ORIGINAL ARTICLE Annals of Nuclear Medicine Vol. 12, No. 1 , 29-33, 1998 Cerebral blood flow changes in the primary motor and premotor cortices during hyperventilation Kazunari ISHII, Masahiro SASAKI, Shigeru YAMAJI, Setsu SAKAMOTO and Kiyoshi MAEDA Division of Neuroimaging Research and Radiology Service, Hyogo Institute for Aging Brain and Cognitive Disorders (HI-ABCD) The aim of this study was to clarify the regional differences in cerebral blood flow (CBF) change during hyperventilation by using H215O and positron emission tomography (PET). Eight healthy volunteers (age: 63.0 +- 8.9 yr.) were studied. Regional CBF was measured by the H215O autoradiographic method and PET. Statistical parametric maps (SPM) and conventional regions of interest (ROI) analysis were used for estimating regional CBF differences in the normocapnic state with normal breathing and the hypocapnic state induced by hyperventilation. Total CBF decreased during the hypocapnic state. The SPM revealed that primary motor and premotor cortices were significantly activated by hyperventilation. In these areas absolute CBF values were significantly higher than those in the temporal, occipital and parietal lobes in the hypocapnic state, but there were no significant regional differences in the normocapnic state. In the hypocapnic state induced by hyperventilation, the primary motor and premotor CBF shows combined changes with vasoreaction to hypocapnia and increase in activation due to hyperventilation. Key words: PET, 15O labeled water, hyperventilation, cerebral blood flow INTRODUCTION THE FACT that an evident decrease in cerebral blood flow (CBF) is acutely induced by hyperventilation is well known. Kety and Schmidt first reported this phenomenon by developing the nitrous oxide washout technique for estimating CBF.1 The mechanism of this effect is caused by vasoconstriction secondary to a respiratory alkalosis, with an increase in pH.2,3 In the past 20 years, regional CBF responses to hypocapnia in patients with cerebral infarction have been widely studied as well as the hypercapnic state in single photon emission computed tomography (SPECT) and positron emission tomography (PET).4-6 Nevertheless, the regional differences in CBF change during hyperventilation in the normal brain tissue Received August 21 , 1997, revision accepted November 12, 1997. For reprint contact: Kazunari Ishii. Hyogo Institute for Aging Brain and Cognitive Disorders, 520 Saisho-Ko, Himeji. Hyogo 670-0981 , JAPAN. E-mail: ishii@hiabcd.go.jp in relative and absolute values have not been studied in normal aged subjects. The purpose of this study was to estimate the regional differences in CBF change during hyperventilation in relative and absolute values especially in the primary motor and premotor areas, measuring regional CBF by the H215O autoradiographic method and PET. METHODS Subject Selection Eight healthy female volunteers (mean +- SD age 63.0 +- 8.9 yr.) were studied. They had no neurological signs or significant medical antecedents and no abnormal magnetic resonance (MR) findings except for age-related hyperintensities on T2-weighted images. Written informed consent was obtained from all the subjects. The PET procedure was approved by our institution's Ethical Committee. Before PET scans, all subjects received MR imaging for anatomical reference, and for PET positioning. Detailed MR procedures have been reported elsewhere.7 Immediately before the PET examination, sagittal gradient-echo images were obtained to determine the coordinates for positioning of the head on the PET table. PET Procedure A Headtome IV (Shimadzu Corp., Kyoto, Japan) PET scanner, which had four rings located 13 mm apart and yielded a transverse resolution of 4.5 mm full-width-half-maximum (FWHM)8 was used in this study. The scanner slice thickness was 11 mm and the slice interval was 6.5 mm when the z-motion mode was used. On the table of the PET scanner, the subject's head was placed horizontally, and the gantry and the table of the PET scanner were adjusted according to the coordinates determined by MR imaging, so that the scans were taken parallel to the AC-PC plane.9,10 A transmission scan was performed with a 68Ga/68Ge pin source for absorption correction after each subject was positioned. PET studies were performed in the supine position with eyes closed and ears unplugged. The CBF was calculated by an autoradiographic technique with a table look up procedure over a 90 sec accumulation after the intravenous injection of H215O.11 The usual amount of the tracer was 5 ml and the dose of radioactivity of H215O was 740-1110 MBq. Details of the procedure are reported elsewhere. 12.13 Two trials of regional CBF were examined for each subject with normal breathing (normocapnia) and voluntary hyperventilation (hypocapnia). Voluntary hyperventilation by deep breathing was started 60 sec before H215O injection and ended after arterial blood sampling. The PaCO2 level was measured 30 sec before H215O injection and immediately after the end of the scan, and the two measures were averaged. Data Analysis PET and MR image data sets were directly transmitted to a workstation (Indigo2, SGI, Mountain View, CA, USA) from the PET and MR imaging units, and image analysis was performed on the workstation. For the relative regional CBF change analysis, statistical processing was performed with SPM 95 software (MRC Cyclotron Unit, London, United Kingdom). Calculations and image matrix manipulations were performed in MATLAB (Mathworks Inc., MA, USA). The original 14 contiguous, 6.5-mm scan slices were interpolated to 43 planes with approximately cubic voxels. The data were then transformed into a standard stereotactic space14,15 and the images were smoothed with an isotropic Gaussian filter to compensate for intersubject gyral variability and to reduce high frequency noise. The stereotactically normalized regional CBF images were then adjusted for individual differences in global blood flow using an analysis of covariance (ANCOVA).16 This algorithm scales all images to a global mean regional CBF of 50 ml/100 ml/min. Finally between the two conditions comparisons were performed on a pixel-by-pixel basis for all voxels common to all subjects. The subset of voxels exceeding a threshold of p < 0.001 in omnibus comparisons and remaining significant after correction for multiple comparisons (p < 0.05) was displayed as a volume image rendered in three orthogonal projections. Quantitative analyses were performed with conventional region of interest (ROI) settings and image analysis software Dr. View (Asahi Kasei Joho System, Tokyo, Japan) referencing the results of the SPM analysis. We determined two or three circular ROIs ( 10 mm diameter) on the cortical ribbon of the temporal lobe, occipital lobe, parietal lobe and primary motor and premotor cortices on the co-registered MR images. The same ROIs were transferred to the CBF images of the normocapnia and hypocapnia. These values were shown as the average of the right and left regional values. For regional differences, one-way analysis of variance (ANOVA) was used and Scheffe's test was used for multiple post hoc comparisons. Differences were considered significant when the p-value was less than 0.05. RESULTS Table 1 summarizes the averaged pH, PaCO2, PaO2 and CBF of each regional cortex in both normal breathing and hyperventilation condition groups. There was a significant difference between the two states in each parameter (p < 0.001). Averaged differences between the normocapnia and hypocapnia in PaCO2, PaO2, pH and whole CBF were 11.6 +- 3.9 mmHg, 26.5 +- 11.8 mmHg, 0.11 +- 0.04 and 14.9 +- 5.1 ml/100 ml/min, respectively. In the SPM study, significant focal relative increases in regional CBF due to hyperventilation are shown in Fig. 1 and Table 2. Bilateral primary motor and premotor areas were included in the activated areas. In addition to the cortical areas shown above, an extracerebral region near the sphenoid sinus was shown as a significantly activated area. In the absolute CBF value, there was no significant difference among the temporal lobe, occipital lobe, parietal lobe and primary motor and premotor cortices in normocapnia, (F = 0.636, p = 0.600), but in hypocapnia there was a significant regional CBF difference (F = 4.549, p = 0.013) and the CBF in the primary motor and premotor areas was significantly higher than that in other regions (p < 0.05). DISCUSSION The present study revealed that in the areas which were activated by hyperventilation there existed a different regional CBF change in hypocapnia. This CBF change should be combined with vasoreaction and neural activation induced by hyperventilation. Bednarczyk et al.17 reported the effects of hypocapnia in ten normal volunteers by using H215O and PET before and after five minutes of hyperventilation and found that CBF decreased by a mean of 49.5%. They showed a significant change over the baseline in PaCO2 and CBF, in the hyperventilated state. Our study also demonstrated similar results, although the magnitude of PaCO2 and CBF changes in ours was smaller than that in their results. This would be due to the duration of the hyperventilation task; that is the hyperventilation time in our study was 2.5 min and theirs was 5 min. Because our subjects were elderly females, age and gender differences also would have affected the magnitude of activation. On the other hand concerning relative CBF change, Ramsay et al.18 reported identified areas of neural activation associated with volitional inspiration and with volitional expiration in five normal male subjects promoting their previous report.19 They performed PET scans on each subject under conditions of volitional inspiration with passive expiration, passive inspiration with volitional expiration and passive inspiration with passive expiration. Regional CBF increases during the volitional and passive ventilation phases, due to increased neural activity associated with either active inspiration or active expiration, were analyzed by SPM. During active inspiration significant increases in regional CBF were found bilaterally in the primary motor cortex, in the supplementary motor area (SMA), in the right lateral premotor cortex and in the left ventrolateral thalamus. In active expiration, significant increases in regional CBF were found in the right and left primary motor cortices, the SMA, the right lateral premotor cortex, the ventrolateral thalamus bilaterally, and the cerebellum. For volitional expiration the areas activated were more extensive, but overlapped with those involved in volitional inspiration. Compared with Ramsay's study, our study demonstrated almost the same activated regions in primary motor and premotor cortices. In our study, some of the superior temporal gyri were also activated. In Ramsay et al.'s study, absolute blood flow values were not measured. On the other hand, in our study we could not demonstrate an activated area in the SMA, which was demonstrated as an activated area by Ramsay et al. in their study, because of the limited total axial field of view of our PET scanner (91 mm). The SMA was outside the field of view of the scanner. We suspect that in our study the SMA concerning respiration would also have been activated as in other studies.17,18 Shimosegawa et al.20 demonstrated that the absolute CBF increase induced by visual stimulation was affected by the PaCO2 level, whereas the fractional CBF increase remained unchanged at different baseline CBF levels, which meant that the CBF increase induced by neural activation changes proportionally with the changes in baseline CBF. Ishii et al.6 reported that the alteration in absolute cerebellar blood flow change to PaCO2 changes was proportional to the level of PaCO2 On the affected and unaffected sides with crossed cerebellar diaschisis (CCD) in 27 cerebrovascular patients and indicated that the rate of change in percent cerebellar blood flow per millimeters of mercury PaCO2 change was uniform across affected and unaffected cerebellar hemispheres with CCD. But these two reports were concerned with not regional cerebral blood changes but only the occipital region, whole brain or cerebellar hemisphere blood flow. Our results indicated that in primary motor and premotor areas the magnitude of CBF decreases, which was a vasoreaction to hypocapnia, was small because those areas were activated by hyperventilation. Therefore, in estimating global CBF reactivity to hypocapnia induced by hyperventilation, we must be careful to set a ROI. Setting ROIs on the primary motor and premotor areas may lead to an error that shows an underestimated vasoreaction to hypocapnia. In the SPM study, ANCOVA removes variance due to differences in global flow, but there may be a problem because the relationship between the global CBF and regional CBF is additive or proportional. Ramsay et al.21 suggested that it is additive and Shimosegawa et al. suggested that it is proportional. In our relative CBF study, we used SPM with ANCOVA to survey the regional CBF changes. Under a condition of dynamic global CBF change such as hypocapnia, there may be a difference in the ability to detect an activated area. To solve these problems, additional research is required. SPM demonstrated activated areas in the extracerebral region near the sphenoid sinus. The finding of a relative increase in sphenoid sinus perfusion due to hyperventilation cannot be explained. We suppose this was caused by an artifact preventing SPM from correctly standardizing extracerebral structures. CONCLUSION There were different regional CBF changes in hypocapnia induced by hyperventilation. In analyzing vasoreaction by means of the voluntary hyperventilation technique, the primary motor and premotor areas which are activated by hyperventilation should be considered as different areas, where the CBF reductions are smaller than in other non-activated areas. These CBF changes are combined with vasoreaction and neural activation. ACKNOWLEDGMENTS We thank Mr. Toru Kida and Mr. Hiroto Sakai (Radiology Service, HI-ABCD) for their technical assistance. REFERENCES 1. Kety S, Schmidt CF. The determination of cerebral blood flow in man by the use of nitrous oxide in low concentrations. Am J Physiol 14: 53-66, 1945. 2. Raichle ME. Posner JB, Plum F. Cerebral blood flow during and after hyperventilation. Arch Neurol 23: 394-403, 1970. 3. Kety S, Schmidt CF. The effects of active and passive hyperventilation on cerebral oxygen consumption, cardiac output, and blood pressure of normal young men. J Clin Invest 25: 107-119, 1946. 4. Melamed E, Lavy S. Portnoy Z. Regional cerebral blood flow response to hypocapnia in the contralateral hemisphere of patients with acute cerebral infarction. Stroke 6: 503-508, 1975. 5. Kanno I. Uemura K. Higano S, Murakami M, Iida H, Miura S, et al. Oxygen extraction fraction at maximally vasodilated tissue in the ischemic brain estimated from the regional CO2 responsiveness measured by positron emission tomography. J Cereb Blood Flow Metab 8: 227-235, 1988. 6. Ishii K. Kanno I. Uemura K, Hatazawa J, Okudera T, Inugami A, et al. Comparison of carbon dioxide responsiveness of cerebellar blood flow between affected and unaffected sides with crossed cerebellar diaschisis. Stroke 25: 826-830, 1994. 7. Ishii K. Sasaki M. Kitagaki H, Sakamoto S, Yamaji S, Maeda K. Regional difference of cerebral blood flow and oxidative metabolism in human cortex. J Nucl Med 37: 1086-1088, 1996. 8. Iida H. Miura S. Kanno I. Murakami M. Yamamoto S, Amano M. Design of evaluation of Headtome IV: a whole body positron emission tomograph. IEEE Trans Nucl Sci NS-37: 1006-1010, 1989. 9. Ishii K. Kitagaki H, Kono M, Mori E. Decreased medial temporal oxygen metabolism in Alzheimer's disease shown by positron emission tomography. J Nucl Med 37: 1159-1165, 1996. 10. Talairach J. Tournoux P. Co-planar Stereotaxic Atlas of the Human Brain. Stuttgart. Thieme Verlag, 1988. 11. Kanno I. Iida H, Miura S, Murakami M, Takahashi K, Sasaki H, et al. A system for cerebral blood flow measurement using H215O autoradiographic method and positron emission tomography. J Cereb Blood Flow Metab 7: 143-153, 1987. 12. 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 emission tomography. J Cereb Blood Flow Metab 9: 874-885, 1989. 13. Hatazawa J, Fujita H. Kanno I. Sato T. Iida H, Miura S, et al. 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. Ann Nucl Med 9: 15-21 , 1995. 14. Friston KJ, Frith CD, Liddle PF, Dolan RJ, Lammertsma AA, Frackowiak RSJ. The relationship between global and local changes in PET scans. J Cereb Blood Flow Metab 10: 458-466, 1990. 15. Friston KJ. Frith CD, Liddle PF, Frackowiak RSJ. Plastic transformation of PET images. J Comp Ass Tomog 15: 631-649, 1991. 16. Friston KJ, Holmes AP, Worsely KJ, Poline JB, Frith CD. Frackowiak RSJ. Statistical parametric maps in functional imaging: a general linear approach. Hum Brain Map 2: 189-210, 1995. 17. Bednarczyk EM, Rutherford WF, Leisure GP, Munger MA, Panacek ED, Miraldi FD, et al. Hyperventilation-induced reduction in cerebral blood flow: assessment by positron emission tomography. DICP Ann Pharmacoth 24: 456-460, 1990. 18. Ramsay SC, Adams L, Murphy, Corfield DR, Grootoonk S, Balley DL, et al. Regional cerebral blood flow during volitional expiration in man: a comparison with volitional inspiration. J Physiol 461 : 85-101 , 1993. 19. Colebatch JG, Adams L, Murphy K. Martin AJ, Lammertsma AA, Tochon-Dangu HJ, et al. Regional cerebral blood flow during volitional breathing in man. J Physiol 443: 91-103, 1991. 20. Shimosegawa E, Kanno I, Hatazawa J, Fujita H, Iida H, Miura S, et al. Photic stimulation study of changing the arterial partial pressure level of carbon dioxide. J Cereb Blood Flow Metab 15: 111-114, 1995. 21. Ramsay SC, Murphy K, Shea SA, Friston KJ, Lammertsma AA, Clark JC, et al. Changes in global cerebral blood flow in humans: effect on regional cerebral blood flow during a neural activation task. J Physiol 471 : 521-534, 1993.