Computational fluid dynamics modelling of atherosclerotic coronary arteries and abdominal aortic aneurysms

Abstract

This thesis aimed to use three-dimensional quantitative coronary angiography (3D-QCA), computational fluid dynamics (CFD), fluid-structure interaction (FSI) and particle image velocimetry (PIV) techniques in idealised and realistic coronary artery models based on data from patients in order to investigate the relation between haemodynamic forces and lesion morphology. The first aim was to assess the effect of lesion severity and eccentricity on important haemodynamic factors including flow recirculation and shear stress using CFD and PIV. The results showed that the extent of flow recirculation is much more sensitive to mild changes in the severity of intermediate stenoses than is peak shear. The second aim was to find the correlation between fractional flow reserve (FFR), a gold standard to measure the functional significance of coronary stenoses, lesion eccentricity and vessel distensibility using 3D-QCA and FSI methods. The results demonstrated that the coronary arteries with similar lesion morphology but different vessel distensibility are likely to experience different FFR. The third aim was to explore the impact of degree of freedom of coronary arteries on important haemodynamic factors. The results showed that modelling of coronary arteries as rigid vessels could result in treating a non-significant lesion as significant because the rigidity assumption overestimates the amount of flow abnormalities in coronary arteries. Overall, the greater the degree of freedom of movement, the lower the maximum wall shear stress (WSS) and the smaller the area of low WSS. Last but not least, the effect of spiral flow on haemodynamic parameters in an elastic model of abdominal aortic aneurysm (AAA) was investigated. The results demonstrated that neglecting the spiral effects of flow in the modelling of AAAs can result in overestimation of wall stress and the artery wall expansion rate.