Stent-based treatment for aortic dissections

Problem statement

A type B aortic dissection is a cardiovascular disease, implying that there is a delamination of the inner part of the aortic wall, over a certain length in the descending aorta. Due to this delamination, a second channel (false lumen), parallel to the normal pathway of the blood (true lumen), is formed through which the blood can flow (see figure 1). The two parallel channels are usually connected by tears in the delaminated membrane. Although this disease is rather rare (1.8/100 000), severe complications can occur, if left untreated (DeMartino et al., 2018).

Nowadays, thoracic endovascular aortic repair (TEVAR) is commonly used to treat type B aortic dissections. This treatment implies that an endoprosthesis, a stent-graft, is implanted in the aorta with the aim to seal off the entry tear (see figure 1). This can, then, lead to thrombus formation of the false lumen and, therefore, healing of the aortic dissection. In reality however, clinicians observed that only a part of the patients developed complete thrombus formation of the false lumen after TEVAR. Moreover, re-interventions within the first year are often required and a persisting dilation of the aorta has been noticed in 30% of the patients as well (Fattori et al., 2013). Although it was found that TEVAR is a promising treatment, this indicates that the effect of TEVAR on the aortic dissection is not yet fully understood.

Therefore, this research aims at gaining insight in and enabling the prediction of the acute and long-term effects of TEVAR on type B aortic dissections by developing biomechanical computational models, on a patient-inspired level.

Figure 1: Type B aortic dissection before and after implanting an endoprosthesis, during TEVAR. After the treatment, partial thrombus formation of the false lumen was observed.

Towards a computational framework for TEVAR in type B aortic dissections

Starting from a database including CT scans of 96 patients with a type B aortic dissection, of whom 45 were treated with TEVAR, idealized as well as patient-inspired 3D geometries of type B aortic dissections are developed. This database was collected in collaboration with the university hospitals of Ghent and Düsseldorf. The geometrical information is used as starting point for the models representing the acute and long-term effects. Although the long-term effects are essential in assessing the benefit of TEVAR for a particular patient, the current focus of the research project is to set up the models that represent the acute effects of the treatment. In particular, models for the stent-graft deployment, the blood flow, the dissected aortic wall, as well as their interaction, are being developed. Hereby, particular attention is given to the robustness of the models. Indeed, it is the aim that the usage of the model set-up is not limited to a few patient-specific geometries, but that the models can be used for a broad range of patients with a type B aortic dissection. Moreover, assessing the sensitivity of multiple assumed model parameters is one of the essential steps in developing a reliable computational model. Therefore, parametric models were created. The first steps of the modeling process were taken in idealized geometries.

Aortic wall
A simplified parametric model of an aortic dissection has been set up. Although the geometry, developed in pyFormex, is strongly simplified, it contains the main features of an aortic dissection, being the false lumen and multiple tears. While an idealized geometry has been used, an advanced hyperelastic anisotropic material model was included for the aortic walls. The finite element analyses were performed in Abaqus.

Stent-graft deployment
A finite element model of the stent-graft deployment, is currently under development. The deployment of a self-expanding one-ring stent has been studied extensively. Some initial steps have been taken towards the expansion of a stent in a dissected wall model, wherein the tissue is assumed to be isotropic and hyperelastic. As an illustration of the technical possibilities, an animation of the stent-graft deployment in an aneurysm model, developed by Dr. Sander De Bock, is shown in figure 2.

Blood flow
Starting from a similar geometry as the one of the aortic wall model, a computational fluid dynamics model of the blood flow has been set up, using Fluent. With the future perspective of going towards a fluid-structure-interaction model, the blood flow is modeled using overset meshes. This technique allows large displacements of the aortic wall and the dissected membrane during the cardiac cycle without the need to adapt the fluid domain during the simulation.

Figure 2: Computational model of the deployment of an endoprosthesis in an abdominal aneurysm.

In conclusion, initial steps have been taken for the three models required to predict the acute effects. Future work will be situated in assessing the sensitivity of the models and increasing the geometrical complexity in order to progress towards a patient-specific setting. The interaction between the different models will be taken into account as well.

IBiTech researchers currently active on the project

• Patrick Segers (contact)
• Nele Famaey (contact)
• Lise Gheysen (contact)

Funding sources

• Doctoral grant strategic basic research (1S48920N) of the research foundation-Flanders (FWO-Vlaanderen)
• Grant research project (G086917N) of the research foundation-Flanders (FWO-Vlaanderen)