Induction of Cerebral Thrombosis With Phenytoin in Rats
Background and Purpose This study was conducted to explore whether intra-arterial infusion of phenytoin causes cerebral ischemia and to examine the mechanism of cerebral ischemia induced by phenytoin.
Methods Ten rats were infused with phenytoin (150 μL, 3.75 mg) retrogradely from the left external carotid artery, followed by perfusion of carbon black transcardially. The removed brain was photographed from above, and the nonperfused area was compared with control rats (n=10) with the use of an image analyzer. Eight animals with or without phenytoin treatment were perfusion-fixed for transmission electron microscopic analyses of cerebral vasculature. To determine the effect of tissue plasminogen activator (TPA) on phenytoin-infused rat cerebrum, 20 rats were treated with or without TPA (120 000 IU) 5 minutes after the phenytoin infusion (n=10 each).
Results All rats suffered from respiratory distress 25 to 40 minutes after the injection and received carbon black transcardially. The nonperfused area was seen in the territory of the left internal carotid artery. Thrombi were observed from arterioles to capillaries. Under electron microscopy, endothelial cells were partially exfoliated, and the vascular lumen was obstructed by thrombi predominantly consisting of platelets. Eight rats with TPA survived more than 60 minutes, whereas only 2 rats survived without the treatment (P<.005). Nonperfused areas were 7±5% and 50±11% of cerebral surface area in rats with and without TPA treatment, respectively (P<.001).
Conclusions Intra-arterial infusion of phenytoin results in a nonperfused area in rat cerebrum primarily due to thrombosis of arterioles and capillaries.
Although cerebral thrombosis of the cerebral vasculature accounts for most of the human stroke cases, currently available animal models of focal cerebral ischemia are induced by surgical occlusion of major arteries,1 embolization with foreign materials,2 blood clot,3 4 5 6 7 8 air,9 or infusion of platelet aggregation factor such as arachidonate10 11 and ADP.12 These models are useful for the study of ischemic changes of the cerebrum but not for the pathological mechanisms of thrombosis and the development of effective preventive treatment for thrombosis. Endothelial damage followed by platelet aggregation is a crucial factor in the pathogenesis of cerebral thrombosis. Since photochemically induced endothelial alteration stimulates platelet activation,13 the animal model of cerebral cortical infarction has been developed by photochemically initiated thrombosis in rats.14 15
Phenytoin,16 a drug commonly used as an anticonvulsant or for brain protection from cerebral ischemic damage, is known as a drug that induces necrosis of tissue in peripheral regions when injected into an artery. To determine whether phenytoin could induce cerebral ischemia as a result of thrombosis, we infused phenytoin into the carotid artery in rats. We used the carbon black method for evaluating cerebral microvascular disorders after phenytoin infusion. Tissue plasminogen activator (TPA) is a native enzyme that converts fibrin-bound plasminogen to plasmin with subsequent clot lysis.3 5 6 7 8 We determined the effect of TPA on the phenytoin-infused rat cerebrum. A reduction in the production of cerebral ischemia would suggest the involvement of the mechanism of thrombosis in phenytoin-induced cerebral ischemia.
This study was conducted to examine whether intra-arterial infusion of phenytoin causes cerebral ischemia and to examine the mechanism of cerebral ischemia induced by phenytoin.
Materials and Methods
Ten male Sprague-Dawley rats each weighing approximately 400 g were anesthetized with ether and immobilized in the supine position with spontaneous breathing. A longitudinal incision approximately 1.5 cm was made in the cervical skin. After exposure of the left common carotid artery and the left internal carotid artery, the left external carotid artery was exposed to the maximum length, and its distal portion and distal branch were ligated with a thread. The vessel was elevated toward the head by gently pulling the thread upward. A 30-gauge injection needle was inserted into the proximal portion of the vessel, and phenytoin (Aleviatin injection, Dainippon Pharmaceutical) diluted with saline at a ratio of 1:1 (150 μL, 3.75 mg) was retrogradely infused as a bolus. The LD50 of phenytoin in rats is 280 mg/kg when administered intraperitoneally. The needle was removed after infusion, and the vessel was ligated at a place slightly proximal to the injection site. The operative wound was then sutured. Body temperature was kept at 37°C with a heating lamp during preparation. The pH of the diluted phenytoin sodium was 10.8. To examine whether the effect was due to alkalinity of the agent, NaOH adjusted to pH 10.8 was retrogradely injected as a control (n=7). The vehicle solution containing propylene glycol (40%) and ethyl alcohol (10.5%) adjusted to pH 10.8 by NaOH was also used as a control (n=3).
One hour after phenytoin infusion, the animals were anesthetized again with ether and perfused transcardially with 10 mL of carbon ink (Genka, Kaimei).17 When respiratory distress occurred (within 1 hour), rats were immediately infused with carbon ink. The brain was then carefully removed from the cranial vault and fixed in 10% buffered formalin. The removed brains were photographed from above. The cerebral surface area and the nonperfused area were examined on the photograph with the use of a photoanalyzer (Luzex 2, Nikon). Each brain was cut into coronal slices, which were then dehydrated, and embedded in paraffin. Paraffin sections were then stained with p-aminosalicylic acid (PAS) for histopathological analysis.
Transmission Electron Microscopic Analyses
Eight animals (4 with intra-arterial phenytoin, 2 with intra-arterial saline, and 2 normal rats) underwent transmission electron microscopic analyses. Animals were perfusion-fixed with half-strength Karnovsky solution 1 hour after phenytoin infusion or when respiratory distress occurred (within 1 hour). The fixed brain was removed from the cranial vault, sectioned into coronal blocks, and placed in the same solution. Each blocked segment containing left parietal region was fixed in 1% osmium tetroxide in buffer. The tissues were dehydrated and embedded in epoxy resin. Thick sections were stained with toluidine blue and examined by light microscopy. From selected regions of the blocks, thin sections were cut and stained with lead citrate and uranyl acetate and examined by transmission electron microscopy (JEM 100C, JOEL).
Effect of TPA
To evaluate the effect of TPA in rats administered intra-arterial phenytoin, 20 rats were randomly assigned to two groups: the phenytoin group and the phenytoin+TPA group. Phenytoin was infused as described above. Five minutes after phenytoin infusion, 200 μL of TPA produced by human gene recombinant (120 000 IU, containing 4 μg of polysorbate 80 and 6 mg of arginine; Mitsubishi Kasei) was retrogradely infused as a bolus at a place slightly proximal to the site of phenytoin injection. After the intra-arterial infusion of phenytoin or phenytoin followed by TPA, the animals were processed as described above.
Survival times were compared with the χ2 test. The percentage of nonperfused area between the two groups was compared by Student’s unpaired t test when the dispersions of the two groups were equal or by Wilcoxon’s unpaired test when they were not. Values of P<.05 were considered statistically significant.
Most of the phenytoin-infused rats suffered from respiratory arrest within 60 minutes and died in the preliminary experiment. Therefore, the rats immediately received carbon ink transcardially when they suffered from dyspnea to prevent inadequate perfusion. These brains showed gross perfusion deficits (Fig 1⇓). The left hemisphere, into which phenytoin was infused, was less stained, whereas the right hemisphere was stained in most animals. The nonperfused area was seen in the region of the left internal carotid artery. The territory of the anterior cerebral artery in the right hemisphere was usually involved in part. Brain stem and cerebellum were well stained in rats treated with phenytoin. Fig 2⇓ shows the coronal section of phenytoin-infused rat brain. The left hemisphere and the territory of the anterior cerebral artery in the right showed the nonperfused area of carbon black. Carbon was found to be absent from many of the terminal vascular branches in such areas, although vessels with a larger diameter were filled with a carbon suspension. The cerebral surface area was larger in the left hemisphere than the right. Table 1⇓ shows the survival time (the time interval between phenytoin infusion and respiratory distress), nonperfused area, cerebral surface area, and percentage of nonperfused area. In both control groups (n=10) all rats survived more than 60 minutes, and the brain was evenly stained with carbon ink. Proper perfusion was achieved in all rats. The coronal section shows symmetrical distribution of carbon black coloration (Figs 1⇓ and 2⇓).
Light Microscopic Findings
Control rat brain infused with carbon black showed extensive carbon deposition along the vessels under a light microscope. In phenytoin-infused rat cortex, amorphous, PAS-positive material was observed from arterioles to capillaries instead of carbon black (Fig 3⇓). The proximal, larger arterioles were frequently filled with carbon black. Numerous vacuoles were apparent within the neuropil in a nonperfused area, indicating significant edema.
Under transmission electron microscopy (Fig 4⇓), aggregates of platelets containing some erythrocytes and fibrin were found in occluding precapillary arterioles and capillaries. Endothelial cells were partially exfoliated, and plasma component containing fibrin infiltrated the intercellular space. In the control group, endothelial cells of arterioles and capillaries were intact, and no thrombi were found inside the vessels. When the experiment was terminated at 5 minutes after phenytoin infusion, various degrees of endothelial injury were observed. The endothelium demonstrated vacuolization and swelling and was frequently partially exfoliated from the basal lamina. Some platelets adhered to the endothelium, but the arterial lumen was not yet obstructed by platelets.
Effect of TPA
Adequate perfusion was achieved in all rats. Animals that received TPA intra-arterially 5 minutes after phenytoin infusion survived longer and showed a smaller nonperfused area than those without it (Fig 5⇓, Table 2⇓). Eight rats that received the treatment survived more than 60 minutes, whereas only 2 rats survived without the treatment (P<.005). Using an image analyzer and a photograph viewed from above, we compared the amount of nonperfused area as a percentage of cerebral surface area. The values were 7±5% and 50±11% (mean±SD) for rats with and without TPA treatment, respectively (P<.001). One rat in each group died during surgery with inadequate anesthesia.
The present study demonstrates that occlusion of microvasculature in a cerebral hemisphere can be produced by intracarotid infusion of phenytoin solution. All rats with phenytoin infusion (150 μL, 3.75 mg) suffered from respiratory arrest (mean, 33 minutes) within 60 minutes and received carbon ink transcardially. The carbon black method revealed that the nonperfused area was produced by phenytoin in the distribution of the ipsilateral internal carotid artery. The change was apparent in the deep cortex and basal ganglia. The carbon was absent from many of the terminal vascular branches in such an area, although vessels with a larger diameter were filled with carbon black. The cerebral surface area of the nonperfused hemisphere, into which phenytoin was infused, was much larger than that of the carbon-stained one, suggesting significant cerebral edema. Neuropathological examination confirmed that arterioles and capillaries in the nonperfused area were obstructed by amorphous, PAS-positive material instead of carbon black. Evidence of significant edema was seen in the involved area under a light microscope.
Electron microscopic observations of vessels in phenytoin-infused rat brain revealed that precapillary arterioles and capillaries were occluded by aggregates of platelets containing some erythrocytes. This formation of thrombus was characterized by endothelial injury. Endothelial cells were always partially exfoliated from basal lamina, with plasma component infiltrating the intercellular space. The strong parenchymal change around the affected vessels, such as astrocytic foot swelling and infiltration of plasma components, suggested a severe microcirculatory failure within a short period of time. When the experiment was terminated at the very early stage (5 minutes after phenytoin infusion), various degrees of the endothelial injury were observed. The morphological alteration of endothelial cells from the very early stage to the completion of thrombosis suggests that phenytoin directly affects the endothelial cells, which may be the pathogenesis of phenytoin-infused thrombosis, since phenytoin in itself does not induce platelet aggregation in vitro (data not shown).
In an earlier preliminary trial, 7.5, 3.25, 2.5, and 1.25 mg of phenytoin were injected retrogradely into the external carotid artery. The effect of phenytoin seemed to be dose dependent. All rats treated with 7.5 mg phenytoin died soon after the surgery or during the surgery. Eight of the 10 rats treated with 2.5 mg phenytoin died within several hours, whereas only 2 of 13 animals treated with 1.25 mg phenytoin died within hours. The rest of the animals survived more than 1 week without any significant neurological deficits, and half of them demonstrated one or more small cerebral infarctions, predominantly in the basal ganglia. We used 150 μL of 3.25 mg phenytoin solution, a dose that could cause a relatively large (≈50%) nonperfused area in the cerebral hemisphere and results in death within an hour if the agent is infused adequately.
TPA administered 5 minutes after the phenytoin infusion mitigated the development of the occlusion of cerebral microvessels. Animals that received TPA survived longer and showed less perfusion deficit. This demonstrates that the coagulation-fibrinolytic system plays an important role in the completion of thrombosis induced by phenytoin and that the intra-arterial infusion of phenytoin itself did not have a significant cytotoxic effect on the cerebral parenchymal cells. TPA is reported to inhibit platelet aggregability uniformly in a dose-dependent manner.18 This may be a factor in the efficacy of TPA, since the thrombus induced by phenytoin does not incorporate much fibrin into the platelet-rich thrombus, as demonstrated by transmission electron microscopy (Fig 4⇑). A reduction in the production of cerebral ischemia suggests the involvement of the mechanism of thrombosis in phenytoin-induced cerebral ischemia.
The carbon black method we used has been accepted for the study of cerebral microvascular disorders.1 14 19 20 21 We perfused a carbon black particle suspension through the left ventricle without preperfusion of physiological saline to avoid artifactual perfusion.17 As shown in the figures, appropriate staining has been performed routinely in our laboratory with the use of carbon ink (Genka, Kaimei Co, Ltd). In the case of moribund animals, perfusion of carbon black could be inadequate if blood pressure decreases below the normal range. The animals in the current experiment were not controlled with artificial breathing or monitored by blood pressure. In early preliminary experiments, we actually measured the blood pressure continuously and found that blood pressure was maintained or was even higher at the time of respiratory distress after phenytoin infusion. Therefore, we believe that our carbon black infusion technique provided no misleading results.
ADP infusion in the internal carotid artery of rabbits has been reported to produce platelet emboli that occluded the cerebral arteries in a number and size to cause cerebral ischemia.12 However, platelet thrombi were almost entirely transient, being fragmented and removed within a very short time of cessation of ADP infusion. No permanent tissue damage ensued in the rat with ADP. Furlow and Bass10 11 reported that an injection of sodium arachidonate into the internal carotid artery produced unilateral cerebrovascular occlusion in heparinized rats within seconds. A majority of animals demonstrated a syndrome of irreversible focal neurological deficits and died within hours after the ictus. Electron microscopic examination of an affected cortex localized the site of occlusion to the microcirculation. Although aggregates of platelets were found occluding the arterioles and capillaries, the vascular endothelium, tunica muscularis, and tunica adventitia were normal, and the perivascular neuropil was intact. The feasibility of studying endothelial injury with subsequent platelet activation has been demonstrated in a variety of animal models of photochemically induced cerebral infarction.13 14 15 There is a general agreement that platelet aggregation progressed in parallel with the development of local endothelial alteration and that the initiation of aggregation could occur without the presence of endothelial denudation.13 15 The aggregating platelets, probably by releasing mediators that are toxic to endothelium, in turn cause the more severe endothelial damage with denudation.22 The final appearance on electron microscopy of widespread denudation may not be a direct consequence of phenytoin but a secondary consequence of platelet aggregation over minimally injured sites. The significant effect of TPA on phenytoin-induced thrombosis supports this hypothesis. Studies on the effect of phenytoin on endothelial cells may help elucidate the mechanism of the interaction between endothelial damage and platelets.
In conclusion, intra-arterial infusion of phenytoin results in a nonperfused area in rat cerebrum primarily due to thrombosis of arterioles and capillaries. Further studies on cerebral thrombosis produced by phenytoin should contribute to a better understanding of the pathological mechanisms of thrombosis and lead to more effective preventive treatment.
The authors thank Yasushi Ohmachi, Shigeo Kurabe, and Akira Yasoshima, Research Laboratory of Drug Metabolism, Tanabe Seiyaku, for their technical assistance.
- Received June 29, 1994.
- Revision received July 5, 1995.
- Accepted July 12, 1995.
- Copyright © 1995 by American Heart Association
Kudo M, Aoyama A, Ichimori S, Fukunaga N. An animal model of cerebral infarction: homologous blood clot emboli in rats. Stroke. 1982;13:505-508.
Zivin JA, Fisher M, DeGirolami U, Hemenway CC, Stashak JA. Tissue plasminogen activator reduces neurological damage after cerebral embolism. Science. 1985;230:1289-1292.
Overgaard K, Sereghy T, Boysen G, Pedersen H, Diemer NH. Reduction of infarct volume and mortality by thrombolysis in a rat embolic stroke model. Stroke. 1992;23:1167-1174.
Garcia JH, Klatzo I, Archer T, Lossinsky AS. Arterial air embolism: structural effects on the gerbil brain. Stroke.. 1981;12:414-421.
Furlow TW, Bass NH. Stroke in rats produced by carotid injection of sodium arachidonate. Science. 1975;187:658-660.
Furlow TW, Bass NH. Arachidonate-induced cerebrovascular occlusion in the rat: the role of platelets and aspirin in stroke. Neurology. 1976;26:297-304.
Fiesta C, Battistini N, Volante F, Zanette E, Weber G, Passero S. Animal model of TIA: an experimental study with intracarotid ADP infusion in rabbits. Stroke. 1975;6:617-621.
Rall TW, Schleifer LS. Drugs effective in the therapy of the epilepsies. In: Goodman LS, Gilman AG, Rall TW, Nies AS, Taylor P, eds. Goodman and Gilman’s the Pharmacological Basis of Therapeutics. 8th ed. New York, NY: Pergamon Press; 1990:439-443.
Karkouly C, Backhauβ J, Nuglisch J, Krieglstein J. The measurement of the infarcted area after middle cerebral artery occlusion in the mouse: a screening model. In: Kriegstein J, Oberpichler H, eds. Pharmacology of Cerebral Ischemia 1990. Stuttgart, Germany: Wissenschaftliche Verlagsgesellschaft mbH; 1990:63-69.
Terres W, Umnus S, Mathey DG, Bleifeld W. Effects of streptokinase, urokinase, and recombinant tissue plasminogen activator on platelet aggregability and stability of platelet aggregates. Cardiovasc Res. 1990;24:471-477.
Ito U, Ohno K, Yamaguchi T, Tomita H, Inaba Y, Kashima M. Transient appearance of ‘no-reflow’ phenomenon in Mongolian gerbils. Stroke. 1980;11:517-521.
Said S, Rosenblum WI, Povlishock JT, Nelson GH. Correlations between morphological changes in platelet aggregates and underlying endothelial damage in cerebral microcirculation of mice. Stroke. 1993;24:1968-1976.