Sleep, 4(4):400-407

© 1981 Raven Press, New York

K-Complexes and Sleep Spindles Before

Transient Activation During Sleep

Downloaded from https://academic.oup.com/sleep/article-abstract/4/4/400/2750238 by guest on 10 April 2019

J. Ehrhart, M. Ehrhart, A. Muzet, J. P. Schieber, and P. Naitoh

Centre d' Etudes Bioc/imatiques du CNRS, Strasbourg, France

Summary: Six subjects spent three consecutive nights in the sleep laboratory.
Activation phases (PATs), spontaneous K-complexes, and sleep spindles were
visually detected in sleep stages 2 and 3 for nights 2 and 3. The K-complex rate
was significantly greater in the 10 sec prior to the PATs than at any other time
spent in stage 2 or 3. K-complexes associated with sleep spindles occurred
significantly less frequently during the epochs just preceding the PATs. In all
subjects, there was a sharp increase of sleep spindles associated with K-
complexes when PATs did not follow within IO sec. These results suggest that
spontaneous K-complexes and sleep spindles act antagonistically with respect
to the occurrence of PATs. These two phasic events are significantly related to
regulating the probability of occurrence of PATs in sleep stages 2 and 3; K-
complexes may reflect an organismic state leading towards PAT, whereas sleep
spindles may inhibit the occurrence of PAT. Key Words: Human sleep---K-
complexes-Sleep spindles-Activation phases.

The physiological features of the phases of transient activation (phases d'acti-

vation transitoire: PAT) have been described previously (Schieber et aI., 1971;
Muzet et aI., 1972, 1973, 1974; Naitoh et aI., 1973; Ehrhart and Muzet, 1974). In
human sleep, PATs are characterized by transient signs of arousal: increased
electro myographic (EMG) tone, reappearance in electroencephalogram (EEG) of
either alpha or low-voltage activity, heart rate increase, decreased finger pulse
volume, and, usually, body movements varying in duration and intensity.
A few recent studies have suggested that sleep spindles and K-complexes pre-
cede PATs. Sassin and Johnson (1968) found that body movements (hence, the
PATs) in non-rapid eye movement (NREM) sleep stage 2 often were preceded by
K-complexes. Ferriere (1970) observed that 80% of the PATs in NREM sleep
stage 2 occurred during periods of decreasing sleep spindle activity.
The electrophysiological characteristics of PATs, sleep spindles, and K-

Accepted for publication July 1981.

Address correspondence and reprint requests to Jean Ehrhart at Centre d'Etudes Bioclimatiques du
CNRS, 21 rue Becquerel, 67087 Strasbourg Cedex, France.
Dr. Naitoh's permanent address is Naval Health Research Center, San Diego, California, U.S.A.
This article was prepared while Dr. Naitoh was on sabbatical from his laboratory.

400
 K-COMPLEXES, SPINDLES, AND ACTIVATION PHASES 401

complexes are fairly well known. However, the functional significance of these
phasic sleep parameters remains virtually unknown, as does the interrelation
among these events. In particular, a point of major interest and controversy has
been the functional role of the sleep spindles.
In 1971, Yamadori reported "suppression" of evoked K-complexes when sen-
sory stimuli (clicks) were presented synchronously with sleep spindles in NREM
sleep stage 2. He concluded that spindles can preserve sleep by inhibiting sensory

Downloaded from https://academic.oup.com/sleep/article-abstract/4/4/400/2750238 by guest on 10 April 2019

input or by raising the arousal threshold. On the contrary, other researchers have
observed an increase in evoked K-complexes to spindle-synchronous stimulation,
and concluded that sleep spindles could be excitatory events. Church et al. (1978)
used 10 msec "clicks" during NREM stage 2 and slow wave sleep (SWS; NREM
stages 3-4) either during spindle bursts or during interburst periods. Cortical
EEG responses showed no evidence of spindle "suppression"; moreover, evoked
K-complexes were potentiated by the spindle-synchronous stimulation, and these
authors suggested that this potentiation of evoked K-complexes could be due to
phasic reductions in inhibitory action during sleep spindles.
The present report deals with the exploration of various interrelationships be-
tween K-complexes, sleep spindles, and PATs in NREM sleep stages 2 and 3.
These relationships have been studied in both undisturbed nights and nights dis-
turbed by acoustical stimulations.
The rationale for this comparison was to test the consistency of the relation-
ships, if any, between the three phasic activities. PATs can be used to explore the
physiological functions of K-complexes and sleep spindles. Two hypotheses were
examined: (1) that K-complexes could reflect an organismic state inducing PATs;
and (2) that sleep spindles might inhibit the occurrence of PATs.

£.E.O. 1 : FR.-OCCIP. O.

E.E.0.2 : MAST.-VERT. O.

E.O.O.

E.K.O.

ACTOORAM

100~vL
hac

FIG. 1. Example of K-complex followed by a PAT within 10 sec. In descending order, the polygraph
channels are: EEG 1 (frontal-parietal derivation), EEG 2 (vertex referenced to the left mastoid), EOG
(electrodes obliquely placed from the left to right eye), electrocardiogram (EKG), and actogram. The
actogram records the movements of a specially constructed actographic bed. In this record, the PAT is
characterized by (1) changes in EEGs from slow-wave-dominant pattern to low-amplitude form, (2)
appearance of muscle and movement artifact on the EEG and EOG channels, (3) increased heart rate,
and (4) body movements (about 90% of the activation phases are accompanied by body movements).
The elapsed time between the upward peak of the K-complex and the first visible movement of the bed
on actogram channel is approximately 8 sec. However, the earliest sign of this PAT is muscle artifact
on the EEG, which occurred within 3 sec after the K-complex.

Sleep, Vol. 4, No.4, 1981

402 1. EHRHART ET AL.

METHODS
Six young adults, ages 21-24 (mean, 22.7), spent three consecutive nights inthe
sleep laboratory. The first night, NO, was treated as an adaptation night. Elec-
trophysiological recordings of sleep (F3-P3 and A 1-Cz EEGs, right and left
electro-oculograms, (EOGs), EMG of the chin, electrocardiogram, and actogram)
were obtained in the next two nights, Nl and N2.

Downloaded from https://academic.oup.com/sleep/article-abstract/4/4/400/2750238 by guest on 10 April 2019

Night Nl was undisturbed, whereas in night N2, the subjects were exposed to
four airplane noises of various intensities and durations randomly distributed
throughout the night at a rate offour noises per hour. The physical characteristics
of these four noises were:

Noise Duration Peak intensity

A 90 sec 77 dB(A)
B 30 sec 80 dB(A)
C 90 sec 82 dB(A)
0 30 sec 96 dB(A).

All sleep records were visually analyzed for detection of PATs, K-complexes,
and sleep spindles in NREM sleep stages 2 and 3 (Fig. 1). These stages were
scored according to the manual of Rechtschaffen and Kales (1968).
K-complexes were defined as slow bi-, tri-, or poly phasic waves, lasting at least
0.5 sec, with amplitude larger than 200 /LV, and appearing as an in-phase wave in
the two EEG channels. Sleep spindles were defined as spindlelike wave forms of
12-15 Hz, lasting more than 0.5 sec, and of amplitude greater than 15JLV. All
PATs were detected visually according to the criteria defined by Schieber et al.
(1971).
Student's two-tailed t-test for correlated means and Wilcoxon's matched-pairs
signed-ranks test were used to determine significance at confidence level of 5% or
better.

RESULTS
The frequency of K-complexes started to increase sharply approximately 20 sec
before PATs. The average frequency of K-complexes calculated for NREM stages
2 and 3 across all subjects was 0.57 K-complexes/min for Nl, and 0.54 K-
complexes/min for N2. Ten sec prior to the PATs, K-complex rates were signifi-
cantly elevated from these basal rates to 2.46 for Nl and 1.92 for N2. This obser-
vation supports the first part of our hypothesis: K-complexes could reflect an
organismic state leading towards transient activation.
Do sleep spindles interact with K -complexes in relation to PATs? For this
analysis, K-complexes were classified into Ko, Kb K2 , and I<:J (Fig. 2). Ko was a
K-complex without contiguously occurring sleep spindles; Kl was a K-complex
associated with sleep spindles that occurredjust prior to the K-complex; K2 was a
K-complex in which sleep spindles occurred during the K-complex; and K3 was a
K-complex in which spindles occurred just after the K-complex. It was

Sleep, Vol. 4, No.4, 1981

 K-COMPLEXES. SPINDLES. AND ACTIVATION PHASES 403

Downloaded from https://academic.oup.com/sleep/article-abstract/4/4/400/2750238 by guest on 10 April 2019

FIG. 2. Different types of associations be-
tween K-complexes and sleep spindles: Kv.
K-complex without contiguous sleep spin-
dle; K t • K-complex associated with a sleep
spindle just prior to the K-complex; K2 •
K-complex associated with a sleep spindle
during the K-complex; K3 • K-complex as-
sociated with a sleep spindle just after the
K-complex.

1 sec

hypothesized that Ko complexes should be more likely followed by PATs than

should complexes K], K2 , and K3, as the spindle components associated with K],
K2 , and K3 complexes would inhibit PATs. The analysis revealed that K-com-
plexes with contiguous sleep spindles (i.e., K1 , K2 , and K3 ) occurred significantly
less frequently during the lO-sec epochs preceding the PATs. For the N 1 records,
only 6.3% of the K-complexes were of the types K], K2 , or K3 in the lO-sec epochs
just preceding PATs, while they were 24.5% in other epochs. Similarly, for the N2

Sleep, Vol. 4. No.4. 1981

404 J. EHRHART ET AL.

TABLE 1. Frequency of the different types of K-complexes:

N I records (undisturbed nights)
K-complexes followed by a PAT K-complexes not followed by a PAT
within 10 sec within 10 sec
K-Comp1ex Ratio"
type SI S2 S3 S4 S5 S6 Total, SI S2 S3 S4 S5 S6 Total, (%)

1<0 4 13 9 11 19 18 74 97 37 29 85 190 112 550 11.9

K, 0 2 1 0 0 0 3 1 5 6 12 6 12 42 6.7

Downloaded from https://academic.oup.com/sleep/article-abstract/4/4/400/2750238 by guest on 10 April 2019

K. 0 0 0 0 0 0 0 0 3 1 12 3 1 20 0.0
Ko 0 1 0 1 0 0 2 2 20 12 45 12 25 116 1.7
Total 4 16 10 12 19 18 79 100 65 48 154 211 150 728 9.8

a [Total,/(Total, + Total,)] X 100.

records, these frequencies were 9.4 and 24.4%, respectively. The difference be-
tween these frequencies was significant for each of the two nights. This result
supports our second hypothesis: sleep spindles can inhibit the occurrence of the
activation phases.
To explore further the interaction of sleep spindles and K-complexes, we
analyzed the different types ofK-complexes not followed by PATs. We expected
to find more of the Ka type, since sleep spindles that occur just after K -complexes
would counteract the "excitatory" effects, thus suppressing PATs.
Therefore, sleep spindles which preceded or were simultaneous with K-
complexes would have less inhibitory effect. To confirm this, we analyzed the
occurrences of sleep spindles 10 sec before and 10 sec after the K-complexes.
The results for NI and N2 are given in Tables 1 and 2, respectively. In all
subjects, there was a sharp increase of sleep spindle activity contiguous with
K-complexes when PATs did not follow within 10 sec (Fig. 3). This increase in
spindles resulted mainly from a significant increase in the number of K3 complexes
in comparison to Kl or K2 complexes.
Figure 3 (Nl and average for the six subjects) shows the distribution of sleep
spindles with respect to the K-complexes that were either followed (broken line)
or not followed (solid line) by a PAT within 10 sec. Results obtained for N2

TABLE 2. Frequency of the different types of K-complexes:

N2 records (noise-disturbed nights)
K-complexes followed by a PAT K-complexes not followed by a PAT
within 10 sec within 10 sec
K-Complex Ratio"
type SI S2 S3 S4 S5 S6 Total, Sl S2 S3 S4 S5 S6 Total, (%)

1<0 0 5 8 11 20 14 58 72 21 49 97 144 171 554 9.5

K, 0 0 0 2 0 1 3 0 4 3 7 16 9 39 7.1
K, 0 0 0 0 1 0 1 0 0 0 2 16 1 19 5.0
Ko 0 1 0 0 1 0 2 3 19 8 35 18 38 121 1.6
Total 0 6 8 13 22 15 64 75 44 60 141 194 219 733 8.0

" [Total,/(Total, + Total,)] x 100.

Sleep. Vol. 4, No.4, 1981

 K-COMPLEXES, SPINDLES, AND ACTIVATION PHASES 405

150
200

K3

OIl

~ 100
c:
150 'Q.

Downloaded from https://academic.oup.com/sleep/article-abstract/4/4/400/2750238 by guest on 10 April 2019

(J)

Co
OIl II
II

-
~
iii
"
.!:
Co 0
(J) ... Kl

~
.
Co

100
II
.0
E
50

-.....
(J) :>
z
0

.0 K2
E
:>
z 0
50
Kl K2 K3
40
30
20
10
0
... -- -
........ ...... .......... ~ ....
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 3 4 5 6 7 8 9 10
K
Seconds
~
Kl+K2 +K3
FIG. 3. Distribution during NI (undisturbed night) of sleep spindles with respect to the K-complexes
that are either followed (broken line) or not followed (solid line) by a PAT within 10 sec.
The following details apply to Figs. 3 and 4. The K-complexes are indicated by arrows (time 0). The
types of temporal association between K-complex and sleep spindles (K" K2 , or K3 ) are not distin-
guished in the main graph but are given in the insert. The Y-axis of the insert is of the same scale as the
one used in the main graph. Whereas the main graph presents the 10-sec epochs preceding and
following the K-complex, the insert is limited to 2 sec centered by the K-complex. The insert, how-
ever, shows the time relationship of sleep spindles to K-complexes, i.e., before (KtI, during (K2 ), or
after (I<,,) the K-complex.

(disturbed night) revealed a similar pattern (Fig. 4). For the six subjects in Nl, the
total number of the Kt. K2 , and K3 not followed by a PAT was 178, whereas only 5
were followed by a PAT. In N2, these numbers were 179 and 6, respectively.
DISCUSSION
Our results seem to agree with Yamadori's hypothesis (1971) that sleep spindles
have an inhibitory effect on the evoked K-complexes, whereas they appear to
contradict those of Church et al. (1978), who attributed to the spindles an activating
effect on the K-complexes. But it certainly should be emphasized that in our
experiment we considered only spontaneous K-complexes (i.e., K-complexes OCe
curring without any detectable external stimulus), whereas in the studies by both

Sleep, Vol. 4, No.4, /98/

406 J. EHRHART ET AL.

150
I 1
200
K3

oil

~ 100
"0
c:
150 '0.

Downloaded from https://academic.oup.com/sleep/article-abstract/4/4/400/2750238 by guest on 10 April 2019

(/)

..
oil
~
..
0-

=0
-.
(/)
c:
0- 0
(/)

.. K1
~

0- .<l
50
~ 100 E

-.
::l
(/)
z
0
~

.<l K2
E
::l
Z 0
50
K1 K2 K3
40
30
20
10
0

FlG.4. Distribution during N2 (disturbed night) of sleep spindles with respect to the K-complexes that
are either followed (broken line) or not followed (solid line) by a PAT within 10 sec. Please see the
legend to Fig. 3 for additional details.

Yamadori and Church, only evoked K-complexes provoked by impulse noise

(clicks) were studied. In our experiment, the noises used in N2 were passing
airplane noises lasting for 30-90 sec. Thirty-two noises were presented at a lim-
ited rate (4/hr). Approximately half of them occurred in NREM sleep stages 2 or 3,
and only a few were contiguous to one of the 797 K-complex-spindle associations
scored. Therefore, the effect of these noises on the EEG (evoked K-complexes)
can be considered as statistically negligible. The noises used in our study are very
different from those used in Yamadori and Church studies. For all these reasons,
our experiment cannot be compared with theirs. K-complexes and sleep spindles
could act differently according to whether their association is spontaneous or
produced by an external stimulus. Such a possibility could explain some of the
differences observed among various studies in the interaction between these
events.
Although this report is not the first to explore the interaction between K-
complexes and sleep spindles, it gives a new approach to the study of their re-

Sleep. Vol. 4, No.4, 1981

 K-COMPLEXES, SPINDLES, AND ACTIVATION PHASES 407

lationship with respect to the occurrence of PATs. Our results suggest that spon-
taneous K-complexes and sleep spindles act antagonistically: spontaneous K-
complexes reflect an organismic state leading towards PAT, while sleep spindles
inhibit the occurrence of PAT. Although we cannot state that K-complexes actu-
ally trigger PATs or that sleep spindles inhibit PATs for a certain time, our results
indicate that the spontaneous occurrence of these two phasic events is signifi-
cantly related to regulating the occurrence of PATs in NREM sleep stages 2 and 3.

Downloaded from https://academic.oup.com/sleep/article-abstract/4/4/400/2750238 by guest on 10 April 2019

REFERENCES
Church MW, Johnson LC, and Seales DM. Evoked K-complexes and cardiovascular responses to
spindle-synchronous and spindle-asynchronous stimulus clicks during NREM sleep. Electroen-
cephalogr Clin Neurophysiol 45:443-453, 1978.
Ehrhart J and Muzet A. Frequence et duree des phases d'activation transitoire au cours du sommeil
normal ou perturbe chez l'homme. Arch Sci PhysioI28:213-260, 1974.
Ferriere PJR. Evolution des frequences respectives des fuseaux de sommeil et des reactions d'eveil
au cours du sommeil normal chez l'homme. These de Medecine, Universite de Strasbourg, 1970.
Muzet A, Naitoh P, Townsend RE, and Johnson LC. Body movements during sleep as a predictor of
stage change. Psychon Sci 29:7-10, 1972.
Muzet A, Schieber JP, Ehrhart J, and Lienhard JP. Les phases d'activation transitoire et les change-
ments de stades electroencephalographiques de sommeil. Rev Electroencephalogr Neurophysiol
Clin 3:219-222, 1973.
Muzet A, Naitoh P, Johnson LC, and Townsend RE. Body movements in sleep during 30-day expo-
sure to tone pulse. Psychophysiology 11:27-34, 1974.
Naitoh P, Muzet A, Johnson LC, and Moses J. Body movements during sleep after sleep loss.
Psychophysiology 10:363-368, 1973.
Rechtschaffen A and Kales A (Eds). A Manual of Standardized Terminology, Techniques and Scoring
System for Sleep Stages of Human Subjects. Brain Information Service/Brain Research Institute,
University of California at Los Angeles, 1968.
Sassin JF and Johnson LC. Body motility during sleep and its relation to the K-complex. Exp Neurol
22:133-144,1968.
Schieber JP, Muzet A, and Ferriere PJR. Les phases d'activation transitoire spontanees au cours du
sommeil normal chez l'homme. Arch Sci Physiol 25:443-465, 1971.
Yamadori A. Role of the spindles in the onset of sleep. Kobe J Med Sci 17:97-11 I, 1971.

Resume: Six sujets ont passe trois nuits consecutives au laboratoire. Les phases d'activa-
tion transitoire (PAT), les complexes K spontanes et les fuseaux de sommeil ont He
detectes visuellement au cours des stades 2 et 3 du sommeil pour les nuits 2 et 3. La
frequence des complexes K est significativement plus elevee dans les dix secondes preced-
ant les PAT que dans toute autre periode pas see en stade 2 ou 3. Les complexes K associes
a des fuseaux de sommeil surviennent avec une frequence significativement moins elevee
dans les periodes precedant les PAT. Pour tous les sujets, on observe une importante
augmentation du nombre de fuseaux associes avec les complexes K lorsque ces derniers ne
sont pas suivis d'une PAT dans un delai de dix secondes. Ces resultats suggerent que les
complexes K spontanes et les fuseaux de sommeil agissent de facon antagoniste sur la
survenue des PAT. L'interaction de ces deux evenements phasiques regule la probabilite
d'apparition des PAT dans les stades 2 et 3 du sommeil; les complexes K seraient I'expres-
sion d'un etat de l' organisme aboutissant a une PAT, tandis que les fuseaux de sommeil
traduiraient une inhibition de la survenue des PAT. Mots-Cles: Sommeil humain-
Complexes K-Fuseaux de sommeil-Phases d'activation.

Sleep, Vol. 4, No.4, 1981