BJA Advance Access published online on November 23, 2007
British Journal of Anaesthesia, doi:10.1093/bja/aem334
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Regional cerebral metabolic rate (positron emission tomography) during inhalation of nitrous oxide 50% in humans
1 Department of Anaesthesiology and Intensive Care
2 Department of Clinical Pharmacology
3 Department of Clinical Neurophysiology
4 Department of Radiation Physics
5 Department of Neurosurgery, University Hospital, S22185 Lund, Sweden
* Corresponding author: Department of Anaesthesia and Intensive Care, University Hospital, S22185 Lund, Sweden. E-mail: peter.reinstrup{at}med.lu.se
Accepted for publication August 27, 2007.
| Abstract |
|---|
|
|
|---|
Background: Recent studies in man have shown that cerebral blood flow increases during inhalation of nitrous oxide (N2O), a finding which is believed to be a result of an increased cerebral metabolic rate (CMR). However, this has not previously been evaluated in man.
Methods: Regional CMRglu (rCMRglu) was measured three dimensionally with positron emission tomography (PET) after injection of 2-(18F)fluoro-2-deoxy-D-glucose in 10 spontaneously breathing men (mean age 31 yr) inhaling either N2O 50% in O2 30% or O2 30% in N2.
Results: Global CMRglu in young men was 27 (3) µmol 100 g–1 min–1 [mean (SD)]. Inhalation of N2O 50% did not change global CMRglu [30 (5) µmol 100 g–1 min–1] significantly, but it changed the distribution of the metabolism in the brain (P<0.0001 analysis of variance). Compared with inhalation of O2 30% in N2, N2O 50% inhalation increased the metabolism in the basal ganglia [14 (17)%, P<0.05] and thalamus [22 (23) %, P<0.05]. There was a prolonged metabolic effect of N2O inhalation seen on a succeeding PET scan with oxygen-enriched air (P<0.0001) performed 1 h after the N2O administration.
Conclusions: Inhalation of N2O 50% did not change global CMRglu, but the metabolism increased in central brain structures, an effect that was still present 1 h after discontinuation of N2O.
Keywords: anaesthetics, gases, N2O; brain, cerebral metabolic rate, regional cerebral metabolic rate; measurement technique, PET
| Introduction |
|---|
|
|
|---|
Nitrous oxide (N2O) has been used for anaesthesia during neurosurgical procedures for half a century, as it was previously thought to be inert.1 However, in the 1970s, it was shown that N2O had potentially striking effects on intracranial pressure (ICP).2 3 The literature regarding the effects of N2O on the brain is still equivocal, probably due to species differences, but also because of interactions with other drugs or interventions. In humans, evidence supporting the conclusion that N2O is a cerebral vasodilator in the absence of other interventions has been obtained from both two- and three-dimensional cerebral blood flow (CBF) studies.4–6 Three-dimensional CBF measurements during N2O 50% inhalation show that CBF increased in all regions, even though not evenly distributed in the brain.6 The reason for this increase in flow is still unknown, but it may theoretically be due to an increased cerebral metabolism7 or an influence of N2O on the cerebral vessels,8 for instance through release of vasoactive mediators.
In order to explore the former hypothesis, the aim of the present study was designed to evaluate the effect of N2O 50% on the cerebral metabolic rate for glucose (CMRglu) and its distribution using a three-dimensional positron emission tomography (PET) technique.
| Methods |
|---|
|
|
|---|
Ten male volunteers, age 25–40 yr (mean 31 yr), participated in the study. The ethics committee for human studies and the isotope committee at the University of Lund approved the protocol. Written informed consent was obtained from each participant.
The participants breathed spontaneously into a facemask held in place by a rubber head strap. After eliminating air leaks, the volunteers inhaled one of two different gas mixtures starting 15 min before and continued 35 min after the injection of 2-(18F)fluoro-2-deoxy-D-glucose (2-18FDG). The inhaled gas mixtures were oxygen-enriched air (O2 30%) and a mixture of N2 20%, O2 30%, and N2O 50%. Oxygen-enriched air was compressed air and O2 mixed with flow meters (Unit 760, Siemens, Elema, Solna, Sweden) whereas N2O 50% with O2 30% in N2 was delivered as a pre-mixed precision gas produced by Alfax (Malmö, Sweden). Each subject served as it own control, starting by either inhaling N2O 50% or O2 30% in a randomized order succeeded by the other gas mixture. A PET scan was performed in each situation within 60 min after the FDG injections. The results were evaluated after six investigations, and as the three subjects who inhaled N2O 50% first had remaining effects of N2O on the succeeding PET scans, all the rest of the investigations (four) were done by starting with O2 30%. The data from the three subjects are included in the physiological values, but the results from their PET scanning are presented separately, and none of these data was included in the main study. Two volunteers breathed air twice in order to control the validity of the method. One participant had discomfort by inhaling N2O and no data from this subject are included in any part of the study.
PET examinations were performed with a PC384-7B scanner (Scanditronix, Uppsala, Sweden). The four ring detector system generated four main slices and three cross slices, with an approximate slice thickness of 13 mm. The mean sensitivity of the scanner was 675 cps (kBq ml–1)–1 for the main slices and 945 cps (kBq ml–1)–1 for the cross slices. Corrections for random and scattered events were made. The attenuation was calculated from contours using a soft-tissue attenuation coefficient.
The 2-18FDG was synthesized using18F from the Lund electrostatic tandem accelerator with a proton beam of 6.0 MeV and produced by the 18O(p,n)18F reaction,9 and by fluorination of the precursor 1,3,4,6-tetra-O-acetyl-2-O-trifluoro-methane-b-D-mannopyranose according to a method described by Toorongian and colleagues.10 The radiochemical purity of FDG was better than 95%.
PET images were acquired within 60 min after administration of 50–150 MBq FDG, with the highest dose at the second measurement, by a 2 min i.v. injection into a peripheral vein in the arm. Venous blood samples were drawn from the other arm at appropriate intervals for quantification (every 20 s for 4 min, every 60 s for 10 min, and every 5 min up to a total of 35 min). When the blood samples had been centrifuged, radioactivity in the plasma (Cp) was measured using a well counter. Blood glucose (Cg) values were obtained, from the same site, immediately before, 10 and 35 min after injection of FDG. Two FDG studies were made on each subject with a time interval of approximately 2 h. The three-dimensional metabolic rate of glucose in the brain was recorded, parallel to the orbito-meatal (OM) line, with the centre of the lowest slice located 1 cm below the OM line. The head position was controlled using light beams on the external auditory meatus and the nasion as landmarks. Each study consisted of two separate scans (2x10 min), in between which the couch was moved half a plane separation (6.75 mm) in order to get a higher axial sampling rate. The CMRglu were calculated using the test–retest method developed by Brooks and colleagues,11 which takes into account the residual activity from the first investigation in the calculation of the second.
A number of blood samples were taken during a 35 min period after each FDG-injection, and the activity in plasma was measured. These curves were extrapolated by fitting a monoexponential function to the tail of the curve (the last 25 min). The first extrapolated curve was adjusted to match a sample taken immediately before the second injection. An example of a plasma activity curve is shown in Figure 1.
|
Tomographic images were reconstructed by filtered back projection using a Hanning reconstruction filter with a cut-off frequency of 2.1 cm–1. The image matrix size was 128x128 and the pixel size 2.2 mm.
The calculation of the second measurement demanded that the images from the two studies were perfectly registered. This was accomplished first by carefully positioning the subjects in the same way for the two studies using line-laser for positioning, and secondly by shifting and rotating the images interactively.
Standard sets of regions of interest (ROIs) corresponding to the brain lobes, cerebellum, pons, thalamus, and basal ganglia were analysed by a ROI analysis program (Amersham, UK). The program works with anatomical templates from a CT brain atlas.12 The templates were applied to the brain slices and scaled to the actual head size according to the external brain diameters. Three-dimensional cerebral ROIs were calculated from adjoining ROIs in different brain slices representing the same structure. With the aid of the ROIs, the mean CMRglu in different brain regions and the global mean CMRglu was calculated.
The quantities of CO2, O2, and N2O in the inspiratory and expiratory gas mixtures were continuously measured along with the pulse rate and arterial haemoglobin oxygen saturation (SaO2), with an Ohmeda 4700 OxiCap (BOC Health Care, Louisville, KY, USA). Non-invasive arterial pressure was recorded with 5 min intervals (Colin Press-Mate, Colin Electronics, Japan) during the inhalation of gas. At the end of the study, a blood sample was withdrawn to measure the haemoglobin concentration.
Global effects of N2O on the CMRglu were analysed by two factor repeated measures analysis of variance (ANOVA) with correction for departure from sphericity for the interaction data.13 As the investigations were test–retest situations with residual activity from the first investigation, this activity had to be subtracted analysing the second measurement. The increased noise level of the subtraction method, in combination with the limited number of subjects, will limit our findings to major CMRglu changes in the order of 10–15% or more, whereas smaller changes will not be detected.
The effects of N2O on the rCMRglu (ROI) was tested with primary component analysis (PCA) in order to identify functionally coupled regions followed by repeated measures ANOVAs of the factor scores. In each ANOVA, the effect of N2O on the rCMRglu (in per cent of mean) was tested. Post hoc testing with paired Student's t-test (two-tailed) was performed when the ANOVA indicated significant effects in order to clarify which regional effect mainly contributed to the significance.
The relative change in distribution was calculated as the differences between the relative rCMRglu values for measurements, with and without N2O.
Physiological data were analysed by two factor repeated measures ANOVA with correction for departure from sphericity for the interaction data.18 Significant ANOVA interactions were further regionally analysed with Student's paired t-test.
P
0.05 was considered statistically significant. All values are given as mean (SD). All statistical analyses were performed using the Stat View program version 5.0.1 for Windows (SAS Institute, Cary, NC, USA).
| Results |
|---|
|
|
|---|
Physiological values for inhalation of oxygen-enriched air and for inhalation of N2O 50% is presented in Table 1. The mean haemoglobin value was 140 g litre–1, range 120–154 g litre–1.
|
All volunteers experienced psychogenic changes in cerebral function during N2O inhalation with reduction in wakefulness, vision, hearing, and touch, whereas there was an increase in impulsiveness.
The volunteers had a global CMRglu of 27 (3) µmol 100 g–1 min–1 when breathing oxygen-enriched air. Inhalation of N2O 50% did not change global CMRglu 30 (5) µmol 100 g–1 min–1 significantly, whereas it induced a significant change in the distribution (F=7.7, P<0.0001) (Fig. 2, Table 2). PCA analysis identified the factors explaining 89% of the variance in the ROIs. These were high negative loading in frontal, parietal, and temporal cortex vs positive loading in pons; high negative loading in cerebellum and positive in thalamus as well and basal ganglia; and finally high negative loading in the occipital region. Only high negative loading in cerebellum and positive in thalamus as well and basal ganglia reached the level of significance (P<0.0001).
|
|
Comparing the individual ROIs, N2O significantly increased metabolism in thalamus by 22 (23%) (t=2.59, P<0.05) and basal ganglia by 14 (17)% (t=2.20, P<0.05) (Fig. 3).
|
In the different ROIs, inhalation of N2O 50% changed the metabolism as shown in Figure 3. Representative PET and SPECT scans (CBF) under the same conditions with and without N2O are presented in Figure 4.
|
There was a significant difference of the results whether N2O or oxygen-enriched air was inhaled first (F=5.48, P<0.0001). In the three volunteers who started with N2O inhalation, the CMRglu was 30 (4) µmol 100 g–1 min–1 during N2O inhalation and 32 (2) µmol 100 g–1 min–1 when air was inhaled. The effects on the different ROIs are presented in Figure 5.
|
| Discussion |
|---|
|
|
|---|
During inhalation of oxygen-enriched air, the global CMRglu of the human brain was 27 µmol 100 g–1 min–1 that is in accordance with former similar studies.14 In the present study, inhalation of N2O 50% in healthy young men resulted in only a minor and non-significant increase in global CMRglu. This is in fact not a surprising finding as no human study has demonstrated any increase in CMR due to N2O inhalation; previous studies were limited by the fact that another anaesthetic drug was present during the investigation15 16 and which has been used to explain the lack of effect of N2O on CMR.17 In the study by Kaisti and colleagues,16 where adjunct of N2O counteracts the CMRO2 and CBF reduction by propofol and sevoflurane, the amounts of propofol and sevoflurane given is lowered when N2O was added, in order to keep the anaesthetic level constant. The effect could be due to a reduction in the basic anaesthesia or an increase in CMR due to the N2O addition. In support of this assumption, N2O generally increased the CMR in animals if administered alone,7 but this increase in metabolism could be reduced or abolished by augmenting the anaesthetic level through addition of other anaesthetics. The animal studies, together with the fact that N2O changes the EEG pattern in humans,15 18 19 have been used to postulate that N2O increases CMR also in man, which our study does not support.
Compared with inhalation of oxygen-enriched air, N2O 50% inhalation made the participants hyperventilate with a significant lowering of the E'CO2 by 0.5 kPa (4 mm Hg), which is in accordance with other studies.4 5 It is generally accepted that changes in CO2 affect the CBF, but not CMR which remains unaffected of CO2 changes within arterial CO2 levels of 2–10 kPa (15–80 mm Hg).20 One possible reason for the reduced effect of N2O on the CMRglu, as suggested by Stapleton and colleagues,21 can be explained by an adaptation to lower anxiety during the second PET scan. However, most of the participants in the present study had also participated in a former study evaluating the effect of N2O on the rCBF,6 which would be expected to relieve the anxiety effect of these participants. Furthermore, N2O 50% inhalation has profound psychogenic effects,4–6 making it difficult to extrapolate the findings of Stapleton and colleagues21 to our experimental condition. We also noted that participants who started with inhalation of N2O 50% tended to increase global CMRglu during the succeeding inhalation of oxygen-enriched air. As volunteers inhaling air twice did not change their metabolism, this observation is best explained by residual effects of N2O on the succeeding CMRglu, and does not favour the view that CMR should generally be decreased at the second measurement.
Inhalation of N2O 50% has been reported to increase global CBF by 22–40% in man.4–6 This effect is mainly believed to be due to the mental activation by N2O as changes in rCBF should normally be directly coupled to changes in rCMR.22 In this study, inhalation of N2O 50% induced a non-significant increase in the global CMRglu of 10% (Fig. 4). This contrasts to the earlier findings where CBF increased by 22–40%, indicating that there might be a direct or indirect effect on the cerebral arteries in addition to the cerebral metabolic effect. In support of this theory, addition of N2O to an isoflurane anaesthesia during normocapnia resulted in an increased CBF without any influence on the rCBF pattern created by isoflurane.23 This uncoupling between CBF and CMR has not been found to be due to a direct effect on the human cerebral vessels by N2O,6 opening up for alternative explanations to this discrepancy.
Inhalation of N2O altered the regional distribution of CMRglu, which increased in thalamus and basal ganglia. Utilizing the same technique to identify ROIs, N2O inhalation was found to augment rCBF in most brain regions, although thalamus and basal ganglia were among the regions with the greatest change.6 Even though we do not find a complete match between flow and metabolism, such conformity has been verified for inhalation of N2O 20%.24 The correlation between flow and metabolism may therefore explain the regional cerebral flow pattern found during N2O inhalation,6 which, however, is generally increased (uncoupled) due to the higher concentration of N2O inhaled. The flow pattern gave the impression that N2O inhalation increased flow through regions anatomically associated with the limbic system.25 In support of the theory of limbic system activation by N2O, we have previously reported6 that N2O stimulated thoughts and emotions, which normally are dealt with by the limbic system. In addition, former reports have shown an increased EEG activity in the limbic regions in cats26 and in humans27 28 during N2O inhalation.
In conclusion, inhalation of N2O 50% resulted in a small and non-significant increase in global CMRglu. This effect on the global metabolism was mainly attributed to an increased metabolism in thalamus and basal ganglia. The metabolic effect of N2O may linger in the brain more than 1 h after clinical recovery from anaesthesia.
| Funding |
|---|
|
|
|---|
Tore Nilssons foundation for medical research; Swedish Medical Research Council (B91-14X-00084-27A); Malmöhus läns landsting; Research founds of the University of Lund; AGA AB Medical Scientific Foundation, Stockholm, Sweden.
| References |
|---|
|
|
|---|
1 Smith AL, Wohllman H. Cerebral blood flow and metabolism. Effects of anesthetics drugs and techniques. Anesthesiology (1972) 36:378–400.[CrossRef][Web of Science][Medline]
2 Henriksen HT, Jørgensen PB. The effect of nitrous oxide on intracranial pressure in patients with intracranial disorders. Br J Anaesth (1973) 45:486–92.
3 Moss E, McDowall DG. I.C.P. increases with 50% nitrous oxide in oxygen in severe head injuries during controlled ventilation. Br J Anaesth (1979) 51:757–61.
4 Deutsch G, Samra SK. Effects of nitrous oxide on global and regional cortical blood flow. Stroke (1990) 21:1293–8.
5 Field LM, Dorrance DE, Krzeminska EK, Barsoum LZ. Effect of nitrous oxide on cerebral blood flow in normal humans. Br J Anaesth (1993) 70:154–9.
6 Reinstrup P, Ryding E, Algotsson L, Berntman L, Uski T. Effects of nitrous oxide on human regional CBF (SPECT) and isolated pial arteries. Anesthesiology (1994) 81:396–402.[Web of Science][Medline]
7 Pelligrino DA, Miletich DJ, Hoffman WE, Albrecht RF. Nitrous oxide markedly increases cerebral cortical metabolic rate and blood flow in the goat. Anesthesiology (1984) 60:405–12.[CrossRef][Web of Science][Medline]
8 Fitzpatrick JH, Gilboe DD. Effects of nitrous oxide on the cerebrovascular tone, oxygen metabolism, and electroencephalogram of the isolated perfused canine brain. Anesthesiology (1982) 57:480–4.[CrossRef][Web of Science][Medline]
9 Ohlsson T, Sandell A, Hellborg R, Håkansson K, Nilsson C, Strand S-E. Clinical useful quantities of [18F]fluoride produced by 6 MeV proton irradiation of a H218O target. Nucl Instr Meth A (1996) 379:341–2.[CrossRef]
10 Toorongian SA, Mulholland GK, Jewett DM, Bachelor MA, Kilbourn MR. Routine production of 2-deoxy-2-[18F]fluoro-D-glucose by direct nucleophilic exchange on a quaternary 4-aminopyridinium resin. Int J Rad Appl Instrum B (1990) 41:273–9.
11 Brooks RA, Di Chiro G, Zukerberg BW, Bairamian D, Larson SM. Test-retest studies of cerebral glucose metabolism using fluorine-18 deoxyglucose: validation of method. J Nucl Med (1987) 28:53–9.
12 Kretschmann HJ, Weirich W. Neuroanatomy and Cranial Computed Tomography (1986) Stuttgart: Thieme Verlag. 92–6.
13 Kirk RI. Experimental Design (1982) Monterey, CA: Brooks/Cole publishing company. 259–62.
14 Alkire MT, Haier RJ, Shah NK, Anderson CT. Positron emission tomography study of regional cerebral metabolism in humans during isoflurane anesthesia. Anesthesiology (1997) 86:549–57.[CrossRef][Web of Science][Medline]
15 Algotsson L, Messeter K, Rosén I, Holmin T. Effects of nitrous oxide on cerebral hemodynamics and metabolism during isoflurane anaesthesia in man. Acta Anaesthesiol Scand (1992) 36:46–52.[Web of Science][Medline]
16 Kaisti KK, Långsjö JW, Aalto S, et al. Effects of sevoflurane, propofol, and adjunct nitrous oxide on regional cerebral blood flow, oxygen consumption, and blood volume in humans. Anesthesiology (2003) 99:603–13.[CrossRef][Web of Science][Medline]
17 Baughman VL. N2O: of questionable value. (Editorial). J Neurosurg Anesthesiol (1995) 7:79–81.[CrossRef][Web of Science][Medline]
18 Hoffman WE, Charbel FT, Edelman G, Albrecht RF, Ausman JI. Nitrous oxide added to isoflurane increases brain artery blood flow and low frequency brain electrical activity. J Neurosurg Anesth (1995) 7:82–8.[CrossRef][Web of Science][Medline]
19 Yli-Hankala A, Lindgren L, Porkkala T, Jantti V. Nitrous oxide-mediated activation of the EEG during isoflurane anaesthesia in patients. Br J Anaesth (1993) 70:54–7.
20 Carlsson A, Holmin T, Lindqvist M, et al. Effects of hypercapnia and hypocapnia on tryptofan and tyrosine hydroxylation in rat brain. Acta Neurol Scand (1977) 99:503–9.
21 Stapleton JM, Morgan MJ, Liu X, et al. Cerebral glucose utilization is reduced in second test session. J Cereb Blood Flow Metab (1997) 17:704–12.[CrossRef][Web of Science][Medline]
22 Raichle ME, Grubb RL, Gado MH, et al. Correlation between regional cerebral blood flow and oxidative metabolism. Arch Neurol (1976) 8:523–6.
23 Reinstrup P, Ryding E, Algotsson L, Berntman L, Uski T. Regional cerebral blood flow (SPECT) during anesthesia with isoflurane and N2O in humans. Br J Anaesth (1997) 78:407–11.
24 Gyulai FE, Firestone LL, Mintum MA, Winter PM. In vivo imaging of human limbic responses to nitrous oxide inhalation. Anesth Analg (1996) 83:291–8.[Abstract]
25 Gyulai FE, Firestone LL, Mintun MA, Winter PM. In vivo imaging of nitrous oxide-induced changes in cerebral activation during noxious heat stimuli. Anesthesiology (1997) 86:538–48.[CrossRef][Web of Science][Medline]
26 Winters WD, Mori K, Spooner CE, Bauer RO. The neurophysiology of anesthesia. Anesthesiology (1967) 28:65–80.[Web of Science][Medline]
27 Ferrer-Allado T, Brechner VL, Dymond A, Cozen H, Crandall P. Ketamine-induced electroconvulsive phenomena in the human limbic and thalamic regions. Anesthesiology (1973) 38:333–44.[Web of Science][Medline]
28 Babb TL, Ferrer-Brechner T, Brechner VL, Crandall PH. Limbic neuronal firing rates in man during administration of nitrous oxide–oxygen or sodium thiopental. Anesthesiology (1975) 43:402–9.[Web of Science][Medline]
![]()
CiteULike
Connotea
Del.icio.us What's this?
This article has been cited by other articles:
![]() |
S. Rex, P. T. Meyer, J.-H. Baumert, R. Rossaint, M. Fries, U. Bull, and W. M. Schaefer Positron emission tomography study of regional cerebral blood flow and flow-metabolism coupling during general anaesthesia with xenon in humans Br. J. Anaesth., May 1, 2008; 100(5): 667 - 675. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||





