Volume 9, Number 7 | July 15, 2013

Transcription

ISSN 1550-9389
Journal of Clinical Sleep Medicine
in this issue:
The Effectiveness of Light/Dark Exposure to Treat Insomnia in Female
Nurses Undertaking Shift Work during the Evening/Night Shift
Huang; Tsai; Chen; Hsu
Commentary on Huang et al.
Bright Light Improves Sleep and Psychological Health in Shift Working
Nurses
Bjorvatn; Waage
Volume 9, Number 7 | July 15, 2013
Pages 637-734
Middle-of-the-Night Hypnotic Use in a Large National Health Plan
Roth; Berglund; Shahly; Shillington; Stephenson; Kessler
Obstructive Sleep Apnea in Patients with End-Stage Lung Disease
Romem; Iacono; McIlmoyle; Patel; Reed; Verceles; Scharf
Defending Sleepwalkers with Science and an Illustrative Case
Cartwright; Guilleminault
July 15, 2013
Volume 9(7) 2013 Pages 637-734
Official Publication of the
American Academy of Sleep Medicine
www.aasmnet.org
Volume 9, Number 7
Official publication of the
American Academy of
Sleep Medicine
July 15, 2013
Pages 637-734
Editor
Stuart F. Quan, M.D., F.A.A.S.M., Boston, MA/Tucson, AZ
Scope
Deputy Editor
Daniel J. Buysse, M.D., F.A.A.S.M., Pittsburgh, PA
JCSM Journal of Clinical Sleep Medicine focuses on clinical sleep medicine.
Its emphasis is publication of papers
with direct applicability and/or relevance
to the clinical practice of sleep medicine.
This includes, clinical trials, clinical reviews, clinical commentary and debate,
medical economic/practice perspectives, case series and novel/interesting
case reports. In addition, the journal will
publish proceedings from conferences,
workshops and symposia sponsored by
the American Academy of Sleep Medicine or other organizations related to improving the practice of sleep medicine.
Associate Editors
Richard B. Berry, M.D.,
F.A.A.S.M., Gainesville, FL
Birgit Högl, M.D.,
Innsbruck, Austria
Amy L. Meoli, M.D., F.A.A.S.M.,
Lee K. Brown, M.D., F.A.A.S.M.,
Conrad Iber, M.D.,
Minneapolis, MN
Robert L. Owens, M.D.
Boston, MA
Rohit Budhiraja, M.D.,
Vishesh Kapur, M.D.,
Sairam Parthasarathy, M.D.,
Andrew L. Chesson, M.D.,
Douglas B. Kirsch, M.D.,
Stephen H. Sheldon, D.O.,
David Dinges, Ph.D.,
Michael Littner, M.D.
Nathaniel F. Watson, M.D.,
Susan M. Harding, M.D.,
W. Vaughn McCall, M.D.,
Albuquerque, NM
F.A.A.S.M., Tucson, AZ
F.A.A.S.M., Shreveport, LA
Philadelphia, PA
F.A.A.S.M., Birmingham, AL
F.A.A.S.M., Seattle, WA
F.A.A.S.M., Brighton, MA
F.A.A.S.M., Sepulveda, CA
Kansas City, MO
F.A.A.S.M., Tucson, AZ
F.A.A.S.M., Chicago, IL
F.A.A.S.M., Seattle, WA
F.A.A.S.M., Augusta, GA
Executive Director
Jerome A. Barrett, Darien, IL
Managing Editor
Andrew Miller, Darien, IL
Editorial Board
Candice A. Alfano, Ph.D.,
Houston, TX
Mexico City, Mexico
Armando Castorena-Maldonado, M.D.
Leila K. Gozal, M.D.,
Matthew T. Naughton, M.D.,
Wolfgang W. Schmidt-Nowara, M.D.,
Fernanda R. Almeida, D.D.S., M.Sc.,
Alejandro D. Chediak, M.D.,
Jason P. Kirkness, Ph.D.,
James M. Parish, M.D.,
Shirin Shafazand, M.D., F.A.A.S.M.,
Monica L. Andersen, M.Sci., Ph.D.,
Ronald D. Chervin, M.D.,
Meir H. Kryger, M.D., F.A.A.S.M.,
West Haven, CT
Thomas Penzel, Ph.D.,
Berlin, Germany
Aaron E. Sher, M.D.,
Donna Arand, Ph.D., F.A.A.S.M.,
Dayton, OH
Madeline Grigg-Damberger, M.D.,
Albuquerque, NM
Clete A. Kushida, M.D., Ph.D.,
Barbara A. Phillips, M.D.,
Lexington, KY
Michael H. Silber, M.B., Ch.B.,
F.A.A.S.M., Rochester, MN
Kristen Archbold, Ph.D., R.N.,
Sally L. Davidson Ward, M.D.,
Peretz Lavie, Ph.D.,
Haifa, Israel
Daniel Picchietti, M.D., F.A.A.S.M.,
Urbana, IL
Edward J. Stepanski, Ph.D.,
Carol M. Baldwin, Ph.D., R.N.,
William C. Dement, M.D., Ph.D.,
Teofilo L. Lee-Chiong, M.D.,
Giora Pillar, M.D., Ph.D.,
Haifa, Israel
Patrick J. Strollo, Jr., M.D.,
F.A.A.S.M., Pittsburgh, PA
Naresh M. Punjabi, M.D., F.A.A.S.M.,
David P. White, M.D., F.A.A.S.M.,
B. Tucker Woodson, M.D., F.A.A.S.M.,
Ph.D., A.B.D.S.M., Vancouver, BC, Canada
Sao Paulo, Brazil
Tucson, AZ
F.A.A.S.M., Miami Beach, FL
Ann Arbor, MI
F.A.A.S.M., Los Angeles, CA
Chicago, IL
Baltimore, MD
F.A.A.S.M., Stanford, CA
Tempe, AZ
F.A.A.S.M., Palo Alto, CA
F.A.A.S.M., Denver, CO
Mary Susan Esther Banks, M.D.,
Lawrence J. Epstein, M.D.,
Ching-Chi Lin, M.D.,
Prahran, Australia
Scottsdale, AZ
F.A.A.S.M., Dallas, TX
Miami, FL
Albany, NY
F.A.A.S.M., Memphis, TN
Charlotte, NC
F.A.A.S.M., Bedford, MA
Taipei, Taiwan
Baltimore, MD
Kenneth R. Casey, M.D., F.A.A.S.M.,
Birgit Frauscher, M.D.,
Innsbruck, Austria
Babak Mokhlesi, M.D., M.Sc.,
F.A.A.S.M., Chicago, IL
Daniel O. Rodenstein, M.D., Ph.D.,
Brussels, Belgium
Milwaukee, WI
Journal of Clinical Sleep Medicine
(JCSM) (Online 1550-9397; Website:
www.aasmnet.org/jcsm) is published online monthly on the 15th of the month by
the American Academy of Sleep Medicine, 2510 North Frontage Road, Darien,
IL 60561-1511, phone (630) 737-9700
and fax (630) 737-9790.
national): $140.00. Prorated subscriptions
are not available. Subscriptions begin with
the January issue of the current year. Renewals should be secured as early in the
year as possible to avoid uninterrupted
service. Questions about subscriptions
(including payments, billing procedures,
or policy matters) should be directed to
the AASM office at (630) 737-9700.
count Executive at advertising@aasmnet.org for complete information.
copyright.com. There are royalty fees associated with such permissions.
PERMISSION TO REPRODUCE: Written permission to reproduce, in print or
electronically, whole articles or any parts
of works, figures or tables published in
JCSM must be obtained prior to publication. Permission for republication must be
arranged through the Copyright Clearance
Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, phone (978) 750-8400 or
fax (978) 646-8600 or URL http://www.
REPRINTS: For author reprints contact
the AASM office. For commercial reprint
orders contact Cenveo Publisher Services,
4810 Williamsburg Road, #2, Hurlock,
MD 21643 or Reprints2@cadmus.com
the authors thereof and not of the American Academy of Sleep Medicine, or of its
officers, regents, members or employees.
The Editor-in-Chief, the American Academy of Sleep Medicine and its officers,
regents, members and employees disclaim all responsibility for any injury to
persons or property resulting from any
ideas or products referred to in articles
contained in this journal.
Cincinnati, OH
ANNUAL SUBSCRIPTION RATES:
Subscription rates for Volume 9, 2013:
Individual Online (US and International):
$75.00; Institutional Online (US and Inter-
ADVERTISING: Digital advertising is
available on www.aasmnet.org/jcsm.
Please contact the National Sales Ac-
DISCLAIMER: The statements and
opinions contained in editorials and articles in this journal are solely those of
Boston, MA
© 2013 American Academy of Sleep
Medicine
Table of Contents
Continuing Medical Education Offerings
640
Vol. 9, No. 7
681
Targeted Case Finding for OSA within the Primary Care Setting
Keith R. Burgess; Adrian Havryk; Stephen Newton; Willis H. Tsai;
William A. Whitelaw
687
Scientific Investigations
641
The Effectiveness of Light/Dark Exposure to Treat Insomnia
in Female Nurses Undertaking Shift Work during the
Evening/Night Shift
Li-Bi Huang; Mei-Chu Tsai; Ching-Yen Chen; Shih-Chieh Hsu
647
Commentary on Huang et al.
Bright Light Improves Sleep and Psychological Health in Shift
Working Nurses
Bjørn Bjorvatn; Siri Waage
649
Serum Brain-Derived Neurotrophic Factor Levels Are
Associated with Dyssomnia in Females, but not Males,
among Japanese Workers
Reiko Nishichi; Yu Nufuji; Masakazu Washio; Shuzo Kumagai
655
Identification of Insomnia in a Sleep Center Population Using
Electronic Health Data Sources and the Insomnia Severity Index
Carl A. Severson; Willis H. Tsai; Paul E. Ronksley; Sachin R. Pendharkar
Obstructive Sleep Apnea in Patients with End-Stage
Lung Disease
Ayal Romem; Aldo Iacono; Elizabeth McIlmoyle; Kalpesh P. Patel;
Robert M. Reed; Avelino C. Verceles; Steven M. Scharf
695
Inter-Observer Reliability of Candidate Predictive
Morphometric Measurements for Women with Suspected
Obstructive Sleep Apnea
John A. Gjevre; Regina M. Taylor-Gjevre; John K. Reid;
Robert Skomro; David Cotton
701
Validating Actigraphy as a Measure of Sleep for
Preschool Children
Marie-Ève Bélanger; Annie Bernier; Jean Paquet; Valérie Simard;
Julie Carrier
707
Heart Rate Variability in Sleep-Related Migraine without Aura
Catello Vollono; Valentina Gnoni; Elisa Testani; Serena Dittoni;
Anna Losurdo; Salvatore Colicchio; Chiara Di Blasi; Salvatore Mazza;
Benedetto Farina; Giacomo Della Marca
661
Case Reports
715
669
717
Thomas Roth; Patricia Berglund; Victoria Shahly; Alicia C. Shillington;
Judith J. Stephenson; Ronald C. Kessler
Association between Sleep Duration and the Mini-Mental Score:
The Northern Manhattan Study
Alberto R. Ramos; Chuanhui Dong; Mitchell S. V. Elkind;
Bernadette Boden-Albala; Ralph L. Sacco; Tatjana Rundek;
Clinton B. Wright
Status Cataplecticus Precipitated by Abrupt Withdrawal of
Venlafaxine
Janice Wang; Harly Greenberg
Nocturnal Diaphoresis Secondary to Mild Obstructive Sleep
Apnea in a Patient with a History of Two Malignancies
Robert Daniel Vorona; Mariana Szklo-Coxe; Mark Fleming; J.
Catesby Ware
675
Characterization of REM Sleep without Atonia in Patients with
Narcolepsy and Idiopathic Hypersomnia using AASM Scoring
Manual Criteria
Lourdes M. DelRosso; Andrew L. Chesson Jr.; Romy Hoque
The current issue podcast and instructions to authors are available online at www.aasmnet.org/jcsm.
Table of Contents
721
Rosalind D. Cartwright; Christian Guilleminault
Sleep Medicine Pearls
727
Ventricular or Pseudo-Ventricular Tachycardia on
Polysomnogram
Romy Hoque; Lourdes DelRosso
Letters to the Editor
731
Is Prediction of CPAP Adherence in Obstructive Sleep Apnea in
the Perioperative Setting Feasible?
Antonio M. Esquinas; Peter Cistulli
733
Response to Esquinas and Cistulli
CPAP Adherence during the Perioperative Period
Babak Mokhlesi; Amy S. Guralnick
The current issue podcast and instructions to authors are available online at www.aasmnet.org/jcsm.
Vol. 9, No. 7
Continuing Medical Education Offerings
Instructions for Earning Credit
Participants may read the selected continuing medical education (CME) articles in this issue of JCSM and complete the online
CME post-test and evaluation at http://www.aasmnet.org/JCSM/ within one year of the date of publication to receive AMA
PRA Category 1 Credits.™ Each activity – journal article, post-test, and evaluation – is estimated to take approximately 30
minutes to complete.
The ACCME mandates that accredited providers only offer AMA PRA Category 1 Credits™ to physicians. Non-physicians
will be provided with a letter of participation indicating the number of AMA PRA Category 1 Credits™ awarded for the activity in which they participated. Non-physicians requesting letters of participation will be assessed the same fees as physicians
requesting AMA PRA Category 1 Credit,™ if applicable; visit the journal website for information on applicable fees.
Each journal article designated for CME is worth 0.5 AMA PRA Category 1 Credits.™ To earn the credit, physicians must read
the article and then correctly answer a minimum of 3 out of 5 post-test questions. Upon successful completion of the post-test,
the learner may download a participation letter indicating the number of credits earned.
Accreditation Statement
The American Academy of Sleep Medicine is accredited by the ACCME to provide continuing medical education for
physicians. The American Academy of Sleep Medicine designates this educational activity for a maximum of 0.5 category 1
credits toward the AMA PRA Category 1 Credits.™ Physicians should only claim credit commensurate with the extent of their
participation in the activity.
Statement of Educational Purpose/Overall Education Objectives
JCSM is a peer-reviewed clinical journal addressing sleep, circadian rhythms, and the diagnosis and treatment of the broad
spectrum of sleep disorders. Its mission and educational purpose is to promote the science and art of sleep medicine and
sleep research. Sleep disorders medicine draws clinical and scientific applications from a wide variety of primary disciplines,
including pulmonology, neurology, psychiatry, psychology, otolaryngology, and dentistry. Readers of JCSM should be able
to: 1) appraise sleep research in basic science and clinical investigation; 2) interpret new information and updates on clinical
diagnosis/treatment and apply those strategies to their practice; 3) analyze articles for the use of sound scientific and medical
problems; and 4) recognize the inter-relatedness/dependence of sleep medicine with primary disciplines.
Articles in this issue that may be read for CME credit
Page
The Effectiveness of Light/Dark Exposure to Treat Insomnia in Female Nurses Undertaking
Shift Work during the Evening/Night Shift
Objective: Learn how the use of bright light therapy in shift workers may improve their insomnia,
anxiety, and depression severity.
641
Objective: Learn more about the incidence of Middle-of-the-Night off-label hypnotic use.
661
Journal of Clinical Sleep Medicine, Vol. 9, No. 7, 2013
640
http://dx.doi.org/10.5664/jcsm.2824
The Effectiveness of Light/Dark Exposure to Treat Insomnia
in Female Nurses Undertaking Shift Work during the
Evening/Night Shift
Li-Bi Huang, B.S.1,2; Mei-Chu Tsai, M.S.1,2; Ching-Yen Chen, M.D.2,4; Shih-Chieh Hsu, M.D.2,3,4
Department of Nursing, Chang Gung Memorial Hospital, Linkou, Taiwan; 2Department of Psychiatry, Chang
Gung Memorial Hospital, Linkou, Taiwan; 3Sleep Center, Chang Gung Memorial Hospital, Linkou, Taiwan;
4
Chang Gung University College of Medicine, Taoyuan, Taiwan
S cientific I nvesti g ations
1
Study Objectives: The present study investigated whether
bright light exposure during the first half of the evening/night
shift combined with light attenuation in the morning is effective in improving sleep problems in nurses undertaking rotating
shift work who suffer from clinical insomnia.
Methods: This was a prospective, randomized control study.
The Insomnia Severity Index (ISI) and the Hospital Anxiety
Depression Scale (HADS) were used to evaluate insomnia
and anxiety/depression severity, respectively. Female hospital nurses on rotating shifts during the evening or night shift
with an ISI score > 14 were enrolled. Subjects in the treatment
group (n = 46) were exposed to bright light at 7,000-10,000 lux
for ≥ 30 minutes. Exposure was continued for at least 10 days
during 2 weeks, and the subjects avoided daytime outdoor sun
exposure after work by wearing dark sunglasses. Subjects in
the control group (n = 46) were not exposed to bright light,
but also wore sunglasses after work. Statistical analyses
were performed to examine group differences and differences
across treatments.
Results: After treatment, the treatment group showed signifi-
cant improvements in the ISI score and the HADS total and
subscale scores as compared with pre-treatment. The ISI,
HADS, and subscales of the HADS scores were significantly
improved across treatments in the treatment group as compared with the control group.
Conclusions: The design of this study is easy to put into practice in the real world. This is the first study to document that
a higher intensity and briefer duration of bright light exposure
during the first half of the evening/night shift with a daytime
darkness procedure performed in rotating shift work female
nurses suffering from clinical insomnia could improve their insomnia, anxiety, and depression severity.
Keywords: Bright light, circadian rhythm, shift work, rotating
shift, insomnia, anxiety, depression, nurses
Commentary: A commentary on this article appears in this
issue on page 647.
Citation: Huang LB; Tsai MC; Chen CY; Hsu SC. The effectiveness of light/dark exposure to treat insomnia in female
nurses undertaking shift work during the evening/night shift.
J Clin Sleep Med 2013;9(7):641-646.
I
nsomnia is common in the general population. From the
review study of Ohayon,1 about one-third of the general
population suffers symptoms of insomnia. The prevalence is
between 9% and 15% when daytime consequences of insomnia
are taken into account. A strong association between insomnia
and anxiety/depression is also found in the community and in
hospitals.2,3 Chronic insomnia may aggravate the severity of
anxiety and depression.4-6 In addition, while insomnia or excessive sleepiness is a risk factor for depression in all individuals,
it is a much greater risk factor in rotating or night shift workers.7
Shift work disorder (SWD) is a circadian rhythm sleep disorder characterized by sleepiness or insomnia that can be attributed to the person’s work schedule. According to second
edition of the International Classification of Sleep Disorders
(ICSD-2),8 the major feature of circadian rhythm sleep disorder
is “a misalignment between the patient’s sleep pattern and the
sleep pattern that is desired or regarded as the societal norm.”
People who work shifts have great difficulty adjusting their
internal clocks and develop SWD due to a mismatch between
the sleep/wake schedule required by their jobs and their own
circadian sleep/wake cycles. It is estimated that around 20%
Brief Summary
Current Knowledge/Study Rationale: Bright light exposure during the
first half of the night shift and daytime darkness have been shown to
improve daytime sleep and nocturnal functioning in nurses working the
night shift. We aimed to investigate the effectiveness of bright light exposure during the first half of the evening/night shift combined with light
attenuation in the morning in nurses undertaking rotating shift work who
suffer from clinical insomnia.
Study Impact: Bright light exposure with morning time darkness is effective to improve insomnia, anxiety, and depression severity in rotating
shift work female nurses suffering from clinical insomnia. The design of
this study is easy to put into practice in the real world.
of the US and 35% of the Taiwan labor force works night,
evening, or rotating shifts,9,10 and that 10% of these individuals suffer from SWD.7 Circadian misalignment can be caused
by shift work. The resulting circadian misalignment associated
with shift work can produce significant morbidity associated
with disturbed sleep11,12 and impaired alertness.13,14 The longitudinal study of Bara and Arber15 found that women’s mental
health and anxiety/depression were more adversely affected by
641
LB Huang, MC Tsai, CY Chen et al
varied shift patterns than by night work. Nurses are the largest
working group in a hospital and most are on a rotating shift
work schedule. Research has shown that shift work, in particular night work, can have negative effects on the health, safety,
and well-being of nurses.16,17 The prevalence of depression is
significantly higher in those who work rotating and night shifts
than in day workers.7 Lin et al.18 also found that female nurses
who have a rotation shift work schedule tend to experience poor
sleep quality and mental health in Taiwan.
Light is the dominant environmental time cue that entrains
the human circadian clock to a 24-h day, and the timing of light
exposure will determine whether the internal clock is phase delayed or advanced.19 Using this principle, bright light exposure
during the first half of the night shift and daytime darkness have
been shown to improve daytime sleep and nocturnal functioning in night shift workers.20,21
However, no studies have investigated whether bright light
exposure could improve sleep problems in nurses working rotating or night shifts with moderate to severe insomnia (clinical insomnia). The aim of the present study was to investigate whether
bright light exposure during the first half of the evening/night shift
combined with light attenuation in the morning would be effective
in improving sleep problems in nurses working rotating shifts during the evening/night shift who suffer from clinical insomnia in a
hospital setting. In addition, we also investigated the changes in
the anxiety and depression scores after bright light intervention.
exposure of 7,000-10,000 lux could be obtained at a distance of
around 70 cm from the light box to the nurse. Light intensity was
measured using a Lutron Electronic LX-1102 light meter. Ward
illumination at night is maintained in the range of 100-400 lux
in our hospital. Treatment was continued for ≥ 10 days during 2
weeks, and daytime outdoor sun exposure after work and before
sleep was avoided by the subjects by wearing dark sunglasses
with UV protection, including on off-days. The subjects in the
control group were not exposed to artificial bright light, but also
wore sunglasses to avoid outdoor sun exposure after work and
before sleep. Other aspects of lifestyle were not changed, including off-days, in either group. Subjects who used sleep medications
did not change the pattern of use across the treatment duration. All
participants were reminded of the study procedure by a telephone
call before and after work to enhance protocol adherence.
Subject Enrollment
This study was conducted at a medical center in northern
Taiwan from May 1, 2009, to March 31, 2010. Around 2,500
three-shift rotating nurses in the most recent 6 months received
information regarding the study by e-mail and during nursing
meetings in the hospital in May 2009. Using t-tests and a onesided type I error of 5%, we estimated that 50 participants in each
group would be necessary to achieve a power of 80% to detect
the effect size at 0.5. One hundred two nurses agreed to join this
study. Ten participants were excluded because their pre-treatment
ISI score was < 15. A total of 92 rotating-shift female hospital
nurses working the evening shift (4 pm to midnight) or night shift
(midnight to 8 am) with an ISI score > 14 were recruited. Written
informed consent was obtained from all participants before study
enrollment. Forty-six subjects were in the treatment group, and
the remainder were in the control group. All subjects completed
the study procedure reported by themselves.
METHODS
Study Design
This was a randomized controlled study performed to assess
the effectiveness of bright light exposure in treating insomnia,
anxiety, and depression in female nurses working a 3-shift rotation
during the evening/night shift. The project was approved by the
Institutional Review Board of Chang Gung Memorial Hospital.
The inclusion criteria were: (1) a score on the Chinese version of
the Insomnia Severity Index (ISI)22 > 14; (2) rotating-shift female
nurses working the evening/night shift; (3) 3-shift rotation including day, evening, and night shifts in the most recent 6 months;
and (4) the same work schedule during treatment. The exclusion
criteria were: (1) substance abuse or dependence according to the
criteria of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV),23 including caffeine, alcohol,
nicotine, and over-the-counter sleeping pills; and (2) unstable
physical conditions. Participants were randomly divided into 2
groups, the treatment group and the control group. Randomization
was performed using a random digit table. An even number was
allocated to the treatment group and an odd number was allocated
to the control group. All subjects completed the Chinese version
of the Hospital Anxiety and Depression Scale (HADS)24 and the
ISI during work before and after intervention. This study was performed in a real workplace. Subjects in the treatment group were
exposed to artificial bright light of 7,000-10,000 lux for ≥ 30 min;
evening shift exposure took place between 19:30 and 20:30, while
night shift exposure occurred between 23:00 and midnight. During exposure, the subjects undertook charting or reading. Light
was delivered by an Apollo briteLITE 6. A light box was placed
at a 45 degree angle from the face, just above eye level. Light
Instruments
Insomnia Severity Index (ISI)
The ISI, developed by Morin, is a 7-item self-rated scale
designed to assess subjective perception of the severity of insomnia.22 The scale contains items including difficulty falling
asleep, difficulty maintaining sleep, early morning awakening,
satisfaction with sleep, concerns about insomnia, and the functional impact of insomnia. The total score, ranging from 0 to 28,
can be used to categorize patients into different levels of insomnia severity (0-7, no clinically significant insomnia; 8-14, subthreshold insomnia; 15-21, clinical insomnia, moderate severity;
> 21, clinical insomnia, severe severity).22 A total score > 14
represents moderate to severe insomnia. The scale was found
to have an adequate internal consistency (Cronbach α = 0.74).25
The Insomnia Severity Index-Chinese version also has a good
internal consistency, with a Cronbach α coefficient of 0.94.26 The
Cronbach α of the ISI before treatment in this study was 0.65.
Hospital Anxiety and Depression Scale (HADS)
The HADS24 is a 14-item self-rated scale designed to assess
clinically relevant anxiety and depression. It is divided into an
Anxiety subscale (HADS-A) and a Depression subscale (HADSD), both containing 7 intermingled items. The original English
version has been translated into and validated in many languages,
642
Light/Dark Exposure to Treat Insomnia in Nurses
including Chinese. In most studies an optimal balance between
sensitivity and specificity was achieved when caseness was defined by a score ≥ 8.28 The HADS, HADS-A, and HADS-D before treatment had good internal consistency in this study, with a
Cronbach α coefficient ranging from 0.72 to 0.82.
27
Table 1—Demographic data of the treatment and control
groups
Treatment
(n = 46)
30.2 ± 4.5
Control
(n = 46)
30.3 ± 4.7
Employment (years), M ± SD
5.1 ± 4.7
5.3 ± 4.8
Education, n (%)
14 years
16 years
14 (30.4)
32 (69.6)
16 (32.6)
30 (67.4)
Marital Status, n (%)
Single/divorced/separated
Married
39 (84.8)
7 (15.2)
36 (78.3)
10 (21.7)
Department, n (%)
Surgery
Internal Medicine
Psychiatry
21 (45.7)
20 (43.5)
5 (10.9)
21 (45.7)
21 (45.7)
4 (8.7)
Character of ward, n (%)
Acute ward
Intensive care unit
42 (91.3)
4 (8.7)
46 (100)
0 (0)
Variables
Age (years), M ± SD
Statistical Analysis
Statistical analyses were performed using the Statistical
Package for the Social Sciences (SPSS 17.0) for Windows 7.
The demographic data of the treatment group and the control
group were compared using Student t-test, the χ2 test, or Fisher
test. The independent-samples t-test was used to compare the
differences in the scores for insomnia, anxiety, and depression
between the 2 groups before and after treatment. We used the
paired-samples t test to assess the effectiveness of bright light
exposure (change from pre-treatment to post-treatment in each
group). All continuous variables were compared across both
groups using an analysis of covariance model (ANCOVA) to
assess the change from pre-treatment, with pre-treatment as
the covariate. We also used ANCOVA to compare the changes
in anxiety and depression severity in both groups, with the ISI
change from pre-treatment to post-treatment as the covariate.
The Pearson correlation was used to assess the correlations between the changes in the ISI, HADS-A, and HADS-D scores.
All statistical tests were 2-sided, and a significance level of 0.05
was used for all comparisons.
RESULTS
During the study period, all subjects completed the study and
did not use psychotropic medication other than sleep medications.
All subjects reported that they did not change their lifestyle, such
as mealtimes, exercise periods, sleep/wake schedule, and pattern
of alcohol/substance/medication use, including off-days, during
the study. Although none of the participants met the diagnosis of
substance dependence or abuse, 19 nurses (20.7%) use one cup of
caffeinated drink per day habitually. Ten subjects were in the treatment group; the remainder were in the control group. There was
no significant difference between the two groups. Most subjects
(93.6%) did not use sleep medications; 6 participants used sleep
medications (zolpidem or lorazepam) prescribed by physicians.
Table 1 shows the demographic data of both groups of participants. All demographic data were comparable between groups,
including previous work schedule and sleep medications use.
Three-shift rotation: Work schedule last month, n (%)
Day shift
16 (34.8)
Evening shift
17 (37.0)
Night shift
13 (28.3)
19 (41.3)
16 (34.8)
11 (23.9)
Three-shift rotation: Current work schedule, n (%)
Evening shift
32 (69.6)
Night Shift
14 (30.4)
30 (65.2)
16 (34.8)
Prescription sleep medications use, n (%)
Yes
1 (2.2)
No
45 (97.8)
5 (10.8)
41 (89.1)
All variables p > 0.05. M ± SD, mean ± standard deviation.
HADS-D, in the control group (paired-samples t test, p < 0.001).
The mean changes in the scores of the ISI, HADS, HADS-A,
and HADS-D from pre-treatment in the treatment group were
-12.2 ± 5.1, -6.6 ± 5.1, -3.8 ± 2.8, and -2.8 ± 3.4, respectively; while those in the control group were -0.2 ± 1.7, 1.5 ± 2.5,
0.2 ± 1.3, and 1.3 ± 1.8, respectively. For subjects in the glassesonly group, the pre-treatment depression score was categorized
as either a HADS-D score < 8 or ≥ 8. The mean changes in the
scores of depression across treatment in the HADS-D < 8 group
and ≥ 8 group were 1.9 ± 1.8 and 0.3 ± 1.0, respectively. The
depression score became worse only in the group with a HADSD < 8 (paired-samples t test, p < 0.001). Relative to the control
group, the scores of the ISI, HADS, HADS-A, and HADS-D
significantly improved across treatment in the treatment group
according to ANCOVA (all p < 0.001; Figure 1). After treatment,
in the treatment group 37 nurses (80.4%) met the criterion for no
insomnia (ISI < 8). Meanwhile, no nurse met the criterion for no
insomnia in the control group.
Differences in ISI, HADS, HADS-A, and HADS-D between
Treatment and Control Groups before and after Treatment
There were no significant differences in the ISI, HADS,
HADS-A, and HADS-D scores between the groups before
treatment. The treatment group had significantly lower scores
on the insomnia, anxiety, and depression scales than the control
group after treatment (Table 2).
Differences across Treatment in ISI, HADS, HADS-A,
and HADS-D between Treatment and Control Groups
Differences across Treatment in the ISI, HADS, HADS-A,
and HADS-D between Evening and Night Shift Nurses in
Treatment and Control Groups
After treatment, the treatment group exhibited significantly improved ISI, HADS, HADS-A, and HADS-D scores as compared
with pre-treatment (paired-samples t test, all p < 0.001). Depression became worse after treatment, as measured by the HADS and
The scales of the ISI, HADS, HADS-A, and HADS-D were
not significantly different between nurses undertaking the evening and night shifts in either group. We analyzed whether the
643
Table 2—Differences in the severity of insomnia, anxiety, and depression between the two groups pre- and post-treatment
Variables
ISI
Pre-treatment
Post-treatment
Treatment group
Control group
95% CI
Statisticsa
17.9 ± 2.5
5.7 ± 5.0
17.1 ± 2.3
16.9 ± 3.2
-0.22, 1.78
-12.91, -9.44
p = 0.123
p < 0.001
HADS
Pre-treatment
Post-treatment
16.2 ± 5.5
9.6 ± 3.9
15.1 ± 6.3
16.6 ± 5.9
-1.33, 3.60
-9.09, -4.95
p = 0.364
p < 0.001
HADS-A
Pre-treatment
Post-treatment
9.1 ± 3.3
6.1 ± 2.7
8.8 ± 3.7
9.0 ± 3.5
-3.43, 2.56
-4.21, -1.62
p = 0.133
p < 0.001
HADS-D
Pre-treatment
Post-treatment
6.3 ± 0.6
3.5 ± 1.9
6.3 ± 3.5
7.6 ± 3.2
-1.45, 1.49
-5.21, -3.01
p = 0.977
p < 0.001
ISI, Chinese version of the Insomnia Severity Index; HADS, Chinese version of the Hospital Anxiety Depression Scale; HADS-A, anxiety subscale of the
HADS; HADS-D, depression subscale of the HADS; CI, confidence interval. aIndependent-samples t test.
correlations between the changes in the severity of insomnia,
anxiety, and depression were high (Pearson correlation between
0.515 and 0.693, p < 0.001). The treatment group nevertheless
had significantly decreased scores on the HADS, HADS-A, and
HADS-D after treatment as compared with the control group,
even when the factor of ISI change was controlled (all p < 0.001).
Figure 1—Mean change in the ISI, HADS, HADS-A, and
HADS-D scores after treatment in the treatment and control
groups
Mean change from pre-treatment
Treatment group
2
1
0
-1
-2
-3
-4
-5
-6
-7
-8
-9
-10
-11
-12
-13
Control group
DISCUSSION
ISI
HADS
HADS-A
To the best of our knowledge, this was the first randomized
control study to investigate the effectiveness of bright light
exposure at night with attenuation of morning light in female
nurses undertaking rotating shift work suffering from clinical
insomnia during the evening/night shift. Our study found that
bright light therapy of 7,000-10,000 lux for at least 30 minutes at night for at least 10 days during 2 weeks significantly
improved the sleep problems of nurses working the evening
or night shift. Studies have found that bright light exposure
improves daytime sleep and nocturnal alertness21,29 in nurses
working the night shift. However, the subjects in these studies
did not necessarily have insomnia. The study of Yoon et al.21
also found that these improvements could be maximized by attenuation of morning light on the way home. Light attenuation
in the morning only was found to be ineffective in improving
insomnia in our study.
There is no consistent design of bright light therapy for evening/night shift workers, including schedule, intensity, and duration. Various light intensities, from 1,200 to 10,000 lux, with
durations of exposure ranging from 3 to 6 hours, have been used
successfully to realign circadian rhythms and improve performance and sleep during the night shift.20,21 Previous studies30,31
have recommended that either intermittent or continuous light
exposure begins early in the shift and terminates approximately
2 hours before the end of shift, with the wearing of sunglasses
outdoor in the morning was efficacious for phase delay. Importantly, a delayed circadian phase was found to be positively
correlated with improved sleep, even in the control group, in
subjects during the night shift.31,32 The design of our treatment
procedure is easy to put into practice in the real world. Higher
intensity and shorter bright light exposure once before work or
HADS-D
All p < 0.001 between groups. Note: p-values reflect the results of the
change from the pre-treatment analyses using ANCOVA. ISI, Chinese
version of the Insomnia Severity Index; HADS, Chinese version of the
Hospital Anxiety Depression Scale; HADS-A, anxiety subscale of the
HADS; HADS-D, depression subscale of the HADS.
changes in the scores for insomnia and mood might differ between evening and night shift workers in the treatment group.
The mean changes in the scores of the ISI, HADS, HADS-A, and
HADS-D from pre-treatment in the evening shift workers were
-11.3 ± 5.5, -6.4 ± 5.1, -3.8 ± 2.8, and -2.7 ± 3.2, respectively;
while those in the night shift workers were -14.1 ± 3.5, -7.1 ±
5.2, -4.0 ± 2.9, and -3.1 ± 3.8, respectively. There were no significant differences in the insomnia/mood scores across treatment
between the evening and night shift workers in either groups.
Changes in HADS, HADS-A, and HADS-D scores between
Groups after Controlling for Severity of Insomnia
We analyzed whether the changes in anxiety and depression
directed the change in insomnia severity. After treatment, the
644
Light/Dark Exposure to Treat Insomnia in Nurses
in a break during the first half of the evening/night shift and
a daytime darkness procedure were implemented in our study.
The study of Drake et al.7 found that insomnia or daytime
sleepiness is a risk factor for major depression, but it is a much
greater risk factor for rotating or night shift workers. A systematic review by Even et al.33 examined the efficacy of light
therapy in nonseasonal depression. They found that bright light
monotherapy is efficacious in treating seasonal depression, but
its efficacy in treating nonseasonal depression is inconsistent.
This is the first study to report that in female nurses working
rotating shifts, anxiety and depression scores were significantly
improved after bright light therapy, even when the change in insomnia severity was controlled. It must be noted that the mean
HADS-D score as a group was not above the cutoff point (≥ 8)
according to the criteria for depression either before or after
intervention in our study, but the scores were high. Besides, the
nurses were not diagnosed with depression, nor was the clinical severity of depression evaluated in this study. Some factors
may contribute to these findings, such as (1) according to the
general criteria of the International Classification of Sleep Disorders, 2nd ed. (ICSD-2)8 for insomnia, depressive symptoms
related to nighttime sleep difficulty are commonly reported by
insomniacs; (2) treating insomnia in patients with major depressive episode improves mood problems other than insomnia.34-36
It must be mentioned that light therapy is not the only choice
of treatment for insomnia or depression. Subjects undertaking
shift work can improve insomnia and mood problems through
pharmacotherapy, behavioral therapy, or other therapy.
There were some primary limitations in our study. First, this
study was not a double-blind study, and “placebo effects” should
be considered. The subjects in both groups might work in the
same unit, and the use of a sham light box (a light box of a much
lower intensity or red light) in the control group would be able
to be detected by the controls, who would discern the difference. Therefore, we considered that the “placebo effect” could
not have been solved by the use of a sham light box. Second,
as this study was performed in a real workplace, it was difficult to measure the differences in endogenous circadian rhythm
change and the real intensity and duration of light exposure in
groups. Aoki et al.37 reported melatonin could be suppressed
by light intensities of around 300 lux. However, compared to
ordinary room light (< 250 lux), Martin and Eastman38 found
that medium- and high-intensity light, approximately 1,230 lux
and 5,700 lux, respectively, for 3 hours, significantly increased
the percentage of subjects who adapted to the night shift, as
measured by temperature rhythm phase shifts. Larger phase
shifts were correlated with more sleep and less fatigue. Third,
we only used questionnaires to measure the severity of insomnia, anxiety, and depression in the nurses working shifts. We
were not able to confirm the diagnosis of insomnia, because
subjects might suffer from shift work disorder (SWD), primary
insomnia, insomnia due to mental disorders, or other insomniarelated disorders. According to data from the Detroit tri-county
population, 32.1% of night workers and 26.1% of rotating shift
workers suffer from insomnia or excessive sleepiness. Thus,
the “true prevalence” of SWD in the night- and rotating-worker
samples was 14.1% and 8.1%, respectively.7 In a study performed in Norway,39 it was found that 44.3% of nurses working on a three-shift rotation met the criteria of SWD by asking
shift work-related symptom questions. Fourth, although the
study was a randomized control study, we did not assess the
differences in rotation patterns, lifestyle during days off, and
off-work periods in the groups. Circadian misalignment might
be aggravated by these factors. Rotation patterns such as the
number of shifts separated by less than 11 hours and the number of nights worked were found to be positively associated
with SWD in the study of Norwegian nurses.39 Lin et al.18 also
reported that rotation shift nurses who had at least two days
off after their most recent night shift showed significantly improved sleep quality and mental health. Finally, 67.4% of the
subjects in the present study were working the evening shift,
and 32.6% were working the night shift. In this study, different
timings of bright light exposure were employed in the evening
and night shift groups. No significant differences were found in
the change of insomnia/mood scales across treatment in treatment and control groups. However, since statistical power is
reduced by dividing the subjects into smaller subgroups by shift
schedule, further studies would be indicated to assess effectiveness under the two different situations.
In conclusion, the design of this study is easy to put into
practice in the real world. Female nurses working the evening/
night shift with insomnia can improve their sleep problems by
higher intensity and shorter bright light exposure once before
work or in a break during the first half of work combined with
a daytime darkness procedure. In addition, anxiety and depression scores can be improved after intervention.
REFERENCES
1. Ohayon MM. Epidemiology of insomnia: what we know and what we still need to
learn. Sleep Med Rev 2002;6:97-111.
2. Hsu SC, Liu CY, Chen NH. Psychiatric diagnoses of patients visiting a sleep
disorder clinic due to dyssomnias. Psychol Rep 2006;98:789-97.
3. Ohayon MM, Roth T. Place of chronic insomnia in the course of depressive and
anxiety disorders. J Psychiatr Res 2003;37:9-15.
4. Buysse DJ, Tu XM, Cherry CR, et al. Pretreatment REM sleep and subjective
sleep quality distinguish depressed psychotherapy remitters and nonremitters.
Biol Psychiatry 1999;45:205-13.
5. Moos RH, Cronkite RC. Symptom-based predictors of a 10-year chronic course
of treated depression. J Nerv Ment Dis 1999;187:360-8.
6. Hsu SC, Wang SJ, Liu CY, Juang YY, Yang CH, Hung CI. The impact of anxiety and migraine on quality of sleep in patients with major depressive disorder.
Compr Psychiatry 2009;50:151-7.
7. Drake CL, Roehrs T, Richardson G, Walsh JK, Roth T. Shift work sleep disorder:
Prevalence and consequences beyond that of symptomatic day workers. Sleep
2004;27:1453-62.
8. American Academy of Sleep Medicine. International classification of sleep disorders, 2nd ed. Diagnostic and coding manual. Westchester, IL: American Academy of Sleep Medicine, 2005.
9. Beers TM. Flexible schedules and shift work: replacing the 9-to-5 workday.
Monthly Lab Rev 2000;123:33-40.
10. Hsu SY, Su SB. Health problems of permanent night-shift and rotating shift workers. Chin J Occup Med 2003;10:71-80. [in Traditional Chinese]
11. Wright Jr KP, Hull JT, Hughes RJ, Ronda JM, Czeisler CA. Sleep and wakefulness out of phase with internal biological time impairs learning in humans. J
Cogn Neurosci 2006;18:508-21.
12. Czeisler CA, Weitzman E, Moore-Ede MC, Zimmerman JC, Knauer RS. Human sleep: its duration and organization depend on its circadian phase. Science
1980;210:1264-7.
13. Folkard S, Lombardi DA, Tucker PT. Shiftwork: safety, sleepiness and sleep. Ind
Health 2005;43:20-3.
14. Smith L, Folkard S, Poole CJ. Increased injuries on night shift. Lancet
1994;344:1137-9.
15. Bara AC, Arber S. Working shifts and mental health—findings from the British Household Panel Survey (1995-2005). Scand J Work Environ Health
2009;35:361-7.
645
16. Monk TH, Buysse DJ, Reynolds CF 3rd, et al. Circadian rhythms in human performance and mood under constant conditions. J Sleep Res 1997;6:9-18.
17. Suzuki K, Ohida T, Kaneita Y, et al. Mental health status, shift work, and occupational accidents among hospital nurses in Japan. J Occup Health
2004;46:448-54.
18. Lin PC, Chen CH, Pan SM, et al. Atypical work schedules are associated with
poor sleep quality and mental health in Taiwan female nurse. Int Arch Occup
Environ Health 2012;85:877-84.
19. Khalsa SB, Jewett ME, Cajochen C, Czeisler CA. A phase response curve to
single bright light pulses in human subjects. J Physiol 2003;549:945-52.
20. Burgess HJ, Sharkey KM, Eastman CI. Bright light, dark and melatonin can promote circadian adaptation in night shift workers. Sleep Med Rev 2002;6:407-20.
21. Yoon IY, Jeong DU, Kwon KB, Kang SB, Song BG. Bright light exposure at night
and light attenuation in the morning improve adaptation of night shift workers.
Sleep 2002;25:351-56.
22. Morin CM, Barlow DH, Dement WC. Insomnia: Psychological assessment and
management. New York: Guilford Press, 1993.
23. Frances A, Princus HA, First MB, eds. Diagnostic and statistical manual of mental disorders, 4th ed. American Psychiatric Press, Washington, DC, 1994.
24. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67:361-70.
25. Bastien CH, Vallières A, Morin CM. Validation of the Insomnia Severity Index as
an outcome measure for insomnia research. Sleep Med 2001;2:297-307.
26. Yang CM, Hsu SC, Lin SC, Chou YY, Chen YM. Reliability and validity of the
Chinese version of insomnia severity index. Arch Clin Psychol 2009;4:95-104.
[in Traditional Chinese]
27. Hung CI, Wang SJ, Liu CY. Validation of the Depression and Somatic Symptoms
Scale by Comparison with the Short Form 36 Scale among psychiatric outpatients with major depressive disorder. Depress Anxiety 2009;26:583-91.
28. Bjelland I, Dahl AA, Haug TT, Neckelmann D. The validity of the Hospital Anxiety and Depression Scale. An updated literature review. J Psychosom Res
2002;52:69-77.
29. Kakooei H, Ardakani ZZ, Ayattollahi MT, Karimian M, Saraji GN, Owji AA. The
effect of bright light on physiological circadian rhythms and subjective alertness
of shift work nurses in Iran. Int J Occup Saf Ergon 2010;16:477-85.
30. Crowley SJ, Lee C, Tseng CY, Fogg LF, Eastman CI. Combinations of bright
light, scheduled dark, sunglasses, and melatonin to facilitate circadian entrainment to night shift work. J Biol Rhythms 2003;18:513-23.
31. Smith MR, Cullnan EE, Eastman CI. Shaping the light/dark pattern for circadian
adaptation to night shift work. Physiol Behav 2008;95:449-56.
32. Smith MR, Eastman CI. Night shift performance is improved by a compromise
circadian phase position: Study 3. Circadian phase after 7 night shifts with an
intervening weekend off. Sleep 2008;31:1639-45.
33. Even C, Schroder CM, Friedman S, Rouillon F. Efficacy of light therapy in nonseasonal depression: a systematic review. J Affect Disord 2008;108:11-23.
34. Fava M, McCall WV, Krystal A, et al. Eszopiclone co-administered with fluoxetine in patients with insomnia coexisting with major depressive disorder. Biol
Psychiatry 2006;59:1052-60.
35. Manber R, Edinger JD, Gress JL, San Pedro-Salcedo MG, Kuo TF, Kalista
T. Cognitive behavioral therapy for insomnia enhances depression outcome
in patients with comorbid major depressive disorder and insomnia. Sleep
2008;31:489-95.
36. Krystal A, Fava M, Rubens R, et al. Evaluation of eszopiclone discontinuation
after cotherapy with fluoxetine for insomnia with coexisting depression. J Clin
Sleep Med 2007;3:48-55.
37. Aoki H, YamadaN, Ozeki Y, Yamane H, Kato N. Minimum light intensity required
to suppress nocturnal melatonin concentration in human saliva. Neurosci Lett
1999;252:91-4.
38. Martin SK, Eastman CI, Medium-intensity light produces circadian rhythm adaptation to simulated night-shift work. Sleep 1998;21:154-65.
39. Flo E, Pallesen S, Magerøy N, et al. Shift work disorder in nurses – assessment,
prevalence and related health problems. PLoS One 2012;7:e33981.
acknowledgments
This study was supported by grants from the Chang Gung Memorial Hospital
(CMRPG-380521).
submission & correspondence Information
Submitted for publication April, 2012
Submitted in final revised form December, 2012
Accepted for publication January, 2013
Address correspondence to: Dr. Shih-Chieh Hsu, Department of Psychiatry, Chang
Gung Memorial Hospital, 5# Fu-Shin Road, Guei-Shan, Tao-Yuan 333, Taiwan
(R.O.C.); Tel: +886-3-3281200 ext. 2479; Fax: +886-3-3280267; E-mail: hsu3160@
cgmh.org.tw
disclosure statement
This was not an industry supported study. The authors have indicated no financial
conflicts of interest.
646
C ommentar y
Bright Light Improves Sleep and Psychological Health in Shift
Working Nurses
Commentary on Huang et al. The effectiveness of light/dark exposure to treat insomnia in female
nurses undertaking shift work during the evening/night shift. J Clin Sleep Med 2013;9:641-648.
Bjørn Bjorvatn, M.D., Ph.D.; Siri Waage, Ph.D.
Norwegian Competence Center for Sleep Disorders, Haukeland University Hospital, Norway;
Department of Global Public Health and Primary Care, University of Bergen, Norway
I
n this issue of Journal of Clinical Sleep Medicine, Huang and
colleagues investigated the effectiveness of bright light exposure in shift working nurses on an evening or night shift schedule.1 Nurses working in a hospital were divided in two groups,
where the treatment group received bright light of 7,000-10,000
lux for 30 minutes at about 8 pm (evening shift) or at about 11:30
pm (night shift), whereas the control group did not receive any
light treatment. Both groups were instructed to wear dark sunglasses to avoid outdoor light after work and before sleep. The
study period was 10-14 days. The outcome measures were wellvalidated self-report instruments: Insomnia Severity Index (ISI)
and Hospital Anxiety and Depression scale (HADS). The results
showed that the treatment group impressively improved sleep.
Furthermore, the scores on the anxiety and depression scale decreased significantly. No effects were seen in the control group.
Many shift workers suffer from poor sleep and sleepiness.
The most afflicted workers may be diagnosed with shift work
disorder (SWD), a circadian rhythm sleep disorder, characterized by a complaint of insomnia or excessive sleepiness that
is temporally associated with a recurring work schedule that
overlaps the usual time for sleep over the course of at least one
month.2 In nurses, the prevalence of SWD varies according to
work schedule, but may be as high as 44% among nurses involved in night shifts.3 On the other hand, it may be surprising
that the majority of shift workers, also those involved in night
work, do not report insomnia or excessive sleepiness. In addition to poor sleep and sleepiness, some studies show that night
workers are at increased risk of anxiety/depression.4
Sleep is regulated by an interaction between homeostatic,
circadian, and behavioral factors.5 Shift work, especially when
night work is part of the schedule, results in a misalignment between the endogenous circadian timing system and the external
24-h environment. The treatment options for circadian rhythm
sleep disorders comprise bright light treatment and exogenous
melatonin administration. Both these chronobiotics need to be
timed according to specific phase-response curves to have the
wanted effect. For instance, bright light before the minimum
(nadir) of the core body temperature will phase delay the circadian rhythm, whereas bright light after nadir produces a phase
advance.5 This means that incorrectly timed bright light will
likely worsen sleep and sleepiness complaints. Nadir is usually
located about 1 to 2 hours before the habitual wake-up time.
In the study by Huang and colleagues,1 the timing of treatment
was appropriate, although even better results may have been expected if bright light had been timed according to the nurses’
individual circadian rhythms.
The findings presented in the study by Huang and coworkers
were surprisingly strong, and more impressive than seen following bright light treatment in, for instance, night workers in the
petroleum industry.6 One reason may be that all participants in
the study by Huang et al. had moderate to severe insomnia before treatment, whereas other studies may include participants
with fewer complaints. However, a reduction in ISI score from
17.9 to 5.7 following treatment, compared to no effect in the
control group, is amazing. In fact, 80% of the nurses did not
have insomnia following treatment, whereas all nurses in the
control group still met the criteria for insomnia. Similarly, the
HADS scores were clearly reduced in the treatment group. It
was somewhat surprising that light attenuation following work
and before sleep (control group) had no effect. Other studies
have shown that such an approach could help night workers with
circadian adaptation.7
One of the major take home messages from the Huang et
al. study is that such a treatment approach is relatively easy to
implement into clinical practice. Bright light exposure of 30
minutes duration seems feasible in the workplace. Furthermore,
there were few inclusion/exclusion criteria, suggesting that the
findings may generalize to large groups of nurses working evening and/or night shifts. The improvements in sleep and psychological health may also have large consequences in terms of
higher productivity, reduced risk of accidents/errors at work, as
well as reduced sickness absence.
Citation
Bjorvatn B; Waage S. Bright light improves sleep and psychological health in shift
working nurses. J Clin Sleep Med 2013;9(7):647-648.
references
1. Huang LB, Tsai MC, Chen CY, Hsu SC. The effectiveness of light/dark exposure
to treat insomnia in female nurses undertaking shift work during the evening/
night shift. J Clin Sleep Med 2013;641-6.
647
B Bjorvatn and S Waage
2. American Academy of Sleep Medicine. International classification of sleep disorders, 2nd ed.: Diagnostic and coding manual. Westchester, Illinois: American
Academy of Sleep Disorders, 2005.
3. Flo E, Pallesen S, Magerøy N, et al. Shift Work Disorder in nurses - Assessment,
prevalence and related health outcome. PLoS One 2012;7:e33981.
4. Bara AC, Arber S. Working shifts and mental health–findings from the British Household Panel Survey (1995-2005). Scand J Work Environ Health
2009;35:361-7.
5. Bjorvatn B, Pallesen S. A practical approach to circadian rhythm sleep disorders.
Sleep Med Rev 2009;13:47-60.
6. Bjorvatn B, Stangenes K, Øyane N, et al. Randomized, placebo-controlled field
study of the effects of bright light and melatonin in adaptation to night work.
Scand J Work Environ Health 2007;33:204-14.
7. Crowley SJ, Lee C, Tseng CY, et al. Combination of bright light, scheduled dark,
sunglasses, and melatonin to facilitate circadian entrainment to night shift work.
J Biol Rhythms 2003;18:513-23.
Submitted for publication May, 2013
Accepted for publication May, 2013
Address correspondence to: Bjørn Bjorvatn, M.D., Ph.D., Department of Global
Public Health and Primary Care, University of Bergen, Kalfarveien 31, Bergen,
N-5018 Norway; Tel: 47 55 58 60 88; E-mail: bjorn.bjorvatn@isf.uib.no
Dr. Bjorvatn has participated in speaking engagements for GlaxoSmithKline, Nycomed, ResMed, Confex, and Medi3. Dr. Waage has indicated no financial conflicts
of interest.
648
Serum Brain-Derived Neurotrophic Factor Levels Are
Associated with Dyssomnia in Females, but not Males,
among Japanese Workers
Reiko Nishichi, M.Sc.1; Yu Nufuji, Ph.D.2; Masakazu Washio, M.D., Ph.D.3; Shuzo Kumagai, M.D., Ph.D.4
Department of Community Health Nursing, Shimane University Faculty of Medicine, Izumo, Shimane, 693-8501, Japan; 2Tokyo
Metropolitan Institute of Gerontology, Itabashi, Tokyo, 173-015, Japan; 3Faculty of Nursing, St. Mary’s College, Kurume, Fukuoka,
830-8558, Japan; 4Institute of Health Science, Kyushu University, Kasuga, Fukuoka 816-8580, Japan
1
Study Objectives: Brain-derived neurotrophic factor
(BDNF) is a member of the neurotrophin family of growth
factors that promote the growth and survival of neurons.
Recent evidence suggests that BDNF is a sleep regulatory
substance that contributes to sleep behavior. However, no
studies have examined the association between the serum
BDNF levels and dyssomnia. The present study was conducted to clarify the association between the serum BDNF
levels and dyssomnia.
Methods: A total of 344 workers (age: 40.1 ± 10.5 years, male:
204, female: 140) were included in the study. The serum BDNF
levels were categorized into tertiles according to sex.
Results: The prevalence of dyssomnia was 35.1% in males
and 30.0% in females. In the females, the BDNF levels were
found to be negatively associated with dyssomnia after adjust-
ing for age, body mass index, hypertension, dyslipidemia, hyperglycemia, depression, smoking, alcohol intake, and regular
exercise. Compared with the females in the high BDNF group,
the multivariate odds ratio (95% CI) of dyssomnia was 2.08
(0.62-6.98) in females in the moderate BDNF group and 8.41
(2.05-27.14) in females in the low BDNF group. No such relationships were found in the males.
Conclusions: The serum BDNF levels are associated with
dyssomnia in Japanese female, but not male, workers.
Keywords: Serum brain-derived neurotrophic factor, dyssomnia, sex, Japanese worker
Citation: Nishichi R; Nufuji Y; Washio M; Shuzo Kumagai S.
Serum brain-derived neurotrophic factor levels are associated
with dyssomnia in females, but not males, among Japanese
workers. J Clin Sleep Med 2013;9(7):649-654.
D
yssomnia is one of the most common health problems in
the Japanese population. Recent surveys by the Japanese
Ministry of Health Labor and Welfare have demonstrated that
21.1% of Japanese adults suffer from dyssomnia.1 Many studies have suggested that dyssomnia is not only linked to mental
disorders, including depression,2 but also to endocrine disorders
(e.g., obesity, diabetes mellitus) and cardiovascular disorders
(e.g., hypertension, heart disease).3-6
Brain-derived neurotrophic factor (BDNF) is a member of
the neurotrophin family of growth factors. In addition to its
neurotrophic and synaptotrophic actions, including the promotion of growth and survival in neurons,7,8 BDNF plays a role in
learning and memory,9 the regulation of food intake,10 glucose
and lipid metabolism and energy homeostasis.11,12 BDNF is
present in the nervous system and peripheral tissues and is also
found in blood.13-15 Accumulating evidence shows the serum
BDNF levels to be associated with psychiatric and metabolic
disorders, including depression,16,17Alzheimer disease,18 and
diabetes mellitus.19-21 However, no studies have examined the
association between the serum BDNF levels and dyssomnia.
Recent evidence interestingly suggests that BDNF is a sleep
regulatory substance.22-24 Faraguna et al.25 showed the degree of
BDNF expression during wakefulness to be causally linked to
the extent of slow wave activity in the subsequent rest period.
Moreover, Martinowich et al.26 demonstrated that a genetic
Brief Summary
Current Knowledge/Study Rationale: BDNF is suggested to contribute
to sleep behavior. However, there is no study on the association between
serum BDNF and dyssomnia.
Study Impact: The serum BDNF levels were inversely associated with
dyssomnia in females, but not in males. However, further studies are
needed to answer whether or not the sex differences in BDNF are related
to sex differences in dyssomnia.
manipulation that leads to disruption of the activity-dependent
BDNF expression results in impairments in sleep regulation.
Based on this evidence showing a biological links between
BDNF and sleep behavior, we hypothesized that serum BDNF
levels may be associated with dyssomnia. In the present study,
we examined this association in Japanese workers.
METHODS
In Japan, the Industrial Safety and Health Law requires all
employers to provide annual health check-ups for their employees. The annual health check-up consists of an interview
regarding lifestyle; measurement of weight, height, and blood
pressure; physical examination; electrocardiogram examinations; chest x-ray; urinalysis; and blood tests. Blood samples
649
R Nishichi, Y Nufuji, M Washio et al
from the study subjects were obtained from 08:00 to 10:00 after overnight fasting. In addition to performing these routine
health check-up examinations, the serum BDNF levels were
measured, and sleep quality and depressive symptoms were assessed with interviews by trained nurses.
mm Hg and/or current treatment with antihypertensive medications. Dyslipidemia was defined as LDL-cholesterol ≥ 140 mg/
dL, triglyceride ≥ 150 mg/dL, HDL-cholesterol < 40 mg/dL and/
or current treatment with antihyperlipidemic drugs. Hyperglycemia was defined as fasting plasma glucose concentrations ≥
110 mg/dL and/or the use of antidiabetic medications.29 Depressive symptoms were evaluated using the Japanese version of the
Center for Epidemiological Studies Depression Scale (CES-D).
Depression was defined as a CES-D score ≥ 16 points.30
Subjects
The subjects of this study were employees of the Creative Research Community (CRC) Company (Fukuoka, Japan), which
provides services such as health check-up support, genetic testing, and clinical testing. A total of 400 workers, 20 years of age
or older underwent an annual health check-up at their company
in 2009. Among these workers, 30 did not agree to participate
and 26 who did not complete the questionnaires or biochemical
tests were excluded from the study. Ultimately, a total of 344
participants (204 males and 140 females) were included in this
study. Two hundred eighty-two study subjects (82%) were day
workers. All participants received oral and written information
about the experimental procedures before giving their written
informed consent. This study was approved by the Ethics Committee of St. Mary’s College and monitored by the institutional
review committee.
Statistical Analyses
The serum BDNF levels were categorized into tertiles according to sex (males: < 10.91, 10.92 to 13.81, > 13.82 ng/mL;
females: < 9.32, 9.33 to 12.12, > 12.13 ng/mL). The crude mean
values and the frequencies of the variables were compared between the groups using the χ2 test and one-way analysis of variance as appropriate. Dunnett test was employed for all post hoc
tests. The odds ratios (OR) and 95% confidence intervals (95%
CI) of dyssomnia for each BDNF tertile group were calculated
by taking the highest tertile as the referent using the logistic regression models. A p-value less than 0.05 was considered to be
statistically significant. All statistical analyses were performed
using the SPSS software program (Statistical Package for Social Sciences, version 18.0, SPSS Inc., Chicago, IL, USA).
Serum BDNF Levels
After the blood was centrifuged 2000 × g for 10 min at 4°C,
the serum was stored at -80°C until the analyses were performed.
The serum BDNF concentrations were measured using an enzyme-linked immunoassay (ELISA) kit (Promega, Madison, WI)
following the manufacturer’s instructions. Briefly, 96 well plates
were coated with anti-BDNF monoclonal antibodies and incubated at 4°C for 16 h. The plates were then incubated in a blocking
buffer for 1 h. All of the incubation stages were conducted at room
temperature. The serum samples were diluted to 1:200, and the
plasma samples were diluted to 1:19 in Block & Sample 1 × Buffer. After adding the samples and the BDNF standard, the plates
were incubated with shaking for 2 h, then washed in washing
buffer. The plates were then incubated with anti-human BDNF
polyclonal antibodies for 2 h. After being washed, the plates were
incubated with anti-IgY HRP conjugate with shaking for 1 h and
washed. Next, TMB One solution was added for 10 min, and the
reaction was stopped with 1 M HCl. The absorbance at 450 nm
was measured within 30 min after stopping the reaction.
RESULTS
Characteristics of Participants
The prevalence of dyssomnia was 35.1% in the males and
30.0% in the females. The serum BDNF levels were significantly higher in the males (12.72 ± 4.08 ng/mL) than in the
females (11.13 ± 3.28 ng/mL, p < 0.001).
Table 1 presents the characteristics of the male participants
by tertile of the serum BDNF levels. There were no significant
differences in PSQI scores or prevalence of dyssomnia among
the 3 groups of males. The frequency of regular exercise was
significantly higher in the low BDNF group than in the high
BDNF group. There were no significant differences in any of
the other parameters among the 3 groups of males. Table 2
presents the characteristics of the female participants by tertile
of the serum BDNF levels. The mean PSQI scores in the low
and moderate BDNF groups were significantly higher than that
in the high BDNF group among females (p < 0.01, p = 0.02,
respectively). Additionally, there were significant differences in
the prevalence of dyssomnia among the 3 groups (p < 0.001).
The prevalence of dyssomnia in the low BDNF group was significantly higher than that in the high BDNF group (p < 0.01).
There were no significant differences in any of the other parameters among the 3 groups of females.
Dyssomnia
Sleep quality was assessed according to the Pittsburgh Sleep
Quality Index (PSQI). The PSQI is used worldwide as a tool
for the assessment of sleep quality. The scores were obtained
according to the PSQI-scoring method (0-1-2-3-4). The cutoff
for the total score of the PSQI is 5.5 points, and scores above
the cutoff are considered to indicate dyssomnia.27
Other Variables
Association between Serum BDNF Levels and
Dyssomnia by Sex
BMI was calculated as the weight in kilograms divided by
the height in meters squared. Obesity was defined as BMI ≥ 25
kg/m2.28 Antihypertensive medication use, antihyperlipidemic
drug use, oral hypoglycemic intake or insulin administration,
and current lifestyle factors, including smoking, alcohol intake,
and regular exercise were determined by interviews with trained
nurses. Hypertension was defined as blood pressure ≥ 140/90
Table 3 shows the association between the serum BDNF
levels and dyssomnia. Compared with the females in the high
BDNF group, the age-adjusted OR (95% CI) of dyssomnia was
2.04 (0.68-6.09) in females in the moderate BDNF group and
8.18 (2.89-23.13) in females in the low BDNF group. These
associations remained statistically significant even after ad650
BDNF and Dyssomnia
Table 1—Characteristics of participants by tertile of serum BDNF levels in men (n = 204)
Serum BDNF level ng/mL
Number of subjects
Age (years)
BMI (kg/m2)
Obesity
SBP (mm Hg)
DBP (mm Hg)
Hypertension
TC (mg/dL)
HDL-C (mg/dL)
LDL-C (mg/dL)
TG (mg/dL)
Dyslipidemia
FBG (mg/dL)
HbA1C (%)
Hyperglycemia
PSQIG score
Dyssomnia
CESD score
Depression
Smoking
Alcohol drinking
Regular exercise
Service form
Day worker
Two shift worker (the day and night)
High (≥ 13.82)
69
43.1 ± 10.0
24.1 ± 4.0
26 (37.7%)
135.2 ± 18.8
84.0 ± 16.6
29 (42.0%)
218.7 ± 33.4
59.2 ± 14.2
126.5 ± 30.2
147.2 ± 110.6
33 (47.8%)
102.9 ± 26.3
5.0 ± 0.8
11 (15.9%)
5.4 ± 2.4
29 (43.3%)
12.0 ± 6.5
16 (23.5%)
27 (39.7%)
41 (60.2%)
25 (36.8%)
52 (77.6%)
15 (22.4%)
Middle (10.92-13.81)
69
41.8 ± 10.9
23.0 ± 2.7
16 (23.2%)
133.6 ± 16.9
82.8 ± 12.3
30 (43.5%)
216.7 ± 31.9
61.7 ± 15.3
126.5 ± 28.9
123.4 ± 67.6
34 (49.3%)
98.1 ± 15.5
4.8 ± 0.5
6 (8.7%)
4.8 ± 2.3
19 (27.5%)
10.1 ± 6.3
11 (15.9%)
30 (43.5%)
43 (62.3%)
22 (32.4%)
55 (80.9%)
13 (19.1%)
Low (≤ 10.91)
69
42.6 ± 11.6
24.0 ± 3.2
22 (33.2%)
136.4 ± 16.4
81.9 ± 12.4
30 (45.5%)
207.2 ± 33.4
60.4 ± 11.4
120.7 ± 27.4
11.2 ± 62.7
26 (39.4%)
101.4 ± 19.6
4.9 ± 0.7
16 (24.2%)
5.0 ± 2.4
23 (34.8%)
11.1 ± 7.1
15 (22.7%)
28 (42.4%)
50 (75.7%)
37 (55.4%)
55 (84.6%)
10 (15.4%)
p value
0.78
0.11
0.17
0.65
0.69
0.92
0.10
0.56
0.41
0.05
0.46
0.40
0.32
0.05
0.32
0.73
0.25
0.49
0.90
0.12
0.02
0.59
Data presented are number (row percentages) or mean value ± standard deviation. BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood
pressure; TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein- cholesterol; LDL-C, low-density lipoprotein-cholesterol; FBG, Fasting blood
glucose; BDNF, brain-derived neurotrophic factor.
justing for age, BMI, dyslipidemia, diabetes mellitus, depression, regular exercise, and so on (moderate: OR 1.73, 95%
CI 0.51-5.90; low: OR 8.77, 95% CI 2.71-28.38). In contrast,
compared with the males in the high BDNF group, the males
in the low BDNF group showed a decreased age-adjusted OR
for dyssomnia (OR 0.47, 95% CI 0.23-0.97). However, this association disappeared after adjusting for confounding factors
(OR 0.58, 95% CI 0.23-1.28).
mellitus,19-21 and regular exercise,33,34 we adjusted the model for
these potential confounding factors. The association between
the serum BDNF levels and dyssomnia remained statistically
significant even after adjusting for these confounders. Among
females, the multivariable-adjusted odds ratio of dyssomnia in
the low BDNF group was eight times higher than that in the
high BDNF group. However, these associations were not observed in the male subjects. To our knowledge, this is the first
study to demonstrate an association between the serum BDNF
levels and dyssomnia.
There are many kinds of dyssomnia, and it is an important
issue to determine what types of dyssomnia correlate with the
serum BDNF levels. Low level of serum BDNF is considered
to associate with intrinsic circadian rhythm disorder, since the
majority of study subjects were day workers. Therefore, the association between serum BDNF levels and extrinsic circadian
rhythm disorder should be investigated in the future. The results
of this study showed that serum BDNF levels were negatively
associated with sleep duration, sleep disturbance, and daytime
dysfunction in the female, although the degrees of these associations seem to be weak. Thus, further large-scale studies are
recommended to confirm how serum the BDNF level correlates
with the occurrence of dyssomnia.
An association between the serum BDNF levels and dyssomnia is biologically plausible. Since BDNF can cross the blood-
Association between Serum BDNF Levels and Patterns
of Dyssomnia
Table 4 shows the association between serum BDNF levels
and the scores of 7 components of PSQIG. Serum BDNF levels in females were significantly inversely correlated with the
score of sleep duration (r = -0.191, p < 0.05), sleep disturbance
(r = -0.179, p < 0.05), daytime dysfunction (r = -0.270, p < 0.01),
and global (r = -0.295, p < 0.001). No such correlations were
found in males.
DISCUSSION
We found the serum BDNF levels to be negatively associated
with dyssomnia in females. Because the serum BDNF levels
have been reported to change according to age,31 body weight,
BMI,32 depression,16,17 metabolic disorders, including diabetes
651
Table 2—Characteristics of participants by tertile of serum BDNF levels in women (n = 140)
Serum BDNF level ng/mL
High (≥ 12.13)
47
37.1 ± 8.8
20.6 ± 2.8
2 (4.3%)
122.3 ± 15.5
75.3 ± 11.1
7 (14.9%)
218.7 ± 33.8
75.4 ± 15.5
114.3 ± 29.2
77.1 ± 50.0
10 (21.3%)
93.1 ± 15.1
4.7 ± 0.3
3 (6.4%)
4.3 ± 2.2
6 (12.8%)
9.7 ± 7.7
6 (12.8%)
8 (17.0%)
21 (44.7%)
16 (34.0%)
Number of subjects
Age (years)
BMI (kg/m2)
Obesity
SBP (mm Hg)
DBP (mm Hg)
Hypertension
TC (mg/dL)
HDL-C (mg/dL)
LDL-C (mg/dL)
TG (mg/dL)
Dyslipidemia
FBG (mg/dL)
HbA1C (%)
Hyperglycemia
PSQIG score
Dyssomnia
CESD score
Depression
Smoking
Alcohol drinking
Regular exercise
Service form
Day worker
Two shift worker (the day and night)
Middle (9.33-12.12)
47
36.0 ± 8.5
20.9 ± 3.8
4 (8.5%)
118.1 ± 11.4
74.8 ± 8.6
4 (8.5%)
206.2 ± 39.5
77.0 ± 14.8
111.9 ± 26.8
64.0 ± 27.0
6 (13.0%)
91.4 ± 14.4
4.7 ± 0.6
4 (8.5%)
3.9 ± 2.2
11 (23.4%)
9.6 ± 7.2
10 (21.3%)
6 (13.0%)
22 (46.8%)
15 (32.6%)
39 (83.0%)
8 (17.0%)
42 (91.3%)
4 (8.7%)
Low (≤ 9.32)
46
36.6 ± 9.6
21.2 ± 2.5
4 (8.7%)
118.1 ± 12.8
74.0 ± 9.1
6 (13.0%)
203.4 ± 26.0
75.1 ± 13.6
110.3 ± 22.6
75.3 ± 44.9
11 (23.4%)
90.2 ± 7.2
4.6 ± 0.2
1 (2.2%)
5.6 ± 2.4
25 (54.3%)
11.5 ± 8.2
13 (28.3%)
4 (8.7%)
28 (60.9%)
16 (36.4%)
39 (84.8%)
7 (15.2%)
p value
0.82
0.69
0.64
0.21
0.74
0.62
0.49
0.37
0.76
0.26
0.41
0.54
0.66
0.41
< 0.01
< 0.001
0.41
0.18
0.49
0.24
0.93
0.47
Data presented are number (row percentages) or mean value ± standard deviation. BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood
pressure; TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein- cholesterol; LDL-C, low-density lipoprotein-cholesterol; FBG, Fasting blood
glucose; BDNF, brain-derived neurotrophic factor.
Table 3—Distribution of Japanese workers with and without dyssomnia according to serum BDNF levels, with corresponding OR
and 95%CI
Serum BDNF level
Male
High
Middle
Low
Female
High
Middle
Low
Number of
participants
Dyssomnia
case
Age- and Sex-adjusted
OR (95%CI)
p value
69
69
66
29
19
23
1
0.67 (0.33-1.36)
0.47 (0.23-0.97)
47
47
46
6
11
25
1
2.04 (0.68-6.09)
8.18 (2.89-23.13)
Multivariable-adjusted
OR (95%CI)
p value
0.32
0.05
1
0.51 (0.23-1.12)
0.58 (0.23-1.28)
0.09
0.18
0.02
< 0.001
1
1.73 (0.51-5.90)
8.77 (2.71-28.38)
0.38
< 0.001
Multivariable-adjusted OR: The odds ratios (OR) and 95% confidence intervals (95% CI) of dyssomnia for each BDNF tertile group were calculated by taking
the highest tertile as the referent using the logistic regression models, adjusted for age, BMI, dyslipidemia, hyperglycemia, depression, smoking, alcohol
drinking, and regular exercise, service form.
brain barrier in both directions35 and brain tissue is the main
contributor to circulating BDNF,36 low serum BDNF levels may
reflect decreased BDNF levels in the brain. An experimental
animal study suggested that BDNF in the brain contributes to
the regulation of sleep behavior and promotes NREM sleep.22
Hence, decreased levels of brain BDNF may be related to poor
control of sleep behavior.
On the other hand, decreased serum BDNF levels may be
caused by dyssomnia. Recent studies in humans suggest that
acute or chronic sleep deprivation affects the hypothalamicpituitary-adrenal (HPA) system and changes the secretion of
cortisol.37,38 Vgontzas et al.38 demonstrated that 24-h mean cortisol secretion in chronic insomniacs is higher than that in normal controls.24 Intriguingly, glucocorticoids have been reported
652
BDNF and Dyssomnia
REFERENCES
Table 4—The association between serum BDNF level and
PSQIG subscores (n = 344)
Sleep quality
Sleep latency
Sleep duration
Sleep efficiency
Sleep disturbance
Hypnotic medication
Daytime dysfunction
Global
All
n = 344
0.038
-0.043
-0.019
0.000
-0.044
-0.031
0.001
-0.044
Male
n = 204
0.111
-0.014
0.072
0.005
0.063
-0.030
0.125
0.079
1. Liu X, Uchiyama M, Kim K, et al. Sleep loss and daytime sleepiness in the general adult population of Japan. Psychiatry Res 2000;93:1-11.
2. Stewart R, Besset A, Bebbington P, et al. Insomnia comorbidity and impact and
hypnotic use by age group in a national survey population aged 16 to 74 years.
Sleep 2006;29:1391-97.
3. Rosmond R, Lapidus L, Mårin P, Björntorp P. Mental distress, obesity and body
fat distribution in middle-aged men. Obes Res 1996;4:245-52.
4. Portaluppi F, Cortelli P, Avoni P, et al. Diurnal blood pressure variation and hormonal correlates in fatal familial insomnia. Hypertension 1994;23:569-76.
5. Wallander MA, Johansson S, Ruigómez A, García Rodríguez LA, Jones R.
Morbidity associated with sleep disorders in primary care: a longitudinal cohort
study. Prim Care Companion J Clin Psychiatry 2007;9:338-345.
6. Resnick HE, Redline S, Shahar E, et al. Diabetes and sleep disturbances: findings from the Sleep Heart Health Study. Diabetes Care 2003;26:702-09.
7. Aloe L, Iannitelli A. Neurotrophic factors and brain damage in hypoxic-ischemic encephalopathy: a role of nerve growth factor? Ann Ist Super Sanita
2001;37:573-80.
8. Thoenen H. Neurotrophins and neuronal plasticity. Science 1995;270:593-98.
9. Ma YL, Wang HL, Wu HC, Wei CL, Lee EH. Brain-derived neurotrophic factor
antisense oligonucleotide impairs memory retention and inhibits long-term potentiation in rats. Neuroscience 1998;82:957-67.
10. Xu B, Goulding EH, Zang K, et al. Brain-derived neurotrophic factor regulates energy balance downstream of melanocortin-4 receptor. Nat Neurosci
2003;6:736-42.
11. Nakagawa T, Tsuchida A, Itakura Y, et al. Brain-derived neurotrophic factor
regulates glucose metabolism by modulating energy balance in diabetic mice.
Diabetes 2000;49:436-44.
12. Tsuchida A, Nonomura T, Nakagawa T, et al. Brain-derived neurotrophic factor ameliorates lipid metabolism in diabetic mice. Diabetes Obes Metab
2002;4:262-9.
13. Fujimura H, Altar CA, Chen R, et al. Brain-derived neurotrophic factor is stored
in human platelets and released by agonist stimulation. Thromb Haemost
2002;87:728-34.
14. Radka SF, Holst PA, Fritsche M, Altar CA. Presence of brain-derived neurotrophic factor in brain and human and rat but not mouse serum detected by a
sensitive and specific immunoassay. Brain Res 1996;709:122-301.
15. Rosenfeld RD, Zeni L, Haniu M, et al. Purification and identification of brain-derived neurotrophic factor from human serum. Protein Expr Purif 1995;6:465-71.
16. Karege F, Perret G, Bondolfi G, Schwald M, Bertschy G, Aubry JM. Decreased
serum brain-derived neurotrophic factor levels in major depressed patients. Psychiatry Res 2002;109:143-8.
17. Shimizu E, Hashimoto K, Okamura N, et al. Alterations of serum levels of brainderived neurotrophic factor (BDNF) in depressed patients with or without antidepressants. Biol Psychiatry 2003;54:70-5.
18. Narisawa-Saito M, Wakabayashi K, Tsuji S, Takahashi H, Nawa H. Regional
specificity of alterations in NGF, BDNF and NT-3 levels in Alzheimer’s disease.
Neuroreport 1996;7:2925-8.
19. Suwa M, Kishimoto H, Nofuji Y, et al. Serum brain-derived neurotrophic factor
level is increased and associated with obesity in newly diagnosed female patients with type 2 diabetes mellitus. Metabolism 2006; 55:852-7.
20. Krabbe KS, Nielsen AR, Krogh-Madsen R, et al. Brain-derived neurotrophic factor (BDNF) and type 2 diabetes. Diabetologia 2007;50:431-8.
21. Navaratna D, Guo SZ, Hayakawa K, Wang X, Gerhardinger C, Lo EH. Decreased cerebrovascular brain-derived neurotrophic factor-mediated neuroprotection in the diabetic brain. Diabetes 2011;60:1789-96.
22. Kushikata T, Fang J, Krueger JM. Brain-derived neurotrophic factor enhances
spontaneous sleep in rats and rabbits. Am J Physiol 1999;276:1334-8.
23. Krueger JM, Rector DM, Churchill L. Sleep and cytokines. Sleep Med Clin
2007;2:161-9.
24. Opp MR. Cytokines and sleep. Sleep Med Rev 2005;9:355-64.
25. Faraguna U, Vyazovskiy VV, Nelson AB, Tononi G, Cirelli C. A causal role for
brain-derived neurotrophic factor in the homeostatic regulation of sleep. J Neurosci 2008;28:4088-95.
26. Martinowich K, Schloesser RJ, Jimenez DV, Weinberger DR, Lu B. Activitydependent brain-derived neurotrophic factor expression regulates cortistatininterneurons and sleep behavior. Mol Brain 2011;9:4-11.
27. Uchiyama S, ed. Sleep disorder for diagnosis and treatment guidelines (13th
ed), Tokyo: Ziho, 2009: 227-35.
28. The examination Committee of Criteria for “Obesity Disease” in Japan. Japan
Society for Study of Obesity. New criteria for ‘obesity disease’ in Japan. Circ J
2002;66:987-92.
Female
n = 140
-0.138
-0.106
-0.191*
-0.049
-0.179*
-0.089
-0.270**
-0.295***
*p < 0.05, **p < 0.01, ***p < 0.001.
to suppress the BDNF expression in the hippocampus.39 Additionally, a human study demonstrated a negative association
between the cortisol levels and the BDNF levels in the blood.40
Therefore, dyssomnia may reduce the BDNF levels in the brain
and the blood by altering the activity of the HPA system to increase the secretion of cortisol.
Many epidemiological studies have suggested gender differences are associated with dyssomnia.5 However, it remains unclear as to whether or not the sex differences in BDNF observed
in the results of the present study are related to sex differences
in dyssomnia.
Several limitations should be noted. First, the cross-sectional
design of the study limits the interpretation of causality between
the serum BDNF levels and dyssomnia. Second, since the sample size was relatively small and the subjects were workers, the
subjects may not be representative of the entire Japanese population. Third, we obtained only one serum sample at morning
for measurement of serum BDNF level from study subjects.
Therefore, we could not investigate the association between the
circadian change of serum BDNF levels and dyssomnia in this
study. This association should be investigated in future study,
since the serum level of BDNF has been demonstrated to be
influenced by several conditions, such as meal intake and level
of activity.10-12 Finally, we did not measure any other hormones
or mediators which were reported to correlate with dyssomnia.
Thus, further study is needed to clarify how serum BDNF levels
associate with those hormones and mediators, such as cortisol,
growth hormone, sex hormones, and melatonin.
Conclusion
In this study, serum BDNF levels were associated with dyssomnia in females but not in males. The association observed
in the female subjects remained statistically significant even
after adjusting for possible confounding factors, including
age, BMI, hypertension, dyslipidemia, diabetes mellitus, depression, smoking, drinking, and regular exercise. Our results
support the emerging concept that BDNF is a sleep regulatory
substance and may contribute to improving understanding of
the pathogenic mechanisms of dyssomnia. Further longitudinal
studies of large populations are required to elucidate the precise
relationship between the serum BDNF levels and dyssomnia.
653
29. Okamura T, Nakamura K, Kanda H, et.al. Effect of combined cardiovascular risk
factors on individual and population medical expenditures: a 10-year cohort study
of national health insurance in a Japanese population. Circ J 2007;71:807-13.
30. Radloff LS, Locke BZ, eds, Center for Epidemiologic Studies Depression Scale
(CES-D). Modified From: Rush J, et al. Psychiatric Measures, Washington, DC:
APA, 2000.
31. Ziegenhorn AA, Schulte-Herbrüggen O, Danker-Hopfe H, et al. Serum neurotrophins--a study on the time course and influencing factors in a large old age
sample. Neurobiol Aging 2007;8:1436-45.
32. Monteleone P, Tortorella A, Martiadis V, Serritella C, Fuschino A, Maj M. Opposite changes in the serum brain-derived neurotrophic factor in anorexia nervosa
and obesity. Psychosom Med 2004;66:744-48.
33. Molteni R, Wu A, Vaynman S, Ying Z, Barnard, RJ, Gomez-Pinilla F. Exercise
reverses the harmful effects of consumption of a high-fat diet on synaptic and
behavioral plasticity associated to the action of brain-derived neurotrophic factor.
Neuroscience 2004;123:429-40.
34. Vaynman S, Ying Z, Gomez-Pinilla F. Hippocampal BDNF mediates the efficacy
of exercise on synaptic plasticity and cognition. Eur J Neurosci 2004;20:2580-90.
35. Pan W, Banks WA, Fasold MB, Bluth J, Kastin AJ. Transport of brain-derived
neurotrophic factor across the blood-brain barrier. Neuropharmacology
1998;37:1553-61.
36. Pilegaard H. Evidence for a release of brain-derived neurotrophic factor from the
brain during exercise. Exp Physiol 2009;4:1062-69.
37. Leproult R, Copinschi G, Buxton O, Van Cauter E. Sleep loss results in an elevation of cortisol levels the next evening. Sleep 1997;20:865-70.
38. Vgontzas AN, Mastorakos G, Bixler EO, Kales A, Gold PW, Chrousos GP. Sleep
deprivation effects on the activity of the hypothalamic-pituitary-adrenal and
growth axes: potential clinical implications. Clin Endocrinol (Oxf) 1999;51:205-15.
39. Smith MA, Makino S, Kvetnansky R, Post RM. Stress and glucocorticoids affect
the expression of brain-derived neurotrophic factor and neurotrophin-3 mRNAs
in the hippocampus. J Neurosci 1995;15:1768-77.
40. Mackin P, Gallagher P, Watson S, Young AH, Ferrier IN. Changes in brain-derived neurotrophic factor following treatment with mifepristone in bipolar disorder
and schizophrenia. Aust N Z J Psychiatry 2007;41:321-26.
acknowledgments
This study was supported financially by St. Mary’s College and the Grants-in-Aid from
Science Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (Kakenhi). The authors thank Dr. Kyoichi Adachi for helpful disscussion,
and the staff members of the Institute of Health Science, Kyushu University and CRC for
their cooperation throughout this study.
Submitted for publication October, 2012
Accepted for publication December, 2012
Address correspondence to: Reiko Nishichi, Department of Community Health
Nursing, Shimane University Faculty of Medicine，89-1 Enya-cho, Izumo, Shimane,
693-8501, Japan; Tel and Fax: +81-853-20-2336; E-mail: rei130@lime.ocn.ne.jp
(nishichi@med.shimane-u.ac.jp)
654
Identification of Insomnia in a Sleep Center Population Using
Electronic Health Data Sources and the Insomnia Severity Index
Carl A. Severson1; Willis H. Tsai, M.D., F.A.A.S.M.2,3; Paul E. Ronksley, M.Sc.2; Sachin R. Pendharkar, M.D.2,3
O’Brien Centre for the Health Sciences Program, University of Calgary, Alberta, Canada; 2Department of Community Health Sciences,
University of Calgary, Alberta, Canada; 3Division of Respirology, Department of Medicine, University of Calgary, Alberta, Canada
1
Study Objectives: To assess the validity and efficacy of using electronic health data to identify a physician diagnosis of
insomnia in a population of patients referred for testing at a
tertiary sleep center.
Methods: Retrospective cohort study in a tertiary sleep
center in Calgary, Alberta, Canada. Cohort consisted of
1,207 patients referred for sleep diagnostic testing and/or
assessment by a sleep physician. Two sleep physicians
independently assigned each patient a primary sleep diagnosis. Univariate logistic regression was used to identify
variables that were predictive for insomnia from online questionnaire and diagnostic testing data. Diagnostic algorithms
derived from these predictors and from the Insomnia Severity Index were evaluated against physician diagnosis as a
reference standard.
Results: The combination of self-reported sleep latency > 20
minutes, total sleep time < 6.5 hours per night, the inability to
fall asleep after waking, BMI < 27 kg/m2, and Epworth Sleepiness Scale score < 9 had very high specificity (99.3%) for diagnosing insomnia; however, sensitivity was poor (11.8%). Other
algorithms derived from these data had either high sensitivity
or high specificity. No combination of variables yielded simultaneous high sensitivity and specificity. Likewise, the Insomnia
Severity Index can be highly sensitive or highly specific at identifying insomnia, but not both.
Conclusions: Diagnostic algorithms derived from electronic
data can provide high specificity or high sensitivity for identifying insomnia.
Keywords: Insomnia, clinical prediction, decision rule, diagnostic algorithm
Citation: Severson CA; Tsai WH; Ronksley PE; Pendharkar
SR. Identification of insomnia in a sleep center population using electronic health data sources and the insomnia severity
index. J Clin Sleep Med 2013;9(7):655-660.
T
he prevalence of chronic insomnia is as high as 30%, but
estimates range considerably, depending on the criteria
used to define insomnia and the sample population used.1-3 Insomnia has also been associated with high levels of healthcare
utilization, and increased direct and indirect healthcare costs.
For instance, Ozminkowski et al.4 estimated that the combined
direct and indirect costs over a 6-month time-period for adults
in the U.S. with insomnia were, per person, $1253.00 more than
matched controls. Similarly, Morin et al.5 estimated the cumulative costs of insomnia in the Canadian province of Quebec
(population approximately 8 million6) to exceed six billion dollars (Cdn) per annum.
A diagnosis of insomnia is typically established through
assessment by an experienced clinician. Several efficacious
treatments for insomnia, such as cognitive behavioral therapy
(CBT), exist. Yet wait times and lack of access to insomnia specialists can be a barrier to diagnosis and treatment. Moreover,
insomnia may occur independently, or may coexist with other
sleep disorders, which can complicate diagnosis and treatment.
Screening tools that can accurately and reliably identify primary insomnia could help from a triage standpoint, as they may
direct newly referred patients to the appropriate specialists and/
or diagnostic testing. In a research setting, an insomnia screening tool would be desirable for case finding. While a breadth
of insomnia questionnaires and screening tools exist, most of
these tools were developed for use in large epidemiologic stud-
Brief Summary
Current Knowledge/Study Rationale: While a breadth of insomnia
questionnaires and screening tools exist, most of these tools were developed for use in large epidemiologic studies and lack validation in a
clinic setting. We sought to assess the validity of using electronic health
data for identifying patients with insomnia in a tertiary sleep centre.
Study Impact: Diagnostic algorithms derived from electronic data can
provide high specificity or high sensitivity for identifying insomnia. When
used to direct patients to the correct provider, or to preclude the need
for polysomnography, this could have significant impact on centralized
triage processes, clinician decision support, and healthcare costs.
ies and lack validation in a clinic setting. Moreover, very few
have used a clinician defined reference standard for insomnia.7-9
The aim of this study was to assess the validity and efficacy
of using electronic health data to identify a physician diagnosis
of insomnia in a population of patients referred for testing at a
tertiary sleep center.
METHODS
Patients
We identified all patients who completed an online questionnaire
and underwent clinical assessment and/or sleep diagnostic testing at
the Foothills Medical Center (FMC) Sleep Center between January
655
CA Severson, WH Tsai, PE Ronksley et al
1, 2009, and January 1, 2011. The FMC Sleep Center is the only
tertiary referral center for Calgary, Alberta, Canada (a catchment
area of approximately 1.3 million people). The referral base consists primarily of family physicians, but also includes physicians
from other disciplines. All referred patients are required to fill out
the online questionnaire at the time of referral. The University of
Calgary Conjoint Health Region Ethics Board approved this study.
Remmers Sleep Recorder (SagaTech Electronics Ltd, Calgary,
Canada) is an ambulatory monitor that measures snoring, oxygen saturation, respiratory airflow (by monitoring nasal pressure), and body position. The RDI is derived from automated
analysis of the oximetry signal using a 4% desaturation threshold. This algorithm uses both shape and magnitude of oxygen
desaturation to score respiratory events.10 Ambulatory studies
are manually reviewed by the interpreting physician with the
flow signal being used for quality assurance purposes. Ambulatory studies are repeated if there are discrepancies between the
automatically scored respiratory events and the airflow channel. This monitor has excellent agreement with the polysomnographically determined AHI.10 It has also been validated as a
clinical management tool.11,12
Determination of Primary Diagnosis
Insomnia
Two American Academy of Sleep Medicine board-certified
sleep physicians independently reviewed all patient charts and
assigned a primary sleep diagnosis to each patient in the cohort.
Insomnia was defined based on either: (a) primary clinical diagnosis by the treating clinician (if treating clinician was ABSM
certified in Behavioral Sleep Medicine); (b) primary clinical
diagnosis of insomnia by the treating physician AND a chart
review of the patient history to determine if patient met International Classification of Sleep Disorders, 2nd edition (ICSD-2),
criteria A-C (if the treating clinician was not an ABSM certified
psychologist). Primary insomnia was determined to be present
if there was consensus from both reviewing physicians on the
diagnosis. If a diagnosis could not be agreed upon, the disagreement was noted and the patient was excluded from analysis.
Polysomnography
Polysomnography was ordered at the discretion of the treating physician. Polysomnography (PSG) data were recorded by
a computerized system (Sandman Elite Version 8.0, Nellcor
Puritan Bennett [Melville] Ltd, Kanata, Ontario, Canada). This
included a standardized montage: 3 electroencephalographic
channels (C4/A1, C3/A2, O1/A2), bilateral electroculograms
(EOG), submental electromyogram (EMG), bilateral leg EMGs,
and electrocardiography (ECG). Airflow was measured using a
nasal pressure transducer (Braebon Medical Corp, Ontario, Canada). Respiratory effort was assessed by inductance plethysmography (Respitrace Ambulatory Monitoring, Ardsley, NY USA),
and oxygen saturation was recorded by oximetry (953 Finger
Flex Sensor; Healthdyne Technologies). The RDI was defined
as the number of apneas and hypopneas/h of sleep. Apnea was
defined as a cessation of airflow ≥ 10 seconds. Hypopnea was
defined as an abnormal respiratory event lasting ≥ 10 sec, with ≥
30% reduction in thorocoabdominal movement or airflow compared to baseline and associated with ≥ 4% oxygen desaturation.
Other
A total of 13 other diagnostic categories were selected prior
to chart review. Eight diagnoses were developed based on the
ICSD-2 criteria: central sleep apnea (CSA); central nervous system (CNS) hypersomnolence; insomnia; obstructive sleep apnea
syndrome (OSAS); parasomnia; restless leg syndrome (RLS);
OSA/hypoventilation; and upper airway resistance syndrome
(UARS). Diagnoses were assigned based on a primary diagnosis by the treating physician AND chart review of the history
and diagnostic testing (to ensure that ICSD-2 criteria were met).
Six non-ICSD-2 diagnoses were also assigned: depression, fatigue, uncomplicated snoring, fibromyalgia, other, and normal.
Given that validated criteria for many of these diagnoses could
not be consistently extracted from the medical record, for nonICSD-2 diagnosis a diagnosis was assigned based on the primary
impression of the treating physician. Disease specific diagnostic
criteria were not employed for non-ICSD-2 diagnoses.
Statistics
The agreement in physician-assigned diagnoses was assessed
by the κ statistic. Simple logistic regression was used to identify predictive variables from the online questionnaire, using
the presence of insomnia as the dependent variable. Receiver
operator characteristic curves and box plots were used to visually select binary cutpoints for predictive continuous variables.
A full model was constructed from univariate binary predictors and reduced by stepwise regression. Diagnostic algorithms
were constructed from predictive variables in the parsimonious
model. Sensitivity, specificity, positive predictive value, and
negative predictive value were determined for each diagnostic
algorithm as well as for the ISI.
All analyses were performed using STATA 9.0 statistical
software (Stata Corporation, College Station, TX). Results are
presented as mean ± standard deviation unless otherwise stated.
An α value of 0.05 was used for all significance calculations.
Electronic Data Elements
Online Questionnaire
The online questionnaire is composed of 108 questions,
which provide a comprehensive overview of a patient’s demographics, anthropometrics, snoring history, daytime function,
and medical history, as well as sleep schedule, behavior, and
complaints. Three smaller, commonly used questionnaires are
also administered within the online questionnaire: the Epworth
Sleepiness Scale (ESS), Patient Health Questionnaire (PHQ),
and the Insomnia Severity Index (ISI).
RESULTS
Patient Characteristics
Ambulatory Monitoring
All patients undergo portable monitoring (level III sleep diagnostic testing) as part of the referral and triage process. The
Figure 1 illustrates the flow of patients. We identified 1,426
patients who underwent clinical assessment with or without
656
Identification of Insomnia in a Sleep Center Population
Figure 1—Patient flow
Table 1—Sleep diagnosis
Primary Diagnosis
OSAS
Insomnia
CNS Hypersomnolence
Normal
Other
Snoring
RLS
OSA/Hypoventilation
UARS
Depression
Fatigued
Parasomnia
Fibromyalgia
CSA
Total
All registered new referrals between January 1, 2009
and January 1, 2011 and completed an online sleep
questionnaire (n = 1,426)
Patients who refused
consent (n = 202)
Patients who provided informed consent (n = 1,224)
Patient with incomplete
questionnaire (n = 1)
Patients for whom
physicians could not agree
on diagnosis (n = 16)
Freq.
554
339
58
50
48
41
28
27
22
13
9
8
6
4
1,207
Percent
45.90
28.09
4.81
4.14
3.98
3.40
2.32
2.24
1.82
1.08
0.75
0.66
0.50
0.33
100
Table 2—Patient characteristics
Patients for whom physicians agreed on diagnosis
(n = 1,207)
Characteristic
Age (years)
Sex (male)
Weight (Lbs)
BMI (kg/m2)
ESS
ISI
Sleep Latency
(hours)
sleep diagnostic testing; 202 did not provide written consent
and were excluded. Of the remaining 1,223 patient charts,
16 were excluded from the analysis due to lack of consensus
over the primary sleep diagnosis, leaving a final cohort size
of 1,207.
Insomnia was the primary diagnosis in 339 patients (28%).
Tables 1 and 2 summarize the distribution of sleep diagnoses
and baseline characteristics of the cohort. The mean age in the
entire cohort was 45.4 ± 12.1 years, 56.8% were men, and mean
body mass index (BMI) was 30.6 ± 7.6 kg/m2. The mean ESS
and ISI scores of all patients were 11.1 ± 5.7 and 3.0 ± 6.8, respectively. The mean self-reported sleep latency for all patients
was 0.47 ± 0.72 hours.
Patients with insomnia were significantly younger and had
lower BMI, weight, and ESS scores than patients without insomnia. Patients with insomnia reported taking longer to fall
asleep and had higher ISI scores than patients without a diagnosis of insomnia.
All Patients
(n = 1,207)
45.4 ± 12.1
56.80%
198.9 ± 52.6
30.6 ± 7.6
11.1 ± 5.7
3.0 ± 6.8
Insomnia
(n = 339)
43.2 ± 12.1
41.30%
173.5 ± 42.2
27.4 ± 6.5
9.0 ± 5.5
4.2 ± 8.0
No Insomnia
(n = 868)
46.2 ± 12.1*
62.80%**
208.8 ± 53.0*
31.8 ± 7.7*
11.9 ± 5.5*
2.6 ± 6.22*
0.47 ± 0.72
0.80 ± 1.15
0.34 ± 0.37*
*p < 0.05 between groups with Insomnia and No Insomnia.
**χ 2 = 45.9, p < 0.001.
Algorithm Performance
Diagnostic algorithm performance is summarized in Table 4.
The combination of self-reporting a sleep latency > 20 min,
sleep time < 6.5 h sleep/night, inability to fall asleep after waking, a BMI < 27, and ESS score < 9 had very high specificity (99.3%) for a diagnosis of insomnia; however, sensitivity
was poor (11.8%). Similarly, a self-reported sleep time < 6.5 h
with concomitant sedative/hypnotic use had a high specificity
(96.7%) at the expense of sensitivity (18.6%). No combination
of variables simultaneously had a high sensitivity and specificity. A model incorporating the use of a sedative/hypnotic or a
reported sleep time < 6.5 h maximized diagnostic performance
(sensitivity 71.4%, specificity 67.4%, positive predictive value
46.1%, negative predictive value 85.8%).
The combination of clinical and diagnostic test data did not
improve diagnostic performance when compared to clinical
data alone. For instance, using a self-reported sleep time < 6.5 h
and an RDI < 5 yields moderate sensitivity (76.4%) and specificity (71.5%). The inability to fall asleep after waking and an
RDI < 5 yields similar levels of sensitivity (70.2%) and specificity (68.1%).
Diagnostic Agreement
Physicians agreed on 98.69% of assigned diagnoses
(1,207/1,223). The un-weighted κ statistic was 0.98 (± 0.016).
Univariate Predictors of Insomnia
Self-reported use of a sleep aid, sleep latency (measured in
hours), and sedative/hypnotic use were predictive of a diagnosis of insomnia (Table 3). The Epworth score, BMI, average
sleep time, and ability to return to sleep after waking during the
night were predictive of a diagnosis other than insomnia. The
following binary cutoffs for continuous predictive variables
were selected from ROC curves and box plots: sleep latency
(20 min), sleep time (6.5 h), BMI (27 kg/m2), and ESS (9/24).
657
12%, specificity 99%). Similar results could be achieved with
a combination of self-reported sleep time of less than 6.5 hours
and sedative or hypnotic use (sensitivity 19%, specificity 97%).
Although lacking simultaneously high sensitivities and specificities, these diagnostic algorithms provide a simple and highly
accurate method of identifying at least a subset of patients with
and without insomnia.
Only two published questionnaire-based tools have been
developed to differentiate between insomnia and other sleep
disorders in a sleep clinic population: the Global Sleep Assessment Questionnaire (GSAQ) and the Holland Sleep Disorders
Questionnaire (HSDQ).13,14 When administered to a combination of primary care and sleep clinic patients, the 11-item
GSAQ could discriminate between insomnia and other sleep
disorders (as diagnosed by a sleep clinician) with a sensitivity of 79% and a specificity of 57%.13 The aim of the HSDQ
is to assess the six different groups of sleep disorders as defined by ICSD-2 criteria (sleep-related breathing disorder, hypersomnia, circadian rhythm sleeping disorder, parasomnias,
sleep related movement disorders, and insomnia). When administered to a population of 891 patients referred for testing
to a sleep center in Holland, their 40-item questionnaire had an
optimized sensitivity of 82% and optimized specificity of 69%
at differentiating between insomnia and the other five classes
of sleep disorders.14 While both of these questionnaires have
moderate levels of combined sensitivity and specificity, neither
measure is maximized.
The Insomnia Severity Index is a brief, well-validated
questionnaire designed to assess insomnia severity and outcomes.15-18 Morin et al. examined the ability of the ISI to identify insomnia in a clinical population, demonstrating that the
ISI could identify patients with physician-diagnosed insomnia
with high sensitivity and specificity: an ISI score ≥ 11 yielded
a sensitivity and specificity of 92.7% and 100%, respectively.15
However, in our clinic-based population, the ISI did not achieve
simultaneously high sensitivity and specificity. The differences
in ISI performance can likely be attributed to the different populations. Though both studies used similarly rigorous definitions of insomnia, Morin et al. looked at the ability of the ISI to
identify patients with insomnia when comparing those patients
to a cohort of healthy controls, whereas we examined the ability of the ISI to identify patients with insomnia in patients who
were referred to a sleep center. Given the overlap in comorbidity and symptoms between different sleep disorders, it is not
surprising that shared symptoms dilute the diagnostic accuracy
of any screening instrument.
The GSAQ, HSDQ, and our data present similarly moderate measures of combined sensitivity and specificity. Additionally though, we present algorithms to maximize either
measure on their own. This is important, as there are situations
when a high sensitivity or specificity is desirable even if the
reciprocal measure is lower. For instance, practice parameters
suggest that polysomnography is not necessary for the routine assessment and diagnosis of insomnia.19 Highly specific
algorithms that rule in a diagnosis of insomnia may reduce the
need for polysomnography in patients positively identified by
the algorithm. Given the cost of polysomnography, the identification of even a small subset of patients as not needing PSG
could lead to large healthcare and insurance savings. How-
Table 3—Univariate predictors
Variable
Odds Ratio
Sleep Aid Use (yes/no)
2.32
Sleep Latency (hours)
2.19
Sedative/Hypnotic Use (yes/no)
1.70
ESS (per point)
0.96
BMI
0.92
Sleep Time (hours)
0.73
Return to Sleep After Waking (yes/no)
0.51
95% CI
1.60 - 3.36
1.57 - 3.07
1.09 - 2.67
0.93 - 0.99
0.90 - 0.95
0.65 - 0.82
0.35 - 0.74
Table 4—Performance of Diagnostic Algorithms
Sens
Spec
PPV
NPV
Sleep latency > 20 min
Sleep time < 6.5 hours
BMI < 27
ESS score < 9
Do not return to sleep after
waking
11.8
99.3
87.0
74.2
Sleep time < 6.5 hours &
Use sedative/hypnotic
18.6
96.7
68.5
75.2
Sleep time < 6.5 hours or
Use sedative/hypnotic
71.4
67.4
46.1
85.8
RDI < 5
RDI < 15
RDI < 30
54.5
94.1
97.6
77.2
43.4
23.8
43.7
35.0
29.4
84.0
95.8
96.9
Sleep time < 6.5 or RDI < 5
76.4
71.5
41.2
86.2
Do not return to sleep after
waking or RDI < 5
70.2
68.1
46.2
85.3
ISI ≥ 8
ISI ≥ 11
ISI ≥ 15
ISI ≥ 22
92.7
87.8
73.2
28.0
16.5
24.7
43.7
86.7
36.5
37.7
40.3
52.3
81.3
79.6
75.8
69.9
The diagnostic performance of the Insomnia Severity Index
(ISI) is also summarized in Table 3. In our clinic population,
the ISI demonstrated either high sensitivity or specificity, but
not both. An ISI score ≥ 8 demonstrates the highest sensitivity (92.7%) that could be achieved using ISI data alone; however specificity was poor (16.5%). An ISI score ≥ 22 yields
the maximum specificity that could be achieved with the ISI
alone (86.7%), but at the expense of sensitivity (28.0%). No
single ISI cutoff resulted in simultaneously high sensitivity
and specificity.
DISCUSSION
To the best of our knowledge, this is the first study to validate
diagnostic algorithms using electronic health data in a clinicbased population. Diagnostic algorithms using electronic health
data from an online questionnaire can achieve high sensitivities
or specificities for identifying insomnia, but not both. A high diagnostic specificity can be achieved using a self-reported sleep
latency of greater than 20 minutes, estimated sleep time of less
than 6.5 hours, inability to fall asleep after waking, BMI of
less than 27 kg/m2, and an ESS score of less than 9 (sensitivity
658
Identification of Insomnia in a Sleep Center Population
ever, polysomnography would still be required in patients
with persistent symptoms despite primary management or in
some patients for whom sleep disordered breathing has not
been ruled out. Conversely, highly sensitive algorithms allow
us to confidently rule out a diagnosis, which could help us
direct patients to the correct provider and thus improve clinic
efficiency. Highly sensitive or specific algorithms are also important in a research setting as they are used in identifying
cohorts, validating comorbidities and monitoring prevalence
and incidence, for example.
The use of an online questionnaire or electronic health data
is compelling both in terms of potential for scalability and
resource demands for data acquisition and analysis. For instance, questionnaires can be disseminated to large numbers
of patients or study cohorts simply by providing a link or web
address. Furthermore, the collection and storage of responses
in a structured and computable manner facilitates linkage to
existing clinical databases. Electronic algorithms can also be
easily integrated with well-structured electronic databases,
allowing for several potential uses. In a triage setting, for
example, such algorithms could be implemented to automatically classify patients and alert triage nurses to this identification. From a research perspective, diagnostic algorithms can
be used to rapidly query massive datasets to find large samples
for study inclusion. As an example, case finding algorithms to
identify outbreaks of influenza have been previously reported
and show the potential use of electronic health data sources
for this purpose.20
We suggest that using highly sensitive or specific algorithms
has the potential to improve clinical efficiency by identifying
subsets of patients and directing them to the correct provider
or clinical test. However, the ability of electronic algorithms
to improve clinical efficiency is largely unexamined in the literature. A recent study by Stein et al.21 used brief self-reported
electronic questionnaires delivered via a computer kiosk to help
assess and treat women presenting to the emergency department for urinary tract infections. These investigators found that
patients randomized to use this system had shorter lengths of
stay in the emergency department than patients who continued
via regular clinical pathways. Though not implementing a diagnostic algorithm electronically, the results of Stein et al. suggest
that implementation of electronic questionnaires have potential
to improve clinical efficiency. Further research is necessary to
assess and quantify how electronic algorithms may improve efficiency and patient flow in a clinical setting.
Our results should be interpreted within the context of the
strengths and limitations of our study. Firstly, our cohort was
selected from referrals to a single academic sleep center. There
are only a few other referral choices for sleep medicine in our
catchment area, and there is no incentive for referring physicians to choose one center over the other. This fact, coupled
with our large sample size, low exclusion rate, and the consistency of our cohort’s demographic characteristics with those of
other referral populations, suggest that selection bias due to a
single center is not a concern.
Secondly, it should be noted that a clinical interview was not
part of the diagnosis process. However, to ensure the integrity of
our reference standard, diagnoses were assigned through independent chart review and required consensus by two board-cer-
tified sleep physicians. The strength of our reference standard is
reflected in the high κ score and percent agreement between the
two raters (0.98 (± 0.016) and 98.69%, respectively).
Finally, although all patients underwent level III sleep diagnostic testing, polysomnography was at the discretion of the
treating physician. Given the use of ambulatory monitoring, it
is unlikely that sleep disordered breathing would be missed.
However, non-respiratory ICSD-2 polysomnographically
based diagnoses could be missed, if not initially suspected by
the treating clinician. Moreover, non-ICSD-2 diagnoses were
assigned based on the impression of the treating clinician and
may not necessarily be valid.
CONCLUSION
Diagnostic algorithms derived from electronic data can provide high specificity or high sensitivity for identifying insomnia. While it is not feasible to simultaneously achieve both high
sensitivity and specificity using these data, it is possible to simply and accurately identify a subset of patients with and without insomnia using only a few simple questions extracted from
online and/or electronic sources. When used to direct patients
to the correct provider, or to preclude the need for polysomnography, this could have significant impact on centralized triage
processes, clinician decision support, and healthcare costs. Furthermore, these algorithms can be used in a research capacity to
identify cohorts and monitor prevalence and incidence, among
other uses. Towards these ends, the ability of these algorithms
to improve clinic efficiency and decision support, and their uses
in a research setting warrant further study and validation.
REFERENCES
1. Ohayon MM. Epidemiology of insomnia: what we know and what we still need to
learn. Sleep Med Rev 2002;6:97-111.
2. Ohayon MM, Reynolds CF. Epidemiological and clinical relevance of insomnia
diagnosis algorithms according to the DSM-IV and the International Classification of Sleep Disorders (ICSD). Sleep Med 2009;10:952-60.
3. Morin CM, LeBlanc M, Bélanger L, Ivers H, Mérette C, Savard J. Prevalence of
insomnia and its treatment in Canada. Can J Psychiatry 2011;56:540-8.
4. Ozminkowski R, Wang S, Walsh J. The direct and indirect costs of untreated
insomnia in adults in the United States. Sleep 2007;30:263-73.
5. Daley M, Morin CM, LeBlanc M, Grégoire J-P, Savard J. The economic burden
of insomnia: direct and indirect costs for individuals with insomnia syndrome,
insomnia symptoms, and good sleepers. Sleep 2009;32:55-64.
6. Canada S. Population by sex and age group, by province and territory. [Internet]. 2011 [cited 2012 Sep 13]. Available from: http://www.statcan.gc.ca/
tables-tableaux/sum-som/l01/cst01/demo31a-eng.htm
7. Novak M, Mucsi I, Shapiro CM, Rethelyi J, Kopp MS. Increased utilization of
health services by insomniacs–an epidemiological perspective. J Psychosom
Res 2004;56:527-36.
8. Kessler R, Berglund P, Coulouvrat C. Insomnia and the performance of US workers: results from the America insomnia survey. Sleep 2011;34:1161-71.
9. Morin CM, LeBlanc M, Daley M, Gregoire JP, Mérette C. Epidemiology of insomnia: prevalence, self-help treatments, consultations, and determinants of helpseeking behaviors. Sleep Med 2006;7:123-30.
10. Vázquez JC, Tsai WH, Flemons WW, et al. Automated analysis of digital oximetry in the diagnosis of obstructive sleep apnoea. Thorax 2000;55:302-7.
11. Mulgrew A, Fox N, Ayas N, Ryan C. Diagnosis and initial management of obstructive sleep apnea without polysomnography. Ann Intern Med 2007;146:157-66.
12. Whitelaw WA, Brant RF, Flemons WW. Clinical usefulness of home oximetry
compared with polysomnography for assessment of sleep apnea. Am J Respir
Crit Care Med 2005;171:188-93.
13. Roth T, Zammit G, Kushida C, et al. A new questionnaire to detect sleep disorders. Sleep Med 2002;3:99-108.
659
14. Kerkhof GA, Geuke ME, Brouwer A, Rijsman RM, Schimsheimer RJ, Van
Kasteel V. Holland Sleep Disorders Questionnaire: a new sleep disorders questionnaire based on the International Classification of Sleep Disorders-2. J Sleep
Res 2013;22:104-7.
15. Morin CM, Belleville G, Bélanger L, Ivers H. The Insomnia Severity Index: psychometric indicators to detect insomnia cases and evaluate treatment response.
Sleep 2011;34:601-8.
16. Okun ML, Kravitz HM, Sowers MF, Moul DE, Buysse DJ, Hall M. Psychometric
evaluation of the Insomnia Symptom Questionnaire: a self-report measure to
identify chronic insomnia. J Clin Sleep Med 2009;5:41-51.
17. Yang M, Morin CM, Wallenstein G V. Original article Interpreting score differences in the Insomnia Severity Index: using health-related outcomes to define
the minimally important difference. Curr Med Res Opin 2009;25:2487-94.
18. Thorndike FP, Ritterband LM, Saylor DK, Magee JC, Gonder-Frederick LA, Morin CM. Validation of the Insomnia Severity Index as a Web-Based Measure.
Behav Sleep Med 2011;9:216-23.
19. Littner M, Hirshkowitz M, Kramer M, et al. Practice parameters for using polysomnography to evaluate insomnia: an update. Sleep 2003;26:754-60.
20. DeLisle S, South B, Anthony JA, et al. Combining free text and structured electronic medical record entries to detect acute respiratory infections. PloS One
2010;5:e13377.
21. Stein JC, Navab B, Frazee B, et al. A randomized trial of computer kiosk-expedited management of cystitis in the emergency department. Acad Emer Med
2011;18:1053-9.
acknowledgments
This work was supported through funding from the O’Brien Center for the Health
Sciences and the FMC Sleep Centre Development Fund.
Submitted for publication November, 2012
Submitted in final revised form January, 2013
Address correspondence to: Willis H. Tsai, Rockyview General Hospital, 7007 14th
Street SW, Calgary, Alberta, Canada, T2V 1P9; E-mail: tsai@ucalgary.ca
660
Thomas Roth, Ph.D.1; Patricia Berglund, M.B.A.2; Victoria Shahly, Ph.D.3; Alicia C. Shillington, Ph.D.4; Judith J. Stephenson, S.M.5;
Ronald C. Kessler, Ph.D.3
Sleep Disorders and Research Center, Henry Ford Health System, Detroit, MI; 2Institute for Social Research, University of
Michigan, Ann Arbor, MI; 3Department of Health Care Policy, Harvard Medical School, Boston, MA; 4EPI-Q, Inc., Oak Brook, IL;
5
HealthCore, Inc., Wilmington, DE
1
Study Objectives: Although difficulty maintaining sleep
(DMS) is the most common nighttime insomnia symptom
among US adults, many FDA-approved hypnotics have indications only for sleep onset, stipulating bedtime administration to offset residual sedation. Given the well-known
self-medication tendencies of insomniacs, concern arises
that maintenance insomniacs might be prone to self-administer their prescribed hypnotics middle-of-the-night (MOTN)
after nocturnal awakenings, despite little efficacy-safety
data supporting such use. However, no US data characterize the actual population prevalence or correlates of MOTN
hypnotic use.
Methods: Telephone interviews assessed patterns of prescription hypnotic use in a national sample of 1,927 commercial
health plan members (ages 18-64) receiving prescription hypnotics within 12 months of study. The Brief Insomnia Questionnaire assessed insomnia symptoms.
Results: 20.2% of respondents reported MOTN hypnotic use,
including 9.0% who sometimes used twice-per-night (once
at bedtime plus once MOTN) and another 11.2% who some-
times used MOTN, but never twice-per-night. The remaining
79.8% used exclusively at bedtime. Among exclusive MOTN
users, only 14.0% used MOTN on the advice of their physician
(52.6% of those seen by sleep medicine specialists and 42.6%
by psychiatrists vs. 5.2% to 13.6% seen by other physicians).
MOTN use predictors included DMS being the most bothersome sleep problem, long duration of hypnotic use, and low
frequency of DMS.
Conclusions: One-fifth of patients with prescription hypnotics used MOTN, only a minority on advice from their physicians. Since significant next-day cognitive and psychomotor
impairment is documented with off-label MOTN hypnotic use,
prescribing physicians should question patients about unsupervised MOTN dosing.
Keywords: Insomnia, sleep maintenance, hypnotics, middleof-the-night, dosing, medication adherence, prevalence
Citation: Roth T; Berglund P; Shahly V; Shillington AC; Stephenson JJ; Kessler RC. Middle-of-the-night hypnotic use in
a large national health plan. J Clin Sleep Med 2013;9(7):661668.
I
nsomnia is the most common nighttime sleep problem, and
sleep maintenance insomnia the most common insomnia
symptom both in the general population1 and among adults
with clinical insomnia.2 An estimated one-fourth of all noninstitutionalized US civilians and two-thirds of US insomniacs
report frequent sleep maintenance problems involving nocturnal awakenings and/or prolonged wake time after nocturnal
awakenings.2 Although insomnia symptoms are highly variable from night to night and frequently co-occur, DMS presents
alone in roughly 20% of transiently or moderately symptomatic
adults and 17% of insomniacs.2 DMS persists in more than 90%
of population-based cases for at least 6 months3 and upwards
of 70% for at least one year.4 Sleep maintenance insomnia is
associated with a variety of impairments, including: daytime
sleepiness3; disruptions in cognition, motor coordination, and
mood3; decrements in perceived health2; and increased healthcare utilization.5 Sleep maintenance insomnia accounts for
more daytime sleepiness6 and poor perceived health2 than any
other nighttime insomnia symptom.
Despite the high prevalence of sleep maintenance insomnia,
the indications of many widely used hypnotics currently approved by the US Food and Drug Administration specify efficacy only for sleep onset.7,8 Furthermore, among those hypnotics
Brief Summary
Current Knowledge/Study Rationale: Although difficulty maintaining
sleep (DMS) is the most common nighttime insomnia symptom among
US adults, many FDA-approved hypnotics have indications only for
sleep onset and specify bedtime administration to offset next-day sedation. Given the well-known self-medication tendencies of insomniacs and
adverse cognitive and psychomotor impacts of hypnotic-related residual
sedation, it is important to assess possible off-label MOTN use among
maintenance insomniacs in the community.
Study Impact: The current study offers a preliminary view of realworld MOTN hypnotic use in a national sample of insured Americans.
Information is also provided on the distribution of insomnia symptoms
of once-a-night MOTN users and recommendations for MOTN use by
prescribing physicians.
that are approved for nocturnal awakenings and/or prolonged
wake time after nocturnal awakenings, all but one stipulate bedtime administration with middle-of-the-night use being explicitly disapproved to minimize risk of residual sedation. In other
words, such bedtime hypnotics are designed to be preventative
treatments for possible nocturnal awakenings rather than active treatments administered after nocturnal awakenings occur.
The single hypnotic accepted by the FDA for as-needed MOTN
661
T Roth, P Berglund, V Shahly et al
use after nocturnal awakenings to date was approved only
very recently (http://www.fda.gov/NewsEvents/Newsroom/
PressAnnouncements/ucm281013.htm. Published November
23, 2011. Accessed November 23, 2011).
Given that 40% to 80% of insomniacs with prominent sleep
maintenance problems do not experience symptoms every
night,9 and many are only symptomatic 3-4 nights a week,3
one potentially important difference between nightly bedtime
hypnotic use to prevent nocturnal awakening and as-needed
MOTN use only after nocturnal awakenings occur is a dramatic reduction in the frequency of hypnotic use with MOTN
dosing. Although it is difficult to know if this reduced frequency of hypnotic use would have beneficial health effects,
as documenting adverse effects of prolonged hypnotic use is
problematic given the well-known physical and psychological
vulnerabilities of long-term hypnotic users. It is noteworthy,
though, within the context of that limitation that prolonged
hypnotic use has been linked with multiple subsequent adverse health outcomes.10
Hypnotics approved for bedtime use to prevent nocturnal
awakenings have demonstrated inconsistent effects on sleep
maintenance parameters during controlled treatment trials.8 In
light of these inconsistent efficacy profiles, experimental studies have begun exploring efficacy-safety of MOTN dosing of
FDA-approved (for bedtime use) hypnotics9,11-13 and investigational agents.14,15 Although these studies have found MOTN
dosing of these medications associated with improvements in
sleep maintenance, trials involving off-label use of approved
hypnotics have also found next-day compromises in psychomotor and cognitive functioning,13,16-18 especially at higher dosages. Concerns have been raised that the tight controls in these
studies may underestimate real-life adverse effects of MOTN
use owing to patient non-compliance regarding hypnotic dosages and timing of doses.17,18 For instance, studies of blood levels
among drivers stopped for driving under the influence (DUI) in
the United States,19 Norway,20 and Sweden21 have found higher-than-expected blood levels of various hypnotic drugs, suggesting that hypnotic misuse involving escalated dosages and/
or improper timing of doses may be associated with impaired
driving.22 This is consistent with other evidence that high proportions of insomniacs self-medicate.23
Since many of the most widely used hypnotics approved by
the FDA have indications that specify efficacy only for sleep
onset symptoms, concern arises that maintenance insomniacs
might be prone to self-administer their prescribed hypnotics
after nocturnal awakenings despite little efficacy-safety data
supporting off-label MOTN use. Given possible adverse shortterm impacts of off-label MOTN dosing on cognitive and psychomotor functioning, possible long-term impacts of prolonged
hypnotic use on health, and the very high prevalence of DMS,
it is important to establish the extent of unsupervised MOTN
hypnotic use in the population. However, we are aware of no
community-based epidemiological data regarding the actual
magnitude or correlates of MOTN dosing. We conducted a survey to provide basic data of this sort in a sample of insured
employees of a large national health plan receiving hypnotic
prescriptions during the 12 months before study. As restrictive
conditions on recruitment imposed by the Health Plan resulted
in a low survey response rate, caution must be used interpreting
results. Nonetheless, the findings are useful given absence of
other data on this issue.
METHODS
The Sample
The sample consisted of adult (ages 18-64 years) members
of a large (over 34 million members) national US commercial
health plan who received prescriptions for one or more FDAapproved hypnotics at some time in the 12 months before the
survey. The sample was restricted to fully insured members enrolled in the Health Plan for at least 12 months in order to allow
claims data to be used in substantive analyses. Sample eligibility was also limited to members who provided the Plan with
a telephone number, spoke English, and had no impairment
that precluded their ability to be interviewed by telephone. The
sample was selected with stratification to match the health plan
distribution on the cross-classification of age (18-34, 35-49, 5064) and sex. In an effort to limit sample burden, attempts to
contact respondent households were limited by the Health Plan
to 2 contacts, except for a 25% subsample of households, in
which up to 9 calls were permitted in order to obtain at least
some information about hard-to-reach respondents. The data
were weighted to adjust for this under-sampling of hard-toreach respondents.
Recruitment and Consent
Survey recruitment began with an advance letter sent to a
probability sample of Plan members meeting eligibility requirements explaining that the survey was designed “to better understand how sleep problems affect the daily lives of people,”
that respondents were randomly selected, that responses were
confidential, that participation was voluntary and would not affect health care benefits, and that a $20 incentive was offered
for participation among eligible respondents. A toll-free number was included for respondents who wanted to ask questions
or opt out. Following initial phone contact, verbal informed
consent was obtained before beginning interviews. The Human
Subjects Committee of the New England IRB (www.neirb.com)
approved these recruitment, consent, and field procedures.
Measures
The survey consisted of two parts. All respondents were administered Part I, which asked them to specify when during the
course of the evening or night they used their sleep medication(s)
in the past 12 months. Part II of the survey was then administered only to (i) all Part I respondents who acknowledged using
FDA-approved hypnotics after nighttime waking in order to resume sleep, but who never used hypnotics twice in one night (i.e.,
both at bedtime to get to sleep and also after waking at night to
resume sleep), whom we refer to throughout this paper as onceper-night MOTN users, and (ii) a random 20% subsample of Part
I respondents who reported using sleep medications exclusively
at bedtime, whom we refer to as exclusive bed-time users. We
excluded from Part II all those who ever used both at bedtime to
get to sleep and also after waking at night to resume sleep, whom
we refer to as twice-per-night MOTN users. The Part II data were
weighted so that the exclusive bedtime users received a weight of
662
Middle-of-the-Night Hypnotic Use
5 (i.e., the reciprocal of 20%) to adjust for their under-sampling
into Part II, making the weighted Part II sample representative of
all once-per-night MOTN users or exclusive bedtime users.
In addition to assessing patterns of hypnotic use, the Part II
survey examined insomnia symptoms using the Brief Insomnia Questionnaire (BIQ),24 a self-report measure of insomnia
symptoms without diagnostic hierarchy rules or organic exclusion rules that has been validated for use in telephone surveys. As respondents were by definition treated insomniacs,
BIQ questions asked how frequently they would have each of
4 nighttime sleep problems if they were unable to use sleep
medications: difficulty initiating sleep (DIS), difficulty maintaining sleep (DMS), early morning awakening (EMA), and
non-restorative sleep (NRS). This was done by prefacing the
questions with the following instruction: Imagine that you
were unable to take your sleep medicine at all. We then asked
respondents to estimate about how many nights out of 7 in a
typical week they would have problems falling asleep, have
problems remaining asleep throughout the night, wake up before you wanted to, and wake up still feeling tired or unrested if
they were unable to take sleep medicine. Follow-up questions
to positive responses then probed for information about typical
duration (e.g., how many minutes or hours it would typically
take them to fall asleep). Respondents reporting more than one
of these sleep problems were asked which one was most bothersome. Respondents were also asked about the duration and
frequency of their hypnotic use. MOTN users were additionally
asked about frequency of this use. Finally, all Part II respondents were administered a battery of standard sociodemographic questions. The complete text of the interview is posted at
http://www.hcp.med.harvard.edu/wmh/AIS_Study.php.
Table 1—The distribution of survey response dispositions
I. Number of subscribers for whom contact was attempted
A. Non-working number
B. Maximum calls were made without a resolution*
a. No household contact
b. Household contact
c. Total maximum calls
C. Refusal
a. Before starting interview
b. After starting interview
c. Total refusals
D. Interview
II. Cooperation and response rates
A. Cooperation rate†
B. Response rate‡
4,106
4,502
7,315
1,817
2,591
218
2,809
1,927
40.7
11.6
*By the term “resolution” we mean either a refusal or a completed
interview. †The cooperation rate was calculated among resolved cases
and equals the number of interviews divided by the sum of interviews
and refusals. This is equivalent to the American Association of Public
Opinion Research (AAPOR) Formula 3 definition of the cooperation
rate (http://www.aapor.org/Response_Rates_An_Overview1.htm). ‡The
response rate was calculated in the total sample and equals the number
of interviews, refusals, and maximum calls. This is equivalent to the
American Association of Public Opinion Research (AAPOR) Formula 1
definition of the response rate (http://www.aapor.org/Response_Rates_
An_Overview1.htm).
estimates. Statistical significance was consistently evaluated
using 0.05-level two-sided tests.
RESULTS
Analysis Methods
Given the low response rate, analysis began by comparing
respondents to non-respondents on key characteristics available
from health plan records. This was done with simple cross-tabulations. A post-stratification weight using the regression-based
propensity score method25 was then used to correct for significant differences between respondents and non-respondents
on these variables prior to carrying out substantive analyses.
Cross-tabulations with these weighted data were then used to
estimate the prevalence of MOTN use in the Part I sample (all
of whom, as noted above, received prescriptions for ≥ 1 FDAapproved hypnotic(s) in the past 12 months), while means were
calculated to estimate the proportion of all instances of hypnotic use that occurred at bedtime versus MOTN. Cross-tabulations and multiple logistic regression analysis were then used
in the weighted Part II sample (which, as noted above, included
once-per-night MOTN users and exclusive bedtime users, but
excluded twice-per-night MOTN users) to study the correlates
of once-per-night MOTN use versus exclusive bedtime use. Logistic regression coefficients and their standard errors were exponentiated for ease of interpretation and are reported as odds
ratios (ORs) with 95% confidence intervals. As survey data are
weighted, the design-based Taylor series linearization method26
implemented in the SAS 9.2 software system27 was used to estimate standard errors of coefficients and to calculate F tests
and Wald χ2 tests. Standard errors of prevalence estimates are
reported in parentheses in the text to the right of the prevalence
The Survey Cooperation Rate
The survey cooperation rate among resolved cases (i.e., the
rate of survey completion among target respondents with known
working telephone numbers who were reached and whose status
was resolved as either completers or refusers, excluding respondents who were never reached and those who were reached but
remained unresolved when data collection ended) was 40.7%
(Table 1). This is comparable to the cooperation rates found in
major government telephone surveys. For example, the 2009
CDC Behavioral Risk Factor Surveillance Survey28 had a cooperation rate, calculated in the same way as here, of 43.1% (ftp://
ftp.cdc.gov/pub/Data/Brfss/2009_Summary_Data_Quality_
Report.pdf; Accessed September 28, 2011).
It should be noted, though, that the Health Plan imposed
rather restrictive conditions on the recruitment process, as any
potential respondent household that was reached twice without
obtaining a resolution (an interview, a refusal, or a confirmation
of ineligibility) could not be contacted a third time. Unresolved
cases included those in which the target respondent was not at
home or said it was an inconvenient time to be interviewed.
Since Plan restrictions on number of phone contacts prevented
follow-up with many unresolved cases, the survey response rate
(i.e., the proportion of all households we attempted to contact
that yielded an interview exclusive of those known not to be eligible) was only 11.6%. This is much lower than response rates
663
Americans (95% CI: 800,000-1,200,000), with once-per-night
MOTN users representing approximately 550,000 (95% CI:
450,000-650,000) Americans in the same age range.
Table 2—Distribution of exclusive bedtime and MOTN
hypnotic use in the Part I sample (n = 1,927)
Exclusive bedtime users
MOTN users
Twice-per-night MOTN users
Once-per-night MOTN users
Exclusively MOTN
Sometimes bedtime, other times MOTN
Total MOTN users
Total sample
% (SE)
79.8 (1.1)
Comparison of Once-per-Night MOTN Users with
Exclusive Bedtime Users
MOTN users in the Part I sample did not differ significantly from exclusive bedtime users in average age (F2,1924 = 2.0,
p = 0.13), percent females (F2,1924 = 4.9, p = 0.08), or the specialty of the physician who most recently prescribed their sleep
medication (F2,1924 = 0.2-1.4, p = 0.49-0.91; Table 3). However,
mean number of years of hypnotic use was significantly longer
for MOTN users (5.5 years for once-per-night MOTN users and
6.2 years for twice-per-night MOTN users) than for exclusive
bedtime users (4.3 years; F2,1924 = 9.8, p < 0.001). MOTN users also differed significantly from exclusive bedtime users in
frequency of hypnotic use. Frequency of use was significantly
lower among once-per-night MOTN users (means of 9.2 nights
in the past month and 105.6 in the past 12 months) than exclusive bedtime users (means of 12.4 nights in the past month,
F1,1925 = 14.4, p < 0.001; and 155.0 in the past 12 months,
F1,1925 = 28.1, p < 0.001), but significantly higher among twiceper-night MOTN users than exclusive bedtime users (means of
18.1 nights in the past month and 228.6 in the past 12 months,
F1,1925 = 22.1-28.1, p < 0.001; Table 3).
In the Part II sample, once-per-night MOTN users were
asked how many times they used MOTN in the past month (30
nights). The mean was 2.8 (0.4) compared to the 9.2 mean for
overall monthly use. This means that the majority (70%) of use
among once-per-night MOTN users is at bedtime rather than
MOTN.
9.0 (0.8)
11.2 (0.8)
2.1 (0.4)
9.1 (0.8)
20.2 (1.1)
100.0 (–)
obtained in surveys in which no such limitations on number of
contact attempts are imposed.
Comparisons between Respondents and NonRespondents
The 1,927 respondents who completed the Part I survey (including both the subsample administered the Part II survey and
the subsample administered only the Part I survey) were compared to non-respondents on key characteristics available from
Health Plan records, including sociodemographics (age, sex),
geographic information obtained by matching the zip code of
household residence with Census data (region of the country,
urbanicity, and median household income in the Block Group of
residence), and global illness severity in the 12 months before
interview as assessed by the Deyo-Charlson score.29 Respondents were somewhat older than non-respondents, somewhat
more likely to be female and to live either in the Midwest or
South, and less likely to live in the West or in major metropolitan areas. Respondents also lived in zip code areas with lower
incomes than non-respondents. (Detailed results are available
on request.) Finally, respondents had higher global illness severity than non-respondents. As noted in the section on analysis
methods, a weight was imposed on the respondent data to adjust
for these differences between respondents and non-respondents.
Doctor Recommendations and Personal Rules for
Once-per-Night MOTN Use
Only a small minority of once-per-night MOTN users (14.0%
[2.9]) reported that the doctor who prescribed their sleep medicine advised them to use it in the middle of the night to resume
sleep. However, the proportion of patients reporting such directions for use varies significantly by type of provider (χ24 = 20.5,
p < 0.001), due to much higher proportions of reported doctor
advice to use MOTN among once-per-night MOTN users treated by a sleep medicine specialist (52.6% [35.3]) or psychiatrist
(42.6% [13.3]) than by a primary care doctor (9.6% [2.9]), pain
specialist (13.6% [7.5]), or other doctor (5.2% [5.2]).
The vast majority of once-per-night MOTN users (86.0%
[2.6]) reported having a personal rule for MOTN use. By far
the most common rules either involved amount of time left in
bed, such as not using MOTN unless expecting to be in bed
≥ 6 h after taking the medication (69.5% [3.5]) and/or involved
next-day demands (such as not using MOTN unless it was possible to sleep in the next morning (73.2% [3.5]). Presence vs.
absence of a rule for use was not significantly related to whether or not MOTN use was based on doctors’ advice (t = 1.8,
p = 0.071). Nor was presence vs. absence of a rule significantly
related either to frequency of MOTN use (3.4 [0.5]/month with
a rule vs. 4.6 [1.6]/month without a rule; t = 0.7, p = 0.48) or
to the mean individual-level proportion of overall hypnotic use
that was MOTN (44.4% [3.4] with a rule vs. 48.6% [10.3] without a rule; t = 0.4, p = 0.70).
Prevalence of MOTN Hypnotic Use
While 79.8% (1.1) of Part I respondents, representing all hypnotic users, reported that they were exclusive bedtime users, the
other 20.2% (1.1) reported being MOTN users (Table 2). The
latter include 11.2% (0.8) once-per-night MOTN users (2.1%
[0.4] exclusively MOTN and 9.1% [0.8] sometimes at bedtime
and other times MOTN) and 9.0% (0.8) twice-per-night MOTN
users. As noted above in the section on analysis methods, the
parenthetical entries to the right of the prevalence estimates are
the standard errors of these estimates.
The health plan from which we selected the survey sample
reported that 2.6% of members in the 18- to 64-year age range
had a prescription sleep medication at some time in the 12
months before the survey. If we assume provisionally that this
rate applies to the total US population in the age range of the
sample and that the sample estimate that 20.2% of prescription
hypnotic users use MOTN applies equally to other hypnotic users in the US population, then the total of such MOTN users in
the population ages 18-64 would be approximately 1 million
664
Table 3—Sociodemographic and treatment profiles of exclusive bedtime users, once-per-night MOTN users, and twice-per-night
MOTN users in the Part I sample (n = 1,927)
Age, mean
Female, %
Specialty of prescribing physician, %
Primary care*
Sleep medicine
Psychiatrist
Pain specialist
All others
Number of years of hypnotic use, mean
Number of nights used in the…
Past month, mean
Past 12 months, mean
n‡
MOTN users
Exclusive
bedtime users
Est (SE)
47.5 (0.3)
62.0 (1.5)
Once per night
Est (SE)
49.2 (0.8)
61.2 (3.9)
Twice per night
Est (SE)
47.3 (1.1)
51.7 (4.6)
Total
Est (SE)
47.7 (0.3)
61.0 (1.3)
F2,1924
2.0
4.9
73.0 (1.3)
1.8 (0.4)
11.1 (0.9)
6.5 (0.7)
7.6 (0.7)
4.3 (0.1)
73.4 (3.6)
1.7 (1.0)
11.4 (2.8)
6.5 (1.6)
7.0 (2.1)
5.5 (0.5)
69.8 (4.1)
1.3 (0.8)
14.8 (3.2)
8.1 (2.3)
6.0 (2.2)
6.2 (0.5)
72.8 (1.2)
1.7 (0.4)
11.5 (0.9)
6.6 (0.6)
7.4 (0.7)
4.6 (0.1)
0.6
0.2
1.4
0.6
0.4
9.8†
12.4 (0.4)
155.0 (4.1)
1,546
9.2 (0.8)
105.6 (8.4)
207
18.1 (1.2)
228.6 (13.1)
174
12.6 (0.3)
156.2 (3.7)
1,927
21.5†
32.6†
*Including reports of primary care, family practice, internal medicine, and OBGYN. †Significant difference across the three subsamples at the 0.05 level, twosided test. ‡Unweighted sample sizes.
Table 4—Distribution of sleep problems and proportional once-per-night MOTN medication use by type of sleep problem in the
Part II sample* (n = 510)
Difficulty initiating sleep (DIS)
Difficulty maintaining sleep (DMS)
Early morning awakening (EMA)
Non-restorative sleep (NRS)
Any of 4 sleep problems
Reported as a problem
% (SE)
53.7 (3.0)
59.4 (3.0)
47.3 (3.0)
63.8 (2.9)
81.2 (2.5)
Reported as the
worst problem
% (SE)
46.1 (3.0)
30.1 (2.7)
5.3 (1.2)
17.4 (2.1)
98.9 (1.7)†
Once-per-night MOTN
use among people
with this problem
% (SE)
10.1 (1.4)
11.8 (1.4)
14.3 (1.8)
11.3 (1.3)
12.1 (1.2)
Once-per-night MOTN use
among people with this
most bothersome problem
% (SE)
8.6 (1.7)
15.4 (2.5)
20.6 (7.0)
8.4 (2.1)
17.5 (2.6)
*The sample includes 303 exclusive bedtime users (weighted up by 5 to reflect the under-sampling of such cases in the Part II survey) plus the 207 once-pernight MOTN users. †The remaining respondents said that they could not pick any one worst sleep problem.
Nighttime Sleep Problems among Exclusive Bedtime
versus Once-per-Night MOTN Users
other nighttime sleep problems (10.1% to 11.8%; Table 4).
The situation is somewhat different in the subsamples of respondents who reported specific sleep problems as their most
bothersome, where the highest proportions with once-per-night
MOTN use are among those whose most bothersome problems
are either EMA (20.6%) or DMS (15.4%). The proportions of
once-per-night MOTN use are much lower among those whose
most bothersome problems are DIS (8.6%) or NRS (8.4%).
Types of sleep problems were assessed only in the Part II
sample. NRS was the most commonly reported nighttime sleep
problem (reported by 63.8% of Part II respondents), followed
by DMS (59.4%), DIS (53.7%), and EMA (47.3%; Table 4)
The sum of these 4 percentages is greater than 100%, which
means that the typical respondent with sleep problems had
more than one symptom. However, only 81.2% of respondents
reported any of these 4 symptoms, the remaining 18.8% reporting that feeling tired during the day was their only sleep problem. DIS was reported as the most bothersome sleep problem
by the largest proportion of respondents (46.1%) followed by
DMS (30.1%), NRS (17.4%), and EMA (5.3%).
The proportion of respondents in Part II of the sample who
are once-per-night MOTN users is 12.3%. (This is higher than
the 11.2% in Table 2 because twice-per-night MOTN users
were included in the Table 1 calculation but are excluded in
Part II of the sample.) This proportion is higher among respondents in the Part II sample who reported EMA (14.3%) than
Predictors of MOTN Use
A logistic regression analysis was carried out among Part II
respondents to examine significant predictors of once-per-night
MOTN use from among the variables considered previously.
Respondents who reported that DMS was their most bothersome sleep problem had a significantly elevated OR (95% CI)
of once-per-night MOTN use (1.9 [1.2-3.1], p = 0.011), indicating that those for whom DMS was most bothersome were
nearly twice as likely as others to use once-per-night MOTN
rather than exclusively at bedtime (Table 5). Number of years
since starting hypnotic use also had a significantly elevated
665
duction that PRN dosing options could lead to a reduction in
nightly hypnotic use among patients with a primary concern
about DMS.9 However, the fact that once per month MOTN
users have a longer duration of use than exclusive bedtime users might indicate an opposite long-term effect. Causal interpretations of these naturalistic associations are inappropriate,
though, and can only be confirmed by controlled studies.
That twice-per-night MOTN users had much higher numbers of uses (an average of 18.1 nights/month compared to
12.4 nights/month among exclusive bedtime users) might be
due to them having more complex insomnia (e.g., high rates
of both DIS and DMS). Alternatively, one might speculate that
twice-per-night MOTN users take hypnotics at bedtime with
the intent of being able to sleep through the night, and when
they awaken, despite having taken the medication at bedtime,
they re-medicate. This possibility is consistent with our finding
that twice-per-night MOTN users have been taking hypnotics
for a longer duration than exclusive bedtime users. Although
laboratory studies suggest that tolerance and dose escalation is
not a significant issue with hypnotics,23 it may be that long-term
users habituate to the sedative effects of prescription hypnotics
and experience “breakthrough” sleep maintenance symptoms,
which then need a second medication dosing for adequate coverage. Such speculations, however, extend beyond the data
available here.
We found that the vast majority of once-per-night MOTN
users switch between bedtime use and MOTN use, with MOTN
use occurring much less often than bedtime use (on average of
2.8 MOTN uses/month compared to an average of 6.4 bedtime
uses/month). Frequency of MOTN use among patients who use
twice a night was not determined, as they were not included in
Part II of the survey. However, as twice-per-night MOTN dosing occurs much more frequently than once-per-night MOTN
dosing (an average of 18.1 in the past 30 days among twiceper-night MOTN users compared to 9.2 among once-per-night
MOTN users) and are almost as numerous as once-per-night
MOTN users (9.0% vs. 11.2% of all hypnotic users), it is not
implausible that the rate of overall MOTN use among twiceper-night MOTN users in the age range of the sample might
accumulate to twice that of once-per-night MOTN users. This
would put the total annual number of MOTN uses in this age
range in the country as a whole at well over 50 million if we
assumed that sample estimates apply to the total population and
that the proportion of the population using prescription hypnotics is consistent with previous national estimates.30-32
This high estimated rate of off-label MOTN use is perhaps expectable in light of broader evidence that insomniacs
frequently use alcohol, over-the-counter medications, and a
variety of prescription medications other than hypnotics to selfmedicate their sleep problems,33 along with evidence that sleep
maintenance insomniacs are particularly prone to self-medication.34 We found that only a small proportion of once-per-night
MOTN users (14.0%) reported that their MOTN use was on
the advice of a physician, although this physician advice was
reported by patients to be much more common among onceper-night MOTN users treated by a sleep medicine specialist
or psychiatrist than by other practitioners. Being mindful that
this result is based on patient self-report, it is possible that specialists in sleep medicine and psychiatry are more sophisticated
Table 5—Predictors of once-per-night MOTN medication
user versus exclusive bedtime use in the Part II sample*
(n = 510)
Number of nights with DMS in a typical week in the
absence of medication
OR (95% CI)
0.8 (0.7-0.9)†
Number of minutes awake on a typical DMS night
Number of times awakening/night on a typical DMS
night
Number of nights with EMA in a typical week in the
absence of medication
Number of years since first starter to take hypnotics
DMS reported as most bothersome sleep problem
1.0 (0.9-1.1)
0.9 (0.8-1.0)
1.1 (1.0-1.2)
1.1 (1.0-1.1)†
1.9 (1.2-3.1)†
*Based on a multivariate logistic regression equation comparing
n = 207 once-per-night MOTN users with n = 303 exclusive bedtime
users.†Significant at the 0.05 level, two-sided test.
OR (95% CI) with once-per-night MOTN use (1.1 [1.0-1.1],
p = 0.014). Number of nights per week respondents typically
experienced nighttime awakenings in the absence of medication, in comparison, was inversely related to once-per-night
MOTN use (0.8 [0.7-0.9], p = 0.002). The other predictors considered in the analysis (number of nights per week with early
morning waking, number of nighttime awakenings on nights
when they occur, mean overall time awake at night) were all
insignificant (χ21 = 0.1-2.5, p = 0.11-0.69).
DISCUSSION
No previous research exists on the epidemiology of MOTN
use in the general population. Such research is warranted,
though, given the high prevalence and persistence of sleep
maintenance insomnia and the suspicion that off-label MOTN
use is common. The current results are the first systematic
large-scale survey data to estimate the prevalence of MOTN
hypnotic use in any detail. Although the external validity of
findings is limited by a low response rate and the fact that the
sample was restricted to insured people in the age range 18-64,
results nonetheless provide a preliminary view of real-world
MOTN hypnotic use patterns in the community.
Within the context of these sample constraints, the results
suggest that approximately one-fifth of hypnotic users in the
age range of 18 to 64 years use hypnotics off-label in the middle
of the night to resume sleep. Nearly half of these MOTN users take hypnotics twice in the same night. The data also suggest that once-per-night and twice-per-night MOTN users are
quite different, in that the former use hypnotics significantly
less frequently than exclusive bedtime users while the latter use
hypnotics significantly more frequently than exclusive bedtime
users. We were unable to make more detailed comparisons of
once-per-night versus twice-per-night MOTN users because the
latter were excluded from Part II of the survey, but we were
able to compare once-per-night MOTN users with exclusive
bedtime users. The fact that once-per-night MOTN users take
hypnotics at bedtime less often than exclusive bedtime users
(averages of 6.4 nights/month vs. 12.4 nights/month, respectively) is indirectly consistent with the suggestion in the introJournal of Clinical Sleep Medicine, Vol. 9, No. 7, 2013
666
than other practitioners regarding sleep psychopharmacology,
more familiar with newer short-acting hypnotic agents, and
more comfortable prescribing them for MOTN use. However,
such speculations are beyond the scope of the current data. One
thing is clear, though, regarding the implications of the overall
low rate of physician recommendation in light of the potential
adverse effects of off-label MOTN hypnotic use: that prescribing physicians should routinely ask patients with hypnotic prescriptions about possible MOTN use and caution them against
off-label MOTN use. Although we are aware of no controlled
studies on the effects of such an intervention, experimental
studies of the basic psychological processes underlying treatment adherence suggest that physician efforts to help patients
understand the rationale for discouraging off-label hypnotic use
could lead to substantial reductions.35
Comparison of once-per-night MOTN users with exclusive
bedtime users found only three significant correlates: DMS as
a most bothersome sleep problem, long duration of hypnotic
use, and low weekly frequency of DMS. The first two of these
three associations are easily interpreted, as we might expect patients to be more aggressive in self-medicating problems they
consider most bothersome and as they become more familiar
with medication effects over time. It is somewhat more difficult to understand the finding that frequency of DMS is lower
among MOTN users than exclusive bedtime users. This might
be a chance finding in the many comparisons made here, or suggest either an especially high rate of habituation among chronic
maintenance insomniacs or a lower severity threshold for selfmedication among MOTN users than other hypnotic users, perhaps due to the intermittent character of symptoms. Evidence
consistent with the possibility of lower symptom tolerance has
been reported in a study of predictors of sham self-medication,34
but future research is needed to determine the extent to which
this accounts for the association of MOTN use with low DMS
frequency. Future research is also needed to examine other predictors of off-label MOTN use, such as the presence and severity of comorbid physical and mental disorders.
11. Farber RH, Burke PJ. Post-bedtime dosing with indiplon in adults and the elderly: results from two placebo-controlled, active comparator crossover studies
in healthy volunteers. Curr Med Res Opin 2008;24:837-46.
12. Zammit G, Wang-Weigand S, Rosenthal M, Peng X. Effect of ramelteon on
middle-of-the-night balance in older adults with chronic insomnia. J Clin Sleep
Med 2009;5:34-40.
13. Zammit GK, Corser B, Doghramji K, et al. Sleep and residual sedation after
administration of zaleplon, zolpidem, and placebo during experimental middleof-the-night awakening. J Clin Sleep Med 2006;2:417-23.
14. Roth T, Hull SG, Lankford DA, Rosenberg R, Scharf MB. Low-dose sublingual
zolpidem tartrate is associated with dose-related improvement in sleep onset
and duration in insomnia characterized by middle-of-the-night (MOTN) awakenings. Sleep 2008;31:1277-84.
15. Roth T, Mayleben D, Corser BC, Singh NN. Daytime pharmacodynamic and
pharmacokinetic evaluation of low-dose sublingual transmucosal zolpidem
hemitartrate. Hum Psychopharmacol 2008;23:13-20.
16. Leufkens TR, Lund JS, Vermeeren A. Highway driving performance and cognitive functioning the morning after bedtime and middle-of-the-night use of gaboxadol, zopiclone and zolpidem. J Sleep Res 2009;18:387-96.
17. Verster JC, Veldhuijzen DS, Patat A, Olivier B, Volkerts ER. Hypnotics and driving safety: meta-analyses of randomized controlled trials applying the on-theroad driving test. Curr Drug Saf 2006;1:63-71.
18. Verster JC, Volkerts ER, Olivier B, Johnson W, Liddicoat L. Zolpidem and traffic
safety - the importance of treatment compliance. Curr Drug Saf 2007;2:220-6.
19. Logan BK, Couper FJ. Zolpidem and driving impairment. J Forensic Sci
2001;46:105-10.
20. Gustavsen I, Al-Sammurraie M, Morland J, Bramness JG. Impairment related
to blood drug concentrations of zopiclone and zolpidem compared to alcohol in
apprehended drivers. Accid Anal Prev 2009;41:462-6.
21. Jones AW, Holmgren A, Kugelberg FC. Concentrations of scheduled prescription
drugs in blood of impaired drivers: considerations for interpreting the results.
Ther Drug Monit 2007;29:248-60.
22. Pressman MR. Sleep driving: sleepwalking variant or misuse of z-drugs? Sleep
Med Rev 2011;15:285-92.
23. Mendelson WB, Roth T, Cassella J, et al. The treatment of chronic insomnia:
drug indications, chronic use and abuse liability. Summary of a 2001 New Clinical Drug Evaluation Unit meeting symposium. Sleep Med Rev 2004;8:7-17.
24. Kessler RC, Coulouvrat C, Hajak G, et al. Reliability and validity of the brief insomnia questionnaire in the America insomnia survey. Sleep 2010;33:1539-49.
25. Rosenbaum PR, Rubin DB. The central role of propensity scores in observational studies for causal effects. Biometrika 1983;70:41-55.
26. Wolter KM. Introduction to variance estimation. New York: Springer-Verlag,
1985.
27. SAS Institute Inc. SAS/STAT software, version 9.2 for Unix. Cary, NC: SAS Institute Inc., 2008.
28. National Center for Chronic Disease Prevention and Health Promotion. Behavioral risk factor survey surveillance system. BRFSS annual survey data. Summary
data quality reports. Atlanta: Centers for Disease Control and Prevention, 2009.
29. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with
ICD-9-CM administrative databases. J Clin Epidemiol 1992;45:613-9.
30. Balter MB, Uhlenhuth EH. The beneficial and adverse effects of hypnotics. J Clin
Psychiatry 1991;52 Suppl:16-23.
31. Mellinger GD, Balter MB, Uhlenhuth EH. Insomnia and its treatment. Prevalence
and correlates. Arch Gen Psychiatry 1985;42:225-32.
32. Patten SB, Williams JV, Lavorato DH, Kassam A, Sabapathy CD. Pharmacoepidemiology of benzodiazepine and sedative-hypnotic use in a Canadian general
population cohort during 12 years of follow-up. Can J Psychiatry 2010;55:792-9.
33. National Institute of Health NIH state-of-the-science conference statement on
manifestations and management of chronic insomnia in adults. NIH Consens
State Sci Statements 2005;22:1-30.
34. Roehrs T, Bonahoom A, Pedrosi B, Rosenthal L, Roth T. Disturbed sleep predicts
hypnotic self-administration. Sleep Med 2002;3:61-6.
35. Meichenbaum D, Turk DC. Facilitating treatment adherence. New York: Plenum,
1987.
REFERENCES
1. Lichstein KL, Taylor DJ, McCrae CS, Ruiter ME. Insomnia: epidemiology and risk
factors. In: Kryger MH, Roth T, Dement WC, eds. Principles and practice of sleep
medicine, 5th ed. St. Louis: Saunders, 2011:827-37.
2. Walsh JK, Coulouvrat C, Hajak G, et al. Nighttime insomnia symptoms and perceived health in the America Insomnia Survey (AIS). Sleep 2011;34:997-1011.
3. Ohayon MM. Nocturnal awakenings and comorbid disorders in the American
general population. J Psychiatr Res 2008;43:48-54.
4. Morin CM, Belanger L, LeBlanc M, et al. The natural history of insomnia: a
population-based 3-year longitudinal study. Arch Intern Med 2009;169:447-53.
5. Bolge SC, Joish VN, Balkrishnan R, Kannan H, Drake CL. Burden of chronic
sleep maintenance insomnia characterized by nighttime awakenings. Popul
Health Manag 2010;13:15-20.
6. Hara C, Stewart R, Lima-Costa MF, et al. Insomnia subtypes and their relationship to excessive daytime sleepiness in Brazilian community-dwelling older
adults. Sleep 2011;34:1111-7.
7. Neubauer DN. The evolution and development of insomnia pharmacotherapies.
J Clin Sleep Med 2007;3:S11-5.
8. Rosenberg RP. Sleep maintenance insomnia: strengths and weaknesses of current pharmacologic therapies. Ann Clin Psychiatry 2006;18:49-56.
9. Roth T, Zammit GK, Scharf MB, Farber R. Efficacy and safety of as-needed, post
bedtime dosing with indiplon in insomnia patients with chronic difficulty maintaining sleep. Sleep 2007;30:1731-8.
10. Phillips BA. Insomnia, hypnotic drug use, and patient well-being: first, do no
harm. Mayo Clin Proc 2012;87:417-8.
ACKNOWLEDGMENTS
The research reported here was funded by Transcept Pharmaceuticals. The study
was designed by Kessler and Roth. The survey instrument was developed by Kessler,
Roth, Shahly, and Shillington. Shahly and Shillington supervised data collection. All
coauthors collaborated in planning data analyses and interpreting results. Kessler
supervised data analyses and Berglund carried out the analyses. Roth and Shahly
prepared the first draft of the manuscript and all coauthors collaborated in making
667
revisions. All coauthors are fully responsible for content and editorial decisions. Transcept played no role in data collection or management other than in posing the initial
research question and providing operational and financial support. Transcept played
no role in data analysis, interpretation of results, or preparation of the manuscript. The
authors thank Marcus Wilson and his staff at HealthCore, Inc. for use of the HealthCore research environment and calculation of population rates of sleep medication
use, the staff of EPI-Q, Inc. for study management, and Marielle Weindorf and her
staff at DataStat, Inc. for carrying out the survey.
Neurocrine, Pfizer, Sanofi, SchoeringPlough, Sepracor, Somaxon, Syrex, Takeda,
Transcept, Wyeth, and Xenoport. Ms. Berglund has participated in research activities supported by Dr. Kessler’s grants, which have included research funding from
Pfizer, Sanofi Aventis, Shire, and Janssen. Ms. Berglund has no financial interest in
any of these organizations. Dr. Shahly is an employee of the Department of Health
Care Policy at Harvard Medical School. That program has received research funding from Pfizer, Sanofi Aventis, Shire Development, Inc., and Janssen Pharmceutica,
N.V. Dr. Shahly has no financial interest in any of these organizations. Dr. Shillington
is an employee and shareholder in the company Epi-Q, Inc. Epi-Q provides project
management service and has received grant and consulting support from the following companies: Merck, Cephalon, Sanofi-Aventis, Pfizer, Biogen-IDEC, Onconova,
AstraZeneca, Glaxo Smith Kline, Novartis, Roche, Ortho McNeil, Takada, Transcept,
Lundbeck Genentech, Bayer, Baxter, and Abbott. Dr. Shillington receives no direct
compensation as a result of grants or contracts, other than her salary from Epi-Q.
Dr. Stephenson is an employee of HealthCore, Inc., a research and consulting organization. All of her research activities are industry-sponsored. However, she receives
no direct compensation as a result of grants or contracts, other than her salary from
Health Core. Dr. Kessler has been a consultant for AstraZeneca, Analysis Group,
Bristol-Myers Squibb, Cerner-Galt Associates, Eli Lilly & Company, GlaxoSmithKline
Inc., HealthCore Inc., Health Dialog, Integrated Benefits Institute, John Snow Inc.,
Kaiser Permanente, Matria Inc., Mensante, Merck & Co, Inc., Ortho-McNeil Janssen Scientific Affairs, Pfizer Inc., Primary Care Network, Research Triangle Institute, Sanofi-Aventis Groupe, Shire US Inc., SRA International, Inc., Takeda Global
Research & Development, Transcept Pharmaceuticals Inc., and Wyeth-Ayerst; has
served on advisory boards for Appliance Computing II, Eli Lilly & Company, Mindsite,
Ortho-McNeil Janssen Scientific Affairs, Plus One Health Management and WyethAyerst; and has had research support for his epidemiological studies from Analysis
Group Inc., Bristol-Myers Squibb, Eli Lilly & Company, EPI-Q, GlaxoSmithKline, Johnson & Johnson Pharmaceuticals, Ortho-McNeil Janssen Scientific Affairs., Pfizer Inc.,
Sanofi-Aventis Groupe, and Shire US, Inc.
Submitted for publication June, 2012
Submitted in final revised form November, 2012
Address correspondence to: Ronald C. Kessler, Ph.D., Department of Health Care
Policy, Harvard Medical School, 180 Longwood Ave., Boston, MA 02115; Tel: (617)
432-3587; Fax: (617) 432-3588; E-mail: kessler@hcp.med.harvard.edu
The research reported here was funded by Transcept Pharmaceuticals and the
work was performed at EPI-Q, Inc. Dr. Roth has served as a consultant for Abbott,
Accadia, Acogolix, Acorda, Actelion, Addrenex, Alchemers, Alza, Ancel, Arena, AstraZenca, Aventis, AVER, Bayer, BMS, BTG, Cephalon, Cypress, Dove, Eisai, Elan, Eli
Lilly, Evotec, Forest, Glaxo Smith Kline, Hypnion, Impax, Intec, Intra-Cellular, Jazz,
Johnson and Johnson, King, Lundbeck, McNeil, MediciNova, Merck, Neurim, Neurocrine, Neurogen, Novartis, Orexo, Organon, Otsuka, Prestwick, Proctor and Gamble,
Pfizer, Purdue, Resteva, Roche, Sanofi, SchoeringPlough, Sepracor, Servier, Shire,
Somaxon, Syrex, Takeda, TransOral, Yanda, Vivometrics, Wyeth, Yamanuchi, and
Xenoport. He has served on speakers bureau for Purdue and Sepracor. He has received research support from Apnex, Aventis, Cephalon, Glaxo Smith Kline, Merck,
668
Association between Sleep Duration and the Mini-Mental
Score: The Northern Manhattan Study
Alberto R. Ramos, M.D., M.S.P.H.1; Chuanhui Dong, Ph.D.1; Mitchell S. V. Elkind, M.D., M.S.2,3; Bernadette Boden-Albala, M.P.H., Dr.P.H.4;
Ralph L. Sacco, M.D., M.S.1,5,6; Tatjana Rundek, M.D., Ph.D.1,5; Clinton B. Wright, M.D., M.S.1,5,7
Evelyn F. McKnight Brain Institute, Department of Neurology, Miller School of Medicine University of Miami, Miami, FL;
Department of Neurology, College of Physicians and Surgeons, Columbia University, New York, NY; 3Department of Epidemiology,
Mailman School of Public Health, Columbia University, New York, NY; 4Division of Social Epidemiology, Department of Health
Policy, Mount Sinai School of Medicine, New York, NY; 5Department of Epidemiology and Public Health, University of Miami, Miami,
FL; 6Department of Human Genetics, University of Miami, Miami, FL; 7Neuroscience Program, University of Miami, Miami, FL
1
2
Background: Short and long sleep duration are associated
with increased mortality and worse global cognitive function,
but is unclear if these relations persist after accounting for the
risk of sleep disordered breathing (SDB). The aim of our study
is determine the association between short and long sleep
duration with worse global cognitive function in a racially/ethnically diverse elderly cohort.
Methods: We examined sleep hours and global cognitive
function cross-sectionally within the population-based Northern Manhattan Study cohort. We conducted nonparametric
and logistic regression to examine associations between
continuous, short (< 6 h) and long (≥ 9 h) sleep hours with
performance on the Mini Mental State Examination (MMSE).
Results: There were 927 stroke-free participants with data
on self-reported sleep hours and MMSE scores (mean age
75 ± 9 years, 61% women, 68% Hispanics). The median (interquartile range) MMSE was 28 (10-30). Sleep hours (centered at 7 h) was associated with worse MMSE (β = -0.01;
SE [0.004], p = 0.0113) adjusting for demographics, vascular
risk factors, medications, and risk for SDB. Reporting long
sleep (≥ 9 h) compared to 6 to 8 h of sleep (reference) was
significantly and inversely associated with MMSE (adjusted β = -0.06; SE [0.03], p = 0.012), while reporting short
sleep was not significantly associated with MMSE performance. Long sleep duration was also associated with low
MMSE score when dichotomized (adjusted OR: 2.4, 95%
CI: 1.1-5.0).
Conclusion: In this cross-sectional analysis among an elderly community cohort, long sleep duration was associated with
worse MMSE performance.
Keywords: Sleep duration, cognition, short sleep, long sleep,
cognitive impairment, mini mental score
Citation: Ramos AR; Dong C; Elkind MSV; Boden-Albala B;
Sacco RL; Rundek T; Wright CB. Association between Sleep
Duration and the Mini-Mental Score: The Northern Manhattan
Study. J Clin Sleep Med 2013;9(7):669-673.
C
ognitive impairment and dementia are disabling conditions
expected to rise in prevalence with the rapidly aging population.1,2 The identification of modifiable risk factors for cognitive impairment can provide important prevention strategies with
significant public health implications. The impact of inadequate
sleep on cognition can be profound. Besides producing sleepiness, which has detrimental effects on mood, job performance,
and accident risk, poor sleep is associated with adverse health
outcomes.3,4 This is particularly relevant to the aging population
as sleep-wake patterns and sleep quality may change throughout
the lifetime, with 50% of the elderly reporting sleep disturbances,
and up to one-third reporting either short sleep or long sleep duration.5,6 Abnormal sleep duration may impair attention/vigilance
and cause executive dysfunction,6,7 but it is unclear if these relationships persist after accounting for the risk of sleep disordered
breathing (SDB). SDB is highly prevalent in the elderly, seen in
up to 62% of those older than 65 years of age and is associated
with poor cognitive function.8-11 Determining the relationship between sleep duration and cognition could lead to novel strategies
to improve health as sleep duration is potentially modifiable.12
The aim of this analysis is to evaluate the association between self-reported sleep hours and short and long sleep dura-
Brief Summary
Current Knowledge/Study Rationale: Cognitive impairment and dementia are expected to rise with the aging population. There is limited
data from elderly race/ethnically diverse cohorts with evaluations of
sleep duration and cognitive function. Determining the relation between
sleep duration and cognitive function could lead to novel strategies to
improve health.
Study Impact: Long sleep duration was associated with worse minimental score, a measure of global cognitive function, after adjusting
for demographic, vascular risk factors and depressive symptoms. The
results of this study suggest that long sleep duration may be an independent predictor of worse cognitive function in the elderly.
tion with worse global cognitive function in an elderly racially/
ethnically diverse population-based cohort.
METHODS
Study Population
The Northern Manhattan Study (NOMAS) enrolled 3,298
stroke-free participants randomly sampled from the Northern
669
IR Ramos, C Dong, MSV Elkind et al
Manhattan population between 1993 and 2001 using the following criteria: (1) resident of Northern Manhattan ≥ 3 months;
(2) from a household with a telephone; (3) age ≥ 40 years at the
time of first in-person assessment; and (4) no history of stroke.13
For the purpose of this analysis, we included participants with
self-reported sleep hours obtained during annual telephone
follow-up evaluation in 2006 and Mini-Mental Examination
(MMSE) scores within one year of reported sleep hours. From
the parent cohort, a total of 2,266 subjects were available for
follow up in 2006. Of the available sample, a total of 927 participants had reports of sleep hours and MMSE within one year
of each assessment. The sample of 927 participants had a similar proportion of women (61%), but a greater proportion of Hispanics (68%) compared to the overall baseline cohort (53%).
NOMAS was approved by the Columbia University Medical
Center and University of Miami, Miller School of Medicine
IRBs, and all participants provided written informed consent.
following items was used to classify participants into high risk
for SDB: (1) frequent snoring (snoring > 3 times per week), (2)
daytime sleepiness (sum score ≥ 10), and (3) presence of hypertension or obesity (BMI > 30 kg/m2).
Risk Factor Assessments
Data were collected through interviews by trained bilingual
research assistants using standardized data collection instruments described elsewhere.13 Race and ethnicity were defined
by self-identification based on questions modeled after the US
census. Race/ethnicity were categorized into mutually exclusive
groups as non-Hispanic Black, non-Hispanic White, and Hispanic. Depressive symptoms were evaluated with the Center for
Epidemiological Studies Depression scale (CES-D). The CESD is a 20-item scale documenting 4 factors: depressive affect,
somatic complaints, positive affect, and interpersonal relations.
Scores on the CES-D range from 0 to 60, with higher scores indicating more symptoms of depression.22 Depressive symptoms
were categorized as present if the sum of the scores was ≥ 16
or if the participant was taking an antidepressant medication.22
Hypertension was defined as a systolic blood pressure > 140
mm Hg or a diastolic blood pressure > 90 mm Hg or a patient’s
self-report of a history of hypertension or use of antihypertensive medications. Diabetes mellitus was defined as fasting blood
glucose ≥ 126 mg/dL or the patient’s self-report of diabetes or
use of insulin or hypoglycemic medications. Cardiac disease
included history of angina, MI, coronary artery disease, atrial
fibrillation, congestive heart failure, or valvular heart disease.
We obtained self-reported medication use at baseline. We
created a dichotomous variable (yes vs. no) based on the use
of the following medications: antidepressant, antiepileptic,
pain, and antipsychotic, that could affect sleep duration and
or cognitive function.17
Cognitive Assessment
Cognitive status was assessed in person by bilingual (English or Spanish) trained research assistants using MMSE.14 The
MMSE is a brief 30-point questionnaire test used to evaluate
cognitive function. The MMSE measures various domains of
cognitive functioning including memory, orientation to place
and time, naming, reading, visuospatial orientation/construction
ability, writing, and the ability to follow a 3-stage command.
It has good sensitivity (71% to 92%) and specificity (56% to
96%) to screen for cognitive impairment and dementia.14
We used the total score of the MMSE as the outcome. Lower
educational levels (≤ 8th grade) can adversely affect the MMSE
scores. We defined low MMSE scores as a dichotomous outcome by adjusting for age and educational level based on established MMSE cutoffs. A cutoff of MMSE < 24 was used for
those with > 8 years education and MMSE < 20 for those with
≤ 8 years of education.15,16
Results are presented as mean ± standard deviation or median
(interquartile range) for continuous variables according to the
variable distribution, and proportion for categorical variables.
The χ2 test was used to compare proportions, while ANOVA,
or the Kruskal-Wallis test if data were not normally distributed,
was used to compare mean or median for continuous variables.
We examined self-reported sleep hours in categories of short
sleep (< 6 h) and long sleep (≥ 9 h), with 6 to 8.9 h of nightly
sleep as the reference.23 We performed nonparametric regression to test for the association between sleep hours centered at
7 h of sleep continuously and comparing categories of < 6 h and
≥ 9 h versus the reference with the non-normally distributed
MMSE using SAS procedure COUNTREG. Sequential models were done to evaluate the unadjusted association between
sleep hours (centered at 7 h) and MMSE. We then adjusted for
demographic factors: age, sex, education, race/ethnicity, and insurance status (Model 2); alcohol consumption, hypertension,
diabetes, depression, medications, and risk for SDB (Model 3).
Logistic regression was performed with the categories for low
MMSE score as the outcome. As a sensitivity analysis, we also
evaluated the relation between sleep hours and memory performance on the MMSE, given that verbal 3-word recall on the
MMSE has been reported as an acceptable estimate of episodic
memory in epidemiologic studies.24 Among participants able
Sleep Hours
We collected self-reported sleep duration as an estimate of
hours of nightly sleep in the four weeks prior to the annual telephone follow-up interview in 2006, using the following question: “During the past 4 weeks, how many hours, on average,
did you sleep each night?” Respondents reported in 30-min increments of each hour.17 The responses ranged from 3 to 12 h of
sleep with a median of 7 hours.
High Risk for Sleep Disordered Breathing
High risk for SDB was estimated by constructing the Berlin
questionnaire,18 based on reports of frequent snoring and daytime sleepiness along with objective information on hypertension and obesity in our sample. Sleep symptoms were derived
from a sleep questionnaire during follow-up examination in
2004-2005.17 The questionnaires were administered in English
or Spanish. Habitual snoring was defined as self-report of snoring > 3 times per week, based on prior definitions of habitual
snoring.19 The Epworth Sleepiness Scale was used and adapted
for relevance to characteristics of people living in northern
Manhattan.20 Daytime sleepiness was categorized as sum score
≥ 10 based on the established definition for daytime sleepiness
from the Epworth Sleepiness Scale.21 The presence of 2 of the 3
670
Sleep Duration and the Mini-Mental Score
Table 1—Demographic and vascular risk factors and cognitive scores across categories of sleep hours
Mean ± SD or N (%) or as indicated
Demographic
Age, years
Women
≤ 8th grade education
Medicaid or no insurance
Total (N = 927)
< 6 h (n = 224, 24%) 6-8.9 h (n = 616, 66%)
≥ 9 h (n = 87, 9%)
75 ± 9
567 (61)
381 (41)
302 (33)
74 ± 8
144 (64)
100 (45)
68 (30)
74 ± 9
371 (60)
241 (39)
199 (32)
77 ± 9*
52 (60)
40 (46)
35 (40)
Race-Ethnicity
Non-Hispanic White
Non-Hispanic Black
Hispanic
127 (14)
163 (18)
615 (68)
25 (11)
42 (19)
152 (69)
93 (15)
104 (17)
405 (67)
9 (11)
17 (20)
58 (69)
Risk Factors
BMI (kg/m2)
Moderate alcohol
Current smoking
Depression
Hypertension
Diabetes
Cardiac disease
Medication£
High risk SDB
28.2 ± 4.7
363 (39)
155 (17)
185 (20)
634 (68)
131 (14)
163 (18)
216 (23)
245 (26)
28.4 ± 5.0
69 (31)
32 (14)
53 (24)
140 (63)
34 (15)
37 (17)
63 (28)
54 (24)
28.1 ± 4.7
259 (42)
104 (17)
111 (18)
429 (70)
75 (12)
105 (17)
131 (21)
166 (27)
27.5 ± 4.7
35 (41)*
19 (22)
21 (24)
65 (75)*
22 (25)**
21 (24)
22 (25)
25 (29)
28 (3)
18 (8)
28 (4)
59 (10)
26 (6)***
16 (18)*
Mini-Mental Score (MMSE)
MMSE, median (interquartile range)
Low MMSE score (Based on age/education cutoff)
28 (10-30)
93 (10)
SDB, sleep disordered breathing. *p < 0.05, **p < 0.001, ***p < 0.0001. £Antidepressant, antiepileptic, pain, and/or antipsychotic medication.
Table 2—Association between sleep hours and Mini-Mental Score Examination
Model 1
Sleep hours
Centered at 7 h
Categorical
<6h
6-8.9 h
≥9h
Model 2
Model 3
β (SE) per hour
-0.01 (0.004)
p
0.0180
β (SE) per hour
-0.009 (0.004)
P
0.0393
0.005 (0.02)
Reference
-0.07 (0.02)
0.74
–
0.0012
0.01 (0.02)
–
-0.05 (0.02)
0.46
–
0.0187
β (SE) per hour
-0.01 (0.004)
0.01 (0.02)
–
-0.06 (0.03)
p
0.0113
0.39
–
0.0120
Model 1: univariate. Model 2: adjusted for age, sex, race-ethnicity, education, and Medicaid or no insurance status. Model 3: adjusted for covariates in model
2 and reported alcohol consumption, depression, diabetes mellitus, hypertension, high risk for SDB, and medications. SE, standard error.
to register 3 initial words, impaired verbal recall was defined
by a score of 0 or 1 obtained on the subsequent 3-word recall
task of the MMSE.24 Additionally, we evaluated the interactions between sleep hours and the covariates. All analyses were
performed using SAS software version 9.3 (SAS Institute Inc,
Cary, NC).
The covariates that were associated with the lower MMSE
scores were increased age (β = -0.004; p < 0.0001), ≤ 8th grade
education (β = -0.14; p < 0.0001), having Medicaid or no insurance (β = -0.14; p < 0.0001), Hispanic race/ethnicity (β = -0.09,
p < 0.0001) compared to non-Hispanic white, depression
(β = -0.05, p = 0.0029), diabetes (β = -0.05, p = 0.012), hypertension (β = -0.05, p = 0.0007), and medications (β = -0.08,
p < 0.0001). Male sex (β = 0.04; p = 0.0065) and moderate alcohol consumption alcohol (β = 0.04, p = 0.009) were positively
associated with the MMSE. The BMI (β = 0.0005, p = 0.74),
non-Hispanic black (β = 0.006, p = 0.71) compared to nonHispanic white race/ethnicity, current smoking (β = 0.001,
p = 0.94), cardiac disease (β = 0.006, p = 0.71), and risk for
SDB (β = -0.02, p = 0.11) were not associated to the MMSE.
Self-reported sleep hours (continuous) was associated with
worse MMSE scores in sequential models (Table 2). In addition, categorical analysis showed that self-reports of ≥ 9 h (long
sleep duration), compared to 6-8.9 h of sleep were associated
with worse MMSE scores in fully adjusted models (Table 2).
When evaluating cognitive scores as a binary outcome, we
RESULTS
The mean age was 75 ± 9 years, with 61% women, 68% Hispanics, and 41% with less than a high school education. Table 1
presents the characteristics of the overall sample and across categories of sleep duration. Self-reports < 6 h were seen in 24%
of the sample, and ≥ 9 h were reported by 9% (n = 87). Participants reporting ≥ 9 h of sleep were older, had greater frequencies of hypertension, diabetes, and lower MMSE scores than
the reference (p < 0.0001). There was no statistical difference
in the frequencies of sex, education, race-ethnicity, Medicaid or
no insurance status, BMI, depression, cardiac disease, or risk of
SDB among the groups.
671
IR Ramos, C Dong, MSV Elkind et al
Additionally, long sleep duration was associated with worse
MMSE score after controlling for depressive symptoms, medications (e.g., antidepressants, antiepileptics) and risk for SDB,
factors that may worsen cognitive function.11,34-36 Our findings
suggest that in the elderly, long sleep duration (≥ 9 h) could be
an independent predictor of worse cognitive function.
It is suggested that the relation between long sleep and adverse health outcomes could be confounded by SDB.30,37 In
our study, there was no difference in risk for SDB among the
sleep duration groups, and high risk for SDB did not modify
the relation between self-reported long sleep and worse MMSE
scores. Our findings are in accordance with an analysis of the
Osteoporotic Fractures in Men study38 that characterized differences in demographic, sleep, and vascular risk factors among
elderly participants (mean age 76.4 years) with long sleep duration compared to average sleepers. In this study there were
no differences in the apnea-hypopnea index between long sleep
compared to 7-8 hours of sleep.38 In addition, self-reported long
sleep duration was positively associated with increased time
in bed and sleep time by actigraphy that was not explained by
sleep disorders, such as SDB or vascular risk factors.
Our findings could be explained by sleep fragmentation.
Fragmented sleep has been linked to long sleep duration.4
Fragmented sleep measured by actigraphy was associated with
worse global cognitive function, independent of sleep duration, in a cross-sectional analysis of the Rush Memory and
Aging Project.39 Sleep fragmentation is directly related to time
in bed,25 and perhaps long sleep duration could be a surrogate
or a compensatory response to fragmented sleep in those with
worse cognitive function.29 Sleep-wake disturbances could also
exacerbate cognitive dysfunction and cause further sleep disturbances, such as advancement of circadian phase with subsequent prolongation of sleep duration.24
In our study, self-reported short sleep duration was not associated with MMSE. Short sleep duration has been associated
with worse global cognitive function, memory impairment, and
psychomotor speed,7,30,40,41 but stronger associations have been
described for long sleep duration. Our findings are in accordance with population based studies where short sleep duration
was not associated, either by self-report28 or actigraphy,27 with
MMSE. Short sleep duration can cause deficits in attention and
vigilance through excessive sleepiness,42 but the mechanisms
by which long sleep duration could affect cognitive function are
not fully understood.
Several limitations should be noted. The current study is
cross-sectional and does not allow assessment of causality between self-reported sleep hours and cognition. Sleep duration
was obtained by subjective reports from a sleep questionnaire.
In particular, it could be that those with cognitive impairment
tend to report longer sleep duration. Also, we were not able to
capture night to night variability of sleep duration, daytime napping, or objective measures of sleep. However, other observational studies of sleep duration and adverse health outcomes are
similarly based on subjective reports from sleep questionnaires.
Self-reports of long sleep duration might represent a greater
sleep time or just more time in bed, which cannot be determined
from the current data. There might be unmeasured confounders (e.g., autoimmune disorders) that could cause fatigue and
sleepiness and in part explain the results of our study. In spite
Table 3—Odds ratio and 95% confidence interval among
categories of sleep duration and low Mini-Mental score
(MMSE)*
<6h
≥9h
Model 1
0.8 (0.5-1.4)
2.1 (1.2-3.9)
Model 2
0.8 (0.4-1.4)
1.9 (1.02-3.7)
Model 3
0.8 (0.4-1.6)
2.4 (1.1-5.0)
*Low MMSE: A cutoff of MMSE < 24 was used for > 8 years education and
MMSE < 20 for ≤ 8 years of education. Reference: 6-8.9 hours. Model
1: unadjusted. Model 2: adjusted for age, sex, race-ethnicity, education,
and Medicaid or no insurance status. Model 3: adjusted for covariates in
model 2 and alcohol consumption, depression, diabetes, hypertension,
high risk for SDB, and medications.
found that long sleep duration (≥ 9 h) was associated with increased odds of low MMSE scores (Table 3). There was no association between sleep hours and delayed verbal memory and
no interactions between sleep hours and demographic, vascular
risk factors, medications, and risk for SDB.
DISCUSSION
In this cross-sectional study, we found that self-reported long
sleep duration was associated with worse global cognitive function
in the elderly, racially/ethnically diverse NOMAS sample, after
adjusting for vascular risk factors, depression, and risk for SDB.
Self-reported long sleep duration is linked to greater mortality, and an increased risk of stroke and cardiovascular disease.3,23,25,26 While very few NOMAS participants had dementia
and our results most likely reflect subtle cognitive differences,
our findings are in agreement with prospective data from an
elderly (≥ 65 years) population-based cohort showing a positive association between long sleep duration (≥ 9 h) and incident dementia.6 A cross-sectional analysis of the Osteoporotic
Fractures in Men Study (MrOS) also demonstrated an association between long sleep (≥ 8 h) by actigraphy and worse global
cognitive scores.27 Population-based studies have reported associations between long sleep duration and worse cognitive performance by measures of global cognition (MMSE),24,28 as well
as verbal fluency, delayed recall,29 and psychomotor speed.30
Most studies on sleep hours and cognitive function have examined homogenous populations, with a paucity of data from
racially/ethnically diverse communities. Studies comparing
self-reported sleep duration in Hispanics and non-Hispanic
blacks have provided inconsistent results, suggesting that habitual sleep duration is possibly dependent on factors other than
race-ethnicity.9,31,32 We previously described greater long sleep
duration in Hispanics compared to non-Hispanic whites17 and
observed an inverse relation between Hispanic race/ethnicity
and MMSE scores. In NOMAS, a greater proportion of Hispanics have less than eight years of formal education as well
as Medicaid or no insurance, both surrogate markers of lower
SES. Lower SES is associated with a number of comorbidities
that could result in long sleep duration.25,26,33
Self-reports of long sleep duration have been associated with
older age, low socioeconomic status (SES), diabetes, and vascular disease,25 but we observed an association between long
sleep duration and MMSE after controlling for these factors.
672
Sleep Duration and the Mini-Mental Score
of these limitations there are several strengths to our study. We
evaluated a relatively large, racially/ethnically diverse, community-based cohort with a high burden of vascular risk factors,
risk of SDB, depression, and systematically applied measures
of cognition with the MMSE.
In conclusion, we found a cross-sectional association between
self-reported long sleep duration, and greater odds of worse cognitive scores that were not explained by high risk of SDB. Prospective studies in racially/ethnically diverse samples are needed
to confirm our findings and determine if long sleep duration is in
the causal pathway and a harbinger of cognitive decline.
25. Grandner MA, Drummond SP. Who are the long sleepers? Towards an understanding of the mortality relationship. Sleep Med Rev 2007;11:341-60.
26. Patel SR, Malhotra A, Gottlieb DJ, White DP, Hu FB. Correlates of long sleep
duration. Sleep 2006;29:881-9.
27. Blackwell T, Yaffe K, Ancoli-Israel S, et al. Association of sleep characteristics
and cognition in older community-dwelling men: the MrOS sleep study. Sleep
2011;34:1347-56.
28. Faubel R, Lopez-Garcia E, Guallar-Castillon P, Graciani A, Banegas JR, Rodriguez-Artalejo F. Usual sleep duration and cognitive function in older adults in
Spain. J Sleep Res 2009;18:427-35.
29. Kronholm E, Sallinen M, Suutama T, Sulkava R, Era P, Partonen T. Self-reported
sleep duration and cognitive functioning in the general population. J Sleep Res
2009;18:436-46.
30. Kronholm E, Sallinen M, Era P, Suutama T, Sulkava R, Partonen T. Psychomotor slowness is associated with self-reported sleep duration among the general
population. J Sleep Res 2011;20:288-97.
31. Hale L, Do DP. Racial differences in self-reports of sleep duration in a population-based study. Sleep 2007;30:1096-103.
32. Loredo JS, Soler X, Bardwell W, Ancoli-Israel S, Dimsdale JE, Palinkas LA.
Sleep health in U.S. Hispanic population. Sleep 2010;33:962-7.
33. Patel SR. Social and demographic factors related to sleep duration. Sleep
2007;30:1077-8.
34. Byers AL, Yaffe K. Depression and risk of developing dementia. Nat Rev Neurol
2011;7:323-31.
35. Mezick EJ, Hall M, Matthews KA. Are sleep and depression independent or overlapping risk factors for cardiometabolic disease? Sleep Med Rev 2011;15:51-63.
36. Yaffe K, Laffan AM, Harrison SL, et al. Sleep-disordered breathing, hypoxia, and risk
of mild cognitive impairment and dementia in older women. JAMA 2011;306:613-9.
37. Lavie P. Self-reported sleep duration--what does it mean? J Sleep Res
2009;18:385-6.
38. Patel SR, Blackwell T, Ancoli-Israel S, Stone KL. Sleep characteristics of selfreported long sleepers. Sleep 2012;35:641-8.
39. Lim AS, Yu L, Costa MD, et al. Increased fragmentation of rest-activity patterns
is associated with a characteristic pattern of cognitive impairment in older individuals. Sleep 2012;35:633-40.
40. Ferrie JE, Shipley MJ, Akbaraly TN, Marmot MG, Kivimaki M, Singh-Manoux A.
Change in sleep duration and cognitive function: findings from the Whitehall II
Study. Sleep 2011;34:565-73.
41. Xu L, Jiang CQ, Lam TH, et al. Short or long sleep duration is associated with
memory impairment in older Chinese: the Guangzhou Biobank Cohort Study.
Sleep 2011;34:575-80.
42. Ohayon MM, Vecchierini MF. Daytime sleepiness and cognitive impairment in
the elderly population. Arch Intern Med 2002;162:201-8.
REFERENCES
1. Korczyn AD. The underdiagnosis of the vascular contribution to dementia. J
Neurol Sci 2005;229-230:3-6.
2. 2010 Alzheimer’s disease facts and figures. Alzheimers Dement 2010;6:158-94.
3. Cappuccio FP, Cooper D, D’Elia L, Strazzullo P, Miller MA. Sleep duration predicts cardiovascular outcomes: a systematic review and meta-analysis of prospective studies. Eur Heart J 2011;32:1484-92.
4. Kripke DF, Garfinkel L, Wingard DL, Klauber MR, Marler MR. Mortality associated with sleep duration and insomnia. Arch Gen Psychiatry 2002;59:131-6.
5. Mesas AE, Lopez-Garcia E, Leon-Munoz LM, Graciani A, Guallar-Castillon P,
Rodriguez-Artalejo F. The association between habitual sleep duration and sleep
quality in older adults according to health status. Age Ageing 2011;40:318-23.
6. Benito-Leon J, Bermejo-Pareja F, Vega S, Louis ED. Total daily sleep duration
and the risk of dementia: a prospective population-based study. Eur J Neurol
2009;16:990-7.
7. Goel N, Rao H, Durmer JS, Dinges DF. Neurocognitive consequences of sleep
deprivation. Semin Neurol 2009;29:320-39.
8. Ancoli-Israel S, Kripke DF, Klauber MR, Mason WJ, Fell R, Kaplan O. Sleepdisordered breathing in community-dwelling elderly. Sleep 1991;14:486-95.
9. Baldwin CM, Ervin AM, Mays MZ, et al. Sleep disturbances, quality of life, and
ethnicity: the Sleep Heart Health Study. J Clin Sleep Med 2010;6:176-83.
10. Young T, Peppard PE, Gottlieb DJ. Epidemiology of obstructive sleep apnea: a
population health perspective. Am J Respir Crit Care Med 2002;165:1217-39.
11. Zimmerman ME, Aloia MS. Sleep-disordered breathing and cognition in older
adults. Curr Neurol Neurosci Rep 2012;12:537-46.
12. Youngstedt SD, Kripke DF. Long sleep and mortality: rationale for sleep restriction. Sleep Med Rev 2004;8:159-74.
13. Sacco RL, Boden-Albala B, Gan R, et al. Stroke incidence among white, black,
and Hispanic residents of an urban community: the Northern Manhattan Stroke
Study. Am J Epidemiol 1998;147:259-68.
14. Boustani M, Peterson B, Hanson L, Harris R, Lohr KN. Screening for dementia in
primary care: a summary of the evidence for the U.S. Preventive Services Task
Force. Ann Intern Med 2003;138:927-37.
15. Tombaugh T, McDowell I, Kristjansson B, Hubley A. Mini-Mental State Examination (MMSE) and the Modified MMSE (3MS): A psychometric comparison and
normative data. Psychol Assess 1996;8:48-59:
16. Wright CB, Elkind MS, Rundek T, Boden-Albala B, Paik MC, Sacco RL. Alcohol intake, carotid plaque, and cognition: the Northern Manhattan Study. Stroke
2006;37:1160-4.
17. Ramos AR, Wohlgemuth WK, Dong C, et al. Race-ethnic differences of sleep
symptoms in an elderly multi-ethnic cohort: the Northern Manhattan Study. Neuroepidemiology 2011;37:210-5.
18. Netzer NC, Stoohs RA, Netzer CM, Clark K, Strohl KP. Using the Berlin Questionnaire to identify patients at risk for the sleep apnea syndrome. Ann Intern
Med 1999;131:485-91.
19. Young T, Shahar E, Nieto FJ, et al. Predictors of sleep-disordered breathing
in community-dwelling adults: the Sleep Heart Health Study. Arch Intern Med
2002;162:893-900.
20. Boden-Albala B, Roberts ET, Bazil C, et al. Daytime sleepiness and risk of stroke
and vascular disease: Findings from the Northern Manhattan Study (NOMAS)
Circ Cardiovas Qual Outcomes 2012;5:500-7.
21. Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep 1991;14:540-5.
22. Salaycik KJ, Kelly-Hayes M, Beiser A, et al. Depressive symptoms and risk of
stroke: the Framingham Study. Stroke 2007;38:16-21.
23. Gallicchio L, Kalesan B. Sleep duration and mortality: a systematic review and
meta-analysis. J Sleep Res 2009;18:148-58.
24. Potvin O, Lorrain D, Forget H, et al. Sleep quality and 1-year incident cognitive
impairment in community-dwelling older adults. Sleep 2012;35:491-9.
acknowledgments
This work was supported by grants from the National Institute of Neurological Disorders and Stroke: Supplement to R37 NS029993 (to Dr. Ramos); R37 NS029993 (to
Drs. Sacco, Elkind, Rundek, Wright, Boden-Albala); K24 NS 062737 (to Dr. Rundek);
and Evelyn F. McKnight Brain Institute (to Drs. Wright, Sacco, Rundek). All authors
had access to the data and contributed substantially to the design, acquisition, and
analysis of data, and writing of manuscript. The Northern Manhattan study cohort is
followed at Columbia University. The administrative support and statistical analysis
was performed at the University of Miami-Miller School of Medicine. The authors are
thankful to the study participants for their collaboration and to all staff of the Northern
Manhattan Study for their efforts to this study, and in particular Edison Sabala and
Janet DeRosa.
Address correspondence to: Alberto Ramos, M.D., M.S.P.H., Sleep Medicine
Program, Department of Neurology, University of Miami Miller School of Medicine,
1120 NW 14th Street, Suite 1350. Miami, Florida 33136, Tel: (305) 243-1305; Fax:
(305) 243-7081
673
Characterization of REM Sleep without Atonia in Patients with
Narcolepsy and Idiopathic Hypersomnia using AASM Scoring
Manual Criteria
Lourdes M. DelRosso, M.D.; Andrew L. Chesson Jr., M.D., F.A.A.S.M.; Romy Hoque, M.D.
Department of Neurology, Division of Sleep Medicine, Louisiana State University School of Medicine Shreveport, LA
Introduction: The AASM Manual for the Scoring of Sleep and
Associated Events (Manual) has provided standardized definitions for tonic and phasic REM sleep without atonia (RSWA).
This study used Manual criteria to characterize REM sleep in
patients with narcolepsy and idiopathic hypersomnia (IH).
Methods: A retrospective review of PSG data from ICSD-2
defined patients with narcolepsy or IH, performed by two
board certified sleep medicine physicians. Data compiled included REM sleep epochs and the presence in REM sleep
of epochs scored as sustained muscle activity (tonic), and
excessive transient muscle activity (phasic) as defined by
Manual criteria.
Results: PSG data from 8 narcolepsy patients (mean age:
27.5 years; age range: 11-55) showed mean ± standard deviation values for: total REM sleep epochs 205 ± 46.1; RSWA/
phasic epochs 56.1 ± 25.4; and RSWA/tonic epochs 15.0 ±
10.7. PSG data from 8 IH patients (mean age: 33.1 years; age
range: 20-57) showed mean ± standard deviation values of
total REM sleep epochs 163.8 ± 67.9; RSWA/phasic epochs
6.2 ± 3.5; and RSWA/tonic epochs 0.2 ± 0.4. Comparison revealed intergroup differences in phasic REM sleep (p < 0.01)
and tonic REM sleep (p < 0.01) were significantly increased in
narcoleptics compared to IH.
Conclusion: Our retrospective analysis showed that RSWA
phasic activity and RSWA tonic activity are significantly increased in patients meeting ICSD-2 criteria for narcolepsy
compared to patients meeting ICSD-2 criteria for IH. This robust difference, with further validation, could be useful as electrophysiological criteria differentiating the two disorders and
understanding the physiological differences.
Keywords: Narcolepsy, idiopathic hypersomnia, rapid eye
movement sleep, REM sleep without atonia, phasic, tonic
Citation: DelRosso LM; Chesson AL Jr; Hoque R. Characterization of REM Sleep without atonia in patients with narcolepsy and idiopathic hypersomnia using AASM scoring manual
criteria. J Clin Sleep Med 2013;9(7):675-680.
T
he International Classification of Sleep Disorders-second
edition (ICSD-2) classifies narcolepsy and idiopathic hypersomnia (IH) under hypersomnias of central origin.1 The
diagnostic criteria for both conditions include at least three
months of subjective excessive daytime sleepiness (EDS) occurring almost daily and a mean sleep latency of less than
eight minutes in the multiple sleep latency test (MSLT). These
two conditions are electrophysiologically differentiated based
on the number of sleep onset REM periods (SOREMPS) in
the MSLT: two or more for narcolepsy and less than two for
IH. The ICSD-2 further subdivides narcolepsy into narcolepsy with cataplexy (N+C) and narcolepsy without cataplexy
(N-C). For both narcolepsy and IH, medical, mental, neurological, or pharmacological causes must also be excluded. The
conditions must also not be accounted for by another sleep
disorder or drug use.
Several earlier reports have presented evidence of REM
dysfunction in patients with narcolepsy including: different
patterns in REM sleep distribution/REM density across the
night, REM sleep phasic activity, and early onset REM sleep
periods.2 REM behavior disorder (RBD) has been reported in
up to 36% of patients with narcolepsy, with higher prevalence
in N+C.3 On rare occasions, RBD has been the presenting
complaint in patients with undiagnosed narcolepsy.4 How-
Brief Summary
Current Knowledge/Study Rationale: Earlier reports have presented
evidence of REM dysfunction in patients with narcolepsy. Our study will
evaluate electrophysiologic differences between Narcolepsy and Idiopathic Hypersomnia.
Study Impact: Our study demonstrates a robust electrophysiologic difference in both tonic and phasic REM sleep without atonia between IH
and narcolepsy, independent of cataplexy status.
ever, the motor manifestations of RBD in narcolepsy have
been found to be both less frequent and less severe than in
idiopathic RBD.5 REM sleep without atonia (RSWA) has been
previously reported in narcoleptics who do not meet criteria
for RBD.6
Different methods have been used to assess motor dysregulation in patients with narcolepsy. Automated computerized scoring has demonstrated increased EMG activity in the chin during
sleep, while video monitoring studies have shown “mild” motor
behaviors in narcolepsy as opposed to full-blown violent/aggressive RBD.7
IH closely resembles narcolepsy both clinically and in polysomnography (PSG) features.8 Furthermore, REM related
symptoms, such as hypnagogic hallucinations and sleep paraly675
LM DelRosso, AL Chesson Jr. and R Hoque
Figure 1—Examples of typical rapid eye movement sleep without atonia (RSWA) epochs
A
B
(A) RSWA - phasic activity. (B) RSWA - tonic activity.
sis, have been found to be higher in a subgroup of patients with
IH than in the general population.9
The goal of this study was to analyze PSG data scored in accordance with the American Academy of Sleep Medicine (AASM)
Manual for the Scoring of Sleep and Associated Events (Manual)
in narcolepsy and IH to evaluate electrophysiologic differences
between these two conditions, with special attention to whether RSWA differs in these two conditions.10 No prior studies of
RSWA in narcolepsy have explicitly utilized the Manual criteria.
METHODS
Selection Criteria
Over a 2-year period (03.01.2010 to 03.01.2012), all patients
with a new diagnosis of narcolepsy or IH were identified from patient records at the Sleep Disorders Center of the Louisiana State
University Health Sciences Center in Shreveport, Louisiana. The
diagnosis of IH, N-C, and N+C was made by a board certified
676
REM Without Atonia in Narcolepsy and IH
Table 1—Demographics
N
Mean ± SD, (range)
27.5 ± 12.73, (11-55)
19.75 ± 2.92 (16-24)
33.69 ± 7.37, (24-43.8)
A
Demographics
Age
ESS
BMI
B
Demographics
Sex (F)
Self reported dream enacting behavior
Self reported sleep talking
RLS
Caffeine intake
IH
Mean ± SD, (range)
32.62 ± 14.34, (20-57)
16.63 ± 3.62, (10-20)
28.69 ± 5.35, (20.9-37.8)
N
Total, % of total
4, 50%
2, 25%
3, 37%
3, 37%
3, 37%
Mann-Whitney p-values:
N/IH
0.674
0.074
0.156
IH
Total, % of total
7, 87%
0
3, 37%
3, 37%
3, 37%
(A) Age, ESS, and BMI. No statistically significant differences where noted in these variables between the two groups. (B) Sex distribution and self-reported
dream enacting behavior, self-reported sleep talking, RLS, and daily caffeine intake. ESS, Epworth Sleepiness Score; BMI, body mass index; N, narcolepsy;
IH, idiopathic hypersomnia; RLS, restless legs syndrome.
Table 2—PSG data from narcoleptics and idiopathic hypersomniacs
p-values
PSG variable
Total REM sleep in
minutes
Total Tonic REM
sleep epochs
Kruskal-Wallis
N
N+C
N-C
IH
Mean ± SD
Mean ± SD
Mean ± SD
Mean ± SD
N+C/N-C/IH
205.75 ± 46.15 206.25 ± 65.06 205.25 ± 27.12 163.88 ± 67.92
0.236
15 ± 10.74
Mann Whitney
N/IH
0.104
N+C/IH N-C/IH N+C/N-C
0.270 0.126 0.564
10.5 ± 5.07
19.5 ± 13.77
0.25 ± 0.46
0.002
0.001
0.007
0.007
0.248
0.001
0.007
0.007
0.564
Total Phasic REM
sleep epochs
56.13 ± 25.74
51.5 ± 27.74
60.75 ± 26.81
6.25 ± 3.54
0.003
RSWAI
42.67 ± 16.39
38.25 ± 22.09
47.09 ± 9.31
5.59 ± 3.48
0.003
0.001
0.007
0.007
0.386
REM sleep onset
21.13 ± 32.95
4.63 ± 1.6
37.63 ± 42.49 119.88 ± 54.89
0.008
0.003
0.007
0.042
0.248
19.56 ± 13.19
9.88 ± 6.86
29.25 ± 10.44
Sleep latency
TST
40.88 ± 26.67
0.019
0.059
0.007
0.734
0.043
419.56 ± 47.96 412.25 ± 35.29 426.88 ± 63.08 461.31 ± 33.35
0.107
0.036
0.062
0.126
0.885
WASO
37.94 ± 23.41
54.25 ± 19.44
21.63 ± 13.82
0.128
0.372
0.107
0.865
0.043
SE
90.89 ± 5.08
87.68 ± 3.97
94.1 ± 4.13
92.49 ± 5.99
32.5 ± 31.91
0.187
0.674
0.174
0.497
0.083
3.06 ± 1.9
0.627
0.793
0.734
0.445
0.386
61.2 ± 10.65
0.458
0.345
0.308
0.610
0.248
N1%
3.36 ± 3.41
4.68 ± 4.37
2.05 ± 1.86
N2%
55.5 ± 14
49.47 ± 18.19
61.53 ± 5.39
N3%
16.38 ± 12.5
20.58 ± 16.39
12.18 ± 6.97
17.83 ± 4.04
0.387
0.294
0.734
0.126
0.386
REM sleep %
24.76 ± 6.43
25.28 ± 9.14
24.24 ± 3.5
17.91 ± 8
0.082
0.027
0.126
0.042
0.773
PLMI
12.99 ± 22.79
5.58 ± 5.59
14.53 ± 27.85
0.967
0.834
0.865
0.865
0.773
20.4 ± 32.15
REM, rapid eye movement; RSWAI, REM sleep without atonia index; TST, total sleep time; WASO, wake after sleep onset; SE, sleep efficiency; PLMI,
periodic limb movement index per hour of sleep; N, narcolepsy; N+C, narcolepsy with cataplexy; N-C, narcolepsy without cataplexy; SD, standard deviation.
sleep medicine physician based on history, clinical symptoms,
and nocturnal PSG and MSLT according to ICSD-2 criteria.1
To these narcolepsy and IH patient records we applied the
following exclusion criteria: presence of obstructive sleep apnea defined by total sleep time (TST) apnea hypopnea index > 5,
circadian disorder, insufficient sleep syndrome based on sleep
diaries, positive drug screen performed with MSLT, REM sleep
altering/cataplexy suppressing medications (e.g., tricyclic antidepressants, selective serotonin and/or norepinephrine reuptake
inhibitors, sodium oxybate), PSG/MSLT done at another facility, psychiatric or neurological comorbidity, and PSG/MSLT recording obscured by significant artifact. The remaining patients
used for the study consisted of 8 patients with IH, 4 patients
with N+C, and 4 patients with N-C.
Data Acquisition
All patients had a nocturnal PSG followed by an MSLT. The
PSG recordings were acquired using the Alice 5 system (Respironics, Inc., Murrysville, PA, USA) and included 6 electroencephalogram channels (F3/A2, F4/A1, C3/A2, C4/A1, O1/
A2, O2/A1), vertical and horizontal electroculograms, chin
and bilateral tibialis anterior muscle electromyogram, 1-channel electrocardiogram, and respiratory monitors (nasal pressure
transducer, thermistor, thoracic and abdominal plethysmogra677
formed using the Mann-Whitney U calculation. P-values less
than 0.05 were considered statistically significant.
Figure 2—Comparison of REM sleep without atonia index
(RSWAI) in narcolepsy with cataplexy, narcolepsy without
cataplexy, and idiopathic hypersomnia
RESULTS
A
Demographics
Demographic data of the 3 patient groups is presented in
Table 1. Mean age for the 8 narcolepsy patients was 27.5 years
(range: 11-55); 4 patients were women. Four patients had N+C; 4
patients had N-C. Mean age for the 8 IH patients was 33.12 years
(range: 20-57); 7 patients were women. Age, body mass index,
and Epworth Sleepiness Scale score were not significantly different between the narcoleptic and IH groups. Two patients with
narcolepsy had questionnaire reports of dream-enacting behavior
consisting of non-injurious nocturnal arm and leg movements.
No patients reported injurious parasomnias or had a primary complaint of parasomnias. Three patients with narcolepsy (all N+C)
and 3 with IH reported daily ingestion of caffeinated beverages.
As the PSG and MSLT studies were to establish a hypersomnia
related diagnosis, all patients were off stimulants and sleep/wake
or REM sleep altering medications for > 2 weeks.
B
PSG Data
PSG data is summarized in Table 2. Total REM sleep in
minutes was not significantly different between narcoleptics
(206.75 min) and IH (163.88 min), but REM sleep % was
significantly different between narcoleptics (24.76%) and IH
(17.91%). The use of MSLT to support a diagnosis of narcolepsy
requires TST > 6 h on the prior night.1 Patients from both groups
exceeded this amount considerably. TST was significantly different between narcoleptics and IH. The mean TST in narcoleptics
was 419.56 min; it was 461.31 min in IH. The narcoleptics TST
ranged from 382.5 min to 517 min. The IH TST ranged from 428
min to 532 min. Wake after sleep onset (WASO), sleep efficiency, N1 percentage, N2 percentage, N3 percentage, and periodic
limb movement indices (PLMI) were not significantly different
between narcoleptics and IH. Neither the narcoleptics nor the
IH patients exhibited abnormal REM sleep behaviors during the
video PSG recording.
Significant intergroup variances were found between IH,
N+C, and N-C in total tonic REM sleep epochs, total phasic
REM sleep epochs, RSWAI, and REM sleep latency at p < 0.01;
and in sleep latency at p < 0.05. When comparing narcolepsy to
IH, N+C to IH, and N-C to IH: significant Mann-Whitney statistical differences at p < 0.01 were found in total tonic epochs,
total phasic epochs, and RSWAI. When comparing N+C to IH:
significant statistical difference at p < 0.01 was found in sleep latency. RSWAI comparisons are shown in Figure 2A. The range
of RSWAI values in narcoleptics and IH do not overlap with one
another and seem to have distinct ranges (Figures 2A and 2B).
No significant statistical difference was found in sleep latency when comparing narcolepsy to IH or when comparing
N-C to IH. Significant statistical difference (p < 0.05) in sleep
latency and WASO was found between N+C and N-C.
Mean REM sleep onset was 21.13 min in all narcoleptics,
4.63 min in N+C, 37.63 min in N-C, and 119.88 min in IH. Statistically significant differences in REM sleep onset at p < 0.01
were seen when comparing narcolepsy to IH, and when com-
(A) Box-plot of RSWAI distributions. (B) RSWAI range in idiopathic
hypersomnia and narcolepsy.
phy belts, microphone and pulse oximetry). PSG scoring was in
accordance with the Manual.
Quantification of RSWA
Each PSG was scored in agreement by 2 board certified sleep
medicine physicians (LD, RH) in accordance with the Manual.
RSWA was scored using section 7: movement rules; Number
6: Scoring PSG features of REM sleep behavior disorder.10
Figures 1A and 1B show sample epochs with RSWA.
Calculation of RSWA Index (RSWAI)
RSWAI, the total number of RSWA epochs per hour of REM
sleep, was calculated using the formula below:
RSWAI (RSWA 30 sec epochs per hour of REM sleep) = 120
(total number of RSWA 30 sec epochs/total REM sleep epochs)
The total number of RSWA epochs is the sum of the total
number of epochs meeting criteria for RSWA by phasic and by
tonic criteria.
Analysis of variance for non-parametric samples was performed using the Kruskal-Wallis calculation, and assessment
of statistical significance for nonparametric samples was perJournal of Clinical Sleep Medicine, Vol. 9, No. 7, 2013
678
REM Without Atonia in Narcolepsy and IH
Figure 3
Figure 4—Epoch distribution of REM sleep without atonia
(RSWA) comparing narcolepsy and idiopathic hypersomnia
A
A
B
B
(A) Comparison of Total REM sleep epoch distribution across the night
in narcolepsy and in idiopathic hypersomnia. (B) REM sleep onset
comparison between narcolepsy with cataplexy, narcolepsy without
cataplexy and idiopathic hypersomnia.
(A) Phasic RSWA epoch distribution. (B) Tonic RSWA epoch distribution.
REM sleep distribution between groups. This finding is consistent with a previous study of the overnight distribution of motor episodes in patients with narcolepsy with cataplexy, which
demonstrated that RBD episodes in these patients are less predictable than in patients with idiopathic RBD and independent
of time of night or REM sleep period length.11 The differences
in sleep latency and REM sleep onset between narcolepsy and
IH shown in our study are consistent with prior investigations.12
Our study demonstrates a robust electrophysiologic difference in both tonic and phasic RSWA between IH and narcolepsy,
independent of cataplexy status. Recent studies in vivo suggest
that the mechanism of REM sleep atonia is dependent on glutamatergic neurons from the sublaterodorsal (SLD) nucleus
projecting to glycinergic neurons on the ventromedial medulla
(VMM) and/or spinal cord.13 Muscle atonia is also maintained
by simultaneous withdrawal of histaminergic, serotonergic, and
noradrenergic input.14 The areas implicated in muscle atonia
during REM sleep include the magnocellular reticular formation, locus ceruleus, subceruleus, pedunculopontine tegmentum, and laterodorsal tegmentum. The presence and complexity
of REM sleep motor behavior exhibited in RBD may depend
on the neuroanatomical site affected.15 For example, a case of
narcolepsy with RBD has been reported in association with an
isolated pontine tegmental lesion.16
paring N+C to IH. Statistically significant differences in REM
sleep onset at p < 0.05 were seen when comparing N-C to IH.
Despite the difference in mean REM sleep onset between N+C
and N-C, this difference was not statistically significant.
Epoch distribution of REM sleep across the night and REM
sleep onset in narcolepsy and IH are shown in Figures 3A and 3B,
respectively. A more even distribution of REM sleep across the
night with more REM sleep earlier in the sleep period was seen
in narcolepsy than in IH. IH showed REM sleep predominance
later in the sleep period.
Phasic RSWA and tonic RSWA are much more prominent
in narcolepsy compared to IH (Figure 4). In both narcolepsy
and IH: phasic RSWA was more prominent than tonic RSWA;
phasic RSWA was most prominent in the third quarter of night
and decreases in the final quarter of the night; and tonic RSWA
was more prominent in the first half of the night and decreased
in the second half of the night.
DISCUSSION
In assessing PSG differences between narcolepsy and IH, this
study revealed a statistically significant difference in overnight
679
Hypocretin containing neurons, localized in the hypothalamus with widespread projections throughout the CNS,17 not
only stabilize the sleep/wake cycle but also play a role in modulating muscle atonia during REM sleep by regulating the activity of lumbar motor neurons through both pre-synaptic and
post-synaptic mechanisms18 in a sleep/wake stage dependent
manner.19 N+C patients have been found to have over 90% deficiency in hypocretin neurons, while N-C patients have been
found to have a 30% deficiency in hypocretin neurons.20 RBD
in narcolepsy has been associated with hypocretin deficiency,
independent of cataplexy status.21
Limitations of our study include: retrospective rather than
prospective analysis, small patient sample, evaluation of single
night PSG data from each patient, patient recruitment from a
single institution, and lack of blinded comparisons. These issues may be addressed in future larger multicenter studies. The
purpose of this study was to compare RSWA in hypersomnias
of central origin, but future studies may also include comparisons to patients who meet ICSD-2 criteria for idiopathic RBD.
Our study may support a common putative pathology for
RSWA in both narcolepsy groups that differs from IH. This
finding may add an extra electrophysiologic measure, RSWAI,
which can potentially be used as an extra marker in aiding in
the differentiation between these two central hypersomnias. If
confirmed by larger numbers of patients and other investigators, a RSWAI cutoff value obtained from a diagnostic PSG
using Manual criteria may be established to help support the
differential diagnosis between narcolepsy or IH.
8. Billiard MM. From narcolepsy with cataplexy to idiopathic hypersomnia without
long sleep time. Sleep Med 2009;10:943-4.
9. Sasai T, Inoue Y, Komada Y, Sugiura T, Matsushima E. Comparison of clinical
characteristics among narcolepsy with and without cataplexy and idiopathic hypersomnia without long sleep time, focusing on HLA-DRB1(*)1501/DQB1(*)0602
finding. Sleep Med 2009;10:961-6.
10. Iber C, Anconi-Israel S, Chesson A, Quan S. The AASM manual for the scoring
of sleep and associated events: rules, terminology, and technical specifications.
Westchester, IL: American Academy of Sleep Medicine; 2007.
11. Cipolli C, Franceschini C, Mattarozzi K, Mazzetti M, Plazzi G. Overnight distribution and motor characteristics of REM sleep behaviour disorder episodes in
patients with narcolepsy-cataplexy. Sleep Med 2011;12:635-40.
12. Martinez-Rodriguez JE, Sabater L, Graus F, Iranzo A, Santamaria J. Evaluation of hypothalamic-specific autoimmunity in patients with narcolepsy. Sleep
2007;30:27-8.
13. Krenzer M, Anaclet C, Vetrivelan R, et al. Brainstem and spinal cord circuitry
regulating REM sleep and muscle atonia. PloS One 2011;6:e24998.
14. Siegel JM. The neurotransmitters of sleep. J Clin Psychiatry 2004;65 Suppl
16:4-7.
15. McCarter SJ, St Louis EK, Boeve BF. REM sleep behavior disorder and REM
sleep without atonia as an early manifestation of degenerative neurological disease. Curr Neurol Neurosci Rep 2012;12:182-92.
16. Mathis J, Hess CW, Bassetti C. Isolated mediotegmental lesion causing narcolepsy and rapid eye movement sleep behaviour disorder: a case evidencing
a common pathway in narcolepsy and rapid eye movement sleep behaviour
disorder. J Neurol Neurosurg Psychiatry 2007;78:427-9.
17. Peyron C, Tighe DK, van den Pol AN, et al. Neurons containing hypocretin
(orexin) project to multiple neuronal systems. J Neurosci 1998;18:9996-10015.
18. Yamuy J, Fung SJ, Xi M, Chase MH. Hypocretinergic control of spinal cord motoneurons. J Neurosci 2004;24:5336-45.
19. Yamuy J, Fung SJ, Xi M, Chase MH. State-dependent control of lumbar motoneurons by the hypocretinergic system. Exp Neurol 2010;221:335-45.
20. Thannickal TC, Nienhuis R, Siegel JM. Localized loss of hypocretin (orexin) cells
in narcolepsy without cataplexy. Sleep 2009;32:993-8.
21. Knudsen S, Gammeltoft S, Jennum PJ. Rapid eye movement sleep behaviour
disorder in patients with narcolepsy is associated with hypocretin-1 deficiency.
Brain 2010;133(Pt 2):568-79.
References
1. American Academy of Sleep Medicine. International classification of sleep disorders; diagnostic and coding manual. 2nd ed. Westchester, IL: American Academy of Sleep Medicine; 2005.
2. Geisler P, Meier-Ewert K, Matsubayshi K. Rapid eye movements, muscle twitches and sawtooth waves in the sleep of narcoleptic patients and controls. Electroencephalogr Clin Neurophysiol 1987;67:499-507.
3. Nightingale S, Orgill JC, Ebrahim IO, de Lacy SF, Agrawal S, Williams AJ. The
association between narcolepsy and REM behavior disorder (RBD). Sleep Med
2005;6:253-8.
4. Bonakis A, Howard RS, Williams A. Narcolepsy presenting as REM sleep behaviour disorder. Clin Neurol Neurosurg 2008;110:518-20.
5. Dauvilliers Y, Rompre S, Gagnon JF, Vendette M, Petit D, Montplaisir J. REM
sleep characteristics in narcolepsy and REM sleep behavior disorder. Sleep
2007;30:844-9.
6. Schenck CH, Mahowald MW. Motor dyscontrol in narcolepsy: rapid-eye-movement (REM) sleep without atonia and REM sleep behavior disorder. Ann Neurol
1992;32:3-10.
7. Franceschini C, Ferri R, Pizza F, et al. Motor events during REM sleep in patients with narcolepsy-cataplexy: a video-polysomnographic pilot study. Sleep
Med 2011;12 Suppl 2:S59-63.
Address correspondence to: Lourdes DelRosso, M.D., Department of Neurology,
Division of Sleep Medicine, Louisiana State University School of Medicine, 1501
Kings Highway, Shreveport, Louisiana 71103; Tel: (318) 762-2930; Fax: (318) 6754404; E-mail: lordydel@yahoo.com
conflicts of interest. This research was presented in part at the 2012 SLEEP 2012
meeting in Boston, Massachusetts.
680
Targeted Case Finding for OSA within the Primary Care Setting
Keith R. Burgess, Ph.D.1; Adrian Havryk, Ph.D.1; Stephen Newton, M.B.A.2; Willis H. Tsai, M.D., F.A.A.S.M.3; William A. Whitelaw, M.D., Ph.D.4
1
Peninsula Respiratory Group, Frenchs Forest, NSW, Australia; 2Healthy Sleep Solutions Pty Ltd, Sydney,
NSW, Australia; 3Department of Community Health Sciences, University of Calgary, Calgary, Alberta, Canada;
4
Department of Medicine, University of Calgary, Calgary, Alberta, Canada
Study Objectives: The aim was to determine the feasibility of
using an unattended 2-channel device to screen for obstructive sleep apnea in a population of high-risk patients using a
targeted, case-finding strategy. The case finding was based
on the presence of risk factors not symptoms in the studied
population.
Methods: The study took place from June 2007 to May 2008
in rural and metropolitan Queensland and New South Wales.
Family doctors were asked to identify patients with any of the
following: BMI > 30, type 2 diabetes, treated hypertension,
ischemic heart disease. Participants applied the ApneaLink+O2
at home for a single night. The device recorded nasal flow and
pulse oximetry. Data were analyzed by proprietary software,
then checked and reported by either of two sleep physicians.
Results: 1,157 patients were recruited; mean age 53 ± 14.6,
M/F% = 62/38, mean BMI = 31.8, obesity = 35%, diabetes = 16%,
hypertension = 39%, IHD = 5%, Mean Epworth Sleepiness
Scale score (ESS) = 8.3. The prevalence of unrecognized OSA
was very high: 71% had an AHI > 5/h, 33% had an AHI > 15/h,
and 16% had an AHI > 30/h. The ApneaLink+O2 device yielded
technically adequate studies in 93% of cases.
Conclusion: The study shows that a “real world” simple low
cost case finding and management program, based on unattended home monitoring for OSA, can work well in a population with risk factors and comorbidities associated with OSA,
independent of the presence of symptoms. The prevalence of
unrecognized OSA was very high.
Keywords: Obstructive sleep apnea, ApneaLink, unattended
sleep study
Citation: Burgess KR; Havryk A; Newton S; Tsai WH;
Whitelaw WA. Targeted case finding for OSA within the primary care setting. J Clin Sleep Med 2013;9(7):681-686.
O
bstructive sleep apnea (OSA) is a modern epidemic of
great health and economic importance. Young et al.1 found
in a population aged 30-60, that 9% of women and 24% of men
had OSA, defined as apnea-hypopnea index (AHI) > 5/h, while
2% of women and 4% of men had AHI > 5/h, in association
with excessive sleepiness (obstructive sleep apnea syndrome
[OSAS]).
Untreated OSA is an important cause of impaired alertness
and daytime sleepiness. Patients with OSA have been found to
consume more health care resources than those without, take
more sick leave, and have more work disability.2,3 Large numbers of cases of OSA remain undiagnosed and untreated, in large
part because of limited resources for case finding and diagnosis.4
Male gender, increasing age, and obesity are known risk factors
for OSA.5 OSA is an independent risk factor for vascular disease,6 stroke,7,8 and probably hypertension,9,10 and contributes to
glucose intolerance11-13 and difficult-to-control atrial fibrillation.14
Patients with such risk factors are more likely to have OSA. They
are also more likely to benefit from treatment for OSA that may
not only improve energy and alertness but also to help control
associated diseases. Currently diagnosis usually results from testing, which is traditionally triggered by patient symptomatology.
If investigation is not triggered unless there are symptoms, many
cases in these high-risk groups may go undiagnosed.
It is important to find and treat patients with OSA. Screening of the whole population would have a relatively low yield.
We wished to assess the feasibility and the yield of a program
to screen high-risk patients for unrecognized OSA, regardless
Brief Summary
Current Knowledge/Study Rationale: It is important to find and treat
patients with OSA, but screening of whole populations would have a
relatively low yield. We wished to assess the feasibility and the yield of a
targeted case finding program to screen high risk patients for unrecognized OSA regardless of whether they had typical symptoms.
Study Impact: The study shows that a simple low cost case finding and
management program, based on unattended home monitoring for OSA,
focused on patients with obesity, hypertension and diabetes can work
well in a population with risk factors and comorbidities associated with
OSA. The prevalence of unrecognized OSA in this population was high,
so these data support the testing for OSA in high risk groups whether
they have the traditional symptoms of OSA or not.
of whether they had typical symptoms, hoping to improve the
health of many of them and to reduce the health care costs associated with untreated OSA. To limit costs, we used an unattended
portable monitor at home that has been shown to give AHI values
similar to those from attended laboratory polysomnography.
The aim of the study was to determine the prevalence of unrecognized OSA in a population of high risk patients, in the
practices of family physicians, using a targeted case finding
strategy.
METHODS
The study used data which was collected from June 2007
to May 2008 at sites in rural and metropolitan Queensland,
(Bundaberg and Brisbane) and 4 regions across the state of
681
KR Burgess, A Havryk, S Newton et al
New South Wales (the Northern, Eastern and North Western
Suburbs of Sydney and the rural Southern Highlands). It was
done with the cooperation of general practitioners located
across these regions and Healthy Sleep Solutions (HSS), a private company that specializes in facilitating home diagnostic
studies and provision of CPAP therapy on behalf of Specialist
Sleep Physicians. The geographic catchment area was approximately equally divided between urban and rural areas, (47%
and 53% respectively), with a total population of approximately
560,000. The de-identified data were provided by HSS. Patients
gave consent to HSS for the data collection. The data analysis was approved by an institutional review board (IRB); the
Northern Sydney Central Coast Human Research Ethics Committee [1101-032M(QA)]. The results of the testing and reporting physician recommendations were conveyed to the patients
by the referring family physicians. Subsequent treatment decisions and implementation were then the result of doctor and
patient decisions, which were beyond the control of the authors.
both 0.933. The new version continues to base its calculation of
AHI on the nasal pressure tracing, but a second channel recording oxygen saturation allows it to count the number of dips per
hour > 4%, which was used as a substitute for AHI if the nasal
pressure probe was lost, and has the potential to reduce the study
failure rate. (This approach is supported by the similar high sensitivities and specificities of both the oximetry and nasal flow
signals [~85%]19). The ApneaLink+O2 displays tracings of nasal
pressure, oxygen saturation, and pulse rate. The pulse oximeter
was a Nonin Xpod 3012 with a Nonin 7000A finger probe and
a sampling rate of 1 Hz. Flow was sampled at 100 Hz. Tracings
were reviewed initially at 5 cm/min, but expanded as necessary.
The device provides statistics, including an apnea-hypopnea index derived from the flow tracing plus minimum saturation, the
number per hour of desaturations of 4%, oxygen desaturation
index (ODI), and mean pulse rate.
To be considered technically “good,” a study must have
lasted ≥ 4 h and must have recorded data from both flow and
oxygen saturation channels ≥ 90% of the time. A second category was considered “acceptable,” which meant that either the
oximetry or the flow signal was missing, or of poor quality for
> 10% of the recording, but the study could still be interpreted
with confidence on data from the remaining channel.
One of two accredited sleep physicians reviewed each tracing, assessed their quality, noted the Epworth score, (a score ≥
11 being considered excessive sleepiness), medications, medical history, and daytime oxygen saturation and issued a report
and recommendation based on an algorithm (Figure 1). AHI
was used as the primary criterion of severity, but on the few
occasions when the oxygen desaturation index was higher than
the AHI because of a problem with the flow signal for part of
the recording, the higher number was used. If the initial test was
inadequate, a repeat test was requested.
Recommendations were based on a modification of published
algorithm.20 Because of previous uncertainty as to the accuracy
of the ApneaLink device when AHI is between 10 and 30, and
because of the recognized lack of certainty about benefits of
treatment in that range, we recommended that these patients be
referred to a respiratory/sleep physician for assessment, (probably including polysomnography), and management. We recommended that all patients with AHI > 30/h undergo a trial of
treatment with continuous positive airway pressure (CPAP), because this degree of OSA is considered likely to be detrimental
to a patient’s health, even if no symptoms are reported. In addition, however, we recommended trials of treatment with CPAP
for patients with an AHI > 20/h who had comorbidities that
can be worsened by the presence of untreated OSA, e.g. type 2
diabetes, hypertension, and ischemic heart disease. These constituted 40% of patients for whom CPAP was recommended.
Referral to a sleep physician was recommended for patients
who had excessive daytime sleepiness and AHI < 10/h. If snoring was an important complaint and AHI < 10/h, a mandibular
advancement device was recommended.
Recruitment
Doctors were approached for participation in the study initially by mail and then by office visits from a specially trained
clinical coordinator representing Healthy Sleep Solutions or
a cooperating representative of a pharmaceutical company.
General practitioners were asked to identify from among their
patients those with a body mass index (BMI) > 30, type 2 diabetes, treated hypertension, ischemic heart disease, or with the
traditional risk factors of snoring, sleepiness, and witnessed
apneas.5 Prior to the study, the consenting subjects were interviewed by one of the HSS clinical coordinators to record clinical data, including height, weight, blood pressure, and details
of their history and medications. Subjects also completed an
Epworth Sleepiness Scale.15
Hypertension was considered present according to self-report or use of antihypertensive medication. Type 2 diabetes was
defined by an abnormal fasting plasma glucose, or use of oral
hypoglycemic medications. Ischemic heart disease was defined
by self-report, or use of cardiac medications.
Each patient was then instructed in the use of the
ApneaLink+O2, took a device home, and applied it for a single
night of recording. Data were extracted and analyzed by proprietary software (ResMed) and were then reviewed and reported
by 2 respiratory physicians experienced in sleep medicine, and
approved by the Health Insurance Commission to report laboratory polysomnograms.
ApneaLink+O2
ApneaLink+O2 is a proprietary monitor based on analysis of
nasal pressure and oxygen saturation signals. An earlier version,
employing only a nasal pressure signal, has been tested against
simultaneously recorded laboratory polysomnography in a study
of 59 subjects, and found to have sensitivity 0.91 and specificity 0.84 for OSA,16 comparable to other similar monitors that
have been validated against polysomnography.17,18 Very recently,
the 2-channel device used in this study has been validated in a
study of 143 subjects against home polysomnography (PSG) in
a Chinese population.19 Those authors found very high sensitivity and specificity for the diagnosis of OSA compared to in home
PSG; the areas under the reader operator curves (ROC) were
Statistics
Patients were described using mean and 95% confidence intervals. A χ2 test for linear trend was used to determine if the
prevalence of comorbidity had a dose dependent relationship
with OSA severity. Patient characteristics were treated as con682
Targeted Case Finding for OSA
Figure 1—OSA management algorithm
High Risk Patients
DM/HT/+BMI
+/No Symptoms
ApneaLink+O2
Low Risk OSA
No Symptoms
AHI < 10/h
Low Risk OSA
+ Symptoms
AHI < 10/h
Mod Risk OSA
No Symptoms
10/h ≤ AHI ≤ 30/h
Severe OSA
AHI > 30/h or
AHI 20-30/h + Symptoms + Risk
factors
CV Risk Management
Standard GP Care
Refer for Mandibular
Device or Specialist
Refer to Specialist
CPAP Trial
Ongoing GP* + HSS follow-up
*If GP is not keen to follow up then patient witll be directed to specialists. Incorporating guidelines produced by the Thoracic Society of Australia & New
Zealand and Australasian Sleep Association. OSA, obstructive sleep apnea; DM, diabetes mellitus; HT, hypertension; AHI, apnea hypopnea index; CV,
cardiovascular disease; GP, general practitioner (family doctor); CPAP, continuous positive airway pressure; HSS, Health Sleep Solutions Pty Ltd.
tinuous variables, but where appropriate, (e.g., Sleepy [ESS >
10], and Obese [BMI > 30]), were also analyzed as categorical
variables. Comorbidities were treated as categorical variables.
The prevalence of OSA was identified in selected populations.
Predictors of OSA (at varying levels of severity) were identified
using multiple logistic regression.
nical inadequacy was complete or partial absence of data, from
both the flow channel and the oximetry channel. The second
most common reason was a study of short duration, (defined
as < 4 h). In most cases, short studies were not due to equipment failures, but the patient’s decision to abandon the test, or
remove the oximeter probe, or the nasal cannulae, because of
discomfort or difficulty sleeping.
RESULTS
Prevalence of Unrecognized OSA
Of patients who agreed and underwent the initial clinical assessment, 95% attended for overnight testing and 92% completed testing. A total of 1,157 patients were referred to the study,
had an initial workup, and underwent a night of monitoring.
Characteristics of the population are listed in Table 1.
Comments about technical adequacy were provided by the
reporting physicians for 1,098, or 95% of the 1,157 studies performed (Table 1). Of those, 79% were technically good and
an additional 13% were acceptable. Only 7% were technically
unsatisfactory, for whom repeat studies were requested. Review
of the studies for which the physicians had not made a comment at the time, shows that they were similar with respect to
distribution of AHI. All but 7 of the patients eventually had a
technically satisfactory test. The most common reason for tech-
Prevalence of OSA at different levels of severity is shown
in Table 1. Mean AHI for the population was 15.8 ± 14.6 and
mean lowest oxygen saturation was 83% ± 7%. Table 1 also
shows the percentage prevalences within the whole group and
various subgroups, (by severity of AHI) of key clinical features:
diabetes, hypertension, obesity, coronary artery disease, and
sleepiness (Epworth score > 10). As OSA severity increased,
there were highly significant increases in prevalence of diabetes (p < 0.001), hypertension (p < 0.001), coronary artery disease (p = 0.005), and sleepiness (p < 0.001), but not obesity
(p = 0.08). Some subjects were not included in the subgroup
analysis because of missing data as shown below Table 1
(Dropped patients: No BMI: 12, Unacceptable tests: 12, No
ESS: 14, No diabetes/HTN recorded: 10).
683
Table 1—Analysis of results by severity of OSA
Moderate OSA
Total
No OSA (n = 322) Mild OSA (n = 418)
(n = 193)
15.74 (14.64, 16.83) 2.37 (2.24, 2.50)
8.56 (8.29, 8.83)
20.34 (19.75, 20.95)
30.82 (30.42, 31.23) 29.8 (29.06, 30.47) 30.48 (29.79, 31.11) 31.69 (30.71, 32.68)
8.36 (8.05, 8.68)
7.44 (6.91, 7.97)
8.25 (7.73, 8.77)
8.63 (7.87,9.39)
53.13 (52.27, 54.00) 48.48 (46.79, 50.18) 54.45 (53.16, 55.74) 55.16 (53.16, 57.15)
61%
52%
57%
68%
182 (16%)
38 (12%)
61 (15%)
43 (22%)
430 (39%)
89 (28%)
169 (41%)
87 (45%)
529 (48%)
140 (43%)
186 (44%)
101 (52%)
50 (4.5%)
12 (3.7%)
16 (3.8%)
12 (6.2%)
383 (34%)
91 (28%)
138 (33%)
72 (37%)
AHI (/h)
BMI (kg/m2)
ESS
Age (years)
Gender (male) %
Diabetes, n(%)
Hypertension, n (%)
Obese, n (%)
CAD, n (%)
“Sleepiness” n (%)†
Severe OSA
(n = 176)
52.18 (49.5, 54.9)
32.71 (31.69, 33.72)
10.02 (9.21,10.8)
56.30 (54.11, 58.50)
77%
40 (23%)
85 (49%)
102 (58%)
10 (5.7%)
82 (47%)
P trend*
0.93
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
0.08
0.005
< 0.001
Total: n = 1,109. *χ2 test for linear trend. (Dropped patients: No BMI: 12, Unacceptable tests: 12, No ESS: 14, No diabetes/HTN recorded: 10). †Epworth
Sleepiness Scale (ESS) score > 10. CAD, coronary artery disease.
Table 2—Analysis of results by selected risk factors for OSA
Table 3—Management recommendations
Severity of OSA
All subjects
Obese
Diabetes
Hypertension
3-variable*
4-variable†
CAD
“Sleepiness”§
AHI > 5/h
(n = 787)
787 (71%)
389 (74%)
144 (79%)
341 (79%)
66 (81%)
30 (88%)
38 (76%)
292 (76%)
AHI > 15/h
(n = 369)
369 (33%)
203 (38%)
83 (46%)
172 (29%)
38 (46%)
22 (64%)
22 (44%)
154 (40%)
Recommendation
Referral to sleep physiciana
CPAP therapyb
MADc
Repeat studyd
Current caree
Missing
AHI > 30/h
(n = 176)
176 (16%)
102 (19%)
40 (22%)
85 (20%)
14 (18%)
10 (29%)
10 (20%)
82 (21%)
a
For patients with 10/h ≤ AHI ≤ 30/h or ESS > 10 not obviously explained
by AHI. bFor AHI (or desaturation index) ≥ 30 or AHI (or desaturation
index) ≥ 20 plus hypertension and/or type 2 diabetes. cFor patients with
AHI < 10/h plus moderate or worse snoring. dFor technically inadequate
studies. e “Current care” patients, with an AHI < 10/h, were generally
referred back to their general practitioner with a recommendation for
weight loss in cases of obesity and for consideration of a dental device if
snoring was more than “moderate” and was a clinical issue.
*Variable: obese, hypertension, diabetes. †Variable: obese, hypertension,
diabetes, and sleepy. §Epworth Sleepiness Scale score > 10. Total: n = 1,109.
Table 2 shows the analysis of results by selected risk factors for OSA. “Sleepiness” indicates an Epworth Sleepiness
score > 10. Note that the patients in the columns marked
“AHI > 15/h” and “AHI > 30/h” have already been included in
the column “AHI > 5/h”.
Management recommendations are summarized in Table 3.
For the purposes of this study we assumed that the device would
reliably detect OSA when present.19 The study was not intended
to validate the device against home or laboratory PSG, because
that had been previously done by others,16,19 The study was intended to test the feasibility of using a targeted case finding
strategy with the device in a primary care setting.
prevalence of OSA remained high even when cases were identified by a single risk factor.
Table 2 shows that for a similar population among people
with all 3 factors of obesity, hypertension, and diabetes, 81%
could be expected to have some OSA (AHI > 5/h), 46% to have
moderate or severe OSA (AHI > 15/h), and 18% to have severe
OSA. If the same analysis is performed using 4 variables (adding in sleepiness), the prevalence increases to 88% for some
OSA, 64% for moderate or severe OSA, and 29% for severe
OSA. However, due to the lower number of patients with multiple risk factors, there was no advantage in using any specific
risk factor or multiple risk factors in combination over case selection based on a single risk factor alone.
In keeping with the previous literature, we found, incidentally, that the prevalence of OSA was higher than for the rest of the
population, in patients with diabetes who are sleepy (ESS > 10),
but not in patients with diabetes who were not sleepy.21 Also,
as expected, there was a dose-dependent relationship between
OSA severity and hypertension or coronary artery disease.
The implications for the management of patients with a high
risk of having OSA, such as the population that we have chosen
to study, are very significant in terms of both health outcome
and economic effects.
DISCUSSION
Prevalence of OSA
The prevalence of unrecognized OSA was very high: 71%
had an AHI > 5/h, 33% had an AHI > 15/h; and 16% had an
AHI > 30/h. Table 2 shows the prevalence of OSA, divided
into the 3 usual grades of severity by AHI (AHI > 5/h, > 15/h,
> 30/h) for the whole study population and for subgroups of
subjects defined by comorbidity or risk factors, (obesity, diabetes, hypertension, coronary artery disease and sleepiness). The
N = 1,157
388 (33.5%)
260 (22.5%)
182 (15.7%)
71 (6.1%)
197 (17.0%)
59 (5%)
684
Targeted Case Finding for OSA
Hillman’s data showed that untreated OSA patients consume more health related dollars than those who are treated or
do not have OSA, and Sivertsen et al. have shown increased
work disability.22 It can be expected, therefore, that a plan to
identify and treat cases early, would reduce health care costs
and thus be of interest to both the community at large and to
funding organizations.
The most widely quoted prevalence data for OSA comes
from the Wisconsin Cohort Study, which took place over two
decades ago.1 Recent survey data from the National Sleep
Foundation, suggests that the prevalence may be higher, most
likely because of the rising prevalence of obesity. Using the
Berlin Questionnaire in 1,506 respondents, Hiestand et al. in
2006 estimated a prevalence of OSA of up to 25%.23
In a population survey in Spain, Duran et al.24 found the overall prevalence of AHI > 10/h in men aged 30 to 70 to be 19%, in
women 15%. Prevalence increased considerably with age, from
8% in the 30-40 decade to 32% in the 60-70 decade in men, and
from 2% to 26% in women. That population had a mean BMI
in men of 26.2 and in women of 25.1. Bixler at al25 conducted
a telephone survey of 4,364 men and chose from them a stratified random sample, taking progressively higher percentages of
subjects, according to how many of four risk factors for OSA
(snoring, daytime sleepiness, obesity, and hypertension) they
had. They found the prevalence of AHI > 15/h increased from
2% in those with no risk factors to 34% for those with 4 risk
factors, (but including only 2 of our risk factors—obesity and
hypertension). Their prevalence of AHI > 15/h increased from
3% in those aged 20-44, to 13% in those over 65, confirming
aging as a separate risk factor for obstructive sleep apnea syndrome (OSAS).
portable monitors used at home, however, is questionable. Unattended portable monitors have been compared in randomized
controlled trials with laboratory polysomnography, as decision
making tools for the investigation and management of patients
with OSA.31,32 These showed no difference in treatment outcome between patients tested by portable monitors at home and
those tested by polysomnography in the laboratory.32
The statistics presented here assume that all cases of sleep
apnea that were found were OSA, rather than central sleep apnea (CSA). The analysis software calculated a central sleep
apnea index for each study based on a proprietary algorithm
(Resmed Pty Ltd, Sydney, Australia). This was available to the
reviewing physicians, but they placed more emphasis on their
interpretation of the raw data tracings than the CSA index, when
deciding whether events were central or obstructive. Although
tracings from ApneaLink can give strong indications, such as
round rather than flat topped inspiratory flow curves, lack of
snoring, and constant cycle length, which favor a diagnosis of
CSA, polysomnography is usually considered essential to verify that. The number of cases of CSA in the study population is
therefore not known. It is likely very small, however, since only
two cases were being treated for congestive heart failure, and
only three were known to have cerebrovascular disease.
Although the technical inadequacy rate for laboratory polysomnography is lower than that for the portable monitor in this
study, it has been reported as 3%.33 A rate of 7% for a portable
monitor should therefore be very acceptable for most purposes,
given the low unit cost of the test.
We did not evaluate the effect of targeted screening on outcome measures such as CPAP adherence or quality of life. The
goal of the study was to determine the feasibility of a targeted
case finding strategy in high-risk patients using simple criteria
likely to be found in the primary care chart. We also wished to
determine whether there was a sufficiently high prevalence of
patients with unrecognized OSA to justify case finding in the
first place. Given the results of this study, further research is
necessary to determine if targeted case finding results in effective treatment of the identified cases.
3
Critique of the Methods
The prevalence found in a program like this would depend
on criteria for selection. The nature of the recruiting process,
where family physicians were invited to refer patients, made
it impossible to define rigorously the criteria for selection. Although the primary criterion was intended to be the presence
of type 2 diabetes, obesity, or hypertension, a small number of
patients were selected by the family physicians primarily because of symptoms that suggested OSA. (The study itself raised
awareness of sleep apnea by participating physicians). The rate
of discovery of previously unsuspected cases will depend on
awareness on the part of both patients and physicians and on
local resources for assessment and treatment.
Portable monitors, similar to the ApneaLink+O2, used unattended at home, have been assessed by many authors over recent
years.17,26-30 Most have been found to reliably identify patients
with severe OSA and those with minimal OSA as determined
by PSG. In general, agreement with PSG for classifying mild
and moderate OSA (AHI 5-15/h range) is not as good. Although
the very recent study by Gantner et al.19 found good correlation
between ApneaLink+O2 and home PSG for both AHI and oxygen desaturation index (ODI) in the normal to mild OSA range,
they found underestimation of severity by the ApneaLink+O2
in the severe range. Despite that, they reported a high level of
diagnostic accuracy in the moderate to severe range of OSA.19
The clinical importance of finding an exact AHI, or of the differences in results between polysomnography and unattended
Conclusions
The study shows that a simple low cost case finding and
management program, based on unattended home monitoring
for OSA, focused on patients with obesity, hypertension, and
diabetes, can work well in a population with risk factors and comorbidities associated with OSA. The prevalence of unrecognized OSA in this population was high, and many patients stand
to benefit from the discovery and treatment of their disease.
These data support the testing for OSA in high-risk groups,
whether they have traditional symptoms of OSAS or not.
REFERENCES
1. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence
of sleep-disordered breathing among middle-aged adults. N Engl J Med
1993;328:1230-5.
2. Reuveni H, Greenberg-Dotan S, Simon-Tuval T, Oksenberg A, Tarasiuk A. Elevated healthcare utilisation in young adult males with obstructie sleep apnea.
Eur Respir J 2008;31:273-9.
685
3. Hillman D, Murphy A, Pezullo L. The economic cost of sleep disorders. Sleep
2006;29:299-305.
4. Flemons W, Douglas N, Kuna S, Rodenstein D, Wheatley J. Access to diagnosis
and treatment of patients with suspected sleep apnea. Am J Respir Crit Care
Med 2004;169:668.
5. Guilleminault C. Principles and practice of sleep medicine. Philadelphia: Saunders, 1989.
6. Wolk R, Kara T, Somers V. Sleep-disordered breathing and cardiovascular disease. Circulation 2003;108:9-12.
7. Artz M, Young T, Finn L, Skatrud J, Bradley T. Association of sleep-disordered breathing and the occurrence of stroke. Am J Respir Crit Care Med
2005;172:1447-51.
8. Yaggi H, Concato J, Kernan W, Lichtman J, Brass L, Mohsenin V. Obstructive
sleep apnea as a risk factor for stroke and death. N Engl J Med 2005;353:2034-41.
9. Peppard P, Young T, Palta M, Skatrud J. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med
2000;342:1378-84.
10. Logan A. Sleep-disordered breathing and hypertension. Am J Respir Crit Care
Med 2009;179:1082-3.
11. Ip M, Lam B, Ng M, Lam W, Tsang K, Lam K. Obstructive sleep apnea is independently associated with insulin resistance. Am J Respir Crit Care Med
2002;165:670-6.
12. Punjabi N, Ahmed M, Polotsky V, Beamer B, O’Donnell C. Sleep-disordered
breathing, glucose intolerance, and insulin resistance. Respir Physiol Neurobiol
2003;136:167-78.
13. Resnick H, Redline S, Shahar E, et al. Diabetes and sleep disturbance: findings
from the Sleep Heart Health Study. Diabetes Care 2003;26:702-9.
14. Gami A, Pressman G, Caples S, et al. Association of atrial fibrillation and obstructive sleep apnea. Circulation 2004;110:364-7.
15. Johns M. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep 1991;14:540-5.
16. Erman M, Stewart D, Einhorn D, Gordon N, Casal E. Validation of the ApneaLink
for the screening of sleep apnea: a novel and simple single-channel recording
device. J Clin Sleep Med 2007;3:387-92.
17. Flemons W, Littner M, Rowley J, et al. Home diagnosis of sleep apnea: a systematic review of the literature. An evidence review cosponsored by the American Academy of Sleep Medicine, The American College of Chest Physicians,
and the American Thoracic Society. Chest 2003;124:1543-79.
18. Nakano H, Tanigawa T, Ohnishi Y, et al. Validation of a single-channel airflow monitor for screening of sleep-disordered breathing. Eur Respir J
2008;32:1060-7.
19. Gantner D, Ge J-Y, Li L-H, et al. Diagnostic accuracy of a questionnaire and
simple home monitoring device in detecting obstructive sleep apnea in a Chinese population at high cardiovascular risk. Respirology 2010;15:952-60.
20. Flemons W. Clinical practice. Obstructive sleep apnea. N Engl J Med
2002;347:498-504.
21. Ronksley P, Hemmelgarn B, Heitman S, et al. Obstructive sleep apnea is associated with diabetes in sleepy subjects. Thorax 2009;64:834-9.
22. Sivertsen B, Overland S, Glozier N, Bjorvatn B, Maeland J, Mykletun A. The effect of OSAS on sick leave and work disability. Eur Respir J 2008;32:1497-503.
23. Hiestand D, Britz P, Goldman M, Phillips B. Prevalence of symptoms and risk of
sleep apnea in the US population: Results from the national sleep foundation in
America 2005 poll. Chest 2006;130:780-6.
24. Duran J, Esnaola S, Rubio R, Iztueta A. Obstructive sleep apnea-hypopnea and
related clinical features in a population-based sample of subjects aged 30 to 70
yr. Am J Respir Crit Care Med 2001;163:685-9.
25. Bixler E, Vgontzas A, Ten H, Tyson K, Kales A. Effects of age on sleep apnea
in men: I. Prevalence and severity. Am J Respir Crit Care Med 1998;157:144-8.
26. Ayappa I, Norman R, Seelall V, Rapoport D. Validation of a self-applied unattended monitor for sleep disordered breathing. J Clin Sleep Med 2008;4:26-37.
27. Collop N, Anderson W, Boehlecke B, et al. Clinical guidelines for the use of unattended portable monitors in the diagnosis of obstructive sleep apnea in adult
patients. Portable Monitoring Task Force of the American Academy of Sleep
Medicine. J Clin Sleep Med 2007;3:737-47.
28. Iber C, Redline S, Kaplan Gilpin A, et al. Polysomnography performed in the
unattended home versus the attended laboratory setting - Sleep Heart Health
Study methodology. Sleep 2004;27:536-40.
29. Kirk V, Bohn S, Flemons W, Remmers J. Comparison of home oximetry monitoring with laboratory polysomnography in children. Chest 2003;124:1702-8.
30. Zou D, Grote L, Peker Y, Lindblad U, Hedner J. Validation a portable monitoring
device for sleep apnea diagnosis in a population based cohort using synchronized home polysomnography. Sleep 2006;29:367-74.
31. Mulgrew A, Fox N, Ayas N, Ryan C. Diagnosis and initial management of obstructive sleep apnea without polysomnography: a randomized validation study.
Ann Intern Med 2007;146:157-66.
32. Whitelaw W, Brant R, Flemons W. Clinical usefulness of home oximetry compared with polysomnography for assessment of sleep apnea. Am J Respir Crit
Care Med 2005;171:188-93.
33. Berry R, Hill G, Thompson L, McLaurin V. Portable monitoring and autotitration
versus polysomnography for the diagnosis and treatment of sleep apnea. Sleep
2008;31:1423-31.
acknowledgments
The authors thank Ms. S. Coulson for her assistance in preparing this manuscript
for publication and Ms. M. Bennett for assistance with data collection. Work for this
study was performed at Peninsula Sleep Laboratory. Author contributions: Dr. Burgess was involved in study design, data collection and manuscript preparation. He
is the Guarantor for the manuscript. Dr. Havryk was involved in data collection and
manuscript preparation. Mr. Newton was involved in study design, data collection
and analysis. Dr. Tsai was involved in data analysis and manuscript preparation. Dr.
Whitelaw was involved in study design, data analysis and manuscript preparation.
Submitted for publication August, 2012
Address correspondence to: Keith R. Burgess, M.B., Ph.D., Peninsula Respiratory
Group, Frenchs Forest, NSW, Australia; Tel: +61 2 9975 4911; Fax: +61 2 9975 4622;
E-mail: krburgess@optusnet.com.au
This was not an industry supported study. Dr. Burgess has received speaker honoraria from Glaxo Smith Kline. He is the director of Peninsula Health Care Pty Ltd. Dr.
Burgess’ Family Trust has shares in a private sleep laboratory and he does sleep
consultations in one of the regions surveyed. Dr. Whitelaw serves as a consultant for
R’ANA Respiratory Care Group. Mr. Newton is a director of Healthy Sleep Solutions
Pty Ltd., a company that provides home diagnostic services and treatment for sleep
apnea. The other authors have indicated no financial conflicts of interest.
686
Obstructive Sleep Apnea in Patients with End-Stage
Lung Disease
Ayal Romem, M.D., M.H.A.; Aldo Iacono, M.D.; Elizabeth McIlmoyle, M.D.; Kalpesh P. Patel, M.D.; Robert M. Reed, M.D.;
Avelino C. Verceles, M.D.; Steven M. Scharf, M.D., Ph.D., F.A.A.S.M.
Division of Pulmonary and Critical Care Medicine, Department of Medicine, University of Maryland, Baltimore, MD
Objectives: Little is known about the rate of obstructive sleep
apnea (OSA) in patients with end stage lung disease (ESLD).
Given the potential deleterious effect of OSA in these patients,
we assessed the case-rate and severity of OSA and described
associated patient characteristics.
Methods: Retrospective survey of 60 patients with ESLD referred for lung transplantation evaluation. Demographic, polysomnographic, spirometric, and medication utilization data
were extracted and analyzed.
Results: As demographic and polysomnographic data did not
differ between obstructive and restrictive patients, we present
analysis of pooled data. Demographics/physiology: median age
was 58.5 years, 52% males, mean BMI 32.3 kg/m2, 52% obstructive. Sleep variables (all medians): total sleep time (TST)
312 min, sleep efficiency 77%, minimal oxygen saturation 84%,
apnea hypopnea (AHI) 9.7, respiratory disturbance index (RDI)
12.7 events/h of sleep. Sixty-seven percent had RDI > 5; 21%
had RDI between 15 and 30; and 21% had RDI > 30. Periodic
limb movement index ≥ 15/h sleep was present in 21.7%. An
independent positive correlation between DLCO% and RDI was
noted (r = 0.41, p < 0.01). The minimal oxygen saturation was
negatively correlated with the RDI (r = -0.34, p < 0.01). The use
of ACE inhibitors was associated with moderate-to-severe OSA
(odd ratio of 4.67, CI 1.45-15.03; p = 0.017).
Conclusions: In patients with ESLD, organic sleep disorders
are common. Greater severity of OSA was associated with the
higher DLCO% and lower oxygen saturation.
Keywords: End-stage lung disease, sleep apnea, obstructive
lung disease, restrictive lung disease, diffusion capacity, lung
transplantation, sleep disorders, oxygen saturation
Citation: Romem A; Iacono A; McIlmoyle E; Patel KP; Reed
RM; Verceles AC; Scharf SM. Obstructive sleep apnea in
patients with end-stage lung disease. J Clin Sleep Med
2013;9(7):687-693.
S
leep disordered breathing (SDB) describes a group of disorders of respiratory pattern or ventilation during sleep.
Obstructive sleep apnea (OSA) is the most common subtype.1
Prevalence estimates of OSA vary widely, depending upon definition used and population studied. In the general population,
prevalence estimates range from 5% to 22%.2-4
Several reports have assessed the epidemiologic relationship between chronic obstructive pulmonary disease (COPD)
and OSA.5-7 Most data suggest that the prevalence of OSA in
patients with COPD is similar to that of the general population, but previously studied cohorts include very few subjects
with advanced lung disease. Patients undergoing evaluation for
lung transplantation constitute a cohort of well-characterized
subjects with advanced lung disease. Few studies have looked
at the case rate of OSA in patients with ESLD being evaluated
for lung transplantation. In one study of 50 patients with ILD,
there was a high prevalence of OSA (88%).9 Both end-stage
lung disease (ESLD) and OSA have been associated with decreased health-related quality of life (HRQOL) and important
comorbidities.10-12 If the rate of OSA in ESLD patients is substantial, some of the associated changes in HRQOL and comorbidities could be due to the presence of concomitant OSA. In
view of the scarcity of data on the case rate of OSA in patients
with ESLD and a possible association between OSA and multiple comorbidities as well as poor HRQOL, we performed a
retrospective review of patients with ESLD referred to our lung
Brief Summary
Current Knowledge/Study Rationale: The concomitant presence of
organic sleep disorders including sleep disordered breathing (SDB) and
periodic limb movement disorder (PLMD) could impact the quality of life
and prognosis of patients with end-stage lung disease. Currently there
are few reports on the prevalence of SDB and PLMD in such patients
being evaluated for lung transplantation.
Study Impact: This study demonstrates a high prevalence of SDB and
PLMD in patients with end-stage lung disease whether obstructive or
restrictive, in a lung transplant clinic. Patients being evaluated for lung
transplant should be evaluated for organic sleep disorders.
transplant service for evaluation. We hypothesized that OSA is
common in this patient group. We also compared the frequency
of OSA between patients with COPD and those with restrictive
lung disease due to interstitial lung disease (ILD).
METHODS
Study Design and Sample
In this study, we retrospectively reviewed the archived data
of 60 subjects with ESLD referred for initial lung transplantation evaluation to the lung transplant clinic of the University of
Maryland. As part of the clinic protocol, patients being evaluated for lung transplant underwent polysomnography (PSG) in
687
A Romem, A Iacono, E McIlmoyle et al
Data Analysis
the sleep disorders center of the University of Maryland, regardless of preexisting risk factors for OSA. Patients also underwent
pulmonary function testing (PFT) as well as extensive clinical
evaluation. Demographic, polysomnographic, spirometric characteristics, comorbidities, and medication utilization data were
extracted from patient records. Patients with Epworth Sleepiness Scale (ESS) scores ≥ 10 were considered to have excessive
daytime sleepiness.13 Comorbidities presented are those present
in ≥ 20% of the cohort. The University of Maryland School of
Medicine Institutional Review Board approved this protocol.
Categorical variables are reported as counts and percentages, and were analyzed by using the Fisher exact test. For
continuous variables, Gaussian distribution was evaluated by
the Kolmogorov-Smirnov normality test. Normally distributed variables are presented as mean (standard deviation), and
non-normally distributed variables are presented as median
(interquartile range). Between-group comparisons employed
non-paired t-tests for normally distributed variables and a
Mann-Whitney rank sum test for skewed variables. The association between medication use and SDB was assessed with
univariate logistic regression. Bivariate linear regression analyses were used to assess the association between clinical correlates with SDB. The following independent variables were
tested: body mass index (BMI), DLCO%, and PSG measures
of sleep architecture and indices of oxygen saturation. In addition, backward stepwise multivariate regression analyses were
conducted while treating AHI/RDI as a dependent variable. Independent variables were chosen if the bivariate associations
were significant. For all comparisons, a two-tailed p < 0.05 was
considered significant. SigmaPlot 12.0 (Systat Software Inc.,
San Jose, CA) was used for all analyses and graph production.
Measurements
Pulmonary Function Testing
Spirometry (FEV1, FVC; FEV1/FVC ratio), measurement
of static lung volumes (total lung capacity [TLC] by body box
plethysmography) and measurement of diffusion capacity of
the lung for carbon monoxide (DLCO% predicted) by the single-breath technique were performed (Vmax22, SensorMedics,
Yorba Linda, CA, USA), with the patient in the seated position
according to approved standards.14
Subjects with a ventilatory defect (defined as FEV1 < 70%
predicted) were included. Subjects were categorized into obstructive (FEV1/FVC ratio < 70%, TLC > 80% predicted) or restrictive (FEV1/FVC > 70% and TLC < 80% predicted) disease.
RESULTS
Subject characteristics are summarized in Table 1. Sixty
subjects meeting inclusion criteria were identified. Comparison
based on their primary ventilatory defect (obstructive versus
restrictive) showed similar demographic indices (age, gender
distribution, BMI, ESS, and supplemental oxygen use) and major comorbid conditions (Tables 1, 2). The combined patient
group was characterized by a median age of 58.5 years, BMI of
32.3, equally balanced gender distribution (52% male), and a
median ESS of 9. Half the subjects were on home supplemental
oxygen treatment, and 40% used supplemental oxygen during
their overnight polysomnography.
No difference was noted in DLCO% predicted between the
obstructive and restrictive subgroups. Both groups included
mostly patients with severely diminished pulmonary function
as indicated by a low FEV1 < 50% predicted and/or DLCO%
< 40% predicted.19
The use of antihypertensive medications and systemic steroid use was equally distributed between both obstructive and
restrictive patient groups (Table 1). Inhaled medication use was
significantly more common in the obstructive group.
PSG data are summarized in Table 3. None of the variables
differed between the obstructive and restrictive groups. Hence, for
further analysis, we have pooled the data for the entire cohort.
Median sleep efficiency was reduced, with half the patients demonstrating sleep efficiency < 77.3%. Slow wave and REM sleep
were also severely reduced. Sleep onset latency was slightly increased. The median AHI (9.7) and RDI (12.7) were in the mild
range. Forty of 60 subjects (67%) had OSA, as defined by an RDI
> 5/h (Figure 1). Fourteen (23%) had mild OSA and 26 (43%) had
moderate-to-severe OSA (RDI > 15/h). Periodic leg movement
index (PLMI) was elevated (> 15/h) in 13 patients (21.7%), with
no differences between obstructive and restrictive patients. Time
with oxygen saturation below 90% (T90%) was 17.7% ± 22.6%
of total sleep time, with a median minimal saturation of 84%.
Polysomnography
All PSGs included ≥ 6 h of overnight sleep in an American
Academy of Sleep Medicine accredited sleep laboratory. The
PSGs were performed according to commonly accepted clinical standards.15,16 The montage included encephalogram leads
O1A2, O2A1, C1A2, C2A1, F1A2, F2A1; electromyogram
leads for left eye, right eye, submentalis, and leg (left and right
separately), electrocardiogram, and respiratory status measures by nasal airflow (nasal air pressure) and oronasal airflow
(thermistor, used for backup), rib cage and abdominal respiratory effort (respiratory impedance plethysmographs), and pulse
oximetry. Sleep scoring was done in 30-sec epochs according
to the system of Rechtschaffen and Kales,17 as modified by the
2007 AASM scoring manual.18
Respiratory events were scored according to the 2007 AASM
scoring manual.18 Obstructive apneas were scored where there
was a decrease in nasal airflow to < 10% of baseline for ≥ 10 s
with continued respiratory effort. Obstructive hypopneas were
scored as a decrease in nasal airflow by 50% to 90% of baseline accompanied by oxygen desaturation > 4% for 10 s with
continued respiratory effort. Respiratory event-related arousals
(RERAs) were scored as a decrease in airflow by 30% to 90%
accompanied by ≥ 3% decrease in oxygen saturation and/or a
terminal arousal. Severity of SDB was quantified in two ways.
First, we calculated the respiratory disturbance index (RDI),
equal to the sum of apneas, hypopneas, and RERAs per hour of
sleep. Second, we calculated the apnea-hypopnea index (AHI,
equal to the sum of apneas and hypopneas per hour of sleep).
For primary clinical purposes, the severity of OSA was defined
as follows: “mild” = RDI 5-14.9, “moderate” RDI 15-29.9, and
“severe” = RDI ≥ 30. Other PSG diagnoses were scored according to the AASM manual.18 Patients used oxygen during their
PSG testing if ordered by their referring physician.
688
OSA in Patients with Lung Disease
Table 1—Demographic, medication, and pulmonary function testing (PFT) characteristics according to underlying breathing
abnormality
Variables
Age, year
Male/Female
BMI, kg/m2
Supplemental home O2
PSG on supplemental O2
ESS
Medications
ACE-I, %
ARB
Diuretic
β-blocker
Calcium channel blocker
β-agonist inhaler
Anticholinergic inhaler
ICS
Systemic steroids
PFT results
FEV1, %
FVC, %
FEV1/FVC
TLC, %
DLCO, %
Total (n = 60)
58.5 (49.5 to 63)
31/29
32.3 (25.4 to 38.7)
30 (50%)
24 (40%)
9 (6 to 11.5)
Obstructive (n = 31)
58 (49 to 63)
19/12
32.1 (25.3 to 35.9)
16 (52%)
13 (42%)
8 (4, 10)
Restrictive (n = 29)
59 (53 to 62.25)
12/17
32.3 (25.6 to 39.4)
14 (48%)
11 (38%)
10 (7 to 12)
p value
ns
ns
ns
ns
ns
ns
19 (32%)
8 (13%)
24 (40%)
14 (23%)
16 (27%)
36 (60%)
30 (50%)
29 (48%)
21 (35%)
12 (39%)
2 (6%)
10 (32%)
6 (19%)
10 (32%)
25 (80%)
23 (74%)
21 (68%)
9 (29%)
7 (24%)
6 (21%)
14 (48%)
8 (28%)
6 (21%)
11 (38%)
7 (24%)
8 (28%)
12 (41%)
ns
ns
ns
ns
ns
< 0.001
< 0.001
0.007
ns
45.2 ± 17.9
53.1 ± 14.5
63.7 ± 20.1
78.7 ± 33.5
38.9 ± 17.1
37.5 ± 14.9
57.1 ± 13.9
47.5 ± 13.3
102.3 ± 30.6
42.1 ± 18.7
53.1 ± 17.6
48.7 ± 14.1
80.5 ± 9
54.3 ± 11.7
35.3 ± 14.6
< 0.001
0.025
< 0.001
< 0.001
ns
Results are presented as number (%), mean ± SD or median (interquartile range) as appropriate. BMI, body mass index; PSG, polysomnography; ESS,
Epworth Sleepiness Score; ACE-I, angiotensin converting enzyme inhibitor; ARB, angiotensin receptor blocker; ICS, inhaled corticosteroids; FEV1, forced
expiratory volume in 1 second; FVC, forced vital capacity; TLC, total lung capacity; DLCO, diffusion capacity of the lung for carbon monoxide; ns, nonsignificant.
Table 2—Major comorbid conditions grouped by ventilator defect and RDI
% Total (n = 60)
% Obstructive (n = 31)
% Restrictive (n = 29)
% RDI < 5 (n = 20)
% RDI ≥ 5 (n = 40)
Hypertension
68
77
59
60
73
Diabetes Mellitus
25
19
31
10
33
IHD
20
23
17
25
18
Hyperlipidemia
38
39
38
30
43
GERD
25
19
31
30
23
Sinusitis
22
19
24
20
23
Observed differences between restrictive versus obstructive and RDI < 5 versus RDI ≥ 5 were all nonsignificant (χ2 and Fisher exact test as appropriate). RDI,
respiratory disturbance index; IHD, ischemic heart disease; GERD, gastroesophageal reflux disease.
Table 4 summarizes the observed correlations to RDI and
AHI. RDI and AHI correlated positively to both predicted
FEV1% and predicted DLCO% values (p < 0.05 and p < 0.01,
respectively). RDI and AHI were inversely correlated with the
minimal oxygen saturation value (p < 0.01) and the percentage
of time spent in slow wave sleep (p < 0.01). Subgroup analysis
of subjects who had PSG done with and without supplemental
oxygen showed persistence of these associations. Neither the
Epworth score nor any of the other spirometric values correlated with the RDI or AHI. None of the major comorbid conditions
was associated with an RDI ≥ 5/h (Table 2). No association
could be found between the PLMI and the AHI, RDI, or ESS.
Multivariable modeling included a backward stepwise regression using significant predictors from Table 4 was performed.
Only DLCO% and minimum O2 saturation were found to be
significant predictors (p = 0.013 and p = 0.023, respectively)
of RDI (Figures 2 and 3, respectively) and AHI (p = 0.034 and
p = 0.023, respectively). A multi-linear regression model incorporating the above predictors generated the following equations:
RDI = 66.606 + (0.497 × DLCO %) −
(0.792 × minimum nocturnal oxygen saturation)
AHI = 67.223 + (0.421 × DLCO %) −
(0.791 × minimum nocturnal oxygen saturation)
The use of ACE inhibitors was strongly associated with the
occurrence of moderate-to-severe OSA (odd ratio of 4.67 CI
1.45-15.03; p = 0.017). This association was not evident for any
other antihypertensive medications, including angiotensin receptor blockers. Only patients with hypertension received ACE
inhibitors. Also, the association between ACE inhibitors and
the occurrence of moderate-to-severe OSA was independent of
comorbid conditions and BMI.
689
Table 3—Sleep architecture characteristics according to underlying breathing abnormality
Polysomnography variables
Total sleep time, min
Sleep efficiency, %
Latency to sleep, min
Stage 1+2 sleep, %
Slow wave sleep, %
REM sleep, %
Minimum saturation,%
T < 90%
T < 90% > 10% of TST
PLMI
PLMI ≥ 10
PLMI ≥ 15
AHI
AHI > 5
AHI ≥ 15
RDI
RDI > 5
RDI ≥ 15
Total (n = 60)
312 (274, 355)
77.3 (68.8, 88.1)
26 (6.0, 49.8)
70.6 (57.2, 81)
4.1 (0, 18.5)
11.8 (4.9, 16.6)
84 (78, 87)
3.5 (0.4, 25)
23 (38.3%)
1.2 (0, 12.1)
19 (31.7%)
13 (21.7%)
9.7 (2.4, 22.1)
36 (60%)
22 (36.7%)
12.7 (2.9, 25.4)
40 (66.6%)
26 (43.3%)
Obstructive (n = 31)
312 (290, 354)
77.3 (72.9, 88.6)
25.5 (9.6, 42.4)
68.6 (54, 79.8)
6.2 (0, 23.5)
12.1 (7.2, 16.7)
86 (74, 88.5)
4.2 (0.4-25)
13 (41.9%)
1.8 (0, 10.3)
9 (29%)
5 (16/1%)
12.2 (2.8, 20.8)
20 (64.5%)
11 (35.5%)
13.1 (3.4, 24.5)
21 (67.7%)
14 (45.2%)
Restrictive (n = 29)
308 (318, 355)
77.9 (59.9, 84.8)
26 (4.3, 60.6)
76 (62.2, 87.4)
0.7 (0, 15.4)
11 (0, 15.9)
83 (78, 86.3)
3.3 (0/3, 21)
10 (34.5%)
0.2 (0, 17.1)
10 (34.5%)
8 (27.6%)
8.9 (2.4, 24.9)
16 (55.1%)
11 (37.9%)
9.7 (2.7, 31.4)
19 (65.5%)
12 (41.4%)
p value
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
Results are presented as absolute number (% of total) or median (25%-75% interquartile range) as appropriate. T < 90%, time with oxygen saturation below
90%; TST, total sleep time; PLMI, periodic leg movement index; AHI, apnea hypopnea index; RDI, respiratory disturbance index; ns, nonsignificant.
Figure 1—Prevalence of obstructive sleep apnea according
to severity
AHI/RDI ≥ 5
5 ≤ AHI/RDI < 15
15 ≤ AHI/RDI < 30
30 ≤ AHI/RDI
cussion we consider these findings in the light of the currently
available literature.
Few studies on sleep disorders have been directed at patients
with ESLD. The majority (67%) of the patients had obstructive
sleep apnea (OSA) as defined by an RDI ≥ 5 events per hour,
with 23 of 40 (57.5%) subjects diagnosed with moderate-to severe OSA (RDI ≥ 15/h). No difference in the case rate of OSA
was noted between the “obstructive” group and the “restrictive”
group, despite the inherent differences in the underlying respiratory mechanical abnormalities. These facts underscore the
complexity and yet poorly understood etiology of OSA among
subjects with lung diseases.
We note that the median BMI among our patients was high.
Patients with higher BMIs are known to have an increased
incidence of OSA.23 Since we did not have a control group
matched for BMI, we cannot say that the rate of OSA was
higher than would be expected on the basis of BMI alone. The
effect of age on polysomnographic respiratory abnormalities
in healthy individuals was previously reported by Pavlova et
al.,24 who showed a positive correlation between age and RDI
with an average RDI of 12.8/h in the 50-65 year age range.
Thus, it is possible that part of the high prevalence of sleep
disordered breathing reported here is related to the age of the
patients we evaluated. Prior reports of lung disease patients
including COPD, idiopathic pulmonary fibrosis (IPF), and
patients on the waiting list for lung transplantation reported
different frequencies of OSA. Some of these are reviewed
in Table 5. Depending on the definition of sleep disordered
breathing and differences in population base, rates of 14% to
88% have been reported in patients with ESLD. The reasons
for differences between reported prevalence estimates between studies are likely attributable to differences in patient
80
70
60
50
%
40
30
20
10
0
AHI
RDI
OSA as defined by an AHI or RDI ≥ 5 events/h was prevalent among 60%
and 68% of the subjects (n = 60) respectively.
DISCUSSION
In this study of patients with ESLD, we demonstrated a high
rate of sleep fragmentation, OSA, and PLM disorder. Factors
such as BMI and ESS failed to predict OSA in patients with
ESLD. Conversely, higher DLCO% and lower saturation during sleep correlated to higher RDI and AHI. In the ensuing disJournal of Clinical Sleep Medicine, Vol. 9, No. 7, 2013
690
Table 4—Univariate coefficients of correlation between
sleep disordered breathing indices and demographic, spirometric, and polysomnographic variables
RDI
Pearson
correlation
Characteristic
(β)
Age
0.130
BMI
0.255
ESS
0.073
FEV1, %
0.262
DLCO, %
0.367
SWS%
-0.290
Minimal Saturation, % -0.343
T < 90%
0.069
Pearson
correlation
(β)
0.143
0.255
0.065
0.274
0.410
-0.334
-0.341
0.057
p
value
ns
ns
ns
< 0.05
< 0.01
< 0.05
< 0.01
ns
140
120
p
value
ns
ns
ns
< 0.05
< 0.01
< 0.01
< 0.01
ns
100
80
RDI
AHI
Figure 2—Bivariate correlation between RDI and CO
diffusion capacity (DCO)
60
40
20
0
0
BMI, body mass index; ESS, Epworth Sleepiness score; FEV1, forced
expiratory volume in 1 second; DLCO, diffusion capacity of the lung for
carbon monoxide; SWS%, slow wave sleep percentage of total sleep
time; T < 90%, time with oxygen saturation below 90%; ns, nonsignificant.
20
40
60
80
100
DCO % predicted
For linear regression: RDI = -3.653 + (0.626 × DCO%); r = 0.410; p = 0.002.
Figure 3—Bivariate correlation between RDI and minimal
O2 saturation during polysomnography
selection, techniques, scoring criteria, definitions, and severity of lung disease.
The potential to predict the occurrence of OSA in patients
with advanced lung disease would facilitate more effective and
efficient screening of this growing patient population. In our
study, associations were found between OSA (as indicated by
higher AHI/RDI scores) and multiple physiological parameters, including higher DLCO%, lower oxygen saturations during sleep, and percent of total sleep time spent in slow wave
sleep. Though all of the above parameters have been found
to correlate with OSA in previous studies, only DLCO% and
minimal oxygen saturation remained independent predictors in
the multivariate analyses. High rates of nocturnal oxygen desaturation have been described extensively in both COPD and
ILD patients,25,26 being further deranged among those with superimposed OSA. The concept of predicting OSA by overnight
pulse oximetry to monitor severity of nocturnal oxygen desaturation was thoroughly reviewed by Netzer et al.27 Those authors concluded that overnight pulse oximetry is a useful tool
for the screening for OSA, although they did not specifically
consider patients with ESLD. Our study is consistent with the
notion that OSA and ESLD have additive negative effects on
oxygenation, and that indices of oxygenation could be useful
for prescreening patients with OSA being considered for lung
transplant. Interestingly, the association between minimum
saturation and RDI was also present in those patients receiving
oxygen at night.
The reasons for the direct correlation between DLCO% and
OSA are not clear. This could be explained by a number of factors. Keens et al.28 have shown that adding external resistance
during inspiration leads to an increase in DLCO in normal subjects. Therefore, increased inspiratory airflow resistance, typical of OSA patients,29 could have resulted in improved DLCO.
It is possible that increased inspiratory effort associated with increased upper airways resistance leads to increased pulmonary
capillary volume, both related to increases in venous return and
140
120
100
RDI
80
60
40
20
0
40
50
60
70
80
90
100
110
Minimal saturation (%)
For linear regression: RDI = 89.934 − (0.850 × minsat); r = -0.341; p = 0.008.
increased afterloading of the left ventricle,30 hence leading to
improved DLCO with worse OSA. Elevated DLCO has been
reported in adults with increased BMI,31 although these findings are not universal.32 It is also possible that in ESLD capillary destruction (lower DLCO), results in a greater degree of
intermittent hypoxemia, which in turn could lead to increased
respiratory drive (peripheral chemoreceptors), including the abductors of the upper airway—thus providing some protection
from OSA. Hence, with worsening DLCO the severity of OSA
might have been mitigated, as we observed (Figure 2).
Excessive sleepiness as assessed by the ESS did not correlate with severity or the occurrence of OSA. The reasons for the
dissociation between excessive sleepiness and severity of OSA
are not clear. Others have also demonstrated that the ESS is a
poor predictor of the severity of sleep disorders and that the ESS
may not reflect sleepiness measured objectively.33 It is possible
691
Table 5—Previously reported prevalences of OSA in lung disease patients (polysomnographic recordings)
Definition SDB
OSA
Reference
Moderate-severe OSA
Mermigkis et al.20
Lancaster et al.9
Pascual et al.
N
Comment
34
Idiopathic pulmonary fibrosis
50
Idiopathic pulmonary fibrosis
n/a*
mean AHI:
patients 6.1 ± 6
control 4.3 ± 4.2
31
Patients on the lung
transplantation waiting list (17)
+ control (14)
RDI ≥ 10 or TST ≥ 10% with oxygen saturation ≤ 90% in
the presence of awake oxygen saturation ≥ 90%
36%
117†
Awaiting lung transplantation
RDI > 10
22%
RDI > 15
14%
AHI > 5
AHI ≥ 15
AHI > 5
AHI > 15
AHI ≥ 10
21
Malouf et al.22
Sanders et al.5
n/a
Hypopnea
Prevalence SDB
Both AASM 2007 hypopnea
rules (VII.4.A/VII.4.B)
59%
AASM 2007 hypopnea rule
VII.4.B only
88%
Reduction of airflow ≥ 50% +
oxygen desaturation ≥ 4% ±
arousal
15%
68%
1,132
Age > 40, FEV1/FVC < 70
Unattended home PSG
*OSA prevalence not reported. †Emphysema, n = 27; cystic fibrosis, n = 44; ILD (interstitial lung disease), n = 25; bronchiectasis, n = 7; other, n = 14.
that sympathetic hyperstimulation, or anxiety could induce a
“hyperalert” state, thereby diminishing the extent of perceived
sleepiness. In their study of sleep quality in patients with COPD,
Scharf et al.34 found that in spite of a high incidence of selfreported poor sleep quality and insufficient sleep, few patients
had excessive sleepiness as measured by ESS. These authors
speculated that these patients also had a “hyperalert” state.
Intriguingly, we found a significant and substantial positive association between moderate-severe OSA and the use
of an angiotensin-converting enzyme inhibitor (odds ratio
4.67). This association did not exist for other antihypertensive
medication categories. While possibly due to chance, these
findings are in agreement with those of Cicolin et al.35 who,
in a small case series, showed that withdrawal of ACE inhibitors can lead to a significant decrease in AHI. These findings
led the authors to suggest that through inhibition of degradation of bradykinin, ACE inhibitors induced vasodilatation and
plasma extravasation in the upper airway in turn increasing
the propensity for OSA.
We failed to find that the presence of any particular RDI cutoff was predictive of the occurrence of self-reported hypertension. The association between OSA and hypertension is well
established.36 In our cohort, it is possible that confounding factors such as obesity and the presence of lung disease may have
weakened the association between hypertension and OSA, so
that it was not detected in our small cohort.
Finally, the finding of the high prevalence of sleep disordered breathing in patients with end-stage lung disease might
appear surprising. The association between lung disease and
OSA could be related to the release of proinflammatory cytokines due to the underlying lung inflammation with subsequent
edema and swelling of the upper airway, which would increase
the likelihood for OSA. Alternatively, during OSA there is release of cytokines and oxidant stress, which could have accelerated the progression of the underlying lung.37
Our study has several limitations. It is limited by its relatively small size and its retrospective nature. Furthermore, clinical
necessity mandated the use of supplemental oxygen during the
overnight sleep study in a large portion of our study population,
potentially masking hypoxic episodes and leading to underestimation of the true prevalence of nocturnal oxygen desaturation—a critical determinant of hypopnea definition.
In summary, we observed a high case rate of OSA among a
group of patients with advanced lung disease, irrespective of
underlying etiology or subjective symptoms of excessive sleepiness. Diagnosis of OSA could well lead to improved quality of
life, better perioperative management, and possibly improved
course of the underlying disease.38 Additional studies will be
necessary to further understand the true prevalence of OSA
in advanced lung disease and its consequences, as well as the
effects of treatment of OSA. There is a large disease burden
borne by patients with ESLD. Addressing a common and underdiagnosed comorbidity such as OSA would appear to be an
important component of their evaluation. Our study suggests
that OSA is common in patients with ESLD, and therefore
these patients might benefit from OSA screening with overnight
polysomnography.
REFERENCES
1. American Academy of Sleep Medicine Task Force. Sleep-related breathing disorders in adults: recommendations for syndrome definition and measurement
techniques in clinical research. Sleep 1999;22:667-89.
2. Young T, Palta M, Dempsey J, et al. The occurrence of sleep disordered breathing in middle-aged adults. N Engl J Med 1993;328:1230-5.
3. Gottlieb DJ, Whitney CW, Bonekat WH, et al. Relation of sleepiness to respiratory disturbance index: the sleep heart health study. Arch Intern Med
2002;162:893-900.
4. Young T, Peppard PE, Gottlieb DJ. Epidemiology of obstructive sleep apnea. Am
J Respir Crit Care Med 2002;165:1217-39.
5. Sanders MH, Newman AB, Haggerty CL, et al. Sleep and sleep-disordered
breathing in adults with predominantly mild obstructive airway disease. Am J
Respir Crit Care Med 2003;167:7-14.
692
6. Bednarek M, Plywaczewski R, Jonczak L, Zielinski J. There is no relationship
between chronic obstructive pulmonary disease and obstructive sleep apnea
syndrome: a population study. Respiration 2005;72:142-9.
7. Chaouat A, Weitzenblum E, Krieger J, et al. Association of chronic obstructive
pulmonary disease and sleep apnea syndrome. Am J Respir Crit Care Med
1995;151:82-6.
8. American Thoracic Society; European Respiratory Society. Idiopathic pulmonary fibrosis: diagnosis and treatment. International consensus statement. Am J
9. Lancaster LH, Mason WR, Parnell JA, et al. Obstructive sleep apnea is common
in idiopathic pulmonary fibrosis. Chest 2009;3:772-8.
10. Kaplan RM, Ries AL. Health-related quality of life in emphysema. Proc Am Thorac Soc 2008;5:561-6.
11. Swigris JJ, Kuschner WG, Jacobs SS, Wilson SR, Gould MK. Health-related
quality of life in patients with idiopathic pulmonary fibrosis: a systematic review.
Thorax 2005;60:588-94.
12. Moyer CA, Sonnad SS, Garetz SL, Helman JI, Chervin RD. Quality of life in
obstructive sleep apnea: a systematic review of the literature. Sleep Med
2001;2:477-91.
13. Patil SP, Schneider H, Schwartz AR, Smith PL Adult obstructive sleep apnea:
pathophysiology and diagnosis. Chest 2007;132:325-37.
14. American Thoracic Society Standardization of spirometry, 1994 update. Am J
15. Friedman M, Wilson MN, Pulver T, et al. Screening for obstructive sleep apnea/
hypopnea syndrome: subjective and objective factors. Otolaryngol Head Neck
Surg 2009;142:531-5.
16. Olivarez SA, Maheshwari B, McCarthy M, et al. Prospective trial on obstructive
sleep apnea in pregnancy and fetal heart rate monitoring. Am J Obstet Gynecol
2010;202:552.e1-7
17. Rechtschaffen A, Kales A. A manual of standardized terminology, techniques
and scoring system of sleep stages in human subjects. Brain Information Service/Brain Research Institute, University of California, Los Angeles, 1968.
18. Iber C, Ancoli-Israel S, Chesson A, Quan SF. The AASM manual for scoring of
sleep and associated events: rules, terminology and technical specifications, 1st
ed. Westchester, IL: American Academy of Sleep Medicine, 2007.
19. Pellegrino R, Viegi G, Brusasco V, et al. Interpretative strategies for lung function
tests. Eur Respir J 2005;26:948-68.
20. Mermgkis C, Stagaki E, Tryfon S, et al. How common is sleep-disordered breathing in patients with idiopathic pulmonary fibrosis? Sleep Breath 2010;14:387-90.
21. Pascual N, Jurado B, Rubio JM, Santos F, Lama R, Cosano A. Respiratory disorders and quality of sleep in patients on the waiting list for lung transplantation.
Transpl Proc 2005;37:1537-9.
22. Malouf MA, Milross MA, Grunstein RR, et al. Sleep-disordered breathing before
and after lung transplantation. J Heart Lung Transplant 2008;27:540-6.
23. Young T, Peppard PE, Taheri S. Excess weight and sleep-disordered breathing.
J Appl Physiol 2005;99:1592-9.
24. Pavlova MK, Duffy JF, Shea SA. Polysomnographic respiratory abnormalities in
asymptomatic individuals. Sleep 2008;31:241-8.
25. Krachman S, Minai OA, Scharf SM. Sleep Abnormalities and treatment in emphysema. Proc Am Thorac Soc 2008;5:536-42.
26. Agarwal S, Richardson B, Krishnan V, Schneider H, Collop NA, Danoff SK. Interstitial lung disease and sleep: What is known? Sleep Med 2009;10:947-51.
27. Netzer N, Eliasson AH, Netzer C, Kristo DA. Overnight pulse oximetry for sleepdisordered breathing in adults: a review. Chest 2001;120:625-33.
28. Keens TG, Mansell A, Krastins IRB, et al. Evaluation of the single-breath diffusing capacity in asthma and cystic fibrosis. Chest 1979;76:41-4.
29. StaufferJL, Zwillich CW, Cadieux RJ, et al. Pharyngeal size and resistance in
obstructive sleep apnea. Am Rev Respir Dis 1987;136:623-7.
30. Robotham J, Scharf SM. The effects of positive and negative pressure ventilation on cardiac performance. Clin Chest Med 1983;4:161-87.
31. Reisin E, Messerli FG, Ventura HO. Renal hemodynamic studies in obesity hypertension. J Hypertens 1987;5:397-400.
32. Collard P, Wilputte JY, Aubert G, Rodenstein DO, Frans A. The single-breath
diffusing capacity for carbon monoxide in obstructive sleep apnea and obesity.
Chest 1996;110:1189-93.
33. Fong SY, Ho CK, Wing YK. Comparing MSLT and ESS in the measurement of
excessive daytime sleepiness in obstructive sleep apnoea syndrome. J Psychosom Res 2005;58:55-60.
34. Scharf SM, Maimon N, Simon-Tuval T, Bernhard-Scharf BJ, Reuveni H, Tarasiuk
A. Sleep quality predicts quality of life in chronic obstructive pulmonary disease.
Int J Chron Obstruct Pulmon Dis 2010;22;6:1-12.
35. Cicolon A, Mangiardi L, Mutani R, Bucca C. Angiotensin-converting enzyme inhibitors and obstructive sleep apnea. Mayo Clin Proc 2006;81:53-5.
36. Peppard PE, Young T, Palta M, Skatrud J. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med
2000;342:1378-84.
37. Williams A, Scharf SM. Obstructive sleep apnea, cardiovascular disease, and
inflammation – is NF-kB the key? Sleep Breath 2007;11:69-76.
38. Marin JM, Soriano JB, Carrizo SJ, Boldova A, Celli BR. Outcomes in patients
with chronic obstructive pulmonary disease and obstructive sleep apnea-the
overlap syndrome. Am J Respir Crit Care Med 2010;182:325-31.
acknowledgments
The study was performed at the University of Maryland Medical Center.
Address correspondence to: Steven M. Scharf, M.D., Ph.D., Sleep Disorders Center,
Division of Pulmonary and Critical Care Medicine, University of Maryland School of
Medicine, 685 West Baltimore Street, MSTF 800, Baltimore, MD 21201-1192; Tel:
(410) 706-4771; Fax: (410) 706-0345; E-mail: sscharf@medicine.umaryland.edu
693
Inter-Observer Reliability of Candidate Predictive
Morphometric Measurements for Women with Suspected
Obstructive Sleep Apnea
John A. Gjevre, M.D., M.Sc.1; Regina M. Taylor-Gjevre, M.D., M.Sc.2; John K. Reid, M.D., F.A.A.S.M.1;
Robert Skomro, M.D., F.A.A.S.M.1; David Cotton, M.D.1
1
Division of Respiratory, Critical Care and Sleep Medicine, and
Department of Medicine, University of Saskatchewan, Saskatoon, Saskatchewan, Canada
2
Objective: Obstructive sleep apnea (OSA) is increasingly recognized as a public health concern. Definitive diagnosis is by
overnight polysomnographic (PSG) examination. Identification
of clinical predictors would be beneficial in helping prioritize
high-risk patients for assessment. Practical application of morphometric predictive variables would require a high level of reproducibility in a clinical setting. In this study, our objective was
to evaluate reliability between observers in measurements of
candidate morphometric parameters in women.
Design and Methods: This was a prospective study of
71 women who had been referred for PSG with suspected OSA. Selected morphometric parameters were measured independently in the sleep laboratory by two trained
sleep physicians.
Results: Neck circumference and truncal measurements
for lower costal, midabdominal, and hip circumferences had
higher reliability coefficients (intraclass correlation coefficients
[ICC] of 0.78, 0.95, 0.95, and 0.81) than the smaller dimension
measurements, including cricomental distance or retrognathia (ICC of 0.04 and 0.17). Of the women participating in this
study, 50 of 71 had apnea-hypopnea indexes (AHI) ≥ 5. Body
mass index (BMI), neck circumference, lower costal girth,
midabdominal girth, and hip girth were all significantly higher
(p < 0.001-0.004) in women with AHI ≥ 5.
Conclusions: There was wide variation in inter-observer
reliability for different physical dimensions. We propose that
any clinical morphologic measurement employed in predictive
modeling should be reliably reproducible in clinical setting conditions. Our findings support the use of several truncal measures, BMI, and neck circumference as predictive measures in
women undergoing evaluation for OSA.
Keywords: Women, sleep, body measures
Citation: Gjevre JA; Taylor-Gjevre RM; Reid JK; Skomro R;
Cotton D. Inter-observer reliability of candidate predictive
morphometric measurements for women with suspected obstructive sleep apnea. J Clin Sleep Med 2013;9(7):695-699.
T
here is growing recognition of obstructive sleep apnea
(OSA) as a public health concern, with up to 20% of
adults estimated to have at least mild OSA.1 Untreated, OSA
may pose substantial risks, contributing to development of
systemic hypertension and mortality from cardio/cerebrovascular disease.2 If associated with hypersomnolence, OSA
may contribute to an increase in motor vehicle or workrelated accidents.3,4
The definitive diagnostic tool for OSA is the overnight
polysomnogram (PSG). Increasing awareness of this disorder among both physicians and the general public is leading
to greater numbers of referrals for sleep physician assessment
and for PSG. Timely access to PSG varies by geographic region. Because of PSG access issues, alternative home-based
sleep diagnostic testing has been validated for use.5 Clinical
predictors for significant OSA are useful tools enabling the
physician to prioritize referrals and diagnostic test scheduling
by degree of urgency. In men, increased BMI (body mass index) and neck circumference are strong predictors of OSA.6
However, gender differences in body fat distribution and OSA
associated symptoms have been reported, and it is unclear that
recognized predictors of OSA in men would also carry the
same significance in women.7-10
Brief Summary
Current Knowledge/Study Rationale: Physical parameters may serve
as clinical predictors of obstructive sleep apnea, assisting physicians in
the diagnostic process and in prioritization of patient referrals for polysomnography. A high level of reproducibility of such physical measurements would be required for practical application; this study evaluates
level of agreement between physicians in such measurements.
Study Impact: The results from this study demonstrate wide variation in
inter-observer reliability for different physical measurements. Our findings support the utilization of body mass index, neck circumference and
several truncal measures in predictive modeling for women undergoing
evaluation for obstructive sleep apnea.
There have been efforts made to identify specific morphometric measurements, which could be employed as clinical
predictors for OSA, either alone or in conjunction with other
parameters.11,12
In order for such dimensions to be employed as practical
screening tools, it is crucial to understand the reliability of
measurements in a clinical environment. A physical dimension measurement that in a particular range is predictive for
OSA may prove to be misleading should the measurement itself be subject to substantial variation between physicians. In
695
JA Gjevre, RM Taylor-Gjevre, JK Reid et al
egorical data were evaluated with χ2 testing and Fisher exact
test when the cell size was < 5. Measures of agreement between
observers were calculated for morphometric measurements,
κ coefficients for categorical data, and intraclass correlation coefficients for continuous data.15,16
For this reliability study using intraclass correlations, we
used conventional values for α of 0.05 and β of 0.20. As each
patient underwent separate measurements by 2 different observers, using a minimum acceptable level of agreement of ρ0 = 0.4,
and an expected level of agreement of ρ1 = 0.9, then approximately 7 subjects would be the estimated sample required per
reliability assessment.17
For 2 group comparisons based on an apnea hypopnea index (AHI) cutoff of 5, mean morphometric values derived from
the 2 observers were utilized. Receiver operator characteristic
(ROC) curves were plotted for the predictive relationships between abnormal range AHI (≥ 5) and morphometric parameters.
this study, we address this concern by examination of interobserver agreement in candidate predictive morphometric
measures in women who are undergoing polysomnography
for suspected OSA.
METHODS
Consecutive women scheduled for routine PSG testing for
evaluation of clinically suspected OSA were invited to participate in this study. Informed consent was obtained. This study
was approved by the institutional research ethics board and is
in accordance with the Helsinki Declaration.
Inclusion criteria included age ≥ 21 years and ability to provide informed consent. Exclusion criteria were: referring sleep
physician’s strong suspicion of another primary sleep disorder
(primary insomnia, narcolepsy, restless legs syndrome, a parasomnia, or nocturnal seizures) as indicated on the patient’s
referral form.
Morphometric variables were assessed by standardized, focused upper airway examination and general examination. For
each participant, this was performed by 2 independent sleep
physician observers to assess inter-observer reliability. Measurements included: neck circumference, lower costal or chest
circumference (lower margin of the antero-lateral ribs, while
standing at functional residual capacity [FRC]), umbilical abdominal (mid-abdominal) circumference (peri-umbilical circumference of the abdomen with abdominal muscles relaxed
at FRC while standing), hip circumference (widest circumference of the buttocks while standing), lateral pharyngeal space
narrowing (grading by Tsai et al.11), vertical pharyngeal space
narrowing (modified Mallampati score),12 cricomental space (in
mm), maxillary over jet (in mm), and retrognathia (in mm). The
presence or absence of tongue ridging, macroglossia, and tonsillar enlargement were also recorded.11,13
The pharyngeal grading system described by Tsai included
a 4-class categorization, with class I designated when the palatopharyngeal arch intersects at the edge of the tongue, class
II when the palatopharyngeal arch intersects at ≥ 25% of the
tongue diameter, class III when the palatopharyngeal arch intersects at ≥ 50% of the tongue diameter, and class IV when the palatopharyngeal arch intersects at ≥ 75% of the tongue diameter.11
All 5 sleep physicians assessing patients for this study participated in a standardized training session on morphometric
measurement techniques prior to study initiation.
Patients were studied overnight in the sleep lab using the
standard 15-channel PSG (Sandman Elite version 8.0 sleep diagnostic software, Ottawa, Canada). Established protocols were
used for all PSG studies.14 This included electroencephalogram
(EEG, 3-channel), electrooculogram (2-channel), electromyogram (chin and leg), electrocardiogram, heart rate, snoring,
thermistor airflow, nasal pressure airflow, oxygen saturation,
chest wall motion, and abdominal motion.
RESULTS
Of 95 consecutive female patients who attended the sleep
lab during the study period, 71 consented to participate. The
means and standard deviations of morphometric measurements
by 2 physicians and the measures of agreement are detailed in
Table 1. A greater degree of inter-observer agreement, as represented by the intra-class correlation coefficients and κ coefficients,15,16 was observed for some measurements compared
to others. The greatest agreement was observed for the truncal
measures of lower costal girth, midabdominal girth, and hip circumferences. The lowest degree of agreement was evident for
the smaller dimension measurements for cricomental distance
and retrognathia. Subjective dichotomous observations for the
presence or absence of tongue enlargement, tongue ridging, or
tonsillar enlargement also had lower measures of agreement between observers.
Of the 71 participants, 50 had AHI ≥ 5. Comparisons of morphometric continuous measurements between the 2 groups are
detailed in Table 2. Mean measurement scores derived from the
2 observers were employed for this 2-group comparison. There
were no significant differences in proportions of participants
designated to have tongue ridging, macroglossia, or tonsillar
abnormalities between those with elevated AHI and those with
normal AHI values. Predictive relationships for abnormal AHI
(≥ 5) with morphometric measures are described in Figures 1
and 2 and Table 3.
DISCUSSION
Although obesity has been linked to OSA and there are
comparable frequencies of obesity between genders, the
prevalence of OSA in women has been lower than in men.1
In men, increased BMI and increased neck circumference
are predictive of OSA.6 It is less clear that these variables
are predictive in women with OSA.8-10,18 Whittle et al. have
demonstrated in magnetic resonance imaging (MRI) studies
of men and women that there are differences in neck fat deposition distribution between the genders and greater overall
soft tissue volume around the airway in men. They speculate that these or other anatomic factors may contribute to the
SPSS v.17.0 was employed for data entry and analysis.
Means and standard deviations were calculated for continuous
data. Proportions were calculated for categorical data. Between
group comparisons of continuous data were evaluated with independent 2-tailed t-tests. Between group comparisons of catJournal of Clinical Sleep Medicine, Vol. 9, No. 7, 2013
696
Morphometric Measurements
Table 1—Comparison of morphometric measurements between observers
Measure
Neck circumference (cm)
Lower costal girth (cm)
Midabdominal girth (cm)
Hip girth (cm)
Retrognathia (mm)
Cricomental distance (mm)
Maxillary overjet (mm)
Lateral pharynx (mm)
Vertical pharynx (mm)
% Tongue ridging
% Tongue enlarged
% Tonsils abnormal
Observer 1, mean (SD)
38.33 (4.76)
100.91 (14.53)
104.85 (18.98)
121.47 (19.22)
5.72 (4.56)
5.45 (5.84)
2.50 (2.01)
2.54 (0.92)
2.32 (0.96)
63.4%
40.8%
9.9%
Observer 2, mean (SD)
38.16 (4.49)
100.25 (15.77)
105.23 (19.84)
120.89 (18.61)
5.00 (5.83)
5.06 (7.05)
2.38 (1.78)
2.61 (0.80)
2.35 (0.95)
76.1%
42.3%
9.9%
Measure of Agreement
0.780
0.952
0.949
0.810
0.174
0.044
0.579
0.327
0.246
0.329*
0.488*
0.363*
SD, standard deviation; cm, centimeter; mm, millimeter. *κ coefficient for categorical variables, intraclass correlation coefficient for continuous variables.
Table 2—Comparison of morphometric measures between groups based on AHI
Measure
Body mass index (kg/m2)
Neck circumference (cm)
Lower costal girth (cm)
Midabdominal girth (cm)
Hip girth (cm)
Retrognathia (mm)
Cricomental distance (mm)
Maxillary overjet (mm)
Lateral pharynx (mm)
Vertical pharynx (mm)
AHI ≥ 5, mean (SD)
39.37 (10.50)
39.39 (4.55)
104.65 (15.18)
110.58 (18.67)
125.07 (18.06)
4.95 (3.61)
4.22 (4.25)
2.30 (1.73)
2.59 (0.76)
2.41 (0.77)
AHI < 5, mean (SD)
30.33 (5.14)
35.54 (2.24)
90.89 (8.96)
91.86 (13.14)
111.93 (14.38)
6.33 (4.80)
7.73 (4.80)
2.88 (1.56)
2.57 (0.56)
2.14 (0.71)
Significance
< 0.001
< 0.001
< 0.001
< 0.001
0.004
0.186
0.003
0.200
0.934
0.175
95% CI
4.230, 13.845
1.766, 5.935
6.664, 20.850
9.772, 27.672
4.277, 21.996
-3.453, 0.684
-5.799, -1.213
-1.479, 0.315
-0.354, 0.384
-0.122, 0.656
SD, standard deviation; cm, centimeter; mm, millimeter; kg, kilogram; m, meter; CI, confidence interval.
Figure 2—ROC curve for prediction of AHI ≥ 5 from upper
airway/mandibular measures
1.0
1.0
0.8
0.8
0.6
0.6
Sensitivity
Sensitivity
Figure 1—ROC curve for prediction of AHI ≥ 5 from truncal
measures
0.4
Source of the Curve
body mass index
neck circumference
lower costal girth
mid-abdominal girth
hip girth
reference line
0.2
0.0
0.0
0.2
0.4
0.6
0.8
Source of the Curve
retrognathia
cricomental
maxillary overjet
lateral pharynx
vertical pharynx
reference line
0.4
0.2
0.0
0.0
1.0
1 - Specificity
0.2
0.4
0.6
0.8
1.0
1 - Specificity
697
JA Gjevre, RM Taylor-Gjevre, JK Reid et al
Table 3—Area under the curve (AUC) for prediction of abnormal AHI
Measure
Body mass index
Neck circumference
Lower costal girth
Midabdominal girth
Hip girth
Retrognathia
Cricomental distance
Maxillary overjet
Lateral pharynx
Vertical pharynx
AUC
0.785
0.785
0.782
0.778
0.715
0.410
0.313
0.396
0.500
0.603
Significance
< 0.001
< 0.001
< 0.001
< 0.001
< 0.004
0.247
0.016
0.183
1.000
0.187
95% CI
0.679, 0.890
0.679, 0.891
0.672, 0.893
0.668, 0.888
0.586, 0.844
0.263, 0.557
0.176, 0.449
0.255, 0.537
0.356, 0.644
0.456, 0.750
CI, confidence interval.
gender disparity in OSA prevalence.7 Identification of clinical predictors for OSA in women would help prioritize PSG
evaluation. Morphometric measures have been examined for
identification of candidate variables to aid in the screening
process. In this group of female patients who had been referred for polysomnography we found significantly different
mean values for a number of measures (BMI, neck circumference, lower costal girth, midabdominal girth, hip girth,
and the cricomental distance) between groups based on AHI
category. However, of these potentially predictive measurements, the cricomental distance has a quite low measure of
inter-observer agreement.
As illustrated by Figure 1 and 2 ROC curves, there is substantial overlap in the predictive relationship between truncal
measures for an abnormal AHI, whereas the upper airway/
mandibular measures have a lesser area under the curve and
are observed to be clustered about the reference line, which
implies lack of predictive contribution. The truncal parameters
each provide a statistically significant predictive measure, with
BMI and neck circumferences having the highest area under
the curve. It is possible that greater predictive capacity may be
achieved by an additive combination of physical measures, or
alternatively by ratio (through adjustment for height as an example). However, identification of such predictive models was
outside the scope of this study.
The extent of inter-observer reproducibility would be expected to influence utilization of any predictive morphometric
parameter. In this study evaluating measure of agreement between observers in a variety of morphometric assessments, we
observe the greatest degree of agreement for truncal dimensions
and neck circumference, and the lowest agreement for smaller
upper airway/mandibular dimensions, such as cricomental distance. We propose that any clinical morphologic measurement
employed in a predictive capacity would need to be one reliably reproduced in clinical settings in order to be of practical
value. Our findings support the use of truncal measures, BMI,
and neck circumference as predictive measures in women undergoing evaluation for OSA.
REFERENCES
1. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence
of sleep-disordered breathing among middle-aged adults. N Engl J Med
1993;32:1230-5.
2. Peker Y, Carlson J, Hedner J. Increased incidence of coronary artery disease in
sleep apnoea: a long-term follow-up. Eur Respir J 2006;28:596-602.
3. Ulfberg J, Carter N, Edling C. Sleep-disordered breathing and occupational accidents. Scand J Work Environ Health 2000;26:237-42.
4. Mulgrew AT, Nasvadi G, Butt A, et al. Risk and severity of motor vehicle crashes
in patients with obstructive sleep apnoea/hypopnoea. Thorax 2008;63:536-41.
5. Gjevre JA, Taylor-Gjevre RM, Skomro R, Reid J, Fenton M, Cotton D. Comparison of polysomnographic and home-based Embletta assessments of obstructive
sleep apnea in Saskatchewan women. Can Respir J 2011;18:271-4.
6. Hoffstein V, Szalai JP. Predictive value of clinical features in diagnosing obstructive sleep apnea. Sleep 1993;16:118-22.
7. Whittle AT, Marshall I, Mortimore IL, Wraith PK, Sellar RJ, Douglas NJ. Neck
soft tissue and fat distribution: comparison between normal men and women by
magnetic resonance imaging. Thorax 1999;54:323-8.
8. Wahner-Roedler DL, Olson EJ, Narayanan S, et al. Gender-specific differences
in a patient population with obstructive sleep apnea-hypopnea syndrome. Gend
Med 2007;4:329-38.
9. Valipour A, Lothaller H, Rauscher H, Zwick H, Burghuber OC, Lavie P. Genderrelated differences in symptoms of patients with suspected breathing disorders
in sleep: a clinical population study using the sleep disorders questionnaire.
Sleep 2007;30:312-9.
10. Millman RP, Carlisle CC, McGarvey ST, Eveloff SE, Levinson PD. Body fat distribution and sleep apnea severity in women. Chest 1995;107:362-6.
11. Tsai WH, Remmers JE, Brant R, Flemons WW, Davies J, Macarthur C. A decision rule for diagnostic testing in obstructive sleep apnea. Am J Respir Crit Care
Med 2003;167:1427-32.
12. Samsoon GLT, Young JRB. Difficult tracheal intubation: a retrospective study.
Anaesthesia 1987;42:487-90.
13. Schellenberg JB, Maislin G, Schwab RJ. Physical findings and the risk for obstructive sleep apnea–The importance of oropharyngeal structures. Am J Respir
Crit Care Med 2000;162:740-8.
14. Iber C, Ancoli-Israel S, Chesson A, Quan S. The AASM manual for the scoring
of sleep and associated events: rules, terminology and technical specifications.
Westchester, IL: American Academy of Sleep Medicine; 2007.
15. Müller R, Büttner P. A critical discussion of intraclass correlation coefficients. Stat
Med 1994;13:2465-76.
16. Rigby AS. Statistical methods in epidemiology. V. Towards an understanding of
the kappa coefficient. Disabil Rehabil 2000;22:339-44.
17. Walter SD, Eliasziw M, Donner A. Sample size and optimal designs for reliability
studies. Stat Med 1998;17:101-10.
18. Dancey DR, Hanly PJ, Soong C, Lee B, Shepard J Jr., Hoffstein V. Gender differences in sleep apnea. The role of neck circumference. Chest 2003;123:1544-50.
698
Morphometric Measurements
acknowledgments
The authors thank Dr. Brian McNab for his contribution to this study and also the
Saskatchewan Health Research Foundation for their support of this work. This study
was supported by a grant from the Saskatchewan Health Research Foundation.
Address correspondence to: John A. Gjevre M.D., M.Sc., FRCP(C), Division of
Respiratory, Critical Care and Sleep Medicine, Department of Medicine, University of
Saskatchewan, Saskatoon, SK, Canada, S7N 0W8; Tel: (306) 966-8299; Fax: (306)
966-8694; E-mail: john.gjevre@usask.ca
699
Validating Actigraphy as a Measure of Sleep for
Preschool Children
Marie-Ève Bélanger, B.Sc.1; Annie Bernier, Ph.D.1; Jean Paquet, Ph.D.2; Valérie Simard, Ph.D.3; Julie Carrier, Ph.D.1,2
Department of Psychology, Université de Montréal, Canada; 2Center of Advanced Research in Sleep Medicine,
Hôpital du Sacré-Coeur de Montréal, Canada; 3Department of Psychology, Université de Sherbrooke, Canada
1
Study Objectives: The algorithms used to derive sleep variables from actigraphy were developed with adults. Because
children change position during sleep more often than adults,
algorithms may detect wakefulness when the child is actually
sleeping (false negative). This study compares the validity of
three algorithms for detecting sleep with actigraphy by comparing them to PSG in preschoolers. The putative influence of
device location (wrist or ankle) is also examined.
Methods: Twelve children aged 2 to 5 years simultaneously
wore an actigraph on an ankle and a wrist (Actiwatch-L, MiniMitter/Respironics) during a night of PSG recording at home.
Three algorithms were tested: one recommended for adults
and two designed to decrease false negative detection of sleep
in children.
Results: Actigraphy generally showed good sensitivity
(> 95%; PSG sleep detection) but low specificity (± 50%;
PSG wake detection). Intraclass correlations between PSG
and actigraphy variables were strong (> 0.80) for sleep
latency, sleep duration, and sleep efficiency, but weak for
number of awakenings (< 0.40). The two algorithms designed for children enhanced the validity of actigraphy in
preschoolers and increased the proportion of actigraphyscored wake epochs scored that were also PSG-identified
as wake. Sleep variables derived from the ankle and wrist
were not statistically different.
Conclusion: Despite the weak detection of wakefulness, Actiwatch-L appears to be a useful instrument for assessing sleep
in preschoolers when used with an adapted algorithm.
Keywords: Actigraphy, polysomnography, validation, children
Citation: Bélanger M; Bernier A; Paquet J; Simard V; Julie
Carrier J. Validating actigraphy as a measure of sleep for preschool children. J Clin Sleep Med 2013;9(7):701-706.
S
leep is considered to be of paramount importance for brain
development during the first two years of life.1 In fact, children spend over half of their first two years of life sleeping, with
daily sleep duration decreasing from 14.5 to about 13 hours between 6 months and 2 years of age.2,3 In the preschool years,
daily sleep needs remain high, decreasing from 13 hours at 2
years to about 11 hours at 5 years.2-4 During infancy and childhood, frequent night awakenings or difficulty falling asleep are
among the most frequent developmental complaints. Studies
estimate that from 10% to 75% of parents report that their children have sleep problems.5 Importantly, sleep problems tend to
persist during childhood6 and are associated with several adverse consequences for behavioral, cognitive, and emotional
health. For example, it has been shown that sleep problems
are associated with behavioral and emotional self-regulation
problems.7 In fact, results suggest that when the sleep of preschoolers is insufficient or fragmented by wakefulness, they
show more difficulty inhibiting emotional responses and more
frequent impulsive and aggressive behavior.8 Poor sleep quality also seems associated with obesity in preschool children.9
Studies further suggest that sleep problems, whether occurring
in infancy10 or at school age,4,11 are associated with lower cognitive performance. In light of the prevalence and the serious
consequences of pediatric sleep problems, it is essential to accurately measure sleep quality in young children.
Studies and clinicians use different methods to assess children’s sleep, each presenting strengths and weaknesses. For
Brief Summary
Current Knowledge/Study Rationale: The algorithms used to derive
sleep variables from actigraphy were developed with adults. Because
children change position during sleep more often than adults, algorithms
may detect wakefulness when the child is actually sleeping (false negative). This study compares the validity of three algorithms for detecting
sleep with actigraphy by comparing them to PSG in preschoolers.
Study Impact: Despite the weak detection of wakefulness, ActiwatchL appears to be a useful instrument for assessing sleep in preschoolers when used with an adapted algorithm. However, further studies are
needed to validate its ability to detect wakefulness in pediatric populations with sleep disturbances.
example, parental retrospective child sleep questionnaires and
prospective sleep diaries are often criticized because parents
can notice that their children awaken only when the children
signal it.12 These measures are also influenced by the reporter’s
perception (usually the mother).13 Videosomnography and direct behavioral observations are often used in home settings, but
they may interfere with family routines and privacy.13 Although
polysomnography (PSG) is the gold standard for measuring
sleep,13,14 it requires considerable equipment and technical resources. Furthermore, this invasive method may interfere with
sleep, and therefore mask habitual sleep quality.13
Actigraphy, which uses a watch-size movement sensor to
determine sleep and wake episodes, provides a useful alternative: the device is small and inexpensive, and it allows for
701
M Bélanger, A Bernier, J Paquet et al
METHODS
Figure 1—Linear regression between activity counts at the
wrist and ankle
800
Subjects
Twelve children (4 boys, 8 girls) aged from 2 to 5 years
(M = 3.1, SD = 1.0) participated in this study. None had sleep
problems, according to their parents. The project was approved
by the institutional review board of the investigators’ university. The parents of all participants signed a consent form that
informed them on the nature and risks of participating, and they
received financial compensation for the study.
Y = 0.513x
R2 = 0.9955
700
600
Wrist
500
400
Procedures
300
Children simultaneously wore an actigraph (Actiwatch-L,
Mini-Mitter/Respironics) on the non-dominant ankle and wrist
during a night of PSG recording in their home. A qualified technician and a research assistant went to the homes 1 h before
each child’s usual bedtime (as previously reported by the parents over the phone) to install the PSG and actigraphy recording equipment. No child refused to wear the equipment. Precise
synchronization between the PSG and actigraph is required to
assess epoch-by-epoch concordance. Prior to each sleep recording, the PSG and the actigraphs were precisely synchronized
with the main server. PSG records and actigraphy activity
bursts were then visually inspected to detect any temporal gaps
between the 2 measures. Once the child was asleep, the technician and the research assistant left the home. The research
assistant returned in the morning to remove the electrodes and
bring the equipment back to the laboratory.
200
100
0
0
250
500
750
1000
1250
1500
Ankle
Equation and percentage of fit are also illustrated.
multiple-day data collection. Actigraphy is also easily used in
a child’s natural environment, thereby conferring ecological
validity to collected sleep data. However, the standard algorithms proposed in the literature were developed with adults,
and have not been definitively validated with children. Because
it has been shown that children change position during sleep
more often than adults,15 it is crucial to develop child-specific
algorithms. Sitnick et al.19 showed that an algorithm commonly
used with adults is too sensitive with a population of young
children, resulting in high false negative rates (i.e., actigraphy
detects wakefulness when the child is probably sleeping). In
fact, although several studies with infants and children have
reported that various actigraphy devices are highly correlated
(> 80%) with PSG or videosomnography,14,16-23 most of these
studies have shown very low ability to correctly identify wakefulness14,17,19,22 and hence, sleep fragmentation.24 Nevertheless,
the American Academy of Sleep Medicine (AASM) states that
the use of actigraphy in normal children and special pediatric
populations is indicated for the assessment of sleep patterns and
response to treatment.25,26 However, the AASM also mentions
that additional research is warranted to further refine and broaden the clinical utilization of actigraphy. Notably, additional research is needed to validate actigraphy against PSG.
Sitnick and colleagues proposed adapting an adult algorithm
to reduce false negatives when the child is sleeping but shows
high activity.19 However, this algorithm has not been validated
in preschoolers against PSG. The present study aims to compare the validity of three algorithms for detecting sleep with
actigraphy by comparing them to PSG in preschoolers.
Three algorithms are tested: one recommended for adults and
two designed to decrease false negative detections of wakefulness. Because preschool is a transition period between infancy
(when the actigraph is worn on the ankle because the wrist is
too small for most devices) and school age (when the device is
most frequently worn on the wrist), we also examined the putative influence of device location (wrist or ankle).
Measures
Actigraphy
Non-dominant wrist and ankle activity were recorded using an Actiwatch-L (Mini Mitter Co., Inc., Respironics, Inc.,
Bend, OR). Actigraphy data were collected in 30-s epochs.
Two Actiwatch-L activity monitors were used. The same
monitor was used on the wrist or the ankle for all children. To
calibrate the 2 monitors, they were fixed to a piece of wood
(3/4” × 3” × 12”) that rotated on a vertical axis at 15 different intensities. Estimated activity counts differed between the
2 actigraphs even if induced movements at the 15 intensities were identical for the 2 actigraphs (see Figure 1). Consequently, a regression was used to adjust the monitor with
higher activity counts (ankle) to activity counts of the other
(wrist; y = 0.513x). Both raw and adjusted data are presented
in this paper.
PSG Recordings
A digital ambulatory sleep recording system (Vitaport-3
System; TEMEC Instruments, Kerkrade, The Netherlands) was
used to record sleep at home. Electroencephalograph (EEG)
electrodes (Cz, Oz) were placed according to the international
10-20 system, using a referential montage with linked ears,
right and left electrooculogram (EOG), and chin electromyogram (EMG). EEG signals were filtered at 70 Hz (low pass)
with 1-s time constant and digitized at a sampling rate of 256
Hz. Sleep stages were scored visually on-screen with 30-s epochs (Stellate System, Montreal) according to the AASM cri702
Validating Actigraphy with Preschoolers
teria, but using only the C4 derivation. The 30-s epochs were
chosen to match the 30-s actigraphy epochs.
Four sleep variables were calculated, with the same definitions for PSG and the 3 actigraphy scoring algorithms. The
sleep variables derived from PSG, from the 2 threshold-based
method algorithms (ACT40 and ACT80) and from the smoothed
algorithm (AlgoSmooth), were calculated using an in-house visual C++ program. Sleep latency was defined as the number
of minutes from the time of lights off to the first 10 successive
sleep epochs (the default criterion for the Actiware program).
Total sleep time (TST) was the number of minutes scored as
sleep from lights off to lights on. The number of awakenings
was equal to the number of wake periods. Sleep efficiency (SE)
was TST/total recording time * 100.
27
Data Analysis
Two sets of analyses were performed to determine PSG and
actigraphy agreement: an epoch-by-epoch agreement analysis
and a sleep variables concordance analysis. The epoch-by-epoch agreement analysis provided sensitivity, specificity, accuracy, and negative predictive value (NPV) parameters. Sensitivity
was defined as the proportion of all epochs scored as sleep by
PSG that were also scored as sleep by actigraphy. Specificity
was the proportion of all epochs scored as wake by PSG that
were also scored as wake by actigraphy. Accuracy was the proportion of all epochs correctly identified by actigraphy. NPV
was the proportion of epochs scored as wake by actigraphy that
were also scored as wake by PSG. The second set of analyses
involved comparisons between sleep variables estimated with
PSG and with actigraphy.
Three methods of scoring the actigraphy-derived sleep/
wake activity counts were applied. The first 2 were thresholdbased method algorithms included in the Actiwatch-L software
(Mini Mitter Inc. Respironics, Inc. Bend, OR). Actiware uses
a weighting algorithm with 3 different thresholds: low (20),
medium (40), and high (80), which were validated on sleep
disordered patients. They score original activity counts by a
weighting scheme that reflects the temporal distance relative to
the scored epoch. Each 30-s epoch is rescored as follows:
Statistical Analyses
Two-way repeated measures ANOVAs with activity type
(ankle, raw wrist, and adjusted wrist) and algorithm (ACT40,
ACT80, and AlgoSmooth) as factors were performed on sensitivity (ability to detect PSG sleep), specificity (ability to detect
PSG wake), accuracy (PSG sleep and PSG wake), and NPV
(percentage of wake detected by actigraphy that is PSG wake).
Similarly, 2-way repeated measures ANOVAs with activity
type (ankle, raw wrist, and adjusted wrist) and scoring method (PSG, ACT40, ACT80, and AlgoSmooth) as factors were
performed on sleep variables. Simple effect analyses were performed when significant activity type by algorithm interactions
were found. The post hoc Tukey HSD test was used for multiple comparisons of means on significant main effects. Since
repeated measures had more than 2 levels, the Huynh-Feldt correction for sphericity was applied, but epsilon values and original degrees of freedom are reported. The Dunnett post hoc test
was also used to determine whether the results derived from the
actigraphy algorithms differed significantly from the PSG-derived results. Finally, to assess PSG and actigraphy agreement,
intraclass correlations were computed on the 4 sleep variables.
Statistical analyses were conducted using SPSS version 17
(SPSS Inc., Chicago, IL). Significance level was set at 0.05.
A = 0.04E-4 + 0.04E-3 + 0.2E-2 + 0.2E-1 + 2E0 + 0.2E+1 + 0.2E+2 + 0.04E+3 + 0.04E+4
where A = the sum of activity counts for the 30-s scored epoch
and the surrounding epochs; and En = the activity counts for
the previous, successive, and scored epoch. If the summed activity count exceeds the defined threshold, the epoch is scored
as wake; otherwise it is scored as sleep. The 40 (ACT40)
and 80 (ACT80) activity count thresholds were used in the
present study because the ACT40 is widely used with adult
populations, whereas the ACT80 requires more movement to
score an epoch as wake (and thus could presumably be more
appropriate for children, who move more frequently than
adults when asleep).
The third actigraphy scoring method (AlgoSmooth) used
in the current study is described in a paper by Sitnick and
colleagues19 and has never been validated with PSG. These
authors rescored or secondarily “smoothed” actigraphy data
derived from the ACT40 sensitivity threshold to reduce the
number of awakenings per night to a range more consistent
with parent diaries and video recordings. More precisely, this
method requires a minimum 2-min awakening period following sleep onset (WASO) to score an awakening. The scoring
criteria are:
1. When ≥ 2 consecutive minutes with activity counts >
100 were immediately preceded by any activity count
above 0, that epoch was considered the start of the
awakening;
2. The end of a wake period, or a return to sleep, was
scored at the first of 3 consecutive 0s (no activity).
This third method was automated using an Excel (Microsoft,
Redmond, WA) spreadsheet.
RESULTS
Epoch-by-Epoch Agreement
Sensitivity, specificity, accuracy, and NPV values (means
and SD) derived from epoch-by-epoch comparisons between
each actigraphy scoring algorithm and PSG for the 3 activity
types are presented in Table 1. Overall, sensitivity was higher
than 88%, whereas specificity was lower (from 57% to 81%).
A 2-way repeated measures ANOVA performed on sensitivity revealed an interaction between algorithm and activity
type, F4,44 = 13.02, p < 0.001; ε = 0.63. AlgoSmooth showed
the highest sensitivity and ACT40 the lowest, with ACT80 in
between, but these differences were more pronounced for adjusted wrist data. A significant interaction between algorithm
and activity type was also found for specificity, F4,44 = 4.08,
p = 0.045, ε = 0.38. ACT40 showed the highest specificity and
AlgoSmooth showed the lowest, with ACT80 in between, for
both adjusted wrist and ankle data. For raw wrist data, ACT40
also showed the highest specificity, but specificity did not differ between ACT80 and AlgoSmooth activity type. The 2-way
repeated measures ANOVA performed on accuracy revealed an
703
Table 1—Means (± SD) for sleep sensitivity, specificity, accuracy, and NPV of epoch-by-epoch comparisons with PSG of the
three actigraphy scoring algorithms with the three activity types
Scoring algorithms
Statistical parameter
Sensitivity (%)
Activity type
Ankle
Raw wrist
Adjusted wrist
ACT40
90.5 (2.8)
92.7 (2.7)
87.9 (2.7)
ACT80
95.0 (1.8)
95.8 (1.9)
93.4 (1.6)
AlgoSmooth
97.6 (2.1)
98.7 (1.3)
97.7 (1.6)
Specificity (%)
Ankle
Raw wrist
Adjusted wrist
75.1 (19.2)
69.9 (16.4)
81.0 (14.8)
65.0 (18.8)
56.7 (18.4)
70.9 (16.3)
57.7 (26.3)
58.7 (21.1)
61.2 (21.1)
Accuracy (%)
Ankle
Raw wrist
Adjusted wrist
89.3 (3.5)
90.7 (3.0)
87.5 (2.8)
92.1 (2.9)
92.4 (2.6)
91.4 (2.1)
94.6 (2.9)
95.6 (2.0)
95.0 (2.2)
NPV (%)
Ankle
Raw wrist
Adjusted wrist
41.7 (13.7)
47.4 (15.8)
39.4 (13.3)
53.7 (15.3)
55.0 (18.8)
49.6 (14.7)
76.8 (17.0)
81.0 (13.0)
72.1 (14.1)
ACT40, Actiware medium threshold algorithm; ACT80, Actiware high threshold algorithm; AlgoSmooth, Sitnick et al.’s smoothing algorithm; NPV, negative
predictive value.
Table 2—Sleep parameters (mean ± SD) scored with PSG and estimated by the three actigraphy scoring algorithms with the
three activity types
Scoring algorithms
Sleep parameters
Sleep latency (min)
Activity type
Ankle
Raw wrist
Adjusted wrist
PSGa
34.5 (20.2)
ACT40
31.9 (22.3)
34.3 (23.8)
34.3 (21.5)
ACT80
29.3 (22.2)
33.0 (24.0)
32.3 (22.2)
AlgoSmooth
28.9 (21.8)
29.9 (22.1)
32.7 (21.5)
Total sleep time (min)
Ankle
Raw wrist
Adjusted wrist
558.8 (49.2)
518.7 (49.0)**
533.8 (49.5)**
500.7 (48.2)**
549.3 (49.5)
558.5 (50.6)
537.3 (50.0)**
564.0 (50.7)
571.8 (54.0)**
565.1 (54.0)
Sleep efficiency (%)
Ankle
Raw wrist
Adjusted wrist
90.9 (3.5)
84.4 (4.9)**
86.8 (3.9)**
81.5 (4.3)**
89.4 (4.3)
90.8 (3.2)
87.4 (3.5)**
92.1 (5.1)
92.8 (3.9)*
91.9 (4.1)
Number of awakenings
Ankle
Raw wrist
Adjusted wrist
23.0 (9.8)
59.0 (9.5)**
60.0 (13.5)**
58.3 (11.0)**
51.1 (12.5)**
49.1 (14.5)**
57.3 (14.1)**
3.1 (2.0)**
2.2 (1.3)**
3.3 (1.4)**
a
Activity type does not apply to PSG. ACT40, Actiware medium threshold algorithm; ACT80, Actiware high threshold algorithm; AlgoSmooth, Sitnick et
al.’s smoothing algorithm. Asterisks denote a significant difference (*p ≤ 0.05 and **p ≤ 0.01) between the sleep parameters derived from PSG and from
actigraphy algorithms.
sented in Table 2. Sleep latency derived from the 3 algorithms
did not differ significantly from PSG. A significant interaction between algorithm and activity type was found for TST
(F6,66 = 10.81, p < 0.01; ε = 0.62) and for SE (F6,66 = 13.42,
p < 0.001; ε = 0.78). Dunnett post hoc results showed that the
ACT40 algorithm underestimated TST by > 25 min and SE by
> 4% (p < 0.001) compared to PSG for the 3 activity types
(ankle, raw wrist, and adjusted wrist). TST and SE derived from
ACT80 and AlgoSmooth did not differ significantly from PSG
when using ankle or raw wrist data. However, when using adjusted wrist data, ACT80 underestimated TST by 21 min and
SE by 3.5% (p < 0.001), whereas AlgoSmooth overestimated
TST by 6 min and SE by 1% (p < 0.001). Finally, a 2-way repeated measures ANOVA performed on number of awaken-
interaction between algorithm and activity type, F4,44 = 5.92,
p = 0.003; ε = 0.70. AlgoSmooth showed the highest accuracy and ACT40 the lowest, with ACT80 in between, but these
differences were more pronounced for adjusted wrist data. A
2-way repeated measures ANOVA performed on NPV showed
an algorithm effect only, F2,22 = 33.0, p < 0.001, ε = 0.52.
Post hoc mean comparisons showed significant differences
(p < 0.05) among the 3 algorithms, with AlgoSmooth showing the highest NPV (AlgoSmooth: 76.6%; ACT80: 52.7%; and
ACT40: 42.8%).
Sleep Variable Concordance
Sleep variables calculated from PSG and estimated with the
3 actigraphy scoring algorithms for the 3 activity types are preJournal of Clinical Sleep Medicine, Vol. 9, No. 7, 2013
704
Validating Actigraphy with Preschoolers
Table 3—Intraclass correlations between sleep parameters (mean ± SD) scored with PSG and estimated by the three actigraphy
scoring algorithms with the three activity types
Scoring algorithms
Sleep parameters
Sleep latency (min)
Activity type
Ankle
Raw wrist
Adjusted wrist
ACT40
0.83*
0.92*
0.96*
ACT80
0.85*
0.92*
0.96*
AlgoSmooth
0.81*
0.75*
0.93*
Total sleep time (min)
Ankle
Raw wrist
Adjusted wrist
0.94*
0.94*
0.94*
0.95*
0.95*
0.97*
0.91*
0.98*
0.98*
Sleep efficiency (%)
Ankle
Raw wrist
Adjusted wrist
0.80*
0.73*
0.76*
0.81*
0.70*
0.82*
0.77*
0.93*
0.90*
No. of awakenings
Ankle
Raw wrist
Adjusted wrist
0.01
0.42
0.28
0.09
0.32
0.36
0.14
0.05
0.36
ACT40, Actiware medium threshold algorithm; ACT80, Actiware high threshold algorithm; AlgoSmooth, Sitnick et al.’s smoothing algorithm. *p ≤ 0.01.
ings showed an algorithm effect only, F6,66 = 139.93, p < 0.001,
ε = 0.62, with ACT40 and ACT80 yielding a significantly higher number of awakenings and AlgoSmooth a lower number of
awakenings than PSG (p < 0.001).
Table 3 shows the intraclass correlations between sleep parameters derived from PSG and estimated from the 3 actigraphy
scoring algorithms for the 3 activity types. In general, the correlation coefficients were high for all algorithms and activity
types regarding sleep latency (ICC > 0.75), TST (ICC > 0.91),
and SE (ICC > 0.70). In contrast, the correlations were low for
all algorithms and activity types regarding number of awakenings (ICC < 0.42).
suggest that ACT80 and AlgoSmooth performed better overall
than ACT40 in preschoolers. Except for number of awakenings,
ACT80 and AlgoSmooth showed no substantial differences
from PSG, and intraclass correlations were high. Consequently,
the results suggest that ACT80 and particularly AlgoSmooth
should be used with populations of preschoolers. The two Actiware algorithms (ACT40 and ACT80) clearly overestimated the
number of awakenings, whereas AlgoSmooth underestimated
them. These results indicate that number of awakenings is not
a valid indicator of sleep quality when assessed with actigraphy in preschoolers. For this reason, we attempted to adapt the
smoothing algorithm (AlgoSmooth) described by Sitnick and
colleagues19 to increase the number of awakenings detected.
The adapted criteria were: (1) when there was 1 or more consecutive minute(s) with activity counts greater than 100, that
epoch was considered to be the start of the awakening; (2) the
end of a wake period, or a return to sleep, was scored at the first
of 3 consecutive 0s (no activity). This method was automated
using a Matlab function and was applied to the wrist data. The
number of awakenings estimated by this adapted algorithm did
not significantly differ from the number of awakenings derived
from PSG (M = 22.6, SD = 6.0 for the adapted algorithm vs M
= 23.0, SD = 9.8 for PSG, t11 = -0.20, p = 0.85). Unfortunately
this was at the expense of sensitivity and accuracy; these were
significantly lower with AlgoSmooth, which had higher values
(sensitivity: M = 88.4; SD = 4.2, t11 = -5.83, p < 0.001; accuracy:
M = 87.8; SD = 3.6), t11 = -7.11, p < 0.001). Hence compared
to PSG, the adapted algorithm showed lower sleep efficiency
(M = 80.6; SD = 6.2 vs M = 90.9, SD = 3.5; t11 = -9.85, p <
0.001) and reduced sleep duration (M = 495.5; SD = 56.3 vs M
= 558.8, SD = 49.2; t11 = -10.07, p < 0.001), and was therefore
discarded. These results further suggest that actigraphy-derived
number of awakenings is not a valid indicator of sleep quality
with preschoolers.
To our knowledge, most laboratories use actigraphy without calibration. In this study, when similar movements were
DISCUSSION
We evaluated the ability of the Actiwatch-L device to detect
sleep/wake in preschool children using three algorithms. Results
clearly showed that the Actiwatch-L is better able to detect sleep
than to detect wake. Importantly, ACT80 and AlgoSmooth enhanced the ability of actigraphy to detect sleep in preschool children compared to ACT40. However, ACT80 and AlgoSmooth
decreased the ability of actigraphy to detect wakefulness compared to ACT40. The low specificity (about 60% of PSG wakefulness is scored as wakefulness by actigraphy) observed in our
data is similar to that found in previous studies that compared
different brands of actigraphy with PSG in infants14,20 and children,17,21 highlighting the difficulty of correctly identifying wake
with actigraphy. Nevertheless, when the actigraphs scored wake,
AlgoSmooth showed higher agreement with PSG (NPV = 76.6%)
than the other two algorithms (42.8% and 52.7%), suggesting
that AlgoSmooth is better suited for this purpose. Importantly,
the general accuracy of actigraphy to detect sleep and wake remained high, despite the low specificity, probably because most
of the assessed intervals consisted of sleep.
Statistical comparisons between sleep variables derived
from actigraphy and PSG as well as intraclass correlations
705
induced, estimated activity counts differed between the two
actigraphs. The criteria for most algorithms to determine wake
and sleep require absolute activity counts. Thus, for similar
movements, actigraphs with higher activity counts will detect
more wakefulness than those with lower activity counts. This
is reflected in our data by lower sleep efficiency with adjusted
than raw wrist data. Importantly however, when using ACT80
or AlgoSmooth, sleep variables derived from ankle, raw wrist,
and adjusted wrist data were comparable.
Overall, the Actiwatch-L appears to be an effective instrument for assessing sleep in preschoolers. However, further
studies are needed to validate its ability to detect wakefulness
in pediatric populations with sleep disturbances.
18. Sadeh A, Lavie P, Scher A, Tirosh E, Epstein R. Actigraphic home-monitoring
sleep-disturbed and control infants and young children: a new method for pediatric assessment of sleep-wake patterns. Pediatrics 1991;87:494 -9.
19. Sitnick SL, Goodlin-Jones BL, Anders TF. The use of actigraphy to study sleep
disorders in preschoolers: some concerns about detection of nighttime awakenings. Sleep 2008;31:395-401.
20. So K, Buckley P, Adamson TM Horne RSC. Actigraphy correctly predicts sleep
behavior in infants who are younger than six months, when compared with polysomnography. Pediatr Res 2005;58:761-5.
21. Spruyt K, Gozal D, Dayyat E, Roman A, Molfese DL. Sleep assessments in
healthy school-aged children using actigraphy: concordance with polysomnography. J Sleep Res 2010;20:223-32.
22. Sung M, Adamson TM, Horne RSC. Validation of actigraphy for determining
sleep and wake in preterm infants. Acta Paediatr 2009;98:52-7.
23. Ward TM, Lentz M, Kieckhefer GM, Landis CA. Polysomnography and actigraphy concordance in juvenile idiopathic arthritis, asthma and healthy children. J
Sleep Res 2011;22:113-21.
24. O’Driscoll DM, Foster AM, Davey MJ, Nixon GM, Horne, RSC. Can actigraphy
measure sleep fragmentation in children? Arch Dis Child 2010;95:1031-3.
25. Littner M, Kushida CA, Anderson WM, et al. Standards of practice committee
of American Academy of Sleep Medicine. Practice parameters for the role of
actigraphy in the study of sleep and circadian rhythms: an update for 2002. An
American Academy of Sleep Medicine report. Sleep 2003;26:337-41.
26. Morgenthaler TI, Lee-Chiong T, Alessi C, et al. Standards of practice committee
of the American Academy of Sleep Medicine. Practice parameters for the clinical
evaluation and treatment of circadian rhythm sleep disorders. An American Academy of Sleep Medicine report. Sleep 2007;30:1445-59.
27. Iber C, Ancoli-Israel S, Chesson A, Quan S, eds. The American Academy Sleep
Medicine manual for the scoring of sleep and associated events: rules, terminology and technical specifications. Westchester, IL: American Academy of Sleep
Medicine; 2007.
REFERENCES
1. Dahl RE. The regulation of sleep and arousal: development and psychopathology. Dev Psychopathol 1996;8:3-27.
2. Davis K, Parker K, Montgomery G. Sleep in infants and young children: part one:
normal sleep. J Pediatr Health Care 2004;18:65-71.
3. Iglowstein I, Jenni OG, Molinari L, Largo RH. Sleep duration from infancy to adolescence: reference values and generational trends. Pediatrics 2003;111:302-7.
4. Touchette E, Petit D, Séguin JR, Boivin M, Tremblay RE, Montplaisir JY. Associations between sleep duration patterns and behavioral/cognitive functioning at
school entry. Sleep 2007;30:1213-9.
5. Mindell J, Sadeh A, Wiegand B, How TH, Goh DYT. Cross-cultural differences in
infant and toddler sleep. Sleep Med 2010;11:272-80.
6. Kataria S, Swanson MS, Trevathan GE. Persistence of sleep disturbances in
preschool children. J Paediatr 1987;110:642-6.
7. Lavigne JV, Arend R, Rosenbaum D, et al. Sleep and behavior problems among
preschoolers. J Dev Behav Pediatr 1999;20:164-9.
8. Bates J, Viken R, Alexander D, Beyers J, Stockton L. Sleep and adjustment in
preschool children: sleep diary reports by mothers relate to behavior reports by
teachers. Child Dev 2002;73:62-74.
9. Jiang F, Zhu S, Yan C, Jin X, Bandla H, Shen X. Sleep and obesity in preschool
children. J Paediatr 2009;154:814-8.
10. Bernier A, Carlson SM, Bordeleau S, Carrier J. Relations between physiological
and cognitive regulatory systems: infant sleep regulation and subsequent executive functioning. Child Dev 2010;81:1739-52.
11. Sadeh A, Gruber R, Raviv A. The effects of sleep restriction and extension on
school-age children: what a difference an hour makes. Child Dev 2003;74:444-55.
12. Sadeh A, Hauri PJ, Kripke DF, Lavie P. The role of actigraphy in the evaluation
of sleep disorders. Sleep 1995;18:288-302.
13. Sadeh A. Sleep assessment methods. In: M. El-Sheikh, ed. Sleep and development: familial and socio-cultural considerations. New York: Oxford University
Press, 2011: 355-371.
14. Insana SP, Gozal D, Montgomery-Downs HE. Invalidity of one actigraphy brand
for identifying sleep and wake among infants. Sleep Med 2010;11:191-6.
15. de Koninck J, Lorrain D, Gagnon P. Sleep positions and position shifts in five age
groups: an ontogenetic picture. Sleep 1992;15:143-9.
16. Gnidovec B, Neubauer D, Zidar J. Actigraphic assessment of sleep-wake rhythm
during the first 6 months of life. Clin Neurophysiol 2002;113:1815-21.
17. Hyde M, O’Driscoll DM, Binette S, et al. Validation of actigraphy for determining sleep and wake in children with sleep disordered breathing. J Sleep Res
2007;16:213-216.
acknowledgments
This study was supported by funding from the Fonds de Recherche en Santé du
Québec (FRSQ) and the Natural Sciences and Engineering Research Council of
Canada (NSERC). We thank Sonia Frenette (project coordinator), Manon Robert (research assistant), Nicolas Pellerin, and Jonathan Godbout (computer programmers)
for help with data collection and analysis.
Accepted for publication February, 2013
Address correspondence to: Julie Carrier, Ph.D., Department of Psychology, University of Montreal, C.P. 6128, Succursale Centreville, Montreal, Québec, Canada H3C
3J7; Tel: (514) 343-5923; Fax: (514) 338-2531; E-mail: julie.carrier.1@umontreal.ca
706
Heart Rate Variability in Sleep-Related Migraine without Aura
Catello Vollono, M.D., Ph.D.1; Valentina Gnoni, M.D.1; Elisa Testani, M.D.1; Serena Dittoni, M.D.1; Anna Losurdo, M.D.1;
Salvatore Colicchio, M.D.2; Chiara Di Blasi, M.D.1; Salvatore Mazza, M.D.1; Benedetto Farina, M.D., Ph.D.3;
Giacomo Della Marca, M.D., Ph.D.1
Institute of Neurology, Catholic University, Rome, Italy; 2Institute of Psychiatry, Catholic University, Rome, Italy;
3
Faculty of Psychology, European University, Rome, Italy
1
Objectives: This is an observational study aimed to investigate the activity of autonomic nervous system during sleep in
patients with sleep-related migraine.
Methods: Eight consecutive migraineurs without aura were
enrolled (6 women and 2 men), aged 30 to 62 years (mean
48.1 ± 9.3 years). Inclusion criteria were: high frequency of
attacks (> 5 per month) and occurrence of more than 75% of
the attacks during sleep causing an awakening. Patients were
compared with a control group of 55 healthy subjects (23 men
and 32 women, mean age 54.2 ± 13.0 years), and with a further control group of 8 age- and gender-matched healthy controls. Patient and controls underwent polysomnography and
heart rate variability analysis.
Results: A significant reduction of the LF/HF ratio during N2
and N3 sleep stages was observed in migraineurs compared
with controls. No differences in sleep macrostructure were
observed; cyclic alternating pattern (CAP) time and CAP rate
were lower in migraineurs than in controls.
Conclusions: These findings indicate a peculiar modification of
the autonomic balance during sleep in sleep-related migraine.
The reduction of LF/HF ratio in NREM sleep was observed in controls, but it was quantitatively much more evident in migraineurs.
Changes in LF/HF could be consequent to an autonomic
unbalance which could manifest selectively (or alternatively
become more evident) during sleep. These findings, together
with the reduction in CAP rate, could be an expression of reduced arousability during sleep in patients with sleep-related
migraine. The simultaneous involvement of the autonomic,
arousal, and pain systems might suggest involvement of the
hypothalamic pathways.
Keywords: Autonomic nervous system, heart rate variability,
hypothalamus, cyclic alternating pattern, sleep-related migraine
Citation: Vollono C; Gnoni V; Testani E; Dittoni S; Losurdo A;
Colicchio S; Di Blasi C; Mazza S; Farina B; Della Marca G.
Heart rate variability in sleep-related migraine without aura.
J Clin Sleep Med 2013;9(7):707-714.
M
igraine attacks are frequently preceded by premonitory
signs and/or associated with symptoms suggesting the
involvement of the autonomic nervous system. In particular,
symptoms of the prodromic phase1 (irritability, increased sensitivity to sounds, light, and smells) and symptoms and signs of
attacks2 (nausea, vomiting, cutaneous vasoconstriction or vasodilatation, piloerection, sweating) have an autonomic basis.
Several previous clinical studies investigated a possible
dysfunction of the autonomic nervous system in migraineurs,
but a large variety of measures have been used to assess autonomic function, and the results are consequently controversial.3-6 Sympathetic hypofunction,4,5,7-10 sympathetic instability
or hyperfunction,11,12 and/or parasympathetic dysfunction4,8,13
have been described or suggested in migraine. Other authors
reported mild sympathetic hyperactivity in migraineurs, without evidence for an impairment of the autonomic cardiovascular control.14 These contradictory results were probably caused
by many factors that can bias autonomic function tests: age,
weight, gender, test selection, test criteria, conditions of testing,
and patient selection. At present, more than 30 years after the
first description,15 the most widely used method for assessing
the status of the cardiovascular sympatho-vagal balance is represented by spectral analysis of heart rate variability (HRV),16,17
which has the advantage of being noninvasive. This analysis,
through the quantification of low-frequency oscillatory com-
Brief Summary
Current Knowledge/Study Rationale: Migraine, as well as sleep, are
associated with modifications in the activity of the autonomic nervous
system (ANS). The aim of this study was to evaluate the activity of ANS
in a population of patients with sleep-related migraine, by means of heart
rate variability analysis.
Study Impact: In sleep-related migraine, there is a reduction of LF/HF
ratio during N2 and N3 sleep stages, as compared with controls. This
reduction occurs in parallel with the decrease of NREM sleep instability
measured with cyclic alternating pattern. The results suggest that sleeprelated migraine is associated with impairment of EEG and autonomic
arousal mechanisms.
ponents (LF) and high-frequency oscillatory components (HF,
synchronous with the respiratory rate, marker of vagal modulation) is used to estimate the respective role, and the balance, of
the orthosympathetic and the parasympathetic components of
the autonomic nervous system.16,17
Migraine has a close relationship with sleep.18 It is well
known that the onset of migraine attacks can occur during sleep;
conversely, in some patients, sleep may relieve the symptoms
of migraine.18 On the basis of the time of onset of the attacks,
some authors3,19-21 have defined a “sleep-related migraine,” in
which the onset of attacks has a close correlation with sleep,
that is, more than 75% of the attacks occur during sleep. In the
707
C Vollono, V Gnoni, E Testani et al
Table 1—Patient clinical and demographic data
Patient
1
2
3
4
5
6
7
8
Age
62
47
46
49
46
48
30
57
Mean
48.1
9.3
SD
Gender
F
F
F
M
F
F
F
M
Attacks
per month
8
10
10
10
6
8
8
8
Duration
of illness
> 15 years
10 years
7 years
6 months
3 years
1 year
5 years
8 years
Treatment
AHI
(events/h)
3.1
2.0
1.7
0.6
1.2
3.2
0.5
2.1
BMI
(kg/m2)
23.7
22.1
19.8
23.2
22.6
20.3
24.8
24.2
8.5
1.8
22.6
1.4
1.0
1.8
Comorbidity
None
None
None
None
None
Moebius syndrome
None
None
Prophylaxis Symptomatic
No
Triptans
No
Indomethacin
No
Triptans
No
NSAIDs
No
Triptans
No
Triptans
No
Triptans
No
Acetaminophen
AHI, apnea-hypopnea index; BMI, body mass index, SD, standard deviation.
International Classification of Sleep disorders (ICSD 2nd edition) sleep-related migraine can be classified among the sleeprelated headaches (Appendix A: sleep disorders associated with
conditions classifiable elsewhere).22 Nevertheless, this form of
migraine, defined on the basis of relation between sleep and the
onset of attacks, is not coded in the International Classification
of Headache Disorders - II.23
Sleep induces deep modifications in the autonomic output, with a peculiar pattern of circadian and ultradian oscillations.24-26 In fact, patterns of autonomic activity undergo
significant modifications through wake and sleep, through
NREM and REM, and in each specific stage of sleep.25 These
findings were confirmed in quantitative EEG and HRV studies during sleep.27,28 For these reasons, it could be hypothesized
that peculiar modifications of autonomic nervous system activity, occurring during sleep, might facilitate the onset of attacks
of migraine in predisposed subjects.
The aim of the present study was to investigate the modification in the autonomic activity during sleep stages in a selected
group of subjects with a very close relation between migraine
attacks and sleep; this condition, in accordance with some previous reports, was called sleep-related migraine.21 We hypothesized that, in these patients, autonomic modification occurring
during sleep stages could predispose to migraine attacks. In order
to clinically define sleep-related migraine, we analyzed the sleep
diaries and selected patients in whom more than 75% of the migraine attacks occurred during sleep and caused an awakening.
Part of this group of patients was the object of a previous study,
in which we analyzed the pattern of arousal in sleep-related migraine.29 Sleep study was performed by means of nocturnal, laboratory-based polysomnography; sympatho-vagal function was
evaluated by means of heart rate variability analysis.
tional Classification of Headache disorders 2nd edition23 for
migraine without aura as well as the criteria of the ICSD22 for
sleep-related headache. Patients were recruited from the Headache Center of the Catholic University in Rome over a period of
12 months. All outpatients, after the first evaluation, were asked
to fill in a diary of headache episodes for a period of 4 weeks;
this constitutes a standard procedure before defining a diagnosis and starting a prophylactic treatment.23 All the patients underwent a full medical and neurological evaluation, and were
asked to complete a migraine diary for 2 weeks before and 2
weeks after the PSG recording.
Inclusion criteria were: (1) high frequency of attacks (≥ 5
per month) and (2) more than 75% of their attacks during sleep,
causing an awakening. Episodes in which the patients presented
headache on morning awakening but were not directly awakened by the pain were excluded. Other inclusion criteria were:
the absence of prophylactic treatment during the study (no patient received drugs of any kind, chronically, in order to prevent
the onset of migraine attacks) or in the previous 3 months; absence of pharmacological treatment of any kind in the month
prior to the sleep study, with the exception of triptans or nonsteroidal anti-inflammatory drugs (NSAIDs) administered for
the acute treatment of attacks. Exclusion criteria were heart
disease, arrhythmias, or intake of cardiovascular active drugs;
diabetes; uncontrolled hypertension; smoking; obesity; chronic
respiratory disease; thyroid disease; psychiatric disorders; severe head trauma; and previous history of sleep disorders of
any type or of other neurological diseases. The main clinical
data concerning the patients’ group are summarized in Table 1.
Controls
Heart rate variability and polysomnographic data obtained in
patients were compared with data recorded in a control group
of 55 healthy subjects (23 men and 32 women, mean age 54.2 ±
13.0); this population of healthy volunteers was previously enrolled to act as controls in previous sleep studies. The same exclusion criteria applied to the patients’ group were also applied
to the controls. Moreover, as requested in the review process, a
further comparison was performed between the patients and an
age- and gender-matched control group composed of 8 subjects
METHODS
Patients
We enrolled in the study 8 consecutive patients of both
genders (6 women and 2 men), aged between 30 and 62 years
(mean 48.1 ± 9.3 years), fulfilling the criteria of the InternaJournal of Clinical Sleep Medicine, Vol. 9, No. 7, 2013
708
HRV in Sleep-Related Migraine
(6 women and 2 men, mean age 46.7 ± 10.7 years). All patients
and controls gave written informed consent to participate. The
study was performed in agreement with the Declaration of Helsinki and was approved by Ethics Committee of the Catholic
University in Rome.
components can be computed: very-low frequency (VLF), lowfrequency (LF), and high-frequency (HF). The HF component
of the spectrum is widely recognized as a measure of vagal activity, whereas the significance of LF component is more debated, and it seems to reflect at the same time both vagal and
sympathetic activity. Overall, the LF/HF ratio may provide a
quantitative esteem of the balance of the 2 branches of the ANS
(sympatho-vagal balance). For a detailed description of the
heart rate variability analysis, see: Task Force of the European
Society of Cardiology and the North American Society of Pacing and Electrophysiology.17
In the present study, HRV analysis was performed during
quiet wake before sleep and in the following sleep stages: N21C (stage N2, first sleep cycle), N2-LC (stage N2, last sleep
cycle), N3, REM. For each sleep and wake stage, we selected
a single time interval lasting 5 min during which no stage shift
occurred. During these intervals the EKG trace was analyzed.
For the analysis, we selected 5-min periods chosen with the
following criteria: (1) 5 consecutive min of quiet wakefulness
(W) before sleep onset, (2) first 5 consecutive min of stage 2 of
NREM sleep in the first sleep cycle (N2-1C), (3) first 5 consecutive min of stage 2 of NREM sleep in the last sleep cycle (N2CL), (4) first 5 consecutive min of stage 3 of NREM (N3), (5)
5 consecutive min of REM sleep (REM). Stage 1 NREM (N1)
was excluded from the analysis because this state is considered,
by definition, a stage of transition, and it is very unlikely to
observe 5 consecutive min of stable N1 in polysomnographic
recordings. We analyzed separately 2 intervals of N2 because
this sleep stage may have deep differences when it occurs in
proximity of SWS or REM sleep. In particular, it has been described as a progressive decrease in HRV sympathetic indexes
during the transition toward SWS, contrasting with high and
stable levels during N2 that evolves toward REM.36 Periods of
EKG recording containing awakenings, arousals, extrasystoles,
or movement artifacts were excluded from the analysis. We
decided to analyze intervals of 5 min because we needed to
select consecutive epochs of homogeneous recording for each
sleep stage, not interrupted by stage shifts, micro-awakenings,
fast-frequency EEG arousals, body movements, extrasystoles,
or artifacts. Longer intervals of stable EEG and EKG recordings can hardly be observed, in particular during stages N3 and
REM. Moreover, 5 min is the minimal length of EKG recording
during which the signal is supposed to be stationary, thus allowing an accurate estimate of the spectral components with the
autoregressive method.17,35
Artifact rejection was performed visually. Dedicated software (Rembrandt SleepView-Medcare) calculated the RR intervals (tachogram). Another software program was used for
automatic evaluation of heart rate variability parameters (HRV
Analysis Software, Biomedical Signal analysis Group, Dept of
Applied Physics, University of Kuopio, Finland).37
HRV analysis was performed both in the time domain and
in the frequency domain. In the time domain, the parameters
calculated were: RMSSD and NN50; geometric measures: the
NN triangular index, determined from the histogram of RR
intervals, in which NN stands for normal-to-normal intervals
(i.e., intervals between consecutive QRS complexes resulting
from sinus node depolarization), SD1 (standard deviation of the
instantaneous beat-to-beat RR interval variability, minor axis of
Polysomnography
Patients and controls underwent a full-night, attended, laboratory-based nocturnal video-polysomnography. In order to
avoid any influence of acoustic stimuli on sleep,30 patients and
controls slept in a partially soundproof room. Polysomnography
were recorded by a Micromed System (Micromed, Mogliano
Veneto, Treviso, Italy) 98 digital polygraph. Montages included
8 EEG leads applied to the following locations: Fp1, Fp2, C3,
C4, T3, T4, O1, O2; reference electrodes applied to the left (A1)
or right (A2) mastoids; 2 electrooculographic electrodes applied
to the cantus of each eye, surface elect myography of submental and intercostal muscles, airflow measured by oronasal thermocouple, thoracic and abdominal effort, EKG (V2 modified
derivation), and peripheral hemoglobin saturation. Impedances
were kept below 5KΩ before starting the recording, and checked
again at the end of the recording. Sampling frequency was 256
Hz. A/D conversion was made at 16 bit. Pre-amplifier amplitude
range was ± 3,200 μV, and pre-filters were set at 0.15 Hz. Sleep
monitoring lasted from 23:00 to 07:00 the next morning. A technician was present for data acquisition, and video monitoring
was performed throughout the registration.
Sleep Analysis
Sleep stages were visually classified by an expert physician
according to the criteria of American Academy of Sleep Medicine.31,32 The analysis of sleep-related respiratory events was
made visually by an expert scorer, according to the criteria established by the AASM.31,33 Cyclic alternating pattern analysis
was performed according to the standardized criteria.34
Heart Rate Variability Analysis
Heart rate variability analysis is the measure of the variations
of the interval between consecutive heart beats. It is widely accepted that heart rate variability represents a quantitative marker of autonomic activity.17,35 The variations in heart rate may
be evaluated by time domain methods and frequency domain
methods.
The time domain methods are based on the detection of the
QRS in a normal EKG and on the determination of normal-tonormal (NN) intervals, which are all the intervals between adjacent QRS sinusal complexes. Time domain variables are: mean
heart rate, heart rate standard deviation, the square root of the
mean squared differences of successive NN interval (RMSSD),
and the number of interval differences of successive NN intervals > 50 ms (NN50).
The frequency domain methods consist in the calculation
of the power spectral density analysis of a plot of consecutive
NN intervals, called tachogram. This power spectral density
can be calculated with nonparametric and parametric methods.
The parametric methods, as the autoregressive method used
in this study, allow an accurate estimation of power spectral
density even on a small number of samples on which the signal is supposed to maintain stationarity. Three major spectral
709
the Poincaré plot), and SD2 (standard deviation of the long term
RR interval variability major axis of the Poincaré plot).
In the frequency domain, HRV was analyzed using the autoregressive model (AR, model 16). The frequency bands
considered were low frequency (LF, 0.04-0.15 Hz) and high
frequency (HF: 0.15-0.4 Hz). The physiological explanation of
the very low-frequency component (VLF, 0-0.04 Hz) is poorly
defined; moreover, the very low-frequency assessed from shortterm recordings is a dubious measure; for this reason the very
low-frequency component was excluded from the analysis.
The frequency domain parameters analyzed were therefore: the
power of the LF and HF bands expressed in absolute values,
normalized units, and the LF/HF ratio.
± 1.6%, controls = 10.8% ± 8.3%; U-test: 325, p = 0.030). As
compared to controls (n = 55), migraineurs showed lower cyclic
alternating pattern (CAP) rate (migraineurs = 22.8% ± 2.5%,
controls = 30.1% ± 6.5%; U-test: 414, p < 0.001) and CAP time
(migraineurs = 78.8 ± 8.0 min, controls = 118.3 ± 36.9 min; Utest: 354, p = 0.006). No significant differences were observed
in the indexes of EEG arousals (number of arousals per hour of
sleep, per hour of NREM, and per hour of REM). Mean values
± standard deviation of sleep macrostructure and microstructure parameters in patients and controls, and results of the statistical comparison, are reported in Table 2. When compared
with the matched controls (n = 8), migraineurs showed no
significant differences in macrostructural parameters, but they
showed lower CAP rate (migraineurs = 22.8% ± 2.5%, matched
controls = 47.01 ± 11.2%; U-test: 8.0, p = 0.004).
Statistical comparisons were performed between migraineurs
and controls (n = 55), as well as between migraineurs and the
restricted group of matched controls (n = 8). Since HRV parameters show a skewed distribution in the general population,38 a
nonparametric Mann-Whitney U-test was used for comparison.
The same test was used to compare sleep parameters between
migraineurs and controls. The comparisons for categorical variables were performed by means of Fisher exact test. In case of
multiples comparison, in order to avoid family-wise type I errors, a formal Bonferroni correction was applied to each family
of comparisons, by dividing the limit of significance by the number of comparisons (for HRV parameters, 5 comparisons were
made, in the conditions Wake, N2-1C, N2-LC, N3, REM; therefore the threshold level for significance was p = 0.05/5 = 0.01).
Statistics were performed using the SYSTAT 12 software, version 12.02.00 for Windows (SYSTAT Software).
HRV: Time Domain Analysis
The most relevant difference observed in time domain concerned mean heart rate. In the migraineurs group, when compared with control group, there was a higher mean heart rate
during wake (migraineurs = 69.6 ± 3.0, controls = 59.0 ± 3.3
beats/min; U-test: 7.0, p = 0.006), stage N2-1C (migraineurs
= 69.3 ± 2.6, controls = 49.0 ± 2.1 beats/min; U-test: 39.0,
p < 0.001), and stage N2-LC (migraineurs = 63.5 ± 7.2, controls = 53.5 ± 10.3 beats/min; U-test: 37.0, p = 0.003). Notably,
no differences were observed in N3 and in REM sleep. The results of the HRV analysis in the time domain in migraineurs
and control groups, with results of U-test and levels of significance, are reported in Table 3. Similar results were observed in
the comparison between migraineurs and the matched control
group (n = 8) (Table 4).
HRV: Frequency Domain Analysis
RESULTS
No significant differences in the measured parameters (LF,
HF, LF/HF) were observed between migraineurs and controls
in wake. In sleep stage N2, in the migraineurs group there was
a statistically significant reduction of LF/HF ratio as compared
to control group (migraineurs = 0.09 ± 0.01; controls = 1.49 ±
2.26; U test: 42.5, p < 0.001); this occurred without significant
modifications of the HF and LF spectral powers. The same result was observed in deep slow wave sleep N3 (migraineurs =
0.09 ± 0.02; controls = 0.78 ± 1.01; U test: 64, p = 0.001). No
significant differences were observed, between the 2 groups
in REM sleep. Detailed results of the statistical comparison
between migraineurs and controls are shown in Table 3; a plot
of the LF/HF values in each sleep stage is shown in Figure 1.
Similar results were observed in the comparison between migraineurs and the matched control group (n = 8) (Table 4).
The diaries of migraine attacks collected by the patients in
the weeks before and after the sleep study showed that all patients had ≥ 5 migraine attacks during this interval (Table 1).
No patient presented migraine attacks in the 48 h before or after
the sleep study. Migraineurs and controls did not differ for age
(migraineurs = 48.1 ± 9.3, controls = 54.2 ± 13.0; U-test: 284.5,
p = 0.183), gender (χ2 = 0.572, p = 0.364) and body mass index
(BMI migraineurs = 22.6 ± 1.8 kg/m2; controls = 22.6 ± 1.8 m/
kg2, p = 0.465; matched controls = 20.9 ± 2.6 kg/m2; p = 0.753).
No patient presented polysomnographic evidence of sleep disordered breathing (AHI migraineurs = 1.8 ± 1.0 events/h; AHI
controls = 2.4 ± 1.3 events/h; AHI matched controls = 2.1 ± 0.8
events/h).
Sleep Structure
All patients had a normal night’s sleep; no patient had a
migraine attack in the night of the sleep study. Patients and
controls did not show snoring or other sleep-related breathing
abnormalities. On average, the patients included in this study
slept for 428.4 ± 43.4 min; their sleep efficiency index (total sleep time/time in bed) was 92.9% ± 3.0%; the number of
awakenings > 1 min was 5.4 ± 4.0. No significant differences
in sleep parameters and sleep stage composition was observed
between patients and controls; only a trend towards decrease in
N1 percentage was observed in patients (migraineurs = 5.5%
DISCUSSION
The main result of this study is the reduction of LF/HF ratio during N2 and N3 sleep stages in migraineurs when compared with controls. As concerns sleep structure, no significant
differences between migraineurs and controls were observed
in sleep macrostructure; migraineurs had a lower degree of
NREM sleep instability measured by CAP time and CAP rate.
These latter findings have been described and discussed in a
previous paper.19
710
Table 2—Results of the PSG study in migraineurs, controls, and matched controls, and results of statistical comparison
Migraineurs
(n = 8)
PSG Parameters
Macrostructure
Sleep onset latency, min
Total sleep time, min
Sleep period time, min
Sleep efficiency index, %
Controls
(n = 55)
Mann-Whitney
Matched Controls
(n = 8)
Mann-Whitney
Mean
SD
Mean
SD
U-test
p
Mean
SD
U-test
p
32.3
428.4
457.6
92.9
24.7
43.4
41.3
3.0
31.3
391.8
443.4
92.0
23.7
53.3
39.8
5.3
205.5
130.0
195.0
215.0
0.765
0.063
0.606
0.918
37.3
419.8
454.1
91.3
29.1
52.8
37.9
5.3
32.0
34.0
32.0
41.0
1.000
0.834
1.000
0.345
Sleep stages, %
REM
N1
N2
N3
Wake
19.3
5.5
42.7
25.2
7.1
5.8
1.6
7.3
7.5
3.0
17.3
10.8
39.5
20.7
33.1
7.8
8.3
11.3
10.5
46.5
187.0
325.0
183.0
148.0
282.0
0.496
0.030
0.445
0.137
0.201
21.0
10.9
36.1
24.2
21.7
6.4
8.0
15.0
14.8
43.5
26.0
14.0
41.0
38.0
36.0
0.529
0.059
0.345
0.529
0.674
Wake parameters
WASO, min
Awakenings > 1 min
33.2
5.4
13.4
4.0
54.2
5.5
40.6
3.5
294.5
235.0
0.124
0.748
40.3
6.0
28.0
4.1
26.0
29.5
0.529
0.791
Microstructure
Arousal index, n
Arousal Index NREM, n
Arousal Index REM, n
CAP time, min
CAP rate, %
10.2
10.6
6.4
78.8
22.8
2.6
2.7
2.4
8
2.5
11.5
10.9
11.6
118.3
30.1
2.2
1.8
5.3
36.9
6.5
286.5
0.167
306.0
0.074
251.0
0.521
354.0
0.006*
414.0 < 0.001*
14.4
14.7
13.0
156.5
47.0
6.1
6.7
4.8
36.8
11.2
19.0
17.0
26.0
16.0
8.0
0.172
0.115
0.528
0.093
0.004*
*Statistically significant differences. WASO, wake after sleep onset; CAP, cyclic alternating pattern; SD, standard deviation.
Table 3—Results of the HRV analysis in migraineurs and controls, and results of the statistical comparison
Migraineurs (n = 8)
Time Domain
Mean HR, bpm
SD, bpm
RMSSD, ms
NN50, count
NNTI
SD1, ms
SD2, ms
W N2-1C N2-LC
69.6 69.3 63.5
3.0
2.6
3.0
24.6 28.6 37.9
21.8 30.4 52.1
0.1
0.0
0.1
17.6 20.2 26.9
47.9 43.3 57.4
N3 REM
68.1 68.8
2.2
3.6
29.6 29.1
43.0 33.1
0.0
0.1
21.0 20.6
35.7 62.5
Frequency Domain
Abs. Power LF, ms2
Abs. Power HF, ms2
LF, n.u.
HF, n.u.
LF/HF
W N2-1C N2-LC N3 REM
69.8 156.2 149.0 93.0 105.4
60.8 128.2 198.7 208.8 133.3
47.7 49.8 45.0 39.2 48.7
33.7 36.3 46.0 52.5 31.8
2.4
0.1
1.1
0.1
4.2
Controls (n = 55)
W N2-1C N2-LC
59.0 49.0 53.5
3.3
2.1
2.5
28.9 27.2 36.8
48.9 52.6 79.4
0.0
0.0
0.0
20.5 19.3 26.1
55.1 39.5 52.1
Mann-Whitney U-test (p)
N3 REM
56.9 67.4
2.4
3.6
32.9 34.7
71.9 59.1
0.0
0.1
23.3 24.6
40.4 71.4
130.1 138.5 132.6 115.6 105.4
140.8 170.1 179.1 101.5 133.3
41.5 33.3 36.6 30.2 55.7
35.2 32.4 42.6 52.3 42.8
2.9
1.5
1.3
0.8
3.0
W
N2-1C N2-LC N3
0.006* < 0.001* 0.003* 0.173
0.665 0.201 0.591 0.836
0.256 0.869 0.901 0.563
0.107 0.403 0.476 0.501
0.445 0.433 0.092 0.025
0.304 0.869 0.901 0.563
0.483 0.665 0.885 0.496
REM
0.853
0.918
0.283
0.353
0.951
0.283
0.680
W
N2-1C
0.099 0.984
0.099 0.536
0.457 0.035
0.084 0.665
0.464 < 0.001*
REM
0.180
0.231
0.757
0.563
0.877
N2-LC
0.757
0.757
0.248
0.421
0.549
N3
0.901
0.695
0.154
0.885
0.001*
*Statistically significant differences. HR, heart rate; SD, standard deviation; RMSSD, root mean square of the differences between consecutive RR intervals;
NN50, number of consecutive RR intervals differing by more than 50 ms; NNTI, NN triangular index (determined from the histogram of RR intervals, in which NN
stands for normal-to-normal intervals [i.e., intervals between consecutive QRS complexes resulting from sinus node depolarization]); SD1, standard deviation of
the instantaneous beat-to-beat RR interval variability, minor axis of the Poincaré plot; SD2, standard deviation of the long-term RR interval variability major axis
of the Poincaré plot; Abs. Power, absolute power; LF, low frequency; HF, high frequency. N2-1C, sleep stage N2, first cycle; N2-LC, sleep stage N2, last cycle;
bpm, beats per minute; n.u., normalized units.
General agreement exists on the functional meaning of spectral component of heart rate variability; nevertheless some
matters of debate still exist. The HF component is universally
considered as a marker of parasympathetic activity; whereas
some doubt exist of the functional meaning of the LF component, which could be an expression of sympathetic tone or a
mixture of sympathetic and parasympathetic activation.17,35
Most authors agree that the LF/HF ratio provides an esteem of
the sympatho-vagal balance and its oscillations17,24,35; although
this point of view has been critically reviewed and questioned
by Eckberg.39 More recently, Burr demonstrated that LF and
HF, expressed in normalized units, are predictable from each
other, and that there is only one degree of freedom inherent in
these two measures.40 With these limitations, we used the LF/
HF ratio as an indicator of fluctuations of the sympatho-vagal
balance during sleep.40
711
Table 4—Results of the HRV analysis in migraineurs and matched controls and results of the statistical comparison
Patients (n = 8)
N2-1C
69.3
2.6
28.6
30.4
0.05
20.2
43.3
N2-LC
63.5
3.0
37.9
52.1
0.1
26.9
57.4
N3
68.1
2.2
29.6
43.0
0.04
21.0
35.7
Matched Controls (n = 8)
Time Domain
Mean HR, bpm
SD, bpm
RMSSD, ms
NN50, count
NNTI
SD1, ms
SD2, ms
W
69.6
3.0
24.6
21.8
0.06
17.6
47.9
REM
68.8
3.6
29.1
33.1
0.06
20.6
62.5
Frequency Domain
Abs. Power LF, ms2
Abs. Power HF, ms2
LF, n.u.
HF, n.u.
LF/HF
69.8 156.2 149.0 93.0 105.4
60.8 128.2 198.7 208.8 133.3
47.7 49.8 45.0 39.2 48.7
33.7 36.3 46.0 52.5 31.8
2.41 0.09 1.1
0.09 4.20
W N2-1C N2-LC
54.2 46.4 65.7
3.5
1.9
2.7
35.4 28.9 35.3
76.9 52.5 58.0
0.1
0.0 131.5
25.1 20.5 25.0
67.6 39.1 46.7
N3
69.3
3.3
29.0
43.1
0.1
20.5
62.6
REM
56.0
2.7
34.9
72.3
0.1
24.7
54.9
301.3 260.3 165.9 276.6 289.4
216.6 164.3 172.5 103.3 262.0
42.7 35.3 43.4 70.2 41.3
40.7 31.9 46.8 22.6 51.9
3.8
2.6
1.9
4.8
1.1
Mann-Whitney U-test (p)
W
0.006*
0.248
0.036
0.008*
0.172
0.036
0.074
N2-1C
0.001*
0.208
0.834
0.345
0.792
0.834
0.834
N2-LC
0.036
0.753
0.753
0.875
0.400
0.753
0.834
N3
0.753
0.345
0.462
0.563
0.012
0.462
0.172
REM
0.916
0.529
0.600
0.674
0.462
0.600
1.000
W
0.074
0.012
0.529
0.529
0.172
N2-1C
0.462
0.462
0.248
0.916
0.001*
N2-LC
0.916
0.834
1.000
0.600
0.916
N3
0.248
0.834
0.753
0.529
0.001*
REM
0.074
0.753
0.248
0.674
0.462
*Statistically significant differences. HR, heart rate; SD, standard deviation; RMSSD, root mean square of the differences between consecutive RR intervals;
NN50, number of consecutive RR intervals differing by more than 50 ms; NNTI, NN triangular index (determined from the histogram of RR intervals, in which NN
stands for normal-to-normal intervals [i.e., intervals between consecutive QRS complexes resulting from sinus node depolarization]); SD1, standard deviation of
the instantaneous beat-to-beat RR interval variability, minor axis of the Poincaré plot; SD2, standard deviation of the long-term RR interval variability major axis
of the Poincaré plot; Abs. Power, absolute power; LF, low frequency; HF, high frequency; N2-1C, sleep stage N2, first cycle; N2-LC, sleep stage N2, last cycle;
bpm, beats per minute; n.u., normalized units
observed in the migraineurs was similar, but it was characterized by a much deeper reduction of LF/HF in NREM sleep;
no difference between N2 and N3; and a greater, though not
significant, rise in REM.
The functional meaning of these autonomic modifications
during sleep in subjects with sleep-related migraine is not defined. It has been reported that during headache-free periods,
migraineurs have a reduction in sympathetic function compared to controls, and that migraine is a disorder characterized by chronic sympathetic dysfunction.42 Seen in this view,
sleep-related migraine could be a peculiar condition in which
the sympathetic impairment occurs selectively during sleep,
hypothetically due to modifications of central nervous system
arousability.19
Migraine attacks during sleep could be facilitated by this autonomic imbalance, and in particular by the relative prevalence
of parasympathetic activity in NREM. In this case, most of the
attacks should be emergent from NREM sleep stages. Alternatively, the trigger could be rapid shift from parasympathetic to
sympathetic predominance which occurs at the NREM-to-REM
transition. In this case, attacks should be more frequent in proximity to REM. Literature data suggest that migraine attacks can
occur both in deep NREM sleep stages and in REM.43,44 No attacks were recorded in our patients.
Whatever the mechanism, these data are in accordance with
our previous observation, concerning a group of patients who
largely overlapped with this present sample.19 In that study we
observed that migraineurs, compared to controls, had a significant reduction of NREM sleep instability (measured with cyclic
alternating pattern) without modifications of the fast-frequency
EEG arousals.19 It is known that CAP reflects a different arousal
mechanism than that measured by fast-frequency EEG arousal.45 Essentially, slow-frequency microarousal (CAP phases
type A1) and fast-frequency microarousal (fast EEG arousal
and CAP phases types A2 and A3) represent state-specific
Figure 1—Plot of the mean values of LF/HF ratio in
migraineurs and controls
Controls
Migraineurs
6
n.s.
5
LF/HF Ratio
4
n.s.
3
p < 0.001
2
p = 0.001
1
0
Wake
n.s., not significant.
N2
N3
REM
Sleep Stages
In the control subjects included in this present study, the time
course of the LF/HF along sleep stages was consistent with a
well-known circadian and ultradian rhythm24-26,36,41: the LF/HF
values decreased progressively from wake to N2 and N3, and
increased again in REM (Figure 1). This pattern of sympathetic-nerve activity has been demonstrated in living humans during sleep by means of a direct, microneurographic recording
of sympathetic fibers.25 The pattern of autonomic oscillations
712
arousal responses, differently distributed along the NREM/
REM cycles.45 Moreover, they differ in the power of autonomic
effect which is associated: hierarchically, an increasing magnitude of vegetative activation is observed from the weaker
slow-frequency microarousals (coupled with mild autonomic
activation) to the stronger fast-frequency microarousal (coupled with a vigorous autonomic activation).45 Our sleep-related
migraine patients seem to differ from controls essentially in
the amount of slow-frequency microarousal, and this in accordance with the reduced amount of autonomic activation during
NREM sleep. Taken together, these data suggest that a close
correlation exists between the activity of arousal systems during sleep and the activity of the autonomic nervous system; and
that in sleep-related migraine, a peculiar modification of both
these systems can be observed.
It could be speculated that the hypothalamus might play a
crucial role in the pathogenesis of sleep-related migraine. First,
hypothalamus has a major role in regulation of autonomic activity. Second, neuroimaging studies have demonstrated that
hypothalamic dysfunction may cause migraine attacks.46 Finally, the hypothalamus is a part of the arousal system.47 Experimental evidence indicates that regulation of autonomic
functions and nociceptive processing are closely coupled in
the hypothalamus and by means of the orexinergic transmission.48,49 Thus, the orexinergic system in the posterior hypothalamus is modulated by the biological clock and the cortex, and is
involved in the modulation of dural nociceptive transmission.49
Moreover, it is well known that the posterior hypothalamus, as
well as the orexinergic pathways, are involved in the regulation
of wake, sleep, and arousal.47 Therefore, we are proposing that a
hypothalamic dysfunction, probably involving the orexinergic
system, is responsible for the link between the pain of primary
neurovascular headaches,50 the autonomic dysfunction, and the
reduced arousability.
In conclusion, in the present study we observed concurrent
modifications of NREM sleep instability (CAP) and autonomic
modulation during sleep in patients with sleep-related migraine.
This is in accordance with a bulk of observations which consider
the hypothalamus, where arousal-related and autonomic relays
are located, a crucial structure in the pathogenesis of migraine
attacks. This is analogous to what happens in another primary, autonomic, closely sleep-related headache, namely cluster
headache. Nevertheless, in order to define a causal relationship
between these phenomena and to clarify the pathogenesis of
migraine attack during sleep, further studies are necessary, and
in particular neurophysiological recordings performed in the
course of migraine attacks emerging from sleep.
6. Tabata M, Takeshima T, Burioka N, et al. Cosinor analysis of heart rate variability
in ambulatory migraineurs. Headache 2000;40:457-63.
7. Fanciullacci M. Iris adrenergic impairment in idiopathic headache. Headache
1979;19:8-13.
8. Gotoh F, Komatsumoto S, Araki N, Gomi S. Noradrenergic nervous activity in
migraine. Arch Neurol 1984;41:951-5.
9. Martin R, Ribera C, Molto JM, Ruiz C, Galiano L, Matias-Guiu J. Cardiovascular
reflexes in patients with vascular headache. Cephalalgia 1992;12:360-4.
10. Mikamo K, Takeshima T, Takahashi K. Cardiovascular sympathetic hypofunction
in muscle contraction headache and migraine. Headache 1989;29:86-9.
11. Appel S, Kuritzky A, Zahavi I, Zigelman M, Akselrod S. Evidence for instability of
the autonomic nervous system in patients with migraine headache. Headache
1992;32:10-7.
12. Zigelman M, Appel S, Davidovitch S, Kuritzky A, Zahavi I, Akselrod S. The effect of verapamil calcium antagonist on autonomic imbalance in migraine:
evaluation by spectral analysis of beat-to-beat heart rate fluctuations. Headache
1994;34:569-77.
13. Thomsen LL, Iversen HK, Boesen F, Olesen J. Transcranial Doppler and cardiovascular responses during cardiovascular autonomic tests in migraineurs during
and outside attacks. Brain 1995;118 (Pt 5):1319-27.
14. Cortelli P, Pierangeli G, Parchi P, Contin M, Baruzzi A, Lugaresi E. Autonomic
nervous system function in migraine without aura. Headache 1991;31:457-62.
15. Akselrod S, Gordon D, Ubel FA, Shannon DC, Berger AC, Cohen RJ. Power
spectrum analysis of heart rate fluctuation: a quantitative probe of beat-to-beat
cardiovascular control. Science 1981;213:220-2.
16. Malliani A, Pagani M, Lombardi F, Cerutti S. Cardiovascular neural regulation
explored in the frequency domain. Circulation 1991;84:482-92.
17. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability: standards of measurement, physiological interpretation and clinical use. Circulation 1996;93:1043-65.
18. Jennum P, Jensen R. Sleep and headache. Sleep Med Rev 2002;6:471-9.
19. Della Marca G, Vollono C, Rubino M, Di Trapani G, Mariotti P, Tonali PA. Dysfunction of arousal systems in sleep-related migraine without aura. Cephalalgia
2006;26:857-64.
20. Eli R, Fasciano JA. A chronopharmacological preventive treatment for sleeprelated migraine headaches and chronic morning headaches: Nitric oxide supersensitivity can cause sleep-related headaches in a subset of patients. Med
Hypotheses 2006;66:461-5.
21. Gori S, Lucchesi C, Morelli N, Maestri M, Bonanni E, Murri L. Sleep-related
migraine occurrence increases with aging. Acta Neurol Belg;112:183-7.
22. American Academy of Sleep Medicine. International classification of sleep disorders: diagnostic and coding manual. 2nd ed. Westchester, IL: American Academy of Sleep Medicine, 2005.
23. Headache Classification Subcommittee of the International Headache Society.
The International Classification of Headache Disorders: 2nd edition. Cephalalgia
2004;24 Suppl 1:9-160.
24. Boudreau P, Dumont G, Kin NM, Walker CD, Boivin DB. Correlation of heart
rate variability and circadian markers in humans. Conf Proc IEEE Eng Med Biol
Soc 2011:681-2.
25. Somers VK, Dyken ME, Mark AL, Abboud FM. Sympathetic-nerve activity during
sleep in normal subjects. N Engl J Med 1993;328:303-7.
26. Stein PK, Lundequam EJ, Oliveira LP, et al. Cardiac autonomic modulation. Analyzing circadian and ultradian rhythms. IEEE Eng Med Biol Mag 2007;26:14-8.
27. Ehrhart J, Toussaint M, Simon C, Gronfier C, Luthringer R, Brandenberger G. Alpha activity and cardiac correlates: three types of relationships during nocturnal
sleep. Clin Neurophysiol 2000;111:940-6.
28. Gronfier C, Simon C, Piquard F, Ehrhart J, Brandenberger G. Neuroendocrine
processes underlying ultradian sleep regulation in man. J Endocrinol Metab
1999;84:2686-90.
29. Della Marca G, Vollono C, Rubino M, Capuano A, Di Trapani G, Mariotti P. A
sleep study in cluster headache. Cephalalgia 2006;26:290-4.
30. Terzano MG, Parrino L, Fioriti G, Orofiamma B, Depoortere H. Modifications of
sleep structure induced by increasing levels of acoustic perturbation in normal
subjects. Electroencephalogr Clin Neurophysiol 1990;76:29-38.
31. Iber C, Ancoli-Israel S, Chesson A, Quan SF. The AASM manual for scoring of
sleep and associated events: rules, terminology and technical specifications.
Westchester, IL: American Academy of Sleep Medicine, 2007.
32. Silber MH, Ancoli-Israel S, Bonnet MH, et al. The visual scoring of sleep in
adults. J Clin Sleep Med 2007;3:121-31.
33. Duchna HW. [Sleep-related breathing disorders--a second edition of the International Classification of Sleep Disorders (ICSD-2) of the American Academy of
Sleep Medicine (AASM)]. Pneumologie 2006;60:568-75.
References
1. Blau JN. Migraine: theories of pathogenesis. Lancet 1992;339:1202-7.
2. Shechter A, Stewart WF, Silberstein SD, Lipton RB. Migraine and autonomic
nervous system function: a population-based, case-control study. Neurology
2002;58:422-7.
3. Cirignotta F, Coccagna G, Sacquegna T, et al. Nocturnal headache: systemic
arterial pressure and heart rate during sleep. Cephalalgia 1983;3 Suppl 1:54-7.
4. Havanka-Kanniainen H, Tolonen U, Myllyla VV. Autonomic dysfunction in migraine: a survey of 188 patients. Headache 1988;28:465-70.
5. Mosek A, Novak V, Opfer-Gehrking TL, Swanson JW, Low PA. Autonomic dysfunction in migraineurs. Headache 1999;39:108-17.
713
34. Terzano MG, Parrino L, Smerieri A, et al. Atlas, rules, and recording techniques
for the scoring of cyclic alternating pattern (CAP) in human sleep. Sleep Med
2002;3:187-99.
35. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Eur Heart J 1996;17:354-81.
36. Brandenberger G, Ehrhart J, Buchheit M. Sleep stage 2: an electroencephalographic, autonomic, and hormonal duality. Sleep 2005;28:1535-40.
37. Niskanen JP, Tarvainen MP, Ranta-Aho PO, Karjalainen PA. Software for advanced HRV analysis. Comput Methods Programs Biomed 2004;76:73-81.
38. O’Brien IA, O’Hare P, Corrall RJ. Heart rate variability in healthy subjects: effect
of age and the derivation of normal ranges for tests of autonomic function. Br
Heart J 1986;55:348-54.
39. Eckberg DL. Sympathovagal balance: a critical appraisal. Circulation
1997;96:3224-32.
40. Burr RL. Interpretation of normalized spectral heart rate variability indices in
sleep research: a critical review. Sleep 2007;30:913-9.
41. Yotsukura M, Sasaki K, Kachi E, Sasaki A, Ishihara T, Ishikawa K. Circadian
rhythm and variability of heart rate in Duchenne-type progressive muscular dystrophy. Am J Cardiol 1995;76:947-51.
42. Peroutka SJ. Migraine: a chronic sympathetic nervous system disorder. Headache 2004;44:53-64.
43. Dexter JD. The relationship between stage III + IV + REM sleep and arousals
with migraine. Headache 1979;19:364-9.
44. Dexter JD, Weitzman ED. The relationship of nocturnal headaches to sleep
stage patterns. Neurology 1970;20:513-8.
45. Halasz P, Terzano M, Parrino L, Bodizs R. The nature of arousal in sleep. J Sleep
Res 2004;13:1-23.
46. Alstadhaug KB. Migraine and the hypothalamus. Cephalalgia 2009;29:809-17.
47. Jones BE. Arousal systems. Front Biosci 2003;8:s438-51.
48. Holland P, Goadsby PJ. The hypothalamic orexinergic system: pain and primary
headaches. Headache 2007;47:951-62.
49. Holland PR, Goadsby PJ. Cluster headache, hypothalamus, and orexin. Curr
Pain Headache Rep 2009;13:147-54.
50. Rainero I, Rubino E, Gallone S, et al. Evidence for an association between migraine and the hypocretin receptor 1 gene. J Headache Pain 2011;12:193-9.
Submitted for publication September, 2012
Address correspondence to: Catello Vollono, M.D., Ph.D., Department of
Neurosciences, Unit of Neurophysiopathology and Sleep Medicine, Catholic
University, Policlinico Universitario ‘A. Gemelli’, Largo A. Gemelli, 8 - 00168 - Rome,
Italy; Tel. +39 0630154279; Fax: +39 0635501909; E-mail: lvol@libero.it
714
C ase R E ports
Status Cataplecticus Precipitated by Abrupt
Withdrawal of Venlafaxine
Janice Wang, M.D.; Harly Greenberg, M.D.
Division of Pulmonary, Critical Care and Sleep Medicine, Hofstra North Shore-LIJ School of Medicine, New Hyde Park, NY
Keywords: Status cataplecticus, narcolepsy, cataplexy,
venlafaxine withdrawal
Citation: Wang J; Greenberg H. Status cataplecticus precipitated by abrupt withdrawal of venlafaxine. J Clin Sleep Med
2013;9(7):715-716.
Status cataplecticus is a rare manifestation of narcolepsy with
cataplexy episodes recurring for hours or days, without a refractory period, in the absence of emotional triggers. This case
highlights a narcoleptic patient who developed status cataplecticus after abrupt withdrawal of venlafaxine.
C
ataplexy is a symptom of narcolepsy triggered by strong
emotion that causes sudden muscle atonia with preserved
consciousness. It may represent intrusion of REM sleep phenomena into wakefulness.1 A refractory period of up to several
hours typically follows a cataplexy attack.2 Persistent cataplexy,
known as status cataplecticus, is a rare, often misdiagnosed
manifestation of cataplexy. Failure to recognize this condition
can lead to unnecessary diagnostic testing that delays appropriate therapy.2
Tricyclic antidepressants (TCAs), selective serotonin reuptake inhibitors (SSRIs), and serotonin-norepinephrine reuptake
inhibitors (SNRIs) prevent cataplexy.3 Enhancement of noradrenergic and serotonergic activity may reduce cataplexy by
inhibiting “REM-on” neurons in the lateral dorsal tegmentum
(LDT) and pedunculopontine tegmentum (PPT).3 The precipitous decrease in noradrenergic and serotonergic tone occurring
upon abrupt discontinuation of therapy may elicit status cataplecticus, presumably by disinhibition of “REM-on” neurons.3
Interestingly, discontinuation of sodium oxybate, another anticataplexy medication that may exert its effect via GABAB receptors, has not been associated with status cataplecticus.4
Initial treatment consisted of imipramine and methylphenidate. Persistent cataplexy and hypersomnolence prompted a
change to sodium oxybate and modafinil with almost complete
resolution of symptoms. Sodium oxybate (9 grams nightly) was
continued for seven years until she reported somnambulism and
nocturnal sleep eating with multiple falls. As these were likely
NREM parasomnias associated with sodium oxybate, anticataplexy therapy was changed to venlafaxine ER 75 mg/day.
Cataplexy persisted, but further increase in venlafaxine was
precluded by worsening hypertension, an adverse effect of venlafaxine. Fluoxetine 20 mg was suggested. The following day,
she developed gastroenteritis; while she continued venlafaxine,
she reported emesis shortly after taking medication. The next
day, emotional upset triggered cataplexy, consisting of slurred
speech and weakness in all extremities; this was witnessed by
her family. Unlike her usual cataplexy episodes which rapidly
resolve, she experienced continuous cataplectic attacks over
the next 4 h, despite absence of further emotional triggers. She
maintained consciousness throughout but recalled vivid hallucinations. Fluoxetine was ineffective; upon resuming venlafaxine, cataplexy resolved within several hours. Neurologic
examination was not performed during status cataplecticus but
was normal the following day after resolution of cataplexy.
Report of Case
A 76-year-old female was evaluated 20 years ago for hypersomnolence and “irresistible sleep attacks” that began during
adolescence. She experienced sleep paralysis but denied hypnogogic or hypnopompic hallucinations. Strong emotion such
as anger, happiness, or excitement often precipitated slurring
of speech, flattening of facial expression, and leg weakness.
She cannot recall whether cataplexy appeared concurrently or
after onset of somnolence. Medical history was significant for
hypertension treated with lisinopril. Polysomnography demonstrated a normal apnea-hypopnea index. Multiple sleep latency
testing revealed a mean sleep onset latency of 3 minutes with
3 sleep-onset REM periods. She was HLA-DQB1*0602 positive. CSF hypocretin levels were not obtained. The diagnosis of
narcolepsy with cataplexy was made at age 56, nearly 40 years
after symptom onset.
DISCUSSION
Our patient experienced status cataplecticus within 48 hours of
abrupt withdrawal of venlafaxine, due to diminished absorption
from gastroenteritis. The resultant rapid decrease in noradrenergic
and serotonergic tone may have precipitated status cataplecticus.3
Status cataplecticus has been reported in narcoleptic patients after abrupt withdrawal of clomipramine, a serotonergic reuptake
inhibitor that augments adrenergic tone via its metabolite, desmethylimipramine.3 Protracted episodes of cataplexy have also been
reported after gradual withdrawal of TCAs and SSRIs, peaking 40
to 60 days after discontinuation.3 Status cataplecticus may also occur with administration of prazosin, an α-adrenergic antagonist.5
While the neurophysiologic basis of cataplexy remains unclear, decreased hypocretinergic activity with reduced noradren715
J Wang and H Greenberg
ergic tone, possibly from decreased activity of locus coeruleus
neurons, may contribute to decreased motoneuron excitation during cataplexy. Dysfunction of other neurotransmitters, including
dopaminergic systems, may also contribute to cataplexy.6
Status cataplecticus is a rare complication in narcolepsy
resulting from abrupt discontinuation of noradrenergic and
serotonergic medications. Behavioral management restricting
social interaction to minimize cataplexy triggers, and limiting
ambulation, may be helpful during status cataplecticus. Anticataplexy medications, with individualized risk-benefit analysis
regarding adverse effects, such as parasomnias and fluid retention from sodium oxybate, hypertension from venlafaxine, and
anticholinergic effects from clomipramine are necessary. First
line therapy for cataplexy is sodium oxybate; however in this
case, risk of falls from somnambulism in an elderly woman outweighed benefits. Venlafaxine ER was reinitiated and gradually
increased to 150 mg/day with resolution of cataplexy; escalation of antihypertensive therapy controlled blood pressure.
2. Calabro RS, Savica R, Lagana A, et al. Status cataplecticus misdiagnosed as
recurrent syncope. Neurol Sci 2007;28:336-38.
3. Ristanovic RK, Liang H, Hornfeldt CS, Lai C. Exacerbation of cataplexy following gradual withdrawal of antidepressants: manifestation of probable protracted
rebound cataplexy. Sleep Med 2009;10:416-21.
4. US Xyrem Multicenter Study Group. Sodium oxybate demonstrates long-term
efficacy for the treatment of cataplexy in patients with narcolepsy. Sleep Med
2004;5:119-23.
5. Aldrich MS, Rogers AE. Exacerbation of human cataplexy by prazosin. Sleep
1989;12:254-56.
6. Peever J. Control of trigeminal motoneuron behavior and masseter muscle tone
during REM sleep, REM sleep behavior disorder and cataplexy/narcolepsy. Arch
Ital Biol 2011;149:454-66.
Accepted for publication November, 2012
Address correspondence to: Janice Wang, M.D., 410 Lakeville Road, Suite 107,
New Hyde Park, NY 11040; Tel: (516) 465-5400; Fax: (516) 465-5454; E-mail:
janiceywang@gmail.com
References
1. Broughton R, Valley V, Aguirre M, et al. Excessive daytime sleepiness and
the pathophysiology of narcolepsy-cataplexy: a laboratory perspective. Sleep
1986;9:205-15.
716
Nocturnal Diaphoresis Secondary to Mild Obstructive Sleep
Apnea in a Patient with a History of Two Malignancies
Robert Daniel Vorona, M.D., F.A.A.S.M.1; Mariana Szklo-Coxe, Ph.D.2; Mark Fleming, M.D.3; J. Catesby Ware, Ph.D., F.A.A.S.M.1
Division of Sleep Medicine, Department of Internal Medicine, Eastern Virginia Medical School, Norfolk, VA; 2Old Dominion
University, College of Health Sciences, Norfolk, VA; 3Virginia Oncology Associates, Hampton, VA
C ase R eports
1
Numerous medical disorders, including obstructive sleep apnea,
may cause nocturnal diaphoresis. Previous work has associated severe obstructive sleep apnea with nocturnal diaphoresis.
This case report is of import as our patient with severe nocturnal
diaphoresis manifested only mild sleep apnea, and, for years,
his nocturnal diaphoresis was ascribed to other causes, i.e., first
prostate cancer and then follicular B-cell lymphoma. Additionally,
it was the nocturnal diaphoresis and not more common symp-
toms of obstructive sleep apnea, such as snoring, that led to the
definitive diagnosis of his sleep apnea and then to treatment with
a gratifying resolution of his onerous symptom.
Keywords: Diaphoresis, sleep apnea
Citation: Vorona RD; Szklo-Coxe M; Fleming M; Ware JC.
Nocturnal diaphoresis secondary to mild obstructive sleep
apnea in a patient with a history of two malignancies. J Clin
Sleep Med 2013;9(7):717-719.
D
iverse medical disorders including malignancies, infections, and endocrine abnormalities have been associated
with nocturnal diaphoresis.1 Nocturnal diaphoresis may occur
in obstructive sleep apnea syndrome (OSAS).2 Nevertheless,
research has been limited and findings mixed.3,4 One study
found night sweats to be associated with snoring and sleepiness but not apnea-hypopnea index (AHI).3 Excessive nocturnal sweating occurred in 34% of patients with severe OSAS
([AHI] = 52).4 Our case is instructive as our patient’s severe
nocturnal diaphoresis occurred with mild OSAS and was not
clearly connected to his two malignancies.
hol and no caffeine. He ceased smoking cigarettes in 1975 and
reported a comfortable bedroom temperature.
Past medical history: Prostate cancer history (treated first
with surgery, then with radiation therapy in 2006 for micrometastatic disease), follicular B-cell lymphoma, radiation cystitis, coronary heart disease, supraventricular and ventricular
dysrhythmias (dronederone previously stopped given concerns
it contributed to diaphoresis), pacemaker implantation, adenomatous colon polyps with dysplasia, pneumonia, previous left
upper extremity thrombophlebitis, gastroesophageal reflux
disease (GERD), Barrett’s esophagus, and distant tonsillectomy history.
Medications (at time of initial sleep center visit) included
clopidogrel, niacin, atorvastatin, aspirin, flecainide, pantoprazole, lubiprostone, glucosamine/chondroitin, and fish oil.
Review of systems demonstrated no recent change in weight,
no history of diabetes mellitus or thyroid disease, and some difficulties with memory and concentration.
Exam demonstrated a tall, slender male with normal respiratory rate and pulse. Upper airway exam revealed no frank anterior septal deviation. He had a large tongue, borderline class III
malocclusion, Mallampati IV, retrognathia, and 15-inch neck
circumference. Neck exam revealed normal thyroid, and no
cervical, supraclavicular, or axillary adenopathy. Cardiopulmonary exam: normal lung sounds and heart tones, and a pacemaker. There was no cyanosis, clubbing, or edema.
The patient’s diagnostic and continuous positive airway pressure (CPAP) polysomnographic data are presented in Table 1.
By numeric indices, the patient’s diagnostic study manifested
mild apnea that was more pronounced in stage R sleep. Pre and
post blood pressures were normal at 109/67 and 104/64, respectively. Esophageal pressure monitoring and ambulatory blood
pressure monitoring were not done as part of the patient’s clinical evaluation.
Report of Case
A 67-year-old male first manifested diaphoresis in 2002 following a radical prostatectomy for T2N0MX prostate cancer.
His clinicians reassured him that diaphoresis, common after
surgery, should resolve.
It worsened; the patient often soaked through clothing and
slept on a towel. Endocrine evaluation was negative. A negative QFT Gold Test for tuberculosis, negative HIV test, normal
erythrocyte sedimentation rate test (1), and normal testosterone
level (444) were obtained in 2010. However, follicular B-cell
lymphoma was diagnosed (1/2011) and thought the likely cause
of profuse, almost nightly sweats. The Rituximab treatment for
lymphoma (3/2011) did not improve his diaphoresis. The patient’s oncologist recommended sleep medicine consultation.
The patient presented 6/2011 to sleep medicine with a chief
complaint of “severe night sweats that have gotten progressively worse over the last 12 months.” Sweating was most apparent
02:30 until 04:00. His registered nurse wife noted infrequent
snoring and pauses in respiration. He denied restless sleep. Epworth Sleepiness Scale score was normal (5), though he felt
tired upon arising after 7-8 hours in bed. He used minimal alco717
RD Vorona, M Szklo-Coxe, M Fleming et al
Table 1—Summary of polysomnographic data
Study Date
06/09/11
06/22/11
Max.
CPAP
0
9b
AHIa
7
1
Supine
REM AHI
AHI
27
7
0
1
Low O2
%
89
94
Length
(s)
21
26
Max
Length
(s)
52
29
Arousal
Index
8
14
Total
Sleep
Sleep Efficiency
(min)
%
381
93
326
78
Min
REM
53
29
PLMS/h
0
0
BMI
(kg/m2)
22.7
23.1
Hypopneas defined as: ≥ 30% decline in tidal volume for ≥ 10 seconds associated with ≥ 4% oxygen desaturation. bThe AHI on the CPAP titration study of
6/22/11 is averaged over all CPAP pressures.
a
The CPAP titration study revealed control of the patient’s
apnea and improved oxygen nadir. Blood pressures pre and post
study, respectively, were again normal at 101/63 and 99/57. In
neither study was there evidence for other sleep pathology such
as periodic limb movements of sleep.
The patient began CPAP at 9 cwp. At 1-month follow-up, he
noted near-disappearance of nocturnal diaphoresis. Diaphoresis
had taken a few days after CPAP initiation to dissipate. Medications including niacin remained unchanged. A CPAP download revealed that during the first month (7/12/11-8/10/11), the
patient used the machine 5 h 45 min per night, with 87% of
the time ≥ 4 h and with an AHI of 2.9 on 9 cwp. His daily records for the preceding month indicated no night sweats save
for “moist” neck area and armpits on 3 nights.
Endocrinology work up (8/2011), requested by the patient’s
oncologist, revealed normal renal, hepatic, and glucose levels.
The overall thyroid panel suggested “borderline low” thyroid
function, yet TSH was normal (2.94).
His last visit (8/2012) revealed continued control of his diaphoresis with CPAP therapy. He noted some diaphoresis and
snoring with CPAP interface slippage.
A CPAP download from 7/8/11-8/1/12 revealed that the patient used his CPAP machine an average of 5 h 8 min per night,
with 77% of the time ≥ 4 h and with an AHI of 1.6 on 9 cwp.
At one and a half years post CPAP initiation, on phone communication, the patient believed his diaphoresis was still dramatically improved. The patient stated that if he stopped CPAP
that the first night he would sweat only modestly. With CPAP
discontinuation for a few days, he noted that increasingly severe diaphoresis ensued.
for RERAs. For example, the patient’s breathing was usually inphase, and the intercostal electromyogram signals were almost
exclusively quiet. We did not see subtle out-of-phase breathing
terminating in arousals. The normal arousal index also argued
against our missing subtle upper airway events.
Sleep stages have been associated with different predispositions to diaphoresis. For that reason, we looked for changes in
stage N3 sleep (the stage with greatest diaphoresis) and stage
R sleep (the stage ostensibly with the least) that might have
explained his diaphoresis and response to therapy. We found no
explanation in our review of the sleep architecture (please see
minutes of stage R for each study in Table 1. The patient had no
stage N3 on either study).
We did not analyze heart rate variability (HRV), a measure
of autonomic function. With only one night in each condition,
selecting samples for analysis while controlling for proximity
of apnea events, proximity of arousals, sleep stage, and time of
night was beyond our capability.
The severe night sweats in this patient with mild OSAS
were noteworthy. Few studies3,4 have investigated diaphoresis in OSAS. Electrodermal activity has also been utilized to
probe this relationship in patients with severe apnea (AHI =
45.3).5 These findings, however, may not be generalizable to
our patient, who had much milder apnea. Future studies aimed
at assessing the prevalence of diaphoresis in mild OSAS and
clarifying its pathophysiological basis may be warranted. Perhaps this severe diaphoresis reflects individual variability, just
as some patients with mild OSAS by numeric indices may have
severe sleepiness. The patient’s sympathetic nervous system
may have been more sensitive to the impact of OSAS, and the
sweating may have reflected heightened autonomic activity.
The association of autonomic dysfunction with mild OSAS was
previously reported in 2004,6 and, in 2008, systolic non-dipping
of blood pressure was reported to be associated with sleep apnea, even mild apnea. 7
Finally, this patient had a history of GERD. GERD has itself
been reported to cause nocturnal sweats,8 to be comorbid with
OSAS, and improved by CPAP.9 While the improvement of diaphoresis by CPAP may have been due to GERD reduction, the
patient presented to the sleep center with diaphoresis notwithstanding use of a proton pump inhibitor. Thus, CPAP treatment
of OSAS was likely primarily responsible for the gratifying response described herein.
DISCUSSION
Firstly, this report illustrates the importance of not excluding OSAS from the differential diagnosis of nocturnal diaphoresis, despite other more obvious potential etiologies. Secondly,
this case indicates that even mild OSAS may result in severe
diaphoresis, extending findings regarding its presence in severe
OSAS. 4 Thirdly, CPAP both treated mild OSAS and resolved
the troubling night sweats. This case emphasizes the onerous
nature of diaphoresis in OSAS, even when unaccompanied by
more typical symptoms, such as prolific snoring and sleepiness.
In this case, severe nocturnal diaphoresis was our primary clue
to the presence of OSAS.
It is possible that the AHI underestimated the patient’s degree of respiratory instability and that the patient may have had
concomitant respiratory event related arousals (RERAs). However, a careful review of the raw data did not reveal evidence
REFERENCES
1. Viera AJ, Bond M, Yates SW. Diagnosing night sweats. Am Fam Physician
2003;67:1019-24
718
Case Report
2. Guilleminault C, Bassiri A. Clinical features and evaluation of obstructive sleep
apnea–hypopnea syndrome and upper airway resistance syndrome. In: Kryger
MH, Roth T, Dement WC, eds. Principles and practice of sleep medicine. 4th ed.
Philadelphia, PA: Elsevier Saunders, 2005; 1043-52.
3. Mold JW, Goodrich S, Orr W. Associations between subjective night sweats and
sleep study findings. J Am Board Fam Med 2008;21:96-100.
4. Cruz IA, Drummond M, Winck JC. Obstructive sleep apnea symptoms beyond
sleepiness and snoring: effects of nasal APAP therapy. Sleep Breath 2012
Jun;16:361-6.
5. Arnardottir ES, Thorliefsdottir B, Svanborg E, Olafsson I, Gislason T. Sleeprelated sweating in obstructive sleep apnoea: association with sleep stages and
blood pressure. J Sleep Res 2010;19:122-30.
6. Woodson BT, Brusky LT, Saurajen A, Jaradeh S. Association of autonomic
dysfunction and mild obstructive sleep apnea. Otolaryngol Head Neck Surg
2004;130:643-8.
7. Hla KM, Young T, Finn L, Peppard PE, Szklo-Coxe M, Stubbs M. Longitudinal
association of sleep-disordered breathing and nondipping of nocturnal blood
pressure in the Wisconsin sleep cohort study. Sleep 2008;31:795-800.
8. Young P, Finn BC, Bruetman JE, Trimarchi H. Gastroesophageal reflux as a
cause of night sweating. An Med Interna 2007;24:285-8.
9. Tawk M, Goodrich S, Kanasewitz G, Orr W. The effect of 1 week of continuous
positive airway pressure treatment in obstructive sleep apnea patients with concomitant gastroesophageal reflux. Chest 2006;130:1003-8.
acknowledgments
Eastern Virginia Medical School Division of Sleep Medicine supported this work.
Submitted for publication December, 2012
Submitted in final revised form March, 2013
Accepted for publication March, 2013
Address correspondence to: Robert Daniel Vorona, M.D., Associate Professor,
Division of Sleep Medicine/ Department of Internal Medicine/ Eastern Virginia Medical
School, 301 Riverview Avenue Suite 670, Norfolk, VA 23510; Tel: 1-757-625-0172;
Fax: 1-757-452-4374; E-mail: voronard@evms.edu
719
S pecial A rticle
Rosalind D. Cartwright, Ph.D., F.A.A.S.M.1; Christian Guilleminault, M.D., D.Sc., F.A.A.S.M.2
Rush University Medical Center, Graduate College, Neuroscience Program, Chicago, IL;
2
Stanford University, Sleep Medicine Division, Redwood City, CA
1
Objective: To test whether laboratory-based research differentiating sleepwalkers (SW) from controls (C) can be applied in an
uncontrolled forensic case as evidence the alleged crime was
committed during an arousal from sleep in which the mind is not
fully conscious due to a SW disorder.
Methods: A PSG study recorded 8 months after the defendant
was charged was analyzed independently by spectral analysis.
Slow wave activity (SWA) and cyclic alternating pattern (CAP)
rates were computed. Clinical interviews and police records
were reviewed for data re: the defendant’s sleep prior to the
event and use of drugs, alcohol, and stimulants.
Results: The SWA distribution was abnormally low and flat,
significantly lower than published controls; in the first NREM
cycle, CAP rate 55 was above normal. Two weeks of prior sleep
deprivation was confirmed from interviews and defendant’s observed daytime sleepiness. Caffeine intake the day before the
event was calculated at 826 mg over 14 hours. Snoring and a
mild breathing disorder were present in the PSG.
Conclusion: Testimony based on spectral analysis of PSG
recorded following an alleged criminal event supported a
SW explanation for the non-rational behaviors charged.
The defendant was acquitted of all charges and has been
successfully treated.
Keywords: Sleepwalking, spectral analysis, slow wave activity,
sleep deprivation, caffeine
Citation: Cartwright RD; Guilleminault C. Defending sleepwalkers with science and an illustrative case. J Clin Sleep
Med 2013;9(7):721-726.
shows sleep to be in slow wave sleep (SWS), (also called delta
sleep or stages 3 and 4 sleep). It is one of this class, SW, that
has been the focus of concern about Sleep Medicine clinicians
testifying in legal cases of adults. The basic issue is that SW is
episodic; therefore, there can be no certainty that an individual,
even one genetically vulnerable to such events,7,8 was sleepwalking at the time of the offence.
In fact, prior to 2000, the PSG was of limited value in differentiating an adult SW from a normal sleeper. Although SW
have more frequent arousals from SWS than normal sleepers,9
this sleep instability had been observed in patients with other
diagnoses such as obstructive sleep apnea (OSA).10 Therefore,
once the many other possible diagnoses that might account for
an alleged non-conscious episode of a defendant had been ruled
out,11 sleep experts pre-2000 had to decide for themselves is
this case likely to be one of sleepwalking or not. Most often the
cases that came to trial were those involving aggressive behaviors inflicted on another person. More recently, persons charged
with non-consensual sexual behavior (sexsomnia)12 have also
become court cases, as have others in which the charge was
unlawful entry with intent to commit robbery or rape.13 Other
activities common during a SW such as sleep eating, protecting
others, and exploring are unlikely to result in a criminal charge.
Without a clear diagnostic sign in the PSG, sleep experts acting for the defense were likely to base their testimony on their
judgment of the accused’s truthfulness during a pre-trial interview, a history of prior witnessed SW, and the similarity of their
behaviors before, during, and following the event to those in
published case studies. Some of these characteristics were based
on formal research, such as early SWS arousal with long lasting
confusion and amnesia following,6 while others were extracted
Introduction: Background Prior to 2000
In 1987 and 1997, two highly publicized cases of first degree
murder occurred in which there was no doubt the accused was
the one who committed the crime.1,2 Nonetheless, a question
was raised: were they guilty under the law? The question of guilt
is based on the principle of mens rea; was the person’s mind
conscious at the time? In both cases, the aggression took place
following an arousal from sleep and was followed by profound
amnesia and regret for what had happened. In both cases, the
defense argued the accused was in a non-conscious state due to
a sleepwalking (SW) disorder. One case was acquitted the other
convicted. Since then, both prosecuting and defending lawyers
have sought the advice of Sleep Medicine specialists for their
opinion asking: is this a case of non-conscious behavior due to
a sleep disorder? This has prompted a strong push-back from
some in the Sleep Medicine community, denying there is a valid
basis in sleep science for an opinion in such cases.3-5
Sleep Medicine was then relatively new as a clinical profession. The research supporting a diagnosis of several major sleep
disorders was strong enough that experienced clinicians, supported by data from the polysomnogram (PSG) revealing the
presence of specific sleep abnormalities, could make a diagnosis
and recommend a treatment with confidence. An exception was
a group of disorders that, although common in young children,
were rarely sustained into adulthood, and so escaped systematic
attention. These are the non-rapid eye movement (NREM) parasomnias. A landmark study by Broughton6 identified a number
of their common features. Primary is that these arousals occur
early in the major sleep period, prior to REM, when the PSG
721
RD Cartwright and C Guilleminault
Figure 1—Example of CAP incidents prior to a sleepwalking event.
(Courtesy of C. Guilleminault).
from a review of published case reports, showing non-rational
behavior during the episode, amnesia, perplexity, and regret afterward.14 These became the basis of the formal clinical diagnosis of SW for the American Academy of Sleep Medicine15 and of
the American Psychiatric Association.16
Sleep experts who acted for the prosecution were more likely to base their testimony on their judgment that the accused’s
actions were premeditated; motivated by a prior negative relationship with the person attacked; or, in the case of sexual behaviors, that the accused took advantage of an opportunity that
presented itself. In some cases the prosecution’s sleep expert
argued the act was not sleep-related but planned during wakefulness and carried out while the accused was fully conscious.17
If the accused had been drinking alcohol prior to the event, the
prosecution’s expert held that this was a voluntary behavior and
therefore the accused was legally responsible for any aggressive
or sexual acts that took place subsequently, even if these followed an arousal from SWS.18
ferent countries. All three reported the same two significant
differences.
SW demonstrated more disrupted sleep, whether reported
as microarousals, wake after sleep onset (WASO), arousals,
awakenings, or an abnormal amount of cyclic alternating pattern (CAP A2 and A3).22 This sleep fragmentation was significantly higher in the SWS of the SW than in the C groups.
Figure 1 shows CAP episodes preceding a SW event recorded
in the Stanford laboratory. Difficulty maintaining sleep in the
first third of the night could now be considered a characteristic
of a NREM parasomnia, as it was observed in all three studies
to be significantly higher in SW than in C subjects only in the
first third of the night, when most SW events occur. Those who
arouse from REM sleep, REM behavior disorder (RBD), may
also be aggressive but differ in demographics and PSG characteristics from NREM SW, and none have to date become published forensic cases.
The second finding is a difference in the amount of SWA. This
is lower throughout the night in SW than in matched controls
and is significantly lower in the first NREM cycle. The analysis
of the PSG responsible for this finding is not the Rechtschaffen
and Kales (R&K)23 delta percent but the more precise spectral
analysis scoring (fast Fourier transform [FFT]) yielding the
count of SWA in each NREM cycle.
These new findings did not, however, settle the concern expressed about sleep experts testifying in court, as stated in a
recent publication, “…there is absolutely no after-the-fact poly-
Sleep Science Post 2000
This difficulty concerning proof of SW changed in 2000
with the publication of two independent studies reporting new
PSG findings in sleepwalkers not found in age- and gendermatched controls.19,20 A third study reporting the same major
findings followed in 2001.21 These studies were carried out
by independent investigators in different laboratories, in difJournal of Clinical Sleep Medicine, Vol. 9, No. 7, 2013
722
Special Article
somnographic finding that could possibly have any relevance as
to whether the accused was sleepwalking at the time of the event
in question.”24 This overlooks the extensive literature that spectral analysis scoring shows high reliability within individuals
to reproduce their profile of the frequency of the various EEG
wave forms across non-consecutive nights, including the delta
power, under normal and sleep deprivation conditions.25-28 The
objective of this study is to test the SWA in a forensic case to
determine if it was significantly low in the first NREM cycle and
if that indicated the presence of a predisposing condition and
possibly a sleep problem which could be treated.
unusually high caffeine intake by his need to stay alert at work
following his considerable sleep loss the week before and the
week following the birth, when he took over responsibility for
the nighttime care of his 3- and 5-year-old children. The 5-yearold had multiple medical problems and difficulty sleeping requiring nighttime attention and medication.
On the Rush Sleep Center intake forms, CD listed he consumed ten large caffeinated drinks the day of the alleged SW
event: two mugs of coffee at breakfast, two additional coffees
between 9 am and 11 am, and two Diet Cokes in the afternoon;
and during the evening program he drank “large tumblers” of
Diet Pepsi. The total caffeine consumption was estimated to be
826 mg over a 14-h period.29 Studies of the effect of 600 mg of
slow-release caffeine show it promotes wakefulness following
sleep deprivation and improves vigilance.30 As caffeine blocks
adenosine receptors, it inhibits sleep onset and sleep maintenance and reduces the amount of slow wave sleep and SWA in
the first sleep cycle.31,32 High caffeine intake has been implicated in SW with violence.2,33
An Illustrative Case
In 2007 a patient, CD, on advice of his lawyer, presented to a
sleep laboratory for an evaluation of SW. He was turned down
as a patient because his SW was the issue in a pending legal
case. The patient then applied to the Sleep Disorder Service at
Rush University Medical Center where he was seen by two senior clinicians, both boarded in Sleep Medicine, a neurologist
and a psychologist. The patient’s wife (LD) also attended the
intake interview.
The initial exam showed CD to be a 39-year-old Caucasian
male, height 70 inches, weight 187 pounds, with body mass index
(BMI) = 26.8, married for 10 years and father of 3 children. He
gave a childhood history of frequent episodes of SW, as did his
elder brother. These were witnessed by both parents and by each
other (as they shared a bedroom). The patient reported his SW
persisted into his adult years, with frequent episodes witnessed
by his wife and by her parents. Another witness, a physician, observed CD during a SW episode when they shared a hotel room
during a trip to attend an athletic event. Although the episode
was benign, CD walked about muttering to himself in a confused
state, the physician reported that episode to their local police
as a safety precaution. LD reported her husband rarely walked
outside prior to the episode for which he was now charged. No
violence was involved in any of his episodes. His usual behavior
was described as patrolling the house after having been aroused
by a noise not heard by others. He sometimes moved objects
in a non-rational manner. For example- before the expected
third child was born, CD placed a clock into an empty cradle
and tucked it in tenderly. Most often he looked out of windows
searching for possible intruders. These appear to be motivated
to protect the family from unrealistic dangers. This is one of five
common types of behavior characteristic of adult sleepwalkers.13
The (Alleged) Sleepwalking Event
A reconstruction of the event presented at the trial was based
on the detectives’ interviews of CD, the complainant, the daughter (EF) of the family living opposite him and her boyfriend. CD
recalled having difficulty sleeping that night. He remembers arising at about 3 am and again later on hearing a noise. On looking
out a window, believed he saw lights on in the house facing him
and the front door standing partly ajar. He felt duty-bound to investigate. He crossed the street and entered through the unlocked
front door. He then wandered from room to room checking for
intruders. He believes he turned off the kitchen light then looked
into the master bedroom. The room was very dark, but he heard
EF whisper and then shout to the sleeping boyfriend to turn on
the light. When the light came on CD ducked down at the foot of
the bed but then arose and identified himself as her neighbor. She
told him to get out. He left but wandered into the kitchen. She
then got up, guided him to the door, watched while he crossed
the street and entered his house. He stated he then went back
to sleep, woke in the morning with no memory of the incident,
and went to work. EF reported that she awoke feeling someone
stroking her abdomen under the covers. She confirmed that he
wandered into the kitchen, and she then led him out via the front
door and went back to sleep. She did not call 911, did not tell
her boyfriend about the “touching,” and was not afraid of CD
when guiding him out. When her mother returned next morning,
EF told her about the intruder. She described the touching as a
tickling sensation, then as soft stroking under her pajamas but
not under her panties. Her mother phoned the police who contacted CD to get his statement. He was surprised to hear that he
was reported to have made a sexual attack, which he repeatedly
denied. He waived his Miranda rights and told the police that he
was not near her when the light came on. The boyfriend reported
his impression of someone feeling his bedcovers. CD explained
that he was being a Good Samaritan, wanted to apologize for the
intrusion, and about his history of sleepwalking. The police had
the physician’s report of CD’s previous event. CD believed that
would be the end of it. Instead he was charged with criminal trespassing, criminal sexual abuse, and a felonious attempt to commit rape. His lawyer advised a sleep study.
Background of the November 2006 Event
The events that led to the charges against CD occurred on
11/25/06. The family was visiting her parents shortly before the
expected birth of third child. The baby arrived during that visit.
The family stayed on another week; then CD, whose work required his presence, returned home alone. This was Thanksgiving weekend; neighbors on either side of his house were away
and had asked him to check on their homes as he usually did
during their absence turning on lights to discourage intruders.
CD performed this task the evening following work. Because
he was responsible for a special program from 6 to 9 pm, he did
this later than usual. CD rarely drank alcohol and had none that
day but did drink many caffeinated beverages. He explained his
723
Figure 2—Hypnogram and distribution of arousals in PSG of CD
Arousals
Score
Patient: CD
W
REM
1
2
3
4
Graphic Display
00:00
01:00
02:00
03:00
04:00
05:00
06:00
00:00
01:00
02:00
03:00
04:00
05:00
06:00
Idiopathic
Respiratory
PLM
Time (hours)
The PSG Performed at Rush Sleep Center
best evidence of the time the intrusion occurred was given by
the complainant, EF, who checked the clock when she awoke
feeling she was being touched. It was between 4:30 and 5 am.
CD’s estimated bedtime around midnight and an awakening at 3
am and another later, might suggest that his confusional arousal
was at 4:30 from the second NREM cycle. However, that was
too speculative to serve as evidence.
To help clarify whether the arousal was from SWS, the second author, who had conducted one of the first studies to use
spectral analysis scoring of SW, was asked to score the clinical
night using the same software and criteria as used in the 2001
publication.21
Two nights of recording were ordered; the first a standard
clinical night to rule out obstructive sleep apnea (OSA), as
CD had a history of loud snoring. That study, performed on
8/16/2007, was scored according to the criteria of Rechtschaffen
and Kales,23 and the arousals by the American Sleep Disorders
Association Atlas 34 (Figure 2).The scoring was approved by the
attending neurologist. His report in summary read: “The study
and clinical history is consistent with mild positional obstructive sleep apnea syndrome.” The sleep efficiency was low (60%)
and sleep latency long (31 minutes); SWS% was low (3.6%),
and arousals per hour moderately high (18.1). The oxyhemoglobin desaturation nadir was normal at 93%, and apnea+ hypopnea index 4.1 occurred only when supine. No periodic limb
movements were recorded. The second night was to have CD
undergo 25 hours of sleep deprivation while matching his reported excessive caffeine followed by 7 hours of recovery sleep.
This would mimic the conditions the night of his intrusion event
and maximize the possibility he would exhibit a SW episode
in the laboratory. However, this plan required approval of the
University Research Committee, and CD and his wife could not
wait for this approval as the trial date was imminent.
The defense attorney requested the senior author, who had
previous experience at SW trials, to act as a sleep expert for the
defense. In the absence of further data, she reviewed the data in
hand against the criteria for sleepwalking. The one central to the
diagnosis—that the arousal typically occurs from SWS within
the first or second NREM cycle—could not be confirmed. CD
had no episode during his diagnostic night and was not able to
state conclusively his bedtime on the night of his event. His best
guess was that this was later than usual as he had worked late
and checked both neighboring houses before entering his home.
He thought this was probably sometime close to or after midnight. Given the amount of caffeine he reported drinking that
day, it was not possible to estimate the time to sleep onset nor
the amount or depth of sleep he achieved before his behavioral
arousal. CD’s wife was not able to report definitively whether
his SW episodes typically took place early in the night or later,
as she frequently slept separately due to his loud snoring. The
Independent Blind Scoring of SWA
A disk of the digitized sleep study was sent to Stanford where
it was scored by three physicians; first by the second author, then
by a visiting assistant professor of neurology and by a visiting
research fellow who scored the record for the CAP frequency.
Their final report received on 3/03/08 read: “Here are the data
from the patient you can use as the official report. The delta
power was calculated to determine the total per sleep cycle.
The FFT was performed on the C3/A2 signal with a Hamming
window applied. Two second windows were averaged over
30 epochs. Artifacts were first rejected. The results are clearly
abnormal in distribution and with low delta power during the
first sleep cycle. This is similar to what has been described by
Gaudreau et al.19 and what we ourselves found in sleepwalkers
(Guilleminault et al.21).” Attached were the data in graphic form
(Figure 3). These data points were plotted against those reported in the Gaudreau study (Figure 4). Additional findings were
“The CAP rate was abnormally high at 55. The rate for normals
in our lab is 32-35 for those of similar age.”
The report confirmed CD had abnormally low SWA in the
first NREM cycle. Espa et al.20 hypothesized that a low SWA
results in an overt SW arousal if two further conditions are met:
(1) there is pressure for more SWS (as follows sleep deprivation), and (2) there is a concurrent stimulus for increased arousal
from SWS. This may be from a medication or substance35 such
as excessive caffeine,2,31,32 from an untreated breathing disor724
Special Article
Figure 3—Total delta power for four NREM cycles in CD
Figure 4—Absolute slow wave activity across four NREM
cycles for controls, sleepwalkers and CD
Delta Power Total
9,000
3,500
Absolute SWA (æV2)
3,000
µV2/Hz
2,500
2,000
1,500
1,000
7,000
6,000
5,000
4,000
3,000
2,000
1,000
500
0
Controls
Sleepwalkers
CD
8,000
0
1
2
3
1
2
3
4
NREM cycles
4
NREM Cycle
The data for Controls and Sleepwalkers are taken from Gaudreau et al.
Dynamics of slow-wave activity during NREM sleep of sleepwalkers and
control subjects. Sleep 2000;23:755-60.
der, or an external stimulus such as noise, auditory tone,36 or a
touch.37 These two conditions were present in CD’s history—the
two weeks of prior SD and excessive caffeine still actively affecting his sleep. This provided the grounds for the senior author
to agree to be a sleep expert for the trial.
The trial began on 10/24/08. The defense lawyer was able to
have Figure 4 admitted into evidence and copies distributed to
the jury while he questioned the sleep expert about how this related to the SW disorder that was the alleged cause of the event.
This figure showed the distribution of the average SWA across
the four NREM cycles for 15 sleepwalkers and 15 age- and gender-matched controls in the Gaudreau et al.19 study and for CD.
Both he and the SW sample had significantly lower SWA in the
first NREM cycle than the C group. Given the reliability of SWA
within a person between nights, it is highly probable that CD’s
low SWA and high CAP constituted a predisposing condition
in his sleep prior to his intrusion into the neighbor’s home. His
history indicated he met the two additional conditions needed to
initiate a SW event.20 Sleep deprivation, which was the priming
condition, and the excessive caffeine and mild OSA22 both may
have precipitated the arousal into a non-rational state of mind.
The jury listened attentively. The prosecuting attorney crossexamined to clarify whether CD was fully conscious when he
responded to the light being turned on by first hiding then identifying himself. These points were addressed, citing CD’s continued mental confusion and failure to recall the event on waking.
The jury returned a verdict of not guilty on all counts.
These data constituted the basis for the opinion that the defendant was likely in a non-rational state due to SW at the time of
the events charged, even though the PSG was conducted eight
months after the event. Those who have warned sleep medicine
clinicians not to testify in forensic cases, stating this would be
tantamount to practicing “junk science,” may not have been
aware of the research establishing the reliability of the sleeping brain wave profile using spectral analysis. The application
of relevant science, including spectral analysis scoring of PSG,
should become integrated into the guidelines recommended for
those serving as sleep experts. An unfortunate aim of the critical literature has been to discourage research given the description of these efforts as “attempts to ‘stimulate’ sleepwalking in
the laboratory (by sleep deprivation, medication administration,
or alcohol ingestion) are completely worthless and totally inappropriate.”24 There is, for example, a strong need for research
involving larger samples to clarify disparate findings between
studies with small samples.21,36,37
REFERENCES
1. Broughton R, Billings R, Cartwright R, et al. Homicidal somnambulism: a case
report. Sleep 1994;17:253-64.
2. Cartwright R. Sleepwalking violence: a sleep disorder, a legal dilemma, and a
psychological challenge. Am J Psychiatry 2004;161:1149-58.
3. Mahowald M, Schenck C. Complex motor behavior arising during the sleep period: forensic science implications. Sleep 1995;18:724-7.
4. Mahowald M, Schneck C. Sleep-related violence and forensic medicine issues.
In: Chokroverty S, ed. Sleep disorders medicine. Boston: Butterworth Heinemann, 1999:729-39.
5. Mahowald M, Schenck C. Parasomnnias: sleepwalking and the law. Sleep Med
Rev 2000;4:321-39.
6. Broughton RJ. Sleep disorders: disorders of arousal? Science 1968;159:1070-8.
7. Hublin C, Kaprio J, Partinen M, Heikkila K, Koskenvuo M. Prevalence and genetics of sleepwalking: a population-based twin study. Neurology 1997;48:177-81.
8. Kales A, Soldatos C, Bixler E. et al. Hereditary factors in sleepwalking and night
terrors. Brit J Psychiat 1980;137:111-8.
Discussion
Laboratory-based research identifying SWA as differentiating SW from normal sleep using spectral analysis was replicated in a forensic case. The additional history of snoring and
mild breathing disorder validated in the PSG may be a contributing cause of his low SWA, high CAP rate, and arousal into
non-conscious acts when sleep deprived and over-caffeinated.
725
9. Blatt I, Peled R, Gadoth N, Lavie P. The value of sleep recording in evaluating somnambulism in young adults. Electroencephalogr Clin Neurophysiol
1991;78:407-12.
10. Nofzinger E, Wettstein R. Homicidal behavior and sleep apnea: a case report and
medicolegal discussion. Sleep 1995;18:776-82.
11. Mahowald M, Schenck C. Violent parasomnias: forensic medicine issues In:
Kryger M, Roth T, Dement W, eds. Principles and practice of sleep medicine.
Philadelphia: WB Saunders, 2000:786-95.
12. Shapiro C, Trajanovic N, Fedoroff P. Sexsomnia- a new parasomnia? Can J Psychiatry 2003; 48:311-7.
13. Cartwright R. The twenty-four hour mind: The role of sleep and dreaming in our
emotional lives. New York: Oxford University Press 2010.
14. Bonkalo A. Impulsive acts and confusional states during incomplete arousal from
sleep: criminological and forensic implications. Psychiatr Q 1974;48:400-9.
15. American Sleep Disorders Association. International classification of sleep disorders revised: diagnostic and coding manual. Rochester, MN: American Sleep
Disorders Association, 1997.
16. American Psychiatric Association. Diagnostic and statistical manual of mental
disorders IV-TR Washington, DC: American Psychiatric Association, 2000.
17. Pressman M. in Arizona v Falater 1999: Court Case of State of Arizona.
18. Pressman M, Mahowald M, Schenck C, Bornemann MC. Alcohol-induced sleepwalking or confusional arousal as a defense to criminal behavior: a review of scientific evidence, methods and forensic considerations. J Sleep Res 2007;16:198-212.
19. Gaudreau H, Joncas S, Zadra A, Montplaisir J. Dynamics of slow- wave activity during the NREM sleep of sleepwalkers and control subjects. Sleep 2000;23:755-60.
20. Espa F, Ondze B, Deglise P, Billiard M, Besset A. Sleep architecture, slow wave
activity, and sleep spindles in adult patients with sleepwalking and sleep terrors.
Clin Neurophysiol 2000;111:929-39.
21. Guilleminault C, Poyares D, Abat F, Palombini L. Sleep and wakefulness in somnambulism: a spectral analysis study. J Psychosomatic Res 2001;51:411-6.
22. Guilleminault C, Kirisoglu C, da Rosa A, Lopes C, Chan A. Sleepwalking, a disorder of NREM sleep instability. Sleep Med 2006;7:163-70.
23. Rechtschaffen A, Kales A. A manual of standardized terminology techniques and
scoring system for sleep stages of human subjects. Bethesda, MD: U.S. Dept. of
Health Education and Welfare 1968.
24. Mahowald M, Bornemann MC, Schenck C. Finally- sleep science for the courtroom. Sleep Med Rev 2007;11:1-3.
25. Preud’homme XA, Lanquart J-P, Mendlewicz J, Linkowski P. Distribution of delta
activity across nonrapid eye movement sleep episodes in healthy young men.
Sleep 1997;20: 313-30.
26. Tan X, Campbell G, Palagini L, Feinberg I. High internight reliability of computermeasured NREM delta, sigma, and beta: biological implications. Biol Psychiatry
2000;48:1010-9.
27. Tan X, Campbell G, Feinberg I. Internight reliability and benchmark values for
computer analyses of non-rapid eye movement (NREM) and REM EEG in normal
young adult and elderly subjects. Clini Neurophysiol 2001;112:1540-52.
28. Tucker A, Dinges D, Van Dongen H. Trait interindividual differences in the sleep
physiology of healthy young adults. J Sleep Res 2007;16:170-80.
29. University of Utah. Caffeine content of popular drinks. http://www.math.utah.
edu/~yplee/fun/caffeine.html
30. Patat A, Rosenzweig P, Enslen M. Effects of a new slow release formulation of
caffeine on EEG, psychomotor and cognitive functions in sleep-deprived subjects. Hum Psychopharmacol Clin Exp 2005;15:153-70.
31. Landolt H, Werth E, Borbely A, Dijk D-J. Caffeine intake (200 mg) in the morning
affects human sleep and EEG power spectra at night. Brain Res 1995;675:67-74.
32. Landolt H, Dijk D-J, Gaus S, Borbely A. Caffeine reduces low-frequency delta
activity in the human sleep EEG. Neuropsychopharmacology 1995;12:229-38.
33. Moldofsky H, Gilbert R, Lue F, MacLean A. Sleep-related violence. Sleep
1995;18:731-9.
34. American Sleep Disorders Association Atlas Task Force: EEG arousals: scoring
rules and examples. Sleep 1992;15:173-84.
35. Cartwright R. Parasomnias due to medications or substances. In: Thorpy M,
Plazzi G, eds. Parasomnias and other sleep-related movement disorders. Cambridge University Press, 2010: 42-53.
36. Pilon M, Montplaisir J, Zadra A. Precipitating factors of somnambulism: impact of
sleep deprivation and forced arousals. Neurology 2008;70:2284-90.
37. Pressman M. Disorders of arousal from sleep and violent behavior: the role of
physical contact and proximity. Sleep 2007;30:1039-47.
38. Perrault R, Carrier J, Desautels A, Montplaisir J, Zadra A. Slow wave activity and
slow oscillations in sleepwalkers and controls: effects of 38 h of sleep deprivation.
J Sleep Res 2013 Feb 11. [Epub ahead of print].
Acknowledgments
The authors thank Babak Mokhlesi, M.D., for his additional literature search, Eric
Frenette, M.D., for providing an independent scoring of the SWA in the PSG of CD,
and Jacques Montplaisir, M.D., for permission to reprint Figure 1 from Gaudreau et
al. Dynamics of slow-wave activity during NREM sleep of sleepwalkers and control
subjects. Sleep 2000;23:755-60.
Submitted for publication February, 2013
Submitted in final revised form April, 2013
Accepted for publication April, 2013
Address correspondence to: Rosalind Cartwright, 680 N. Lake Shore Dr., Chicago, IL
60611; Tel: (312) 642-1065; Fax: (312) 642-1371; E-mail: drrdcartwright@gmail.com
DISCLOSURE STATEMENT
conflicts of interest. The senior author acted for the defense pro bono.
726
S leep M edicine P earls
Ventricular or Pseudo-Ventricular Tachycardia on
Polysomnogram
Romy Hoque, M.D.; Lourdes DelRosso, M.D.
Division of Sleep Medicine, Department of Neurology, Louisiana State University School of Medicine, Shreveport, LA
A
30-year-old man with past medical history of
asthma presented for evaluation of nocturnal
shortness of breath and daytime fatigue. His medications included an albuterol inhaler used as needed,
fluticasone-salmeterol discus inhaler used daily, and
cetirizine used daily. He smoked half a pack of cigarettes each day. Physical exam revealed an obese man
with a body mass index of 33. Cardiac and respiratory
exam were unremarkable.
A split-night polysomnogram (PSG) was performed. During the diagnostic portion of the PSG,
total sleep time (TST) was 173 min, sleep onset was 3
min, sleep efficiency was 90%, wake after sleep onset
was 16 min, TST apnea hypopnea index (AHI) was
5.9, and REM AHI was 85.7. During the titration portion of the PSG, sleep disordered breathing resolved
at continuous positive airway pressure of 11 cm.
PSG review revealed an electrocardiogram (ECG)
finding shown in Figure 1A, with an enlarged view of
the relevant ECG shown in Figure 1B.
QUESTION:
What is the significance of this ECG finding?
Figure 1
(A) Electrocardiogram (ECG) finding on polysomnogram (PSG). (B) Enlarged view of the ECG2 channel from the PSG.
727
R Hoque and L DelRosso
upper chest and left lower chest electrodes. The augmented leads
are calculated post-acquisition and do not require the use of additional electrodes (Table 1). The use of multiple ECG channels
allows the sleep physician the opportunity to more fully evaluate
ECG rhythms and avoid electrodes contaminated by artifact.
Pseudo-ventricular tachycardia (VT) associated with tremor
like movements has been previously reported.4 Three signs are
useful in the identifying tremor/movement induced pseudo-VT
on ECG: the sinus sign, the spike sign, and the notch sign.5
The sinus sign in pseudo-VT is the presence of a normal sinus rhythm with normal P, QRS, and T waves in either a bipolar
lead or an augmented lead during an apparent episode of VT.
This is due to the fact that one of the upper limb electrodes is
free of movement artifact.
The spike sign in pseudo-VT is the presence of regular or
irregular spikes among the wide-complex QRS artifact that
represents the superimposition of a normal sinus rhythm QRS
complex during the vent.
The notch sign in pseudo-VT is the presence of a superimposed notch on the wide-complex QRS artifact that also represents the superimposition of a normal sinus rhythm QRS
complex during the event.
In Figure 3 we can see the spike sign interspersed among the
pseudo-VT wide-complex QRS artifact. In PSG interpretation,
prompt evaluation of both the video recording and the patient’s
status is also recommended. Our patient was medically stable
throughout the course of the recording.
ANSWER:
Pseudo-ventricular tachycardia due to nocturnal scratching
of the precordial electrocardiogram electrodes (Figures 2, 3).
Discussion
Cardiac arrhythmias are the most common severe adverse effect encountered on nocturnal PSG.1 The severity of the cardiac
arrhythmia noted on PSG may correlate with the apnea severity,
especially in men, and may be seen more frequently in those
with comorbid cardiac disease.2 It is critical for sleep technologists to recognize these events and take appropriate action.
The American Academy of Sleep Medicine (AASM) Manual
for Scoring of Sleep and Associated Events (Scoring Manual)
recommends the use of a modified Lead II (ECG2) for cardiac
evaluation with two ECG electrodes: one on the right upper chest
and another on the left lower chest.3 In our sleep laboratory, we
apply three ECG electrodes: on the right upper chest, the left upper chest, and the left lower chest. The software in our PSG system (Alice 5, Respironics, Inc., Murrysville, PA, USA) allows
the display of six ECG channels—three bipolar channels and
three augmented channels (Figure 3). The three bipolar channels are Lead I (ECG1) consisting of both upper chest electrodes;
Lead II (ECG2) consisting of the right upper chest and the left
lower chest electrodes; and Lead III (ECG3) consisting of the left
Figure 2—Scratching artifact on electrocardiogram channel
728
Sleep Medicine Pearls
Figure 3—Scratching artifact visualized on electrocardiogram using 6 channels: 3 bipolar leads and 3 augmented leads
Table 1—Augmented electrocardiogram (ECG) leads in polysomnography
Augmented ECG leads
Augmented voltage right arm (aVR)
Positive electrode
Right arm electrode
Composite Negative
Electrode Source
Left arm electrode
Left leg electrode
Post-Acquisition Formula in
Augmented ECG lead calculation
–(Lead I + Lead II) / 2
Augmented voltage left arm (aVL)
Left arm electrode
Right arm electrode
Left leg electrode
(Lead I – Lead III) / 2
Augmented voltage left foot (aVF)
Left leg electrode
Left arm electrode
Right arm electrode
(Lead II + Lead III) / 2
Clinical Pearls
References
1. Mehra R, Strohl KP. Incidence of serious adverse events during nocturnal polysomnography. Sleep 2004;27:1379-83.
2. Szaboova E, Holoubek D, Tomori Z, Szabo P, Donic V, Stancak B. Severity of
nocturnal cardiac arrhythmias correlates with intensity of sleep apnea in men.
Adv Exp Med Biol 2013;755:155-68.
3. Iber C, Anconi-Israel S, Chesson A, Quan S. American Academy of Sleep Medicine. The AASM manual for the scoring of sleep and associated events : rules,
terminology, and technical specifications. Westchester, IL: American Academy
of Sleep Medicine; 2007.
4. Riaz A, Gardezi SK, O’Reilly M. Pseudo ventricular tachycardia: a case report.
Irish J Med Sci 2010;179:295-6.
5. Huang CY, Shan DE, Lai CH, et al. An accurate electrocardiographic algorithm
for differentiation of tremor-induced pseudo-ventricular tachycardia and true
ventricular tachycardia. Int J Cardiol 2006;111: 163-5.
1. Scratching can produce a pseudo-ventricular tachycardia
artifact on electrocardiography (ECG) monitoring
2. The sinus sign, the spike sign, and the notch sign are
useful in the identification of movement induced pseudoventricular tachycardia.
3. Evaluation of the PSG video recording for movement
and assessment of patient clinical status are also
recommended.
4. The use of an extended ECG montage composed of three
bipolar channels and three augmented channels may
allow more extensive visualization of ECG abnormalities
when compared to single ECG channel.
Citation
Submitted in final revised form February, 2013
Accepted for publication February, 2013
Hoque R; DelRosso L. Ventricular or pseudo-ventricular tachycardia on
polysomnogram. J Clin Sleep Med 2013;9(7):727-730.
729
R Hoque and L DelRosso
Address correspondence to: Romy Hoque, M.D., Division of Sleep Medicine,
Department of Neurology, Louisiana State University School of Medicine, Shreveport,
LA; E-mail: romy.hoque@gmail.com
This work was performed at the Louisiana State University School of Medicine in
Shreveport, LA. The authors have no conflicts of interest, financial support, or offlabel/investigational uses to disclose.
730
Is Prediction of CPAP Adherence in Obstructive Sleep Apnea in
the Perioperative Setting Feasible?
Antonio M. Esquinas, M.D., Ph.D.1; Peter Cistulli, M.D., F.A.A.S.M.2
Intensive Care Unit, Hospital Morales Meseguer, Murcia, Spain; 2Centre for Sleep Health & Research, Department of Respiratory
Medicine, Royal North Shore Hospital, University of Sydney, Australia
L etter to the E ditor
1
T
here is growing interest in the of continuous
positive airway pressure (CPAP) in the perioperative setting with untreated obstructive sleep apnea
(OSA).1
Guralnick et al. describe CPAP adherence in patients with newly diagnosed OSA prior to elective surgery.2 The major findings were that African American
race, male gender and depressive symptoms were associated with reduced CPAP adherence. However, we
consider that some aspects of this study need clarification in order to extrapolate the findings into clinical
practice.
First, we believe that the overall results could be
separated according to the OSA severity categories of
moderate or severe. This separation into two different
groups could clarify the overall message of the study,
since the literature suggests that more severe OSA is
associated with greater CPAP adherence compared to
milder severities.3 Moreover, previous studies have
shown that a higher baseline apnea-hypopnea index
(AHI) was the only significant independent predictor
of better CPAP compliance.4
Secondly, the finding of male gender being a risk
factor for poor CPAP adherence warrants further
consideration.2 We suggest that male gender may be
a surrogate for risk factors or comorbidities such as
smoking, obesity, or anthropometric parameters such
as neck circumference.
Thirdly, depressive symptomatology has been
reported to be an independent predictor of reduced
CPAP adherence.5 This aspect may have a dual interpretation in the population tested in this study and is
not clearly in the same direction of previous studies.
Fourth, there is lack information about some relevant practical aspects that was not examined: (a) type
of surgery and extent of postoperative pain; (b) the
auto-set CPAP settings were atypical, with a pressure
range of only 5 around the optimal pressure derived
from polysomnography (PSG); (c) they do not report
on the efficacy of CPAP from the machine download;
(d) 50% of patient who scored highly on the ques-
tionnaire refused PSG—were their postoperative
outcomes different to those who were treated with
CPAP?
We propose to include a more accurate assessment
of possible psychological disorders that may interfere
with patient adherence to CPAP and hospital education program before surgery.
Citation
Esquinas AM; Cistulli P. Is prediction of cpap adherence in obstructive
sleep apnea in the perioperative setting feasible? J Clin Sleep Med
2013;9(7):731.
References
1. Porhomayon J, El-Solh A, Chhangani S, Nader ND. The management of surgical patients with obstructive sleep apnea. Lung
2011;189:359-67.
2. Guralnick AS, Pant M, Minhaj M, Sweitzer BJ, Mokhlesi B. CPAP
adherence in patients with newly diagnosed obstructive sleep apnea prior to elective surgery. J Clin Sleep Med 2012 15;8:501-6.
3. Oksenberg A, Arons E, Froom P. Does the severity of obstructive
sleep apnea predict patients requiring high continuous positive airway pressure? Laryngoscope 2006;116:951-5.
4. Hui DS, Choy DK, Li TS, et al. Determinants of continuous positive
airway pressure compliance in a group of Chinese patients with
obstructive sleep apnea. Chest 2001;120:170-6.
5. Fuchs FS, Pittarelli A, Hahn EG, Ficker JH. Adherence to continuous positive airway pressure therapy for obstructive sleep apnea:
impact of patient education after a longer treatment period. Respiration 2010;80:32-7.
submission & correspondence
Information
Submitted for publication March, 2013
Address correspondence to: Antonio M. Esquinas, M.D., Ph.D, Avenida
Marques de los Velez s/n, Murcia, 30008 Spain; Tel: +34609321966;
Fax: +34968232484; E-mail: antmesquinas@gmail.com
The authors have indicated no financial conflicts of interest.
731
CPAP Adherence during the Perioperative Period
Response to Esquinas and Cistulli. Is prediction of CPAP adherence in obstructive sleep
apnea in the perioperative setting feasible? J Clin Sleep Med 2013;9:731.
Babak Mokhlesi, M.D., M.Sc., F.A.A.S.M.; Amy S. Guralnick, M.D.
L etter to the E ditor
University of Chicago Pritzker School of Medicine, Chicago, IL
W
or reported in the context of CPAP adherence, two
prior studies have reported an association between
lower psychological well-being and reduced CPAP
adherence.7,8 As we discussed in the limitations, the
findings in our inner-city urban cohort may not be
applicable to other populations, and as such, further
studies are needed to confirm our findings. We agree
that our study was limited by lacking information on
the extent of postoperative pain and sedative-narcotic
use. However, the types of surgery are described in
the first paragraph of the results.
Our entire cohort was directly referred from the
Anesthesia Perioperative Medicine Clinic to the sleep
laboratory for a diagnostic polysomnogram. On average, the patients underwent in-laboratory split-night
polysomnograms just 4 days before the scheduled date
of surgery. Given the time constraints and the inability of the sleep clinicians to fully evaluate the patients
prior to surgery, we implemented a program in which
the patients would receive an auto-PAP device upon
awakening in the sleep laboratory. In regards to the
pressure settings of the auto-PAP devices, we believe
that providing a very wide range of pressures (e.g.,
4-20 cm H2O) was not necessary since our patients had
all been manually titrated during the polysomnogram.
We set the upper limit of the auto-pap pressure just a
few cm of H2O above the optimal pressure with the rationale that in the immediate postoperative period, the
patients may need a higher pressure due to the effect
of sedatives and narcotics on the upper airway collapsibility. We agree that having additional information on
residual AHI and mask leak as estimated by the CPAP
units would have been of interest, but unfortunately
not all CPAP units had the capability of reporting these
variables. We also agree that having postoperative outcomes would have strengthened our study, but as we
pointed out in our limitations, given that overall serious postoperative complications due to OSA are rare,
our study was neither powered nor designed to ascertain rates of postoperative complications.
Of interest, in a recent randomized controlled trial
of patients undergoing elective hip/knee arthroplasty,
patients suspected of having moderate or severe OSA
were randomized to auto-PAP therapy during the
e thank Drs. Esquinas and Cistulli for their
thoughtful comments and careful review of our
recent study in the Journal of Clinical Sleep Medicine.1 We believe the most important finding of our
study was that the majority of patients referred from
the Anesthesia Perioperative Medicine Clinic and
subsequently diagnosed with obstructive sleep apnea
(OSA) were poorly adherent to auto-PAP therapy during the perioperative period (median adherence of 2.5
h per night). As we pointed out in our discussion, the
overall adherence was significantly lower than what
we have reported in our clinical practice as well as
what has been widely reported in the literature.2-5 In
multivariate linear regression modeling, OSA severity
was not a predictor of CPAP adherence. Esquinas and
Cistulli note that increasing OSA severity is associated with better CPAP adherence. However, there are
conflicting data on the association of OSA severity
and the presence of daytime sleepiness with adherence to CPAP therapy. Although disease severity is
frequently identified as influential on CPAP adherence, the relationships are relatively weak, and when
other factors are included, disease severity and sleepiness are less contributory to CPAP adherence.6,7 In
another independent cohort of 403 non-presurgical
patients seen at our institution, we also found that the
severity of OSA and the Epworth Sleepiness Scale
were not predictive of CPAP adherence.2 In fact, the
only predictors of reduced CPAP adherence in that cohort were African American race and non-sleep specialists ordering polysomnograms and CPAP therapy.
We agree that male gender may simply be a marker
of other important predictors of CPAP adherence, but
unfortunately, our data do not allow us to better delineate it. Our regression model shows that a score
above 16 on the validated Center for Epidemiologic
Studies Depression Scale was independently associated with 65 minutes lower mean CPAP adherence per
night during the first 30 days of therapy. Esquinas and
Cistulli contend that this association has “dual interpretation in the population tested in this study and is
not clearly in the same direction of previous studies.”
Although we acknowledge that validated measures
of depressive symptoms are not routinely measured
733
B Mokhlesi and AS Guralnick
postoperative period vs. standard of care. Even in this rigorous
clinical trial, the median postoperative daily auto-PAP usage
was suboptimal at 184.5 minutes per night. Moreover, empiric
auto-PAP therapy led to a 1 day increase in median length of
stay in those that were adherent to therapy.9 This study raises
new questions about the role for empiric postoperative autoPAP therapy. Therefore, we wholeheartedly agree with Esquinas and Cistulli that further research is needed to identify
barriers to CPAP adherence in this patient population, or efforts
directed towards diagnosis are likely to be wasted. Given the
large volume of elective surgeries performed globally, implementation of systematic screening and empiric auto-PAP therapy in patients at risk for OSA would impose a significant cost
burden. This underlines the need for further clinical research to
determine the most efficient methods to identify presurgical patients that would benefit from CPAP therapy as well as the utility of education programs before surgery that aim to improve
perioperative CPAP adherence and patient outcomes.
4. Engleman HM, Martin SE, Douglas NJ. Compliance with CPAP therapy in patients with the sleep apnoea/hypopnoea syndrome. Thorax 1994;49:263-6.
5. Reeves-Hoche MK, Meck R, Zwillich CW. Nasal CPAP: an objective evaluation
of patient compliance. Am J Respir Crit Care Med 1994;149:149-54.
6. Sawyer AM, Gooneratne NS, Marcus CL, Ofer D, Richards KC, Weaver TE.
A systematic review of CPAP adherence across age groups: clinical and empiric insights for developing CPAP adherence interventions. Sleep Med Rev
2011;15:343-56.
7. Poulet C, Veale D, Arnol N, Levy P, Pepin JL, Tyrrell J. Psychological variables
as predictors of adherence to treatment by continuous positive airway pressure.
Sleep Med 2009;10:993-9.
8. Edinger JD, Carwile S, Miller P, Hope V, Mayti C. Psychological status, syndromatic measures, and compliance with nasal CPAP therapy for sleep apnea.
Percept Mot Skills 1994;78:1116-8.
9. O’Gorman SM, Gay PC, Morgenthaler TI. Does auto-titrating positive airway
pressure therapy improve postoperative outcome in patients at risk for obstructive sleep apnea syndrome? a randomized controlled clinical trial. Chest 2013.
Jan 3. http://dx.doi.org/10.1378/chest.12-0989. [Epub ahead of print]
Submitted for publication March, 2013
Address correspondence to: Babak Mokhlesi, M.D., M.Sc., Associate Professor
of Medicine, Section of Pulmonary and Critical Care Medicine, Director of Sleep
Disorders Center, The University of Chicago Pritzker School of Medicine, 5841
S. Maryland Ave, MC 6076, Room M630, Chicago, IL 60637; E-mail: bmokhles@
medicine.bsd.uchicago.edu
Citation
Mokhlesi B; Guralnick AS. CPAP adherence during the perioperative period. J Clin
Sleep Med 2013;9(7):733-734.
References
1. Esquinas AM, Cistulli P. Is prediction of CPAP adherence in obstructive sleep
apnea in the perioperative setting feasible? J Clin Sleep Med 2013;9:731.
2. Pamidi S, Knutson KL, Ghods F, Mokhlesi B. The impact of sleep consultation
prior to a diagnostic polysomnogram on continuous positive airway pressure
adherence. Chest 2012;141:51-7.
3. Kribbs NB, Pack AI, Kline LR, et al. Objective measurement of patterns of nasal CPAP use by patients with obstructive sleep apnea. Am Rev Respir Dis
1993;147:887-95.
Dr. Mokhlesi has received consultant fee from Philips/Respironics. Dr. Guralnick
has indicated no financial conflicts of interest.
734

Volume 9, Number 7 | July 15, 2013

Transcription

Similar documents

Marlton Office – New Patient Forms

SUMMERHOUSE OF SHOREVIEW

Media Release - The Black Sorrows

Flexi Puf New 10x8 Brochure

Visual Memory and Imagery

CareFusion - MedDirect

Obstructive Sleep Apnoea Syndrome

The Chopra Center presents 7 Steps to Radical Wellbeing by

25% off - Bensons for Beds

Treatments for Obstructive Sleep Apnea

The Tonsils and Adenoids in Pediatric Patients