In Vivo Imaging of Human Limbic Responses to Nitrous Oxide... : Anesthesia & Analgesia (original) (raw)
Despite its limited potency, nitrous oxide is the most widely used inhaled general anesthetic [1]. Subanesthetic concentrations (i.e., 20%-30% atm) are also used extensively in dentistry and minor surgery for sedation, amnesia, and analgesia. Human behavioral studies indicate that nitrous oxide, in this same concentration range, depresses psychomotor function [2], cognition [3], learning, and memory [4], but the neuroanatomic substrates of these effects are essentially unknown. Quantitative local cerebral glucose utilization has been measured in animals during nitrous oxide inhalation [5-7], but only minimal regional changes were revealed due to substantial interindividual variance in global glucose metabolic rate.
Recent advances in functional brain imaging now provide an opportunity to identify, noninvasively, brain areas affected by nitrous oxide in humans, and positron emission tomography (PET) is among the best tools for measuring changes in brain activity in vivo. Based on the Kety-Schmidt technique [8], where flow-dependent accumulation or disappearance of a bolus-injected, freely diffusable tracer is used to measure regional cerebral blood flow (rCBF) [9], PET uses15 O-water as the tracer because its relatively short half-life (approximate 2 min) translates into a relatively high temporal resolution for rCBF alterations. The alterations in rCBF, in turn, are interpreted as reflecting temporally related changes in neuronal activity [10]. Thus, we undertook to identify brain regions where activity changed in response to nitrous oxide inhalation and to correlate these regions with areas associated with behaviors most affected by nitrous oxide. A low concentration (20%) was used to avoid global metabolic changes. To ensure that alterations in rCBF truly represented alterations in neuronal activity, the effect of 20% nitrous oxide on cerebral blood flow-metabolism coupling was separately assessed by18 F-deoxyglucose (18) FDG) PET scanning. This adaptation of the technique of Sokoloff et al. [11] to measure regional cerebral metabolic rate (rCMR) uses deoxyglucose labeled with the positron emitter18 F; this substrate is trapped within neurons at a rate directly proportional to their metabolic activity [12]. Since18 F has a half-life of 110 minutes and is only slowly taken up into neurons,18 FDG PET scanning monitors rCMR with much less temporal resolution than rCBF is monitored by15 O-water. However, this disparity does not limit the validity of rCBF-rCMR comparisons, since in the present experiment a constant brain concentration of nitrous oxide is maintained for the entire duration of18 FDG PET scanning.
Methods
Written, informed consent was obtained from eight right-handed healthy volunteers (five women, three men) with a mean age of 29.5 yr (range, 21-46 yr). The protocol was approved by the University of Pittsburgh Institutional Review Board (IRB #931063) and Radioactive Drug Review Committee. Subjects had no history of a psychotic or mood disorder or substance abuse. Serum pregnancy tests were negative for all female volunteers.
Subjects were studied under two experimental conditions. During experiments all subjects were supine with their gaze fixed on a crosshair image projected on an overhead computer screen. In the control condition, subjects inhaled room air, while in the nitrous oxide condition, the inhaled mixture contained 20% atm of nitrous oxide, 30% atm oxygen, balance room air.
To assess whether nitrous oxide affected rCBF-rCMR coupling, both were separately assayed under control and nitrous oxide conditions in the initial four subjects. Specifically, two rCBF PET scans and one rCMR scan were obtained during both conditions using15 O-water [13] and18 FDG [12] as radiotracers, respectively. In each of the remaining four subjects, only two rCBF scans were obtained during the control and nitrous oxide conditions. Values obtained during the two rCBF scans were averaged for each subject for each condition, and these averages were then statistically compared. In all subjects, the control condition always preceded nitrous oxide administration.
Cranial PET scans were performed with a Siemens 951R/31 scanner (Siemens Medical Systems, Hoffman Estates, IL), with an inherent detector resolution of approximately 6 mm full-width half-maximum (FWHM), by collecting 31 parallel slices over an axial field of view of 10.8 cm in both axial and transaxial directions. Each subject was positioned in the scanner to optimize cerebral images, with the lowest image plane approximately parallel to, and 1 cm above, the canthomeatal line. A transmission scan, collected during exposure of a68 Ge/(68) Ga ring source, was used to correct for radiation attenuation by the tissues of the head. Scans were reconstructed with a Hanning 0.5 filter, giving a transaxial resolution of 8.5 mm FWHM. The reconstructed PET images contained 128 times 128 pixels, each 2.05 times 2.05 mm.
For rCBF scans, 20 s after bolus injection of 50 mCi H215 O, a 60-s data acquisition interval was started [14]. At least 12 min elapsed between each H215 O scan and between H215 O and subsequent18 FDG scans to allow radioactivity to decay to background levels. For each rCMR scan, 5 mCi of18 FDG was injected by intravenous bolus followed by a 40-min period during which uptake and metabolic trapping of18 FDG in the brain are nearly complete [11,15]. A 20-min scanning period was then begun.
Nitrous oxide administration began at least 15 min before PET scanning to achieve steady-state maximum pulmonary alveolar and brain concentrations, which were maintained during the entire scanning session and verified by a laser gas monitor (Rascal II; Ohmeda, Salt Lake City, UT). Inhaled gas mixtures were delivered through a conventional semiclosed anesthesia circuit system using a tight-fitting, soft plastic face mask and monitored during both inspiration and expiration. Other physiologic variables monitored included noninvasive blood pressure, electrocardiogram, arterial oxygen saturation, and end-tidal carbon dioxide.
PET scan images are reconstructed using measured attenuation factors and aligned to each other to correct for small head movements during the session [16]. rCBF and rCMR images are processed using statistical parametric mapping (SPM), which allows image averaging across subjects for a given experimental condition [17,18]. Specifically, to reduce differences in brain position, size, and shape, individual rCBF and rCMR images are stereotactically normalized (rescaled) to conform to the standard stereotactic space defined by the atlas of Talairach and Tournoux [19]. This stereotactic space is now the international standard for communicating PET results. Positional differences are also removed by reorienting the individual rCBF and rCMR images to a standard line passing through the anterior and posterior commissures (AC-PC line). Images are then smoothed with a three-dimensional Gaussian filter (20 mm times 20 mm times 12 mm in the x, y, and z axes, respectively) to increase the signal-to-noise ratio and accommodate normal variability in functional and gyral anatomy. The next component of intersubject rCBF and rCMR averaging is normalization for differences in global CBF and CMR between subjects and conditions using analysis of covariance with global CBF and CMR as independent variables and rCBF and rCMR as dependent variables [17]. This process, together with the stereotactic normalization, generates group mean rCBF and rCMR images for the control and nitrous oxide conditions. The pixel values from these images, with their adjusted variances, are used for further statistical analysis.
To identify brain areas of nitrous oxide-related activity changes, comparisons between rCBF and rCMR for the control and nitrous oxide conditions are performed on a pixel-by-pixel basis using t statistics, as reported previously [18]. For regions of activation or deactivation to be considered significant, three separate criteria had to be fulfilled: 1) P values of differences in pixel values between conditions had to be < 0.005; 2) the number of pixels achieving significance in the pixel-by-pixel t-test had to exceed the number of significant pixels expected to reach significance by chance, as tested by chi squared analysis; and 3) the significant pixels also had to form a confluent group extending over more than one transverse image slice. T statistic values (t scores) are converted to z scores to allow comparisons independent of degrees of freedom.
To detect whether there is rCBF-CMR uncoupling in areas where significant nitrous oxide-associated blood flow increases and decreases are identified in the averaged rCBF image of all eight volunteers, the averaged rCBF and rCMR scans of the first four subjects are assessed for differences on a pixel-by-pixel basis with "point analysis." Specifically, in the averaged rCBF and rCMR data sets of the initial four subjects, the mean percentage changes of rCBF scan pixels, which show the most robust changes in the averaged rCBF scan pixels of all eight volunteers, are compared with those of stereotactically corresponding rCMR scan pixels by t-tests, using a P level of 0.05 to define significance.
In addition to the standard SPM method for statistical analysis, which identifies areas of significant changes according to the standard stereotactic coordinates of Talairach and Tournoux and is also used for point analysis, rCBF scans of all eight subjects are aligned with magnetic resonance images (MRIs) to more accurately verify and display the anatomic locations of nitrous oxide-associated activated and deactivated areas. Specifically, this yields horizontal PET image slices depicting statistically significant activated or deactivated areas associated with nitrous oxide administration superimposed upon the corresponding MRI slices. The PET images are actually parametric maps resulting from the pixel-by-pixel comparison of group rCBF PET images for both conditions; the coregistered MRI scan is derived from the volunteer whose brain size and shape differs least from the standard stereotactic atlas of Talairach and Tournoux based on visual inspection. Two-dimensional cranial MRI scans of each subject are obtained using a 1.5-T G.E. Signa system. Three-millimeter thick axial slices, without interslice gap, are obtained using a multiechomultiplanar pulse sequence with the following parameters: repetition time, 400 ms; echo time, 18 ms; acquisition matrix, 256 times 196; number of echos, 2; field of view, 20 cm; spatial resolution, 7 mm FWHM.
Intersubject averaging of nitrous oxide's effects requires stereotactic normalization of individual rCBF PET images, as well as normalization of global CBF differences between subjects and experimental conditions. Such stereotactic normalization is performed by mathematical registration of PET image data from each subject to their own MRI data set using the algorithm and software described by Woods et al. [20]. Next, one subject's MRI data set is selected as a reference MRI for all the other subjects, reoriented to the AC-PC line using Ecat software (Siemens/CTI PET Systems, Knoxville, TN), and resliced to yield the same pixel size, slice thickness, and orientation as the PET images. Each of the other MRI data sets is then mathematically registered to this reference image using the transformation technique developed by Woods et al. [21]. This is followed by registration of individual PET images into the reference MRI coordinate system with matrix transformation, using the parameters for the intrasubject PET-to-MRI registration and the intersubject MRI-to-MRI transformation [21]. After conversion of each individual PET and MRI data set to this common MRI coordinate system, all PET images are normalized to an arbitrary global value and smoothed by a three-dimensional Gaussian filter, similar to the process used with SPM. The spatial and global CBF normalization facilitates averaging of rCBF pixel value differences and associated standard deviations between the conditions and across subjects. This process yields effect-size maps of nitrous oxide-associated activation and deactivation, where pixel values are derived from the mean rCBF increases and decreases, respectively, in each pixel, divided by the standard deviation of the change. Finally, the averaged functional data are superimposed on the common (anatomical) MRI, and a color scale is used to indicate the significance level of differences between conditions. z scores and P values are converted to colors to facilitate comparisons with SPM data.
Vital signs, consisting of blood pressure, heart rate, respiratory rate, arterial oxygen saturation, and endtidal carbon dioxide concentration, measured before and after each scan, were averaged across individuals for each experimental condition. These averages were compared between conditions with analysis of variance.
Results
Point analysis of rCBF and rCMR scans revealed no significant (P < 0.05) differences between the 20% nitrous oxide-induced percentage increases in randomly selected pixels in the anterior cingulate cortex images Table 1. Similarly, no significant differences were found between the 20% nitrous oxide-induced percentage decreases in randomly selected pixels in the parahippocampal gyrus, hippocampus, and posterior cingulate cortex images Table 2. The cardiovascular and respiratory physiologic parameters were not significantly affected (P < 0.05) by administration of nitrous oxide Table 3.
Areas of Regional Cerebral Activation (n = 8) and Point Analysis Comparison of Regional Cerebral Blood Flow (rCBF) and Regional Cerebral Metabolic Rate (rCMR) Percentage Increases in the Same Areas (n = 4) During Nitrous Oxide Inhalation
Areas of Regional Cerebral Deactivation (n = 8) and Point Analysis Comparison of Regional Cerebral Blood Flow (rCBF) and Regional Cerebral Metabolic Rate (rCMR) Percentage Decreases in the Same Areas (n = 4) During Nitrous Oxide Inhalation
Vital Signs
Comparison of the rCBF profiles obtained during control and nitrous oxide conditions revealed multiple areas of significant (P < 0.005) nitrous oxide-induced activation Table 1 and Figure 1A. These were localized by effect-size maps to the anterior cingulate cortex (areas 24 and 32; vertical extent: +4 to +36 mm relative to the AC-PC line) in both hemispheres. Comparison also revealed loci of significant bilateral deactivation (P < 0.001) in the posterior cingulate (areas 23, 29, and 30; vertical extent: +8 to +12 mm relative to the AC-PC line), hippocampus, parahippocampal gyrus (areas 19 and 36; vertical extent relative to the AC-PC line: -20 to +4 mm), and visual association cortices (areas 18 and 19; vertical extent relative to AC-PC line: -8 to +28 mm; Table 2 and Figure 1B and C).
Effect-size map of regional cerebral blood flow (rCBF) increases (A) and decreases (B and C) during the nitrous oxide condition, superimposed on axial slices of co-registered magnetic resonance images for anatomic reference. Maps were obtained by dividing the mean (n = 8) difference in rCBF between the control and nitrous oxide conditions by the standard deviation from the means. Representative slices were chosen to illustrate localization of the largest horizontal section of all rCBF increases and decreases. The standard stereotactic coordinates of contributing pixels, with the statistically most robust increases and decreases, are specified in Results as well as in Table 1 and Table 2, respectively. Positron emission tomography and magnetic resonance imaging pixel size, slice thickness, and orientation are the same. The color scale has been adjusted to z score minimums of 2.32 (panel A; P = 0.01) and 2.57 (panels B and C; P = 0.005) and a maximum of 3.09 (panels A, B, and C; P = 0.001). Note that the lower z score minimums allow visualization of the extent of activated and deactivated regions, while Table 1 and Table 2 show only regions with the conservative thresholds of 2.57 and 3.09, respectively. A, Image slice 32 mm, above anterior commissure-posterior commissure (AC-PC) line; B, -20 mm, below AC-PC line; C, 8 mm, above AC-PC line. AC = anterior cingulate cortex; HC = hippocampus; PHC = parahippocampal gyrus; PC = posterior cingulate cortex; VA = visual association area; L = left; R = right. Data are derived from all eight subjects.
Although subjective experiences were not measured, every volunteer reported some degree of sedation and euphoria, as well as altered perception of time and self-image, during nitrous oxide inhalation.
Discussion
This functional brain imaging study demonstrates that, although rCBF remained unaltered in most of the brain, there were certain localized brain responses noted during inhalation of 20% nitrous oxide in humans. Specifically, neuronal activation was identified in the anterior cingulate cortex bilaterally (areas 24 and 32) and deactivation was noted in the posterior cingulate (areas 23, 29, and 30), hippocampus, and parahippocampal (areas 19 and 36) and visual association (areas 18 and 19) cortices. Radioactivity dosimetry constraints precluded repeated PET measurements; thus, the results are limited to responses observed with a single concentration of nitrous oxide.
These findings contrast with the large, global cerebral metabolic rate (CMR) changes associated with higher concentrations of nitrous oxide by metabolic brain mapping in animals [6]. In addition to the presence of other drugs [5,7] and a possible species-related confounder, methodologic differences prevent further useful comparisons between these studies. For example, since mean absolute CMR was used to map neuronal activity in these experiments [5-7], the variance in global CMR among individuals might eclipse any smaller regional changes. Nitrous oxide 70% has been reported variously to increase CMR significantly [22] or not [23], but it consistently increased CBF in these same studies to a similar degree, suggesting that flow and metabolism diverge with higher concentrations. In contrast, during 20% nitrous oxide inhalation, we found that rCBF and rCMR remain tightly coupled, validating that the rCBF changes associated with nitrous oxide administration properly reflect altered cerebral activity. Finally, it is unlikely that a physiological artifact stemming from changes in arterial carbon dioxide levels [24] could account for the results, since end-tidal carbon dioxide never changed.
Nitrous oxide 20% activated neurons in the anterior cingulate cortex. Recent human PET studies by our group [25] and others [26] have demonstrated a correlation between opioid-related anterior cingulate activation and pain relief, suggesting this area's involvement in opioid analgesia. The same might be true for nitrous oxide analgesia [27], since nitrous oxide acts to release methionine-enkephalin, beta-endorphin, and other opioid peptides [28,29]. Several recent PET studies indicate that the anterior cingulate is also activated during cognitive tasks that involve words, spatial objects, or motor learning [30-32]. Thus, perturbations at this locus could lead to the psychomotor and cognitive impairment observed in humans during inhalation of even low nitrous oxide concentrations [2,3].
Activation of the anterior cingulate could be due to either local modulation or activation of afferents to the cingulate. The two primary limbic afferents to the anterior cingulate, the posterior cingulate and hippocampus, are deactivated by nitrous oxide, but increased activity could still result from activation of ventral tegmental area afferents as was demonstrated in animal experiments [33].
Nitrous oxide inhalation was also associated with deactivation in the posterior cingulate and the surrounding visual associative cortices. The posterior cingulate has been implicated in visual learning and memory, based on electrophysiological [34] and lesion studies [35]. Human PET studies have documented activation in the retrosplenial cortex during visual pattern recognition tests [36] and memorization of historical dates [37]. Thus, nitrous oxide-induced impairment of visual [38], word [4], and instruction recall [39] may be related to deactivation in the areas we describe.
Although few studies have addressed nitrous oxide's mechanism of action on a neural circuit level, there are data suggesting that depression of visual afferents by nitrous oxide might account for deactivation of the visual association cortex. Extracellular single-unit recordings [40] show that nitrous oxide proportionally decreases the firing of excitatory cells in the primary somatosensory and visual cortices and corresponding thalamic relay nuclei and increases the activity of inhibitory neurons in the same areas. It follows that net metabolic activity would not change in these areas, rendering them invisible to PET. The next level of processing, however, would demonstrate decreased activity due to suppressed afferent input, which, in principle, would be readily detectable by PET. Our findings generally support this mechanism. Specifically, subjects' gazes were fixed at a crosshair during scanning, but the room was otherwise quiet and comfortable, so the main sensory input was visual. The net metabolic activity of the thalamus (corpus geniculatum laterale) and primary visual cortex underwent little change, whereas that of the next processing level (i.e., visual associative cortex) decreased. The latter area is linked to the posterior cingulate cortex via extensive anatomic connections [41], which could account for our observation of deactivation.
The most significant decrease in activity we found, however, was in the hippocampus and parahippocampal gyrus (areas 19 and 36) bilaterally. Multiple lines of evidence indicate that these areas are integral to memory. In human PET studies, the hippocampal area is activated when subjects memorize the shape and color of a visual stimulus [42], as well as during tasks testing declarative memory [43]. Depression in these regions by nitrous oxide may well explain the memory impairment, such as word free recall and recognition of emotional words, noted with this drug.
Depression of hippocampal activity by nitrous oxide could be primary or secondary to depression of afferent circuits. The principal afferents to the hippocampus are the septum and the entorhinal cortex. Selective ablation of the medial septal nucleus, although it abolishes hippocampal electroencephalographic activity [44], increases anterior cingulate activity in a pattern similar to that produced by subanesthetic nitrous oxide inhalation. Based on this, one pathway that could mediate hippocampal deactivation by nitrous oxide is the medial septal system. Alternatively, nitrous oxide could also locally depress hippocampal inhibitory and excitatory synaptic activity similarly to potent volatile anesthetics [45].
In summary, we report localized and specific neuronal activity changes in human limbic areas during inhalation of a low concentration of nitrous oxide. These areas are known to be involved in pain processing, reinforcement, psychomotor activity, cognitive function, visuospatial and semantic learning, and memory. These limbic areas may be the neuroanatomic sites of behavioral impairment during nitrous oxide inhalation.
We thank Marsha Dachille, Donna Milko, James Ruszkiewicz, Louise Smith, Don Sashin, Tom Nichols, and Norman Simpson of the UPMC PET Facility for technical assistance; Drs. Jan Smith, Barbara DeRiso, and our many colleagues at UPMC's Department of Anesthesiology for helping to arrange research time; Lisa Cohn for editorial comments; Melissa Sampson for research accounting; and Sandra Higgins for preparation of the manuscript.
REFERENCES
1. Marshall DE, Longnecker DE. General anesthetics. In: Gilman AG, Rall TW, Nies AS, Taylor P, eds. Goodman and Gilman's The pharmacological basis of therapeutics. New York: Pergamon, 1990:298-300.
2. Korttila K, Ghoneim MM, Jacobs L, et al. Time course of mental and psychomotor effects of 30 per cent nitrous oxide during inhalation and recovery. Anesthesiology 1981;54:220-6.
3. McMenemin IM, Parbrook GD. Comparison of the effects of subanaesthetic concentrations of isoflurane or nitrous oxide in volunteers. Br J Anaesth 1988;60:56-63.
4. Block RI, Ghoneim MM, Hinrichs JV, et al. Effects of a subanaesthetic concentration of nitrous oxide on memory and subjective experience: influence of assessment procedures and types of stimuli. Hum Psychopharmacol 1988;3:257-65.
5. Cole DJ, Shapiro HM. Different 1.2 MAC combinations of nitrous oxide-enflurane cause unique cerebral and spinal cord metabolic responses in the rat. Anesthesiology 1989;70:787-92.
6. Crosby G, Crane AM, Sokoloff L. A comparison of local rates of glucose utilization in spinal cord and brain in conscious and nitrous oxide- or pentobarbital-treated rats. Anesthesiology 1984;61:434-8.
7. Sakabe T, Tsutsui T, Maekawa T, et al. Local cerebral glucose utilization during nitrous oxide and pentobarbital anesthesia in rats. Anesthesiology 1985;63:262-6.
8. Kety SS, Schmidt CF. The nitrous oxide method for the quantitative determination of cerebral blood flow in man; theory, procedures and normal values. J Clin Invest 1948;27:476-83.
9. Herscovitch P, Markham J, Raichle ME. Brain blood flow measured with intravenous H215 O. I. Theory and error analysis. J Nucl Med 1983;14:782-9.
10. Meyer E.15 O studies with PET. In: Diksic M, Reba RC, eds. Radiopharmaceuticals and brain pathology studies with PET and SPECT. Boca Raton, FL: CRC Press, 1991:166-97.
11. Sokoloff L, Reivich M, Kennedy C, et al. The [(14) C]deoxyglucose method for the measurement of local cerebral glucose utilization: theory, procedure, and normal values in the conscious and anesthetized albino rat. J Neurochem 1977;28:897-916.
12. Phelps ME, Huang SC, Hoffman EJ, et al. Tomographic measurement of local cerebral glucose metabolic rate in humans with (F-18)2-fluoro-2-deoxy-D-glucose: validation of method. Ann Neurol 1979;6:371-88.
13. Raichle ME. Circulatory and metabolic correlates of brain function in normal humans. In: Mountcastle VB, Plum F, eds. Handbook of physiology, section I. The nervous system. V. Higher functions of the nervous system, part 2. Bethesda, MD: American Physiological Society, 1987.
14. Raichle ME, Martin WRW, Herscovitch P, et al. Brain blood flow measured with intravenous H215 O. II. Implementation and validation. J Nucl Med 1983;24:790-8.
15. Huang SC, Phelps ME, Hoffman EJ, et al. Noninvasive determination of local cerebral metabolic rate of glucose in man. Am J Physiol 1980;238:E69-92.
16. Woods RP, Cherry SR, Mazziotta JC. Rapid automated algorithm for aligning and reslicing PET images. J Comput Assist Tomogr 1992;16:620-33.
17. Friston KJ, Frith CD, Little PF, et al. The relationship between global and local changes in PET scans. J Cereb Blood Flow Metab 1990;10:458-66.
18. Friston KJ, Passingham RE, Nutt JG, et al. Localization of PET images: direct fitting of the intercommissural (AC-PC) line. J Cereb Blood Flow Metab 1991;9:690-5.
19. Talairach J, Tournoux P. Co-planar stereotactic atlas of the human brain. 3-dimensional proportional system: an approach of cerebral imaging. New York: Theime Medical Publishers, 1988:10-55.
20. Woods RP, Mazziotta JC, Cherry SR. MRI-PET registration with automated algorithm. J Comp Assist Tomogr 1993;17:536-46.
21. Woods RP, Mazziotta JC, Cherry SR. Automated image registration. In: Uemura K, Lassen NA, Jones T, et al., eds. Quantification of brain function: tracer kinetics and image analysis in brain PET. London: Elsevier Science Publishers, 1993:391-400.
22. Pelligrino DA, Miletich DJ, Hoffman WE. Nitrous oxide markedly increases cerebral cortical metabolic rate and blood flow in the goat. Anesthesiology 1984;60:405-12.
23. Kaieda R, Todd MM, Warner DS. The effects of anesthetics and PaCO (2) on the cerebrovascular, metabolic, and electroencephalographic responses to nitrous oxide in the rabbit. Anesth Analg 1989;68:135-43.
24. Dubbink DA. Physiologic effects of hyper- and hypocarbia. In: Faust RJ, ed. Anesthesiology review. New York: Churchill Livingstone, 1991:24-6.
25. Adler L, Firestone L, Mintun M, Winter P. Central mechanisms of pain and opioid analgesia elucidated by positron emission tomography [abstract]. Anesthesiology 1994;81:A917.
26. Jones AK, Friston KJ, Qi LY, et al. Sites of action of morphine in the brain. Lancet 1991;338:825.
27. Chapman CR, Benidetti C. Nitrous oxide effects on cerebral evoked potential to pain: partial reversal with a narcotic antagonist. Anesthesiology 1979;51:135-8.
28. Quock RM, Kouchich FJ, Tseng L-F. Does nitrous oxide induce release of brain opioid peptides? Pharmacology 1985;30:95-9.
29. Zuniga JR, Joseph SA, Knigge KM. The effects of nitrous oxide on the central endogenous pro-opiomelanocortin system in the rat. Brain Res 1987;420:57-65.
30. Pardo JV, Pardo PJ, Janer KW, Raichle ME. The anterior cingulate cortex mediates processing selection in the Stroop attentional conflict paradigm. Proc Natl Acad Sci USA 1990;87:256-9.
31. Corbetta M, Meizin FM, Dobmeyer S, et al. Selective and divided attention during visual discriminations of shape, color, and speed: functional anatomy by positron emission tomography. J Neurosci 1991;11:2385-402.
32. Raichle ME. Exploring the mind with dynamic imaging. Semin Neurosci 1990;2:307-15.
33. Porrino LJ, Esposito RU, Seeger TF, et al. Metabolic mapping of the brain during rewarding self-stimulation. Science 1984;224:306.
34. Olson CR, Musil SY, Goldberg ME. Posterior cingulate cortex and visuospatial cognition: properties of single neurons in the behaving monkey. In: Vogt BA, Gabriel M, eds. Neurobiology of cingulate cortex and limbic thalamus. Boston: Birkheauser, 1993.
35. Murray EA, Davidson M, Gaffan D, et al. Effects of fornix transection and cingulate cortical ablation on spatial memory in rhesus monkeys. Exp Brain Res 1989;74:173-86.
36. Roland PE, Gulyas B, Seitz RJ, et al. Functional anatomy of storage, recall, and recognition of a visual pattern in man. Neuroreport 1990;1:53-6.
37. Pawlik G, Heiss WD, Beil C, et al. Three-dimensional patterns of speech-induced cerebral and cerebellar activation in healthy volunteers and in aphasic stroke patients studied by positron emission tomography of (18) F)-fluorodeoxyglucose. In: Meyer JS, Lechner H, Reivich M, Ott EO, eds. Cerebral vascular disease. New York: Excerpta Medica, 1987:207-10.
38. Vallerand WP, Vallerand AH, Hall MB. Evaluation of the sedative combination of nitrous oxide and orally administered diazepam. Oral Surg Oral Med Oral Pathol 1991;71:660-4.
39. File SE, Balakrishnan J, Murray A, et al. Effects of nitrous oxide on memory for instructions in dental patients. Hum Psychopharmacol 1992;7:37-44.
40. Angel A. The GL Brown lecture. Adventures in anaesthesia. Exp Physiol 1991;76:1-38.
41. Vogt BA, Sikes RW, Swadlow HA, Weyand TG. Rabbit cingulate cortex: cytoarchitecture, physiological border with visual cortex, and afferent cortical connections of visual, motor, postsubicular, and intracingulate origin. J Comp Neurol 1986;248:74-94.
42. Corbetta M, Miezin FM, Dobmeyer S, et al. Attentional modulation of neural processing of shape, color, and velocity in humans. Science 1990;248:1556-9.
43. Squire LR, Ojemann JG, Miezin FM, et al. Activation of the hippocampus in normal humans: a functional anatomical study of memory. Proc Natl Acad Sci USA 1992;89:1837-41.
44. Borst JG, Leung L-WS, Macfabe DF. Electrical activity of the cingulate cortex. II. Cholinergic modulation. Brain Res 1987;407:81-93.
45. Yoshimura M, Higashi H, Fujita S, et al. Selective depression of hippocampal inhibitory postsynaptic potentials and spontaneous firing by volatile anesthetics. Brain Res 1985;340:363-8.
© 1996 International Anesthesia Research Society