Transcript Quantitative and qualitative analysis of sleep

Quantitative and qualitative
analysis of sleep: past, present
and future
V. Mohan Kumar
Sree Chitra Tirunal Institute for Medical
Sciences & Technology
Thiruvananthapuram, Kerala, India
Sleep in Hindu & Islamic Philosophy
• Vedas & Upanishadas consider sleep as part of consciousness:
– Jagrat (or Vaisvanara)- awake.
– Svapna (or Taijasa)- dream.
– Sushupti – dreamless sleep.
– Turiya - spiritual state of consciousness.
• The Quran considers sleep as part of “state of self”:
– Consciousness.
– “Wafat” (death).
• Wafat is divided into:
– sleep (temporary death).
– death (real death).
• Quran refers to “dream” as ru’ya (vision), hulm (dream), manam
(sleep) and bushra (tidings).
Sleep in Christian & Jewish Philosophy
• According to the Bible, sleep is influenced by physical,
psychological and spiritual factors.
– Sleep is a gift from God
– Hard labor can induce the deepest form of sleep.
– Worry about material possessions results in sleeplessness.
• According to Jewish (Talmud) philosophy sleep is 1/60 of
death.
– Sleep is physical separation of the soul from the body.
– Soul comes back after sleep. It does not return after death.
– Refraining from sleep is a sin because sleep is enforced by
nature.
– God reaches out to us through dreams; bringing guidance,
nourishing our spirits, healing and refreshing us.
What is sleep?
Behaviorally sleep is characterized by:
• Reduced motor activity.
• Decreased response to stimulation.
• Stereotyped posture.
• Relatively easy reversibility
(this distinguishes it from coma, hibernation,
torpor & estivation).
Sleep investigations
• Modern sleep research is multidisciplinary.
• EEG has been the backbone of sleep research
and sleep medicine since 1930s.
• Analysis of sleep should ensure that it is:
– Compatible with existing scientific evidence.
– Based on biological principles.
– Applicable to clinical disorders.
– Easily used by sleep clinicians, sleep
scientists and technologists.
Modern classification of sleep
• Though EEG has been the backbone of sleep analysis, it is
used in combination with EMG and EOG.
• Polysomnographic recording includes electrocardiogram,
respiratory effort, nasal/oral airflow, blood oxygen saturation,
body position, limb movements, and video recording.
• Modern definition & classification
– Nathaniel Kleitman (1939) Book “Sleep and
Wakefulness”.
– Aserinsky & Kleitman (1953) REM sleep.
– Rechtschaffen & Kales (1966) A Manual of Standardized
Terminology, Techniques, and Scoring System for Sleep Stages of
Human Subjects.
• American Academy of Sleep Medicine (AASM) modified the
staging rules in 2007.
Major issues in sleep analysis
• Our understanding of sleep and its neural mechanism is
grossly inadequate.
• Modern sleep classification is based on the age old belief
that sleep consists of REM sleep and Non-REM sleep.
• Classification of sleep into REM and Non-REM started with
the assumption that REM sleep is sleep with dream, and
Non-REM is dreamless sleep.
• Use of computer has helped in fast analysis of large data.
• But computer analysis has not been able to take care all
aspects of available information.
• There are major differences between manual (visual)
scoring, and computer scoring.
Sleep-wakefulness defined
electrophysiologically
Stages of sleep & wakefulness
Sleep-wake changes every night
EEG Recordings
Sleep Pattern
REM
Sleep
Awake
Stage 1
Stage 2
Stage 3
NREM
Sleep
Stage 4
0
1
2
3
4
5
Time (hrs)
6
7
Polysomnography
•
•
•
•
•
•
Apart from EEG,EOG and
EMG, other variables include:
Electrocardiogram (ECG) with
two or three chest leads.
Respiratory effort, by chest-wall
and abdominal movements.
Nasal and/or oral airflow via
thermistor or
pneumotachograph.
Oxygen saturation via pulse
oximetry.
Body position via mercury
switches or by direct
observation.
Limb movements (arms and
legs) via EMG.
Hypnogram
23°C
25°C
27°C
Electroencephalography
• Modern sleep research began with electrophysiological
monitoring of sleep.
• The first person to record electric currents of the brain
was Richard Caton in1875.
• Hans Berger (1929) is generally credited with the
discovery of the EEG. He showed that the EEG differs
between sleep and waking.
• Aserinsky and Kleitman (1953) showed that sleep could
be further differentiated into two distinct states, ie REM
sleep and Non-REM sleep.
• EEG recorded from frontal, central and occipital regions
using 6 electrodes in10-20 system.
EEG during sleep & wakefulness
Features of EEG used for analysis
• NREM sleep stages1 to 4 represent successively deeper
stages of sleep, with EEG showing increasing voltage &
decreasing frequency.
• Relaxed awake EEG shows alpha waves, alert EEG shows
beta waves.
• Drop in EEG voltage at sleep onset (Stage 1 NREM).
• EEG spindles and K complexes at Stage 2 NREM.
• Delta wave dominance at deep NREM sleep (Stages 3&4
NREM).
• REM sleep EEG resembles wake stage EEG in animals
(low voltage fast activity). So it’s called paradoxical sleep.
• In humans, REM-EEG resembles that of stage 1 NREM
(low voltage mixed waves).
Importance of power spectral
analysis of EEG
• Scoring more than 1400 pages of record is very time
consuming.
• Visual inspection & analysis of EEG bands, frequently do
not indicate the depth of sleep, and do not help to quantify
sleep.
• Therefore, automated scoring is essential.
• When a computer makes a hypnogram, it uses a lot of
parameters, such as alpha rhythm, sleep spindles, sleep
delta waves, rapid eye movements or tonic chin EMG
levels.
• The depth of sleep is reflected by EEG slow waves
produced during NREM sleep and tells us something about
its recuperative value.
• In all mammalian species delta waves increase as a
function of prior waking duration.
Power spectral analysis of EEG
• Computer assisted power spectral analysis of EEG
involves digitization of an analog EEG signal.
• EEG is filtered for fast Fourier transform (FFT) and power
spectral analysis.
Localization of dominant band in
the EEG
• Power spectral analysis can also be used to
show the area of dominance of each band in
the EEG
Assessment of depth of anesthesia
• The 90% spectral edge frequency (SEF 90) of EEG
power is used for assessing the depth of anesthesia.
• SEF 90 shift in from 16 to 12 Hz in adequate anesthesia.
(MDF: mean dominant frequency; PF: peak frequency; SMF: spectral
median frequency).
EEG power in the delta band
• Elevated EEG power in the delta band (0.5–4.0 Hz)
during NREM sleep during recovery sleep after
modafinil, reflect augmented sleep.
(Edger and Seidel, 1997)
Principal Component Analysis of EEG
• EEG after Fourier Transformation can be
subjected to Principal Component Analysis
• Principal components of different sleep-wakeful
states are:
NREM: 1-8 Hz & 5-15 Hz
REM: 1-9 Hz & 10-15 Hz
Wakefulness: 1-7 Hz, 7-11 Hz, & 12-15 Hz
• Alpha band was identified only during
wakefulness.
• Traditional division of theta band in the human
cortical EEG is artificial.
(Corsi-Cabrera et al, 2000)
Sigma versus beta
• FFT spectral analyses showed that sigma (12–
16 Hz) versus beta (20–28 Hz) EEG can
discriminate between NREM and REM sleep.
• NREM had high sigma and low beta.
• REM showed low sigma and high beta.
• EEG of REM and NREM sleep are composed of
two sets of EEG frequency components,
perhaps reflecting different neuronal pools.
(Sunao Uchida et al,1994)
Gamma oscillations
Magnetoencephalography
(MEG) mapping of magnetic
fields produced by electrical
currents in the brain. MEG is
recorded using arrays of
SQUIDs (superconducting
quantum interference devices).
• Amplitude of gamma (35–45) oscillations is markedly
diminished in NREM sleep compared to wakefulness and
REM sleep.
(Llinas and Ribary, 1993)
EOG (Electroocculogram):
• EOG of REM sleep is characterised by bursts of Rapid Eye
Movements (REM).
• EOG - Awake or Stage N1 - regular, sinusoidal, initial deflection
>500msec
• EOG - Stage N3- not typically seen.
• EOG - REM sleep- irregular, sharp, initial deflection ≤ 500msec.
• Correlated with REM, there are PGO waves (best recorded with
depth electrodes in animals).
One electrode is placed above
and to the outside of the right
eye, and another placed below
and to the outside of the left eye.
EMG (Electromyogram)
• Muscles are progressively relaxed during deeper NREM sleep.
• Maximum loss of muscle tone during REM sleep.
• Muscle relaxation is produced by progressive hyperpolarisation
of lower motor neurons.
• During REM sleep limb muscles show sudden twitches in
between.
Three leads are placed on the chin (one in the front
and center and the other two underneath and on the
jawbone).
Two leads are placed on the inside of each calf
muscle 2-4cm apart.
Sleep & body temperature
• Body temperature is slightly reduced during NREM
sleep. It is actively maintained at this lower level.
• There is decreased thermoregulatory ability during REM
sleep. Body temperature drifts towards ambient
temperature.
(Kaushik, Mallick and Kumar, 2009)
Changes in sleep & brain
temperature
W1
W2
S1
S2
• During REM sleep brain temperature and brain
metabolism are increased.
(Thomas and Kumar, 2002)
PS
EKG/ECG (Electrocardiogram)
• Two electrodes are placed on the
upper chest near the right and
left arms.
• These record the heart rate and
rhythm and serve to alert the
technician to a possible
emergency situation.
• They also demonstrate whether
apneic desaturation leads to
arrhythmias or not.
Heart-rate variability
• Normal ECG exhibits periodic variation in R-R
intervals, called Heart rate variability (HRV).
• Cardio-acceleration during inspiration, and
deceleration during expiration is known as
respiratory sinus arrhythmia (RSA).
• RSA is mediated by parasympathetic efferent
activity.
Spectral analysis of HRV
• Power spectra of R-R intervals in sleep states
(Vanoli,1995; Otzenberger, 1998).
• Ratio of the low-to-high frequency spectra is used as
an index of parasympathetic-sympathetic balance.
Sleep & Sympathetic-Nerve Activity
Phasic REM
Tonic REM
• During NREM: Sympathetic-Nerve Activity (SNA) reduced.
• Increase in SNA with K complexes (during N2).
• In REM: SNA frequency & amplitude, increased.
• Inhibition of SNA during tonic REM
(Somers et al,1993).
Tonic & phasic REM
• During tonic REM: No PGO in lateral geniculate nucleus
(LGN).
• No theta rhythm in hippocampal (CA1) leads.
• Changes in HR & respiration
(Verrier et al, 1966).
Sympathetic activity & cardiovascular
physiology during sleep
• HR & BP lower during NREM sleep than during wakefulness.
• SNA lower during NREM (stages 3 & 4), increased during REM.
• BP & HR similar during REM & wakefulness.
(Murali, Svatikova & Somers, 2003).
Sleep & gene encoding
• Immediate-early gene mapping shows neuronal activity changes in
forebrain, on sleep deprivation (Thompson et al, 2010).
• Decrease in delta activity parallel increase in retinoic acid receptor beta
gene expression in rats (Maret et al. , 2005)
• Protein synthesis in the brain is increased during NREM sleep.
• Quantification of sleep, on the basis of imaging of gene activity, is a
distant possibility.
PET studies during sleep
• PET is silent and relatively non-invasive, and can be
acquired during natural sleep in healthy humans.
• Biologically active positron-emitting radionuclide is
injected, and the concentrations of tracer indicate tissue
metabolic activity, or cerebral blood flow in terms of
regional activity.
Result of PET studies
During NREM sleep
• Blood flow is reduced in orbital, dorsolateral prefrontal and inferior
parietal heteromodal association cortices, brainstem, thalamus
and basal forebrain.
(Braun et al., 1997)
During REM sleep
• Blood flow is increased in pontine tegmentum, left thalamus, both
amygdaloid complexes, anterior cingulate cortex and right parietal
operculum.
• Blood flow is reduced bilaterally, in dorsolateral prefrontal cortex,
in parietal cortex (supra-marginal gyrus) posterior cingulate cortex
and precuneus.
(Maquet et al., 1996)
Functional Magnetic Resonance
Imaging (fMRI) during sleep
Preoptic area, known as the regulator of the sleep-wake
cycle, showed fMRI changes during NREM sleep.
(Khubchandani, Mallick, Jagannathan & Kumar, 2003 & 2005).
Human fMRI during sleep
• Though fMRI provides real-time imaging of neuronal
activity, this technique in sleep research has suffered
due to high noise levels in the scanner.
(Kaufmann et al, 2006).
Sleep assessment: future-1
• Many physiological changes do occur during
sleep.
• Though we know the functions of some
physiological changes, there are many that we
still do not know.
• Present polysomnographic assessment is helpful
in clinical situations and in healthy subjects.
• But we are still far from adequate assessment of
sleep.
• Electrophysiologically and behaviorally defined
sleep do not explain all the aspects of sleep.
• Incorporation of more physiological parameters
will help in better analysis of sleep.
Sleep assessment: future -2
• The present analysis of sleep does not provide
any quantification of dreams.
• It is essential to develop a technique to record
and quantify dream phase of sleep.
• There may be some basis in the ancient wisdom
of putting emphasis on dreams.
• Better quantitative and qualitative analysis of
sleep is essential for understanding its role in
health and survival.
• It also is essential for better clinical diagnosis.
• It is essential to continue with the manual scoring
and checking till a better system is evolved.
Thank You