A Reproducible Model of Middle Cerebral Infarcts, Compatible with Long-term Survival, in Aged Rats
Background and Purpose Stroke is a disease associated with aging, but experimental stroke studies are generally done in young male animals. Because there are numerous differences associated with aging, such as an altered immune system and altered neurochemistry, that could affect the outcome of these experiments, a model of reproducible cerebral infarction in aged rats is needed.
Methods We attempted to produce middle cerebral artery (MCA) infarcts in aged (22 months of age) rats using two standard methods. A nylon suture with a heat-induced bulb was passed through the external carotid artery in seven animals, with an attempt to place the tip at the origin of the MCA. The MCA was ligated through a craniotomy just proximal to the internal cerebral vein in 14 rats. Survival potential was tested by attempting 2-week survival in four rats and 2-month survival in one rat.
Results The suture model failed to produce MCA infarcts, even when the bulb of the suture was properly placed in the MCA. The intracranial MCA occlusion resulted in reproducible MCA infarcts. There were no deaths, including the animals allowed to survive 2 weeks and 2 months.
Conclusions We conclude that reproducible MCA infarcts can be produced in aged rats by craniotomy and that these lesions may be compatible with long-term survival. This should be a useful technique for studying therapeutic interventions and rehabilitation strategies in an animal model that immunologically and neurochemically more closely resembles humans at risk for stroke.
There is considerable concern about the lack of correlation between treatments for stroke developed in the animal laboratory and the efficacy of those treatments in patients with stroke. Although “species differences” have been implicated,1 the failure to use age- and sex-matched animal models in the studies may be even more important.2
Stroke occurs more often in elderly people, and there are many age-related changes in the brain that could influence the outcome of stroke and of potential interventional strategies. Neurochemical changes occur in the old brain, such as alterations in levels of and sensitivity to dopamine,3 4 glutamate,5 and nerve growth factor.6 Parallel anatomic changes occur in aged brains of many species,7 including alterations of the blood-brain barrier,8 vascular thickening with accumulation of β-amyloid,9 and lipofuscin accumulation in neurons.10 11
The design of an animal model of stroke with relevance to human disease has always been difficult. Attempts to produce MCA infarcts in mature adult (12 months of age) rats resulted in death of the animals within 12 hours.12 We have succeeded in producing embolic strokes in aged rats, but our previous technique results in infarcts that often are not reproducible in terms of number, size, and location.13 Although this model has yielded important information on the decreased inflammatory response within cerebral infarcts in aged rats,13 the variability limits its usefulness in studying therapy.
The ability to adequately test therapeutic interventions in aged animals will depend on the availability of a model of stroke in aged animals that is reproducible in terms of size and location as well as compatible with survival of the animal. To develop such a model, we attempted to produce MCA infarcts in aged animals by methods that are standardly used to produce MCA infarcts in young rats.
Materials and Methods
The procedures followed were in accordance with institutional guidelines. Twenty-one male Fisher rats of 22 months of age weighing between 350 and 450 g were used. Rats were anesthetized with 4.5% chloral hydrate by intraperitoneal injection. The initial dose of 8 mg/kg was supplemented as needed. The surgical areas were shaved and washed three times with 2% iodine solution, and aseptic procedures were used. Seven animals underwent attempted suture occlusion of the right MCA, and 14 animals underwent intracranial occlusion of the right MCA. A rectal thermometer and heating unit were used to constantly maintain the body temperature at 37±0.5°C.
The method of Longa et al14 was used for attempted suture occlusion of the MCA in 7 animals. A 2-cm ventral midline incision was made in the neck. The right CCA was isolated and dissected distally to the bifurcation of the right ICA and ECA. The superior thyroid artery and the ascending pharyngeal artery, both branches of the ECA, were dissected and coagulated, and the ECA was mobilized. A 5-0 silk suture was tied loosely around the ECA, and a microvascular clip was placed on the CCA and the ICA adjacent to the ECA origin. A monofilament nylon suture, its tip rounded by heating near a flame, was introduced into the ECA through a microsurgical incision. A 4-0 nylon suture was used in 4 animals, and a 3-0 nylon suture was used in 3 animals. As the nylon suture was introduced, the 5-0 silk suture was tightened around the ECA stump to prevent bleeding through the ECA incision. The microvascular clip was then removed. The nylon suture was gently advanced from the ECA to the ICA, with visual observation documenting that the suture reached the base of the skull. When the suture was advanced 17.5 to 18 mm, resistance was felt, which suggested proper placement of the suture.14
A variation of the method of Brint et al15 was used for intracranial MCA occlusion in 14 animals. The right CCA was permanently occluded by double ligation with two 4-0 silk sutures. The neck wound was then sutured. The rat was placed in a prone position, and the head was turned to the left lateral position and stabilized by a head holder (Stoelting). Between the right orbit and the right external auditory canal, a 3-cm skin incision was made. The temporalis muscle was exposed and removed. All of the connective tissue was dissected from the bone around the surgical area. To clear the craniotomy field, part of the zygomatic arch was removed. Under the operating microscope, a 4-mm burr hole was drilled 2 to 3 mm rostral to the fusion of the zygomatic arch with the squamosal bone. When only a thin layer of temporal bone remained, the final removal of bone was completed with a fine dental probe. Next, the dura was opened, and the arachnoid membrane was separated with the tip of the probe. A fine wire with the tip at a 90° angle was held by a three-dimensional micromanipulator (Stoelting). Just proximal to the internal cerebral vein, the right MCA was lifted by the wire. The wire was heated with a thermocautery tip (Geiger-NY, model 100), occluding and then transecting the MCA. A small piece of gel foam was used to cover the burr hole. The skin incision was then sutured.
Animals were killed by transcardiac perfusion with 500 mL 0.9% NaCl, followed by 300 mL 4% paraformaldehyde. Three of the 7 suture model animals (3-0 suture) were killed immediately after the surgery. The 4 suture animals with successful passage of the 4-0 suture and the first 9 intracranial MCA animals were killed 24 hours after the surgery. Four animals with intracranial MCA occlusion were killed at 2 weeks, and 1 animal was killed at 2 months. Brains were removed immediately after perfusion and placed in the fixative solution for 1 hour. Gross coronal sections of the hemispheres, five 3-mm-thick blocks per brain, were made with a razor blade. Each block of tissue was soaked in fixative for 4 hours, washed with cold water for 20 minutes, processed without heat for 21 hours, and embedded in paraffin. A microtome was used to cut coronal sections of brain tissue of 7 μm in thickness. Semiserial sections were cut at intervals of 300 μm, producing an average of eight slides per block, which had shrunken to about 2.4 mm with processing. The slides were stained with hematoxylin and eosin and were studied under a light microscope.
Infarct volume was measured using the MCID M4 Image Analysis System (Imaging Research). The microscopic image was digitized with a Cohu camera, and the border of the infarct was manually outlined using the auto-outline tool, outlining the area of tissue necrosis. Distance between samples was then entered into the computer, which calculated an infarct volume. The first evaluation included all slides, with a distance of 300 μm between samples used to calculate infarct volume (Table⇓). To determine the optimal sampling distance, we omitted evenly spaced slides, increasing the sampling distance. We then calculated mean infarct volume, and standard deviation was calculated using three different protocols, with the number of slides per brain varying from 9 to 30 evenly spaced sections (see Table⇓).
The suture model failed to produce cerebral infarcts in all of the 7 rats. With the 4-0 suture, the suture was successfully advanced until resistance was felt at approximately 18 mm. It was noted that the passage of the suture was more difficult in the aged rats, with a stronger resistance felt as the bulb entered the carotid canal, compared with our experience in young rats. This suggested that the carotid canal was narrower in the aged rats. Suture placement was checked at death. In 1 rat, the suture was in the distal ICA, with a kink in the artery preventing further advancement into the MCA. In the other 3 rats, proper suture placement at the takeoff of the MCA was confirmed, but there were no infarcts by histology. Suture size was increased to 3-0 in 3 rats, but the bulb was too large to pass through the carotid canal in any of the 3 animals.
The intracranial MCA model produced infarcts in all of the 9 animals killed at 24 hours. The animals tended to move slowly after the experiment, with a trace of left hemiparesis and the head turned slightly to the right. There was no circling behavior. Soft food and water were kept available on the floor of the cage for the first 3 weeks in the long-term survival animals. After that, as the movement of the animals was normalized, they ate normally. All 5 long-term survival animals lost weight after the surgery (approximately 10% of body weight over 2 weeks). The 2-month survival animal began to regain weight after 1 month. Mean infarct volumes and standard deviations were measured using varying sample intervals (see Table⇑). The optimal sample number was 20, producing a mean infarct volume of 61.53±10.15 mm3. Decreasing the distance between samples, which increased the total number of sections sampled to 30, did not decrease the standard deviation (see Table⇑). The location of the infarct is illustrated in the Figure⇓.
MCA occlusion through a craniotomy results in a reproducible cerebral infarct in aged rats. On the basis of the infarct mean infarct volume and standard deviation, we can assume that the nontreated animals that would serve as a control group for testing a therapy would have a mean infarct volume of 61.53±10.15 mm3. This means that a sample size of 10 per group would be required to detect a change of 25% or larger in mean infarct volume between a control and an experimental group. This calculation assumes a two-sided test, α=.05, and power of at least 80%. The sample size and power calculations are based on a two-sample t test to compare mean infarct volume between the two groups. Such a high level of reproducibility should make this model very useful for testing potential therapeutic agents.
Models of MCA occlusion have been plagued by poor reproducibility in many species, with SHR having more consistent infarcts.15 This is thought to be due to the reduction in collateral blood flow that has been seen in SHR.16 This reduction in collaterals in SHR is also responsible for larger lesions after bilateral carotid occlusion when compared with normotensive rats.17 Since aging and hypertension produce similar changes in autoregulation,18 we expect similar mechanisms are at work in aged rats and SHR.
This model is also compatible with long-term survival after MCA infarction. The survival for as long as 2 months after MCA occlusion in aged animals is in contrast to the findings of others; previous attempts to produce stroke from MCA occlusion in mature adult (12 months of age) rats resulted in death within 12 hours.12 This led to the conclusion that only very young animals would survive after intracranial MCA occlusion. An alternate explanation for these early deaths is that the combination of the ischemia with autonomic effects of the urethane anesthesia,19 rather than the effects of the ischemia alone, resulted in death of the animals. Our results demonstrate that aged rats (which were even older at 22 months of age) can survive intracranial MCA occlusion produced under chloral hydrate anesthesia.
The main disadvantage of this model is the invasiveness of the technique. The less invasive suture model was unsuccessful in producing MCA occlusion. Most likely, this is due to age changes in connective tissues, with little elastin and collagen remaining in the aged vessels.20 Tortuous vessels have been reported with aging,21 22 consistent with our observation of dilated and tortuous carotid arteries in most of the old animals (N. Futrell, unpublished data, 1991 to 1995). One of these tortuous areas in the intracranial portion of the ICA prevented the advancement of the suture into the MCA. In the other animals, we presume the reduced elastin and collagen contents of aged vessels,20 which result in an increased diameter of the MCA, resulted in a lumen too large to be occluded by the same bulb that successfully occluded the MCA of younger animals.
The attempt to overcome this problem with the use of a larger suture was also unsuccessful. Hypertrophy of the bones in the skull, which is part of the normal aging process and reduces the temporal window and occasionally hampers the ability to perform transcranial Doppler in aging patients,23 would be expected to narrow the carotid canal. This fits with our observation of increased resistance in passing the 4-0 suture through the carotid canal in aged rats compared with young rats. The passage of the larger 3-0 suture, which might have compensated for the less elastic and larger MCA, was not possible because of the size of the carotid canal.
The other disadvantage of this model is the cost of the aged animals, which is near $100 per rat. Given the need for 10 animals per group, the cost of animals for a treatment trial would be $2000 compared with $340 to $400 for young rats (10 animals per group, two groups). Although this is a major increase in animal costs, animals represent a small fraction of the cost of doing research. When the entire cost of research is considered (including technical help and supplies), the use of old animals would increase the cost of most experiments by less than 10% to 15%.
An important consideration in using aged animals is the altered response to and metabolism of multiple drugs. Doses of all agents, including anesthetics and potential therapeutic agents, need to be recalculated for aged animals. Old rats are adequately anesthetized with 8 mg/kg of 4.5% chloral hydrate, while young rats require 10 to 12 mg/kg.13 In our initial experience with aged rats, we found that the chloral hydrate dose had to be decreased to prevent mortality.13 The age-related similarities in drug metabolism will be another way in which this model will better represent human disease, as both the therapeutic effects and side effects of particular doses of therapeutic agents for stroke will be better tested in aged animals.
MCA occlusion can be successfully accomplished via craniotomy in geriatric rats and is compatible with long-term survival, making studies of recovery and plasticity of the aged brain possible. The reproducibility of infarct volume is high enough to make this a very practical model for testing therapeutic interventions for stroke in an animal appropriately age-matched for the human population at risk for stroke.
Selected Abbreviations and Acronyms
|CCA||=||common carotid artery|
|ECA||=||external carotid artery|
|ICA||=||internal carotid artery|
|MCA||=||middle cerebral artery|
|SHR||=||spontaneously hypertensive rats|
This work was sponsored by US Public Health Service grant RO1AG11759. The authors thank Dr Andrew Slivka for his instruction in the production of MCA occlusion in rats via craniotomy. Sheryl Orok and Lori Wolinsky assisted in preparation of the manuscript. Clark Millikan, MD, was an advisor on this project and reviewed the manuscript.
- Received March 20, 1995.
- Revision received July 13, 1995.
- Accepted July 13, 1995.
- Copyright © 1995 by American Heart Association
Wiebers DO, Adams HP, Whisnant JP. Animal models of stroke: are they relevant to human disease? Stroke. 1990;21:1-3.
Millikan C. Animal stroke models. Stroke. 1992;23:795-797.
Arai Y, Kinemuchi H. Differences between monoamine oxidase concentrations in striatum and forebrain of aged and young rats. J Neural Transm. 1988;72:99-105.
Dayan AD. Comparative neuropathology of ageing: studies on the brains of 47 species of vertebrates. Brain. 1971;94:31-42.
Dahme E. Pathologische befunde an den hirngefäßen bei tieren. Acta Neuropathol (Berl). 1962;suppl 1:54-60.
Brizzee KR, Ordy JM, Kaack A. Regional lipofuscin in primate brain. J Gerontol. 1974;29:366-381.
Futrell N, Garcia JH, Peterson E, Millikan C. Embolic stroke in aged rats. Stroke. 1991;22:1582-1591.
Longa EZ, Weinstein PR, Carlson S, Cummins R. Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke. 1989;20:84-91.
Coyle P, Odenheimer BA, Sing CF. Cerebral infarction after middle cerebral artery occlusion in progenies of spontaneously stroke-prone and normal rats. Stroke. 1984;15:711-716.
Ogata J, Fujishima M, Morotomi Y, Omae T. Cerebral infarction following bilateral carotid artery ligation in normotensive and spontaneously hypertensive rats: a pathological study. Stroke. 1976;7:54-60.
Cechetto D. Experimental cerebral ischemic lesions and autonomic and cardiac effects. Stroke. 1993;12(suppl I):I-6-I-10.
Bogdahn U, Becker G, Winkler J, Greiner K, Perez J, Meurers B. Transcranial color-coded real-time sonography in adults. Stroke. 1990;21:1680-1688.