The mechanisms by which general anesthetics suppress consciousness remain elusive.1,2 One of the most important and salient manifestation of the effects of general anesthetics on the brain is an increase in low-frequency (≤4 Hz) electroencephalographic (EEG) power. 3–5 This applies to the vast majority of general anesthetics, including modern agents such as propofol6,7 and isoflurane.8 The increased low-frequency EEG power reflects the emergence of slow oscillations caused by the phasic hyperpolarization of cortical and thalamic neurons.9 Concurrently, anesthetic action also results in a decrease in high-frequency (30–80 Hz) EEG power, often referred to as gamma range.10 The hypothesis that attenuation of thalamocortical EEG rhythms in the gamma range contributes to the hypnotic effect of general anesthetics has been proposed.10–13 Gamma rhythms are an important element of the activity of the waking brain, reflecting the depolarization of thalamic and cortical neurons.14 They occur in all sensory areas and have been associated with a number of high-level neurocognitive processes, including general arousal, sensory function, and attention.15,16 Early observations of high-frequency rhythms were confined to the gamma range, but recent developments have also drawn attention to the broadband high-gamma (80–200 Hz) range.17,18
We have recently shown, with recordings from the somatosensory (barrel) cortex and ventroposteromedial (VPM) thalamus of the rat that IV propofol causes a robust linear concentration-dependent attenuation of cortical EEG power in the 76 to 200 Hz range and of thalamic power in the 30 to 200 Hz range.19 Furthermore, the concentration-effect slopes for the thalamus were markedly steeper than for the cortex. These observations provide strong evidence that general anesthesia is associated with the attenuation of high-frequency thalamocortical EEG rhythms in the gamma (30–80 Hz) and high-gamma (80–200 Hz) ranges17,18 in both cortex and thalamus.
If similar, or more extensive, effects could be demonstrated with inhaled anesthetics, the hypothesis that anesthesia is associated with impairment of high-frequency thalamocortical rhythms would be strengthened. Such findings would also inform the debate on the relative role of cortical and thalamic dysfunction in anesthetic-induced unconsciousness.1,20 Furthermore, comparative studies of the electrophysiologic effects of general anesthetics may contribute to translational research efforts in view of the accumulating evidence that the distinct brain states produced by different anesthetics are readily visible in the patient’s EEG.21
Our aim was to characterize the effects of the inhaled anesthetic, isoflurane, on high-frequency (30–200 Hz) thalamocortical EEG rhythms. The 30 to 200 Hz range was divided into the following bands chosen on the basis of our recent study and other works22,23: 30 to 50, 51 to 75, 76 to 125, and 126 to 200 Hz. For each frequency band, we evaluated the concentration-effect relationship of isoflurane on the power of the EEG recorded from the somatosensory (barrel) cortex and VPM thalamic nucleus of the rats with chronically implanted electrodes, and we compared the steepness of the cortical and thalamic regression lines. We also compared the potency of isoflurane for attenuating high-frequency power with that of propofol as reported in our recently published work. 19
All procedures adhered to the guidelines of the Canadian Council on Animal Care and were approved by the Animal Ethics Boards of Concordia University. Male Long-Evans rats (300–350 g) were acquired from Charles River Laboratories (Senneville, Quebec, Canada).
They were housed individually, provided food and water ad libitum, and maintained on a 12-hour reversed light cycle.
Animals were anesthetized with isoflurane (2.5%) in an induction chamber. Once anesthetized, they were transferred to a stereotaxic apparatus with a heated surgical plate. An ophthalmic ointment was used to protect the corneas. Anesthesia was maintained throughout the entire procedure with isoflurane (1.5%–2.5% via a nose cone) adjusted to prevent movement in response to manual toe pinch. A midline incision was made along the scalp, and the periosteum was retracted with hemostats. Bipolar electrodes made of twisted, Teflon-coated stainless steel wires (125 μm diameter; vertical separation between the wire tips: 0.5 mm; AM Systems, Sequim, WA) were then positioned in the VPM thalamic nucleus (Paxinos atlas24 coordinates: anterior/posterior [A/P]: −3.5, lateral/medial [L/M]: 2.7, ventral/dorsal [V/D]: −6.6, relative to bregma) and sensory (barrel) cortex (A/P: −2.3, L/M: 5.0, V/D: −3.6) for local field potential (LFP) recordings of EEG activity within the target regions.25 The cortical electrode was adjusted vertically to optimize the amplitude of the field excitatory postsynaptic potentials evoked by VPM stimulation. Thicker wires (280 μm diameter; AM Systems) soldered to stainless steel screws were placed in the contralateral parietal bone and the ipsilateral frontal bone to serve as the reference and ground electrodes, respectively. Electrode leads were then fastened to gold-plated Amphenol pins and inserted into a 9-pin ABS connector (Ginder Scientific, Nepean, ON, Canada). This assembly was then secured to the skull by using acrylic dental cement. All animals were administered carprofen (5 mg/kg, subcutaneous at end of surgery and repeated once a day for 2 days) and buprenorphine (0.5 mg/kg, subcutaneous at end of surgery and repeated every 12 hours for up to 3 days if needed). Animals were routinely monitored to minimize any suffering or postoperative complications. One week was allotted for recovery and observation.
The animals were placed in a 40 × 40 × 60 cm airtight Plexiglas chamber, containing separate ports for fresh gas flow, gas sampling and scavenging, and a silicone rubber seal opening allowing manual access to the animal without introduction of ambient air in the testing chamber. There was a continuous oxygen flow at a rate of 1 to 5 L/min. Oxygen, carbon dioxide, and anesthetic concentration were measured with a Capnomatic Ultima Analyzer (Datex-Ohmeda, Helsinki, Finland) calibrated before each testing session.
The experimental sessions were preceded by 20-minute acclimation sessions, which occurred daily for 3 days, allowing the animals to explore the testing chamber without exposure to anesthetic. Each rat underwent a single testing session with 5 conditions (baseline, isoflurane concentrations of 0.75%, 1.1%, and 1.5%, and recovery defined as return of spontaneous ambulation). These concentrations were based on a pilot study that showed that loss of righting occurred at 0.9% to 1.1% isoflurane. Ten minutes were needed to reach the desired inspired concentrations, and 30 minutes were allowed for equilibration at each concentration. If the animal was not ambulating at the end of equilibration, the righting reflex was tested by placing the animal on its side and by observing attempts to right. The animal was considered unconscious if no movements or righting attempts occurred. We limited the number of anesthetic concentrations to 3 to complete the main part of the experiment in 2 hours 15 minutes, a duration roughly similar to the 2 hours required for the main part of the experiment in our propofol study.19
Both referential and bipolar recordings were obtained. We report only the results of the bipolar recordings. LFPs from the cortex and thalamus were amplified (0.1–500-Hz pass band; AM Systems, Model 1700), digitized at 1024 Hz, and stored for off-line analysis. For each period, 2 minutes of high-quality data devoid of artifacts was obtained.
Spectral power was computed with the Welch method26 by using 4-second segments and a Hanning window with 50% overlap (Matlab Signal Processing Toolbox, version 6; MathWorks Inc., Sherborn, MA). Segments containing >5% outliers (defined as values outside mean ± 3 SDs of the entire recording of approximately 2 minutes) were excluded from analysis. The frequency bands were defined as 30 to 50, 51 to 75, 76 to 125, and 126 to 200 Hz. To minimize the impact of interference from external electrical sources, power at 59 to 61 Hz, 119 to 121 Hz, and 179 to 181 Hz were excluded from analysis. Because baseline power differs between animals, normalization of the spectra is required to avoid an undue influence of animals with high-baseline power. Power spectra were normalized for each animal and recording site by dividing the power values by the average power from 0.1 to 200 Hz during baseline. Thus, for each animal and recording site, the average power during baseline was equal to unity, and the spectra for the other periods were scaled with the same parameter.
Perfusion and Histology
After completing the testing session, animals were killed with a lethal dose of urethane and perfused via the left ventricle with heparinized saline followed by 10% neutral buffered formalin. The isolated brains were postfixed in 4% paraformaldehyde for histologic processing. Accurate electrode placements could be confirmed in 5 animals based on the termination of the electrode tracks. In the other animal, damage to the brain during retrieval or processing prevented identification of the electrode tracks or surrounding structures.
Based on the data of our prior report,19 we aimed for a sample size of ≥8 animals. Animals were instrumented in groups of 2 or 3, and the raw traces and spectra were inspected after testing each rat to confirm that valid data had been obtained. No interim statistical analyses were conducted. When we had obtained valid data from 8 animals, we stopped instrumenting animals but tested animals that had already been instrumented. Ten animals were instrumented, but a problem with the electrode assembly made it impossible to test 1 animal. Valid data were, therefore, obtained from 9 animals.
All power values were transformed with log base 10. Multiple regression was used to model the changes of log power as a linear function of concentration, with a common slope used for all animals. A separate analysis was conducted for each frequency band (30–50, 51–75, 76–125, and 126–200 Hz). For each analysis, there were 4 repeated measurements (corresponding to the baseline and 3 isoflurane concentrations: x j, j = 1, 2, 3, and 4, corresponding to 0, 0.75%, 1.1%, and 1.5%) for each recording site (k = 1 for cortex and k = 2 for thalamus). We modeled each outcome y ijk as follows:
- yijk denotes the spectral power of rat i (i = 1, 2, …, 9) at concentration j for site k
- αik is the difference between the random intercept of rat i and β0k
- β0k is the regression constant for site k
- β1k is the regression coefficient for the linear term (i.e., slope) for site k.
- εijk is the error term, which accounts for unexplained errors (e.g., measurement error).
The model contains both fixed effects (the influence of concentration and recording site) and a random effect (the baseline values of each rat). This mixed model was estimated with PROC MIXED in SAS Statistical Software (version 9.2; SAS Institute Inc., Cary, NC) by using an autoregressive covariance structure. The choice of a mixed, more complex regression model is required because the experimental design involves serial measurements in the same animals (see Ref. 27 for a detailed account).
Inspection of the residuals revealed no anomalies. The P values for the linear term were corrected for multiple comparisons (8 tests) with the Hommel procedure.28 The goodness of fit was assessed with the concordance correlation coefficient (CCC), which was developed for mixed models as an alternative to the traditional R 2 in linear regression.29,30 This analysis was done for each site separately. The value of CCC ranges from −1 (perfect discordance) to 1 (perfect concordance). A value of 0 reflects no relationship at all.
We compared the cortical and thalamic regression coefficients for the linear term (β1) for each frequency range to assess the significance of differences in the slope of concentration-effect plots by modeling signals from 2 sites simultaneously based on approaches for multivariate longitudinal data.31 We tested the null hypothesis as follows:
follows a normal distribution, where
is the SE of the difference of the coefficients. (The underline denotes the estimate of a parameter: e.g.,
is the estimate, derived from the data, of β1Cortex.). To obtain the SE, we used the variance of
computed with the following standard formula32:
where s u-v 2 is the variance of the difference between the variables u and v, s u and s v is their SD, and r is the correlation coefficient between u and v. Paired t tests were used to compare spectral power between 1.5% isoflurane and recovery.
To compare the regression coefficients for the linear term (β1) obtained for isoflurane with those recently published for propofol,19 we assumed that a target concentration 9 μg/mL propofol was equivalent to 1.1% isoflurane because they are the lowest concentrations that abolished the righting reflex in all animals. The mean and SE of the linear terms obtained from SAS Proc Mixed for propofol were multiplied by 9 and those for isoflurane were multiplied by 1.1. This is equivalent to computing the regression using 0, 0.33, 0.67, 1, and 1.33 for concentration (instead of 0, 3, 6, 9, and 12 μg/mL for propofol) and 0, 0.68, 1, and 1.36 (instead of 0, 0.75%, 1.1%, and 1.5% for isoflurane). The linear terms were compared with t tests for independent samples.
All tests were adjusted for multiple comparisons (4 tests for each site) with Hommel procedure28 implemented in SAS Statistical Software (version 9.2; SAS Institute Inc.).
After equilibration at the 0.75% isoflurane plateau, animals ambulated spontaneously with poor coordination and all promptly righted after attempts to place them on their side. Ambulation and righting attempts were completely abolished at the 1.1% plateau.
Figure 1 shows that isoflurane caused a highly significant (P < 0.001) linear decrease of logarithmic EEG power as a function of concentration for both the cortex and thalamus in all frequency ranges. Table 1 details the results of the regression analysis of the effect of isoflurane on spectral power in each frequency band, including the regression coefficients for the linear term (β1, i.e., the slope of the regression line), along with their SE and the probability that they differ from 0. In addition, the CCCs, which provide a measure of how well the regression parameters fit the data, are all >0.5, an indication that the regression analysis provided a satisfactory fit of the data. Furthermore, the slope of the regression line (β1 parameter in Table 1) for the thalamus is significantly steeper than for the cortex in the 51 to 75 Hz (P = 0.029), 76 to 125 Hz (P < 0.001), and 126 to 200 Hz (P < 0.001) bands. The difference between slopes for the 30 to 50 Hz band was not significant (P = 0.020).
The concentration-dependent effect of isoflurane on the power of fast rhythms is readily noticeable on the average spectra with all animals combined (Fig. 2 top panels) as well as on the power spectra from a single animal (Fig. 2 bottom panels). The distance between the traces in these spectra is greater for the thalamus than for the cortex, an observation that is consistent with the steeper negative slope of the concentration-effect plots for the thalamus.
The attenuation of the power of fast rhythms is also readily apparent in spectrograms (Fig. 3B) based on 30-second recordings (shown in Fig. 3A) from a single animal. The decrease in high-frequency power is revealed by the predominance of the blue color for the 0.75%, 1.1%, and 1.5% panels, and this effect was more pronounced for the thalamus than for the cortex, an observation suggesting a more pronounced effect on the thalamus. The spectrograms also indicate stable frequency content of the EEG within each period, and this pattern was maintained during the entire recording.
Table 2 shows the comparison of the isoflurane slope parameters with those recently reported for propofol.19 The comparison shows that the slope for the cortex is significantly less steep with propofol than for isoflurane for all frequency ranges (P = 0.033). The slopes for the thalamus were steeper with isoflurane for the 51 to 200 Hz range, but the difference was significant only for the 126 to 200 Hz band and before correction for multiple comparisons.
Recovery and Paired Comparisons
The transition from the last concentration studied (1.5%) to recovery was associated with a significant increase in power in all instances (P values ranged from 0.0001 to 0.0495 for cortex and from 0.0001 to 0.0014 for thalamus after corrections for multiple comparisons). Recovery recordings were obtained to confirm a return toward baseline of power values after cessation of anesthetic administration and to rule out other possible causes for the observed power changes such as duration of experiment, technical fault, etc. The study was not designed to identify the precise moment when the animals regained consciousness.
The main new finding of this study is that isoflurane causes a concentration-dependent attenuation of the power of high-frequency (30–200 Hz) EEG in the cerebral cortex and thalamus and that this effect is stronger for the thalamus than for the cortex (in the 51–200-Hz range). We are aware of no prior reports of alterations of thalamic high-frequency rhythms by isoflurane. The present observations demonstrate, for the first time, important similarities between the effects of isoflurane and those we recently reported for propofol.19 Both drugs caused a comparable attenuation of high-frequency rhythms in the sensory thalamus, and their effect on the thalamus was stronger than on the cortex. These effects may possibly constitute an electrophysiologic signature of the anesthetized state, similar to that revealed by functional brain imaging in humans.33 This is not surprising because high-frequency LFP power correlates tightly with changes in regional cerebral blood.34 The present observations provide strong evidence that anesthetic-induced unconsciousness is associated with impairment of high-frequency thalamocortical rhythms and also strongly reaffirm the role of impaired thalamic function in anesthetic action.20,35
There are interesting differences between isoflurane and propofol with regard to their effect on the cerebral cortex. First, the attenuation by isoflurane of cortical power occurred over the entire frequency range studied (30–200 Hz), whereas the effect of propofol was limited to the 76 to 125 and 126 to 200 Hz bands. Second, the slope of the attenuation of cortical power was significantly steeper for isoflurane in all frequency bands. Isoflurane thus appears to cause more cortical suppression than propofol in the concentration range studied. A similar difference was reported in humans using somatosensory-evoked potentials.36 The more intense suppression observed with isoflurane in the present study cannot be attributed to use of higher concentrations for isoflurane because the range of concentrations (normalized for each drug by the lowest concentration that abolished righting) were similar (see Statistical Analysis).
Although the concentration ranges of propofol and isoflurane were chosen by aiming for approximately equivalent effects on the loss of the righting reflex, the results of the comparison between these agents require careful interpretation. First, the comparison of potency of these drugs was solely based on their ability to abolish righting and may not apply to their other effects, such as suppression of movement in response to pain. Second, the demonstration of a steeper concentration-effect curve for isoflurane provides no information on the maximal level of cortical suppression achievable with these drugs if given at higher concentrations. Thus, although we can conclude that, in the concentration range relevant for loss of righting, isoflurane causes more suppression of cortical activity than propofol, this does not mean that isoflurane is intrinsically more potent than propofol in reducing cortical activity.
The most likely explanation for the steeper concentration-effect slope of isoflurane is that this agent, similar to other volatile anesthetics, has additional molecular targets, notably the 2-pore potassium channels and the N-methyl-D-aspartate receptor.37 As noted by Purdon et al.21 in their recent review, the similarity of the spectral signatures (in the 0–30 Hz range) of propofol and isoflurane suggest that potentiation of the γ-aminobutyric acid (GABA) type A receptors is likely a key determinant of their mechanisms of action. The more pronounced effect of both propofol and isoflurane on thalamic EEG is also compatible with the effects mediated by GABA type A receptors, given the pivotal role of the reticular thalamic nucleus, which influences thalamic function via GABAergic inhibition.
In planning the experiment, we assumed that the EEG would remain stable at each anesthetic concentration plateau, and inspection of the data revealed stable EEG patterns within each period. This finding contrasts with those reported by Hudson et al.38 who found that ≥2 activity patterns with distinct power frequency distributions could occur at a stable anesthetic concentration. Hudson et al. used prolonged recording periods (approximately 1 hour per concentration) and obtained their recordings in an acute setting. These factors may explain why we did not observe changes of EEG patterns.
The present report confirms prior observations on the effects of volatile anesthetics on spontaneous (as opposed to stimulus induced) fast (30–200 Hz) EEG rhythms in the cerebral cortex. Hudetz et al.23 showed, in rats, that isoflurane attenuates spectral power in the 70 to 140 Hz range in neocortical and hippocampal recordings in a concentration-dependent manner. Silva et al.39 showed that isoflurane 0.8% decreases power in the 50 to 100 Hz range in rat primary sensory cortex. These 2 groups, however, found no evidence of attenuation for the 30 to 50 Hz band as was observed in the present study, but Ma et al.40 showed a reduction of 30 to 100 Hz power by halothane in hippocampal EEG.
In contrast to the present findings, Vanderwolf41 reported that the amplitude of cortical gamma (30–90 Hz) activity in rats during isoflurane anesthesia is variable and often greater than during waking state. However, Vanderwolf did not examine concentration-effect relationships, but rather studied effects of low concentrations of isoflurane combined with noxious stimulation sufficient to cause righting, and also used a different method to quantify gamma activity. An important aspect of the study by Vanderwolf is the evidence that the amplitude of gamma activity is influenced by the phase of the slow EEG oscillations, an observation that deserves further inquiry.
The concentration of isoflurane that caused loss of righting in all animals (1.1%) is greater than the 0.8% concentration reported by Hudetz et al.23 and Silva et al.39 Two factors may account for the difference. We defined loss of righting as absence of any righting attempts or movements after placing the animal on its side, a definition that is more stringent. Second, Hudetz et al. and Silva et al. used Sprague-Dawley rats and we used Long-Evans. There is evidence that the minimal anesthetic concentration of isoflurane is higher for Long-Evans than for Sprague-Dawley rats,42 suggesting that the Long-Evans may be more resistant to the effects of isoflurane.
We propose that the attenuation of power of ≥80 Hz by isoflurane reflects a decrease in neuronal firing rates based on the evidence that high-gamma power (80–150 Hz) in LFPs is strongly correlated with average firing rate18,43 and that isoflurane markedly decreases spontaneous neuronal firing rates in the rat barrel cortex44 and VPM.45 Concurrent recordings of both LFPs and of action potentials with the same electrodes would allow verification of this interpretation.
We conclude that isoflurane, similar to propofol, causes a concentration-dependent attenuation of the spectral power of thalamocortical rhythms in the 30 to 200 Hz range. This effect is more consistent and more pronounced for the thalamus than for the cortex for frequencies >50 Hz. These observations provide strong support for the hypothesis that anesthetic-induced unconsciousness is associated with impairment of high-frequency thalamocortical rhythms and rhythms that reveal increased neuronal activity and possibly reflect mechanisms of information coding and network function.17 Further work, thus, is warranted to determine how, and to what extent, the physiologic changes revealed by the attenuation of high-frequency spectral power contribute to the hypnotic action of general anesthetics.
Name: Gilles Plourde, MD.
Contribution: This author helped design the study, perform the surgery and testing, analyze the data, and write the manuscript.
Attestation: Gilles Plourde has seen the original study data, performed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.
Name: Sean J. Reed, MSc.
Contribution: This author helped design the study, perform the surgery and testing, analyze the data, and write the manuscript.
Attestation: Sean J. Reed has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: C. Andrew Chapman, PhD.
Contribution: This author helped design the study, provided assistance for initial testing sessions, and participated to the writing the manuscript.
Attestation: C. Andrew Chapman has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
This manuscript was handled by: Markus W. Hollmann, MD, PhD, DEAA.
We thank Chantale Porlier for help with testing and Xianming Tan (Biostatistics Core Facilities, McGill University Health Center Research Institute) for advice on regression analysis and SAS programming.
1. Alkire MT, Hudetz AG, Tononi G. Consciousness and anesthesia. Science 2008;322:876–80.
2. Franks NP. General anaesthesia: from molecular targets to neuronal pathways of sleep and arousal. Nat Rev Neurosci 2008;9:370–86.
3. Gibbs FA, Gibbs LE, Lennox WG. Effects on the electroencephalogram of certain drugs which influence nervous system activity. Archiv Intern Med 1937;60:154–66.
4. Clark DL, Rosner BS. Neurophysiologic effects of general anesthetics: 1. The electroencephalogram and sensory evoked responses in man (Review article). Anesthesiology 1973;38:564–82.
5. Stockard J, Bickford R. Gordon E, The neurophysiology of anaesthesia. In: A Basis and Practice of Neuroanaesthesia. 1981:1st ed. Amsterdam, North Holland, The Netherlands: Biomedical Press, 3–46.
6. Ní Mhuircheartaigh R, Warnaby C, Rogers R, Jbabdi S, Tracey I. Slow-wave activity saturation and thalamocortical isolation during propofol anesthesia in humans. Sci Transl Med 2013;5:208ra148.
7. Purdon PL, Pierce ET, Mukamel EA, Prerau MJ, Walsh JL, Wong KF, Salazar-Gomez AF, Harrell PG, Sampson AL, Cimenser A, Ching S, Kopell NJ, Tavares-Stoeckel C, Habeeb K, Merhar R, Brown EN. Electroencephalogram signatures of loss and recovery of consciousness from propofol. Proc Natl Acad Sci U S A 2013;110:E1142–51.
8. Clark DL, Hosick EC, Adam N, Castro AD, Rosner BS, Neigh JL. Neural effects of isoflurane (forane) in man. Anesthesiology 1973;39:261–70.
9. Steriade M, McCormick DA, Sejnowski TJ. Thalamocortical oscillations in the sleeping and aroused brain. Science 1993;262:679–85.
10. John ER, Prichep LS. The anesthetic cascade: a theory of how anesthesia suppresses consciousness. Anesthesiology 2005;102:447–71.
11. Madler C, Keller I, Schwender D, Pöppel E. Sensory information processing during general anaesthesia: effect of isoflurane on auditory evoked neuronal oscillations. Br J Anaesth 1991;66:81–7.
12. Plourde G. Depth of anesthesia. Clinical use of the 40-Hz auditory steady state response. Int Anesthesiol Clin 1993;31:107–20.
13. Mashour GA. Consciousness unbound: toward a paradigm of general anesthesia. Anesthesiology 2004;100:428–33.
14. Steriade M. Central core modulation of spontaneous oscillations and sensory transmission in thalamocortical systems. Curr Opin Neurobiol 1993;3:619–25.
15. Bartos M, Vida I, Jonas P. Synaptic mechanisms of synchronized gamma oscillations in inhibitory interneuron networks. Nat Rev Neurosci 2007;8:45–56.
16. Fries P. Neuronal gamma-band synchronization as a fundamental process in cortical computation. Annu Rev Neurosci 2009;32:209–24.
17. Crone NE, Korzeniewska A, Franaszczuk PJ. Cortical γ responses: searching high and low. Int J Psychophysiol 2011;79:9–15.
18. Ray S, Maunsell JH. Different origins of gamma rhythm and high-gamma activity in macaque visual cortex. PLoS Biol 2011;9:e1000610.
19. Reed SJ, Plourde G. Attenuation of high-frequency (50-200 Hz) thalamocortical EEG rhythms by propofol in rats is more pronounced for the thalamus than for the cortex. PLoS One 2015;10:e0123287.
20. Verdonck O, Reed SJ, Hall J, Gotman J, Plourde G. The sensory thalamus and cerebral motor cortex are affected concurrently during induction of anesthesia with propofol: a case series with intracranial electroencephalogram recordings. Can J Anaesth 2014;61:254–62.
21. Purdon PL, Sampson A, Pavone KJ, Brown EN. Clinical electroencephalography for anesthesiologists: part I: background and basic signatures. Anesthesiology 2015;123:937–60.
22. Breshears JD, Roland JL, Sharma M, Gaona CM, Freudenburg ZV, Tempelhoff R, Avidan MS, Leuthardt EC. Stable and dynamic cortical electrophysiology of induction and emergence with propofol anesthesia. Proc Natl Acad Sci U S A 2010;107:21170–5.
23. Hudetz AG, Vizuete JA, Pillay S. Differential effects of isoflurane on high-frequency and low-frequency γ oscillations in the cerebral cortex and hippocampus in freely moving rats. Anesthesiology 2011;114:588–95.
24. Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates. 20045th ed. San Diego, CA: Academic Press.
25. Buzsáki G, Anastassiou CA, Koch C. The origin of extracellular fields and currents—EEG, ECoG, LFP and spikes. Nat Rev Neurosci 2012;13:407–20.
26. Welch PD. The use of fast Fourier transform for the estimation of power spectra: a method based on time averaging over short, modified periodograms. IEEE Trans Audio Electroacoust 1967;15:70–3.
27. Glantz SA, Slinker BK. Primer of Applied Regression & Analysis of Variance. 2001Toronto, ON, Canada: McGraw-Hill Inc..
28. Hommel G. A Stagewise rejective multiple test procedure based on a modified Bonferroni test. Biometrika 1988;75:383–6.
29. Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics 1989;45:255–68.
30. Vonesh EF, Chinchilli VM, Pu K. Goodness-of-fit in generalized nonlinear mixed-effects models. Biometrics 1996;52:572–87.
31. Gao F, Thompson P, Xiong C, Miller JP. LeBouton KJ, Analyzing multivariate longitudinal data using SAS®. In: Thirty-first Annual SAS® Users Group International (SUGI) Conference San Francisco, 2006.
32. Sokal RR, Rohlf FJ. Biometry. 1969San Francisco, CA: Freeman.
33. Kaisti KK, Metsähonkala L, Teräs M, Oikonen V, Aalto S, Jääskeläinen S, Hinkka S, Scheinin H. Effects of surgical levels of propofol and sevoflurane anesthesia on cerebral blood flow in healthy subjects studied with positron emission tomography. Anesthesiology 2002;96:1358–70.
34. Niessing J, Ebisch B, Schmidt KE, Niessing M, Singer W, Galuske RA. Hemodynamic signals correlate tightly with synchronized gamma oscillations. Science 2005;309:948–51.
35. Fiset P, Paus T, Daloze T, Plourde G, Meuret P, Bonhomme V, Hajj-Ali N, Backman SB, Evans AC. Brain mechanisms of propofol-induced loss of consciousness in humans: a positron emission tomographic study. J Neurosci 1999;19:5506–13.
36. Liu EH, Wong HK, Chia CP, Lim HJ, Chen ZY, Lee TL. Effects of isoflurane and propofol on cortical somatosensory evoked potentials during comparable depth of anaesthesia as guided by bispectral index. Br J Anaesth 2005;94:193–7.
37. Franks NP. Molecular targets underlying general anaesthesia. Br J Pharmacol 2006;147(suppl 1):S72–81.
38. Hudson AE, Calderon DP, Pfaff DW, Proekt A. Recovery of consciousness is mediated by a network of discrete metastable activity states. Proc Natl Acad Sci U S A 2014;111:9283–8.
39. Silva A, Cardoso-Cruz H, Silva F, Galhardo V, Antunes L. Comparison of anesthetic depth indexes based on thalamocortical local field potentials in rats. Anesthesiology 2010;112:355–63.
40. Ma J, Shen B, Stewart LS, Herrick IA, Leung LS. The septohippocampal system participates in general anesthesia. J Neurosci 2002;22:RC200.
41. Vanderwolf CH. Are neocortical gamma waves related to consciousness? Brain Res 2000;855:217–24.
42. Russell GB, Graybeal JM. Differences in anesthetic potency between Sprague-Dawley and Long-Evans rats for isoflurane but not nitrous oxide. Pharmacology 1995;50:162–7.
43. Ray S, Crone NE, Niebur E, Franaszczuk PJ, Hsiao SS. Neural correlates of high-gamma oscillations (60-200 Hz) in macaque local field potentials and their potential implications in electrocorticography. J Neurosci 2008;28:11526–36.
44. Hentschke H, Schwarz C, Antkowiak B. Neocortex is the major target of sedative concentrations of volatile anaesthetics: strong depression of firing rates and increase of GABAA
receptor-mediated inhibition. Eur J Neurosci 2005;21:93–102.
© 2016 International Anesthesia Research Society
45. Detsch O, Vahle-Hinz C, Kochs E, Siemers M, Bromm B. Isoflurane induces dose-dependent changes of thalamic somatosensory information transfer. Brain Res 1999;829:77–89.