Review Article
* These authors contributed equally
This review systematically compares experimental CVST animal models, evaluating induction methods, species, reversibility, severity, feasibility, reproducibility, limitations, and translational value to guide rational selection among existing models.
Cerebral venous sinus thrombosis (CVST) is a rare but life-threatening cerebrovascular disorder for which animal models are essential for mechanistic and translational research. This review systematically synthesizes experimental CVST models across species and induction techniques using predefined criteria, including thrombus composition, reversibility, severity, reproducibility, technical feasibility, and clinical relevance. The successful establishment of CVST models is defined by direct evidence of sinus occlusion, with parenchymal lesions considered optional. Thrombosis induced by FeCl₃ or photochemical methods is generally reversible, whereas models based on permanent ligation, autologous clot injection, balloon catheter combined with thrombin, or water-swellable rubber result in irreversible occlusion.
Most models induce mild pathology, while severe phenotypes usually require combined interventions such as balloon catheter inflation with thrombin, complete ligation with thrombin, or partial ligation with thrombin injection and FeCl₃ application. Ligation-only models exhibit low reproducibility, whereas photochemical and balloon catheter-based models are limited by specialized equipment and technical expertise requirements. In terms of translational relevance, large-animal balloon catheter models combined with thrombogenic agents and rodent clot-injection models are most clinically applicable. Thrombolysis or recanalization studies should preferentially employ reversible models while avoiding permanent ligation or rigid occlusive devices. Water-swellable rubber-based models are particularly suitable for studying slowly progressive forms of CVST. Intravascular device validation studies should prioritize swine models using balloon catheters in combination with thrombogenic agents. In addition, developing etiology-specific models may facilitate the preclinical evaluation of targeted, mechanism-based therapeutic approaches, while non-human primate models may further enhance translational relevance.
Cerebral venous sinus thrombosis (CVST) is a rare but potentially life-threatening cerebrovascular disorder characterized by thrombosis in dural venous sinuses or cerebral veins1. Unlike arterial stroke, CVST predominantly affects young adults, especially women of childbearing age, and exhibits diverse clinical presentations, ranging from headache and papilledema to seizures, focal neurological deficits, and venous infarction2,3. The underlying pathophysiology is multifactorial, involving venous stasis, endothelial injury, and hypercoagulability. Additional contributing factors include local mechanical obstruction, inflammatory disorders, and systemic prothrombotic conditions4,5,6.
Despite advances in imaging and anticoagulant therapy, CVST continues to exhibit substantial morbidity and mortality, particularly in cases of severe or extensive thrombosis7,8. Clinical investigations remain limited due to the disease's rarity and heterogeneity, making animal models essential for studying pathophysiological mechanisms, testing novel treatments, and exploring hemodynamic consequences9. Experimental CVST models enable controlled induction of thrombosis, precise timing of interventions, and systematic evaluation of thrombus resolution, recanalization, and secondary brain injury.
Over the past decades, multiple approaches have been developed to model CVST in animals, including mechanical ligation10, ferric chloride (FeCl3)-induced chemical injury11,12, photochemical thrombosis13, thrombogenic agents, or clot injection14,15, and implantable or expandable devices16,17,18. These models vary in thrombus composition, anatomical site, reversibility, severity, and species suitability, ranging from genetically tractable mice to anatomically comparable swine19. However, most existing models fail to replicate the complex etiological spectrum of human CVST, including systemic inflammation, hypercoagulability, and local venous compression. Understanding the advantages, limitations, and translational potential of these models is essential for designing experiments that can reliably inform clinical practice.
This review provides a comprehensive overview of available CVST animal models, discussing induction techniques, species selection, classification based on thrombus characteristics, and translational considerations. It also highlights existing gaps, including the absence of disease- or etiology-specific models, and offers perspectives for developing clinically relevant experimental platforms.
Access restricted. Please log in or start a trial to view this content.
Induction techniques
Multiple induction techniques have been developed to establish experimental models of CVST. The following section summarizes the major approaches, emphasizing their advantages, limitations, and appropriate applications. These induction techniques are schematically illustrated in Figure 1.
Ligation
Ligation involves the mechanical obstruction of venous outflow, typically through partial or complete ligation of a targeted sinus, thereby inducing localized blood stasis. Although permanent ligation alone has been employed in gerbils and cats, rat CVST models are typically generated using a combination of partial or complete ligation and injection of thrombogenic agents20,21,22,23. This is because rats possess extensive collateral veins that drain into alternative sinuses, mitigating superior sagittal sinus (SSS) thrombosis and parenchymal injury. This technique is valuable for studying hemodynamic consequences of venous obstruction and parenchymal injury secondary to venous stasis. Its main limitations include an overemphasis on venous stasis, which may not accurately represent the multifactorial etiology of clinical CVST, and its irreversibility, which makes it unsuitable for evaluating antithrombotic or thrombolytic therapies.
FeCl3
FeCl3 induces thrombosis by causing oxidative endothelial injury at the exposed sinus wall. The resulting endothelial denudation triggers platelet adhesion, aggregation, and coagulation, forming rapid and reproducible thrombi11. FeCl3-induced thrombosis has been widely applied in rodents and rabbits11,24. However, this method directly damages the endothelium, which may not accurately reproduce clinical triggers of human CVST, such as inflammation, hypercoagulability, or venous compression. In addition, FeCl₃-induced thrombosis undergoes spontaneous recanalization, rendering this model unsuitable for long-term studies. Despite these limitations, the FeCl3 model remains advantageous for investigations of pathophysiological processes downstream of CVST and for evaluating antithrombotic or thrombolytic therapies.
Photochemical thrombosis
Photochemical models, most commonly using Rose Bengal dye activated by green light, generate platelet-rich thrombi in selected surface vessels25. Mechanistically, illumination of Rose Bengal produces reactive oxygen species that cause localized endothelial injury and oxidative stress, exposing subendothelial matrix proteins that promote platelet adhesion, aggregation, and coagulation25. This approach allows precise spatial control of thrombus location and size, and in some cases, partially reversible thrombi can be created. Photochemical models are particularly useful for simulating cortical vein thrombosis, or SSS thrombosis, when combined with thrombin, thereby enabling detailed investigation of localized hemodynamic alterations and microvascular responses13. This model can be applied independently only to cortical veins and is not suitable for inducing thrombosis in major venous sinuses. The resulting injury is not strictly confined to the targeted veins, as surrounding vessels may also be exposed to light or affected by the diffusion of reactive oxygen species. Thrombosis is induced primarily through direct endothelial injury, which does not fully recapitulate the pathophysiology of spontaneous venous sinus thrombosis.
Thrombogenic agents or clot injection
Direct intraluminal administration of thrombin or thromboplastin into the venous sinus, causing rapid clot formation characterized by distinct fibrin and platelet composition24,26. Injection of a kaolin-cephalin suspension into the SSS also induces thrombosis formation in rats and rabbits27,28. However, the injection of thrombogenic agents is rarely used alone and is more commonly combined with other methods, such as ligation, FeCl₃ application, or photochemical thrombosis13,23,27. Autologous or pre-formed clot injection can also be utilized in mice and swine15,29. In this approach, prepared clots are introduced directly into the sinus lumen, generating thrombi closely resembling naturally formed clots with erythrocyte- and fibrin-rich regions29. The bilateral transverse sinus or jugular vein was subsequently ligated. Thrombogenic agents or clot injection are particularly suitable for assessing thrombolytic therapies, as clot size, composition, and location can be standardized. Limitations include the risk of distal embolization and technical challenges in achieving precise clot placement.
Implantable devices
Implant-based rat models represent an important but previously underemphasized category of experimental venous thrombosis models. These models typically induce thrombosis by placing intravascular implants, leading to localized flow disturbance and thrombus formation, rather than by direct vessel ligation or chemically induced endothelial injury. Devices designed for implantation or expansion, including plastic grafts, silicone rods, balloon catheters, and water-swellable polymers, induce venous obstruction by occupying the sinus lumen and exerting a mechanical mass effect14,16,17,18. Mu et al.16 reported a reproducible and irreversible rat model of CVST established by inserting a water-swellable rubber device into the SSS. Following implantation, the water-swellable polymer gradually expands within the sinus lumen, thereby simulating slow-onset and progressive venous obstruction. A silicone rod with a diameter of 1.2 mm was inserted into the SSS until its posterior edge was confluent with the sinuses, thereby inducing venous outflow obstruction17. Similarly, a plastic graft consisting of a conical anterior segment with a maximum diameter of 0.12 cm and a gradually tapering and flattened posterior segment with a width of 0.2 cm and a length of 0.1 cm was placed within the SSS to induce venous outflow obstruction18. Balloon catheters can be positioned and inflated in the venous sinuses to create temporary or gradually intensifying luminal narrowing. They are often used in combination with thrombin or other pro-thrombotic agents to promote thrombus formation in swine and canine models14,30. Such implant-based models should be distinguished from traditional ligation or FeCl3-induced models, as their mechanisms of thrombogenesis and translational implications differ substantially.
Species considerations
Mouse models
In mice, CVST models are predominantly created in the superior sagittal sinus (SSS), though cortical veins may also be affected. Several induction approaches are available. Photochemical injury with Rose Bengal and green light induces platelet-enriched thrombi in the SSS or superficial veins, offering precise spatial regulation and partial reversibility31. FeCl3 application generates oxidative endothelial injury along the sinus wall, producing consistent and reproducible thrombi32,33. Alternatively, injection of prepared clots into the SSS quickly forms intraluminal blockages29. The mouse model's key advantage is its genetic manipulability and the wide availability of knockout strains, which facilitate detailed investigations of immune and inflammatory processes. However, the small size of murine sinuses poses technical difficulties, complicating surgery and in vivo imaging.
Rat models
In rats, CVST typically affects the SSS, although the transverse sinuses may also be involved. Multiple induction strategies exist. Topical FeCl3 exposure on the exposed SSS results in endothelial damage and thrombosis11. Direct injection of thrombogenic agents into the SSS rapidly produces fibrin-rich thrombi, allowing reproducible results34. Partial or complete ligation combined with injection of thrombogenic agents, with or without the adjunctive application of FeCl₃, induces more extensive multi-sinus obstruction23,35. Photochemical thrombosis, alone or paired with thrombin, can lead to SSS occlusion13,36. Moreover, the insertion of a water-swellable polymer device or a plastic graft into the SSS can induce prolonged venous outflow obstruction, enabling the study of chronic impairments16,18. In addition, the insertion and removal of a silicone rod can achieve acute occlusion and subsequent recanalization of the SSS17. Compared with mice, rats possess a larger sinus diameter, improving surgical accessibility and enabling detailed imaging such as MRI or MRV. However, the limited availability of genetic tools constrains mechanistic studies.
Gerbil models
Gerbils have also been utilized to model CVST, providing a complementary small-animal platform. In these models, stasis is typically induced by ligating the posterior segment of the SSS22,37. In animals with fewer bridging veins, SSS occlusion resulted in a marked reduction in regional cerebral blood flow and hemoglobin oxygen saturation, whereas individuals with a greater number of bridging veins maintained relatively better blood flow22. These findings suggest that inter-individual anatomical variations can significantly influence cerebral hemodynamics, potentially reducing the reproducibility and stability of experimental outcomes in this model.
Rabbit models
Rabbit CVST models typically involve the transverse sinus, although the sigmoid sinus is sometimes targeted. Induction methods include FeCl3 application to the sinus wall, causing local endothelial disruption and clot formation, as well as direct injection of thrombin or kaolin-cephalin suspension into the lumen24,28. The larger venous caliber in rabbits compared with rodents supports intravascular interventions and drug administration, making them suitable for assessing anticoagulant and thrombolytic therapies. However, the lack of genetic modification tools and limited suitability for behavioral or neurological testing restricts their broader use.
Cat models
Cats have occasionally been used as a medium-sized animal model to investigate the pathophysiological consequences of CVST. To date, only one study has established a model of cerebral venous infarction by microsurgically ligating the posterior portion of the SSS in cats. In this model, occlusion of the posterior SSS produced a significant and reproducible reduction in regional cerebral blood flow, affecting approximately 45% of the brain, and histological examination revealed subacute venous infarction, thereby recapitulating key features of impaired venous drainage observed in severe CVST patients21. However, this approach represents mechanical obstruction rather than spontaneous thrombus formation, which limits its suitability for studies of thrombogenesis or thrombolysis. Variations in bridging veins may also undermine the reproducibility and stability of this model.
Swine models
In swine, CVST models typically target the SSS, and occasionally the internal jugular vein38. Several induction methods have been proposed, including balloon catheter inflation combined with endovascular administration of thrombin, thromboplastin, or fibrin glue14,26,30,39, as well as a combined approach involving autologous clot injection into the SSS and ligation of the bilateral transverse sinus15. Swine display venous architecture and hemodynamics that closely resemble those of humans, making them ideal for validating imaging modalities, testing medical devices, and evaluating translational pharmacological interventions. However, high operational costs, intensive perioperative requirements, and limited accessibility restrict their widespread application.
Canine models
In dogs, SSS thrombosis was induced using double-lumen balloon catheters inserted via each internal jugular vein and positioned at the torcular Herophili. Inflation of the balloons caused venous stasis, which was further promoted by intraluminal thrombin injection, leading to thrombus formation that could be visualized by CT40.
Non-human primate models
Although non-human primates display remarkable anatomical similarity to humans in cerebral venous structures, ethical constraints greatly hinder model development, and to date, no CVST model in non-human primates has been reported41. Researchers should carefully weigh feasibility, including the high experimental costs and limited availability of non-human primates, ethical approval, and scientific justification when considering non-human primate models.
Model classifications
Reversible versus irreversible models
Experimental CVST models are classified as reversible or irreversible, depending on whether thrombotic occlusion can be naturally or mechanically alleviated. Reversible models permit controlled restoration of blood flow, making them ideal for studying thrombus resolution, fibrinolysis, and therapeutic reperfusion. FeCl₃-induced thrombi exhibited partial spontaneous recanalization by day 711. Thrombi induced by thrombogenic agents such as thromboplastin showed spontaneous recanalization, with a recanalization rate of 40% at 4 weeks34. Although neurological deficits showed improvement by day 7, it remains unknown whether the sinus occlusion undergoes spontaneous recanalization in the photochemically induced CVST model13. Since photochemical thrombosis has been shown to spontaneously recanalize in the middle cerebral arteries, a similar reversible process may potentially occur in the CVST model, although this has yet to be experimentally confirmed42. In addition, the implantable silicone rod and plastic graft can be either left in the SSS to generate an irreversible model or removed to create a reversible model. These reversible models are valuable for investigating the timing and efficacy of anticoagulant and thrombolytic therapies, as they allow the observation of spontaneous or induced recanalization.
In contrast, irreversible models induce persistent venous sinus occlusion with minimal or no spontaneous recanalization, making them particularly suitable for studying end-stage thrombotic obstruction, secondary tissue injury, and long-term sequelae. Common examples include permanent ligation, either alone or in combination with thrombin and FeCl323. FeCl₃ application combined with thrombin injection also demonstrated a marked decrease in the recanalization rate24. Additionally, the obstruction induced by injected clots or by balloon catheters combined with thrombin is generally considered irreversible, as no spontaneous recanalization was observed during the first 7 days14,29. Water-swellable rubber likewise induces irreversible occlusion during this time frame16. However, it should be noted that the irreversibility of these occlusions has not yet been validated over longer time periods. These approaches generate stable, adherent thrombi that persist for extended periods, providing a reliable platform for investigating chronic venous congestion, infarction, and downstream neuropathological alterations. While these irreversible models are highly valuable for evaluating long-term pathophysiological outcomes of venous obstruction, they are less appropriate for studies focusing on spontaneous recanalization, fibrinolysis, or short-term thrombolytic interventions, as the thrombi are designed to remain largely permanent. The overall reversibility of these models is illustrated in Figure 2.
Mild versus severe models
Experimental CVST models can also be categorized according to the severity of thrombotic burden and the extent of parenchymal injury. Mild models, such as photochemical thrombosis, clot injection, silicone rod, plastic graft, or localized FeCl3 application, produce limited venous congestion and minimal parenchymal injury. These models are particularly useful for investigating subtle inflammatory responses, early microvascular alterations, or compensatory hemodynamic adaptations without inducing severe neurological deficits or high mortality.
In contrast, severe models involve multi-sinus thrombosis and more extensive venous occlusion, which are often achieved through a combination of balloon catheter inflation and thrombin injection, complete ligation combined with thrombin injection, photochemical thrombosis combined with thrombin injection, or partial ligation combined with thrombin injection and FeCl₃ application13,14,23,35. These approaches result in widespread venous infarction, cerebral edema, seizures, and higher mortality, providing a robust platform to study the pathophysiology of hemorrhagic transformation, severe brain injury, and aggressive therapeutic interventions. Recent rat studies have demonstrated that such combined models consistently reproduce severe phenotypes, making them highly valuable for translational investigations of high-risk CVST and for testing novel anticoagulant or thrombolytic therapies. The overall severity of these models is illustrated in Figure 3.
Validation of CVST animal model
The success of CVST animal models is primarily evaluated across two complementary domains: cerebral venous flow and parenchymal injuries. Comprehensive assessment of venous flow and parenchymal alterations ensures that CVST models accurately replicate the hemodynamic and tissue-level characteristics of human disease, thereby providing a solid foundation for model validation and translational research.
Assessment of venous sinus flow
Confirming sinus occlusion and quantifying alterations in cerebral venous hemodynamics are essential for evaluating model success. Non-invasive imaging modalities, such as magnetic resonance venography (MRV) or micro-computed tomography (CT) in rodents, enable direct visualization of thrombus presence and residual flow43,44. Laser speckle contrast imaging (LSCI) provides real-time, high-resolution assessment of relative blood perfusion in the SSS and parasinusal cortical regions, enabling dynamic monitoring of venous flow alterations45. Doppler ultrasonography enables in vivo assessment of thrombus size and blood flow velocity within the residual lumen of the SSS46. Together, these methods determine the extent, completeness, and stability of induced thrombi. Successful establishment of the CVST model requires direct evidence of sinus occlusion, characterized by luminal filling defects, interruption of venous flow, or histopathological evidence of intraluminal thrombus formation.
Assessment of brain parenchymal injury
CVST frequently leads to secondary parenchymal injury because of venous congestion, ischemia, and hemorrhagic transformation. Magnetic resonance imaging (MRI) enables in vivo detection of edema, infarction, and hemorrhage, whereas histopathological techniques such as hematoxylin and eosin (H&E) staining, 2,3,5-triphenyltetrazolium chloride (TTC) staining, and immunohistochemistry for neuronal or neuroglial markers permit detailed assessment of tissue damage11,43,45. Quantitative analyses, including measurements of infarct or hemorrhagic volume and neuronal loss, further characterize the extent of injury. Evaluation of blood-brain barrier integrity can be performed using Evans blue extravasation, whereas brain edema can be evaluated using brain water content24,35. Behavioral tests such as neurological score, open field test, balance beam test, and rotarod provide evidence of neurological dysfunction, parenchymal compromise35,45. Brain parenchymal injury represents a downstream consequence of CVST and is therefore considered optional for model establishment.
Rodent models offer excellent genetic tractability and are well-suited for mechanistic studies of immune and inflammatory pathways47. However, their small venous structures present challenges for precise surgical manipulation and hemodynamic evaluation. Larger species, such as canine and swine, provide more accessible venous anatomy, allowing the induction of thrombosis in multiple venous sinuses, as well as detailed hemodynamic assessment and pharmacological testing15. Nevertheless, logistical complexity, high costs, and the limited availability of genetic tools hinder their widespread use. Although non-human primates more closely resemble humans in terms of physiology and behavior, they remain largely underutilized due to a lack of reproducible models. The lack of standardized non-human primate models restricts investigations into complex hemodynamic interactions and device-based interventions directly translatable to human CVST. Developing reproducible models in larger species would help bridge this gap and substantially enhance the translational relevance of preclinical research. Animal burden across species should be considered, as smaller animals have lower physiological stress but very small venous structures, whereas larger animals experience higher physiological stress and greater surgical complexity despite larger venous anatomy.
Reversible models are suitable for studying acute-phase mechanisms, thrombus resolution, and drug efficacy, whereas irreversible models are more appropriate for exploring downstream pathophysiological processes following CVT in the acute or chronic phase, as they simulate gradual or persistent venous obstruction. The severity of thrombosis also affects model utility: mild models induce partial or complete sinus occlusion with subtle parenchymal injury, whereas severe models lead to extensive parenchymal infarction and edema. Collectively, these models provide a comprehensive framework for investigating the temporal evolution and pathophysiological mechanisms of CVST, as well as for evaluating targeted therapeutic interventions. Ethical considerations and animal welfare should guide the selection of the CVST model and the experimental design. Model severity, including the extent of venous occlusion, secondary parenchymal injury, and neurological deficits, should be predefined to minimize animal suffering. Researchers should also account for humane endpoints and prioritize the least severe valid model aligned with the scientific question to avoid excessive injury. Integrating these ethical safeguards with model development enhances scientific rigor and translational relevance.
High translational value models include large-animal models using balloon catheters combined with thrombogenic agents, and rodent clot-injection models, as these generate intraluminal fibrin-erythrocyte-rich thrombi, permit imaging validation, and allow testing of pharmacological therapies. Moderate translational value models include FeCl3-induced and photochemical models, which are reproducible and useful for downstream mechanisms and therapeutic screening but rely on artificial endothelial injury. Low translational value models include permanent ligation-based approaches, which overemphasize venous stasis and are therefore discouraged for studies of etiology, anticoagulation, or thrombolysis.
All these models can be applied to studies focusing on the downstream pathophysiological consequences of CVST. Studies of thrombolysis should preferentially employ clot-injection models or balloon catheters combined with thrombogenic agents, and should avoid permanent ligation, silicone rod insertion, and plastic graft insertion. Mechanistic investigations centered on inflammation, endothelial injury, or recanalization may use FeCl3-induced or photochemical-based models. Water-swellable rubber-based models are particularly suitable for studying slowly progressive forms of CVST, as gradual expansion of the implant within the venous sinus mimics progressive luminal narrowing and delayed venous outflow obstruction. Intravascular device validation and translational imaging studies should prioritize swine models using balloon catheters in combination with thrombogenic agents. Table 1 summarizes the recommended applications for commonly used induction techniques.
FeCl3-induced models, clot-injection models, ligation combined with thrombogenic agents, and silicone rod insertion are generally highly feasible in routine laboratory settings. In contrast, photochemical models and balloon catheters combined with thrombogenic agents are more dependent on specialized equipment and technical infrastructure. Photochemical induction of CVST requires a stable light source of a specific wavelength, precise optical focusing and positioning systems, and controlled administration of photosensitizing agents, making this approach highly dependent on specialized optical equipment and technical expertise. Similarly, balloon catheter-based models require endovascular instrumentation and fluoroscopic guidance for accurate catheter placement, and advanced operator skills in intravascular manipulation, thereby limiting their feasibility to laboratories with dedicated interventional facilities and expertise. The feasibility of these induction methods is provided in Table 1.
Reproducibility varies substantially across CVST models. Ligation-only models exhibit low reproducibility, largely due to anatomical variability in venous drainage and collateral circulation. Balloon catheter combined with thrombogenic agents shows moderate reproducibility, as procedural outcomes are highly operator-dependent, and photochemical models are similarly sensitive to illumination range and targeting precision. In contrast, FeCl3-induced models, clot-injection models, and silicone rod insertion generally demonstrate high reproducibility when standardized protocols are applied. The reproducibility of these induction methods is summarized in Table 1.
However, several limitations remain that constrain the clinical relevance of current approaches. Current induction techniques reliably produce venous occlusion but rarely capture the complex, multifactorial etiologies observed in human CVST, such as genetic thrombophilia, systemic inflammation, infection, autoimmune disorders, or venous compression. Consequently, most models primarily reflect the downstream consequences of thrombus formation rather than initial pathological events. This limitation restricts their ability to fully recapitulate disease mechanisms19.
Developing etiology-specific models represents an important future direction. Techniques such as thrombin or clot injection, FeCl3 application, and photochemical injury generate thrombi of reproducible size and composition but do not reflect patient-relevant risk factors or systemic conditions. Incorporating factors such as hypercoagulable states, inflammatory stimuli, hormonal influences, or combined multifactorial conditions could significantly increase the clinical relevance of preclinical models, facilitating targeted therapeutic research. Despite these limitations, existing models remain crucial for evaluating pharmacological therapies, such as anticoagulants and thrombolytics, and for validating imaging techniques and interventional devices. Future studies should integrate genetic and systemic risk factors, multimodal induction methods, and advanced imaging techniques to more accurately model human CVST pathophysiology, thereby improving translational potential and supporting the development of targeted, mechanism-based therapies.
Access restricted. Please log in or start a trial to view this content.
The successful establishment of CVST models is defined by direct evidence of sinus occlusion, with parenchymal lesions considered optional. Thrombosis induced by FeCl₃ or photochemical methods is generally reversible, whereas models based on permanent ligation, autologous clot injection, balloon catheter combined with thrombin, or water-swellable rubber typically result in irreversible occlusion. Most models induce mild pathology, while severe phenotypes usually require combined interventions such as balloon catheter inf...
Access restricted. Please log in or start a trial to view this content.
The authors have no conflicts of interest to declare.
This work was supported by the National Natural Science Foundation of China (82401527).
Access restricted. Please log in or start a trial to view this content.
Access restricted. Please log in or start a trial to view this content.
Request permission to reuse the text or figures of this JoVE article
Request Permission