A computational study of red blood cell trafficking in tumor/angiogenic capillary vessel networks

Capillary vessels together with arterioles, venules, and vascular junctions, form a highly complex network of narrow pathways for blood flow. Blood in such vessels flows as a dense suspension made of red blood cells (RBC) which are extremely deformable and can easily pass through capillaries less than their size. The distribution of RBCs in the vascular networks is highly heterogeneous and is critical to the healthy functioning of the body. An accurate prediction of RBC trafficking in such networks is a challenging computational problem as it requires resolving 3D deformation of each RBC in the dense suspension flowing through the complex geometry of the vascular networks. We have developed a high-fidelity 3D direct simulation method to predict flow of thousands of deformable RBCs through physiologically realistic microvascular networks comprised of many blood vessels and vascular bifurcations. The complexity of the vascular networks increases significantly in tumor and angiogenesis as characterized by the emergence of abnormal geometry, non-circular and collapsed vessels, multi-furcations, and tessellated architecture. We use in vivo images to create such vascular networks in silico and then predict RBC trafficking and capillary hemodynamics using our direct simulation. We provide quantitative differences between healthy networks and tumor/angiogenic networks in terms of RBC distribution, near-wall cell-free layer and wall shear stress.