Editorial I: Mapping the anatomy of unconsciousness—imaging anaesthetic action in the brain

Abstract

The issue of how and where general anaesthetics act in the brain continues to intrigue and occupy researchers. Studies that address these issues can be focused at different levels. One reductionist approach is to investigate anaesthetic interactions with specific receptors or other molecular targets in a variety of models.1Van Swinderen B Shook DR Ebert RH et al.Quantitative trait loci controlling halothane sensitivity in Caenorhabditis elegans.Proc Natl Acad Sci USA. 1997; 94: 8232-8237Crossref PubMed Scopus (28) Google Scholar, 2Krasowski MD Koltchine VV Rick CE et al.Propofol and other intravenous anesthetics have sites of action on the γ-aminobutyric acid type A receptor distinct from that for isoflurane.Mol Pharmacol. 1998; 53: 530-538Crossref PubMed Scopus (245) Google Scholar, 3Franks NP Lieb WR Selectivity of general anesthetics: a new dimension.Nat Med. 1997; 3: 377-378Crossref PubMed Scopus (11) Google Scholar, 4Lees G Molecular mechanisms of anaesthesia: light at the end of the channel?.Br J Anaesth. 1998; 81: 491-493Crossref PubMed Scopus (15) Google Scholar, 5Franks NP Lieb WR Which molecular targets are most relevant to general anaesthesia?.Toxicol Lett. 1998; 100-101: 1-8Crossref PubMed Scopus (129) Google Scholar, 6Delia Belelli R Pistis M Peters JA Lambert JJ General anaesthetic action at transmitter-gated inhibitory amino acid receptors.Tr Pharmacol Sci. 1999; 20: 496-502Abstract Full Text Full Text PDF PubMed Scopus (132) Google Scholar While such studies are important, they are difficult to interpret in the context of functional neuroanatomical models of consciousness.7Drummond JC Monitoring depth of anesthesia.Anesthesiology. 2000; 93: 876-882Crossref PubMed Scopus (239) Google Scholar 8Angel A Central neuronal pathways and the process of anaesthesia.Br J Anaesth. 1993; 71: 148-163Crossref PubMed Scopus (126) Google Scholar An alternative approach is to consider the differential effects of anaesthesia on different brain areas, with a view to identifying specific brain regions that are important for anaesthesia (and by inference, for the generation of consciousness). Such an assessment of spatial variations in general anaesthetic effects in the brain is not easy. While clinical measurement of anaesthetic effect has substantially depended on measuring spontaneous or evoked electrical responses,7Drummond JC Monitoring depth of anesthesia.Anesthesiology. 2000; 93: 876-882Crossref PubMed Scopus (239) Google Scholar current implementations cannot provide the tomographic visualization of subcortical physiology required to understand sites of anaesthetic action.8Angel A Central neuronal pathways and the process of anaesthesia.Br J Anaesth. 1993; 71: 148-163Crossref PubMed Scopus (126) Google Scholar 9Smythies J The functional neuroanatomy of awareness: with a focus on the role of various anatomical systems in the control of intermodal attention.Consc Cognit. 1997; 6: 455-481Crossref PubMed Scopus (90) Google Scholar Magneto-encephalography (MEG) can provide detailed information regarding foci of brain activation,10Simos PG Papanicolaou AC Breier JI et al.Insights into brain function and neural plasticity using magnetic source imaging.J Clin Neurophysiol. 2000; 17: 143-162Crossref PubMed Scopus (27) Google Scholar but while the technique has been applied to pre-surgical planning,11Alberstone CD Skirboll SL Benzel EC et al.Magnetic source imaging and brain surgery: presurgical and intraoperative planning in 26 patients.J Neurosurg. 2000; 92: 79-90Crossref PubMed Scopus (70) Google Scholar there are no reports of MEG studies of anaesthetic action. This inability to map primary brain function (i.e. electrical activity) has led researchers to image secondary physiological phenomena associated with neuronal excitation. As neuronal activity is closely coupled to glucose metabolism (and in many circumstances to blood flow), imaging the effect of anaesthetic agents on cerebral blood flow (CBF) and cerebral metabolic rates for glucose (CMRgluc) may provide surrogate measures of regional effects. Indeed, these indices have been investigated in several experimental studies of anaesthetic action in animal models using autoradiography with [14C]iodoantipyrine and [14C]deoxyglucose, respectively.12Hodes JE Soncrant TT Larson DM Carlson SG Rapoport SI Selective changes in local cerebral glucose utilization induced by phenobarbital in the rat.Anesthesiology. 1985; 63: 633-639Crossref PubMed Scopus (40) Google Scholar, 13Maekawa T Tommasino C Shapiro HM Keifer-Goodman J Kohlenberger RW Local cerebral blood flow and glucose utilization during isoflurane anesthesia in the rat.Anesthesiology. 1986; 65: 144-151Crossref PubMed Scopus (127) Google Scholar, 14Ori C Dam M Pizzolato G Battistin L Giron G Effects of isoflurane anesthesia on local cerebral glucose utilization in the rat.Anesthesiology. 1986; 65: 152-156Crossref PubMed Scopus (58) Google Scholar, 15Dam M Ori C Pizzolato G et al.The effects of propofol anesthesia on local cerebral glucose utilization in the rat.Anesthesiology. 1990; 73: 499-505Crossref PubMed Scopus (62) Google Scholar, 16Lenz C Rebel A van Ackern K Kuschinsky W Waschke KF Local cerebral blood flow, local cerebral glucose utilization, and flow-metabolism coupling during sevoflurane versus isoflurane anesthesia in rats.Anesthesiology. 1998; 89: 1480-1488Crossref PubMed Scopus (97) Google Scholar, 17Cavazzuti M Porro CA Biral GP Benassi C Barbieri GC Ketamine effects on local cerebral blood flow and metabolism in the rat.J Cereb Blood Flow Metab. 1987; 7: 806-811Crossref PubMed Scopus (118) Google Scholar More recent experimental studies have used the newer technique of functional magnetic resonance imaging (fMRI) to map changes in regional CBF associated with anaesthetic effects.18Burdett NG Menon DK Carpenter TA Jones JG Hall LD Visualisation of changes in regional cerebral blood flow (rCBF) produced by ketamine using long TE gradient-echo sequences: preliminary results.Magn Reson Imag. 1995; 13: 549-553Abstract Full Text PDF PubMed Scopus (32) Google Scholar Similar studies are now possible in humans with techniques such as positron emission tomography (PET) and fMRI. The most common implementation of fMRI in this context uses blood oxygen level dependent (BOLD) contrast,19Howseman AM Bowtell RW Functional magnetic resonance imaging: imaging techniques and contrast mechanisms.Phil Trans Biol Sci. 1999; 354: 1179-1194Crossref PubMed Scopus (54) Google Scholar 20Ugurbil K Chen W Zhu X-H Kim S-G Georgopoulos A Functional mapping in the human brain using high magnetic fields.Phil Trans: Biol Sci. 1999; 354: 1195-1213Crossref PubMed Scopus (131) Google Scholar which depends on coupled increases in regional CBF associated with neuronal activation. It is now known that such CBF recruitment is associated with a reduction in proportional oxygen extraction, and a relative increase in regional oxygen saturation. The use of appropriate image acquisition protocols or ‘sequences’ translates these changes in regional oxygen saturation into areas of high signal intensity on subtraction images. These changes can be submitted to statistical analysis to identify areas of significant blood flow change, and inferences drawn about neural activation. The microcirculatory changes responsible for such fMRI contrast are crucially dependent on the maintenance of normal flow-metabolism coupling. While flow-metabolism coupling is retained to a substantial extent with many anaesthetics,21Hansen TD Warner DS Todd MM Vust LJ The role of cerebral metabolism in determining the local cerebral blood flow effects of volatile anesthetics: evidence for persistent flow-metabolism coupling.J Cereb Blood Flow Metab. 1989; 9: 323-328Crossref PubMed Scopus (79) Google Scholar the relative difficulty of providing anaesthesia in MR environments has limited the application of fMRI to studies of anaesthetic action in humans. However, one recent fMRI study in human volunteers suggested that isoflurane impairs thalamo-cortical transmission of sensory information,22Antognini JF Buonocore MH Disbrow EA Carstens E Isoflurane anesthesia blunts cerebral responses to noxious and innocuous stimuli: a fMRI study.Life Sci. 1997; 61: PL349-PL354Crossref Scopus (67) Google Scholar in keeping with previous electrophysiological studies in animal models.8Angel A Central neuronal pathways and the process of anaesthesia.Br J Anaesth. 1993; 71: 148-163Crossref PubMed Scopus (126) Google Scholar Positron emission tomography studies the distribution and kinetics of molecules that incorporate positron-emitting isotopes23Saha GB MacIntyre WJ Go RT Radiopharmaceuticals for brain imaging.Semin Nucl Med. 1994; 24: 324-349Abstract Full Text PDF PubMed Scopus (89) Google Scholar, 24Hoffman JM Coleman RE Perfusion quantitation using positron emission tomography.Invest Radiol. 1992; 27: S22-S26Crossref PubMed Scopus (7) Google Scholar, 25Jones T The role of positron emission tomography within the spectrum of medical imaging.Eur J Nucl Med. 1996; 23: 207-211Crossref PubMed Scopus (82) Google Scholar such as H215Dam M Ori C Pizzolato G et al.The effects of propofol anesthesia on local cerebral glucose utilization in the rat.Anesthesiology. 1990; 73: 499-505Crossref PubMed Scopus (62) Google ScholarO, which is used to study blood flow, and 18fluoro-deoxyglucose (18FDG), which provides images of cerebral glucose uptake. Such studies provide maps of cerebral physiology and metabolism with sub-centimetre spatial resolution and a sufficiently small radiation burden (typically less than 5 mSv in the context of an annual background radiation in East Anglia of about 2 mSv) that allows application of the technique to healthy volunteers. While concomitant sampling of radioactivity in arterial blood can provide quantitative results, it is far more common in volunteer studies to acquire non-quantitative images. This approach has been widely used by neuropsychologists.26Frackowiak RS Friston KJ Functional neuroanatomy of the human brain: positron emission tomography – a new neuroanatomical technique.J Anat. 1994; 184: 211-225PubMed Google Scholar Blood flow maps are acquired during a selected control task and during a test task, which imposes a small and well-defined additional cognitive burden on the brain. Differences in CBF patterns observed between the two tasks can then be attributed to neuronal activation and the synaptic activity required to service the additional cognitive burden of the test task27Jueptner M Weiller C Review: does measurement of regional cerebral blood flow reflect synaptic activity? Implications for PET and fMRI.Neuroimage. 1995; 2: 148-156Crossref PubMed Scopus (428) Google Scholar. While this process of cognitive subtraction has occasionally been criticized,28Friston KJ Price CJ Fletcher P Moore C Frackowiak RS Dolan RJ The trouble with cognitive subtraction.Neuroimage. 1996; 4: 97-104Crossref PubMed Scopus (410) Google Scholar 29Price CJ Friston RJ Cognitive conjunction: a new approach to brain activation experiments.Neuroimage. 1997; 5: 261-270Crossref PubMed Scopus (721) Google Scholar it (or one of its variants) still forms the basis of most functional imaging studies. This technique can be extended to studies of drug effects on the brain, where the cognitive burden is invariant, but images are acquired before and after a drug is administered. As with conventional functional imaging, subtraction images can be submitted to careful statistical analysis to identify significant increases or decreases in CBF or CMRgluc associated with administration of a drug. Several research groups have attempted to use PET to identify sites of anaesthetic action in the human brain,30Alkire MT Pomfrett CJ Haier RJ et al.Functional brain imaging during anesthesia in humans: effects of halothane on global and regional cerebral glucose metabolism.Anesthesiology. 1999; 90: 701-709Crossref PubMed Scopus (166) Google Scholar, 31Alkire 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, 32Alkire 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 (283) Google Scholar, 33Fiset 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, 34Veselis RA Reinsel RA Beattie BJ et al.Midazolam changes cerebral blood flow in discrete brain regions: an H2(15)O positron emission tomography study.Anesthesiology. 1997; 87: 1106-1117Crossref PubMed Scopus (137) Google Scholar and have provided interesting, though not always concordant results. Alkire and colleagues used 18Burdett NG Menon DK Carpenter TA Jones JG Hall LD Visualisation of changes in regional cerebral blood flow (rCBF) produced by ketamine using long TE gradient-echo sequences: preliminary results.Magn Reson Imag. 1995; 13: 549-553Abstract Full Text PDF PubMed Scopus (32) Google ScholarFDG PET to show that halothane30Alkire MT Pomfrett CJ Haier RJ et al.Functional brain imaging during anesthesia in humans: effects of halothane on global and regional cerebral glucose metabolism.Anesthesiology. 1999; 90: 701-709Crossref PubMed Scopus (166) Google Scholar and isoflurane31Alkire 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 resulted in global reductions in cerebral metabolism, with relatively prominent effects on the basal forebrain structures, thalamus, limbic system, and cerebellum. In contrast, propofol32Alkire 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 (283) Google Scholar produced more metabolic depression in the cortex than in subcortical structures. However, these data are at odds with a more recent study that used H215O PET to study changes in CBF during graded propofol anaesthesia.33Fiset 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 These authors showed global reductions in CBF with relatively prominent effects in the medial thalamus, in addition to specific cortical areas. They also demonstrated a close correlation between midbrain and thalamic blood flow, suggesting concordant changes in arousal systems. Interestingly, in another study, high dose midazolam34Veselis RA Reinsel RA Beattie BJ et al.Midazolam changes cerebral blood flow in discrete brain regions: an H2(15)O positron emission tomography study.Anesthesiology. 1997; 87: 1106-1117Crossref PubMed Scopus (137) Google Scholar resulted in CBF reductions in multiple cortical areas, but also selectively reduced thalamic CBF. Specific cortical areas that were commonly affected by both agents across three studies included the angular gyrus, anterior cingulate area, and parietal and temporal association cortices, areas that are commonly implicated in arousal and information processing. While it is tempting to suggest that these data reveal the anatomical sites at which propofol exerts its anaesthetic effects, it is also possible that these changes are a consequence of the anaesthetic state produced by propofol acting at a more strategic and focused site. The functional imaging studies described thus far address the ‘where’ of anaesthetic action in neuroanatomical terms, but it may be possible to provide additional specificity in terms of the receptor systems involved. Data regarding the regional distribution of receptor subtypes are available from post-mortem and PET ligand studies. Synthesis of such data with changes in regional CBF or CMRgluc would allow us to correlate the distribution of a putative site of anaesthetic action with the documented distribution of regional pharmacodynamic effects. The paper by Alkire and colleagues35Alkire MT Haier RJ Correlating in vivo anaesthetic effects with ex-vivo receptor density supports a GABAergic mechanism of action for propofol, but not for isoflurane.Br J Anaesthesia. 2001; 86: 618-626Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar in this issue of the journal attempts to do this by using PET data that has, in large part, been previously published by these authors. They relate the cerebral metabolic effects of isoflurane and propofol in human subjects (measured using 18FDG PET)31Alkire 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 32Alkire 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 (283) Google Scholar to the regional distribution of selected neurotransmitter systems in the brain (using literature data from post-mortem immunohistochemistry36Braestrup C Albrechtsen R Squires RF High densities of benzodiazepine receptors in human cortical areas.Nature. 1977; 269: 702-704Crossref PubMed Scopus (344) Google Scholar 37Zezula J Cortes R Probst A Placacios JM Benzodiazepine receptor sites in the human brain: Autoradiographic mapping.Neuroscience. 1988; 25: 771-795Crossref PubMed Scopus (123) Google Scholar to define receptor densities). In principle, the approach is a valid one, and has produced interesting results, with dissociation demonstrated between patterns of metabolic suppression seen with propofol and isoflurane. The authors also demonstrate a correlation of regional metabolic suppression by propofol with historical data on regional [3H]diazepam and [3H]flunitrazepam binding. Unexpectedly, isoflurane induced reductions in CMRgluc did not correlate with benzodiazepine binding, but showed an inverse correlation with muscarinic receptor density. These data led the authors to conclude that ‘the most logical interpretation for this effect is to suggest that the diazepam binding site on the GABAA complex is likely to be strongly regionally co-localized with the GABAergic site that mediates propofol’s in vivo effects on brain metabolism’. They suggest that the absence of such a correlation with isoflurane may be because of a variety of factors, including the possibility that it acts on several receptor systems, and to use common parlance, is a ‘dirtier’ agent. They suggest that the inverse correlation seen with muscarinic receptor binding can be accounted for by the well documented arousal effects of this receptor system, which would antagonize the metabolic suppression produced by isoflurane in proportion to its local density. While these data are interesting, several caveats need to be voiced regarding the details of the methodology used for functional imaging studies of anaesthetic action in general, and the current paper in particular. These techniques are based on the assumption that coupled increases in blood flow and glucose metabolism can provide a marker of regional neuronal activation. While this may be true for physiological activation, its relevance to pharmacological deactivation remains unclear. In particular, caution needs to be exercised regarding the implicit but central assumption that a reduction in regional brain metabolism by a drug is necessarily the consequence of a modulation of synaptic activity by the agent in that region. While this may well be the case, this is not always so, as suppression of strategic neuronal pathways can clearly affect metabolic activity in their projection areas. For example, marked suppression of thalamic transmission of somatosensory inputs can decrease cortical activity in the somatosensory cortex. Indeed, Angel and colleagues8Angel A Central neuronal pathways and the process of anaesthesia.Br J Anaesth. 1993; 71: 148-163Crossref PubMed Scopus (126) Google Scholar have shown that such thalamic activity is an important site of action for several anaesthetic agents, including those thought to act primarily at the GABAA complex. These electrophysiological data are concordant with the fMRI data from Antognini22Antognini JF Buonocore MH Disbrow EA Carstens E Isoflurane anesthesia blunts cerebral responses to noxious and innocuous stimuli: a fMRI study.Life Sci. 1997; 61: PL349-PL354Crossref Scopus (67) Google Scholar who demonstrated lack of cortical somatosensory activation following anaesthesia despite preserved thalamic activation, implying arrest of inputs at a thalamic ‘gate’.22Antognini JF Buonocore MH Disbrow EA Carstens E Isoflurane anesthesia blunts cerebral responses to noxious and innocuous stimuli: a fMRI study.Life Sci. 1997; 61: PL349-PL354Crossref Scopus (67) Google Scholar 38Alkire 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 unconsciousness.Consc Cognit. 2000; 9: 370-386Crossref PubMed Scopus (329) Google Scholar In this situation, the visible deactivation is in the cortex, but the site of anaesthetic action is the thalamus. The direct vascular effects of anaesthetics present us with another confounder in such studies, especially when the paradigm involves higher doses of halogenated volatile agents.39Matta BF Heath KJ Tipping K Summors AC Direct cerebral vasodilatory effects of sevoflurane and isoflurane.Anesthesiology. 1999; 91: 677-680Crossref PubMed Scopus (286) Google Scholar The varying degrees of direct cerebral vasodilation observed with these agents may modulate the CBF responses to activation and deactivation. However, available evidence suggests that even drugs that alter neurovascular coupling do not alter changes in cerebral glucose metabolism, other than by direct neuronal effects.40Tsukada H Kakiuchi T Ando I Shizuno H Nakanishi S Ouchi Y Regulation of cerebral blood flow response to somatosensory stimulation through the cholinergic system: a positron emission tomography study in unanesthetized monkeys.Brain Res. 1997; 749: 10-17Crossref PubMed Scopus (37) Google Scholar Studies also need to take account of the variability in functional neuroanatomy between subjects. While these considerations may not be relevant to studies addressing gross patterns of activation and deactivation, they may be important when testing the effects of low or residual levels of anaesthesia in functional imaging paradigms that assess effects on complex cognitive tasks. In these circumstances, functional images should be coregistered to each individual subject’s anatomy (using a high resolution MRI), and image averaging undertaken after ‘normalizing’ these images to a generic template. Ideally, regions of interest that define neuroanatomical structures should be identified on such high-resolution images. Finally, the source of data for receptor densities in comparisons of the kind used by Alkire and colleagues35Alkire MT Haier RJ Correlating in vivo anaesthetic effects with ex-vivo receptor density supports a GABAergic mechanism of action for propofol, but not for isoflurane.Br J Anaesthesia. 2001; 86: 618-626Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar remains a difficult issue. There are problems using in vitro data on [3Franks NP Lieb WR Selectivity of general anesthetics: a new dimension.Nat Med. 1997; 3: 377-378Crossref PubMed Scopus (11) Google ScholarH]diazepam binding36Braestrup C Albrechtsen R Squires RF High densities of benzodiazepine receptors in human cortical areas.Nature. 1977; 269: 702-704Crossref PubMed Scopus (344) Google Scholar as a basis for assessing regional GABA receptor density. Diazepam binds to both neuronal (GABA/Cl– channel associated) and mitochondrial benzodiazepine receptors (MBRs).41Parola AL Yamamura HI Laird 3rd, HE Peripheral-type benzodiazepine receptors.Life Sci. 1993; 52: 1329-1342Crossref PubMed Scopus (100) Google Scholar MBRs are typically expressed by inflammatory cells, including activated microglia and blood derived macrophages. Their use of flunitrazepam binding to corroborate these data substantially addresses this criticism, as this is a selective neuronal benzodiazepine ligand.37Zezula J Cortes R Probst A Placacios JM Benzodiazepine receptor sites in the human brain: Autoradiographic mapping.Neuroscience. 1988; 25: 771-795Crossref PubMed Scopus (123) Google Scholar However, regardless of ligand used, post-mortem data cannot account for inter-individual variability, and an ideal study would have compared regional benzodiazepine receptor density in volunteers with regional reductions in 18FDG uptake induced by anaesthetic agents in the same subjects. This is a technically feasible exercise, since [11Alberstone CD Skirboll SL Benzel EC et al.Magnetic source imaging and brain surgery: presurgical and intraoperative planning in 26 patients.J Neurosurg. 2000; 92: 79-90Crossref PubMed Scopus (70) Google ScholarC]flumazenil is a well-characterized PET ligand in humans.42Frey KA Holthoff VA Koeppe RA et al.Parametric in vivo imaging of benzodiazepine receptor distribution in human brain.Ann Neurol. 1991; 30: 663-672Crossref PubMed Scopus (93) Google Scholar Even if we accept that post-mortem benzodiazepine binding provides an acceptable approximation of central benzodiazepine receptor distribution, caution is required in equating this to GABAA receptor distribution or to the regional metabolic effects of potentiating GABAergic receptors. Several papers43Ableitner A Herz A Influence of meprobamate and phenobarbital upon local cerebral glucose utilzation: parallelism with effects of the anxiolytic diazepam.Brain Res. 1987; 403: 82-88Crossref PubMed Scopus (8) Google Scholar 44Kelly PA McCulloch J Effects of the putative GABAergic agonists, muscimol and THIP, upon local cerebral glucose utilisation.J Neurochem. 1982; 39: 613-624Crossref PubMed Scopus (58) Google Scholar (including that by Braestrup and colleagues36Braestrup C Albrechtsen R Squires RF High densities of benzodiazepine receptors in human cortical areas.Nature. 1977; 269: 702-704Crossref PubMed Scopus (344) Google Scholar) make the point that regional [3H]diazepam binding does not necessarily correlate with GABA receptor distribution (though there is a correlation with bicuculline binding sites). More importantly, administration of GABA agonists such as muscimol and THIP (and diazepam) to animals does not produce changes in glucose metabolism in patterns that correlate to receptor density.43Ableitner A Herz A Influence of meprobamate and phenobarbital upon local cerebral glucose utilzation: parallelism with effects of the anxiolytic diazepam.Brain Res. 1987; 403: 82-88Crossref PubMed Scopus (8) Google Scholar 44Kelly PA McCulloch J Effects of the putative GABAergic agonists, muscimol and THIP, upon local cerebral glucose utilisation.J Neurochem. 1982; 39: 613-624Crossref PubMed Scopus (58) Google Scholar Despite these reservations, the authors’ findings are intriguing. The relative selectivity of propofol induced cortical metabolic suppression (in contrast to that produced by isoflurane, which was more global) may well be caused by variations in the regional density of receptor systems. However, the data provided do not prove this (and to be fair, the authors do not claim it). What the authors have done is to provide a clear experimental basis for an interesting and testable hypothesis. Using modern PET cameras, it should be possible to obtain 18Burdett NG Menon DK Carpenter TA Jones JG Hall LD Visualisation of changes in regional cerebral blood flow (rCBF) produced by ketamine using long TE gradient-echo sequences: preliminary results.Magn Reson Imag. 1995; 13: 549-553Abstract Full Text PDF PubMed Scopus (32) Google ScholarFDG studies at baseline and with anaesthetics, [11C]flumazenil PET images and anatomical MR, for coregistration in a cohort of volunteers with an acceptable radiation burden (≈6 mSv) to participants. Indeed, modern PET ligand technology already offers, or promises to provide, access to a wide variety of receptors45Duncan JS Positron emission tomography receptor studies.Adv Neurol. 1999; 79: 893-899PubMed Google Scholar (including opioid46Jones AK Kitchen ND Watabe H et al.Measurement of changes in opioid receptor binding in vivo trigeminal neuralgic pain using [11C]diprenorphine and positron emission tomography.J Cereb Blood Flow Metab. 1999; 19: 803-808Crossref PubMed Scopus (81) Google Scholar, 47Casey KL Svensson P Morrow TJ et al.Selective opiate modulation of nociceptive processing in the human brain.J Neurophysiol. 2000; 84: 525-533Crossref PubMed Scopus (157) Google Scholar, 48Zubieta JK Dannals RF Frost JJ Gender and age influences on human brain mu-opioid receptors by PET.Am J Psychiatr. 1999; 156: 842-848Crossref PubMed Scopus (280) Google Scholar and cholinergic49Yoshida T Kuwabara Y Sasaki M et al.Sex-related differences in the muscarinic acetylcholinergic receptor human brain – a positron emission tomography study.Ann Nucl Med. 2000; 14: 97-101Crossref PubMed Scopus (26) Google Scholar, 50Zubieta JK Koeppe RA Mulholland GK Kuhl DE Frey KA Quantification of muscarinic cholinergic receptors with [11C]NMPB and positron emission tomography: method development and differentiation of tracer delivery from receptor binding.J Cereb Blood Flow Metab. 1998; 18: 619-631Crossref PubMed Scopus (37) Google Scholar, 51Grunwald F Biersack HJ Kuschinsky W Nicotine receptor mapping.Eur J Nucl Med. 1996; 23: 1012-1014Crossref PubMed Google Scholar) that would be of interest in this context. Future results from such studies promise to be of great interest. D. K. Menon Professor of Anaesthesia, University of Cambridge Director, Neurocritical Care Unit Addenbrooke’s Hospital Cambridge UK