Fluid-structure interaction for heart attack prediction

Abstract

Cardiovascular disease is one of the largest and most economically burdensome challenges of modern society. Despite continuous imaging, intervention and pharmacological improvements, this burden continues to grow. Biomechanical stress is known to play a role in the natural history of the largest facet of this burden, coronary atherosclerosis. Yet a detailed understanding or capability to determine these stresses in any great detail eludes both clinical research and practice. With consistent improvements in computational power and numerical complexity enabling advancements in coronary simulation, this thesis now aims to address this challenge through the development of fluid-structure interaction (FSI) frameworks. Beginning with a detailed reconstruction of an entire vessel from invasive coronary angiography (ICA), including all epicardial branches and replication of the unique dynamic coupling between coronary arteries and cardiac function, traditional stress measures were assessed, namely wall shear (WSS) and von Mises stress. The reconstruction approach was then further enhanced with high-resolution intravascular optical coherence tomography (OCT) to assess changes in plaque morphology between two time points and their associations with baseline biomechanical stress. To expand the detail in biomechanical results a further 11 multidirectional and topological WSS metrics were developed for use in FSI simulations for the first time. Their relationship with coronary dynamics was also assessed to determine if dynamic factors may play a role in atherosclerosis development. Taking inspiration from these WSS metrics, several novel structural metrics were then proposed to specifically provide information on the directional properties of the Cauchy stress tensor in coronary simulations. These metrics were assessed against changes in the coronary vasculature in six patients. Finally, to combine the developed knowledge, a complete FSI framework, including a novel patientspecific dynamic profile (PDP), was proposed to culminate with a biomechanical stress profiling index (BSPI) which used a total of 66 metrics to target key changes in lipid plaques and vessel remodeling. This holistic modelling approach significantly builds upon current biomechanical simulation knowledge, with preliminary results already suggesting mechanisms of plaque destabilisation and remodelling not previously identified. It is hoped the details encompassed in the BSPI will provide both a better understanding of the natural history of atherosclerosis and lead to novel approaches to prevent its significant morbidity, mortality, and economic burden when studied in larger patient cohorts.