(Stroke. 2001;32:783.)
© 2001 American Heart Association, Inc.
Original Contributions |
From the Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University, School of Medicine, Baltimore, Md.
Correspondence to Richard J. Traystman, PhD, Distinguished University Research Professor Vice Chairman, Department of Anesthesiology and Critical Care Medicine, Johns Hopkins Medical Institutions, Blalock 1408, 600 N Wolfe St, Baltimore, MD 21287. E-mail rtraystm{at}@jhmi.edu
| Abstract |
|---|
|
|
|---|
MethodsHalothane-anesthetized male Wistar rats were subjected to 2 hours of MCA occlusion by the intraluminal occlusion technique. Physiological variables were controlled, and ipsilateral cortical perfusion was monitored by laser Doppler flowmetry throughout ischemia. At onset of reperfusion, rats received intravenous LTG 5, 10, or 20 mg/kg or PBS (n=9 to 11 per group) during 15 minutes. Behavioral assessment was completed at 3 and 7 days after stroke, and the brain was harvested for histology (triphenyltetrazolium chloride staining).
ResultsValues are mean±SE. Cortical infarction volumes were unchanged in LTG-treated animals: 14±6% of contralateral cortex at 5 mg/kg LTG, 17±7% at 10 mg/kg, and 30±6% at 20 mg/kg, versus saline-treated cohorts (12±3%; P=0.19; n=9). Caudate-putamen infarction injury was also unchanged (37±11% of contralateral caudate-putamen at 5 mg/kg LTG, 44±8% at 10 mg/kg, and 65±9% at 20 mg/kg versus saline (38±11%; P=0.18). Total infarction was not different among groups (P=0.15). Consistent with histology, behavioral outcomes were unimproved by treatment.
ConclusionsHistological damage and behavioral recovery at 7 days after MCA occlusion was not altered by LTG treatment over the dosage range used in the present study.
Key Words: anticonvulsant cerebral ischemia, focal cerebral ischemia, transient stroke, experimental stroke, ischemic triazine
| Introduction |
|---|
|
|
|---|
The antiepileptic drug lamotrigine (LTG) is a phenyltriazine derivative that acts by stabilizing voltage-sensitive sodium channels in a usage-dependent manner, preventing glutamate and aspartate release and reversibly blocking excitatory neurotransmission7 8 (for review, see Tauboll and Gjerstad9 ). In addition, LTG also inhibits calcium currents in corticostriatal and hippocampal neurons, which suggests action at high voltage, release-coupled calcium channels.10 11 12 LTG has been used as a neuroprotective agent in several models of global4 8 13 14 and permanent focal cerebral ischemia,15 with varying degrees of efficacy. Most of these studies used preischemic or intraischemic drug administration, and the therapeutic utility and timing of LTG in stroke resuscitation remains unclear. Because the agent has not been evaluated in reversible focal stroke, it is not known whether LTG is efficacious when administered after ischemia and once perfusion has been restored. We hypothesized that if it is the dampening of intraischemic glutamate release that is critical to stroke outcome, then LTG administered even immediately on reperfusion would be too delayed to salvage tissue from infarction. Alternatively, a secondary and delayed rise in extracellular glutamate concentration has been characterized in transient ischemia that may represent a second window for antiglutamatergic intervention.16 17 The aim of the present study was to evaluate 7-day behavioral recovery and infarction volume when LTG is administered on recirculation after middle cerebral artery (MCA) occlusion.
| Subjects and Methods |
|---|
|
|
|---|
Behavioral testing to establish baseline parameters was conducted 1 to 3 days before MCAO. In brief, behavioral tests were conducted in a quiet room, during the light phase (between 1 and 3 PM EST). Between trials, testing apparatus was cleaned with a 75% alcohol solution. All trials were scored by the same experimentally uninformed observer.
Locomotor Balance and Coordination
Rats were placed at the center of a wooden bridge or
pole that was suspended approximately 60 cm above a foam pillow.
Animals were tested in random order by use of a 2-cm-wide bridge or a
2-cm-diameter pole. Latency to fall from the bridge or pole in seconds
was recorded. Animals that did not fall received a score of
120 s.
Strength and Agility
Rats were placed in the middle of a mesh screen
maintained at a 45° angle, facing downward. Time for each rat to
complete a 90° turn was recorded. In addition, ability to turn in
a narrow alley was measured as a measure of coordinated muscle movement
and agility. Animals were tested in a 12-cm-wide blind alley, with each
animal placed to face the back wall. Time to turn around and face the
open end of the alley was recorded. Last, forelimb strength and
grasping ability was evaluated by suspending the rats by their
forelimbs on a wire. The wire was stretched between 2 posts at a height
of 50 cm over a foam pillow. Time in seconds until the animal fell was
recorded. A score of zero was assigned to animals that did not
grasp the wire or that fell immediately. The trial lasted a maximum of
90 s.
On the day of surgery, each animal was anesthetized
with 1% to 1.5% halothane delivered by means of face mask in
oxygen-enriched air. A femoral artery and venous catheter were placed
for arterial blood pressure and blood gas measurement and
for infusion of drug and fluids. Rectal and temporalis muscle
temperatures were controlled at 36°C to 38°C with heat lamps.
Cortical perfusion by laser Doppler flowmetry (LDF; model
MBF3D, Moor Instruments Ltd) was measured as previously
described, with probe placement at 6 mm lateral and 2 mm
posterior to bregma. Focal cerebral ischemia was accomplished
using the intraluminal filament model (4-0 nylon monofilament suture)
of proximal MCAO as previously
described.18 19 The
right common carotid artery was exposed through a lateral incision,
separated from the vagus nerve, and ligated. The external carotid
artery was ligated, the occipital branch was cauterized, and the
pterygopalatine artery was ligated. An occluding filament was advanced
through the common carotid artery until the LDF signal displayed an
abrupt and significant reduction, confirming ongoing ischemia,
and then the filament was secured in place. Rats that did not
demonstrate a significant reduction in LDF signal (
45% of baseline
values) were excluded from study. Ischemic LDF was determined
over 5-minute periods at 5, 15, 30, 60, 90, 105, and 120 minutes after
MCAO, and the suture was withdrawn, with prompt restoration of blood
flow. LTG 5, 10, or 20 mg/kg or PBS vehicle was infused through the
femoral venous catheter during 15 minutes (infusion rate, 6 mL/h)
beginning at the onset of reperfusion as the occluding filament was
removed. Arterial blood pressure and LDF were recorded
continuously during vehicle or drug infusion. Arterial
blood gases were again measured at the end of the drug infusion, and
then all instrumentation was removed. The animal was allowed to recover
and was evaluated for behavioral deficits at 3 and 7 days of
reperfusion. On day 7, the brain was harvested under deep halothane
anesthesia. Tissue was sliced into seven 2-mm-thick coronal
sections for 2,3,5-triphenyltetrazolium
chloride staining and quantification through standard photography and
digital planimetry (SigmaScan Pro, Jandel). The infarcted area was
numerically integrated across each section and over the entire
ischemic hemisphere. Infarct volume was measured separately in
the cortex and caudate-putamen and expressed as volume percentage of
the contralateral structure. Ipsilateral total infarction was also
measured and expressed as a percentage of the contralateral structures
(sum of contralateral cortex and caudate putamen).
All values are reported as mean±SEM unless otherwise
indicated. Data from the pole and wire were not normally distributed;
therefore, data were transformed [log10(Y+1)] before
analysis. Transformation was not successful at normalizing the
data from the alley, inclined screen, and bridge. Thus, these data were
analyzed by Kruskal-Wallis 1-way ANOVA on ranks. If no
significant difference occurred between the experimental groups on any
of the test days, groups were collapsed and Kruskal-Wallis 1-way ANOVA
on ranks was conducted across time with post hoc comparisons conducted
by use of the method of Dunn. Data points >3 SDs from the mean were
removed before analysis (alley, n=3).
Physiological parameters and LDF were
subjected to 2-way ANOVA and post hoc Newman-Keuls test. Differences in
infarction volumes, mean-ischemic LDF, and plasma hormone
levels were determined by 1-way ANOVA. If significant differences were
found, a post hoc Newman-Keuls test was applied. Criterion for
statistical significance was
P
0.05.
| Results |
|---|
|
|
|---|
Physiological data are summarized in
Table 1
. Arterial blood pressure and
blood gases were similar among groups during MCAO and early
reperfusion. Furthermore, LTG administration did not alter temperature,
arterial blood gases, or arterial blood
pressure in the reperfusing animals at any of the 3 doses
(Table 1
). The ipsilateral LDF signal during MCAO decreased
rapidly to approximately 25% of baseline values and remained depressed
in each animal throughout occlusion. On reperfusion, LDF returned to
near baseline by 30 minutes. No difference existed in LDF among
treatment groups during MCAO occlusion or early reperfusion. Infusion
of saline or LTG did not alter recovery of the LDF signal. Five animals
were excluded from study because of inadequate reduction of the
intraischemic LDF signal, which suggested ineffective
occlusion. Two of these animals were saline treated, and 1 animal per
group was excluded from the LTG 5, 10, and 20 mg/kg groups.
|
Histological damage at 7 days after stroke
was not altered by LTG treatment. Cortical and caudate-putamen
infarction volumes were also unchanged by LTG- versus saline-treated
cohorts at any dosage level
(P=0.19 and
P=0.18, respectively, for
cortex and caudate-putamen analysis;
Figure
).
Total infarction volume was not different among groups: 16±4%,
18±7%, 22±7%, and 35±6% of contralateral volume for saline and
LTG 5, 10, and 20 mg/kg groups, respectively
(P=0.15).
|
MCAO-induced deficits in sensorimotor function were readily
apparent in all animals. Poststroke function was generally
characterized by depression of locomotor activity and prolonged latency
in performing tasks. When behavioral recovery was compared among
all treatment groups, performance of sensorimotor tasks at 3
and 7 days after MCAO was not improved by LTG treatment. The only
difference among groups in any of the behavioral variables was a
somewhat longer latency to fall when suspended from a wire on
postocclusion day 7 in the LTG 20 mg/kg-treated animals
(P=0.03). At 7 days after
occlusion, latency to drop from the wire was 22.6±9.1, 25.8±8.6,
32.2±5.6, and 49.8±10.8 in saline-treated and LTG 5, 10, and 20 mg/kg
groups, respectively. Because few differences occurred among the
saline- or drug-treatment groups at any dosage level, we evaluated the
data from all animals treated with MCAO to better understand temporal
poststroke recovery. When sensorimotor testing was summarized in
aggregate across all treatment groups, significant MCAO-induced
deficits in sensorimotor tests were apparent at 3 and 7 days
(Table 2
). Latency to fall when suspended from a wire was
significantly longer on days 3 and 7 post-MCAO as compared with
baseline performance
(P=0.001). Latency to fall from
a square bridge was longer on both post-MCAO testing days
(P=0.001), and latency to fall
from the pole was significantly longer on day 7
(P=0.02) compared with
baseline. No significant differences existed across test days in
latency to turn in the alley or on the inclined screen
(P=0.07 and
P=0.08, respectively). A
separate cohort of sham-operated, experimentally unmanipulated female
rats was tested in the same series of sensorimotor tasks to determine
whether repeated testing altered behavior. No significant change was
seen in task performance across trials in any of the
sensorimotor tests. Therefore, MCAO-induced behavioral changes are
unlikely to be artifacts of repeated testing.
|
| Discussion |
|---|
|
|
|---|
We used a full dose-response protocol in the present
study, with 3 intravenous dosages of LTG: 5, 10, and 20
mg/kg. The choice of dosages was based on previous studies that used
parenteral rather than oral drug formulations and showed some degree of
efficacy in rodent models. Early reports emphasized that LTG has a
potentially narrow range of effectiveness. The agent reduced neuronal
loss after cardiac arrest and infarction size after permanent MCAO in
rat over a range of 8 to 20
mg/kg.13 15 In a
previous report, intravenous LTG administration of 10 mg/kg
resulted in plasma levels of 8 to 13 µg/mL in rat, levels that are
similar to plasma levels of patients who receive LTG as an
antiepileptic.13 Brain
concentrations of LTG in rat rose within minutes of
intravenous administration and remained elevated for
5
hours after dosing.13 On the
basis of these data, we used a middle dosage level of 10 mg/kg.
Furthermore, in our preliminary studies that assessed
cardiovascular effects of LTG at 50 to 100 mg/kg, we
observed undesirable blood pressure fluctuations and
cardiovascular instability in rat, as have been
reported previously in pig.6
Therefore, we limited our highest dosage to 20 mg/kg for treatment
after MCAO.
Previous animal investigations using preischemic or intraischemic LTG administration have demonstrated reduction of tissue injury or neurological deficits in many4 13 14 15 but not all6 studies. Cerebroprotective effects of LTG have been credited to its inhibition of presynaptic neuronal sodium channels, thus reducing inward sodium currents and preventing excessive depolarizations. The drug also modulates potassium outward transient current, potentially inhibiting pathological excitation.21 Ischemic glutamate release is not thought to be through vesicular exocytosis, but through a reversal of electrogenic, sodium-coupled amino acid transporters. Increases in internal Na+ and external K+ during ischemia, in conjunction with membrane depolarization, prompt reversal of glutamate transporters, releasing cytosolic glutamate. LTG may stabilize glutamatergic neurons under these pathological circumstances through blockade of inward sodium and possibly calcium currents. Other potentially neuroprotective mechanisms have been identified, such as indirect inhibition of veratrine-stimulated nitric oxide synthesis and consequent increase in cGMP.22
However, the timing of the action of LTG relative to ischemic pathophysiology is important. One explanation for the lack of efficacy in our rat focal stroke model is that the agent was supplied as a single infusion during early reperfusion. Few data are available regarding efficacy of LTG as a resuscitation agent; ie, administered during recovery from a previous ischemic insult. In experimental cardiac arrest, LTG improved selective neuronal injury within hippocampus when the agent (10 mg/kg) was administered either before or after arrest.13 Although return of EEG activity was suppressed in LTG-treated rats, CA1 pyramidal cell counts were increased relative to untreated animals.13 Larger doses of LTG (30 to 100 mg/kg) supplied hippocampal neuroprotection in gerbil with a postischemic treatment paradigm.4 14 LTG improved CA1 neuronal loss after transient forebrain ischemia in gerbil, even when administered up to 1 hour after ischemia.14 Although postischemic LTG was not tested, water-maze performance was improved when LTG was administered in a combination preischemic and postischemia paradigm.14
Our data indicate that the low dosages that are tolerated without hemodynamic sequelae in rat (5 to 20 mg/kg) do not provide protection in either cortex or striatum when given during reperfusion. These results are consistent with those of Smith and Meldrum15 in that delayed LTG administration during permanent MCAO by 2 hours was ineffective in reducing damage. A likely explanation is that the therapeutic window for antiglutamatergic neuroprotection is defined by the period of anoxic depolarization and peak excitatory neurotransmitter release, typically within the first 1 to 2 hours of focal cerebral ischemia in rat.23 If secondary increases in extracellular glutamate concentration occur during the first reperfusion hours after ischemic stroke,16 17 modulation of neuronal sodium channels by LTG at this time does not alter the ultimate progress of tissue damage. A less likely explanation is that LTG has dose-dependent adverse effects that obscure its antiglutamatergic potential during reperfusion. In humans, common side effects are predominately neurological (eg, ataxia and visual disturbances) and dermatologic (for review, see Matsuo24 ). Adverse neurological effects such as loss of locomotor coordination have also been reported in rat but at higher doses than used in the present study.15
As we reported previously,20 the present model of transient MCAO is characterized by altered performance of some, but not all, sensorimotor tasks during the first week of recovery. Non-drug-treated, sham-operated rats did not show deficits or improvements in performance despite repeated task performance over days. Therefore, we think it is unlikely that selective recovery or loss of motoric behavior is related to learning or skill acquisition. Instead, a depression of locomotor activity is present, as reflected in the increased latency to fall from the square bridge and pole after MCAO. Stroked animals spent less time exploring the entire length of the bridge and pole and were therefore less likely to misstep and fall. Latency to fall from the wire also increased dramatically after MCAO and may be due to an abnormal grasp reflex in the forelimbs after damage to the frontal cortex. Abnormal grasp reflexes have been reported in humans with cortical damage.25 Similar alterations in sensorimotor performance on the bridge, pole, wire, and elevated plus maze, in addition to an increased latency to turn on an inclined screen, have been previously reported in male rats20 and mice26 subjected to MCAO. LTG did not improve this latency to perform tasks, and dosing at 20 mg/kg resulted in even more prolonged latency to drop from the wire. This result suggests that LTG did not ameliorate and might at high dosage levels enhance depressed cortical function after experimental stroke.
All animals were randomly assigned to treatment group, and surgery was performed by a single investigator (J.K.). Previous studies with the intraluminal filament MCAO technique in our laboratory resulted in premature death rates in placebo or vehicle-treated rats of 0% to 33% with a 22-hour recovery endpoint.17 18 27 28 Premature deaths in the present study with a 7-day recovery endpoint were 33% in the high-dosage LTG group and ranged from 41% to 48% in the other groups. The relative underrepresentation of premature deaths in high-dosage LTG was not statistically significant. The increase in premature deaths over 7 days versus 1-day survival in our previous studies probably was due not only to stroke maturation and action of late-phase inflammatory injury mechanisms, but also to multiple systemic and environmental factors pertinent to convalescence from large cerebral infarction. No pattern was observed for time of death among the treatment groups or in vehicle-treated animals. Furthermore, cause of death was not rigorously studied by full-body autopsy in all animals. Similar premature death rates have been reported in other MCAO studies that evaluated histological and behavioral damage 4 to 7 days after injury.29 30 31
In conclusion, LTG administered immediately on reperfusion over a range of dosages did not improve tissue or functional outcome from transient focal cerebral ischemia in rat. We conclude that efficacy of the agent may be limited to experimental or clinical settings in which preischemic or intraischemic LTG administration is feasible.
| Acknowledgments |
|---|
Received July 25, 2000; revision received December 1, 2000; accepted December 5, 2000.
| References |
|---|
|
|
|---|
1-receptor ligand, PPBP [4-phenyl-1-(4-phenylbutyl)
piperidine] affords neuroprotection from focal ischemia with
prolonged reperfusion. Stroke. 2000;31:976982.This article has been cited by other articles:
![]() |
P. Ryvlin, A. Montavont, and N. Nighoghossian Optimizing therapy of seizures in stroke patients Neurology, December 26, 2006; 67(12_suppl_4): S3 - S9. [Abstract] [Full Text] |
||||
![]() |
A. M. Naidech, K. T. Kreiter, N. Janjua, N. Ostapkovich, A. Parra, C. Commichau, E. S. Connolly, S. A. Mayer, and B.-F. M. Fitzsimmons Phenytoin Exposure Is Associated With Functional and Cognitive Disability After Subarachnoid Hemorrhage Stroke, March 1, 2005; 36(3): 583 - 587. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Stroke Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2001 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |