A multiscale model will be developed to describe in a physically accurate manner the dynamics of blood as it flows through constricted geometries. This study is important because in cardiovascular prostheses (heart valves, stents, and ventricular assist devices) such constricted geometries (narrow leakage gaps, hinges) are locations where platelet damage can be caused by high fluid shear stresses;platelet activation and collisions can then trigger aggregation and subsequent thrombus formation. The key idea underlying this proposal is that the phenomena that lead to thrombus initiation are intrinsically multiscale in nature. Device scales are ~centimeters, while the constricted geometries are ~100 microns wide and the blood cells are order of microns in size. Blood flowing through 100 micron gaps cannot be treated as a homogeneous fluid and platelets cannot be treated as passive point particles. In fact, the larger, deformable and relatively numerous red blood cells play a crucial role in the mechanics of platelets. Therefore, a physically correct model of shear-induced platelet activation (SIPA) must account for the dynamics of blood cells, the flow geometry and the flow of plasma. To better understand the micro-scale interaction that leads to SIPA and to guide beter design of prostheses, accurate computational modeling of blood flow through constricted and contorted micro-geometries will be extremely useful. The challenge for modeling is that even in constricted micro-geometries there are millions of blood cells, thus well resolved 3-D computations are presently infeasible. Therefore, the proposed multiscale model: (1) will be 2-dimensional and (2) will couple fully resolved micro-scale models of red blood cells and platelets with coarse-grained meso-scale models in order to make possible the tracking of multitudes of (~104) cells. To ensure physical validity, model predictions (for example cell trajectories, cell-cell interactions and flow field characteristics) will be compared with tandem micro-particle image velocimetry (m-PIV) experiments. Furthermore, platelets will be tracked in the m-PIV visualizations and biological assays will be employed to determine the levels of activation and aggregation of platelets. By intimately combining computations and experiments, we seek to obtain: (1) a clear picture of the micro-mechanics of blood cells and (2) quantitative correlation between local shear stresses and shear gradients and the tendency of platelets to exhibit the signals that mark thrombus initiation. The work is highly challenging and it will advance the state-of-the-art by: (1) computing the dynamics of large numbers of cells immersed in blood flow;(2) devising techniques for transferring information from fully-resolved micro-scale to coarse-grained meso-scale calculations;(3) intimately coupling experimental and computational studies to understand micro-scale dynamics;and (4) developing the ability to perform truly large scale parallel computations with broad application to a variety of biomedical systems. These cutting-edge techniques will provide unprecedented quantitative information on the micro-dynamics of blood cells and the impact of micro-geometry and flow patterns on thrombus initiation.
This proposal seeks to develop multiscale computational models of blood flow in specific regions of cardiovascular prostheses (heart valves, stents, ventricular assist devices) that are implicated in initiation of potentially dangerous thrombi. By advancing the understanding of and the ability to model the mechanics of blood cells under thrombogenic conditions, the proposed work will not only further our understanding of the pathology, but also develop predictive capabilities that can be used in designing safer prostheses.