Effects of non-Newtonian models on viscosity of unsteady aortic blood flow

Computational hemodynamics plays a crucial role in reproducing the details of aortic blood flow. However, the application of the non-Newtonian viscosity model is still controversial. The objective of this study is to demonstrate the effects of different non-Newtonian models on the viscosity of blood flow in healthy aorta. First, we reconstructed the three-dimensional geometric models of two healthy aortas based on computed tomography angiography images. The blood flow waveform with parabolic distribution and the three-element Windkessel model were adopted as boundary conditions. Then, the interaction between the blood flow and hyperelastic aortic vessel wall was considered by the two-way fluid–structure interaction method. Finally, we chose four commonly used non-Newtonian viscosity models: the Quemada model, Casson model, Carreau, and Carreau–Yasuda models. Results show that the instantaneous low shear strain rate (SSR < 100 s−1) cannot be neglected considering its relatively high proportion in the aortic wall (50%) and cardiac cycle (33%). We find that the Quemada model can predict the shear-thinning properties of aortic blood flow, especially the relatively low viscosity distribution. Besides, the high-viscosity iso-surface is observed in the descending aorta throughout the cardiac cycle. The phenomena further underline that the Newtonian assumption is not suitable for predicting the viscosity distribution of aortic blood flow. In conclusion, the non-Newtonian viscosity model is suggested to be adopted in aortic computational hemodynamics, and the performance of the Quemada model is satisfactory.