Computational Fluid Dynamics Simulations at Micro-Scale Stenosis for Microfluidic Thrombosis Model Characterization

Platelet aggregation plays a central role in pathological thrombosis, preventing healthy physiological blood fl ow within the circulatory system. For decades, it was believed that platelet aggregation was primarily driven by solu-ble agonists such as thrombin, adenosine diphosphate and thromboxane A2. However, recent experimental fi nd-ings have unveiled an intriguing but complementary biomechanical mechanism — the shear rate gradients generated from fl ow disturbance occurring at sites of blood vessel narrowing, otherwise known as stenosis, may rapidly trigger platelet recruitment and subsequent aggregation. In our Nature Materials 2019 paper [1], we employed micro fl uidic devices which incorporated micro-scale stenoses to elucidate the molecular insights underlying the prothrombotic effect of blood fl ow disturbance. Nevertheless, the rheological mechanisms associated with this stenotic micro fl uidic device are poorly characterized. To this end, we developed a computational fl uid dynamics (CFD) simulation approach to systematically analyze the hemodynamic in fl uence of bulk fl ow mechanics and fl ow medium. Grid sensitivity studies were performed to ensure accurate and reliable results. Interestingly, the peak shear rate was signi fi cantly reduced with the device thickness, suggesting that fabrication of micro fl uidic devices should retain thicknesses greater than 50 µm to avoid unexpected hemodynamic aberra-tion, despite thicker devices raising the cost of materials and processing time of photolithography. Overall, as many groups in the fi eld have designed micro fl uidic devices to recapitulate the effect of shear rate gradients and investigate platelet aggregation, our numerical simulation study serves as a guideline for rigorous design and fabrication of micro fl uidic thrombosis models. blood fl ow rendering was colored by shear rate at constant viscosity. Note the shear rate maxima occurs at the stenosis apex and near the wall while the center forms a low shear pocket. (B and F) Peak shear rate γ max linearly correlates with input bulk shear rate γ 0 for both eccentric and concentric stenoses respectively. Note the concentric stenosis geometry demonstrates better linearity. The shear rate γ (C and G) and shear rate gradient γ ’ (D and H) distribution were analyzed along a sample streamline 1µm above stenosis apex spanning the shear acceleration ( x = − 100 – 0 µm) and deceleration ( x = 0 – 100 µm) zones. Note the γ and γ ’ present equal-space increment in regard to γ 0 . The γ max is located at x = 0 µm while the γ ’ max is located at x = − 5 µm