Multi-Slice MRI of Rat Brain Perfusion During Amphetamine Stimulation Using Arterial Spin Labeling

Multi-Slice MRI of Rat Brain Perfusion During Amphetamine Stimulation Using Arterial Spin Labeling Afonso C. Silva, Weiguo Zhang, Donald S. Williams, Alan P. Koretsky

When a single coil is used to measure perfusion by arterial spin labeling, saturation of macromolecular protons occurs during the labeling period. Induced magnetization transfer contrast (MTC) effects decrease tissue water signal intensity, reducing the sensitivity of the technique. In addition, MTC effects must be properly accounted for in acquiring a control image. This forces the image to a single slice centered between the labeling plane and the control plane. In this work, a two-coil system is presented as a way to avoid saturation of macromolecular spins during arterial spin labeling. The system consists of one small surface coil for labeling the arterial water spins, and a head coil for MRI, actively decoupled from the labeling coil by using PIN diodes. It is shown that no signal loss occurs due to MTC effects when the two-coil system is used for MRI of rat brain perfusion, enabling three-dimen- sional perfusion imaging. Using the two-coil system, a multi- slice MRI sequence was used to study the regional effects of amphetamine on brain perfusion. Amphetamine causes significant increases in perfusion in many areas of the brain including the cortex, cingulate, and caudate putamen, in agreement with previous results using deoxyglucose uptake to monitor brain activation. Key words: NMR; cerebral blood flow; MTC effects; catecholamines.

INTRODUCTION

Tissue perfusion can be noninvasively measured by MRI using saturation (1, 2) or inversion (3-5) of arterial water spins. For the measurement of rat brain perfusion, the arterial water spins are tagged by applying selective RF irradiation in the presence of a longitudinal field gradient corresponding to a labeling plane situated in the neck. Labeled arterial water spins flow into the brain and exchange with tissue water, causing a decrease in tissue water magnetization. This change in signal intensity can

MRM 33:209-214 (lB95) From the Pittsburgh NMR Center for Biomedical Research (A.C.S., W.Z., D.S.W.. A.P.K.), Biomedical Engineering Program (A.C.S.), and Department of Biological Sciences (W.Z., A.P.K.), Carnegie Mellon University, Pitts- burgh, Pennsylvania. Address corrapondence to: Alan P. Koretsky, Ph.D., Department of Bio- logical Sciences, Carnegie Mellon University, 4400 Fifth Avenue, Pittsburgh, PA 15213. Received June 27,1994; revised October 18,1994; accepted October 19, 1994. Current address (W.Z.): Radiologic Imaging Laboratory, Toshiba America MRI, Inc.. 400 Grandview Drive, South San Francisco, CA 94080. This work was supported by an NIH Research Resource Award (RR-03631) to the Pittsburgh NMR Center for Biomedical Research, an NIH Research Career Development Award (5 KO4 HL 02847-02) to APK, and a Brazilian Research and Development Council (CNPq - 201 088/92-4) fellowship to ACS. We thank the Richard King Mellon Foundation, the Lucille Markey Charitable Trust, the Ralph M. Parsons Foundation, and the Ben Franklin Partnership Program of the Commonwealth of Pennsylvania for providing financial support for the establishment of the Pittsburgh NMR Center for Biomedical Research.

Copyright Q 1995 by Williams & Wilkins All rights of reproduction in any form reserved.

0740-31 94/95 $3.00

be used to quantitate perfusion. The Bloch equations can be modified to include the effects of perfusion so that any sequence that alters arterial water magnetization can be used to derive a measure of perfusion (5).

The longitudinal field gradient used during the labeling period of an experiment to measure perfusion allows selective labeling of arterial water spins as long as the frequency offset of the FW irradiation is much larger than the tissue water bandwidth. This condition is, indeed, very easily fulfilled. For example, as shown in Fig. la , the labeling plane for rat brain perfusion is located in the neck about 2 cm apart from the detection slice. When a longitudinal gradient of 1 Gauss/cm is used, the frequency offset at the labeling plane is 8.5 kHz, well be- yond the tissue water 'H bandwidth of 850 Hz for a 2-mm thick coronal slice. However, macromolecular 'H spins can have line widths as large as 50 kHz (6). Thus, i f a single coil is used to label the arterial spins and to detect the brain water magnetization as well, macromolecules in the detection region are saturated by the RF irradiation employed for arterial spin labeling, causing magnetization transfer contrast (MTC) effects (6). MTC effects can have adverse consequences on the measurement of perfusion by arterial spin labeling. Saturation of macromolecules causes a large loss in tissue water signal intensity (5, 6). For example, in rat brain at 4.7 T, the use of an RF field strength of 100 mG placed 8.5 kHz off the water resonance causes the water signal at steady-state to decrease by 75% compared with the value obtained without irradiation (5).

Another disadvantage is due to the fact that MTC effects are frequency dependent (6). To correct for MTC effects when imaging perfusion using arterial spin labeling, a control image must be obtained, with the RF irradiation placed in a plane symmetrically opposed to the labeling plane with respect to the imaging slice (Fig. la), so that saturation of macromolecules occurs without arterial spin labeling. This limits the perfusion mapping to a single slice at a time, centered between the labeling plane and the control plane, with the slice orientation restricted to being parallel to the labeling and control planes.

If the saturation of the macromolecular spins could be avoided, the MTC effects would be eliminated, and in principle the sensitivity of the arterial spin labeling technique for perfusion MRI would be improved. Also, mapping of perfusion could be extended to the whole organ of interest. Slices could be acquired with any orientation, and multi-slice or 3D MRI of perfusion could also be performed. One possible scheme for performing arterial spin labeling while avoiding MTC effects in the organ of interest is to use two coils, one for labeling and the other for imaging. If the RF from the labeling coil does not physically reach the organ of interest, MTC effects can be avoided.

209

210 Silva et al.

In this work, a two-coil system is presented as a way to obtain MRI of rat brain perfusion without saturation of macromolecular spins. This system consists of a small surface coil used exclusively for labeling of the arterial spins, and a volume imaging coil used for MRI of the brain. Experiments are performed to show that, when the two coils are decoupled, the B, field generated by the labeling coil does not reach the brain, so that saturation of the macromolecules does not occur during the arterial labeling phase before the imaging sequence. The degree of arterial spin labeling, a, is determined for the small surface coil. Multi-slice perfusion images of the rat brain are obtained in different orientations and the effects of amphetamines on perfusion in different regions of the brain are studied using the two-coil system.

MATERIALS AND METHODS

Male Sprague-Dawley rats (250-350 g, Taconic Farms, Germantown, NY) were initially anesthetized with 5 % halothane. The rats were then orally intubated and kept on 0.5% halothane and a 1:1 N,O/O, gas mixture using a rodent ventilator (Harvard Apparatus, South Natick, MA). A dose of 0.5 mgI200 g of pancuronium bromide in saline (2 mg/ml) (Organon Inc., W. Orange, NJ) was initially administered to paralyze the rats and followed by a dose of 0.25 mg/200 g every 90 min intraperitoneally to maintain immobilization. A femoral arterial line was used to monitor blood pressure and sample arterial gases.

NMR experiments were performed on a 4.7 T, 40-cm bore Bruker BIOSPEC instrument equipped with a 15-cm gradient insert. As shown in Fig. lb , the two-coil system was made up of a small 0.5-cm diameter butterfly-shaped surface coil, which was used for labeling the arterial water spins in the neck region, and a 3-cm saddle-shaped volume imaging coil. These two coils were actively decoupled by the use of PIN diodes (UM 9415, Unitrode Corp., Lexington, MA) (7). The PIN diodes were driven by a computer-controlled driver, so that when one of the coils was tuned (PIN diode in the conducting state), the other was detuned (PIN diode in the nonconducting state). RF chokes (2-144, Ohmite Manufacturing Co., Skokie, IL) were used to prevent RF leakage through the DC pathways. Despite the orthogonal arrangement between the labeling and the imaging coils, the use of an active decoupling scheme was necessary to prevent the RF power applied to the labeling coil from reaching the brain via interactions with the volume imaging coil and causing perturbation of the macromolecular spins.

A gradient-echo sequence with TE = 10 ms, TR = 530 ms, FOV = 3 cm and slice thickness of 2 mm was used in the experiments to measure the degree of labeling, a. The slices were located 1.3 cm downstream from the labeling plane shown in Fig. Ib. Arterial spin labeling was accomplished with the use of RF irradiation through the labeling coil under the presence of a longitudinal gradient of 1 Gausslcm during the TR time. The RF pulse amplitude was varied from 0 to 100 mG to alter the degree of spin labeling, which was measured according to Q = (Mao/ - M,)/2Mao, where M, is the arterial water magnetization intensity in the carotid arteries with RF power applied to the labeling coil, and Mao is the arterial water intensity without RF power applied.

For perfusion measurements, MR images were taken using a spin-warp sequence with TE = 30 ms, TR = 2 S,

matrix size 128 X 64 and FOV = 3 cm. The slice thickness for all images was 2 mm. Labeling of the arterial spins was accomplished during the TR time before each phase-encoding step by applying off-resonance RF power to the labeling coil under the presence of a longitudinal gradient of 1 Gauss/cm. Images without arterial spin labeling were acquired without RF power applied to the labeling coil during the TR time. Perfusion was calculated according to f = A/Tlb[Mbo - Mbss/Mbss + (2a - l)Mbo], where f i s the tissue perfusion rate in ml g-' s-'; is the brain:blood partition coefficient; Tlb is the spin- lattice relaxation time of brain water magnetization in the absence of perfusion and cross-relaxation; Mbo is the brain water signal intensity from the image without arterial spin labeling; Mbss is the brain water signal intensity from the image with arterial spin labeling; and a is the degree of arterial spin labeling. In this work, a value of Tlb = 1.7 s was used in all perfusion calculations (5). The expression above ignores the influence of cross-relaxation between tissue water and macromolecular spins when saturation of the macromolecules is avoided. The full theory for this case is presented in the accompanying paper, which shows that this effect is measurable but small (8).

D-Amphetamine sulfate (Sigma Chemical Co., St. Louis, MO) was administered intraperitoneally at a dose of 20 mg/kg in saline solution. Five sets of five-slice images with and without arterial spin labeling were acquired in a total scanning time of 50 min before the injection of d-amphetamine sulfate, for measuring the baseline perfusion. Another five sets of five-slice images with and without arterial spin labeling were acquired after the intraperitoneal administration of amphetamine. Four rats were used in this study. Amphetamines are known to stimulate blood flow for over an hour. Regions of interest corresponding to different structures in the brain were delineated by eye using either the images with or without arterial spin labeling. The selected regions were then overlaid onto the perfusion images to extract the average perfusion in each region before and after the administration of amphetamine.

RESULTS

A schematic of the two-coil system used to avoid MTC effects during arterial spin labeling is shown in Fig. 1b. The 0.5-cm diameter butterfly surface coil was placed perpendicularly to the 3-cm diameter saddle coil to improve the decoupling achieved with the use of PIN diodes, The isolation between the two coils, measured as the voltage induced at the saddle coil when RF power was applied to the labeling coil, was of the order of -74 dB .

Figure 2a shows a series of magnitude calculated coronal gradient echo images of a rat neck, acquired 1.3 cm downstream from the labeling plane. This is as far away from the labeling plane where the carotid arteries can still be imaged before they bifurcate into the brain micro- vasculature. For each image, arterial spin labeling was accomplished with the 0.5-cm butterfly surface coil at varying RF irradiation amplitudes of approximately: (i) 0 ,

Multi-Slice MRI of Rat Brain Perfusion 211

a 1

* 0.8 P

% 0.4 nl

10.2

FIG. 1. (a) Horizontal MR image of a rat head and neck acquired with a Bruker 7-cm diameter volume coil, showing the labeling plane and the control plane for RF irradiation, as well as the location of the detection slice for MRI of perfusion. (b) Same image showing the location of the labeling coil and the detection coil when the two-coil system is used for perfusion MRI.

(ii) 5, (iii) 10, (iv) 20, (v) 50, and (vi) 100 mG. The signal intensities in the carotid arteries (arrows) decrease and then increase as the B, strength increases. Because the images are magnitude calculated, the increase in signal intensities after passing through a minimum indicates the inversion of the arterial spins. Figure 2b shows a plot of the degree of labeling, a, as a function of the labeling pulse strength with the butterfly surface coil for the same rat of Fig. 2a. Saturation of the arterial spins (a = 0.5) occurs at a B, strength of approximately 15 mG, and the

0 20 40 60 80 100 120 BI Strength

0

b (mG)

FIG. 2. (a) Magnitude-calculated coronal MR images of a rat neck with different degrees of arterial spin labeling using the butterfly surface coil. The approximate B, strengths used were: (i) 0, (ii) 5, (iii) 10, (iv) 20, (v) 50, and (vi) 100 mG. (b) Plot of the degree of labeling as a function of the labeling pulse strength with the butterfly surface coil. In this case, a maximum value of a = 0.84 was obtained for the degree of labeling.

degree of labeling plateaus at a = 0.84 for B, strengths greater than 30 mG. The degree of spin labeling obtained with the butterfly surface coil from two rats under normocarbic conditions was a = 0.82 & 0.03. This value of a is decreased by relaxation of labeled arterial spins from the labeling plane to the region of interest in the brain. It has previously been shown that this decrease in a due to transit-time is approximately 15% at normal perfusion and decreases at elevated flow (5, 10). Therefore, in the following experiments, the effect of transit time on a was ignored and the value of 0.82 for a was used.

To see if any MTC effects occur with the two-coil system, horizontal images were obtained. Figure 3 shows a horizontal slice of a rat brain acquired with a single coil corresponding to the difference between an image with RF power applied at a frequency conforming to the control plane in Fig. l a and an image with RF power applied at a frequency conforming to the labeling plane in Fig. la. Perfect subtraction occurs only at the center of the image, showing that only a single coronal slice centered between the labeling plane and the control plane can be used for imaging of perfusion by arterial spin labeling with a single coil. Figure 4a shows a horizontal slice of a rat brain obtained using a standard spin-warp sequence. Figure 4b shows the same slice acquired with a single

212 Silva et al.

FIG. 3. Difference image obtained with a single coil from two horizontal images of a rat brain acquired with RF irradiation at the control plane and with RF irradiation at the labeling plane indi- cated in Fig. la. MTC effects are controlled for only at the center of the image, demonstrating that only a single coronal slice, centered between the labeling plane and the control plane, can be used for imaging of perfusion by arterial spin labeling with a single coil.

FIG. 4. (a) Horizontal spin-warp M R image of a rat brain acquired with a single coil and without RF irradiation for arterial spin labeling. (b) The same image acquired with RF irradiation applied at the labeling plane in Fig. la during the TR time. The variation in signal intensity across the image is due to MTC effects. The difference between the image in (b) and the image in (a) is shown in (c). (d) Horizontal image of a rat brain without RF power applied to the labeling coil in the two-coil system. (e) Same image with RF power applied to the labeling coil. (0 Difference image ((d)-(e)).

effects (compare with Fig. 4b). Figure 4f shows the difference between the image in Fig. 4d and the image in Fig. 4e. Perfect subtraction occurs in the muscle through- out the image, showing that no MTC effects are present (compare with Fig. 4c). There is a small difference in the brain due to perfusion.

Figure 5a shows a horizontal image from a rat brain under slightly hypercapnic [pCO, = 52.5 mmHg) conditions. The corresponding perfusion image is shown in Fig. 5b. The five consecutive coronal perfusion slices corresponding to the prescriptions given in Fig. 5a are shown in Fig. 5c, demonstrating that multi-slice perfusion images can be obtained.

Figure 6a shows a five-slice coronal spin-warp M R image acquired from a rat brain under normocarbic conditions (pC0, = 30.6 mmHg). The corresponding perfusion images before (Fig. 6b) and after [Fig. 6c) an intraperitoneal administration of 20 mg/kg of D-amphetamine in saline are also shown. The difference between the perfusion images before and after the administration of amphetamines is shown in Fig. 6d. It can be clearly seen that amphetamines increase cerebral blood flow in a region specific manner. For example, the cortex shows an increase in perfusion, which gets larger toward the fron- tal slices. Table 1 summarizes the effects of amphetamines on cerebral blood flow to different regions in four rats. Besides the cortex, the cingulate, the caudate putamen, and the olfactory tubercle also show significant increases in perfusion after the administration of amphetamine. The hippocampus shows a slight decrease in perfusion after the administration of amphetamine. How- ever, the thalamus, hypothalamus, and the amigdala show no significant response to amphetamine.

DISCUSSION

Arterial water spins are labeled by using arterial blood flow in the presence of a B, field gradient and a B, RF irradiation to fulfill the conditions for adiabatic fast passage inversion (9). One important parameter to be determined to enable quantitation of perfusion with the arterial spin labeling method is the degree of labeling, a. This parameter is, in principle, dependent on the blood flow velocity, the gradient strength, and the RF power used for arterial spin labeling. It has been shown that, with the

coil in which RF power was applied during the TR time at a frequency that corresponds to the labeling plane of Fig. la. The variation in signal intensity across the image due to MTC effects is clearly noticeable. Figure 4c shows the difference between the image in Fig. 4a and the image in Fig. 4b. The RF power employed in Fig. 4b to label the arterial spins saturates macromolecules in the whole head of the rat, causing a decrease in tissue water signal due to MTC effects, and preventing proper subtraction of the images obtained with (Fig. 4b) and without (Fig. 4a) RF irradiation for arterial spin labeling. The same experiment was repeated using the two-coil system. Figure 4d shows a horizontal image of a rat brain obtained with the two-coil system when no RF Power was to *e labeling coil. Figure 4e shows the same slice orientation with RF power applied to the labeling coil. No variations in signal intensity are seen across the image due to MTC

FIG. 5. (a) Horizontal spin-warp image of a rat brain under slightly hypercapnic conditions @co, = 52.5 mmHg). (b) Corresponding perfusion image. (c) Transverse perfusion images of the same rat corresponding to the prescriptions given in (a).

Multi-Slice MRI of Rat Brain Perfusion 213

FIG. 6. Five-slice perfusion images of a rat brain before and after the intraperitoneal administration of 20 mg/kg of D-amphetamine in saline: (a) anatomical spin-echo slices; (b) perfusion images before the administration of amphetamine; (c) perfusion images after the administration of D-amphetamine; (d) difference images found by subtracting (b) from (c).

Table 1 Comparison of Perfusion Values (mean 2 SD) in ml g-' min-' for Different Regions of the Brain before and after the Intraperitoneal Administration of 20 mg/kg of D-Amphetamine in Saline

Region Cingulate Cortex Caudate Hippocampus Hypothalamus Thalamus Arnigdala put a m e n Olfactory tubercle

Control 1.28 5 0.41 0.84 5 0.28 0.83 5 0.43 1.07 5 0.40 0.88 2 0.39 0.89 5 0.37 1.10 t 0.39 1.26 t 0.37 Amphetamine 1.91 t 0.28' 1.79 ? 0.50a 1.21 ? 0.35" 0.81 t 0.44' 0.78 2 0.41 0.95 2 0.38 1.17 5 0.32 1.59 5 0.42' Data shown is from four rats.

Significant change in perfusion compared with control (P < 0.1).

gradient strength and RF power normally used in these experiments, the degree of arterial spin labeling is not affected by the changing blood velocity over a broad range (10). However, it is important to determine the RF power levels at the transmitter output necessary to achieve spin inversion every time a different coil geom- etry is used. The value found for the degree of arterial spin labeling of a butterfly coil, a = 0.82 ? 0.03, shows that this coil is capable of efficiently inverting arterial water spins. This value of a is 10% higher than the degree of labeling found for a 7-cm volume coil (10). This slight increase could be explained, in part, by the fact that the two-coil scheme avoids MTC effects in blood as well as in brain, so that no signal decrease of tagged arterial spins due to saturation of macromolecules in blood occur before their exchange with tissue water. If this is the case, a better labeling efficiency can be achieved with the use of the two-coil scheme. The use of a small surface coil to achieve arterial spin labeling also decreases the amount of RF power deposition associated with arterial spin labeling, making the technique safer for use in humans.

The images in Figs. 4e and 4f and the rat brain perfusion images shown in Fig. 5 demonstrate that MTC effects can be avoided by physically isolating the RF power used for labeling arterial spins from the organ of interest.

Perfusion can be measured with any slice orientation and the whole organ of interest can be imaged with a 3D or multi-slice MFU sequence. There is no loss in tissue signal intensity due to MTC effects when a two-coil scheme is used for perfusion measurements. Therefore, there is substantially more signal increasing the sensitivity of the arterial spin labeling technique. However, MTC effects decrease the apparent TI of tissue water (5, 6). Thus, the MRI sequence used for measuring perfusion with a single coil is less affected by short TR times than when a two- coil system is used. The competition between signal intensity and apparent TI means that some of the improve- ment in sensitivity when the two-coil system is used for measuring perfusion is lost if short TR times are used. The two-coil system should have better sensitivity in a one-shot rapid imaging sequence.

Previously the theory for quantitating perfusion has been presented for the case when arterial spin labeling causes MTC effects (5). This theory needs to be extended for the case when macromolecules are not saturated. Even though there are no MTC effects, the presence of cross-relaxation between intracellular water and macromolecules affects water relaxation. This theory is presented with some of the ramifications for perfusion imaging in the accompanying paper (8). The two-coil scheme is of easy design and implementation for mea-

2 14 Silva et al.

suring rat brain perfusion. For other organs, such as kidney or heart, this approach may be difficult. It would be interesting to develop labeling schemes that use a single coil but manage to avoid MTC effects.

D-Amphetamine is a powerful stimulant that induces increased locomotor activity and stereotyped behavior in humans and animals (11). Its mechanism of action is to affect the release and uptake of neurotransmitters at cathecolaminergic synapses in the brain. D-Amphet- amine has been shown to inhibit uptake of dopamine in the corpus striatum as well as to affect neurotransmitters in noradrenergic systems. Using deoxyglucose uptake, it has been shown that D-amphetamine stimulates brain metabolism in a region specific manner (11). Increases in deoxyglucose uptake were measured in cortical regions as well as the subthalamic nucleus and the substantia nigra. Small but significant decreases in deoxyglucose uptake were measured in regions of the thalamus and hypothalamus. In addition to the deoxyglucose work, it is well known that amphetamine stimulates whole brain blood flow in rats (12, 13) and that cortical blood flow is increased (14). Using the two-coil approach, we have been able to look at the effects of amphetamines on blood flow in a region specific manner. In agreement with deoxyglucose results, amphetamine caused a significant increase in flow in the cortex, cingulate, and caudate putamen. Interestingly small decreases in flow to the hippocampus were detected in agreement with the deoxyglucose results. Unfortunately the multi-slice imaging technique used here required long acquisition times. It will be interesting to extend arterial spin labeling tech- niques to enable rapid 3D determination of perfusion so that the time course of amphetamine effects on blood flow can be examined. There is great interest in the cascade of events coupling electrical activity to neural metabolism to blood flow, because this is the cascade that is the basis for functional imaging by MRI (15, 16). Amphetamine administration may be a good model to study this cascade in rat brain. Chronic administration of D-amphetamines induces a psychotic state in humans. Therefore, it would be interesting to see how chronic administration of amphetamines affects cerebral blood flow in the rat. The noninvasive method of perfusion MRI using arterial spin labeling will enable us to study alterations that occur in single animals over time.

In conclusion, the use of a two-coil system for the measurement of perfusion by arterial spin labeling shows the advantages of eliminating the signal decrease due to MTC effects so that 3D or multi-slice imaging of perfusion in the rat brain can be performed. The experiments described in this work have shown that good decoupling can be achieved between the labeling and the imaging coils to prevent saturation of macromolecules, a good degree of labeling can be delivered by a small surface coil and no constraints regarding position or orientation of the regions of interest exist when a two-coil scheme is used for perfusion imaging. This technique has been used to investigate the regional increases in blood flow that accompany amphetamine administration and should be useful for imaging changes in cerebral blood flow associated with pharmacological interventions, brain injury due to stroke or trauma, and task activation of specific regions of the brain.

ACKNOWLEDGMENTS

The authors thank Mr. Donald Bennett and Ms. Maryann Butowitz for technical assistance, and Dr. Steven D. For- man for helpful discussions.

REFERENCES 1. J. A. Detre, J. S. Leigh, D. S. Williams, A. P. Koretsky,

Perfusion imaging. Magn. Reson. Med. 23, 37-45 (1992). 2. E. G. Walsh, K. Minematsu, J. Leppo, S . C. Moore, Radioac-

tive microsphere validation of a volume localized continu- ous saturation perfusion measurement. Magn. Reson. Med.

3. D. S. Williams, J. A. Detre, J. S. Leigh, A. P. Koretsky, Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proc. Natl. Acad. Sci. (USA) 89, 212- 216 (1992).

4. D. S. Williams, W. Zhana, A. P. Koretskv, S. Adler, Perfusion

31, 147-153 (1994).

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16,

imaging of the rat kidney with MR. Ridiology 190(3), 813- 818 (1994). W. Zhang, D. S. Williams, J. A. Detre, A. P. Koretsky, Mea- surement of brain perfusion by volume-localized NMR spec- troscopy using inversion of arterial water spins: accounting for transit time and cross-relaxation. Magn. Reson. Med. 25,

S. D. Wolff, R. S. Balaban, Magnetization transfer contrast (MTC) and tissue water proton relaxation in vivo. Magn. Reson. Med. 10, 135-144 (1989). A. C. Silva, W. Zhang, D. S. Williams, A. P. Koretsky, Multi- slice brain perfusion imaging of the rat by arterial spin labeling using two actively decoupled RF coils, in “Proc., SMRM, 12th Annual Meeting, New York, 1993,” p. 632. W. Zhang, A. C. Silva, D. S. Williams, A. P. Koretsky, NMR measurement of perfusion using arterial spin labeling without saturation of macromolecular spins. Magn. Reson. Med., in press. W. T. Dixon, L. N. Du, D. D. Faul, M. Gado, S. Rossnick, Projection angiograms of blood labeled by adiabatic fast passage. Magn. Reson. Med. 3, 454-462 (1986). W. Zhang, D. S. Williams, A. P. Koretsky, Measurement of rat brain perfusion by NMR using spin labeling of arterial water: in vivo determination of the degree of arterial spin labeling. Magn. Reson. Med. 29, 416-421 (1993). L. R. Wechsler, H. E. Savaki L. Sokoloff, Effects of d- and 1-amphetamine on local cerebral glucose utilization in the conscious rat. J. Neurochem. 32, 15-22 (1979). C. Carlsson, M. HBgerdal, B. K. Seisjo, Influence of amphetamine sulfate on cerebral blood flow and metabolism. Acta Physiol. Scand. 94, 128-129 (1975). L. Berntman, C. Carlsson, M. Hhgerdal, B. K. Seisjo, Circu- lation and metabolic effects in the brain induced by amphetamine sulfate. Acta Physiol. Scand. 102, 310-323 (1978). J. A. Detre, D. S . Williams, A. P. Koretsky, Nuclear magnetic resonance determination of flow, lactate, and phosphate metabolites during amphetamine stimulation of the rat brain. NMR Biomed. 3(6), 272-278 (1990). K. K. Kwong, J. W. Belliveau, D. A. Chesler, I. E. Goldberg, R. M. Weisskoff, B. P. Poncelet, D. N. Kennedy, B. E. Hoppel, M. S. Cohen, R. Turner, H.-M. Cheng, T. J. Brady, B. R. Rosen, Dynamic magnetic resonance imaging of human brain activity during primary sensory stimulation. Proc. Natl. Acad. Sci. (USA) 89, 5675-5679 (1992). S. Ogawa, D. W. Tank, R. Menon, J. M. Ellermann, S. G. Kim, H. Merkle, K. Ugtirbil, Intrinsic signal changes accompanying sensory stimulation: functional brain mapping with magnetic resonance imaging. Proc. Natl. Acad. Sci. (USA)

362-371 (1992).

89, 5951-5955 (1992).

Multi-Slice MRI of Rat Brain Perfusion During Amphetamine Stimulation Using Arterial Spin Labeling

Documents

Transcript of Multi-Slice MRI of Rat Brain Perfusion During Amphetamine Stimulation Using Arterial Spin Labeling