SEARCH

SEARCH BY CITATION

Keywords:

  • autonomic function;
  • cyclic alternating pattern;
  • heart rate variability;
  • sleep;
  • spectral analysis

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

The natural arousal rhythm of non-rapid eye movement (NREM) sleep is known as the cyclic alternating pattern (CAP), which consists of arousal-related phasic events (Phase A) that periodically interrupt the tonic theta/delta activities of NREM sleep (Phase B). The complementary condition, i.e. non-CAP (NCAP), consists of a rhythmic electroencephalogram background with few, randomly distributed arousal-related phasic events. Recently, some relation between CAP and autonomic function has been preliminarily reported during sleep in young adults by means of spectral analysis of heart rate variability (HRV). The present study was aimed at analysing the effects of CAP on HRV in a group of normal children and adolescents. Six normal children and adolescents (age range 10.0–17.5 y) were included in this study. All-night polygraphic recordings were performed after adaptation to the sleep laboratory. Six 5-min epochs were selected from sleep Stage 2 and six from Stages 3 and 4 (slow-wave sleep), both in CAP and NCAP conditions. From such epochs, a series of parameters describing HRV was then calculated, in both time and frequency domains, on the electrocardiographic R–R intervals. Statistical comparison between CAP and NCAP epochs revealed a significant difference for most of the frequency domain parameters (increase of the low-frequency band, increase of the low-frequency/high-frequency ratio and decrease in the high-frequency band during CAP) both in Stage 2 and in slow-wave sleep. Our results demonstrate that the physiological fluctuations of arousal during sleep described as CAP are accompanied by subtle, but significant, changes in balance between the sympathetic and vagal components of the autonomic system.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

The natural electroencephalographic (EEG) arousal rhythm of non-rapid eye movement (NREM) sleep is known as the cyclic alternating pattern (CAP) ( Terzano et al. 1985 , 1988). CAP consists of arousal-related phasic events (Phase A) that interrupt, at intervals of 20–40 s, the tonic theta/delta activities of NREM sleep (Phase B). Functionally, CAP translates a condition of sustained arousal instability while the complementary EEG pattern, i.e. non-CAP (NCAP), characterized by a rhythmic background activity with few, randomly distributed arousal-related phasic events, reflects a condition of stable arousal.

The CAP sequences, formed by the regular succession of Phases A and B, constitute the microstructural component of sleep that accompanies the dynamic shifts between the NREM stages and seems to play a role in the transition from NREM to REM sleep ( Terzano et al. 1985 ). Instead, NCAP is the EEG expression of consolidated sleep within the NREM stages. The physiological balance between CAP and NCAP varies across the life span ( Parrino et al. 1998 ) and has been found to be altered in a number of sleep-disturbed conditions, such as insomnia ( Terzano and Parrino 1993), epilepsy ( Terzano et al. 1989 , 1992), parasomnias ( Zucconi et al. 1995 ) and mood disorders ( Parrino et al. 1994 ).

CAP and NCAP not only constitute the dynamic scaffolds for the microstructural organization of sleep, but also have repercussions for motor and vegetative activities, which fluctuate during CAP and remain quiescent during NCAP. Recently, some relation between CAP/NCAP and autonomic function has been preliminarily reported in young adults ( Ferini-Strambi et al. 1997 ) by means of spectral analysis of heart rate variability (HRV) during sleep. Heart rate is under the control of efferent sympathetic and vagal activities directed to the sinus node, which are modulated by central brainstem and peripheral oscillators ( Malliani et al. 1991 ). Spectral analysis of HRV is a quantitative, reliable method for analysing the modulatory effects of neural mechanisms on the sinus node ( [25]Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology 1996) and two main components are currently considered, high-frequency (HF) and low-frequency (LF). The vagal activity is the major contributor to the HF component, while the LF component is considered by some authors to be a marker of sympathetic modulation and by others to be a parameter including both vagal and sympathetic influences.

The aim of this study was to analyse the effects of the CAP and NCAP conditions on HRV during NREM sleep in a group of normal children and adolescents.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Subjects

Ten subjects (mean age 13.4 y, range 10.0–17.5 y) were admitted to this study. The protocol was approved by our ethical review committee and informed consent was obtained from the families of all subjects. All individuals were carefully evaluated from the neurological, otorrhino-laryngoiatric and cardiovascular points of view and failed to show significant abnormalities.

Recordings

All subjects slept in the laboratory for two consecutive nights. The data were analysed during the second night. To determine sleep stages, EEG (six channels), electro-oculogram (EOG) and mental electromyogram were recorded by means of an Oxford Medilog 9000-II recorder. Other physiological variables, such as electrocardiogram (CM4 derivation: anode in position V4 and cathode attached to the manubrium of the sternum), peripheral oxygen saturation, chest wall movement by thoracic impedance and oro-nasal airflow with thermistors, were recorded using an Oxford MPA-II recorder. All signals were sampled at a rate of 128 Hz and were also reproduced on paper by means of a Siemens Mingograf EEG 21 polygraph.

Sleep and HRV analysis

Sleep staging (macrostructure of sleep) was accomplished on paper recordings, following standard criteria ( Rechtschaffen and Kales 1968). In particular, sleep staging was carried out by visually analysing the EEG (C3-right earlobe derivation), EOG (left and right outer canthi referred to the left earlobe) and electromyogram (submentalis muscle). Body position was also controlled by means of a video-camera; during all epochs chosen for HRV analysis patients rested in a supine position.

In order to study sleep-stage-related HRV, a series of 5-min epochs was chosen from quiet wakefulness (W) and the following stages during the first two sleep cycles: sleep Stage 1 comprising the preceding 2 min of quiet wakefulness (W + S1), sleep Stage 2 (S2), sleep Stages 3 and/or 4 (slow-wave sleep, SWS) and REM sleep. This first-step selection was carried out regardless of the ongoing microstructural condition. Subsequently, CAP and NCAP sequences were detected in each recording, during S2 and SWS, according to the rules defined by Terzano et al. (1985 , 1988). Therefore, HRV was additionally studied on the recording night in three different 5-min epochs from CAP and in three different epochs from NCAP periods during both S2 and SWS. For each subject, 12 epochs were selected (three from S2-CAP, three from S2-NCAP, three from SWS-CAP and three from SWS-NCAP).

In order to avoid gross effects on HRV, only CAP sequences and NCAP periods without transient activation phases ( Schieber et al. 1971 ) or arousal ( American Sleep Disorders Association 1992) were selected. Moreover, because of the age range of our subjects, children and adolescents, low amounts of arousal were expected; in fact, the number of arousal events shows a linear increase with age ( Mathur and Douglas 1995; Boselli et al. 1998 ). Within the structure of sleep, most of the CAP sequences selected were identified during the transition from light to deep NREM sleep, in which the A phases are basically composed of K-complexes or delta bursts ( Ferrillo et al. 1997 ), which are accompanied by less evident heart rate changes. Finally, the eventual presence of apneas and hypopneas was also carefully controlled so that the epochs selected for analysis were free of respiratory events.

In each 5-min epoch, ECG signals were analysed for automatic detection of R waves using a self-made program utilizing a simple threshold plus first and second derivative algorithms; however, careful visual inspection for possible errors was performed on all epochs. In order to overcome the problem of the low sampling rate of our recorders (128 Hz), which might have caused a bias in the estimation of the R-wave fiducial point ( [25]Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology 1996) and a consequent alteration of the spectrum, a parabolic interpolation was used to refine its evaluation ( Merri et al. 1990 ; Bianchi et al. 1993 ). A series of time domain measures was calculated: mean R–R-value, standard deviation of all R–R intervals (SDNN), the square root of the mean of the sum of the squares of differences between adjacent R–R intervals (RMSSD), number of pairs of adjacent R–R intervals differing by more than 50 ms in the entire epoch (NN50) and percentage of NN50 among the total R–R intervals (pNN50). The first 256 R–R intervals from each epoch were utilized for all subsequent analysis steps. The R–R interval tachograms were processed by means of a FFT algorithm and the following spectral parameters were obtained: power in very low-frequency range, < 0.04 Hz (VLF), power in low-frequency range, 0.04–0.15 Hz (LF), power in high-frequency range, 0.15–0.4 Hz (HF), total power (VLF + LF + HF), LF power in normalized units: LF/(total power-VLF) × 100 (LF%), HF power in normalized units: HF/(total power-VLF) × 100 (HF%), ratio of LF to HF (LF/HF), frequency of highest peak in the VLF range (VLF peak), frequency of highest peak in the LF range (LF peak) and frequency of highest peak in the HF range (HF peak).

Statistical analysis

The comparison between HRV parameters obtained from CAP and NCAP epochs, during each sleep stage considered in this study, was performed by means of the Wilcoxon test for paired data sets.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

On average, the subjects included in this study slept for 8 h, their sleep efficiency index (total sleep time/time in bed) was good (0.892) and the number of awakenings/h was very low (0.770). A normal sleep macrostructure was observed in all cases. The mean length of the CAP cycle during CAP sequences was 25.2 s (Phase A 8.4 s and Phase B 16.8 s).

Figure 1 illustrates the stage-related changes in R–R interval and HRV. R–R interval increased (therefore, heart rate decreased) throughout sleep without relevant differences between the sleep stages. LF% and LF/HF showed a common trend towards a decrease with increasing sleep depth, from wakefulness to SWS; during REM sleep, LF% and LF/HF values approximated those of wakefulness. HF% underwent complementary variations in wakefulness and in the different sleep stages.

image

Figure 1. Figure 1.  Sleep stage associated changes in R–R interval and HRV spectral components.

Download figure to PowerPoint

The results of the comparison between HRV during CAP and NCAP epochs in S2 and SWS, in the absence of sleep apnea or arousal, are shown in Tables 1 and 2.

Table 1.  Comparison between HR findings during sleep Stage 2 CAP and NCAP epochs Thumbnail image of
Table 2.  Comparison between HR findings during slow-wave sleep CAP and NCAP epochs Thumbnail image of

The mean R–R-values were very similar in both CAP and NCAP conditions, with a limited interindividual variability. During both S2 (Table 1) and SWS (Table 2) a significant difference between CAP and NCAP conditions emerged for LF% and LF/HF ratio (higher in CAP than in NCAP) and for HF% (lower in CAP than in NCAP); in SWS also VLF showed values significantly higher during CAP than during NCAP states.

None of the time-domain parameters describing HRV considered in this study (SDNN, RMSSD, NN50, pNN50) showed any significant difference between CAP and NCAP epochs.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

In the second century, Galen observed that sleep is accompanied by a slowing in heart rate ( Furley and Wilkie 1984); however, it was necessary to wait for almost 2000 years in order to achieve further insight into the sleep-related changes of heart rate with the report by MacWilliam (1923) of increases in heart rate and blood pressure with dreaming. Entering the EEG and ECG era, sleep-stage-related changes in heart rate started to be described in detail ( Baust and Bonhert 1969; Malpas and Purdie 1990; Cajochen et al. 1994 ; Pivik et al. 1996 ). More recently, spectral analysis of HRV has been used to study fluctuations in the activity of the autonomic nervous system during sleep. Using this technique, it has been reported that LF shows a decrease during sleep, reaching minimal values during SWS; in contrast, REM sleep is accompanied by elevated LF values, similar to those of wakefulness, both in children ( Baharav et al. 1995 ) and young adults ( Vaughn et al. 1995 ). In children, the opposite is true for HF, which increases with sleep onset and reaches its highest values during SWS ( Baharav et al. 1995 ); in young adults HF has been reported to be maximal during sleep Stage 2 ( Vaughn et al. 1995 ). The ratio between these two components of HRV (LF/HF) shows changes similar to those of LF ( Baharav et al. 1995 ).

Thus, spectral analysis of HRV has been considered to be a reliable noninvasive method of quantifying changes in the sustained tonic autonomic influences on the heart during sleep. The results obtained from the present analysis of sleep-stage influences on heart rate show good agreement with the findings of previous reports ( Baharav et al. 1995 ; Vaughn et al. 1995 ). Moreover, it must be emphasized that time-domain parameters were unable to disclose significant CAP-related changes in HRV, being less sensitive to subtle modifications of the autonomic control of heart rate during sleep.

It must be noted that the age of the subjects included in this study ranged over a period of life in which significant changes in sleep structure are to be expected. In particular, teenagers show higher percentages of SWS and lower amounts of arousal than older controls ( Boselli et al. 1998 ; Parrino et al. 1998 ). However, in this study we did not aim to evaluate developmental aspects of HRV.

In addition to the correlation between depth of sleep and variations of vegetative activity, there is also consolidated evidence of transient changes in autonomic activity in association with spontaneous arousal-related phasic events. Among such events, the most powerful effect on sympathetic activity is exerted by arousal periods, which are actually accompanied by robust increases in muscle sympathetic activity, heart rate and blood pressure ( Guilleminault and Stoohs 1995). However, a less prominent, but still significant, acceleration of heart rate and rises in muscle sympathetic activity and blood pressure are observed during NREM sleep in the seconds that follow a K-complex ( Hornyak et al. 1991 ) or a delta burst ( Church et al. 1978 ). In other words, all arousal-related phasic events occurring during NREM sleep are associated with short-lasting increases in sympathetic activity and constitute a sort of continuum from weaker (K-complexes and delta bursts) to powerful (arousal) EEG features. In order to cast light exclusively on the impact of the weaker events, in the present study, arousal periods were excluded from the correlation between microstructural conditions and HRV.

In a preliminary report, Ferini-Strambi et al. (1997 ) have already shown that in young adults CAP and NCAP are accompanied by significant changes in HRV parameters. However, these authors made no distinction between the different types of A phases included in their study and probably also analysed epochs containing arousal or transient activation phases. Our results indicate that such an effect is also evident in children and adolescents, and that it reaches statistical significance even when the microanalysis of sleep is cleansed of arousal. In particular, the increase in LF and LF/HF ratio during S2–CAP and SWS–CAP (at the expense of a complementary HF reduction) implies that during CAP the sympatho-vagal balance is shifted towards sympathetic prevalence. Even if CAP-associated changes are of small amplitude, they might be able to influence the results of studies on HRV during sleep in which they are not taken into account due to the omitted scoring of CAP. No statistical comparison was possible between epochs in which CAP was ignored and only macrostructural analysis was accomplished, and those in which CAP was evaluated. However, if we roughly compare the HRV values expressed during the NREM sleep stages (Fig. 1) and during the CAP/NCAP conditions (Tables 1 and 2, it can be seen that the CAP-related increase in sympathetic activity is always higher than the mean value expressed by the corresponding sleep stage as a whole. It is also interesting to note that significant autonomic differences between CAP and NCAP were found both in Stage 2, which generally occupies ≈ 50% of the total sleep time in children and adolescents, and in SWS, which generally occupies ≈25% of the total sleep time. In other words, we are dealing with a microstructural phenomenon that may have extensive HRV repercussions throughout sleep. Therefore, the presence or absence of CAP should be verified, whenever possible, when studying HRV during sleep.

As a constituent manifestation of NREM sleep, CAP occurs even in the absence of sleep-perturbing factors (e.g. noise, pain, myoclonic jerks, respiratory events). Basically, internal and environmental disturbances vary the physiological amount of CAP, but have limited effects on the 20–40-s alternating rhythm of Phases A and B. Previous investigation has ascertained that this periodicity is involved in the modulation of a number of repetitive phenomena, such as periodic limb movements during sleep ( Parrino et al. 1996 ), sleep bruxism ( Macaluso et al. 1998 ) and sleep-related respiratory events ( Terzano et al. 1996 ). With respect to the present study, it can be suggested that the same cerebral mechanisms underlying the regulation of arousal instability might be involved in the genesis and enhancement of the sympathetic activity, since the two phenomena (CAP and LF) operate within a common 20–40-s scale.

In conclusion, our data are in tune with the idea that CAP modulates not only EEG activities, but also a series of other sleep-related phenomena including autonomic fluctuations ( Terzano and Parrino 1993; Terzano et al. 1996 ). Further studies are needed to clarify the vegetative impact of both spontaneous and evoked CAP sequences, including those containing arousal, and to investigate the role of CAP-related autonomic changes in different pathological conditions.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References
  • 1
    American Sleep Disorders Association. Arousals. Scoring rules and examples. A preliminary report from the sleep disorders atlas task force of the American Sleep Disorders Association. Sleep, 1992, 15: 174 184.
  • 2
    Baharav, A., Kotagal, S., Gibbons, V., Rubin, B. K., Pratt, G., Karin, J., Akselrod, S. Fluctuations in autonomic nervous activity during sleep displayed by power spectrum analysis of heart rate variability. Neurology, 1995, 45: 1183 1187.
  • 3
    Baust, W. & Bonhert, B. The regulation of heart rate during sleep. Exp. Brain Res., 1969, 7: 169 180.
  • 4
    Bianchi, A. M., Mainardi, L. T., Petrucci, E., Signorini, M. G., Mainardi, M., Cerutti, S. Time-variant power spectrum analysis for the detection of transient episodes in HRV signal. IEEE Trans. Biomed. Eng., 1993, 40: 136 144.
  • 5
    Boselli, M., Parrino, L., Smerieri, A., Terzano, M. G. Effects of age on EEG arousals in normal sleep. Sleep, 1998, 21: 351 357.
  • 6
    Cajochen, C., Pischke, J., Aeschbach, D., Borbély, A. A. Heart rate dynamics during human sleep. Physiol. Behav., 1994, 55: 769 774.
  • 7
    Church, M. W., Laverne, C. W., Seales, D. M. Evoked K-complexes and cardiovascular responses to spindle-synchronous and spindle-asynchronous stimulus clicks during NREM sleep. Electroenceph. Clin. Neurophysiol., 1978, 45: 443 453.
  • 8
    Ferini-Strambi, L., Mattioli, S., Bianchi, A., Zucconi, M., Oldani, A., Castronovo, C., Smirne, S. The impact of cyclic alternating pattern (CAP) on the spectral analysis of heart rate variability during sleep in normal subjects. Sleep Res., 1997, 26: 7474.
  • 9
    Ferrillo, F., Gabarra, M., Nobili, L., Parrino, L., Schiavi, G., Stubinski, B., Terzano, M. G. Comparison between visual scoring of cyclic alternating pattern (CAP) and computerized assessment of slow EEG oscillations in the transition from light to deep non-REM sleep. J. Clin. Neurophysiol., 1997, 14: 210 216.
  • 10
    Furley, D. & Wilkie, J. S. (Eds). Galen on Respiration and the Arteries. Princeton University Press, Princeton, NJ, 1984.
  • 11
    Guilleminault, C. & Stoohs, R. Arousal, increased respiratory efforts, blood pressure and obstructive sleep apnoea. J. Sleep. Res., 1995, 4 (Suppl. 1): 117 124.
  • 12
    Hornyak, M., Cejnar, M., Elam, M., Matousek, M., Wallin, G. Sympathetic muscle nerve activity during sleep in man. Brain, 1991, 114: 1281 1295.
  • 13
    Macaluso, G., Guerra, P., Di Giovanni, G., Boselli, M., Parrino, L., Terzano, M. G. Sleep bruxism is a disorder related to periodic arousal during sleep. J. Dental Res., 1998, 77: 565 573.
  • 14
    MacWilliam, J. A. Some applications of physiology to medicine. III. Blood pressure and heart action in sleep and dreams: their relation to haemorrhages, angina, and sudden death. Br. Med. J., 1923, 2: 1196 1200.
  • 15
    Malliani, A., Pagani, M., Lombardi, F., Cerutti, S. Cardiovascular neural regulation explored in the frequency domain. Circulation, 1991, 84: 1482 1492.
  • 16
    Malpas, S. C. & Purdie, G. L. Circadian variation of heart rate variability. Cardiovasc. Res., 1990, 24: 210 213.
  • 17
    Mathur, R. & Douglas, N. J. Frequency of EEG arousal from nocturnal sleep in normal subjects. Sleep, 1995, 18: 330 333.
  • 18
    Merri, M., Farden, D. C., Mottley, J. G., Titlebaum, E. L. Sampling frequency of the electrocardiogram for the spectral analysis of heart rate variability. IEEE Trans. Biomed. Eng., 1990, 37: 99 106.
  • 19
    Parrino, L., Boselli, M., Buccino, G. P., Spaggiari, M. C., Di Giovanni, G., Terzano, M. G. The cyclic alternating pattern plays a gate-control on periodic limb movements during non-rapid eye movement sleep. J. Clin. Neurophysiol., 1996, 13: 314 323.
  • 20
    Parrino, L., Boselli, M., Spaggiari, M. C., Smerieri, A., Terzano, M. G. Cyclic alternating pattern (CAP) in normal sleep: polysomnographic parameters in different age groups. Electroenceph. Clin. Neurophysiol., 1998, 107: 439 450.
  • 21
    Parrino, L., Spaggiari, M.C., Boselli, M., Di Giovanni, G., Terzano, M.G. Clinical and polysomnographic effects of trazodone CR in chronic insomnia associated with dysthymia. Psychopharmacology, 1994, 116: 389 395.
  • 22
    Pivik, R. T., Busby, K. A., Gill, E., Hunter, P., Nevins, R. Heart rate variations during sleep in preadolescents. Sleep, 1996, 19: 117 135.
  • 23
    Rechtschaffen, A. & Kales, A. (Eds). A Manual of Standardized Terminology, Techniques and Scoring System of Sleep Stages of Human Subjects. Washington Public Health Service. US Government Printing Office, Washington D.C., 1968.
  • 24
    Schieber, J. P., Muzet, A., Ferriere, P. J. R. Les phases d’activation transitoire spontanées au cours du sommeil chez l’homme. Arch. Sci. Physiol., 1971, 25: 443 465.
  • 25
    Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability. Standards of measurements, physiological interpretation and clinical use. Circulation, 1996, 93: 1043 1065.
  • 26
    Terzano, M. G., Mancia, D., Salati, M. R., Costani, G., Decembrino, A., Parrino, L. The cyclic alternating pattern as a physiologic component of normal NREM sleep. Sleep, 1985, 8: 137 145.
  • 27
    Terzano, M. G., Parrino, L., Anelli, S., Boselli, M., Clemens, B. Effects of generalized interictal EEG discharges on sleep stability: assessment by means of cyclic alternating pattern. Epilepsia, 1992, 33: 317 326.
  • 28
    Terzano, M. G., Parrino, L., Anelli, S., Halasz, P. Modulation of generalized spike-and-wave discharges during sleep by cyclic alternating pattern. Epilepsia, 1989, 30: 772 781.
  • 29
    Terzano, M. G., Parrino, L., Boselli, M., Spaggiari, M. C., Di Giovanni, G. Polysomnographic analysis of arousal responses in OSAS by means of the cyclic alternating pattern (CAP). J. Clin. Neurophysiol., 1996, 13: 145 155.
  • 30
    Terzano, M. G., Parrino, L., Spaggiari, M. C. The cyclic alternating pattern sequences in the dynamic organization of sleep. Electroenceph. Clin. Neurophysiol., 1988, 69: 437 447.
  • 31
    Terzano, M. G. & Parrino, L. Clinical application of cyclic alternating pattern. Physiol. Behav., 1993, 54: 807 813.
  • 32
    Terzano, M. G. & Parrino, L. Evaluation of EEG cyclic alternating pattern during sleep in insomniacs and controls under placebo and acute treatment with zolpidem. Sleep, 1993, 15: 64 70.
  • 33
    Vaughn, B. V., Quint, S. R., Messenheimer, J. A., Robertson, K. R. Heart period variability in sleep. Electroenceph. Clin. Neurophysiol., 1995, 94: 155 162.
  • 34
    Zucconi, M., Oldani, A., Ferini-Strambi, L., Smirne, S. Arosual fluctuations in non-rapid eye movement parasomnias: the role of cyclic alternating pattern as a measure of arousal instability. J. Clin. Neurophysiol., 1995, 12: 147 154.