Download SLEEP-WAKE Research in The Netherlands

SLEEP-WAKE Research in The Netherlands
Volume 10
1999
This publication was sponsored by
The Netherlands
1
Dutch Society for Sleep-Wake Research
Founded at Leiden, The Netherlands at June 7, 1990
Board
Prof. dr. A.M.L. Coenen
Prof. dr. G.A. Kerkhof
Dr. A. Knuistingh Neven
Dr. J.H.M. de Groen
Dr. A.L. van Bemmel
chairman
vice-chairman
secretary
treasurer
second secretary
Scientific committee
Dr. A.L. van Bemmel
Dr. D.G.M. Beersma
Prof. Dr. H. Folgering
Dr. W.F. Hofman
Dr. G.S.F. Ruigt
chairman
member
member
member
member
2
SLEEP-WAKE Research in The Netherlands
Volume 10
1999
Published by
Dutch Society for Sleep-Wake Research
Edited by
Alex L. van Bemmel
Maastricht University
Domien G.M.Beersma
University of Groningen
Hans Folgering
University of Nijmegen
Winnie F. Hofman
University of Amsterdam
Gé S.F. Ruigt
Organon International B.V.
3
1999
Dutch Society for Sleep-Wake Research, Maastricht, The Netherlands
ISBN
90-73675-10-3
Lay-0ut:
Bureau van de Manakker, Maastricht
4
Preface
Sleep-Wake Research in The Netherlands, volume 10 1999, contains a set of
papers presenting the on-going research and progress in this field (regular papers). The content is the result of the interactions between the authors and the
members of the scientific committee of the Dutch Society for Sleep-Wake Research (NSWO), who served as referees.
The society started the year 1999 on November 21 1998 with the scientific
meeting at Kempenhaeghe, Heeze. This meeting was organised at the occasion
of the retirement of our colleague dr. Guus Declerck from the epilepsy and
sleep-wake centre Kempenhaeghe. Also for this reason the meeting was a joint
meeting with our Belgian colleagues. Proceedings of this first joint meeting are
also included in the present volume. In the general meeting for members dr.
Domien Beersma announced that he preferred to leave the board of the society. Dr. Alex van Bemmel was chosen as a new board member and as the new
chairman of the scientific committee.
The Dutch Handbook Sleep and Sleep Disturbances is in steady progress and
will appear in the year 2000. The consensus reports on ‘obstructive sleep apnoea syndrome’ and ‘non-pharmaceutical sleep therapy’ are ready. On request
these publications are available. New consensus committees will be installed
on other topics of societal and scientific interest, such as the assessment of
vigilance. The society is trying to get a module sleep and sleep disturbances in
the education for health psychologist. Furthermore, the society will develop
criteria for the formal recognition of genuine sleep-wake centres.
Our main sponsor the pharmaceutical company Synthelabo joined with Sanofi.
Fortunately, this has no consequences for our partnership. Also the societal 3S (‘sleep line service’) project will be continued. In this project both partners,
together with the Dutch Society of Sleep Apnoea Patients and the Dutch Society for Narcolepsy, give information and education to citizens on the field of
sleep and sleep disturbances. The home page on the Internet is updated and is
more accessible with many information for a large audience. It contains also
many interesting links to other sleep sites worth visiting (www.socsci.kun.nl/
psy/nswo).
The annual spring meeting was organised in Amsterdam on June 12 1999 by
Prof. Gerard Kerkhof and dr. Winni Hofman. At this occasion Prof. Piet Visser
was appointed honorary member of the society. Prof. Visser, the ‘grandfather’
of Dutch sleep research, was of eminent importance for the development of
this type of research in The Netherlands. The autumn meeting, which is the
closing meeting of the year will take place in Maastricht on November 26 1999.
The organiser is dr. Alex van Bemmel. In the mean time the fourth conference
5
of the World Federation of Sleep Research Societies was organised in Dresden
from October 5 till 9, with many participants of the Dutch federation.
Finally, the society will already announce its spring meeting in 2000. This
meeting will take place in Luxemburg, in Esch-sur-Alzette on May 19 and 20,
2000, together with our colleagues from Belgium and Luxemburg. Finally, it is
worthwhile to note that the Dutch Society for Sleep-Wake Research can celebrate its 10 years of existence on June 7 2000. First, however, it is with pride
and pleasure that our federation presents its tenth volume Sleep-Wake Research
in The Netherlands. The society is again indebted to our sponsor SanofiSynthelabo, who makes this annual publication feasible.
Prof. Anton M.L. Coenen, chairman
Prof. Piet Visser
6
7
Contents
Nathaniel Kleitman 1895-1999: a legend in sleep research
A.M.L. Coenen ............................................................................................. 13
A survey of SLEEP-WAKE research in The Netherlands anno 1999
A.M.L. Coenen (editor) ................................................................................ 15
Proceedings of the first joint scientific meeting of the Belgian and Dutch Sleep
Societies in honor of the retirement of dr. A.C. Declerck ........................ 21
Center for SLEEP-WAKE disorders, Kempenhaeghe: past, present, future
J.H.M. de Groen ........................................................................................... 23
Circadian control of hormonal rhythms: important roles for suprachiasmatic
nucleus efferents and the autonomic nervous system
A.Kalsbeek, I.F. Palm, S.A. la Fleur and R.M.Buijs ....................................... 25
Body temperature and sleep-wake rhythms torn apart: implications for the regulation of mood (abstract)
D.G.M. Beersma ........................................................................................... 37
Hormonal and behavioral effects of prenatal stress: focus on circadian rhythms
and sleep
O. van Reeth, C. Dugivic, M. Koehl, L. Weibel and S.Maccari ..................... 39
EEG-Topography during slow-wave sleep, normative data and modifications
induced by thalamic lesions
D. Dive, R. Poirrier, F. Claes and G. Franck ................................................. 51
Interindividual phase differences in circadian rhythms and sleep (abstract)
G.A. Kerkhof ................................................................................................ 59
Neurophysiology of sleep-wake states in relation to consciousness and information processing
A.M.L. Coenen ............................................................................................. 63
Pathophysiology of narcolepsy (abstract)
M. Billiard ..................................................................................................... 77
8
Laryngeal dysfunction during sleep (abstract)
D.Pevernagie ................................................................................................ 79
Surgical treatment for sleep-related breathing disorders: possibilities and limitations
A. Boudewijns .............................................................................................. 81
Regular papers ....................................................................................... 87
Arousal detection in sleep
F.W. Bes, H. Kuykens and A.Kumar ............................................................. 89
Sleep monitoring equipment affects the assessment of nocturnal oxygenation
in patients with COPD
F. Brijker, F.J.J. van den Elshout, Y.F.Heijdra and H.Th.M. Folgering .......... 93
Similarities between deep slow wave sleep and abscence epilepsy
A.M.L. Coenen ............................................................................................. 99
Functional assessment and treatment of sleeping problems in developmentally
disabled children: case studies
R.Didden and L.M.G.Curfs ......................................................................... 105
Diurnal characteristics of coagulation and fibrinolysis in exhausted subjects
R. van Diest and K. Hamulyák ................................................................... 109
A case study of the free-running period of a morning type subject in time isolation
H.P.A. van Dongen, M.G.C.E. Kuijpers, H. Duindam and G.A. Kerkhof ..... 113
The relation between motor activity and daily stress in remitted bipolar outpatients
R.Havermans, A.L. van Bemmel and N.A.Nicolson .................................... 119
Multiple Sleep Latency Test: are four tests necessary to diagnose hypersomnia ?
S.E. Jenkins, R.J.Schimsheimer, G.A. Kerkhof and A.W. de Weerd .......... 123
9
Do non-benzodiazepine-hypnotics prove a valuable alternative to
benzodiazepines for the treatment of insomnia ?
A.Knuistingh Neven ................................................................................... 127
A psychophysiological study of sleep onset by means of dynamic spectral analysis
and ERP
A.R. Koning, W.F. Hofman and K.R. Ridderinkhof ..................................... 135
Ambulatory assessment of sympathovagal heart activity in primary insomniacs
R. Mullaart, W.F. Hofman, G.A. Kerkhof and A.Knuistingh Neven............. 139
P300 in sleep state misperception
R.J. Schimsheimer, M.M.R. Verhelst and D.Zeeman .................................. 143
The effect of treatment with melatonin for chronic sleep onset insomnia in
children with attention deficit hyperactivity disorder: randomized placebo-controlled trial
M.G. Smits, E.J. Nagtegaal, G.A. Kerkhof, S.Valentijn
and A.L.M. Coenen .................................................................................... 147
Modelling the relation of body temperature and sleep: importance of the circadian rhythm in skin temperature
E.J.W. van Someren ................................................................................... 153
Dynamics of cortical EEG power decrease rate during entry into natural hypothermia in European ground squirrels
A.M.Strijkstra, T. Deboer and S.Daan ........................................................ 157
Exploring daytime sleepiness during migraine
J.H.M. Tulen, D.L. Stronks, L. Pepplinkhuizen and J. Passchier ............... 163
Effect of non-pharmacological treatment on polysomnography, sleep/wake diary and questionniaires in patients with primary insomnia
I. Verbeek and Y. Sweere ............................................................................ 169
Night-to-night variability of apnea indices
M.M.R. Verhelst, R.J. Schimsheimer, C. Kluft and A.W. de Weerd ............ 175
10
11
12
Nathaniel Kleitman 1895-1999:
a legend in sleep research
ANTON M.L. COENEN
Nathaniel Kleitman died on Friday august 13 1999 on the blessed age of 104
years in Santa Monica (California). Of Jewish origin he was born in 1895 in
Kishinev (Russia), now the capital of Moldavia. He studied mathematics and
sciences and, for one year, medicine in Beirut (Libanon). His study was interrupted by the first World War and he escaped to New York in 1915. Kleitman
graduated at the City College in 1919 and completed his PhD in physiology at
Columbia on ‘Studies on the physiology of sleep’. In 1921 he moved to professor A.J. Carlson’s laboratory at the University of Chicago and stayed briefly in
Henri Pieron’s laboratory in Paris. Kleitman worked on several fields of sleep
research. He was deeply interested in consciousness and he reasoned that he
could get insight in consciousness by studying the unconsciousness of sleep.
Kleitman’s master work was the ‘bible’ of sleep research ‘Sleep and Wakefulness’, University of Chicago, first edition in 1939, and a revised and updated
edition in 1963. This edition contains 4337 references and Kleitman apologised in the Preface that his reading ability was limited to English, French,
German, Italian and Russian. In 1938 he performed research to circadian
rhythms and tried to reveal whether the 24-hour rhythm is an independent
phenomenon or just an adjustment to the alternation of day and night. In 1938
he stayed, together with his young assistant Bruce Richardson 32 days in a
chilly and wet cave, the Mammouth Cave, in Kentucky. The question was
whether an adaptation to a 28-hours day should take place. Unfortunately,
while Richardson achieved to adapt to this long, artificial day, the poor Kleitman
did not adapt to this new rhythm.
A famous student of Kleitman was Eugene Aserinsky, born in 1921 in New York
and killed in an accident in 1998. He started to investigate the eye motility
during sleep and discovered so the regularly occurring phases with ‘rapid eye
movements’. This discovery meant a revolution in sleep research, than it implied the existence of an active kind of sleep. It took Aserinsky time to convince the sceptical Kleitman, a fervent supporter of the uniform, passive character of sleep. In 1953, however, Aserinsky as the first author and Kleitman as
the senior author announced the phenomenon of REM sleep. Intuitively, they
felt that REM sleep could be the type of sleep that was associated with dream-
13
ing. Another famous student of Kleitman, William Dement, started to investigate this relationship. Many data were in favour of a unique relationship which
caused a firm belief in this concept. Disputes on this point, however, are heard
from that time. A point of continuously interest for Kleitman was the issue of
the ‘basic rest activity cycle’. Kleitman believed in a permanent cycle of 90
minutes of rest and activity during the 24 hours. During the night this is expressed in an active REM period followed by a period of slow wave sleep, while
this rest-activity cycle is more covered during the day. Also this cycle is still a
matter of debate.
Kleitman was married with Paulena and had two daughters, Hortense and Esther.
The world of sleep research owes many thanks to this legendary sleep researcher.
This photograph is taken on the occasion of the 100th birthday of Nathaniel
Kleitman (middle right). He is accompanied by famous REM sleep and dream
researchers: Michel Jouvet (left), William Dement (middle left) and Eugene
Aserinsky (right).
14
A survey of sleep-wake research in
The Netherlands anno 1999
A.M.L. COENEN (EDITOR)
This 1999 survey of sleep-wake research in The Netherlands is edited in cooperation with members of the Dutch Society for Sleep-Wake Research. Institutes and centres engaged in sleep-wake research as well as their researchers
are mentioned in alphabetical order. Academic and non-academic centres are
not distinguished and also fundamental, applied and clinical research are listed
without any distinction. Although this survey is carefully composed, the editor
apologises for eventual errors in the text and for possible omissions.
Amsterdam: Netherlands institute for brain research
Sleep-wake reseach in this institute is concentrated on hypothalamic mechanisms. A main line deals with the role of the suprachiasmatic nucleus in the
regulation of circadian rhythmicity. Furthermore, investigations are performed
towards the early development of circadian rhythms in rats and children. Also
rhythm research in ageing and neurodegenerative disorders (Alzheimer and
Parkinson) are topics of interest. The institute contains expertise in actigraphic
assessment of circadian rhythms and tremor. The institute leads a multicentre
clinical trial on the long-term effects of daily melatonin and exposure to bright
light on the course of development in behavioural, emotional and cognitive
disturbances in demented elderly people. Moreover, the relation between temperature regulation and sleep is a main issue.
Researchers: Dr. R. van Hutten, Drs. R. Riemersma, Dr. E. van Someren, Prof.
D. Swaab,
Amsterdam: University Hospital ‘Vrije Universiteit’
The centre for sleep disorders located at the Department of Clinical Neurophysiology is managed by a multidisciplinary team. It has an outpatient clinic
and extensive facilities for ambulatory and clinical sleep registration. Important interests are the obstructive sleep apnoea syndrome, narcolepsy and epilepsy during sleep.
15
Researchers: Dr. A, Boonstra, Dr. L. Cohen, Prof. dr. E. Jonkman, Dr. E. van
Nieuwkerk, Dr. R. Strijers
Amsterdam: University of Amsterdam
The Department of Psychonomics is involved in sleep-wake research in the
following areas: the influence of chronic insomnia on performance, the influence of disturbing effects on sleep and circadian rhythms, the inter-individual
differences in adaptation to shifted sleep and, furthermore, the vulnerability to
stress of circadian rhythms. Core-variables are sleep-EEG, performance and
24-hour recording of body temperature.
Researchers: Drs. A. van Eekelen, Dr. W. Hofman, Prof. dr. G. Kerkhof, Dr. J.
Snel
Breda: Medical Centre ‘de Klokkenberg’ and Amsterdam: Academic
Medical Centre
In co-operation with the Academic Medical Centre in Amsterdam, sleep research is currently in preparation. A main issue will be sleep disorders in relation to respiratory disorders.
Researchers: Dr. A.R.J. van Keimpema, Dr. R. van Steenwijk
Ede: Sleep Centre of the Hospital ‘de Gelderse Vallei’
This sleep centre, situated in the midst of the Netherlands, has extensive possibilities to study disorders in outpatients as well as in hospitalised patients.
Treatments are recommended that can be supervised by the patients’ own general practitioner. Special interests are circadian rhythm disorders and sleep
disorders in blind people, in autistic patients, in hyperactive children, as well
as in elderly people. Effects of melatonin treatment on circadian rhythm disorders are presently evaluated.
Researchers: Dr. M. Majoor, Drs. E. Nagtegaal, Dr. M. Smits, Dr. A. Vos.
Enschede: Sleep Centre of the ‘Medisch Spectrum Twente’ Hospital
This sleep centre located in the east part of The Netherlands is an example of
symbiosis between the Clinical Department of Respiratory Medicine and the
16
Department of Clinical Neurophysiology. It fulfils a supraregional function in
the diagnosis of sleep disorders and nightly respiratory disturbances.
Researchers: Drs. M. Eysvogel, Mr. B. Hilhorst, Drs. G. Wilts
Groningen: University of Groningen
Since a long time the Department of Biological Psychiatry is engaged in sleep
research. The department investigates the relationship between sleep and depression and the anti-depressive effects of sleep deprivation as well as light
therapy. Together with the Zoological Laboratory and in collaboration with the
Institute of Pharmacology in Zürich, the two-process model of sleep regulation
is under continuous evaluation. A time isolation facility is available in which
experiments are performed under conditions of forced desynchronisation.
Researchers: Dr. D. Beersma, Dr. N. Bouhuys, Drs. M. Gordijn, Drs. E. Hiddinga,
Prof. dr. R. van den Hoofdakker, Drs. K. Koorengevel
The Research Group BCN-Behavioural Biology (Zoological Laboratory) has a
long-standing history of research in the field of causation, function and timing
of behaviour in animals. Theoretical work on the temporal organisation of sleepwake behaviour in humans has recently been given new impetus to the creation of a new temporal isolation facility. This is used to study timing and function of sleep. In addition, animal research focuses on the generation and function
of circadian rhythms and sleep in nocturnal and diurnal as well as subterranean rodents, and on the role of circadian rhythms in growth, learning and
ageing.
Researchers: Dr. D.G.M. Beersma, Ir. P.E. Boon, Prof. dr. S. Daan, Drs. L.G.
Everts, Dr. M.P. Gerkema, Drs. R.A. Hut, Drs. K. Jansen, Dr. P. Meerlo, Mr. M.
Oklejewicz, Drs. A.M. Strijkstra, Drs. K.E. Visser, Dr. E.A. van der Zee
At the Department of Medical Psychology research is performed to non-pharmacological treatments of sleep-disturbances. Psychotherapeutic, behavioural
and conditioning methods are applied and evaluated.
Researchers: Prof. dr. E. Klip, in co-operation with Dr. A. Oosterhuis, affiliated
to the Hans Berger Clinic in Breda.
Heeze: Centre for Sleep-Wake Disorders ‘Kempenhaeghe’
The Centre for Sleep-Wake Disorders ‘Kempenhaeghe’ situated in the south of
The Netherlands near the city of Eindhoven, is a clinical sleep centre for diag-
17
nosis and treatment of disorders of sleep and wakefulness and related complaints. The centre is specialised in narcolepsy, obstructive sleep apnoea syndrome, periodic leg movement syndrome, parasomnia, sleep epilepsy and nonpharmacological treatment of insomnia. The centre is located at, and affiliated
with, the epilepsy centre ‘Kempenhaeghe’.
Researchers: Dr. A.C. Declerck, Dr. M.G. van Erp, Dr. J.H.M. de Groen, Drs.
G.M.L.G. Konings, Dr. K.E. Schreuder, Drs. I. Verbeek
Leiden: State University of Leiden
Sleep-wake research at the Department of Physiology and Experimental Psychology concentrates on the following topics: inter-individual differences in
circadian rhythms and sleep, stress and circadian rhythms, seasonal variation
in circadian rhythms, circadian disorders. The applied methodology includes
laboratory-based (e.g. constant-routine, time-isolation) as well as ambulatory
24-hour-recordings of body temperature, sleep, hormonal and cardiovascular
variables.
Researchers: Drs. H.P.A. van Dongen, Prof. dr. G.A. Kerkhof
The Department of General Practice is currently investigating sleep disorders
and the prevalence of sleep apnoea in general practice. The Dutch standard for
general practitioners is being evaluated.
Researchers: Drs. P.R. Eijkelenboom, Drs. A.W. Graffelman, Dr. A. Knuistingh
Neven, Prof. dr. M.P. Springer
Leiden: University Hospital
The Department of Neurology evaluates fundamental and clinical aspects of
narcolepsy, periodic limb movements and vigilance, using ambulatory monitoring techniques. Actigraphy is a major research tool in this centre.
Researchers: Dr. J. van Dijk, Drs. G.J. Lammers, Dr. H. Middelkoop
Maastricht: Maastricht University
Effects of antidepressants on sleep in depressives, the effects of light therapy
on depression and sleep, and the predictive value of changes in the rest/activity cycle in bipolar mood disorder are main issues at the Academic Mood Disor-
18
ders Clinic and Sleep Laboratory. Furthermore, in association with the Department of Medical Psychology sleep characteristics associated with exhaustion in cardiovascular disorders are also studied.
Researchers: Dr. A. L. van Bemmel, Dr. R. van Diest , Drs. R. Havermans
Nijmegen: University of Nijmegen
At the Department of Psychology the neurophysiology and neuropsychology of
sleep, including REM-sleep, is the central topic. This together with information
processing during the various sleep-wake states, using the concept of ‘sensory
gating’. This is partly done in co-operation with Organon Laboratories in
Newhouse, Scotland. This is mainly approached in rats, using EEG, psychoactive
drugs and deprivation techniques. Also the relationship of EEG sleep waves
with epileptic phenomena is presently studied. The classic benzodiazepine
hypnotics are studied for their sedative and amnestic properties to unravel
their working mechanisms and their effects on cognitive processes. The latter
topic is human research.
Researchers: Prof. dr. A. Coenen, Dr. W. Drinkenburg, Drs. H. van Lier, Dr. E.
van Luijtelaar
The Department of Pulmonary Diseases ‘Dekkerswald’ is specialised in sleep
disordered breathing in patients with pulmonary diseases, such as asthma,
chronic obstructive pulmonary disease, chest wall deformations, respiratory
muscle failure, problems with control of breathing and obstructive sleep apnoea syndrome. Research and patient care are performed on administration of
nocturnal oxygen, on respiratory muscle training, on respiratory stimulants
and on continuous positive air pressure treatment.
Researchers: Drs. F. Brijker, Dr. F. van den Elshout, Prof. dr. H. Folgering, Dr. Y.
Heydra, Drs. M. Wagenaar, Dr. P. Vos
Oss: Pharmaceutical Company ‘Organon’
Effects of newly developed drugs on the electroencephalogram of waking and
sleeping is assessed in healthy volunteers and patients suffering from depressive disorders as well as sleep disorders.
Researchers: J. van Proosdij, Dr. G. Ruigt
19
Rotterdam: Erasmus University and University Hospital ‘Dijkzigt’
The Department of Psychiatry studies the psychological, clinical and cardiovascular effects of psychoactive drugs in relation to 24-hour behavioural patterns as assessed by means of accelerometry (postural changes, physical as
well as locomotor activity), in healthy subjects, psychiatric patients and patients with cardiovascular dysfunction’s.
Researcher: Dr. J.H.M. Tulen, Drs. A.C. Volkers
The Hague: Centre for Sleep and Wake Disorders of the ‘Westeinde’
Hospital
The Centre for Sleep and Wake Disorders in The Westeinde Hospital is a clinical oriented sleep centre, The centre is specialised in the diagnosis and treatment of sleep disturbances of all kinds. Emphasis is laid on the treatment of
sleep apnoeas, periodic limb movements as well on the non-pharmacological
treatment of insomnia. Research topics are sleep disturbances in metabolic
disease, validation of existing sleep questionnaires and the development of new
ones, non-pharmacological treatment of insomnia. development of sleep algorithms and neuropathy and sleep disturbances. More details can be found on
their home-page: http://www.ziekenhuis.nl/domeinen/westeinde/slc/research/
index.htm.
Researchers: Dr. ir. B. Kemp, Prof. Dr. G.A. Kerkhof, Drs. C. Kluft, Drs. R.M.
Rijsman, Dr. R.J. Schimsheimer, M. Verhelst, Dr. A.W de Weerd
The Hague: ‘Parnassia’ Psycho-Medical Centre
The aim of this centre is to evaluate the subjective and objective aspects of
sleep, mental status and circadian rhythms in the research program
‘chronobiology and psychiatry’. The links between chronobiological disturbances
and psychiatric disorders are examined in a clinical setting. Emphasis is laid
on the treatment with melatonin and bright light therapy in depression, sleep
disturbances, chronic fatigue syndrome, dementia and post partum depression. Moreover the role of melatonin in the human reproductive system is studied.
Researchers: Dr. P.M.J. Haffmans, Drs. S.A.P. Lucius
20
Proceedings of the first joint scientific meeting
of the Belgian and Dutch Sleep Societies,
in honor of the retirement
of Dr. A.C. Declerck.
Center for Sleep/Wake Disorders
Kempenhaeghe, Heeze
November 21, 1998
Organizing committee:
D.G.M. Beersma, chair scientific committee NSWO
R.Cluydts, vice-president BASS
J.H.M. de Groen, board member NSWO,
local organizer
Dr. A.C. Declerck
Presentations:
J.H.M. de Groen: CSW: past, present, future
A. Kalsbeek: SCN-morphology, physiology and pathology
D.G.M. Beersma: Body temperature and sleep-wake rhythms torn apart: implications for
the regulation of mood
O. van Reeth: Effects of prenatal stress on circadian rhythms and sleep
D. Dive: EEG-topography during slow-wave sleep
G.A. Kerkhof: Interindividual differences in circadian rhythms
A.M.L Coenen: Neurophysiology of sleep-wake states in relation to information processing
M. Billiard: Invited lecture: Pathophysiology of narcolepsy
D. Pevernagie: Laryngeal dysfunction during sleep
A. Boudewijns: Surgical treatment in SRBD: possibilities and limitations
21
22
CSW: past, present, future
J.H.M. DE GROEN,
CSW KEMPENHAEGHE HEEZE
It is a great honour for the Center for Sleep/Wake Disorders of Kempenhaeghe
to be host of the first joint scientific meeting of the Dutch and Belgian Sleep
Societies NSWO and BASS, in honour of the retirement of Dr. Declerck, founder
and first head of the CSW.
The history of the CSW of Kempenhaeghe goes back to as early as the first of
May 1975, when Dr. Declerck became head of the department of Clinical Neurophysiology of the Epilepsy Center Kempenhaeghe and started his work on
clinical-neurophysiologic aspects of sleep and epilepsy. He introduced new techniques (ambulatory monitoring, EEG-video registration), and methods (sleepdeprivation EEG), and published on several important issues on this field.
Because of his interest in sleep-wake disorders, Dr. Declerck concentrated more
and more on sleep research, both scientific and clinical. To the department of
Clinical Neurophysiology of Kempenhaeghe, traditionally directed into EEGdiagnostics in epilepsy, he added a qualitatively outstanding facility for
polysomnography in sleep disorders.
In the late 1980s he started his so-called Sleep-Wake project, in order to explore the feasibility of a separate department for sleep-wake disorders within
Kempenhaeghe, with accents both on patient care and sleep research. Treatment facilities were enlarged with CPAP-therapy (OSAS) and non-pharmacological treatment of insomnia. Moreover, Dr. Declerck paid much attention to
advising and training physicians in the first and second line.
In 1995 Dr. Declerck turned over his responsibilities as head of the department
of Clinical Neurophysiology and since then he has dedicated all of his time to
the Sleep-Wake project. The Sleep-Wake project became a success and led to
the formal start of the CSW on the first of January 1997.
In October 1997 Dr. Declerck turned over his responsibilities as head of the
CSW, to become advisor of our center. His advisorship will be continued now
that he is retired.
Nowadays the CSW Kempenhaeghe is expertise center in the Netherlands for
diagnosis and treatment of patients with severe or complex sleep-wake disorders, both somatic and non-somatic, covering the whole field of sleep medi-
23
cine, including OSAS, PLMD, RLS, the various subtypes of insomnia and of
parasomnia, sleep-related epilepsy and circadian rhythm disorders. The CSW
has a multidisciplinary medical staff, consisting of two neurologists, a specialist for social diseases, a psychiatrist, a medical biologist, and a psychologist.
They are working in firm collaboration with the neurophysiologists and technicians of the department of Clinical Neurophysiology.
Patients are referred by all medical echelons, mostly by medical specialists of
general and academic hospitals, but also by family doctors or by psychiatric
institutions or nursing homes.
Current research topics are alertness and narcolepsy, modafinil-treatment of
narcolepsy, HLA-typing in narcolepsy, non-pharmacological treatment of insomnia, upper airway impedance in OSAS, pergolide-treatment of RLS/PLMD,
and sleep disturbances in PTSD.
All these developments had not been possible without the work of Dr. Declerck,
his expertise, his intense involvement with the patients, the way he stimulated
his co-workers, his loyalty to Kempenhaeghe and “his” CSW, the enormous drive
and the enthusiasm with which he promoted sleep research, his network of contacts, nationally and internationally, and last but not least his pleasant and charming manner, such as has been illustrated by the short presentations of Dr Arends,
head of the dep. of Clinical Neurophysiology, Drs. I. Verbeek and Drs. K.S.
Schreuder, his first co-workers and staff members of the former Sleep-Wake
project, and by the words of Ir. N. Bomer, General Director of Kempenhaeghe
and Prof. Dr. P. Boon, Director Research, at the end of the meeting.
We are happy that Dr. Declerck -although after his retirement from now on in a
more indirect way- will continue to be interested in the CSW and its further
developments in the future, which undoubtedly will proceed along the lines set
out by him so successfully in the past.
The presentations of the meeting of today will give an overview of various aspects of sleep research such as is going on in Belgium and the Netherlands.
Moreover, we are greatly honoured with the fact that Prof. M. Billiard, president of the ESRS, accepted the invitation for joining us today and for giving a
lecture on narcolepsy.
We thank the speakers for their presentations, and for their willingness of contributing to the proceedings of the meeting, as an abstract or as a full paper.
24
Circadian control of hormonal rhythms:
important roles for suprachiasmatic nucleus
efferents and the autonomic nervous system
ANDRIES KALSBEEK, INGE F. PALM, SUSANNE E. LA FLEUR AND RUUD M. BUIJS.
NETHERLANDS INSTITUTE FOR BRAIN RESEARCH
MEIBERGDREEF 33, 1105 AZ AMSTERDAM
THE NETHERLANDS.
1. Introduction
Hormonal rhythms express themselves in a variety of forms. The two major
processes controlling the shape of a hormonal rhythm are the circadian signal
generated by the central pacemaker and the behaviour of an animal. The central neural pacemaker in humans and other mammals is localized in the
suprachiasmatic nuclei of the hypothalamus (SCN). Little is known, however,
about the neural mechanisms responsible for the implementation of rhythmic
information generated by the SCN into behavioural and physiological rhythms.
The endogenous pacemaker capacity of the SCN is evidenced by the rhythmic
metabolic and electrical activity of its neurons 1,9, but also in the release of its
transmitters. Best studied in this regard is the secretion of vasopressin (VP),
showing a peak release during (subjective) daytime both in vivo and in vitro
9,16
. But rhythmic fluctuations in the SCN tissue content of peptide and mRNA
have also been reported for somatostatin, VIP, GRP and GABA 12. Therefore, we
put forward the following hypothesis:“The rhythmic release of SCN transmitters within its target areas is responsible for the expression of circadian
rhythms in locomotor activity, body temperature, autonomic functions, and
hormone secretion”. To test the above hypothesis additional information is
needed about the brain structures towards which SCN output is directed,
i.e. what are the target areas of the SCN efferents. Since the discovery of the
SCN as “master clock” several attempts have been made to elucidate it neuronal projections (eg. 2,11,22,34). Taken together all animal studies indicate that the
outflow of SCN information is mainly directed to medial hypothalamic target
areas such as preoptic, paraventricular (PVN), dorsomedial (DMH) and ventromedial nuclei, with the exception of the paraventricular thalamus and the
intergenicular leaflet.
25
2.1 Experimental strategy
The main experimental strategy employed to test the above hypothesis included
the intracerebral administration of SCN transmitters or their antagonists in
SCN target areas of SCN-lesioned and intact animals, and the concomitant
measurement of its consequences for the release pattern of the hormonal rhythm
under study. We first concentrated on the SCN projection to the neuroendocrine center of the hypothalamus, i.e. PVN/DMH area, and the possible implication of this projection for the circadian control of the corticosterone rhythm.
In order to test the general applicability of the hypothesis, additional experiments were extended to other hormonal rhythms and other SCN target areas.
In the present chapter experiments will be described investigating the circadian control of corticosterone, melatonin, luteinizing hormone and insulin release.
2.2 Corticosterone
Circadian fluctuations in circulating glucocorticoid levels have been reported
for many species. In nocturnal animals such as the laboratory rat, plasma glucocorticoid levels are high at the onset of darkness and then decline, reaching
a nadir in the morning. Our initial experiments using microinfusions in the
PVN/DMH area showed a strong inhibitory effect of exogenous VP, but not VIP,
on the release of corticosterone 17. These results suggested an inhibitory effect
of VP released by SCN terminals in the PVN/DMH area on the activity of the
hypothalamo-pituitary-adrenal (HPA)-axis. The initial data were further substantiated by microdialysis mediated administration of VP and its V1-antagonist. Stress-free infusion of the VP-antagonist in the DMH of freely moving,
undisturbed animals during the middle of the light period (i.e. the trough of the
corticosterone rhythm), caused an immediate dose-dependent increase of circulating plasma corticosterone. On the other hand, similar infusions of VP at
the end of the light period completely prevented the diurnal rise in plasma
corticosterone 23. In the above studies only single time points were investigated. Therefore, to further specify the nature of the SCN control, intracerebral infusions of VP-antagonist were performed at different times of the L/Dcycle. One hour infusions of the VP-antagonist were performed using in vivo
microdialysis at CT2, 6, 10, 14 and 21, revealing the existence of a stimulatory
SCN input to the HPA-axis as well, next to the inhibitory control of the VPergic
SCN projection 24. The chemical identity of this stimulatory SCN transmitter is
not know as yet, presently we only have preliminary data for a stimulatory
action of VIP in the PVN/DMH area 14.
Comparing congruent corticosterone and ACTH responses showed that whereas
pronounced changes in circulating levels of corticosterone were induced, ACTH
levels only showed minor changes. In addition, anatomical studies showed that
26
indeed the number of SCN fibers making synpatic contacts with CRH containing neurons in the PVN was very small 4,35. These results indicated that the
SCN does not control the daily rhythm of corticosterone release solely via the
HPA-axis (i.e. via the subsequent release of CRH and ACTH). Comparing the
ACTH - corticosterone relation in different experimental conditions 7,24 revealed
to us an alternative pathway for SCN control over corticosterone release. As
indicated in Figure1 the circadian release of corticosterone does not depend
heavily on the release of ACTH, contrary to the stress-related corticosterone
release. Instead of stimulating ACTH release the SCN seems to change adrenal
sensitivity for ACTH, probably via the sympathetic innervation of the adrenal
gland. In order to investigate if indeed there is a neural connection between
the SCN and the adrenal gland we employed the viral transneuronal tracing
technique.
Figure 1:
Peak plasma corticosterone values during Ringer’s (ɀ, solid line) or VPantagonist (័, stippled line) administration at different times of the L/Dcycle, as a function of plasma ACTH. Contrary to novelty-stress (ᔤ, dashed
line), circadian corticosterone excursions show minimal ACTH increments.
This technique is based on the demonstrated ability of the virus to invade axon
terminals, replicate in neurons and pass retrogradely through a multisynpatic
circuit. After injection in the adrenal gland immunohistochemical localization
of the viral antigen revealed the progressive appearance of infected neurons in
the spinal cord, parvicellular subdivisions of the PVN and subsequently the
SCN 6. The neurons in the spinal cord were shown to receive an input from
27
oxytocin fibers 30. Some of the virus-containing neurons in the PVN also exhibited oxytocin immunoreactivity. These latter neurons were shown to receive
an input from VP- or VIP-containing SCN fibers 31. Furthermore, it was shown
that denervation of adrenal gland by removal of its sympathetic innervation
resulted in increased secretory activity of the adrenal cortex during the daytime 13, thus supporting the existence of an important daytime inhibition via
the neural input to the adrenal gland. Together these results indicated that
SCN inputs to PVN neurons with descending projections to preganglionic neurons of the sympathetic branch of the autonomic nervous system might be an
important effector pathway for the SCN to impose its control on the corticosterone rhythm. Interestingly, SCN control on the circadian melatonin rhythm
has already been known for a long time to involve the sympathetic innervation
to the pineal gland.
2.3 Lutinizing Hormone
In female rats the pre-ovulatory surge of luteinizing hormone (LH) occurs at
regular 4 to 5 day intervals, and induces ovulation on the day of estrus. Previous studies had indicated two brain structures of crucial importance for the
occurrence of LH surges. First, the medial preoptic area (MPO) containing the
estrogen-receptor containing neurons necessary for the positive estrogen feedback. Secondly, the SCN providing the timing of the LH surge on the day of
proestrus. Both MPO- and SCN-lesioned animals are completely incapable of
showing LH surges 33. Anatomical studies have shown that estrogen-receptor
containing neurons in the MPO receive direct synaptic inputs from SCN fibers,
probably containing VP as a neurotransmitter 11,22,36. We hypothesized that VP
release in the MPO would act as the daily timing signal of the SCN necessary
for the induction of an LH surge. Indeed, a 5-h perfusion of the MPO with VP
elicited a surge-like LH pattern in SCN-lesioned animals, whereas constant,
basal levels were found in SCN-lesioned animals with control infusions 27.
These results show that with regard to the LH surge VP may act as a stimulatory timing signal from the SCN, contrary to its inhibitory effect on the corticosterone rhythm. More importantly, however, these experiments indicate that
VP can act as an output signal of the biological clock in different SCN target
areas and completely different hormonal rhythms.
2.4 Melatonin
Previous experiments with SCN-lesioned animals yielded equivocal results with
respect to the abolishment of the circadian release of the pineal hormone melatonin. In our own experiments, using the transpineal microdialysis technique to
measure melatonin release 15, 30% of the behaviourally arhythmic animals still
showed nocturnal elevations of pineal melatonin levels. Effectively SCN-lesioned
28
animals, however, showed fairly constant extracellular melatonin levels, which
were comparable to night time levels in control animals (Figure 2). These results
indicated that SCN-lesions removed not only the circadian input to the pineal
gland but also an important inhibitory input. The transneuronal viral tracing
technique showed that also in the case of the melatonin rhythm SCN projections
to autonomic PVN neurons were an important link 25,32.
Figure 2:
Melatonin release patterns during the first part of the dark period in SCNintact (ɂ) and SCN- lesioned (ɀ) animals. The shaded area shows the
complete nocturnal melatonin profile of a previous group of control animals. Only animals were included which showed peak levels of melatonin significantly above those of control animals during the light period
(> 400 pg/ml.).
Our initial studies investigated the effect of both VP and VIP inputs to the
dorsal hypothalamus on melatonin release. However, neither SCN peptide
showed a clear inhibitory effect 20. One of the most abundant inhibitory transmitters in the hypothalamus is GABA, eg. in some hypothalamic nuclei about
half of the total synpatic input may be accounted for by GABA-containing terminals. At that same time both anatomical and electrophysiological data indicated the presence and release of GABA from SCN terminals 3,5,10. Accordingly,
experiments were initiated to determine whether the GABAergic projection
from the SCN might be involved in depressing the activity of neurons in the
dorsal hypothalamus that are associated with the control of the circadian me-
29
latonin rhythm. Infusion of the GABA-agonist muscimol in the dorsal
hypothalamus completely blocked the daily increase of plasma melatonin during darkness and experiments in SCN-lesioned animals showed that also the
elevated plasma melatonin levels in these animals were decreased by the
muscimol administration 19. On the other hand, infusion of the GABA-antagonist bicuculline had no effect on the nocturnal melatonin increase. Muscimol
infusions elicited a decrease of extracellular noradrenaline levels in the pineal
concomittant with decreased melatonin release showing clearly that the inhibitory influence of the SCN is mediated via the sympathetic input to the
pineal gland 19. In conjunction with the strong resemblance between the acute
inhibitory effect of muscimol and light on melatonin release, these data suggested to us that the GABA-containing projections from the SCN might serve to
tr?ÿsmit the inhibitory effect of light onto the melatonin-rhythm-generatingsystem in the dorsal hypothalamus. Therefore, we set out to test the hypothesis that the light-induced inhibition of melatonin is mediated by the endogenous release of GABA within the dorsal hypothalamus. Animals were confronted with a 1-min nocturnal light exposure in the middle of a bilateral 2-h
perfusion of the dorsal hypothalamus with either Ringer or Ringer + Bicuculline.
It is clear from Figure 3 that bicuculline completely prevented the inhibitory
effect of light on melatonin release. This experiment thus documents that
retina-mediated photic activation of SCN neurons induces the release of GABA
from efferent SCN nerve terminals, resulting in an inhibition of melatonin release by the pineal gland 18. At present it is not clear yet if GABA release from
SCN terminals is also involved in the circadian control of the daily melatonin
rhythm. In this aspect the reported co-localization of GABA and vasopressin in
SCN neurons and their efferent projections is intriguing5, since the circadian
rhythm in synthesis and release of vasopressin, with peak activity found during (subjective) day time, is well-known16.
2.5 Insulin
The daily rhythm of insulin release seems to be mainly driven by a rise in
blood glucose levels as a consequence of a circadian pattern in feeding behaviour, and a direct circadian modulation of insulin release is not evident. After
SCN lesions glucose and insulin responses are similar during day and night, but
also the distribution of the meals across the light/dark cycle is equal now 28.
Therefore, it might still be so that the distribution of feeding activity is the
primary determinant of daily variations in insulin and glucose responses, and
not the circadian oscillator located in the SCN. Lesion experiments did indicate, however, that the SCN exerts an important inhibitory control on insulin
release 28, as previously found for corticosterone and melatonin. To further
investigate SCN control of insulin release we designed an experimental setup
30
that would enable us to unmask a possible direct control of the endogenous
pacemaker on the circadian aspects of insulin release. Rats were subjected to a
Figure 3:
Inhibitory effect of a 1-min light exposure at ZT17 on melatonin levels
(n=6). During light exposure (stippled line) hypothalamic microdialysis
probes aimed at the PVN/DMH area were perfused with either Ringer alone
(ɂ) or Ringer + Bicuculline (ɀ). The gray box indicates the 2-h
hypothalamic perfusion period.
feeding regimen of 6 meals a day, spaced 4 hours apart. A significant daily
variation in feeding responses was present despite the fact that feeding conditions, such as prior fasting period, amount of food intake and intake rate, were
very much similar now for all six meals 21. It is evident that the most prominent
(inhibitory) effects of the circadian timing system on feeding-induced insulin
release are present at the end of the light period (Figure 4). With regard to the
basal insulin levels, however, a daily rhythm was not evident. On the other
hand basal glucose levels did show a significant circadian variation with highest glucose levels being reached at the end of the light period 21, i.e. shortly
before the onset of the dark period and the main period of (feeding) activity of
the rat. Together these data indicate that the SCN controls glucose homeostasis
at two levels. First basal glucose levels are controlled to ensure sufficient glucose supply to active tissues at times when the animal awakens but is not feeding yet. Secondly, feeding-induced, but not basal, insulin release is stimulated
(or disinhibited) only at the animals usual feeding time to prevent hypoglycemia
due to increased insulin release before the animal has been able to find food.
At present we have no indications yet which SCN-projections and/or -transmit-
31
ters might be involved in the control of basal glucose and feeding-induced insulin responses. As described in the foregoing we think SCN control over sympathetic inputs to the adrenal is an important mechanism to control circadian
corticosterone release. A similar mechanism may account for the control on
glucose homeostasis. Increased sympathetic input to the liver and fat tissue
enhances glucose output and stimulates lipolysis, thus promoting hyperglycemia.
Electrical stimulation of the SCN too results in hyperglycemia 8,26. Therefore,
increased SCN activity at the end of the light period may stimulate sympathetic activity in different tissues resulting in both a corticosterone and a glucose peak. At the same time, as becomes clear from Figure4, the stimulatory
effect of the SCN on corticosterone secretion and plasma glucose levels coincides exactly with its inhibitory effect on the feeding-induced insulin responses.
Our previous experiments have shown that the pronounced daily rhythmicity
Figure 4:
Maximal feeding-induced insulin increments (bars) during 6 meals equally
dispersed over the 24- hour L/D-cycle and maximal corticosterone increases
(open symbols) due to a 1-hour blockade of the VPergic SCN output to the
DMH (by hypothalamic administration of VP-antagonist) at five different
times along the L/D-cycle.
in insulin responses is most likely effectuated via an inhibitory action of the
SCN on the parasympathetic input to the pancreas 21,28. Thus, congruent with
a stimulatory effect of SCN output on sympathetic activity, parasympathetic
activity is inhibited. Indeed, a considerable proportion of PVN neurons with
32
descending projections send divergent axon collaterals to both the spinal cord
and the dorsal vagal complex in the brainstem 29. Therefore, an important aspect of SCN control on daily rhythms may involve its inputs to PVN neurons
with descending projections and its effect on the sympathovagal balance of the
autonomic nervous system 30,31.
3. Summary
The results obtained so far show that the rhythmic release of VP and GABA from
SCN terminals are important signals for the control of hormonal rhythms such
as corticosterone, melatonin and LH. The ultimate circadian waveform of a hormonal rhythm seems to be determined by an alternation of inhibitory and stimulatory signals from the SCN. For a number of rhythms evidence has been obtained that an important part of the circadian message from the SCN is transferred
by modulating the sympathovagal balance of the autonomic input to hormonal
glands or other peripheral organs. The presently proposed mechanism for the
implementation of circadian rhythms may hold not only for hormonal rhythms,
but also for SCN control on behavioral (i.e. sleep-wake) and physiological rhythms,
such as body temperature and cardiovascular regulation.
Acknowledgements
We gratefully acknowledge support from IRIS (grant No. PHA-614-NLD) and
the superb technical assistence from J. Van der Vliet, J.J. van Heerikhuize
and J. Wortel.
References
1 Bos N.P.A. and Mirmiran M. (1990) Circadian rhythms in spontaneous neuronal discharges of the cultured suprachiasmatic nucleus. Brain Res. 511, 158-162.
2 Buijs, R.M. The anatomical basis for the expression of circadian rhythms: the efferent projections of the suprachiasmatic nucleus. In: Progress in Brain Research,
Vol.111, Hypothalamic Integration of Circadian Rhythms, edited by Buijs, R.M.,
Kalsbeek, A., Romijn, H.J., Pennartz, C.M.A. and Mirmiran, M. Amsterdam: Elsevier
Science BV., 1996, p. 229-240.
3 Buijs R.M., Hou Y.X., Shinn S. and Renaud L.P. (1994) Ultrastructural evidence for
intra- and extranuclear projections of GABAergic neurons of the suprachiasmatic
nucleus. J. Comp. Neurol. 340, 381-391.
4 Buijs R.M., Markman M., Nunes-Cardoso B., Hou Y.-X. and Shinn S. (1993) Projections of the suprachiasmatic nucleus to stress-related areas in the rat hypothalamus:
A light and electron microscopic study. J. Comp. Neurol. 335, 42-54.
5 Buijs R.M., Wortel J. and Hou Y.X. (1995) Colocalization of gamma-aminobutyric
acid with vasopressin, vasoactive intestinal peptide, and somatostatin in the rat
33
suprachiasmatic nucleus. J. Comp. Neurol. 358, 343-352.
6 Buijs R.M., Wortel J., Van Heerikhuize J.J., Feenstra M.G.P., Ter Horst G.J., Romijn
H.J. and Kalsbeek A. (1999) Anatomical and functional demonstration of a
multisynaptic suprachiasmatic nucleus adrenal (cortex) pathway. Eur. J. Neurosci.
7 Buijs R.M., Wortel J., VanHeerikhuize J.J. and Kalsbeek A. (1997) Novel environment induced inhibition of corticosterone secretion: Physiological evidence for a
suprachiasmatic nucleus mediated neuronal hypothalamo-adrenal cortex pathway.
Brain Res. 758, 229-236.
8 Fujii T., Inoue S., Nagai K. and Nakagawa H. (1989) Involvement of adrenergic mechanism in hyperglycemia due to SCN stimulation. Horm. Metab. Res. 21, 643-645.
9 Gillette M.U. and Reppert S.M. (1987) The hypothalamic suprachiasmatic nuclei:
circadian patterns of vasopressin secretion and neuronal activity in vitro. Brain Res.
Bull. 19, 135-139.
10 Hermes M.L.H.J., Coderre E.M., Buijs R.M. and Renaud L.P. (1996) GABA and glutamate mediate rapid neurotransmission from suprachiasmatic nucleus to hypothalamic
paraventricular nucleus in rat. J. Physiol-London. 496, 749-757.
11 Hoorneman E.M.D. and Buijs R.M. (1982) Vasopressin fiber pathways in the rat brain
following suprachiasmatic nucleus lesioning. Brain Res. 243, 235-241.
12 Inouye, S.I.T. Circadian rhythms of neuropeptides in the suprachiasmatic nucleus.
In: Progress in Brain Research, Vol.111, Hypothalamic Integration of Circadian
Rhythms, edited by Buijs, R.M., Kalsbeek, A., Romijn, H.J., Pennartz, C.M.A. and
Mirmiran, M. Amsterdam: Elsevier Science BV., 1996, p. 75-90.
13 Jasper M.S. and Engeland W.C. (1994) Splanchnic neural activity modulates ultradian
and circadian rhythms in adrenocortical secretion in awake rats. Neuroendocrinology 59, 97-109.
14 Kalsbeek, A. and Buijs, R.M. Peptidergic transmitters of the suprachiasmatic nuclei
and the control of circadian rhythmicity. In: Progress in Brain Research, Vol.92, The
Peptidergic Neuron, edited by Joosse, J., Buijs, R.M. and Tilders, F.J.H. Amsterdam:
Elsevier Science BV, 1992, p. 321-333.
15 Kalsbeek, A. and Buijs, R.M. Rhythms of inhibitory and excitatory output from the
circadian timing system as revealed by in vivo microdialysis. In: Progress in Brain
Research, Vol.111, Hypothalamic Integration of Circadian Rhythms, edited by Buijs,
R.M., Kalsbeek, A., Romijn, H.J., Pennartz, C.M.A. and Mirmiran, M. Amsterdam:
Elsevier Science BV, 1996, p. 271-291.
16 Kalsbeek A., Buijs R.M., Engelmann M., Wotjak C.T. and Landgraf R. (1995) In vivo
measurement of a diurnal variation in vasopressin release in the rat suprachiasmatic
nucleus. Brain Res. 682, 75-82.
17Kalsbeek A., Buijs R.M., Van Heerikhuize J.J., Arts M. and Van Der Woude
T.P. (1992) Vasopressin-containing neurons of the suprachiasmatic nuclei
inhibit corticosterone release. Brain Res. 580, 62-67.
18 Kalsbeek A., Cutrera R.A., Van Heerikhuize J.J., Van Der Vliet J. and Buijs R.M.
34
(1998) GABA release from SCN terminals is necessary for the light-induced inhibition of nocturnal melatonin release in the rat. Neuroscience
19 Kalsbeek A., Drijfhout W.J., Westerink B.H.C., Van Heerikhuize J.J., Van Der Woude
T., Van Der Vliet J. and Buijs R.M. (1996) GABA receptors in the region of the
dorsomedial hypothalamus of rats are implicated in the control of melatonin. Neuroendocrinology 63, 69-78.
20 Kalsbeek A., Rikkers M., Vivien-Roels B. and Pévet P. (1993) Vasopressin and Vasoactive Intestinal Peptide Infused in the Paraventricular Nucleus of the Hypothalamus
Elevate Plasma Melatonin Levels. J. Pineal Res. 15, 46-52.
21 Kalsbeek A. and Strubbe J.H. (1998) Circadian control of insulin secretion is independent of the temporal distribution of feeding. Physiol. Behav. 63, 553-560.
22 Kalsbeek A., Teclemariam-Mesbah R. and Pévet P. (1993) Efferent projections of the
suprachiasmatic nucleus in the golden hamster (Mesocricetus auratus). J. Comp.
Neurol. 332, 293-314.
23 Kalsbeek A., Van Der Vliet J. and Buijs R.M. (1996) Decrease of endogenous vasopressin release necessary for expression of the circadian rise in plasma corticosterone: a reverse microdialysis study. J. Neuroendocrinol. 8, 299-307.
24 Kalsbeek A., Van Heerikhuize J.J., Wortel J. and Buijs R.M. (1996) A diurnal rhythm
of stimulatory input to the hypothalamo-pituitary-adrenal system as revealed by timed
intrahypothalamic administration of the vasopressin V1 antagonist. J. Neurosci. 16,
5555-5565.
25 Larsen P.J., Enquist L.W. and Card J.P. (1998) Characterization of the multisynaptic
neuronal control of the rat pineal gland using viral transneuronal tracing. Eur. J.
Neurosci. 10, 128-145.
26 Nagai K., Fujii T., Inoue S., Takamura Y. and Nakagawa H. (1988) Electrical stimulation of the suprachiasmatic nucleus of the hypothalamus causes hyperglycemia. Horm.
Metab. Res. 20, 37-39.
27 Palm I.F., Van Der Beek E.M., Wiegant V.M., Buijs R.M. and Kalsbeek A. (1999) Vasopressin induces an LH surge in ovariectomized, estradiol-treated rats with lesion of
the suprachiasmatic nucleus. Neuroscience
28 Strubbe J.H., Alingh Prins A.J., Bruggink J. and Steffens A.B. (1987) Daily variation
of food-induced changes in blood glucose and insulin in the rat and the control by
the suprachiasmatic nucleus and the vagus nerve. J. Auton. Nerv. Syst. 20, 113-119.
29 Swanson L.W. and Kuypers H.G.J.M. (1980) The paraventricular nucleus of the
hypothalamus: Cytoarchitectonic subdivisions and organization of projections to the
pituitary, dorsal vagal complex, and spinal cord as demonstrated by retrograde fluorescence double-labeling methods. J. Comp. Neurol. 194, 555-570.
30 Teclemariam-Mesbah R., Kalsbeek A., Buijs R.M. and Pévet P. (1997) Oxytocin innervation of spinal preganglionic neurons projecting to the superior cervical ganglion in
the rat. Cell Tiss. Res. 287, 481-486.
31 Teclemariam-Mesbah R., Kalsbeek A., Pévet P. and Buijs R.M. (1997) Direct vasoac-
35
tive intestinal polypeptide-containing projection from the suprachiasmatic nucleus
to spinal projecting hypothalamic paraventricular neurons. Brain Res. 748, 71-76.
32 Teclemariam-Mesbah R., Ter Horst G.J., Postema F., Wortel J. and Buijs R.M. (1999)
Anatomical demonstration of the suprachiasmatic nucleus - pineal pathway. J. Comp.
Neurol.
33 Van Der Beek, E.M. Circadian control of reproduction in the female rat. In: Progress
in Brain Research, Vol.111, Hypothalamic Integration of Circadian Rhythms, edited by Buijs, R.M., Kalsbeek, A., Romijn, H.J., Pennartz, C.M.A. and Mirmiran, M.
Amsterdam: Elesevier Science BV, 1996, p. 295-320.
34 Van Der Beek E.M., Horvath T.L., Wiegant V.M., Van Den Hurk R. and Buijs R.M.
(1997) Evidence for a direct neuronal pathway from the suprachiasmatic nucleus to
the gonadotropin-releasing hormone system: Combined tracing and light and electron microscopic immunocytochemical studies. J. Comp. Neurol. 384, 569-579.
35 Vrang N., Larsen P.J. and Mikkelsen J.D. (1995) Direct projection from the
suprachiasmatic nucleus to hypophysiotrophic corticotropin-releasing factor immunoreactive cells in the paraventricular nucleus of the hypothalamus demonstrated by
means of Phaseolus vulgaris-leucoagglutinin tract tracing. Brain Res. 684, 61-69.
36 Watson R.E., Langub M.C., Engle M.G. and Maley B.E. (1995) Estrogen-receptive
neurons in the anteroventral periventricular nucleus are synaptic targets of the
suprachiasmatic nucleus and peri-suprachiasmatic region. Brain Res. 689, 254-264.
36
Body temperature and
sleep wake rhythms torn apart:
implications for the regulation of mood
DOMIEN G.M. BEERSMA
DEPARTMENT OF BIOLOGICAL PSYCHIATRY AND ZOOLOGICAL LABORATORY, UNIVERSITY OF GRONINGEN, GRONINGEN, THE NETHERLANDS
The prior history of sleep and wakefulness and the current phase position of
the circadian pacemaker (biological clock) both have a major impact on all
kinds of physiological and psychological processes, including the course of body
temperature and many aspects of performance and well being. Under normal
circumstances all sleep episodes occur within a restricted range of circadian
phases. Therefore, it is difficult to discriminate between those two major influences. Experimentally, the range of circadian phases at which sleep occurs can
be increased, for instance by scheduling sleep at 20 hour intervals. At low light
levels it turns out that the circadian pacemaker will no longer run in synchrony
with the 20 hour sleep-wake cycle. Instead it will free run with a period slightly
longer than 24 hours. Under such conditions of forced desynchrony sleep systematically scans through all circadian phases. Forced desynchrony experiments revealed that mood in healthy subjects varies systematically with circadian phase and with the time since wake-up. The circadian component shows
maximum feelings of depression around the minimum in the endogenous rhythm
in core body temperature. The component related to the timing of sleep shows
a progressive increase in feelings of depression in the course of prolongued
wakefulness.
Inspired by these findings we embarked to investigate the circadian regulation
of mood in seasonal affective disorder (SAD) patients, since it has been suggested that the mood disturbances in SAD are due to circadian abnormalities.
The results of the very first forced desynchrony studies in an SAD patient will
be presented.
37
38
Hormonal and behavioral effects
of prenatal stress:
focus on circadian rhythms and sleep
VAN REETH, O.1, DUGOVIC, C.2, KOEHL, M.3,
WEIBEL, L. AND MACCARI, S.4
1
CENTER FOR THE STUDY OF BIOLOGICAL RHYTHMS,
SCHOOL OF MEDICINE, ERASME HOSPITAL,
UNIVERSITÉ LIBRE DE BRUXELLES,
BRUSSELS, BELGIUM
2
3
4
DEPARTMENT OF NEUROPSYCHOPHARMACOLOGY,
JANSSEN RESEARCH FOUNDATION,
BEERSE, BELGIUM
PSYCHOBIOLOGIE DES COMPORTEMENTS ADAPTATIFS,
INSERM U259,
UNIVERSITÉ VICTOR SEGALEN,
BORDEAUX, FRANCE
LABORATOIRE DES NEUROSCIENCES DU COMPORTEMENT
UNIVERSITÉ DE LILLE
VILLENEUVE D’ASCQ
FRANCE
Key words:
prenatal stress, restraint, corticosterone, corticosteroid receptors, circadian
rhythms, sleep, paradoxical, depression.
Address for correspondance:
Dr. Olivier Van Reeth
Center for the Study of Biological Rhythms
School of Medicine-Université Libre de Bruxelles
Hópital Erasme - Route de Lennik, 808 - 1070 Brussels, Belgium
T 32-2-5556427
F 32-2-5553569
E-mail: [email protected]
39
Abstract
In humans, prenatal stress can induce psychological difficulties, mental retardation and sleep disturbances in the infants. In animals, dams stressed during
pregnancy can bear offspring with reduced male sexual activity, enhanced
emotional reactivity, modifications of glucocorticoid secretion and increased
propensity to selfadminister drugs. In adult rats submitted to prenatal stress,
we studied the stress-induced corticosterone secretion response and hippocampal corticosterone receptors, their hormonal and behavioral circadian
rhythms, and their sleep-wake parameters. We found that prenatal stress prolongs stress-induced corticosterone secretion and decreases hippocampal corticosterone receptors in adult offspring. Under entrainment to a regular lightdark cycle, prenatally-stressed rats of both genders show significant phase
advances in the circadian rhythms of locomotor activity (running-wheel
behavior) and plasm corticosterone secretion. Compared to controls, PNS rats
exhibit higher amounts of paradoxical sleep during both the light and the dark
phase (increased number of REM episodes). Light slow-wave sleep is increased,
number of stage shifts more frequent and duration of wakefulness episodes
reduced. Taken together, those results indicate that prenatal stress in rats is
associated with long-term alterations in various hormonal and behavioral parameters, which are comparable to those described in depressed patients, suggesting the usefulness of prenatally-stressed rats as an animal model of human
depression.
Influence of perinatal environment changes on individual’s
development and HPA axis activity
Changes of prenatal and postnatal environments can exert complex influences
on the development of an organism. In particular, life events occurring during
those two early periods of life can have different long-term behavioral effects.
For example, in humans, prenatal stress (PNS) can induce mental retardation
and sleep disturbances in the infant (17). In animals, dams stressed during
pregnancy can bear offspring with reduced male sexual activity, enhanced
emotional reactivity (22) and an increased propensity to self-administer drugs
(2). Conversely, postnatal stimulation has been found to improve the performance of aged offspring in cognitive tasks (10). Although prenatal and postnatal
events can have different behavioral consequences, they may also impinge on
the same behavioral response, and postnatal manipulations can reverse the
behavioral effects of prenatal stress. For example, it has been shown that postnatal handling or adoption (8,9) can reverse the increase in emotional reactivity induced by prenatal stress (21).
40
Several observations indicate that glucocorticoid secretion could be a substrate
of the different long-term behavioral effects of prenatal or postnatal events.
Prenatal stress increases stress-induced corticosterone secretion peak in preweaning rats (4) and attenuates its habituation over repeated exposure to stress
in the adult (3). In contrast, postnatal handling reduces stress-induced corticosterone secretion in adult and aged rats, probably by strengthening corticosterone feedback (10, 19). Finally, impairment in glucocorticoid feedback, resulting in an increased glucocorticoid secretion, is associated with behavioral
disorders in depression (6). We therefore thought to determine whether a
modification of fetal hormonal environment by stressing the mother can influence the development of the activity of the hypothalamo-pituitary-adrenal (HPA)
axis. It has already be shown that stress during pregnancy sensitizes different
neuroendocrine systems, such as gonadal and HPA axis (4). However, it remains unclear which mechanisms are involved in the dysregulation of corticosterone secretion in prenatally-stressed adult rats. Given that hippocampal
type I and type II corticosteroid receptors appear to be major regulating factors
in corticosterone secretion (15), we assessed stress-induced corticosterone
secretion and hippocampal corticosteroid receptors in adult rats that had been
submitted to prenatal manipulations (4,8).
Prenatal stress procedure
Prenatal stress was performed daily during the last week of pregnancy until
delivery. Pregnant females were individually restrained three times a day (at
09:00, 12:00 and 17:00 h) for 45 min in transparant plastic cylinders (7 cm
diameter, 19 cm long). Control pregnant females were left undisturbed in their
home cages. Offspring were weaned 21 days after birth and housed in samesex groups of four until the age of two months. Only litters of 8-13 pups with
similar numbers of males and females were utilized for the study. Only two
male pups per litter were studied as adults to minimize any possible “litter
effects” (1) on the measured variables.
Effects on the HPA axis
Repeated restraint of the mother during the last week of pregnancy induces prolonged corticosterone secretion in adult offspring (90 days of age), which was
indicative of impaired corticosterone feedback. Indeed, corticosterone levels in
either basal conditions or 30 min after stress did not differ between the control
and prenatally-stressed rats, but two hours after stress, corticosterone secretion
was higher in the prenatally stressed than in the control rats (Figure 1a) (2, 8).
Prenatal stress also decreased hippocampal type I corticosteroid receptors (Figure 1b) and, in contrast, as described by other authors (23), prenatal stress failed
to modify type II corticosteroid receptors (Figure 1c) (8).
41
Figure 1:
Plasma corticosterone secretion after restraint stress (a), type I (b) and
type II (c) corticosteroid receptors in control or prenatally-stressed adult
rats. a: corticosterone levels 120 min after restraint stress remained elevated in prenatally-stressed rats, whereas they returned to preexposure
values in controls. b: Prenatally-stressed rats showed a lower binding
capacity of type I corticosteroid receptors compared to controls. c: Prenatal stress did not significantly modify binding capacities of type II corticosteroid receptors. Affinities of type I and type II receptors were not modified by prenatal stress. Mean affinities were: type I = 1.14 ± 0.11 nM, type
II = 0.6 ± 0.12 nM. **p<0.01. Vertical line shows S.E.M.
The decrease in hippocampal type I corticosteroid receptors observed in
prenatally-stressed rats could account for their prolonged stress-induced corticosterone secretion. It has been shown that a selective reduction in hippocampal corticosteroid receptors is accompanied by a prolonged corticosterone
secretion in response to stress (13). In view of their affinities for corticosterone, it is generally thought that type II receptors are involved in stress-induced
feedback mechanisms, while type I receptors are involved in the tonic regulation of corticosterone release under basal conditions (11). Thus, the observed
decrease in hippocampal type I receptors might not be expected to be involved
in stress-modulated feedback control. However, there is also evidence that
both receptor types are involved in feedback control mechanisms (14). The
prolonged corticosterone secretion observed in prenatally-stressed animals could
also account for the behavioral alterations, as for example the increased propensity to amphetamine self-administration observed in prenatally-stressed
adult rats (2).
42
Influence of prenatal stress on circadian rhythms of locomotor
activity and corticosterone secretion in adult offspring
The circadian system plays a major physiological role to insure optimal functioning of the organism and its adaptation to the various changes in the environment (18). Alterations in circadian rhythmicity has been associated with
aging (18,20), sleep disorders (24) and affective disorders (12). Perinatal events
can also influence the functioning of the circadian system. Postnatal maternal
environment can influence circadian oscillations in plasma corticosterone: blind
rat pups adopted by dams out-of-phase with respect to their original mothers
do modify their corticosterone rhythm to reflect that of the foster mother (5).
An imposed restricted access to the natural mother can induce a shift in the
corticosterone rhythm of rat pups (16). Despite the fact that postnatal events
can have complex influences on circadian function, little is known on the longterm effects of prenatal manipulations on circadian rhythms in adult male rats.
Therefore, we sought to determine in adult offspring the long-term effects of
PNS on circadian rhythms of corticosterone and locomotor activity, two robust
markers of the circadian clock.
Locomotor activity rhythm
Control (N = 6) and PNS (N = 6) adult female offspring were kept in individual
cages equipped with a running-wheel for continuous recording of locomotor
activity via an on-line computer (Chronobiology Kit, Stanford Softwares Systems, CA, USA), and exposed to a 12:12 light-dark (LD) cycle with light dimmed
to 20 - 50 lux to reduce possible masking effects of light on actual onset of
locomotor activity. Individual 24-hour activity profiles were constructed for
each animal over 10 day-intervals. From those activity profiles, individual
onsets of locomotor activity were estimated by determining the time of the
first 15-min bin when the number of rotations exceeded 10% of the profile’s
highest level (peak) and remained at that level for at least 50% of the time in
the next 30 minutes.
Mean (± s.e.m.) onset of locomotor activity in the group of Control rats occurred 15 ± 12 min after lights-off, while it occurred 39 ± 22 min before lightsoff in the group of PNS rats (p < 0.02, unpaired Student’s t test). At the time of
lights-off, relative amplitude of locomotor activity (i.e.; % of peak activity value)
averaged, respectively, 15 ± 5 % and 35 ± 10 %, in Control and PNS rats (p <
0.05, unpaired Student’s t test).
43
Figure 2:
Mean (± s.e.m.) onset of locomotor activity relative to time of lights-off in
control adult female rats (Control) and prenatally-stressed rats (Prenatal
Stress). A value above zero line indicaties that activity occurred before
the time of lights-off, a value below this line indicaties that it occurred
after the time of lights-off.
Rhythm of corticosterone secretion
Control (N = 6) and PNS (N = 6) adult female offspring were housed in individual cages under a dim 12/12 LD cycle (light intensity = 150 lux) and allowed
2 weeks of habituation to this lighting condition. Rats were then implanted
with chronic intraveneus cathethers. After at least 12 days of recovery, blood
samples were collected at nine different points over the 24-hour cycle. After
each withdrawn, blood was immediately replaced with the same volume of saline, and the dead volume of the catheter was filled with heparin. Corticosterone levels were determined by radioimmunoassay using a highly specific corticosterone antiserum (Kit ICN Biomedicals Inc.) with a detection threshold of
0.1 µg/100ml.
PNS rats secreted higher corticosterone levels than Control rats two hours before the onset of the dark period (F(1,10) = 9.12, p<0.01). Corticosterone levels culminated at 45.79 + 6.9 ug/100 ml in Control females at the beginning of
the dark period, and in PNS females at 51.54 + 4.86 µg/100 ml 2 hours before
lights-off, indicating that prenatal stress induced a phase advance of corticosterone secretion in female offspring. Furthermore, the area under the curve
was significantly larger in PNS female rats (527 + 59 µg/100ml/24 h) compared
to controls (375 ± 20 (µg/100ml/24 h) (F(1,10)=5.816; p<0.05).
44
Figure 3:
Circadian fluctuations of plasma corticosterone levels measured at 9 points
over a 12 / 12 light / dark cycle in 6 control and 6 prenatally-stressed rats
: male on the left panel, female on the right panel. Times are shown at the
bottom, with the thick black bar representing the dark phase of the lightdark cycle. Prenatal stress induced a significant phase advance of the
corticosterone rhythm relative to the light/dark cycle (p < 0.01) and a
larger area under the curve (p < 0.05).
Prenatal stress can be responsible for major changes in the phase angle of
entrainement of circadian rhythms of locomotor activity and corticosterone
secretion. Our finding of a similar change (i.e.; a phase advance) in two rhythms
driven by the hypothalamic suprachiasmatic nuclei (SCN), the location of the
circadian clock in mammals (18) , raises the possibility that the circadian clock
of those animals has been altered by prenatal stressful events. Various clinical
observations in humans suggest a possible pathophysiological link between affective disorders (such as depression) and disturbances in circadian rhythmicity
(12). One of the current hypothesis on the neuroendocrinology of depression
involves a flattened (and advanced) circadian cortisol rhythm with
hypercortisolism, possibly due to an increased sensitivity of the adrenal cortex
(6) thought to normalize pituitary ACTH release in spite of an enhanced drive
from the hypothalamic CRH neurons (6).
45
Influence of prenatal environment modifications on sleep-wake
parameters in adult offspring
One strong marker of human depression is an alteration in the sleep-wake
cycle, including a shortened REM sleep latency, increased amount and frequency of REM sleep during the first part of the night, increased sleep fragmentation, and decreased slow wave sleep amount (7). In the present study we
studied sleep in the adult PNS male rat under baseline conditions.
Adult Control and PNS rats were implanted, under deep anaesthesia, with
chronic electrodes for polygraphic recordings of fronto-parietal electroencephalogram (EEG), electro-oculogram (EOG) and nuchal electromyogram (EMG).
The animals were then habituated to the sleep recording procedure over the
next for 14 days. The rats were connected with a cable to a rotating swivel
allowing free movements and EEG, EOG and EMG activities were recorded on
a polygraph (Nihon~Khoden, EEG-4414 A/K) with an output connected to a
computer for on-line spectral analysis of the EEG. Habituation consisted of
two sessions of recording for 8 hours and two sessions of 24 hours. At the end
of the habituation period, sleep was recorded for a period of 24 hrs beginning at
the onset of the light phase.
Figure 4:
Distribution per 12-h intervals (light phase, dark phase) of vigilance states
in 8 control (CONT) and 8 prenatally-stressed (PNS) rats under baseline
conditions. Mean (± s.e.m.) values of wake (W), light slow-wave sleep
(SWSI), deep slow-wave sleep (SWS2) and paradoxical sleep (PS) are
expressed as percentage of recording time. *p < 0.05, **p < 0.01, ***p <
0.001 (two-tail unpaired Student’s t-test) for between-groups comparisons.
Prenatal stress induced changes in both the structure and the continuity of
sleep in adult offspring. Paradoxical sleep (PS) was the most altered state (Fig-
46
ure 4). Compared to control rats, PNS rats showed increased total PS time, as
well as an increase in the percentage of total sleep time in PS, during both the
light and the dark phases. The increase in time spent in PS was due to an
increase in the number of PS episodes (+31% over the 24-h interval, p = 0.0014)
while the mean duration of PS episodes remained the same between the two
groups. PNS also induced an increase in total light slow wave sleep (SWS1)
time that was restricted to the dark phase. In addition, sleep was more fragmented during the dark phase in PNS animals, as indexed by a larger number
of episodes in each vigilance state and a shorter duration of wake (W) episodes.
This resulted in less time spent in W during the dark phase in the PNS animals
that was particularly pronounced during the last 4 hrs of darkness (Figure 4).
Quantitative analysis of EEG activity during specific vigilance states over the
12-h of light phase revealed minor differences in power spectral values (1 - 32
Hz) between the two groups of rats. EEG slow wave activity (SWA; EEG power
in the 1-5 Hz range), an indicator of sleep intensity, was evaluated during SWS2
in the light phase. SWA progressively decreased over consecutive 2-h intervals
during the light phase and the time course for this decrease was comparable in
the two groups. PNS rats showed a non significant tendency for higher SWA
levels in each 2-h interval; this tendency was also observed in power intensities
within the higher frequency ranges (5-32 hz). During PS, similar tendency for
enhanced power spectral values were observed in PNS rats as compared to
CONT rats in the entire frequency range studies (1-32 Hz).
Our results demonstratie pronounced effects of PNS on sleep in the adult rat
that parallel to some extent changes in sleep architecture found in depressed
humans.
Conclusions
Added to our previous findings in PNS rats of high anxiety and emotionality,
dysfunction of the HPA axis and circadian timing abnormalities, the observation of long-term changes in their sleep homeostasis supports the validity of
the “prenatal stress” model as a new animal model of depression. The persistence of all induced abnormalities after removal of the imposed stressors should
be seen as advantageous for the design and testing of new therapeutical strategies in mood and sleep disorders.
References
1 Chapman R., and J. Stern. Failure of severe maternal stress or ACTH during pregnancy to affect emotionality of male rat offspring: implications of litter effects for
prenatal studies. Dev Psychobiol. 12: 255-269, 1979.
2 Deminière J. M., P. V. Piazza, G. Guegan, N. Abrous, M. S., L. M. M., and H. Simon.
47
Increased locomotor response to novelty and propensity to intraveneus amphetamine self-administration in adult offspring of stressed mothers. Brain Res. 586: 135139, 1992.
3 Fride E., Y. Dan, J. Feldon, G. Halevy, and M. Weinstock. Effects of prenatal stress on
vulnerability to stress in prepubertal and adult rats. Physiol. Behav. 37: 681-687,1986.
4 Henry C., M. Kabbaj, S. H., M. Le Moal, and S. Maccari. Prenatal stress increases the
hypothalamo-pituitary-adrenal axis response in young and adult rats. J. of
Neuroendocrinol. 6: 341-345, 1994.
5 Hiroshige T., K. Honma, and K. Watanabe. Prenatal onset and maternal modifications of the circadian rhythm of plasma corticosterone in blind infantile rats. J Physiol
(Lond). 325: 521-32, 1982.
6 Holsboer F., U. Bardeleben, A. Gerken, G. Stalla, and 0. Muller. Blunted corticotropin and normal cortisol response to human corticotropin-releasingfactor in depression. N Engl J Med. 311: 1127, 1984
7 Kupfer D. J., and C. F. Reynolds.Sleep and affective disorders. In: Handbook of affective disorders, edited by E. S. Paykel. Edinburgh: Churchill Livingstone, 1992, p.
311-323
8 Maccari S., P. V. Piazza, M. Kabbaj, A. Barbazanges, H. Simon, and M. Le Moal. Adoption reverses the long term impairement in glucocorticoid feedback induced by prenatal stress. J. Neuroscience. 15: 110-116, 1995
9 Meaney M. J., D. H. Aitken, C. Van Berkel, S. Bhatnagar, and M. Sapolsky. Effects of
neonatal handling on age-related impairments associated with the hippocampus.
Science. 239: 766-768.,1988.
10 Meaney M. J., V. Viau, D. H. Aitken, and S. Bhatnagar. Stress-induced occupancy
and translocation of hippocampal glucocorticoid receptors. Brain Res. 445: 198203, 1988.
11 Reul J. M., F. R. van den Bosch, and E. R. de Kloet. Relative occupation of type-1 and
type-11 corticosteroid receptors in rat brain following stress and dexamethasone
treatment: functional implications. J Endocrinol. 115: 459-67, 1987.
12 Rosenwasser A., and A. Wirz-justice.Circadian Rhythms and Depression: Clinical
and Experimental Models. In: Physiology and pharmacology of biological rhythms,
edited by P. H. Redfern, and B. Lemmer. Berlin: Springer Verlag, 1997, p. 457-486.
13 Sapolsky R. M. The adrenocortical axis in the aged rat: impaired sensitivity to both
fast and delayed feedback inhibition. Neurobiol Aging. 7: 331-335, 1986.
14 Sapolsky R. M. The Adrenocortical Axis. Handbook of the Biology of Aging,3rd Edition.: 330-346,1990 .
15 Sapolsky R. M., L. C. Krey, and B. S. McEwen. The adrenocortical axis in the aged
rat: impaired sensitivity to both fast and delayed feedback inhibition. Neurobiol
Aging. 7:331-5,1986.
16 Shimoda K., K. Hanada, N. Yamada, and K. Takahashi. Restricted access to natural
mother shifted endogenous rhythm of rat pups. Brain and Development. 8:366-372,1986.
48
17 Stott D. N. Follow-up study from birth of the effects of prenatal stress. Dev. Med.
Child Neurol. 15: 770-787,1973
18 Turek F. W., P. Penev, Y. Zhang, 0. Van Reeth, J. Takahashi, and P. C. Zee. Alterations
in circadian system in advanced age. CIBA Foundation Symposium, Pitman Press,
London, 212-234,1995.
19 Vallee M., W. Mayo, F. Dellu, M. Le Moal, H. Simon, and S. Maccari. Prenatal stress
induces high anxiety and postnatal handling induces low anxiety in adult
offspring: role of corticosterone. J. of Neurosci. 17: 2626-2636, 1997.
20 Van Reeth O., Y. Zhang, P. Zee, and F. Turek. Aging alters the feedback effects of the
activity-rest cycle on the circadian clock. Am. J. Physiol. 263: R981-R986, 1992.
21 Wakshlak A., and M. Weinstock. Neonatal handling reverses behavioral abnormalities induced in rats by prenatal stress. Physiol. Behav. 48: 289-292, 1990.
22 Weinstock M., E. Fride, and R. Hertzberg. Prenatal stress effects on functional development of the offspring. Progress in Brain Research. 73: 319-331, 1988.
23 Weinstock M., E. Matlina, G. 1. Maor, H. Rosen, and B. S. McEwen. Prenatal stress
selectively alters the reactivity of the hypothalamic-pituitary adrenal system in the
female rat. Brain Res. 595: 195-200, 1992.
24 Weitzman E. D., C. A. Czeisler, R. M. Coleman, A. J. Spielman, J. C. Zimmerman, W.
C. Dement, G. S. Richardson, and C. P. Pollak. Delayed sleep phase insomnia: A
chronobiologic disorder associated with sleep onset insomnia. Arch. Gen. Psychiatry. 38: 737-746, 1981.
49
50
EEG topography during slow wave sleep,
normative data and modifications
induced by thalamic lesions
D. DIVE, R. POIRRIER, F. CLAES AND G. FRANCK
DEPARTMENT OF NEUROLOGY
C.H.U. LIÈGE, BELGIUM
Correspondence to:
Dr. D. DIVE, Department of Neurology, CHU Sart-Tilman (B35), University of
Liège, 4000 Liège, Belgium
E-Mail: [email protected]
Introduction
Recent studies brought us a lot of information about slow wave sleep physiological mechanisms and disclosed several oscillating elements integrated in
networks whose functional properties are greatly modified by comparison with
waking state. We used topographical quantified EEG analysis in order to study
slow waves and spindles activities during sleep in the presence of thalamic
lesions compared with control data.
Slow waves
Since RECHTSCHAFFEN and KALES definitions [27], unicellular
electrophysiological investigations lead to dissociate delta activities between a
slow rhythm (below 1 Hz) and truly intrinsic delta oscillations (between 1 and
4 Hz) generated in cortical neurons or as a clock-like rhythm in thalamo-cortical neurons [5].
A slow oscillation (<1Hz) described in intracellular recordings from cortical
and thalamic neurons [35,41,42] is able to synchronise and group other sleep
rhythms like spindles and delta into complex sequences [32]. The interplay
between such oscillations reflects at the cortical level and yields to patterns
that take the shape of polymorphic waves [5]. The slow rhythm is essentially
induced by alternative sequences of long-lasting depolarisation and hyperpolarisation phases [11,42]. It was initially observed in cats during anaesthesia
and has been more recently demonstrated during natural sleep in animals
51
[33,34] and humans [1,4,6,42]. A lot of experiments point the cortex as the
site of genesis of this slow oscillation where it appears over large areas in a
synchronous manner [2,3,5,41,43]. Combined unicellular intracortical recordings and frequency analysis have established the relation between K-complexes
and this slow oscillation [4,6].
Intrinsic cortical and thalamo-cortical neurons properties are responsible for
delta activities distributed in the 1 to 4 Hz frequency range. The basic mechanism relies on the interplay between two currents (Ih and It) which are activated at hyperpolarized membrane potentials [5,23,30,31,40].
Spindles
Since 1950, visual EEG analysis separated two distinct spindles activities on
the basis of frequency and topography [13]. The slower spindles (11.5 — 14.0
Hz) are distributed in the frontal region whereas posterior ones are more
focalised in the parietal region and are of higher frequency (14.0 — 16.0 Hz)
[16-18,25,29,44]. On the counterpart, intracellular recordings led to a homogenous view of a spindling phenomenon largely distributed in a wide frequency
range from 7 to 14 Hz [11,32,36-38,40]. The EEG spindles are the epitome of
brain synchronisation at the onset of sleep, mainly due to the pacing role of the
rostral part of the thalamic reticular nucleus [37,38,40]. GABAergic reticular
neurons inhibit large numbers of thalamo-cortical cells through IPSPs that are
able to trigger low threshold calcium spikes (LTS) and associated bursts of
action potentials. Integrated in a large thalamo-cortico-thalamic network, these
action potentials are transferred to the cortex where they generate spindles
waves [12,20,21,32,36,39,40].
Patients and methods
Seven patients with vascular thalamic lesions (six patients with unilateral lesions and one patient with a bilateral anterior lesion, 62.86 ± 6.96 (S.D.) years
old) were compared to six age-matched control subjects (64.50 ± 6.80 (S.D.)
years old). 32 polygraphic whole night recordings were acquired (28 EEG channels) with a NICOLET PATHFINDER I system. Earlobe controlateral to the studied hemisphere was used as reference in order to avoid false lateralisations or
temporal region signal contamination due to the reference. We used FFT analyses (0.147 Hz resolution) with spectrum averaging (80 epochs of 6819 msec.
duration each) in order to obtain a representative picture of the frequency domain for stages 2 and 4. Spatial normalisation allowed intrinsic topographical
comparisons (Z scores with significant threshold established at a p<0.05 level).
We applied generalised linear models [19] for classical frequency bands [15]
statistical analysis. These methods allowed us to evaluate patient, electrode and
sleep stage interactions using a statistical arbitrary left frontal reference point.
52
Figure
Comparison of slow wave sleep EEG results in case of thalamic lesions
Top: MRI or CT-scan data of seven cases (6 patients with unilateral lesions and 1 patient
with a bilateral dorso-median lesion)
Middle: topographic maps of EEG rhythms during slow wave sleep. 0.5 to 7.5 Hz signals
were acquired during stage 4 and 7.5 to 17.9 Hz activities were analysed during stage 2.
N : no significant power difference between patient data compared to normal controls
↑ or P↑: increased power
↓: power at a lower level than for control subjects
N.P. : no significant difference between patient power level compared with control subjects
n.s. : no significant
Results
Slow wave activities
Slow delta rhythms (0.5 — 2.0 Hz) are distributed in the frontal anterior and
frontal lateral regions during stage 2 or 4. The absolute power level is greatly
53
higher during stage 4 than during stage 2. Whereas topographical distribution
is exactly the same throughout whole slow wave sleep period, statistical
focalisation is proportional to the power level, and is significant only during
stage 4.
Delta rhythms between 2.0 to 3.5 Hz reveal a topographical focalisation in the
medial central region that is significantly different than for the frontal slower
delta activities. Those frequency range rhythms are also focalised in proportion with their power level.
Patient’s data disclosed no significant change in relation with a thalamic unilateral or bilateral lesion. Only minor increased delta power was observed in a
case with voluminous haemorrhagic thalamic infarction with probably extrathalamic involvement (Figure case 1). It was specially interesting to note that
a patient with a bilateral dorso-median thalamic infarct responsible for important clinical symptoms and frontal glucose hypometabolism during waking state
([18F]FDG PET scan), had strictly normal delta activities during slow wave sleep
(Figure case 7).
Spindles
Multiple peaks are observed in the normal mean spectra within the frequency
range of spindle activities (7.5 — 15.2 Hz). Two main groups are distinguished on
the basis of power level and topographical distribution. Anterior spindle rhythms
are observed in the medial frontal and central regions within 10.0 to 12.5 Hz frequency range. They are statistically different from the posterior spindles that are
disclosed in the medial central and parietal areas within the 12.5 to 15.2 Hz ranges.
Differential absolute power statistical maps clearly define opposite topographical
distributions between these two rhythms. The same results are obtained with
normalised power spectra that show slower activities on Fz than faster ones on Pz.
Statistical focalisation is only seen during stage 4 when power level is lower and
topographic extension more restricted than during stage 2. This aspect is completely different that what happens for delta activities.
Patients with voluminous unilateral thalamic lesions involving the rostral part of
the lateral portion of the thalamus had a dramatic modification of the spindle
activities (Figure cases 1 and 5). Absolute power was increased on the controlateral
side of the lesion and the topographical distribution of either frontal or parietal
spindles was largely extended. One patient with a smaller lesion restricted to the
posterior and lateral part of the thalamus had lateralised spindle activities on the
safe side without significant power level or topographic extension modification
(Figure case 2).
Patients with unilateral dorsal thalamic or anterior bilateral small lesions had normal spindle activities parameters (Figure cases 3, 4 and 7).
54
Conclusions
Our results indicate that slowest delta activities during slow wave sleep are
distributed over frontal regions without any significant change in the presence
of a unilateral thalamic lesion. These rhythms are preserved even if there is a
lesion in the thalamic nuclei for which frontal cortex is the projection area. We
hypothesise that the slow rhythm described at the unicellular level is the basis
of these slowest delta activities and that they have a frontal cortex substrate.
This relation leads us to evoke a direct link between frontal slow activities and
fronto-basal combined with frontal cingular regions cerebral blood flow preferential decrease described with PET studies [9,14,22] during slow wave sleep.
Delta activities in the 2.0 to 3.5 Hz frequency range probably reflect the whole
network interplay of dorsal thalamic and cortical neurons which share intrinsic oscillating properties at a hyperpolarized membrane voltage level, reflecting preferentially in the central cortical region. Spindles activities are distributed over a wide frequency window segmented into two compartments in regard
with their frontal or parietal median distribution.
Our results obtained with patients are in good accordance with a pacemaker
role devoted to the rostral part of the reticular thalamic nucleus. Moreover, we
observed power increase in the spindle frequency range on the controlateral
side of a thalamic lesion if the anterior and lateral portion of the thalamus was
involved in the pathologic process. Those data raise the hypothesis of a
bithalamic functional counterbalance in the regulation of the cortical expression of spindling activities. Anatomical animal data demonstrated direct
bithalamic relationships involving reticulo-reticular or reticulo-dorso-thalamic
pathways [7,8,10,24,26,28] that could be the substrate of such control.
Quantified EEG topographical studies during sleep give data that can be integrated in the whole field of functional evaluation of the brain during a state that
is characterised by important electrophysiological and metabolic modifications
by comparison with wake. They allow human global cortical functional measurements even in the presence of cerebral lesions. Our studies have to be
developed in the future with coregistration with other functional techniques
like PET scan or MRI. REM-sleep study is mandatory. Other mathematical
signal analyses, like time-frequency methods, would be of a great interest in
order to avoid FFT temporal resolution compression.
EEG signal analysis remains thus one interesting way of functional investigation in closed relationship with neuronal firing greatly modified during sleep
that is characterised by a brain environmental disconnection.
55
Bibliography
1 ACHERMANN P, BORBÉLY AA: Low-frequency (<1Hz) oscillations in the human
sleep electroencephalogram. Neurosc 1997, 81:213-222.
2 AMZICA F, STERIADE M: Disconnection of intracortical synaptic linkages disrupts
synchronisation of a slow oscillation. J Neurosci 1995, 15:4658-4677.
3 AMZICA F, STERIADE M: Short- and long-range neuronal synchronization of the
slow (<1Hz) cortical oscillation. J Neurophysiol 1995, 75:20-38.
4. AMZICA F, STERIADE M: The K-complex : its slow (<1Hz) rhythmicity and relation
to delta waves. Neurology 1997, 49:952-959.
5. AMZICA F, STERIADE M: Electrophysiological correlates of sleep delta waves.
Electroenceph Clin Neurophysiol 1998, 107:69-83.
6 AMZICA F, STERIADE M: Cellular substrates and laminar profile of sleep K-complex.
Neurosc 1998, 82:671-686.
7 BATTAGLIA G, LIZIER C, COLACITTI C, PRINCIVALLE A, SPREAFICO R: A
reticuloreticular commissural pathway in the rat thalamus. The Journal of Comparative Neurology 1994, 347:127-138.
8 BATTAGLIA G, LIZIER C, COLACITTI C, REGONDI C, SPREAFICO R: Commissural pathways in the rat thalamus : evidence for a reticulo-reticular connection. In
Thalamic networks for relay and modulation, Edited by Minciacchi D, Molinari M,
Macchi G, Jones EG. Oxford: Pergamon Press; 1993:337-346.
9 BRAUN AR, BALKIN TJ, WESENSTEN NJ, CARSON RE, VARGA M, BALDWIN P,
SELBIE S, BELENKY G, HERSCOVITCH P: Regional cerebral blood flow throughout
the sleep-wake cycle. An H215O study. Brain 1997, 120:1173-1197.
10 CHEN S, RAOS V, BENTIVOGLIO M: Connections of the thalamic reticular nucleus
with the controlateral thalamus in the rat. Neurosci Lett 1992, 147:85-88.
11 CONTRERAS D, STERIADE M: Cellular basis of EEG slow rhythms: a study of dynamic corticothalamic relationships. J Neurosci 1995, 15:604-622.
12 DESCHENES M, PARADIS M, ROY JP, STERIADE M: Electrophysiology of neurons of
lateral thalamic nuclei in the cat : resting properties and burst discharges. J
Neurophysiol 1984, 51:1196-1219.
13 GIBBS EL, GIBBS FA: Atlas of electroencephalography, Cambridge: Addison-Wesley;
1950.
14 HOFLE N, PAUS T, REUTENS D, FISET P, GOTMAN J, EVANS AC, JONES BE: Regional cerebral blood flow changes as a function of delta and spindle activity during
slow wave sleep in humans. J Neurosci 1997, 17:4800-4808.
15 INTERNATIONAL PHARMACO-EEG GROUP (IPEG): Recommendations for EEG and
evoked potential mapping. Neuropsychobiology 1989, 22:170-176.
16 JANKEL WR, NIEDERMEYER E: Sleep spindles. [Review]. J Clin Neurophysiol 1985,
2:1-35.
17 JOBERT M, POISEAU E, JAHNIG P, SCHULZ H, KUBICKI S: Pattern recognition by
matched filtering: an analysis of sleep spindle and K-complex density under the in-
56
fluence of lormetazepam and zopiclone. Neuropsychobiology 1992, 26:100-107.
18 JOBERT M, POISEAU E, JAHNIG P, SCHULZ H, KUBICKI S: Topographical analysis
of sleep spindle activity [published erratum appears in Neuropsychobiology
1993;27(4):236]. Neuropsychobiology 1992, 26:210-217.
19 LITTLE RC, MILLIKEN GA, STROUP WW, WOLFINGER RD: SAS system for mixed
models. SAS Institute Inc , Cary, North Carolina (version 6.11), 1996.
20 LLINAS RR, YAROM Y: Electrophysiology of mammalian inferior olivary neurons in
vitro. Different types of voltage-dependent ionic conductances. J Physiol (Lond )
1981, 315:549-567.
21 LLINAS RR, YAROM Y: Properties and distribution of ionic conductances generating
electroresponsiveness of mammalian inferior olivary neurons in vitro. J Physiol (Lond
) 1981, 315:569-584.
22 MAQUET P, DEGUELDRE C, DELFIORE G, AERTS J, PÉTERS JM, LUXEN A, FRANCK
G: Functional neuroanatomy of human slow wave sleep. J Neurosci 1997, 17:28072812.
23 McCORMICK DA, PAPE HC: Properties of a hyperpolarization-activated cation current and its role in rhythmic oscillation in thalamic relay neurones. J Physiol (Lond
) 1990, 431:291-318.
24 PARE D, STERIADE M: The reticular thalamic nucleus projects to the contralateral
dorsal thalamus in macaque monkey. Neurosci Lett 1993, 154:96-100.
25 POISEAU E, JOBERT M, SCHULZ H, KUBICKI S: Localization of sleep spindle activity. Sleep Research 1991, 20:60
26 RAOS V, BENTIVOGLIO M: Crosstalk between the two sides of the thalamus through
the reticular nucleus: a retrograde and anterograde tracing study in the rat. Journal
of Comparative Neurology 1993, 332:145-154.
27 RECHTSCHAFFEN A, KALES A: A manual of standardized terminology, techniques
and scoring system for sleep stages of human subjects, Los Angeles: Brain Information Service / Brain Research Institute; 1968.
28 RINVIK E: Thalamic commissural connections in the cat. Neurosci Lett 1984, 44:311316.
29 SCHEULER W, KUBICKI S, SCHOLZ G, MARQUARDT J: Two different activities in
the sleep spindle frequency band-discrimination based on the topographical distribution of spectral power and coherence. In Sleep ‘90, edited by Horne J. Bochum:
Pontenagel Press; 1990:13-16.
30 SOLTESZ I, LIGHTOWLER S, LERESCHE N, JASSIK-GERSCHENFELD D, POLLARD
CE, CRUNELLI V: Two inward currents and the transformation of low-frequency
oscillations of rat and cat thalamocortical cells. J Physiol (Lond ) 1991, 441:175197.
31 STERIADE M: Cellular substrates of brain rhythms. In Electroencephalography :
basic principles, clinical applications and related fields, 3rd edn. Edited by
Niedermeyer E, Lopes da Silva FH. Baltimore: Williams & Wilkins; 1993:27-62.
57
32 STERIADE M, AMZICA F: Coalescence of sleep rhythms and their chronology in
corticothalamic networks. Sleep Research Online 1998, 1:1-10.
33 STERIADE M, AMZICA F, CONTRERAS D: Synchronization of fast (30-40 Hz) spontaneous cortical rhythms during brain activation. J Neurosci 1996, 16:392-417.
34 STERIADE M, CONTRERAS D, AMZICA F, TIMOFEEV I: Synchronisation of fast (3040 Hz) spontaneous oscillations in intrathalamic and thalamocortical networks. J
Neurosci 1996, 16:2788-2808.
35 STERIADE M, CONTRERAS D, CURRO DOSSI R, NUNEZ A: The slow (< 1 Hz) oscillation in reticular thalamic and thalamocortical neurons: scenario of sleep rhythm
generation in interacting thalamic and neocortical networks. J Neurosci 1993,
13:3284-3299.
36 STERIADE M, DESCHENES M: The thalamus as a neuronal oscillator. Brain Res
1984, 8:1-63.
37 STERIADE M, DESCHENES M, DOMICH L, MULLE C: Abolition of spindle oscillations in thalamic neurons disconnected from nucleus reticularis thalami. J
Neurophysiol 1985, 54:1473-1497.
38 STERIADE M, DOMICH L, OAKSON G, DESCHENES M: The deafferented reticular
thalamic nucleus generates spindle rhythmicity. J Neurophysiol 1987, 57:260-273.
39 STERIADE M, GLOOR P, LLINAS RR, LOPES DA SILVA FH, MESULAM M-M: Basic
mechanisms of cerebral rhythmic activities. Electroenceph Clin Neurophysiol 1990,
76:481-508.
40 STERIADE M, McCORMICK DA, SEJNOWSKI TJ: Thalamocortical oscillations in the
sleeping and aroused brain. [Review]. Science 1993, 262:679-685.
41 STERIADE M, NUNEZ A, AMZICA F: Intracellular analysis of relations between the
slow (< 1 Hz) neocortical oscillation and other sleep rhythms of the electroencephalogram. J Neurosci 1993, 13:3266-3283.
42 STERIADE M, NUNEZ A, AMZICA F: A novel slow (< 1 Hz) oscillation of neocortical
neurons in vivo: depolarizing and hyperpolarizing components. J Neurosci 1993,
13:3252-3265.
43 TIMOFEEV I, STERIADE M: Low-frequency rhythms in the thalamus of intact-cortex and decorticated cats. J Neurophysiol 1996, 76:4152-4168.
44 ZEITLHOFER J, GRUBER G, ANDERER P, ASENBAUM S, SCHIMICEK P, SALETU B:
Topographic distribution of sleep spindles in young healthy subjects. J Sleep Res
1997, 6:149-155.
58
Inter-individual phase differences
in circadian rhythms and sleep
GERARD KERKHOF
STATE UNIVERSITY OF LEIDEN
Morning-type (M-type) individuals reach an earlier phase position of their circadian rhythms than evening-type (E-type) individuals. This has not only been
observed under natural conditions (e.g. for body temperature, subjective alertness and mood), where masking influences may have been involved, but also
under constant conditions, where direct masking influences were eliminated
or kept constant. Thus, evidence exists for the endogenous nature of the circadian phase difference between M-types and E-types.
When allowed to choose their preferred sleep times, E-types generally have
their sleep onset at an earlier phase of their temperature rhythm than M-types.
Given the well-documented relationship between the phase of the temperature
rhythm and REM-sleep propensity, a M-type vs E-type difference would be expected with respect to the time-course of REM-sleep. Contrary to expectation,
E-types appear to have a higher REM-sleep propensity in the first part of their
sleep than M-types. As a likely result of the interactive inhibition between
REM-sleep and slow wave sleep, E-types also show a relative curtailment of
slow wave sleep during their first NREM-REM cycle.
The impact of a relatively small difference in the phase relationship between
sleep and body temperature was investigated by systematically shifting the
timing of the sleep period of M-types and E-types, while keeping constant the
length of their prior wakefulness. Rectal temperature recordings confirmed
that the sleep shifts had not affected the phase of the body temperature rhythm.
The results showed that: 1. the sleep shifts had a similar influence upon the
sleep structure of the two groups of subjects: a relative advance of the sleep
period was associated with an increased REM latency, an increased percentage
of slow wave energy during the first nonREM period, and a compensatory reduction of the percentage of slow wave energy during the later half of sleep;
and 2. the E-types had a higher overall REM propensity than the M-types. The
latter difference may be associated with a difference in personality and
behavioral characteristics of the two groups of subjects.
59
60
61
62
Neurophysiology of sleep-wake states
in relation to consciousness and
information processing
ANTON M.L. COENEN
NICI, DEPARTMENT OF PSYCHOLOGY, UNIVERSITY OF NIJMEGEN,
P.O. BOX 9104, 6500 HE NIJMEGEN, THE NETHERLANDS,
TEL + 31 24 36 12545, FAX + 31 24 36 16066,
E-MAIL: Coenen@ NICI.KUN.NL
Summary
Wakefulness is accompanied by a low amplitude high frequency electroencephalogram, due to the fact that thalamocortical neurons fire in a state of tonic
depolarization. In this situation information can easily pass the low-level threshold of these neurons, leading to a high transfer ratio. The complexity of the
electroencephalogram during conscious waking is high as expressed in a high
correlation dimension. Accordingly, the level of information processing is high.
Spindles mark the transition from wakefulness to sleep. These phenomena are
related to drowsiness, associated with a reduction in consciousness. Drowsiness occurs when cells undergo moderate hyperpolarizations. Increased inhibitions result in a reduction of afferent information with a lowered transfer
ratio. Information processing subsides, which is also expressed in a diminished
correlation dimension. Consciousness is further decreased at the onset of slow
wave sleep. This sleep is characterized by a high voltage, low frequency electroencephalogram. Slow wave sleep becomes manifest when neurons undergo
a further hyperpolarization. Inhibitory activities are so strong that the transfer
ratio further drops, as well as the correlation dimension. Thus, sensory information is largely blocked and information processing is on a low level. Nevertheless, subconscious information processing and stimulus evaluation can take
place during slow wave sleep. Similar to the patterns in the electroencephalogram, the architecture of evoked potentials is dependent on the state of alertness. During waking, components in event related potentials are moderate in
amplitude, while during slow wave sleep larger waves are visible. This is caused
by more synchronized unit responses with sharper phases of excitations and
inhibitions, which results from increased hyperpolarizations. In contrast, event
related potentials belonging to rapid eye movement sleep closely resemble those
of wakefulness. This type of sleep is associated with a ‘wake-like’
63
electroencephalographic pattern. Just as during wakefulness, this is the expression of a depolarization of thalamocortical neurons, The transfer ratio of
rapid eye movement sleep has not yet been determined, but seems to vary.
Evidence exists that rapid eye movement sleep, associated with dreaming, with
some kind of perception and consciousness, is involved in processing of ‘internal’ information. In line with this, rapid eye movement sleep has higher correlation dimensions than slow wave sleep. It is assumed that the ‘near-the-threshold’ depolarized state of neurons in thalamus and cerebral cortex is a necessary
condition for perceptual processes and consciousness, such as occurs during
waking and in an altered form during rapid eye movement sleep.
Waking and sleeping
A key structure in the regulation of sleeping and waking and thus of consciousness is the reticular formation, a meshwork of nuclei and tracts located in the
brainstem (e.g. Steriade and McCarley, 1990). The brainstem reticular formation roughly consists of two systems. The first is located in the rostral mesencephalic part of the brainstem and is called the midbrain reticular formation.
The second system, positioned in the caudal medullar part of the brainstem, is
known as the medullar or bulbar reticular formation. There is overwhelming
evidence for the view that the midbrain reticular formation controls the process of wakefulness and sets the general level of activation of the brain (Singer,
1977; Steriade et al, 1990). The activating influence from the mesencephalic
reticular formation on the cerebral cortex is transmitted through two ways: a
dorsal pathway to the thalamus and a ventral route to the basal forebrain
(Jones, 1990; Szymusiak, 1995). Activation of the midbrain reticular formation leads to an enhancement of the spontaneous discharges of neurons, both
in thalamus and in extended areas of the cerebral cortex (Jones, 1990; Siegel,
1990). In short, the anatomical system essential for wakefulness and vigilance
is roughly located where mesencephalon, thalamus and hypothalamus meet.
This implies that this junction and its vicinity is crucial for preparing the cerebral cortex towards a condition conducive to information processing and consciousness.
Sleep is controlled by the medullar or bulbar reticular formation. In this area a
population of neurons is maximally active when slow wave sleep occurs (Siegel,
1990). It is thought that this part influences GABA-containing neurons in the
dorsolateral pontine tegmentum lying there intermingled with excitatory cholinergic neurons. These GABA-ergic cells could dampen the activity of adjacent
projection neurons, such as the cholinergic ones, or directly influence distant
structures, such as thalamus or cerebral cortex (Jones, 1990). Thalamic inhibitory interneurones located in the thalamic reticular nucleus as well as the
short-axoned interneurons located in the thalamic relay nuclei itself, also play
64
part in the decrease of activation of thalamocortical neurons (Steriade and
McCarley, 1990). Parts of the hypothalamic region such as the preoptic area
and even adjacent basal forebrain areas are further involved in the regulation
of slow wave sleep (Szymusiak, 1995). Thus, slow wave sleep generation involves interaction among several brainstem, diencephalic and forebrain cell
groups (Siegel, 1990).
Neuronal activities and the electroencephalogram
A relationship exists between the patterns in the electroencephalogram (EEG)
and the level of vigilance and consciousness. Active wakefulness is accompanied by low amplitude high frequency (beta) waves in the EEG, whereas the
EEG of slow wave sleep is composed of high voltage, low frequency (delta)
waves. During waking, thalamocortical cells are in a state of tonic depolarization with relatively stable membrane potentials of around -60 mV. Neurons fire
in the ‘tonic’ or ‘relay’ mode, implying a sustained and high spontaneous activity (Glenn and Steriade, 1982; Steriade and McCarley, 1990). This variable
discharge pattern with a low synchronization between cells is the reason why
EEG electrodes, which summate the electrical activity of numerous cells, record
small but irregular waves with a high frequency of fluctuations. The tonic mode
of firing is the substrate of beta waves (Figure 1).
Figure 1:
At the left side the EEG of alert wakefulness is shown in the upper trace
and the EEG of deep slow wave sleep in the lower trace. At the right side
the spontaneous activity of a corticothalamic neuron in the same states is
shown. Note the ‘tonic’ firing mode of wakefulness in the upper trace and
the ‘burst’ firing mode of slow wave sleep in the lower trace. Bursts have
the same frequency as the mean frequency of the large delta waves in the
EEG. [The right panel is taken from Glenn and Steriade, 1982].
The occurrence of spindles, often a sign for drowsiness or light sleep, marks
the transition from alert wakefulness, with its low voltage high frequency waves,
to slow wave sleep, with its large amplitude low frequency (delta) waves. Spindles become manifest when thalamocortical relay cells undergo a moderate
65
hyperpolarization with membrane potentials lower than -60 mV (Steriade, 199l).
This firing mode can be called the ‘oscillatory’ mode. The high voltage, irregular and low frequency waves of slow wave sleep, become manifest when neurons undergo a further hyperpolarization to about -70 till -90 mV. Delta waves
have a large amplitude, which implies that extended populations of neurons
fire rather synchronously in bursts, interspersed with prolonged
hyperpolarizations (Figure 1). Inhibitory interneurones play a role in the lowering of the membrane potential of relay cells; they are also responsible for the
strong synchronization of these cells by tying them together by powerful inhibitory activities. In contrast to spindles, delta waves are not rhythmical but
highly irregular. This mode of activation results in pause-burst discharges of
many cells and is called the ‘burst’ mode (Steriade, 1991; Steriade et al, 1993).
‘Rapid eye movement’ (REM) sleep is controlled by the pontine reticular formation. This is a third system in the brainstem roughly located between the
mesencephalic and medullar reticular formation (Jouvet, 1967; Hobson and
Steriade, 1986). Characteristic for REM sleep are the ‘ponto-geniculo-occipital’
(PGO) waves originating in this part of the pons. These waves propagate rostrally
and project through the lateral geniculate nucleus and other thalamic relay
nuclei to the cerebral cortex (Siegel, 1990). PGO waves are the pacemakers for
the activation of thalamus and extended cortical areas. As during waking, neurons become tonically depolarized and start to fire in the tonic mode (Glenn
and Steriade, 1982). Although the high spontaneous activity characteristic of
REM sleep is not limited to the sensory areas of the cortex but also include the
motor areas, the activity of the latter parts is not expressed at a bodily level. A
deep hyperpolarization of neurons in the peripheral motor system is the underlying mechanism for muscular relaxation (Chase and Morales, 1989). This
mechanism prevents gross overt movements during REM sleep and with exception of the tiny muscles of eyes and extremities, all muscles are relaxed.
Transfer of information
All information is encoded in electric impulses by the sensory organs. The
transfer of this information over the sensory pathways to the thalamus and
further to cortical areas is dependent of the state of vigilance. The concept of
the ‘transfer ratio’ was introduced by Coenen and Vendrik in 1972. They performed research on the flow of transmission from the peripheral visual organs
to the visual cortical areas during sleeping and waking. Using intracellular recordings of cells in the lateral geniculate body of the cat, Coenen and Vendrik
(1972) showed that during wakefulness the ratio between the output and the
input of a thalamocortical relay neuron varies between 0.7 and 1.0. The latter
value is reached under circumstances of alert wakefulness and implies that all
action potentials of retinal ganglion cells produce excitatory postsynaptic
66
potentials (EPSPs), which easily pass the low-level threshold of geniculate neurons firing in the tonic or relay mode. Obviously, this can be regarded as the
underlying process of the high transfer ratio. All EPSPs generate outgoing action potentials and the transmission occurs in a way of ‘one input to one output
action potential’ (Coenen, 1995, 1998, Steriade et al, 1993). This means that
the complete message as coded by the peripheral receptors, reaches the sensory parts of the cortex in its entirety. A massive thalamocortical and corticothalamic traffic is the result (Steriade et al, 1993).
The transfer ratio goes down till about 0.7 when the animal becomes drowsy.
Then the output decreases while the input remains identical. Nevertheless, the
peripheral sense organs transform sensory stimuli in series of impulses just as
during waking and independently from the state of vigilance of the organism.
The information reaching the cortical level strongly decreases during slow wave
sleep. This process is called ‘thalamic’ or ‘sensory gating’ (McCormick and Bal,
1994). Thus, a decline of the processing of information becomes manifest when
the states of drowsiness and light slow wave sleep appear. A further increase of
the hyperpolarization of thalamocortical neurons is associated with slow wave
sleep. This results in a further inability of the EPSPs, produced by the incoming action potentials, to reach the increased threshold. The burst mode in which
the system fires, continuously blocks a major part of the afferent information
at the thalamic level. The blocking of the incoming series of spikes becomes so
strong that the transfer ratio drops further to about 0.3 or 0.4. Thus, the transfer ratio or sensory gating strongly depends of the level of vigilance (Figure 2).
Livingstone and Hubel (1981) extended and extrapolated these findings to the
primary visual cortex.
When the animal is awakened by an external arousing stimulus, the transfer
ratio immediately returns shortly back to about one (Figure 2) (Coenen and
Vendrik, 1972). This was confirmed by electrical stimulation of the mesencephalic reticular system of a cat, which also causes an arousal response (Singer,
1977). Singer could also demonstrate that electrical stimulation of the rostral
part of the brainstem was associated with a reduction of intrathalamic inhibition and with an induction of a depolarized state in cells of the lateral geniculate body, initiating a high and sustained spontaneous activation and an opening of the sensory channels. Thus, facilitation of the transfer of impulses through
the thalamic relay nuclei is controlled by the activating reticular system of the
brainstem.
67
Figure 2:
Responsiveness of a neuron in the lateral geniculate body of a cat to visual
stimuli. A flash of light is given every second and the vertical lines represent the spike response of the cell (output). The upper, horizontal interrupted line indicates the input to the cell. Following an arousal stimulus
to the cat (arrow), the sleep EEG desynchronizes into a wake EEG, which
is accompanied by an increase of the spike response to the flash. At that
point the transfer ratio increases from about 0.5 to about 1, shortly later
followed by a small decline to 0.7 or 0.8. [Adapted from Coenen and
Vendrik, 1972].
During slow wave sleep there is no conscious perception which, however, does
not imply that cognitive activities must completely be excluded. The transfer
ratio is low but does not come beyond 0.3 till 0.4, which indicates that some
information still reaches the cortical levels. The analysis of this information
remains intact to a certain degree during slow wave sleep. Stimulus evaluation,
for example, is an automatic process that can take place under lower levels of
consciousness. Thoroughly investigated is stimulus evaluation which takes place
during sleep. It is well known, for example, that the threshold for awakening is
lower for a relevant stimulus than for a physically identical stimulus which has
no relevance for the individual (Langford et al, 1974). This shows that during
the state of slow wave sleep only a shade of the original information reaches
cortical levels. It might be possible that the amount of information passing
through the thalamus during sleep is just enough for a shallow, subconscious
evaluation.
68
At this time the transfer ratio of REM sleep has not been determined. Nevertheless, anecdotal data can be gathered, suggesting that this ratio varies with the
fluctuating threshold of awakening during REM sleep. Presumably, the ratio is
quite high when the awakening threshold is low at the end of a REM sleep
period. Evidence for this view is the presence of a depolarized state of the
neurons analogous to waking (Steriade and McCarley, 1990) and further the
identical shape of visual evoked potentials made during REM sleep and waking
(van Hulzen and Coenen, 1984). Obviously, a depolarized state of thalamocortical cells is a necessary condition for perceptual processes such as occur during waking and dreaming, the latter being associated with REM sleep. The easy
integration of external events in an on-going dream also points to a high transfer ratio during REM sleep.
Correlation dimension and evoked potentials
A reasonable supposition suggests that the small amplitude high frequency ‘wake’
EEG, generated by a large number of relatively non-synchronized independently firing cells, is more complex than the large amplitude low frequency ‘sleep’
EEG. As mentioned above, the latter EEG pattern is the result of the synchronized pause-burst firing mode of neurons, which are coupled by inhibitory
interneurons. The complexity of the EEG can be expressed by correlation dimension, a quantitative estimation of the degrees of freedom contributing to
the generation of the signal under study. The number of factors contributing to
the composition of the EEG varies with the states of sleeping and waking
(Babloyantz et al, 1985). In general, it appeares that the beta EEG of wakefulness needs more than ten degrees of freedom for its description. This number
decreases during the relaxed state accompanied by alpha activity in the EEG
to around six or eight, and further decreases during slow wave sleep (Lopes da
Silva, 1991). By applying this nonlinear analysis, it soon became clear that the
dimension of REM sleep was significantly higher than that of slow wave sleep,
implying that during REM sleep more modes are activated in the brain.
Achermann et al (1994) also showed that correlation dimensions were rather
high during REM sleep. Values of 9 were no exceptions. In short, the values of
correlation dimension closely follow the level of vigilance.
Van Hulzen and Coenen (70) registered the visual evoked potential (VEP) in
the cortical EEG of the rat during the main states of alertness (Figure 3). It is
remarkable that the VEP produced during REM sleep is almost similar to that
obtained during wakefulness, which again underlines the consonant characters of these brain states. In contrast, the VEP derived during slow wave sleep
reveals the differential nature of this state. The most striking distinction is the
shape and amplitude of the N1-P2-N3 complex. This complex is substantially
larger in amplitude during slow wave sleep, while the small N2-P3 complex on
69
Figure 3:
The influence of wakefulness (W), slow wave sleep (SWS) and REM sleep
(PS) on the shape of the averaged visual evoked potential (VEP) of a rat
(left). To the right the cortical EEGs associated with the states of alertness
are presented, together with the individual evoked potentials. Flashes are
indicated by points. Note in the VEP the large N1-P2-N3 complex of slow
wave sleep and the similarity between the VEP of wakefulness and that of
REM sleep. Note also the various background EEGs with the prominent
theta-rhythm in the lowest trace, indicating REM sleep. Negativity is upwards directed. [Adapted from Van Hulzen and Coenen, 1984].
the slope of the large P2-N3 wave, is no longer observable. The increased synchronization of thalamocortical unit discharges during slow wave sleep is expressed in an enlargement of evoked potential components. The stimulus acts
70
as a trigger pulse producing a resonance in the synchronized neural assembly.
For the reason that cells are already firing in a bursting mode during slow wave
sleep, this resonance is more prominent as that produced by the same triggering of the neural net, when neurons fire in the asynchronous, tonic mode during wakefulness and REM sleep. In the latter situations, the waves of the N1-P2N3 complex are smaller and less sharp, whereas the tiny N2-P3 complex is
manifest. The response on a flash of a neuron expressed in a poststimulus time
histogram (PSTH) shows that the discharge frequency is lower during slow
wave sleep. But what is still more relevant for the building of evoked potentials
is that the slow wave sleep PSTH is composed of sharper peaks. Both the primary and the secondary excitations are smaller and more time-locked, giving
rise to larger amplitude but narrower components in the evoked potential. A
comparison between flash responses of both kinds of neurons in the visual part
of the thalamus with a cortical VEP, reveals a striking temporal correspondence. This comparison strongly suggests that N1, the first wave of the VEP, is
comprised of the primary discharges of the ON-cells. On the other hand, P2 is
produced by the joint and simultaneously inhibitory actions of both ON- and
OFF-cells, whereas N3 is the result of the secondary firing of the ON-neurons
facilitated by the primary bursts of the OFF-cells (Figure 4) (Coenen, 1995).
Figure 4:
Responses of two
visual thalamic neurons of a cat to light
flashes, expressed in
poststimulus time
histograms (PSTHs).
In the upper traces
the PSTH-response of
both a lateral geniculate ON- and OFFneuron are shown. In
the bottom trace a
visual evoked potential (VEP) of a rat is
presented. The temporal correspondence is
discussed in the text.
[Adapted
from
Coenen, 1995].
71
Neuronal phenomena associated with consciousness: an integration
A survey of the main characteristics of the neuronal activity of the brain in
relation to the states of vigilance and consciousness is presented in Table I
(Coenen, 1998).
Table 1:
Neuronal characteristics and states of vigilance. [Adapted from Coenen,
1998].
Membrane
potential
Firing
mode
< - 60 mV
‘tonic’ or ‘relay’
0.7 - 1.0
‘oscillatory’
> -70 mV
< - 60 mV
- 60 - -70 mV
Transfer Correlation
ratio
dimension
EEG
pattern
State of
Vigilance
8 - 10
beta
wakefulness
0.5 - 0.7
6-8
alpha or spindles
drowsiness
‘burst’
0.3 - 0.5
4-6
delta
slow wave sleep
‘tonic’
?
6 - 10
beta-like
REM sleep
In this table several neuronal characteristics, such as the membrane voltages
of the neurons and associated firing modes, the transfer ratio’s and the correlation dimensions of the different EEG patterns of the main sleep-wake states
(wakefulness, drowsiness, slow wave sleep and REM sleep) are shown. It should
be stressed that data are gathered from both human and animal research and
are rather simplified. Nevertheless, the consistent correlations of parameters
characteristic for the several states of vigilance are remarkable. When the activity of the mesencephalic reticular formation drops under a critical level, the
inhibitory interneurons start to inhibit the thalamocortical neurons. These begin
to oscillate and when inhibition is still further carried upwards, resulting in low
membrane voltages, neurons start to discharge irregularly in a corresponding
burst-pause firing mode. Sensory information is largely blocked and information processing is at a low level. The burst firing mode of large groups of highly
synchronized neurons results in the typical high amplitude low frequency EEG,
associated with slow wave sleep. The number of factors involved in generating
the EEG is accordingly low. Perceptive processes do not occur and the level of
vigilance and consciousness is low. This does, however, not exclude that stimu-
72
lus evaluation is still possible to a certain degree during this brain state. On the
other end of the sleep-wake continuum there is wakefulness, with its characteristic low amplitude high frequency beta-pattern in the EEG, composed of a
relatively spontaneous activity of nerve cells with membrane voltages near the
threshold, firing in the tonic mode. This results in an easy passing of sensory
information through the thalamic relay nuclei to the sensory cortex, leading to
a high transfer ratio. The level of information processing is high and many
processes, such as perception, evaluation and recognition take place. In all,
this implies that a great number of factors play a role in the composition of the
EEG. The correlation dimension estimates the degrees of freedom to be about
ten.
A mysterious state of existence is REM sleep, not fitting into the normal continuum from deep slow wave sleep to alert wakefulness. Neuronal characteristics
show that there is a high spontaneous neuronal activity comparable to that of
wakefulness. Muscle paralysis prevents that the high brain activity is overtly
expressed. During dreaming mentation, which is associated with REM sleep, there
are perceptual phenomena mostly of an internal origin. In all, REM sleep is associated with some kind of perception and consciousness. It has to be regarded as
a third state of existence, next to wakefulness and slow wave sleep. Contrary to
these states, REM sleep is not a uniform state. Awakening thresholds and presumably transfer ratio’s heavily fluctuate, dependent on time of the night and the
stage in a particular REM phase. This might also true for correlation dimensions.
Sometimes, REM sleep is more ‘sleep-like’ and sometimes more ‘wake-like’. This
curious state is accompanied by a state of consciousness which can currently
best be regarded as a physiological form of altered consciousness.
References
Achermann, P., Hartmann, R., Gunzinger, A., Guggenbühl, W. and Borbély, A.A. (1994).
Correlation dimension of the human sleep electroencephalogram: cyclic changes in the
course of the night. European Journal of Neuroscience 6, 497-500.
Babloyantz, A., Salazar, J.M. and Nicolis, C. (1985). Evidence of chaotic dynamics of
brain activity during the sleep cycle. Physics Letters 111A, 152-156.
Chase, M.H. and Morales, F.R. (1989). The control of motoneurons during sleep. In
Kryger, M.H., Roth, T. & Dement, W.H. (Eds.), Principles and practices of sleep medicine, pp. 74-85, Philadelphia: Saunders.
Coenen, A.M.L. (1995). Neuronal activities underlying the electroencephalogram and
evoked potentials of sleeping and waking: implications for information processing. Neuroscience and Biobehavioral Reviews 19, 447-463.
73
Coenen, A.M.L. (1998). Neuronal phenomena associated with vigilance and consciousness: From cellular mechanisms to electroencephalographic patterns. Consciousness
and Cognition 7, 42-53.
Coenen, A.M.L. and Vendrik, A.J.H. (1972). Determination of the transfer ratio of cat’s
geniculate neurons through quasi-intracellular recordings and the relation with the level
of alertness. Experimental Brain Research 14, 227-242.
Glenn, L.L. and Steriade, M. (1982). Discharge rate and excitability of cortically projecting intralaminar thalamic neurons during waking and sleep states. Journal of Neuroscience 2, 1387-1404.
Hobson, J.A. and Steriade, M. (1986). The neuronal basis of behavioural state control:
internal regulatory systems in the brain. In F. Bloom (Ed.), Handbook of Physiology, vol.
4, pp 701-823, Bethesda: American Physiological Society.
Jones, B.E. (1990). Influence of the brainstem reticular formation, including intrinsic
monoaminergic and cholinergic neurons, on forebrain mechanisms of sleep and waking. In M. Mancia & G. Marini (Eds.), The Diencephalon and Sleep, pp 31-48, New York:
Raven Press.
Jouvet, M. (1967). Neurophysiology of the states of sleep. Physiological Reviews 47,
117-176.
Langford, G.W., Meddis, R. and Pearson, A.J.D. (1974). Awakening latency from sleep
for meaningful and non-meaningful stimuli. Psychophysiology 11, 1-5.
Livingstone, M.S. and Hubel, D.H. (1981). Effects of sleep and arousal on the processing
of visual information in the cat. Nature 291, 554-561.
Lopes da Silva, F. (1991). Neural mechanisms underlying brain waves: from neural
membranes to networks. Electroencephelography and Clinical Neurophysiology 79, 8193.
McCormick, D.A. and Bal, T. (1994). Sensory gating mechanisms of the thalamus. Current Opinion in Neurobiology 4, 550-556.
Siegel, J. M. (1990). Mechanisms of sleep control. Journal of Clinical Neurophysiology
7, 49-65.
74
Singer, W. (1977). Control of thalamic transmission by corticofugal and ascending
reticular pathways in the visual system. Physiological Reviews 57, 386-420.
Steriade, M. (1991). Alertness, quiet sleep, dreaming. In A. Peters (Ed.), Cerebral Cortex, Vol. 9, pp 279-357, New York: Plenum Press.
Steriade, M. and McCarley, R. W. (1990). Brainstem control of wakefulness and sleep.
New York: Plenum Press,
Steriade, M., Jones, E.G. and Llinás R.R. (1990). Thalamic oscillations and signalling.
New York: John Wiley & Sons.
Steriade, M., McCormick, D.A. and Sejnowski, T.J. (1993). Thalamocortical oscillations
in the sleeping and aroused brain. Science 262, 697-685.
Szymusiak, R. (1995). Magnocellular nuclei of the basal forebrain: substrates of sleep
and arousal regulation. Sleep 18, 478-500.
van Hulzen, Z.J.M. and Coenen, A.M.L. (1984). Photically evoked potentials in the visual
cortex following paradoxical sleep deprivation in rats. Physiology and Behavior 32, 557563.
75
76
Pathophysiology of narcolepsy
M. BILLIARD
GUI DE CHAULIAC HOSPITAL, MONTPELLIER, FRANCE
The pathophysiology of narcolepsy is still far from being totally understood.
However significant progress has been made within the last feefteen years.
In this presentation we will first focus on the dysregulation of the states of sleep
and wakefulness in narcolepsy and on the monoamine - cholinergic imbalance
at the origin of excessive daytime sleepiness and cataplexy. Then, in a second
part, we will consider the pathogeny of the condition including genetic and
environmental factors.
77
78
Laryngeal dysfunction during sleep
D. PEVERNAGIE
COORDINATOR OF THE SLEEP DISORDERS CENTRE
UNIVERSITY HOSPITAL GENT – BELGIUM
Sleep related respiratory dysfunction is most often related to increased resistance of the upper airway (UA) at the level of pharynx. The prevalence of laryngeal obstruction, as a cause of disturbed breathing during sleep, is rather exceptional compared with the frequency of snoring and obstructive sleep apnea,
but may be worse in terms of severity and prognosis. Failure to abduce the
vocal cords (i.e. opening the glottis) during inspiration is the basic pathophysiological mechanism. The clinical hallmark of this condition is “stridor”. Typically, the patient with laryngeal obstruction produces a harsh, high-pitched
inspiratory noise, that is different from but may be confused with snoring. Both
anatomic and functional factors may be responsible for the maintenance of
glottic closure during inspiration. Paroxysmal sleep-related laryngospasm is a
benign, infrequent disorder, characterised by sudden awakening with a sense
of choking. The patient is in agony because he feels that his breath has been
cut of. The respiration is very noisy. Typically, the attack subsides within seconds or minutes. In most cases no evident cause can be detected, though gastroesophageal reflux may aggravate the problem. Patients with sleep-related laryngeal stridor (SLS) suffer from severe UA obstruction every night. SLS is a
frequent complication in neurodegenerative diseases, e.g. multiple system atrophy (Shy-Dräger syndrome) and olivio-ponto-cerebellar atrophy. SLS in rheumatoid arthritis is due to crico-arythenoiditis. SLS is a potentially life-threatening condition. Tracheostomy may be required for permanent relief of the UA
obstruction.
79
80
Surgical treatment for sleep-related breathing
disorders: possibilities and limitations
BOUDEWYNS A, MD.
DEPARTMENT OF ENT, UNIVERSITY OF ANTWERP, BELGIUM.
Introduction
The physiologic spectrum of sleep-related breathing disorders (SRBD) ranges
from an increased upper airway resistance over partial airway collapse to complete upper airway obstruction. Accordingly, there is a wide spectrum of clinical manifestations: asymptomatic snoring, snoring and excessive daytime sleepiness, the upper airway resistance syndrome, sleep apnea syndrome and periodic
breathing.
Sleep-related breathing disorders are highly prevalent. Data from the Wisconsin Sleep Cohort Study indicate that 4% of men and 2% of women (30-60 years
old) meet the diagnostic criteria for sleep apnea syndrome 1. In addition to the
high prevalence, there is increasing evidence that SRBD are associated with
increased morbidity and mortality 2.
The cause of upper airway obstruction is multifactorial and the following factors contribute to the pathogenesis of this condition: alterations in ventilatory
drive, increased collapsibility of upper airway tissues, abnormalities in the
arousal response to upper airway occlusion and anatomical narrowing of the
upper airway.
A variety of treatment modalities are currently available. All patients should be
advised to avoid muscle relaxants such as benzodiazepines and alcohol and to
take care of their body weight. Positional training may be indicated for those
patients with a position dependent form of sleep apnea. The use of medication
such as ventilatory stimulants, oxygen therapy or the use of oral appliances
may be indicated in selected patients. There is no doubt that continuous positive airway pressure (CPAP) treatment is today the most widely used treatment for moderate to severe obstructive sleep apnea syndrome. However, despite the beneficial effects of CPAP, its success is hampered by poor compliance
by many patients due to side effects such as nasal problems or lack of perceived benefit. In addition, in West-European countries, CPAP reimbursement
by the social security system is only available for patients with moderate to
severe sleep apnea syndrome. This despite the fact that some patients with
more mild disease also could benefit from this kind of therapy.
Different options are also available for upper airway surgery as a treatment for
81
snoring and obstructive sleep apnea syndrome: nasal surgery; various forms of
palatal surgery, maxillo-facial surgery, tongue base surgery etc.
Instead of merely giving an overview of the surgical treatment for SRBD, I would
like to focus on some more recent insights and new surgical treatment options.
For a broader overview I would like to refer to a recent review on SRBD by Van
de Heyning et al.3.
Nasal surgery
Experimental and clinical data support the notion that nasal obstruction may
contribute to sleep-disordered breathing. Zwillich et al. introduced upper airway obstruction in normal subjects during sleep by inflating a nasal balloon 4.
Significantly more apneas, arousals and awakenings were observed during the
nasal occlusion compared to unoccluded sleep periods. Data from the Wisconsin Sleep Cohort study revealed that patients with chronic nighttime symptoms of rhinitis, were 2.1 times more likely to be habitual snorers 5.
Despite the potential role of nasal obstruction in the occurrence of snoring and
obstructive sleep apnea, nasal surgery may improve snoring in only a selected
group of patients 6 and poor results of nasal surgery with respect to the improvement of SRBD are often reported. Sériès et al. evaluated the effect of
nasal surgery (septoplasty, turbinectomy, and/or polypectomy) on the occurrence of respiratory events in 20 subjects with moderate obstructive sleep apnea
7
. The authors observed a significant improvement of nasal resistance but this
was not associated with a decrease in respiratory disturbance index (RDI) or
with an improvement in oxygen saturation or sleep quality. Inspite the lack of
improvement in SRBD for the majority of these patients, the decrease in nasal
resistance resulted in a better tolerance of CPAP in those patients who could
not tolerate this treatment because of nasal stuffiness. Similar findings were
reported by Verbraecken et al. and the authors suggested that treatment of
nasal obstruction might improve CPAP compliance 8.
Uvulopalatopharyngoplasty
Uvulopalatopharyngoplasty consists of the removal of redundant pharyngeal
tissue. UPPP enlarges the retropalatal and oropharyngeal airway through excision of the tonsils, if present; trimming of the anterior tonsillar pillars and
excision of the uvula together with the posterior portion of the soft palate. This
is combined with a suspension technique lifting the oropharyngeal palatal
mucosa. Some authors advocate reorienting the posterior tonsillar pillars. Resection or trimming of the posterior pillars is contra-indicated as it increases
the risk of nasaopharyngeal stenosis.
About 90% of patients report a marked improvement in snoring after UPPP.
However, these results should be interpreted with some caution since most
82
studies did not attempt to measure snoring objectively and a discrepancy between objective and subjective measures of snoring 9, has been frequently observed. In addition, there are yet no standardized methods available to report
snoring or to measure snoring objectively 10. When considering the results of
UPPP for the treatment of obstructive sleep apnea (OSA), the results are often
perceived to be disappointing. In unselected OSA patients, the success rate is
generally around 50%. There is increasing evidence that the site of upper airway obstruction is an important predictor of the success of UPPP in the treatment of obstructive sleep apnea 11. We investigated the effect of UPPP on sleepmicrostructure (alpha-EEG arousals) in a group of 10 nonapneic snorers 12. A
significant improvement in both snoring and the arousal index was observed.
These findings support the hypothesis that UPPP may also indicated for patients with the upper airway resistance syndrome 13 and justifies further studies to explore this hypothesis.
Radiofrequency volume reduction of the soft palate (RF palatoplasty)
The use of radiofrequency (RF) applications for both medical and surgical purposes has been explored previously in different fields of medicine such as neurology, cardiology, oncology and urology. Possible advantages of this technique
are the use of low heat energy (up to 80-90_ Celsius) and the possibility to
control lesion size by carefully selecting diameter and length of the electrode,
treatment duration and total energy delivery.
The application of RF volume reduction of the soft palate (RF-palatoplasty) has
been introduced in 1997 as an outpatient procedure to treat simple snoring or
mild OSA and the first clinical results have been published recently by Powel
et al. 14. In this feasibility study of 22 subjects, snoring score fell by a mean of
77% accompanied by a significant improvement in the Epworth sleepiness scale.
Patients had minimal pain and there were no major complications.
Last-year, a European multi-center study was set-up to investigate the effect of
RF-palatoplasty on snoring and polysomnographic variables. Over 90 subjects
participated in this trial, coordinated by the ENT department of the Antwerp
University. Outcome results of this study are expected by the end of 1998.
Functional electrical stimulation of the hypoglossal nerve
Although the mechanisms underlying upper airway collapse are incompletely
understood, this collapse has been attributed to a decline in pharyngeal neuromuscular activity during sleep. This knowledge has supported the notion
that stimulation of upper airway muscles may represent a specific approach to
the treatment of OSA. It is know widely accepted that the genioglossus muscle
is the most important upper airway dilator muscle and that this muscle dilates
the upper airway by pulling the tongue forward.
83
Consequently, methods have been explored to selectively stimulate upper airway dilator muscles, in particular the genioglossus.
In 1988 Miki et al. reported their experience with submental electrical stimulation using an apnea-demand type stimulator in OSA patients 15. An improvement in sleep-disordered breathing and sleep-architecture could be obtained.
However, these early favorable results could not be reproduced by subsequent
studies in other centers; stimulation was either inefficient to relieve upper airway obstruction or caused arousal from sleep. Smith et al. 16 outlined the different consecutive steps undertaken in the development of functional electrical stimulation of the hypglossal nerve, as we are currently using.
Since 1996, the Unit for SRBD of the Antwerp University has been involved in
a multi-center study aimed to investigate the feasibility of treating obstructive
sleep apnea by stimulating the hypoglossal nerve unilaterally. Worldwide, 8
patients are actually treated with a fully implantable system (Inspire I TM,
Medtronic Inc, Minneapolis), 3 of them at the University Hospital Antwerp. A
programming system is used by the investigators to determine and adjust the
stimulation parameters.
A significant improvement in sleep-disordered breathing can be documented
during stimulation 17. The stimulation was found not to cause any patient discomfort or arousal from sleep. Careful selection of the surgical candidates seems
however of paramount importance. Patients with very severe OSA (high upper
airway collapsibility) or with predominantly central sleep apnea are no suitable candidates for this kind of treatment. In addition, good cooperation between patient and physician is important because many patient visits are often
required to define optimal stimulation parameters.
Conclusions
We believe that nasal surgery plays an adjunctive role in treatment of sleepdisordered breathing. The major aim of nasal surgery is not the relief of snoring
or OSA but the improvement of nose breathing. By doing so, it may improve
CPAP tolerance and increase CPAP compliance.
Our results suggest that patients with socially disturbing snoring and excessive
daytime sleepiness (nonapneic snorers) are good surgical candidates for UPPP.
In this patients a reduction of snoring and an improvement in sleep microstructure, can be expected. Evidence exists that those OSA patients with
retropalatal obstruction are more likely to have a favorable outcome after UPPP.
We therefore recommend that the site of upper airway obstruction is determined during sleep in all OSA patients considered to be candidates of UPPP. In
those patients having multiple sites of upper airway obstruction, a treatment
other that UPPP or a combination of different surgical procedures should be
considered.
84
Because RF- palatoplasty can be used in an outpatient setting, under local
anesthesia, with minimal patient discomfort and allows accurate control of
lesion size, it may have substantial advantages over other forms of treatment
for snoring. The feasibility of this new treatment has been demonstrated. Studies are going on to investigate the effectivness of this treatment and to determine which patients are good candidates for this procedure.
Functional electrical stimulation of the hypoglossal nerve by the Inspire I TM
system results in an improvement of sleep-related breathing disorders in carefully selected patients. Hypoglossal nerve stimulation stimulation may be considered as an alternative to CPAP treatment in patients with dominantly obstructive sleep apnea and moderately increased upper airway collapsibility.
For all types of upper airway surgery in the treatment SRBD, a consensus should
be reached among physicians as to the definition of treatment success and
efforts should be made to develop standardized methods for the recording and
reporting of snoring. In addition, the effect of upper airway surgery on quality
of life in patients with SRBD merits further investigation.
Bibliography
1 Young TB, Palta M, Dempsey JA, Skatrud JB, Weber SA, Badr MS. The occurrence of
sleep-disordered breathing among middle-aged adults. N Engl J Med 1993;328:12301235.
2 Peter JH, Koehler U, Grote L, Podszus T. Manifestations and consequences of obstructive sleep apnoea. Eur Respir J 1995;8:1572-1583.
3 Van de Heyning PH, De Backer WA, Boudewyns A, Verbraecken J, Hamans E. Obstructive sleep apnea syndrome and snoring: A physiological approach. In: Van
Cauwenberge P, Wang DY, Ingels K, Bachert C, eds. The Nose. The Hague/The Netherlands: Kugler Publications, 1998;87-103.
4 Zwillich CW, Pickett C, Hanson FN, Weil JV. Disturbed sleep and prolonged apnea
during nasal obstruction in normal men. Am Rev Respir Dis 1981;124:158-160.
5 Young TB, Finn L, Kim H. Nasal obstruction as a risk factor for sleep-disordered
breathing. J Allergy Clin Immunol 1997;99:S757-S762.
6 Elsheuff I, Hussein S. The effect of nasal surgery on snoring. Am J Rhinol 1998;12:77-79.
7 Sériès F, St.Pierre S, Carrier G. Effects of surgical correction of nasal obstruction in
the treatment of obstructive sleep apnea. Am Rev Respir Dis 1992;146:1261-1265.
8 Verbraecken J, Willemen M, De Cock W, Wittesaele W, Van de Heyning PH, De Backer
WA. Nasal resistance and patient compliance in obstructive sleep apnoea treated
with CPAP. Eur Respir J 1995;8, suppl 19:325s
9 Hoffstein V, Mateika S, Nasch S. Comparing perceptions and measurements of snoring. Sleep 1996;19:783-789.
10 Young TB. Some methodologic and practical issues of reported snoring validity. Chest
85
1991;99:531-532.
11 Sher AE, Schechtman KB, Piccirillo JF. The efficacy of surgical modifications of the
upper airway in adults with obstructive sleep apnea syndrome. Sleep 1996;19:156177.
12 Boudewyns A, De Cock W, Willemen M, Wagemans M, De Backer WA, Van de Heyning
PH. Influence of uvulopalatopharyngoplasty on alpha-EEG arousals in nonapnoeic
snorers. Eur Respir J 1997;10:129-132.
13 De Backer WA, Van de Heyning PH. Is the role of UPPP in nonapnoeic snorers
understimated? Eur Respir J 1994;7:843-844.
14 Powel NB, Riley RW, Troell R, Li KK, Blumen MB, Guilleminault C. Radiofrequency
volumetric tissue reduction of the palate in subjects with sleep-disordered breathing. Chest 1998;113:1163-1174.
15 Miki H, Hida W, Inoue H, Takishima T. A new treatment for obstructive sleep apnea
syndrome by electrical stimulation of submental region. Tokohu J exp Med
1988;154:91-92.
16 Smith PL, Eisele DW, Podszus T, et al. Electrical stimulation of upper airway musculature. Sleep 1996;19:S284-S287.
17 De Backer WA, Boudewyns A, Van de Heyning PH. Hypoglossal nerve (HGN) stimulation using a fully implantable pulse generator. Clinical experience in 2 OSA patients.
Am J Respir Crit Care Med 1998;157:A284
86
Regular Papers
87
88
Arousal detection in sleep
FW BES, H KUYKENS AND A KUMAR
MEDCARE AUTOMATION, OTTHO HELDRINGSTRAAT 27
1066XT AMSTERDAM, THE NETHERLANDS
Introduction
Arousals are part of normal sleep. They become pathological if the frequency
of occurrence passes a limit, beyond which the normal, dynamic course of the
sleep process is disturbed. This results in the subjective experience of shallow
sleep and daytime fatigue or somnolence, often accompanied by complaints
about insufficient proper daytime functioning. Next to spontaneously occurring, arousals very often are caused by an underlying pathology like e.g. the
obstructive sleep apnea syndrome (OSAS) or periodic limb movement syndrome (PLMS). In 1992, the ASDA made proposals about criteria how to score
arousals in NREM and REM sleep1. Using these guidelines, we developed a computerized method for scoring and evaluation of arousals in sleep. Given a logical order and an appropriate time frame, the algorithm also may link detected
arousals with possible underlying pathological events like desaturations, snores,
apneas/hypopneas or limb movements.
Methods
In short, the EEG arousal detection is based on a weighted score from the
evaluation, in one channel of EEG, of
a) EEG frequency shifts, with zero-crossing,
b) the ratio slow vs. fast EEG activity, with the Alpha-Slow-wave Index (ASI)2 and
c) mean frequency, with the Hjorth-parameter Mobility.
During REM-sleep, user-defined detectable changes in chin EMG level are also
considered. We used the arousal detection method in combination with
REMbrandt, a computerized system for polysomnography and evaluation of
sleep recordings. We processed polysomnographic data of 5 full-night recordings (from three female patients A, B and C, resp. 75, 35 and 24 yrs, and from
one healthy male D, 49 yrs; patient A was recorded twice, the others for one
night). All recordings had been visually scored on sleep stages by two experienced raters, as well as on the occurrence of arousals according to the ASDA
scoring criteria. To facilitate the visual arousal scoring, the raters supportively
used EEG spectral analysis from the evaluation module of the REMbrandt system. Also, apnea- and PLM detection and labelling was made with REMbrandt.
In addition, the automatic arousal analysis was performed on the same EEG
channels used for visual evaluation. A time window of 6 s was chosen as de-
89
fault, to make a causal link between detected arousals and apnea/hypopnea- or
PLM events, i.e. a link was made, if the start of the arousal overlapped with the
event or occurred maximally 6 s after the end of the event.
90
Results and discussion
The figure depicts a typical example of an arousal event that was linked, in this
case, with a leg movement. The pane labeled ‘Signal window’ displays 30 seconds of the sleep recording and shows the following traces, from top to bottom:
a) the C3-A2 EEG; this signal was used for the arousal evaluation, b) the arousal,
as detected by the algorithm and automatically labeled ‘LM’, to indicate the
causal relationship with a leg movement, c) the arousal, as indicated by the
human rater, d) the C4-A1 EEG, e) EOG left, f) EOG right, g) leg movement
signal, obtained from piezo sensors on both legs, h) the leg movement, as detected automatically by the REMbrandt system and indicated with ‘LM-B’ (the
‘B’ stands for ‘both legs’), i) the EMG from the chin, f) the EKG. The pane
labeled ‘Overview window’ displays from top to bottom respectively, a) the visually rated hypnogram with time axis, b) the automatically detected arousals in
the whole recording and c) the visually rated arousal events. The arousals under b) and c) are both indicated with a maximum of 2 events per epoch.
If there was any overlap of a visually evaluated arousal event with an automatically detected one, the detection was rated as ‘concordant’. A visually scored
arousal without automatical detection was was rated as ‘negative’ and an automatically detected arousal without visual indication was rated as ‘false positive’. Comparison of the automatically detected arousals with the visually scored
ones gave the following results.
Recording
Pt. A, night 1
Pt. A, night 2
Pt. B, night 1
Pt. C, night 1
Pt. D, night 1
Concordance of
visually det. with
automatically det.
(aut.det.=100%)
90%
88%
92%
92%
94%
False
positives
10%
12%
8%
8%
6%
Concordance of
automatically det.
with visually det.
(vis.det.=100%)
78%
81%
76%
81%
74%
Negatives
22%
19%
24%
19%
26%
The EEGs of the subjects showed age dependent differences, as well as specific
individual traits. The present results were however obtained with similar, default detection settings. An optimalization of the detected parameters with respect to every individual night recording (a user-controlled option, integrated
in the arousal detection method) will therefore certainly increase the quality of
detection.
91
1
2
ASDA Task force. ASDA report. EEG arousals: Scoring rules and examples. Sleep
1992, 15:173-184.
Jobert M, Schulz H, Jähnig P, Tismer C, Bes F and Escola H. A computerised method
for detecting episodes of wakefulness during sleep based on the Alpha Slow-Wave
Index (ASI). Sleep 1994; 17:37-46.
92
Sleep monitoring equipment affects the
assessment of nocturnal oxygenation
in patients with COPD
BRIJKER F1, VAN DEN ELSHOUT FJJ1, HEIJDRA YF2, FOLGERING HTHM2
DEPARTMENT OF PULMONOLOGY, RIJNSTATE HOSPITAL ARNHEM1 AND
DEKKERSWALD, UNIVERSITY OF NIJMEGEN2, THE NETHERLANDS.
Introduction
Patients with Chronic Obstructive Pulmonary Disease (COPD) run a risk of
developing nocturnal oxygen desaturations, predominantly during rapid eye
movement (REM) sleep (1-4). When measuring nocturnal oxygen saturation,
the monitoring conditions may cause a less efficient sleep. Studies on
polysomnography (PSG) show a delayed sleep onset, sleep fragmentation, frequent arousals and shortened periods of REM sleep as compared to a common
sleep (5-8). When sleep is disturbed, less oxygen desaturations may occur in
patients with COPD. It was hypothesised that, as a result, an overestimation of
nocturnal oxygenation will be found. The aim of this trial was to determine the
impact of the monitoring equipment and of the unfamiliar hospital environment on the assessment of nocturnal oxygen saturation in COPD.
Methods
The study was performed in patients with COPD, with a stable disease and with
a daytime arterial oxygen tension (PaO2) < 10 kPa. COPD was defined according to the standards of the American Thoracic Society (9). Stability of the disease was defined as a fluctuation in FEV1 < 10% in the last 3 months and an
absence of an exacerbation in the last 8 weeks preceding the study. Subjects
were considered suitable for evaluation if the mean nocturnal SaO2 during
oxymetry at home was below 92%. Subjects with a history of a sleep apneas
were excluded.
Sixty subjects were screened during the first night of the study: oximetry at
home. This showed a mean nocturnal SaO2 < 92% in 17 subjects. These subjects were then invited for the second night: PSG in the hospital sleep laboratory. Three subjects had to be excluded at this phase of the study because sleep
apneas were found. The remaining 14 subjects completed the study with the
third night: PSG at home. The latter population consisted of 7 males and 7
females, with a mean (± SD) age of 70 (7) years and a mean (± SD) FEV1 of 39
93
(17) % of predicted. The study protocol was approved by the hospital ethical
committee. An informed consent was obtained from all subjects.
Oximetry was performed with the same portable pulse oximeter (Nonin 8500M)
during the different nights. SaO2 values and the heart rate were stored in the
memory. Data were processed by the PROFOX software (PROFOX associates,
inc. Version NN856 03/95).
PSG, performed in hospital (Nicolet Voyageur), included EEG, EOG, EMG,
thoraco-abdominal movements, airflow thermistor registration, leg movements
and simultaneous oximetry.
PSG at home (Oxford Medical 9000) included EEG, EMG, EOG and simultaneous oximetry. Sleep states were assessed manually, according the guidelines of
Rechtschaffen and Kales (10).
For statistical analysis, the SAS package (SAS Institute Inc., Cary, North Carolina, USA) was used. Differences between dependent variables were analysed
by the Wilcoxon signed rank test. Correlation analysis was performed by
Spearman correlation. Results all expressed as mean (± SD).
94
Results
Oximetry at home versus polysomnography at home
The mean nocturnal oxygen saturation (NSaO2) was lower during oxymetry at
home than during PSG at home; The mean NSaO2 was 87,2 (3) % during
oxymetry versus 88.4 (4) % during PSG (p<0.05). This was associated with a
higher fraction of time spent at SaO2<90% and a higher fraction of desaturation
time (table 1).
Table 1:
NSaO2 during different monitoring conditions.
NSaO2 variables
NSaO2, recorded time
Mean SaO2 %
SaO2<90 %
Desaturation time %
Baseline awake (%)
Lowest value %
NSaO2, actual sleep
Mean SaO2 %
SaO2< 90 %
Desaturation time %
NSaO2, REM sleep
Mean SaO2 %
SaO2<90 %
Desaturation time %
Oxymetry
PSG at home
PSG in hospital
87,2 (3)
62,7 (29)
49,2 (28)
92,2 (2)
71,6 (12)
88,4 (4)*
47,6 (35)**
25,2 (28)**
91,4 (3)
71,6 (10)
88,9 (4) *
47,7 (38)
22,6 (23)**
92,0 (3)
72,6 (12)
87,6 (5)
52,9 (35)
31,4 (25)
88,1 (4)
59,8 (37)
30,2 (26)
85,4 (6)
66,9 (29)
48,7 (35)
85,9 (4)
75,6 (26)
70,8 (29)†
Results represent mean (± SD). PSG: polysomnography.
Statistics: oxymetry versus PSG: * p<0.05, ** p<0.01, PSG at home versus PSG in hospital: † p<0.05.
Polysomnography at home verus polysomnography in hospital
No significant difference was found between the mean NSaO2 during PSG at
home and PSG in hospital (table 1).
The fraction of REM sleep was higher at home than in hospital: 17.0% versus
12.7% (p<0.05). The other variables of sleep architecture were not significantly
different (table 2).
95
Table 2:
Sleep architecture during polysomnography
Sleep architecture
Recorded time, hrs
Total sleep time, hrs
Sleep latency, min
Sleep efficiency, %
REM efficiency, %
PSG at home
7,3 (0,8)
5,6 (1,3)
32 (29)
69,4 (15)
17,0 (6)
PSG in hospital
7,0 (1,4)
5,8 (1,7)
49 (35)
69,9 (15)
12,7 (5)†
Results represent mean (± SD). PSG: polysomnography.
Statistics: † p<0.05.
Discussion
The purpose of this study was to determine the impact of the sleep monitoring
conditions on the validity of the assessment of nocturnal SaO2 in patients with
COPD. NSaO2 was lower during oximetry than during polysomnography. The
difference in the mean NSaO2 of 1.2 % was accompanied by a larger difference
in hypoxic time (15.1%) and a larger difference in desaturation time (24%).
Hypoxic time was defined as the time with NSaO2 values lower than 90% and
the desaturation time was defined as the time with NSaO2 values 4% lower than
the baseline in the wake state. It was also hypothesised that hospital environmental factors affect the assessment of NSaO2. A lower fraction of REM sleep
was demonstrated in hospital than at home, but no differences in NSaO2 were
found. The variation in sleep efficiency between both polysomnographic nights
was apparently too small to cause significant differences in the NSaO2. It seems
that unfamiliar environment has substantial less confounding effect on the assessment of NSaO2 than monitoring equipment.
We believe that screening of nocturnal oxygenation in patients with COPD can
best be performed by oximetry. When NSaO 2 is measured during
polysomnography, one needs to considered that the NSaO2 values will probably
be higher due to abnormal sleep architecture. Less efficient sleep can affect the
assessment of NSaO2. Consequently, the severity of nocturnal hypoxaemia may
be masked.
Acknowledgement: This study was supported by GlaxoWellcomme, The Netherlands
96
References
1 Fletcher E.C., D. Scott, W. Qian, R.A. Luckett, C.C. Miller, and S. Goodnight-White.
1991. Evolution of nocturnal oxyhemoglobin desaturation in patients with chronic
obstructie pulmonary disease ans a daytime PaO2 above 60 mm Hg. Am. Rev. Respir.
Dis. 144: 401-405.
2 Stradling J.R., and D.J. Lane. 1982. Nocturnal hypoxaemia in chronic obstructive
pulmonary disease. Clin. Sci. 64: 213-222.
3 Fletcher E.C., B.A. Gray, and D.C. Levin. 1983. Non-apneic mechanisms of arterial
oxygen desaturation during rapid-eye movement sleep. J. Appl. Physiol. 54: 632639.
4 Hudgel D.W., R.J. Martin, M. Capehart, B. Johnson, and P. Hill. 1983. Contribution of
hypoventilation to sleep oxygen desaturation in chronic obstructive pulmonary disease. J. Appl. Physiol. 55: 669-67.
5 Agnew H.W., W.B. Webb, and R.L. Williams. 1966. The first night effect: an EEG
study of sleep. Psychophysiology 2: 263-266.
6 Mendels J., and D.R. Hawkins. Sleep laboratory adaptation in normal subjects and
depressed patients (“first night effect”). Electroencephalog Clin Neurophysiol 1967;
22: 556-558.
7 Coble P., R.J. McPartland, W.J. Sila, and D.J. Kupfer. 1974. Is there a first night effect
(a revisit). Biol. Psychiatry. 9: 215-219.
8 Aber W.R., A.J. Block, D.W. Hellard, and W.B. Webb. 1989. Consistency of respiratory measurements from night to night during the sleep of elderly men. Chest 96:
747-751.
9 American Thoracic Society. 1987. Standards for the diagnosis and care of patients
with chronic obstructive pulmonary disease (COPD) and asthma. Am Rev Respir Dis
136 225-44:
10 Rechtschaffen A., and Kales A. A manual of standardized terminology, techniques
and scoring system for sleep stages of human subjects. 1968. Los Angeles: UCLA,
Brain information service / Brain Research Institute.
97
98
Similarities between deep slow wave sleep
and absence epilepsy
A.M.L. COENEN
NICI, DEPARTMENT OF PSYCHOLOGY
UNIVERSITY OF NIJMEGEN
P.O. BOX 9104
6500 HE NIJMEGEN
THE NETHERLANDS
Prologue
Deep slow wave sleep is a normal physiological phenomenon, while absence
epilepsy is regarded as a brain disturbance. Nevertheless, both phenomena
share common characteristics and the decrease in consciousness is most striking. In this paper the similarities between the two brain states are shown,
whereby data of humans as well as rats are taken into consideration.
The lowering of consciousness
Consciousness is decreased during sleep. This is expressed by a lack of a response towards an external stimulus. This does not exclude that sleep can be
terminated by stimuli and two factors play a role in awakening: the strength of
the stimulus and its relevance. While waking threshold for auditory stimuli is
dependent on the stage of sleep, awakening starts by stimuli of about 57 till 60
dB. However, from classical experiments it is well-known that the threshold for
awakening can be much lower for a relevant stimulus than for a physical identical but neutral stimulus. Obviously, evaluation of incoming stimuli still takes
place during sleep and the putative relevance can even be established. This
common sense phenomenon was investigated by Westerhuis et al (1996) by
measuring waking thresholds in rats. They compared the arousal reaction to
relevant and irrelevant stimuli during slow wave sleep, during the presence of a
spike-wave discharge (the electrophysiological manifestation of a seizure characterising generalised absence epilepsy) and during REM sleep. They found
that rats were more often aroused after the presentation of a conditioned stimulus than after an unconditioned stimulus, irrespective of the behavioural condition. This suggests that the information flow through the brain is never stopped,
even not during the highly synchronised epileptoformic activity. Westerhuis et
al (1996) also noticed a lower awakening threshold for spike-wave discharges
99
than for sleep (taken light and deep slow wave sleep together), suggesting that
there are gradients in what the brain is able to do in the various states.
In concordance with sleep it is shown that during the occurrence of a spikewave discharge the evaluation of incoming stimuli is still possible. When physically identical stimuli with differential impact, induced by previous conditioning, are offered during the presence of a spike-wave discharge, the relevant
stimulus terminates this activity more frequently than the less relevant one.
This not only means that during this aberrant state, just as during sleep, some
sensory information reaches the sensory cortex, but also that the brain still
has the ability to evaluate this activity. This despite the lowering in consciousness, as indicated by the fact that no abortion of a spike-wave discharge occurs
on a neutral stimulus, unless this stimulus is powerful enough (Drinkenburg et
al, 1996).
Time estimation
The general feeling of the course of time during sleep is that the time stands
still. Brief lapses in time also occur during absences in which the patient is
unable to maintain contact with the environment. When a patient is counting
this counting is stopped when an attack begins, but continues correctly when
the absence is finished. To study this in more detail an experiment was performed with children suffering from absence epilepsy and the same experiment was achieved in epileptic WAG/Rij rats. All these rats show abundantly
spike-wave discharges of the absence type. Children were asked to press a button when they thought that a fixed period of time had elapsed. The performance of the subjects in trials with and without spike-wave discharges was compared. Brief discharges, shorter than 3 seconds, prolonged the duration of the
estimated time period, while longer ones reduced its duration. The prolongation was longer than could be anticipated from the duration of the spike-wave
discharge. Moreover, it seemed that after long discharges patients behave differently and were perhaps more severely disturbed. The duration of misperception of the interval after the short discharges was larger than could be predicted from the duration of the paroxysms, which seem to contrast with the
view that cognitive impairments are only limited to the period of the epileptic
phenomena. Correspondingly, it was studied in epileptic rats whether differences in timing could be noticed in the estimation of a fixed time period with
and without spike-wave discharges. Rats were extensively trained to press a
lever for food in a fixed interval task until a stable response pattern emerged.
The time the rats waited till they made a response, the post-reinforcement
pause, lasting about half of the interval. During the task, the post-reinforcement pause was significantly enhanced in trials with spike-wave discharges
compared to trials without discharges, indicating a clear change in perform-
100
ance. The difference in pause was about 15 seconds, while the absence lasted
about 5 seconds. Again it was found that the duration of the misperception was
longer than the duration of the EEG paroxysm. This result shows that the rat
does not count the time of the spike-wave, and even more, indeed suggests that
the time stands still during an absence. This is fully comparable to that what
happens in epileptic children when they have short lasting absences (van
Luijtelaar et al, 1991a, 1991b).
Underlying neuronal activities
The EEG of deep slow wave sleep in humans is characterised by large amplitude, low frequency, irregular (delta) waves, while waking is associated with
high frequency, low amplitude (beta) waves. This is also true for rats. The ‘tonic’
or ‘relay’ firing mode of wakefulness explains the EEG characteristics during
waking. Numerous thalamo-cortical neurones spontaneously show a high neuronal firing during waking. This is caused by depolarised cells which polarisation voltage is near the threshold. This also implies that information can easily
pass the thalamic nuclei on the way to the higher cortical areas. The high
spontaneous activity together with a low synchronisation between thalamic
and cortical cells causes the low amplitude, high frequency waves. How different is the situation during slow wave sleep! Thalamo-cortical cells are than
firing in the so-called ‘burst’ mode. Extended populations of neurones fire rather
synchronously with short bursts, interspersed with prolonged pauses. This firing mode is the substrate of the high voltage, low frequency and irregular waves
characterising deep slow wave sleep. During this firing mode sensory information from the sensory receptors in the periphery through the thalamus on the
way to the cortex is largely blocked and does not reach the higher perceptual
cortical centres. This process is called ‘sensory gating’ (Coenen, 1995, 1998).
The EEG associated with human generalised absence epilepsy also shows large
amplitude waves, but this activity is quite regular with 3 Hz spike-wave activity. In rats also spike-wave activity is present during an attack of absence epilepsy. Instead of 3 Hz the frequency in rats is fluctuating around 8 Hz. This
difference is considered as a not well understandable species difference.
Single unit measurements of cells in thalamus and cortex reveal that these
neurones are firing during absences in the burst mode, with short bursts underlying the spikes and the pauses underlying the waves (Inoue et al, 1993).
There is a heavy synchronisation between cells, causing the sharp spikes. This
implies again that the neuronal characteristics of slow wave sleep and absence
epilepsy are rather identical, although the amount of regularity and synchronisation is different. The bursting mode of firing during an absence also means
that the incoming afferent information is blocked to a certain degree. The fact,
however, that a relevant stimulus can easily abort a spike-wave discharge im-
101
plies that also during this aberrant brain activity incoming information is, just
as during sleep, evaluated.
Figure
An evoked potential study to the characteristics of the brain states was performed in WAG/Rij rats. Visual evoked potentials (VEP) were constructed during active and passive wakefulness, during deep slow wave sleep and REM sleep
and during the occurrence of spike-wave discharges (see figure) (Meeren et al,
1998). In general two types of evoked potentials were recorded. The first type
was seen during active wakefulness and REM sleep, while the second type was
recorded during deep slow wave sleep and spike-wave discharges. The VEP of
passive wakefulness takes a middle position. VEP’s recorded during active wakefulness and REM sleep are smaller in amplitude and are more detailed and
complex, compared to those registered during deep slow wave sleep and spikewave discharges. In the latter VEP’s larger components can be seen and the
shape is less detailed. Obviously this is related to the electrophysiological state
of the brain. Active during alertness and REM sleep, and less active but bursting during the two other states. This finding again underlines the comparable
aspects of neuronal activities during slow wave sleep and absence epilepsy.
Epilogue
From neuropsychological studies it is concluded that the physiological phenomenon of deep slow wave sleep shares common characteristics with the aberrant state of absence epilepsy. Most pronounced is a relatively comparable
lowering in consciousness. This is expressed in an unresponsiveness to external stimuli, but both states can be terminated by strong stimuli. Another correspondence is the fact that unconscious stimulus evaluation is still possible.
Relevant stimuli can terminate both slow wave sleep and absence attacks much
102
easier than neutral stimuli. Thus, the process of sensory gating is still intact
during absences. All phenomena can be related to the underlying neuronal
mechanisms. In both the sleep and the absence state neurones are firing in the
‘burst’ mode. Single unit studies in rats as well as evoked potential studies
reveal this firing mode. A difference is the regular and spiky character of the
spike-wave discharges, contrasting the large irregular sleep waves. This is due
to the even stronger burst mode during absences. Perhaps this can also explain
the small differences in behavioural aspects. During absences small muscle
twitches of eyes and extremities can often be seen, which is not the case during
slow wave sleep. To uncover this difference is a matter of future research.
References
Coenen, A.M.L.: Neuronal activities underlying the electroencephalogram and evoked
potentials of sleeping and waking: implications for information processing. Neuroscience
and Biobehavioral Reviews 19:447-463, 1995
Coenen, A.M.L.: Neuronal phenomena associated with vigilance and consciousness: from
cellular mechanisms to electroencephalographic patterns. Consciousness and Cognition 7: 42-53, 1998
Drinkenburg, W.H.I.M., van Luitelaar, E.L.J.M., Coenen, A.M.L.: Information processing
during aberrant electroencephalographic activity in rats: motor responses and stimulus
evaluation. In: Bonke, B., Bovill, J.G., Moerman, N., eds. Memory and Awareness in
Anaesthesia III. Van Gorcum: Assen, 186-379, 1996
Inoue, M., Duysens, J., Vossen, J.M.H., Coenen, A.M.L.: Thalamic multiple-unit activity
underlying spike-wave discharges in anesthetized rats. Brain Research 612: 35-40, 1993
van Luijtelaar, E.L.J.M., de Bruijn, S.F.T.M., Declerck, A.C., Renier, W.O., Vossen, J.M.H.,
Coenen, A.M.L.: Disturbances in time estimation during absence seizures in children.
Epilepsy Research 9: 148-153, 1991a
van Luijtelaar, E.L.J.M., van der Werf, S.J., Vossen, J.M.H., Coenen, A.M.L.: Arousal,
performance and absence seizures in rats. Electroencephalography and clinical Neurophysiology 79: 430-434, 1991b
Westerhuis, M.F., Coenen, A.M.L., van Luijtelaar, E.L.J.M.: Tone discrimination during
sleep-wake states and spike-wave discharges in rats. In: Sleep-Wake Research in The
Netherlands, ed. D.G.M. Beersma, vol. 7, 155-159, 1996
103
104
Functional assessment and treatment of
sleeping problems in developmentally disabled
children: case studies
R. DIDDEN (1) AND L.M.G. CURFS (2)
(1) DEPT. OF SPECIAL EDUCATION, UNIVERSITY OF NIJMEGEN; (2) CLINICAL GENETICS CENTER, MAASTRICHT UNIVERSITY, THE NETHERLANDS.
Sleeping problems of children with developmental disabilities are highly prevalent as well as persistent. Such problems can be distinghuished as settling difficulties (e.g., prolonged sleep latencies, disruptive behaviors at bedtime, cosleeping with parents) and waking up frequently during the night or early
morning and disturb the parents or siblings (e.g., calling out, crying, co-sleeping). Recent prevalence studies show that the prevalence of sleeping problems
among children with biologically determined intellectually handicapping disorders such as Prader-Willi syndrome, Angelman syndrome, Smith-Magenis
syndrome, Rett syndrome or autism, is particularly high compared to children
without such syndromes. The mechanisms causing phenotypic features like
sleeping problems remain for most of the syndromes largely unknown.
Disruptive sleep patterns are of considerable concern for several reasons, the
first being the adverse effects on the child. For example, results from a small
number of studies indicate that there is a strong correlation between chronic
sleeping problems and daytime behavioral difficulties. Next to this, chronic
and severe sleeping problems may place enormous stress on the family.
There are several factors that may be related to the emergence and maintenance of sleeping problems in our target group. Medical factors encompass
apnea (of central origin or caused by upper airway obstruction), nighttime seizure activity, pain as a result of e.g., otitis media and/or discomfort due to e.g.,
exzema, asthma or nocturnal enuresis/encopresis. A disordered sleep-waking
cycle may be due to visual impairments. Psychological factors include inconsistent bedtime rules of parents and positive reinforcement of abnormal sleeping behavior by parental attention. Sleeping problems may also be related to
traumatic events experienced by the child. Finally, factors pertaining to the
physical environment may be related to the level of light and noise and opportunities of toy play.
Only a few studies have documented the effectiveness of treatment of chronic
sleeping problems of our target group. A number of treatment modalities have
been developed and validated. Behavioral procedures encompass (graduated)
105
extinction, stimulus control, desensitization (fading), and bedtime fading with
response cost. Medical treatment include pharmacological approaches (e.g.,
melatonin, anti-epileptics) and surgery (e.g., tonsillectomy). Recently a procedure called chronotherapy has been described in order to adjust a disordered
sleep-waking cycle.
Controlled case studies
In our ongoing research project up to now 13 children with syndromal abnormalities and/or multiple handicaps who show also chronic and severe sleeping
problems have participated (see References). The results of the treatment with
8 children have been published. Shortterm implementation of behavioral procedures such as extinction and desensitization have been found to be highly
effective in establishing a normalized sleep pattern in each child.
One child was a six-year-old girl with Wolf-Hirschhorn syndrome (see Curfs et
al., in press). She was severely mentally handicapped and had a seizure disorder, which was controlled by anticonvulsive medication. She was referred to
the Clinical Genetics Center of Maastricht for analysis and treatment of her
longstanding sleeping problems. Since about one year she showed problems in
settling as well as frequent nighttime wakings. Her mother would take her out
of bed and she was then allowed to ly on a bench in the living room where her
mother provided favorite toys and allowed her to watch a videotape. This had a
calming effect upon the child and mother would take her to bed again. Usually,
she then started to cry again. She showed no signs of fatigue during the day.
The process of analysis and treatment encompass a number of distinctive steps.
First, interviews are conducted with the child’s parents during which information is obtained on medical history, emergence of sleep problems, type and
duration of sleep problems as well as antecedents and consequences of the
target behaviors. Subsequently, parents are asked to registrate the number of
minutes of target behaviors during about seven nights and to describe the circumstances under which the target behaviors occur. During baseline the parents are instructed to continue their usual techniques. Following the baseline
phase, hypotheses are generated about the occurrence and maintenance of the
sleeping problems and the parents are informed about the results of the functional assessment. Finally, a treatment approach that is based on the hypothesis is presented and parents are asked to carry out the treatment in the home
setting. The data are being collected until the end of the follow-up phase.
In our case example the results of the functional assessment suggested that the
girl’s nighttime disruptive behaviors were reinforced by parental attention. No
medical factors, such as eczema, nighttime seizures, or pain were found to be
involved. The role of reinforcement in maintaining the target behaviors and
the importance of achieving stimulus control with a bedtime ritual was ex-
106
plained to the mother. An extinction procedure was chosen and the principles
as well as possible side-effects of this treatment option were explained to the
mother who would carry out the treatment. The mean duration of disruptive
behaviors during baseline was 166 minutes and the treatment resulted in a
decrease of the number of minutes of disruptive behavior to zero after 35 nights
of treatment. The therapeutic effects were maintained during follow-up and
thereafter.
Conclusion
Children with developmental disabilities do not simply outgrow chronic sleeping problems without treatment. It is our experience that a considerable number
of families do not receive appropriate professional help. In our ongoing project,
behavioral and/or pharmacological treatment is based upon the results of functional assessment and the (side)effects of treatment are continuously evaluated. Treatment is carried out by parents in the natural setting of the child
(mediation therapy). Functional assessment and shorttime treatment may result in a normalized sleep pattern as well as remediation of parental stress. The
childrens’ parents in our ongoing research project found the behavioral treatment approach to be safe, very helpful and acceptable.
References Research Project
Curfs, L.M.G., Didden, R., Sikkema, S.P.E., & De Die-Smulders, C.E.M. (1999). Management of sleeping problems in Wolf-Hirschhorn syndrome: A case study. Genetic
Counseling, in press.
Didden, R., Curfs, L.M.G., Van Driel, S., & De Moor, J. Extinction effectiveness in the
treatment of sleep problems with developmentally disabled children. Manuscript in
progress.
Didden, R., Curfs, L.M.G., Sikkema, S.P.E., & De Moor, J. (1998). Functional assessment
and treatment of sleeping problems with developmentally disabled children: Six case
studies. Journal of Behavior Therapy and Experimental Psychiatry, 29, 85-97.
Didden, R., De Moor, J., & Wichink Kruit, I. (1999). The effects of extinction in the
treatment of sleep problems with a child with a physical disability. International Journal of Disability, Development, and Education, 46, 247-252.
107
108
Diurnal characteristics of coagulation and
fibrinolysis in exhausted subjects
R. VAN DIEST, PHD. & K. HAMULYÁK, MD. PHD., CARIM,
MAASTRICHT UNIVERSITY, THE NETHERLANDS.
Introduction
Increased activation of coagulation and decreased fibrinolysis have been forwarded to underly the association between acute myocardial infarction (MI)
and chronic psychological stress [1]. Chronic stress ultimately leads to exhaustion [2], a state characterized by fatigue, irritability and poor morale. To
describe this fatigue state, the concept Vital Exhaustion (VE) has been developed. VE has been shown to be an independent risk indicator of first MI, and of
new cardiac events after coronary angioplasty [3-4]. Impaired sleep and life
stressors are important factors in the etiology of VE [5,6], whereas extent of
coronary atherosclerosis and left ventricular impairment are only marginally
associated with VE [7]. Recent studies have shown that VE is indeed associated with decreased fibrinolysis [8,9], and possibly with an increased coagulation[9]. Normally, these two mechanisms are in balance, such that acute forms
of stress usually result in an increased coagulation and increased fibrinolysis
[10]. The net effect is an unchanged balance that may be considered to reflect
a healthy, adaptive response. A decreased fibrinolysis and an increased coagulation in VE, however, suggest an imbalance that may enhance thrombus formation, thus promoting the risk of MI [10]. Unfortunately, the evidence is based
upon a limited assessment of coagulation, and circadian influences were not
included. Various measures of coagulation and fibrinolysis, however, exhibit
circadian fluctuations that may contribute to the early morning increase of MI
[11]. Because the interplay between VE and circadian variation in hemostasis
has not yet been investigated, this study tested the hypothesis that VE is characterized by increased coagulation, and decreased fibrinolysis, and examined
whether the association between VE and hemostasis is dependent upon the
time of day.
Methods
Participants
Questionnaires to screen for VE [12], and current disease status, medication,
smoking, and alcohol/coffee were available of 577 males (40-65 yr). Due to
various diseases and/or current smoking, 313 subjects were excluded. From
the remaining pool, 87 subjects were interviewed to evaluate frequency, sever-
109
ity, and duration of VE [13]. The interview further included the assessment of
mood disorders [14], sleep problems [15], and weekly physical exercise.
Experimental procedure
Subjects slept 2 consecutive nights at the hospital, night 1 being an adaptation
night. Pulse/blood pressure were measured on day 2 after 15 min. of bed rest
and venipuncture 1 was performed at 18.00h. After an overnight fast, venipuncture 2 was performed at 7.00h.
Laboratory methods
Coagulation measures were prothrombin fragment 1+2 (F1+2), thrombin-antithrombin complexes (TAT), activated factor VII (VII:a), factor VII and VIII coagulant activity (VII:c, VIII:c), von Willebrand factor (vWf), and fibrinogen
(FNG). Fibrinolytic measures were tissue plasminogen activator (tPA), plasminogen activator inhibitor (PAI-1), and tPA-PAI complexes (tPA-PAI). The
complete blood count was also measured at 18.00h and 7.00h, while control
blood chemistry was measured at 7.00h.
Statistical analysis
Skewed distributions were log-normalized. Group differences were analysed
with t- tests, or 2 (VE/controls) by 2 (18.00/7.00) MANOVA’s of repeated measures. Significance levels were based on two-tailed tests, with α ≤ .05.
Results
22 subjects were excluded because they did not meet VE criteria (N=19), or
because they met criteria for major depression or other current mood disorders (N=3). None of the control subjects failed to meet selection criteria. Five
selected subjects were unable to participate on the scheduled dates, and 1 subject was excluded post-hoc due to hypercholesterolemia. The final sample consisted of 29 VE and 30 control subjects, all nonsmokers and apparently healthy.
Selected characteristics are shown in Table 1. The two groups were similar in
age, body mass index, alcohol, and physical exercise (data not shown). Pulse,
blood pressure, blood count and blood chemistry (data not shown) were also
similar. All these measures were within their normal ranges. VE subjects used
significantly more coffee, reported a significantly shorter habitual sleep duration, a worse sleep quality, and significantly more sleep problems. Significant
time effects were found in F1+2, TAT and VII:a, and in all fibrinolytic measures
(data not shown). Significant group*time interactions were found in fibrinolytic measures, but not in coagulation measures (data not shown). The significant interactions were further examined with post-hoc t-tests. Finally, significantly higher levels of F1+2 and FNG were found in VE both at 18.00h and at
110
Table 1:
Selected characteristics of exhausted and control subjects
Characteristic
Age (years)
Body mass index (kg/m2)
Alcohol (N/day)
Coffee (N/day)
Habitual sleep (hours)
Habitual sleep quality
Problems Sleep onset (%)
Sleep maintenance (%)
Early morning awakening (%)
Difficulty waking up (%)
Fatigue upon awakening (%)
Daytime sleepiness (%)
VE
Mean
51.1
26.3
1.2
7.1
6.3
5.5
17.2
37.9
31
10.3
69
24.1
(SD)
(4.5)
(2.6)
(1.3)
(5.1)
(1.2)
(3.4)
Control
Mean (SD)
52.2
(5.1)
26.8
(3.3)
1.9
(1.7)
3.9
(2.1)
7.1
(0.9)
0.6
(1.1)
3.3
-
t
p
0.85 NS
0.65 NS
1.65 NS
3.11 0.00
2.65 0.01
7.48 0.00
7.00h (Table 2), while VII:c tended to be higher. The post-hoc t-tests revealed
that all fibrinolytic measures in VE subjects were significantly higher in the
morning but were similar to those of the controls in the evening (Table 2).
After adjusting for triglycerides and (HDL and LDL) cholesterol, the results
remained essentially the same, although VII:c now reached statistical significance (p=0.02).
Table 2: Hemostatic characteristics of exhausted and control subjects
Coagulation Measures
F1+2 (nMol/l)
FNG (g/l)
Fibrinolytic measures
tPA:ag (ng/ml)
tPA-PAI:ag (ng/ml)
PAI-1:act (U/ml)
7 am
6 pm
7 am
6 pm
VE
Mean
0.59
0.73
3.14
2.94
(SD)
(0.20)
(0.27)
(0.86)
(0.47)
7 am
6 pm
7 am
6 pm
7 am
6 pm
5.19
3.71
3.72
2.12
10.14
2.48
(1.83)
(1.59)
(2.53)
(1.88)
(6.59)
(1.99)
111
Control
Mean (SD)
F
p
0.51
(0.12) 5.61 0.02
0.59
(0.15)
2.70
(0.47) 6.43 0.01
2.68
(0.46)
Post-hoc t-test
4.29
(1.34) 2.15 0.04
3.32
(1.36) 1.01 NS
2.55
(1.60) 2.15 0.04
1.67
(1.15) 1.01 NS
6.20
(5.27) 2.54 0.01
2.13
(2.46) 0.60 NS
Discussion
These results provide substantial evidence for an increased activation of coagulation in VE, an increase that may be present throughout the day in these
subjects, as higher levels of F1+2, VII:c and FNG were not only found just after
awakening, prior to rising, but also in the late afternoon. Furthermore, the
higher levels of PA-1, tPA, and tPA-PAI in VE support the association of this
fatigue state with a decreased fibrinolysis [8,9]. The present results, however,
also revealed that this decreased fibrinolysis is particularly present in VE just
after awakening, prior to rising, but not in the late afternoon. As increased
activation of coagulation is present both after awakening and in the afternoon
in VE, the combined findings suggest the existence of a maladaptive imbalance
between these two systems in VE subjects that is particularly prominent early
in the morning. The resulting risk for thrombus formation in these subjects
may, therefore, also be higher early in the morning, and suggests an intermediary role of VE with respect to this high risk period of MI.
References
1 Markovitz J, et al. Psychosom Med 1991; 53:643-668.
2 Thayer R. The biopsychology of mood and arousal. Oxford, Oxford University Press,
1989.
3 Appels A, et al. Eur Heart J 1988; 9: 758-764.
4 Kop W, et al. Psychosom Med 1994; 56:281-287.
5 van Diest R, et al. Psychosom Med 1994; 56:28-35.
6 Falger P, et al. J Psychosom Res 1992; 36:777-786.
7 Kop W, et al. J Psychosom Res 1996; 40:397-405.
8 Räikkönen K, et al. Arterioscler Thromb Vasc Biol 1996; 16:363-367.
9 Kop W, et al. Psychosom Med 1998; 60:352-358.
10 Jern Ch, et al. Thromb Haemostas 1989; 62:767-771.
11 Goldberg R. Cardiol Clin 1996; 14:175-184.
12 Appels A, et al. Int J Cardiol 1987; 17:15-24.
13 Meesters C, et al. Psychol Health 1996; 11:557-581.
14 Robins L, et al. Psychol Med 1982; 12:855-870.
15 van Diest R. Psychosom Med 1990; 52:603-609.
This study was supported by grant 97.084 from the Dutch Heart Foundation
112
A case study of the free-running
circadian period of a morning type subject
in time isolation
HANS P. A. VAN DONGEN1, MONIQUE G. C. E. KUIJPERS2,
HANS DUINDAM2 & GERARD A. KERKHOF3
1
DIVISION OF SLEEP AND CHRONOBIOLOGY, UNIVERSITY OF PENNSYLVANIA, U.S.A.
2
DEPARTMENT OF PHYSIOLOGY, LEIDEN UNIVERSITY, THE NETHERLANDS
3
DEPARTMENT OF PSYCHOLOGY, LEIDEN UNIVERSITY, THE NETHERLANDS
Supported by NWO grants 575-65-068 and 575-10-020, and in part by NASA
grant NCC 9-58 (NSBRI) and the Institute for Experimental Psychiatry Research Foundation
Introduction
Morning type and evening type individuals (M-types and E-types) differ an average of 2.1h in the phase of their endogenous circadian rhythmicity (Kerkhof
& Van Dongen, 1996). It has also been hypothesized (Kerkhof, 1985) that Mtypes and E-types may differ in their free-running circadian period (tau). In
order to address this issue, we investigated the free-running circadian period of
an M-type subject. The free-running period depends on the conditions under
which it is measured (e.g., Czeisler et al., 1999). In a traditional time isolation
experiment, which is the paradigm we used, the general population average of
tau equals 24.93h (95% confidence interval 24.85 – 25.01; Wever, 1979, p.74).
It is hypothesized that, compared to this population average, M-types have a
somewhat smaller tau value (but still greater than 24.0h), and E-types have a
somewhat larger tau value. To examine this in an M-type subject, while simultaneously checking that time isolation is actually achieved, it must be shown
that the observed tau is both significantly less than 24.93h and significantly
different from 24.0h.
Methods
A healthy, non-smoking, 40-year old female subject, who took no medications
and used no steroidal means of contraception, was initially identified as an Mtype person on the basis of a validated Dutch morningness/eveningness questionnaire and ambulatory recordings of oral temperature (Kerkhof, 1984). The
subject underwent a 26h baseline constant routine, as described in Kerkhof &
113
Van Dongen (1996), to verify her classification as an M-type individual. Core
body temperature (CBT) was recorded continuously (1 sample per 2min) with
a rectal probe, and subjective alertness (ALERT) was measured hourly by means
of the Activation factor of the Activation–Deactivation Adjective Check List
(AD-ACL; Thayer, 1978). Both time series were fitted with a double (i.e., 24h
fundamental and 12h harmonic) sine wave to obtain the endogenous circadian
minimum and maximum. The CBT minimum occurred at 04:00, which is within
the range of 04:38 ± 46min (mean ± s.d.) observed for M-types under constantroutine conditions; the maximum of ALERT fell at 12:44, which is early relative to the previously observed constant-routine value of 15:25 ± 102min (mean
± s.d.) for M-types (Kerkhof & Van Dongen, 1996). Thus, the subject classified
as an M-type individual.
Following the constant routine, the subject entered time isolation for a period
of 20 subjective days in the Sleep Laboratory of the Department of Psychiatry
at Leiden University. The fully equipped laboratory area available to her consisted of a living room with built-in kitchen, a bathroom with shower, a bedroom, and a lock room with an opaque double-door exit to the laboratory’s
control room (which was outside the time isolation area). There were no other
exits and no windows, the laboratory was acoustically isolated, and all possible
sources of information about the time of day were removed from the laboratory. The laboratory’s own air conditioning unit provided fresh air of constant
temperature (18°C ± 1°C). Lighting conditions were at the control of the subject, but did not exceed 100 lux. At the start of the experiment, the subject was
allowed to bring personal items into the laboratory, to the exclusion of any
items revealing the time of day.
The control room was staffed in shifts of random, irregular duration; there
were strict instructions not to reveal the time of day to the subject (watches
were not allowed inside the control room). The subject was continuously monitored with a camera system (except when in the bedroom) that also allowed
continual vision inside the lock room by means of an infra-red light source.
The subject could contact the control room through a two-way intercom for
brief messages; contact was initiated by the subject only. She notified the control room immediately when she went to bed and when she got up (which she
was instructed to do upon final awakening). All intercom messages were recorded with reference to the date and time shown on the (hidden) master clock
in the control room. Food and other supplies were delivered and garbage was
collected via the lock room when the subject was asleep, after a random and
unspecified delay that varied from case to case. The subject was informed about
all these procedures before the start of the experiment.
114
Wrist actigraphy (ACT; 1 averaged sample per 30 min) and CBT (1 averaged
sample per 6 min) were recorded continuously; artifacts associated with bathroom visits were labeled missing values. One out of every three subjective days,
the subject collected urine samples that she measured for volume and stored in
pre-numbered tubes in a refrigerator in the lock room. Through the intercom,
the subject notified the control room immediately after taking a urine sample;
the tube number and volume were then recorded with reference to the date
and time on the control room clock. Every urine sample was later analyzed for
cortisol metabolite concentration, which was multiplied by the corresponding
volume and divided by the integration interval (i.e., the time passed since the
previous sample) to obtain a measure of average cortisol excretion (CORT);
the midpoint of the integration interval was used as the matching reference
time.
Lomb-Scargle periodogram analysis (Van Dongen et al., 1999) was applied to
the CBT, ACT and CORT data, yielding direct estimates of tau and the corresponding 95% confidence interval. Additionally, for visual representation of the
results only, a double (i.e., 24h fundamental and 12h harmonic) sine wave was
fitted to the CBT, ACT and CORT data to estimate the clock times of the circadian minimum and maximum in every 24h day of the experiment. The
periodogram analysis results were used to verify the appropriateness of this
visual representation.
After 20 subjective days of time isolation, the subject underwent another constant routine, while time isolation was maintained. The procedure was identical to that of the constant routine preceding the time isolation experiment,
with the exception that all recurring events (such as filling out the AD-ACL)
were now initiated by the subject rather than timed on the laboratory clock.
The endogenous circadian phases (i.e., timing of CBT minimum and ALERT
maximum) in the two constant routines were compared to assess the phase
differences. These were then divided by the number of free-running circadian
cycles of CBT (based on the tau value determined with periodogram analysis)
separating the two constant routines. This procedure provided additional, partially independent estimates of the circadian period (as a deviation of tau from
24.0h), that could be used for verification of the periodogram analysis results.
Results
Figure 1 shows the bedtime periods recorded during time isolation. The 20
subjective days of time isolation turned out to take 19 laboratory (i.e., 24h)
days, for on laboratory day 15 the subject judged two brief sleep periods occurring within a 24h interval to be actual subjective nights. This resulted in two
115
consecutive subjective days of very short duration. Furthermore, on laboratory day 18, a subjective nap preceded the subjective night. The data for the
first day of time isolation were discarded because of potential adaptation effects.
Figure 1: Clock times (on the abscissa) of bedtime periods,
plotted across laboratory days (downward on the
ordinate).
Figure 2: Clock times (on the abscissa) of the circadian CBT
minimum (dots), ACT maximum (stars) and CORT
minimum (circles), plotted across laboratory days
(downward on the abscissa). These clock times
were assessed for the purpose of visual representation of the data only.
Figure 2 shows the clock times of CBT minimum, ACT maximum and CORT
minimum as they occurred during time isolation. (The clock times for ACT
maximum on laboratory days 2 and 16 and for CORT minimum on laboratory
day 12 were missing because no statistically significant double sine wave could
be fitted.) With Lomb-Scargle periodogram analysis, significantly free-running
(i.e., longer than 24.0h) periods were found for CBT (tau=24.49h; 95% confidence interval 24.46 – 24.52) and ACT (tau=24.45h; 95% confidence interval
24.26 – 24.65). These free-running circadian periods were also significantly
shorter than Wever’s general population average of 24.93h (95% confidence
interval 24.85 – 25.01). The tau value for CORT was not significantly different
from 24.0h (cf. Van Dongen et al., 1998), but neither could a significant difference from the tau values observed for CBT and ACT be demonstrated.
The endogenous circadian CBT minimum during the constant routine at the
end of the time isolation study fell at 11:05, and the maximum of ALERT occurred at 01:53. These values were compared to the baseline (i.e., before time
116
isolation) constant-routine values of 04:00 and 12:44, respectively, for verification of the tau results obtained during time isolation. It was taken into account
that 19 free-running circadian cycles of CBT passed between the two constant
routines (as determined with periodogram analysis, see above). Thus, circadian periods of tau=24.4h for CBT and tau=24.7h for ALERT could be derived,
showing that the constant-routine results matched the periodogram analysis
results presented above.
Discussion
A free-running circadian period of approximately 24.5h was found for CBT,
and similar values were observed for other variables, in the M-type subject
exposed to time isolation. These tau values were smaller than the population
average of 24.93h assessed by Wever (1979, p.74). Wever’s results were obtained at slightly higher wake-time light intensities (139 lux on average; p.88),
but he demonstrated that there was very little variation in the value of tau
associated with different light intensities of the same order of magnitude (p.87).
Hence, a comparison of his results with ours appears to be valid, and supports
the hypothesis that M-types have a free-running circadian period shorter than
the general population average.
Czeisler et al. (1999) showed that free-running circadian periods measured
during forced desynchrony are generally much closer to 24.0h (namely, 24.18h
on average), the difference having to do with subjects’ self-selected light exposure during conventional time isolation. In a recent forced-desynchrony study
(Duffy et al., 1999), a preliminary M-type vs. E-type comparison of tau values
was made. On an imposed 28h rest/activity cycle, a –0.59 correlation (P<0.02)
between circadian period length and morningness score on a questionnaire
was found in 17 male subjects, with tau values ranging from about 23.8h to
24.4h. This result provided further evidence for the hypothesis that M-types
have a free-running circadian period shorter than the general population average. It would be valuable to repeat this forced-desynchrony experiment with
endogenous circadian phase rather than questionnaire criteria for the selection of M-types and E-types, to investigate whether a difference in free-running
circadian period may present an underlying mechanism for the difference in
circadian phase between M-types and E-types (Kerkhof & Van Dongen, 1996).
References
Czeisler, C.A., Duffy, J.F., Shanahan, T.L., Brown, E.N. et al. (1999). Stability, precision,
and near-24-hour period of the human circadian pacemaker. Science 284: 2177-2182.
117
Duffy, J.F., Dijk, D.-J., Hall, E.F. & Czeisler, C.A. (1999). Relationship of endogenous circadian melatonin and temperature rhythms to self-reported preference for morning or evening
activity in young and older people. Journal of Investigative Medicine 47: 141-150.
Kerkhof, G.A. (1984). Een Nederlandse vragenlijst voor de selectie van ochtend- en
avondmensen. Nederlands Tijdschrift voor de Psychologie 39: 281-294.
Kerkhof, G.A. (1985). Inter-individual differences in the human circadian system: a
review. Biological Psychology 20: 83-112.
Kerkhof, G.A. & Van Dongen, H.P.A. (1996). Morning-type and evening-type individuals
differ in the phase position of their endogenous circadian oscillator. Neuroscience Letters 218: 153-156.
Thayer, R.E. (1978). Factor analytic and reliability studies in the Activation-Deactivation Adjective Check List. Psychological Reports 42: 747-756.
Van Dongen, H.P.A., Mullington, J. & Dinges, D.F. (1998). Circadian phase delay during
88-hour sleep deprivation in dim light: Differences among body temperature, plasma
melatonin and plasma cortisol. Sleep-Wake Research in The Netherlands 9: 33-36.
Van Dongen, H.P.A., Olofsen, E., Van Hartevelt, J.H. & Kruyt, E.W. (1999). A procedure
of multiple period searching in unequally spaced time-series with the Lomb-Scargle
method. Biological Rhythm Research 30: 149-177.
Wever, R.A. (1979). The circadian system of man. Results of experiments under temporal isolation. Springer-Verlag, New York.
118
The relation between motor activity and daily
stress in remitted bipolar outpatients
R. HAVERMANS, A.L. VAN BEMMEL AND N.A. NICOLSON
DEPARTMENT OF PSYCHIATRY AND NEUROPSYCHOLOGY,
MAASTRICHT UNIVERSITY, THE NETHERLANDS
The course of a bipolar mood disorder is characterized by changes in motor
activity. Most studies have reported a decreased level of motor activity in depressed bipolar patients, while manic patients are more active compared to
normal controls (for a review see Teicher 1995). Most of these observations
have been made in hospitalized patients during a depressive or manic episode.
Little is known about motor activity in remitted bipolar patients. Klein et al.
(1991; 1992) studied motor activity in small groups of bipolar outpatients
during discontinuation of lithium treatment. They found that a higher daytime
activity level and a dysregulated rest/activity cycle preceded manic relapse.
These findings suggest that disturbances of the rest/activity cycle in remitted
bipolar patients are potential markers of increased risk for future relapse. The
rest/activity cycle reflects aspects of the sleep/wake cycle and disturbed sleep
is a well-known risk factor for manic relapse. It has been hypothesized that a
variety of psychological, interpersonal, environmental, and pharmacological
factors that are thought to trigger mania, do so through their capacity to cause
sleep disturbance (Wehr et al., 1987). The interaction between motor activity
variables and potential triggers, such as daily stress, is largely unknown. In the
present pilot study motor activity and its relation to daily stress are explored in
remitted bipolar patients in comparison with age- and sex-matched healthy
controls. This study forms part of a larger ongoing study on stress reactivity in
the daily life of bipolar patients in relation to the course of the disorder.
Methods
The patient group consisted of six men and four women with a DSM-IV diagnosis of Bipolar I (n= 9) or Bipolar II (n= 1) disorder. All were outpatients receiving prophylactic treatment with pharmacological mood stabilisers for at least
four months. For inclusion, patients had to be in partial or full remission for at
least two months. Mean age in this group was 40.7 years (SD = 9.7; range = 2755). The mean number of previous episodes was 5.4 (SD = 3.3; range=1-13).
An age- and sex-matched control group of ten healthy volunteers was recruited
among hospital staff and through local newspaper advertisements. Mean age
119
was 42.1 years (SD = 9.6; range = 30-56). Exclusion criteria were a history of
psychiatric or neurological disorder, use of psychotropic drugs, and a positive
family history for psychiatric disorders.
Motor activity was recorded continuously and sampled every 30-seconds over
six successive 24-hour periods, preceded by an adaptation 24-hour period, with
an activity monitor (Gaehwiler Electronic) worn on the non-dominant wrist.
The subjects were instructed to note the times the activity monitor was removed
(e.g., when swimming or taking a bath). Only 24-hour periods with a total ‘activity monitor off’ time less than 2 hours were used for the analysis. Each 24-hour
period was divided into a diurnal and a nocturnal period. The nocturnal period
was defined on the basis of a daily sleep log as the time period between going to
bed in the evening and getting up in the morning. The motor activity data were
analyzed with the software program ACTPLAN (Kramer et. al., 1992). Based on
the counts obtained for each 30-sec sampling interval (‘epoch’), the mean level
of motor activity was calculated for each diurnal and nocturnal period. The program also provides an algorithm for the calculation of the following variables,
which are thought to reflect sleep disturbance: in each subject the movement
index (MI), defined as the percentage of nocturnal epochs with an activity count
> 0; and the fragmentation indices F30 and F60, defined as the percentage of one
(F30) and two (F60) successive nocturnal epochs with activity, with respect to
the total number of immobility periods of all durations. At the end of each diurnal period, just before going to bed, subjects rated how stresssful that day had
been on a 7-point Likert scale with endpoints 1 ‘not at all’ and 7 ‘very much’. For
each subject, a single estimate for each measure was obtained by averaging across
the six diurnal and nocturnal periods. Group mean values were compared with ttests. Pearson correlations were used to examine the relation between perceived
daily stress level and motor activity variables.
Results
As shown in Table 1 patients reported spending more time in bed (nocturnal
period) than controls. Moreover, although patients did not differ significantly
from control on levels of overall diurnal activity, higher scores on the fragmentation indices F30 and F60 and a (non-significantly) higher movement index
suggest that patients experienced longer or more frequent periods of sustained
activity at night than controls. Daily stress ratings were similar in the two groups.
The correlations between mean daily stress and nocturnal activity variables
were low and not statistically significant (all r < .27; all p > .44).
Discussion
The results of this pilot study indicate some differences between the rest/activity patterns of remitted bipolar patients and those of healthy controls. No rela-
120
tion was found between daily stress ratings and any of the motor activity variables. Although the results of studies on motor activity cannot simply be extrapolated to conclusions about sleep per se, the finding that the patients showed
a less sustained immobility during the nocturnal period might reflect a disturbed sleep pattern, which might have clinical significance. However, it cannot be excluded to what extent medication affected the different nocturnal
motor activity levels and the durations spent in bed in our patients. Because of
the small number of subjects our findings are preliminary and have to be
evaluated in a larger group of patients.
Table 1:
Motor activity and stress measures in remitted bipolar patients and controls.
Patients (n = 10)
Controls (n=10)
Activity and Stress measures
Mean
SD
Mean
SD
t
p
Activity Monitor Off time (min)
Diurnal activity level
(counts / 30 sec)
Nocturnal period (min)
Nocturnal activity level
(counts / 30 sec)
Movement Index (%)
Fragmentation Index 30 sec (%)
Fragmentation Index 60 sec (%)
Perceived daily stress
(range 1 – 7)
16.09 14.19
32.71 15.63
20.39
40.78
7.40
9.68
.85
1.39
.41
.18
529.78 34.17
4.31
2.11
471.24
4.17
34.26
2.24
3.83
.15
.001
.89
13.31
13.69
23.61
2.82
3.90
5.91
8.44
1.03
1.60
2.08
2.24
.09
.13
.05
.04
.93
15.93
19.53
31.95
2.79
3.50
6.62
8.20
0.71
References
Klein, E., Meiraz, R., Pascal, M., Hefez, A., Lavie, P., 1991. Discontinuation of lithium
treatment in remitted bipolar patients: relationship between clinical outcome and changes
in sleep-wake cycles. The Journal of Nervous and Mental Disease 179, 499-501.
Klein, E., Lavie, P., Meiraz, R., Sadeh, A., Lenox, R.H., 1992. Increased motor activity
and recurrent manic episodes: predictors of rapid relapse in remitted bipolar patients
after lithium discontinuation. Biological Psychiatry 31, 279-284.
121
Kramer, C.G.S., Middelkoop, H.A.M., van Hilten, J.J., Kamphuisen H.A.C., 1992.
ACTPLAN and ACTSTAT: software programs for analysing motor activity data as obtained with commercial actigraphs. Journal of Ambulatory Monitoring 5, 331-340.
Middelkoop H.A.M.,1994. Actigraphic assessment of sleep and sleep disorders. PhD
Thesis. Eburon Publishers, Delft, The Netherlands.
Teicher, M.H., 1995. Actigraphy and motion analysis: new tools for psychiatry. Harvard
Rev Psychiatry 3, 18-35.
Wehr, T.A., Sack, D.A., Rosenthal, N.E., 1987. Sleep reduction as a final common pathway in the genesis of mania. American Journal of Psychiatry 144, 201-204
122
Multiple Sleep Latency Test:
are four tests necessary to
diagnose hypersomnia?
S. E. JENKINS*, R. J. SCHIMSHEIMER, G. A. KERKHOF, A. W. DE WEERD
CENTRE FOR SLEEP AND WAKE DISORDERS, MCH,
WESTEINDE HOSPITAL, THE HAGUE
*TRAINEE INTERN, DUNEDIN SCHOOL OF MEDICINE,
UNIVERSITY OF OTAGO, NEW ZEALAND
Introduction
The Multiple Sleep Latency Test (MSLT) is a standardised way to assess the
severity of excessive daytime sleepiness (EDS). 1 Currently, the recommendations are for four to five tests to be performed at two hourly intervals starting
from 9 am. This results in the test being time consuming and expensive. The
aim of this study is to determine whether we can decrease the number of tests
to obtain the same result.
Patients and Methods
The records of 170 patients were included in this study. We selected all patients who had an MSLT at the Westeinde Centre for Sleep and Wake Disorders
from 1994 until May 1999. Repeat MSLT’s and those recorded using old guidelines were excluded. The ages of the patients ranged from 9 to 74 years, the
average being 42.5 (standard deviation 14.8). The male to female sex ratio was
approximately 2:1 with 111 males and 59 females.
Stage 1b latency values for each patient’s four sleep periods (registrations) were
recorded and then categorised into three groups:
Group 1.
Group 2.
Group 3.
Latency <=5 minutes
Latency >5 <=10 minutes
Latency >10 minutes
(hypersomnia)
(‘grey’ area – diagnosis uncertain)
(no excessive daytime sleepiness)
Those who did not sleep were given the value 20 minutes or Group 3. All four
registrations were averaged in order to obtain the final average latency, which
123
is normally used in the diagnosis. This average latency was used as the control
value.
Results
The registrations were grouped into all possible combinations with the average
latency calculated for each. These averages were compared to the control
value to obtain sensitivity and specificity values for each category in each of
the registration combinations. (Table I)
The percentage of each category in each of the registrations was also calculated to observe any time-of-day variation. (Figure I)
Sensitivity and Specificity of Each Registration
Combination in Each Category
Registration
Combinations
1.
2.
3.
4.
1&2
1&3
1&4
2&4
2&3
3&4
1&2&4
2&3&4
1&2&3
1&3&4
Control Value
Latency
<=5
Sens.
0.79
0.89
0.92
0.91
0.81
0.92
0.89
0.92
0.94
0.91
0.96
0.94
0.98
0.94
1.00
Spec.
0.89
0.85
0.78
0.80
0.97
0.91
0.91
0.91
0.88
0.86
0.97
0.92
0.97
0.92
1.00
Latency
>5 <=10
Sens.
0.53
0.65
0.47
0.45
0.76
0.64
0.62
0.71
0.71
0.65
0.85
0.84
0.87
0.87
1.00
Spec.
0.81
0.88
0.83
0.90
0.87
0.88
0.90
0.94
0.91
0.87
0.97
0.95
0.94
0.94
1.00
Sens. = Sensitivity Spec. = Specificity
Table I
124
Latency
>10
Sens.
0.81
0.84
0.68
0.85
0.92
0.84
0.92
0.95
0.87
0.84
0.98
0.95
0.90
0.90
1.00
Spec.
0.87
0.96
0.93
0.92
0.91
0.92
0.90
0.95
0.97
0.97
0.95
1.00
0.96
0.96
1.00
Overall
Latency
Average
Sens.
0.71
0.79
0.69
0.74
0.84
0.80
0.81
0.86
0.84
0.80
0.94
0.91
0.92
0.86
1.00
Spec.
0.86
0.90
0.84
0.87
0.92
0.90
0.91
0.93
0.92
0.90
0.97
0.96
0.96
0.94
1.00
Overall the registration combination 1,2&4 gave the highest number of correct
values with 94% of the latency averages representing the control value. This
trend was not seen in the other categories.
Not one of the four individual registrations was more specific and more sensitive in any of the categories. However, of the different combinations, that of
1,2&3 gave the highest sensitivity and specificity (98% and 97% respectively)
for a latency <=5. As noted in Table I, the sensitivity and specificity for this
combination were also high in the other categories.
Figure I
A time variation during the course of the MSLT was noted. The third registration, which occurs at approximately one p.m., had the highest proportion of
sleep latencies’ occurring in the first category, (latency <= 5 minutes), with an
according decrease in the proportions of patients with latencies in the other
two categories.
Discussion
Not one of the four individual registrations was more specific and more sensitive in any of the categories suggesting that not one of the individual registrations can be given more weight in the overall analysis.
125
The registration combination of 1,2&3 was shown to be the most sensitive and
specific in category 1 (latency <=5). This is a very helpful finding considering
that this combination is clinically the most useful, enabling the removal of the
fourth registration (unless the MSLT is being performed for SOREM’s to diagnose narcolepsy or other related sleep disorders). A recent article by Sarodia
et al 2 also tried to determine how many registrations are needed in the MSLT.
Although they addressed this same problem in a different manner they came
up with a similar conclusion – that the “final MSLT result is highly predictable
from the sleep latency results of the first three naps”.
The time variation during the day of each of the categories showed a trend in
the third registration to an increased percentage of the patients sleeping within
five minutes. The last registration showed a sharp increase in the number of
patients who took more than ten minutes to fall asleep. One suggestion for this
trend is that this represents the ‘last registration effect’, (increased alertness
due to anticipation of finishing the MSLT).
Conclusion
The first three registrations together give a sensitivity of 98% and a specificity
of 97% for latency less than or equal to five. This suggests that only these three
registrations may need to be performed in patients with suspected hypersomnia.
References
1] Dement W. C., Mitler M. M., Roth T., Westbrook P. R., Keenan S. Guidelines for the
Multiple Sleep Latency Test (MSLT): A Standard Measure of Sleepiness Sleep 1986;
9:519-524
2] Sarodia b. D., Golish J., Yu N., Perry M., Foldvary N., Dinner D. Multiple Sleep Latency Test: How Many Naps Are Needed? Sleep 1999;22(supplement):15
126
Do non-benzodiazepine-hypnotics prove a
valuable alternative to benzodiazepines
for the treatment of insomnia?
A. KNUISTINGH NEVEN,
DEPARTMENT OF GENERAL PRACTICE,
LEIDEN UNIVERSITY MEDICAL CENTER
Introduction
In this article insights with regard to sleep physiology and insomnia are discussed, as well as the pharmacological approach of insomnia. The main pharmacological mechanisms of action of benzodiazepines (BZDs) and the newest
generation of non-benzodiazepine hypnotics, cyclopyrrolone- and
imidazopyridine-derivatives, are evaluated.
Sleep: physiological insights
Macrostructure of sleep
During sleep, 5 to 6 sleepcycles occur during the night, lasting approximately
1,5 hours. Every sleep cycle consists of one REM stage, characterised by rapid
eye movements, paradoxical cortical activation and muscular atonia, and four
non-REM stages (‘macrostructure’ of sleep). The first three to five hours of the
night are considered to be most important and responsible for the restorative
function of sleep, since during these hours the largest amount of slow wave
sleep (non-REM 3&4) occurs. The sleep in the second part of the night is dominated by superficial REM sleep and sleep stage 2.1
Microstructure of sleep
Additional microstructural characteristics such as the cyclic alternating patterns (CAPs) can also be obtained in insomnia patients for more detailed evaluation of sleep. CAP is a physiological microcomponent of sleep that undergoes
measurable changes in several sleep disorders and is rated in %ratio to NREM
sleep.2 The CAP rate is decreased during night sleep recovery, whereas it is
increased by any perturbing factor, especially noise. The CAP rate is significantly correlated with the subjective appreciation of sleep quality and sleep
stability.
127
&Pharmacological management of sleep disorders
Insomnia is best characterized as a complaint of difficulty in initiating or maintaining sleep or having a non-restorative sleep. Over the last century a variety
of hypnotic agents have been introduced for the treatment of insomnia of which
the BZDs have been used on a large scale for a long time now. Their main
therapeutic aims are: a shortening of sleep onset latency, a reduction of
awakenings and a decrease of arousability.
The side effects can be an important reason to restrict prescription. The most
significant are: a negative impact on daytime alertness, memory, cognitive functions performance, and mood (residual effects), tolerance (reduction of hypnotic efficacy after several weeks), difficulties in discontinuing treatment due
to withdrawal symptoms and/or rebound insomnia (causing a non-negligible
risk of dependence and abuse). The anxiolytic effects of BZD’s may last for
months. Therefore patients might still consider the hypnotics to be effective;
this may also be responsible for chronic use or abuse by patients. Furthermore, the BZD’s can affect sleep architecture: 1) an increase of non-REM 2
sleep, 2) a reduction of the amount of non-REM 3/4 sleep (SWS or deep sleep)
and 3) suppression of REM-sleep.
The most recently developed hypnotic compounds are the non-benzodiazepines
(non-BZD): The cyclopyrrolone zopiclon.and the imidazopyridine zolpidem.
Their profiles are described below.
Pharmacological profile
BZDs, zolpidem and zopiclone bind to GABAA-receptors in the brain and are
effective by potentiating the activity of the inhibitory neurotransmitter GABA.
However, they show a different binding affinity for the GABA-A receptors.3 The
GABAA - receptor sites are each made up of three distinct sets of high-affinity
binding sites recently designated as ␻1-, ␻2-, and ␻3-subtypes, which are modulatory sites of the GABAA- receptor complex. BZDs and the cyclopyrrolone
zopiclone present no selectivity for the ␻-receptor subtypes. BZDs bind to all
three receptor-subtypes (␻1, ␻2, ␻3), zopiclone to ␻1 and ␻2 subtypes. The
BZD’s have sedative, anticonvulsant, anxiolytic and myorelaxant properties.
Zopiclone has a similar profile, although it causes less myorelaxation. In contrast, zolpidem shows high selectivity for the ␻1-receptor subtype which mainly
can be found in the cerebellum, cortex and subcortex. This receptor selectivity probably explains why the hypnotic action of zolpidem occurs at doses
much lower than those needed for its anticonvulsant or myorelaxant activities
and its lack of anxiolytic properties.4
128
Zopiclone
Hypnotic effects
Zopiclone is indicated for the short-term treatment of insomnia. It has a short
half-life (3.5-7 hours), which is being prolonged in elderly and patients with
liver cirrhosis. The drug is metabolized by the liver to inactive metabolites that
are eliminated for 75%-80% by renal excretion. Zopiclone, like most BZDs, demonstrates no selectivity for the ␻-receptor subtypes of the GABAA-receptor complex. Zopiclone acts favourably on sleep parameters such as sleep onset latency and total sleep time, but doesn’t affect the total duration of REM-sleep.
Zopiclone’s pharmacological profile resembles to a greater extent that of BZDs
than zolpidem.5
Comparative clinical trials with zopiclone
Results from placebo-controlled trials show that zopiclone significantly improved
quality of sleep in patients with insomnia as well as in healthy subjects who
were working in night-shifts. The effectiveness of zopiclone was superior to
placebo in both treatment groups measured in a subjective and objective manner. In several smaller studies zopiclone also demonstrated an improvement of
sleep quality which was similar to the effect demonstrated by temazepam,
nitrazepam, flunitrazepam, flurazepam and triazolam. Comparative trials including other short-acting BZDs have not been performed. In five comparative
studies zopiclone demonstrated no adverse effects on daily functioning during
the following day and did not promote rebound-insomnia, whereas nitrazepam
and flurazepam did.
Tolerance & dependency
Use of zopiclone doesn’t promote tolerance during the first 4 weeks of treatment. However, some data indicate that chronic use in higher dosages can
induce dependency.6
Withdrawal of treatment and rebound insomnia
Discontinuation of clinical administration of zopiclone to healthy volunteers during a two-week treatment period hardly led to any withdrawal symptoms. However, rebound-insomnia occurred after discontinuation of the drug after a 3-week
treatment period. In three other, short studies with zopiclone withdrawal of treatment did not induce marked rebound-insomnia, possibly due to the short period
of observation (7, 10 and 14 days). In another clinical trial, a gradual withdrawal of
medication was investigated in persons who had used zopiclone for at least 3 consecutive months. Trial results mainly showed the occurrence of sleep disorders
following withdrawal of zopiclone, with minor other withdrawal symptoms.
129
Zolpidem
Hypnotic effects
Zolpidem is indicated for the short-term treatment of insomnia. It shows a high
selectivity for the ␻1-receptor and is considered to be a pure hypnotic without
other significant effects. It has a short half-life of 2.4 hours, and has no active
metabolites. No accumulation occurs during repeated administration. It is oxidized and hydroxylated by the liver to inactive metabolites that are eliminated
primarily by renal excretion.7 After oral intake of zolpidem, the onset of sleep
is within 12-25 minutes and the hypnotic effect has a duration of approximately 6 hours. Zolpidem in the recommended dose of 10 mg has no negative
influence on overall sleep architecture, both in adults and elderly. A prolonged
restorative non-REM3/4 sleep (slow wave sleep) has been observed, which can
be considered as a beneficial effect. The duration and latency of REM sleep
stays unmodified. Furthermore, the drug has a favourable influence on nocturnal awakenings and increases the number of sleep cycles.8,9
Comparative clinical trials
It was confirmed that 10 mg zolpidem is superior to placebo with, in contrast
to most BZD hypnotics, no or minimal impact on sleep architecture. In studies
involving insomniac patients, zolpidem induced a significantly higher quality
of sleep compared to placebo, subjectively (questionnaires) as well as objectively (polysomnography). The same effect was observed in elderly patients
with psychiatric disorders. Furthermore, the comparative efficacy of zolpidem
has been established for diazepam, flunitrazepam, lorazepam, midazolam and
many other BZDs in controlled studies, showing that zolpidem had at least
similar or superior activity in terms of sleep onset in insomniac patients.
Tolerance & dependency
It has been hypothesized that the rate at which the GABA-receptor complex is
occupied pharmacologically plays a role in the development of physical dependency. The low receptor occupancy (around 14%) necessary for zolpidem
to induce hypnotic activity could imply that the risk of physical dependency
will be lesser.10 Furthermore, zolpidem lacks anxiolytic activity. Persistent
anxiolytic effects, as seen with BZD’s are perceived as effectiveness by patients
and could cause chronic use. This suggests that the risk of chronic use/abuse
with zolpidem is lower. Indeed, studies have shown that the efficacy of zolpidem
does’nt diminish even after longer use (up to 360 days) However, hypnotic
drugs are not meant to be taken for such long periods.
130
Residual effects
Due to its pharmacological profile, zolpidem has a limited effect on psychomotor performance.9 Twelve double-blind comparative trials (with placebo/and or
benzodiazepine) have been performed, including a total of 1521 subjects/patients and of whom approximately 900 subjects received zolpidem treatment
in order to assess daytime impact on alertness, attention, psychomotor skills
and memory. Several standard techniques and procedures were used to obtain
objective data. Daytime alertness after drug intake has been studied and no
significant impairment was found after zolpidem in the recommended dose.
Assessments of attention and psychomotor skills (for instance driving tests)
objectively showed the lack of residual effects of zolpidem on vigilance, concentration and coordination performances on the morning after intake of
zolpidem 5 or 10 mg. Also, no significant memory impairment has been observed six hours after intake of zolpidem 5 or 10 mg. In addition, zolpidem
intake did not cause significant subjective residual effects the following morning.
Withdrawal symptoms and rebound insomnia
Five placebo-controlled studies focussing on the objective assessment of rebound symptoms in 229 insomniac patients, of whom 115 received zolpidem
treatment for a period of 14-35 days, demonstrated that up to 5 weeks treatment in the recommended dose of 10 mg, no objective evidence of rebound
insomnia after abrupt cessation of the drug. In one study, withdrawal of triazolam
was compared with withdrawal of zolpidem in 22 insomniac patients. Only
patients in the triazolam treatment group suffered a marked rebound-insomnia. The above mentioned studies show that abrupt discontinuation with
zolpidem is easy to manage in the vast mojority of patients: rebound insomnia
is unlikely to occur if zolpidem is used within the current recommendations.
Summary and conclusions
Zopiclone and zolpidem exert their hypnotic effects through a interaction with
the ␻-receptor in the central nervous system. It is suggested that the selectivity of zolpidem for the ␻1-receptor determines its different pharmacological
profile compared to BZDs and zopiclone. Zolpidem has a greater potency in
producing hypnotic effects relative to anticonvulsant effects and myorelaxant
effects and is devoid of anxiolytic effects. Zopiclon has demonstrated no clinically significant advantages in controlled clinical trials comparing the drug with
BZDs such as nitrazepam, flunitrazepam, temazepam, triazolam and
flurazepam.The adverse events are comparable to those of BZDs. Whether
zopiclone leads to less rebound insomnia, cannot be concluded from available
data of clinical trials. Should be decided for prescription, a maximum treat-
131
ment period of 2-4 weeks is advised. Comparison of zolpidem with reference
BZDs in controlled studies showed that zolpidem had at least similar efficacy
in terms of sleep onset in insomniac patients, whilst having a very limited impact on sleep architecture (especially on REM sleep stages) and on cognitive
functioning. Furthermore, zolpidem preserved or may prolong deep, restorative sleep NREM 3/4. In contrast to most BZDs hypnotics, zolpidem lacks important mechanisms concerning dependency potential and the risk of chronic
use : the abrupt discontinuation of treatment does not induce rebound insomnia, tolerance seems to be very unlikely to appear and it lacks anxiolytic effects. Therefore, the dependency potential and the risk of chronic use/abuse
are considered to be low.
Acknowledgement:
This article is based on a report on this subject in the Dutch Pharmaceutical
Reports by the author: Geneesmiddelen Bulletin 1999; 33(2): 13-7. The author thanks Mrs. Denise van den Berg, Medical Advisor Sanofi-Synthélabo
for her great help in preparing the article.
References
1 Horne JA. Functional aspects of human slow wave sleep (hSWS). In: Slow wave sleep:
Physiological, Pathophysiological and functional aspects, edited by A. Wauquier et
al. Raven Press, Ltd., New York.1989: 109-18.
2 Parrino L et al. Multidrug comparison in situational insomnia: PSG Analysis by means
of the Cylcic Alternating Pattern. Clinical Neuropharmacology 1997; vol. 20, No. 3,
253-63.
3 Perrault G, Morel E, Sanger DJ, Zivkovic B. Differences in pharmacological profiles
of a new generation of benzodiazepine and non-benzodiazepine hypnotics. Eur J
Pharmacol 1990; 187: 487-94.
4 Besnard F, et al. GABAA receptor subtypes and the mechanism of action of Zolpidem.
In: Freeman H, Puech AJ, Roth T (eds.): Zolpidem; an update of its pharmacological
properties and therapeutic place in the management of insomnia. Paris, Elsevier
1996.
5 Wadworth AN, et al. Zopiclone. A review of its pharmacological properties and therapeutic efficacy as an hypnotic. Drugs and Aging 1993; 3; 441-59.
6 Jones IR, Sullivan G. Physical dependence on zopliclone: case reports. BMJ 1998;
316: 117.
7 Fraisse J, Garrigou-Gadenne D, Thenot JP. Pharmacokinetic and metabolic profile of
zolpidem. In: Freeman H, Puech AJ, Roth T (eds.): Zolpidem; an update of its pharmacological properties and therapeutic place in the management of insomnia. Paris,
Elsevier 1996.
132
8 Walth J, Roehrs T, Declerck AC. Polysomnographic studies of the effects of zolpidem
in patients with insomnia. In: Freeman H, Puech AJ, Roth T (eds). Zolpidem: an
update of its pharmacological properties and therapeutic place in the management
of insomnia. Paris: Elsevier, 1996.
9 Sanger D.J., Depoortere H. The pharmacology and mechanism of action of zolpidem.
CNS Drug Reviews, 1998;vol. 4, no.4, 323-40.
10 Darcourt G., Pringuey D., Salliere D. and Lavoisy J. The safety and tolerability of
zolpidem - an update.J of Psychopharmacol 1999;13 (1): 81-93.
133
134
A psychophysiological study of sleep onset by
means of dynamic spectral analysis and ERP
A.R. KONING, W.F. HOFMAN & K. R. RIDDERINKHOF
DEPARTMENT OF PSYCHOLOGY,
UNIVERSITY OF AMSTERDAM, AMSTERDAM
Introduction
Sleep onset (SO), as defined by Rechtschaffen and Kales (1968), is identified
by the first sleep spindle or K-complex. Polysomnographic scoring of all night
EEG recordings makes it possible for researchers to identify this moment of
SO in subjects. Using a sleep spindle or K-complex as the start of the second
sleep stage (S2), and hence the start of sleep, only serves as a rule of thumb.
When subjects are in the preceding first stage of sleep (S1), they are not considered to be sleeping. However, several physiological changes do occur before
the start of S2 and should be the focus of sleep onset research. The use of a
sleep spindle or K-complex as a marker for the start of sleep does not take into
consideration the interaction of physiological systems responsible for the lowering of the level of consciousness in the process of falling asleep.
The interaction of the ascending reticular activating system (ARAS), with an
activating input, and neurocortical neurons, with a synchronizing input, seems
to be important for falling asleep. In order to examine the level of
(de)synchronization, and its change across time during the wake/sleep transition Fast Fourier Transformation (FFT) can be used. FFT was used by Ogilvie
et al. (1997). Both the alfa and beta frequency bands decreased in power in S2
compared to S1. The slower theta and delta frequency bands both increased in
power in S2 compared to S1. Using behavioral responses in sleep research provides additional insight into the level of arousal. The experiment of Harsh et al.
(1994) showed an increase in latency and a simultaneous lowering of amplitude for the P3 component during S2 compared to S1. Both FFT and ERP will
be used to examine physiological changes across time during the wake/sleep
process. By calculating the FFTs for each trial (standard tones only) a dynamic
change of power can be observed.
Methods
Thirteen first year students (3 males, 10 females) of the University of Amsterdam participated in the experiment. Average age was 21,27 (SD 2,43). Five
tasks were administered consecutively in one session in the evening. A tenminute rest period was taken between each task. Tones were presented
135
binaurally through a set of earplugs using an oddball paradigm. Deviant tones
were randomly presented in a train of standard tones. An inter-stimulus interval of 2000 ms was used. For the first task, the instructions required the subjects to react whenever a deviant tone was heard. A rubber ball with an electronic switch inside was strapped to the subjects’ preferred hand and registered
each reaction. The second task required the subjects to ignore all tones presented. The third task required the subjects to go to sleep, but at the same time
to try to react whenever a deviant tone was heard. Sleep, however, should not
be resisted. The behavioral response required from the subject enables monitoring of the level of consciousness. For the fourth task, the subjects were asked
to go to sleep, but no reaction to the presented tones needed to be given. During the third and the fourth tasks the subjects were told that should they fall
asleep they would be awakened in order not to take away delta pressure for the
following tasks. Online monitoring of the EEG activity was performed during
these tasks. The appearance of a sleep spindle or a K-complex was used to
awaken the subjects. The earplugs were removed for the last task and the subjects were simply asked to go to sleep. Wake and sleep tasks were administered, in conjunction with a response/no-response condition, to examine changes
in the level of arousal. No tones were presented in the last task to eliminate
possible interference of the tones with sleep. ERP and FFT analyses will be
performed in all tasks. FFT will be performed on each trial (standard tones
only) in order to examine changes in power across time.
Results
ERP: P3 amplitude and latency and the MMN amplitude and latency did not
reveal any changes across time during the sleeping tasks. In addition, the twoway interactions between the wake and sleep tasks were not significant. No
clear distinction in the level of arousal could be made between the wake and
sleep tasks. Neither for the response condition, nor for the no-response condition.
Spectral Analyses: The spectral analyses revealed that the alfa and theta bands
give the most unambiguous information concerning the approach to sleep onset. A third order polynomial best describes the rise and fall of both frequency
bands before the start of S2 (p<.001 in all tasks). The third order polynomials
of both the alfa and beta bands show an initial rise in power, followed by a drop
in power across time. The third order polynomials of the theta and delta bands
show an increment in power across time. An example of the averaged regression curves for the theta band can be seen in figure 1. In figure 1a, the averaged
regression curves of the theta band of the third task are presented. In this task,
the subjects were asked to give a response to each deviant heard. Figure 1b
shows the averaged regression curves of the theta band for the fourth task. No
136
behavioral response needed to be given here. As can be seen in figure 1 the lack
of a behavioral response in the fourth task (figure 1b) results in a sharp increment in power during the wake/sleep transition.
Figure 1:
Averaged regression curves of the theta frequency band for task 3 (1a)
and task 4 (1b).
Discussion
The focus of this experiment was to reveal the dynamic character of the sleep
onset period. Whether or not the sleep onset process is a continuous process
remains unclear from the ERP point of view in this experiment. Since the subjects were awakened when a sleep spindle or K-complex appeared, the lowering of the level of consciousness could have been insufficient to discriminate
the wake and sleep tasks on the basis of the ERP. The results of the spectral
analyses show that the theta band seems to be the best frequency band, together with the alfa band, to examine abrupt changes in power (rise of theta,
and a simultaneous drop in alfa) across time. These results were not corroborated by the last task where no tones were presented, but this could be the
result of a drop in delta pressure. Although the subjects were awakened when a
sleep spindle or K-complex appeared there was only a ten-minute rest period,
and all tasks were administered during a single session. A suggestion for further
research is to spread out the tasks over different session days and to let the
subjects enter the second sleep stage and then compare the first two stages
with each other using the border between the two stages (sleep spindle or Kcomplex) to reduce the variance of time of sleep onset. As for the design of the
experiment, the presentation of tones might overemphasize the subject’s levels
of the beta frequency band.
137
References
Harsh, J.R. Event-related potential changes during the wake-to-sleep transition. In: Sleep
Onset: Normal and Abnormal Processes. Rochester Hills, MI: American Psychological
Association (1994) Eds. Ogilvie R.D. & Harsh, J.R.
Ogilvie R.D., Simons, I.A., Kuderian, R.H., MacDonald, T. & Rustenburg, J. Behavioral,
event-related potential and EEG/FFT changes at sleep onset. Psychophysiology 28 (1991),
54-64.
Rechtschaffen, A. & Kales, A. A manual of standardized terminology, techniques and
scoring systems for sleep stages of human subjects. U.S. Govt. Print. Office, Washington, DC, (1968).
138
Ambulatory assessment of sympathovagal
heart activity in primary insomniacs
R.MULLAART1, W.F. HOFMAN1, G.A. KERKHOF1 AND A. KNUISTINGH NEVEN2
1
DEPARTMENT OF PSYCHOLOGY, UNIVERSITY OF AMSTERDAM, AMSTERDAM
2
DEPARTMENT OF GENERAL PRACTICE
OF THE LEIDEN UNIVERSITY MEDICAL CENTRE
Introduction
The concept of arousal plays an important role in most explanations of insomnia. According to the hypothesis of Espie (1991) external stimuli can cause
insomnia by a conditioning process which subsequently produces chronic
hyperarousal. This hyperarousal remains present in the absence of the external stimuli and leads therefore to chronic insomnia.
Hyperarousal is reflected in the level of activity of the autonomic nervous system. Heart activity is an important index of the activity of the autonomic nervous system. Activity of the heart is influenced by the two branches of the autonomic nervous system, the sympathetic chain and the parasympathetic chain.
If it is possible to examine the two branches separately this would result in a
better understanding of the mediation of arousal through the two branches of
the autonomic nervous system. There could be a difference in activity of the
two branches at different moments during the wake and sleep cycle.
The measurement of the heart activity was done by means of the VU-AMS(Vrije
Universiteit Ambulatory Monitoring System). This device is able to measure,
ambulant and continuously, the sympathovagal heart activity in combination
with breathing variables.
The present study focuses on the measurement of hyperarousal in primary
insomniacs by continuous monitoring of the activity of the autonomic nervous
system.
Method
The insomnia group consisted of 3 males and 4 females (age range: 40 to 68
years). The patients were compared with a group of healthy sleepers which
consisted of 4 males and 5 females (age range: 45 to 61 years). First, all subjects were screened for anxiety and depression using the SCL-90 questionnaire. Subjects who scored more than a specific threshold on SCL-90 where
excluded. Second, all subjects where screened for sleep variables using the
subjective measuring device: the ‘Ritmelog’ (Kerkhof, 1984). And finally, the
139
physiological data were collected for 24 hours, ambulatory, using the VU-AMS
combined with wrist activity measurements.
The following physiological variables were calculated:
- Heart rate activity (HRA) was calculated from R-top intervals. HRA was used
as an gross index of arousal.
- Delta heart rate activity (DHRA) was calculated by comparing individuals
with their 24-hour mean heart rate.
- Pre-ejection period (PEP), which was used as an index of sympathetic influence on the heart, was calculated from the ICG complexes (Geus de &
Doornen van, 1994).
- Delta pre-ejection period (DPEP) was calculated by comparing individuals
with their 24-hour mean pre-ejection period.
- Respiratory sinus arrhythmias (RSA), which was used as an index of
parasympathetic influence on the heart, was calculated from respiratory intervals using the peak-to-through method (Grossman, van Beek & Wientjes,
1990) for each breath.
Results
During the day insomniacs showed a significantly higher relative average heartbeat (DHRA) than healthy sleepers (F(6,7)=6,919, p<.025). See Figure 1.
Figure 1
Differences in sympathetic influence on the heart activity were not found (see
Figure 2). Also no statistical significant differences in parasympathetic influence on the heart activity were found. (see Figure 3).
140
Figure 2
Figure 3
Discussion
Contrary to what is reported in previous studies, no differences in absolute
heart rate was found. However, it is more interesting to examine the relative
findings of heart rate data (DHRA) in this study because of large differences
between individuals in absolute heart rate. There was a difference during the
day in relative average heartbeat between the groups. Insomniacs seem to have
a higher relative average heartbeat during the day than healthy sleepers. When
we compare this finding with the theory of hyperarousal in insomniacs it seems
logical to conclude that insomniacs are hyperaroused during the day. But when
we look at the two mechanisms influencing the heart, sympathetic and
parasympathetic, there were no differences between the groups. There seemed
to be a difference in vagal tonus between the two groups, but the results were
probably influenced by a lack of statistical power.
The absence of difference in sympathovagal heart activity can also be influenced by the lack of difference between the groups.
In summary, it remains questionable if hyperarousal plays a role in the cause
and ongoing of insomnia.
Reference
Espie, C.A. (1991). The psychological treatment of insomnia. John Wiley and Sons,
Chichester.
Geus de, E.J.C., & Doornen van, L.J.P. (1994). Ambulatory assessment of
parasympathetic/sympathetic balance by impedance cardiography. 141-163.
Grossman, P, & Karemaker, J., & Wieling, W. (1991). Prediction of tonic parasympathetic
cardiac control using respiratory sinus arrhythmia: the need for respiratory control.
Psychophysiology, 28, 201-216
141
142
P300 in sleep state misperception
R. J. SCHIMSHEIMER, M.M.R. VERHELST, D. ZEEMAN
CENTRE FOR SLEEP AND WAKE DISORDERS, MCH, WESTEINDE HOSPITAL,
THE HAGUE
Sleep State Misperception is a disorder in which a complaint of insomnia occurs without objective evidence of sleep disturbance. Excessive mentation during
sleep has been suggested as a possible explanation for this disorder. A higher
mentation during sleep in Sleep State Misperception indicates that there might
be a difference in information processing during sleep in these patients.
Event related potentials; especially the P300 can be used to study information
processing during wake and sleep (1).
It may be postulated that P300 amplitude and latency during wake and sleep
behave different in patients with Sleep State Misperception. However, in sleep
other event related potentials could appear for example SII-P3 that cannot easily be discerned from P300 (1). Therefore we studied the P300 in 3 patients
with Sleep State Misperception and 10 controls in 3 conditions:
a during wakefulness, at 22:00 hours - W condition
b during a forced awakening from the first period of Slow Wave Sleep – SWS
condition
c during a forced awakening from the first period of REM sleep – REM condition
Methods
A simple auditory oddball paradigm was used with 20% rare high tones (40 dB
2 kHz) and with 80% normal low tones (40 dB 1 kHz). The stimulus frequency
was 0,5 Hz. Patients had to press a button when they heard the rare “target”
tones. During wakefulness two trials of 200 stimuli were offered in one trial by
a standard headphone. Rare tones were at random mixed with the frequent
tones. After the forced awakenings only 200 stimuli were offered in one trial in
order to prevent the patient from falling asleep.
Signals were derived from Fz, Cz and Pz with linked ears as reference electrode. Signals were averaged with a band pass of 0.1 – 50 Hz and with an analysis time of 1 s. The EOG was monitored with two electrodes, one supra orbital
and one on the outer canthus both of the left eye. Signals contaminated with
eye movements were automatically rejected.
143
Subjects
The control group consisted of 10 healthy persons, 5 males and 5 females, age
range 23-49 years, who had no sleep problems.
The patient group consisted of 3 persons, in the age of 16, 36 and 39, who
suffered from total Sleep State Misperception (they reported no sleep at all in
their sleepdiary in contrast with their normal 2x24 hours polysomnography).
Results
The results of the latency and amplitude measurements are given in the following tables. Standard deviations were not calculated for the patient group due to
the small number of patients.
Derived from Fz
P300 latency controls
P300 latency patients
P300 amplitude controls
P300 amplitude patients
Wake
301 ± 34 ms
324 ms
17 ± 9 µV
8 µV
SWS
312 ± 44 ms
337 ms
15 ± 5 µV
8 µV
REM
320 ± 38 ms
367 ms
11 ± 6 µV
4 µV
Wake
301 ± 36 ms
319 ms
21 ± 9 µV
10 µV
SWS
310 ± 40 ms
337 ms
16 ± 4 µV
11 µV
REM
322 ± 39 ms
379 ms
14 ± 6 µV
3 µV
Wake
305 ± 38 ms
315 ms
17 ± 9 µV
13 µV
SWS
311 ± 36 ms
341 ms
15 ± 4 µV
12 µV
REM
325 ± 45 ms
379 ms
13 ± 6 µV
8 µV
Derived from Cz
P300 latency controls
P300 latency patients
P300 amplitude controls
P300 amplitude patients
Derived from Pz
P300 latency controls
P300 latency patients
P300 amplitude controls
P300 amplitude patients
144
Conclusion
Although the lower amplitudes and longer latencies in patients compared to controls suggest lower vigilance, the tendency to lower amplitudes in the SWS condition in controls was
not seen in patients. This probably indicates higher vigilance/mentation in the patients
during the SWS condition. In REM condition this phenomenon did not occur. Although 3
patients are too few to draw definite conclusions, the results suggest that excessive mentation during Slow Wave Sleep may be responsible for Sleep State misperception.
1) H. Bastuji, L. Garcia Larrea. What do studies on P300 tell us about information processing during sleep? J Sleep Res 1998,7; supp. 2: p. 18
145
146
The effect of treatment with melatonin for
chronic sleep onset insomnia in children with
attention deficit hyperactivity disorder:
randomized placebo-controlled trial
M.G. SMITS*, E.J. NAGTEGAAL**, G.A. KERKHOF***,
S. VALENTIJN****, A.L.M. COENEN****
HOSPITAL “DE GELDERSE VALLEI”, EDE: DEPARTMENT OF NEUROLOGY AND SLEEPWAKE DISORDERS*, DEPARTMENT OF HOSPITAL PHARMACY RIJNSTAETE HOSPITAL
ARNHEM, UNIVERSITY OF LEIDEN, DEPARTMENT OF PHYSIOLOGY***,
UNIVERSITY OF NIJMEGEN, NICI****. THE NETHERLANDS.
Introduction
Generally chronic sleep onset insomnia in children with attention deficit hyperactivity disorder (ADHD) does not respond well to pharmacological and nonpharmacological treatments (1). Melatonin advances sleep onset in adults with
chronic sleep onset insomnia and delayed onset of endogenous melatonin (2).
Children with delayed sleep onset and bedtime resistance, may also wake up
later (3). This suggests that their sleep-wake rhythm is delayed. Early school
times or early wake-up times of family members can prevent delayed wake up
time. In ADHD children this lack of difficulty to arise in the morning probably
is associated with the hyperarousal conditions of this disorder (4). Consequently a delayed sleep-wake rhythm can easily be masked.
Endogenous melatonin, a hormone produced by the pineal gland during the dark
phase of the day-night cycle, plays a major role in the synchronisation of circadian rhythms (5). As early as the second half of the first year of life melatonin is
involved in the evolution of the sleep-wake system (6). The circadian rhythm of
melatonin is highly reproducible and generally not easily altered (7). The endogenous 24-h. melatonin profile is a reliable marker for circadian phase position.
The time at which melatonin starts to rise in dim light, the Dim Light Melatonin
Onset (DLMO), is shown to be particularly convenient, to assess circadian phase
position, as it can usually be obtained before sleep (8). In adults with chronic
sleep onset insomnia and a delayed sleep-wake rhythm, exogenous administered
melatonin, 5 mg, advances both sleep onset and DLMO (9).
147
We hypothesized that administration of exogenous melatonin advances also
sleep onset in children with ADHD and chronic sleep onset insomnia.
Methods.
This randomized placebo-controlled study evaluated the efficacy of oral melatonin, 5 mg day-1, at 18:00 h. in primary school children with ADHD and chronic
sleep onset insomnia. 40 children with chronic insomnia were randomly allocated to melatonin or placebo treatment. In 26 ADHD was diagnosed ( DSM-IV
criteria). 11 children were randomly assigned to melatonin and 14 to placebo
treatment during 4 weeks (fig.1.).
40 participants randomized
26 with ADHD
16 placebo
1 excluded
Figure 1:
14 without ADHD
10 melatonin
15 placebo
11 sleep onset*
10 sleep onset*
13 lights off time*
10 lights off time*
12 wake up time*
10 wake up time*
13 sleep latency*
10 sleep latency*
12 total sleep time*
10 total sleep time*
11 sleep onset**
8 sleep onset**
12 DLMO
6 DLMO
13 sustained attention
9 sustained attention
13 Connors Scale
4 Connors Scale
Trial profyle. *: diary. **: actigraphy. DLMO: Dim Light Melatonin Onset.
Numbers are numbers of patients involved.
In the week before the start of the treatment and during the fourth treatment
week, sleep was assessed with diaries and actigraphy, severity of ADHD with
the Connors Scale, and sustained attention with the Bourdon -Vos test by meas-
148
uring Rule Completion Time (RCT) (10). Salivary melatonin was collected hourly
between 18:00 and 23:00 h. in dim light at the last night of the baseline week
and at the last night of the fourth treatment week. The children did not take
melatonin tablets at the last night of the fourth treatment week. After the trial
all children received “open” melatonin. Treatment effects were analyzed with
ANOVA repeated measures procedure with 2 factors, i.e. treatment (melatonin
versus placebo) and measurement (baseline versus fourth treatment week).
Significant treatment x measurement interactions, indicating a differential treatment effect across groups, were followed by post hoc comparisons of the baseline with treatment measurements, using paired T-Test; the between-group differences in the mean change from baseline to fourth treatment week were
analysed using T-Test.
Results
Melatonin treatment advanced lights-off time, diary and actigraphic sleep onset, and melatonin onset, decreased sleep latency and increased total sleep
time (table 1). Wake up time, Connor score and RCT did not change. One year
after the trial all children still used melatonin. In 15 children behavior had
improved remarkably. Adverse effects did not occur.
baseline week
melatonin
fourth treatment week
melatonin
F
P
Lights off time* (h:min) 21:21 (0:37)
20:56 (0:33)
20:51 (0:53) 2057 (1:00)
6.12
0.022
Sleep onset* (h:min)
22:26 (0:45)
22:02 (0:28)
21:51 (1:10) 21:51 (0:34)
10.28
0.004
Sleep latency* (min)
63.3 (29.2)
65.4 (32.2)
25.5 (24.3) 55.5 (35.3)
7.19
0.014
Wake up time* (h:min)
7:30 (0:17)
7:19 (0:22)
7:13 (0:28)
7:01 (0:21)
0.13
n.s.
Total sleep time* (h:min) 9:02 (0:39)
9:22 (0:21)
9:56 (0:43)
9:10 (0:46)
21.04
0.001
Sleep onset** (h:min)
22:36 (1:32)
21:21 (0:43)
21:04 (0:54) 22:06 (1:53)
5.56
0.035
DLMO (h:min)
21:19 (2:00)
20:57 (0:42)
20:04 (2:01) 21:10 (1:06)
5.36
0.001
Mean RCT(s)
16.4 (3.9)
17.8 (4.8)
13.7 (2.8)
15.10 (3.9)
0.74
n.s.
Connors Scale
1.64 (0.27)
1.78 (0.15)
1.98 (0.37)
2.02 (0.44)
0.39
n.s.
Table 1:
placebo
treatment interaction
placebo
Mean (SD) sleep parameters, Dim Light Melatonin Onset (DLMO) and
Mean Rule Completion Time (RCT) at baseline and fourth treatment week.
*: Diary. **: Actigraphy.. Treatment interaction determined by ANOVA with
repeated measures statistics. F: F ratio. P: level of significance. N.S.: not
significant (P >0.05)
149
Discussion
The present study showed that one month melatonin treatment advanced mean
lights-off time, sleep onset, sleep latency and DLMO and increased sleep duration in the children with chronic sleep onset insomnia. Sustained attention
and Connors score was not affected. Serious side effects did not occur.
Seeing that normal values of DLMO in 6 – 12 years-old children have not been
published, it is not known whether the mean baseline DLMO, which occurred in
our group children around 21 p.m. is abnormal. However, Carskadon et al (11).
found that mean DLMO occurs at 20:24h in 14-years old adolescents. One year
later DLMO was delayed about 40 minutes. So the mean DLMO before treatment, which we found in our patients, probably is later than what might be expected. This late DLMO suggests that the endogenous circadian pacemaker might
be delayed (8) in at least some ADHD children with chronic sleep onset insomnia. This can be due to dysfunction of clock genes (12, 13), enzymes involved in
the melatonin synthesis (9) or neural connections between the retina and pineal
gland (14). Melatonin, at least in the short term, seems to be an effective and safe
treatment for chronic insomnia in children with ADHD. As long as toxicity and
long term effects of melatonin not have been studied, we recommend prescribing melatonin only in well-performed trials.
Acknowledgement
This study was financially supported by the “Jan Dekker en dr. Ludgardine
Bouwman” Foundation.
Reference List
1 Dahl RE, Pelham WE, Wierson M. The role of sleep disturbances in attention deficit disorder symptoms: a case study. J Pediatr Psychology 1991; 16: 229-239.
2 Nagtegaal JE, Kerkhof GA, Smits MG, Swart ACW, van der Meer YG. Delayed sleep phase
syndrome: A placebo-controlled cross-over study on the effects of melatonin administered
five hours before the individual dim light melatonin onset. J Sleep Res 1998; 7: 135-143.
3 Blader JC, Koplewicz HS, Abikoff H, Foley CA. Sleep problems in elementary school children. A community survey. Arch Pediatr Adolesc Med 1997; 151: 473-480.
4 Ring A, Stein D, Barak Y, et al. Sleep disturbances in children with attention-deficit/hyperactivity disorder: a comparative study with health siblings. J Learn Disabil 1998; 31: 572578.
5 Arendt J. Melatonin and the mammalian pineal gland. London: 1995; 161-281.
6 Sadeh A. sleep and melatonin in infants: a preleminary study. sleep 1997; 20: 185-191.
7 Reiter RJ, Richardson BA. Some perturbations that disturb the circadian melatonin rhythm.
Chronobiol.Int. 1992; 9: 314-321.
150
8 Lewy AJ, Cutler NJ, Sack RL. endogenous melatonin profile as a marker for circadian
phase position. J.Biol.Rhythm. 1999; 14: 227-236.
9 Nagtegaal JE, Kerkhof GA, Smits MG, Swart ACW, van der Meer YG. Delayed sleep phase
syndrome: A placebo-controlled cross-over study on the effects of melatonin administered
five hours before the individual dim light melatonin onset. J Sleep Res 1998; 7: 135-143.
10 Vos PG. Bourdon-Vos test. Manual. Swets & Zeitlinger; Swets Test Services (Rev. ed.).
Lisse, The Netherlands: 1998;
11 Carskadon MA, Wolfson AR, Acebo C, Tzischinsky O, Seifer R. Adolescent sleep paterns,
circadian timing and sleepiness at a transition to early school days. sleep 1999; 21: 871881.
12 Kay A. PAS, present and future:clues to the origins of circadian clocks. Science 1997; 276:
753-754.
13 Horst van der TJ, Muijtjens M, Kobayashi K, et al. Mammalian Cry1 and Cry2 are essential
for maintenance of circadian rhythms. Nature 1999; 398: 627-630.
14 Nagtegaal JE, Kerkhof GA, Smits MG, Swart ACW, van der Meer YG. Traumatic brain
injury-associated delayed sleep phase syndrome. Functional Neurology 1997; 12: 345348.
151
152
Modelling the relation of body temperature
and sleep: importance of the
circadian rhythm in skin temperature
EUS J.W. VAN SOMEREN
NETHERLANDS INSTITUTE FOR BRAIN RESEARCH, AMSTERDAM
A close relation between thermoregulatory and arousal related processes in
basal forebrain and hypothalamic areas has been substantiated in many studies. It has been demonstrated especially - but not exclusively - in the preoptic
area of the anterior hypothalamus (POAH), that a subpopulation of warm-sensitive neurons (WSNs) spontaneously increases firing rate at sleep onset, and
that experimental local warming of the POAH induces a similar increase in
firing rate and facilitates sleep onset. It has consequently been proposed that
brain temperature may be involved in physiological regulation of sleep4. However, contrary to the experimental findings of increased sleep onset probability
with a locally increased hypothalamic temperature, the likeliness of sleep onset in unmanipulated conditions is actually minimal at the time when the circadian rhythm in temperature reaches its peak. In fact, sleep onset probability
increases on the falling limb of the circadian core temperature rhythm. How
can this discrepancy of increased sleep with well-controlled experimental local
POAH warming be reconciled with increased sleep with local cooling under
natural conditions?
A suitable explanation has been put forward5: in modelling based on local warming experiments, it has generally been ignored that many of the locally
thermosensitive neurons also respond, in a similar way, to changes in skin
temperature. Thus, the very changes in cell membrane properties leading to
sleep-related alterations in firing rate as induced by local warming in experimental conditions, may be induced by warming of the skin under natural conditions.
Several considerations support the importance of skin temperature. First, after
warming of the skin, preoptic WSNs show a markedly increased firing rate to a
level that can be reached only by extreme (non-physiological) local warming.
Moreover, with elevated skin temperature, WSNs have a high firing rate irrespective of the local temperature, i.e. temperature input from the skin appears
to dominate when competing signals are present1, 2. Second, sleep appetitive
behavior like lying down, covering etc. is associated with a redistribution of
warm blood to the extremities, thus increasing their temperature. Third, and
153
most important, the increase of sleep onset probability that occurs on the falling limb of the circadian core temperature rhythm can now be understood.
Since the circadian drop in core temperature is mainly due to increased dissipation of body heat, skin temperature is actually elevated when core temperature is falling.
The preoptic area is by no means exclusive in containing thermosensitive neurons: they are present in many of the brain areas involved in arousal regulation. When sleep onset probability was modelled with human constant routine
data3 of the circadian profiles of core and peripheral temperature as inputs to
thermosensitive neurons in these brain areas, we found agreement between
habitual sleep onset time and peak modelled onset probability, as shown in the
figure. A series of studies is presently being undertaken in order to reveal the
relative importance of core and peripheral temperature in the regulation of
arousal state transitions.
Figure:
The upper panel shows the
average 24-hour profile of
human rectal temperature
and temperature at the extremities under constant
routine conditions3. For convenience, the curves have
been moved freely on the vertical axis, omitting absolute
values on the vertical axis,
while the relative magnitude
of the amplitudes is preserved in the figure. Both
curves affect the probability
of sleep-type neuronal firing
patterns of thermosensitive
neurons in many parts of the
brain involved in sleep regulation. High skin temperature is associated with increased probability of
sleep-type firing patterns in
most brain areas. High core
temperature is associated with increased probability in some areas, but
decreased probability in other areas. The lower panel shows a straightforward (normalized) summation of the resulting sleep probability curves
154
of six sleep-related brain structures for which thermosensitivity has been
demonstrated. Depending on the setting of a threshold level, a temperature window favourable to sleep onset can thus be determined (indicated with dashed lines). Note that maximal sleep onset probability is
indeed at usual bedtime.
Acknowledgement
Research supported by NWO, The Hague, Successful Aging Project 014-90-001
References
1 Boulant JA and Bignall KE (1973) Hypothalamic neuronal responses to peripheral
and deep-body temperatures. Am J Physiol 225:1371-1374.
2 Boulant JA and Hardy JD (1974) The effect of spinal and skin temperatures on the
firing rate and thermosensitivity of preoptic neurones. J Physiol 240:639-660.
3 Kräuchi K and Wirz-Justice A (1994) Circadian rhythm of heat production, heart
rate, and skin and core temperature under unmasking conditions in men. Am J Physiol
267:R819-R829.
4 McGinty D and Szymusiak R (1990) Keeping cool: a hypothesis about the mechanisms and functions of slow-wave sleep. Trends Neurosci 13:480-487.
5 Van Someren EJW (1997) Rest-Activity Rhythms in Aging, Alzheimer’s Disease and
Parkinson’s Disease. Methodological Developments and Therapeutic Interventions.
Ph.D. Thesis, Medical Faculty of the University of Amsterdam, pp 303, University of
Amsterdam, Amsterdam.
155
156
Dynamics of cortical EEG power decrease rate
during entry into natural hypothermia
in European ground squirrels
ARJEN M. STRIJKSTRA*, TOM DEBOER& & SERGE DAAN*.
&
* ANIMAL BEHAVIOR, UNIVERSITY OF GRONINGEN, THE NETHERLANDS
DEPARTMENT OF PHARMACOLOGY, UNIVERSITY OF ZÜRICH, SWITZERLAND
CORRESPONDENCE TO ARJEN M. STRIJKSTRA, ANIMAL BEHAVIOR, UNIVERSITY OF
GRONINGEN, P.O. BOX 14, NL-9750 AA, HAREN, THE NETHERLANDS.
TEL (31) 50 3637658; FAX (31) 50 3635205;
E-MAIL [email protected]
Introduction
Hibernation has interested sleep researchers for decades. The cortical EEG
patterns during NREM sleep are similar to EEG’s during the first stages of entrance into natural hypothermia (also called torpor). In spite of this resemblance in EEG patterns, torpor and sleep are functionally dissimilar. Short daily
torpor bouts (circa 6 hours) in Djungarian hamsters induced an increase in
slow wave activity (SWA), which can be interpreted as the result of a sleep debt
increase during prior torpor (Deboer & Tobler 1994). Longer torpor bouts (212 days) in hibernating ground squirrels induced changes in spectral EEG power
during subsequent euthermy, including the spindle (~10-14Hz) and slow wave
(~1-4Hz) ranges (Trachsel et al. 1991; Strijkstra & Daan 1997ab). The spectral
changes in hibernators were not related to a sleep debt increase caused by
prior torpor (Strijkstra & Daan 1998; Larkin & Heller 1999), and may be based
on changes in neurotransmission capacity and neuronal connectivity in the
brain (see Strijkstra 1999). The findings in both Djungarian hamsters and ground
squirrels disprove the idea that torpor and sleep have similar physiological
functions.
The physiological changes in the brain of hibernating ground squirrels are most
likely produced by effects of the low temperature of neurons during deep torpor. Cortical EEG recordings during the temperature cline at the entry into
torpor may reveal information on these effects. We report here on gradual and
more abrupt temperature dependencies of cortical EEG power decrease rates
during entry into torpor in European ground squirrels.
157
Methods
Animals
We used EEG recordings of 8 animals in the hibernation seasons of 1993-1994
and 1994-1995. During hibernation, the animals were housed individually in
60*40*40(l*w*h) cm steel cages on sawdust bedding at an ambient temperature of 5.5°C and continuous dim light (<1 Lux). Hay was provided as nest
material, and animals had continuous access to water and food.
Prior to hibernation, EEG and EMG electrodes were implanted under deep
anesthesia (1993: Pentobarbital, 60mg/kg i.p., 1994: Halothane, 2.5% in air). Silver EEG electrodes were placed on the dura above the parietal cortex and the
cerebellum. Stainless steel EMG electrodes (MS303, Plastics One inc.) were placed
under the skin on the neck muscles. Two stainless steel screws were placed through
the skull above the frontal cortex, and served as ground. A pre-calibrated thermistor, sealed in a glass capillary (1 mm diameter), was placed on the dura above
the frontal cortex. All leads were fixed to a connector, which was anchored with
dental cement to the two ground screws, and an additional screw placed in the
skull above the parietal cortex on the contra-lateral side of the EEG electrodes.
At least two weeks at 20°C were allowed for recovery. Recordings were made
during the second half of hibernation, from December until early March.
Recordings
All recordings took place in the animal’s home cage. Cortical temperature and
cortical EEG’s were recorded simultaneously during entry into torpor. Animals
were connected with flatcable via a slip ring swivel (Air Precision, France) to an
amplifier system (EEG: 0.2mV/V, 0.2-80Hz; EMG: 0.5mV/V, 20-600Hz; Twente
Technology Transfer, The Netherlands). Data acquisition, storage and additional
software filtering (-3dB at 17Hz, 35dB per octave) was performed with the commercial PC based EEG data acquisition and processing package POLY (vs.4.7; Inspector Research Systems, The Netherlands). EEG’s were sampled at 100Hz and
analyzed on line with Fast Fourier Transformation per 10 sec epoch.
Visual scoring of the vigilance states wakefulness, NREM sleep and REM sleep was
done on all EEG’s at cortical temperatures of >25°C. Between cortical temperatures of 27-25°C, signs of REM sleep (e.g., a reduced amplitude and a concurrent
acceleration of the EEG frequency, combined with a low EMG level) rarely occurred. Below 25°C, long lasting (>20 sec) reduction of EEG amplitude coincided
with an increase in EMG. EMG activity was used to differentiate between wakefulness and sleep EEG’s in those conditions. Because REM sleep did not occur below
25°C (see also Deboer & Tobler 1994, 1995), we treated the sleep EEG below 25°C
as NREM sleep EEG in the analyses. NREM sleep cortical EEG power over the 0.215Hz range was integrated and averaged over 30 min intervals for analysis.
158
Results
Cortical temperature and vigilance states
Figure 1 shows the average cortical temperature (Figure 1A) and the visually
scored vigilance state percentages (Fig. 1B) during entry into torpor, averaged
over 30 min intervals. Cortical temperature (Fig. 1A) showed a monotonic decrease from the near euthermic level of circa 32°C towards the (near to the
expected asymptotic) value of circa 7.5°C. During cooling, the EEG pattern
changed from a regular NREM to REM sleep alternation to an alternation between a state with NREM like features, e.g., relatively low frequency waves, and
a stage with wakefulness features, e.g., relatively low amplitude waves and an
increase in EMG activity.
REM sleep was observed at cortical temperatures of >25°C, during the first
2.5h of entry into torpor (Fig. 1B). A substantial amount of wakefulness occurred during the first 9h of entry into torpor (Fig. 1B), until cortical temperature was below circa 10°C.
Figure 1:
Cortical temperature
(mean ± SD) and vigilance
states assigned to cortical
EEG patterns (mean value)
during entry into torpor in
hibernating European
ground squirrels (contributing data per point range
between 5 and 8).
Cortical temperature and cortical EEG power
In Figure 2A, the logarithm of cortical EEG power for the cooling stage is shown
as a function of the average cortical temperature per 30 min interval. Log(EEG
power) decreased monotonically, but this decrease showed two discontinuities.
(1) Log(EEG power) decreased at a low rate between euthermy and 25°C, but
the decrease rate appeared to accelerate below 25°C. (2) A similar acceleration
of decrease rate of log(EEG power) appeared to occur at a cortical temperature
159
of 15°C. At 8°C, log(EEG power) stabilized at a very low level: the cortical EEG
is near iso-electric at temperatures below circa 10°C.
The first derivative of the log(EEG power), indicating the cortical EEG decrease
rate, is shown in Figure 2B. The dotted line represents an expected decrease
rate based on the effect of temperature on biochemical processes, e.g., a Q10
effect. The expected Q10 value of 2.5 was the average of observed Q10 values for
temperature dependent reduction of cortical EEG frequencies in REM and NREM
sleep EEG’s during torpor in Djungarian hamsters (Deboer & Tobler 1995). At
cortical temperatures of >27°C and <9°C, the cortical EEG power decrease
rate was not significantly different from the expected Q10 value. Significantly
higher cortical EEG power decrease rates were found between 27°C and 9°C
(solid line at bottom of Figure 2B).
Figure 2:
Logarithm of cortical
NREM sleep EEG power
(panel A) and its first derivative (panel B) as a
function of cortical temperature. Panel B: dotted
line indicates a Q10 value
of 2.5, solid lines indicate
significant deviation of
data from the plotted Q10
value (t-tests, p<0.05).
Discussion
Cortical EEG power during entry into torpor was not only simply affected by
Q10 effects. The data indicate that at specific temperatures, the decrease rate
in cortical EEG power of European ground squirrels during entrance into natural hypothermia or torpor, was enhanced.
In our ground squirrels, REM sleep was reduced at sub-euthermic temperatures, and was absent at temperatures below circa 25°C. Deboer and Tobler
(1995) found that in sleep EEG’s during entry into torpor, the frequency of the
theta (~7Hz) peak shifted to lower frequencies as a function of temperature,
and could not be recognized in the signal at lower temperatures. Neurons are
apparently unable to maintain theta like activity at temperatures below 25°C.
This may explain the enhanced decrease rate of total EEG power at circa 25°C
160
in our data. The (more extreme) enhancement of EEG power decrease rate at
circa 15°C suggests that also another activity mode of neurons is affected at
that temperature.
Krilowicz et al. (1988) studied activity of thalamic neurons of ground squirrels,
in vivo. Firing patterns of single units were followed during torpor bouts and in
euthermy. A gradual reduction in the occurrence of REM sleep firing patterns
in single units was found between euthermic temperatures down to 21°C, and
REM sleep firing patterns were absent during torpor at temperatures below
21°C. Wakefulness and NREM sleep firing patterns were found during torpor
between near euthermic temperatures down to 15°C. Below 15-18°C, single
units spontaneously ceased spike activity during entry into torpor. Our results
on cortical EEG’s parallel these findings. The faster EEG power decrease rate at
15°C may be related to the observed inactivation of spike activity at that temperature.
The reduction in electrical activity of neurons is based on temperature dependent changes in the functional activity of proteins. Inactivation of the Na+K+-pump has been suggested to play an important role (Deboer & Tobler 1995).
The electrical inactivity during deep torpor in turn may be responsible for
reduced connectivity in deep hibernators. Disuse may cause functional deterioration by a lack of dynamic stabilization (Kavanau 1997). Indeed, besides
observations on anatomical changes (Popov et al. 1992; Strijkstra 1999), also
torpor related deterioration of brain function (e.g. memory performance) has
been found in European ground squirrels (Millesi & Fieder 1998)
In conclusion, electrical activity of the brain is reduced at low temperatures.
Reduction takes place by gradual and more qualitative ‘stepwise’ temperature
effects. The ‘stepwise’ reductions may represent limitations of specific functional activities of neurons at specific temperatures. At 25°C, temperature effects may compromise production of theta frequencies, thus preventing REM
sleep occurrence. At 15°C, temperature effects cause a major impairment of
neuronal activity, which may lead to an impairment of neuronal connections
during torpor. This type of brain damage may thus be specific for deep hibernators, which tolerate extremely low body and brain temperatures, in contrast to
daily torpor animals and shallow hibernators, such as bears.
References
Deboer, T. & I. Tobler (1994) Sleep EEG after daily torpor in the Djungarian hamster :
similarity to the effects of sleep deprivation. Neurosci. Lett. 166:35-38.
Deboer, T. & I. Tobler (1995) Temperature dependence of EEG frequencies during natural hypothermia. Brain Res. 670:153-156.
161
Geiser, F. & T. Ruf (1995) Hibernation versus daily torpor in mammals and birds: physiological variables and classification of torpor patterns. Physiol. Zool. 68:935-966.
Kavanau, J.L. (1997) Memory, sleep and the evolution of mechanisms of synaptic efficacy maintenance. Neurosci. 79:7-44.
Krilowicz, B., S.F. Glotzbach & H.C. Heller (1988) Neuronal activity during sleep and
complete bouts of hibernation. Am. J. Physiol. 255:R1008-R1019.
Larkin, J.E. & H.C. Heller (1999) Sleep after arousal from hibernation is not
homeostatically regulated. Am. J. Physiol. 276:R522-R529.
Millesi, E. & M. Fieder (1998) Effects of hibernation on memory in European ground
squirrels. Advances in Ethology 33:38.
Popov, V.I., L.S. Bocharova & A.G. Bragin (1992) Repeated changes in dendritic morphology in the hippocampus of ground squirrels in the course of hibernation. Neurosci.
48:45-51.
Strijkstra, A.M. & S. Daan (1997a) Ambient temperature during torpor affects NREM
sleep EEG during arousal episodes in hiberating European ground squirrels. Neurosci.
Lett. 221:177-180.
Strijkstra, A.M. & S. Daan (1997b) Sleep during arousal episodes as a function of prior
torpor duration in hibernating European ground squirrels. J. Sleep Res. 6:36-43.
Strijkstra, A.M. & Daan, S. (1998) Dissimilarity of slow wave activity enhancement by
torpor and sleep deprivation in a hibernator. Am. J. Physiol. 275:R1110-R1117.
Strijkstra, A.M. (1999) Periodic euthermy during hibernation in the European ground
squirrel: causes and consequences. PhD. thesis, University of Groningen, The Netherlands.
Trachsel, L., D.M. Edgar & H.C. Heller (1991) Are ground squirrels sleep deprived during hibernation? Am. J. Physiol. 260:R1123-R1129.
Acknowledgments
We thank Roelof Hut for his useful comments on the manuscript.
162
Exploring daytime sleepiness during migraine
J.H.M. TULEN1, D.L. STRONKS1,2, L. PEPPLINKHUIZEN1, J. PASSCHIER2
DEPARTMENTS OF PSYCHIATRY1 AND MEDICAL PSYCHOLOGY & PSYCHOTHERAPY2,
UNIVERSITY HOSPITAL ROTTERDAM – DIJKZIGT AND ERASMUS UNIVERSITY
ROTTERDAM, THE NETHERLANDS
Introduction
Migraine is a paroxysmal disorder, with attacks of unilateral, pulsating headache associated with nausea, vomiting, photophobia and phonophobia (1). It is
estimated to occur in about 10% of the Dutch population, with females being
more affected than males. The repeated recurrence of migraine attacks significantly reduces quality of life and leads to impaired functioning (physically,
emotionally, and socially) both at home and at work (2,3). Dependent on the
severity, a migraine attack can necessitate bed rest and/or reduced activity
level, thereby disturbing normal diurnal rest-activity patterns (and sleep-wake
patterns). The research done on daily functioning during an actual migraine
attack is limited and exclusively based on self-report measures (4). We do not
know how these subjective reports of migraine-induced disability relate to the
behavioral aspects of daily functioning. Yet, quantification of the patient’s actual behavior in connection to their subjective perspective of level of functioning during a migraine attack can provide a better understanding of the relationships between diurnal mobility patterns, migraine and effectiveness of
therapeutic interventions. Sofar, no standardized studies in this area of research have been performed. We evaluated the feasibility of using accelerometry
(5,6) as a method to assess the effects of migraine attacks on normal daily
activities (body posture, physical and locomotor activities) by performing ambulatory 24-hr measurements during a headache free baseline period, as well
as during and after an actual attack (acute treatment varied per patient) in the
habitual environment of migraine patients. Simultaneously, repeated subjective assessments of sleep quality, mood, pain, level of functioning, and daytime
sleepiness were obtained by means of daily logs. In this report we focus on our
findings regarding daytime sleepiness during migraine with the aim to explore
whether a migraine attack induces changes in daytime sleepiness, because
sleepiness may also influence daily functioning. At present, we have no knowledge about the relationships between migraine attacks and feelings of sleepiness, with the exception of reports of increased sedation as a possible sideeffect of various anti-migraine drugs (e.g., 9). We will interpret the sleepiness
163
findings in relation to the effects of the migraine attacks on subjective level of
functioning and daytime fatigue.
Methods
Subjects
Six female migraine patients (mean age: 39.8 years , range: 29 - 49), recruited
by means of advertisements, participated in this study. The diagnosis ‘migraine’
(criteria: Headache Classification Committee of the IHS, 1) was confirmed by a
neurologist of the University Hospital Rotterdam – Dijkzigt. Patients with a
positive history of drug abuse or psychiatric illness, or current medical illness
other than migraine, were excluded from the study. The patients used their
habitual anti-migraine medication for the treatment of the migraine attack,
but did not use prophylactic anti-headache medication during the study; two
patients used no medication during their migraine attack (patients 1 and 3).
Procedures and measurements:
Repeated measurements of 24-hour patterns of physical activities (activity
monitoring by means of body-mounted accelerometers on the upper legs and
trunk), as well as repeated subjective assessments of mood, level of functioning, and sedation (daily logs), were obtained during a headache free baseline
period, as well as during and after a migraine attack, in the habitual environment of the patients. A spontaneous migraine attack was documented by measuring migraine patients from the onset of their migraine attack until two days
after the attack. Apart from these measurements, two consecutive 24-hour recordings (baseline) of the same patients were made during a headache free
period. These baseline measures were scheduled before the measurements
during the migraine attack.
Within the daily log, the following self-rating scales were filled in by the patients at breakfast, lunch, dinner, before sleep, and at onset of migraine and
after 2 and 4 hrs of migraine: a) the Profile Of Mood States (POMS; validated
Dutch version; 7): the POMS comprises 5 subscales: vigor, fatigue, tension,
anger, and depression, b) the Level of Functioning scale (LOF; Passchier et al.,
in preparation): the LOF consists of a short Guttman scale with items ranging
from ‘is only capable of lying on bed’ to ‘is capable to perform heavy physical
activities’, and c) the Stanford Sleepiness Scale (SSS, Dutch translation; 8): the
SSS consists of a 7-point scale with items ranging from ‘feeling active, alert’ to
‘cannot stay awake, sleep onset appears imminent’. We will report the findings
of the SSS, the LOF scale, and the subscale Fatigue of the POMS, of the baseline period and at the onset of measurements during the migraine attack (when
all patients were still drug-free).
164
Statistical analysis
Per patient, the migraine data were evaluated versus the mean values obtained during the second day of the baseline period (first day was considered an habituation
period, although the mean data of the first day did not significantly differ from the
mean data of the second day). Per parameter, differences between the two conditions were analyzed by means of Wilcoxon tests. Because patients 1 and 3 used no
anti-migraine treatment, the complete SSS data of these patients during baseline,
migraine and recovery periods are presented as illustrations of changes in sleepiness
during untreated migraine.
Results
Effect of migraine on sleepiness, fatigue and level of functioning (fig.1)
The migraine attacks of the patients varied in severity at the onset of measurements
between moderate (patients 2 and 4) and severe (patients 1, 3, 5, and 6). Patients 1
and 2 were able to continue their daily activities, whereas patients 3, 4, 5, and 6 had
to lie down during at least part of their attack period. Migraine caused an increase in
Sleepiness, as compared to baseline, in all patients (increase always > 25%; baseline
mean[sd]: 2.6[0.8], migraine: 5.5[1.1]; Wilcoxon test: p<.05; fig.1). Although Level
Of Functioning decreased significantly during migraine in comparison with baseline
values (baseline: 15.6[7.1], migraine: 2.3[1.1]; p<.05; fig.1), it did not lower much in
patients 1 and 2 (about 10%) who were able to continue their activities. Migraine also
significantly increased subjective feelings of fatigue (baseline: 2.6[1.2], migraine:
15.0[7.4]; p<.05; fig.1); only in patient 2 this increase was relatively small (13%).
Figure 1:
Per patient, the change during a migraine attack (versus baseline) in sleepiness, level of functioning, and fatigue, expressed as percentage of the
maximal range of each scale. SSS: Stanford Sleepiness Scale; LOF: Level
Of Functioning scale; Fatigue: subscale fatigue of the Profile Of Mood States.
165
Subjective Sleepiness during untreated migraine (fig.2)
Patient 1: The migraine attack of patient 1 was monitored from about 13:00 hrs
onwards. Headache at starting point was severe, but soon reduced to mild intensity (after 2 hrs); the patient reported a significant improvement of headache after 4 hrs. During the migraine attack sleepiness was increased, although
the patient was able to continue with her daily activities. The sleepiness scores
of the recovery day resembled the data obtained under baseline conditions
(fig.2a).
Patient 3: This patient suffered from a very severe migraine attack, lasting 2
days; the patient stayed in bed from the onset of measurements (at about 13:00
hrs) till most of the next day. A significant improvement of headache was reported after about 10 hrs of measurement. During this migraine period the
patient scored maximal on the SSS (maximum score=7); also during the first
recovery day, when the patient still spend 84% of the daytime period in bed,
the sleepiness scores remained elevated in relation to the daytime period during baseline (fig.2b).
Figure 2:
Sleepiness during untreated migraine: SSS scores during baseline, migraine and subsequent recovery periods of patients 1 (A) and 3 (B).
Discussion
Migraine constitutes a complex set of symptoms, the most important one being
severe pain. Apart from this feature, migraine-induced disability consists of a
range of complaints comprising reduced quality of life, reduced level of functioning, and lowered mood. In this exploratory study, we have provided some
first evidence showing the presence of sleepiness during untreated migraine.
Increased sleepiness occurred in all patients. It appeared not unequivocally
related to requirements of bed rest or reduced level of functioning, yet, to a
certain extent it did correspond with increased feelings of fatigue. The impor-
166
tance of increased sleepiness and fatigue during migraine and their possible
connection with alterations in diurnal patterns of behavioral activities need to
be further clarified in studies with more subjects before and during (acute treatment of) migraine. The fact that some effective anti-migraine drugs may induce sedation as a side-effect (9) makes the study of these relationships more
relevant but also more complex.
The study was supported by a grant from Glaxo Wellcome BV.
References
1 Headache Classification Committee of the International Headache Society: Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial
pain. Cephalalgia 8(suppl 7):10-73, 1988.
2 Kryst S, Scherl E. A population-based survey of the social and personal impact of
headache. Headache 34:344-350, 1994.
3 Passchier J, Andrasik F. Migraine, Chapter 31. Psychological Factors. In: Olesen J,
Tfelt-Hansen P, Welch KMA, eds. The Headaches New York, Raven Press, 1993, pp:233245.
4 Stewart WF, Lipton RB, Kolodner K, Liberman J, Sawyer J. Reliability of the migraine
disability assessment score in a population-based sample of headache sufferers. Cephalalgia 19:107-114, 1999.
5 Bussmann JBJ, Reuvekamp R, Veltink PH, Martens WLJ, Stam HJ. Validity of an
instrument for ambulatory measurement of mobility activities. Pain 74:153-161, 1998.
6 Tulen JHM, Bussmann JBJ, van Steenis HG, Pepplinkhuizen L, Man in ‘t Veld AJ. A
novel tool to quantify physical activities: ambulatory accelerometry in psychopharmacology. J Clin Psychopharmacol 17:202-207, 1997.
7 Wald FDM, Mellenbergh GJ. De verkorte versie van de nederlandse vertaling van de
Profile Of Mood States (POMS). Ned T Psychol 45:86-90, 1990.
8 Hoddes E, Zarcone V, Smythe H, Phillips R, Dement W. Quantification of sleepiness:
a new approach. Psychophysiology 10:431-436, 1973.
9 Schoenen J. Acute migraine therapy: the newer drugs. Current Opinion in Neurology 10:237-243, 1997.
167
168
Effect of non-pharmacological treatment
on polysomnography, sleep/wake diary
and questionniares in patients
with primary insomnia
INGRID VERBEEK (1) & YVON SWEERE (2)
1 CENTER FOR SLEEP AND WAKE DISORDERS KEMPENHAEGHE, HEEZE
2 CENTER FOR SLEEP AND WAKE DISORDERS MCH WESTEINDE HOSPITAL,
THE HAGUE
Introduction
Primary insomnia is a subjective complaint of disturbed duration, quality and efficiency
of sleep. When the criteria of the International Classification of Sleep Disorders (ICSD)
and the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) are combined,
the definition of primary insomnia is as follows:
• primary complaint of sleep onset and/or sleep maintenance insomnia or non-refreshing
sleep during at least one month
• the sleep complaints lead to dysfunction during the day (social, work)
• the sleep disorder does not occur exclusively in the course of narcolepsy, sleep related breathing disorder, circadian rhythm disorder or parasomnie.
• the sleep disorder does not occur exclusively in the course of a psychiatric disorder
(depression, general anxiety disorder)
• the sleep disorder is not the result of direct physiological effects of drugs or medicine
or a somatic disorder
• there are indications of learned sleep-preventing associations (trying too hard too
sleep, conditioned arousal to bed-room or sleep-related activities).
Primary insomnia is sustained by behavioral, cognitive and physiological factors. Non pharmacological therapy (NPT) has been proven effective in treating
primary insomnia (1-6). Most studies used subjective measures to measure the
effects of NPT on sleep. Despite the fact that insomnia in the first place is a
subjective complaint, objective evaluation of NPT is important because
polysomnography (PSG) is still seen as the golden standard in the evaluation of
sleep. Furthermore, the influence of NPT on quality of life and psychological
well being has only been established in a few studies.
In this ongoing two-center study we investigated the effect of short-term NPT
on polysomnography (PSG), sleep/wake diary (SD) and various questionnaires.
169
Subjects
Subjects were 22 hypnotic-independent primary insomniacs (13 females, 9
males) with a mean age of 44.1 years (range 25-58) and a mean duration of
insomnia of 16.9 years (range 1-50). All patients were referred to the sleep
center by their general practitioner or specialist. Inclusion criteria were a sleep
onset latency and/or wake after sleep onset of ≥ 30 minutes and a sleep efficiency ≤ 85% on more than 3 nights a week. Moreover, the insomnia had to
exist ≥ 1 year. Exclusion criteria were excessive alcohol consumption (> 22
consumption’s a week), hypnotic use (≥ 3 nights a week) and the use of antidepressants. Psychopathology was excluded by anamnesis and questionnaires.
Other sleep disorders were excluded by sleep history and PSG. During the study,
patients were not allowed to follow other treatments for insomnia.
Methods
Treatment
All subjects underwent cognitive-behavioral therapy (sleep hygiene, stimulus
control, sleep restriction, relaxation exercises, and cognitive therapy) in 6 weekly
sessions.
Measures
Two nights of polysomnography, one week sleep/wake diary, and questionnaires
were examined at 3 time points: at baseline, after 3 months waiting period, and
one month after therapy. Outcome measures were sleep onset latency, total
sleep time, sleep efficiency, wake after sleep onset and number of times awake
as measured by PSG and SD, and the amount of slow wave sleep as measured
by PSG (mean of two nights). In the questionnaires quality of life (QOL) was
measured by selected items from two validated Dutch questionnaires: the sickness impact profile (SIP, 7-8) and the RAND-36 (9-10). Sleep was subjectively
evaluated with a self developed Sleep Evaluation Form (SEF), psychological
well-being with the Symptom Check List (SCL-90,11) and Beliefs and Attitudes about Sleep (BAS,1).
Statistics
General Linear Model factorial analysis was used to determine variances over
different time points in PSG and SD. Mann-Whitney test was used for the data
from the questionnaires. The level of significance was set at p≤0.05.
Results
Results are shown in table 1-3: table 1 shows polysomnographic changes, table
2 the results from the sleep/wake diaries and table 3 the results from the questionnaires. Results are shown at baseline (A), after 3 months waiting (B) and
170
after therapy (C). The number of PSG data is limited because the protocol was
quite demanding for patients.
No significant improvements were seen during the waiting period in either
polysomnography, sleep/wake diaries or questionnaires (table 1-3: A-B).
After treatment, PSG showed significant decrease in sleep onset latency, wake
after sleep onset and total sleep time (table 1). The number of times awake was
decreased nearly significant.
In the SD, sleep onset latency and sleep efficiency were significantly improved
after therapy (table 2). Large discrepancies are seen between sleep parameters
measured by PSG or SD (table 1,2).
The QOL data after therapy show significant improvements in problems with
work, social occupation, the way patients feel, the number of social interactions and the number of recreational activities (table 3).
The SEF shows large significant improvements in satisfaction and coping with
sleep, concern about sleep and daytime complaints (tiredness, mood). Concentration problems during the day (concentration) was improved nearly significant.
The psychological data as assessed by the SCL-90 show significant improvements in anxiety, depression, somaticization and sensibility.
The BAS shows that attitudes and beliefs about sleep are significantly improved
in the way that patients experience more control over the sleep problem, are
more realistic in the consequences and causes of insomnia and have improved
bad sleep habits. Interestingly, no significant improvement is seen in ‘unrealistic expectations’.
171
Table 1:
Polysomnography at baseline (A), after 3 months waiting (B), and after
therapy (C). SOL= sleep onset latency, TST= total sleep time, SE= sleep
efficiency, WASO= wake after sleep onset, AWAK= number of awakenings,
SWS= amount of slow wave sleep. GLM=general linear model (general
factorial). *p≤0.05
SOL (min.)
TST (min.)
SE (%)
WASO (min)
AWAK
SWS (min.)
Table 2:
A: baseline
B: 3 months
(n=10)
waiting (n=16)
C: after therapy
(n=14)
18.45
407.05
83.02
43.20
10.20
79,25
17.22
417.75
81.66
48.06
9.56
73,75
9,50
384.61
87.55
17,64
5,82
82,46
GLM
A-B A-C B-C
*
*
*
*
0.059
Sleep diaries at baseline (A), after 3 months waiting (B), and after therapy
(C). SOL= sleep onset latency, TST= total sleep time, SE= sleep efficiency,
WASO= wake after sleep onset, AWAK= number of awakenings. *p≤0.05,
ᔤp≤0.01
SOL (min.)
TST (min.)
SE (%)
WASO (min)
AWAK
A: baseline
B: 3 months
(n=22)
waiting (n=21)
(n=22)
77.74
255.11
49.54
63.65
1.61
45.66
267.14
55.59
66.43
1.48
29.18
321.75
74.16
39.26
1.27
172
C: after therapy
GLM
A-B A-C B-C
*
ᔤ
*
Table 3:
Questionnaires at baseline, after 3 months waiting and after therapy.
Quality of life is measured by RAND-36 and Sickness Impact Profile (SIP).
SEF= sleep evaluation form, SCL-90= symptom check list, BAS= beliefs
and attitudes about sleep. Improvement on RAND-36 and SEF show higher
scores. Improvement on SIP, SCL-90 and BAS show lower scores. *p≤0.05,
ᔤp≤0.01
(min-max)
A: baseline
B: 3 months
C: after therapy
(n=21)
waiting (n=22)
(n=22)
health (1-5)
3.35
3.05
3.09
work (5-10)
7.30
7.73
Mann-Whitney
A-B
A-C
B-C
8.55
*
0.057
RAND-36
social occupation (1-5)
3.76
4.05
4.40
*
feeling (9-54)
34.48
35.91
39.45
*
3.90
3.64
2.05
ᔤ
*
SIP
social interaction (0-20)
alertness 0-10)
3.52
3.23
2.32
recreation (0-8)
1.90
1.91
1.00
*
*
satisfaction with sleep
1.95
1.86
3.32
ᔤ
ᔤ
coping with sleep
2.05
2.14
3.55
ᔤ
ᔤ
concern about sleep
2.10
2.18
3.18
ᔤ
ᔤ
tiredness
2.24
1.95
2.95
ᔤ
concentration
2.62
2.55
3.09
mood
3.05
3.18
3.55
*
SEF (all 1-5)
ᔤ
0.056
SCL-90
anxiety (10-50)
15.43
13.38
12.36
*
depression (16-80)
25.86
23.38
21.32
ᔤ
somaticization (12-60)
17.95
16.81
14.95
*
insufficiency (9-45)
26.19
24.57
23.68
sensibility (18-90)
19.95
18.29
16.27
0.063
ᔤ
BAS
expectations (2-10)
7.52
4.50
3.73
control (7-35)
19.71
22.23
15.95
*
ᔤ
consequences (6-30)
19.29
18.82
16.55
*
*
causes (1-5)
4.71
2.73
2.36
*
sleep habits (7-35)
14.43
15.91
12.95
173
*
Conclusions
We conclude that short-term non-pharmacological treatment of insomnia improves both subjective and objective sleep. Subjective improvements are seen
in quality of life, attitudes and beliefs about sleep, psychological distress and
sleep evaluation in general. The subjective improvements in quality of life and
psychological distress seem to be related to a better sleep and better coping
with the sleep problem. Improvements in sleep lead to an improvement in the
quality of the day, therefore, insomnia has to be seen as a 24-hour problem.
references
1 Morin CM. Insomnia: Psychological Assessment and Management. New York, NY:
Guilford Press; 1993.
2 Espie CA. The psychological treatment of insomnia. Chichester: John Wiley & Sons
Ltd, 1991.
3 Klip EC, Zwart FM, Jansma K. Leren slapen in groepen. Single-group experimenten
in de behandeling van slaapstoornissen. Nederlands Tijdschrift voor de Psychologie.
1978; 33: 1-15.
4 Oosterhuis A, Klip EC. Beter slapen: effectiviteit van de gedragstherapeutische
slaapcursus van de Kruisverenigingen. Gedragstherapie 1993; 26: 191-201.
5 Verbeek I, Schreuder KE, Declerck AC. Evaluation of short-term non-pharmacological treatment of insomnia in a clinical setting. Journal of Psychosomatic Research
(in press).
6 Hauri PJ. Can we mix behavioral therapy with hypnotics when treating insomnia?
Sleep; 20(12):1111-1118, 1997.
7 Bergner M, Bobbitt RA, Pollard WE, Martin DP, Gilson BS. The Sickness Impact
Profile: Validation of a health status measure. Medical Care, 14, 57-67, 1976a.
8 Vakgroep Huisartsgeneeskunde, Universiteit Utrecht. De Sickness Impact Profile,
Nederlandse versie, 1987.
9 Brazier JE, Harper R, Jones NMB, Cathain A, Thomas KJ, Usherwood T, Westlake L.
Validating the SF-36 health survey questionnaire: outcome measure for primary care.
British Medical Journal, 305, 160-164, 1992.
10 Van der Zee KI, Sanderman R. Het meten van de algemene gezondheidstoestand met
de RAND-36, een handleiding. Noordelijk Centrum voor Gezondheidsvraagstukken,
Rijksuniversiteit Groningen, 1993.
11 SCL-90: Symptom Check List, Dutch version. Swets and Zeitlinger, Lisse, 1986.
174
Night-to-night variability of apnea indices
M.M.R. VERHELST, R.J. SCHIMSHEIMER, C. KLUFT, A.W. DE WEERD
CENTRE FOR SLEEP AND WAKE DISORDERS, MCH, WESTEINDE HOSPITAL,
THE HAGUE
In our centre, the diagnosis of OSAS is made by 2x24 hours ambulatory
polysomnography. This approach has the advantage that it provides insight not
only in respiratory parameters but also in sleep quality and excessive daytime
sleepiness. In view of financial restrictions by the government, the question
arose if a single night screening might be sufficient. The answer to this question is also important for another reason : CPAP treatment is given when the
apnea-index (AI) is ᔤ15. Can such an important therapeutical decision be taken
on the basis of a single night recording?
On the other hand little is known about night-to-night variability of apnea indices in ambulatory polysomnography. This study therefore investigates the night
to night variability of apnea indices.
Methods
The polysomnographic records of 50 patients with an apnea index ᔤ5 in at
least one of the two nights were retrospectively screened (45 men and 5 women).
We compared the mean duration of the apneas, the maximum duration of the
apneas, the AI in the total sleep time (TST) and the AI in the total sleep period
(TSP) in both nights. The intra subject apnea variability between night1 and
night 2 was calculated as a relative percentage by the following formula : (AI
TST 1 – AI TST 2) : (AI TST 1 + AI TST 2) x 100.
Results
Table 1 lists the mean values with standard deviations of the analysed characteristics in both the first and second night and the absolute difference between
first and second night.
175
Mean duration
Maximum duration
AI TST
AI TSP
Night 1
Night 2
Night 2 – Night 1
22 ±
50 ±
21 ±
16 ±
23 ±
52 ±
24 ±
17 ±
0.5 ± 5
2 ± 20
10 ± 15
1.3 ± 7
7
26
21
15
8
32
27
16
Table 1
Absolute differences where analysed by Student’s t-test. There was no significant difference for all parameters even if individual differences were quite substantial. Figure 1 shows the relative difference of apnea indices in TST between
the two nights. Most patients differed in the range 10-30 %.
For example: a patient with an apnea index of 15 (TST) in one night may have
an index of 10 or 20 in the following night.
The cut-off point for CPAP treatment was reached by 23 patients in the first
night. From the other 27 patients only 3 reached this criterium in the second
night.
Figure 1
Conclusion
Night-to-night intra subject variability of the apnea index was small for this
group. As substantial individual variability does exist, single night recording
induces a certain risk. For the decision to treat a patient with CPAP (at a cutoff point AI=15) the risk is 13-17%. This means for example that a patient with
clear clinical syptoms of OSAS but with an apnea index of only 10 in the first
night a second polysomnography is strongly recommended.
176
The transition from neonatal to infantile sleep
A.W.DE WEERD, R.J.SCHIMSHEIMER, B.KEMP
JULIANA CHILDREN HOSPITAL AND CENTRE FOR SLEEP AND WAKE DISORDERS,
MCH, WESTEINDE HOSPITAL, THE HAGUE
The electro-encephalogram (EEG) of the full term born baby is well developed,
but has many features that disappear in the first three months of life. On the
other hand new elements emerge in this period that represent the transition to
the infantile EEG. During sleep, spindles are example of new EEG phenomena.
Features from the record during sleep that disappear are REM sleep onset (mostly
called Active Sleep onset, AS, at this age), background patterns consisting of
short periods with alternating high and low amplitude activity (Tracé Alternant,
TA) and triangular slow waves with highest amplitude over frontal regions
(Encoches Frontales, EF). Elements from the EEG that are seen in the prematurely born child and are sometimes still present at term in a normal baby,
disappear completely during these first months. Examples are so called delta
brushes (DB) and short runs of rhythmic 4-6 Hz activity over temporal regions
(Premature Temporal Theta activity, PTT). For more details, see 1.
The elements mentioned are well described, but knowledge about the time
course of the transition from the neonatal sleep EEG to that characteristic for
the two to three month old child is far less detailed.
The EEG is an important tool in the assessment of cerebral function in the first
months of life. Sleep spindles (SS), a continuous background pattern (instead
of TA) and no sleep onset AS are not only characteristic for the somewhat older
infant, but are the first manifestations of human sleep as seen in the older child
and adult as well. Thus, more details on the development of the sleep EEG in
this period of life is of interest from a scientific point of view as well as for daily
practice in pediatric medicine. Up to now this course in time of important
aspects of the neonatal and infantile EEG has been studied in small groups of
children only (for example 2, 3, 4), leading to sometimes conflicting results.
The results presented in this report are taken from a large study of the EEG in
babies with a Conceptional Age (CA, i.e. the age in weeks after conception; for
example CA at term birth is 38-42 weeks) in the range between 30 and 53
weeks. Except for prematurity in the children born before CA 37 weeks, all
were normal at clinical examination at the time of recording of the EEG and all
had a good long-term outcome. The group of children with a CA in which the
transition from neonatal to infantile sleep could be expected (CA 43-53 weeks)
was studied in detail for the features of sleep mentioned above.
177
Patients and Methods
Eighty-one children (44 boys) from the database met the criteria for inclusion
in the study. They had a CA between 43 and 53 weeks, slept at least part of the
recording time and were normal at clinical examination at the time of EEG
recording and during follow-up. Admittedly, they underwent EEG registration
for clinical reasons (short lasting asphyxia during birth: 2; unexplained breathing disturbances, so called ALTE’s: 54; febrile convulsion: 1; slight head trauma:
1; and miscellaneous, mostly movements that might be interpreted as of epileptic origin, but never proven so: 23). The children studied were not completely normal in the strict sense of the word, but they represented the group
encountered in daily practice that can be considered as very low risk and defacto normal. Seven children had a second EEG registration, resulting in 88
records for analysis. Differentiation for CA resulted in the following categories
and numbers of patients: CA 43-44 weeks: N=24; CA 45-46 weeks: N=20; 47-48
weeks: N=15; 49-50 weeks: N=19 and CA 51 weeks or more: N=10.
The EEG and accompanying polygraphy of ECG, respiration, eye movements
and muscular tone was recorded and assessed according to internationally standardized criteria (5). In addition to this basic evaluation the EEG features relevant in the transition from neonatal to infantile sleep were studied in detail.
From the 88 records 1287 minutes of sleep EEG were available for analysis
(range: 90-2290 s.; median: 925 s. per recording). For the EEG in the waking
condition these figures were 1713 minutes total recording time; range: 2703460 s.; median 1055 s. per registration
Results
Discontinuity in background activity, the TA pattern, was seen up to the age of
CA 47 weeks and was not encountered in the EEGs recorded after that age. TA
is characterized by alternating periods of activation and relative suppression of
background activity. When TA was present, mean and maximal duration of the
suppression and the ratio between amplitudes during activation and suppression were independent from the age of the child.
REM (AS) sleep onset occurred in only nine of the EEGs, in all cases before the
age of CA 49 weeks. In seventy recordings NREM was the mode of first sleep.
For the remaining eleven differentiation in sleep onset was not certain, but
rapid eye movements, which are the hallmark of AS, were not seen at all.
Frontal sharp waves, EF, occurred up to the age of 49 weeks. In particular
bilateral EF, which is the main variant of this EEG phenomenon in the young
child, gradually disappeared with age.
EEG transients (DB and PTT) that are seen mostly in the prematurely born
child and are still sometimes present at CA around 40 weeks did not occur in
the study group.
178
From the age of CA 43 weeks on, there were recordings containing sleep spindles. In the youngest age category of CA 43-44 weeks this EEG phenomenon
was seen in 6 out of 24 records (25%) with maximum duration of the spindles
of three seconds. For the children with CA of 49 and more weeks (who all had
periods of NREM sleep, see above) these figures were 25 out of 29 recordings
(86%) and duration up to 12 seconds. From the age of CA 52 weeks all recordings contained sleep spindles.
Conclusion. See the figure.
Figure
Development of sleep patterns
Conceptional Age (weeks)
43 44 45 46 47 48 49 50 51 52 53
Disappearing patterns
(no recording
contains the pattern)
Trace Alternant
[———————————
Active Sleep onset
[—————————
Encoches Frontales
[———————
Emerging patterns
(all recordings
show the pattern)
Sleep spindles
[————
References
1 de Weerd AW. Atlas of EEG in the first months of life. Amsterdam, Elsevier Scientific
Publishers 1995.
2 Ellingson RJ, Peters JF. Development of EEG and daytime sleep in normal full-term
infants during the first 3 months of life: longitudinal observations. Electroenceph clin
Neurophysiol 1980;49:112-124.
3 Jankel WR, Niedermeyer E. Sleep spindles. J Clin Neurophysiol 1985;2:1-35.
179
4 Statz A, Dumermuth G. Transient EEG patterns during sleep in healthy newborns.
Neuropediatrics 1982;13:115-122.
5 de Weerd AW, Despland PA, Plouin P. Neonatal EEG. In: G. Deuschl, A.Eisen eds.
Recommendations for the practice of Clinical Neurophysiology. International Federation for Clinical Neurophysiology 1999 (ter perse en verschijnt begin september
1999; de exakte beschrijving volgt).
Acknowledgement: The EEGs were recorded by I. Keyzers, J. Vreeburg and E. Dijkstra.
180
List of members
Honorary members
Dr. A.C. Declerck
Prof. dr. R.H. van den Hoofdakker
Prof. dr. H.A.C. Kamphuisen
Prof. dr. P. Visser
Regular members
M. Anema
Instituut voor Slaap en Ontspanning
W.A. Scholtenlaan 126
6865 VZ DOORWERTH
T 026 333 71 70
F 026 33 90 327
privé 026 33 90 327
Dr. J.B.A.M. Arends, neuroloog
Epilepsie-centrum Kempenhaeghe
Sterkselseweg 65
5591 VE HEEZE
T 040 227 90 22
F 040 226 49 24
J.W. Barendrecht-Kooijman
Van Eesterenplein 309
3315 KX DORDRECHT
T 078 621 37 02
F 078 621 37 05
Dr. D.G.M. Beersma, bio-fysicus
Rijksuniversiteit Groningen
Afd. Biol. Psychiatrie
Postbus 30.001
9700 RB GRONINGEN
T 050 361 20 34
F 050 361 16 99
E-mail [email protected]
Dr. A.L. van Bemmel, psychiater
Vakgr. Psychiatrie & Neuropsychologie
Universiteit Maastricht
Postbus 616
6200 MD MAASTRICHT
T 043 3685444
F 043 3685331
E-mail [email protected]
Dr. E.J.F.M. ten Berge
Polikliniek longziekten
Geerdinksweg 141
7555 DL HENGELO
T 074 2475290
F 074 2475686
Dr. F.W. Bes, psychofysioloog
KNSM-laan 588
1019 LP AMSTERDAM
T 020 4194389 (privé)
Medcare Automation
Stadhouderskade 2
1054 ES AMSTERDAM
T 020 6071367
F 020 6163002
E-mail [email protected]
Dr. F. Boer
p/a Curium,
Endegeesterstraatweg 27
2342 AK OEGSTGEEST
181
H. Boot
Adm.Helfrichplein 14
2665 AG BLEISWIJK
T 010 522 08 49
H.J. Dalewijk
Nieuwstad 56
1381 CD WEESP
T 0294 41 27 47
M. ten Caat-Swart
Mallinckrodt Benelux b.v.
Hambakenwetering 1
Postbus 745
5201 AS ‘s HERTOGENBOSCH
T 073 648 52 00
F 073 641 09 15
Dr. A.C. Declerck
Centrum voor Slaap/Waak
Stoornissen
Sterkselseweg 65
5591 VE HEEZE
T 040 227 92 02
F 040 227 93 99
Prof.dr. R. Cluijdts
Faculteit voor Psychologie
Vrije Universiteit Brussel
Pleinlaan 2
1050 BRUSSEL (België)
Dr. R. van Diest
Afd. Psychiatrie
Psycho Med.Streekcentrum Vijverdal
Vijverdalseweg 1
6226 NM MAASTRICHT
T 043 368 53 30
F 043 368 53 31
Prof.dr. A.M.L. Coenen, psycholoog
NICI-verg. Fysiol.Psychologie
Psychologisch Laboratorium
Postbus 9104
6500 HE NIJMEGEN
T 024 361 25 45
F 024 361 60 66
E-mail [email protected]
Dr. L. Cohen
Ac.Ziekenhuis, V.U.
Med. Psychologie
Postbus 7057
1007 MB AMSTERDAM
T 020 548 34 73
F 020 548 59 49
PHM. Van Dongen, psychiater
Gezondheidscentrum Oost Brabant
Postbus 3
5427 ZG Boekel
Dr. H.P.A. van Dongen
per 1/06 4200 Chester AV. 3rd fl
Philadelphia, PA 19104
United States of America
+1 215 573 5866 (werk)
+1 215 573 6410 (fax)
E-mail [email protected]
Haarlemmerstraat 209
2312 DR LEIDEN
T 512 4555 (prive)
Dr. W.H.I.M. Drinkenburg
NICI/Vergelijkende en Fysiologische Psychologie KUN
Postbus 9104
6500 HE NIJMEGEN
T 024 361 25 54
F 024 361 60 66
E-mail [email protected]
182
Dr. H. van Duyn
St. Lucas Ziekenhuis
Afd. KNF
J. Tooropstraat 164
1061 AE AMSTERDAM
T 020 510 89 11
F 020 618 84 68
Dr. M. Dzoljic
Zwanenkade 124
2925 AT KRIMPEN A/D IJSSEL
3000 DR ROTTERDAM
T 010 408 75 31
F 010 436 68 39
B. van Egdom
Air Liquide Vital Aire
De Witbogt 1
5652 AG EINDHOVEN
Dr. M. Elton
Faculteit der Psychologie
Vakgroep Psychonomie
Roetersstraat 15
1018 WB AMSTERDAM
T 020 525 68 45
F 020 639 16 56
E-mail [email protected]
Dr. M.G. van Erp
Afd. Klinische Neurofysiologie
Kempenhaeghe
Sterkelseweg 65
5591 VE HEEZE
T 040 2279022
F 040 2264924
E-mail [email protected]
P.R. Eykelenboom
Akkerpad 12
6081 HC HAELEN
T 0475 591767
E-mail [email protected]
M.M.M. Eysvogel
Medisch Spectrum Twente
Polikliniek Longziekten
Postbus 50.000
7500 KA ENSCHEDE
Prof.dr. H. Folgering
Universitair Longcentrum Dekkerswald
Postbus 9001
6560 GB GROESBEEK
T 024 685 9250
F 024 6959290
E-mail [email protected]
Prof. Dr. K. Gill
Canadalaan 13
2498 MS Alphen aan de Rijn
Dr. M.C.M. Gordijn
Zoological Laboratory
Rijksuniversiteit Groningen
Postbus 14
9750 AA GRONINGEN
T 050 363 20 73
F 050 363 52 05
E-mail [email protected]
Dr. J.H.M. de Groen
Centrum voor Slaap- en Waakstoornissen
Sterkelseweg 65
5591 VE HEEZE
T 040 2279202
F 040 2279399
183
Dr. P.M.J. Haffmans
Psych. Centrum Bloemendaal
Monsterseweg 93
2553 RJ DEN HAAG
T 070 391 66 38
F 070 391 64 65
Y.F. Heydra
ULC Dekkerswald
Nijmeegsebaan 31
6561 KE GROESBEEK
T 024 685 99 11
F 024 685 92 90
P.Th.M. van der Ham
Thorbeckelaan 21
5463 BM VEGHEL
T 0413 36 70 38
J.H. van der Hoeven
afd.Neurologie, Academisch Z’huis
Postbus 30.001
9700 RB GRONINGEN
T 050 361 25 99
F 050 369 67 40
privé: Cantersveen 2
9753 KK Haren (Gn)
T 050 534 72 35
Dr. H.L. Hamburger
Afd. Neurologie Slotervaartziekenhuis
Louwesweg 6
1066 EC AMSTERDAM
T 020 512 44 64
J. Hanraads
Comcare Medical
Postbus 711
8440 AS HEERENVEEN
T 0513 650423
E. Heemskerk
Houtstraat 45
6511 JL NIJMEGEN
T 024 3601455
J.C. van Hemert-van der Poel
Afd. Neurologie
Postbus 9020
7200 GZ ZUTPHEN
T 0575 592452
F 0575 521206
H.L. Hermes
Schering Nederland bv
Postbus 116
1380 AC WEESP
Dr. W. Hofman
Vakgroep Psychonomie
Universiteit van Amsterdam
Roetersstraat 15
1018 WB AMSTERDAM
T 020 525 68 48
F 020 639 16 56
Prof.dr. R.H. van den Hoofdakker, psychiater
Afd. Biologische Psychiatrie
Psychiatrische Universiteitskliniek
Oostersingel 59
9713 EZ GRONINGEN
T 050 361 91 11
F 050 369 67 27
C.H. Hoog
Instituut voor Slaap en Ontspanning
p/a W.A. Scholtenlaan 126
6865 VZ DOORWERTH
T 026 333 71 70
184
S. Jongman
Weinmann bv
Postbus 64
1140 AB MONNICKENDAM
T 0299 656977
F 0299 656980
Prof.dr. H.A.C. Kamphuisen
Poelgeesterweg 2
2341 NM OEGSTGEEST
Dr. A. van Keimpema, longarts
Long- en asthmakliniek
De Klokkenberg
Postbus 90108
4800 RA BREDA
Dr. Ir. B. Kemp, med.fysicus
Westeinde Ziekenhuis, Slaapcentrum
Postbus 432
2501 CK DEN HAAG
T 070 330 22 05
F 070 388 26 36
Prof.dr. G.A. Kerkhof, psycholoog/
fysioloog
Lab. voor Fysiologie
Rijksuniversiteit Leiden
Postbus 9604
2300 RC LEIDEN
T 071 527 68 10/6783
F 071 527 67 82
E-mail [email protected]
Prof.dr. E. Klip
G.Borgesiuslaan 50
9722 RL GRONINGEN
Dr. A. Knuistingh Neven, huisarts
Leids Universitair Medisch Centrum
Afdeling Huisartsgeneeskunde en
Verpleeghuisgeneeskunde
Postbus 2088
2301 CB LEIDEN
T 071 527 53 18
F 071 527 53 25
Praktijk:
Brederodestraat 3
2931 XB KRIMPEN A/D LEK
T 0180 586 586
F 0180 586 587
E-mail [email protected]
G.M.L.G. Konings,
klinisch psycholoog i.o.
psychotherapeut i.o.
Koning Clovisstraat 1
6226 AE MAASTRICHT
T 043 36 39 802
K.W. van Kralingen
Bronovolaan 24
2597 AZ DEN HAAG
tel. 070 3240553
Dr. C. Kramer
privé v. Spilbergenstraat 147-III
1057 RD AMSTERDAM
Ir. A. Kumar
Medcare Automation
O. Heldringstraat 27
1066 XT AMSTERDAM
T 020 3460120
F 020 6152795
185
G.J. Lammers, neuroloog
Afd. KNF, Gebouw 20
Ac. Ziekenhuis
Postbus 9600
2300 RC LEIDEN
T 071 526 21 43/526 21 04
P.B. Luursema, longarts
‘Het Nieuwe Spittaal’
Postbus 9020
7200 GZ ZUTPHEN
T 0575 592592
F 0575 521206
M. Laurant
Stieltjesstraat 131
6511 AA NIJMEGEN
T 024 322 06 69
F 030 272 97 85
Dr. E.L.J.M. van Luytelaar
Fysiol.Psychol.en Neuropsych.
Psychologisch Lab.
Postbus 9104
6500 HE NIJMEGEN
T 024 351 56 21
F 024 361 59 38
W.H.G. Lieuwens
Oosterschelde Ziekenhuis
Afd. Neurologie
‘s Gravenpolderseweg 114
4462 RA GOES
T 0113 23 44 63
F 0113 23 04 24
Dr. F. Lobbezoo
Vakgroep Orale Functieleer/CMD
ACTA
Louwesweg 1
1066 EA AMSTERDAM
T 020 5188412/384
F 020 5188414
E-mail [email protected]
Drs. S.A.P. Lucius, psycho-fysioloog
Hooigracht 13
2312 KM LEIDEN
werk:
Psychiatrisch Ziekenhuis Bloemendaal
Monsterseweg 93
2553 RJ DEN HAAG
T 070 39 16 582
F 070 39 16 146
Dr. M.H.J.M. Majoor
KNO-arts
Vossenberg 59
6721 BM BENNEKOM
Ir. W.L.J. Martens
Kromstraat 3
5421 XZ GEMERT
T 0492 36 52 90
Dr. P. Meerlo
Rijksuniversiteit Groningen
Zoölogisch Laboratorium
Postbus 14
9750 AA HAREN
T 050 363 20 51
F 050 363 52 05
E-mail [email protected]
Dr.ir. J.W.H. Meijs
Algemeen directeur SBI
De Hengmolen 16,
6932 BP WESTERVOORT
T 026 3116175
186
Dr. F.W. v/d Meulen
KNO-arts, A.M.C.
Polikliniek KNO
Meibergdreef 9
1105 AZ AMSTERDAM
T 020 566 91 11, sein no. 096
privé: Graaf Wichmanlaan 37
1405 GZ BUSSEM
T 035 694 16 99
Rogier Mullaart
Somnio Slaap Consultancy
Nieuwe Kerkstraat 116-4
1018 VM AMSTERDAM
T 020 420 74 47/06 263 187 17
E-mail [email protected]
V.A. Muray
Händelstraat 55
4003 LA TIEL
T 0344 62 47 61
Dr. J. Meulstee
Canisius Wilhelmina Ziekenhuis
Afd. C86, Klinische Neurofysiologie
Postbus 9015
6500 GS NIJMEGEN
T 024 3 65 83 35
E-mail meulstee@
Dr. H.A.M. Middelkoop
Academisch Ziekenhuis
Afdeling Neurologie
Sectie Neuropsychologie J3R
Postbus 9600
2300 RC LEIDEN
T 071 526 21 25
F 071 524 82 53
E-mail
[email protected]
P.I. Milder
Ned. Ver.voor Slaap Apneupatiënten
Dintel 74
5172 CS KAATSHEUVEL
T 0416 54 32 32
F 0416 54 31 99
F.J. Mud
Afd. KNO
“Het Spittaal”
Postbus 9020
7200 GZ ZUTPHEN
M.H.A.M. Mutsaers
Product Manager
Farmadomo b.v.
Postbus 28
5390 AA NULAND
Bezoekadres Industriestraat 20
T 073 53 43 434
F 073 53 43 430
E-mail [email protected]
J.E. Nagtegaal, ziekenhuisapotheker
Ziekenhuis Rijnstate
Ziekenhuisapotheek
Postbus 9555
6800 TA ARNHEM
T 026 37 86 309
privé:S.vd Hagenlaan 2
3818 HD AMERSFOORT
T 033 46 12 564
Nederlandse Vereniging voor Narcolepsie
t.a.v. de heer B.J. van Rooijen, penningmeester
Brechtzijde 37
2725 NR ZOETERMEER
187
T. van Noort
Pegasusplaats 101
6525 JJ NIJMEGEN
T .024 3563924
A. Oosterhuis
Monteverdistraat 97
2901 KE CAPELLE AAN DEN IJSSEL
T 010 447 29 01
E-mail [email protected]
P. Ossenblok
medisch fysicus
Epilepsie-centrum Kempenhaeghe
Sterkselseweg 65
5591 VE HEEZE
J. van Proosdij
CNS Farmacologie, afd.RE 2134
Postbus 20
5340 BH OSS
T 0412 66 24 49
F 0412 66 25 42
R.H.U. Rammeloo, longarts
‘Het Nieuwe Spittaal’
Postbus 9020
7200 GZ ZUTPHEN
T 0575 592592
F 0575 521206
Dr.ir. J.P.H. Reulen
Academisch Ziekenhuis Maastricht
Afd. KNF
Postbus 1918
6201 BX MAASTRICHT
T 043 387 72 74
F 043 387 78 78
R.M. Rijsman
Afd. Neurologie
Westeinde Ziekenhuis
Lijnbaan 32
2512 VA DEN HAAG
T 070 388 26 36
F 070 330 30 17
Dr. G.S.E. Ruigt, bioloog
Afdeling Neurofarmacologie
Organon International BV, kamer RE
2130
Postbus 20
5340 BH OSS
T 0412 66 24 49
F 0412 66 25 42
Dr. R.J. Schimsheimer
klin.neurofysioloog
Westeinde Ziekenhuis, Afd.KNF
Postbus 432
2501 CK DEN HAAG
M.H.T. Schouten
TeFa Portanje b.v.
Wipmolenlaan 3
3440 AG WOERDEN
T 0348 49 57 00
S. Schreiner-Polster MAP GMBH
Frauenhoferstrasse 16
D-82152 MARTINRIED
Germany
K.E. Schreuder
Centrum voor Slaap/Waakstoornissen
Postbus 61
5590 AB HEEZE
T 040 227 92 23/02
F 040 227 93 99
188
Chr.F. Sepmeijer
Afd. KNO
“Het Spittaal”
Postbus 9020
7200 GZ ZUTPHEN
Dr. M.G. Smits, neuroloog
Ziekenhuis ‘De Gelderse Vallei’
Stationsweg 8
6711 PV EDE
T 0318 65 72 90
F 0318 61 39 44
E-mail [email protected]
Dr. J. Snel
Vakgroep Psychonomie
Universiteit van Amsterdam
Roetersstraat 15
1018 WB AMSTERDAM
T 020 525 68 40
F 020 639 16 56
Dr. E.J.W. van Someren
Ned.Inst.v. Hersenonderzoek
Meibergdreef 33
1105 AZ AMSTERDAM
T 020 566 55 00
F 020 691 84 66
E-mail [email protected]
Dr. R.L.M. Strijers
Afd. KNF AZVU
Academisch ziekenhuis, V.U.
De Boelelaan 1117
1081 HV AMSTERDAM
T 020 444 07 31
F 020 444 28 00
E-mail [email protected]
Dr. A.M. Strijkstra
Zoologisch Laboratorium
Rijks Universiteit Groningen
Postbus 14
9750 AA HAREN
T 050 363 20 73
F 050 363 52 05
E-mail [email protected]
Sunrise Medical bv
t.a.v. de heer J.M. Gerritse
Pascalbaan 3
3439 MP NIEUWEGEIN
Y. Sweere, psycholoog
Centrum voor Slaap- en Waakstoornissen
Westeinde Ziekenhuis
Postbus 432
2501 CK DEN HAAG
T 070 330 20 04
F 070 388 26 36
E-mail
[email protected]
T.J. Tacke
Streekziekenhuis Midden-Twente
Postbus 546
7550 AM HENGELO
T 074 247 52 75
M.M.J.B. Thijssen
Regionaal Centrum voor Slaap en Waakstoornissen
Westeinde Ziekenhuis
Postbus 432
2501 CK DEN HAAG
T 070 330 27 22
van Boetselaerlaan 134A
2581 AX DEN HAAG
189
Dr. J.H.M. Tulen, psychofysiologe
Afd. Psychiatrie
Dijkzigt Ziekenhuis
Dr. Molewaterplein 40
3015 GD ROTTERDAM
T 010 463 59 74
F 010 463 32 17
E-mail [email protected]
Dr. R. van Uffelen
Air Liquide Vitalaire
De Witbogt 1
5652 AG EINDHOVEN
privé Drechtsteden 2H
Dordrecht
T 078 6541111
Prof. dr. A.M. Vein
Moscow Medical Academy
Rossolimo str. 11
Moscow
RUSSIA
Drs. H.M.J.C. Verbeek, med.biologe
Centrum voor Slaap/Waak
Stoornissen
Sterkselseweg 65
5591 VE HEEZE
T 040 227 92 39
F 040 227 93 99
E-mail [email protected]
M.M.R. Verhelst
Regionaal Centrum voor Slaap en Waakstoornissen
Westeinde Ziekenhuis
Postbus 432
2501 CK DEN HAAG
T 070 330 27 22
F. Visscher, neuroloog
Oosterscheldeziekenhuizen
‘s Gravenpolderseweg 114
4460 BB GOES
T 0113 23 44 63
F 0113 23 04 24
Prof. dr. P. Visser
J. v. Eycklaan 276
3723 BC BILTHOVEN
T 05730 1463
Dr. P. Vos
A.Romein-Verschoorlaan 8
6532 SH NIJMEGEN
Dr. J.W. Vredeveld
De Wever Ziekenhuis
Afd. KNF
Postbus 4446
6401 CX HEERLEN
T 045 576 63 71
F 045 576 60 55
Dr. J.A.S. Verbraeken, longarts
Academisch Ziekenhuis Maastricht
Postbus 5800
6202 AZ MAASTRICHT
T 043 387 50 44
F 043 387 50 51
E-mail [email protected]
190
Drs. F. de Vries
tandarts-gnatholoog
Subfac. Tandheelkunde
Centrum Bijzondere Tandheelkunde
Nijmegen
Het Lage Erf 1
6816 RK ARNHEM
T 024 323 38 82 (praktijk)
T 024 442 44 60 (prive)
F 024360 04 42
E-mail [email protected]
J.M.A. Vugts
UCB Pharma bv
Postbus 6851
4802 HW BREDA
R. Warmerdam
Weyth-CNSm AHP-Pharma b.v.
Postbus 255
3130 AG HOOFDDORP
Dr. D. Waterman
Faculteit der Psychologie
Vakgroep Psychonomie
Afd. Psychofysiologie
Roetersstraat 15
1018 WB AMSTERDAM
T 020 525 68 48
F 020 639 16 56
priv’e: Borssenstraat 29III
1078 VA AMSTERDAM
T 020 6755967
Dr. G. Wilts
Medisch Spectrum Twente
Postbus 50000
7500 KA ENSCHEDE
T 053 4872000
F 053 4873072
Dr. P.G.W.M. Wuisman
Maasland Ziekenhuis
Afd. KNF
Walramstraat 23
6131 BK SITTARD
T 046 459 77 12
F 046 459 79 70
D. Zeeman-Ruyters
Zambiastraat 21
2622 DG DELFT
T 015 261 67 84
Prof.dr. M.J. Zwarts
AZN St. Radboud
Afdeling KNF
Postbus 9101
6500 HB NIJMEGEN
Prof. Dr. K. Gill
Canadalaan 13
2498MS Alphen aan de Rijn AG
Dr. A.W. de Weerd
Afd.Klinische Neurofysiologie
Westeinde Ziekenhuis
Postbus 432
2501 CK DEN HAAG
T 070 330 20 05
F 070 380 94 59
191
List of sponsors
AHP Pharma BV
Danica Nederland BV
MAP GMBH
Medtronic Upper Airway
OSG
Sanofi - Synthelabo
Schering Nederland BV
Somnus Europa SMT
Stopler Instrumenten en Apparaten
TEFA-Portanje BV
UCB Farma Nederland BV
VitalAire, Aire Liquide BV
192