In vitro and in vivo dynamics of abdominal aortic aneurysms: A fluid-structure interaction study

Abdominal aortic aneurysm (AAA) rupture occurs when the stress of the in vivo environment exceeds the resistance provided by the arterial wall, causing a failure of the wall to retain the blood that passes through the aorta to the systemic circulation. It is the 13th leading cause of death in the United States, affecting an estimated 8% of the population over the age of 65. The current state of the clinical assessment of AAAs includes using a diameter-based criterion or growth rate to determine if surgical intervention is necessary. Typical repair includes endovascular (EVAR) or open surgery where a graft device is placed within the AAA to restore normal blood flow through the aorta while alleviating the stresses acting on a diseased arterial wall. Unfortunately as many as 33% of AAAs that rupture have a diameter smaller than the 5.5cm critical value used in the clinical setting for assessment. Since AAAs are a largely asymptomatic condition, the screening necessary to measure growth rate often cannot be completed. This has lead to the development of several proposed approaches that establish improved criteria which can be used to determine rupture risk.;Chapter 1 describes in detail these approaches and the resulting criteria which are recommended for the assessment of AAAs. The majority of the research has applied the predicted wall stress from computational solid stress (CSS) analyses to determine the effects of parameters such as asymmetry, wall thickness, and tortuosity on the AAA rupture risk. However there are several factors which affect the ability to develop an accurate rupture assessment tool, including the mechanical properties of the patient's AAA, the presence and consistency of intraluminal thrombus (ILT), and the arterial pressure within the AAA sac. One factor which may be overlooked is the fluid dynamics that exists within the aneurysm, which has been shown to correlate with biological as well as mechanical changes to the AAA wall.;To this end, the use of fluid-structure interaction (FSI) may provide additional characteristics that contribute to AAA rupture. The investigation presented in Chapter 2 forms the hypothesis that the dynamic interaction of the blood flow and the arterial wall significantly affects the maximum wall stress and the conditions under which it occurs. Three different patient-specific morphologies are reconstructed using semi-automatic edge detection software to include the lumen, common iliac arteries, intraluminal thrombus, and arterial wall with uniform thickness. Computational solid stress (CSS) as well as partially and fully-coupled fluid-structure interaction (FSI) analyses are completed using a generalized inlet velocity and outlet pressure waveform. The arterial wall is modeled with an isotropic, hyperelastic material property as is the localized intraluminal thrombus. The results show that the shape of the lumen affects the pressure gradients within the AAA and therefore the magnitude of the maximum wall stress. The maximum CSS- and FSI-predicted wall stresses for a patient with a uniform luminal cross-section were within 3%, while those patients with a protruding intraluminal thrombus differed by as much as 26% between the computational approaches. The higher FSI-predicted wall stress in these cases is attributed to the increased wall pressure resulting from the fluid dynamics within the AAA sac.;Demonstrating the importance of fluid dynamics on the resulting wall stress is a significant contribution to the study of AAAs, in part because it represents the first fully-coupled FSI that utilizes a hyperelastic material model for the wall and ILT, and includes the common iliac bifurcation. However the results need to be validated both for physical as well as physiological accuracy. In Chapter 3, an idealized AAA glass model is used in experimental and computational analyses to compare the fluid dynamics obtained with particle image velocimetry (PIV) with those from a CFD and FSI analysis. The glass model is scanned and converted into a CAD model so that an accurate lumen geometry is created. The CAD models are then used to form a destructible core which is coated with Silbione(R) elastomer to form a compliant model. Similarly the CAD models are imported into meshing software to create identical computational domains. Two PIV experiments were conducted, one with the glass model and the other with the compliant model. These results were then compared to computational fluid dynamics (CFD) and f-FSI simulations that applied boundary conditions derived from the in vitro flow loop. The effect of the AAA dilation was captured by all three approaches, which showed the development and dissipation of vortices. Furthermore the f-FS I demonstrated the effect of the compliant wall on this flow dynamics, including a longer duration of attached flow at the lateral walls. However the wall motion was not well replicated in the f-FSI, in part because of the two-dimensional visualization of the three-dimensional flow fields from both the PIV and f-FSI systems. The relative consistency between the f-FSI results and the compliant model PIV experiments, coupled with the comparable dynamics in the CFD and glass model flow visualization, showed the ability to numerically predict the fluid dynamics within the AAA geometry, with the success of the f-FSI dependent on the wall properties and external conditions. (Abstract shortened by UMI.)...