- Email: [email protected]
Large animal canine endovascular ischemic stroke models: A review
Accepted Manuscript Title: Large Animal Canine Endovascular Ischemic Stroke Models: A Review Author: Kunakorn Atchaneeyasakul MD Luis Guada MD Kevin Ramdas MD Mitsuyoshi Watanabe MD Pallab Bhattacharya PhD Ami P. Raval PhD Dileep R. Yavagal MD PII: DOI: Reference:
S0361-9230(16)30158-7 http://dx.doi.org/doi:10.1016/j.brainresbull.2016.07.006 BRB 9055
To appear in:
Brain Research Bulletin
Received date: Revised date: Accepted date:
31-3-2016 10-7-2016 12-7-2016
Please cite this article as: Kunakorn Atchaneeyasakul, Luis Guada, Kevin Ramdas, Mitsuyoshi Watanabe, Pallab Bhattacharya, Ami P.Raval, Dileep R.Yavagal, Large Animal Canine Endovascular Ischemic Stroke Models: A Review, Brain Research Bulletin http://dx.doi.org/10.1016/j.brainresbull.2016.07.006 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Title Large Animal Canine Endovascular Ischemic Stroke Models: A Review
Kunakorn Atchaneeyasakul MD1, Luis Guada MD1, Kevin Ramdas MD1, Mitsuyoshi Watanabe MD1, Pallab Bhattacharya PhD1, Ami P. Raval PhD1, Dileep R. Yavagal MD1 1.
Neurology Department/Interventional Division, University of Miami Miller School of
Medicine. Miami, FL 33136
Corresponding author: Dileep Yavagal, M.D. 1120 NW 14 ter. Miami, FL 33136 Email: [email protected] Phone: +1 9513578338
Page 1
Highlights - Canine models help us understand stroke process and develop novel treatment. - Newer endovascular techniques for canine stroke model are less invasive than surgical techniques. - This is a comprehensive review of endovascular canine stroke models
Page 2
ABSTRACT BACKGROUND Stroke is one of the leading causes of death and long-term disability worldwide. Recent exciting developments in the field with endovascular treatments have shown excellent outcomes in acute ischemic stroke. Prior to translating these treatments to human populations, a large-animal ischemic stroke model is needed. With the advent of new technologies in digital subtraction angiography, less invasive endovascular stroke models have been developed. Canines have gyrencephalic brain similar to human brain and accessible neurovascular anatomy for stroke model creation. Canine stroke model can be widely utilized to understand the disease process of stroke and to develop novel treatment. Less invasive endovascular internal carotid emboli injection and coil embolization methods can be used to simulate transient or permanent middle cerebral artery occlusion. Major restriction includes the extensive collateral circulation of canine cerebral arteries that can limit the stroke size. Transient internal carotid artery occlusion can decrease collateral circulation and increase stroke size to some degree. Additional method of manipulating the extent of collateral circulation needs to be studied. Other types of canine stroke models, including vertebral artery occlusion and basilar artery occlusion, can also be accomplished by endovascular thrombi injection. CONCLUSIONS We extensively review the literature on endovascular technique of creating canine ischemic stroke models and their application in finding new therapies for ischemic stroke.
Page 3
KEYWORDS Middle cerebral artery occlusion; Canine model; Stroke; Vertebral artery occlusion; Basilar artery occlusion
ABBREVIATIONS ACA: Anterior Cerebral Artery MCA: Middle Cerebral Artery PCA: Posterior Cerebral Artery ICA: Internal Carotid Artery ECA: External Carotid Artery VA: Vertebral Artery BA: Basilar Artery MCAo: Middle Cerebral Artery Occlusion pMCAo: Permanent Middle Cerebral Artery Occlusion tMCAo: Transient Middle Cerebral Artery Occlusion
Page 4
1. INTRODUCTION
Stroke is one of the leading causes of death and long-term disability worldwide. [1] With its significant disease burden, the treatment of stroke has been extensively studied and it has undergone significant changes throughout the years. Recent exciting developments in the field with endovascular treatments have shown excellent outcomes in acute ischemic stroke; this includes the use of the novel stent retriever. [2, 3] Prior to translating these treatments to human populations, a large-animal ischemic stroke model is needed to gather important information about the safety and efficacy of these endovascular procedures.
A variety of large-animal stroke models, such as non-human primates (NHP), canines, felines, ovis, swine, and leporidae, have been extensively used over the course of several decades to understand the disease process involved in stroke, and also to develop novel treatments that can translate to humans. [4-8] In choosing animal models that closely resemble humans, there are several factors that need to be considered, not only economically and anatomically, but also ethically. This is especially the case for NHP, where ethical components are the most common barrier to using this model, next to the high costs associated with this particular population. [9] This is why canine models are preferred among many research groups when studying various facets of stroke.
Historically, canine stroke models were established by directly injecting cells into the carotid artery to occlude the vessels. [10] With the advent of new technologies in digital
Page 5
subtraction angiography, less invasive endovascular stroke models have been developed. Multiple methods have been proposed to induce stroke in canine subjects, including angiography-guided embolus injection and coil embolization of the cerebral arteries. [11, 12]
In this review, we aim to evaluate all of the available literature associated with the endovascular technique of inducing stroke in a canine model.
2. WHY CANINES?
One of the most common forms of acute ischemic stroke results from large artery occlusion. This disease state is difficult to mimic ex vivo, and the data on the safety and feasibility of intra-arterial techniques cannot be obtained from non-animal studies. To date, there have been large numbers of animal stroke model trials that showed promising results for future therapeutic techniques, but they failed to translate into successful treatments in humans. [13] Per the recommendations from the Stroke Therapy Academic Industry Roundtable (STAIR), testing in a large-animal species with anatomy that more closely resembles that of humans shows promise in addressing this failed translational gap. [14]
Canines exhibit a number of characteristics that render them one of the ideal species for use after small-animal studies, but before beginning human clinical trials. Canines have a gyrencephalic brain with a proportion of gray/white matter that more closely
Page 6
resembles that of humans than the lissencephalic brains of small animals. [15, 16] The anatomical disparity between rodent and human brains makes rodent stroke model research difficult to translate to humans.
The neurovascular anatomy of canines enables their arteries to be catheterized in a similar manner to humans under angiographic guidance, allowing for minimally invasive endovascular stroke induction and stem cell delivery surgeries. [12] Endovascular procedure evaluation is very limited in small-animal models due to the small size of the cervical and cranial arteries in these animals; as such, the procedure cannot be visualized in real time by angiography. Other large-animal species such as felines, ovis, and swine have a well-developed carotid rete mirabile, a fine vascular network that supplies the cerebral arteries. [17, 18] This is an obstacle when advancing endovascular catheters and devices to the cranial arteries, particularly in the middle cerebral artery (MCA), which is the target of most stroke models. In these species, it is necessary to perform invasive craniotomy to access the MCA. [19] Structures such as the eye (which often is enucleated), temporalis muscle, and zygomatic arch suffer damage under this method. [20] Other disadvantages include the fact that the skull needs to be opened, ultimately affecting cerebral pressures; moreover, the dura mater is breached and cerebrospinal fluid is lost.
NHP are an ideal model for ischemic stroke research due to their close relation to humans and the similarity of their brains. Table 1 summarized the differences in utilizing canine and NHP as stroke models. The disadvantages of using NHP models
Page 7
include the costly maintenance fees, and the use of NHPs also raises more ethical concerns than the use of canines. [9, 21, 22] Even so, medical experiments on canines still carry major ethical considerations in the current era, as canines are considered “man’s best friend”, and society expresses concerns about the usage of canines when compared to other animals, such as rodents. [23]
3. CANINE CEREBRAL ARTERY SUPPLY
The canine brain is supplied by a cerebral arterial circle system, consisting of anterior and posterior circulation. [25] The anterior circulation consists of the anterior/rostral cerebral arteries (ACA), anterior communicating artery, internal carotid arteries (ICA), and MCA. The ICA is approximately half the size of the external carotid artery (ECA). The ICA enters the skull through the posterior foramen lacerum, travels through the carotid canal, and continues to be surrounded by the cavernous sinus, where it is joined by the ramus anastomicus, a branch of the maxillary artery. Beyond this point, the ICA forms an Sshaped curve and joins the cerebral arterial circle. Approaching the MCA via the ICA is relatively difficult due to the tortuosity of the canine ICA. [12] The canine MCA also receives more significant leptomeningeal circulation from the ipsilateral ACA and posterior/caudal cerebral artery (PCA), and significant maxilla–carotid collateral circulation is also present in canines when compared to humans, which is one difficulty that is faced when trying to repeat lesion sizes. [26]
Page 8
The posterior circulation is formed by the basilar artery (BA) caudally and the PCA. [25] The BA is formed after the vertebral arteries (VA) fuse at the level of the first cervical nerve origin. The canine BA has a uniform diameter throughout and is straight enough to accommodate endovascular devices, which is why this route is used more commonly than the anterior circulation to reach the MCA. Superior/Rostral cerebellar arteries and inferior/caudal cerebellar arteries both branch from the BA with some variations. Figure 1 shows the canine cerebral arterial supply.
The most common canine breeds used in canine endovascular stroke models are mongrels (weight range: 13–37 kg) or beagles (weight range: 9–15 kg). Many factors should be considered when choosing canine breed and size for an endovascular stroke model. Smaller canines have many disadvantages; for instance, it is more difficult to gain femoral access, and these animals also have a smaller caliber of intracranial vasculature. Conversely, larger canines can be more costly.
4. METHODS TO CREATE ENDOVASCULAR STROKE MODELS
4.1 Middle Cerebral Artery Occlusion Model Models of focal ischemia generally include permanent MCA occlusion (pMCAo) and transient MCA occlusion (tMCAo) techniques. pMCAo, or non-reperfusion models, can simulate an acute ischemic stroke without reperfusion, and it serves as the ideal model for studying recanalization, as well as reperfusion techniques involving endovascular instrument and procedures. tMCAo, or recanalization models, will simulate endovascular
Page 9
recanalization after an acute ischemic stroke; additionally, it could serve to test future therapies following reperfusion. The default study model recommended from the STAIR committee is the pMCAo; the exception to this rule is the use of tMCAo, which serves as a model to study possible reperfusion treatments – including stem cell therapy – where previous studies have found that the cells likely approach the infarcted area through the ipsilateral MCA, and no cell engraftment is seen in subjects with occluded MCA. [14, 27] Minimally invasive endovascular techniques for MCA occlusion (MCAo) in canines may be performed via multiple methods, including emboli injection via ICA for permanent MCAo and coil embolization of the MCA for both permanent and transient MCAo. [11, 12]
The two conventional routes used to access the MCA via an endovascular approach are the ICA or VA. Approaching the MCA via the ICA is relatively difficult due to the tortuosity of the petrous loop, which differs considerably between subjects and needs an experienced interventionalist to perform this procedure. Also, the size of the ICA is difficult to accommodate, even using small-diameter microguidewire (0.010–0.014 inches) and microcatheter systems. [12] The ICA route is mostly utilized for emboli injection, as the catheter used for the injection is placed a few centimeters above the origin of the ICA; it does not need to go through the tortuous sections of the ICA. The VA system tends to be utilized more frequently as a direct access point where the catheter is placed into the MCA. Furthermore, the BA is adequately straight and easier to navigate than the ICA. Potential adverse events include the tendency of mechanical vasospasm, which leads to permanent occlusion and results in a larger than expected infarction;
Page 10
however, this could be prevented by administering a vasodilating agent. Also, arterial perforation can occur during coil deployment. [28] 4.1.1
Internal Carotid Artery Emboli Injection for MCA Occlusion
The MCA is considered to be the natural continuation of the ICA; thus, an optimally sized embolus injection via the ICA can occlude the MCA. [29] This method is performed by an endovascular approach or via direct puncture of the ICA to acquire arterial access. The catheter is targeted towards 2 cm of the ascending part of the ICA using digital subtraction angiography. Emboli are injected through a syringe, which is directly connected to the catheter and flows through the ICA to occlude the MCA. The most commonly used emboli are autologous clots, created from the subject’s blood mixed with bovine thrombin. [11, 30, 31] Other types of emboli are synthetic contents, such as silk suture portions. [32] Autologous clot embolus has the advantage of being amenable to study the effects of pharmaceutical or mechanical thrombolysis. The injection of a single autologous clot 1.7 mm in diameter and 5 mm long has been proven to establish complete proximal MCAo with 100% efficiency (20/20 beagles) in 2 studies. [27, 31] This clot size was chosen due to the fact that the M1 segment in beagles is approximately 1 mm in diameter and 13 mm long. [31] MCAo with a clot 1.7 mm in diameter and 5 mm long resulted in acute stroke at the subcortical area (caudate, internal capsule), resembling a lacunar infarct with small average stroke volumes on magnetic resonance imaging (MRI) at 24 hours (150.01±153.61 mm3). [31] Larger clots with an approximate diameter of 2.33 mm have been shown to occlude the distal ICA with extension into the proximal MCA. Stroke volumes at 4 hours in each subject vary between 120–12,530 mm3. [33] The major drawback seen with an embolus injection MCAo model is that the precise site
Page 11
of the occlusion cannot be controlled. The infarct size resulting from complete MCAo is also relatively small compared to human MCAo given the extensive leptomeningeal collateral circulation in canines. [26]
Additional ipsilateral ICA transient occlusion has been proposed to reduce leptomeningeal collateral circulation in an emboli injection MCAo model to create larger infarct volumes. This can be accomplished by keeping the catheter in the ICA for an additional period to occlude blood flow. MCAo with an autologous clot embolus 1.7 mm in diameter in addition to 2 hours of ICA occlusion achieved an average stroke volume of 599.45±262.50 mm3 at 24 hours. [27] In one study, two autologous clots of 1.4 and 1.7 mm in diameter were injected consecutively to occlude the distal M1 and proximal M1, respectively, with occlusion of ICA blood flow for 2 hours using same technique as described above. This method was able to create larger MCA territory stroke with an average stroke volume of 4,173.23±603.92 mm3 at 24 hours. [30]
An alternative approach to create larger stroke volume can be performed by clot injection via the ICA terminus to occlude the MCA and ACA. Access through the VA system is preferred to bypassing the tortuous ICA. Boulos et al. injected 1 mL of autologous clot through the ICA terminus, and they were able to accomplish MCA and ACA occlusion, where the average stroke volume was 9,540±1,800 mm3. MCA, ACA, and PCA occlusion were seen in some subjects with a mean stroke volume of 15,870±3,840 mm3. [34] Figure 2 shows the ICA emboli injection method for MCAo.
Page 12
The disadvantage of this technique is that it may not be reproducible in different dog breeds that feature smaller and more tortuous ICAs. This model was also utilized to evaluate for mesenchymal stem cell delivery via intra-arterial method in one study. [27] 4.1.2
Coil Occlusion for MCA Occlusion
A soft platinum coil used to treat intracranial aneurysms can be utilized for a canine tMCAo model. The coil can be precisely positioned by advancing the microcatheter to the proximal M1, and it can be repositioned until appropriate complete MCAo is achieved. This precise positioning method cannot be achieved by emboli injection via ICA. Transient occlusion is also possible without detaching the coil, and by retrieving it after a selected duration. The most common route to access the MCA is via the VA and by gaining access into the cerebral arterial circle through the BA. ICA access is generally not performed due to tortuosity, as mentioned. Even though posterior circulation access has been well described, it might not be recommended for use in every dog breed. To avoid deadly complications, such as vessel perforation, it is imperative for the surgeon to consider the anatomical variations between each breed; in some cases, access via the ICA maybe a better choice. In multiple reported MCAo models, the platinum coil was delivered into the MCA, including the entire M1 segment. Angiography was repeated every 15 minutes to confirm complete occlusion, and the total occlusion time was 60 minutes; the coil was then retrieved. This method was able to achieve a hemispherical infarct volume of 30.9%±2.1%. [12, 28, 35] In one study, the coil was positioned to occlude the proximal MCA, ACA, and distal ICA with the aim of creating a larger stroke; however, in 3/8 of the animals faced premature mortality within 24 hours. [28] Figure 3
Page 13
shows the coil occlusion method for MCAo. Table 2. summarized the comparison of Ischemic Stroke Size by Each MCA Occlusion Techniques.
4.2 Vertebral Artery Occlusion Model Autologous clot injection is a potential method that is used to create VA occlusion in a canine model. The obstacle to this is that the canine VA features strong blood flow and it is difficult for the clot to remain anchored. A recent study proposed the usage of a selfexpanding thrombus filter preinstalled in the delivery catheter. A clot 3 mm × 20 mm in size was injected through a catheter that was proximally positioned 2 cm from the thrombus filter; the filter was left for 15 minutes to allow implantation of the clot. This resulted in successful complete occlusion of the VA in 12/12 subjects. [36]
This model may be useful to further evaluate novel endovascular therapy for acute VA occlusion in humans. Further imaging and clinical outcome studies of this canine model are still needed.
4.3 Basilar Artery Occlusion Model This model was created to test the effectiveness of intraarterial tPA therapy versus intravenous tPA therapy for acute basilar occlusion. Autologous clot injection is utilized. Two clots were injected consecutively via a microcatheter placed in the proximal portion of the BA. All 13 subjects showed successful BA occlusion; 11 subjects had proximal BA occlusion and 2 subjects had mid-BA occlusion. In this study, the subjects were used to test recanalization by intravenous versus intraarterial thrombolysis, and they were
Page 14
humanely killed 6 hours after therapy. [37] Figure 4 shows the BA occlusion method that was used.
The disadvantages of this technique include the fact that it poses major ethical challenges and more extensive care of the canines is required post-procedure. Complete BA occlusion will likely cause a severe and fatal neurological deficit, potentially leading to canine suffering, which may require prolonged sedation to achieve the desired survival time.
5. BEHAVIORAL ASSESSMENT The National Institutes of Health Stroke Scale (NIHSS) is widely used as a tool to quantify the level of stroke impairment, and it has been shown to strongly predict outcomes after stroke. [38] Many research groups have developed a stroke scale similar to the NIHSS to assess canine neurological function by observing canine behavior following the induction of stroke. [28, 34, 39] Boulos et al. proposed a canine stroke score ranging from 0 (normal) to 18 (comatose). [34] This score assesses the canine’s level of consciousness, vocalization, gait, behavior, motor function, and sensory function. In their study, this score was correlated with infarct volume. The proposed canine stroke scale could potentially be used as a clinical assessment tool for successful stroke model creation, or as a clinical indicator of improvement with stroke therapy trials. Assessment of canine post-stroke induction should be performed at multiple times. This should include early assessment following recovery from anesthesia, as many canines may experience rapid recovery soon after.
Page 15
Myriad other factors must also be considered for canine recovery post-stroke induction, including stroke location, duration of the occlusion, and time under anesthesia. It is important to mention that during the first 6–8 hours, canines should be under intensive care with strict monitoring of vital signs and fluid resuscitation as part of their further recovery and stroke assessment. A post-procedure care protocol should be standardized, as this may impact future neurological recovery and assessment.
6. DISCUSSION
With the advancements in digital subtraction angiography technology, the less invasive endovascular method (as compared to the surgical approach) has been preferred for stroke model creation in recent years. The guided fluoroscopy approach also allows for real-time visualization of MCAo. One of the earliest reported endovascular canine stroke models, where microfibrillar collagen is injected and delivered via femoral catheterization, has shown success in creating a canine stroke model without the need for surgery, which involves cervical and intracranial vasculature. [40] Since then, multiple techniques for endovascular canine stroke model creation have been developed.
An emboli injection technique is the most preferred for a pMCAo model. Drawbacks include difficulty controlling the site of the emboli occlusion. Multiple studies suggested that using a single clot 1.7 mm in diameter and 5 mm long successfully occludes the proximal MCA with a high success rate. [27, 31] This should provide a more predictable
Page 16
occlusion at the proximal MCA than using multiple random sized clot injections. Even then, we should assume that the proximal MCA size is slightly different between each subject and breed; thus, 100% successful occlusion in all subjects may not be achieved.
A transient MCAo or recanalization model is ideal for studying the effect of future treatments that require proper reperfusion of the occluded cerebral vessel in order to be effective, which includes cell-based therapy. In recent years, the study of regenerative medicine in ischemic stroke patients has shown tremendous advancements, with attempts to find future novel therapies to reduce stroke morbidity. The use of stem cells has been a topic of interest in many research groups, and a tMCAo canine model by coil occlusion is a suitable point of focus. One major limitation noted in literature reviews, and from our own experience, in the use of this tMCAo canine model at our institution includes the inability to create a reliable infarct size. This is mainly due to the variability in the arterial anatomy between subjects, even within the same age range. [12, 28, 35] Brain infarct volume is reported to be a major predictor of stroke outcome, and the reduction of infarct volume is a major primary outcome in stem cell therapy studies. [41, 42] Larger scale prospective studies of a novel method to create a canine tMCAo models with reliable stroke size are needed to further study the efficacy of stem cell therapy.
Although there are multiple advantages of choosing canines over other species for producing preclinical ischemic stroke models, notable disadvantages and challenges are seen in the literature. Canine MCA has significant maxilla–carotid collateral circulation and higher amounts of leptomeningeal collateral circulation from the ipsilateral ACA and
Page 17
PCA when compared to humans. [26] If a larger stroke size is desired, additional methods to minimize collateral flow is recommended by transient occlusion of the ICA. [27] Supplementary ipsilateral ACA and/or PCA occlusion can also increase infarct size. [34] In both a transient and permanent MCAo model, infarct volume is difficult to control and varies extensively in each canine, even though the same technique is used throughout. This is likely due to the different extent in collateral circulation between each canine subject. In subjects with extensive infarct size, premature mortality can occur. Assessment of pial collateral recruitment by digital subtraction angiography prior to recanalization in the tMCAo model is possible and may be able to predict outcomes and final infarct volumes based on the extent of the visualized leptomeningeal collaterals. [28] Multiple factors can influence the leptomeningeal collaterals, including systemic blood pressure, which affects flow through the anastomoses. [43, 44] Modifying the systemic blood pressure during stroke induction may be a possible method to reduce collateral flow. Permanent occlusion of the proximal ACA may be able to create a larger stroke size and limit collateral circulation, which can be performed prior to tMCAo with a retrievable coil. Further studies are needed to prove these hypotheses.
Large-animal models, including canines, are associated with high purchase costs, which are generally related to the breed and age; these animals also include high accommodation fees. [45, 46] Due to limitations in research funding, fewer canine subjects can be studied when compared to using small-animal models. Even with costly maintenance in canine subjects, NHP are generally associated with higher costs and more ethical considerations. Choosing an endovascular approach for a stroke model also
Page 18
requires an animal research facility that can accommodate digital subtraction angiography, which may not be available at every institution. There is also a steep learning curve for surgeons to navigate canine cerebral vasculature, particularly among those practitioners with prior training in human neuro–endovascular procedures, due to the significant differences mentioned in our review. Thus, surgical access through the cervical area for an ICA clot injection is still preferred by some research groups, particularly in recent years. [39, 47]
7. CONCLUSIONS
Canine has gyrencephalic brain similar to human brain and accessible neurovascular anatomy for stroke model creation. Canine stroke model is widely utilized to understand the disease process of stroke and to develop novel treatment. Less invasive endovascular internal carotid emboli injection and coil embolization methods can be used to simulate transient or permanent MCA occlusion. Major restriction includes the extensive collateral circulation of canine cerebral arteries that can limit the stroke size. Transient ICA occlusion can decrease collateral circulation and increase stroke size to some degree. Additional method of manipulating the extent of collateral circulation needs to be studied. Other types of canine stroke models, including VA occlusion and BA occlusion, can also be accomplished by endovascular thrombi injection.
ACKNOWLEDGEMENTS
Page 19
We are grateful to Aisha Khan, Ph.D. and team at University of Miami Interdisciplinary Stem Cell Institute to allow the use of digital subtraction angiography images.
Page 20
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
Feigin, V.L., et al., Global and regional burden of stroke during 1990-2010: findings from the Global Burden of Disease Study 2010. Lancet, 2014. 383(9913): p. 245-54. Gounis, M.J., A.K. Wakhloo, and J.Y. Chueh, Preclinical investigations for thrombectomy devices--does it translate to humans? Stroke, 2013. 44(6 Suppl 1): p. S7-S10. Saver, J.L., et al., Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke. N Engl J Med, 2015. 372(24): p. 2285-95. Cook, D.J. and M. Tymianski, Nonhuman primate models of stroke for translational neuroprotection research. Neurotherapeutics, 2012. 9(2): p. 371-9. Culp, W.C., et al., Three variations in rabbit angiographic stroke models. J Neurosci Methods, 2013. 212(2): p. 322-8. Duberstein, K.J., et al., Gait analysis in a pre- and post-ischemic stroke biomedical pig model. Physiol Behav, 2014. 125: p. 8-16. Boltze, J., et al., Permanent middle cerebral artery occlusion in sheep: a novel large animal model of focal cerebral ischemia. J Cereb Blood Flow Metab, 2008. 28(12): p. 1951-64. Hayakawa, T. and A.G. Waltz, Immediate effects of cerebral ischemia: evolution and resolution of neurological deficits after experimental occlusion of one middle cerebral artery in conscious cats. Stroke, 1975. 6(3): p. 321-7. Coors, M.E., et al., The ethics of using transgenic non-human primates to study what makes us human. Nat Rev Genet, 2010. 11(9): p. 658-62. Hill, N.C., et al., Studies in cerebrovascular disease. VII. Experimental production of cerebral infarction by intracarotid injection of homologous blood clot; preliminary report. Proc Staff Meet Mayo Clin, 1955. 30(26): p. 625-33. Shaibani, A., et al., First results in an MR imaging--compatible canine model of acute stroke. AJNR Am J Neuroradiol, 2006. 27(8): p. 1788-93. Rink, C., et al., Minimally invasive neuroradiologic model of preclinical transient middle cerebral artery occlusion in canines. Proc Natl Acad Sci U S A, 2008. 105(37): p. 14100-5. Kahle, M.P. and G.J. Bix, Successfully Climbing the "STAIRs": Surmounting Failed Translation of Experimental Ischemic Stroke Treatments. Stroke Res Treat, 2012. 2012: p. 374098. Stroke Therapy Academic Industry, R., Recommendations for standards regarding preclinical neuroprotective and restorative drug development. Stroke, 1999. 30(12): p. 2752-8. Ferrer, I., I. Fabregues, and E. Condom, A Golgi study of the sixth layer of the cerebral cortex. II. The gyrencephalic brain of Carnivora, Artiodactyla and Primates. J Anat, 1986. 146: p. 87-104. Wynshaw-Boris, A., et al., Lissencephaly: mechanistic insights from animal models and potential therapeutic strategies. Semin Cell Dev Biol, 2010. 21(8): p. 823-30.
Page 21
17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34.
Gillilan, L.A., Extra- and intra-cranial blood supply to brains of dog and cat. Am J Anat, 1976. 146(3): p. 237-53. Gralla, J., et al., A dedicated animal model for mechanical thrombectomy in acute stroke. AJNR Am J Neuroradiol, 2006. 27(6): p. 1357-61. Wells, A.J., et al., A surgical model of permanent and transient middle cerebral artery stroke in the sheep. PLoS One, 2012. 7(7): p. e42157. Howells, D.W., et al., Different strokes for different folks: the rich diversity of animal models of focal cerebral ischemia. J Cereb Blood Flow Metab, 2010. 30(8): p. 1412-31. Tong, F.C., et al., An Enhanced Model of Middle Cerebral Artery Occlusion in Nonhuman Primates Using an Endovascular Trapping Technique. AJNR Am J Neuroradiol, 2015. 36(12): p. 2354-9. Quigley, M., Non-human primates: the appropriate subjects of biomedical research? J Med Ethics, 2007. 33(11): p. 655-8. Hasiwa, N., et al., Critical evaluation of the use of dogs in biomedical research and testing in Europe. ALTEX, 2011. 28(4): p. 326-40. de Crespigny, A.J., et al., Acute studies of a new primate model of reversible middle cerebral artery occlusion. J Stroke Cerebrovasc Dis, 2005. 14(2): p. 807. Budras, K.-D., et al., Anatomy of the dog : an illustrated text. 4th ed. 2002, Hannover: Schlütersche. 217 p. Symon, L., Observations on the leptomeningeal collateral circulation in dogs. J Physiol, 1960. 154(1): p. 1-14 2. Lu, S.S., et al., In vivo MR imaging of intraarterially delivered magnetically labeled mesenchymal stem cells in a canine stroke model. PLoS One, 2013. 8(2): p. e54963. Christoforidis, G.A., et al., An endovascular canine middle cerebral artery occlusion model for the study of leptomeningeal collateral recruitment. Invest Radiol, 2011. 46(1): p. 34-40. Purdy, P.D., et al., Reversible middle cerebral artery embolization in dogs without intracranial surgery. Stroke, 1989. 20(10): p. 1368-76. Zu, Q.Q., et al., An endovascular canine stroke model: middle cerebral artery occlusion with autologous clots followed by ipsilateral internal carotid artery blockade. Lab Invest, 2013. 93(7): p. 760-7. Liu, S., et al., A novel embolic stroke model resembling lacunar infarction following proximal middle cerebral artery occlusion in beagle dogs. J Neurosci Methods, 2012. 209(1): p. 90-6. Yasuda, R., et al., C-arm CT measurement of cerebral blood volume using intraarterial injection of contrast medium: an experimental study in canines. AJNR Am J Neuroradiol, 2012. 33(9): p. 1696-701. van der Bom, I.M., et al., Quantitative evaluation of C-arm CT cerebral blood volume in a canine model of ischemic stroke. AJNR Am J Neuroradiol, 2012. 33(2): p. 353-8. Boulos, A.S., et al., Tamoxifen as an effective neuroprotectant in an endovascular canine model of stroke. J Neurosurg, 2011. 114(4): p. 1117-26.
Page 22
35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47.
Rink, C., et al., Tocotrienol vitamin E protects against preclinical canine ischemic stroke by inducing arteriogenesis. J Cereb Blood Flow Metab, 2011. 31(11): p. 2218-30. Zhang, Y., et al., A Novel Canine Model of Acute Vertebral Artery Occlusion. PLoS One, 2015. 10(11): p. e0142251. Qureshi, A.I., et al., Randomized comparison of intra-arterial and intravenous thrombolysis in a canine model of acute basilar artery thrombosis. Neuroradiology, 2004. 46(12): p. 988-95. Adams, H.P., Jr., et al., Baseline NIH Stroke Scale score strongly predicts outcome after stroke: A report of the Trial of Org 10172 in Acute Stroke Treatment (TOAST). Neurology, 1999. 53(1): p. 126-31. Kang, B.T., et al., Canine model of ischemic stroke with permanent middle cerebral artery occlusion: clinical and histopathological findings. J Vet Sci, 2007. 8(4): p. 369-76. Purdy, P.D., et al., Microfibrillar collagen model of canine cerebral infarction. Stroke, 1989. 20(10): p. 1361-7. Li, Y., et al., Treatment of stroke in rat with intracarotid administration of marrow stromal cells. Neurology, 2001. 56(12): p. 1666-72. Vogt, G., et al., Initial lesion volume is an independent predictor of clinical stroke outcome at day 90: an analysis of the Virtual International Stroke Trials Archive (VISTA) database. Stroke, 2012. 43(5): p. 1266-72. Brozici, M., A. van der Zwan, and B. Hillen, Anatomy and functionality of leptomeningeal anastomoses: a review. Stroke, 2003. 34(11): p. 2750-62. Symon, L., S. Ishikawa, and J.S. Meyer, Cerebral arterial pressure changes and development of leptomeningeal collateral circulation. Neurology, 1963. 13: p. 237-50. Davis, P.R. and E. Head, Prevention approaches in a preclinical canine model of Alzheimer's disease: benefits and challenges. Front Pharmacol, 2014. 5: p. 47. Chu, C.R., M. Szczodry, and S. Bruno, Animal models for cartilage regeneration and repair. Tissue Eng Part B Rev, 2010. 16(1): p. 105-15. Kang, B.T., et al., Three-dimensional time-of-flight magnetic resonance angiography of intracranial vessels in a canine model of ischemic stroke with permanent occlusion of the middle cerebral artery. Comp Med, 2009. 59(1): p. 72-7.
Page 23
Figure 1 shows canine cerebral arterial supply by illustration and digital subtraction angiography imaging. ACA-anterior cerebral artery, MCA-middle cerebral artery, ICAinternal carotid artery, BA-basilar artery, VA-vertebral artery.
Figure 2 shows the internal carotid artery emboli injection method for middle cerebral artery occlusion. Catheter is targeted at 2 cm in the ascending part of ICA for injection. ICA-internal carotid artery, CCA-common carotid artery, ECA-external carotid artery.
Figure 3 shows the coil occlusion method for middle cerebral artery occlusion. MCA is accessed via the vertebrobasilar system and coil is delivered to occlude the entire M1 segment. ACA-anterior cerebral artery, MCA-middle cerebral artery, BA-basilar artery, VA-vertebral artery.
Figure 4 shows basilar artery occlusion method. Microcatheter is placed in the proximal BA for injection of autologous clots. ACA-anterior cerebral artery, MCA-middle cerebral artery, BA-basilar artery, VA-vertebral artery.
Table 1. Differences in utilizing canine and NHP as stroke models. ICA = internal carotid artery, MCA = middle cerebral artery, ECA = External carotid artery, NHP = NonHuman Primate
Table 2. Comparison of Ischemic Stroke Size by Each Middle Cerebral Artery Occlusion Techniques.
Page 24
Table 1. Differences in utilizing canine and NHP as stroke models. ICA = internal carotid artery, MCA = middle cerebral artery, ECA = External carotid artery, NHP = NonHuman Primate Canine Breed, Species - Mongrel - Beagle
Non-Human Primate - Rhesus [21] - Cynomolgus [24]
ICA
- Smaller luminal diameter than human - ECA is larger - High tortuosity - MCA inaccessible by most endovascular methods via ICA route (Many endovascular thrombectomy devices cannot be studied)
- Smaller luminal diameter than human - Branching similar to human - MCA accessible via ICA route (can study thrombectomy devices via this route)
MCA
- Smaller luminal diameter than human (similar to NHP) - Extensive maxillo-carotid anastomosis - Stroke size less predictable
- Smaller luminal diameter than human (similar to canine) - More similar to human
- ICA thrombus injection - Reversible Coil Occlusion
- ICA thrombus injection - Endovascular trapping technique [21] - Reversible Coil/Balloon Occlusion
- Lower than NHP
- Higher than canine
- Lower than NHP
- Higher than canine
Cerebral Vasculature Collateral Anastomosis Endovascular MCA occlusion Method Financial Burden Ethical consideration
Page 25
Table 2 Comparison of Ischemic Stroke Size by Each Middle Cerebral Artery Occlusion Techniques. Stroke Size ICA Emboli Injection - clot 1.7mm - clot 2.3mm - clot 1.7mm + transient ICA occlusion - clot 1.4mm and 1.7mm + transient ICA occlusion Coil Occlusion - 60 min occlusion
150.01±153.61 mm3 120-12530 mm3 599.45±262.50 mm3 4173.23±603.92 mm3
30.9±2.1% hemispherical infarct volume
Page 26
S0361-9230(16)30158-7 http://dx.doi.org/doi:10.1016/j.brainresbull.2016.07.006 BRB 9055
To appear in:
Brain Research Bulletin
Received date: Revised date: Accepted date:
31-3-2016 10-7-2016 12-7-2016
Please cite this article as: Kunakorn Atchaneeyasakul, Luis Guada, Kevin Ramdas, Mitsuyoshi Watanabe, Pallab Bhattacharya, Ami P.Raval, Dileep R.Yavagal, Large Animal Canine Endovascular Ischemic Stroke Models: A Review, Brain Research Bulletin http://dx.doi.org/10.1016/j.brainresbull.2016.07.006 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Title Large Animal Canine Endovascular Ischemic Stroke Models: A Review
Kunakorn Atchaneeyasakul MD1, Luis Guada MD1, Kevin Ramdas MD1, Mitsuyoshi Watanabe MD1, Pallab Bhattacharya PhD1, Ami P. Raval PhD1, Dileep R. Yavagal MD1 1.
Neurology Department/Interventional Division, University of Miami Miller School of
Medicine. Miami, FL 33136
Corresponding author: Dileep Yavagal, M.D. 1120 NW 14 ter. Miami, FL 33136 Email: [email protected] Phone: +1 9513578338
Page 1
Highlights - Canine models help us understand stroke process and develop novel treatment. - Newer endovascular techniques for canine stroke model are less invasive than surgical techniques. - This is a comprehensive review of endovascular canine stroke models
Page 2
ABSTRACT BACKGROUND Stroke is one of the leading causes of death and long-term disability worldwide. Recent exciting developments in the field with endovascular treatments have shown excellent outcomes in acute ischemic stroke. Prior to translating these treatments to human populations, a large-animal ischemic stroke model is needed. With the advent of new technologies in digital subtraction angiography, less invasive endovascular stroke models have been developed. Canines have gyrencephalic brain similar to human brain and accessible neurovascular anatomy for stroke model creation. Canine stroke model can be widely utilized to understand the disease process of stroke and to develop novel treatment. Less invasive endovascular internal carotid emboli injection and coil embolization methods can be used to simulate transient or permanent middle cerebral artery occlusion. Major restriction includes the extensive collateral circulation of canine cerebral arteries that can limit the stroke size. Transient internal carotid artery occlusion can decrease collateral circulation and increase stroke size to some degree. Additional method of manipulating the extent of collateral circulation needs to be studied. Other types of canine stroke models, including vertebral artery occlusion and basilar artery occlusion, can also be accomplished by endovascular thrombi injection. CONCLUSIONS We extensively review the literature on endovascular technique of creating canine ischemic stroke models and their application in finding new therapies for ischemic stroke.
Page 3
KEYWORDS Middle cerebral artery occlusion; Canine model; Stroke; Vertebral artery occlusion; Basilar artery occlusion
ABBREVIATIONS ACA: Anterior Cerebral Artery MCA: Middle Cerebral Artery PCA: Posterior Cerebral Artery ICA: Internal Carotid Artery ECA: External Carotid Artery VA: Vertebral Artery BA: Basilar Artery MCAo: Middle Cerebral Artery Occlusion pMCAo: Permanent Middle Cerebral Artery Occlusion tMCAo: Transient Middle Cerebral Artery Occlusion
Page 4
1. INTRODUCTION
Stroke is one of the leading causes of death and long-term disability worldwide. [1] With its significant disease burden, the treatment of stroke has been extensively studied and it has undergone significant changes throughout the years. Recent exciting developments in the field with endovascular treatments have shown excellent outcomes in acute ischemic stroke; this includes the use of the novel stent retriever. [2, 3] Prior to translating these treatments to human populations, a large-animal ischemic stroke model is needed to gather important information about the safety and efficacy of these endovascular procedures.
A variety of large-animal stroke models, such as non-human primates (NHP), canines, felines, ovis, swine, and leporidae, have been extensively used over the course of several decades to understand the disease process involved in stroke, and also to develop novel treatments that can translate to humans. [4-8] In choosing animal models that closely resemble humans, there are several factors that need to be considered, not only economically and anatomically, but also ethically. This is especially the case for NHP, where ethical components are the most common barrier to using this model, next to the high costs associated with this particular population. [9] This is why canine models are preferred among many research groups when studying various facets of stroke.
Historically, canine stroke models were established by directly injecting cells into the carotid artery to occlude the vessels. [10] With the advent of new technologies in digital
Page 5
subtraction angiography, less invasive endovascular stroke models have been developed. Multiple methods have been proposed to induce stroke in canine subjects, including angiography-guided embolus injection and coil embolization of the cerebral arteries. [11, 12]
In this review, we aim to evaluate all of the available literature associated with the endovascular technique of inducing stroke in a canine model.
2. WHY CANINES?
One of the most common forms of acute ischemic stroke results from large artery occlusion. This disease state is difficult to mimic ex vivo, and the data on the safety and feasibility of intra-arterial techniques cannot be obtained from non-animal studies. To date, there have been large numbers of animal stroke model trials that showed promising results for future therapeutic techniques, but they failed to translate into successful treatments in humans. [13] Per the recommendations from the Stroke Therapy Academic Industry Roundtable (STAIR), testing in a large-animal species with anatomy that more closely resembles that of humans shows promise in addressing this failed translational gap. [14]
Canines exhibit a number of characteristics that render them one of the ideal species for use after small-animal studies, but before beginning human clinical trials. Canines have a gyrencephalic brain with a proportion of gray/white matter that more closely
Page 6
resembles that of humans than the lissencephalic brains of small animals. [15, 16] The anatomical disparity between rodent and human brains makes rodent stroke model research difficult to translate to humans.
The neurovascular anatomy of canines enables their arteries to be catheterized in a similar manner to humans under angiographic guidance, allowing for minimally invasive endovascular stroke induction and stem cell delivery surgeries. [12] Endovascular procedure evaluation is very limited in small-animal models due to the small size of the cervical and cranial arteries in these animals; as such, the procedure cannot be visualized in real time by angiography. Other large-animal species such as felines, ovis, and swine have a well-developed carotid rete mirabile, a fine vascular network that supplies the cerebral arteries. [17, 18] This is an obstacle when advancing endovascular catheters and devices to the cranial arteries, particularly in the middle cerebral artery (MCA), which is the target of most stroke models. In these species, it is necessary to perform invasive craniotomy to access the MCA. [19] Structures such as the eye (which often is enucleated), temporalis muscle, and zygomatic arch suffer damage under this method. [20] Other disadvantages include the fact that the skull needs to be opened, ultimately affecting cerebral pressures; moreover, the dura mater is breached and cerebrospinal fluid is lost.
NHP are an ideal model for ischemic stroke research due to their close relation to humans and the similarity of their brains. Table 1 summarized the differences in utilizing canine and NHP as stroke models. The disadvantages of using NHP models
Page 7
include the costly maintenance fees, and the use of NHPs also raises more ethical concerns than the use of canines. [9, 21, 22] Even so, medical experiments on canines still carry major ethical considerations in the current era, as canines are considered “man’s best friend”, and society expresses concerns about the usage of canines when compared to other animals, such as rodents. [23]
3. CANINE CEREBRAL ARTERY SUPPLY
The canine brain is supplied by a cerebral arterial circle system, consisting of anterior and posterior circulation. [25] The anterior circulation consists of the anterior/rostral cerebral arteries (ACA), anterior communicating artery, internal carotid arteries (ICA), and MCA. The ICA is approximately half the size of the external carotid artery (ECA). The ICA enters the skull through the posterior foramen lacerum, travels through the carotid canal, and continues to be surrounded by the cavernous sinus, where it is joined by the ramus anastomicus, a branch of the maxillary artery. Beyond this point, the ICA forms an Sshaped curve and joins the cerebral arterial circle. Approaching the MCA via the ICA is relatively difficult due to the tortuosity of the canine ICA. [12] The canine MCA also receives more significant leptomeningeal circulation from the ipsilateral ACA and posterior/caudal cerebral artery (PCA), and significant maxilla–carotid collateral circulation is also present in canines when compared to humans, which is one difficulty that is faced when trying to repeat lesion sizes. [26]
Page 8
The posterior circulation is formed by the basilar artery (BA) caudally and the PCA. [25] The BA is formed after the vertebral arteries (VA) fuse at the level of the first cervical nerve origin. The canine BA has a uniform diameter throughout and is straight enough to accommodate endovascular devices, which is why this route is used more commonly than the anterior circulation to reach the MCA. Superior/Rostral cerebellar arteries and inferior/caudal cerebellar arteries both branch from the BA with some variations. Figure 1 shows the canine cerebral arterial supply.
The most common canine breeds used in canine endovascular stroke models are mongrels (weight range: 13–37 kg) or beagles (weight range: 9–15 kg). Many factors should be considered when choosing canine breed and size for an endovascular stroke model. Smaller canines have many disadvantages; for instance, it is more difficult to gain femoral access, and these animals also have a smaller caliber of intracranial vasculature. Conversely, larger canines can be more costly.
4. METHODS TO CREATE ENDOVASCULAR STROKE MODELS
4.1 Middle Cerebral Artery Occlusion Model Models of focal ischemia generally include permanent MCA occlusion (pMCAo) and transient MCA occlusion (tMCAo) techniques. pMCAo, or non-reperfusion models, can simulate an acute ischemic stroke without reperfusion, and it serves as the ideal model for studying recanalization, as well as reperfusion techniques involving endovascular instrument and procedures. tMCAo, or recanalization models, will simulate endovascular
Page 9
recanalization after an acute ischemic stroke; additionally, it could serve to test future therapies following reperfusion. The default study model recommended from the STAIR committee is the pMCAo; the exception to this rule is the use of tMCAo, which serves as a model to study possible reperfusion treatments – including stem cell therapy – where previous studies have found that the cells likely approach the infarcted area through the ipsilateral MCA, and no cell engraftment is seen in subjects with occluded MCA. [14, 27] Minimally invasive endovascular techniques for MCA occlusion (MCAo) in canines may be performed via multiple methods, including emboli injection via ICA for permanent MCAo and coil embolization of the MCA for both permanent and transient MCAo. [11, 12]
The two conventional routes used to access the MCA via an endovascular approach are the ICA or VA. Approaching the MCA via the ICA is relatively difficult due to the tortuosity of the petrous loop, which differs considerably between subjects and needs an experienced interventionalist to perform this procedure. Also, the size of the ICA is difficult to accommodate, even using small-diameter microguidewire (0.010–0.014 inches) and microcatheter systems. [12] The ICA route is mostly utilized for emboli injection, as the catheter used for the injection is placed a few centimeters above the origin of the ICA; it does not need to go through the tortuous sections of the ICA. The VA system tends to be utilized more frequently as a direct access point where the catheter is placed into the MCA. Furthermore, the BA is adequately straight and easier to navigate than the ICA. Potential adverse events include the tendency of mechanical vasospasm, which leads to permanent occlusion and results in a larger than expected infarction;
Page 10
however, this could be prevented by administering a vasodilating agent. Also, arterial perforation can occur during coil deployment. [28] 4.1.1
Internal Carotid Artery Emboli Injection for MCA Occlusion
The MCA is considered to be the natural continuation of the ICA; thus, an optimally sized embolus injection via the ICA can occlude the MCA. [29] This method is performed by an endovascular approach or via direct puncture of the ICA to acquire arterial access. The catheter is targeted towards 2 cm of the ascending part of the ICA using digital subtraction angiography. Emboli are injected through a syringe, which is directly connected to the catheter and flows through the ICA to occlude the MCA. The most commonly used emboli are autologous clots, created from the subject’s blood mixed with bovine thrombin. [11, 30, 31] Other types of emboli are synthetic contents, such as silk suture portions. [32] Autologous clot embolus has the advantage of being amenable to study the effects of pharmaceutical or mechanical thrombolysis. The injection of a single autologous clot 1.7 mm in diameter and 5 mm long has been proven to establish complete proximal MCAo with 100% efficiency (20/20 beagles) in 2 studies. [27, 31] This clot size was chosen due to the fact that the M1 segment in beagles is approximately 1 mm in diameter and 13 mm long. [31] MCAo with a clot 1.7 mm in diameter and 5 mm long resulted in acute stroke at the subcortical area (caudate, internal capsule), resembling a lacunar infarct with small average stroke volumes on magnetic resonance imaging (MRI) at 24 hours (150.01±153.61 mm3). [31] Larger clots with an approximate diameter of 2.33 mm have been shown to occlude the distal ICA with extension into the proximal MCA. Stroke volumes at 4 hours in each subject vary between 120–12,530 mm3. [33] The major drawback seen with an embolus injection MCAo model is that the precise site
Page 11
of the occlusion cannot be controlled. The infarct size resulting from complete MCAo is also relatively small compared to human MCAo given the extensive leptomeningeal collateral circulation in canines. [26]
Additional ipsilateral ICA transient occlusion has been proposed to reduce leptomeningeal collateral circulation in an emboli injection MCAo model to create larger infarct volumes. This can be accomplished by keeping the catheter in the ICA for an additional period to occlude blood flow. MCAo with an autologous clot embolus 1.7 mm in diameter in addition to 2 hours of ICA occlusion achieved an average stroke volume of 599.45±262.50 mm3 at 24 hours. [27] In one study, two autologous clots of 1.4 and 1.7 mm in diameter were injected consecutively to occlude the distal M1 and proximal M1, respectively, with occlusion of ICA blood flow for 2 hours using same technique as described above. This method was able to create larger MCA territory stroke with an average stroke volume of 4,173.23±603.92 mm3 at 24 hours. [30]
An alternative approach to create larger stroke volume can be performed by clot injection via the ICA terminus to occlude the MCA and ACA. Access through the VA system is preferred to bypassing the tortuous ICA. Boulos et al. injected 1 mL of autologous clot through the ICA terminus, and they were able to accomplish MCA and ACA occlusion, where the average stroke volume was 9,540±1,800 mm3. MCA, ACA, and PCA occlusion were seen in some subjects with a mean stroke volume of 15,870±3,840 mm3. [34] Figure 2 shows the ICA emboli injection method for MCAo.
Page 12
The disadvantage of this technique is that it may not be reproducible in different dog breeds that feature smaller and more tortuous ICAs. This model was also utilized to evaluate for mesenchymal stem cell delivery via intra-arterial method in one study. [27] 4.1.2
Coil Occlusion for MCA Occlusion
A soft platinum coil used to treat intracranial aneurysms can be utilized for a canine tMCAo model. The coil can be precisely positioned by advancing the microcatheter to the proximal M1, and it can be repositioned until appropriate complete MCAo is achieved. This precise positioning method cannot be achieved by emboli injection via ICA. Transient occlusion is also possible without detaching the coil, and by retrieving it after a selected duration. The most common route to access the MCA is via the VA and by gaining access into the cerebral arterial circle through the BA. ICA access is generally not performed due to tortuosity, as mentioned. Even though posterior circulation access has been well described, it might not be recommended for use in every dog breed. To avoid deadly complications, such as vessel perforation, it is imperative for the surgeon to consider the anatomical variations between each breed; in some cases, access via the ICA maybe a better choice. In multiple reported MCAo models, the platinum coil was delivered into the MCA, including the entire M1 segment. Angiography was repeated every 15 minutes to confirm complete occlusion, and the total occlusion time was 60 minutes; the coil was then retrieved. This method was able to achieve a hemispherical infarct volume of 30.9%±2.1%. [12, 28, 35] In one study, the coil was positioned to occlude the proximal MCA, ACA, and distal ICA with the aim of creating a larger stroke; however, in 3/8 of the animals faced premature mortality within 24 hours. [28] Figure 3
Page 13
shows the coil occlusion method for MCAo. Table 2. summarized the comparison of Ischemic Stroke Size by Each MCA Occlusion Techniques.
4.2 Vertebral Artery Occlusion Model Autologous clot injection is a potential method that is used to create VA occlusion in a canine model. The obstacle to this is that the canine VA features strong blood flow and it is difficult for the clot to remain anchored. A recent study proposed the usage of a selfexpanding thrombus filter preinstalled in the delivery catheter. A clot 3 mm × 20 mm in size was injected through a catheter that was proximally positioned 2 cm from the thrombus filter; the filter was left for 15 minutes to allow implantation of the clot. This resulted in successful complete occlusion of the VA in 12/12 subjects. [36]
This model may be useful to further evaluate novel endovascular therapy for acute VA occlusion in humans. Further imaging and clinical outcome studies of this canine model are still needed.
4.3 Basilar Artery Occlusion Model This model was created to test the effectiveness of intraarterial tPA therapy versus intravenous tPA therapy for acute basilar occlusion. Autologous clot injection is utilized. Two clots were injected consecutively via a microcatheter placed in the proximal portion of the BA. All 13 subjects showed successful BA occlusion; 11 subjects had proximal BA occlusion and 2 subjects had mid-BA occlusion. In this study, the subjects were used to test recanalization by intravenous versus intraarterial thrombolysis, and they were
Page 14
humanely killed 6 hours after therapy. [37] Figure 4 shows the BA occlusion method that was used.
The disadvantages of this technique include the fact that it poses major ethical challenges and more extensive care of the canines is required post-procedure. Complete BA occlusion will likely cause a severe and fatal neurological deficit, potentially leading to canine suffering, which may require prolonged sedation to achieve the desired survival time.
5. BEHAVIORAL ASSESSMENT The National Institutes of Health Stroke Scale (NIHSS) is widely used as a tool to quantify the level of stroke impairment, and it has been shown to strongly predict outcomes after stroke. [38] Many research groups have developed a stroke scale similar to the NIHSS to assess canine neurological function by observing canine behavior following the induction of stroke. [28, 34, 39] Boulos et al. proposed a canine stroke score ranging from 0 (normal) to 18 (comatose). [34] This score assesses the canine’s level of consciousness, vocalization, gait, behavior, motor function, and sensory function. In their study, this score was correlated with infarct volume. The proposed canine stroke scale could potentially be used as a clinical assessment tool for successful stroke model creation, or as a clinical indicator of improvement with stroke therapy trials. Assessment of canine post-stroke induction should be performed at multiple times. This should include early assessment following recovery from anesthesia, as many canines may experience rapid recovery soon after.
Page 15
Myriad other factors must also be considered for canine recovery post-stroke induction, including stroke location, duration of the occlusion, and time under anesthesia. It is important to mention that during the first 6–8 hours, canines should be under intensive care with strict monitoring of vital signs and fluid resuscitation as part of their further recovery and stroke assessment. A post-procedure care protocol should be standardized, as this may impact future neurological recovery and assessment.
6. DISCUSSION
With the advancements in digital subtraction angiography technology, the less invasive endovascular method (as compared to the surgical approach) has been preferred for stroke model creation in recent years. The guided fluoroscopy approach also allows for real-time visualization of MCAo. One of the earliest reported endovascular canine stroke models, where microfibrillar collagen is injected and delivered via femoral catheterization, has shown success in creating a canine stroke model without the need for surgery, which involves cervical and intracranial vasculature. [40] Since then, multiple techniques for endovascular canine stroke model creation have been developed.
An emboli injection technique is the most preferred for a pMCAo model. Drawbacks include difficulty controlling the site of the emboli occlusion. Multiple studies suggested that using a single clot 1.7 mm in diameter and 5 mm long successfully occludes the proximal MCA with a high success rate. [27, 31] This should provide a more predictable
Page 16
occlusion at the proximal MCA than using multiple random sized clot injections. Even then, we should assume that the proximal MCA size is slightly different between each subject and breed; thus, 100% successful occlusion in all subjects may not be achieved.
A transient MCAo or recanalization model is ideal for studying the effect of future treatments that require proper reperfusion of the occluded cerebral vessel in order to be effective, which includes cell-based therapy. In recent years, the study of regenerative medicine in ischemic stroke patients has shown tremendous advancements, with attempts to find future novel therapies to reduce stroke morbidity. The use of stem cells has been a topic of interest in many research groups, and a tMCAo canine model by coil occlusion is a suitable point of focus. One major limitation noted in literature reviews, and from our own experience, in the use of this tMCAo canine model at our institution includes the inability to create a reliable infarct size. This is mainly due to the variability in the arterial anatomy between subjects, even within the same age range. [12, 28, 35] Brain infarct volume is reported to be a major predictor of stroke outcome, and the reduction of infarct volume is a major primary outcome in stem cell therapy studies. [41, 42] Larger scale prospective studies of a novel method to create a canine tMCAo models with reliable stroke size are needed to further study the efficacy of stem cell therapy.
Although there are multiple advantages of choosing canines over other species for producing preclinical ischemic stroke models, notable disadvantages and challenges are seen in the literature. Canine MCA has significant maxilla–carotid collateral circulation and higher amounts of leptomeningeal collateral circulation from the ipsilateral ACA and
Page 17
PCA when compared to humans. [26] If a larger stroke size is desired, additional methods to minimize collateral flow is recommended by transient occlusion of the ICA. [27] Supplementary ipsilateral ACA and/or PCA occlusion can also increase infarct size. [34] In both a transient and permanent MCAo model, infarct volume is difficult to control and varies extensively in each canine, even though the same technique is used throughout. This is likely due to the different extent in collateral circulation between each canine subject. In subjects with extensive infarct size, premature mortality can occur. Assessment of pial collateral recruitment by digital subtraction angiography prior to recanalization in the tMCAo model is possible and may be able to predict outcomes and final infarct volumes based on the extent of the visualized leptomeningeal collaterals. [28] Multiple factors can influence the leptomeningeal collaterals, including systemic blood pressure, which affects flow through the anastomoses. [43, 44] Modifying the systemic blood pressure during stroke induction may be a possible method to reduce collateral flow. Permanent occlusion of the proximal ACA may be able to create a larger stroke size and limit collateral circulation, which can be performed prior to tMCAo with a retrievable coil. Further studies are needed to prove these hypotheses.
Large-animal models, including canines, are associated with high purchase costs, which are generally related to the breed and age; these animals also include high accommodation fees. [45, 46] Due to limitations in research funding, fewer canine subjects can be studied when compared to using small-animal models. Even with costly maintenance in canine subjects, NHP are generally associated with higher costs and more ethical considerations. Choosing an endovascular approach for a stroke model also
Page 18
requires an animal research facility that can accommodate digital subtraction angiography, which may not be available at every institution. There is also a steep learning curve for surgeons to navigate canine cerebral vasculature, particularly among those practitioners with prior training in human neuro–endovascular procedures, due to the significant differences mentioned in our review. Thus, surgical access through the cervical area for an ICA clot injection is still preferred by some research groups, particularly in recent years. [39, 47]
7. CONCLUSIONS
Canine has gyrencephalic brain similar to human brain and accessible neurovascular anatomy for stroke model creation. Canine stroke model is widely utilized to understand the disease process of stroke and to develop novel treatment. Less invasive endovascular internal carotid emboli injection and coil embolization methods can be used to simulate transient or permanent MCA occlusion. Major restriction includes the extensive collateral circulation of canine cerebral arteries that can limit the stroke size. Transient ICA occlusion can decrease collateral circulation and increase stroke size to some degree. Additional method of manipulating the extent of collateral circulation needs to be studied. Other types of canine stroke models, including VA occlusion and BA occlusion, can also be accomplished by endovascular thrombi injection.
ACKNOWLEDGEMENTS
Page 19
We are grateful to Aisha Khan, Ph.D. and team at University of Miami Interdisciplinary Stem Cell Institute to allow the use of digital subtraction angiography images.
Page 20
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
Feigin, V.L., et al., Global and regional burden of stroke during 1990-2010: findings from the Global Burden of Disease Study 2010. Lancet, 2014. 383(9913): p. 245-54. Gounis, M.J., A.K. Wakhloo, and J.Y. Chueh, Preclinical investigations for thrombectomy devices--does it translate to humans? Stroke, 2013. 44(6 Suppl 1): p. S7-S10. Saver, J.L., et al., Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke. N Engl J Med, 2015. 372(24): p. 2285-95. Cook, D.J. and M. Tymianski, Nonhuman primate models of stroke for translational neuroprotection research. Neurotherapeutics, 2012. 9(2): p. 371-9. Culp, W.C., et al., Three variations in rabbit angiographic stroke models. J Neurosci Methods, 2013. 212(2): p. 322-8. Duberstein, K.J., et al., Gait analysis in a pre- and post-ischemic stroke biomedical pig model. Physiol Behav, 2014. 125: p. 8-16. Boltze, J., et al., Permanent middle cerebral artery occlusion in sheep: a novel large animal model of focal cerebral ischemia. J Cereb Blood Flow Metab, 2008. 28(12): p. 1951-64. Hayakawa, T. and A.G. Waltz, Immediate effects of cerebral ischemia: evolution and resolution of neurological deficits after experimental occlusion of one middle cerebral artery in conscious cats. Stroke, 1975. 6(3): p. 321-7. Coors, M.E., et al., The ethics of using transgenic non-human primates to study what makes us human. Nat Rev Genet, 2010. 11(9): p. 658-62. Hill, N.C., et al., Studies in cerebrovascular disease. VII. Experimental production of cerebral infarction by intracarotid injection of homologous blood clot; preliminary report. Proc Staff Meet Mayo Clin, 1955. 30(26): p. 625-33. Shaibani, A., et al., First results in an MR imaging--compatible canine model of acute stroke. AJNR Am J Neuroradiol, 2006. 27(8): p. 1788-93. Rink, C., et al., Minimally invasive neuroradiologic model of preclinical transient middle cerebral artery occlusion in canines. Proc Natl Acad Sci U S A, 2008. 105(37): p. 14100-5. Kahle, M.P. and G.J. Bix, Successfully Climbing the "STAIRs": Surmounting Failed Translation of Experimental Ischemic Stroke Treatments. Stroke Res Treat, 2012. 2012: p. 374098. Stroke Therapy Academic Industry, R., Recommendations for standards regarding preclinical neuroprotective and restorative drug development. Stroke, 1999. 30(12): p. 2752-8. Ferrer, I., I. Fabregues, and E. Condom, A Golgi study of the sixth layer of the cerebral cortex. II. The gyrencephalic brain of Carnivora, Artiodactyla and Primates. J Anat, 1986. 146: p. 87-104. Wynshaw-Boris, A., et al., Lissencephaly: mechanistic insights from animal models and potential therapeutic strategies. Semin Cell Dev Biol, 2010. 21(8): p. 823-30.
Page 21
17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34.
Gillilan, L.A., Extra- and intra-cranial blood supply to brains of dog and cat. Am J Anat, 1976. 146(3): p. 237-53. Gralla, J., et al., A dedicated animal model for mechanical thrombectomy in acute stroke. AJNR Am J Neuroradiol, 2006. 27(6): p. 1357-61. Wells, A.J., et al., A surgical model of permanent and transient middle cerebral artery stroke in the sheep. PLoS One, 2012. 7(7): p. e42157. Howells, D.W., et al., Different strokes for different folks: the rich diversity of animal models of focal cerebral ischemia. J Cereb Blood Flow Metab, 2010. 30(8): p. 1412-31. Tong, F.C., et al., An Enhanced Model of Middle Cerebral Artery Occlusion in Nonhuman Primates Using an Endovascular Trapping Technique. AJNR Am J Neuroradiol, 2015. 36(12): p. 2354-9. Quigley, M., Non-human primates: the appropriate subjects of biomedical research? J Med Ethics, 2007. 33(11): p. 655-8. Hasiwa, N., et al., Critical evaluation of the use of dogs in biomedical research and testing in Europe. ALTEX, 2011. 28(4): p. 326-40. de Crespigny, A.J., et al., Acute studies of a new primate model of reversible middle cerebral artery occlusion. J Stroke Cerebrovasc Dis, 2005. 14(2): p. 807. Budras, K.-D., et al., Anatomy of the dog : an illustrated text. 4th ed. 2002, Hannover: Schlütersche. 217 p. Symon, L., Observations on the leptomeningeal collateral circulation in dogs. J Physiol, 1960. 154(1): p. 1-14 2. Lu, S.S., et al., In vivo MR imaging of intraarterially delivered magnetically labeled mesenchymal stem cells in a canine stroke model. PLoS One, 2013. 8(2): p. e54963. Christoforidis, G.A., et al., An endovascular canine middle cerebral artery occlusion model for the study of leptomeningeal collateral recruitment. Invest Radiol, 2011. 46(1): p. 34-40. Purdy, P.D., et al., Reversible middle cerebral artery embolization in dogs without intracranial surgery. Stroke, 1989. 20(10): p. 1368-76. Zu, Q.Q., et al., An endovascular canine stroke model: middle cerebral artery occlusion with autologous clots followed by ipsilateral internal carotid artery blockade. Lab Invest, 2013. 93(7): p. 760-7. Liu, S., et al., A novel embolic stroke model resembling lacunar infarction following proximal middle cerebral artery occlusion in beagle dogs. J Neurosci Methods, 2012. 209(1): p. 90-6. Yasuda, R., et al., C-arm CT measurement of cerebral blood volume using intraarterial injection of contrast medium: an experimental study in canines. AJNR Am J Neuroradiol, 2012. 33(9): p. 1696-701. van der Bom, I.M., et al., Quantitative evaluation of C-arm CT cerebral blood volume in a canine model of ischemic stroke. AJNR Am J Neuroradiol, 2012. 33(2): p. 353-8. Boulos, A.S., et al., Tamoxifen as an effective neuroprotectant in an endovascular canine model of stroke. J Neurosurg, 2011. 114(4): p. 1117-26.
Page 22
35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47.
Rink, C., et al., Tocotrienol vitamin E protects against preclinical canine ischemic stroke by inducing arteriogenesis. J Cereb Blood Flow Metab, 2011. 31(11): p. 2218-30. Zhang, Y., et al., A Novel Canine Model of Acute Vertebral Artery Occlusion. PLoS One, 2015. 10(11): p. e0142251. Qureshi, A.I., et al., Randomized comparison of intra-arterial and intravenous thrombolysis in a canine model of acute basilar artery thrombosis. Neuroradiology, 2004. 46(12): p. 988-95. Adams, H.P., Jr., et al., Baseline NIH Stroke Scale score strongly predicts outcome after stroke: A report of the Trial of Org 10172 in Acute Stroke Treatment (TOAST). Neurology, 1999. 53(1): p. 126-31. Kang, B.T., et al., Canine model of ischemic stroke with permanent middle cerebral artery occlusion: clinical and histopathological findings. J Vet Sci, 2007. 8(4): p. 369-76. Purdy, P.D., et al., Microfibrillar collagen model of canine cerebral infarction. Stroke, 1989. 20(10): p. 1361-7. Li, Y., et al., Treatment of stroke in rat with intracarotid administration of marrow stromal cells. Neurology, 2001. 56(12): p. 1666-72. Vogt, G., et al., Initial lesion volume is an independent predictor of clinical stroke outcome at day 90: an analysis of the Virtual International Stroke Trials Archive (VISTA) database. Stroke, 2012. 43(5): p. 1266-72. Brozici, M., A. van der Zwan, and B. Hillen, Anatomy and functionality of leptomeningeal anastomoses: a review. Stroke, 2003. 34(11): p. 2750-62. Symon, L., S. Ishikawa, and J.S. Meyer, Cerebral arterial pressure changes and development of leptomeningeal collateral circulation. Neurology, 1963. 13: p. 237-50. Davis, P.R. and E. Head, Prevention approaches in a preclinical canine model of Alzheimer's disease: benefits and challenges. Front Pharmacol, 2014. 5: p. 47. Chu, C.R., M. Szczodry, and S. Bruno, Animal models for cartilage regeneration and repair. Tissue Eng Part B Rev, 2010. 16(1): p. 105-15. Kang, B.T., et al., Three-dimensional time-of-flight magnetic resonance angiography of intracranial vessels in a canine model of ischemic stroke with permanent occlusion of the middle cerebral artery. Comp Med, 2009. 59(1): p. 72-7.
Page 23
Figure 1 shows canine cerebral arterial supply by illustration and digital subtraction angiography imaging. ACA-anterior cerebral artery, MCA-middle cerebral artery, ICAinternal carotid artery, BA-basilar artery, VA-vertebral artery.
Figure 2 shows the internal carotid artery emboli injection method for middle cerebral artery occlusion. Catheter is targeted at 2 cm in the ascending part of ICA for injection. ICA-internal carotid artery, CCA-common carotid artery, ECA-external carotid artery.
Figure 3 shows the coil occlusion method for middle cerebral artery occlusion. MCA is accessed via the vertebrobasilar system and coil is delivered to occlude the entire M1 segment. ACA-anterior cerebral artery, MCA-middle cerebral artery, BA-basilar artery, VA-vertebral artery.
Figure 4 shows basilar artery occlusion method. Microcatheter is placed in the proximal BA for injection of autologous clots. ACA-anterior cerebral artery, MCA-middle cerebral artery, BA-basilar artery, VA-vertebral artery.
Table 1. Differences in utilizing canine and NHP as stroke models. ICA = internal carotid artery, MCA = middle cerebral artery, ECA = External carotid artery, NHP = NonHuman Primate
Table 2. Comparison of Ischemic Stroke Size by Each Middle Cerebral Artery Occlusion Techniques.
Page 24
Table 1. Differences in utilizing canine and NHP as stroke models. ICA = internal carotid artery, MCA = middle cerebral artery, ECA = External carotid artery, NHP = NonHuman Primate Canine Breed, Species - Mongrel - Beagle
Non-Human Primate - Rhesus [21] - Cynomolgus [24]
ICA
- Smaller luminal diameter than human - ECA is larger - High tortuosity - MCA inaccessible by most endovascular methods via ICA route (Many endovascular thrombectomy devices cannot be studied)
- Smaller luminal diameter than human - Branching similar to human - MCA accessible via ICA route (can study thrombectomy devices via this route)
MCA
- Smaller luminal diameter than human (similar to NHP) - Extensive maxillo-carotid anastomosis - Stroke size less predictable
- Smaller luminal diameter than human (similar to canine) - More similar to human
- ICA thrombus injection - Reversible Coil Occlusion
- ICA thrombus injection - Endovascular trapping technique [21] - Reversible Coil/Balloon Occlusion
- Lower than NHP
- Higher than canine
- Lower than NHP
- Higher than canine
Cerebral Vasculature Collateral Anastomosis Endovascular MCA occlusion Method Financial Burden Ethical consideration
Page 25
Table 2 Comparison of Ischemic Stroke Size by Each Middle Cerebral Artery Occlusion Techniques. Stroke Size ICA Emboli Injection - clot 1.7mm - clot 2.3mm - clot 1.7mm + transient ICA occlusion - clot 1.4mm and 1.7mm + transient ICA occlusion Coil Occlusion - 60 min occlusion
150.01±153.61 mm3 120-12530 mm3 599.45±262.50 mm3 4173.23±603.92 mm3
30.9±2.1% hemispherical infarct volume
Page 26