Neurological Deficit and Extent of Neuronal Necrosis Attributable to Middle Cerebral Artery Occlusion in Rats
Background and Purpose Occluding a large intracranial artery in rats produces a brain lesion that grows in terms of an increase in both surface area and number of necrotic neurons. The present study investigated whether reperfusing the ischemic territory 30 to 60 minutes after the arterial occlusion would have a beneficial effect on either the clinical or the histological outcome of the lesion.
Methods One hundred four adult rats (including appropriate controls) were used; 97 had a middle cerebral artery occluded by inserting a nylon monofilament via the right external carotid artery. The arterial occlusion was transient in two groups and permanent in another; survival times were comparable for all groups. Control animals were subjected to a sham operation during which the artery was occluded for less than 1 minute. The outcome was evaluated by measuring the extent of the neurological deficit and the severity of the histological injury.
Results Mean neurological score and mean number of necrotic neurons in the cortex were more favorable after transient (30- to 60-minute) compared with permanent arterial occlusion (P<.005). Moreover, the correlation between mean neurological score and mean number of necrotic neurons was highly significant: r=.951; P<.001.
Conclusions The histological effects of an intracranial arterial occlusion in the adult rat can be predicted on day 1 by the neurological score described in this report. Significant improvement can be obtained in these animals by reestablishing arterial flow 60 minutes or sooner after the ictus. The pattern of cortical pannecrosis observed after permanent occlusion (≥72 hours) was transformed into incomplete ischemic injury in most instances of transient occlusion.
Reperfusing the ischemic myocardium by means of either thrombolysis or angioplasty is a well-accepted method of treatment for human coronary artery occlusions.1 Ischemic myocardial changes induced by occluding a coronary artery in dogs can be completely reversed if reperfusion of the ischemic myocardium begins within 15 minutes; in contrast, “severe” ischemia induced by the same method and lasting 30 to 40 minutes leads to biologically irreversible injury or myocardial infarct.2 3 Few comparable studies have tried to ascertain whether similar improvement in a brain lesion could be obtained in experimental situations involving the transient occlusion of a single intracranial artery.4 5 6 7 8 9 None of those experiments attempted to correlate the duration of the arterial occlusion with an evaluation of neurological function or with a detailed, sequential histopathologic quantitation of the brain lesions.
The objectives of this study were to answer these questions: (1) Are the methods we used to test neurological function in animals with transient occlusion of a middle cerebral artery (MCA) sufficiently sensitive to predict differences in the histopathologic degree of brain injury? (2) Can significant differences in the histopathologic features of the brain lesion be demonstrated by changing the time of arterial occlusion but maintaining a constant survival time? For example, is a period of 30-minute arterial occlusion and 7-day survival better (in terms of preserving brain functions and histopathology) than a period of 60-minute arterial occlusion followed by 7 days of reperfusion? (3) Is there, in this model, evidence of a “maturation phenomenon” of the type described in rodents with forebrain ischemia?10
Using a method to occlude the MCA that is minimally invasive and is compatible with reperfusion of the ischemic territory,11 12 we evaluated three groups of animals: (1) those with permanent arterial occlusion (24 to 96 hours); (2) those with transient arterial occlusion (30 to 60 minutes); and (3) appropriate controls, including sham-operated animals. The outcome was evaluated by calculating in each of these groups the number of necrotic and intact neurons and the degree or severity of the neurological deficit.
The results suggest that in this model of single-artery occlusion, beneficial effects can be obtained by reperfusing the brain as late as 60 minutes after the arterial occlusion. Moreover, the beneficial effects can be predicted (on day 1) by means of the neurological tests described in this report. A modest increase in the number of necrotic neurons was recorded in a few specimens 3 to 4 days after transient (30- to 60-minute) arterial occlusion; however, this histological change did not occur consistently, and the daily changes in the number of necrotic neurons were not detectable by the methods we used to test neurological function.
Materials and Methods
All experiments were conducted according to the guidelines issued by the Institutional Animal Care Committee and were in compliance with regulations formulated by the US Department of Agriculture.13 Outbred rats were purchased from Charles River Laboratory (Wilmington, Mass).
One hundred four Wistar rats (weight, 270 to 300 g) were housed individually in stainless plastic cages with wire-mesh bottoms. Neurobehavioral tests and scoring of these tests were carried out on a daily basis starting the day before surgery. Details of the method to occlude the artery are described elsewhere.14 15 Briefly, under general anesthesia (halothane and nitrous oxide) the right MCA was occluded for either 30 minutes (n=40) or 60 minutes (n=36) with a 4-0 nylon monofilament inserted into the right external carotid artery and advanced via the internal carotid artery. Reperfusion was achieved in the subjects with transient MCA occlusion by pulling out the filament until its tip was visible in the cervical segment of the internal carotid. All animals with transient MCA occlusion were allowed to survive up to 7 days. Twenty-one rats were subjected to permanent MCA occlusion of 24 to 96 hours’ duration (without reperfusion). Sham operations (n=7) were also conducted by pulling back the filament within 60 seconds.
Neurological evaluations were carried out once a day starting the day before the surgery and continuing until the end of each experiment. Each subject was examined daily in the late afternoon hours so that rats that had been operated on during the morning had fully recovered from the effects of anesthesia. Adherence to a predetermined time excluded behavioral changes based on circadian rhythm. The examiners (S.W. and X.H.) had no knowledge of the procedure that the rat had undergone. The neurobehavioral study consisted of the following six tests (Table 1⇓).
1. Spontaneous Activity
The animal was observed for 5 minutes in its normal environment (cage). The rat’s activity was assessed by its ability to approach all four walls of the cage. Scores indicate the following: 3, rat moved around, explored the environment, and approached at least three walls of the cage; 2, slightly affected rat moved about in the cage but did not approach all sides and hesitated to move, although it eventually reached at least one upper rim of the cage; 1, severely affected rat did not rise up at all and barely moved in the cage; and 0, rat did not move at all.
2. Symmetry in the Movement of Four Limbs
The rat was held in the air by the tail to observe symmetry in the movement of the four limbs. Scores indicate the following: 3, all four limbs extended symmetrically; 2, limbs on left side extended less or more slowly than those on the right; 1, limbs on left side showed minimal movement; and 0, forelimb on left side did not move at all.
3. Forepaw Outstretching
The rat was brought up to the edge of the table and made to walk on forelimbs while being held by the tail. Symmetry in the outstretching of both forelimbs was observed while the rat reached the table and the hindlimbs were kept in the air. Scores indicate the following: 3, both forelimbs were outstretched, and the rat walked symmetrically on forepaws; 2, left side outstretched less than the right, and forepaw walking was impaired; 1, left forelimb moved minimally; and 0, left forelimb did not move.
The rat was placed on the wall of a wire cage. Normally the rat uses all four limbs to climb up the wall. When the rat was removed from the wire cage by pulling it off by the tail, the strength of attachment was noted. Scores indicate the following: 3, rat climbed easily and gripped tightly to the wire; 2, left side was impaired while climbing or did not grip as hard as the right side; and 1, rat failed to climb or tended to circle instead of climbing.
5. Body Proprioception
The rat was touched with a blunt stick on each side of the body, and the reaction to the stimulus was observed. Scores indicate the following: 3, rat reacted by turning head and was equally startled by the stimulus on both sides; 2, rat reacted slowly to stimulus on left side; and 1, rat did not respond to the stimulus placed on the left side.
6. Response to Vibrissae Touch
A blunt stick was brushed against the vibrissae on each side; the stick was moved toward the whiskers from the rear of the animal to avoid entering the visual fields. Scores indicate the following: 3, rat reacted by turning head or was equally startled by the stimulus on both sides; 2, rat reacted slowly to stimulus on left side; and 1, rat did not respond to stimulus on the left side.
The score given to each rat at the completion of the evaluation is the summation of all six individual test scores. The minimum neurological score is 3 and the maximum is 18.
Ninety-two Wistar rats of the 104 subjected to neurological testing were used for the histopathologic quantitations (Table 2⇓). Forty rats were subjected to 30-minute MCA occlusion followed by 1 to 7 days of survival (group A in Table 2⇓); 36 rats had transient MCA occlusion (60 minutes) followed by 1 to 7 days of reperfusion; 12 animals were subjected to permanent MCA occlusion and allowed to survive 1 to 4 days (group C); and 4 animals were used in the sham-operated group (group D).
All experiments were terminated with the use of analeptics (ketamine and xylazine) by cardiovascular perfusion with a paraformaldehyde fixative, according to methods described in detail elsewhere.14 After brains were removed from the skull and five coronal sections (each 2 to 3 mm thick) were made, the entire specimen was embedded in paraffin, and 4- to 5-μm-thick histology sections were obtained from each slab and stained with hematoxylin-eosin. In each animal a 4- to 5-μm-thick histology section derived from the caudal surface of coronal slice B and representing the level of the anterior commissure was examined at a magnification of ×400. Additional sections from block B were reacted with antiserum for demonstration of glial fibrillary acidic protein (Dako) according to the avidin-biotin complex method (dilution 1:1000); additional details are provided elsewhere.14 15
An Image Measure System (Microscience Inc) was used to quantitate the number of necrotic neurons in the territory of the occluded MCA. The Im 2500 morphometry program of this system was chosen to calculate the number and percentage of either necrotic or intact neurons in the various studies. Using a computer-controlled display video camera interfaced with an Olympus microscope, we collected 20 nonoverlapping microscopic fields each measuring 150×200 μm in area from the neocortex and the caudoputamen in each animal. The definition of necrotic neuron was derived from criteria formulated by Farber et al16 and Eke et al17 and is based on the identification of either pyknotic/eosinophilic neurons or ghost neurons.14 18 The tempo of neuronal necrosis is different in the cortex and lateral caudoputamen; necrotic neurons are very abundant in the caudoputamen, while the cortex contains few necrotic neurons during the initial 1 to 2 hours after MCA occlusion.18 For this reason necrotic neurons were counted in the cortex and intact neurons were counted in the lateral caudoputamen. The evaluator (K.-F.L.) was blinded to the results of the neurological score or to the experimental group to which each rat had been assigned.
The nonparametric neurological scores for each experimental group were added to obtain a mean±SD number. Comparisons were made among individual groups and also against the preoperative scores by means of the Wilcoxon test.
The mean number and the percentage of either intact or necrotic neurons per microscopic field were obtained from cortex and caudoputamen, respectively. ANOVA followed by Bonferroni’s corrected Student’s t test was applied to evaluate the significance of these neuronal counts among the various groups. Pearson correlation matrix was applied when we compared the mean neurological scores for each experimental group against the mean numbers of necrotic neurons (±SD) for each experimental subgroup.
Two groups of rats with ischemia-reperfusion (30- or 60-minute MCA occlusion followed by variable periods of reperfusion), one group with permanent MCA occlusion (24 to 96 hours), and a sham-operated group were compared (Table 3⇓). The mean preoperative neurological score in each of the groups studied ranged from 17.7 to 17.9 (Table 3⇓). On the afternoon of the day of surgery, the neurological score of rats with permanent MCA occlusion was significantly lower than that of those with transient (30-minute) occlusion (P<.001) (Table 3⇓; Fig 1⇓). In the initial 3 days, the neurological scores of rats with either transient occlusion of 60 minutes or permanent MCA occlusion were not significantly different from one another (Table 3⇓). After 4 days of survival, the neurological scores of rats subjected to transient (either 30- or 60-minute) MCA occlusion were not significantly different from one another but were different from scores of those with permanent occlusion (P<.05) (Table 3⇓).
Transient MCA Occlusion (30 Minutes)
The neurological scores of rats that had 30-minute MCA occlusion followed by up to 7 days of survival were slightly lower than those recorded preoperatively (P<.05) (Table 3⇑). The neurological scores of rats with different survival periods were not statistically different, except that on days 4 and 5 the neurological scores were better than the mean neurological score on the day of the operation and not significantly different from those obtained preoperatively.
Transient MCA Occlusion (60 Minutes)
The neurological scores of subjects with 60-minute MCA occlusion followed by up to 7 days of survival were lower than those noted before surgery (P<.05). The mean neurological scores on days 5, 6, and 7 were not statistically different from those obtained preoperatively, suggesting a belated improvement (Table 3⇑). The neurological scores of rats with different survival periods were not significantly different from one another.
Permanent MCA Occlusion
The neurological scores of rats subjected to 24 to 96 hours of MCA occlusion (without reperfusion) were significantly lower than those recorded preoperatively (P<.05) and also were much lower than those obtained in groups A, B (transient occlusion) (P<.05), and D (P<.05) (Table 3⇑).
Among four rats subjected to a sham operation (MCA occlusion lasting <1 minute), the mean number of necrotic neurons in the cortex was 0.7±0.4 per microscopic field (Table 4⇓). The mean number of necrotic neurons in rats with permanent MCA occlusion and up to 7 days of survival was 37.4±9.3 per microscopic field, and almost 80% of neurons in the cortex were necrotic (Table 2⇑). In contrast, in the striatum less than 1 neuron per microscopic field appeared morphologically intact (Table 4⇓). In rats with transient MCA occlusion and 7 days of survival it was difficult to outline an “area of pallor” in the hematoxylin-eosin preparations; these areas could be better defined in the caudoputamen by glial fibrillary acidic protein immunostaining, but the same could not be outlined in the cortex (Fig 2⇓).
Among 40 rats with 30-minute MCA occlusion followed by 1 to 7 days of survival, 9 had few or no necrotic neurons in the cortex; an additional 28 specimens showed neuronal necrosis involving isolated cortical groups (Fig 3A⇓), and only 3 specimens of 40 had small cortical foci where neuronal necrosis was widespread but still incomplete (Fig 3B⇓). In this group A, the mean number of necrotic neurons in the cortex was 6±6.3 per microscopic field (12.7%) (Tables 2⇑ and 4⇑). Only 1 rat in group A, surviving 6 days, had a very large number of necrotic neurons. Comparisons of group A with experimental groups B (60-minute MCA occlusion and up to 7 days of survival) and C (permanent MCA occlusion lasting up to 7 days) revealed significant differences in the mean numbers of necrotic neurons (P<.05) (Table 4⇑).
Among 36 rats with 60-minute MCA occlusion followed by 1 to 7 days of survival, only 2 rats had few or no necrotic neurons in the cerebral cortex; an additional 26 specimens showed neuronal necrosis involving isolated cortical neurons, and 8 specimens had small, focal areas of pannecrosis in the cortex (Fig 3C⇑). The mean number of necrotic neurons in the cortex was 13±12.3 per microscopic field (27.6%) (Tables 2⇑ and 4⇑). In both groups with transient MCA occlusion, an area of pannecrosis or infarct of variable size was found in the lateral striatum. At this location the necrotic neurons were so abundant that it became more manageable to count the number of intact neurons; the mean number of neurons with preserved normal histological features was approximately 8.1±7.5 and 6.0±8.0 per microscopic field after transient MCA occlusion of 30 and 60 minutes, respectively (Table 4⇑). There were no significant differences between these two groups in terms of injury to striatum.
The number of necrotic neurons in the cortex increased moderately in some rats with transient (30-minute) MCA occlusion as a function of days of survival. In animals surviving 5 days or more, the mean numbers were highly variable, although almost each individual subgroup had fewer necrotic neurons than those in which the arterial occlusion lasted 60 minutes or more (Fig 4⇓). The Pearson correlation matrix between mean numbers of necrotic neurons for each group and mean neurological scores for each group was highly significant: r=.951 (P<.0001). Fig 5⇓ illustrates the time-dependent increase in number of necrotic neurons and the changes in neurological scores for each experimental group.
First, the neurological function as reflected in mean neurological score is significantly different in each of the four groups; the differences are well defined as early as the first 2 days after MCA occlusion. Differences between groups A and B tend to become blurred on days 5 or 6, but the differences between them and groups C and D persist. Second, there are also clear-cut differences in the extent of brain injury (quantitated by percentage of necrotic neurons) in each of the four groups studied. Brain specimens with 30-minute MCA occlusion have fewer necrotic neurons than those with 60-minute MCA occlusion, and these two groups are better (in terms of numbers of intact neurons) than the groups that had permanent MCA occlusion (Table 2⇑). Wide fluctuations in the daily neurological scores and in the number of necrotic neurons suggest that there is considerable variability in individual responses to MCA occlusion. Nonetheless, overall mean numbers for each group are significantly different from one another (Tables 2⇑ and 4⇑). Third, the answer to the question of whether there is “maturation,” or continued neuronal necrosis, after reopening an artery that was occluded between 30 and 60 minutes is ambiguous. There is considerable variation in responses obtained in individual animals; some had an increasing number of necrotic neurons as a function of the duration of reperfusion, but others did not (Table 2⇑).
Neurological function in rodents with MCA occlusion has been evaluated by grading sensorimotor deficits. Most previous studies tested simple reflex and motor function by observing forelimb movement, resistance to a lateral push, and tendency to circle.19 20 21 In those studies the neurological deficit correlated with the volume or size of the infarct, which was defined as the area of pallor appearing 24 to 72 hours after the arterial occlusion.19 22 23 24 More complex and detailed neurobehavioral evaluations in rodents with MCA occlusion included the Y maze test, water maze (hidden platform) test, modified open-field test, and other memory or learning tests. The behavioral changes observed in those experiments, with the exception of one, were appropriate to the corresponding neuropathological findings, including infarct size and brain water or Ca2+ content.25 26 27 28
Many neurological evaluations of rats with MCA occlusion were completed during the early hours after the occlusion, whereas pathological examinations were based on measuring the area of pallor 2 to 3 days later.19 22 23 24 In this paradigm of brain infarct, the area of pallor grows at a predictable rate; it covers the territory of the MCA 3 to 4 days after the permanent arterial occlusion,14 but it remains to be proven whether area of pallor and number of necrotic neurons measure the same type of brain injury. In these experiments it would have been impossible to measure area of pallor in the brains of rats with transient occlusion; there was no clearly demarcated area of pallor in any of those specimens (Fig 2⇑).
There are significant differences in the severity of histological changes between group A (30-minute transient MCA occlusion) and the two other experimental groups (B and C). However, analysis based on daily assessments shows considerable numerical fluctuations; for example, 30-minute MCA occlusion plus 6-day survival is worse than other subgroups (Tables 3⇑ and 5), and 60-minute MCA occlusion plus 6-day survival is also significantly worse than other subgroups (Table 3⇑). This might be a reflection of the variability in biological responses attendant to the occlusion of a major brain artery; comparable heterogeneity in the biological responses has been observed among humans with a similar type of arterial occlusion.29
Based on studies of MCA occlusion in primates, Astrup et al30 introduced the concept of a penumbra; in vivo measurements of regional cerebral blood flow (CBF) (hydrogen clearance) showed a center or core of the lesion where blood flow values were extremely low (<10 mL/100 g per minute). According to this concept the marginal or penumbral tissues are sufficiently ischemic to lose their ability to generate either spontaneous or evoked action potentials but also are sufficiently perfused to maintain plasma membrane functions and control of ionic gradients. We assume that MCA occlusion in rats (by the method we used) creates a core in the caudoputamen and a penumbra in the frontoparietal cortex.
Cerebral infarcts (pannecrosis) can be avoided in animals with MCA occlusion when reperfusion is instituted within 2 to 3 hours in the cat31 and between 3 to 6 hours in monkeys.4 5 Cats with MCA occlusion (lasting between 2 minutes and 3 hours) undergo complete reversal of the neurological deficit, and brain sections show only scattered areas of necrosis.32 Neuronal survival and functional recovery after MCA occlusion are dependent not only on the residual blood flow but also on the duration of the ischemic event.33 Further correlation between local values of CBF and extent of histological injury has been recorded in cats subjected to MCA occlusion of 2 hours’ duration followed by 2 hours of recirculation: 8 of 24 cats that had a marked decrease in regional CBF values showed “severe” cortical damage. In contrast, minimal or no cortical damage was seen in 16 cats in which regional CBF values had recovered in a prompt and uniform manner.34
We interpret the results of our experiments as follows: After MCA occlusion caudoputamenal tissues are more ischemic than the cortex probably because they are situated farthest from the collateral arterial connections on the brain surface; this may explain why reperfusion as early as 60 minutes does not stop the necrotic process at this site. Only 8 of 36 subjects with transient MCA occlusion (60 minutes) had small foci of pannecrosis in the cortex; we suggest that probably the end-to-end arterial anastomoses at those sites in those animals were not as efficient as in other animals and the local CBF values promptly fell to very low levels. In the group with transient MCA occlusion (60 minutes) and 7-day survival there was consistent infarct or widespread necrosis in the striatum. One of the following three types of response was recorded in the cortex: occasional small infarcts, occasional selective or incomplete neuronal injury, and no detectable necrosis in a few animals. These patterns of histopathologic damage correspond to an incomplete form of infarct and correlate with topographically related estimates of regional CBF measured by the [14C]iodoantipyrine method: 1 to 6 hours after MCA occlusion CBF values were lower in the caudoputamen area compared with the frontoparietal cortex; moreover, local CBF values in the cortex at 6 hours were 50% lower than similar values had been 1 hour after MCA occlusion.35 With diffusion-weighted magnetic resonance imaging, a hyperintense area, encompassing the entire territory of the MCA, has been demonstrated as early as 30 minutes after the occlusion.7 Withdrawing the arterial filament results in disappearance of the hyperintensity from the cortical area, although the abnormal signal persists in the striatum.7 Finally, metabolic abnormalities (high-energy phosphates) that develop in the rat brain in the territory of the occluded MCA undergo better recovery when reperfusion starts after 1 hour compared with an MCA occlusion of 2 hours’ duration.36
This study was supported in part by US Public Health Service grant NS 31631 (Dr Garcia). Lisa Pietrantoni, HTL (ASCP), and Jun Xu were responsible for most of the technical preparations; Paula McGee and Kathy Zajas provided valuable secretarial support; we thank them all.
- Received July 28, 1994.
- Revision received November 4, 1994.
- Accepted December 29, 1994.
- Copyright © 1995 by American Heart Association
Minematsu K, Li L, Sotak CH, Davis MA, Fisher M. Reversible focal ischemic injury demonstrated by diffusion-weighted magnetic resonance imaging in rats. Stroke. 1992;23:1304-1311.
del Zoppo GJ, Schmid-Schönbein GW, Mori E, Copeland BR, Chang CM. Polymorphonuclear leukocytes occlude capillaries following middle cerebral artery occlusion and reperfusion in baboons. Stroke. 1991;22:1276-1283.
Zea Longa E, Weinstein PR, Carlson S, Cummins R. Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke. 1989;20:84-91.
Animal welfare: final rules (9CFR parts 1, 2, and 3). Federal Register. August 31, 1989;54:36112-36163. US Dept of Agriculture, Animal and Plant Health Inspection Service.
Eke A, Conger K, Anderson M, Garcia JH. Histologic assessment of neurons in rat models of cerebral ischemia. Stroke. 1990;21:299-304.
Garcia JH, Joseph R, Herpolsheimer BC, Liu KF, Ho KL. Neuronal ischemic injury to the brain after focal ischemia: apoptosis or necrosis? Stroke. 1994;25:260. Abstract.
Bederson JB, Pitts LH, Tsuji M, Nishimura MC, Davis RL, Bartkowski H. Rat middle cerebral artery occlusion: evaluation of the model and development of a neurologic examination. Stroke. 1985;17:472-476.
Persson L, Hardemark HG, Bolander HG, Hillered L, Olsson Y. Neurologic and neuropathologic outcome after middle cerebral artery occlusion in rats. Stroke. 1988;20:641-645.
Obana WG, Pitts LH, Nishmura MC. Effects of opiate antagonists on middle cerebral artery occlusion infarct in the rat. J Neurosurg. 1969;69:98-103.
Leshley KS. Integrative functions of cerebral cortex. Physiol Rev. 1933;13:1-42.
Yamamoto M, Tamura A, Kirino T, Shimizu M, Sano K. Behavioral changes after focal cerebral ischemia by left middle cerebral artery occlusion in rats. Brain Res Stroke. 1988;452:323-328.
Tominaga T, Ohnishi ST. Interrelationship of brain edema, motor deficits, and memory impairment in rats exposed to focal ischemia. Stroke. 1989;20:513-518.
Wahl F, Allix M, Plotkine M, Goulu RG. Neurological and behavioral outcomes of focal cerebral ischemia in rats. Stroke. 1991;23:267-272.
Mohr JP, Gautier JC, Hier DB. Middle cerebral artery disease. In: Barnett HJM, Mohr JP, Stein BM, Yatsu FM, eds. Stroke. New York, NY: Churchill Livingstone, Inc; 1992:361-418.
Astrup J, Symon L, Siesjö BK. Thresholds in cerebral ischemia: the ischemic penumbra. Stroke. 1981;12:723-725.
Astrup J, Symon L, Branston NM, Lassen NA. Cortical evoked potential and extracellular K+ and H+ at critical levels of brain ischemia. Stroke. 1977;8:51-57.
Weinstein PR, Anderson GG, Telles DA. Neurological deficit and cerebral infarction after temporary middle cerebral artery occlusion in unanesthetized cats. Stroke. 1986;17:318-324.
Heiss WD. Experimental evidence of ischemic thresholds and functional recovery. Stroke. 1992;23:1668-1672.
Tamura A, Asano T, Sano K. Correlation between rCBF and histological changes following temporary middle cerebral artery occlusion. Stroke. 1980;11:487-493.
Nagasawa H, Kogure K. Correlation of cerebral blood flow and histologic changes in a new rat model of middle cerebral artery occlusion. Stroke. 1989;20:1037-1043.
Selman WR, Crumrine RC, Ricci AJ, LaManna JC, Ratcheson RA, Lust WD. Impairment of metabolic recovery with increasing periods of middle cerebral artery occlusion in rats. Stroke. 1990;21:467-471.