Abstract – Introduction: Arterial thrombosis is a multifaceted process characterized by platelet aggregation and fibrin deposition, leading to the occlusion of blood vessels. It plays a central role in cardiovascular conditions such as myocardial infarction and ischemic stroke. Gaining insight into the mechanisms underlying arterial thrombosis is essential for developing effective treatments aimed at preventing thrombotic events and reducing associated health burdens. In vitro and ex vivo models serve as critical tools for investigating the pathophysiology of arterial thrombosis by providing controlled environments to study thrombus formation and characteristics. This systematic review provides a comprehensive overview of in vitro and ex vivo flow-based models used to study arterial thrombosis, classifying them by scale (macro vs. micro) and evaluating their design principles, physiological relevance, and experimental utility. Methods: A systematic search of Medline, Embase, and Web of Science was conducted using broad and specific terms related to arterial thrombosis models incorporating flow or shear stress. Articles were screened by two independent reviewers. Studies were included if they described in vitro or ex vivo models with dynamic flow; models limited to static or venous conditions or in vivo studies were excluded. In total, 82 studies met the inclusion criteria. Results: Macro-scale models can mimic complex flow patterns in larger arterial conditions and enable the formation of thrombi comparable in size to clinical specimens. Microfluidic models allow precise control over shear conditions and geometry with minimal blood volumes and are suitable for high-resolution imaging and customization, including endothelialization and patient-specific designs. While, both model types present limitations in replicating complex in vivo hemodynamics, standardization, and scalability, they offer valuable, controllable platforms for mechanistic studies and drug testing in arterial thrombosis. Conclusions: While no single model fully recapitulates the in vivo environment, ongoing innovations, particularly in microfabrication and model standardization, continue to improve physiological relevance and clinical translatability.

The fibrin network is one of the main components of thrombi. Altered fibrin network properties are known to influence the development and progression of thrombotic disorders, at least partly through effects on the mechanical stability of fibrin. Most studies investigating the role of fibrin in thrombus properties prepare clots under static conditions, missing the influence of blood flow which is present in vivo. In this study, plasma clots in the presence and absence of flow were prepared inside a Chandler loop. Recitrated plasma from healthy donors were spun at 0 and 30 RPM. The clot structure was characterized using scanning electron microscopy and confocal microscopy and correlated with the stiffness measured by unconfined compression testing. We quantified fibrin fiber density, pore size, and fiber thickness and bulk stiffness at low and high strain values. Clots formed under flow had thinner fibrin fibers, smaller pores, and a denser fibrin network with higher stiffness values compared to clots formed in absence of flow. Our findings indicate that fluid flow is an essential factor to consider when developing physiologically relevant in vitro thrombus models used in researching thrombectomy outcomes or risk of embolization. Graphical Abstract: [Figure not available: see fulltext.].