- Email: [email protected]
Sleep profile in anorexia and bulimia nervosa female patients
Accepted Manuscript Sleep profile in Anorexia and bulimia nervosa female patients Tarek Asaad Abdou, Professor, Heba Ibrahim Esawy, Professor, Ghada Abdel Razek Mohamed, Professor, Hanan Hussein Ahmed, Professor, Mahmoud Mamdouh ElHabiby, Assisstant Professor, Sherien Ahmed Khalil, Assisstant Professor, Dr Yomna Ahmed El-Hawary, Lecturer PII:
S1389-9457(18)30160-6
DOI:
10.1016/j.sleep.2018.03.032
Reference:
SLEEP 3687
To appear in:
Sleep Medicine
Received Date: 13 November 2017 Revised Date:
20 February 2018
Accepted Date: 22 March 2018
Please cite this article as: Asaad Abdou T, Esawy HI, Abdel Razek Mohamed G, Hussein Ahmed H, ElHabiby MM, Khalil SA, El-Hawary YA, Sleep profile in Anorexia and bulimia nervosa female patients, Sleep Medicine (2018), doi: 10.1016/j.sleep.2018.03.032. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Sleep profile in Anorexia and bulimia nervosa female patients. Short running title: Sleep profile in females with eating disorders Authors
RI PT
Professor Tarek Asaad Abdou, professor of neuropsychiatry, faculty of medicine, Ain Shams University.
SC
Professor Heba Ibrahim Esawy, professor of neuropsychiatry, faculty of medicine, Ain Shams University. Professor Ghada Abdel Razek Mohamed, professor of neuropsychiatry, faculty of medicine, Ain Shams University.
M AN U
Professor Hanan Hussein Ahmed, professor of neuropsychiatry, faculty of medicine, Ain Shams University. Assisstant Professor Mahmoud Mamdouh El-Habiby, assisstant professor of neuropsychiatry, faculty of medicine, Ain Shams University.
TE D
Assisstant Professor Sherien Ahmed Khalil, assisstant professor of neuropsychiatry, faculty of medicine, Ain Shams University. Dr Yomna Ahmed El-Hawary, lecturer of neuropsychiatry, faculty of medicine, Ain Shams University.
AC C
EP
Corresponding author: Sherien Ahmed Khalil, Phone no: +02012223430493 Email: [email protected], [email protected]. Address: Postal code 25b Omar lofty st, 8th area, Nasr city, 11471, Cairo, Egypt.
ACCEPTED MANUSCRIPT
Introduction
SC
Efforts to identify the sleep architecture of patients with eating disorders have yielded conflicting results. Some studies found rapid eye movement (REM) latency to be shortened in some eating disorder patients [5,6] similar to that which was found in major depression patients [7]; others studies reported normal REM latencies [8,9]. Polysomnography (PSG) in anorexia patients has shown fragmented sleep [8], but the relationship of these abnormalities with concurrent depression is not clear and needs further research. Some studies suggest that altered sleep is related to basal metabolic index and not to psychopathological symptoms, with a positive correlation between BMI (basal metabolic index) and SWA (slow wave activity). This is in agreement with the neurobiological consequences of malnutrition [10].
RI PT
It was recently discovered that hypothalamic neuropeptides (such as orexin A and B) are involved in sleep regulation, feeding and the endocrinal system, proving that sleep and eating are related homeostatic functions [1]. Eating disorders are often associated with sleep complaints and alterations [2]; patients with eating disorders tend to complain of many physical disturbances, particularly in sleep. Studies have shown that patients with eating disorders complain of sleep problems, which is why sleeprelated eating disorders have been recently described as independent clinical syndromes [3]. The anatomical proximity of hypothalamic centers controlling sleep and appetite could be the cause of this correlation between sleep and eating disorders [4].
M AN U
The aim of this study was to fully investigate sleep architecture in anorexia and bulimia patients using a structured sleep disorder questionnaire and all-night PSG, and to find the clinical correlates that could be affecting sleep architecture.
Methods Participants
AC C
Materials
EP
TE D
This was a cross-sectional observation study of female patients with a diagnosis of anorexia nervosa or bulimia, who were either drug naive or non-compliant on treatment for the last month, aged 18–45 years, with or without co-morbid depression, who sought help at the outpatient clinic and inpatient ward of the Institute of Psychiatry, Ain Shams University in the period between January 2012 and May 2014. All cases had no co-morbid physical illness or any neurological problems. Anorexia nervosa patients’ BMI was not less than 15, with no current medical complications to avoid any confounding factors that could affect PSG results. The control group comprised 20 healthy individuals matched for age and sex, with no psychiatric or medical problems. All cases gave a written consent after an explanation of the full study procedure, expected benefits and possible drawbacks. The study was approved by the ethical committee of scientific research, Faculty of Medicine, Ain Shams University.
ACCEPTED MANUSCRIPT
For diagnosis, the Structured Clinical Interview for DSM-IV (SCID-I) [11] was applied to patients to confirm the diagnosis of eating disorders and to detect any comorbid psychiatric disorders, and applied to controls to exclude any psychiatric disorders.
For the assessment of sleep (applied to patients and controls), the following were used: (1) the Structured Sleep Disorder Questionnaire – a comprehensive sleep disorder questionnaire (in Arabic) [13] for the subjective assessment of sleep complaints; (2) all-night PSG for objective assessment of sleep profile performed at the sleep laboratory unit of the Institute of Psychiatry, Ain Shams University, after obtaining consent from all of the participants (a second night assessment was needed for patients who failed to acclimatize during the first night assessment). The apparatus used was Neurofax EEG 2110, Digital Electroencephalograph, Nihon Kohden Corporation, and Tokyo, Japan. The assessment included electroencephalography, electrooculography, electromyography of chin and leg, electrocardiography, respiratory effort snoring, oxygen saturation, and body position. Sleep assessment was performed when the participants were medication free for at least 7 days to exclude the effects of any psychotropic medications on their sleep architecture.
SC
Statistical analysis
RI PT
The Beck Depression Inventory (BDI-II) [12] was applied to the patients group only to assess the presence and severity of depressive status.
M AN U
The collected data was handled using the Statistical Package for Social Science (SPSS 15.0.1 for Windows; SPSS Inc, Chicago, IL, 2001). The following statistical measures were used: mean ± standard deviation (SD) for descriptive statistics and range for parametric numerical data, while median and interquartile range (IQR) were used for non-parametric numerical data.
AC C
EP
TE D
For analytical statistics, Student’s t-test was used to assess the statistical significance of the difference between the two study-group means. Chi-squared test was used to examine the relationship between the two qualitative variables. Fisher’s exact test was used to examine the relationship between the two qualitative variables, Mann– Whitney U-test was used to assess the statistical significance of the difference of a non-parametric variable between the two study groups. A p-value <0.05 was considered significant.
ACCEPTED MANUSCRIPT
Results Participant characteristics
Twenty-three patients underwent the full procedure; all were female, in the age range 20–40 years (mean ± SD = 30 ± 6 years. They were divided into 39.1% (N=9 participants) anorexia nervosa and 60.9% (N=14 participants) bulimia patients.
With BDI-II, 34.7% of the sample were found to have minimal depressive symptoms, while 34.7% (N=8) were in a mild depressive state and 30.43% of cases (N=7) were in a moderate depressive state. Comparing patients with anorexia to those with bulimia regarding depressive symptoms revealed that there was highly statistically significant difference between both groups as depressive symptoms were greater in bulimic patients with BDI-II mean score (15.42±5.96) versus (9.75±1.49) in the anorexic group (p=0.08).
M AN U
SC
Subjective sleep assessment The results of the Sleep Questionnaire are given in Table 1. Significant differences between cases and control groups included insomnia of all types (initial, middle, and late), as well as excessive daytime somnolence and parasomnias in the form of nightmares, sleep-related panic and bruxism indicating significant sleep problems in participants with eating disorders compared to controls.
RI PT
A control group of 20 healthy volunteers matched with the patient group for age and sex and with no physical or psychiatric morbidity according to SCID I (non-patient version). They were selected from the employees of the Institute of Psychiatry, Ain Shams University.
Objective sleep assessment The results of the PSGs are shown in Tables 2 and 3. Regarding sleep latency, sleep efficiency and arousal index, marked effects were found in the eating disorders group: there was lengthening of sleep latency, reduction in sleep efficiency, and significant increases in arousal index in patients as compared
to healthy controls. There were no significant differences in stage 1, but there were
highly significant differences in stage 2 of non-REM (NREM) sleep in patients as
TE D
compared to healthy controls, while stages 3 and 4 as well as (slow wave sleep) SWS% were reduced. Analysis of REM parameters revealed significant differences with respect to REM%, REM density and density of first REM, yet no significant differences with respect to REM latency and duration of first REM. There were no significant differences between cases and controls regarding respiratory variables
EP
(total apnea/hour, obstructive apnea, central apnea and mixed apnea, as well as
AC C
respiratory desaturation index, desaturation/hour and periodic leg movement index).
ACCEPTED MANUSCRIPT
Factors affecting the PSG results
On comparing anorexia nervosa patients to bulimia patients, there was no significant difference regarding any of the PSG findings (NREM and REM parameters or respiratory variables). With respect to the BDI-II scores in anorexia patients there was a significant positive correlation with sleep latency, stage I sleep and REM latency, and a negative correlation to stage III and IV sleep. In bulimia patients there was positive correlation between BDI-II scores and stage I sleep only as shown in Table 4.
The study revealed that all sleep complaints including all types of insomnia, daytime hypersomnolence and parasomnias were significantly high in both anorexia and bulimia patients compared to controls. The most common was initial insomnia (56.5%), followed by interrupted sleep (47.8%), 21.7% with late insomnia and daytime hypersomnolence, and 39% with parasomnias. These results are in agreement with Kim et al., who found that 50.3% of eating disorders patients have sleep problems with no significant differences between bulimia and anorexia, as our results revealed, and also that the most common complaint was initial insomnia followed by mid-sleep awakenings [14].
M AN U
SC
Objective sleep studies performed by PSG revealed that there were marked effects on sleep architecture in patients compared to controls, including sleep latency, sleep efficiency and arousal index; ie there was significant lengthening of sleep latency, significant reduction in sleep efficiency, and significant increase in arousal index in patients of eating disorders compared to healthy controls. Levy et al. found increased sleep latency and reduced sleep efficiency in anorexic patients compared to controls, but no difference was found between bulimic and control subjects [15].
RI PT
Discussion
NREM parameters of patients and control subjects showed that there was significant increase in stage 2 of NREM sleep in eating disorder patients compared to controls, while stages 3, 4 and SWS% were significantly reduced. Regarding REM, our study revealed reduced REM%, decreased REM density and density of first REM, and no significant difference in REM latency or duration. Results of many other studies were controversial, with some in agreement with our results, Marca et al. found prolonged stage II and reduced REM%, but they found reduced REM latency [16]; however, Levy et al. found prolonged stage II and reduced SWS with no difference in REM latency or density, or in density of first REM [15].
AC C
EP
TE D
Regarding the role of depression in affecting sleep, the current study found that late insomnia, prolongation of sleep latency, increase in REM latency and stage I sleep were significantly associated with severity of depressive symptoms, as recorded by Beck in anorexic patients. In the same manner greater reduction in stage III and stage IV in anorexic subjects was associated with more depressive symptoms with BDI-II. In bulimia patients there was positive correlation between BDI-II scores and stage I sleep only. Yet it was not related to any other parameter, so not all sleep problems can be explained by depression; this is in agreement with Marca et al. who found that not all REM alterations were related to depression on correlating them to Hamilton depression score results [17].
ACCEPTED MANUSCRIPT In conclusion, our study took the preliminary steps towards determining and understanding changes in sleep architecture in anorexia and bulimia patients, using both subjective and objective tools. The study found that sleep was affected in both anorexia and bulimia patients equally, and that the changes in sleep cannot be explained by depression alone, which confirms that both sleep and eating disorders are highly correlated. Sleep can be a clinical marker in eating disorder patients. Further research is needed in this area, not only on the co-morbidity between them, but also on why they co-exist and how they affect one another.
1. Salin Pasqual R., Gerashchenco D., Greco M., Blanco-Centurion C., Shiromani P.J. (2001). Hypothalamic regulation of sleep. Neuropsychopharmacology, 25 (suppl. 5), s21-7. 2. Benca, R.M., Casper, R.C. (2000). Eating disorders. In: Kryger M.H., Roth T., Dement W.C. (Eds.). Principles and Practice of Sleep Medicine, 3rd ed. Philadelphia, Saunders, pp. 1140-57.
SC
3. Birketvedt, G.S., Florholmen, J., Sundsfjord, J., Osterud, B., Dinges, D., Bilker, W., Stunkard, A., (1999). Behavioral and neuroendocrine characteristics of the night eating syndrome. JAMA 282, 657–63.
RI PT
References
M AN U
4. McSherry, J., Ashman, G., (1990). Bulimia and sleep disturbance. J Fam Prac 30, 102–3.
5. Neil, JF, Merikangas, JR, Foster, FG, Merikangas, KR, Spiker, DG, Kupfer, DJ (1980). Waking and all-night sleep EEGs in anorexia nervosa. Clin Electroencephalogr, 11:9-15. 6. Katz, JL, Kuperberg, A, PoIIack, CP, WaJsh, BT, Zumoff, B, Weiner, H (1984). Is there a relationship between eating disorders and affective disorder? New evidence from sleep recordings, Am J Psychiatry 141:753-9.
TE D
7. Kupfer, DJ, Foster, FG (1972). Interval between sleep onset and rapid eye movement sleep as an indicator of depression. Lancet ii684-6.
8. Walsh, BT, Goetz, R, Roose, SP, Fingeroth, S, Classman, AH (1985). EEG monitored sleep in anorexia nervosa and bulimia. Biol Psychiatry 20:47-56. 9. Levy, AB, Dixon, KN, Schmidt, H (1987). REM and delta sleep in anorexia nervosa and bulimia. Psychiatry Res. 20:189-97.
AC C
EP
[10] Nobili, L., Baglietto, M.C., De Carli, F., Savoini, M., Schiavi, G., Zanotto, E., et al. (1999). A quantified analysis of sleep electroencephalography in anorectic adolescents. Biol. Psychiatry, 45, 771-5.
ACCEPTED MANUSCRIPT 11. First MB, Spitzer RL and Gibbon M. (1997): Structured Clinical Interview for DSM-IV Axis I Disorders-Clinical Version (SCID-CV). Washington DC: American Psychiatric Press. 12. Beck, A. T., Steer, R. A., Brown, G. K. (1996). Manual for the Beck Depression Inventory-II. San Antonio, TX: Psychological Corporation.
14. Kim, KR, Jung, YC, Shin, MY, Namkoong, K, Kim, JK, Lee, JH (2010). Sleep disturbance in women with eating disorders: Prevalence and clinical characteristics. Psychiatry Res; 176: 88–90. 15. Levy, A, Dixon, K, Schmidt, H. (1988). Sleep architecture in anorexia nervosa and bulimia. Biol Psychiatry 23:99-101. 16. Marcus, MD, Wildes, JE.(2009). Obesity: is it a mental disorder? Int J Eat Disord; 42:739753.
SC
17. Marca, GD, Farina, B, Mennuni, GF, Mazza, S, Giannantonio, MD, Spadini, V, et al. (2004). Microstructure of sleep in eating disorders: preliminary results. Eating Weight Disord; 9:77-80.
RI PT
13. Asaad T, Kahla O (2001). Psychometric Sleep Assessment Instrument: An Arabic Version for Sleep Evaluation. El Faggala, Cairo: El Nahda Book Center.
M AN U
Allison, K. C., Lundgren, J. D., O’Reardon, J. P., Geliebter, A., Gluck, M. E., Vinai, P. (2010). Proposed diagnostic criteria for night eating syndrome. Int J Eat Disord, 43, 241–7. American Academy of Sleep Medicine—AASM (2005). International classification of sleep disorders. Diagnostic and coding manual (2nd Ed.). Westchester, IL: American Academy of Sleep Medicine.
Howell, M.J., Shenck, C.H., & Crow, S.J. (2009). A review of nighttime eating disorders. Sleep Med Rev, 13, 23–34. Tables
(1) Intial insomnia
Controls (N=20)
p
Yes
13
1
0.0001**
No
10
Yes
11
AC C
(2) Middle
Cases (N=23)
EP
Sleep complaints
TE D
Table 1. Case–control comparison regarding sleep questionnaire items.
19 1
0.002**
ACCEPTED MANUSCRIPT
insomnia No
12
Yes
5
0
No
18
20
(4) Daytime hypersomnolen ce
Yes
5
0
No
18
20
(5) Parasomnias
Yes
9
0
No
14
20
0.05*
0.05*
0.002**
*p<0.01. **p<0.001.
Cases
Controls
(N=23)
(N =20)
M AN U
Table 2. Case–control comparison regarding sleep parameters of polysomnography. p
Mean
SD
Mean
SD
Sleep latency
22.69
2.57
18.05
2.68
Sleep efficiency
79.06
4.56
91.58
2.24
Arousal index
3.06
0.79
1.55
0.67
Stage I%
3.53
0.72
3.26
0.99
Stage II%
54.82
1.99
50.44
0.93
0.0001**
Stage III%
9.78
0.87
10.54
0.82
0.0001**
Stage IV%
10.38
1.03
11.13
1.2
0.041*
(slow 20.16
1.73
21.62
1.66
0.01*
1.97
29.45
1.57
0.027*
30.75
REM%
21.79
1.08
24.64
AC C
SWS latency
EP
wave sleep)
0.0001**
0.0001**
0.0001**
0.338
TE D
SWS%
1.21
SC
RI PT
(3) Late insomnia
19
0.0001**
ACCEPTED 6.08 0.178MANUSCRIPT
REM latency
67.55
6.13
70.20
REM density
19.34
0.81
20.02
1.07
0.031*
1.03
7.20
0.96
0.175
0.85
9.28
1.04
0.0001**
0.62
1.14
0.39
0.073
Duration of first 7.63 REM Density
first 10.52
REM PLM1
1.48
PLMI, Periodic Leg Movement Index; REM, rapid eye movement; SD, standard
Table 3. Case–control comparison regarding respiratory variables of sleep. Median
Mann–Whitney U-test p
0.18 0.08
0.32 0.08
0.00 0.00
0.305
0.16 0.07
0.29 0.14
0.00 0.00
0.00 0.01
0.00 0.02
0.00 0.00
0.03 0.06
0.07 0.00
0.00 0.00
0.30 0.04
0.55 0.16
0.00 0.00
0.459
0.317
0.948
0.064
TE D
0.03 0.02
M AN U
SD
0.08 0.05
0.00 0.00
AC C
EP
Total apnea Cases Controls Obstructive apnea Cases Controls Central apnea Cases Controls Mixed apnea Cases Controls Respiratory disturbance Cases Controls Desaturation/hour Cases Controls SD, standard deviation.
Mean
SC
*p<0.01. **p<0.001.
RI PT
deviation; SWS, slow wave sleep.
0.931
ACCEPTED MANUSCRIPT
Stage II Stage III Stage IV SWS% SWS latency
r 0.971 p 0.001* REM density r 0.104 p 0.807 Duration first REM r -0.194 p 0.645 Density first REM r 0.173 p 0.682 REM, rapid eye movement; SWS, slow-wave sleep.
r p r p r p r p
-0.208 0.475 0.481 0.114 -0.194 0.546 -0.158 0.623
AC C
EP
TE D
REM latency
SC
Stage I
Bulimia patients (N=14) Beck score r 0.413 p 0.143 r 0.676 p 0.008* r -0.522 p 0.056 r -0.374 p 0.188 r -0.386 p 0.173 r -0.279 p 0.379 r -0.195 p 0.543
M AN U
Sleep latency
Anorexia patients (N=9) Beck score r 0.950 p 0.0001* r 0.937 p 0.0001* r -0.541 p 0.133 r -0.943 p 0.0001* r -0.907 p 0.001* r -0.238 p 0.570 r -0.070 p 0.870
RI PT
Table 4. Correlation of Beck scores and polysomnography in anorexia and bulimia patients.
ACCEPTED MANUSCRIPT
Highlights
AC C
EP
TE D
M AN U
SC
RI PT
• Sleep and eating disorders are highly correlated and considered a rich area for research. • Insomnia, excessive daytime somnolence and parasomnias were found to be significantly high in the Structured Sleep Disorder Questionnaire of Anorexia and Bulimia patients in comparison to the control group. • Polysomnography revealed multiple areas of sleep affection in both bulimia and anorexia nervosa patients in comparison to controls that could not be explained by depression alone. • Further research is needed on larger samples.
ACCEPTED MANUSCRIPT
The ICMJE Uniform Disclosure Form for Potential Conflicts of Interest
AC C
EP
TE D
M AN U
SC
RI PT
The authors have nothing to declare, there is no conflict of interest.
S1389-9457(18)30160-6
DOI:
10.1016/j.sleep.2018.03.032
Reference:
SLEEP 3687
To appear in:
Sleep Medicine
Received Date: 13 November 2017 Revised Date:
20 February 2018
Accepted Date: 22 March 2018
Please cite this article as: Asaad Abdou T, Esawy HI, Abdel Razek Mohamed G, Hussein Ahmed H, ElHabiby MM, Khalil SA, El-Hawary YA, Sleep profile in Anorexia and bulimia nervosa female patients, Sleep Medicine (2018), doi: 10.1016/j.sleep.2018.03.032. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Sleep profile in Anorexia and bulimia nervosa female patients. Short running title: Sleep profile in females with eating disorders Authors
RI PT
Professor Tarek Asaad Abdou, professor of neuropsychiatry, faculty of medicine, Ain Shams University.
SC
Professor Heba Ibrahim Esawy, professor of neuropsychiatry, faculty of medicine, Ain Shams University. Professor Ghada Abdel Razek Mohamed, professor of neuropsychiatry, faculty of medicine, Ain Shams University.
M AN U
Professor Hanan Hussein Ahmed, professor of neuropsychiatry, faculty of medicine, Ain Shams University. Assisstant Professor Mahmoud Mamdouh El-Habiby, assisstant professor of neuropsychiatry, faculty of medicine, Ain Shams University.
TE D
Assisstant Professor Sherien Ahmed Khalil, assisstant professor of neuropsychiatry, faculty of medicine, Ain Shams University. Dr Yomna Ahmed El-Hawary, lecturer of neuropsychiatry, faculty of medicine, Ain Shams University.
AC C
EP
Corresponding author: Sherien Ahmed Khalil, Phone no: +02012223430493 Email: [email protected], [email protected]. Address: Postal code 25b Omar lofty st, 8th area, Nasr city, 11471, Cairo, Egypt.
ACCEPTED MANUSCRIPT
Introduction
SC
Efforts to identify the sleep architecture of patients with eating disorders have yielded conflicting results. Some studies found rapid eye movement (REM) latency to be shortened in some eating disorder patients [5,6] similar to that which was found in major depression patients [7]; others studies reported normal REM latencies [8,9]. Polysomnography (PSG) in anorexia patients has shown fragmented sleep [8], but the relationship of these abnormalities with concurrent depression is not clear and needs further research. Some studies suggest that altered sleep is related to basal metabolic index and not to psychopathological symptoms, with a positive correlation between BMI (basal metabolic index) and SWA (slow wave activity). This is in agreement with the neurobiological consequences of malnutrition [10].
RI PT
It was recently discovered that hypothalamic neuropeptides (such as orexin A and B) are involved in sleep regulation, feeding and the endocrinal system, proving that sleep and eating are related homeostatic functions [1]. Eating disorders are often associated with sleep complaints and alterations [2]; patients with eating disorders tend to complain of many physical disturbances, particularly in sleep. Studies have shown that patients with eating disorders complain of sleep problems, which is why sleeprelated eating disorders have been recently described as independent clinical syndromes [3]. The anatomical proximity of hypothalamic centers controlling sleep and appetite could be the cause of this correlation between sleep and eating disorders [4].
M AN U
The aim of this study was to fully investigate sleep architecture in anorexia and bulimia patients using a structured sleep disorder questionnaire and all-night PSG, and to find the clinical correlates that could be affecting sleep architecture.
Methods Participants
AC C
Materials
EP
TE D
This was a cross-sectional observation study of female patients with a diagnosis of anorexia nervosa or bulimia, who were either drug naive or non-compliant on treatment for the last month, aged 18–45 years, with or without co-morbid depression, who sought help at the outpatient clinic and inpatient ward of the Institute of Psychiatry, Ain Shams University in the period between January 2012 and May 2014. All cases had no co-morbid physical illness or any neurological problems. Anorexia nervosa patients’ BMI was not less than 15, with no current medical complications to avoid any confounding factors that could affect PSG results. The control group comprised 20 healthy individuals matched for age and sex, with no psychiatric or medical problems. All cases gave a written consent after an explanation of the full study procedure, expected benefits and possible drawbacks. The study was approved by the ethical committee of scientific research, Faculty of Medicine, Ain Shams University.
ACCEPTED MANUSCRIPT
For diagnosis, the Structured Clinical Interview for DSM-IV (SCID-I) [11] was applied to patients to confirm the diagnosis of eating disorders and to detect any comorbid psychiatric disorders, and applied to controls to exclude any psychiatric disorders.
For the assessment of sleep (applied to patients and controls), the following were used: (1) the Structured Sleep Disorder Questionnaire – a comprehensive sleep disorder questionnaire (in Arabic) [13] for the subjective assessment of sleep complaints; (2) all-night PSG for objective assessment of sleep profile performed at the sleep laboratory unit of the Institute of Psychiatry, Ain Shams University, after obtaining consent from all of the participants (a second night assessment was needed for patients who failed to acclimatize during the first night assessment). The apparatus used was Neurofax EEG 2110, Digital Electroencephalograph, Nihon Kohden Corporation, and Tokyo, Japan. The assessment included electroencephalography, electrooculography, electromyography of chin and leg, electrocardiography, respiratory effort snoring, oxygen saturation, and body position. Sleep assessment was performed when the participants were medication free for at least 7 days to exclude the effects of any psychotropic medications on their sleep architecture.
SC
Statistical analysis
RI PT
The Beck Depression Inventory (BDI-II) [12] was applied to the patients group only to assess the presence and severity of depressive status.
M AN U
The collected data was handled using the Statistical Package for Social Science (SPSS 15.0.1 for Windows; SPSS Inc, Chicago, IL, 2001). The following statistical measures were used: mean ± standard deviation (SD) for descriptive statistics and range for parametric numerical data, while median and interquartile range (IQR) were used for non-parametric numerical data.
AC C
EP
TE D
For analytical statistics, Student’s t-test was used to assess the statistical significance of the difference between the two study-group means. Chi-squared test was used to examine the relationship between the two qualitative variables. Fisher’s exact test was used to examine the relationship between the two qualitative variables, Mann– Whitney U-test was used to assess the statistical significance of the difference of a non-parametric variable between the two study groups. A p-value <0.05 was considered significant.
ACCEPTED MANUSCRIPT
Results Participant characteristics
Twenty-three patients underwent the full procedure; all were female, in the age range 20–40 years (mean ± SD = 30 ± 6 years. They were divided into 39.1% (N=9 participants) anorexia nervosa and 60.9% (N=14 participants) bulimia patients.
With BDI-II, 34.7% of the sample were found to have minimal depressive symptoms, while 34.7% (N=8) were in a mild depressive state and 30.43% of cases (N=7) were in a moderate depressive state. Comparing patients with anorexia to those with bulimia regarding depressive symptoms revealed that there was highly statistically significant difference between both groups as depressive symptoms were greater in bulimic patients with BDI-II mean score (15.42±5.96) versus (9.75±1.49) in the anorexic group (p=0.08).
M AN U
SC
Subjective sleep assessment The results of the Sleep Questionnaire are given in Table 1. Significant differences between cases and control groups included insomnia of all types (initial, middle, and late), as well as excessive daytime somnolence and parasomnias in the form of nightmares, sleep-related panic and bruxism indicating significant sleep problems in participants with eating disorders compared to controls.
RI PT
A control group of 20 healthy volunteers matched with the patient group for age and sex and with no physical or psychiatric morbidity according to SCID I (non-patient version). They were selected from the employees of the Institute of Psychiatry, Ain Shams University.
Objective sleep assessment The results of the PSGs are shown in Tables 2 and 3. Regarding sleep latency, sleep efficiency and arousal index, marked effects were found in the eating disorders group: there was lengthening of sleep latency, reduction in sleep efficiency, and significant increases in arousal index in patients as compared
to healthy controls. There were no significant differences in stage 1, but there were
highly significant differences in stage 2 of non-REM (NREM) sleep in patients as
TE D
compared to healthy controls, while stages 3 and 4 as well as (slow wave sleep) SWS% were reduced. Analysis of REM parameters revealed significant differences with respect to REM%, REM density and density of first REM, yet no significant differences with respect to REM latency and duration of first REM. There were no significant differences between cases and controls regarding respiratory variables
EP
(total apnea/hour, obstructive apnea, central apnea and mixed apnea, as well as
AC C
respiratory desaturation index, desaturation/hour and periodic leg movement index).
ACCEPTED MANUSCRIPT
Factors affecting the PSG results
On comparing anorexia nervosa patients to bulimia patients, there was no significant difference regarding any of the PSG findings (NREM and REM parameters or respiratory variables). With respect to the BDI-II scores in anorexia patients there was a significant positive correlation with sleep latency, stage I sleep and REM latency, and a negative correlation to stage III and IV sleep. In bulimia patients there was positive correlation between BDI-II scores and stage I sleep only as shown in Table 4.
The study revealed that all sleep complaints including all types of insomnia, daytime hypersomnolence and parasomnias were significantly high in both anorexia and bulimia patients compared to controls. The most common was initial insomnia (56.5%), followed by interrupted sleep (47.8%), 21.7% with late insomnia and daytime hypersomnolence, and 39% with parasomnias. These results are in agreement with Kim et al., who found that 50.3% of eating disorders patients have sleep problems with no significant differences between bulimia and anorexia, as our results revealed, and also that the most common complaint was initial insomnia followed by mid-sleep awakenings [14].
M AN U
SC
Objective sleep studies performed by PSG revealed that there were marked effects on sleep architecture in patients compared to controls, including sleep latency, sleep efficiency and arousal index; ie there was significant lengthening of sleep latency, significant reduction in sleep efficiency, and significant increase in arousal index in patients of eating disorders compared to healthy controls. Levy et al. found increased sleep latency and reduced sleep efficiency in anorexic patients compared to controls, but no difference was found between bulimic and control subjects [15].
RI PT
Discussion
NREM parameters of patients and control subjects showed that there was significant increase in stage 2 of NREM sleep in eating disorder patients compared to controls, while stages 3, 4 and SWS% were significantly reduced. Regarding REM, our study revealed reduced REM%, decreased REM density and density of first REM, and no significant difference in REM latency or duration. Results of many other studies were controversial, with some in agreement with our results, Marca et al. found prolonged stage II and reduced REM%, but they found reduced REM latency [16]; however, Levy et al. found prolonged stage II and reduced SWS with no difference in REM latency or density, or in density of first REM [15].
AC C
EP
TE D
Regarding the role of depression in affecting sleep, the current study found that late insomnia, prolongation of sleep latency, increase in REM latency and stage I sleep were significantly associated with severity of depressive symptoms, as recorded by Beck in anorexic patients. In the same manner greater reduction in stage III and stage IV in anorexic subjects was associated with more depressive symptoms with BDI-II. In bulimia patients there was positive correlation between BDI-II scores and stage I sleep only. Yet it was not related to any other parameter, so not all sleep problems can be explained by depression; this is in agreement with Marca et al. who found that not all REM alterations were related to depression on correlating them to Hamilton depression score results [17].
ACCEPTED MANUSCRIPT In conclusion, our study took the preliminary steps towards determining and understanding changes in sleep architecture in anorexia and bulimia patients, using both subjective and objective tools. The study found that sleep was affected in both anorexia and bulimia patients equally, and that the changes in sleep cannot be explained by depression alone, which confirms that both sleep and eating disorders are highly correlated. Sleep can be a clinical marker in eating disorder patients. Further research is needed in this area, not only on the co-morbidity between them, but also on why they co-exist and how they affect one another.
1. Salin Pasqual R., Gerashchenco D., Greco M., Blanco-Centurion C., Shiromani P.J. (2001). Hypothalamic regulation of sleep. Neuropsychopharmacology, 25 (suppl. 5), s21-7. 2. Benca, R.M., Casper, R.C. (2000). Eating disorders. In: Kryger M.H., Roth T., Dement W.C. (Eds.). Principles and Practice of Sleep Medicine, 3rd ed. Philadelphia, Saunders, pp. 1140-57.
SC
3. Birketvedt, G.S., Florholmen, J., Sundsfjord, J., Osterud, B., Dinges, D., Bilker, W., Stunkard, A., (1999). Behavioral and neuroendocrine characteristics of the night eating syndrome. JAMA 282, 657–63.
RI PT
References
M AN U
4. McSherry, J., Ashman, G., (1990). Bulimia and sleep disturbance. J Fam Prac 30, 102–3.
5. Neil, JF, Merikangas, JR, Foster, FG, Merikangas, KR, Spiker, DG, Kupfer, DJ (1980). Waking and all-night sleep EEGs in anorexia nervosa. Clin Electroencephalogr, 11:9-15. 6. Katz, JL, Kuperberg, A, PoIIack, CP, WaJsh, BT, Zumoff, B, Weiner, H (1984). Is there a relationship between eating disorders and affective disorder? New evidence from sleep recordings, Am J Psychiatry 141:753-9.
TE D
7. Kupfer, DJ, Foster, FG (1972). Interval between sleep onset and rapid eye movement sleep as an indicator of depression. Lancet ii684-6.
8. Walsh, BT, Goetz, R, Roose, SP, Fingeroth, S, Classman, AH (1985). EEG monitored sleep in anorexia nervosa and bulimia. Biol Psychiatry 20:47-56. 9. Levy, AB, Dixon, KN, Schmidt, H (1987). REM and delta sleep in anorexia nervosa and bulimia. Psychiatry Res. 20:189-97.
AC C
EP
[10] Nobili, L., Baglietto, M.C., De Carli, F., Savoini, M., Schiavi, G., Zanotto, E., et al. (1999). A quantified analysis of sleep electroencephalography in anorectic adolescents. Biol. Psychiatry, 45, 771-5.
ACCEPTED MANUSCRIPT 11. First MB, Spitzer RL and Gibbon M. (1997): Structured Clinical Interview for DSM-IV Axis I Disorders-Clinical Version (SCID-CV). Washington DC: American Psychiatric Press. 12. Beck, A. T., Steer, R. A., Brown, G. K. (1996). Manual for the Beck Depression Inventory-II. San Antonio, TX: Psychological Corporation.
14. Kim, KR, Jung, YC, Shin, MY, Namkoong, K, Kim, JK, Lee, JH (2010). Sleep disturbance in women with eating disorders: Prevalence and clinical characteristics. Psychiatry Res; 176: 88–90. 15. Levy, A, Dixon, K, Schmidt, H. (1988). Sleep architecture in anorexia nervosa and bulimia. Biol Psychiatry 23:99-101. 16. Marcus, MD, Wildes, JE.(2009). Obesity: is it a mental disorder? Int J Eat Disord; 42:739753.
SC
17. Marca, GD, Farina, B, Mennuni, GF, Mazza, S, Giannantonio, MD, Spadini, V, et al. (2004). Microstructure of sleep in eating disorders: preliminary results. Eating Weight Disord; 9:77-80.
RI PT
13. Asaad T, Kahla O (2001). Psychometric Sleep Assessment Instrument: An Arabic Version for Sleep Evaluation. El Faggala, Cairo: El Nahda Book Center.
M AN U
Allison, K. C., Lundgren, J. D., O’Reardon, J. P., Geliebter, A., Gluck, M. E., Vinai, P. (2010). Proposed diagnostic criteria for night eating syndrome. Int J Eat Disord, 43, 241–7. American Academy of Sleep Medicine—AASM (2005). International classification of sleep disorders. Diagnostic and coding manual (2nd Ed.). Westchester, IL: American Academy of Sleep Medicine.
Howell, M.J., Shenck, C.H., & Crow, S.J. (2009). A review of nighttime eating disorders. Sleep Med Rev, 13, 23–34. Tables
(1) Intial insomnia
Controls (N=20)
p
Yes
13
1
0.0001**
No
10
Yes
11
AC C
(2) Middle
Cases (N=23)
EP
Sleep complaints
TE D
Table 1. Case–control comparison regarding sleep questionnaire items.
19 1
0.002**
ACCEPTED MANUSCRIPT
insomnia No
12
Yes
5
0
No
18
20
(4) Daytime hypersomnolen ce
Yes
5
0
No
18
20
(5) Parasomnias
Yes
9
0
No
14
20
0.05*
0.05*
0.002**
*p<0.01. **p<0.001.
Cases
Controls
(N=23)
(N =20)
M AN U
Table 2. Case–control comparison regarding sleep parameters of polysomnography. p
Mean
SD
Mean
SD
Sleep latency
22.69
2.57
18.05
2.68
Sleep efficiency
79.06
4.56
91.58
2.24
Arousal index
3.06
0.79
1.55
0.67
Stage I%
3.53
0.72
3.26
0.99
Stage II%
54.82
1.99
50.44
0.93
0.0001**
Stage III%
9.78
0.87
10.54
0.82
0.0001**
Stage IV%
10.38
1.03
11.13
1.2
0.041*
(slow 20.16
1.73
21.62
1.66
0.01*
1.97
29.45
1.57
0.027*
30.75
REM%
21.79
1.08
24.64
AC C
SWS latency
EP
wave sleep)
0.0001**
0.0001**
0.0001**
0.338
TE D
SWS%
1.21
SC
RI PT
(3) Late insomnia
19
0.0001**
ACCEPTED 6.08 0.178MANUSCRIPT
REM latency
67.55
6.13
70.20
REM density
19.34
0.81
20.02
1.07
0.031*
1.03
7.20
0.96
0.175
0.85
9.28
1.04
0.0001**
0.62
1.14
0.39
0.073
Duration of first 7.63 REM Density
first 10.52
REM PLM1
1.48
PLMI, Periodic Leg Movement Index; REM, rapid eye movement; SD, standard
Table 3. Case–control comparison regarding respiratory variables of sleep. Median
Mann–Whitney U-test p
0.18 0.08
0.32 0.08
0.00 0.00
0.305
0.16 0.07
0.29 0.14
0.00 0.00
0.00 0.01
0.00 0.02
0.00 0.00
0.03 0.06
0.07 0.00
0.00 0.00
0.30 0.04
0.55 0.16
0.00 0.00
0.459
0.317
0.948
0.064
TE D
0.03 0.02
M AN U
SD
0.08 0.05
0.00 0.00
AC C
EP
Total apnea Cases Controls Obstructive apnea Cases Controls Central apnea Cases Controls Mixed apnea Cases Controls Respiratory disturbance Cases Controls Desaturation/hour Cases Controls SD, standard deviation.
Mean
SC
*p<0.01. **p<0.001.
RI PT
deviation; SWS, slow wave sleep.
0.931
ACCEPTED MANUSCRIPT
Stage II Stage III Stage IV SWS% SWS latency
r 0.971 p 0.001* REM density r 0.104 p 0.807 Duration first REM r -0.194 p 0.645 Density first REM r 0.173 p 0.682 REM, rapid eye movement; SWS, slow-wave sleep.
r p r p r p r p
-0.208 0.475 0.481 0.114 -0.194 0.546 -0.158 0.623
AC C
EP
TE D
REM latency
SC
Stage I
Bulimia patients (N=14) Beck score r 0.413 p 0.143 r 0.676 p 0.008* r -0.522 p 0.056 r -0.374 p 0.188 r -0.386 p 0.173 r -0.279 p 0.379 r -0.195 p 0.543
M AN U
Sleep latency
Anorexia patients (N=9) Beck score r 0.950 p 0.0001* r 0.937 p 0.0001* r -0.541 p 0.133 r -0.943 p 0.0001* r -0.907 p 0.001* r -0.238 p 0.570 r -0.070 p 0.870
RI PT
Table 4. Correlation of Beck scores and polysomnography in anorexia and bulimia patients.
ACCEPTED MANUSCRIPT
Highlights
AC C
EP
TE D
M AN U
SC
RI PT
• Sleep and eating disorders are highly correlated and considered a rich area for research. • Insomnia, excessive daytime somnolence and parasomnias were found to be significantly high in the Structured Sleep Disorder Questionnaire of Anorexia and Bulimia patients in comparison to the control group. • Polysomnography revealed multiple areas of sleep affection in both bulimia and anorexia nervosa patients in comparison to controls that could not be explained by depression alone. • Further research is needed on larger samples.
ACCEPTED MANUSCRIPT
The ICMJE Uniform Disclosure Form for Potential Conflicts of Interest
AC C
EP
TE D
M AN U
SC
RI PT
The authors have nothing to declare, there is no conflict of interest.