ORIGINAL ARTICLE 
Annals of Nuclear Medicine Vol. 9, No.l, 15-21 l995 

Regional cerebral blood flow, blood volume, oxygen extraction fraction, and oxygen utilization rate in normal volunteers measured by the autoradiographic technique and the single breath inhalation method 

Jun HATAZAWA, Hideaki FUJITA, Iwao KANNO, Takao SATOH, Hidehiro IIDA, Shuhichi MIURA, Matsutaroh MURAKAMI, Toshio OKUDERA, Atsushi INUGAMI, Toshihide OGAWA, Eku SHIMOSEGAWA, Kyo NOGUCHI, Yasuaki SHOHJI and Kazuo UEMURA 

Department of Radiology and Nuclear Medicine, Akita Research Institute of Brain and Blood Vessels 

By means of a high resolution PET scanner, the regional cerebral blood flow (rCBF), cerebral blood volume (rCBV), oxygen extraction fraction (rOEF), and metabolic rate of oxygen (rCMRO2) for major cerebral gyri and deep brain structures were studied in eleven normal volunteers during an eye-covered and ear-unplugged resting condition. Regional CBF was measured by the autoradiographic method. after intravenous administration of H2^15O. Regional OEF and rCMR02 Were measured by the single inhalation of 15O2. With MR T1-weighted images as an anatomical reference, thirteen major cerebral gyri, caudate nucleus, lentiform nucleus, thalamus, midbrain, pons, cerebellum and vermis were defined on the CMRO2 images. Values were read by using circular regions of interest 16 mm in diameter. The posterior part of the cingulate gyri had the highest rCBF and rCMRO2 Values among brain structures, followed by the lentiform nucleus, the cerebellum, the caudate nucleus, and the thalamus. Parahippocampal gyri had the lowest rCBF and rCMR02 values among the cortical gyri. Regional OEF for the pontine nuclei (0.34 +- 0.04), the midbrain (0.35 +- 0.05), the parahippocampal gyri (0.35 +- 0.04 for the right and 0.37 +- 0.05 for the left), and the thalami (0.37 +- 0.05 for the right and 0.36 +- 0.04 for the left) were significantly lower than the mean OEF for the cerebral cortices (0.42 +- 0.04) (p < 0.05 or less). The global CBF and CMRO2 Were consistent with those obtained by the Kety-Schmidt method. Although several limitations to the quantification derived from an inadequate spacial resolution remain unsolved, the performance of the present PET scanner and the method for the quantification employed provide regional estimates of brain circulation and oxygen metabolisrn more acurately than the PET system and the steady state method previously used. 

Key words: rCBF, rCBV, rOEF, rCMR02, PET 

INTRODUCTION 
SINCE THE DEVELOPMENT of PET and the oxygen-15 steadystate (SS) technique,1-3 regional values for cerebral blood flow (rCBF), the extraction fraction of oxygen (rOEF) and the metabolic rate of oxygen (rCMR02) have been studied in normal volunteers to obtain control values for comparison with patients and to examine the effect of age4-8 but mean CBF values obtained by the SS technique were considerably lower than those obtained by the Kety-Schmidt (KS) method9 and the radioactive inert gas clearance method.10,ll Mean CBF for whole brain by the KS method was estimated to be 54 ml/100 g/min in adults, whereas the rCBF in the lentiform nucleus defined as "pure" gray matter and white matter by the SS method8 was 43.0 ml/100 ml/min and 21.9 ml/100 ml/min, respectively, which may produce a mean global CBF of around 30 ml/100 ml/min by assuming that the gray to white matter volume ratio is 1. By means of a 133Xe clearance method, the blood flow in cortical gray matter and deep white matter was reported to be 78.2 +- 9.8 ml/100 g/min and 18.7 +- 1.6 ml/100 g/min, respectively.11 The major reason for lower values in the initial PET measurement was inadequate spacial resolution of the PET scanner employed.12 Due to a partial volume effect, a concentration of radioactivity accumulated in a small structure or thin structure in the cortical gray matter is underestimated. In addition, the SS model itself includes the underestimation factor derived from tissue mixture. 13 

Recently, several improvements in measuring rCBF, rOEF, and rCMR02 have been made. Firstly, the autoradiographic method with H2^15O for rCBF measurement14 and the 15O2 single inhalation for rCMR02 measurementl5 were developed instead of the SS technique. Secondly, a Headtome IV scanner with an improved spacial resolution16 was employed for clinical use. Thirdly, MR images which clearly depict cortical gyri are available as an anatomical reference instead of CT. Although anatomical regions had been defined by CT in our previous report on normals,6 it was still difficult to identify some major cerebral gyri such as the angular gyrus and supramarginal gyrus. 

The normal values obtained with a high resolution tomograph by the autoradiographic technique and the single inhalation method have not yet been published. Here we give the normal rCBF, rCBV, rOEF, and rCMR02 values estimated by means of these improved methods. The values are compared with values in the literature obtained by the KS method and the SS technique. 

MATERIALS AND METHODS 
Subjects Eleven normal volunteers (nine males and two females) were studied. Their age ranged from 24 to 68 years (mean and I SD: 42.5 +- 18.2 years). All of them had always been free from neurological and psychiatric abnormalities. No organic lesions were found in an examination of brain MR imaging (T1-weighted images) before a PET study. All the subjects gave written informed consent. This project was approved by the PET Research Committee of the Institution. 

MRI 
Prior to PET measurements, the subjects were studied with a 0.5 T whole body MR scanner (Magnex 50; Shimadzu, Kyoto). After taking midsagittal images, sixty T1-weighted sequence axial images (TR 300 msec/TE 9 msec) were obtained parallel to an anterior commissure-posterior commissure (AC-PC) line with an axial interval of 2 mm. Axial T1-weighted images were used as an anatomical reference. 

Setting of scan slice in PET 
All the PET images were obtained parallel to the AC-PC line. The AC-PC line in the PET measurement was first 
identified in a mid-sagittal image of T1-weighted MRI as follows. Each patient was lying on a bed for the PET study. A lateral cranial X-ray photograph was taken with a metal line landmark placed parallel to the scanning slices. By fitting the cranial X-ray photograph to the MR-mid sagittal image and by measuring the angle produced by a metal line landmark and the AC-PC line of the MRI, the tilting angle of the PET gantry was determined. PET measurements PET scans were performed with a whole body 4 ring-7 slice positron tomograph (Headtome IV, Shimadzu, Kyoto).16 The spacial resolution in the tomographic plane and axial resolution were 8 mm (FWHM) and 10 mm, respectively. The center to center spacing of slices was 1 5 mm. By moving the detector ring 7.5 mm in the axial direction, two different positions were scanned. A total of 14 axial images with 7.5 mm spacing were produced in each measurement. The regional cerebral blood flow (rCBF), regional extraction fraction for oxygen (rOEF), regional metabolic rate for oxygen (rCMR02) and regional cerebral blood volume (rCBV) were measured by administering oxygen-15 labeled water,14 and by inhaling oxygen-15 labeled molecular oxygenl5 and oxygen-15 labeled carbon monoxide,15 respectively. The subject lay on the PET bed in the supine position, and the catheter was inserted into the radial artery . The catheter end was connected to a manometer tube of 0.5 mm inner diameter. The manometer tube was coiled two turns 20 cm from the catheter end and was taped on the face of the beta-ray detector. The other end of the manometer tube was connected to a Harvard pump to withdraw arterial blood during the 1502 and H2^15O study. After setting the tilting angle of the gantry, transmission scan was performed to correct tissue attenuation. The subjects inhaled trace doses of oxygen-15 labelled C15O through a small plastic face mask. The tracers were delivered continuously in a constant flow of air (0.5 l/min) and at a constant concentration of C15O. The radioactive doses in the air delivered were 100 to 140 mCi/min or 3700 to 5180 MBq/min. The static scans for two positions were initiated at 3 min after C15O inhalation for 1 min to measure rCBV. The data aquisition time was 2 min for each position. Arterial blood was taken three times during the Cl5O study to measure the radioactivity in whole blood. Following a 15-min period to allow for the decay of 150 radioactivity in the brain to the background level, the patient inhaled 15O2 gas delivered through the face mask. The radioactive doses in the air delivered were 1 50 to 1 80 mCi/min (5550 to 6660 MBq/min). At 15 sec after starting inhalation, a dynamic scan with 5 sec data aquisition was repeated 24 times for each position. The arterial radioactivity concentration during 1502 inhalation was monitored by detecting p-rays with a plastic scintillator. The arterial blood was withdrawn at a constant rate of 5 ml/min. The arterial plasma concentration of 15O labeled metabolic water was calculated according to the method previously described by Iida et al.17 
Following the C15O and 1502 inhalation study, oxygen-l5 labeled water was administered intravenously as a bolus. The usual amount of the tracer was 5 ml and the dose of radioactivity was 30 to 40 mCi. Sequential scanning of 5 sec data aquisition repeated 9 times for each position was started at 10 to 20 sec after injection when the radioactivity was detected in the brain. The arterial radioactivity concentration was monitored in the same way as in the 1502 study. The dispersion was corrected by deconvolving the beta probe curve with a fixed dispersion time constant of 3 sec.18 Beta probe curve delay in comparison with radiotracer appearance in the brain was corrected by adjusting the initial rise in radioactivity in the blood and brain. The delay was 3 to 8 sec among the subjects. 

Arterial partial pressure of O2 and CO2, hematocrit, and pH were measured in a blood gas tension analyzer (IL-1303, Instrumental Laboratory, USA). Mean values were 87.5 +- 7.2 mmHg, 38.5 +- 2.1 mmHg, 39.5 +- 3.0%, and 7.412 +- 0.112, respectively. The arterial hemoglobin concentration was measured with a hemoglobin analyzer (MLK- 1100, NIHON KODEN Ltd., Japan). The mean value was 12.0 g/dl. Systemic blood pressure and heart rate were monitored with a 2300 Finapress BP monitor (Omeda, USA) during the study. 

Data analysis 
Regional CBF, rCBV, rOEF, and rCMR02 were calculated by using tissue radioactivity and radioactivity of the blood input function, and the model equations. The blood to tissue partition coefficient of water was set at I .O for the CBF calculation. The functional data were transfered to a conventional unix work station system (TITAN 750, Kubota Computer, Tokyo) for further analysis. 

The first step is to identify cerebral gyri in the MR images of individual subject by comparing them with the human brain atlas by Duvernoy.19 Although the rectal gyri, orbital gyri, inferior temporal gyri, and fusiform gyri were easily found in MR images, these cortical gyri were not always depicted in the PET images, and were excluded from the analysis. The second step is to set regions of interest on the CMRO2 images. The circular regions of interest (ROI) with a 16 mm diameter were manually located in the following regions on the PET images with MR reference: pons, midbrain, cerebellar cortices (2 ROIs), cerebellar vermis, caudate nucleus, lentiform nucleus, thalamus, superior frontal gyrus, middle frontal gyrus, inferior frontal gyrus (pars orbitalis), superior temporal gyrus, middle temporal gyrus, entorhrinal area of parahippocampal gyrus, parahippocampal gyrus, supramarginal gyrus, angular gyrus, middle occipital gyrus, superior occipital gyrus, occipital cuneus, precuneus, cingulate gyrus (anterior part and posterior part), and semioval center. Except for midline structures such as pons, midbrain, and vermis, ROIs were set on both hemispheres. The location of ROI in the reference MRI and the CMRO2 image are shown in Figure I and Figure 2, respectively. The mean values for rCBF, rCBV, rOEF and rCMRO2 in 11 subjects were calculated for each brain structure. Student's t-test was employed for the statistics. 

Estimation of global CBF and CMRO2 
We calculated the global CBF and CMRO2 by using the "pure" gray matter and white matter values in the PET measurement comparable with the Kety-Schmidt values. The values for the lentiform nucleus defined as "pure gray matter" and semioval center defined as white matter were combined with the 57 : 43 mass ratio for gray and white matter reported by Miller et al.20 

RESULTS 

Averages of regional values for rCBF, rCBV, rOEF, and rCMRO2 in 11 subjects are shown in Table1 . The mean cortical CBF, CBV, OEF, and CMRO2 were 55.4 ~ 8.9 ml/100 ml/min, 4.80 +- 0.51 ml/100 ml, 0.42 +- 0.04 and 3.86 +- 0.42 ml/100 ml/min, respectively. The posterior part of the cingulate gyri showed the highest values for rCBF and rCMRO2 among the brain structures, followed by the lentiform nucleus, the cerebellum, the caudate nucleus, and the thalamus. The parahippocampal gyri showed the lowest values among cortical gyri. Right-left asymmetry was not significantly found among the supratentorial structures. Regional CBV showed a high value in areas neighboring the venous sinuses, such as the transverse sinus, superior sagittal sinus and cavernous sinus. 

Regional OEF values in the pontine nuclei, the midbrain, the parahippocampal gyri and the thalami were significantly lower than the mean OEF for cerebral cortices (0.42 +- 0.04) (p < 0.05 or less). There was no significant difference in rOEF among the cortical gyri. 

The regional CBF of "pure" gray matter (lentiform nucleus, mean for right and left) and white matter (semioval center, mean of right and left) in this study was 73.5 ml/100 ml/min and 22.8 ml/100 ml/min, respectively. The CMRO2 for the lentiform nucleus and semioval center were 5.22 ml/LOO ml/min and 1.54 ml/100 ml/min, respectively. The global CBF and CMRO2 calculated with these values were 51.7 ml/100 ml/min and 3.63 ml/100 ml/min, respectively. 

DISCUSSION 
The oxygen-15 steady state technique has been widely employed to measure rCBF, rOEF, and rCMR02 in clinical studies. It is based on a simple model and the procedure is not complicated1-3 but it is not suitable for measuring the vasoreactivity to PaCO2 and physiological stimulations because it takes 15 to 20 min for a single measurement. On the other hand, the ARG method is a quick way to measure rCBF which provides multiple measurement in a single study. However, as will be indicated below, the ARG method is sensitive to the dispersion and the delay of the arterial input function 18,21,22 The rCBF and rCMRO2 Values in the present study were higher in most gray matter structures than those obtained in previous PET measurements.4-8 For example, the mean rCBF of lentiform nucleus was 53.1 ml/100 ml/ min with the Headtome III and the SS technique6 but 73.8 ml/100 ml/min with the Headtome IV and the autoragiographic method in this study. In early studies, the scanner employed was the ECAT 114,7,8 or Headtome 1115,6 which had a spacial resolution as great as 1 2 mm in FWHM in the tomographic plane. For example, in-plane and axial spacial resolution of Headtome IV were 8 mm and 10 mm respectively in FWHM but those of Headtome III were 10.5 mm and 13.1 mm respectively.23 The improved spacial resolution of the present scanner may contribute most to the high values for rCBF and rCMR02 by reducing a partial volume effect in the transaxial and axial directions. Another reason for the increased CBF values is the difference in the method employed. As shown previously, the SS method inherently underestimates rCBF owing to tissue heterogeneity.13 In the area of an equal mixture of gray and white matter, underestimation may reach around 18%.14 Kanno et al.24 demonstrated that this underestimation is less for the autoradiographic method (9% in the same situation). The cortical mean CBF with the SS technique and Headtome III was 43.0 ml/100 ml/ min for the cerebrum, and 52.0 ml/100 ml/min for the cerebellum.5 In the present study, the values were 55.4 ml/100 ml/min for the cerebrum and 66.3 ml/100 ml/min (mean of right and left) for the cerebellum. In the cortical ROI where a mixture of gray and subcortical white matter is expected, the autoradiographic measurement is much less affected by underestimation due to tissue heterogeneity especially in high flow areas.24 

Correction for the dispersion and the delay of the arterial input function is critically important in measuring rCBF accurately. We previously reported that rCBF by 
the ARG method was equal to that by the SS technique 
when a dispersion time constant was around 10 sec.21 The study also predicted that a smaller constant produced a higher CBF value by the ARG method. Later, we estimated the dispersion time constant to be 3 sec in the present beta probe system with a manometer tube 20 cm in length and 0.5 mm inner diameter.18 The employment of a dispersion time constant of 3 sec in the present study may be another reason for the systematic increase in rCBF compared with that obtained by Headtome 111 and the SS method. 

Relatively high rCBF and rCMR02 Values were found in the cingulate gyri, cerebellar hemisphere, lentiform nucleus, caudate nuclei, superior temporal gyri, superior occipital gyri, and occipital cuneus. In contrast, parahippocampal gyri had the lowest values among the cortical gyri measured. Although the results might be greatly affected by the condition of the study (i.e., eye-covered or not), this distribution of high rCBF and rCMRO2 was consistent with previous reports.4,5,8 

In contrast to the remarkable increase in rCBF and rCMR02, rCBV values for most brain structures were not changed compared with previous results. For example, mean cortical CBV was 4.8 ml/100 ml in the present study but 4.3 ml/100 ml in the study by Yamaguchi et al.5 and 4.7 ml/100 ml in the territory of the middle cerebral artery as reported by Leenders et al.8 

Regional OEF for cerebral cortices and the caudate and the lentiform nuclei was simillar to that in previous reports. The values ranged from 0.40 to 0.45 although there were remarkable regional changes in rCBF and rCMRO2 among structures. This confirmed the coupling of rCBF and rCMRO2 under physiological conditions. It is interesting that the pontine nuclei, midbrain, parahippocampal gyri, and thalami had lower OEF than the cerebral cortices. This was not found in previous reports. Although we do not have an explanation for this, it is noteworthy that the perfusion of these areas is maintained by the posterior circulation through vertebro-basilar arteries (pons, mid brain, thalamus) and the branches of the posterior cerebral artery (parahippocampal rus). 

The global CBF and CMRO2 for the whole brain were originally estimated by Kety and Schmidt9 to be 54 ml/100 ml/min and 3.30 ml/100 ml/min, respectively. Their values were confirmed by many studies using the radioactive inert gas method. It is now considered that normal CBF for the whole brain is between 45 and 60 ml/100 g/min, and normal CMRO2 is between 2.9 and 4.0 ml/100 g/min.25 26 The calculated values for global CBF (51.7 ml/100 ml/min) and CMRO2 (3.63 ml/100 ml/min) in this study were in close agreement with these estimates. 

Each region of interest was set on the PET image by using a corresponding T1-weighted image. In a previous study, Yamaguchi et al. used the CT scan and standard brain atlas as an anatomical reference5 but it was still difficult to identify some gyri, especially the marginal and angular gyri in the parietal lobe. By using MR images, it is easier to find the gyral structure. 

In conclusion, the rCBF and rCMRO2 estimated by means of Headtome IV and the autoragiographic and the single inhalation method were higher in most of the brain structures than those measured by means of a PET scanner with lower spacial resolution. The global rCBF and rCMRO2 estimated by using the "pure" gray matter and white matter values for the present PET measurements were consistent with the KS method. Although several limitations to the quantification derived from inadequate spacial resolution remain to be clarified, the performance of the present PET scanner and the method employed in measurements may provide rCBF, rCBV, rOEF, and rCMRO2 Values with anatomical and physiological accuracy. 

ACKNOWLEDGMENTS 

The authors are much indebted to the technical staff of the Department of Radiology and Nuclear Medicine, the Akita Research Institute of Brain and Blood Vessels for their assistance. This work was partly supported by a grant-in-aid for science research from the Ministry of Health and Welfare, Japan. 

REFERENCES 
l. Frackowiak RSJ, Lenzi GL, Jones T, Heather JD. Quantitative measurement of regional cerebral blood flow and oxygen metabolism in man using 150 and positron emission tomography : theory, procedure, and normal values. J Comput Assist Tomogr 4: 727-736, 1980. 
2. Lammertsma AA. Wise RJS. Heather JD, Gibbs JM, Leenders KL. Frackowiak RSJ, et al. Correction for the presence of intravascular oxygen-15 in the steady state technique for measuring regional oxygen extraction ratio in the brain. 2. Results in normal subjects and brain tumor and stroke patients. J Cerebr Blood Flow Metab 3: 425-424, 1983. 
3. Lammertsuma AA, Correia JA, Jones T. Stability of arterial concentrations during continuous inhalation of Cl5O and 15O2 and the effects on computed values of CBF and CMRO2 J Cerebr Blood Flow Metab 8: 411-417, 1988. 
4. Pantano P. Baron JC. Lebrun-Grandie P, Duquesnoy N, Bousser MG, Comar D. Regional cerebral blood flow and oxygen consumption in human aging. Stroke 15: 635-641, 1984. 
5. Yamaguchi T, Kanno I, Uemura K, Shishido F. Inugami A, Ogawa T, et al. Reduction in regional cerebral metabolic rate of oxygen during human aging. Stroke 17: 1220-1228, 1986. 
6. Sasaki H. Kanno I. Murakami M, Shishido F, Uemura K. Tomographic mapping of kinetic rate constants in the fluorodeoxyglucose model using positron emission tomography. J Cerebr Blood Flow Metab 6: 447-454, 1986. 
7. Hatazawa J. Itoh M. Ido T, Matsuzawa T. Watanuki S. Measurement of the ratio of cerebral oxygen consumption to glucose utilization by positron emission tomography: Its consistency with the values determined by the Kety-Schmidt method in normal volunteers. J Cerebr Blood Flow Metab 8: 426-432, 1988. 
8. Leenders KL, Perani D, Lammertsma AA, Heather JD, Buckingham P, Healy MJR, et al. Cerebral blood flow, blood volume and oxygen utilization normal values and effect of age. Brain 113: 27-47, 1990. 
9. Kety SS, Schmidt CF. The nitrous oxygen method for the quantitative determination of cerebral blood flow in man: theory, procedure and normal values. J Clinical Investigation 27: 476-483, 1948. 
10. Lassen NA, Munck O. The cerebral blood flow in man determined by the use of radioactive krypton. Acta Physiol Scand 33: 30-49, 1955. 
11. Meyer JS. Ishihara N. Deshmukh VD, Naritomi H. Sakai F, Hsu MC, et al. Improved method for noninvasive measurement of regional cerebral blood flow by 133Xe Inhalation. Part 1 : Description of method and normal values obtained in healthy volunteers. Stroke 9: 195-204, 1978. 
12. Hoffman EJ, Huang SC, Phelps ME. Quantitation in positron emission computed tomography: I . Effect of object size. J Comput Assist Tomogr 3: 299-308, 1979. 13. Herscovitch P, Markham J, Raichle ME. Brain blood flow measured with intravenous H2^15O. 1. Theory and error analysis. J Nucl Med 24: 782-789, 1983. 
14. Raichle ME, Martin WRW, Herscovitch P, Mintun MA, Markham J. Brain blood flow measured with intravenous H2^15O. II. Implementation and validation. JNucl Med 24: 790-798, 1983. 
15. Mintun MA, Raichle ME, Martin WRW, Herscovitch P. Brain oxygen utilization measured with O-15 radiotracers and positron emission tomography. J Nucl Med 25: 177187, 1984. 
16. Iida H, Miura S, Kanno I, Murakami M, Yamamoto S, Amano M. Design and evaluation of HEADTOME IV, a new whole-body positron emission tomograph. IEEE Trans Nucl Sci NS-37: 1006-1010, 1989. 
17. Iida H, Jones T, Miura S. Modeling approach to eliminate the need to separate arterial plasma in Oxygen-15 inhalation positron emission tomography. J Nucl Med 34: 1333-1340, 1993. 18. Iida H, Kanno I, Miura S, Murakami M, Takahashi K, Uemura K. A determination of the regional brain/blood partition coefficient of water using dynamic positron ernission tomography. J Cerebr Blood Flow Metab 9: 87l-885, 1989. 
l9. Duvernoy H. The Human Brain, Surface, Three-dimensional Sectional Anatomy and MRI. Wien, New York, Springer-Verlag, 1991. 
20. Miller AKS, Alson RL, Corsellis J. Variation with age in the volume of grey and white matter in the cerebral hemispheres of man: measurement with an image analyzer. Neuropathol Appl Neurobiol 6: 118-132, 1980. 
21. Kanno I, Iida H, Miura S, Murakami M, Takahashi K, Sasaki H, et al. A system for cerebral blood flow measurement using an H2^15O autoradiographic method and positron emission tomography. J Cerebr Blood Flow Metab 7: 143-153, 1987. 
22. Meyer E. Simultaneous correction for tracer arrival delay and dispersion in CBF measurements by the H2^15O autoradiographic method and dynamic PET. J Nucl Med 30: 1069-1078, 1989. 23. Kanno l, Miura S, Yamamoto S, Iida H, Murakami M, Takahashi K, et al. Design and evaluation of a positron emission tomograph: HEADTOME 111. J Comput Assist Tomogr 9: 931-939, 1985. 
24. Kanno I, Lammertsma AA. Heather JD, Gibbs JM, Rhodes CG, Clark JC, et al. Measurement of cerebral blood flow using bolus inhalation of C^l5O2 and positron emission tomography : description of method and its compari son with the C^l5O2 continuous inhalation method. J Cerebr Blood Flow Metab 4: 224-234, 1984. 
25. Siesjo BK. Brain Energy Metabolism, Chichester, New York, Brisbane, Tronto, John Wiley & Sons, pp. 56-100, 1978. 
26. Lassen NA. Normal average value of cerebral blood flow in younger adults is 50 ml/100 g/min. J Cerebr Blood Flow Metab 5: 347-349, 1985.