NIGHT SLEEP ELECTROENCEPHALOGRAM POWER SPECTRAL
ANALYSIS IN EXCESSIVE D A Y T I M E SLEEPINESS DISORDERS
RUBENS REIMÃO *
SUMMARY — A group of 53 patients (40 míales, 13 females) with mean age of 49 years,
ranging from 30 to 70 years, was evaluated in the. following excessive daytime sleepiness
(EDS) disorders : obstructive sleep apnea syndrome (B4a), periodic movements in sleep (B5a),
affective disorder (B2a), functional psychiatric non affective disorder (B2b). We considered
all adult patients referred to the Center sequentially with no other distinctions but these
three criteria: (a) EDS was the main complaint; (b) right handed ; (c) not using psychotropic
drugs for two weeks prior to the all-night polysomnography. EEG (C3/A1, C4/A2) samples
from 2 to 10 minutes of each stage of the first REM cycle were chosen. The data was
re-corded simultaneously in magnetic tape and then fed into a computer for power spectral
analysis. The percentage of power ( P P ) in each band calculated in relation to the total EEG
power was determined of subsequent sections of 20.4 s for the following frequency bands:
delta, theta, alpha and beta. The PP in all EOS patients sample had a tendency to decrease
progressively from the slowest to the fastest frequency bands, in every sleep stage. PP
dis-tribution in the delta range increased progressively from stage 1 to stage 4; stage REM
levels were close to stage 2 levels. In an EDS patients interhemispheric coherence was high in
every band and sleep stage. B4a patients sample PP had a tendency to decrease progressively
from the slowest to the fastest frequency bands, in¡ every sleep stage; PP distribution in the
delta range increased progressively from stage 1 to stage 4; stage REM levels were between
stage 1 and stage 2 levels. B2a patients sample PP had a tendency to decrease
progressi-vely from the slowest to the fastest frequency bands, in every sleep stage; PP distribution
in the delta range increased progressively from stage 1 to stage 4; stage REM levels were
close to stage 2 levels. B2b patients sample PP had a tendency to decrease progressively
from the slowest to the fastest frequency bands, in every sleep stage; PP distribution in
the delta range increased progressively from stage 1 to stage 3; stage 4 levels were close
to stage 3 levels; stage REM levels were close to stage 2. B5a patients sample PP had a
tendency to decrease progressively from the slowest to the fastest frequency bands, in every
sleep stage; PP distribution in the delta range increased progressively from stage 1 to
stage 3; stage REM levels were close to stage 2 levels, Interhemispheric coherences of B4a,
B2b, and B5a groups were high in, every band and sleep stage. B4a, B2a, B2b, and B5a
power spectral analysis comparison showed higher PP in B2b stage 1 alpha band, as well as,
higher PP in B5a stage 2 theta band. The B4a versus. B2a power spectral analysis
compa-rison showed higher PP in B4a stages 1 and REM alpha bands, as well as higher PP in
B2a stage REM delta band.
Análise do espectro de potência do eletroencefalograma durante o sono: estudo de pacientes
com sonolência excessiva diurna.
RESUMO — Foram avaliados 53 pacientes (média de idade — 49,0 anos; variação de 30-70
anos) com o objetivo de analisar o espectro de potencia do EEG durante o sono nos seguintes
distúrbios que provocam sonolência excessiva diurna: síndrome de apnéia do sono tipo
obstru-tivo, movimentos periódicos do sono, distúrbios afetivos, distúrbios psiquiátricos funcionais
não afetivos. Os dados obtidos indicam primeiramente que os diversos estágios do sono
po-* Sleep Disorders Center, Baptist Memorial Hospital, Memphis, Tennessee, USA. This
research was partially supported by Coordenadoria de Aperfeiçoamento de Pessoal de
En-sino Superior (CAPES).
d e m s e r d i s t i n g u i d o s q u a n t i t a t i v a m e n t e e n t r e o s p a c i e n t e s c o m s o n o l ê n c i a e x c e s s i v a d i u r n a . E v i d e n c i a m , e m s e g u n d o l u g a r , q u e o s g r u p o s d i a g n ó s t i c o s t ê m p a d r õ e s p r ó p r i o s d e d i s t r i -b u i ç ã o d o e s p e c t r o d e p o t ê n c i a . E m t e r c e i r o , m o s t r a m a a u s ê n c i a d e a s s i m e t r i a s m a n t i d a s .
T h e all-night polysomnogram study is usually made by visual inspection, fol-lowing standardized parameters 16. T h i s is an accurate tool for clinical purposes. However, the computerized study of the electroencephalogram ( E E G ) during sleep revealed new aspects and details not visually detectable. T h e power spectral analysis was the first system utilized for quantitative E E G research during sleep, by Grass & col.8 in 1938, followed by Knott & col.10 in 1942. T h e power spectrum (also called auto-spectrum of variance) shows the distribution of average intensity (mean square value or variance) of the E E G in relation to its frequency 5. It may be studied in its original unit ( ¿ ¿ V2
) but, more often, due to its methodological and statistical a d v a n t a g e s1 4
, it is applied as the percentage of power ( P P ) in a given frequency in relation to the power of a given spectrum. Sleep discorders accompanied by excessive daytime sleepiness ( E D S ) form the main group of patients who look for Sleep Disorders Centers, comprising 3 6 . 0 ot 4 2 . 2 % of t h o s e2
. In this respect, it is remarkable the lack of literature in connection with the use of this quantitative tool (the E E G power spectral analysis) in such pathological group.
T h e objective of this paper was to determine the E E G power spectrum during sleep in the following E D S disorders: sleep apnea syndrome ( B 4 a ) , periodic move-ments in sleep ( B 5 a ) , affective disorders ( B 2 a ) , functional psychiatric non affective disorders ( B 2 b ) . T h i s nomenclature and codes follow the international classification i.
M E T H O D S
We evaluated 53 consecutive patients (40 men and 13 women) with E D S (Table 1). Mean age was 49.0 years (range 30 to 70 years). The selection criteria systematically applied were: (a) E D S was the main complaint; (b) right handed; (c) not using psychotropic drugs for two weeks prior to the polysomnography.
All-night standard p o l y s o m n o g r a p h y recordings included E E G ( C 3 / A 1 ; C4/A2) accor-ding to 10-20 System 9 criteria; electrooculogram using common reference ( L . E / F P ; F P / R E ) ; electromyogram of submentalis and anterior tibialis muscles; buccal and nasal airflow mea-sured by thermocouples; respiratory effort detected by thoracic and abdominal pneumograms; continuous transcutaneous oxygen saturation monitoring. E E G sensitivity of 7.5 ^ V / m m , high
frequency filter of 90.0 H z , and low frequency filter of 1.0 Hz were used.
E E G samples of each sleep stage of the first R E M cycle were obtained. Sample dura-tion varied from 2 to 10 min. An effort was made to obtain the longest possible time for each stage without the interference of artifacts, however, we established a maximum limit of 10 min. Slesep staging followed the standardized criteria of Rechtschaffen, & col. 16. Data was simultaneously tape recorded, and liater fed into a computer (Med-80, Nicolet Inst. Co.), using a power spectral analysis software 15. Each B E G frequency band PP was calculated 19 in relation ot the total power for 20.4 s consecutive) segments, for the following bands: delta (0-3 H z ) , theta (4-7 H z ) , alpha (8-12 H z ) land beta (13-20 H z ) .
T h e analyzed samples did not include awakenings, arousals, artifacts, apneas nor pe-riodic leg movements. Such a selection was obtained employing three distinct quality controls:
(1) during the visual scoring on the paper E E G records; (2) during visual scoring of the computer screen; (3) by means of computer filters for excessively high amplitude waves 11.
and statistically applicable, pairs of variables were compared using the Newman-Keules test; thé Analysis of Variance w a s only applied after considered adequate by homogeneity tests of Cochrans and Bartlett. ( d ) . PP distribution differences between diagnostic groups were compared using the A n a l y s i s of Variance test, for each stage and band ( T a b l e 5 ) ; Newman-Keules test w a s also used when statistically adequate, ( e ) . T h e t test, for each stage and frequency band w a s also used to establish a detailed comparison between diagnostic groups; this test was only applicable to B4a and B2a groups; groups B2b and B5a did not have sufficient sample size to use the! t test at its best.
R E S U L T S
Data obtained follows the siame five steps described above:
a. The PP distribution in, the whole 53 E D S patients sample, showed a progressive decrease of PP from the highest to the lowest frequency bands, in every sleep stage ( T a b l e 2 ) . PP in delta band increased progressively from stage 1 to 4, and the R E M values were close to stage 2. PP in theta band decreased progressively from stiage 1 to 4, and the values of stage R E M were higher than stage 1. PP in alpha band decreased progressively from stage 1 to 4, and the values of stage R E M were close to stage 2. PP in beta band decreased progressi-vely from stage 1 to 4, and the values of stiage R E M were close to stage 2. T h e SD varied from 1.4 to 14.0 with a tendency of the higher SD to occur in the lowest frequency bands. The PP interhemispheric differences varied from —2.4 to 1.2, with a tendency of the higher differences to occur in the lowest frequency bands (Table 3 ) .
b. PP distribution in each diagnostic group was determined in this second step. The B4a group showed PP distribution similar to the whole E D S patients sample, with a progressive increase from the highest to* the lowest frequency bands, in every stage. PP in each fre-quency band also showed distribution similar to the whole E D S sample except for the R E M values in delta band that were between stages 1 and 2 values. SD varied between 0.8 and 18.8, with a tendency of the higher SD to occur in the lowest frequency bands. T h e PP interhemispheric differences varied from —3.8 to 1.5, With a tendency of the higher diffe-rences to occur in the lowest frequency bands.
The B2a group showed PP distribution similar to the whole E D S patients sample, with a progressive increase from the highest to the lowest frequency bands, in every stage,
except for the values of stage 4 theta and alpha bands that were close together. PP in each frequency band also showed distribution similar to the whole E D S sample except for: (1) stages 3 and 4 delta band were) close together; (2) stages 1 and 2, as well as, stages 3 and 4 alpha bands w e r e close together; (3) stage R E M alpha band was close to stages 3 and 4. SD varied between 0.5 and 13.2, with a tendency of the higher SD to occur in the lowest frequency bands. T h e PP interhemispheric differences varied from —7.4 to 3.9, with a ten-dency of the higher differences to occur in the lowest frequency bands.
T h e B 2 b group showed PP distribution similar to the whole E D S patients sample, with a progressive increase from the highest to the lowest frequency bands, in every stage, except for the stage 1 delta and theta bands values that were lower than the alpha band values. PP in each frequency band also showed distribution similar to the whole E D S sample except for: (1) stages 3 and 4 delta bands w e r e close together; (2) stage 2 theta band w a s slightly higher and stages 1, 3, and 4 were close together; (3) stages 3 and 4 beta bands asi well as alpha bands were close together. SD varied between 0.4 and 17.4 w i t h a tendency of the higher SD to occur in the lowest frequency bands. T h e PP interhemispheric differences va-ried from —2.5 to 4.2, with a tendency of the higher differences to occur in the lowest frequency bands.
The B5a group showed distribution similar to the whole E D S patients sample, with a progressive increase from the highest to the lowest frequency bands1
, in every stage, except for the stage 1 theta band that had a tendency to be lower than the alpha band. PP in each frequency band also showed similar distribution to the whole E D S sample except for: (1) stages 3 and 4 delta bands werei close together, with a slight tendency to be lower in, stage 4;
c. PP was compared between, the sleep stages, for every frequency band, according to the diagnostic groups ( T a b l e 4 ) , showing significant differences in every comparison, except for:
(1) B2a alpha band; (2) B2b theta band; (3) B5a theta and alpha bands.
Submitting only the B4a group to N e w m a n - K e u l e s test, based on all 4 frequency bands, all the sleep stages were significantly different from each other, except for the distinction beween stages 3 and 4.
Submitting only the B2a group to Newmian-Keules test, based on all 4 frequency bands, all the sleep stages were significantly different from each other, except for the distinctions between stages 2 and 4, as well as, stages 3 and 4. There were no significantly different PP pairs in the alpha band, at the p < [ 0 . 0 5 level.
Submitting only the B2b group to Newman-Keules test, based on all 4 frequency bands, only stage 1 could be differentiated from the others. There were no significantly different PP pairs in the theta band, at the p < 0 . 0 5 level.
Submitting only the B5a group to Newman-Keules test, based on all 4 frequency bands, all the sleep stages were significantly different from the others, except for the distinctions between the following pairs of stages: 1—2, 1—REM, 2 — R E M , 3—4, 2—4 and 4—REM, in
the delta band; 1—2, 1—£, 1—REM, 3—4, 3 — R E M land 4 — R E M , in the beta band. There were no significantly different PP pairs in theta and alpha bands, at the p < 0 . 0 5 level. d. PP in every, band and stage was compared between the four diagnostic groups ( T a b l e 5). T h e y showed significant differences restricted to stage 1 alpha and stage 2 theta bands; W h e n taking exclusively stage 1 alpha band -and comparing the diagnostic groups, it became evident that the significant difference w a s between B2a and B2b. W h e n taking exclusively stage 2 theta band and comparing the diagnostic groups, it became evident that the signi-ficant difference was between B5a and all the others.
C O M M E N T S
T h e power spectral analysis methodology is particularly adequate for studying sleep due to its capacity of detecting a consistent oscilation, phasic or tonic, in a limited time period, even if this activity has small a m p l i t u d e1 2
. Gasser & c o l .7
have demonstrated that the relative potency, such as we have used in this study, gives more reliable results than the use of absolute potency, which is subject to individual variations. M a t o u s e k1 4
also recommends to present it as PP in each band, as it was done in this report. In this research, the E E G derivations followed the classical ones that have been used in sleep for the last two d e c a d e s1 6
. T h e option to register the spectrum potency without common reference w a s due to the smaller distortion of the potentials, as recommended by Fein & c o l A
T h e literature reveals a lack of studies with similar methodology, utilizing the power spectral analysis in E D S patients. T h e PP distribution pattern in our total E D S patients sample showed PP reduction from the lowest to the highest frequency bands, in all sleep stages. T h i s pattern w a s previously described in normal adults 5.8.10. T h e same pattern w a s found in our subgroups, when divided according to the diagnosis.
Our total E D S sample showed increase of the PP in delta band from stage 1 to 4; this band values in stage R E M are similar to stage 2. T h i s increase w a s also previously described in normal adults 5. Such finding indirectly shows the sleep stages (particularly stages 2,3 and 4) definition criteria, which takes into conside-ration the proportion of high amplitude delta waves in a given time-limited sample 16.
In the total E D S sample, theta band had progressive decrease from stage 1 to 4, with R E M values higher than stage 1. In the alpha band, there w a s also a progressive reduction from stage 1 to 4, with stage R E M values similar to stage 2.
In the beta band, there w a s a decrease from stage 1 to 4, and the stage R E M values were similar to stage 2. T h e difficulty in distinguishing between stage R E M and stage 1, by this m e t h o d ^ , wa s usually attributed to the similarity of their power
spectral distributions. However, our data suggests that the differentiation between these two stages, or between stage R E M and stage 2 should be considered in a dynamic perspective, as it changes according to the frequency band, as well as, the pathology taken into consideration. Stage R E M is heterogeneous in its time course, e.g. showing sawtooth waves that are detected in the power spectral analysis as a PP increase in theta band. T h e s e irregularly distributed waves along R E M sleep change the PP in a given sample and so does the comparison with stages 1 and 2.
T h e beta band peak in stage 2, around 13-15 Hz, corresponding to sleep spindles 5, can not be detected by the method we have used as the potency is distri-buted for the whole beta band. T h e absence of peaks does not prevent the compa-rison between stages because the beta band potency will tend to elevate reflecting the spindles. It also should be considered that the increase in sigma band in non-REM sleep does not always correspond to spindles, and an increase in this band may occur even without visually detectable spindles 3
»4
.
T h e progressive increase in delta band P P , from stage 1 to 4 allows a quan-titative distinction between these stages and is partially related to the definition criteria that considers the percentage of delta contingent 16. However, the power spectral analysis determines the delta potency in a given time and not the duration of delta w a v e s in this given sample. In spite of being distinct parameters, some previous data suggests that the power spectral analysis may be utilized for sleep staging with the usual visual standardization 16. T h e two methods show an agreement of 8 5 - 9 2 % in normal adults records 17.
T a n g u a y & col.20 described the E E G power spectrum in normal children. In the 3 4 - 5 6 months old, the 11-15 Hz band had PP of 3.3 in stage 2 and 1.8 in the same band in stage R E M . Comparing with our results, beta band PP are respectively 8.9 at right and 8.5 at left hemispheres, as well as 9.3 at right and 8.7 at left. Regarding the 0-3 Hz band, those children 20 had stage 2 PP of 6 6 . 5 and stage R E M PP of 73.6. In our sample, delta band PP w a s 50.8 at right and 5 2 . 3 at left in stage 2; 45.6 at right and 47.2 at left in stage R E M . T h i s comparison supports and objecti-vely quantifies the well known decline of delta band from infancy to adulthood.
Our power spectral analysis data of the four major E D S diagnostic groups
points out that the sleep stages m a y be quantitatively distinguished by this method.
Second, it shows that different diagnostic groups have distinct power spectral patterns.
Third, it gives quantitative support to the lack of marked sustained asymetries concept,
in every diagnostic group. Finally, it suggests that broader use of quantitative E E G
power spectral analysis could help differentiating and understanding the E D S diseases.
A c k n o w l e d g e m e n t s — O u r g r a t i t u t e t o H e l i o L e m m i , M . D . , D i r e c t o r o f t h e B a p t i s t M e -m o r i a l H o s p i t a l S l e e p D i s o r d e r s C e n t e r f o r h i s a d v i c e a n d s u p p o r t . W e t h a n k a l s o A r o n D i a m e n t , M . D . , R o g e r V a n d e r Z w a a g , P h . D . , H a g o p A k i s k a l , M . D . , A l d o B e v i l a c q u a , M . D . , J o s e p h B e l l u o m i n i , R o n a l d C o w a n a n d S h a r o n B u r t f o r i n v a l u a b l e h e l p , statistical, a n d t e c h
-n i c a l a s s i s t a -n c e .
R E F E R E N C E S
1 . A s s o c i a t i o n o f S l e e p D i s o r d e r s C e n t e r s . C l a s s i f i c a t i o n o f s l e e p a n d a r o u s a l d i s o r d e r s . S l e e p 1979, 2 : 1 .
2 . C o l e m a n R M . D i a g n o s i s , t r e a t m e n t , a n d f o l l o w - u p o f a b o u t 8,000 s l e e p / w a k e d i s o r d e r
p a t i e n t s . I n G u i l l e m i n a u l t C , L u g a r e s i E ( e d s ) : S l e e p / W a k e D i s o r d e r s : N a t u r a l H i s t o r y , E p i d e m i o l o g y , a n d L o n g - t e r m E v o l u t i o n . N e w Y o r k . R a v e n P r e s s 1983, p 87-98.
3 . D u m e r m u t h G , G a s s e r T , L a n g e B . A s p e c t s o f E E G a n a l y s i s i n t h e f r e q u e n c y d o m a i n . I n D o l c e G , K u n k e l H ( e d s ) : C o m p u t e r i z e d E E G A n a l y s i s . S t u t t g a r t : G u s t a v F i s h e r V e r l a g , 1975, p 429-457.
4 . D u m e r m u t h G , S c o l l o - L a v i z z a r i G . S p e c t r a l a n a l y s i s o f E E G i n n o r m a l a d u l t s . I n K o e l a W P , L e v i n P ( e d s ) : S l e e p — P h y s i o l o g y , B i o c h e m i s t r y , P s y c h o l o g y , P h a r m a c o l o g y , C l i n i c a l I m p l i c a t i o n s . B a s e l : K a r g e r , 1973, p 246-249.
5 . D u m e r m u t h G , W a l t z W , S c o l l o L a v i z z a r i G , K l e i n e r B . S p e c t r a l a n a l y s i s o f E E G a c t i -v i t y i n d i f f e r e n t s l e e p s t a g e s i n n o r m a l a d u l t s . E u r N e u r o l 1972, 7:265.
6 . F e i n G , R a z J , B r o w n F F , M e r r i n E L . C o m m o n r e f e r e n c e c o h e r e n c e dato, a r e c o u n - ¬ f o u n d e d b y p o w e r a n d p h a s e e f f e c t s . E l e c t r o e n c e p h a l o g r C l i n N e u r o p h y s i o l 1988, 69:581.
7 . G a s s e r T , V e r l e g e r R , B a c h e r P , S r o k a L . D e v e l o p m e n t o f t h e E E G o f s c h o o l a g e c h i l -d r e n a n -d a -d o l e s c e n t s : I . A n a l y s i s o f b a n -d p o w e r . E l e c t r o e n c e p h a l o g r C l i n N e u r o p h y s i o l 1988, 69:91.
8 . G r a s s A M , G i b b s F A . A F o u r i e r t r a n s f o r m o f t h e e l e c t r o e n c e p h a l o g r a m . J N e u r o p h y -s i o l 1938, 1 : 521.
9 . J a s p e r H H . R e p o r t o f t h e c o m m i t t e e o n m e t h o d s o f c l i n i c a l e x a m i n a t i o n i n e l e c t r o -e n c -e p h a l o g r a p h y . E l -e c t r o -e n c -e p h a l o g r C l i n N -e u r o p s y s i o l 1958, 10:371.
10. K n o t t J R , G i b b s A , H e n r y C E . F o u r i e r t r a n s f o r m o f e l e c t r o e n c e p h a l o g r a m d u r i n g s l e e p . J E x p e r P s y c h o l 1942, 31:465.
11. K t o n a s P Y , O s o r i o P L , E v e r e t t R L . A u t o m a t e d d e t e c t i o n o f E E G a r t i f a c t s d u r i n g s l e e p : p r e p r o c e s s i n g f o r a l l - n i g h t s p e c t r a l a n a l y s i s . E l e c t r o e n c e p h a l o g r Clin N e u r o p h y s i o l 1979, 46 : 382.
12. K u w a h a r a H , H i g a s h i H , M i z u k i Y , M a t s u n a r i S , T a n a k a M , I n a n a g a K . A u t o m a t i c r e a l t i m e a n a l y s i s o f h u m a n s l e e p s t a g e s b y a n i n t e r v a l h i s t o g r a m m e t h o d . E l e c t r o e n c e p h a -l o g r C -l i n N e u r o p h y s i o -l 1988, 70:220.
13. L a r s e n L E , W a l t e r D O . O n a u t o m a t i c m e t h o d s o f s l e e p s t a g i n g b y E E G s p e c t r a l p o w e r d u r i n g R E M p e r i o d s . S l e e p R e s 1981, 10:151.
14. M a t o u s e k M . F r e q u e n c y a n a l y s i s : p r o c e s s i n g o f o u t p u t d a t a . I n R e m o n d A , ( e d ) : H a n -d b o o k o f E l e c t r o e n c e p h a l o g r a p h y a n -d C l i n i c a l N e u r o p h y s i o l o g y . A m s t e r -d a m : E l s e v i e r 197), 5A61-5A66.
15. N i c o l e t I n s t r u m e n t C o r p o r a t i o n . N i c o l e t M e d 8 0 F r e q u e n c y A n a l y s i s S o f t w a r e . W i s c o n -s i n : N i c o l e t , 1975.
16. R e c h t s c h a f f e n A , K a l e s A . M a n u a l o f S t a n d a r d i z e d T e r m i n o l o g y , T e c h n i q u e s a n d S c o r i n g f o r S l e e p S t a g e s o f H u m a n S u b j e c t s . L o s A n g e l e s : B r a i n I n f o r m a t i o n S e r v i c e , U n i -v e r s i t y o f C a l i f o r n i a , 1968.
17. S a l i n s k y M , G o i n s S , S u t u l a T , R o s c o e D , W e v e r S . C o m p a r i s o n o f s l e e p s t a g i n g b y p o l y g r a p h a n d c o l o r d e n s i t y s p e c t r a l a r r a y . S l e e p 1988, 11:131.
18. S c h u l t e F J , B e l l B. B i o e l e c t r i c b r a i n d e v e l o p m e n t . N e u r o p e d i a t r i c s 1973, 4 : 30.
19. S c i a r e t t a G , E r c u l i a n i P . A p r o p o s a l f o r c a l i b r a t i n g E E G s p e c t o g r a m s . E l e c t r o e n c e -p h a l o g r C l i n N e u r o -p h y s i o l 1978, 45 : 674.
