(H-291) Estimation of Pressure in Aortic Dissection to Study the Development of Aneurysmal Degeneration

Introduction:: Aortic dissection is a life-threatening condition associated with high morbidity and mortality rates due to the complex nature of the disease. As such, prognosis is often challenging, unpredictable, and highly variable on a patient-to-patient basis. One of the complications associated with aortic dissection is the development of aneurysmal degeneration in the outer wall of the false lumen (FL). Although changes in the pressure in FL is perceived to play a key role in such a development, there are currently no non-invasive methods of monitoring the changes in pressure over time, which can directly affect the rate at which aneurysmal degeneration occurs. In this study, our primary objective was to utilize computational fluid dynamics (CFD) in conjunction with machine learning (ML) to explore the possibility of non-invasive characterization of the intraluminal pressure. Ultimately, we aim to investigate the effects of pressure over time as a factor in driving aneurysmal degeneration in aortic dissection to improve risk-stratification in aortic dissection patients.


Materials and Methods:: A three-dimensional geometry of the type B dissected aorta was reconstructed using patient-specific computed tomography angiography images. The segmentation was created by introducing one centerline. As such, the aorta was segmented as having one true lumen and a false lumen (Fig 1, a). Aortic branches, both minor and major, were included in the model and introduced as resistances. Two-dimensional phase contrast-magnetic resonance images were used to further inform patient-specific boundary condition, and CFD simulations were performed assuming rigid-wall aortas, incompressible flow, and Newtonian fluid. Four-dimensional-flow MRI (4D-MRI) was used for a quantitative comparison between the in-vivo patient-specific and CFD-derived pressures. The simulations were performed by altering the aortic branch outflow resistance, and pressures were measured as a function of resistances. Using geometric and hemodynamic parameters, such as branch resistances, vessel dimensions, flow rate and the resulting flow pressure, an artificial neural network is being trained. Validation will be performed using synthetically generated images, and corresponding aortas.

Results, Conclusions, and Discussions:: Preliminary investigations show that there is a predictable pattern in the increase in pressure by inducing changes in aortic branches. There was effectively a 55 mmHg increase in the lumen pressure compared to the pressure immediately past the aortic valve when increasing the branch resistance. At the max percent cardiac cycle of 96.10%, there was a total flow volume of 85.65 ml with a maximum flow of 370 ml/s. Also, the CFD simulation pressures offered considerably similar values to that of in-vivo 4D-MRI (please refer to figure 1 for simulation results). Through CFD approach, relationships between intraluminal pressure, resistances, and vessel dimensions were observed which can ultimately inform of the relationship between pressure variations and aneurysmal degeneration. In this work, we present the possibility of pressure calculation using an integrated multimodality imaging-CFD framework to improve the understanding of the progression of aneurysmal degeneration in aortic dissection patients. This pipeline can be extended in the future to incorporate two separate lumens and distinguish pressure progression in each lumen. Ultimately, our objective is to understand the role of time course variations in pressure in the development of aneurysmal degeneration.

Acknowledgements (Optional): :

References (Optional): :