The transport behaviour of the haematocrit in the larger arteries is important in defining the variations in viscosityof blood. In this study, a finite volume method is used in order to simulate the blood flow and haematocrit transportthrough a large 3D human-like 90-degree bifurcation. The simulations are carried out to investigate the importance ofexplicitly modelling the non-Newtonian viscosity of blood regarding defining the flow. It is expected to be especiallyimportant in the regions surrounding a bifurcation. The main focus is to compare non-Newtonian to Newtonianbehaviour of the flow through important parameters such as pressure losses, mean viscosity variations and bulktransport properties of haematocrit. The study considers a broad range of physiological and pulsatile flow conditions,and displays the importance of modelling blood flow as a non-Newtonian fluid. The results have a relevant impactregarding the possible discrepencies in important physiological parameters such as wall shear stress (WSS), whencoupling the haematocrit field data back to the viscosity models.

Complex and slow interaction of different mechanical and biochemical processes in hemodynamics is believed to govern atherogenesis. Over the last decades studies have shown that fluid mechanical factors such as the Wall Shear Stress (WSS) and WSS gradients can play an important role in the pathological changes of the endothelium. This study provides further indications that the effects of fluid mechanical aspects are correlated with the diseased regions of the larger arteries. Unsteady high temporal WSS gradients (TWSSG), a function of the shear-thinning property of the non-Newtonian viscosity, move with the separation bubble. Red Blood Cell (RBC) dilution due to the secondary flows determines the magnitudes of the WSS and TWSSG. The results indicate that the focal nature of the TWSSG may have implications on the response of the endothelium.

Cardiovascular defects characterized by geometrical anomalies of the aorta and its eecton the blood ow is the focus of this study. Not only are the local ow characteristicsgeometry dependent, but they are also directly connected to the rheological properties ofblood. Flow characteristics such as wall shear stress are often postulated to play a centralrole in the development of vascular disease.In this study, blood is considered to be a non-Newtonian uid and modeled via theQuemada model, an empirical model that is valid for dierent red blood cell loading.Three patient-specic geometries of the aortic arch are investigated numerically. Thethree geometries investigated in this study all display malformations that are prevalent inpatients having the genetic disorder Turner syndrome. The results show a highly complexow with regions of secondary ow that are enhanced in two of the three aortas. Moreover,blood ow is clearly diverted due to the malformations, moving to a larger extent throughthe branches of the arch instead of through the descending aorta. The geometry havingan elongated transverse aorta is found to be subjected to larger areas of highly oscillatorylow wall shear stress.

When studying disease development in arteries, it is important to understand the local variations in blood rheology. Blood flow in large arteries is often assumed to behave as a homogeneous fluid, an assumption that is not entirely correct. The local viscosity changes with the local concentration of Red Blood Cells (RBCs) and the rate of shear strongly influences the Wall Shear Stress (WSS) and its gradients, physiological parameters important in the study of atherosclerosis. Moreover, the flow behavior of RBCs is influenced by the geometric structure of the flow environment. In experiment, rheological properties across a tube cross-section are difficult to measure if non-invasive techniques are to be used. Therefore, rheometric devices are constructed of simple geometries to measure the bulk rheology. In this study, the Lattice Boltzmann Method is used to model the blood as a particle suspension of RBCs. The RBC Volume Fractions (VF) investigated corresponds to 1, 2 and 5%, and both a channel and a tube flow are considered. The results display large differences in RBC distributions and velocity profiles. Estimated from existing viscosity models, the viscosity distributions are found to display variations of up to 5% when comparing the two geometries. This is of importance since errors in quantifying the viscosity can lead to miscalculations of the physiological variables.

The identification of regions prone to atherogenesis in the arterial network is compounded by the complex, slow interaction of mechanical and biomechanical processes. In recent times simplifications to the analysis of the near wall hemodynamics have been sought-after to identify plaque prone regions. Mean parameters have been defined to analyze the common fluid mechanical hypotheses considering the role of wall shear stress (WSS) variations in the pathological changes to the endothelium. In this study well known WSS indicators are applied to varying flow conditions of blood-like fluids in a 90-degree arterial bifurcation. The conventional indicators identify two distinct, focal regions that correlate with a known plaque prone location near arterial bifurcations. The results however demonstrate that the interpretation of the indicators can be difficult under varying flow conditions unless complementary parameters are considered simultaneously. A new indicator is also suggested that extracts the peaks of the temporal WSS gradients (PTWSSGs) and is shown to co-incide well with plaque prone regions. The PTWSSG could be used as a complimentary atherogenic indicator in bifurcating arteries, thereby expanding cardiovascular disease studies to the consideration of alternative fluid mechanical hypotheses. The inclusion of a non-Newtonian model is important in predicting the WSS and temporal WSS gradient distributions near the bifurcation due to the separation bubble induced fluctuations in the shear. Atherogenic indicators could be misleading if non-Newtonian effects are excluded.