al.³⁷

Statistical analysis

Statistical power analysis for defining the adequate sample size had revealed that, to test for an effect size equal to odds ratio (OR) = 0.5 at significance level of 5% and achieving power higher than 80%, 55 subjects in each treatment group were needed [calculations performed in STATA software (Texas)].

Continuous variables are presented as mean value ± standard deviation and categorical variables as absolute frequencies. The normality of the data was assessed using Shapiro–Wilk test and graphically through histograms. Chi-squared test, two-sample independent t test, or Mann–Whitney U test was used, where appropriate, to test differences between patients and controls. Logistic regression analysis was used to estimate the association between lifestyle parameters and the likelihood of NAFLD presence and liver stiffness severity, after controlling for multiple confounders. All reported P-values were based on two-sided tests and compared to a significance level of 5%. Statistical Package for Social Sciences software (version 21.0; SPSS 2012, Chicago) was used for all statistical analyses.

Results

Descriptive characteristics of both patients and controls are presented in Table 1. Mean age of both patients and controls was 45 years, whereas the majority of the participants were males and overweight (Table 1). Although the two groups did not differ in age, sex, and BMI, patients exhibited higher WC, higher systolic and diastolic blood pressure, higher HOMA-IR, TGs, TNF-a, and hsCRP levels and lower adiponectin and HDL-C levels (all P < 0.05). As expected, liver enzymes, namely ALT, AST, and GGT, were higher in patients (all P < 0.05). Liver biopsies were available only for 32 patients. From them, 10 were diagnosed with simple fatty liver, 21 with NASH, and 1 with cirrhosis. Patients with NASH compared to patients with simple fatty liver had higher BMI (29.11 ± 3.02 kg/m² vs. 27.05 ± 2.73 kg/m²) and waist circumference (102.6 ± 10.0 cm vs. 96.2 ± 7.8 cm), and higher levels of insulin (15.33 ± 5.15 μIU/mL vs. 10.0 ± 5.34 μIU/mL) and HOMA-IR (3.51 ± 1.33 vs. 2.19 ± 1.25) (all P < 0.05).

Table 1.

Descriptive Characteristics of Patients with Nonalcoholic Fatty Liver Disease and Healthy Controls

Variables	NAFLD patients (n = 100)	Controls (n = 55)	P^a
Age (years)	45.51 ± 11.6	44.82 ± 12.06	0.73
Gender, males, n (%)	72 (72.0)	34 (62.0)	0.21
Weight (kg)	84.01 ± 12.08	81.85 ± 12.25	0.29
Body mass index (kg/m²)	28.32 ± 2.6	27.89 ± 3.51	0.11
Waist circumference (cm)
Males	101 ± 8	97 ± 8	0.02
Females	96 ± 8	96 ± 11	0.89
Increased waist circumference,^b n (%)	56 (56.0)	22 (40.0)	0.04
Smoking habits (yes), n (%)	27 (27)	24 (43.6)	0.05
Former smokers, n (%)	26 (26)	12 (21.8)	0.15
Education level
≤Secondary school level	17 (17)	7 (12.7)	0.57
Primary and technical school level	38 (38)	25 (45.5)
≥University level	45 (45)	23 (41.8)
Clinical and biochemical characteristics
Hypertension, n (%)	31 (31.0)	5 (9.1)	0.002
Metabolic syndrome, n (%)	33 (33.0)	5 (9.1)	0.001
Antihypertensive drugs (yes), n (%)	25 (26.9)	6 (10.9)	0.02
Hypolipidemic drugs (yes), n (%)	12 (13.3)	5 (9.3)	0.59
Total cholesterol (mmol/L)	5.38 ± 1.06	5.32 ± 1.15	0.73
HDL-C (mmol/L)	0.98 ± 0.47	1.29 ± 0.43	<0.001
LDL-C (mmol/L)	3.73 ± 1.03	3.49 ± 0.95	0.25
Triglycerides (mmol/L)	1.43 ± 0.63	1.15 ± 0.67	0.001
HOMA-IR	2.90 ± 1.84	1.92 ± 1.55	<0.001
ALT (IU/L)	69.25 ± 45.96	15.27 ± 5.78	<0.001
AST (IU/L)	37.79 ± 19.38	24.79 ± 6.93	<0.001
GGT (IU/L)	105.23 ± 113.13	20.24 ± 12.20	<0.001
Adiponectin (μg/mL)	6.16 ± 4.73	9.87 ± 7.83	0.001
TNF-a (pg/mL)	4.21 ± 2.98	2.35 ± 3.52	<0.001
hsCRP (mg/L)	1.95 ± 2.63	1.24 ± 1.14	0.02
Liver stiffness (kPa) (n = 85)	7.2 ± 4.7	—	—

Data are presented as mean ± SD or otherwise mentioned.

P-value as derived by chi-square test for nominal variables, two-sample independent t-test for normal distribution variables, or Mann–Whitney U test for nonparametric variables. Figures in bold are statistically significant.

>102 cm for males and >88 cm for females.

ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma-glutamyl transferase; HDL-C, high-density lipoprotein cholesterol; HOMA-IR, Homeostatic Model Assessment of Insulin Resistance; hsCRP, high-sensitivity C-reactive protein; LDL-C, low-density lipoprotein cholesterol; NAFLD, nonalcoholic fatty liver disease; TNF-a, tumor necrosis factor-a.

Comparisons of lifestyle characteristics between patients and controls are presented in Table 2. Patients compared to controls consumed less vegetables and nuts, more sweets, and drank less coffee and alcohol (all P < 0.05). According to total MET-min/day, patients exhibited a lower level of total PA (P = 0.006) as well as of moderate intensity PA (P = 0.004) compared to controls. Both patients and controls did not report vigorous-intensity physical activities. Moreover, patients reported a higher duration of sedentary activities compared to controls (P = 0.044). No differences were observed in sleeping habits, although patients tended to sleep less compared to controls (P = 0.12). Regarding smoking habits, 43.6% of controls were current smokers compared to 27% of NAFLD patients (P = 0.049), whereas percentage of former smokers in the patients group (26%) tended to be higher than in controls (21.8%) (P = 0.15). Regarding lifestyle differences between patients with simple fatty liver and NASH, patients with NASH tended to watch television for more hours per day (3.8 ± 1.9 hr vs. 2.6 ± 1.5 hr, P = 0.09) than patients with simple fatty liver.

Table 2.

Differences in Lifestyle Characteristics Between Nonalcoholic Fatty Liver Disease Patients and Healthy Controls

Variables	NAFLD patients (n = 100)	Control (n = 55)	P^a
Physical activity
Total MET-min/day	2045 ± 286	2165 ± 259	0.006
MET-min/day of moderate intensity physical activity	229 ± 273	330 ± 269	0.004
Sedentary activities/day (hr)	3.3 ± 1.8	2.6 ± 1.4	0.044
Sleep habits
Total sleep hours	7.3 ± 1.4	7.8 ± 1.9	0.12
Short siesta (yes),^b n (%)	79 (79.0)	42 (76.4)	0.69
Optimal sleep duration,^c n (%)	69 (69.0)	32 (58.0)	0.21
Food groups and beverages intake
Full-fat dairy (serv./day)	2.11 ± 2.20	2.95 ± 2.62	0.03
Low-fat dairy (serv./day)	1.28 ± 1.50	0.90 ± 1.02	0.13
Refined cereals (serv./day)	5.85 ± 4.32	5.84 ± 3.38	0.39
Nonrefined cereals (serv./day)	1.01 ± 1.37	1.15 ± 1.35	0.24
Red meat (serv./day)	1.77 ± 1.19	1.61 ± 1.54	0.14
Poultry (serv./day)	0.61 ± 0.44	0.57 ± 0.57	0.55
Fish (serv./day)	0.70 ± 0.58	0.66 ± 0.55	0.89
Legumes (serv./day)	0.50 ± 0.36	0.40 ± 0.31	0.05
Vegetables (serv./day)	3.06 ± 1.96	3.72 ± 1.81	0.006
Fruits (serv./day)	1.98 ± 1.76	2.25 ± 2.20	0.67
Sweets (serv./day)	1.51 ± 1.33	0.90 ± 0.79	0.003
Soft drinks (serv./day)	0.15 ± 0.29	0.12 ± 0.27	0.25
Nuts (serv./day)	0.36 ± 0.48	0.93 ± 1.60	0.03
Alcohol consumption (g ethanol/day)	7.47 ± 11.10	13.23 ± 17.75	0.01
Coffee consumption (times/day)	1.62 ± 1.09	2.17 ± 1.59	0.01

Data are presented as mean ± SD or otherwise mentioned.

P-value as derived by chi-square test for nominal variables, two-sample independent t test for normal distribution variables, or Mann–Whitney U test for nonparametric variables. P ≤ 0.05.

≤1 hr/day.

≥7 and ≤9 hr/day.

MET-min/day, metabolic equivalent of task per minutes per day; serv./day, servings per day according to dietary guidelines for Greek adults.²⁸

The impact of lifestyle habits on NAFLD presence was further explored in logistic regression analyses models. From the lifestyle variables examined, PA level, optimal sleep duration, as well as consumption of nuts and coffee intake were associated with a lower NAFLD likelihood, whereas sweets consumption was associated with an increased likelihood, after adjustment for age and sex confounders (Table 3, model 1). When adjustment was also made for waist circumference and variables related to the pathophysiologic mechanisms of insulin resistance and inflammation (i.e., HOMA-IR, adiponectin, and TNF-a), the variables PA and sweets consumption maintained a significant association with NAFLD presence (all P < 0.05). Moreover, optimal sleep duration was marginally associated with less likelihood of having NAFLD (P = 0.05), however, the association of coffee intake was weakened (P = 0.07) (Table 3, model 3). When energy intake was also included in the independent variables, the results remained unchanged (data not shown).

Table 3.

Logistic Regression Analysis Models, Exploring the Association Between Lifestyle Characteristics and the Likelihood of the Presence of Nonalcoholic Fatty Liver Disease (n = 100 Cases and 55 Controls)

	Model 1 ^a			Model 2 ^b			Model 3 ^c
	OR	95% CI	P	OR	95% CI	P	OR	95% CI	P
Physical activity (per 100 MET-min)	0.78	0.67–0.91	0.001	0.77	0.66–0.91	0.002	0.74	0.61–0.89	0.002
Optimal sleep duration^d	0.50	0.15–0.97	0.04	0.52	0.22–1.24	0.14	0.38	0.14–1.01	0.05
Nuts (servings/day)	0.56	0.33–0.94	0.03	0.61	0.38–0.98	0.04	0.72	0.41–1.25	0.24
Vegetables (servings/day)	0.94	0.77–1.15	0.57	0.95	0.77–1.18	0.64	1.01	0.79–1.29	0.96
Sweets (servings/day)	2.13	1.30–3.48	0.003	2.24	1.33–3.78	0.003	2.13	1.22–3.71	0.008
Coffee (times/day)	0.67	0.48–0.93	0.01	0.68	0.49–0.94	0.02	0.72	0.49–1.04	0.07

Figures in bold are statistically significant.

Model 1: adjusted for age, sex.

Model 2: adjusted for age, sex, waist circumference, HOMA-IR.

Model 3: adjusted for age, sex, waist circumference, HOMA-IR, adiponectin, and TNF-a.

Optimal sleep duration: ≥7 and ≤9 hr sleep per day.

CI, confidence interval; OR, odds ratio.

Lifestyle habits were also explored in relation to liver stiffness values. Mean LSM in 85 patients who underwent elastography was 7.2 ± 4.7 kPa. NAFLD patients were divided according to elastography results, to those with LSM ≤6.6 (n = 47) and LSM >6.6 kPa (n = 38), based on current published cutoffs, and differences in lifestyle parameters were explored. Among dietary, sleep, and PA variables examined, only percentage of patients with optimal sleep duration differed between the two groups (76.6% in those with low stiffness vs. 55.3% in those with high stiffness, P = 0.037). Results from logistic regression analysis showed that patients with optimal sleep duration were less likely to have high liver stiffness compared to those sleeping less or more than the optimal hours [OR = 0.25 95% confidence interval (CI) 0.09–0.74, P = 0.012], after adjustment for age, sex, BMI, HOMA-IR, sweets consumption, and MET-min/day. Due to the small sample and the unbalanced distribution of NAFLD stages, no further exploration of the study results, according to the stage of the disease, was possible.

Discussion

This study aimed at exploring potential associations between lifestyle characteristics, namely PA, sleep habits, and dietary intake, and NAFLD presence in a case–control analysis. Based on the results, PA was associated with a lower likelihood of having NAFLD, whereas sweets consumption was associated with an increased likelihood. These associations remained unchanged even after taking into account important confounders such as waist circumference and energy intake. Optimal sleep duration (i.e., ≥7 and ≤9 hr per day) was marginally and inversely associated with NAFLD presence in the multivariate models, whereas the inverse association between coffee intake and the presence of NAFLD, initially observed, was attenuated after adjustment for inflammatory indices, namely adiponectin and TNF-a.

In this study, patients exhibited a lower level of total PA as well as of moderate-intensity PA, and reported a higher duration of sedentary activities than controls. Furthermore, a 100 MET-min/day increase in PA, corresponding, for example, to 20 min walking per day, was associated with 26% lower likelihood of having NAFLD. Several studies have indicated the protective effect of physical exercise interventions in the treatment of NAFLD in terms of intrahepatic lipid content and aminotransferase blood levels,³⁸ some of them showing protection independent of the presence of obesity.³⁹ Moreover, recent studies have shown that exercise exerts its beneficial effects on improving intrahepatic lipid profile even without the need for weight reduction.^40
–42 The underlying mechanism for the direct hepatic effect of exercise on liver fat and liver enzymes is not fully understood. Apart from well-known improvements in general measures of glucose tolerance and insulin sensitivity,^43,44 experiments in rats show that exercise induces several hepatic metabolic adaptations.⁴⁵ Specifically, regular PA decreases serum TG and free fatty acids by increasing mitochondrial oxidation of lipid and downregulating lipogenic proteins such as fatty acid synthase (Fas) and acecyl-Coa carboxylase (Acc).⁴⁵

In this study, although patients and controls did not actually differ regarding sleeping habits, patients tented to sleep less compared to controls. Furthermore, optimal sleep duration was associated with less likelihood of having high values of liver stiffness (i.e., >6.6 kPa). Therefore, we decided to further explore the association of optimal sleep duration with the presence of NAFLD through multivariate analyses. Optimal sleep duration was not significantly associated with the likelihood of having NAFLD when HOMA-IR was included in the multivariate model, but it was marginally associated with less likelihood of having NAFLD independent of weight status when inflammation markers (namely adiponectin and TNF-a) were also included. Evidence regarding the association between sleep habits and the development of NAFLD is sparse. In the study of Kim et al.,²⁵ short sleep duration (≤5 hr vs. >7 hr) and poor sleep quality (Pittsburgh Sleep Quality Index score >5) were associated with an increased risk of NAFLD, although these associations seemed to be mediated by weight status in men, but not in women. Similarly, in the study of Imaizumi et al.,²³ short sleep duration (≤6 hr vs. 6 to ≤7 hr) was linked to an increased risk of NAFLD only in Japanese women and this association was lost after BMI adjustment. Moreover, in the study of Bernsmeier et al.²⁴ daytime sleepiness was further associated with insulin resistance and histologic features of NAFLD.

Several studies^46,47 have shown that sleep duration might be an important risk factor for obesity and other health-related problems, such as metabolic syndrome, cardiovascular diseases, type 2 diabetes, and insulin resistance,⁴⁸ and both short (<7 hr) and long (>9 hr) sleep durations are associated with increased risk of morbidity and mortality,⁴⁹ following a U-shaped association. It has been suggested that reduced sleep not only increases food intake by altering specific hormone secretions affecting appetite, that is, leptin and ghrelin,⁵⁰ but also increases preference for high carbohydrate content, energy-rich foods, including sweets, salty snacks, and starchy foods.⁵¹ Sleep deprivation seems to share common metabolic pathways with stress system by activating hypothalamic-pituitary-adrenal axis.⁵² Moreover, increases in habitual sleep or short sleep durations have been associated with higher levels of inflammation.⁵³ Thus, according to our results, we speculate that the potential beneficial effect of sleep on the likelihood of having NAFLD may be associated with inflammation status, and future studies are needed to elucidate and confirm our observations.

Regarding dietary factors, sweets consumption showed the strongest and most powerful association with NAFLD presence, from all other foods or food groups explored in this study. This association was also independent of total daily energy intake and BMI. Several studies^54,55 have shown that excess carbohydrate intake, mainly as soft drinks and sugared beverages, and especially rich in fructose or high-fructose corn syrup and sucrose, accompanied with excessive energy intake is associated with higher liver fat, steatosis, and fibrosis.¹⁶ Carbohydrate-induced hypertriacylglycerolemia, as well as the sucrose and fructose upregulation of hepatic de novo lipogenesis can induce NAFLD.⁵⁶ Moreover, while glucose, after its absorption from the gut, is released to peripheral tissues, fructose retains and is metabolized by the liver, where it bypasses the inhibition feedback control point of phosphofructokinase, providing energy and fat, not regulated by insulin.^56,57

Coffee consumption was associated with a lower likelihood of having NAFLD, an association that might be mediated through suppression of inflammation, as further adjustment for adiponectin and TNF-a weakened this association. In a recent meta-analysis, regular caffeine intake through coffee consumption was associated with reduced hepatic fibrosis in NAFLD patients.⁵⁸ Coffee has been found to decrease proinflammatory TNF-a and increase the expression of anti-inflammatory adiponectin in animal models of NASH.⁵⁹ Moreover, it is postulated that several antioxidant compounds of coffee might mediate its protective effect against NAFLD by reducing reactive oxygen species concentrations, which can otherwise damage DNA causing liver fibrosis.⁶⁰

Our study has some limitations. Most importantly, its cross-sectional nature does not allow reaching a causal inference or elucidating underlying mechanisms. Given that patients and controls differed in several clinical and biochemical characteristics and although several confounders were carefully selected for the analysis adjustments, the possibility of residual confounding cannot be ruled out. In addition, the study was also vulnerable to recall bias. FFQs, although easy to administer and cost-effective, require recall of food use in the past, consist of a difficult cognitive task for the respondents, do not quantify foods very precisely, and aggregate foods in groups and closed lists limiting the usual diversity of dietary intake.⁶¹

Moreover, the assesment of NAFLD severity was mostly based on shear-wave elastography instead of liver biopsy, which is the gold standard method.⁶² However, because liver biopsy is an invasive procedure carrying potential risks of several complications, the use of noninvasive methods for the assessment of liver stiffness and hence fibrosis has become increasingly popular, particularly after the recent introduction of commercially available devices of liver elastography. In particular, shear-wave elastography provides improved diagnostic accuracy compared to other elastographic methods and restricts several physical limitations, including obesity.⁶² Finally, inadequate availability of liver biopsies did not allow further investigation of lifestyle habits and NAFLD stage associations. On the contrary, this is the first study combining assessment of several aspects of lifestyle, namely diet, PA, and sleep habits, while exploring their potential associations with NAFLD presence in adults.

In conclusion, a higher PA level was associated with lower likelihood, whereas sweets consumption with a higher likelihood of having NAFLD, independent of body weight status and energy intake. Optimal sleep duration showed a borderline inverse association with NAFLD presence and an inverse association with liver stiffness in NAFLD patients. Current findings could be further explored and integrated in lifestyle interventions aiming at NAFLD prevention.

Footnotes

Author Disclosure Statement

All authors declare that “no competing financial interests exist.” C.N.K. has received a Fellowship of Excellence for Postgraduate Research in Greece-Siemens Program, awarded by the State Scholarships Foundation (IKY). However, the State Scholarships Foundation had no role in the design, analysis, or writing of this article.

References

Neuschwander-Tetri

, Caldwell

. Nonalcoholic steatohepatitis: Summary of an AASLD Single Topic Conference. Hepatology, 2003; 37:1202–1219.

Matteoni

, Younossi

, Gramlich

, et al. Nonalcoholic fatty liver disease: A spectrum of clinical and pathological severity. Gastroenterology, 1999; 116:1413–1419.

Ratziu

, Bellentani

, Cortez-Pinto

, et al. A position statement on NAFLD/NASH based on the EASL 2009 special conference. J Hepatol, 2010; 53:372–384.

Masarone

, Federico

, Abenavoli

, et al. Non alcoholic fatty liver: Epidemiology and natural history. Rev Recent Clin Trials, 2014; 9:126–133.

Marchesini

, Brizi

, Bianchi

, et al. Nonalcoholic fatty liver disease: A feature of the metabolic syndrome. Diabetes, 2001; 50:1844–1850.

Miele

, Targher

. Understanding the association between developing a fatty liver and subsequent cardio-metabolic complications. Expert Rev Gastroenterol Hepatol, 2015; 9:1243–1245.

Kasturiratne

, Weerasinghe

, Dassanayake

, et al. Influence of non-alcoholic fatty liver disease on the development of diabetes mellitus. J Gastroenterol Hepatol, 2013; 28:142–147.

Levene

, Goldin

. The epidemiology, pathogenesis and histopathology of fatty liver disease. Histopathology, 2012; 61:141–152.

EASL–EASD–EASO. Clinical Practice Guidelines for the management of non-alcoholic fatty liver disease. J Hepatol, 2016; 64:1388–1402.

10.

Madero

, Arriaga

, Jalal

, et al. The effect of two energy-restricted diets, a low-fructose diet versus a moderate natural fructose diet, on weight loss and metabolic syndrome parameters: A randomized controlled trial. Metabolism, 2011; 60:1551–1559.

11.

Chiu

, Sievenpiper

, de Souza

, et al. Effect of fructose on markers of non-alcoholic fatty liver disease (NAFLD): A systematic review and meta-analysis of controlled feeding trials. Eur J Clin Nutr, 2014; 68:416–423.

12.

Georgoulis

, Kontogianni

, Tileli

, et al. The impact of cereal grain consumption on the development and severity of non-alcoholic fatty liver disease. Eur J Nutr, 2014; 53:1727–1735.

13.

Zelber-Sagi

, Nitzan-Kaluski

, Goldsmith

, et al. Long term nutritional intake and the risk for non-alcoholic fatty liver disease (NAFLD): A population based study. J Hepatol, 2007; 47:711–717.

14.

Zelber-Sagi

, Ratziu

, Oren

. Nutrition and physical activity in NAFLD: An overview of the epidemiological evidence. World J Gastroenterol, 7 2011; 17:3377–3389.

15.

Oliveira

, de Lima Sanches

, de Abreu-Silva

, et al. Nutrition and physical activity in nonalcoholic fatty liver disease. J Diabetes Res, 2016; 2016:4597246.

16.

Marchesini

, Petta

, Dalle Grave

. Diet, weight loss, and liver health in nonalcoholic fatty liver disease: Pathophysiology, evidence, and practice. Hepatology, 2016; 63:2032–2043.

17.

Hallsworth

, Thoma

, Moore

, et al. Non-alcoholic fatty liver disease is associated with higher levels of measured sedentary behaviour and lower levels of physical activity than matched healthy controls. Frontline Gastroenterol, 2015; 6:44–51.

18.

Ryu

, Chang

, Jung

, et al. Relationship of sitting time and physical activity with non-alcoholic fatty liver disease. J Hepatol, 2015; 63:1229–1237.

19.

Gerber

, Otgonsuren

, Mishra

, et al. Non-alcoholic fatty liver disease (NAFLD) is associated with low level of physical activity: A population-based study. Aliment Pharmacol Ther. 2012; 36:772–781.

20.

Nielsen

, Danielsen

, Sorensen

. Short sleep duration as a possible cause of obesity: Critical analysis of the epidemiological evidence. Obesity Rev, 2011; 12:78–92.

21.

Mesarwi

, Polak

, Jun

, et al. Sleep disorders and the development of insulin resistance and obesity. Endocrinol Metab Clin North Am, 2013; 42:617–634.

22.

Badran

, Yassin

, Fox

, et al. Epidemiology of sleep disturbances and cardiovascular consequences. Can J Cardiol, 2015; 31:873–879.

23.

Imaizumi

, Takahashi

, Tanji

, et al. The association between sleep duration and non-alcoholic fatty liver disease among Japanese men and women. Obes Facts, 2015; 8:234–242.

24.

Bernsmeier

, Weisskopf

, Pflueger

, et al. Sleep disruption and daytime sleepiness correlating with disease severity and insulin resistance in non-alcoholic fatty liver disease: A comparison with healthy controls. PLoS one, 2015; 10:e0143293.

25.

Kim

, Yun

, Jung

, et al. Sleep duration and quality in relation to non-alcoholic fatty liver disease in middle-aged workers and their spouses. J Hepatol, 2013; 59:351–357.

26.

World Medical Association declaration of Helsinki. Recommendations guiding physicians in biomedical research involving human subjects. JAMA, 1997; 277:925–926.

27.

Bountziouka

, Bathrellou

, Giotopoulou

, et al. Development, repeatability and validity regarding energy and macronutrient intake of a semi-quantitative food frequency questionnaire: Methodological considerations. Nutr Metab Cardiovasc Dis, 2012; 22:659–667.

28.

Ministry of Health & Welfare of Greece. Dietary guidelines for adults in Greece. Arch Hellenic Med, 1999; 16:516–524.

29.

Kavouras

, Maraki

, Kollia

, et al. Development, reliability and validity of a physical activity questionnaire for estimating energy expenditure in Greek adults. Science Sports, 2016; 31:e47–e53.

30.

Mukherjee

, Patel

, Kales

, et al. An Official American Thoracic Society Statement: The importance of healthy sleep. Recommendations and future priorities. Am J Respir Crit Care Med, 2015; 191:1450–1458.

31.

Yumuk

, Tsigos

, Fried

, et al. European guidelines for obesity management in adults. Obesity Facts, 2015; 8:402–424.

32.

Friedewald

, Levy

, Fredrickson

. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem, 1972; 18:499–502.

33.

Matthews

, Hosker

, Rudenski

, et al. Homeostasis model assessment: Insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia, 1985; 28:412–419.

34.

Alberti

, Eckel

, Grundy

, et al. Harmonizing the metabolic syndrome: A joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation, 2009; 120:1640–1645.

35.

Abenavoli

, Beaugrand

. Transient elastography in non-alcoholic fatty liver disease. Ann Hepatol, 2012; 11:172–178.

36.

Brunt

, Janney

, Di Bisceglie

, et al. Nonalcoholic steatohepatitis: A proposal for grading and staging the histological lesions. Am J Gastroenterol, 1999; 94:2467–2474.

37.

Kleiner

, Brunt

, Van Natta

, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology, 2005; 41:1313–1321.

38.

Orci

, Gariani

, Oldani

, et al. Exercise-based interventions for nonalcoholic fatty liver disease: A meta-analysis and meta-regression. Clin Gastroenterol Hepatol, 14:1398–1411.

39.

Bae

, Suh

, Park

, et al. Regular exercise is associated with a reduction in the risk of NAFLD and decreased liver enzymes in individuals with NAFLD independent of obesity in Korean adults. PLoS One, 2012; 7:e46819.

40.

Keating

, Hackett

, George

, et al. Exercise and non-alcoholic fatty liver disease: A systematic review and meta-analysis. J Hepatol, 2012; 57:157–166.

41.

Hallsworth

, Fattakhova

, Hollingsworth

, et al. Resistance exercise reduces liver fat and its mediators in non-alcoholic fatty liver disease independent of weight loss. Gut, 2011; 60:1278–1283.

42.

Sullivan

, Kirk

, Mittendorfer

, et al. Randomized trial of exercise effect on intrahepatic triglyceride content and lipid kinetics in nonalcoholic fatty liver disease. Hepatology, 2012; 55:1738–1745.

43.

van der Heijden

, Toffolo

, Manesso

, et al. Aerobic exercise increases peripheral and hepatic insulin sensitivity in sedentary adolescents. J Clin Endocrinol Metab, 2009; 94:4292–4299.

44.

Borghouts

, Keizer

. Exercise and insulin sensitivity: A review. Int J Sports Med, 2000; 21:1–12.

45.

Trefts

, Williams

, Wasserman

. Exercise and the regulation of hepatic metabolism. Prog Mol Biol Transl Sci, 2015; 135:203–225.

46.

, Choi

. Sleep duration and metabolic syndrome in adult populations: A meta-analysis of observational studies. Nutr Diabetes, 2013; 3:e65.

47.

Cappuccio

, Cooper

, D'Elia

, et al. Sleep duration predicts cardiovascular outcomes: A systematic review and meta-analysis of prospective studies. Eur Heart J, 2011; 32:1484–1492.

48.

Cappuccio

, D'Elia

, Strazzullo

, et al. Quantity and quality of sleep and incidence of type 2 diabetes a systematic review and meta-analysis. Diabetes Care, 2010; 33:414–420.

49.

Cappuccio

, D'Elia

, Strazzullo

, et al. Sleep duration and all-cause mortality: A systematic review and meta-analysis of prospective studies. Sleep, 2010; 33:585–592.

50.

Copinschi

, Leproult

, Spiegel

. The important role of sleep in metabolism. Front Horm Res, 2014; 42:59–72.

51.

Dashti

, Scheer

, Jacques

, et al. Short sleep duration and dietary intake: Epidemiologic evidence, mechanisms, and health implications. Adv Nutr, 2015; 6:648–659.

52.

Hirotsu

, Tufik

, Andersen

. Interactions between sleep, stress, and metabolism: From physiological to pathological conditions. Sleep Sci, 2015; 8:143–152.

53.

Patel

, Zhu

, Storfer-Isser

, et al. Sleep duration and biomarkers of inflammation. Sleep, 2009; 32:200–204.

54.

Abdelmalek

, Suzuki

, Guy

, et al. Increased fructose consumption is associated with fibrosis severity in patients with nonalcoholic fatty liver disease. Hepatology, 2010; 51:1961–1971.

55.

Chung

, Ma

, Patel

, et al. Fructose, high-fructose corn syrup, sucrose, and nonalcoholic fatty liver disease or indexes of liver health: A systematic review and meta-analysis. Am J Clin Nutr, 2014; 100:833–849.

56.

Moore

, Gunn

, Fielding

. The role of dietary sugars and de novo lipogenesis in non-alcoholic fatty liver disease. Nutrients, 2014; 6:5679–5703.

57.

Griffin

. Relevance of liver fat to the impact of dietary extrinsic sugars on lipid metabolism. Proc Nutr Soc, 2015; 74:208–214.

58.

Shen

, Rodriguez

, Shiani

, et al. Association between caffeine consumption and nonalcoholic fatty liver disease: A systemic review and meta-analysis. Therap Adv Gastroenterol, 2016; 9:113–120.

59.

Vitaglione

, Morisco

, Mazzone

, et al. Coffee reduces liver damage in a rat model of steatohepatitis: The underlying mechanisms and the role of polyphenols and melanoidins. Hepatology, 2010; 52:1652–1661.

60.

Yesil

, Yilmaz

. Review article: Coffee consumption, the metabolic syndrome and non-alcoholic fatty liver disease. Aliment Pharmacol Ther, 2013; 38:1038–1044.

61.

EFSA. General principles for the collection of national food consumption data in the view of a pan-European dietary survey. EFSA J, 2009; 7:1435–1451.

62.

Papagianni

, Sofogianni

, Tziomalos

. Non-invasive methods for the diagnosis of nonalcoholic fatty liver disease. World J Hepatol, 2015; 7:638–648.

Associations Between Lifestyle Characteristics and the Presence of Nonalcoholic Fatty Liver Disease: A Case–Control Study

Abstract

Background:

Methods:

Results:

Conclusion:

Background

Materials and Methods

Lifestyle assessment

Anthropometric assessment

Biochemical and inflammatory markers

Shear wave elastography

Liver histology

Statistical analysis

Results

Discussion

Footnotes

Author Disclosure Statement

References