While three determinants for thrombosis—hypercoagulability, endothelial dysfunction, and hemodynamics—are outlined in Virchow's triad (1), platelets play an essential role in arterial thrombosis. Their thrombotic functions are governed by a range of biomechanical factors and the underlying mechanobiology is a rapidly evolving field and topical research area (2). Intriguingly, platelets respond to flow disturbance in complex vessel architectures with respect to bifurcation, branching, stenosis and aneurysm (3), which inevitably produce local shear gradients (4–6), vorticity (7) and even turbulence (4, 8). The biomechanical platelet adhesion and aggregation are primarily mediated by its two mechanoreceptors—glycoprotein Ib (GPIb) and glycoprotein IIb/IIIa (GPIIb/IIIa or integrin α_IIbβ₃) (9, 10). Furthermore, to facilitate thrombus stabilization, platelets generate contractile forces which are mediated by cytoskeletal components (actin, tubulin, Arp2/3) and motor proteins (myosin and dynein) in response to the mechanical microenvironment (11, 12). Meanwhile, platelets interact with red blood cells, neutrophils, plasma proteins [von Willebrand factor (VWF), fibrinogen], and the vessel wall (endothelial cells) to exert their hemostatic and thrombotic functions (13). Increasingly, microfluidic technologies have emerged as powerful and indispensable approaches to investigate the mechanosensitive behaviors of platelets (2, 12, 14). Numerous endothelialized microfluidics have been invented to model the interplay between platelets, endothelium and VWF under physiologically relevant hemodynamic microenvironments (6, 15, 16). In the context of platelet mechanobiology, we summarize these state-of-the-art microfluidic methodologies that have recently been invented in the field.

Single-cell biomechanical nanotools such as atomic force microscopy (17), optical tweezers (18) and micropipette based adhesion assays (19) significantly advanced our understanding of platelet mechanobiology (20). However, a major shortcoming of these in vitro techniques is their inability to recapitulate the physiologically relevant hemodynamics in the context of platelet adhesion and aggregation (21). To this end, animal models of thrombosis and intravital microscopies have become more popular and accessible approaches for examining thrombus initiation, progression and propagation in vivo (22). Nevertheless, multiple fundamental limitations still exist with the animal approaches (23, 24): (i) inter-species variabilities which prevent complete recapitulation of human disease pathogenesis; (ii) physiological variations between individual animals; (iii) lengthy ethical approval processes.

Shear-based in vitro assays, such as parallel plate flow chambers (25), cone-and-plate viscometers (26) and thromboelastometers (27) have been widely used for investigations of human blood samples under physiologically relevant shear rates and shear stresses (5, 28–30). Although these methods were instrumental in the current understanding of shear dependent platelet thrombosis, to some extent, they are limited due to the need for large sample volumes, low throughput, and their inability to accurately emulate vascular architectures and mechanical properties of the circulatory system (31–33). To these points, microfluidic techniques have rapidly emerged as complementary humanized models for investigations of platelet mechanobiology and biomechanical thrombosis. More recently, the International Society of Thrombosis and Hemostasis (ISTH) have even made recommendations for producing reliable microfluidic devices with reproducible thrombogenic coating and consistent hemodynamic design (34). This represents a major step forward in promoting microfluidic technologies for future pre-clinical and diagnostic testing methods in hemostasis and thrombosis.

Moreover, while most of the research interests focus on large vessel thrombosis, there is an emerging trend toward understanding thrombosis in the microcirculation which is significantly implicated in a variety of systemic disorders such as: Thrombotic Thrombocytopenia Purpura (TTP) and Hemolytic Uremic Syndrome (HUS) which lead to multi-organ dysfunction syndrome and death (35, 36); and localized organ injuries resulting from trauma, ischemia-reperfusion, transplant rejection, disseminated intravascular coagulation, sickle cell disease and more recently COVID-19 (35, 37, 38). The microfluidic approach can effectively recapitulate the anatomical structures within such small scale as opposed to conventional shear-based assays (39). The recent advancements of microfluidic designs and the micro and nano fabrication—particularly the shift from devices with a single layer to those with multiple-layered structures (40), and from parallel straight channels to complex geometries (15)—have enabled the investigation of platelet mechanobiology in more complicated vascular biomechanical microenvironments.

In the past decade, soft lithography has rapidly grown to enable high-fidelity micropatterning. Parallel straight channels can simply be fabricated by the single-layered photolithography and poly(dimethylsiloxane) (PDMS) casting processes, resulting in a vascular geometry with laminar flow (5, 16, 41, 42). When connected with external syringe pumps, such simple microfluidic setting enables systematic investigation of shear dependent platelet adhesion (43) and aggregation (44). Additionally, vascular cell cultures can be incorporated; specifically, endothelialized microchannels can model the interaction between blood components and the vessel wall under physiological hemodynamic parameters (45), or even when subjected to proinflammatory or other endothelial diseased conditions (16).

Furthermore, single-layered microfluidics can emulate more complex vessel structures, such as bifurcation (39), 2D stenosis (46), network (47), and micropost array (40) to recapitulate flow disturbance induced platelet activation, adhesion, contraction and aggregation (48, 49). The mechanical stimuli investigated include elevated shear stress (50, 51), shear rate gradient (52–54), vorticity (7) and turbulence (8, 55) at different locations of the microchannel. Moreover, lining the microchannel with endothelial cells enables the investigation of prothrombotic synergistic effects of hemodynamic forces, blood cells and the vessel wall (56, 57). Rapid prototyping methods such as vertical milling molding (58) and 3D printing (56) enable fabrication of microfluidics with circular structures and variations in the z-direction. PDMS casting around poly(methyl methacrylate) (PMMA) optical fibers can also create circular channels (59, 60). As such, these recent microfluidic advancements provide great means for examining how shear stress, shear rate gradients, vorticity, blood viscosity and contractility affect platelet thrombotic functions.

In the past 3 years, in vitro lung (61), blood vessel (62), and tumor models (63) were made on microfluidics (64). These biomimetic organ-on-a-chip models recapitulate the structural, functional, and mechanical aspects of vascular microenvironments. Obviously, emulating such hierarchical organs requires emerging microfabrication methods including 3D bioprinting (65), double-layered soft lithography (66), injection molding (67), and micropattern stamping (68). This allows the production of multiple-layered structures via one-time or assembled fabrication. The composite microsystems can support tissue coculture and the channel-channel interface, thereby enabling studies of bidirectional tissue signaling across the endothelial barrier (61). More importantly, external mechanical manipulation apparatus (e.g., stiff ECM matrices, actuation chambers) can be integrated for disease-specific studies and rapid drug screening (55, 69). Although multiple-layered microfluidics incur higher fabrication requirements including precise alignment and assembly, they present physiologically relevant hemodynamic microenvironments for better biomimetic performance.

Last but not least, the transparency of PDMS material is greatly compatible with advanced microscopies to visualize thrombotic dynamics in real time. These PDMS or equivalent microfluidics thus offer a versatile way to evaluate platelet activation, adhesion, aggregation, morphological change, and soluble agonist secretion under hemodynamic control. A range of studies have delved into reproducing diverse vessel architectures to investigate how biomechanical factors (e.g., shear stress, shear gradients and traction forces) and the circulatory components (e.g., vessel endothelium and VWF) regulate platelet functions. As shown in the Figure 1 and Table 1, we summarize these recent lab-on-chip approaches in the context of platelet mechanobiology and thrombosis.

Increased usages of shear-based microdevices by different laboratories for examining platelet functions lead to heterogenous device characteristics. Lack of standardization in microfluidic designs and operating methods complicates the result comparison and reproducibility, which impedes translation into both clinical practices and industrial settings (89, 90). The ISTH Biorheology subcommittee has advocated for standardization of flow chamber-based thrombus formation assays (34, 89, 91, 92). Requirements for microfluidic standardization can be extended to the following considerations.

The geometry of microfluidic devices is of the utmost relevance to their application in platelet function studies and their extension to antiplatelet drug screening platforms. Commercially available microfluidic systems have been able to standardize more simple, quasi-2D planar geometries but are still limited by their inability to create complex geometry in a high-throughput manner. A summary of studies looking at platelet biology in thrombosis using commercialized microfluidic slides is presented in Table 2.

For effective commercialization of more complicated microfluidic structures, the fabrication protocol need to be high-throughput, robust, reproducible and cost-efficient (108). Such fabrication techniques including nanoimprint lithography, 3D printing and anodic aluminum oxidation are promising in supporting rapid-prototyping of highly tailorable complex microfluidics with replicable specifications (90, 108). With increased standardization and advancements in the commercial capacity, biomimetic microfluidic systems will accelerate novel antiplatelet therapeutic development and have great potential in patient screening (109).

Emerging microfluidic technologies have advanced our understanding of platelet mechanobiology and its role in hemostasis and thrombosis. The ease of manufacturing complex geometries, versatility and the high throughput of microfluidic experiments have enabled researchers to investigate how platelets respond to their hemodynamic microenvironment and interact with the circulatory system in a well-controlled biomechanical milieu. The varieties in designs, fabrications, operation procedures and analytical approaches of customized microfluidics have urged the standardization need for reproducibility. The increasing use of commercially available microfluidics accelerates the translation of platelet mechanobiology to pre-clinical and industrial applications. In the next decade, we foresee microfluidic technologies to be used for patient-specific disease management, diagnosis, as well as antiplatelet drug screening.

YZ and LJ conceived the study and wrote the manuscript. VC, RP, and SR co-wrote the manuscript. HL and LJ provided critical comments, suggestions, and text. LJ designed and supervised the study. All authors contributed to the article and approved the submitted version.

This work was supported by the Australian Research Council (ARC) Discovery Project (DP200101970 - LJ), the National Health and Medical Research Council (NHMRC) of Australia Ideas Grant (APP2003904 – LJ), NSW Cardiovascular Capacity Building Program (Early-Mid Career Researcher Grant - LJ), Sydney Research Accelerator prize (SOAR - LJ), and Ramaciotti Foundations Health Investment Grant (2020HIG76 - LJ). LJ is an ARC DECRA fellow (DE190100609) and National Heart Foundation Future Leader fellow (FLF2 105863).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.