Computational haemodynamic analysis of thoracic endovascular aortic repair for complex aortic arch diseases
File(s)
Author(s)
Sengupta, Sampad
Type
Thesis
Abstract
The aortic arch is a region of great interest due to its complex anatomical features as it carries blood to the upper limbs and the brain, and then leads down into the descending aorta. Much like the remaining vasculature, it is susceptible to pathologies such as aneurysms and dissections. Aortic arch surgery has come a long way from its first reported case in 1964, with treatment procedures continually evolving. This has led to the emergence of thoracic endovascular aortic repair (TEVAR), which provides a minimally invasive and efficient means of treatment, becoming one of the most popular choices for treating thoracic aortic diseases involving the aortic arch.
A vital component of TEVAR is the use of endografts, which serve as an artificial lumen for blood to flow through by covering primary entry tears in aortic dissection or separating blood from the aneurysm sac. A significant challenge for TEVAR of the aortic arch is to ensure adequate perfusion of the supra-aortic branches. Branched and fenestrated devices have been designed to meet this requirement, but little is known about their haemodynamic performances and long-term durability. The overarching aim of this thesis was to perform computational analysis of blood flow through the aortic arch and to evaluate the haemodynamic conditions following TEVAR using novel branched endografts, specifically single- and double-branched devices for Zone 0 implantation.
In order to elucidate the intricate haemodynamic responses to branched endografts in complex cases, a combination of idealised and patient-specific models were created. The effect of geometric variation on flow and the influence of inner tunnel branches inside the ascending aorta and arch have been investigated in detail. Simulation results for double-branched endografts using both idealised and patient-specific models demonstrate that the selection of tunnel branches should be made to follow the patient’s anatomy as closely as possible. It is important to match the tunnel branch diameters to their respective emerging arch branches to minimise the risk of thrombosis or impaired flow. The performance of a single-branched endograft was investigated utilising both computational fluid dynamics (CFD) and finite element analysis (FEA) for structural mechanics. CFD analyses were performed at different stages of the treatment, allowing better understanding of the haemodynamic changes caused by TEVAR and how they evolve during follow-up assessments. It was noted that small changes in outflow conditions did not have an effect on overall flow conditions and wall shear stress, but the displacement forces experienced by the vessel wall increased by 15%, potentially compromising the seal of the device and led to migration and endoleaks. FEA was performed to investigate the wall stresses experienced at the post-intervention stage. This was then coupled with the fluid flow analysis to perform two-way fluid-structure interaction (FSI) simulations which shed further light on the local haemodynamics and biomechanics of the region. These findings were compared with rigid walled CFD and static FEA simulations to evaluate the efficacy of the modelling procedure. It was found that the CFD simulations tend to over-predict flow velocities, thereby affecting wall shear stress estimation, especially at the regions experiencing low values. The rigid walled CFD predicted higher peak velocities at regions of interest by approximately 9 – 11%, and higher TAWSS by 40.5%. The FEA simulations accurately predicted the spatial distribution of wall stresses experienced in the model, but their static nature failed to capture the time-varying evolution of the vessel wall displacement, under-predicting the peak displacement by 7.6%. FSI was deemed to be more physiologically representative, incorporating the pulsatile time-varying nature of flow and the compliant vessel walls. However, the use of FSI results in a greater level of complexity in the pre-processing stages and increases the computational time required. Therefore, it should be used judiciously based on the requirements of the study and the output desired. The workflow demonstrated in this thesis has been utilised to successfully carry out physiologically representative computational simulations for aortic arch flow.
A vital component of TEVAR is the use of endografts, which serve as an artificial lumen for blood to flow through by covering primary entry tears in aortic dissection or separating blood from the aneurysm sac. A significant challenge for TEVAR of the aortic arch is to ensure adequate perfusion of the supra-aortic branches. Branched and fenestrated devices have been designed to meet this requirement, but little is known about their haemodynamic performances and long-term durability. The overarching aim of this thesis was to perform computational analysis of blood flow through the aortic arch and to evaluate the haemodynamic conditions following TEVAR using novel branched endografts, specifically single- and double-branched devices for Zone 0 implantation.
In order to elucidate the intricate haemodynamic responses to branched endografts in complex cases, a combination of idealised and patient-specific models were created. The effect of geometric variation on flow and the influence of inner tunnel branches inside the ascending aorta and arch have been investigated in detail. Simulation results for double-branched endografts using both idealised and patient-specific models demonstrate that the selection of tunnel branches should be made to follow the patient’s anatomy as closely as possible. It is important to match the tunnel branch diameters to their respective emerging arch branches to minimise the risk of thrombosis or impaired flow. The performance of a single-branched endograft was investigated utilising both computational fluid dynamics (CFD) and finite element analysis (FEA) for structural mechanics. CFD analyses were performed at different stages of the treatment, allowing better understanding of the haemodynamic changes caused by TEVAR and how they evolve during follow-up assessments. It was noted that small changes in outflow conditions did not have an effect on overall flow conditions and wall shear stress, but the displacement forces experienced by the vessel wall increased by 15%, potentially compromising the seal of the device and led to migration and endoleaks. FEA was performed to investigate the wall stresses experienced at the post-intervention stage. This was then coupled with the fluid flow analysis to perform two-way fluid-structure interaction (FSI) simulations which shed further light on the local haemodynamics and biomechanics of the region. These findings were compared with rigid walled CFD and static FEA simulations to evaluate the efficacy of the modelling procedure. It was found that the CFD simulations tend to over-predict flow velocities, thereby affecting wall shear stress estimation, especially at the regions experiencing low values. The rigid walled CFD predicted higher peak velocities at regions of interest by approximately 9 – 11%, and higher TAWSS by 40.5%. The FEA simulations accurately predicted the spatial distribution of wall stresses experienced in the model, but their static nature failed to capture the time-varying evolution of the vessel wall displacement, under-predicting the peak displacement by 7.6%. FSI was deemed to be more physiologically representative, incorporating the pulsatile time-varying nature of flow and the compliant vessel walls. However, the use of FSI results in a greater level of complexity in the pre-processing stages and increases the computational time required. Therefore, it should be used judiciously based on the requirements of the study and the output desired. The workflow demonstrated in this thesis has been utilised to successfully carry out physiologically representative computational simulations for aortic arch flow.
Version
Open Access
Date Issued
2023-02-22
Date Awarded
01/08/2023
License URL
Advisor
Xu, Xiao Yun
Sponsor
Engineering and Physical Sciences Research Council (Great Britain)
Grant Number
EP/R513052/1
Publisher Department
Chemical Engineering
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)