Directory UMM :Data Elmu:jurnal:B:Brain Research:Vol887.Issue2.Dec2000:
www.elsevier.com / locate / bres
Research report
Effect of a neuronal sodium channel blocker on magnetic resonance
derived indices of brain water content during global cerebral ischemia
a,b b c b ,
*
Herbert Koinig
, John P. Williams , Michael J. Quast , Mark H. Zornow
a
Department of Anesthesiology and General Intensive Care, University of Vienna, Vienna, Austria b
Department of Anesthesiology, The University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-0591, USA c
Marie Hall Magnet Laboratory, Marine Biomedical Institute and Department of Anatomy and Neuroscience,
The University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-0591, USA Accepted 19 September 2000
Abstract
Diffusion-weighted magnetic resonance imaging (DWI) with calculation of the apparent diffusion coefficient (ADC) of water is a widely used noninvasive method to measure movement of water from the extracellular to the intracellular compartment during cerebral
1
ischemia. Lamotrigine, a neuronal Na channel blocker, has been shown to attenuate the increase in extracellular concentrations of excitatory amino acids (EAA) during ischemia and to improve neurological and histological outcome. Because of its proven ability to reduce EAA levels during ischemia, lamotrigine should also minimize excitotoxic-induced increases in intracellular water content and therefore attenuate changes in the ADC. In this study, we sought to determine the effect of lamotrigine on intra- and extracellular water shifts during transient global cerebral ischemia. Fifteen New Zealand white rabbits were anesthetized and randomized to one of three groups: a control group, a lamotrigine-treated group, or a sham group. After being positioned in the bore of the magnet, a 12-min 50-s period of global cerebral ischemia was induced by inflating a neck tourniquet. During ischemia and early reperfusion there was a similar and significant decrease of the ADC in both the lamotrigine and control group. The ADC in the sham ischemia group remained at baseline throughout the experiment. Lamotrigine-mediated blockade of voltage-gated sodium channels did not prevent the intracellular movement of water during 12 min 50 s of global ischemia, as measured by the ADC, suggesting that the ADC decline may not be mediated by voltage-gated sodium influx and glutamate release. 2000 Elsevier Science B.V. All rights reserved.
Theme: Disorders of the nervous system
Topic: Ischemia
Keywords: Diffusion-weighted magnetic resonance imaging; Global cerebral ischemia; Sodium channel blocker
1. Introduction apparent diffusion coefficient (ADC) of water molecules correlates with the changes of extracellular water volume
During cerebral ischemia, a compromised energy supply [17,18,21]. Ischemia leads to a decline of the ADC with a
in the neurons leads to failure of energy-dependent ion return to normal values during reperfusion in global [5,10]
exchange pumps. This results in anoxic depolarization of and focal ischemia [22]. Measuring the ADC during
cells, a run-down in transmembrane ion gradients, a ischemia allows for real-time, noninvasive monitoring of
massive release of excitatory amino acids (EAAs), and a intra- and extracellular water movements, which may
fluid shift from the extracellular to the intracellular com- result from acute excitotoxicity [9,26]. DWI is therefore
partment [29]. considered to be a powerful tool for evaluating ischemic
Numerous previous studies have demonstrated that damage during acute cerebral ischemia, and has been used
diffusion-weighted imaging (DWI) and calculation of the to assess therapeutic drug effects in several animal models
of focal cerebral ischemia [3,6,32,38].
1
The ability of lamotrigine, a neuronal Na channel
*Corresponding author. Tel.: 11-409-772-1221; fax: 1
1-409-772-blocker, to attenuate the release of EAAs has been shown
1224.
E-mail address: [email protected] (M.H. Zornow). in vitro [37], and various studies in animal models have
0006-8993 / 00 / $ – see front matter 2000 Elsevier Science B.V. All rights reserved. P I I : S 0 0 0 6 - 8 9 9 3 ( 0 0 ) 0 3 0 1 2 - 2
(2)
demonstrated the ability of lamotrigine to attenuate EAA neck for the later production of cerebral circulatory arrest.
accumulation in the extracellular space and its resulting At the end of the experiment, the animals were euthanized
excitotoxicity after cerebral ischemia [1,7]. Recently, by increasing the inspired halothane concentration to 5%
lamotrigine was shown to improve neurobehavioral and and injecting an intravenous dose of potassium chloride.
histologic outcome following global cerebral ischemia
[20]. 2.3. Drug administration
In the present study, we evaluated the use of an easily
reproducible model of transient global cerebral ischemia in Lamotrigine (Glaxo Wellcome Inc, Greenville, NC) was
the rabbit. This animal model requires minimal surgical administered intravenously at a dose of 50 mg / kg diluted
preparation and allows for single or multiple episodes of in 20 ml of deionized water in the lamotrigine group. This
cerebral ischemia of any desired duration. Using this dose of lamotrigine has been shown to prevent any
model, we investigated the effect of the preischemic increase of glutamate during 10 min of ischemia [1], and
intravenous administration of lamotrigine on ischemia- improved neurobehavioral outcome after a 6.5 min
is-induced cytotoxic brain edema through serial monitoring chemic episode [20]. The drug was infused over 20 min
of the ADC. starting 90 min before the onset of ischemia. The animals
in the control group received an identical volume of deionized water without lamotrigine. The infusion medium
2. Materials and methods was prepared by an investigator not involved in image acquisition and data analysis. To avoid any biasing of the
2.1. Animals results, blinding of the investigators was maintained until
data analysis was completed. This study was performed under a protocol approved by
the Institutional Animal Care and Use Committee at the 2.4. Induction of ischemia
University of Texas Medical Branch. Fifteen New Zealand
white rabbits, aged 4 months and weighing 3.560.25 kg After the animals had been positioned in the bore of the
(mean6S.D.), were randomly assigned to one of three magnet, baseline images were recorded. Seventy minutes
groups: control (n56), lamotrigine (n56), or sham (n53). after completion of the lamotrigine infusion, the MAP was
The animals were fasted for 24 h before the start of the decreased to between 25 and 50 mmHg with an
intraven-experiment and housed one per cage at the institutional ous bolus of trimethaphan (5 mg). The neck tourniquet
Animal Resource Center where they received routine was inflated to a pressure of 700 mmHg within 0.5 s using
veterinary care. a regulated source of compressed air. This protocol reliably
results in profound cerebral ischemia as evidenced by the
2.2. Surgical procedure prompt (,30 s) appearance of an isoelectric
electroence-phalogram [1,20]. After 12 min and 50 s of ischemia, the
The animals were anesthetized in a Plexiglas box with neck tourniquet was deflated and MAP was restored to
5% halothane in oxygen. After loss of the righting reflex, between 80 and 100 mmHg with a bolus of phenylephrine
endotracheal intubation was performed and the lungs were (5–10mg given intravenously).
mechanically ventilated (FiO251.0). The inspired
halothane concentration was reduced to 1% as soon as 2.5. Brain temperature measurements
mechanical ventilation was established. After infiltration
with 0.25% bupivacaine, the groin was incised and PE-90 Brain temperature measurements were carried out in an
catheters were inserted into the femoral artery and vein. additional two animals to monitor the actual change during
Mean arterial pressure (MAP) was measured throughout cerebral ischemia while in the magnet. A burr hole, 2 mm
the study. Serial arterial blood samples were intermittently in diameter, was drilled 4 mm posterior and 4 mm lateral
obtained during the experiment to measure pH, PO , and2 to the bregma in order to place a temperature probe into
PCO2 (1306 pH / Blood Gas Analyzer, Instrumentation the dorsal hippocampus. After securing the temperature
Laboratory, Lexington, MA). Mechanical ventilation was probe with dental acrylic the animals were treated
accord-adjusted to maintain the PCO between 35 and 40 mmHg.2 ing to the protocol described above.
Hemoglobin concentration was determined with a
CO-oximeter (482 CO-Oximeter, Instrumentation Laboratory, 2.6. Magnetic resonance imaging
Lexington, MA). Body temperature was monitored with a
rectal temperature probe and maintained at 388C with a Following surgical preparation, the animals were
se-circulating water heating pad (Gaymar Industries Inc., cured in the prone position on a Plexiglas cradle. The head
Orchard Park, NY) which was placed around the animal’s of the rabbit was fixed in a nonmagnetic head holder and
body throughout the experiment. An inflatable neck tour- positioned in the bore of the magnet. Magnetic resonance
(3)
Instruments Ltd., Oxford, UK) and a custom-built surface intravenously during the sixth frame of each movie. A
1
coil with a 9-cm diameter tuned to the H resonance washout period of 30 min was used after each bolus
frequency (200.056 MHz). The magnetic field homogen- tracking movie to reduce contrast agent concentration. The
eity throughout the sample volume was then maximized by sequence of DWIs, bolus tracking movies, cerebral
is-shimming on the water free induction decay (FID) using a chemia, and the measurement of physiologic variables are
Varian Unity Inova NMR console coupled to a Sun depicted in Fig. 1.
Microsystems host computer (Ultra Sparc-Station 10)
running VNMR 6.1B software (Varian Inc., Palo Alto, 2.7. Data analysis
CA). Sagittal and coronal pilot scans were acquired for the
selection of a set of five transverse imaging slices. DWIs Computation of quantitative ADC images was
per-were obtained using a multi-slice spin-echo diffusion formed on a Sun Microsystems computer (Ultra
Sparc-sequence with a diffusion gradient applied along the Station 10). Regional evaluations of ADC mean values
transverse horizontal (‘X’) axis. Imaging acquisition pa- were carried out in two regions of interest (ROI’s), i.e.,
rameters were as follows: five consecutive slices centered bilateral hippocampus (Fig. 2). Size and location of the
on the slice of interest, 1.6-mm slice thickness, repetition / ROI’s were selected with the reference to an atlas of the
echo times of 3000 / 65 ms, 10310 cm field of view, rabbit brain [34] and high-resolution images acquired using
using128 phase encoding steps, and one echo was averaged a conventional spin-echo sequence.
per phase encoding step. For quantitative determination of Phase files were made for each frame of the bolus
the ADC, DWIs with different-weighting factors (b-val- tracking experiment. These were converted into individual
2
ues50, 293, 661 s / mm ) were recorded before ischemia frames of the bolus tracking movie in a proprietary format
and after 10, 30, 60, and 90 min of reperfusion. To using an in-house developed software package (Transit).
improve temporal resolution in the periischemic period, An artery (pixel) at the base of the brain and the sagittal
2
additional single DWIs (b5661 s / mm ) were acquired vein (outlined) were then selected to provide arterial (A )
RT
every 6 min and 25 s. These images and the unweighted and venous residue times (V ). From these data the total
RT 2
spin echoes (b50 s / mm ) of the previously recorded signal observed (V ) was determined for each vessel.
AUC
baseline measurements were used for calculation of the The parenchyma of the whole brain was then outlined and
ADC during ischemia and early reperfusion. the average flow curve determined to give WB and
AUC
Bolus track imaging was used to assess cerebral blood parenchymal residue time (WB ). These data were then
RT
flow (CBF) during baseline and reperfusion. Two bolus used to calculate the cerebral blood flow (CBF) before and
tracking movies were acquired using a flash sequence with after ischemia using the following method:
the following parameters: 50 single slices in rapid
succes-V ?WB
sion, total acquisition time per frame of 520 ms, repetition / 21 21 AUC AUC
]]]]]
CBF (ml?100 g ?min )5 21
echo time of 8 / 3 ms, and field of interest equal to 11311 WB T2A
R RT
cm. The movies were recorded 25 min before ischemia and
at the end of the experimental protocol. A bolus of 0.5 ml 2.8. Statistical analysis
gadopentetate dimeglumine contrast agent (Magnevist ,
Berlex Imaging Laboratories, Wayne, NJ) was injected Data were analyzed using a commercially available
Fig. 1. Time-course of the experimental protocol. The small arrows indicate times when temperature and mean arterial blood pressure (MAP) readings were recorded. Plus (1) marks indicate times when arterial blood gas samples were obtained. B, baseline images; I, ischemia images; Re, reperfusion images.
(4)
Fig. 2. Regions of interest were drawn on the image by a single observer. The hippocampus was selected for each animal using the help of both an atlas of the rabbit brain [34] and a high-resolution image acquired with a conventional spin-echo sequence.
computer program (StatView 5.0; SAS Institute Inc., San groups during the entire experiment. Brain temperature
Francisco, CA). Physiologic parameters (pH, PO , PCO ,2 2 measurements in two animals showed a 3.0560.358C
hemoglobin, MAP, and body temperature) were compared decrease at the end of the 12 min and 50 s ischemic
with repeated-measures analysis of variance (ANOVA) and episode. As per the protocol, MAP during ischemia was
Scheffe’s test. ADC and CBF values were compared using lowered to 25 to 50 mmHg in the control and lamotrigine
factorial ANOVA and Dunnett’s test. A paired t-test was groups. After deflating the neck tourniquet, MAP returned
used to test for statistically significant changes between to baseline values. No significant differences in MAP were
baseline and minimum ADC values as well as pre- and observed between groups. CBF (ml / 100 g / min) during
postischemic CBF values in both groups. Differences were baseline was 54617 in the control group, 52622 in the
considered statistically significant at P,0.05; data are lamotrigine group, and 5363 in the sham group. During
presented as mean6S.D. reperfusion CBF was 5464 in the control group, 49612 in
the lamotrigine group, and 5765 in the sham group. There
were no significant differences in CBF between groups or
3. Results over time.
3.1. Physiologic data 3.2. ADC — hippocampus
There were no significant differences in pH, PO , PCO ,2 2 Fig. 3 shows ADCs versus time curves for each of the
and hemoglobin among groups and no significant changes groups during the experiment. The mean value of the ADC
26 2
were observed over time (Table 1). Body temperature within the hippocampus was 989625310 mm / s in the
26 2
(5)
Table 1 the lamotrigine group. In the second image during is-a
Summary of physiologic data chemia (I2), the mean ADC decreased to 87% of baseline Control Lamotrigine Sham in the control group and to 85% in the lamotrigine group. group group group In the first image during reperfusion (Re1), ADCs con-Weight (kg) 3.5160.24 3.3860.22 3.8260.07 tinued to decline to 80% and 83% of baseline in the
Rectal temperature (8C) control and lamotrigine groups, respectively. The mini-Baseline 37.960.1 38.060.2 38.060.1 mum observed ADC during ischemia was not different Ischemia 37.960.3 38.060.2 37.960.2
between control and lamotrigine groups. Twenty-five
Early reperfusion 37.960.2 37.960.2 38.160.0
minutes of reperfusion returned mean ADCs to baseline.
Late reperfusion 37.960.2 38.060.2 38.160.0
pH The mean ADC in the sham group remained at baseline
Baseline 7.3660.06 7.3460.05 7.3860.09 throughout the entire experiment. Early reperfusion 7.3360.07 7.3460.06 7.3960.01
Late reperfusion 7.3360.08 7.3560.05 7.3960.01 PCO (mmHg)2
4. Discussion
Baseline 36.461.3 35.461.5 37.261.3 Early reperfusion 38.262.1 35.761.4 38.862.6
Late reperfusion 38.261.1 36.461.0 38.761.2 This manuscript describes the use of a model of
PO (mmHg)2 transient global cerebral ischemia ideally suited for use in
Baseline 417641 377650 482616
MR studies. Using this model, we sought to investigate the
Early reperfusion 388641 427642 434692
effects of lamotrigine on ischemia-induced cytotoxic brain
Late reperfusion 384631 429647 414671
Hb (g / dl) edema using DWI. Despite the expected ischemia-induced
Baseline 13.161.3 11.661.0 12.261.3 decrease in the ADC, there was no significant difference in Early reperfusion 12.461.5 12.161.8 12.362.8 the mean hippocampal ADC after 12 min and 50 s of Late reperfusion 13.062.1 12.461.7 12.663.2
global cerebral ischemia between the control or
lamot-MAP (mmHg)
rigine-treated groups. The ADC recovered to baseline
Baseline 7565 69612 7363
Early reperfusion 7768 69610 7263 following reperfusion within 25 min in both groups.
Late reperfusion 7263 66610 7263 The validity of our ischemia model has been proven in
b 21 21
CBF (ml?100 g ?min ) several previous studies. Reproducible neurologic deficits
Baseline 54617 52622 5363
[2,20] and histopathology [20], as well as a consistent
Reperfusion 5464 49612 5765
increase in glutamate release [1] have been found using
a
Values are mean6S.D.
this model. In contrast to cardiac arrest or surgical arterial
b
For the calculation method of CBF, see Materials and methods.
occlusion models, this model of transient global cerebral
Hb, hemoglobin; MAP, mean arterial pressure; CBF, cerebral blood flow.
ischemia is easy to implement in the MRI environment and
26 2 allows fine control of the duration of ischemia. Our
group and 95067310 mm / s in the sham group. In the
ischemia model produced a reliable ADC decrease during first ischemic image (I1), the ADC in the control group
ischemia in this study, and is therefore ideally suited for was numerically less (but not statistically so) than that for
further MRI studies.
Although many neuroscientists have focussed their attention on drugs which may have neuroprotective prop-erties when administered after an ischemic episode, there is a very large population of patients who are at significant risk for neurologic injury and who could benefit from preischemic drug therapy. Nearly 1 million patients under-go coronary artery bypass surgery every year [27]. Adverse cerebral outcome after coronary bypass continues to be a significant clinical problem, causing strokes in 1–5% of patients, transient ischemic attacks in 4–14%, and neuro-cognitive deficits in as many as 30 – 88% [25,27,30,36]. Preoperative treatment of these patients with an effective neuroprotective agent which has few side effects could dramatically improve outcome and reduce the high neuro-logic morbidity of this type of surgery.
Neuronal ischemia results in a rapid decrease in ATP
Fig. 3. Changes in the apparent diffusion coefficient (ADC) during the concentration, a subsequent loss of ion homeostasis, and experiment. B, baseline images; I, ischemia images; Re, reperfusion
the release of excitatory amino acids and other
neuro-images. Solid bar represents a 12-min 50-s ischemic period. *P,0.05
transmitters [29]. EAAs acting at postsynaptic receptors
when compared with the sham group (analysis of variance, Dunnett’s test)
1 1 and to baseline (paired t-test). Values are mean6S.D. and depletion of ATP stores result in Na / K pump
(6)
1 2
dysfunction, increased Na influx, and cotransport of Cl occur immediately after onset of ischemia, even before the
and water, resulting in cytotoxic brain edema [14]. DWI loss of ion hemostasis and anoxic depolarization [15,33].
1 1
and calculation of the ADC can detect changes in the brain The Na / H exchange transporter, which is activated
1
tissue within minutes after the onset of ischemia. It is during cerebral ischemia [33], cycles Na into the cell and
1
widely accepted that the ischemic ADC decrease is due to H to the extracellular compartment [19]. This active
water protons moving from the extracellular to the intracel- proton transfer causes an osmotic water shift into the cell.
lular compartments, which have different diffusion prop- Other possible mechanisms for water movement into the
erties. Restricted intracellular proton diffusion occurs intracellular compartment and subsequent ADC changes
because of the high protein content and the presence of include generation of free radicals and membrane
break-hydrophobic membranes hinder the free movement of down [19] or changes in the membrane water permeability
water molecules within the cell [18]. [16].
1
Lamotrigine inhibits Na influx by blocking voltage- To achieve better temporal resolution, ADC maps in this
1 1 2
sensitive Na channels. The neuronal Na channel bloc- study were calculated using single DWIs (b5661 s / mm )
kade decreases the frequency of action potentials, reduces acquired during ischemia and an additional previously
2
presynaptic glutamate release, and diminishes accumula- recorded single DWI (b50 s / mm ). This enabled us to
1 1
tion of intracellular Na . Reduced intracellular Na load- decrease acquisition time to approximately 6.5 min. To
1 1
ing decreases the activity of energy-dependent 3Na / 2K assure the validity of this method, which has already been
1
adenosine triphosphatase and preserves a sufficient Na used in previous studies [10], we calculated ADC maps
1
gradient used for the Na glutamate uptake carrier, which using two different methods during baseline conditions and
forms an important reuptake mechanism, thus terminating compared ADC values calculated from three different
2
glutamate or other EAA transmitter action. In addition, DWIs (b50, 293, 661 s / mm ) with the method described
1
inhibition of Na influx also prevents the periischemic above in each of the animals. There was no difference
release of glutamate by reversed operation of the glutamate between these ADC values indicating that our approach is
uptake carrier, which is the main mechanism of glutamate valid.
release during ischemia [31,35]. Glutamate acting at The changes observed in the ADC measured during
postsynaptic NMDA receptors is known to increase the ischemia and the first reperfusion period could potentially
movement of sodium and water into the postsynaptic be influenced by changes in the water T and T relaxation1 2
neuron. parameters. Using a similar period of transient global
Lamotrigine has been approved for use in humans for ischemia in rats it has been shown that T relaxation times2
the treatment of epilepsy. It has excellent bioavailability remain constant during ischemia and reperfusion [12]. It
when taken orally and is associated with few side effects may therefore be concluded that changes in T relaxation2
[11]. Several studies using our model of transient global do not contribute to signal alterations of cerebral
metabo-cerebral ischemia in rabbits have proven the validity of this lites detected by DWI [12]. In this study it has also been
model and demonstrated that lamotrigine attenuates gluta- demonstrated that no significant changes in major cerebral
mate concentrations during 10 min of cerebral ischemia [1] metabolite signal intensity occur during ischemia and
and improves histological and neurobehavioral outcome reperfusion [12]. Apart from depletion of brain glucose
[20]. Lamotrigine has also shown neuroprotective effects pools and the accumulation of lactic acid, it has been
in a focal ischemia model measuring the lesion volume previously demonstrated that levels of these metabolites
with DWI [28]. remain unchanged during brief periods of transient global
However, in this study, lamotrigine did not attenuate ischemia [4,24]. It can then be inferred that the water
decreases in the ADC during ischemia. It may be that the relaxation properties remain unchanged as well. This
duration of ischemia used in this study was excessively conclusion is supported by the fact that T1 and T2
long for lamotrigine to demonstrate a beneficial effect. relaxation times of water did not change during focal
Suggestive evidence in support of this possibility is found ischemia [13,23]. Hence, we believe that the changes we
in the numerically higher ADC in the lamotrigine group observed in the calculated ADC’s accurately reflect the
during the first ischemic image (I1). Alternatively, our internal environment of the cerebral tissue under study.
study may suggest that ischemic ADC changes are not Other important factors that may affect the diffusion
simply mediated by glutamate activated, voltage-gated coefficient of water are brain temperature and CBF [17].
1
Na channels, given the fact that the same dose of To quantify the influence of the brain temperature on ADC
lamotrigine prevented any increase in cerebral glutamate values, we measured brain temperature changes during
during 10 min of transient global cerebral ischemia [1]. ischemia. We found a mean maximum brain temperature
This explanation is supported by a recent study [15] decrease of 38C during ischemia. In previous studies, an
26 2
observing a decrease of the ADC of water during cerebral ADC drop of 10310 mm / s per 18C was calculated for
ischemia even before the onset of anoxic depolarization. rats and cats [8]. Assuming the same value for the rabbit
26 2
Ischemic ADC decline may be a result of mechanisms brain, a decrease of 30310 mm / s could be due to
26
(7)
2 26 2 DeWitt, W.E. Johnston, Lamotrigine attenuates cortical glutamate
mm / s in the control group and 167310 mm / s in the
release during global cerebral ischemia in pigs on cardiopulmonary
lamotrigine group, only 15% to 17% of the mean ADC
bypass, Anesthesiology 90 (1999) 844–854.
change during ischemia may have been caused by tempera- [8] D. Davis, J. Ulatowski, S. Eleff, M. Izuta, S. Mori, D. Shungu, P.C.
ture decrease. van Zijl, Rapid monitoring of changes in water diffusion coefficients
Because blood flow may also interfere with ADC during reversible ischemia in cat and rat brain, Magn. Reson. Med.
31 (1994) 454–460.
measurements, we measured pre- and postischemic CBF.
[9] R.M. Dijkhuizen, R.A. de Graaf, K.A. Tulleken, K. Nicolay,
CBF values showed no difference, thus indicating adequate
Changes in the diffusion of water and intracellular metabolites after
cerebral perfusion during baseline and reperfusion. excitotoxic injury and global ischemia in neonatal rat brain, J.
In this study, lamotrigine did not attenuate the decrease Cereb. Blood Flow Metab. 19 (1999) 341–349.
in ADC during ischemia despite the fact that an identical [10] M. Fischer, K. Bockhorst, M. Hoehn-Berlage, B. Schmitz, K.A.
Hossmann, Imaging of the apparent diffusion coefficient for the
dose of lamotrigine in the same animal model prevented
evaluation of cerebral metabolic recovery after cardiac arrest, Magn.
increases in extracellular glutamate during 10 min of
Reson. Imaging 13 (1995) 781–790.
ischemia [1]. This finding suggests either that the duration [11] A. Fitton, K.L. Goa, Lamotrigine. An update of its pharmacology
of ischemia exceeded the ability of lamotrigine to prevent and therapeutic use in epilepsy, Drugs 50 (1995) 691–713.
excitoxic-induced movement of water or that anoxic [12] H. Fujimori, T. Michaelis, M. Wick, J. Frahm, Proton T2 relaxation
of cerebral metabolites during transient global ischemia in rat brain,
depolarization and elevation of glutamate levels do not
Magn. Reson. Med. 39 (1998) 647–650.
necessarily need to precede changes in the ADC. In this
[13] I.M. Germano, L.H. Pitts, I. Berry, M. Moseley, Magnetic resonance
case, mechanisms other than voltage-gated sodium influx
imaging and 31P magnetic resonance spectroscopy for evaluating
and EAA release may be responsible for the observed focal cerebral ischemia, J. Neurosurg. 70 (1989) 612–618.
ADC decline. Additional studies using the finer temporal [14] M.P. Goldberg, D.W. Choi, Combined oxygen and glucose
depriva-tion in cortical cell culture: calcium-dependent and
calcium-in-resolution provided by echo planar imaging may help
dependent mechanisms of neuronal injury, J. Neurosci. 13 (1993)
resolve these issues.
3510–3524.
[15] N.G. Harris, E. Zilkha, J. Houseman, M.R. Symms, T.P. Ob-renovitch, S.R. Williams, The relationship between the apparent
Acknowledgements diffusion coefficient measured by magnetic resonance imaging, anoxic depolarization, and glutamate efflux during experimental cerebral ischemia, J. Cereb. Blood Flow Metab. 20 (2000) 28–36.
This study was supported in part by the National
[16] J.A. Helpern, R.J. Ordidge, R.A. Knight, The effect of cell
mem-Institute of Health grant 2 RO1 NS 29403 to Dr Zornow,
brane water permeability on the apparent diffusion coefficient of
and a stipend from the Max Kade Foundation to Dr water, in: Proceedings SNMR, 11th Annual Meeting, 1992, p. 1201.
Koinig. [17] K.A. Hossmann, M. Fischer, K. Bockhorst, M. Hoehn-Berlage,
NMR imaging of the apparent diffusion coefficient (ADC) for the evaluation of metabolic suppression and recovery after prolonged cerebral ischemia, J. Cereb. Blood Flow Metab. 14 (1994) 723–731.
References [18] K.A. Hossmann, M. Hoehn-Berlage, Diffusion and perfusion MR
imaging of cerebral ischemia, Cerebrovasc. Brain Metab. Rev. 7 [1] A. Bacher, M.H. Zornow, Lamotrigine inhibits extracellular gluta- (1995) 187–217.
mate accumulation during transient global cerebral ischemia in [19] H.K. Kimelberg, Current concepts of brain edema. Review of rabbits, Anesthesiology 86 (1997) 459–463. laboratory investigations, J. Neurosurg. 83 (1995) 1051–1059. [2] A.J. Baker, M.H. Zornow, M.S. Scheller, T.L. Yaksh, S.R. Skilling, [20] J.Y. Kwon, A. Bacher, D.J. Deyo, M.R. Grafe, J.F. Disterhoft, T.
D.H. Smullin, A.A. Larson, R. Kuczenski, Changes in extracellular Uchida, M.H. Zornow, Effects of hypothermia and lamotrigine on concentrations of glutamate, aspartate, glycine, dopamine, serotonin, trace-conditioned learning after global cerebral ischemia in rabbits, and dopamine metabolites after transient global ischemia in the Exp. Neurol. 159 (1999) 105–113.
rabbit brain, J. Neurochem. 57 (1991) 1370–1379. [21] K. Minematsu, L. Li, M. Fisher, C.H. Sotak, M.A. Davis, M.S. [3] L. Belayev, W. Zhao, P.M. Pattany, R.G. Weaver, P.W. Huh, B. Lin, Fiandaca, Diffusion-weighted magnetic resonance imaging: rapid R. Busto, M.D. Ginsberg, Diffusion-weighted magnetic resonance and quantitative detection of focal brain ischemia, Neurology 42 imaging confirms marked neuroprotective efficacy of albumin (1992) 235–240.
therapy in focal cerebral ischemia, Stroke 29 (1998) 2587–2599. [22] K. Minematsu, L. Li, C.H. Sotak, M.A. Davis, M. Fisher, Reversible [4] A. Bizzi, A. Righini, R. Turner, D. Le Bihan, K.H. Bockhorst, J.R. focal ischemic injury demonstrated by diffusion-weighted magnetic
Alger, Imaging focal reperfusion injury following global ischemia resonance imaging in rats, Stroke 23 (1992) 1304–1310.
with diffusion-weighted magnetic resonance imaging and 1H-mag- [23] M.E. Moseley, Y. Cohen, J. Mintorovitch, L. Chileuitt, H. Shimizu, netic resonance spectroscopy, Magn. Reson. Imaging 14 (1996) J. Kucharczyk, M.F. Wendland, P.R. Weinstein, Early detection of 581–592. regional cerebral ischemia in cats: comparison of diffusion- and [5] A. Bizzi, A. Righini, R. Turner, D. LeBihan, D. DesPres, G. Di T2-weighted MRI and spectroscopy, Magn. Reson. Med. 14 (1990)
Chiro, J.R. Alger, MR of diffusion slowing in global cerebral 330–346.
ischemia, AJNR Am. J. Neuroradiol. 14 (1993) 1347–1354. [24] Y. Nagatomo, M. Wick, F. Prielmeier, J. Frahm, Dynamic moni-[6] S. Chandra, R.F. White, D. Everding, G.Z. Feuerstein, R.W. Coatney, toring of cerebral metabolites during and after transient global S.K. Sarkar, F.C. Barone, Use of diffusion-weighted MRI and ischemia in rats by quantitative proton NMR spectroscopy in vivo, neurological deficit scores to demonstrate beneficial effects of NMR Biomed. 8 (1995) 265–270.
isradipine in a rat model of focal ischemia, Pharmacology 58 (1999) [25] M.F. Newman, N.D. Croughwell, J.A. Blumenthal, W.D. White, J.B.
292–299. Lewis, L.R. Smith, P. Frasco, E.A. Towner, R.M. Schell, B.J.
(8)
cardiopulmonary bypass. Association with postoperative cognitive M. Fisher, A novel endothelin antagonist, A-127722, attenuates dysfunction, Circulation 90 (II) (1994) 243–249. ischemic lesion size in rats with temporary middle cerebral artery [26] C. Pierpaoli, J.R. Alger, A. Righini, J. Mattiello, R. Dickerson, D. occlusion: a diffusion and perfusion MRI study, Stroke 29 (1998)
Des Pres, A. Barnett, G. Di Chiro, High temporal resolution 850–857.
diffusion MRI of global cerebral ischemia and reperfusion, J. Cereb. [33] D.L. Taylor, T.P. Obrenovitch, L. Symon, Changes in extracellular Blood Flow Metab. 16 (1996) 892–905. acid-base homeostasis in cerebral ischemia, Neurochem. Res. 21 [27] G.W. Roach, M. Kanchuger, C.M. Mangano, M. Newman, N. (1996) 1013–1021.
Nussmeier, R. Wolman, A. Aggarwal, K. Marschall, S.H. Graham, [34] I. Urban, P. Richard, A Stereotaxic Atlas of the New Zealand C. Ley, Adverse cerebral outcomes after coronary bypass surgery. Rabbit’s Brain 1-86, Thomas, Springfield, IL, 1972.
Multicenter Study of Perioperative Ischemia Research Group and [35] J. Urenjak, T.P. Obrenovitch, Pharmacological modulation of
volt-1
the Ischemia Research and Education Foundation Investigators, New age-gated Na channels: a rational and effective strategy against Engl. J. Med. 335 (1996) 1857–1863. ischemic brain damage, Pharmacol. Rev. 48 (1996) 21–67. [28] J.B. Schulz, R.T. Matthews, B.G. Jenkins, P. Brar, M.F. Beal, [36] G.W. Wahl, A.J. Swinburne, A.J. Fedullo, D.K. Lee, K. Bixby,
Improved therapeutic window for treatment of histotoxic hypoxia Long-term outcome when major complications follow coronary with a free radical spin trap, J. Cereb. Blood Flow Metab. 15 (1995) artery bypass graft surgery. Recovery after complicated coronary
948–952. artery bypass graft surgery, Chest 110 (1996) 1394–1398.
[29] B.K. Siesjo, Pathophysiology and treatment of focal cerebral [37] P.C. Waldmeier, P.A. Baumann, P. Wicki, J.J. Feldtrauer, C. Stierlin, ischemia. Part II: Mechanisms of damage and treatment, J. Neuro- M. Schmutz, Similar potency of carbamazepine, oxcarbazepine, and surg. 77 (1992) 337–354. lamotrigine in inhibiting the release of glutamate and other neuro-[30] K.A. Sotaniemi, Long-term neurologic outcome after cardiac opera- transmitters, Neurology 45 (1995) 1907–1913.
tion, Ann. Thorac. Surg. 59 (1995) 1336–1339. [38] M.A. Yenari, J.T. Palmer, G.H. Sun, A. de Crespigny, M.E. Mosely, [31] M. Szatkowski, D. Attwell, Triggering and execution of neuronal G.K. Steinberg, Time-course and treatment response with SNX-111, death in brain ischaemia: two phases of glutamate release by an N-type calcium channel blocker, in a rodent model of focal different mechanisms, Trends Neurosci. 17 (1994) 359–365. cerebral ischemia using diffusion-weighted MRI, Brain Res. 739 [32] T. Tatlisumak, R.A. Carano, K. Takano, T.J. Opgenorth, C.H. Sotak, (1996) 36–45.
(1)
Instruments Ltd., Oxford, UK) and a custom-built surface intravenously during the sixth frame of each movie. A
1
coil with a 9-cm diameter tuned to the H resonance washout period of 30 min was used after each bolus frequency (200.056 MHz). The magnetic field homogen- tracking movie to reduce contrast agent concentration. The eity throughout the sample volume was then maximized by sequence of DWIs, bolus tracking movies, cerebral is-shimming on the water free induction decay (FID) using a chemia, and the measurement of physiologic variables are Varian Unity Inova NMR console coupled to a Sun depicted in Fig. 1.
Microsystems host computer (Ultra Sparc-Station 10)
running VNMR 6.1B software (Varian Inc., Palo Alto, 2.7. Data analysis CA). Sagittal and coronal pilot scans were acquired for the
selection of a set of five transverse imaging slices. DWIs Computation of quantitative ADC images was per-were obtained using a multi-slice spin-echo diffusion formed on a Sun Microsystems computer (Ultra Sparc-sequence with a diffusion gradient applied along the Station 10). Regional evaluations of ADC mean values transverse horizontal (‘X’) axis. Imaging acquisition pa- were carried out in two regions of interest (ROI’s), i.e., rameters were as follows: five consecutive slices centered bilateral hippocampus (Fig. 2). Size and location of the on the slice of interest, 1.6-mm slice thickness, repetition / ROI’s were selected with the reference to an atlas of the echo times of 3000 / 65 ms, 10310 cm field of view, rabbit brain [34] and high-resolution images acquired using using128 phase encoding steps, and one echo was averaged a conventional spin-echo sequence.
per phase encoding step. For quantitative determination of Phase files were made for each frame of the bolus the ADC, DWIs with different-weighting factors (b-val- tracking experiment. These were converted into individual
2
ues50, 293, 661 s / mm ) were recorded before ischemia frames of the bolus tracking movie in a proprietary format and after 10, 30, 60, and 90 min of reperfusion. To using an in-house developed software package (Transit). improve temporal resolution in the periischemic period, An artery (pixel) at the base of the brain and the sagittal
2
additional single DWIs (b5661 s / mm ) were acquired vein (outlined) were then selected to provide arterial (A )
RT
every 6 min and 25 s. These images and the unweighted and venous residue times (V ). From these data the total
RT 2
spin echoes (b50 s / mm ) of the previously recorded signal observed (V ) was determined for each vessel.
AUC
baseline measurements were used for calculation of the The parenchyma of the whole brain was then outlined and ADC during ischemia and early reperfusion. the average flow curve determined to give WB and
AUC
Bolus track imaging was used to assess cerebral blood parenchymal residue time (WB ). These data were then
RT
flow (CBF) during baseline and reperfusion. Two bolus used to calculate the cerebral blood flow (CBF) before and tracking movies were acquired using a flash sequence with after ischemia using the following method:
the following parameters: 50 single slices in rapid
succes-V ?WB
sion, total acquisition time per frame of 520 ms, repetition / 21 21 AUC AUC
]]]]]
CBF (ml?100 g ?min )5 21
echo time of 8 / 3 ms, and field of interest equal to 11311 WB T2A
R RT
cm. The movies were recorded 25 min before ischemia and
at the end of the experimental protocol. A bolus of 0.5 ml 2.8. Statistical analysis
gadopentetate dimeglumine contrast agent (Magnevist ,
Berlex Imaging Laboratories, Wayne, NJ) was injected Data were analyzed using a commercially available
Fig. 1. Time-course of the experimental protocol. The small arrows indicate times when temperature and mean arterial blood pressure (MAP) readings were recorded. Plus (1) marks indicate times when arterial blood gas samples were obtained. B, baseline images; I, ischemia images; Re, reperfusion images.
(2)
Fig. 2. Regions of interest were drawn on the image by a single observer. The hippocampus was selected for each animal using the help of both an atlas of the rabbit brain [34] and a high-resolution image acquired with a conventional spin-echo sequence.
computer program (StatView 5.0; SAS Institute Inc., San groups during the entire experiment. Brain temperature Francisco, CA). Physiologic parameters (pH, PO , PCO ,2 2 measurements in two animals showed a 3.0560.358C hemoglobin, MAP, and body temperature) were compared decrease at the end of the 12 min and 50 s ischemic with repeated-measures analysis of variance (ANOVA) and episode. As per the protocol, MAP during ischemia was Scheffe’s test. ADC and CBF values were compared using lowered to 25 to 50 mmHg in the control and lamotrigine factorial ANOVA and Dunnett’s test. A paired t-test was groups. After deflating the neck tourniquet, MAP returned used to test for statistically significant changes between to baseline values. No significant differences in MAP were baseline and minimum ADC values as well as pre- and observed between groups. CBF (ml / 100 g / min) during postischemic CBF values in both groups. Differences were baseline was 54617 in the control group, 52622 in the considered statistically significant at P,0.05; data are lamotrigine group, and 5363 in the sham group. During presented as mean6S.D. reperfusion CBF was 5464 in the control group, 49612 in the lamotrigine group, and 5765 in the sham group. There were no significant differences in CBF between groups or
3. Results over time.
3.1. Physiologic data 3.2. ADC — hippocampus
There were no significant differences in pH, PO , PCO ,2 2 Fig. 3 shows ADCs versus time curves for each of the and hemoglobin among groups and no significant changes groups during the experiment. The mean value of the ADC
26 2
were observed over time (Table 1). Body temperature within the hippocampus was 989625310 mm / s in the
26 2
(3)
Table 1 the lamotrigine group. In the second image during
is-a
Summary of physiologic data chemia (I2), the mean ADC decreased to 87% of baseline
Control Lamotrigine Sham in the control group and to 85% in the lamotrigine group.
group group group In the first image during reperfusion (Re1), ADCs
con-Weight (kg) 3.5160.24 3.3860.22 3.8260.07 tinued to decline to 80% and 83% of baseline in the
Rectal temperature (8C) control and lamotrigine groups, respectively. The
mini-Baseline 37.960.1 38.060.2 38.060.1
mum observed ADC during ischemia was not different
Ischemia 37.960.3 38.060.2 37.960.2
between control and lamotrigine groups. Twenty-five
Early reperfusion 37.960.2 37.960.2 38.160.0
minutes of reperfusion returned mean ADCs to baseline.
Late reperfusion 37.960.2 38.060.2 38.160.0
pH The mean ADC in the sham group remained at baseline
Baseline 7.3660.06 7.3460.05 7.3860.09 throughout the entire experiment.
Early reperfusion 7.3360.07 7.3460.06 7.3960.01 Late reperfusion 7.3360.08 7.3560.05 7.3960.01 PCO (mmHg)2
4. Discussion
Baseline 36.461.3 35.461.5 37.261.3
Early reperfusion 38.262.1 35.761.4 38.862.6
Late reperfusion 38.261.1 36.461.0 38.761.2 This manuscript describes the use of a model of
PO (mmHg)2 transient global cerebral ischemia ideally suited for use in
Baseline 417641 377650 482616
MR studies. Using this model, we sought to investigate the
Early reperfusion 388641 427642 434692
effects of lamotrigine on ischemia-induced cytotoxic brain
Late reperfusion 384631 429647 414671
Hb (g / dl) edema using DWI. Despite the expected ischemia-induced
Baseline 13.161.3 11.661.0 12.261.3 decrease in the ADC, there was no significant difference in
Early reperfusion 12.461.5 12.161.8 12.362.8
the mean hippocampal ADC after 12 min and 50 s of
Late reperfusion 13.062.1 12.461.7 12.663.2
global cerebral ischemia between the control or lamot-MAP (mmHg)
rigine-treated groups. The ADC recovered to baseline
Baseline 7565 69612 7363
Early reperfusion 7768 69610 7263 following reperfusion within 25 min in both groups.
Late reperfusion 7263 66610 7263 The validity of our ischemia model has been proven in
b 21 21
CBF (ml?100 g ?min ) several previous studies. Reproducible neurologic deficits
Baseline 54617 52622 5363
[2,20] and histopathology [20], as well as a consistent
Reperfusion 5464 49612 5765
increase in glutamate release [1] have been found using
a
Values are mean6S.D.
this model. In contrast to cardiac arrest or surgical arterial
b
For the calculation method of CBF, see Materials and methods.
occlusion models, this model of transient global cerebral Hb, hemoglobin; MAP, mean arterial pressure; CBF, cerebral blood flow.
ischemia is easy to implement in the MRI environment and
26 2 allows fine control of the duration of ischemia. Our
group and 95067310 mm / s in the sham group. In the
ischemia model produced a reliable ADC decrease during first ischemic image (I1), the ADC in the control group
ischemia in this study, and is therefore ideally suited for was numerically less (but not statistically so) than that for
further MRI studies.
Although many neuroscientists have focussed their attention on drugs which may have neuroprotective prop-erties when administered after an ischemic episode, there is a very large population of patients who are at significant risk for neurologic injury and who could benefit from preischemic drug therapy. Nearly 1 million patients under-go coronary artery bypass surgery every year [27]. Adverse cerebral outcome after coronary bypass continues to be a significant clinical problem, causing strokes in 1–5% of patients, transient ischemic attacks in 4–14%, and neuro-cognitive deficits in as many as 30 – 88% [25,27,30,36]. Preoperative treatment of these patients with an effective neuroprotective agent which has few side effects could dramatically improve outcome and reduce the high neuro-logic morbidity of this type of surgery.
Neuronal ischemia results in a rapid decrease in ATP Fig. 3. Changes in the apparent diffusion coefficient (ADC) during the concentration, a subsequent loss of ion homeostasis, and experiment. B, baseline images; I, ischemia images; Re, reperfusion
the release of excitatory amino acids and other neuro-images. Solid bar represents a 12-min 50-s ischemic period. *P,0.05
transmitters [29]. EAAs acting at postsynaptic receptors when compared with the sham group (analysis of variance, Dunnett’s test)
1 1
(4)
1 2
dysfunction, increased Na influx, and cotransport of Cl occur immediately after onset of ischemia, even before the and water, resulting in cytotoxic brain edema [14]. DWI loss of ion hemostasis and anoxic depolarization [15,33].
1 1
and calculation of the ADC can detect changes in the brain The Na / H exchange transporter, which is activated
1
tissue within minutes after the onset of ischemia. It is during cerebral ischemia [33], cycles Na into the cell and
1
widely accepted that the ischemic ADC decrease is due to H to the extracellular compartment [19]. This active water protons moving from the extracellular to the intracel- proton transfer causes an osmotic water shift into the cell. lular compartments, which have different diffusion prop- Other possible mechanisms for water movement into the erties. Restricted intracellular proton diffusion occurs intracellular compartment and subsequent ADC changes because of the high protein content and the presence of include generation of free radicals and membrane break-hydrophobic membranes hinder the free movement of down [19] or changes in the membrane water permeability water molecules within the cell [18]. [16].
1
Lamotrigine inhibits Na influx by blocking voltage- To achieve better temporal resolution, ADC maps in this
1 1 2
sensitive Na channels. The neuronal Na channel bloc- study were calculated using single DWIs (b5661 s / mm ) kade decreases the frequency of action potentials, reduces acquired during ischemia and an additional previously
2
presynaptic glutamate release, and diminishes accumula- recorded single DWI (b50 s / mm ). This enabled us to
1 1
tion of intracellular Na . Reduced intracellular Na load- decrease acquisition time to approximately 6.5 min. To
1 1
ing decreases the activity of energy-dependent 3Na / 2K assure the validity of this method, which has already been
1
adenosine triphosphatase and preserves a sufficient Na used in previous studies [10], we calculated ADC maps
1
gradient used for the Na glutamate uptake carrier, which using two different methods during baseline conditions and forms an important reuptake mechanism, thus terminating compared ADC values calculated from three different
2
glutamate or other EAA transmitter action. In addition, DWIs (b50, 293, 661 s / mm ) with the method described
1
inhibition of Na influx also prevents the periischemic above in each of the animals. There was no difference release of glutamate by reversed operation of the glutamate between these ADC values indicating that our approach is uptake carrier, which is the main mechanism of glutamate valid.
release during ischemia [31,35]. Glutamate acting at The changes observed in the ADC measured during postsynaptic NMDA receptors is known to increase the ischemia and the first reperfusion period could potentially movement of sodium and water into the postsynaptic be influenced by changes in the water T and T relaxation1 2
neuron. parameters. Using a similar period of transient global Lamotrigine has been approved for use in humans for ischemia in rats it has been shown that T relaxation times2
the treatment of epilepsy. It has excellent bioavailability remain constant during ischemia and reperfusion [12]. It when taken orally and is associated with few side effects may therefore be concluded that changes in T relaxation2
[11]. Several studies using our model of transient global do not contribute to signal alterations of cerebral metabo-cerebral ischemia in rabbits have proven the validity of this lites detected by DWI [12]. In this study it has also been model and demonstrated that lamotrigine attenuates gluta- demonstrated that no significant changes in major cerebral mate concentrations during 10 min of cerebral ischemia [1] metabolite signal intensity occur during ischemia and and improves histological and neurobehavioral outcome reperfusion [12]. Apart from depletion of brain glucose [20]. Lamotrigine has also shown neuroprotective effects pools and the accumulation of lactic acid, it has been in a focal ischemia model measuring the lesion volume previously demonstrated that levels of these metabolites with DWI [28]. remain unchanged during brief periods of transient global However, in this study, lamotrigine did not attenuate ischemia [4,24]. It can then be inferred that the water decreases in the ADC during ischemia. It may be that the relaxation properties remain unchanged as well. This duration of ischemia used in this study was excessively conclusion is supported by the fact that T1 and T2
long for lamotrigine to demonstrate a beneficial effect. relaxation times of water did not change during focal Suggestive evidence in support of this possibility is found ischemia [13,23]. Hence, we believe that the changes we in the numerically higher ADC in the lamotrigine group observed in the calculated ADC’s accurately reflect the during the first ischemic image (I1). Alternatively, our internal environment of the cerebral tissue under study. study may suggest that ischemic ADC changes are not Other important factors that may affect the diffusion simply mediated by glutamate activated, voltage-gated coefficient of water are brain temperature and CBF [17].
1
Na channels, given the fact that the same dose of To quantify the influence of the brain temperature on ADC lamotrigine prevented any increase in cerebral glutamate values, we measured brain temperature changes during during 10 min of transient global cerebral ischemia [1]. ischemia. We found a mean maximum brain temperature This explanation is supported by a recent study [15] decrease of 38C during ischemia. In previous studies, an
26 2
observing a decrease of the ADC of water during cerebral ADC drop of 10310 mm / s per 18C was calculated for ischemia even before the onset of anoxic depolarization. rats and cats [8]. Assuming the same value for the rabbit
26 2
Ischemic ADC decline may be a result of mechanisms brain, a decrease of 30310 mm / s could be due to
26
(5)
2 26 2 DeWitt, W.E. Johnston, Lamotrigine attenuates cortical glutamate
mm / s in the control group and 167310 mm / s in the
release during global cerebral ischemia in pigs on cardiopulmonary lamotrigine group, only 15% to 17% of the mean ADC
bypass, Anesthesiology 90 (1999) 844–854.
change during ischemia may have been caused by tempera- [8] D. Davis, J. Ulatowski, S. Eleff, M. Izuta, S. Mori, D. Shungu, P.C. ture decrease. van Zijl, Rapid monitoring of changes in water diffusion coefficients Because blood flow may also interfere with ADC during reversible ischemia in cat and rat brain, Magn. Reson. Med.
31 (1994) 454–460. measurements, we measured pre- and postischemic CBF.
[9] R.M. Dijkhuizen, R.A. de Graaf, K.A. Tulleken, K. Nicolay, CBF values showed no difference, thus indicating adequate
Changes in the diffusion of water and intracellular metabolites after cerebral perfusion during baseline and reperfusion. excitotoxic injury and global ischemia in neonatal rat brain, J.
In this study, lamotrigine did not attenuate the decrease Cereb. Blood Flow Metab. 19 (1999) 341–349.
in ADC during ischemia despite the fact that an identical [10] M. Fischer, K. Bockhorst, M. Hoehn-Berlage, B. Schmitz, K.A. Hossmann, Imaging of the apparent diffusion coefficient for the dose of lamotrigine in the same animal model prevented
evaluation of cerebral metabolic recovery after cardiac arrest, Magn. increases in extracellular glutamate during 10 min of
Reson. Imaging 13 (1995) 781–790.
ischemia [1]. This finding suggests either that the duration [11] A. Fitton, K.L. Goa, Lamotrigine. An update of its pharmacology of ischemia exceeded the ability of lamotrigine to prevent and therapeutic use in epilepsy, Drugs 50 (1995) 691–713. excitoxic-induced movement of water or that anoxic [12] H. Fujimori, T. Michaelis, M. Wick, J. Frahm, Proton T2 relaxation
of cerebral metabolites during transient global ischemia in rat brain, depolarization and elevation of glutamate levels do not
Magn. Reson. Med. 39 (1998) 647–650. necessarily need to precede changes in the ADC. In this
[13] I.M. Germano, L.H. Pitts, I. Berry, M. Moseley, Magnetic resonance case, mechanisms other than voltage-gated sodium influx
imaging and 31P magnetic resonance spectroscopy for evaluating and EAA release may be responsible for the observed focal cerebral ischemia, J. Neurosurg. 70 (1989) 612–618. ADC decline. Additional studies using the finer temporal [14] M.P. Goldberg, D.W. Choi, Combined oxygen and glucose
depriva-tion in cortical cell culture: calcium-dependent and calcium-in-resolution provided by echo planar imaging may help
dependent mechanisms of neuronal injury, J. Neurosci. 13 (1993) resolve these issues.
3510–3524.
[15] N.G. Harris, E. Zilkha, J. Houseman, M.R. Symms, T.P. Ob-renovitch, S.R. Williams, The relationship between the apparent
Acknowledgements diffusion coefficient measured by magnetic resonance imaging,
anoxic depolarization, and glutamate efflux during experimental cerebral ischemia, J. Cereb. Blood Flow Metab. 20 (2000) 28–36. This study was supported in part by the National
[16] J.A. Helpern, R.J. Ordidge, R.A. Knight, The effect of cell mem-Institute of Health grant 2 RO1 NS 29403 to Dr Zornow,
brane water permeability on the apparent diffusion coefficient of and a stipend from the Max Kade Foundation to Dr water, in: Proceedings SNMR, 11th Annual Meeting, 1992, p. 1201. Koinig. [17] K.A. Hossmann, M. Fischer, K. Bockhorst, M. Hoehn-Berlage, NMR imaging of the apparent diffusion coefficient (ADC) for the evaluation of metabolic suppression and recovery after prolonged cerebral ischemia, J. Cereb. Blood Flow Metab. 14 (1994) 723–731.
References [18] K.A. Hossmann, M. Hoehn-Berlage, Diffusion and perfusion MR
imaging of cerebral ischemia, Cerebrovasc. Brain Metab. Rev. 7 [1] A. Bacher, M.H. Zornow, Lamotrigine inhibits extracellular gluta- (1995) 187–217.
mate accumulation during transient global cerebral ischemia in [19] H.K. Kimelberg, Current concepts of brain edema. Review of rabbits, Anesthesiology 86 (1997) 459–463. laboratory investigations, J. Neurosurg. 83 (1995) 1051–1059. [2] A.J. Baker, M.H. Zornow, M.S. Scheller, T.L. Yaksh, S.R. Skilling, [20] J.Y. Kwon, A. Bacher, D.J. Deyo, M.R. Grafe, J.F. Disterhoft, T.
D.H. Smullin, A.A. Larson, R. Kuczenski, Changes in extracellular Uchida, M.H. Zornow, Effects of hypothermia and lamotrigine on concentrations of glutamate, aspartate, glycine, dopamine, serotonin, trace-conditioned learning after global cerebral ischemia in rabbits, and dopamine metabolites after transient global ischemia in the Exp. Neurol. 159 (1999) 105–113.
rabbit brain, J. Neurochem. 57 (1991) 1370–1379. [21] K. Minematsu, L. Li, M. Fisher, C.H. Sotak, M.A. Davis, M.S. [3] L. Belayev, W. Zhao, P.M. Pattany, R.G. Weaver, P.W. Huh, B. Lin, Fiandaca, Diffusion-weighted magnetic resonance imaging: rapid R. Busto, M.D. Ginsberg, Diffusion-weighted magnetic resonance and quantitative detection of focal brain ischemia, Neurology 42 imaging confirms marked neuroprotective efficacy of albumin (1992) 235–240.
therapy in focal cerebral ischemia, Stroke 29 (1998) 2587–2599. [22] K. Minematsu, L. Li, C.H. Sotak, M.A. Davis, M. Fisher, Reversible [4] A. Bizzi, A. Righini, R. Turner, D. Le Bihan, K.H. Bockhorst, J.R. focal ischemic injury demonstrated by diffusion-weighted magnetic
Alger, Imaging focal reperfusion injury following global ischemia resonance imaging in rats, Stroke 23 (1992) 1304–1310.
with diffusion-weighted magnetic resonance imaging and 1H-mag- [23] M.E. Moseley, Y. Cohen, J. Mintorovitch, L. Chileuitt, H. Shimizu, netic resonance spectroscopy, Magn. Reson. Imaging 14 (1996) J. Kucharczyk, M.F. Wendland, P.R. Weinstein, Early detection of
581–592. regional cerebral ischemia in cats: comparison of diffusion- and
[5] A. Bizzi, A. Righini, R. Turner, D. LeBihan, D. DesPres, G. Di T2-weighted MRI and spectroscopy, Magn. Reson. Med. 14 (1990) Chiro, J.R. Alger, MR of diffusion slowing in global cerebral 330–346.
ischemia, AJNR Am. J. Neuroradiol. 14 (1993) 1347–1354. [24] Y. Nagatomo, M. Wick, F. Prielmeier, J. Frahm, Dynamic moni-[6] S. Chandra, R.F. White, D. Everding, G.Z. Feuerstein, R.W. Coatney, toring of cerebral metabolites during and after transient global S.K. Sarkar, F.C. Barone, Use of diffusion-weighted MRI and ischemia in rats by quantitative proton NMR spectroscopy in vivo, neurological deficit scores to demonstrate beneficial effects of NMR Biomed. 8 (1995) 265–270.
isradipine in a rat model of focal ischemia, Pharmacology 58 (1999) [25] M.F. Newman, N.D. Croughwell, J.A. Blumenthal, W.D. White, J.B.
292–299. Lewis, L.R. Smith, P. Frasco, E.A. Towner, R.M. Schell, B.J.
(6)
cardiopulmonary bypass. Association with postoperative cognitive M. Fisher, A novel endothelin antagonist, A-127722, attenuates dysfunction, Circulation 90 (II) (1994) 243–249. ischemic lesion size in rats with temporary middle cerebral artery [26] C. Pierpaoli, J.R. Alger, A. Righini, J. Mattiello, R. Dickerson, D. occlusion: a diffusion and perfusion MRI study, Stroke 29 (1998)
Des Pres, A. Barnett, G. Di Chiro, High temporal resolution 850–857.
diffusion MRI of global cerebral ischemia and reperfusion, J. Cereb. [33] D.L. Taylor, T.P. Obrenovitch, L. Symon, Changes in extracellular Blood Flow Metab. 16 (1996) 892–905. acid-base homeostasis in cerebral ischemia, Neurochem. Res. 21 [27] G.W. Roach, M. Kanchuger, C.M. Mangano, M. Newman, N. (1996) 1013–1021.
Nussmeier, R. Wolman, A. Aggarwal, K. Marschall, S.H. Graham, [34] I. Urban, P. Richard, A Stereotaxic Atlas of the New Zealand C. Ley, Adverse cerebral outcomes after coronary bypass surgery. Rabbit’s Brain 1-86, Thomas, Springfield, IL, 1972.
Multicenter Study of Perioperative Ischemia Research Group and [35] J. Urenjak, T.P. Obrenovitch, Pharmacological modulation of
volt-1
the Ischemia Research and Education Foundation Investigators, New age-gated Na channels: a rational and effective strategy against Engl. J. Med. 335 (1996) 1857–1863. ischemic brain damage, Pharmacol. Rev. 48 (1996) 21–67. [28] J.B. Schulz, R.T. Matthews, B.G. Jenkins, P. Brar, M.F. Beal, [36] G.W. Wahl, A.J. Swinburne, A.J. Fedullo, D.K. Lee, K. Bixby,
Improved therapeutic window for treatment of histotoxic hypoxia Long-term outcome when major complications follow coronary with a free radical spin trap, J. Cereb. Blood Flow Metab. 15 (1995) artery bypass graft surgery. Recovery after complicated coronary
948–952. artery bypass graft surgery, Chest 110 (1996) 1394–1398.
[29] B.K. Siesjo, Pathophysiology and treatment of focal cerebral [37] P.C. Waldmeier, P.A. Baumann, P. Wicki, J.J. Feldtrauer, C. Stierlin, ischemia. Part II: Mechanisms of damage and treatment, J. Neuro- M. Schmutz, Similar potency of carbamazepine, oxcarbazepine, and surg. 77 (1992) 337–354. lamotrigine in inhibiting the release of glutamate and other neuro-[30] K.A. Sotaniemi, Long-term neurologic outcome after cardiac opera- transmitters, Neurology 45 (1995) 1907–1913.
tion, Ann. Thorac. Surg. 59 (1995) 1336–1339. [38] M.A. Yenari, J.T. Palmer, G.H. Sun, A. de Crespigny, M.E. Mosely, [31] M. Szatkowski, D. Attwell, Triggering and execution of neuronal G.K. Steinberg, Time-course and treatment response with SNX-111, death in brain ischaemia: two phases of glutamate release by an N-type calcium channel blocker, in a rodent model of focal different mechanisms, Trends Neurosci. 17 (1994) 359–365. cerebral ischemia using diffusion-weighted MRI, Brain Res. 739 [32] T. Tatlisumak, R.A. Carano, K. Takano, T.J. Opgenorth, C.H. Sotak, (1996) 36–45.