Hematocrit, defined as the volume percentage of red blood cells (RBCs) in blood, is an important indicator of human health status, which demonstrates the capability of blood to deliver oxygen. It has been studied over many decades using in vivo, in vitro, and in silicon experiments, and recent studies have shown that its major feature in microvascular networks is the temporal-spatial heterogeneity. The present work is a numerical study of such temporal-spatial heterogeneity, based on direct simulations of cellular-scale blood flow in complex microvascular networks. The simulations take into account the cell deformation and aggregation, and thus are able to capture both the three-dimensional (3D) dynamics of each individual cell and the temporal-spatial distribution of cell population. The results showed that the temporal-spatial heterogeneity is more pronounced in the network that has the vessels with smaller diameters or with more complex geometry. Such heterogeneity is largely attributed to the existence of bifurcations, where the positively-correlated hypotactic (feeding-branch) and paratactic (branch-branch) relations are generally observed in both the time-averaged hematocrit and temporal hematocrit range. This suggests that the successive bifurcations have a substantial impact on the temporal-spatial heterogeneity of hematocrit. However, these positive correlations may be broken up if the diameter of the feeding vessel is small enough or the bifurcation is asymmetric extremely, due to the vessel blockage. The present study is of great clinical significance to help doctors make more accurate diagnosis and treatment, by providing more information about the temporal-spatial distribution of the hematocrit in microvascular networks.

No abstract

Computational modeling and simulation of cellular blood ow is highly desirable for understanding blood microcirculation and blood-related diseases, such as anemia, thrombosis and tumor, but it remains a challenge because the blood requires to be described as a dense suspension of di_erent types of cells and the microvessels continually bifurcate or merge into a complex network. A smoothed dissipative particle dynamics-immersed boundary method (SDPD-IBM) has been developed, integrating the uid ow and cell behavior to simulate physiological and pathological phenomena involved in blood ow. The SDPD is used to model the uid ow, the IBM is used to model the interactions between the uid and cells, and three phenomena are taken into account, cell deformation, aggregation and adhesion. The simulations consist of two parts: validation studies for the _delity of the SDPD-IBM, and case studies for its potential Computational modeling and simulation of cellular blood ow is highly desirable for understanding blood microcirculation and blood-related diseases, such as anemia, thrombosis and tumor, but it remains a challenge because the blood requires to be described as a dense suspension of di_erent types of cells and the microvessels continually bifurcate or merge into a complex network. A smoothed dissipative particle dynamics-immersed boundary method (SDPD-IBM) has been developed, integrating the uid ow and cell behavior to simulate physiological and pathological phenomena involved in blood ow. The SDPD is used to model the uid ow, the IBM is used to model the interactions between the uid and cells, and three phenomena are taken into account, cell deformation, aggregation and adhesion. The simulations consist of two parts: validation studies for the _delity of the SDPD-IBM, and case studies for its potential and usefulness. The validation studies consider the ow of pure uid, the mechanical behavior of cells, and the multi-outlet cellular ow, while the case studies include cells passing through simple vessels, successive bifurcations, and even a complex microvascular network. These studies concern the formation of a thrombus, the partitioning of red blood cells, and the metastasis of tumor cells. The SDPD-IBM has special advantages in modeling uid ows in complex domains and with uid-structure interactions, because the SDPD is convenient to model a complex domain by discrete particles, while the IBM is exible to model the interactions between the uid and structures.and usefulness. The validation studies consider the ow of pure uid, the mechanical behavior of cells, and the multi-outlet cellular ow, while the case studies include cells passing through simple vessels, successive bifurcations, and even a complex microvascular network. These studies concern the formation of a thrombus, the partitioning of red blood cells, and the metastasis of tumor cells. The SDPD-IBM has special advantages in modeling uid ows in complex domains and with uid-structure interactions, because the SDPD is convenient to model a complex domain by discrete particles, while the IBM is exible to model the interactions between the uid and structures.

During hematogenous metastasis, the arrest of tumor cells in the microvasculature is a prerequisite for extravasation from the circulation to a distant host organ. To reveal such arrest behavior, we implement three-dimensional numerical simulations on the motion of a single tumor cell in microvessels at the cellular scale and mainly investigate the interactions among mechanical entrapment, adhesion, and cell stiffness, and their effects on the tumor cell arrest. Two types of vascular configurations qualifying for mechanical entrapment are considered, the constriction and bifurcation structures that are comparable in diameter with the tumor cell. The main results indicate that in the constriction tube, as the constriction radius is increased, the tendency that number of adhesion bonds increases with increasing shear modulus becomes more and more obvious. However, the adhesion behavior has little effect on the tumor cell arrest in the constriction region, regardless of the number of adhesion bonds. The mechanical entrapment plays a more important role than the cell stiffness in the tumor cell arrest in the constriction tube. In the bifurcated tube, the tumor cell is more likely to be arrested in the bifurcation region with a small bifurcation angle. Moreover, as the bifurcation angle or shear modulus is decreased, the effect of adhesion behavior on the tumor cell arrest becomes increasingly obvious. These results are helpful in understanding the biomechanism of tumor metastasis.