Abstract
Aims
To define the change in power in standard waveband frequencies of quantitative cortical electroencephalogram (EEG) data over a 24 h period, in a drug free representative healthy volunteer population.
Methods
This was an open, non randomised study in which 18 volunteers (9 male and 9 female) were studied on 1 study day, over a 24 h period. Volunteers had a cortical EEG recording taken at 0, 2, 4, 6, 8, 10, 12, 16 and 24 h. Each recording lasted for 6 min (3 min eyes open, 3 min eyes closed). All EEG recordings were taken in a quietened ward environment with the curtains drawn round the bed and the volunteer supine. During the 3 min eyes open, volunteers were asked to look at a red circle on a screen at the foot of the bed, and refrain from talking.
Results
Plots produced of geometric mean power by time of the standard wave band frequencies gave some indication of a circadian rhythm over the 24 h period for θ (4.75–6.75 Hz), α1 (7.0–9.5 Hz) and β1 (12.75–18.50 Hz) wavebands. Mixed models were fitted to both the eyes open and eyes closed data which confirmed a change in mean waveband power with time with statistical significance at the conventional 5% level (P < 0.05).
Conclusions
These data indicate the presence of a diurnal variation in the cortical quantitative EEG. They support the use of a placebo control group when designing clinical trials which utilize quantitative EEG to screen for central nervous system (CNS) activity of pharmaceutical agents, to control for the confounding variable of time of day at which the EEG recordings were made.
Keywords: diurnal variation, quantitative EEG, volunteers
Introduction
Computerized quantitative electroencephalography (EEG) has become increasingly popular as a tool to demonstrate central nervous system (CNS) activity of new molecular entities of several classes of psychoactive drugs in early volunteer [1–4] and patient [5] studies – (for reviews see [6, 7]). In addition to screening for desirable CNS activity, quantitative EEG has also been used to demonstrate a lack of effect of drugs at therapeutic doses on cortical electrical activity as predictive of a lack of CNS side-effects [8]. Whilst guidelines on processing and interpretation of computer generated EEG data exist [9], the environment of data capture can differ considerably between published reports. Many such ‘environmental’ factors have been demonstrated to alter the quantitative EEG including the pharmacological agents caffeine [10, 11], nicotine and alcohol [12]. The subjects age [13, 14] and mental load [15–17] at the time of data capture as well as subjects being allowed to talk [18] during data acquisition, have also been shown to alter the subjects' quantitative EEG; as have the recent use of a mobile phone [19] and exposure to a static magnetic field [20]. Underlying diseases such as dementia [21, 22] and schizophrenia [23, 24] can also profoundly affect the EEG. Although published reports control for at least some of these factors, little attention has been paid to the possibility that a diurnal variation also might exist as a further confounding variable. There are reports of such an effect in animals [25] but clinical studies addressing this are sparse, with the majority of 24 h EEG studies being primarily focused on the well documented effects of sleep and sleep deprivation on the quantitative EEG [26–28] or reporting continuous EEG recordings in disease states such as epilepsy [29], narcolepsy [30], depression [31] or sleeping sickness [32]. Previous attempts to characterize a diurnal variation in EEG in healthy volunteers have been limited to the assessment of a single spectral frequency [33, 34] or recording eyes closed data only [35, 36] or using a limited number of subjects and sampling points [37].
The purpose of this study was to define the change in power in the standard frequency bands of the computerized quantitative EEG over a 24 h period in 18 healthy volunteers under standardized conditions, for eyes open and eyes closed data. This would provide a baseline against which quantitative EEG data from future drug studies can be interpreted by an investigator blinded to the treatment. In addition, knowing the magnitude of change which occurs in each waveband frequency over a 24 h period, will better guide the determination of a ‘clinically meaningful’ change in waveband frequency when seeking a real drug effect.
Methods
Ethical considerations
The study protocol was approved by Zeneca Pharmaceuticals independent research ethics committee. All volunteers gave their written informed consent following a verbal explanation of the study and after reading a detailed information sheet.
Design
This study was open and nonrandomised. It did not involve the administration of any drug. Volunteers were studied on 1 study day when they had nine quantitative EEGs recorded at 0, 2, 4, 6, 8, 10, 12, 16 and 24 h. Each record lasted 6 min and comprised of 3 min eyes open and 3 min eyes closed.
Subjects
The study was conducted in 18 day-time shift worker healthy volunteers (9 male and 9 female) who were all Caucasian, aged 20–47 years (mean 32.6 years), of weight 52–86 kg (mean 70.4 kg) and height 157–183 cm (mean 170.2 cm). Each subject prior to entry into the study had completed an acceptable health screening questionnaire. In addition all female volunteers had a negative pregnancy test at the time of their study day. For inclusion into the study all volunteers had to attend a prestudy EEG recording unless one had been performed in the previous 24 months; the purpose of this was to familiarize the volunteers with the procedure and to exclude any potential volunteers with abnormalities suggesting underlying disease on their quantitative EEG trace. At this session an impedance test was performed to ensure good electrical contact between scalp and electrode. Any volunteers with multiple electrodes exhibiting > 50 kω resistance were excluded. Subjects had not participated in drug studies within 3 months of the start of the present study. Volunteers were requested to stop smoking (2 females smoked 3 or 10 cigarettes/day; 2 male volunteers each smoked 2 cigarettes/day and 1 male volunteer smoked 3 cigars/day at the time of study enrolment) and taking caffeine containing drinks or food and alcohol (weekly alcohol intake: Male: Mean 13.3 units, range 6–20 units; Female: Mean 7.1 units, range 1–10 at time of study enrolment) for at least 12 h before the study day and female volunteers were required to have a negative pregnancy test on the study day and were not to be breast feeding during the course of the study. Volunteers reported to the Clinical Pharmacology Unit (CPU) at Zeneca Pharmaceuticals after an overnight fast from midnight of the day before. All subjects indicated compliance with these requests.
Quantitative electroencephalography
On the study day each volunteer's quantitative EEG using the CATEEM™[38] apparatus (Proscience Private Research Institute, Linden, Germany) was recorded nine times at 0, 2, 4, 6, 8, 10, 12, 16 and 24 h for 6 min (3 min eyes open, 3 min eyes closed). Time 0 corresponds to 08 00 h. All recordings took place in a quietened ward environment with the curtains drawn around the bed and the volunteer supine. During the 3 min eyes open, volunteers were asked to look at a red circle on a screen at the foot of their bed and not to talk.
On the morning of each study day the volunteers attended the CPU having been asked to wash their hair, and were fitted with an electrode cap approximately 30 min before the first EEG recording. The electrode cap consisted of a shaped cloth cap on which 17 hollow metal electrodes are positioned according to the international 10 : 20 system [39]. These electrodes were back filled with electrode jelly prior to placing the cap on the head. The cap was positioned so that the front and back were equidistant from the nose and the ears. Electrical contact with the scalp was made by top filling the electrodes with electrode jelly until the volunteer felt the jelly on the scalp. As part of the artefact rejection procedure, an electro-oculogram electrode was placed on the temple, and ECG electrodes were strapped to the wrists. EEG artefacts were rejected automatically by the CATEEM™ programme. Only those recordings which were greater than 30% artefact free were used in subsequent analysis. Micro-voltage potentials recorded from each of the 17 scalp electrodes were displayed continuously on a video monitor. The analogue signals were processed by the CATEEM™ computer to provide power spectra [9] for each recording session at each of the frequency bands listed in Table 1.
Table 1.
Frequency bands for power analysis of the quantitative EEG (Hz).
| Band | Frequency (Hz) |
|---|---|
| δ | 1.25–4.50 |
| θ | 4.75–6.75 |
| α1 | 7.0–9.50 |
| α2 | 9.75–12.50 |
| β1 | 12.75–18.50 |
| β2 | 18.75–35.00 |
The amount of activity in each frequency band was taken to be the power in that frequency band, summed across the 17 electrodes. The power in each frequency band was calculated for eyes open and eyes closed data separately.
Adverse event monitoring
Any symptoms the volunteers experienced were recorded by the volunteers at 30 min before dosing and at 0, 2, 4, 6, 8, 10, 12, 16 and 24 h on the study day. The investigator subsequently coded the maximum severity, time of onset and duration, seriousness and whether treatment was required, together with an assessment of causality to the procedure.
Endpoints and methods of analysis
In the analysis of the EEG power spectra the total amount of electrical activity was assessed in each of the six frequency bands δ, θ, α1, α2, β1, β2 and summed across the 17 electrodes. No topographical analyses were performed. The EEG data were log transformed and plots of the geometric mean by time were presented. Following inspection of the plots of geometric mean by time, the frequency bands of θ, α1 and β1 gave some indication of a circadian rhythm over the 24 h period. To assess further whether the time of day at which these three frequency bands were measured had a statistically significant effect on the values observed, mixed models were fitted. A model including terms up to quartic time was found to adequately fit the data for the θ band (eyes closed). A model including terms up to cubic time was found to adequately fit the data for the θ band (eyes closed), and the β1 band (eyes open and closed) and the α1 band (eyes closed). A model including terms up to quadratic time was found to adequately fit the data for the α1 band (eyes open). In all the frequency bands analysed, assumptions of normality and constancy of variance were met.
Results
The actual frequency waveband power, with standard deviation with coefficient of variation expressed as a percentage of the geometric mean (CV(%)) at time 0 are presented in Table 2. For each frequency waveband power, the geometric mean (gmean) at time 0, its standard deviation and coefficient of variation expressed as a percentage of the gmean (CV(%)) are presented in Table 2.
Table 2.
Quantitative EEG power by frequency band at 08.00 h (Time 0).
| Frequency band | Geometric mean power (µV2) | Geometric mean power ± 1 s.d. (µV2) | CV (%) | Minimum (µV2) | Maximum (µV2) |
|---|---|---|---|---|---|
| δ Eyes open | 134.14 | 90.46, 198.90 | 40.97 | 66.73 | 290.99 |
| δ Eyes closed | 148.74 | 110.35, 200.48 | 30.53 | 83.73 | 234.55 |
| θ Eyes open | 34.35 | 23.80, 49.57 | 37.95 | 18.49 | 57.76 |
| θ Eyes closed | 49.51 | 32.17, 76.20 | 45.20 | 19.87 | 103.27 |
| α1 Eyes open | 60.92 | 30.20, 122.88 | 79.76 | 23.41 | 214.21 |
| α1 Eyes closed | 169.42 | 86.12, 333.30 | 76.20 | 46.52 | 684.70 |
| α2 Eyes open | 52.77 | 23.13, 120.40 | 98.72 | 13.30 | 259.64 |
| α2 Eyes closed | 151.29 | 63.87, 358.37 | 105.05 | 37.56 | 741.78 |
| β1 Eyes open | 34.90 | 22.92, 53.16 | 44.01 | 19.27 | 89.31 |
| β1 Eyes closed | 44.62 | 28.92, 68.84 | 45.49 | 21.69 | 148.86 |
| β2 Eyes open | 51.61 | 29.37, 90.67 | 61.15 | 20.32 | 152.33 |
| β2 Eyes closed | 50.44 | 32.40, 78.53 | 46.52 | 24.17 | 159.72 |
The proportional change in waveband power with time is presented in Figures 1 and 2.
Figure 1.
Proportional change in power of each frequency band of the quantitative EEG over 24 h (Eyes closed data). ▴ θ, • β1, ▪ α1, ○ β2, ♦ δ, □ α2.
Figure 2.
Proportional change in power of each frequency band of the quantitative EEG over 24 h (Eyes open data). ▴ θ, • β1, ▪ α1, ○ β2, ♦ δ, □ α2.
Following inspection of the plots of proportional change over time for all volunteers (Figures 1 and 2) the frequency bands of θ, α1 and β1 (bold lines) gave an indication of a diurnal variation for both eyes open and eyes closed data.
The θ, α1 and β1 frequency bands were analysed using a mixed model, which allowed for the fact that repeated measurements had been taken on each volunteer to be taken into account. Each model was fitted using PROC MIXED in SAS. Each frequency band was log transformed prior to analysis as previous experience had shown them to be log-normally distributed. In all of the frequency bands analysed, assumptions of normality and constancy of variance were met. The results of the analyses to investigate the statistical significance of time are presented in Table 3
Table 3.
Assessment of the effect of the time of day on EEG frequency bands.
| Band | Eyes open/closed | Type of relationship with time | P value for relationship with time |
|---|---|---|---|
| θ | Eyes open | Cubic | 0.001 |
| Eyes closed | Quartic | 0.046 | |
| β1 | Eyes open | Cubic | < 0.001 |
| Eyes closed | Cubic | 0.007 | |
| α1 | Eyes open | Cubic | < 0.001 |
| Eyes closed | Quadratic | 0.031 |
For the three frequency wavebands θ, α1 and β1, there was statistical evidence to support the clinical review of the data that these three wavebands significant changed their mean power over a 24-h period.
Sex effects
With the exception of the α1 waveband, the nine female volunteers had larger geometric mean power values than the nine male volunteers for all frequency bands (Table 4). This difference was not investigated statistically, as the differences were typically small and lay within 1 standard deviation of the other genders mean (other than for α2) and were considered to be clinically insignificant.
Table 4.
Quantitative EEG power by frequency band and by gender at 08.00 h (Time 0).
| Frequency band | Geometric mean power (µV2) | Geometric mean power ± 1 s.d. (µV2) | CV (%) | Minimum (µV2) | Maximum (µV2) | |
|---|---|---|---|---|---|---|
| δ Eyes open | male | 128.37 | 90.33, 182.43 | 36.26 | 70.30 | 203.48 |
| female | 140.16 | 89.43, 219.67 | 47.3 | 66.73 | 290.99 | |
| δ Eyes closed | male | 148.37 | 111.29, 197.81 | 29.36 | 83.73 | 192.25 |
| female | 149.10 | 107.56, 206.7 | 33.55 | 85.33 | 234.55 | |
| θ Eyes open | male | 33.61 | 22.43, 50.36 | 42.15 | 18.49 | 57.76 |
| female | 35.11 | 24.78, 49.73 | 35.91 | 20.83 | 55.51 | |
| θ Eyes closed | male | 46.35 | 26.73, 80.36 | 59.47 | 19.87 | 103.27 |
| female | 52.88 | 39.69, 70.47 | 29.3 | 36.17 | 102.27 | |
| α1 Eyes open | male | 66.57 | 30.47, 145.43 | 91.74 | 23.41 | 195.17 |
| female | 55.74 | 29.2, 106.41 | 72.04 | 33.06 | 214.21 | |
| α1 Eyes closed | male | 172.94 | 94.77, 315.59 | 66.03 | 55.24 | 310.03 |
| female | 165.97 | 75.99, 362.48 | 91.7 | 46.52 | 684.7 | |
| α2 Eyes open | male | 37.96 | 14.36, 100.31 | 125.34 | 13.30 | 259.64 |
| female | 73.36 | 44.17, 121.84 | 54.18 | 38.71 | 169.97 | |
| α2 Eyes closed | male | 103.25 | 41.72, 255.53 | 112.84 | 37.56 | 365.68 |
| female | 221.70 | 115.02, 427.29 | 73.35 | 96.05 | 741.78 | |
| β1 Eyes open | male | 32.14 | 20.04, 51.53 | 49.98 | 19.71 | 89.31 |
| female | 37.91 | 26.15, 54.96 | 38.46 | 19.27 | 81.56 | |
| β1 Eyes closed | male | 37.7 | 26.09, 54.48 | 38.09 | 21.69 | 75.45 |
| female | 52.80 | 33.75, 82.6 | 47.1 | 28.13 | 148.86 | |
| β2 Eyes open | male | 42.96 | 26.61, 69.37 | 50.80 | 20.32 | 109.78 |
| female | 61.99 | 33.75, 113.88 | 66.89 | 22.89 | 152.33 | |
| β2 Eyes closed | male | 41.17 | 29.65, 57.15 | 33.71 | 24.17 | 82.57 |
| female | 61.81 | 38.85, 98.35 | 49.06 | 30.21 | 159.72 |
Adverse events
Five adverse events were reported by four volunteers. Four of the events were headaches which were all mild in intensity, commencing 6.5–8 h after time 0, with a mean duration of 4 h. One volunteer reported an event of vomiting which was mild in intensity and occurred 11 h and 50 min after the first EEG recording and lasted 10 min. All adverse events recovered and were considered by the investigator as not atypical of the adverse event profile expected of incarcerated, dietry restricted volunteers. These subjects were not eliminated from the qEEG waveband analyses, since a primary objective of this study was provide a drug-free qEEG profile against which quantitative EEG data from future drug studies conducted with similar environmental restrictions can be interpreted by an investigator blinded to the treatment
Discussion
This study has demonstrated that for three of the frequency bands (θ, α1and β1) there is evidence of a time effect which was statistically significant. The diurnal variation in theta power being greatest around the middle of the day confirms an earlier report [33]. Therefore, the time of day at which the frequency bands were measured had a statistically significant effect for these frequency bands for both eyes open and eyes closed. These data provide firm support for the use of a placebo group in any clinical trial utilizing quantitative EEG to assess whether a pharmacological agent has any CNS activity-the comparison being with placebo rather than a predose value. In such volunteer studies drugs are typically administered during the morning, with subsequent quantitative EEG pharmacodynamic recordings performed over the ensuing 1–4 h to best capture the time of maximal concentration of the drug. This is precisely the part of the day over which the change in the middle frequency wavebands (θ to β1) is greatest. This study did not assess whether night-time shift workers who have a reverse sleep-wake cycle to the day-time shift workers who were the subjects in this study, have a similar spectral frequency power profile to that reported here, but phase shifted by time and whether this should be an additional selection criteria in future drug trials.
Furthermore, these 24 h normal ranges for quantitative EEG change in healthy male and female volunteers clearly demonstrate that changes of + 20% in group mean data are within biological circadian variation. Searching or finding drug effect on quantitative EEG frequency waveband power of less than 20% would appear to be of questionable biological relevance. This is also supported by the sampling variability of quantitative EEG data between individuals exemplified by the time 0 data (Table 1) showing a coefficient of variation typically in excess of 40%.
Particularly for the β2 frequency band, the proportional change at 24 h lay some 20% different from that at time 0, suggesting that the ‘diurnal’ variation in this waveband may alter over longer than a 24 h period. Clearly, control data measured over a longer sampling period would determine whether this was a real phenomena.
In conclusion, the data presented in this study characterize normal ranges for the change in quantitative EEG waveband power over a 24 h period. In addition to standardizing subjects with respect to caffeine, nicotine and alcohol, these data suggest that attention should also be given to the time of day when computerized quantitative EEG recordings are made, strongly supporting the need for placebo-controlled trials when using quantitative EEG as a pharmacodynamic tool in future trials.
References
- 1.Herrmann WM. Electroencephalography in drug research. Stuttgart: Fischer; 1982. [Google Scholar]
- 2.Saletu R, Grunberger J, Anderer P, Linzmayer L, Semlitsch HV, Magni G. Pharmacodynamics of venlafaxine evaluated by EEG brain mapping, psychometry and psychophysiology. Br J Clin Pharmacol. 1992;33:589–601. doi: 10.1111/j.1365-2125.1992.tb04087.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hughes AM, Lynch P, Rhodes J, Ervine CM, Yates RA. Electroencephalographic and psychomotor effects of chlorpromazine and risperidone relative to placebo in normal healthy volunteers. Br J Clin Pharmacol. 1999. in press. [DOI] [PMC free article] [PubMed]
- 4.Barbanoj MJ, Anderer P, Antonijoan RM, Torrent J, Saletu B, Jane F. Topographic pharmaco-EEG mapping of increasing doses of buspirone and its comparison with diazepam. Human Psychopharmacol. 1994;9:101–109. [Google Scholar]
- 5.Schellenberg R, Milch W, Schwarz A, Schober F, Dimpfel W. Quantitative EEG and BPRS data following Haldol-decanoate administration in schizophrenics. Int Clin Psychopharmacol. 1994;9:17–24. doi: 10.1097/00004850-199400910-00003. [DOI] [PubMed] [Google Scholar]
- 6.Saletu B. The use of pharmaco-EEG in drug profiling. In: Hindmarch I, Stonier PD, editors. Human Psychopharmacology Measures and Methods. Vol. 1. Chichester, New York: John Wiley & Sons Ltd; 1987. pp. 173–200. [Google Scholar]
- 7.Itil TM. Psychotrophic drugs and the human EEG. In: Ban TA, Freyhan FA, Pichot P, Poldinger W, editors. Modern Problems of Pharmacopsychiatry. Vol. 8. Basel: S Karger; 1974. [Google Scholar]
- 8.Hughes AM, Dixon R, Dane A, Kemp J, Cummings L, Yates RA. Effects of zolmitriptan (Zomig TM) on central serotonergic neurotransmission as assessed by active oddball auditory event-related potentials in volunteers without migraine. Cephalagia. 1999;19:100–106. doi: 10.1046/j.1468-2982.1999.019002100.x. [DOI] [PubMed] [Google Scholar]
- 9.Nuwer MR, Lehmann D, Lopes da Silva F, Matsuoka S, Sutherling W, Vibert JF. IFCN guidelines for topographic and frequency analysis of EEG's and EP's. Report of an IFCN committee. Electroencephalogr Clin Neurophysiol. 1994;91:1–5. doi: 10.1016/0013-4694(94)90011-6. [DOI] [PubMed] [Google Scholar]
- 10.Dimpfel W, Schoeber F, Spuler M. The influence of caffeine on human EEG under resting conditions. Clin Invest. 1993;71:197–207. doi: 10.1007/BF00180102. [DOI] [PubMed] [Google Scholar]
- 11.Etevenon P, Peron- Magnan P, et al. EEG cartography profile of caffeine in normals. Clin Neuropharmacol. 1986;9:538–540. [PubMed] [Google Scholar]
- 12.Murphree HB. Electroencephalographic effects of caffeine, nicotine, tobacco smoking and alcohol. In: Itil TM, editor. Psychotrophic Drugs and the Human EEG: Modern Problems in Pharmacopsychiatry. Vol. 8. Basel: Karger; 1974. pp. 22–36. [Google Scholar]
- 13.Duffy FH, Albert MS, McAnulty G, Garvey AJ. Age related differences in brain electrical activity of healthy subjects. Ann Neurol. 1984;16:430–438. doi: 10.1002/ana.410160403. [DOI] [PubMed] [Google Scholar]
- 14.Dustman RE, Shearer DE, Emmerson RY. EEG and event-related potentials in normal ageing. Prog Neurobiol. 1993;41:369–401. doi: 10.1016/0301-0082(93)90005-d. [DOI] [PubMed] [Google Scholar]
- 15.Schober F, Dimpfel W. Relation between psychometric tests and quantitative topographic EEG in pharmacology. Int J Clin Pharmacol Ther Toxicol. 1992;30:428–430. [PubMed] [Google Scholar]
- 16.Schoeber F, Schellenberg R, Dimpfel W. Reflection of mental exercise in the dynamic quantitative topographical EEG. Neuropsychobiology. 1995;31:98–112. doi: 10.1159/000119179. [DOI] [PubMed] [Google Scholar]
- 17.Grunau RV, Low MD. Cognitive and task-related EEG correlates of arithmetic performance in adolescents. J Clin Exp Neuropsychol. 1987;9:563–574. doi: 10.1080/01688638708410769. [DOI] [PubMed] [Google Scholar]
- 18.Rugg MD, Dickens AM. Dissociation of alpha and theta activity as a function of verbal and visuo-spacial tasks. Electroencephalogr Clin Neurophysiol. 1982;53:210–217. doi: 10.1016/0013-4694(82)90024-4. [DOI] [PubMed] [Google Scholar]
- 19.Reiser HP, Dimpfel W, Schober F. The influence of electromagnetic fields on human brain activity. Eur J Med Res. 1995;1:27–32. [PubMed] [Google Scholar]
- 20.Klitzing L. Static magnetic fields increase the power intensity of EEG of man. Brain Res. 1989;483:201–203. doi: 10.1016/0006-8993(89)90056-5. [DOI] [PubMed] [Google Scholar]
- 21.Schellenberg R, Todorova A, Dimpfel W, Schoeber F. Pathophysiology and psychopharmacology of dementia—a new study design. Neuropsychobiology. 1995;32:81–97. doi: 10.1159/000119219. [DOI] [PubMed] [Google Scholar]
- 22.Leuchter AF, Cook IA, Newton TF, et al. Regional differences in brain electrical activity in dementia: Use of spectral power and spectral ratio measures. Electroencephalogr Clin Neurophysiol. 1993;87:385–393. doi: 10.1016/0013-4694(93)90152-l. [DOI] [PubMed] [Google Scholar]
- 23.Lacroix D, Rappelsberger P, Rescher B, Steinberger K, Thau K, Petsche H. Amplitude and coherence mapping of cognitive processing in healthy and schizophrenic males. In: Rother M, Zweiner U, editors. Quantitative EEG Analysis: Clinical Utility and New Methods. Jena: Universitatsverlag; 1993. pp. 187–190. [Google Scholar]
- 24.Itil TM. Qualitative and quantitative EEG findings in schizophrenics. Schizophrenia Bull. 1977;3:61–79. doi: 10.1093/schbul/3.1.61. [DOI] [PubMed] [Google Scholar]
- 25.Grasing K, Szeto H. Diurnal varaiation in continuous measures of the rat EEG power spectra. Physiol Behav. 1992;51:249–254. doi: 10.1016/0031-9384(92)90138-r. [DOI] [PubMed] [Google Scholar]
- 26.Johnson LC. The EEG during sleep as viewed by a computer. In: Redmond A, editor. EEG Informatics. A Didactic Review of Methods and Applications of EEG Data Processing. Amsterdam: Elsevier; 1977. [Google Scholar]
- 27.Corsi-Cabrera M, Ramos J, Arce C, Guevara MA, de Ponce-Leon M, Lorenzo I. Changes in the waking EEG as a consequence of sleep and sleep deprivation. Sleep. 1992;15:550–555. doi: 10.1093/sleep/15.6.550. [DOI] [PubMed] [Google Scholar]
- 28.Kecklund G, Akerstedt T. Sleepiness in long distance truck driving: an ambulatory EEG study of night driving. Ergonomics. 1993;36:1007–1017. doi: 10.1080/00140139308967973. [DOI] [PubMed] [Google Scholar]
- 29.Vignaendra V, Walsh J, Burrows S. The application of prolonged EEG telemetry and videotape recording to the study of seizures and related disorders. Clin Exp Neurol. 1979;16:81–94. [PubMed] [Google Scholar]
- 30.Vincent D, Favarel-Garrigues F, Faure J, Julien J. On a case of narcolepsy with paroxysmal phenomena. A diurnal and nocturnal study. Electroencephalogr Clin Neurophysiol. 1967;23:284. [PubMed] [Google Scholar]
- 31.Berger-Gross P, Bruder GE, Quitkin F, Goetz R. Auditory laterality in depression: relation to circadian patterns and EEG sleep. Biol Psychiatry. 1985;20:611–622. doi: 10.1016/0006-3223(85)90096-4. [DOI] [PubMed] [Google Scholar]
- 32.Buguet A, Gati R, Sevre JP, Develoux M, Bogui P, Lonsdorfer J. 24 hour polysomnographic evaluation in a patient with sleeping sickness. Electrtoencephalogr Clin Neurophysiol. 1989;72:471–478. doi: 10.1016/0013-4694(89)90223-x. [DOI] [PubMed] [Google Scholar]
- 33.Takii O. Diurnal rhythm in appearance of frontal midline theta activity. Jap J Psychiatry Neurol. 1986;40:609–615. doi: 10.1111/j.1440-1819.1986.tb03175.x. [DOI] [PubMed] [Google Scholar]
- 34.Scheich H. Interval histograms and periodic diurnal changes of human alpha rhythms. Electroencephalogr Clin Neurophysiol. 1969;26:442. [PubMed] [Google Scholar]
- 35.Gundel A, Hilbig A. Circadian acrophases of powers and frequencies in the waking EEG. Int J Neurosci. 1983;22:125–133. doi: 10.3109/00207459308987391. [DOI] [PubMed] [Google Scholar]
- 36.Goldenberg F. Refinement of an EEG diurnal vigilance index. Application to drug trials. Rev Electroencephalogr Neurophysiol Clin. 1986;16:39–48. doi: 10.1016/s0370-4475(86)80017-x. [DOI] [PubMed] [Google Scholar]
- 37.Cacot P, Tesolin B, Sebban C. Diurnal variations of EEG power in healthy adults. Electroencephalogr Clin Neurophysiol. 1995;94:305–312. doi: 10.1016/0013-4694(94)00298-y. [DOI] [PubMed] [Google Scholar]
- 38.Dimpfel W. The new route to brain mapping. Med Technol Int Yearbook. 1993:24–25. [Google Scholar]
- 39.Jasper H. The 10: 20 Electrode System of the International Federation. Electroencephalogr Clin Neurophysiol. 1958;10:371–375. [PubMed] [Google Scholar]