Background: Alongside obesity, insomnia and depression are common public health problems. Sleep problems are currently believed to be associated with excessive food intake and metabolic disturbances. Therefore, we aimed to explore a relationship between insomnia, depressive symptoms and eating habits as well as metabolic parameters in bariatric surgery candidates. Methods: A total of 361 unrelated obese subjects were included in this study. Severity of sleep problems was measured with Athens Insomnia Scale (AIS) and the severity of depressive symptoms was assessed with the Beck Depression Inventory (BDI-II). Obstructive sleep apnea (OSA) was assessed by the Apnea Hypopnoea Index (AHI). Information was obtained about demographics, eating habits and lifestyle. Blood samples were collected to measure concentration of lipids (cholesterol, triglyceride, HDL-cholesterol, LDL-cholesterol), and glucose. Results: The median (interquartile range) score for AIS in the study participants was 5 (3–8) with a range of 0–24 and 47% (171) participants scored ≥6 (met criteria for diagnosis of insomnia). Statistically significant correlations were found between the AIS scores and serum triglycerides and glucose concentrations, and BDI-II total scores. The highest scores on AIS and BDI-II were found in participants with high frequency of snack food consumption, in physically inactive individuals as well as in those who self-reported eating at night or who declared more than 3 intense emotions associated with a desire-to-eat. Adjusted multivariate logistic regression analysis revealed that clinical insomnia was most strongly associated with daily consumption of snack foods, with the odds ratio of 3.26 (95% CI: 1.74–6.11), while depressive symptoms were strongly associated with both eating in response to ≥3 specific emotions with OR = 2.93 (95% CI: 1.26–6.78) as well as with daily consumption of snack foods with OR = 2.87 (95% CI: 1.16–5.14). Conclusions: The results indicate that insomnia and depression in obese individuals are associated with eating habits, and suggest that in some patients these associations appears as major factors affecting obesity development. Keywords: Obesity, Insomnia, Depression, Eating habits Background number of morbidly obese patients for whom bariatric Obesity, which reflects energy imbalance related to surgery is the only method leading to a significant increased dietary intake and low energy expenditure, has weight loss. Many patients, however, fail to maintain the become one of the world’s most significant health prob- achieved weight and become morbidly obese again. lems in recent few decades . The effectiveness of Therefore, a better understanding of factors that impede obesity treatment based on reduction of food intake and obesity treatment is needed to develop comprehensive increase in energy expenditure is low, both at population and effective therapy. Eating and sleeping are two kinds and patients’ levels . It results in an increase in the of behavior that are essential for the survival of humans . Sleep is a major modulator of hormone release and * Correspondence: email@example.com glucose metabolism regulation. Food intake is controlled Department of Biochemistry and Pharmacogenomics, and Center for by the neuroendocrine system and the central nervous Preclinical Studies, Medical University of Warsaw, Banacha 1, 02–097 Warsaw, system . Previous studies have demonstrated that Poland Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Wrzosek et al. BMC Psychiatry (2018) 18:153 Page 2 of 10 short sleep duration as well as experimental sleep depressive symptoms and it was shown that the deprivation in healthy humans increase hunger and nocturnal snacking was higher among the bariatric appetite [5, 6]. In a cross-sectional study in Japanese surgery candidates than among participants from females poor sleep quality was found to be significantly general community . associated with consumption of energy drinks and Insomnia co-occurs with depression  and can have sugar-sweetened beverages, skipping breakfast, and eat- serious consequences . Several recent studies have ing irregularly . This finding suggests that unhealthy demonstrated that insomnia is associated with metabolic food habits may be associated with insomnia symptoms, disorders. [23–26]. It was shown that both sleep distur- and prospective weight gain in individuals with sleep bances (difficulty in initiating or maintaining sleep) and disorders. Thus, the crosstalk between sleep quality and sleep duration (< 5 h or more than 9 h) are risk factors metabolism plays a key role in the regulation of food in- for type 2 diabetes [27, 28] and in patients with type 2 take and energy balance, affects obesity development diabetes poor sleep quality as assessed by the Pittsburg and should be taken into account in both obesity treat- Sleep Quality Index (PSQI) was associated with longer ment and prevention. Insomnia is the most common duration of diabetes [29, 30] and poor glycemic control sleep disorder . The insomnia diagnosis according to . On the other hand, insomnia symptoms were rec- the fourth version of the Diagnostic and Statistical Ma- ognized to be associated with physical inactivity [31, 32]. nual of Mental Disorders (DSM-IV) includes difficulties Reduced motivation to exercise may be caused partly by in initiating or maintaining sleep or experiencing tiredness. There is some evidence supporting physical non-restorative sleep for a period of 1 month or more activity as a nonpharmacologic treatment for sleep and the symptoms result in a significant impairment in disturbance . daily functioning . In clinical practice to measure Improving sleep quality appears as an important point insomnia symptoms Athens Insomnia Scale (AIS) is of intervention, which can enhance effectiveness of obes- commonly used [10, 11]. Various risk factors for insom- ity therapy including effectiveness of bariatric surgery nia have been identified in the general population. associated weight loss. The evidence of adverse effects of Female gender and obesity were reported to increase the poor sleep on dietary intake is based mainly on studies risk of chronic insomnia . Previous studies have in- of experimental reduction of sleep duration [34, 35], and dicated that individuals with obesity are significantly there is a lack of studies in which objective measures of more likely to report insomnia [13, 14], which could sleep were used. For the present study, we hypothesized suggest that some individuals sleep worse because that sleep problems as assessed by Athens Insomnia they are obese. This could be due to obstructive Scale (AIS) are associated with unhealthy eating habits sleep apnea (OSA) that often coexists with obesity in bariatric surgery candidates. Moreover, the relation- . It was shown that among individuals presenting ships between the insomnia and depressive symptoms, complaints related to sleep apnea, the co-occurrence sedentary lifestyle, and obstructive sleep apnea (OSA) of insomnia varies between 6 and 84% . On the that often coexists with obesity may be of importance other hand, insomnia may predispose to overcon- for effective obesity treatment. sumption of energy or night eating, thus leading to weight gain . Unhealthy behaviors, including un- Methods healthy food choices and consuming excessive Participants amounts of food after sleepless night, are probably The current study comprises a group of patients who driven by coping mechanisms and hedonic stimuli were recruited by clinical staff members at the time of a processing in the brain. Thus snacking unhealthy routine evaluation prior to bariatric surgery. Data were food is not performed to satisfy hunger or thirst, but retrospectively collected based on clinical examinations to compensate for sleep disturbances and to improve between March 2013 and March 2016. All participants mood . Thereisanevidencefor acombinedre- were Caucasian from European ancestry. Written in- lationship between unhealthy eating and depressive formed consent was obtained from each participant after symptoms [18, 19]. Snacking on food and/or beve- a full explanation of the study. The study protocol was rages was associated with an increased odds ratio for approved by the Institutional Bioethics Committees (KB/ depression in 24,697 Japanese adults after adjusting 127/2012 at the Medical University of Warsaw; 7/PB/ for sleep problems . Whileina studyin376 2015 at the Medical Centre of Postgraduate Education). Japanese adults, participants with more than two Criteria for exclusion from the study were as follows: unhealthy eating behaviors had a higher incidence of use of antipsychotics or antidepressants, current depressive symptoms compared to those with fewer pharmacotherapy for insomnia, acute endocrine than two unhealthy eating behaviors . Also dysfunction, chronic kidney disease, and alcohol use nocturnal snacking was associated with greater disorder. Wrzosek et al. BMC Psychiatry (2018) 18:153 Page 3 of 10 Procedures and assessment measures the past month participants had had an episode of eating Overnight fasting blood samples were taken from all after they had gone to bed (i.e. night eating) . We rated participants. Standard assays were used to measure total night eating frequency using a 0 to 3 scale: 0 = Not eaten; 1 cholesterol, high-density lipoprotein cholesterol, triglyc- = Eaten seldom, 2 = Eaten often, 3 = Eaten every day. Partici- erides, and glucose. Low-density lipoprotein-cholesterol pants were also asked about snack-type foods and requested levels were calculated using the Friedewald formula . to circle snacks (energy-dense, nutrient-poor foods such as In all subjects, anthropometric measurements (body biscuits, cakes, sweets, chocolate, crisps, nuts, ice-cream) that weight and height) were taken and body mass index (BMI) they usually eat. Then, they rated the frequency of eating was calculated as the ratio of weight (kilograms) to the those snacks over the past month on the following 3-point square of height (meters). Obstructive sleep apnea (OSA) scale: 0 = Not eaten, 1 = Eaten sometimes, 2 = Eaten every was assessed by the Apnea Hypopnoea Index (AHI), which day. To measure the number of specific emotions as- is the number of complete (apneas) or incomplete (hypop- sociated with a desire-to-eat we used items developed neas) obstructive events per hour of sleep. All participants with language similar to that used in the Emotional underwent standard overnight assessment (polysomnogra- Overeating Questionnaire (EOQ) ; we focused on phy, PSG) to evaluate presence of OSA. OSA was defined the most commonly experienced emotions according as an Apnea Hypopnea Index of five or greater (AHI ≥ 5) to the previous research [45, 46]. Studied individuals . The OSA group was divided into 3 severity stages: who reported eating in response to various emotional mild (5 ≤ AHI < 15), moderate (15 ≤ AHI ≤ 30), and severe states, were asked to mark the type of emotions, (AHI > 30) . All-night hemoglobinoxygensaturation choosing from: happy, lonely, depressed, bored, sad, (SpO2) was obtained with a finger oximeter. Polysomno- angry, stressed, frightened, love, surprised, upset, and graphic variables were not valid to be used to evaluate anxious . insomnia . Finally, variables in the questionnaire included a Information on age, education, and a detailed clinical description of weekly aerobic physical activities . All history, including history of obesity  was obtained for subjects were asked to self-defined their physical activity each patient, and a full physical examination was per- and answer a few questions to give better insight in formed as a part of the preoperative evaluation process. working day and weekend physical activity. Patients de- The participants, under supervision of a multidisciplinary scribed their physical activity (PA) using the following team, completed a battery of self-reported measures. At 4-point scale: 0 = Inactive, 1 = Low PA, 2 = Moderate PA, baseline all participants completed a questionnaire packet, 3 = High PA, and were asked about amount and intensity including the Athens Insomnia Scale (AIS) and the Beck of activities such as walking, jogging, cycling, swimming, Depression Inventory (BDI-II). Details on Polish transla- gym, gymnastics, rehabilitation, fitness, dance, team tion of these instruments were described elsewhere . games . Patients physically inactive or with low The AIS was used to determine the presence of clinical in- physical activity reported also if they were aware of the somnia. The AIS is a validated, effective, self-assessment importance of physical activity for health, and pointed to psychometric instrument designed for quantifying severity various reasons which, in their opinion, made them of sleep problems based on the ICD-10 criteria . The difficult to follow advices to enhance their activity. AIS items measure awakenings during the night, early morning awakening, total sleep duration, sleep quality and Data analysis sleepiness during the day. The scale has eight questions. Statistical analysis was performed with the Statistica Each question could be rated from zero (no problem) to software package, version 12.0. The concordance with three (very serious problem). AIS total score is the sum of normal distribution for all variables was calculated with the scores on each question and may vary from zero to 24. the Shapiro-Wilk and Kolmogorov-Smirnov tests. If the Scores of six or higher indicate the presence of insomnia data were not normally distributed, we used nonpara- (AIS sore ≥6). Depressive symptoms were measured using metric tests. Descriptive statistics were presented as the Beck Depression Inventory (BDI-II) . The BDI-II medians and interquartile range (quartiles 1 and 3). Cat- has 21 items rated on an intensity scale of 0–3with a egorical variables were described with number (percent- maximum score of 63, and its reliability and validity in age). Correlations between variables and AIS scores were mental health contexts are well established. A BDI-II analyzed using Spearman’s correlation coefficient. score ≥ 14 is considered a positive screen for depressive Association tests were performed using Kruskal-Wallis symptoms . tests and Mann-Whitney rank tests with the eating A number of introductory questions designed to obtain habits (unhealthy snack consumption, eating at night, general picture of participant’s eating habits, similar to ques- and eating in response to specific emotions), physical tions of the Eating Disorders Examination-Questionnaire activity and the AIS and BDI-II scores entered as (EDE-Q), were used andwerecognizedhow many timesin dependent variables. Logistic regression was used to Wrzosek et al. BMC Psychiatry (2018) 18:153 Page 4 of 10 compute crude odds ratios (ORs) with 95% confidence Metabolic parameters and the prevalence of OSA intervals (CIs) for variables associated with insomnia In the study participants, 29% were diagnosed with type (AIS score ≥ 6) or depression (BDI-II score ≥ 14). 2 diabetes, and in 14% increased fasting glucose concen- Variables that were significantly related with insomnia or trations were found, indicating disturbances in glucose depression in univariate analyses were then entered into metabolism. In patients with T2D, 43% patients received the multivariate logistic regression model to estimate ad- hypoglycemic drugs, 32% patients received insulin, while justed ORs with 95% CIs. In all analyses, a p-value < 0.05 25% patients received insulin and hypoglycemic drugs. was considered statistically significant. Patients with dyslipidemia who were taking hypolipi- demic drugs received statins (74%), fibrates (16%) or sta- tins plus fibrates (10%). Despite treatment, in 42% of the Results patients, total cholesterol was above 190 mg/dL; in 46%, Out of 436 severely obese individuals who agreed to par- LDL-cholesterol exceeded 115 mg/dL; and in 21%, ticipate in this study and provided informed consent, triglycerides levels were above 200 mg/dL. Weak but sta- 361 subjects successfully completed the Athens Insom- tistically significant correlations were found between the nia Scale (AIS) and Beck Depression Inventory (BDI-II), AIS scores and triglycerides and glucose concentrations and returned all other questionnaires; thus a response (Table 1). rate was 83%. The median (interquartile range) Apnea Hypopnea Index for the entire sample was 5 (1–15) with 50% par- ticipants with AHI ≥ 5, referred to as the OSA group Patients characteristics . PSG revealed that 25% (n = 90) patients had mild The mean age of the study group was 43.6 ± 11.5 years OSA (5 ≤ AHI < 15), 14% (n = 50) had moderate OSA and the mean BMI was 42.3 ± 6.4 kg/m . The majority of (15 ≤ AHI ≤ 30), and 11% (n = 40) had severe OSA the participants (63%) had a BMI above 40 kg/m , 28% (AHI ≥ 30). No significant differences in the AIS scores had a BMI in the range of 35–39.9 kg/m , and 9% of the between OSA group (AHI ≥ 5) and non-OSA group subjects had class I obesity (BMI: 30.0–34.9 kg/m ). The (AHI < 5) were found (Mann-Whitney U test, p = 0.850). study group had a significantly higher representation of In addition, no significant differences in the AIS scores women (73%) than men (27%; p < 0.001). About 28% of between 3 stages in OSA group (mild, moderate and participants had a Master’s Degree, 19% had Bachelor’s severe OSA) were found (Kruskal-Wallis test: H = 0.535, Degree, 46% had some tertiary education, 1% had sec- p = 0.765). Likewise, no correlation was observed be- ondary education, and the remaining 6% had primary tween AIS scores and AHI (p = 0.772, Table 1). AIS did education. not correlate with number of desaturations or with average (Av SaO2) and minimum (Min SaO2) oxygen saturation (Table 1). Insomnia and depression The median (interquartile range) score for AIS in the Eating habits, sedentary lifestyle, insomnia and study participants was 5 (3–8). Based on the AIS, in- depressive symptoms somnia was diagnosed in 47% (171) participants scor- Sixty-six participants (18%) reported daily consumption ing ≥6 (higher cut-off score). No statistically of snack foods and they had the highest AIS and BDI-II significant correlation between the AIS scores and scores (Table 2). Of the studied patients with obesity, education level was found (Rho = 0.058, p = 0.293). No 65% described their physical activity as low or reported significant differences in the AIS scores between no psychical activity and classified their lifestyle as sed- BMI-based categories of obesity were found (Kruskal-- entary. We found that individuals who reported “no”, Wallis test: H = 0.535, p =0.765). Likewise, no correl- “low” or “moderate” physical activity had significantly ation was found between insomnia severity (AIS) and higher AIS sores than those with self-reported “high” BMI (Table 1). Strong correlation was recognized physical activity (Table 2). between the AIS and BDI-II total scores (Table 1). Of the studied patients, 2% reported daily episodes of BDI-II scores analysis revealed that 3% (n = 11) of the eating at night, which was associated with high AIS and study participants scored 29–63, which indicates severe BDI-II scores (Table 3). Moreover, 67% (n = 243) of stu- depression, 14% (n = 52) scored 20–28 indicating moder- died individuals reported eating in response to various ate symptoms, 19% (n = 67) participants scored 14–19 emotional states, and significantly higher AIS and BDI-II corresponding to mild depression, and 64% (n = 231) scores were found in this group in comparison to the scored 0–13 corresponding to minimal symptoms subjects who did not report eating when emotional. In according to generally accepted BDI-II scores inter- addition, participants who reported more emotions (3 or pretation . more) associated with a desire-to-eat had higher AIS Wrzosek et al. BMC Psychiatry (2018) 18:153 Page 5 of 10 Table 1 Correlations between Athens Insomnia Scale (AIS) scores and clinical variables in patients with obesity Median (IQR) Athens Insomnia Scale (AIS) Rho p BDI-II 10 (5–16) 0.576 0.000 Age (years) 43 (35–53) 0.105 0.054 Weight (kg) 118 (103–133) −0.061 0.262 Height (cm) 168 (162–174) −0.107 0.050 BMI (kg/m2) 41 (37–45) −0.021 0.698 Fat (%) 44 (39–47) 0.029 0.624 Glucose (mg/dl) 98 (90–115) 0.115 0.035 Total cholesterol (mg/dl) 183 (154–211) 0.011 0.836 LDL-cholesterol (mg/dl) 111 (84–133) −0.074 0.184 HDL-cholesterol (mg/dl) 40 (34–47) −0.002 0.964 Triglycerides (mg/dl) 139 (101–188) 0.119 0.028 Apnea Hypopnea Index (AHI) 5 (1–15) −0.0167 0.772 Number of desaturations 24 (6–72) −0.003 0.960 Average oxygen saturation (%) 94 (92–95) −0.098 0.076 Minimum oxygen saturation (%) 86 (81–89) −0.061 0.274 Spearman’s correlations, p values were considered significant when p > 0.05 (in bold). Data is presented as median, interquartile range (IQR) AIS Athens Insomnia Scale, BDI-II Beck Depression Inventory, BMI body mass index, HDL high-density lipoprotein, LDL low-density lipoprotein Table 2 Eating habits and physical activity in relation to the Athens Insomnia Scale and Beck Depression Inventory N Athens Insomnia Scale (AIS) Beck Depression Inventory (BDI-II) Eating snack-type foods Every day 66 7 (5–10) 15 (10–23) ** *** Occasionally 260 4 (3–8) 10 (5–15) ** *** Not at all 35 4 (2–8) 8(4–14) Statistics p = 0.0001 p = 0.0000 Number of emotions associated with desire to eat 3 or more 29 9 (6–13) 19 (11–25) 6(3–9) 11 (6–16) < 3 214 Statistics p = 0.0009 p = 0.0002 Night eating Every night 7 9 (3–13) 19 (12–21) ns ns Often 56 7 (4–11) 10 (6–20) ns ns Rarely 96 6 (3–9) 11 (7–16) * * Never 202 5 (3–7) 10 (5–15) Statistics p = 0.0009 p = 0.0131 Physical activity *** *** Inactive 50 7 (4–10) 14 (8–20) *** * Low 185 5 (4–8) 11 (6–16) ** ns Moderate 116 4 (3–9) 9(5–14) High 10 2 (1–3) 3 (0–8) Statistics p = 0.0000 p = 0.0001 Data are presented as median, interquartile range Analysis was performed using Kruskal-Wallis tests and Mann-Whitney rank tests, where appropriate; p values were considered significant when p > 0.05 (in bold); Post-hoc analyses were conducted to determine differences in AIS and BDI-II scores between the groups and group eating snack-type foods every day, group eating every night, group with high physical activity * p < 0.05, **p < 0.01, ***p < 0.001, ns – non-significant Wrzosek et al. BMC Psychiatry (2018) 18:153 Page 6 of 10 Table 3 Results of a logistic regression - crude and adjusted odds ratios for factors associated with insomnia and depression in bariatric surgery candidates Variable Athens Insomnia Scale sore ≥6 Beck Depression Inventory-II sore ≥14 Crude OR p Adjusted p Crude OR p Adjusted p a b a b (95% CI) OR (95% CI) (95% CI) OR (95% CI) Daily consumption of 3.76 (1.58–8.97) 0.0022 3.26 (1.74–6.11) 0.0002 3.42 (1.96–5.96) 0.0000 2.87 (1.16–5.14) 0.0007 snack foods Eating in response 3.56 (1.47–8.50) 0.0046 2.86 (1.14–5.13) 0.0235 3.53 (1.57–7.94) 0.0021 2.93 (1.26–6.78) 0.0117 to ≥3 specific emotions Night eating 1.54 (1.01–2.35) 0.0434 1.24 (0.79–1.93) 0.3472 1.54 (0.99–2.38) 0.0416 1.22 (0.76–1.94) 0.3989 Physical inactivity 2.22 (1.19–4.12) 0.0104 1.48 (0.75–2.90) 0.2492 2.14 (1.17–3.93) 0.0131 1.44 (0.75–2.79) 0.2699 OR odds ratio, CI Confidence interval, p-value < 0.05 are bolded Crude logistic regression analysis Adjusted logistic regression analysis (after controlling for all other significant factors) and BDI-II scores than those who reported fewer emo- bidirectional relation between sleep disturbances and tions (see Table 2). eating habits can be suggested. Nevertheless, it is not In univariate analysis, we found that daily consumption clear from our cross-sectional study whether sleep dis- of snack foods, self- reported eating in response to more turbances influence consumption of snack foods or vice than 3 emotions, night eating and physical inactivity were versa. Very few previous findings suggest that unhealthy all significantly associated with clinical insomnia (AIS food habits are associated with insomnia symptoms. The score ≥ 6) and depression (BDI-II score ≥ 14) (Table 3). relationship between dietary intake and sleep was exam- Adjusted multivariate logistic regression analysis revealed ined in Japanese female workers. The results showed that clinical insomnia was most strongly associated with that low intake of vegetables, high intake of confection- daily consumption of snack foods, with the odds ratio of ary, and unhealthy eating habits were associated with 3.26 (95% CI: 1.74–6.11), while depressive symptoms were poor sleep quality assessed using the Pittsburgh Sleep strongly associated with both eating in response to ≥3spe- Quality Index (PSQI) . On the contrary, mainly weak cific emotions with OR = 2.93 (95% CI: 1.26–6.78) as well and inconsistent associations between insomnia symp- as with daily consumption of snack foods with OR = 2.87 toms and poor food habits in Helsinki Health Study (95% CI: 1.16–5.14) (Table 3). Additionally, in adjusted were found . However, the Finnish study was unable analysis insomnia was the strongest predictor of daily con- to confirm any associations probably due to general sumption of snack foods (Table 4); this association was questions on food habits and suggestive categorization even stronger than for depression. This indicates a bidir- of food habits as healthy or unhealthy. On the other ectional relation between the daily consumption of snack hand, an experimental reduction of sleep duration was foods and insomnia. accompanied by increased intake of calories from snacks [34, 48], increased food purchasing in normal-weight Discussion men , and increased hunger and appetite, especially The present study shows that the participants reporting for calorie-dense foods with high carbohydrate content daily consumption of snack foods and eating in response . These associations seem to be related to the alter- to 3 or more emotions had the highest AIS and BDI-II ations in appetitive brain signaling . Neural activa- scores. Unhealthy eating may be a coping mechanism tion was measured by functional magnetic resonance for sleep deficit as well hard digestible food may imaging (fMRI) in twenty-three healthy participants ex- deteriorate the quality of sleep. Therefore, the amined on two sessions: a night of normal sleep and a Table 4 Results of the logistic regression - crude and adjusted odds ratios for factors significantly associated with a daily consumption of snack foods in bariatric surgery candidates Variable Daily consumption of snack foods Crude OR (95% CI) p Adjusted OR (95% CI) p a b Athens Insomnia Scale sore ≥6 3.95 (2.16–7.20) 0.0000 3.06 (1.56–5.86) 0.0007 Beck Depression Inventory-II sore ≥14 2.56 (1.47–4.47) 0.0000 2.11 (1.14–3.93) 0.0168 Eating in response to ≥3 specific emotions 2.78 (1.21–6.33) 0.0153 1.83 (0.77–4.37) 0.1693 OR odds ratio, CI Confidence interval, p-value < 0.05 are bolded Crude logistic regression analysis Adjusted logistic regression analysis (after controlling for all other significant factors) Wrzosek et al. BMC Psychiatry (2018) 18:153 Page 7 of 10 night of total sleep deprivation. Sleep deprivation signifi- reported daily episodes of eating at night, which was associ- cantly decreased activity in the anterior cingulate cortex, ated with the highest AIS scores (median = 9) and BDI-II lateral orbital frontal cortex and anterior insular cortex scores (median = 19). However, interpretation of our find- (brain regions known to be instrumental in appetitive ings is limited due to the small group of every night eating desire and food stimulus evaluation) and increased the individuals. amygdala responsivity to desirable food items. Addition- In the present study, the AIS and BDI-II scores were ally, increase in desire for high-calorie foods, which posi- high in participants who self-reported eating in response tively correlated with the severity of sleep deprivation, to more than 3 different emotions. Emotions may affect was demonstrated . Thus, deeper insight in potential human eating behavior. Indeed, in the current study we determinants of eating behaviors is needed to identify demonstrated that depression was associated with a ten- new effective strategies for obesity prevention and treat- dency to eat in response to more than 3 emotions as well ment. The prevalence of insomnia in our study (47% as with unhealthy snack consumption among individuals participants scoring ≥6 on the AIS) is similar to insom- with obesity. Our data support the previous findings nia frequency assessed in the population of 6079 Latin showing that emotional eating can be associated with in- American women aged 40 to 59 years (43.6% had insom- creased BMI  and with depressive symptoms . nia, AIS) . Also sedentary lifestyle, defined as fewer Adequate sleep quality and duration are important for than three weekly 30-min periods of physical activity, the appetite regulation and normal functioning of meta- was very common in surveyed American women (63.9% bolic and hormonal processes . Moreover, many ex- of them were self-defined as sedentary). Sedentary perimental and epidemiologic studies link sleep women had more depressive and insomnia symptoms as disturbance with alterations in glucose homeostasis . compared with non-sedentary women. As expected, in For instance, it was shown that short sleep time (4 h per our report 65% studied patients with obesity described night for 6 nights) in young, healthy men was associated their physical activity as low or declared psychical in- with lower glucose tolerance than in the fully rested men activity and classified their lifestyle as sedentary. The . In healthy individuals, glucose tolerance varies statistical analysis revealed that these individuals had sig- throughout the day; glucose concentrations are markedly nificantly higher AIS and BDI-II scores than those with higher in the evening than in the morning, and glucose self-reported “high” physical activity. It has been consid- tolerance is reduced in the night . Circadian rhythm is ered that cultivating exercise habits, reducing sedentary important modulator of glucose homeostasis and sleep time and improving sleep quality may be important loss may result in metabolic alterations. In our study, high strategies for obesity prevention [51–53]. Our present AIS scores were associated with high glucose and triglyc- research indicates that insomnia and depression symp- erides concentrations in bariatric surgery candidates. This toms commonly coexist in obese individuals, and it is finding is consistent with the hypothesis that changes in consistent with findings of a meta-analysis of 19 original the quantity or quality of sleep may affect carbohydrate papers reporting that sleep deprivation alters mood  and lipid metabolism. The amount of research investigat- and the results of a large prospective population-based ing the relationship between insomnia symptoms and glu- study — the Nord-Trøndelag Health Studies (HUNT2 cose and lipid abnormalities in obese patients is limited. and HUNT3) showing that insomnia is associated with However, prior reports demonstrated that shift workers more than twice the odds for depression . from two plants of northern France had significantly Sleep disturbances are considered as a risk factor of obes- higher levels of serum triglyceride than day workers, but ity [55, 56]. However, available findings are inconsistent and there was no influence of shift work on total cholesterol high BMI was reported to be not associated with the in- and HDL cholesterol concentrations . somnia symptoms or insomnia as a syndrome [57, 58]. We Obstructive sleep apnea (OSA) is a highly prevalent also found no correlation between BMI and Athens Insom- respiratory disorder, characterized by recurrent episodes nia Scale scores. Nonetheless, it should be emphasized that of upper airway obstruction occurring during sleep, and participants in our study had extremely high BMI. In associated with recurrent cycles of desaturation and addition, insomnia might be directly associated with re-oxygenation . Although OSA is characterized by increased consumption of palatable food, but not with a fragmented sleep, not all affected patients complain of general tendency to eat more. This suggests that insomnia insomnia [12, 16, 68]. In our study, no correlation be- symptoms may contribute to unhealthy eating patterns or tween AHI and insomnia symptoms as assessed with increased amounts of energy consumed at inappropriate AIS was observed, however, 25% of the study partici- time points, i.e., at night. It is important to note that other pants had mild OSA and only 11% had severe OSA. Our studies revealed late-night eating in response to sleep re- results support the hypothesis that patients with OSA of striction , insomnia ordepressivesymptoms . mild to moderate severity may not necessarily report Similarly, 2% of the studied bariatric surgery candidates low quality of sleep, as it was previously suggested . Wrzosek et al. BMC Psychiatry (2018) 18:153 Page 8 of 10 The current findings should be considered in the con- Acknowledgements We would like to thank all the participants who took part in the study and text of several limitations. First, to assess insomnia we the staff at the Orłowski Hospital in Warsaw, Poland. We would like to thank used the Athens Insomnia Scale, a self-report validated Marek Głowala, Marzanna Kieszek, Elżbieta Malinowska from the Medical questionnaire, which is a subjective instrument to evalu- University of Warsaw for technical assistance. ate sleep disturbances. Unfortunately, we were not able Funding to utilize PSG to measure objective electrophysiological This work was supported by the Medical University of Warsaw under Grants parameters of sleep quantity and quality. Second, this FW113/NM1/17, FW113/NM1/14, FW113/PM32/14, and carried out through CePT infrastructure financed by the European Union (the European Regional study did not allow assessing insomnia symptoms in a Development Fund within the Operational Programme ‘Innovative economy’ relationship to amount of calories consumed, because for 2007–2013). we did not assess the overall caloric intake for each par- ticipant. We used self-report measurements of physical Availability of data and materials The dataset supporting the conclusions of this article is available with the activity and there might be a concern that the recall of corresponding author and will be made available on reasonable request. physical activity may be subjective in some cases. Our study also did not assess all the aspects of night eating, Authors’ contributions MW1 was responsible for the conception and design of the study; MW1, GN, which may be relevant when diagnosing night eating MW2 participated in drafting and final approval of the article; AS, MT syndrome (NES), i.e., consuming ≥25% of total daily cal- contributed to the recruitment and evaluation of patients; MW1, AS ories after the evening meal, or the type of food con- performed analysis, collection of data, GN were responsible for funding of the study, MW1, GN, MW2 were responsible for interpretation of data sumed . Instead, we have assessed night eating analysis, important contribution for intellectual content. All authors read and frequency using the Eating Disorder Examination Ques- approved the final article. tionnaire (EDE-Q) . Another limitation of our study Ethics approval and consent to participate may be related to a larger number of females than males. This study was approved by the Institutional Bioethics Committees (KB/127/ Therefore, future studies should focus upon identifying 2012 and KB/67/2017 at the Medical University of Warsaw; 7/PB/2015 at the any potential gender differences. Our study participants Medical Centre of Postgraduate Education). Before data collection, study objectives were explained to patients and their written informed consent had extremely high BMI and obesity-associated meta- was obtained. bolic abnormalities, and some of them were under pharmacological treatment for dyslipidemia and/or type Competing interests 2 diabetes. Despite taking medications that improve lipid The authors declare they have no competing interests. or glucose profile we still could demonstrate in our pa- tients the influence of sleep disturbances on studied bio- Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in chemical parameters. published maps and institutional affiliations. The dramatic increase in obesity prevalence overlaps the reduced sleep duration and quality, which are hall- Author details Department of Biochemistry and Pharmacogenomics, and Center for marks of the modern society. Sleep disturbances and de- Preclinical Studies, Medical University of Warsaw, Banacha 1, 02–097 Warsaw, pressive symptoms may be considered as a risk factor, Poland. Department of Psychiatry, Medical University of Warsaw, Warsaw, which contributes to unhealthy eating behaviors that Poland. Department of Geriatrics, Internal Medicine and Metabolic Bone Diseases, Medical Centre of Postgraduate Education, Prof. W. Orlowski might result in the development of obesity. Hospital, Warsaw, Poland. Received: 12 September 2017 Accepted: 11 May 2018 Conclusions The results of the study show new data on the associa- References tions between insomnia, depression and eating behaviors 1. Popkin BM, Adair LS, Ng SW. Global nutrition transition and the pandemic in bariatric surgery candidates, and highlight the import- of obesity in developing countries. Nutr Rev. 2012;70(1):3–21. ance of the potential consequences of poor sleep or de- 2. Williams RL, Wood LG, Collins CE, Callister R. Effectiveness of weight loss interventions–is there a difference between men and women: a systematic pression in obesity. This should stimulate further review. Obes Rev. 2015;16(2):171–86. research in this area leading to the development of in- 3. Steiger A. Eating and sleeping–their relationship to ghrelin and leptin. novative and effective strategies for obesity prevention American journal of physiology Regulatory, integrative and comparative physiology. 2004;287(5):R1031–2. and treatment. 4. Valassi E, Scacchi M, Cavagnini F. Neuroendocrine control of food intake. Nutrition, metabolism, and cardiovascular diseases : NMCD. Abbreviations 2008;18(2):158–68. AHI: Apnea Hypopnoea Index; AIS: Athens Insomnia Scale; BDI-II: Beck 5. Spiegel K, Tasali E, Penev P, Van Cauter E. Brief communication: sleep Depression Inventory; BMI: Body Mass Index; CIs: Confidence intervals; EDE- curtailment in healthy young men is associated with decreased leptin Q: Eating Disorders Examination-Questionnaire; EOQ: Emotional Overeating levels, elevated ghrelin levels, and increased hunger and appetite. Ann Questionnaire; fMRI: functional magnetic resonance imaging; HDL-C: High- Intern Med. 2004;141(11):846–50. density lipoprotein cholesterol; LDL-C: low-density lipoprotein-cholesterol; 6. Taheri S, Lin L, Austin D, Young T, Mignot E. Short sleep duration is OR: Odds Ratio; OSA: Obstructive sleep apnea; PA: Physical activity; associated with reduced leptin, elevated ghrelin, and increased body mass PSG: Polysomnography index. PLoS Med. 2004;1(3):e62. Wrzosek et al. BMC Psychiatry (2018) 18:153 Page 9 of 10 7. Katagiri R, Asakura K, Kobayashi S, Suga H, Sasaki S. Low intake of 32. Morgan K, Clarke D. Risk factors for late-life insomnia in a representative vegetables, high intake of confectionary, and unhealthy eating habits are general practice sample. Br J Gen Pract. 1997;47(416):166–9. associated with poor sleep quality among middle-aged female Japanese 33. Kline CE. The bidirectional relationship between exercise and sleep: workers. J Occup Health. 2014;56(5):359–68. implications for exercise adherence and sleep improvement. Am J Lifestyle 8. Silva JA CE. Sleep disorders in psychiatry. Metabolism: clinical and Med. 2014;8(6):375–9. experimental. 2006;55(10 Suppl 2):S40–4. 34. Hogenkamp PS, Nilsson E, Nilsson VC, Chapman CD, Vogel H, Lundberg LS, 9. Ohayon MM, Reynolds CF 3rd. Epidemiological and clinical relevance of Zarei S, Cedernaes J, Rangtell FH, Broman JE, et al. Acute sleep deprivation insomnia diagnosis algorithms according to the DSM-IV and the increases portion size and affects food choice in young men. international classification of sleep disorders (ICSD). Sleep Med. 2009; Psychoneuroendocrinology. 2013;38(9):1668–74. 10(9):952–60. 35. Chapman CD, Nilsson EK, Nilsson VC, Cedernaes J, Rangtell FH, Vogel H, 10. Soldatos CR, Dikeos DG, Paparrigopoulos TJ. Athens insomnia scale: Dickson SL, Broman JE, Hogenkamp PS, Schioth HB, et al. Acute sleep validation of an instrument based on ICD-10 criteria. J Psychosom Res. 2000; deprivation increases food purchasing in men. Obesity (Silver Spring, Md). 48(6):555–60. 2013;21(12):E555–60. 11. Chung KF, Kan KK, Yeung WF. Assessing insomnia in adolescents: 36. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of comparison of insomnia severity index, Athens insomnia scale and sleep low-density lipoprotein cholesterol in plasma, without use of the quality index. Sleep Med. 2011;12(5):463–70. preparative ultracentrifuge. Clin Chem. 1972;18(6):499–502. 12. Singareddy R, Vgontzas AN, Fernandez-Mendoza J, Liao D, Calhoun S, 37. Mbata G, Chukwuka J. Obstructive sleep apnea hypopnea syndrome. Annals Shaffer ML, Bixler EO. Risk factors for incident chronic insomnia: a general of medical and health sciences research. 2012;2(1):74–7. population prospective study. Sleep Med. 2012;13(4):346–53. 38. Littner M, Hirshkowitz M, Kramer M, Kapen S, Anderson WM, Bailey D, Berry RB, 13. Pearson NJ, Johnson LL, Nahin RL. Insomnia, trouble sleeping, and Davila D, Johnson S, Kushida C, et al. Practice parameters for using complementary and alternative medicine: analysis of the 2002 national polysomnography to evaluate insomnia: an update. Sleep. 2003;26(6):754–60. health interview survey data. Arch Intern Med. 2006;166(16):1775–82. 39. Wrzosek M, Sawicka A, Wrzosek M, Piątkiewicz P, Tałałaj M, Nowicka G. Age 14. Palmese LB, DeGeorge PC, Ratliff JC, Srihari VH, Wexler BE, Krystal AD, Tek C. at onset of obesity, transcription factor 7-like 2 (TCF7L2) rs7903146 Insomnia is frequent in schizophrenia and associated with night eating and polymorphism, adiponectin levels and the risk of type 2 diabetes in obese obesity. Schizophr Res. 2011;133(1–3):238–43. patients. Arch Med Sci. 2017. https://doi.org/10.5114/aoms.2017.69638. 15. Young T, Peppard PE, Taheri S. Excess weight and sleep-disordered 40. Wojnar M, Ilgen MA, Jakubczyk A, Wnorowska A, Klimkiewicz A, Brower KJ. breathing. J Appl Physiol. 2005;99(4):1592–9. Impulsive suicide attempts predict post-treatment relapse in alcohol- 16. Ong JC, Crawford MR. Insomnia and obstructive sleep apnea. Sleep dependent patients. Drug Alcohol Depend. 2008;97(3):268–75. medicine clinics. 2013;8(3):389–98. 41. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for 17. Trinko R, Sears RM, Guarnieri DJ, DiLeone RJ. Neural mechanisms underlying measuring depression. Arch Gen Psychiatry. 1961;4:561–71. obesity and drug addiction. Physiol Behav. 2007;91(5):499–505. 42. Wang YP, Gorenstein C. Assessment of depression in medical patients: a 18. Furihata R, Uchiyama M, Takahashi S, Konno C, Suzuki M, Osaki K, Kaneita Y, systematic review of the utility of the Beck Depression Inventory-II. Clinics Ohida T. Self-help behaviors for sleep and depression: a Japanese (Sao Paulo, Brazil). 2013;68(9):1274–87. nationwide general population survey. J Affect Disord. 2011;130(1–2):75–82. 43. Fairburn CG, Cooper Z, O’Connor M. Eating disorder examination 19. Huang C, Momma H, Cui Y, Chujo M, Otomo A, Sugiyama S, Ren Z, Niu K, questionnaire. In: Fairburn CG, editor. Cognitive behavior therapy and eating Nagatomi R. Independent and combined relationship of habitual unhealthy disorders. New York: Guilford Press; 2008. p. 1–48. eating behaviors with depressive symptoms: a prospective study. Journal of 44. Masheb RM, Grilo CM. Emotional overeating and its associations with eating epidemiology. 2017;27(1):42–7. disorder psychopathology among overweight patients with binge eating 20. Colles SL, Dixon JB, O'Brien PE. Night eating syndrome and nocturnal disorder. The International journal of eating disorders. 2006;39(2):141–6. snacking: association with obesity, binge eating and psychological distress. 45. Bennett J, Greene G, Schwartz-Barcott D. Perceptions of emotional eating Int J Obes. 2007;31(11):1722–30. behavior. A qualitative study of college students. Appetite. 2013;60(1):187–92. 21. Tsuno N, Besset A, Ritchie K. Sleep and depression. The Journal of clinical 46. Arnow B, Kenardy J, Agras WS. The emotional eating scale: the psychiatry. 2005;66(10):1254–69. development of a measure to assess coping with negative affect by eating. 22. Sivertsen B, Lallukka T, Salo P, Pallesen S, Hysing M, Krokstad S, Simon O. The International journal of eating disorders. 1995;18(1):79–90. Insomnia as a risk factor for ill health: results from the large population- 47. Blumel JE, Chedraui P, Aedo S, Fica J, Mezones-Holguin E, Baron G, based prospective HUNT study in Norway. J Sleep Res. 2014;23(2):124–32. Bencosme A, Benitez Z, Bravo LM, Calle A, et al. Obesity and its relation to 23. Taylor DJ, Mallory LJ, Lichstein KL, Durrence HH, Riedel BW, Bush AJ. depressive symptoms and sedentary lifestyle in middle-aged women. Comorbidity of chronic insomnia with medical problems. Sleep. 2007; Maturitas. 2015;80(1):100–5. 30(2):213–8. 48. Nedeltcheva AV, Kilkus JM, Imperial J, Kasza K, Schoeller DA, Penev PD. 24. Vgontzas AN, Liao D, Bixler EO, Chrousos GP, Vela-Bueno A. Insomnia with Sleep curtailment is accompanied by increased intake of calories from objective short sleep duration is associated with a high risk for snacks. Am J Clin Nutr. 2009;89(1):126–33. hypertension. Sleep. 2009;32(4):491–7. 49. Greer SM, Goldstein AN, Walker MP. The impact of sleep deprivation on 25. Vgontzas AN, Liao D, Pejovic S, Calhoun S, Karataraki M, Bixler EO. Insomnia food desire in the human brain. Nat Commun. 2013;4:2259. with objective short sleep duration is associated with type 2 diabetes: a 50. Blumel JE, Fica J, Chedraui P, Mezones-Holguin E, Zuniga MC, Witis S, Vallejo population-based study. Diabetes Care. 2009;32(11):1980–5. MS, Tserotas K, Sanchez H, Onatra W, et al. Sedentary lifestyle in middle- 26. Spiegel K, Tasali E, Leproult R, Van Cauter E. Effects of poor and short aged women is associated with severe menopausal symptoms and obesity. sleep on glucose metabolism and obesity risk. Nat Rev Endocrinol. Menopause (New York, NY). 2016;23(5):488–93. 2009;5(5):253–61. 51. Chen LJ, Lai YJ, Sun WJ, Fox KR, Chu D, Ku PW. Associations of exercise, 27. Kawakami N, Takatsuka N, Shimizu H. Sleep disturbance and onset of type 2 sedentary time and insomnia with metabolic syndrome in Taiwanese older diabetes. Diabetes Care. 2004;27(1):282–3. adults: a 1-year follow-up study. Endocr Res. 2015;40(4):220–6. 28. Ayas NT, White DP, Al-Delaimy WK, Manson JE, Stampfer MJ, Speizer FE, 52. Dolezal BA, Neufeld EV, Boland DM, Martin JL, Cooper CB. Interrelationship Patel S, Hu FB. A prospective study of self-reported sleep duration and between sleep and exercise: a systematic review. Adv Prev Med. 2017;14. incident diabetes in women. Diabetes Care. 2003;26(2):380–4. Article ID 1364387. 29. Shamshirgaran SM, Ataei J, Malek A, Iranparvar-Alamdari M, Aminisani N. 53. Baron KG, Reid KJ, Zee PC. Exercise to improve sleep in insomnia: exploration Quality of sleep and its determinants among people with type 2 diabetes of the bidirectional effects. Journal of clinical sleep medicine : JCSM : official mellitus in northwest of Iran. World J Diabetes. 2017;8(7):358–64. publication of the American Academy of Sleep Medicine. 2013;9(8):819–24. 30. Cunha MC, Zanetti ML, Hass VJ. Sleep quality in type 2 diabetics. Revista 54. Pilcher JJ, Huffcutt AI. Effects of sleep deprivation on performance: a meta- latino-americana de enfermagem. 2008;16(5):850–5. analysis. Sleep. 1996;19(4):318–26. 31. Haario P, Rahkonen O, Laaksonen M, Lahelma E, Lallukka T. Bidirectional 55. Fatima Y, Doi SA, Mamun AA. Sleep quality and obesity in young subjects: a associations between insomnia symptoms and unhealthy behaviours. J meta-analysis. Obesity reviews : an official journal of the International Sleep Res. 2013;22(1):89–95. Association for the Study of Obesity. 2016;17(11):1154–66. Wrzosek et al. BMC Psychiatry (2018) 18:153 Page 10 of 10 56. Vgontzas AN, Lin HM, Papaliaga M, Calhoun S, Vela-Bueno A, Chrousos GP, Bixler EO. Short sleep duration and obesity: the role of emotional stress and sleep disturbances. International journal of obesity (2005). 2008;32(5):801–9. 57. LeBlanc M, Merette C, Savard J, Ivers H, Baillargeon L, Morin CM. Incidence and risk factors of insomnia in a population-based sample. Sleep. 2009;32(8): 1027–37. 58. Gureje O, Oladeji BD, Abiona T, Makanjuola V, Esan O. The natural history of insomnia in the Ibadan study of ageing. Sleep. 2011;34(7):965–73. 59. Spaeth AM, Dinges DF, Goel N. Phenotypic vulnerability of energy balance responses to sleep loss in healthy adults. Sci Rep. 2015;5:14920. 60. Meule A, Allison KC, Platte P. Emotional eating moderates the relationship of night eating with binge eating and body mass. European eating disorders review : the journal of the Eating Disorders Association. 2014;22(2):147–51. 61. van Strien T, Konttinen H, Homberg JR, Engels RC, Winkens LH. Emotional eating as a mediator between depression and weight gain. Appetite. 2016; 100:216–24. 62. Van Cauter E, Spiegel K, Tasali E, Leproult R. Metabolic consequences of sleep and sleep loss. Sleep Med. 2008;9(Suppl 1):S23–8. 63. Ip M, Mokhlesi B. Sleep and glucose intolerance/diabetes mellitus. Sleep medicine clinics. 2007;2(1):19–29. 64. Spiegel K, Leproult R, Van Cauter E. Impact of sleep debt on metabolic and endocrine function. Lancet (London, England). 1999;354(9188):1435–9. 65. Van Cauter E, Polonsky KS, Scheen AJ. Roles of circadian rhythmicity and sleep in human glucose regulation. Endocr Rev. 1997;18(5):716–38. 66. Romon M, Nuttens MC, Fievet C, Pot P, Bard JM, Furon D, Fruchart JC. Increased triglyceride levels in shift workers. Am J Med. 1992;93(3):259–62. 67. Garvey JF, Pengo MF, Drakatos P, Kent BD. Epidemiological aspects of obstructive sleep apnea. Journal of thoracic disease. 2015;7(5):920–9. 68. Fernandez-Mendoza J, Vgontzas AN, Bixler EO, Singareddy R, Shaffer ML, Calhoun SL, Karataraki M, Vela-Bueno A, Liao D. Clinical and polysomnographic predictors of the natural history of poor sleep in the general population. Sleep. 2012;35(5):689–97. 69. Macey PM, Woo MA, Kumar R, Cross RL, Harper RM. Relationship between obstructive sleep apnea severity and sleep, depression and anxiety symptoms in newly-diagnosed patients. PLoS One. 2010;5(4):e10211. 70. Allison KC, Lundgren JD, O'Reardon JP, Martino NS, Sarwer DB, Wadden TA, Crosby RD, Engel SG, Stunkard AJ. The night eating questionnaire (NEQ): psychometric properties of a measure of severity of the night eating syndrome. Eat Behav. 2008;9(1):62–72.
– Springer Journals
Published: May 29, 2018