The influence of geometric factors on the wall shear stress distribution in realistic human coronary arteries

Dissertação apresentada na Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa para a obtenção do grau de Mestre em Engenharia Biomédica. A presente dissertação foi desenvolvida no Erasmus Medical Center em Roterdão, Holanda

Background: Atherosclerosis is the main cause of death in the Western society. It is a geometrically focal disease, affecting preferentially vessel areas of low wall shear stress (SS), which induces the expression of atherogenic genes. To predict wall SS several options are available. Among them Computational Fluid Dynamics (CFD) simulations on 3D reconstructed coronaries using Finite Element Modeling (FEM). However, to perform CFD a 3D representation is needed. To obtain a 3D representation of the coronary under study different methods can be applied. Methods: CFD calculations were performed using FEM on ten 3D reconstructed coronary arteries by the state-of-the-art ANGUS method (biplane angiography + Intravascular Ultrasound (IVUS)). The SS outcomes of the CFD calculations were compared with SS calculated by the Poiseuille equation, and with the SS outcomes of CFD simulations of the same 3D reconstructed arteries by QCA-3D (biplane angiography – no cross-sectional information) and Straight (IVUS images stacked on a straight centerline – no curvature information) methods. Results: The Poiseuille equation did not have any sensitivity in predicting any low SS (<0.5 Pa) per cross-section. However, the average correlation coefficient between the average SS per cross section from the Angus geometries and SS based on the Poiseuille equation was r2 = 0.65  0.09. A strong correlation was obtained for the SS from the ANGUS and the Straight method, while only an average correlation was obtained between ANGUS and QCA-3D average SS. Bland-Altman analysis was performed to confirm the results agreement. The sensitivity and specificity of the QCA-3D and Straight method in predicting low and high SS was measured. Geometric factors, such as local curvature, area gradient and torsion were found to be related to the presence of SS peaks or to regions prone to plaque development. These geometric risk factors were utilized to give some guidelines on meshing optimization. Conclusions: The use of a simpler 3D reconstruction approach, such as the QCA-3D or the Straight method, in combination with the optimization of meshing based on the geometric features of the coronaries, has the potential to, in the future, bring CFD calculations of wall SS from bench to bedside.