Anesthetics produce unconsciousness through an as yet unknown mechanism. Significant strides have been made in understanding the receptors at which some anesthetics act to produce their effects. For example, propofol and etomidate seem to affect the γ-aminobutyric acid type A receptor (GABAA).1–3 Unfortunately, less is known about the anatomical sites of action.
Some evidence suggests that the cerebral cortex, as compared with subcortical structures, is more sensitive to anesthetics. Velly et al.4 found that, at the point of loss of consciousness, propofol changed the surface electroencephalogram (EEG) (reflecting cortical activity) much more than the electrical activity recorded from subcortical electrodes. However, others have reported that propofol has a greater effect on the thalamus,5 whereas Alkire et al.6 found that propofol produced global depression, with only minor differences in metabolic depression at cortical versus subcortical sites. Nevertheless, there are significant similar findings among studies that have used different techniques and different species, especially in regard to anesthetic depression of neuronal function and excitability.7–9
We sought to investigate in an animal model the effects of propofol and etomidate on cortical, thalamic, and reticular formation neurons. We hypothesized that propofol and etomidate would depress cellular discharge activity (as measured by extracellular recordings) similarly in the cerebral cortex, thalamus, and reticular formation.
This study was approved by the animal care and use committee at the University of California, Davis, and the methods conformed to the humane care and use of laboratory animals as outlined by the National Academy of Sciences' Institute for Laboratory Animal Research. We studied 5 cats (3 male, 2 female; weight, 4–7 kg) that were bred for research purposes at the University of California, Davis. The cats first underwent surgery to implant a recording well and EEG electrodes as described below. The procedures were modifications of those used by Steriade et al.10 After recovery, the cats were studied on a chronic basis.
During isoflurane anesthesia, the head was secured to a stereotaxic frame (Model 1430; Kopf, Tujunga, CA) using ear bars and a mouthpiece. A custom-made head frame was attached to the stereotaxic frame; this head frame was used to secure bolts that would later be incorporated into a dental acrylic head cap (described below).
A craniotomy (approximately 1 cm diameter) was made. The center of the craniotomy was approximately 5 mm lateral to the midline and 5 mm anterior to the interaural line. Six to eight small stainless steel anchoring screws were inserted into the skull, 3 to 4 on each side, along the lateral aspects of the skull. A custom-made recording well was placed into the craniotomy. Five EEG electrodes were placed: 2 on each side of the skull overlying the frontal region of the brain, 2 overlying the occipital brain region, and the fifth electrode (ground) was inserted into the anterior aspect of the skull. The ends of the wires were fitted to a computer com-port. Stainless steel bolts were incorporated into the implant by surrounding the bolts with dental acrylic. These bolts were attached to the head frame.
When all surgical procedures were finished, 5-fluorouracil (5-FU, 1 mL of a 25 mg/mL solution; Sigma, St. Louis, MO) was placed on the dural surface in the recording well. The 5-FU acted as an antimitotic to prevent dural thickening.11 The cats were recovered and placed into a room with water and food.
Chronic Recording Sessions
Five to 7 days later, the cats were brought back into the laboratory and studied according to the methods of Steriade et al.10; the cats showed no signs of distress. The head was secured to the custom head frame of the stereotaxic frame using the steel bolts that had been incorporated into the dental acrylic. A 22-gauge IV catheter was inserted into a forelimb vein. The IV line was flushed with saline or Ringer lactate solution. The com-port that was part of the implant and contained the EEG wires was attached to a recording system consisting of amplifiers and a computer with recording software (see below).
A hydraulic microdrive was attached to an electrode micropositioner, which was attached to the stereotaxic frame. A tungsten microelectrode (approximately 10 MΩ; FHC, Bowdoinham, ME) was used to record single-unit activity. Extracellular action potentials were amplified (×5000) and filtered (10–3000 Hz) using a headstage and amplifier system (Tucker-Davis, Alachua, FL); the signals were recorded using Chart5 software (ADInstruments, Colorado Springs, CO) and Spike2 (Version 5.06; CED, Oxford, UK). The thalamus and reticular formation were targeted (see below for details) using the coordinates of the stereotaxic frame and the interaural and midline markers previously placed.
We recorded single-unit activity from areas 7, 18 and 19 of the cortex (coordinates: 4–8 mm lateral, 2–6 mm rostral to the interaural line); the ventral posterolateral and ventral posteromedial nuclei of the thalamus (coordinates: 7 mm lateral, 8 mm rostral to the interaural line, 16 mm inferior to the cortical surface); and the reticular formation at the region of the periaqueductal gray matter and medial geniculate body (coordinates: 4 mm lateral, 4 mm rostral to the interaural line, 20 mm inferior to the cortical surface).12,13 Most recorded neurons were within 1 mm of these coordinates. However, because the reticular formation extends across several millimeters of the brainstem, we recorded from reticular formation neurons as far caudal as −1 mm.
We usually recorded activity of 1 neuron at a time, although we often were able to record multiunit activity and were then able to later separate the action potentials from different neurons using spike discrimination software. Recording stability was assumed if the signal-to-noise ratio was >4:1 and the neuronal firing rate was either relatively constant or had recurring variability (i.e., was regularly irregular, waxing and waning in a recurring manner). When we had stable neuronal activity, we recorded continuously for at least 5 minutes, after which we administered either etomidate 0.4 to 0.5 mg/kg followed by 0.06 to 0.1 mg/kg/min or propofol 0.8 to 1 mg/kg followed by 0.1 to 0.2 mg/kg/min. After 10 minutes, the infusion was stopped; immediately after this point, the sedative state of the cat was assessed, including eye blink reflex, whisker reflex, and withdrawal to noxious tail pinching. The cat was permitted to recover while neuronal activity was recorded. If the neuronal activity was stable, then we administered the drug that had not been given previously (i.e., we gave etomidate if we had first given propofol) using the dose described above. In many cases, we searched for a new neuron because the previously studied neuron had been lost. Thus, the time between administration of one drug and another averaged 64 minutes (range, 24–218 minutes). In some cases, we lost the spike during the recovery period, in which case we searched for another neuron. If another neuron was found, then we obtained control neuronal activity for 3 to 5 minutes before giving the next anesthetic. If a neuron had been studied using both anesthetics, then we searched for another neuron, usually in a separate brain region (i.e., if we had studied a cortical neuron, we would then study a thalamic neuron). We generally studied only 2 to 3 neurons per session.
The frontal and occipital EEGs were recorded using an amplifier and software (Bio Amp, ADInstruments, Sydney, Australia; Chart5 and Spike2). The impedance was <5000 Ω. The EEG was digitized (1000 Hz) and stored on the computer hard drive for later analysis.
After the recording session was finished, the recording site was cleansed with Ringer lactate solution and dilute Betadine solution. Approximately every 1 to 2 weeks, 1 mL (25 mg/mL) 5-FU was placed in the well and left for 5 to 10 minutes to minimize dural growth. The cat was removed from the frame and bag and placed into a cage for transport back to the animal housing unit. We studied the cat again 1 to 3 days later, and continued to study the cats on an every 1- to 3-day basis for as long as 3 to 4 months, depending on the clinical course of each cat. In general, however, dural thickening was the limiting factor that prevented stable recordings. When we were unable to obtain stable recordings, the cats were euthanized with propofol followed by potassium chloride.
EEG data were filtered (0.5–50 Hz), subjected to Fourier transform, and the median edge frequency (MEF) and spectral edge frequency (SEF) were calculated, representing the frequencies below which 50% and 90% of the power resided. A spectral analysis was performed on 60-second segments of the EEG before and during propofol or etomidate anesthesia.
Action potentials were counted in the 3-minute period before drug administration and the 7- to 10-minute period immediately before discontinuing the drug. Recovery period data were analyzed for a 3-minute period, usually 25 to 30 minutes after discontinuing the drug. Action potentials were discriminated using the discriminator function in the Chart5 or Spike2 software. This process permitted us to analyze data when recording from 2 neurons simultaneously. We further analyzed cortical neuronal activity by identifying neurons that were fast spiking (FSNs) versus those that were regular spiking (RSNs). We used criteria according to Vijayan et al.,14 i.e., FSNs were identified by narrow spike width (<370 μm) and firing frequency ≥10 Hz whereas units with spike width >370 μm were classified as RSNs.
The interspike interval (ISI) was determined by measuring the time between action potentials. These time data were pooled for each anatomical location (cortex, thalamus, and reticular formation) and histograms were made for the control (awake) condition and drug exposure.
The EEG data (wakefulness versus anesthesia) were compared using a t test. Action potential data (firing rate) were log-transformed and analyzed using analysis of variance followed by Tukey post hoc test. A P value <0.05 was considered significant. Thalamic and reticular formation recording sites were reconstructed from electrolytic lesions made in the brain of a sixth cat anesthetized with isoflurane. The lesions were made using the coordinates listed above.
The etomidate and propofol infusions resulted in unconsciousness as demonstrated by decreased or absent eyelid and whisker reflexes; the animals withdrew or moved in response to noxious stimulation, indicating that the drug infusion rates used achieved plasma concentrations below that needed to produce immobility. Etomidate infusion was associated with spontaneous movements, including twitching. In addition, the infusions resulted in changing the EEG from a high-frequency, low-amplitude pattern to one with lower frequencies and greater amplitude (Fig. 1A). Quantitative analysis showed that the MEF and SEF shifted to lower frequencies (Fig. 1B). Compared with values during wakefulness, the SEF was lower during propofol infusion (P < 0.001; df = 143) and etomidate infusion (P < 0.001; df = 78). The MEF was lower during etomidate infusion (P < 0.001; df = 78) as compared with the MEF during wakefulness. Despite propofol and etomidate both producing similar clinical end points, etomidate caused more significant EEG changes, as shown by greater changes in the MEF and SEF (Fig. 1B). The MEF and SEF during etomidate anesthesia were lower than corresponding values during propofol anesthesia (P < 0.02; df = 221) and (P < 0.0001; df = 221), respectively.
Spectral analysis of the EEG showed small peaks at 12 Hz during wakefulness. During propofol infusion, there was a marked increase in power with a peak at 12 to 13 Hz, whereas during etomidate infusion, there were 2 dominant peaks: one at 12 to 14 Hz and another at 7 to 8 Hz (Fig. 2).
Cortical, thalamic, and reticular formation neurons were active during wakefulness but slowed during etomidate and propofol infusions (Figs. 3–6). Examples of the effects of propofol and etomidate on cortical and thalamic neurons are shown in Figure 3. Overall, the cortical firing rate decreased approximately 37% with etomidate (Figs. 4 and 5; P < 0.001; df = 25). Similarly, propofol depressed the cortical firing rate by 41% (Figs. 4 and 5; df = 50). Ten FSNs and 41 RSNs were studied during propofol anesthesia, whereas 8 FSNs and 18 RSNs were studied during etomidate anesthesia (Fig. 5). The FSNs had a firing frequency of 20.1 ± 2.3 Hz, whereas the RSNs fired at 11.1 ± 1.3 Hz. The FSNs and RSNs were similarly affected by propofol and etomidate administration. Firing frequencies of FSNs and RSNs during propofol infusion were depressed 34% and 44%, respectively, whereas during etomidate infusion, the firing frequencies were depressed 37% and 37%, respectively (Fig. 5). The resulting average firing frequencies of RSNs and FSNs during anesthesia roughly corresponded to the peak power frequencies of the EEG (Fig. 2).
Thalamic neurons were similarly affected by both drugs: 49% depression by etomidate (P < 0.01; df = 30) and 48% depression by propofol (Figs. 4 and 6; P < 0.01; df = 62). Reticular neurons were also depressed: 49% depression by etomidate (P < 0.01; df = 24) and 30% depression by propofol (Figs. 4 and 6; P < 0.01; df = 52).
Most neurons fired in an irregular pattern, with short ISIs interspersed between longer ISIs, resulting in ISI histograms with peaks in the shorter ISI times (Fig. 7). Because propofol and etomidate decreased the neuronal firing rate, the area under the curve of the ISI histograms (Fig. 7) was decreased as compared with the awake condition. In the thalamic neurons, the peaks at the shorter ISI times increased during propofol and etomidate administration. Reticular formation neurons had 2 peaks in the ISI histogram that decreased, but were otherwise preserved, during anesthesia (Fig. 7). Although most neurons responded to anesthesia with generalized slowing of an irregular firing rate, a small minority (approximately 10%) of cortical, thalamic, and reticular neurons began “bursting” during etomidate and propofol anesthesia (Fig. 8).
The cortical neurons were recorded at a depth of 1199 ± 106 μm. Reconstruction of recording sites revealed that recording of neurons in the thalamus was centered in the region of the ventral posterolateral and ventral posteromedial nuclei and medial geniculate body; reticular neurons were recorded in the midbrain region within and near the mesencephalic reticular nucleus and central tegmental field (Fig. 9).
The main findings of this study were that etomidate and propofol infusions that resulted in unconsciousness and decreased spontaneous neuronal activity in the cortex, thalamus, and reticular formation, coincident with a change in the EEG. We discuss these findings in relation to the effect of these anesthetics on global brain activity, as well as in vitro data on receptor systems.
Several recent reviews summarize the current state of knowledge regarding anesthetic mechanisms, especially related to specific clinical end points such as unconsciousness, amnesia, and immobility.7–9 The present study was focused on the first of these end points. The “state-of-the-art” body of information paints a picture of an emerging understanding of how anesthetics alter consciousness. The thalamus is clearly an important site of anesthetic action, and the present study, along with earlier and recent data, support the notion that anesthetics act in the thalamus to impede consciousness. Alkire et al.15,16 have shown that injection of antibodies directed against specific potassium channels or nicotine into the thalamus partially reverses or even awakens rats anesthetized with sevoflurane. Angel17–19 showed that propofol depressed centripetal transmission of sensory information by affecting cortico-inhibitory processes, in particular, depressing thalamic relay neurons. Ching et al.20 developed a thalamocortical model to show that propofol alters thalamocortical loops in part by action in the thalamus, resulting in synchronous α activity in the EEG. We presently found that both propofol and etomidate markedly increased EEG power in the α range, consistent with the data of Ching et al.20 and what is observed in humans during propofol anesthesia.21 Such synchronous activity within the cortex might prevent the thalamus from relaying information forward from the periphery to the cortex, thereby affecting consciousness. Indeed, we presently found that the firing frequencies of RSNs and FSNs during anesthesia were similar to the peak power frequencies on the EEG.
Etomidate and propofol act primarily at the GABAA receptor. Mutation of the β3 subunit results in mice that are resistant to both of these anesthetics.1 These drugs enhance current flow when GABA binds to the receptor, and presumably this increased current flow hyperpolarizes the cell membrane rendering the neuron more resistant to depolarizing influences.
Several investigators have used imaging techniques to probe brain sites where anesthetics might act. Alkire et al.6 reported that propofol produced depression of cortical and subcortical glucose metabolism as measured with positron emission tomography, although there was slightly greater depression of cortical structures. Some structures were more affected than others (e.g., left anterior cingulate and the inferior colliculus). Fiset et al.,5 using positron emission tomography, also found that propofol resulted in global depression. Similar to Alkire et al., they found regional differences. For example, Fiset et al.5 reported that propofol actions in the reticulothalamic system likely resulted in unconsciousness, although cortical metabolism was also significantly depressed.
Nelson et al.3 reported that propofol acted in the tuberomammillary nucleus to produce unconsciousness. The authors reached that conclusion based on the fact that injection of gabazine (a GABAA antagonist) into the tuberomammillary nucleus attenuated the effects of propofol. Others have reported that the mesopontine tegmental region is an important site of anesthetic action.22 Collectively, these data suggest that subcortical structures likely have an important role in anesthetic-induced unconsciousness.
Little work has been published on the effects of anesthesia on neuronal activity in intact animals. Hentschke et al.23 found that volatile anesthetics (isoflurane, enflurane, and halothane) decreased the cortical neuronal firing rate approximately 75% at the concentrations that produce unconsciousness and EEG changes (e.g., high-amplitude slow waves). Furthermore, these authors concluded that cortical neurons, compared with subcortical neurons, were more sensitive to inhaled anesthetics. The large depression described by Hentschke et al. was significantly more than what we presently observed with propofol and etomidate. We have found a similar depression with isoflurane (data not reported). We speculate that because isoflurane acts at multiple receptors, it can depress neurons both directly and indirectly via presynaptic mechanisms, for example, by decreasing synaptic release of neurotransmitters.24 Regardless of the mechanism, the disparate effects of isoflurane, propofol, and etomidate on cortical neuronal activity suggest that unconsciousness is not produced solely by a depressant effect on neuronal firing, as we would expect all anesthetics to produce similar amounts of depression. Steriade et al.25–27 studied EEG and neuronal activity during various anesthetic regimens (ketamine, urethane, nitrous oxide, barbiturates), finding waxing and waning EEG patterns and neuronal activity, including spindles and a slow EEG (<1 Hz) oscillation. The animals in those studies, however, seemed to be much more deeply anesthetized than the animals in the present study.
Another way that anesthetics might produce unconsciousness is to disrupt the normal firing pattern of neurons. This presently seemed to be the case, at least partially, because we found that propofol and etomidate caused some neurons to fire in a bursting pattern. This is akin to what occurs during the transition from wakefulness to sleep.10 Thus, slow wave sleep is similar in many ways to anesthetic-induced unconsciousness, with fluctuating neuronal firing being one characteristic.
In computo modeling of thalamocortical activity has provided some insight into how altered states of consciousness arise. Hill and Tononi28 found that sleep could be modeled by altering the potassium leak current in cortical and thalamic neurons. Using the same model, we found that we could depress the cortical firing rate approximately 35% to 45% using changes in GABA conductance that would likely occur with free etomidate concentrations that are consistent with loss of consciousness.29 Thus, our in vivo data that we presently report are similar to the in computo data.
Our dataset obviously represents a tiny fraction of neurons. Yet, single neurons can have significant impact on the behavior and function of the entire brain. So-called “grandmother” neurons respond to specific stimuli, suggesting that these neurons encode for those stimuli alone (although this is not universally accepted).30 Furthermore, during anesthesia, intracellular stimulation of a single cortical neuron can change global brain function in the intact rat.31
The present study has several limitations. We recorded spontaneous neuronal activity and did not specifically examine evoked activity, although the “spontaneous” activity likely had some evoked component to the extent that complete sensory deprivation is difficult to eliminate completely. The propofol and etomidate infusions seemed to achieve similar clinical end points; however, we did not rigorously test to determine whether subtle differences occurred. Thus, we cannot exclude the possibility that a different clinical end point was achieved with one drug compared with another, or across administrations of the same drug. Although FSNs are presumably inhibitory interneurons, and RSNs are primarily excitatory pyramidal neurons,14 we did not otherwise identify the function of the neurons that we studied, which further limits data interpretation. We cannot exclude an animal main effect because, in many of the comparisons, most of the data were obtained in 2 to 3 animals. However, it is not uncommon in neuroscience studies using animals in chronic conditions to report data from just a few animals.
We found that propofol and etomidate similarly depressed neuronal firing in the cortex, thalamus, and reticular formation. RSNs and FSNs were similarly affected. Peak power frequencies in the EEG during propofol and etomidate anesthesia corresponded with neuronal firing rates. These effects occurred at anesthetic doses that produced EEG changes consistent with unconsciousness. We conclude that disruption of neuronal activity is one possible mechanism by which anesthetics could produce unconsciousness.
Name: Jason Andrada, MS.
Contribution: This author helped collect and analyze data and write the manuscript.
Attestation: The author approved the manuscript.
Name: Preetha Livingston, BS.
Contribution: This author helped collect and analyze data.
Attestation: The author approved the manuscript.
Name: Bong Jae Lee, MD, PhD.
Contribution: This author helped collect data.
Attestation: The author approved the manuscript.
Name: Joseph Antognini, MD.
Contribution: This author helped collect and analyze data and write the manuscript.
Attestation: The author approved the manuscript.
This manuscript was handled by: Gregory J. Crosby, MD.
1. Jurd R, Arras M, Lambert S, Drexler B, Siegwart R, Crestani F, Zaugg M, Vogt KE, Ledermann B, Antkowiak B, Rudolph U. General anesthetic actions in vivo strongly attenuated by a point mutation in the GABA(A) receptor beta3 subunit. FASEB J 2003;17:250–2
2. Krasowski MD, Jenkins A, Flood P, Kung AY, Hopfinger AJ, Harrison NL. General anesthetic potencies of a series of propofol analogs correlate with potency for potentiation of gamma-aminobutyric acid (GABA) current at the GABA(A) receptor but not with lipid solubility. J Pharmacol Exp Ther 2001;297:338–51
3. Nelson LE, Guo TZ, Lu J, Saper CB, Franks NP, Maze M. The sedative component of anesthesia is mediated by GABA(A) receptors in an endogenous sleep pathway. Nat Neurosci 2002;5:979–84
4. Velly LJ, Rey MF, Bruder NJ, Gouvitsos FA, Witjas T, Regis JM, Peragut JC, Gouin FM. Differential dynamic of action on cortical and subcortical structures of anesthetic agents during induction of anesthesia. Anesthesiology 2007;107:202–12
5. 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
6. Alkire MT, Haier RJ, Barker SJ, Shah NK, Wu JC, Kao YJ. Cerebral metabolism during propofol anesthesia in humans studied with positron emission tomography. Anesthesiology 1995;82:393–403
7. Brown EN, Lydic R, Schiff ND. General anesthesia, sleep, and coma. N Engl J Med 2010;363:2638–50
8. Mashour GA, Orser BA, Avidan MS. Intraoperative awareness: from neurobiology to clinical practice. Anesthesiology 2011;114: 1218–33
9. Alkire MT, Hudetz AG, Tononi G. Consciousness and anesthesia. Science 2008;322:876–80
10. Steriade M, Timofeev I, Grenier F. Natural waking and sleep states: a view from inside neocortical neurons. J Neurophysiol 2001;85:1969–85
11. Spinks RL, Baker SN, Jackson A, Khaw PT, Lemon RN. Problem of dural scarring in recording from awake, behaving monkeys: a solution using 5-fluorouracil. J Neurophysiol 2003;90:1324–32
12. Berman AL. The Brain Stem of the Cat: A Cytoarchitectonic Atlas with Stereotaxic Coordinates. Madison, WI: University of Wisconsin Press, 1968
13. Berman AL, Jones EG The Thalamus and Basal Telencephalon of the Cat: A Cytoarchitectonic Atlas with Stereotaxic Coordinates. Madison, WI: University of Wisconsin Press, 1982
14. Vijayan S, Hale GJ, Moore CI, Brown EN, Wilson M. Activity in the barrel cortex during active behavior and sleep. J Neurophysiol 2010;103:2074–84
15. Alkire MT, McReynolds JR, Hahn EL, Trivedi AN. Thalamic microinjection of nicotine reverses sevoflurane-induced loss of righting reflex in the rat. Anesthesiology 2007;107:264–72
16. Alkire MT, Asher CD, Franciscus AM, Hahn EL. Thalamic microinfusion of antibody to a voltage-gated potassium channel restores consciousness during anesthesia. Anesthesiology 2009;110:766–73
17. Angel A, LeBeau F. A comparison of the effects of propofol with other anaesthetic agents on the centripetal transmission of sensory information. Gen Pharmacol 1992;23:945–63
18. Angel A. The G. L. Brown lecture: adventures in anaesthesia. Exp Physiol 1991;76:1–38
19. Angel A. Central neuronal pathways and the process of anaesthesia. Br J Anaesth 1993;71:148–63
20. Ching S, Cimenser A, Purdon PL, Brown EN, Kopell NJ. Thalamocortical model for a propofol-induced alpha-rhythm associated with loss of consciousness. Proc Natl Acad Sci USA 2010;107:22665–70
21. Feshchenko VA, Veselis RA, Reinsel RA. Propofol-induced alpha rhythm. Neuropsychobiology 2004;50:257–66
22. Devor M, Zalkind V. Reversible analgesia, atonia, and loss of consciousness on bilateral intracerebral microinjection of pentobarbital. Pain 2001;94:101–12
23. 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
24. Schlame M, Hemmings HC Jr. Inhibition by volatile anesthetics of endogenous glutamate release from synaptosomes by a presynaptic mechanism. Anesthesiology 1995;82:1406–16
25. Steriade M, Contreras D, Curro DR, Nunez A. The slow (<1 Hz) oscillation in reticular thalamic and thalamocortical neurons: scenario of sleep rhythm generation in interacting thalamic and neocortical networks. J Neurosci 1993;13:3284–99
26. Steriade M, Nunez A, Amzica F. Intracellular analysis of relations between the slow (<1 Hz) neocortical oscillation and other sleep rhythms of the electroencephalogram. J Neurosci 1993;13:3266–83
27. Steriade M, Nunez A, Amzica F. A novel slow (<1 Hz) oscillation of neocortical neurons in vivo: depolarizing and hyperpolarizing components. J Neurosci 1993;13:3252–65
28. Hill S, Tononi G. Modeling sleep and wakefulness in the thalamocortical system. J Neurophysiol 2005;93:1671–98
29. Talavera JA, Esser SK, Amzica F, Hill S, Antognini JF. Modeling the GABAergic action of etomidate on the thalamocortical system. Anesth Analg 2009;108:160–7
30. Gelbard-Sagiv H, Mukamel R, Harel M, Malach R, Fried I. Internally generated reactivation of single neurons in human hippocampus during free recall. Science 2008;322:96–101
31. Li CY, Poo MM, Dan Y. Burst spiking of a single cortical neuron modifies global brain state. Science 2009;324:643–6