Automatic classification algorithms of the EEG monitor Narcotrend for routinely recorded EEG data from general anaesthesia: a validation study.

Impacts of hypnotic drugs on brain function are reflected in the EEG. The EEG monitor Narcotrend performs an automatic classification of the EEG using a scale which was proposed by Kugler for visual evaluation of the EEG. In this article the results of a validation study of the automatic classification algorithms implemented in the EEG monitor Narcotrend are presented. Visual and automatic classification of EEG data recorded in routine clinical practice were compared. The correlation between visual and automatic assessment was high (Spearman rank correlation r = 0.90, prediction probability Pk = 0.90) and a sufficient agreement between visual and automatic assessment was achieved for 92% of the analysed EEG epochs. The results of the study suggest that the automatic classification algorithms implemented in the EEG monitor Narcotrend yield a reliable assessment of the depth of hypnosis.     

EEG power spectrum analysis for monitoring depth of anaesthesia during experimental surgery

The first attempts to introduce computerized power spectrum analysis of the electroencephalogram (EEG) as an intraoperative anaesthesia monitoring device started approximately 30 years ago. Since that time, the effects of various anaesthetic agents, sedative and analgesic drugs on the EEG pattern have been addressed in numerous studies in human patients and different animal species. These studies revealed dose-dependent changes in the EEG power spectrum for many intravenous and volatile anaesthetics. Moreover, EEG responses evoked by surgical stimuli during relative light levels of surgical anaesthesia have been classified as 'arousal' and 'paradoxical arousal' reaction, previously referred to as 'desynchronization' and 'synchronization', respectively. Contrasting reports on the correlation between quantitative EEG (QEEG) variables derived from power spectrum analysis (i.e. spectral edge frequency, median frequency) and simultaneously recorded clinical signs such as movement and haemodynamic responses, however, limited the routine use of intraoperative EEG monitoring. In addition, the appearance of EEG burst suppression pattern and isoelectricity at clinically relevant concentrations/doses of newer general anaesthetics (i.e. isoflurane, sevoflurane, propofol) may have weakened the dose-related EEG changes previously reported. Despite these findings, the EEG power spectrum analysis may still provide valuable information during intraoperative monitoring in the individual subject. The information obtained from EEG power spectrum analysis may be further supplemented by newer EEG indices such as bispectral index and approximate entropy or other neurophysiological monitors including auditory evoked potentials or somatosensory evoked potentials.     

A Pharmacokinetics-Neural Mass Model (PK-NMM) for the Simulation of EEG Activity during Propofol Anesthesia

Modeling the effects of anesthetic drugs on brain activity is very helpful in understanding anesthesia mechanisms. The aim of this study was to set up a combined model to relate actual drug levels to EEG dynamics and behavioral states during propofol-induced anesthesia. We proposed a new combined theoretical model based on a pharmacokinetics (PK) model and a neural mass model (NMM), which we termed PK-NMM—with the aim of simulating electroencephalogram (EEG) activity during propofol-induced general anesthesia. The PK model was used to derive propofol effect-site drug concentrations (C _eff) based on the actual drug infusion regimen. The NMM model took C _eff as the control parameter to produce simulated EEG-like (sEEG) data. For comparison, we used real prefrontal EEG (rEEG) data of nine volunteers undergoing propofol anesthesia from a previous experiment. To see how well the sEEG could describe the dynamic changes of neural activity during anesthesia, the rEEG data and the sEEG data were compared with respect to: power-frequency plots; nonlinear exponent (permutation entropy (PE)); and bispectral SynchFastSlow (SFS) parameters. We found that the PK-NMM model was able to reproduce anesthesia EEG-like signals based on the estimated drug concentration and patients’ condition. The frequency spectrum indicated that the frequency power peak of the sEEG moved towards the low frequency band as anesthesia deepened. Different anesthetic states could be differentiated by the PE index. The correlation coefficient of PE was 0.80±0.13 (mean±standard deviation) between rEEG and sEEG for all subjects. Additionally, SFS could track the depth of anesthesia and the SFS of rEEG and sEEG were highly correlated with a correlation coefficient of 0.77±0.13. The PK-NMM model could simulate EEG activity and might be a useful tool for understanding the action of propofol on brain activity.     

Adaptive method for EEG spectrum component decomposition

A new computerized method for EEG rhythms extraction is proposed as a development of the idea of adjustable boundaries of frequency components that was put forward in previous investigations. Principle component analysis of the correlation matrix of EEG spectra with subsequent rotation of factor solutions was used for decomposition of a spectrum into physically meaningful spectral components. The method was tested on EEG of 14 healthy subjects recorded in 17 functional waking states. Fourteen independent spectral components in the spectral range from 0 to 100 Hz were extracted and their frequency boundaries were consistent with the current knowledge on frequency components of EEG oscillations. Main advantage of the described method is the adjustable estimation of EEG frequency oscillators taking into account characteristic properties of individual EEGs. Possible area of application might be the correct evaluation of spectral power of the EEG rhythms, EEG coherence and other spectral characteristics in clinical and experimental research, studies of the frequency characteristics of the EEG rhythms in different human functional states, changes in frequency characteristics of the EEG rhythms during maturation and in mental pathology.     

Automated EEG preprocessing during anaesthesia: new aspects using artificial neural networks.

The computer-aided detection of artefacts became an essential task with increasing automation of quantitative electroencephalogram (EEG) analysis during anaesthesiological applications. The different algorithms published so far required individual manual adjustment or have been based on limited decision criteria. In this study, we developed an artificial neural networks-(ANN-)aided method for automated detection of artefacts and EEG suppression periods. 72 hr EEG recorded before, during and after anaesthesia with propofol have been evaluated. Selected parameterized patterns of 0.25 s length were used to train the ANN (22 input, 8 hidden and 4 output neurons) with error back propagation. The detection performance of the ANN-aided method was tested with processing epochs between 1 to10 s. Related to examiner EEG evaluation, the average detection performance of the method was 72% sensitivity and 80% specificity for artefacts and 90% sensitivity and 92% specificity for EEG suppression. The improvement in signal-to-noise ratio with automated artefact processing was 1.39 times for the spectral edge frequency 95 (SEF95) and 1.89 times for the approximate entropy (ApEn). We conclude that ANN-aided preprocessing provide an useful tool for automated EEG evaluation in anaesthesiological applications.     

Interactions between electrical activities of the sensomotor cortex and the hippocampus during 'animal hypnosis' in rabbits

The electrical activity of the left and right sensorimotor cortex and left and right dorsal hippocampus (CA3 fields) was recorded during "animal hypnosis" in rabbits. The "animal hypnosis" produced asymmetry in the spectral power of the hippocampal electrical activity due to an increase in the power of delta 1, delta 2, and theta 1 components in the left-hippocampus and decrease in the spectral power in the same ranges in the right-hippocampus. Hemispheric asymmetry in the electrical activity during the "animal hypnosis" was also expressed in the indices of coherence between the sensorimotor cortex and hippocampus. EEG coherence between the left sensorimotor cortex and left hippocampus in the delta 1, theta 1, and theta 2 ranges was higher than that between the right-side structures.     

Human mediotemporal EEG characteristics during propofol anesthesia

Evidence for a response-control-related kind of declarative memory during deep propofol anesthesia has recently been reported. Connectivity within the mediotemporal lobe (MTL), and in particular rhinal–hippocampal synchronization within the gamma band, has been shown to be crucial for declarative memory formation. Thus, we analyzed EEG recordings obtained from the scalp, as well as directly from within the hippocampus and from the anterior parahippocampal gyrus, which is covered by rhinal cortex, in patients with unilateral temporal lobe epilepsy during propofol anesthesia, which preceded electrode explantation. For the gamma band a power decrease starting with induction of anesthesia was observed at scalp position Cz, but a power increase was detected at MTL locations. In contrast to prior results for sleep recordings, rhinal–hippocampal coherence did not decrease within the gamma band at deeper levels of anesthesia. These findings may represent an indirect electrophysiological correlate of partially intact declarative memory formation during deep propofol sedation. Furthermore, we investigated how well the plasma propofol level, as well as different stages of anesthesia including the burst suppression phase, could be monitored by different spectral as well as by nonlinear EEG measures. We observed that conventional spectral power measures, most prominently those recorded from mediotemporal locations, are most closely correlated with the plasma propofol level, whereas different stages of anesthesia can be distinguished best by nonconventional spectral as well as nonlinear measures.     

Differences in brain activity during olfactory testing of female body odors between young men with and without sexual experience

Olfactory testing was performed in two groups of young men aged 18–22 years with and without sexual experience. Multichannel EEG was recorded at rest and upon the presentation of 45 natural body odors of young women in different phases of the menstrual cycle. The odors were either intact or mixed with J’Adore L’Absolu perfume (Christian Dior) with or without a female pheromone. Changes in spectral power within narrow frequency ranges in response to olfactory stimulation were estimated by multivariate analysis of variance. It was found that background electrical activity of the brain in young men with regular sex life had lower spectral power in all frequency ranges. The EEG responses to the intact olfactory stimuli in all young men were associated with a decrease in spectral power in the majority of the frequency ranges and especially in the θ₂ and high- and low-frequency α rhythms. Presentation of female odors masked with a perfume resulted in bidirectional EEG responses; i.e., the spectral power in the θ₁ and β₁ ranges increased or decreased depending on the sexual experience of men and the menstrual cycle phase of a female donor as compared with an intact odor. The comparison of the EEG responses to intact or perfume-masked odors showed that individual odor remains dominant.     

Association between individual EEG characteristics and the level of intelligence

The aim of this study was to investigate the relationship between individual electroencephalogram (EEG) characteristics in the resting state and the level of nonverbal intelligence. The study involved 77 students of Demidov Yaroslavl State University. Analysis of the relationship between IQ and spectral parameters of EEG theta, alpha, and two subbands of beta oscillations revealed that the amplitude and power of alphaband EEG oscillations and low frequency beta-band EEG oscillations were positively correlated with the performance in the nonverbal intelligence test. The variety of brain periodic regimes was assessed using the correlation dimension (CD) of EEG. The correlation dimension can be used to quantify the degree of complexity of the nonlinear dynamical system. It was found that the value of the EEG correlation dimension was positively associated with the level of intelligence. The periodicity of the EEG signal was studied using autocorrelation analysis. It was shown that the autocorrelogram duration was negatively associated with IQ and the autocorrelogram amplitude was positively associated with IQ. A regression equation for predicting the level of nonverbal intelligence based on the power of theta- and beta-band oscillations, alpha-band oscillation indexes, and the amplitude and autocorrelation characteristics of the EEG signal was obtained.     

A correlation of EEG characteristics of pregnant women with the level of their anxiety

It has been shown that the EEG of pregnant women with high anxiety level is characterized by a lower occipital alpha and theta rhythm spectral power if compared to the EEG of women with low anxiety level. The frequency of the alpha rhythm of their EEG was reliably higher. Pregnant women with high anxiety level with a pregnancy interruption threat diagnosis have an essentially lower occipital alpha rhythm spectral power than women of this group without such a diagnosis. And vice versa, the occipital alpha rhythm spectral power in the EEG of pregnant women with low anxiety level with a pregnancy interruption threat diagnosis is essentially higher and its frequency essentially lower than the EEG of women without that diagnosis. The data received are interpreted as a change in hormone regulation during the pregnancy period, as well as psychogenic influence on the pregnancy.     

JM-1232(−) and propofol,a new combination of hypnotics with short-acting and non-cumulative preferable properties

Drug interactions are significant in anesthesiology because drug combinations can potentially possess novel properties. The pharmacological advantages of a new combination of the benzodiazepine receptor agonist JM-1232(−) and propofol were investigated in mice. Male adult mice were administered JM-1232(−) or propofol or combinations of the two drugs intravenously. Loss of the righting reflex was evaluated as achieving hypnosis, and the time until recovery of the reflex was measured as hypnosis time. After determining the ED₅₀, doses double and triple the ED₅₀ of propofol were injected with JM-1232(−) to compare hypnosis time. The injections were repeated four times, and the hypnosis times were compared. Flumazenil was administered separately immediately after the last dose was injected. The ED₅₀ values ([95% confidence interval]) for hypnosis were 3.76 [3.36–4.10] for JM-1232(−) and 9.88 [8.03–11.58] mg kg⁻¹ for propofol. Co-administration of 0.5 and 1 mg kg⁻¹ JM-1232(−) reduced the ED₅₀ values of propofol to 1.76 [1.21–2.51] and 1.00 [0.46–1.86] mg kg⁻¹, respectively. The drug combination for hypnosis produced a supra-additive interaction. Hypnosis time was significantly shorter in the groups given the mixtures compared to each hypnotic administered alone. After repeated injections, hypnosis time with the mixtures showed smaller prolongation than that with the hypnotic alone. Flumazenil completely restored the recovery time after anesthesia. The combination of JM-1232(−) and propofol showed a supra-additive interaction, and the reduced hypnotic dose contributed to a faster recovery even after multiple injections.     

Effects of the anesthetic agent propofol on neural populations

The neuronal mechanisms of general anesthesia are still poorly understood. Besides several characteristic features of anesthesia observed in experiments, a prominent effect is the bi-phasic change of power in the observed electroencephalogram (EEG), i.e. the initial increase and subsequent decrease of the EEG-power in several frequency bands while increasing the concentration of the anaesthetic agent. The present work aims to derive analytical conditions for this bi-phasic spectral behavior by the study of a neural population model. This model describes mathematically the effective membrane potential and involves excitatory and inhibitory synapses, excitatory and inhibitory cells, nonlocal spatial interactions and a finite axonal conduction speed. The work derives conditions for synaptic time constants based on experimental results and gives conditions on the resting state stability. Further the power spectrum of Local Field Potentials and EEG generated by the neural activity is derived analytically and allow for the detailed study of bi-spectral power changes. We find bi-phasic power changes both in monostable and bistable system regime, affirming the omnipresence of bi-spectral power changes in anesthesia. Further the work gives conditions for the strong increase of power in the δ-frequency band for large propofol concentrations as observed in experiments.     

EEG parameters and their combination as indicators of depth of anaesthesia.

EEG parameters for assessment of depth of anaesthesia are typically based on different signal processing methods, such as spectral and complexity analysis. In the present study, the parameters investigated (WSMF, qWSMF, approximate entropy and Lempel-Ziv complexity) do not correlate monotonically to depth of anaesthesia. To obtain this correlation, parameters are combined based on fuzzy inference, whereby each parameter only operates in a specific range. Fuzzy inference seems to be a suitable approach, as the indicator designed separates wakefulness from unconsciousness as well as the best single parameter does and correlates to the depth of anaesthesia.     

Relationship of bispectral index values, haemodynamic changes and recovery times during sevoflurane or propofol anaesthesia in rabbits

The aim of this study was to determine and compare the degree of hypnosis achieved during propofol or sevoflurane anaesthesia in rabbits using bispectral index (BIS), and to evaluate its usefulness as a predictor of both haemodynamic changes during anaesthesia and recovery times. Twenty adult male New Zealand White rabbits, average weight 4.4 +/- 0.4 kg, were used for this study. Animals were randomly allocated to one of two groups with 10 rabbits/group. An electroencephalographic recording was obtained from each conscious rabbit prior to drug administration. All animals received buprenorphine as a preanaesthetic medication (0.05 mg/kg, intravenous [i.v.]). Anaesthesia was induced with propofol (8 mg/kg, i.v.) in all animals; 10 rabbits were maintained with sevoflurane via inhalation (1 minimum alveolar concentration--end-tidal sevoflurane concentration of 3.7%--at a fresh gas flow rate of 3 L/min; group I), and 10 were maintained with i.v. propofol (0.6 mg/kg/min; group II). The rabbits were orotracheally intubated and spontaneous ventilation was maintained throughout the study (100% oxygen). After abdominal surgery through a ventral midline laparotomy, rabbits were allowed to recover from anaesthesia. Cardiovascular variables and BIS values were recorded at intervals throughout the procedure, as was the duration of recovery from anaesthesia. In both groups, mean BIS values were significantly decreased immediately after induction, compared with baseline values obtained during consciousness. Anaesthetic depth (evaluated by clinical observation) was similar in both groups; however, group II rabbits had significantly higher (P<0.001) BIS values from 30 s before incision until anaesthesia was discontinued. There was no significant difference in BIS recorded 1 and 5 min after incision as compared with values obtained 30 s before incision in either group. During sevoflurane or propofol administration, correlations were found between BIS values and mean arterial blood pressure (MABP), and between BIS values and heart rate (HR). Mean BIS values at discontinuation of administration of the anaesthetic agent were greater in group II (69.1 +/- 6.0) than in group I (49.3 +/- 2.2). However, recovery from anaesthesia was significantly longer in group II (38.4 +/- 7.2 min) than in group I (11.5 +/- 2.5 min). In conclusion, BIS can be used to differentiate between conscious and unconscious states during anaesthesia in rabbits. BIS values derived from an electroencephalogram at the end of anaesthesia were not useful for predicting the speed of anaesthetic recovery in sevoflurane or propofol-anaesthetized rabbits undergoing abdominal surgery. Despite the correlation found between BIS and haemodynamic parameters, its usefulness as a predictor of clinically important changes in arterial blood pressure and HR in anaesthetized rabbits was limited.     

Relationship between cortical electrical and cardiac autonomic activities in the awake lizard, Gallotia galloti

ECG and EEG signals were simultaneously recorded in lizards, Gallotia galloti, both in control conditions and under autonomic nervous system (ANS) blockade, in order to evaluate possible relationships between the ANS control of heart rate and the integrated central nervous system activity in reptiles. The ANS blockers used were prazosin, propranolol, and atropine. Time-domain summary statistics were derived from the series of consecutive R-R intervals (RRI) of the ECG to measure beat-to-beat heart rate variability (HRV), and spectral analysis techniques were applied to the EEG activity to assess its frequency content. Both prazosin and atropine did not alter the power spectral density (PSD) of the EEG low frequency (LF: 0.5-7.5 Hz) and high frequency (HF: 7.6-30 Hz) bands, whereas propranolol decreased the PSD in these bands. These findings suggest that central beta-adrenergic receptor mechanisms could mediate the reptilian waking EEG activity without taking part any alpha(1)-adrenergic and/or cholinergic receptor systems. In 55% of the lizards in control conditions, and in approximately 43% of the lizards under prazosin and atropine, a negative correlation between the coefficient of variation of the series of RRI value (CV(RRI)) and the mean power frequency (MPF) of the EEG spectra was found, but not under propranolol. Consequently, the lizards' HRV-EEG-activity relationship appears to be independent of alpha(1)-adrenergic and cholinergic receptor systems and mediated by beta-adrenergic receptor mechanisms.     

Attenuation of High-Frequency (50-200 Hz) Thalamocortical EEG Rhythms by Propofol in Rats Is More Pronounced for the Thalamus than for the Cortex

Background

Thalamocortical EEG rhythms in gamma (30-80 Hz) and high-gamma (80-200 Hz) ranges have been linked to arousal and conscious processes. To test the hypothesis that general anesthetics attenuate these rhythms, we characterized the concentration-effect relationship of propofol on the spectral power of these rhythms. In view of the ongoing debate about cortex versus thalamus as the primary site of anesthetic action for unconsciousness, we also compared the relative sensitivity of cortex and thalamus to this effect propofol.

Methods

Adult male Long-Evans rats were chronically implanted with electrodes in somatosensory (barrel) cortex and ventroposteromedial thalamus. Propofol was delivered by a computer-controlled infusion using real-time pharmacokinetic modeling to obtain the desired plasma concentration. Spectral power was assessed during baseline, at four stable propofol plasma-concentrations (0, 3,6,9,12 μg/ml) and during recovery over four frequency ranges (30-50, 51-75, 76-125, 126-200 Hz). Unconsciousness was defined as complete loss of righting reflex. Multiple regression was used to model the change of power (after logarithmic transformation) as a function of propofol concentration and recording site.

Results

Unconsciousness occurred at the 9 μg/ml concentration in all animals. Propofol caused a robust linear concentration-dependent attenuation of cortical power in the 76-200 Hz range and of thalamic power in the 51-200 Hz range. In all instances the concentration-effect slope for the thalamus was markedly steeper than for the cortex. Furthermore the lowest concentration causing unconsciousness significantly reduced cortical power in the 126-200 Hz range and thalamic power in the 51-200 Hz range.

Conclusions

Propofol causes a concentration-dependent attenuation of the power of thalamocortical rhythms in the 51-200 Hz range and this effect is far more pronounced for the thalamus, where the attenuation provides a robust correlate of the hypnotic action of propofol.     

Coupling between Intrinsic Prefrontal HbO2 and Central EEG Beta Power Oscillations in the Resting Brain

There is increasing interest in the intrinsic activity in the resting brain, especially that of ultraslow and slow oscillations. Using near-infrared spectroscopy (NIRS), electroencephalography (EEG), blood pressure (BP), respiration and heart rate recordings during 5 minutes of rest, combined with cross spectral and sliding cross correlation calculations, we identified a short-lasting coupling (duration [Formula: see text] s) between prefrontal oxyhemoglobin (HbO2) in the frequency band between 0.07 and 0.13 Hz and central EEG alpha and/or beta power oscillations in 8 of the 9 subjects investigated. The HbO2 peaks preceded the EEG band power peaks by 3.7 s in 6 subjects, with moderate or no coupling between BP and HbO2 oscillations. HbO2 and EEG band power oscillations were approximately in phase with BP oscillations in the 2 subjects with an extremely high coupling (squared coherence [Formula: see text]) between BP and HbO2 oscillation. No coupling was identified in one subject. These results indicate that slow precentral (de)oxyhemoglobin concentration oscillations during awake rest can be temporarily coupled with EEG fluctuations in sensorimotor areas and modulate the excitability level in the brains' motor areas, respectively. Therefore, this provides support for the idea that resting state networks fluctuate with frequencies of between 0.01 and 0.1 Hz (Mantini et.al. PNAS 2007).     

Selection of valid and reliable EEG features for predicting auditory and visual alertness levels

A selection procedure with three rules, high efficiency, low individual variability, and low redundancy, was developed to screen electroencephalogram (EEG) features for predicting behavioral alertness levels. A total of 24 EEG features were derived from temporal, frequency spectral, and statistical analyses. Behavioral alertness levels were quantified by correct rates of performance on an auditory and a visual vigilance task, separately. In the auditory task study, a subset of three EEG features, the relative spectral amplitudes in the alpha (alpha%, 8-13 Hz) and theta (theta%, 4-8 Hz) bands, and the mean frequency of the EEG spectrum (MF), was found to be the best combination for predicting the auditory alertness level. In the visual task study, the mean frequency of the beta band (Fbeta, 13-32 Hz) was the only EEG feature selected. The application of an averaging subwindow procedure within a moving time window to EEG analysis increased the predictive power of EEG features and decreased the disturbing effect of movement artifacts on the EEG data.     

Segmental structure of the EEG alpha activity in adolescents with disorders of schizophrenic spectrum

The EEG records of 39 healthy adolescents and 45 age-matched schizophrenic patients were analyzed. The broad-band EEG spectral analysis and segmental analysis of the alpha-activity revealed significant differences between the groups. Schizophrenics differed in a decreased spectral power for the alpha2 and betal frequency bands and increased power for the delta and theta bands. Also, in schizophrenic adolescents, quasi-stationary alpha-rhythm segments were longer, and within-segmental EEG amplitudes were higher than in the healthy subjects; the amplitude variability and the steepness of transitions between neighbor segments were increased. The results of the EEG segmental analysis suggest a disintegration of local cortical neuronal ensembles in schizophrenia.     

Robust closed-loop control of hypnosis with propofol using WAVCNS index as the controlled variable

This paper presents a robust closed-loop strategy for control of depth of hypnosis. The proposed method regulates the electroencephalogram (EEG)-derived WAV_CNS index as measure of hypnosis by manipulating intravenous propofol administration. In contrast to many existing closed-loop control methods for hypnosis drug delivery, the control design presented in this paper produces stability and robustness against uncertainty by explicitly accounting for the pharmacokinetic (PK) and pharmacodynamic (PD) variability between individuals, as well as the unexpected surgical stimulation and anesthetic–analgesic interaction that the closed-loop control is required to tolerate. This robust closed-loop controller was evaluated in comparison with a heuristically tuned proportional-derivative-integral (PID) controller using a simulated surgical procedure on 44 patients whose PK and PD models were identified using real clinical data. The results demonstrate that the robust control strategy can deliver propofol to yield consistent and acceptable closed-loop induction and maintenance phase responses over wide-ranging PK and PD differences (mean rise and settling times of 4 min and 7 min and mean overshoot of less than 8%, which meets anesthesiologists’ response specifications), whereas its PID control counterpart exhibits limitations in performance.