Portal de Periódicos da CAPES

In vitro networks: cortical mechanisms of anaesthetic action

2002; Elsevier BV; Volume: 89; Issue: 1 Linguagem: Inglês

10.1093/bja/aef154

1471-6771

B. Antkowiak,

Photoreceptor and optogenetics research

Brain slices are well-established tools in neuroscience research. In the last decade scientists succeeded in isolating viable brain slices from many different regions of the mammalian central nervous system. The brain slice preparation was introduced by Henry MacIlwain in the 1950s.40Li CH McIlwain H Maintenance of resting membrane potentials in slices of mammalian cerebral cortex and other tissues in vitro.J Physiol. 1957; 139: 178-190Crossref PubMed Scopus (80) Google Scholar 45McIlwain H Buchel L Cheshire JD The inorganic phosphate and phosphocreatine of brain especially during metabolism in vitro.Biochem J. 1951; 48: 12-20Crossref PubMed Scopus (35) Google Scholar 46McIlwain H Buddle HL Techniques in tissue metabolism. I. A mechanical chopper.Biochem J. 1953; 53: 412-420Crossref PubMed Google Scholar Milestones in the development of this technique are listed in Table 1. Brain slices can be viewed as intact local microcircuits, lacking synaptic inputs from other parts of the central nervous system. In using this type of preparation, drug actions on specific types of neurones have been investigated on the network, cellular, and molecular level. Studies on brain slices provided interesting insights into the mechanisms by which general anaesthetics affect cortical neurones. In the last decade, brain slices also proved to be helpful for analysing patterns of neuronal activity, observed during the anaesthetic state in experimental animals. There is increasing interest in uncovering the contribution of specific local microcircuits to the overall changes in neuronal activity evoked by general anaesthetics. Electroencephalographic (EEG) recordings have shown that anaesthetic agents dramatically alter the firing mode of cortical neurones, even when applied in a range of rather low concentrations.20Clark DL Rosner BS Neurophysiologic effects on general anesthetics: I. The electroencephalogram and sensory evoked responses in man.Anesthesiology. 1973; 38: 564-582Crossref PubMed Scopus (238) Google Scholar 68Thornton C Evoked potentials in anaesthesia.Eur J Anaesth. 1991; 8: 89-107PubMed Google Scholar As well as compound-specific actions, almost each anaesthetic that has been investigated transfers high-frequency low-voltage EEG activity, present during the awake state and rapid eye movement (REM) sleep, into low-frequency high-voltage activity. The latter is commonly taken as evidence for the presence of a hypnotic, delta sleep-like state. However, interpretation of EEG recordings turned out to be rather difficult, as they provide a complex correlate of the summed synaptic activity of neurones located in the upper layers of the cerebral cortex.53Pulvermüller F Birbaumer N Lutzenberger W et al.High-frequency brain activity: its possible role in attention, perception and language processing.Prog Neurobiol. 1997; 52: 445Crossref Scopus (179) Google Scholar As yet, the physiological mechanisms that underlie EEG activity during anaesthesia remain largely unknown. Recent imaging studies on human subjects demonstrated that, besides altering synchronized activity, some general anaesthetics strongly depress cortical metabolism and blood flow.1Alkire MT Quantitative EEG correlations with brain glucose metabolic rate during anesthesia in volunteers.Anesthesiology. 1998; 89: 323-333Crossref PubMed Scopus (134) Google Scholar 2Alkire MT Haier RJ Barker SJ et al.Cerebral metabolism during propofol anesthesia in humans studied with positron emission tomography.Anesthesiology. 1995; 82: 393-403Crossref PubMed Scopus (261) Google Scholar 4Alkire MT Haier RJ Shah NK et al.Positron emission tomography study of regional cerebral metabolism in humans during isoflurane anesthesia.Anesthesiology. 1997; 86: 549-557Crossref PubMed Scopus (169) Google Scholar 5Alkire MT Pomfrett CJD Haier RJ et al.Functional brain imaging during anesthesia in humans. Effects of halothane on global and regional glucose metabolism.Anesthesiology. 1999; 90: 701-709Crossref PubMed Scopus (157) Google Scholar 24Fiset P Paus T Daloze T et al.Brain mechanisms of propofol-induced loss of consciousness in humans: a positron emission tomographic study.J Neurosci. 1999; 19: 5506-5513Crossref PubMed Google Scholar Again, the physiological processes underlying these phenomena remain to be elucidated. In this article, anaesthetic actions in brain slices are discussed. Experimental findings will be related to recent work on living animals and human subjects.Table 1Development of brain slice methodsDateDiscoveryReferences1951Preparation of viable slices45McIlwain H Buchel L Cheshire JD The inorganic phosphate and phosphocreatine of brain especially during metabolism in vitro.Biochem J. 1951; 48: 12-20Crossref PubMed Scopus (35) Google Scholar, 46McIlwain H Buddle HL Techniques in tissue metabolism. I. A mechanical chopper.Biochem J. 1953; 53: 412-420Crossref PubMed Google Scholar1957First intracellular recording40Li CH McIlwain H Maintenance of resting membrane potentials in slices of mammalian cerebral cortex and other tissues in vitro.J Physiol. 1957; 139: 178-190Crossref PubMed Scopus (80) Google Scholar1966First study on anaesthetic modulation of electrical activity in brain slices75Yamamoto C McIlwain H Electrical activities in thin sections from the mammalian brain maintained in chemically-defined media in vitro.J Neurochem. 1966; 13: 1333-1343Crossref PubMed Scopus (248) Google Scholar1981Organotypic brain slice cultures28Ga¨hwiler BH Organotypic monolayer cultures of nervous tissue.J Neurosci Methods. 1981; 4: 329-342Crossref PubMed Scopus (731) Google Scholar, 29Ga¨hwiler BH Capogna M Debanne D et al.Organotypic slice cultures: a technique has come of age.Trends Neurosci. 1997; 20: 471-477Abstract Full Text Full Text PDF PubMed Scopus (709) Google Scholar1989Blind patch-clamping in brain slices18Blanton MG Lo Turco JJ Kriegstein AR Whole cell recording from neurons in slices of reptilian and mammalian cerebral cortex.J Neurosci Methods. 1989; 30: 2103-2110Crossref Scopus (815) Google Scholar1993Patch clamping on visually identified neurones22Dodt HU Zieglga¨nsberger W Infrared videomicroscopy: a new look at neuronal structure and function.Trends Neurosci. 1994; 17: 453-458Abstract Full Text PDF PubMed Scopus (93) Google Scholar, 65Stuart GJ Dodt HU Sakmann B Patch-clamp recordings from the soma and dendrites of neurons in brain slices using infrared video microscopy.Pflügers Arch. 1993; 423: 511-518Crossref PubMed Scopus (633) Google Scholar Open table in a new tab What happens in the cerebral cortex when anaesthesia is commenced and patients lose consciousness? A considerable number of studies, making use of different methodological approaches, have tackled this question. For example, it has been shown that in lightly anaesthetized subjects the metabolism of most cortical areas is decreased by roughly 50%.1Alkire MT Quantitative EEG correlations with brain glucose metabolic rate during anesthesia in volunteers.Anesthesiology. 1998; 89: 323-333Crossref PubMed Scopus (134) Google Scholar 2Alkire MT Haier RJ Barker SJ et al.Cerebral metabolism during propofol anesthesia in humans studied with positron emission tomography.Anesthesiology. 1995; 82: 393-403Crossref PubMed Scopus (261) Google Scholar 4Alkire MT Haier RJ Shah NK et al.Positron emission tomography study of regional cerebral metabolism in humans during isoflurane anesthesia.Anesthesiology. 1997; 86: 549-557Crossref PubMed Scopus (169) Google Scholar 5Alkire MT Pomfrett CJD Haier RJ et al.Functional brain imaging during anesthesia in humans. Effects of halothane on global and regional glucose metabolism.Anesthesiology. 1999; 90: 701-709Crossref PubMed Scopus (157) Google Scholar 24Fiset P Paus T Daloze T et al.Brain mechanisms of propofol-induced loss of consciousness in humans: a positron emission tomographic study.J Neurosci. 1999; 19: 5506-5513Crossref PubMed Google Scholar Furthermore, the transition from wakefulness to unresponsiveness is accompanied by a depression of high-frequency electrical activity in the gamma band.69Uchida S Nakayama H Maehara T et al.Suppression of gamma activity in the human medial temporal lobe by sevoflurane anesthesia.Neuroreport. 2000; 11: 39-42Crossref PubMed Scopus (32) Google Scholar Low-frequency high-voltage activity becomes more pronounced as anaesthesia is deepened.20Clark DL Rosner BS Neurophysiologic effects on general anesthetics: I. The electroencephalogram and sensory evoked responses in man.Anesthesiology. 1973; 38: 564-582Crossref PubMed Scopus (238) Google Scholar As with many anaesthetics, unresponsiveness occurs at drug concentrations far below those abolishing painful stimuli-induced movements, some aspects of information processing in the cerebral cortex seem to be rather sensitive to anaesthetic treatment. How can this be explained? Do cortical cells themselves possess molecular targets, mediating neuronal depression and hypnosis? A number of arguments indeed support this view. There is now broad agreement that GABAA and NMDA receptors are important sites of general anaesthetic actions.7Antkowiak B The molecular basis of general anaesthesia.Naturwissenschaften. 2000; 88: 201-213Crossref Scopus (77) Google Scholar 16Belelli D Pistis M Peters JA et al.General anaesthetic action at transmitter-gated inhibitory amino acid receptors.Trends Pharmacol Sci. 1999; 20: 496-502Abstract Full Text Full Text PDF PubMed Scopus (128) Google Scholar 25Franks NP Lieb WR Molecular and cellular mechanisms of general anaesthesia.Nature. 1994; 367: 607-614Crossref PubMed Scopus (1590) Google Scholar 33Harris RA Mihic SJ Dildy-Mayfield JE et al.Actions of anesthetics on ligand-gated ion channels: role of receptor subunit composition.FASEB J. 1995; 9: 1454-1462Crossref PubMed Google Scholar 67Tanelian DL Kosek P Mody I et al.The role of the GABAA receptor/chloride channel complex in anesthesia.Anesthesiology. 1993; 78: 757-776Crossref PubMed Scopus (304) Google Scholar The density of these receptors in the cerebral cortex is higher than in most other parts of the central nervous system. Furthermore, a role of cortical GABAA receptors in hypnosis is also suggested by the action of sedative drugs. Benzodiazepines are well known to cause their therapeutic effects by enhancing GABAA receptor-function in a rather specific manner. Recently, evidence has been provided that benzodiazepine-induced sedation in mice is mediated via a GABAA receptor-subtype largely restricted to the cerebral cortex.47McKernan MG Rosahl TW Reynolds DS et al.Sedative but not anxiolytic properties of benzodiazepines are mediated by the GABAA receptor α1 subtype.Nature Neurosci. 2000; 3: 587-592Crossref PubMed Scopus (809) Google Scholar 57Rudolph U Crestani F Benke D et al.Benzodiazepine actions mediated by specific γ-aminobutyric acidA receptor subtypes.Nature. 1999; 401: 796-800Crossref PubMed Scopus (1009) Google Scholar Taken together, these findings indicate that hypnosis is to a substantial degree caused by drug actions on neurones located in the cortex. However, in order to explain the hypnotic aspect of the anaesthetic state, many textbooks and review articles refer to a different theory, emphasizing similarities between anaesthesia and natural sleep. Pronounced high-voltage low-frequency EEG activity is present during the non-REM stages of natural sleep. The pioneering experiments of Moruzzi and Magoun demonstrated that the states of wakefulness and sleep as well as their electrical correlates are influenced by brain stem nuclei.48Moruzzi G Magoun HW Brain stem reticular formation and activation of the EEG.Electroenceph Clin Neurophysiol. 1949; 1: 455-473Abstract Full Text PDF PubMed Scopus (2332) Google Scholar These sub-cortical structures also might be potential sites for general anaesthetic actions.10Antognini JF Carstens E Isoflurane blunts electro encephalographic and thalamic-reticular formation responses to noxious stimulation in goats.Anesthesiology. 1999; 91: 1770-1779Crossref PubMed Scopus (34) Google Scholar 21Detsch O Vahle-Hinz C Kochs E et al.Isoflurane induces dose-dependent changes of thalamic somatosensory information transfer.Brain Res. 1999; 829: 77-89Crossref PubMed Scopus (72) Google Scholar 37Keifer JC Baghdoyan HA Lydic R Pontine cholinergic mechanisms modulate the cortical electroencephalographic spindles of halothane anesthesia.Anesthesiology. 1996; 84: 945-954Crossref PubMed Scopus (66) Google Scholar 50Nagi SH Cheney DL Finck AD Acetylcholine concentrations and turnover in rat brain structures during anesthesia with halothane, enflurane, and ketamine.Anesthesiology. 1978; 48: 4-10Crossref PubMed Scopus (35) Google Scholar 55Ries CR Puil E Mechanism of anesthesia revealed by shunting actions of isoflurane on thalamocortical neurons.J Neurophysiol. 1999; 81: 1795-1999Crossref PubMed Scopus (99) Google Scholar It may be that general anaesthetics create a brain state similar to delta sleep by modulating arousal systems.3Alkire MT Haier RJ Fallon JH Toward a unified theory of narcosis: brain imaging evidence for a thalamocortical switch as the neurophysiologic basis of anesthetic-induced uncon sciousness.Conscious Cogn. 2000; 9: 370-386Crossref PubMed Scopus (309) Google Scholar 19Cariani P Anesthesia, neural information processing, and conscious awareness.Conscious Cogn. 2001; 9: 387-395Crossref Scopus (27) Google Scholar 49Munglani R Jones GJ Altered consciousness: pharmacology and phenomenology. BAP summer meeting, York, July 1991. Sleep and general anaesthesia as altered states of consciousness.J Psychopharmacol. 1992; 6: 399-409Crossref PubMed Scopus (3) Google Scholar In summary, there are good reasons for assuming that anaesthetic-induced synchronization and depression of neuronal activity in the cerebral cortex involve sub-cortical arousal systems. However, as discussed above, we are also faced with data strongly supporting the possibility that direct actions on cortical neurones come into play. How do we evaluate the importance of different potential pathways of drug action? How do we establish a more precise hypothesis that distinguishes between cortical and sub-cortical mechanisms? What type of experiments can be conducted to elucidate the specific contributions made by different local networks? A major problem arises from the fact that distributed local microcircuits are extensively interconnected in intact brains: if we assume that a particular part of the brain, say the neocortex (termed network B in the following) receives excitatory input from a sub-cortical structure (network A), neuronal activity in B may be reduced either because of direct effects on drug targets located in B, or because of a decrease in excitatory drive provided by A. Thus, it is necessary to study drug effects in isolated local networks, which do not receive synaptic input from different brain areas. If it turns out that network B is equally sensitive to drug treatment, regardless of whether synaptic input provided by A is present or not, B is most probably the substrate of drug action. If B is rendered insensitive by removing synaptic input provided by A, the more important effects take place in network A. At this point, the topics addressed above have been considered in order to establish some type of conceptual framework. Such a framework seems to be helpful for recognizing the benefit of brain slice studies and for integrating the results into a general understanding of how anaesthetics work. The research discussed in the following is centred on work dealing with the question of how general anaesthetics alter activity patterns in hippocampal and neocortical brain slices. The focus is on drug actions occurring at the network level. It is commonly assumed that anaesthetic-induced unconsciousness is causally related to changes in the firing characteristics of cortical neurones. Thus, monitoring the activity patterns of these cells takes a central place in understanding the physiological processes related to the hypnotic state. Today, different methods are in use. Some basic information concerning measurements of electrical brain activity is summarized in Figure 1. In the following I will consider what particular aspect of neuronal activity can be obtained by the specific methodological approaches. In the 1920s Hans Berger succeeded in recording small, time-varying changes in the electrical potential from electrodes fixed on the scalp of human subjects.17Berger H U¨ber das Elektroenkephalogramm des Menschen.Arch Psychiatr Nervenkr. 1929; 87: 527-570Crossref Scopus (2121) Google Scholar These spontaneous fluctuations were caused by the activity of synapses located on dendrites in the upper layer of the neocortex. Berger reported that specific patterns of bioelectrical activity were closely related to specific brain states: oscillatory activity in the alpha range (9–12 Hz) dominates during eye closure, whereas low-frequency high-voltage activity developed as soon as the subject falls asleep. The frequency bands commonly used to classify EEG activity are listed in Table 2. Many neurones contribute to the voltage changes detected during EEG recordings. A realistic number is hard to estimate.53Pulvermüller F Birbaumer N Lutzenberger W et al.High-frequency brain activity: its possible role in attention, perception and language processing.Prog Neurobiol. 1997; 52: 445Crossref Scopus (179) Google Scholar The bioelectrical signal monitored by electrodes placed on the scalp is heavily low-pass filtered. Another drawback of EEG recordings is the fact that the shapes of the voltage traces largely depend on the placement of the reference electrode. In conclusion, EEG recordings are based on synchronized synaptic activity of many neurones in the neocortex. Information about average firing rates cannot be derived from such data. The EEG provides a good temporal but poor spatial resolution.Table 2Frequency bands for synchronized cortical activityFrequency bandCycles per secondGamma>31Beta13–30Alpha9–12Theta4–8Delta0.5–3 Open table in a new tab In animal and brain slice studies, bioelectrical activity generated by neocortical neurones is frequently recorded by electrodes with tiny tips, positioned extracellularly within the cortical layers. Using this approach, synaptic and action potential activity can be detected simultaneously. For separating these different correlates of neuronal information processing, filter techniques are applied.35Hentschke H Antkowiak B NMDA receptor-mediated changes of spontaneous activity patterns in thalamocortical slice cultures.Brain Res. 1999; 830: 123-137Crossref PubMed Scopus (6) Google Scholar Extracellularly recorded action potentials are short events, lasting usually 0.5–2 ms. Thus, band-pass filtering the signal at 100–5000 Hz does not affect the shape of fast action potentials whereas slower voltage fluctuations, indicative of synaptic activity, are removed. The filtered signal either displays multi-unit activity (it is composed of action potentials produced by several different cells located in the vicinity of the electrode) or single-unit activity (fast voltage deflections are generated by a single neurone). Band-pass filtering the recordings at 0.1–100 Hz removes action potentials: the residual voltage changes display correlated synaptic activity. This type of signal is commonly termed the local field potential (LFP) or the micro-EEG. However, in contrast to the EEG recorded from humans and animals, the LFP integrates the synaptic activity over much fewer neurones: thus the spatial resolution is far better. How are the LFP and action potential firing related in the time domain? If oscillatory synaptic activity, for example in the gamma or delta range is present in the LFP, it is most commonly also evident in multi-unit firing. If action potential firing and LFPs are recorded simultaneously, a stable-phase relationship can be observed in most cases. However, LFP recordings do not provide information about changes in the average firing rates of cortical neurones. In the past few years, imaging studies have shown that at light levels of anaesthesia, just sufficient to render human subjects unresponsive, cortical metabolism and blood flow are depressed by as much as 30–50% compared with the awake state. Quantitatively, these findings fit well with recent recordings of the mean firing rates in rats.26Gaese BH Ostwald J Anesthesia changes frequency tuning of neurons in the rat primary auditory cortex.J Neurophysiol. 2001; 86: 1062-1066Crossref PubMed Scopus (136) Google Scholar Taken together, it seems that the depression of cortical metabolism induced by anaesthetic agents is related to a strong decrease in the mean firing rates and synaptic activity. Because of the poor time resolution, imaging studies neither provide information about the dominant rhythms nor about the degree of synchronization characterizing neuronal activity during the awake or anaesthetized state. Most anaesthetic agents alter the EEG activity patterns in a similar, although not identical way. The study of MacIver and collaborators is a good example for illustrating commonly observed hallmarks of anaesthetic effects, occurring in a range of clinically relevant concentrations.41Lukatch HS MacIver MB Synaptic mechanisms of thiopental-induced alterations in synchronized cortical activity.Anesthesiology. 1996; 84: 1425-1434Crossref PubMed Scopus (60) Google Scholar 42MacIver M Mandema JW Stanski DR et al.Thiopental uncouples hippocampal and cortical synchronized electroencephalographic activity.Anesthesiology. 1996; 84: 1411-1424Crossref PubMed Scopus (55) Google Scholar The key results are summarized in Table 3. At different thiopental concentrations, the authors correlated the behavioural state in the rat with the corresponding EEG pattern and the molecular mechanisms of anaesthetic action. They observed increased power in all frequency bands of the EEG (a pattern termed activation) during sedation, delta oscillations during hypnosis, and a pattern termed burst suppression during anaesthesia. Burst suppression is characterized by episodes of low-frequency high-voltage activity, which are separated by periods of neuronal silence. Very similar concentration-dependent effects appear with a number of barbiturates and volatile anaesthetics in animals and humans. The finding that general anaesthetics induce delta activity in the EEG has prompted the suggestion that the brain states of anaesthesia and natural sleep are comparable with regard to several aspects. This hypothesis is mainly based on the fact that synchronized cortical delta activity is present during non-REM sleep. A number of studies on sleep mechanisms provided evidence that synchronized cortical delta activity originates in the thalamus. During delta sleep, cortical neurones seem to be driven and synchronized by thalamic relay neurones.44McCormick DA Bal T Sleep and arousal: thalamocortical mechanisms.Annu Rev Neurosci. 1997; 20: 185-215Crossref PubMed Scopus (956) Google Scholar In this sleep state, thalamic relay cells fire rhythmic bursts of fast sodium action potentials, as indicated in Figure 2. Ca2+ currents play a major role in generating the rhythm.64Steriade M McCormick DA Sejnowski TJ Thalamocortical oscillations in the sleeping and aroused brain.Science. 1993; 262: 679-685Crossref PubMed Scopus (2590) Google Scholar In the awake state and during REM sleep, Ca2+ channels remain inactivated, as relay neurones receive a strong depolarizing synaptic input originating in brain stem nuclei. Some of the latter are involved in arousal and in controlling the sleep–wake cycle. With a view on the mechanism thought to underlie natural sleep, cortical delta rhythms during anaesthesia could be explained by a depression of neuronal activity at the sub-cortical level. Assuming that general anaesthetics reduce the excitability of thalamic relay cells either by direct actions on these neurones or by reducing their synaptic input, it seems possible that the cells are forced to enter the burst mode, thereby producing synchronized delta activity in the thalamo-cortical loop. In this scenario, anaesthetic-induced delta activity is largely related to direct or indirect drug actions on thalamic relay neurones.Table 3Stages of thiopental anaesthesia in the rat. The table summarizes results published by MacIver and colleagues41 42BehaviourEEG patternPutative molecular mechanismBlood concentration (μg ml−1)Awake: explorative behaviourDesynchronized0Sedation: rats are sedated and lose their righting reflexIncrease in frequency and amplitude?15–20Hypnosis: no response to moderate sensory stimuli like loud sounds and whisker strokes. Tail pinch reflex is lostSlow wave delta activity dominatesGABAA receptor: potentiation of agonist40Anaesthesia: corneal reflex is lostBurst suppressionGABAA receptor: tonic inhibition50Intubation: motor responses to intubation are blockedIsoelectric activityDepression of glutamatergic synaptic transmission70 Open table in a new tab Although the explanation outlined so far sounds quite convincing, MacIver and co-workers have shown that this might only be a part of the story.41Lukatch HS MacIver MB Synaptic mechanisms of thiopental-induced alterations in synchronized cortical activity.Anesthesiology. 1996; 84: 1425-1434Crossref PubMed Scopus (60) Google Scholar 42MacIver M Mandema JW Stanski DR et al.Thiopental uncouples hippocampal and cortical synchronized electroencephalographic activity.Anesthesiology. 1996; 84: 1411-1424Crossref PubMed Scopus (55) Google Scholar Their data, obtained from neocortical and hippocampal brain slices, provide evidence that anaesthetic-induced delta activity may also be related to a cortical site of action. By recording LFPs in neocortical slices, the authors observed synchronized oscillatory network activity in the theta range under drug-free conditions. Upon thiopental treatment, theta oscillations slowed, thus shifting the dominant frequency towards the delta range. At the time these investigations were published, it was uncertain how they relate to humans as theta activity is very difficult to detect in the human EEG. Because pronounced theta activity is well known to occur in the hippocampus of rodents, two questions have been debated: does theta activity also occur in the neocortex? And, is it specific for rodents? Surprisingly, these questions have been answered to a large extent in recent work on human subjects.54Raghavachari S Kahana MJ Rizzuto DS et al.Gating of human theta oscillations by a working memory task.J Neurosci. 2001; 21: 3175-3183Crossref PubMed Google Scholar Recordings were carried out with electrodes placed in direct contact with the surface of the neocortex. The authors discovered prominent theta rhythms, closely correlated to cognitive performance, especially working memory tasks. The study also has shown that theta activity is very difficult to detect by EEG recordings, as oscillatory activity is phase-shifted at different locations and thus is not apparent with electrodes with a poor spatial resolution. In rat neocortical slices a burst suppression-like pattern can be observed at thiopental concentrations somewhat higher than those shifting theta oscillations towards the delta range. This finding indicates that burst suppression involves a cortical site of action. In fact, recordings in the neocortex in experimental animals, before and after removing the ascending inputs by cutting along the white matter, demonstrated that burst suppression survives this procedure.34Henry CE Scoville WB Supression-burst activity from isolated cerebral cortex in man.Electroenceph Clin Neurophyiol. 1952; 4: 1-22Abstract Full Text PDF PubMed Scopus (96) Google Scholar 66Swank RL Synchronization of spontaneous electrical activity of cerebrum by barbiturate narcosis.J Neurophysiol. 1949; 12: 161-172PubMed Google Scholar Anaesthetic concentrations exceeding those causing burst suppression fully depress neuronal activity in the cortex: in vivo, an isoelectric EEG is seen during this state. In vitro, neither spontaneous changes in the LFP nor action potential activity can be observed. What are the molecular mechanisms underlying the different stages of anaesthesia described above? Do different patterns of network activity result from anaesthetic modulation at a single molecular locus? Or, alternatively, does a multi-site theory of anaesthetic action provide a better explanation of the experimental data mentioned so far? MacIver and co-workers report that at 10 μmol litre−1 thiopental the period of spontaneous theta oscillations is increased by about 2- to 3-fold.41Lukatch HS MacIver MB Synaptic mechanisms of thiopental-induced alterations in synchronized cortical activity.Anesthesiology. 1996; 84: 1425-1434Crossref PubMed Scopus (60) Google Scholar 42MacIver M Mandema JW Stanski DR et al.Thiopental uncouples hippocampal and cortical synchronized electroencephalographic activity.Anesthesiology. 1996; 84: 1411-1424Crossref PubMed Scopus (55) Google Scholar The same concentration alters the kinetic properties of GABAA receptor-mediated synaptic events, thus causing a 3-fold increase in inhibitory postsynaptic current (IPSC) decay time, leaving the amplitude of spontaneous IPSCs unchanged. At anaesthetic concentrations causing burst suppression, besides the effects on IPSC decay time, neurones were tonically inhibited. This inhibition most probably resulted from direct activation of GABAA receptors, which was consistent with the finding that the GABAA receptor-agonist muscimol induced a similar pattern of network activity. The suggestion that lengthening of IPSC decays vs tonic inhibition produce different effects on the network level has been supported