Recent progress in brain imaging offers great potential for anaesthesia research.46Menon DK Mapping the anatomy of unconsciousness'imaging anaesthetic action in the brain.Br J Anaesth. 2001; 86: 607-610Abstract Full Text Full Text PDF PubMed Scopus (10) Google Scholar Among the most important imaging modalities for mapping human brain activation are positron emission tomography (PET) and functional magnetic resonance imaging (fMRI). Both techniques are based on mapping haemodynamic or metabolic changes that are a direct consequence of alterations in neuronal activity.54Raichle M Behind the scenes of functional brain imaging: a historical and physiological perspective.Proc Natl Acad Sci USA. 1998; 95: 765-772Crossref PubMed Scopus (490) Google Scholar 59Villringer A Dirnagel U Coupling of brain activity and cerebral blood flow: basis of functional neuroimaging.Cerebrovasc Brain Metab Rev. 1995; 7: 240-276PubMed Google Scholar In comparison with electroencephalographic techniques, which are able to demonstrate that a given anaesthetic agent has central nervous system activity, PET or fMRI provide information about the pathways and the anatomical localization of the drug effect. On one hand, fMRI offers a superior spatial resolution to PET. The theoretical limit for PET is expected to be 2 mm isotropic resolution for the human head,17Budinger TF PET instrumentation: what are the limits?.Semin Nucl Med. 1998; 28: 247-267Abstract Full Text PDF PubMed Scopus (78) Google Scholar whereas a recent fMRI study has demonstrated an in-plane resolution of 0.55 mm.28Goodyear BG Menon RS Brief visual stimulation allows mapping of ocular dominance in visual cortex using fMRI.Hum Brain Mapp. 2001; 14: 210-217Crossref PubMed Scopus (120) Google Scholar Furthermore, fMRI does not require the injection of exogenous radioactive tracers. Therefore, a subject can perform a variety of different tasks during various experimental conditions within the same imaging session. Besides these methodological factors, fMRI is of advantage in terms of lower costs and clinical availability. On the other hand, PET offers the unique possibility of in vivo receptor imaging, only shared in part with magnetic resonance spectroscopy, and it provides absolute values of physiological variables. Using the complementary features of both techniques allows us to obtain information about how the brain works from the molecular to the complex neural network level61Volkow ND Rosen B Farde L Imaging the living human brain: magnetic resonance imaging and positron emission tomography.Proc Natl Acad Sci USA. 1997; 94: 2787-2788Crossref PubMed Scopus (64) Google Scholar and to explore regional drug effects on integrated brain processes at these various levels. Both imaging modalities have the ability to reveal new insights in altered brain processes during anaesthesia and may contribute to our understanding of where, why, and how brain functions collapse in the presence of anaesthetic drugs. This review will focus mainly on previously reported results obtained with human brain imaging techniques related to the general mechanism of anaesthetic action. It will first discuss some methodological aspects of PET and fMRI and then comment on results obtained mostly from volunteer studies. PET enables non-invasive measurements of physiological and biochemical processes in any part of the human body. PET is based on the i.v. injection of a radioactive tracer that typically consists of endogenous molecules labelled with positron emitting isotopes. In the tissues, these positrons are wiped out by electrons causing the emission of two collinear gamma rays. These events are detected by PET cameras and subsequent data processing provides three-dimensional images indicating the location and density of the positron sources. Depending on the biochemical and biophysical properties of the tracer, PET can be used to map a variety of different biochemical processes and physiological variables. In cognitive neuroscience, functional brain imaging with PET is often used to map changes in regional cerebral blood flow (CBF) or cerebral metabolic rate of glucose (CMRGlu) caused by well-defined stimuli or tasks. These studies are usually called activation studies. The analysis of the measured data is based on statistical comparison of the measurements obtained during different conditions (e.g. performing a task vs resting) reveals regional stimulus or task-related changes in CBF or CMRGlu. In contrast to these cognitive activation studies, PET studies of anaesthetic action in the human brain are often based on comparing physiological variables measured under different states of anaesthesia. Regional alterations in CBF and CMRGlu are assumed to reflect changes in regional neural activity and may provide neuroanatomical evidence for behavioural responses associated with sedation and anaesthesia. The theoretical basis for regional (r)CBF measurements with PET is the work by Kety,38Kety SS The theory and applications of the exchange of inert gas at the lungs and tissues.Pharmacol Rev. 1951; 3: 1-41PubMed Google Scholar 39Kety SS Measurements of local cerebral flow by the exchange of an inert, diffusible substance.Methods Med Res. 1960; 8: 228-236Google Scholar which allows assessment of rCBF in laboratory animals. The animal tissue autoradiographical methods of rCBF measurements are adapted for use with PET.35Herscovitch P Markham J Raichle ME Brain blood flow measured with intravenous H215O. I. Theory and error analysis.J Nuclear Med. 1983; 24: 782-789PubMed Google Scholar In human studies, the tracer of choice for CBF mapping is 15O-labelled water (H215O).36Herscovitch P Can [15O]water be used to evaluate drugs?.J Clin Pharmacol. 2001; 41: 11-20Crossref Google Scholar This molecule can diffuse freely across the blood–brain barrier. Because of its short half-time (2 min) the tracer allows the performance of repeated measurements in a single study session and the use of a variety of experimental designs. After injection of a single dose of H215O PET images are immediately acquired to obtain the flow-dependent accumulation and disappearance of the radiotracer and to calculate maps of rCBF. PET measurements of CMRGlu in man are based usually on [18F]fluorodeoxyglucose (18FDG).2Aine JC A conceptual overview and critique of functional neuroimaging techniques in humans: 1. MRI/fMRI and PET.Crit Rev Neurobiol. 1995; 9: 229-309PubMed Google Scholar This tracer is taken up by brain neurones depending on their functional state, as if it were glucose. It is phosphorylated in the brain tissue by hexokinase to FDG-6-phosphate. Unlike glucose, FDG-6-phosphate becomes intracellularly trapped for at least 45 min without being further metabolized. Uptake of FDG and metabolic trapping of the FDG-6-phosphate is nearly complete about 30 min after injection of the tracer. After this period PET scans are obtained and blood samples are collected to measure glucose and FDG concentrations in plasma as a function of time. The amount of radioactivity in each region of the brain is related to the glucose uptake and the glucose metabolism in this region. As glucose is the primary substrate for brain cells, the measured radioactivity in any particular region represents the glucose metabolism in this area. However, this technique has some limitations. Because of the long half-time (110 min) and the long uptake period, repeated scans are difficult to perform and a constant neurobehavioural state has to be ensured over a long period. For this reason, many drug studies have been performed using CBF rather than CMRGlu as an index of local cerebral activity.36Herscovitch P Can [15O]water be used to evaluate drugs?.J Clin Pharmacol. 2001; 41: 11-20Crossref Google Scholar Because of the biochemical selectivity of PET, neurotransmission processes can be investigated that occur at very low concentrations, typically in the nanomolar to the picomolar range. Specific radioactive tracers can be designed that bind selectively to molecular targets, such as receptors or proteins involved in the synthesis or metabolism of neurotransmitters. A variety of radiolabelled ligands have been designed for tracing different signalling systems in the human brain, including the GABAergic/benzodiazepine, cholinergic, opioid, and monoaminergic system.29Grasby P Malizia A Bench C Psychopharmacology—in vivo neurochemistry and pharmacology.Br Med Bull. 1996; 52: 513-526Crossref PubMed Scopus (11) Google Scholar PET therefore offers the unique possibility to identify putative molecular targets of anaesthetics and to probe their action on neurochemical processes in vivo. Most fMRI studies are based on measuring changes in the blood oxygenation level-dependent (BOLD) contrast that arise from the paramagnetic properties of the deoxyhaemoglobin molecule.25Forster BB MacKay AL Whittall KP et al.Functional magnetic resonance imaging: the basics of blood-oxygen-level-dependent (BOLD) imaging.Can Assoc Radiol J. 1998; 49: 320-329PubMed Google Scholar 48Ogawa S Lee TM Kay AR Tank DW Brain magnetic resonance imaging with contrast dependent on blood oxygenation.Proc Natl Acad Sci USA. 1990; 87: 9868-9872Crossref PubMed Scopus (4576) Google Scholar A decrease in the regional deoxyhaemoglobin concentration increases the effective transverse relaxation time, T2*, that can be mapped by using T2*-weighted MRI techniques. Neural activation is accompanied by a regional increase in blood volume, blood flow, and oxygen consumption.14Bandettini PA Wong EC Magnetic resonance imaging of human brain function. Principles, practicalities, and possibilities.Neurosurg Clin N Am. 1997; 8: 345-371PubMed Google Scholar An increase in blood flow produces a decrease in deoxyhaemoglobin concentration, whereas an increase in oxygen consumption or venous blood volume has the opposite effect. The BOLD contrast therefore depends on changes in both cerebral haemodynamics and oxidative metabolism. When periods of neural stimulation and of rest are compared, the BOLD signal typically is increased during periods of neural stimulation indicating that the haemodynamic response to the stimulus is the dominant variable in BOLD-based fMRI. Several investigators have used PET techniques to measure rCBF15Bonhomme V Fiset P Meuret P et al.Propofol anesthesia and cerebral blood flow changes elicited by vibrotactile stimulation: a positron emission tomography study.J Neurophysiol. 2001; 85: 1299-1308Crossref PubMed Scopus (105) 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 57Veselis RA Reinsel R Feshenko V et al.Dose related decreases in rCBF in R and L prefrontal cortices (PFC) during memory impairment with midazolam and propofol.NeuroImage. 2001; 6 (Abstract): 13Google Scholar and CMRGlu to identify neural structures related to anaesthetic actions in the human brain.3Alkire 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 (282) Google Scholar 4Alkire MT Haier RJ Shah NK Anderson CT Positron emission tomography study of regional cerebral metabolism in humans during isoflurane anesthesia.Anesthesiology. 1997; 86: 549-557Crossref PubMed Scopus (179) Google Scholar 6Alkire MT Pomfrett CJ Haier RJ et al.Functional brain imaging during halothane anesthesia in humans: effects of halothane on global and regional cerebral glucose metabolism.Anesthesiology. 1999; 90: 701-709Crossref PubMed Scopus (166) Google Scholar Using the 18FDG PET technique Alkire and co-workers studied the effects of isoflurane, halothane, and propofol on brain glucose metabolism.3Alkire 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 (282) Google Scholar 4Alkire MT Haier RJ Shah NK Anderson CT Positron emission tomography study of regional cerebral metabolism in humans during isoflurane anesthesia.Anesthesiology. 1997; 86: 549-557Crossref PubMed Scopus (179) Google Scholar 6Alkire MT Pomfrett CJ Haier RJ et al.Functional brain imaging during halothane anesthesia in humans: effects of halothane on global and regional cerebral glucose metabolism.Anesthesiology. 1999; 90: 701-709Crossref PubMed Scopus (166) Google Scholar In these studies the anaesthetics were incrementally titrated to the point of unresponsiveness and brain glucose metabolism was investigated during steady-state conditions. At the same clinical endpoint, similar results in whole-brain glucose metabolism were obtained for the three agents. Halothane decreased whole-brain glucose metabolism by 40%, isoflurane by 46%, and propofol by 55%. This general decrease in metabolic activity is believed to reflect the reduced synaptic activity across the brain in the anaesthetic state. This hypothesis is confirmed by a linear correlation of the metabolic reduction in various EEG variables during propofol and isoflurane anaesthesia.5Alkire MT Quantitative EEG correlations with brain glucose metabolic rate during anesthesia in volunteers.Anesthesiology. 1998; 89: 323-333Crossref PubMed Scopus (141) Google Scholar It is therefore reasonable to assume that the general metabolic reduction associated with the anaesthetic state is a determinant of anaesthetic depth. However, research by Alkire's group did not show only a general cerebral depression, it also revealed subtle differences in regional metabolic reduction caused by the volatile agents. In addition, in comparison with volatile anaesthetics these differences were more pronounced for propofol, consistent with the proposal that volatile agents and propofol act through different molecular mechanisms.27Franks NP Lieb WR Molecular and cellular mechanisms of general anaesthesia.Nature. 1994; 367: 607-614Crossref PubMed Scopus (1632) Google Scholar 42Lees G Molecular mechanism of anaesthesia: light at the end of the channel?.Br J Anaesth. 1998; 81: 491-493Crossref PubMed Scopus (15) Google Scholar Isoflurane anaesthesia caused a nearly uniform cortical and subcortical metabolic reduction.4Alkire MT Haier RJ Shah NK Anderson CT Positron emission tomography study of regional cerebral metabolism in humans during isoflurane anesthesia.Anesthesiology. 1997; 86: 549-557Crossref PubMed Scopus (179) Google Scholar Regional shifts in the pattern of brain glucose metabolism were observed for halothane with a more specific metabolic suppression in the thalamus, basal forebrain, cerebellum, occiput, and limbic system.6Alkire MT Pomfrett CJ Haier RJ et al.Functional brain imaging during halothane anesthesia in humans: effects of halothane on global and regional cerebral glucose metabolism.Anesthesiology. 1999; 90: 701-709Crossref PubMed Scopus (166) Google Scholar In contrast to volatile anaesthetics, the metabolic reduction caused by propofol was not uniform throughout the brain.3Alkire 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 (282) Google Scholar It was more pronounced in the cortex (58%) than in subcortical regions (48%). Furthermore, propofol tended to suppress glucose metabolism in the cerebral cortex, especially in the temporal and occipital regions, to a greater extent than inhalation agents. In addition, compared with halothane, propofol caused significantly less metabolic suppression in the basal ganglia and midbrain regions. Inhaled anaesthetics may cause unconsciousness by altering neuronal activity in specific regions of the CNS. Therefore, the next logical step taken by Alkire and his colleagues was to compare the regional patterns of metabolic depression caused by isoflurane and halothane.7Alkire 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 (329) Google Scholar Such an analysis was expected to reveal brain regions that were affected to a similar extent. If those brain areas had existed, it would have been reasonable to assume that these regions would mediate the state of inhalation anaesthesia. And, in fact, Alkire's analysis revealed brain regions, which differed in their functional activity between the awake state and the drug-induced unconsciousness, irrespective of each particular effect of the agent on regional cerebral metabolism. Both agents caused a specific relative reduction of regional cerebral glucose metabolism primarily in the thalamus and also in the midbrain reticular formation, basal forebrain, cerebellum, and occipital cortex. These findings clearly underline the importance of specific neural structures in mediating drug-induced loss of consciousness and reflect the importance of the thalamus as a target in inhibition of the flow of information to the cortex. Similar mechanisms in mediating anaesthesia were proposed by Angel on the basis of in vitro experiments using rats.10Angel A Central neuronal pathways and the process of anaesthesia.Br J Anaesth. 1993; 71: 148-163Crossref PubMed Scopus (126) Google Scholar 11Angel A How do anaesthetcis work?.Curr Anaesth Crit Care. 1993; 4: 37-45Abstract Full Text PDF Scopus (6) Google Scholar The results of recent H215O PET studies further underpin the hypothesis that specific neural networks contribute to the behavioural changes produced by anaesthetics.15Bonhomme V Fiset P Meuret P et al.Propofol anesthesia and cerebral blood flow changes elicited by vibrotactile stimulation: a positron emission tomography study.J Neurophysiol. 2001; 85: 1299-1308Crossref PubMed Scopus (105) 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 Fiset and co-workers investigated rCBF during graded changes of propofol anaesthesia. Consistent with the general reduction in glucose metabolism reported for propofol by Alkire,3Alkire 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 (282) Google Scholar Fiset found a 20.2% overall decrease in absolute CBF, indicating an overall decrease in cortical neural activity. This effect was particularly pronounced in the medial thalamus, posterior cingulate, basal forebrain, and in the occipitoparietal association cortices. In addition, the authors found a strong correlation between the level of consciousness and the reduction in CBF in the thalamus, basal forebrain, and occipitoparietal regions. Furthermore, a significant covariation between the thalamic and midbrain blood flow changes was observed, suggesting a close relationship between these structures. These results indicate that propofol preferentially affects brain activity in areas that are known to be linked to the control of consciousness, associative functions and autonomic control.51Paus T Functional anatomy of arousal and attention systems in the human brain.Prog Brain Res. 2000; 126: 65-77Crossref PubMed Scopus (129) Google Scholar These findings have been confirmed recently by the same group of authors.15Bonhomme V Fiset P Meuret P et al.Propofol anesthesia and cerebral blood flow changes elicited by vibrotactile stimulation: a positron emission tomography study.J Neurophysiol. 2001; 85: 1299-1308Crossref PubMed Scopus (105) Google Scholar In addition to the previous investigation a stimulus-induced study design was chosen. Changes in rCBF were measured in order to determine whether different levels of propofol anaesthesia would affect subcortical and cortical processing of vibrotactile stimuli in a different way. This study showed a dose-dependent impairment of the processing of vibrotactile stimuli in the brain. Sedative concentrations of propofol already caused a suppression of the rCBF response in the somatosensory cortex, whereas the stimulus-induced thalamic activation was still present. Suppression of rCBF response in this area was merely found when volunteers had lost consciousness. These differential effects may reflect sequential effects on the complex thalamocortical system and may be responsible for dose-dependent behavioural changes. The finding that thalamic activation only disappeared in unconscious subjects provides further evidence for the importance of this particular part of the brain in mediating drug-induced unconsciousness. Specific changes in rCBF were also reported for midazolam.13Bagary M Fluck E File SE et al.Is benzodiazepine-induced amnesia due to deactivation of the left prefrontal cortex?.Psychopharmacology. 2000; 150: 292-299Crossref PubMed Scopus (26) Google Scholar 55Reinsel RA Veselis RA Dnistrian AM et al.Midazolam decreases cerebral blood flow in the left prefrontal cortex in a dose-dependent fashion.Int J Neuropsychopharmacol. 2000; 3: 117-127Crossref PubMed Scopus (55) Google Scholar 56Veselis RA Reinsel RA Beattie BJ et al.Midazolam changes cerebral blood flow in discrete brain regions.Anesthesiology. 1997; 87: 1106-1117Crossref PubMed Scopus (137) Google Scholar Beside the well-known global decrease in CBF, midazolam infusion resulted in CBF reductions in discrete brain regions, including the prefrontal cortex, the superior frontal gyrus, the anterior cingulate gyrus, parietal and temporal association areas, the insular cortex, and the thalamus.35Herscovitch P Markham J Raichle ME Brain blood flow measured with intravenous H215O. I. Theory and error analysis.J Nuclear Med. 1983; 24: 782-789PubMed Google Scholar 56Veselis RA Reinsel RA Beattie BJ et al.Midazolam changes cerebral blood flow in discrete brain regions.Anesthesiology. 1997; 87: 1106-1117Crossref PubMed Scopus (137) Google Scholar These areas subserve arousal and memory processes, particularly the prefrontal cortex.51Paus T Functional anatomy of arousal and attention systems in the human brain.Prog Brain Res. 2000; 126: 65-77Crossref PubMed Scopus (129) Google Scholar It was proposed that the affected regions were most likely involved in mediating amnesia produced by midazolam.56Veselis RA Reinsel RA Beattie BJ et al.Midazolam changes cerebral blood flow in discrete brain regions.Anesthesiology. 1997; 87: 1106-1117Crossref PubMed Scopus (137) Google Scholar Most of the brain imaging studies cited above have underscored the importance of the thalamus as a putative macroscopic target of anaesthetic action. It is therefore not surprising that benzodiazepines may also produce their substantial sedative effects via that target. In the study by Veselis the largest decrease in rCBF during deep midazolam sedation was found in the thalamus. Other human PET studies, related to the action of benzodiazepines, also demonstrated large decreases in regional neural activity in the thalamus after lorazepam45Matthew E Andreason P Pettigrew K et al.Benzodiazepine receptors mediate regional blood flow changes in the living human brain.Proc Natl Acad Sci USA. 1995; 92: 2775-2779Crossref PubMed Scopus (47) Google Scholar 60Volkow ND Wang GJ Hitzemann R et al.Depression of thalamic metabolism by lorazepam is associated with sleepiness.Neuropsychopharmacology. 1995; 12: 123-132Crossref PubMed Scopus (76) Google Scholar or midazolam13Bagary M Fluck E File SE et al.Is benzodiazepine-induced amnesia due to deactivation of the left prefrontal cortex?.Psychopharmacology. 2000; 150: 292-299Crossref PubMed Scopus (26) Google Scholar administration. Like midazolam, low doses of propofol are known to produce substantial anterograde memory impairment. PET imaging revealed significant dose-related CBF decreases in the prefrontal and parietal cortices similar to those obtained for midazolam.57Veselis RA Reinsel R Feshenko V et al.Dose related decreases in rCBF in R and L prefrontal cortices (PFC) during memory impairment with midazolam and propofol.NeuroImage. 2001; 6 (Abstract): 13Google Scholar This may indicate that both drugs, though structurally different, produce their behavioural effects by affecting the same neural networks. In summary, besides a dose-dependent general decrease in whole-brain activity during anaesthesia, the PET imaging studies indicated similar key brain structures involved in the effects of anaesthetics. Furthermore, specific anatomical regions were identified that appear to be crucial for specific behavioural changes caused by anaesthetics, such as amnesia or loss of consciousness. Very low doses of anaesthetics seemed to affect cortical areas preferentially, mainly in the association cortices, and with a further increase in their dosage the primary sensory cortices. This caused disturbances in attention and memory processes, while subjects were still awake and able to respond to commands. Higher doses of anaesthetics did not only affect cortical but also subcortical brain structures. In particular, the thalamus and the midbrain reticular formation appeared to be key targets for drug-induced loss of consciousness. Although PET studies provide a macroanatomical picture of cerebral anaesthetic action, they provide no information about why certain brain areas are more or less sensitive to anaesthetics. Indirectly, this information can be obtained by comparing the results of in vivo studies with ex vivo studies of receptor distribution but the use of modern PET ligand technology appears to be a more promising approach. There is strong evidence from in vitro studies, that anaesthetics may act via GABAergic mechanism.26Franks NP Lieb WR Selective actions of volatile general anaesthetics at molecular and cellular levels.Br J Anaesth. 1993; 71: 65-76Crossref PubMed Scopus (205) Google Scholar 27Franks NP Lieb WR Molecular and cellular mechanisms of general anaesthesia.Nature. 1994; 367: 607-614Crossref PubMed Scopus (1632) Google Scholar 42Lees G Molecular mechanism of anaesthesia: light at the end of the channel?.Br J Anaesth. 1998; 81: 491-493Crossref PubMed Scopus (15) Google Scholar 49Orser BA Wang LY Pennefather PS MacDonald JF Propofol modulates activation and desensitization of GABAA receptors in cultured murine hippocampal neurons.J Neurosci. 1994; 14: 7747-7760Crossref PubMed Google Scholar 50Orser BA McAdam LC Roder S MacDonald JF General anaesthetics and their effects on GABAA receptor desensitization.Toxicol Lett. 1998; 100 (101): 217-224Crossref PubMed Scopus (44) Google Scholar 52Peduto VA Concas A Santoro G et al.Biochemical and electrophysiologic evidence that propofol enhances GABAergic transmission in the rat brain.Anesthesiology. 1991; 75: 1000-1009Crossref PubMed Scopus (132) Google Scholar Measuring with PET the binding characteristic of 11C-labelled flumazenil, a specific benzodiazepine antagonist, in the presence of anaesthetics allows a direct assessment of anaesthetic effects on GABAA receptors in the intact brain. Using this technique, isoflurane has been shown to increase receptor-specific radioligand binding, dependent on GABAA receptor density.32Gyulai FE Mintun MA Firestone LL Dose-dependent enhancement of in vivo GABA(A)-benzodiazepine receptor binding by isoflurane.Anesthesiology. 2001; 95: 585-593Crossref PubMed Scopus (60) Google Scholar This observation provides strong support for the hypothesis that the GABAA receptor is involved in mediating the action of volatile anaesthetics in humans. In addition, the radioligand binding measured during anaesthesia at 1.5 MAC (minimum alveolar concentration) was significantly greater than at 1.0 MAC, indicating a dose-related effect of isoflurane on GABAA receptor ligand binding (Fig. 1). Alkire and co-workers8Alkire MT Haier RJ Correlating in vivo anaesthetic effects with ex vivo receptor density data supports a GABAergic mechanism of action for propofol, but not for isoflurane.Br J Anaesth. 2001; 86: 618-626Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar correlated the data from their in vivo studies3Alkire 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 (282) Google Scholar 7Alkire 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 (329) Google Scholar reported recently with regional distribution patterns of various human receptor binding site densities obtained from previous ex vivo studies.16Braestrup C Albrechtsen R Squires RF High densities of benzodiazepine receptors in human cortical areas.Nature. 1977; 269: 702-704Crossref PubMed Scopus (344) Google Scholar 22Enna SJ Bennet Jr, JP Bylund DB et al.Neurotransmitter receptor binding: regional distribution in human brain.J Neurochem. 1977; 28: 233-236Crossref PubMed Scopus (80) Google Scholar 64Zezula J Cortes R Probst A Palacios JM Benzodiazepine receptor sites in the human brain: autoradiographic mapping.Neuroscience. 1988; 25: 771-795Crossref PubMed Scopus (123) Google Scholar The regional reduction in glucose metabolism caused by propofol exhibited a significant correlation (r=–0.86, P<0.0005) with the distribution of the benzodiazepine binding site densities, which means that metabolism decreased more in regions with higher receptor levels (Fig. 2). In contrast to the link between changes in glucose metabolism caused by propofol and the benzodiazepine receptors the metabolic reduction observed during isoflurane anaest