AUTHOR(S): Menzies, Stephen A., M.D., Ph.D.; Hoff, Julian T., M.D.; Betz, A. Lorris, M.D., Ph.D. Departments of Surgery (Neurosurgery) (SAM, JTH, ALB), Pediatrics (ALB), and Neurology (ALB), University of Michigan, Ann Arbor, Michigan Neurosurgery 31; 100-107, 1992 ABSTRACT: MIDDLE CEREBRAL ARTERY occlusion (MCAO) in rats produces an infarct of varying size. We examined three factors that may influence this variability: animal weight, vascular anatomy, and extent of occlusion in rats undergoing MCAO. We also developed a four-point neurological evaluation scale and validated its useful-ness by comparing it with a four-grade pathological determination of the size of the infarct. Of 82 animals sub-jected to a standard MCAO, 34 developed small cortical infarcts (pathological grades I-II; infarct size 25 mm2, 20-56% of surface area). We were able to predict the size of infarction from the neurological evaluation in 83% of the animals, and this accuracy reached 91% when grades I and II and III and IV were considered together (P < 0.001). In 41 animals subjected to a more extensive vascular occlusion, 89% exhibited large infarcts. Four vascular patterns were identified but none played a significant role in the incidence or size of the cortical stroke. However, rats weighing 300 g. Our proposed new MCAO technique appears useful in reproducing largesized infarcts of the frontoparietal cortex. KEY WORDS: Focal cerebral ischemia; Middle cerebral artery occlusion; Vascular anatomy Reproducibility of infarct size is essential if the rat model of middle cerebral artery occlusion (MCAO) is to be used for evaluation of treatment efficacy. At the moment, however, there is considerable variability in infarct size between different groups and even variability between experiments undertaken by a single investigator where the technique has been rigidly defined (9). The subtemporal approach and its modifications (20) , whereby the MCA is occluded at the origin of the lateral striate arteries, is associated with consistent infarction of the caudoputamen and the olfactory cortex but with variable involvement of the frontoparietal somatosensory cortex. Bederson et al. (1) proposed occlusion of the MCA over 3 to 6 mm of its length between the rhinal cortex and the inferior cerebral vein and observed a higher success rate of cerebral cortical infarction; however, uniformity of
lesion size was not evaluated. Additional variables that have been reported to affect infarct size include variations in vascular anatomy (19) and the strain, blood glucose level, and age of the rats (7,18). In this study, we further defined the requirements for a predictable lesion size in the Sprague-Dawley rat. We also refined the Bederson et al. (1) clinical scale of evaluating postischemic motor-behavioral abnormalities and validated its accuracy with a comparable pathological grading. Finally, we evaluated the importance of variability in the anatomical pattern of the MCA and of the age of the animal as factors affecting the surface size of the brain infarct. Preliminary results were reported previously (12). MATERIALS AND METHODS Surgical procedure Two hundred eighteen adult male Sprague-Dawley rats weighing 180 to 540 g were anesthetized with ketamine (50 mg/kg) and xylazine (10 mg/kg). They were divided into three groups. In the first group of 133 rats (A), a 6- to 7-mm length of the right MCA was exposed via a subtemporal craniectomy and cauterized from the origin of the lateral striate arteries to the inferior cerebral vein (Fig. 1A). In the second group (B), 41 rats were subjected to an extended MCAO (Fig. 1B). The main trunk of the MCA was cauterized over a greater length (7-9 mm) than in Group A, from and including some of the distal lateral striate arteries to the parietal branch 2 to 3 mm beyond the inferior cerebral vein. In addition, the main branches were cauterized for a distance of approximately 5 mm from their origin. the third group (C) included 44 sham-operated animals in which the surgical exposure of the MCA was identical but without cauterization. Clinical evaluation of ischemia The results presented here are part of a larger study in which postischemic edema and blood-brain barrier permeability were examined at different time intervals ranging from 1, 3, 6, 12, and 24 hours, 2, 3, 5, and 7 days, and 2, 3, 4, 5, and 6 weeks after MCAO (13) . The neurological status of rats was assessed at 4 and 24 hours and 1 week after the MCAO and before death (up to 6 wk), although some rats were killed before the 24-hour and 1-week examinations. At each time, the highest score out of three consecutive trials was used. A scale of 0 to 4 was used to assess the motor and behavioral changes after the MCAO (Table 1 and Fig. 2). This test consisted of the following maneuvers: rats were suspended from the tail approximately 30 cm above the floor and their forelimb posture was noted. Normal animals extended both forelimbs toward the floor and were assigned a score of 0 (Fig. 2A). When the forelimb contralateral to the side of the MCAO was consistently flexed during the suspension and there was no other abnormality, the rat was scored 1 (Fig. 2B). Rats were then placed on absorbent pads and they were gently held by the tail. If animals showed an apparent decrease in grip in the contralateral forelimb when pulled by the tail, then
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Neurosurgery 1992-98 July 1992, Volume 31, Number 1 100 Middle Cerebral Artery Occlusion in Rats: A Neurological and Pathological Evaluation of a Reproducible Model Experimental Study
Pathological evaluation of ischemia For all measurements, animals were killed by decapitation and the brains were quickly removed and prepared for evaluation of the infarct surface area. Although an ischemic lesion has three dimensions, in this study, we used two-dimensional data to estimate surface area of the lesion. In a pilot group of 16 animals, from Group A the brain was removed, placed under a bright light, and using a magnifying glass, the margins between the pale discoloration of the ischemic and the surrounding healthy tissues were delineated. The lesions were primarily elliptical, with their major axes following the course of the main trunk of the MCA (Fig. 3). Using a transparent grid with 1 mm2 pixel size made from chart recording paper, the surface area of each lesion was measured by adding the number of square pixels overlying the infarcted area. Thereafter, the brain was immersed in a 2% solution of 2,3,5-triphenyltetrazolium chloride at room temperature for 60 minutes. The surface area was then similarly recalculated and compared with that obtained earlier by measuring the pale discoloration. Comparison of the two calculations proved identical when lesions of 3 days or older were examined. Ischemic discoloration was inconsistent within the initial 6 to 12 hours after MCAO. By 24 hours, however, ischemic discoloration was clearly evident in more than 95% of the brains examined. In Group A, 82 of 124 animals had their surface infarct area estimated by measuring pale discoloration 3 days after MCAO, and these data were analyzed for correlations between neurological grade and infarct size. From this, a pathological grading of the size of infarction was developed (Fig. 3). Grade I, the smallest sized lesions, were circular and ranged from 8 to 10 mm2 (mean, 9 ± 1 mm2). Grade IV, the largest lesions, ranged from 63 to 84 mm2 (mean, 78 ± 8 mm2) were elliptical in shape, and extended but never exceeded the plane of MCA anastomoses with the anterior and posterior cerebral arteries (Fig. 3). Grade II (range, 11-28 mm2; mean, 20 ± 5 mm2) and Grade III (range, 32-60 mm2; mean, 44 ± 9 mm2) were created arbitrarily by comparison of lesion sizes. Statistical analysis The χ2 test (contingency table analysis) was used for intragroup comparison of the significance of the anatomical vascular patterns of the MCA as factors affecting size of infarction. A Cohen's weighted κ analysis was used to analyze intergroup predictability
of infarct size from the clinical neurological score. The Cohen's κ analysis examines to what extent the two methods (neurological versus pathological) agree, correcting for agreement that would be due to chance alone. The Statview 512+ statistical software package for the Macintosh computer was used throughout. RESULTS Findings related to MCAO Fifteen animals (9 from Group A, 4 from Group B, 2 from Group C) of the total of 218 rats (7%) used in the study died between Day 1 and Day 10 after the MCAO; 4 died with hyperthermia and meningismus 48 hours after surgery; 2 died from intractable diarrhea at Day 3, and the remaining 9 died from unknown causes at Days 7 to 10. These animals were excluded from the study. The body weight of animals operated on declined by an average of 7.5% during the first 24 hours and 12.6% at 48 hours. It continued to decline until the 6th day, when it reached an average loss of 17.4%. Thereafter, the animals gained weight, and they regained their original weight by Day 11. In contrast, the sham-operated animals showed a minor decline in weight during the initial 48 hours but a steady increase thereafter. There was considerable variability in the number and location of the branches of the MCA from its origin to its bifurcation. Several branches constituted the lateral striate arterial complex, which supplies blood to the basal ganglia. For an unknown reason, there usually appeared to be many more lateral striate branches from the right than from the left proximal MCA (Fig. 4). Four major patterns of the MCA were observed (Fig. 5). Type A, with a pyriform and a frontal branch anteriorly and a temporal branch posteriorly, was observed in more than half of the rats (51%). Type B (26.6% of animals) with both pyriform and frontal branches but lacking the temporal one, was the second most common pattern. Type C (14%) with a bifurcating frontal branch and a posterior temporal branch, was the third most common pattern. Type D (7%) showed only a bifurcated frontal branch. Only 1.4% of the MCA vascular anatomy of the Sprague-Dawley rat fell outside of these four standard pat-terns. Clinical and pathological evaluation None of the sham-operated animals showed any motor-behavioral abnormalities as described in Table 1. However, all group A animals with MCAO developed one or more of these features upon recovery from surgical anesthesia approximately 2 hours after MCAO. By the end of 24 hours, 4 rats were scored 1, 30 scored 2, 36 scored 3, and the remaining 12 scored 4. By the first weekly reevaluation, 6 of 82 animals had clinically improved back to normality and 13 of 82 had deteriorated. Five of the 6 animals with the normalized neurological features, however, showed neurological scores at l week similar to those shown at their 24-hour evaluations if examined during induction or recovery from anesthesia. This effect became less apparent in
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
they were assigned a score of 2 (Fig. 2C). Thereafter and while still being held by the tail, the rats were allowed to move freely and were observed for circling behavior. Rats that moved spontaneously in all directions but established a monodirectional circling toward the paretic side when given a slight jerk of the tail were scored 3 (Fig. 2D). Rats that showed a consistent spontaneous contralateral circling were scored 4. Animals that showed a higher clinical score also showed all features of the lower grades. This neurological evaluation was completed within a few minutes.
the inferior cerebral vein resulted in a 100% infarction of the frontoparietal cortex. Using a similar occlusion technique, we observed considerable variability with just over half of the infarcts characterized as large. It was evident in our studies that the severity of the observed pathophysiological changes was proportionate to the size and, therefore, the severity of the cortical lesion. This prompted us to further expand and refine the Bederson modification (1) by extending the thermocoagulatory occlusion to a greater length of the main MCA trunk and to all major branches of the MCA from near its origin to beyond its distal junction with the inferior cerebral vein. As a result, the rate of large-sized infarcts was increased to 89% and more uniform and reproducible lesions were obtained. This more extensive occlusion increased the incidence of maximal lesions (Grade IV) from 15% to 60%, probably related to the effective compromise of the distal cortical interarterial anastomoses.
Effects of vascular anatomy of the MCA and animal age We examined the role of vascular variability as a factor determining the size of infarction in 82 animals of Group A. From the four anatomical vascular patterns identified (Fig. 5), none played a significant role in the variability in severity of ischemia. Similarly, neither the total number of MCA branches nor the number of branches to the frontal and/or parietal cortices played any significant role (contingency table analysis). The age of the rats as reflected by their weight before MCAO, however, appeared to be an important determinant of the extent of infarction (Fig. 8). In particular, rats weighing 300 g. The latter also showed a significantly larger lesion size than the former (P < 0.0004, contingency table analysis).
Clinical assessment The importance of accurately assessing the clinical status of animals with stroke in studies of the efficacy of treatment is obvious. The clinical method of neurological evaluation after MCAO used by Bederson et al. (1) employs three neurological grades: Grades 1 and 2 for motor deficits and Grade 3 for behavioral deficits. The overall sensitivity in predicting histological changes compatible with an infarction is 88%. Grade 2 indicates a consistently reduced resistance to lateral push behind the shoulders toward the paretic side as compared with the normal or mildly dysfunctional animals, which resisted sliding equally in both directions. In our experience, the application of this test is sensitive only to grossly paretic rats, it is subjective among different investigators, and it is rather difficult to reproduce accurately. Our four-score clinical test provides a reproducible Grade 2 as well as an intermediate behavioral grade (score 3), whereby the circling behavior is provoked as compared with the spontaneous circling in score 4, and this is correlated with a different size of infarction. The prediction of small, intermediate, or large size of infarction from this neurological evaluation is 83%; when the motor (1-2) and behavioral (3-4) scores are considered together, the sensitivity reaches 91%.
DISCUSSION Operative technique Rats are a convenient small animal model for the study of pathophysiology and treatment of cerebral ischemia. Tamura et al. (20) developed a subtemporal approach of proximal MCAO at a point near the origin of the lateral striate arteries, which produced infarction of both the cortex and the caudoputamen. The original technique, however, was very invasive and the animals survived only for a few hours (20). Subsequent modifications preserving the zygoma and the masseter muscle improved the postoperative survival for several days and eventually the subtemporal approach emerged as the standard technique of focal cortical ischemia in rats (7,15,16,18). It was apparent, however, in all of these works that the cortical somatosensory infarction was variable in size and location. Bederson et al. (1), using the subtemporal approach, could not produce a 100% cortical infarction by proximal focal occlusion of the MCA; their success rate varied between 13 and 67%. They observed that occlusion of the MCA from its origin medial to the olfactory bulb to its junction with
Pathological assessment Morphometric studies of the infarction and its surface quantification proved easy and comparable with other studies that used computer-assisted algorithms (5,11). In one previous study, the total lesion size in hypertensive stroke-prone rats with a single point ligation of the MCA distal to the rhinal fissure varied from 54 to 89 mm2 (11). A significant role of surface curvature in determining the absolute surface area was emphasized in that study (11). In another study using the same measurement algorithm, it was shown that the average surface area of infarction after MCAO and ipsilateral carotid artery occlusion in Long-Evans rats was 100 mm2. Our method is simpler but useful only in the measurement
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
animals that survived for 4 to 6 weeks after MCAO. In 37 surviving animals with the extended MCAO (Group B), only 2 animals improved and none deteriorated in their neurological scores between the 24 hours and 1 week. Of 82 animals in Group A, 34 developed small infarcts (4 in pathological Grade I and 30 in Grade II) and 48 large infarcts (36 in Grade III and 12 in Grade IV) (Fig. 6). We were able to predict the size of infarction from the neurological evaluation in 83% of the animals (Fig. 7). This accuracy reached 91% when pathological Grades I-II and III-IV and neurological scores 1-2 and 3-4 were considered together (P < 0.001, Cohen's weighted κ). In the 37 surviving animals of Group B, a substantial difference in the lesion size was observed. Indeed the extended occlusion of the MCA trunk and all major branches increased the proportion of the animals having a large (Grade III-IV) infarct from 59 to 89% (Fig. 6).
Factors affecting the size of infarction Spontaneously hypertensive and stroke-prone spontaneously hypertensive rats show approximately 1.5 times the infarct size of the normotensive Fischer 344 strain (7). Among normotensive rats, the Fischer 344 strain scored slightly better than the SpragueDawley strain, showing the largest and most reproducible lesions after MCAO. In Wistar-Kyoto rats, the stroke size was smallest and most variable. Spontaneously hypertensive or stroke-prone rats may have substantial anatomical, biochemical, and metabolic differences compared with the normotensive rats, and consequently, their use as a model may be limited. Among the normotensive strains the Fischer 344 appeared slightly better for predictability and size of stroke compared with Sprague-Dawley rats; however, the technical modifications described in our method using the Sprague-Dawley strain also produce large and reproducible infarcts. Six anatomical vascular patterns of the MCA in rats have been observed in morbidity studies. Their role in the formation of infarction after selective occlusion of these branches has been recently described (19). Distal occlusion of the parietal and frontal branches combined with temporary occlusion of the common carotid artery in the neck did not result in infarction of the areas supplied by these branches. According to the same investigators, this effect was attributed to the existence of an extensive collateral network between these branches and other arterial systems. Obliteration of the pyriform branch resulted in a consistent infarction, which was thought to indicate that the pyriform branch is an end artery without collateral blood supply. However, this conclusion is not supported by evidence from other investigators, who demonstrated collateral networks between the olfactory, pyriform, cingulate, and frontal cortex (3). Using high magnification observation in living animals, we demonstrated four major vascular patterns of the MCA, the most common of which had all major branches. The frontal and, indirectly, the pyriform branches are in collateral communication with the anterior cerebral artery, whereas the parietal and temporal branches collateralize with branches of the posterior cerebral artery. This explains why distal occlusions of the MCA and its branches do not result in infarction. On the other hand, a single point, proximal occlusion below the rhinal fissure produces inconsistent lesions, whereas occlusion of the MCA over 3 to 6 mm of length results in a 100% infarction rate (1). We suggest that the latter model provides a more efficient compromise of the collateral blood supply from retrograde flow via the major MCA branches. In our model, in which the main trunk of the MCA is occluded over a greater length and, in addition, the major MCA branches are also coagulated, the size of the infarction reaches maximal size as determined by the extent of the anastomoses between the MCA-anterior cerebral artery and MCA-
posterior cerebral artery network (3). Perhaps coagulation of the MCA trunk and its branches results in an intraluminal clot, which propagates distally compromising the collateral system to its maximum. The result of such a well described pathophysiological process is the extensive cortical infarctions we observed in our Group B. Although technically more demanding, our technique offers a reproducible and maximal cortical ischemic lesion in 90% of the animals. Cortical anastomoses are the key to the extent of infarction after a MCAO. This was clearly shown from the extent of maximal lesions in the present study, where the margins of infarction never exceeded the cortical interarterial anastomoses sites. It is not surprising, therefore, that none of the vascular anatomical patterns described in this study played a significant role in modulating the extent and severity of cortical ischemia. Similarly neither the total number of MCA branches nor the number of branches to the frontal and/or parietal cortices played any significant role. Another important factor in infarction after MCAO appears to be the body weight and, thus, indirectly, the age of the animals. How does this factor relate to the size of infarction? Probably through the availability and efficiency of the cortical interarterial anastomoses among the MCA, anterior cerebral artery, and posterior cerebral artery. In a study by Coyle (4), no young rats weighing below 150 g developed infarcts after proximal MCAO. In contrast, adult rats, cats, and primates nearly always develop infarcts after acute MCAO, although the size may vary considerably (2,10,14,17). This probably reflects the fact that distal collateral anastomotic vessels in adult primates, including humans, are not distributed uniformly (8) and are not as numerous (21) as in young animals (6). During the fetal and neonatal period of brain development in primates and, to a lesser degree, in species with non-convoluted brains such as the rat, these vascular anastomoses undergo substantial structural changes with age leading to regression (22). The extent of this regression may materially influence the size of brain infarction after MCAO. Thus, younger rats with a better collateral supply of the vascular bed of the MCA would be protected from infarction and neurological deficits. ACKNOWLEDGMENTS The authors thank Dr. Peter Coyle for many helpful discussions. This work was supported by Grants NS-23870 and NS-17760 from the National Institutes of Health. Received, July 22, 1991. Accepted, August 16, 1991. Reprint requests: A. Lorris Betz, M.D., Ph.D., University of Michigan, Department of Pediatrics, D3227 Medical Professional Building, Ann Arbor, MI 48109-0718. REFERENCES: (1-22) 1.
Bederson JB, Pitts LH, Tsuji M, Nishimura
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
of surface infarct. Nevertheless, the infarct size correlated well with the neurological symptoms.
3. 4. 5.
6. 7.
8. 9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
Nakayama H, Ginsberg MD, Dietrich WD: (S)Emopamil, a novel calcium channel blocker and serotonin S2 antagonist, markedly reduces infarct size following middle cerebral artery occlusion in the rat. Neurology 38:1667-1673, 1988. O'Brien MD, Waltz AG: Transorbital approach for occluding the middle cerebral artery without craniectomy. Stroke 4:201-206, 1973. Prado R, Ginsberg MD, Dietrich WD, Watson BD, Busto R: Hyperglycemia increases infarct size in collaterally perfused but not endarterial vascular territories. J Cereb Blood Flow Metab 8:186- 192, 1988. Rubino GJ, Young W: Ischemic cortical lesions after permanent occlusion of individual middle cerebral artery branches in rats. Stroke 19:870-877, 1988. Tamura A, Graham DI, McCulloch J, Teasdale GM: Focal cerebral ischemia in the rat: I. Description of technique and early neuropathological consequences following middle cerebral artery occlusion. J Cereb Blood Flow Metab 1:53-60, 1981. vander Eecken HM, Adams RD: The anatomy and functional significance of the meningeal arterial anastomoses of the human brain. J Neuropathol Exp Neurol 12:132-157, 1953. vander Eecken HM: Anastomoses Between the Leptomeningeal Arteries of the Brain: Their Morphological, Pathological and Clini-cal Significance. Springfield, IL, Charles C Thomas, 1959, pp 1-160.
COMMENT The authors characterize and further refine the rat model for producing cerebral infarction. They used two surgical methods. In one, the middle cerebral artery was cauterized from the lateral branch to the inferior cerebral vein. In the other, which yielded more large infarcts, the artery was cauterized more extensively, even including the main branchings for 5 mm distance. The size of the animal was important in determining infarct size but the cortical anastomotic pattern was not. Infarct size also correlated well with clinical examinations, as illustrated in the paper. This work represents prodigious investigative effort that goes a long way toward standardizing a reliable and inexpensive model for future stroke studies. Most of our other acceptable animal models suffer from the major flaw of non-reproducibility. Although the authors achieve good success rates with cautery, we wonder if even higher infarct rates could not have been obtained by division of the main trunk, because cauterization is sometimes not as complete as we think from external inspection. This might also remove the variable that may be produced by distal embolization. The relative unimportance of the cortical anastomotic pathway in rats is surprising, because in human middle cerebral artery occlusion,
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
2.
MC, Davis RL, Bartkowski H: Rat middle cerebral artery occlusion: Evaluation of the model and development of a neurologic examination. Stroke 17:472-476, 1986. Cohen MM: Animal models of cerebral infarction, in Klawans HL Jr (ed): Models of Human Neurological Diseases. Amsterdam, Excerpta Medica, 1974, pp 205-248. Coyle P: Arterial patterns of the rat rhinencephalon and related structures. Exp Neurol 49:671-690, 1975. Coyle P: Middle cerebral artery occlusion in the young rat. Stroke 13:855-859, 1982. Coyle P: Altered cerebral collaterals and protection from infarction, in Hartmann A, Kuschinsky W (eds): Cerebral Ischemia and Calcium. Berlin, Springer-Verlag, 1989, pp 6978. Coyle P, Jokelainen PT: Dorsal cerebral arterial collaterals of the rat. Anat Rec 203:397-404, 1982. Duverger D, MacKenzie ET: The quantification of cerebral infarction following focal ischemia in the rat: Influence of strain, arterial pressure, blood glucose concentration and age. J Cereb Blood Flow Metab 8:449461, 1988. Gillian LA: Potential collateral circulation to the human cerebral cortex. Neurology 24:941948, 1974. Ginsberg MD, Busto R: Rodent models of cerebral ischemia. Stroke 20:1627-1642, 1989. Hudgins WR, Garcia JH: Transorbital approach to the middle cerebral artery of the squirrel monkey: A technique for experimental cerebral infarction applicable to ultrastructural studies. Stroke 1:107-111, 1970. Jones PG, Coyle P: Microcomputer assisted lesion size measurements in spontaneously hypertensive stroke-prone rats. J Electrophysiol Tech 11:71-78, 1984. Menzies SA, Hoff JT, Betz AL: Clinical and pathological evaluation of focal brain ischemia after MCA occlusion in rats. Stroke 21:162, 1990 (abstr). Menzies SA, Hoff JT, Betz AL: Extravasation of albumin in ischemic brain edema, in Reulen HJ, Baethmann A, Fenstermacher J, Marmarou A, Spatz M (eds): Brain Edema. Vienna, Springer-Verlag, 1990, vol 8, pp 220222. Molinari GF, Moseley JI, Laurent JP: Segmental middle cerebral artery occlusion in primates: An experimental method requiring minimal surgery and anesthesia. Stroke 5:334339, 1974. Nakayama H, Dietrich WD, Watson BD, Busto R, Ginsberg MD: Photothrombotic occlusion of rat middle cerebral artery: Histopathological and hemodynamic sequelae of acute recanalization. J Cereb Blood Flow Metab 8:357-366, 1988.
this variable must account for the differences in the deficits that have been observed. This article constitutes a monumental effort.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Robert R. Smith Jackson, Mississippi
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Figure 1. Methods of direct middle cerebral artery occlusion in rats. A, occlusion of the MCA over 3 to 6 mm of its length between the olfactory bulb and the inferior cerebral vein (1). B, occlusion of the MCA over 7 to 9 mm of its length from and including some of the distal lateral striate arteries to the parietal branch 2 to 3 mm beyond the inferior cerebral vein with the additional occlusion of all its major branches (present study). M.C.A., middle cerebral artery; I.C.V., inferior cerebral vein; L.S.a, lateral striate artery; O.b., olfactory bulb; T.a., temporal artery; P.a., parietal artery; F.a., frontal artery; Pr.a., pyriform artery.
Figure 3. Schematic diagram of the distribution of the MCA and its anastomoses with the anterior and posterior cerebral arteries. The extent of the four grades of ischemia is also shown. The outer margins of Grade IV were always limited by the distal level of MCA anastomoses.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Figure 2. Neurological evaluation after MCAO. A, normal animal extends both forelimbs symmetrically toward the floor (score 0). B, after MCAO, the contralateral forelimb is consistently flexed during suspension (score 1). C, decreased grip of the contralateral forelimb when pulled by the tail (score 2). D, rat established a monodirectional circling toward the paretic side, when a slight jerk of the tail was done (score 3) or showed a consistent spontaneous contralateral circling (score 4).
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Figure 4. Anatomy of major vessels in a typical Sprague-Dawley rat as visualized by latex injection. Top, macroscopic overview of the circle of Willis in rat. Bottom, coronal section at the level of MCA and its lateral striate branches supplying mainly the caudoputamen to which infarction is extended when MCAO is done close to and including the lateral striate arteries. These branches are multiple and are more numerous on the left side as compared with the right.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Figure 5. Four major branching patterns of the MCA as seen through a craniectomy window and their relationship to surrounding anatomical landmarks including the inferior cerebral vein and the olfactory bulb. Type A was the most common, being observed in little more than half of the rats (51%), as compared with Type B (26.6%), Type C (14%), and Type D (7%). Only 1.4% of the MCA vascular anatomy fell outside these standard patterns.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Figure 6. Percentage of pathological grades (top left) with the occlusion of the main MCA trunk from the lateral striate arteries up to its crossing with the inferior cerebral vein (Group A) and (top right) with the extended MCA trunk and additional occlusion of all major branches of the main trunk (Group B). With the latter technique, the proportion of animals having a large (Grade III-IV) infarct was increased from 59% to 89%, and the incidence of maximal lesions (Grade IV) from 15% to 60%.
Figure 8. Correlation between body weight and size of infarction in 82 rats. Larger rats developed the largest infarcts as compared with smaller rats, the majority of which developed small lesions after the MCAO.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Figure 7. Comparison between neurological scores and pathological grades. Grades I and II represent smaller size lesions (25 mm2; 20-56% of surface area) and in addition to the contralateral motor deficits they also showed provoked or spontaneous circling (score 3 or 4).
Table 1. Neurological Evaluation of Rats after MCAO
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
lesion size was not evaluated. Additional variables that have been reported to affect infarct size include variations in vascular anatomy (19) and the strain, blood glucose level, and age of the rats (7,18). In this study, we further defined the requirements for a predictable lesion size in the Sprague-Dawley rat. We also refined the Bederson et al. (1) clinical scale of evaluating postischemic motor-behavioral abnormalities and validated its accuracy with a comparable pathological grading. Finally, we evaluated the importance of variability in the anatomical pattern of the MCA and of the age of the animal as factors affecting the surface size of the brain infarct. Preliminary results were reported previously (12). MATERIALS AND METHODS Surgical procedure Two hundred eighteen adult male Sprague-Dawley rats weighing 180 to 540 g were anesthetized with ketamine (50 mg/kg) and xylazine (10 mg/kg). They were divided into three groups. In the first group of 133 rats (A), a 6- to 7-mm length of the right MCA was exposed via a subtemporal craniectomy and cauterized from the origin of the lateral striate arteries to the inferior cerebral vein (Fig. 1A). In the second group (B), 41 rats were subjected to an extended MCAO (Fig. 1B). The main trunk of the MCA was cauterized over a greater length (7-9 mm) than in Group A, from and including some of the distal lateral striate arteries to the parietal branch 2 to 3 mm beyond the inferior cerebral vein. In addition, the main branches were cauterized for a distance of approximately 5 mm from their origin. the third group (C) included 44 sham-operated animals in which the surgical exposure of the MCA was identical but without cauterization. Clinical evaluation of ischemia The results presented here are part of a larger study in which postischemic edema and blood-brain barrier permeability were examined at different time intervals ranging from 1, 3, 6, 12, and 24 hours, 2, 3, 5, and 7 days, and 2, 3, 4, 5, and 6 weeks after MCAO (13) . The neurological status of rats was assessed at 4 and 24 hours and 1 week after the MCAO and before death (up to 6 wk), although some rats were killed before the 24-hour and 1-week examinations. At each time, the highest score out of three consecutive trials was used. A scale of 0 to 4 was used to assess the motor and behavioral changes after the MCAO (Table 1 and Fig. 2). This test consisted of the following maneuvers: rats were suspended from the tail approximately 30 cm above the floor and their forelimb posture was noted. Normal animals extended both forelimbs toward the floor and were assigned a score of 0 (Fig. 2A). When the forelimb contralateral to the side of the MCAO was consistently flexed during the suspension and there was no other abnormality, the rat was scored 1 (Fig. 2B). Rats were then placed on absorbent pads and they were gently held by the tail. If animals showed an apparent decrease in grip in the contralateral forelimb when pulled by the tail, then
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Neurosurgery 1992-98 July 1992, Volume 31, Number 1 100 Middle Cerebral Artery Occlusion in Rats: A Neurological and Pathological Evaluation of a Reproducible Model Experimental Study
Pathological evaluation of ischemia For all measurements, animals were killed by decapitation and the brains were quickly removed and prepared for evaluation of the infarct surface area. Although an ischemic lesion has three dimensions, in this study, we used two-dimensional data to estimate surface area of the lesion. In a pilot group of 16 animals, from Group A the brain was removed, placed under a bright light, and using a magnifying glass, the margins between the pale discoloration of the ischemic and the surrounding healthy tissues were delineated. The lesions were primarily elliptical, with their major axes following the course of the main trunk of the MCA (Fig. 3). Using a transparent grid with 1 mm2 pixel size made from chart recording paper, the surface area of each lesion was measured by adding the number of square pixels overlying the infarcted area. Thereafter, the brain was immersed in a 2% solution of 2,3,5-triphenyltetrazolium chloride at room temperature for 60 minutes. The surface area was then similarly recalculated and compared with that obtained earlier by measuring the pale discoloration. Comparison of the two calculations proved identical when lesions of 3 days or older were examined. Ischemic discoloration was inconsistent within the initial 6 to 12 hours after MCAO. By 24 hours, however, ischemic discoloration was clearly evident in more than 95% of the brains examined. In Group A, 82 of 124 animals had their surface infarct area estimated by measuring pale discoloration 3 days after MCAO, and these data were analyzed for correlations between neurological grade and infarct size. From this, a pathological grading of the size of infarction was developed (Fig. 3). Grade I, the smallest sized lesions, were circular and ranged from 8 to 10 mm2 (mean, 9 ± 1 mm2). Grade IV, the largest lesions, ranged from 63 to 84 mm2 (mean, 78 ± 8 mm2) were elliptical in shape, and extended but never exceeded the plane of MCA anastomoses with the anterior and posterior cerebral arteries (Fig. 3). Grade II (range, 11-28 mm2; mean, 20 ± 5 mm2) and Grade III (range, 32-60 mm2; mean, 44 ± 9 mm2) were created arbitrarily by comparison of lesion sizes. Statistical analysis The χ2 test (contingency table analysis) was used for intragroup comparison of the significance of the anatomical vascular patterns of the MCA as factors affecting size of infarction. A Cohen's weighted κ analysis was used to analyze intergroup predictability
of infarct size from the clinical neurological score. The Cohen's κ analysis examines to what extent the two methods (neurological versus pathological) agree, correcting for agreement that would be due to chance alone. The Statview 512+ statistical software package for the Macintosh computer was used throughout. RESULTS Findings related to MCAO Fifteen animals (9 from Group A, 4 from Group B, 2 from Group C) of the total of 218 rats (7%) used in the study died between Day 1 and Day 10 after the MCAO; 4 died with hyperthermia and meningismus 48 hours after surgery; 2 died from intractable diarrhea at Day 3, and the remaining 9 died from unknown causes at Days 7 to 10. These animals were excluded from the study. The body weight of animals operated on declined by an average of 7.5% during the first 24 hours and 12.6% at 48 hours. It continued to decline until the 6th day, when it reached an average loss of 17.4%. Thereafter, the animals gained weight, and they regained their original weight by Day 11. In contrast, the sham-operated animals showed a minor decline in weight during the initial 48 hours but a steady increase thereafter. There was considerable variability in the number and location of the branches of the MCA from its origin to its bifurcation. Several branches constituted the lateral striate arterial complex, which supplies blood to the basal ganglia. For an unknown reason, there usually appeared to be many more lateral striate branches from the right than from the left proximal MCA (Fig. 4). Four major patterns of the MCA were observed (Fig. 5). Type A, with a pyriform and a frontal branch anteriorly and a temporal branch posteriorly, was observed in more than half of the rats (51%). Type B (26.6% of animals) with both pyriform and frontal branches but lacking the temporal one, was the second most common pattern. Type C (14%) with a bifurcating frontal branch and a posterior temporal branch, was the third most common pattern. Type D (7%) showed only a bifurcated frontal branch. Only 1.4% of the MCA vascular anatomy of the Sprague-Dawley rat fell outside of these four standard pat-terns. Clinical and pathological evaluation None of the sham-operated animals showed any motor-behavioral abnormalities as described in Table 1. However, all group A animals with MCAO developed one or more of these features upon recovery from surgical anesthesia approximately 2 hours after MCAO. By the end of 24 hours, 4 rats were scored 1, 30 scored 2, 36 scored 3, and the remaining 12 scored 4. By the first weekly reevaluation, 6 of 82 animals had clinically improved back to normality and 13 of 82 had deteriorated. Five of the 6 animals with the normalized neurological features, however, showed neurological scores at l week similar to those shown at their 24-hour evaluations if examined during induction or recovery from anesthesia. This effect became less apparent in
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
they were assigned a score of 2 (Fig. 2C). Thereafter and while still being held by the tail, the rats were allowed to move freely and were observed for circling behavior. Rats that moved spontaneously in all directions but established a monodirectional circling toward the paretic side when given a slight jerk of the tail were scored 3 (Fig. 2D). Rats that showed a consistent spontaneous contralateral circling were scored 4. Animals that showed a higher clinical score also showed all features of the lower grades. This neurological evaluation was completed within a few minutes.
the inferior cerebral vein resulted in a 100% infarction of the frontoparietal cortex. Using a similar occlusion technique, we observed considerable variability with just over half of the infarcts characterized as large. It was evident in our studies that the severity of the observed pathophysiological changes was proportionate to the size and, therefore, the severity of the cortical lesion. This prompted us to further expand and refine the Bederson modification (1) by extending the thermocoagulatory occlusion to a greater length of the main MCA trunk and to all major branches of the MCA from near its origin to beyond its distal junction with the inferior cerebral vein. As a result, the rate of large-sized infarcts was increased to 89% and more uniform and reproducible lesions were obtained. This more extensive occlusion increased the incidence of maximal lesions (Grade IV) from 15% to 60%, probably related to the effective compromise of the distal cortical interarterial anastomoses.
Effects of vascular anatomy of the MCA and animal age We examined the role of vascular variability as a factor determining the size of infarction in 82 animals of Group A. From the four anatomical vascular patterns identified (Fig. 5), none played a significant role in the variability in severity of ischemia. Similarly, neither the total number of MCA branches nor the number of branches to the frontal and/or parietal cortices played any significant role (contingency table analysis). The age of the rats as reflected by their weight before MCAO, however, appeared to be an important determinant of the extent of infarction (Fig. 8). In particular, rats weighing 300 g. The latter also showed a significantly larger lesion size than the former (P < 0.0004, contingency table analysis).
Clinical assessment The importance of accurately assessing the clinical status of animals with stroke in studies of the efficacy of treatment is obvious. The clinical method of neurological evaluation after MCAO used by Bederson et al. (1) employs three neurological grades: Grades 1 and 2 for motor deficits and Grade 3 for behavioral deficits. The overall sensitivity in predicting histological changes compatible with an infarction is 88%. Grade 2 indicates a consistently reduced resistance to lateral push behind the shoulders toward the paretic side as compared with the normal or mildly dysfunctional animals, which resisted sliding equally in both directions. In our experience, the application of this test is sensitive only to grossly paretic rats, it is subjective among different investigators, and it is rather difficult to reproduce accurately. Our four-score clinical test provides a reproducible Grade 2 as well as an intermediate behavioral grade (score 3), whereby the circling behavior is provoked as compared with the spontaneous circling in score 4, and this is correlated with a different size of infarction. The prediction of small, intermediate, or large size of infarction from this neurological evaluation is 83%; when the motor (1-2) and behavioral (3-4) scores are considered together, the sensitivity reaches 91%.
DISCUSSION Operative technique Rats are a convenient small animal model for the study of pathophysiology and treatment of cerebral ischemia. Tamura et al. (20) developed a subtemporal approach of proximal MCAO at a point near the origin of the lateral striate arteries, which produced infarction of both the cortex and the caudoputamen. The original technique, however, was very invasive and the animals survived only for a few hours (20). Subsequent modifications preserving the zygoma and the masseter muscle improved the postoperative survival for several days and eventually the subtemporal approach emerged as the standard technique of focal cortical ischemia in rats (7,15,16,18). It was apparent, however, in all of these works that the cortical somatosensory infarction was variable in size and location. Bederson et al. (1), using the subtemporal approach, could not produce a 100% cortical infarction by proximal focal occlusion of the MCA; their success rate varied between 13 and 67%. They observed that occlusion of the MCA from its origin medial to the olfactory bulb to its junction with
Pathological assessment Morphometric studies of the infarction and its surface quantification proved easy and comparable with other studies that used computer-assisted algorithms (5,11). In one previous study, the total lesion size in hypertensive stroke-prone rats with a single point ligation of the MCA distal to the rhinal fissure varied from 54 to 89 mm2 (11). A significant role of surface curvature in determining the absolute surface area was emphasized in that study (11). In another study using the same measurement algorithm, it was shown that the average surface area of infarction after MCAO and ipsilateral carotid artery occlusion in Long-Evans rats was 100 mm2. Our method is simpler but useful only in the measurement
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
animals that survived for 4 to 6 weeks after MCAO. In 37 surviving animals with the extended MCAO (Group B), only 2 animals improved and none deteriorated in their neurological scores between the 24 hours and 1 week. Of 82 animals in Group A, 34 developed small infarcts (4 in pathological Grade I and 30 in Grade II) and 48 large infarcts (36 in Grade III and 12 in Grade IV) (Fig. 6). We were able to predict the size of infarction from the neurological evaluation in 83% of the animals (Fig. 7). This accuracy reached 91% when pathological Grades I-II and III-IV and neurological scores 1-2 and 3-4 were considered together (P < 0.001, Cohen's weighted κ). In the 37 surviving animals of Group B, a substantial difference in the lesion size was observed. Indeed the extended occlusion of the MCA trunk and all major branches increased the proportion of the animals having a large (Grade III-IV) infarct from 59 to 89% (Fig. 6).
Factors affecting the size of infarction Spontaneously hypertensive and stroke-prone spontaneously hypertensive rats show approximately 1.5 times the infarct size of the normotensive Fischer 344 strain (7). Among normotensive rats, the Fischer 344 strain scored slightly better than the SpragueDawley strain, showing the largest and most reproducible lesions after MCAO. In Wistar-Kyoto rats, the stroke size was smallest and most variable. Spontaneously hypertensive or stroke-prone rats may have substantial anatomical, biochemical, and metabolic differences compared with the normotensive rats, and consequently, their use as a model may be limited. Among the normotensive strains the Fischer 344 appeared slightly better for predictability and size of stroke compared with Sprague-Dawley rats; however, the technical modifications described in our method using the Sprague-Dawley strain also produce large and reproducible infarcts. Six anatomical vascular patterns of the MCA in rats have been observed in morbidity studies. Their role in the formation of infarction after selective occlusion of these branches has been recently described (19). Distal occlusion of the parietal and frontal branches combined with temporary occlusion of the common carotid artery in the neck did not result in infarction of the areas supplied by these branches. According to the same investigators, this effect was attributed to the existence of an extensive collateral network between these branches and other arterial systems. Obliteration of the pyriform branch resulted in a consistent infarction, which was thought to indicate that the pyriform branch is an end artery without collateral blood supply. However, this conclusion is not supported by evidence from other investigators, who demonstrated collateral networks between the olfactory, pyriform, cingulate, and frontal cortex (3). Using high magnification observation in living animals, we demonstrated four major vascular patterns of the MCA, the most common of which had all major branches. The frontal and, indirectly, the pyriform branches are in collateral communication with the anterior cerebral artery, whereas the parietal and temporal branches collateralize with branches of the posterior cerebral artery. This explains why distal occlusions of the MCA and its branches do not result in infarction. On the other hand, a single point, proximal occlusion below the rhinal fissure produces inconsistent lesions, whereas occlusion of the MCA over 3 to 6 mm of length results in a 100% infarction rate (1). We suggest that the latter model provides a more efficient compromise of the collateral blood supply from retrograde flow via the major MCA branches. In our model, in which the main trunk of the MCA is occluded over a greater length and, in addition, the major MCA branches are also coagulated, the size of the infarction reaches maximal size as determined by the extent of the anastomoses between the MCA-anterior cerebral artery and MCA-
posterior cerebral artery network (3). Perhaps coagulation of the MCA trunk and its branches results in an intraluminal clot, which propagates distally compromising the collateral system to its maximum. The result of such a well described pathophysiological process is the extensive cortical infarctions we observed in our Group B. Although technically more demanding, our technique offers a reproducible and maximal cortical ischemic lesion in 90% of the animals. Cortical anastomoses are the key to the extent of infarction after a MCAO. This was clearly shown from the extent of maximal lesions in the present study, where the margins of infarction never exceeded the cortical interarterial anastomoses sites. It is not surprising, therefore, that none of the vascular anatomical patterns described in this study played a significant role in modulating the extent and severity of cortical ischemia. Similarly neither the total number of MCA branches nor the number of branches to the frontal and/or parietal cortices played any significant role. Another important factor in infarction after MCAO appears to be the body weight and, thus, indirectly, the age of the animals. How does this factor relate to the size of infarction? Probably through the availability and efficiency of the cortical interarterial anastomoses among the MCA, anterior cerebral artery, and posterior cerebral artery. In a study by Coyle (4), no young rats weighing below 150 g developed infarcts after proximal MCAO. In contrast, adult rats, cats, and primates nearly always develop infarcts after acute MCAO, although the size may vary considerably (2,10,14,17). This probably reflects the fact that distal collateral anastomotic vessels in adult primates, including humans, are not distributed uniformly (8) and are not as numerous (21) as in young animals (6). During the fetal and neonatal period of brain development in primates and, to a lesser degree, in species with non-convoluted brains such as the rat, these vascular anastomoses undergo substantial structural changes with age leading to regression (22). The extent of this regression may materially influence the size of brain infarction after MCAO. Thus, younger rats with a better collateral supply of the vascular bed of the MCA would be protected from infarction and neurological deficits. ACKNOWLEDGMENTS The authors thank Dr. Peter Coyle for many helpful discussions. This work was supported by Grants NS-23870 and NS-17760 from the National Institutes of Health. Received, July 22, 1991. Accepted, August 16, 1991. Reprint requests: A. Lorris Betz, M.D., Ph.D., University of Michigan, Department of Pediatrics, D3227 Medical Professional Building, Ann Arbor, MI 48109-0718. REFERENCES: (1-22) 1.
Bederson JB, Pitts LH, Tsuji M, Nishimura
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
of surface infarct. Nevertheless, the infarct size correlated well with the neurological symptoms.
3. 4. 5.
6. 7.
8. 9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
Nakayama H, Ginsberg MD, Dietrich WD: (S)Emopamil, a novel calcium channel blocker and serotonin S2 antagonist, markedly reduces infarct size following middle cerebral artery occlusion in the rat. Neurology 38:1667-1673, 1988. O'Brien MD, Waltz AG: Transorbital approach for occluding the middle cerebral artery without craniectomy. Stroke 4:201-206, 1973. Prado R, Ginsberg MD, Dietrich WD, Watson BD, Busto R: Hyperglycemia increases infarct size in collaterally perfused but not endarterial vascular territories. J Cereb Blood Flow Metab 8:186- 192, 1988. Rubino GJ, Young W: Ischemic cortical lesions after permanent occlusion of individual middle cerebral artery branches in rats. Stroke 19:870-877, 1988. Tamura A, Graham DI, McCulloch J, Teasdale GM: Focal cerebral ischemia in the rat: I. Description of technique and early neuropathological consequences following middle cerebral artery occlusion. J Cereb Blood Flow Metab 1:53-60, 1981. vander Eecken HM, Adams RD: The anatomy and functional significance of the meningeal arterial anastomoses of the human brain. J Neuropathol Exp Neurol 12:132-157, 1953. vander Eecken HM: Anastomoses Between the Leptomeningeal Arteries of the Brain: Their Morphological, Pathological and Clini-cal Significance. Springfield, IL, Charles C Thomas, 1959, pp 1-160.
COMMENT The authors characterize and further refine the rat model for producing cerebral infarction. They used two surgical methods. In one, the middle cerebral artery was cauterized from the lateral branch to the inferior cerebral vein. In the other, which yielded more large infarcts, the artery was cauterized more extensively, even including the main branchings for 5 mm distance. The size of the animal was important in determining infarct size but the cortical anastomotic pattern was not. Infarct size also correlated well with clinical examinations, as illustrated in the paper. This work represents prodigious investigative effort that goes a long way toward standardizing a reliable and inexpensive model for future stroke studies. Most of our other acceptable animal models suffer from the major flaw of non-reproducibility. Although the authors achieve good success rates with cautery, we wonder if even higher infarct rates could not have been obtained by division of the main trunk, because cauterization is sometimes not as complete as we think from external inspection. This might also remove the variable that may be produced by distal embolization. The relative unimportance of the cortical anastomotic pathway in rats is surprising, because in human middle cerebral artery occlusion,
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
2.
MC, Davis RL, Bartkowski H: Rat middle cerebral artery occlusion: Evaluation of the model and development of a neurologic examination. Stroke 17:472-476, 1986. Cohen MM: Animal models of cerebral infarction, in Klawans HL Jr (ed): Models of Human Neurological Diseases. Amsterdam, Excerpta Medica, 1974, pp 205-248. Coyle P: Arterial patterns of the rat rhinencephalon and related structures. Exp Neurol 49:671-690, 1975. Coyle P: Middle cerebral artery occlusion in the young rat. Stroke 13:855-859, 1982. Coyle P: Altered cerebral collaterals and protection from infarction, in Hartmann A, Kuschinsky W (eds): Cerebral Ischemia and Calcium. Berlin, Springer-Verlag, 1989, pp 6978. Coyle P, Jokelainen PT: Dorsal cerebral arterial collaterals of the rat. Anat Rec 203:397-404, 1982. Duverger D, MacKenzie ET: The quantification of cerebral infarction following focal ischemia in the rat: Influence of strain, arterial pressure, blood glucose concentration and age. J Cereb Blood Flow Metab 8:449461, 1988. Gillian LA: Potential collateral circulation to the human cerebral cortex. Neurology 24:941948, 1974. Ginsberg MD, Busto R: Rodent models of cerebral ischemia. Stroke 20:1627-1642, 1989. Hudgins WR, Garcia JH: Transorbital approach to the middle cerebral artery of the squirrel monkey: A technique for experimental cerebral infarction applicable to ultrastructural studies. Stroke 1:107-111, 1970. Jones PG, Coyle P: Microcomputer assisted lesion size measurements in spontaneously hypertensive stroke-prone rats. J Electrophysiol Tech 11:71-78, 1984. Menzies SA, Hoff JT, Betz AL: Clinical and pathological evaluation of focal brain ischemia after MCA occlusion in rats. Stroke 21:162, 1990 (abstr). Menzies SA, Hoff JT, Betz AL: Extravasation of albumin in ischemic brain edema, in Reulen HJ, Baethmann A, Fenstermacher J, Marmarou A, Spatz M (eds): Brain Edema. Vienna, Springer-Verlag, 1990, vol 8, pp 220222. Molinari GF, Moseley JI, Laurent JP: Segmental middle cerebral artery occlusion in primates: An experimental method requiring minimal surgery and anesthesia. Stroke 5:334339, 1974. Nakayama H, Dietrich WD, Watson BD, Busto R, Ginsberg MD: Photothrombotic occlusion of rat middle cerebral artery: Histopathological and hemodynamic sequelae of acute recanalization. J Cereb Blood Flow Metab 8:357-366, 1988.
this variable must account for the differences in the deficits that have been observed. This article constitutes a monumental effort.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Robert R. Smith Jackson, Mississippi
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Figure 1. Methods of direct middle cerebral artery occlusion in rats. A, occlusion of the MCA over 3 to 6 mm of its length between the olfactory bulb and the inferior cerebral vein (1). B, occlusion of the MCA over 7 to 9 mm of its length from and including some of the distal lateral striate arteries to the parietal branch 2 to 3 mm beyond the inferior cerebral vein with the additional occlusion of all its major branches (present study). M.C.A., middle cerebral artery; I.C.V., inferior cerebral vein; L.S.a, lateral striate artery; O.b., olfactory bulb; T.a., temporal artery; P.a., parietal artery; F.a., frontal artery; Pr.a., pyriform artery.
Figure 3. Schematic diagram of the distribution of the MCA and its anastomoses with the anterior and posterior cerebral arteries. The extent of the four grades of ischemia is also shown. The outer margins of Grade IV were always limited by the distal level of MCA anastomoses.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Figure 2. Neurological evaluation after MCAO. A, normal animal extends both forelimbs symmetrically toward the floor (score 0). B, after MCAO, the contralateral forelimb is consistently flexed during suspension (score 1). C, decreased grip of the contralateral forelimb when pulled by the tail (score 2). D, rat established a monodirectional circling toward the paretic side, when a slight jerk of the tail was done (score 3) or showed a consistent spontaneous contralateral circling (score 4).
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Figure 4. Anatomy of major vessels in a typical Sprague-Dawley rat as visualized by latex injection. Top, macroscopic overview of the circle of Willis in rat. Bottom, coronal section at the level of MCA and its lateral striate branches supplying mainly the caudoputamen to which infarction is extended when MCAO is done close to and including the lateral striate arteries. These branches are multiple and are more numerous on the left side as compared with the right.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Figure 5. Four major branching patterns of the MCA as seen through a craniectomy window and their relationship to surrounding anatomical landmarks including the inferior cerebral vein and the olfactory bulb. Type A was the most common, being observed in little more than half of the rats (51%), as compared with Type B (26.6%), Type C (14%), and Type D (7%). Only 1.4% of the MCA vascular anatomy fell outside these standard patterns.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Figure 6. Percentage of pathological grades (top left) with the occlusion of the main MCA trunk from the lateral striate arteries up to its crossing with the inferior cerebral vein (Group A) and (top right) with the extended MCA trunk and additional occlusion of all major branches of the main trunk (Group B). With the latter technique, the proportion of animals having a large (Grade III-IV) infarct was increased from 59% to 89%, and the incidence of maximal lesions (Grade IV) from 15% to 60%.
Figure 8. Correlation between body weight and size of infarction in 82 rats. Larger rats developed the largest infarcts as compared with smaller rats, the majority of which developed small lesions after the MCAO.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Figure 7. Comparison between neurological scores and pathological grades. Grades I and II represent smaller size lesions (25 mm2; 20-56% of surface area) and in addition to the contralateral motor deficits they also showed provoked or spontaneous circling (score 3 or 4).
Table 1. Neurological Evaluation of Rats after MCAO
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
