Predicting the Outcome of TEVAR for Aortic Dissection

Aortic dissection is a potentially life threatening disease that begins with an initial tear in the inner most layer of the aortic wall, the intima. When blood passes through this tear the layers of the aortic wall separate creating a secondary, unwanted, channel of blood, known as the false lumen. Secondary and/or reentry tears can form in the distal thoracic aorta, abdominal aorta or the iliac arteries below the bifurcation. The reasoning for why a dissection occurs is not completely understood, however the most common risk factors are uncontrolled hypertension and atherosclerosis. Dissection can also occur due to predisposing factors (Salameh and Ratchford, 2016). As the aorta is the primary vessel for supplying oxygenated blood to the body a fault in the system can lead to a range of complications, including ischemia (and ultimately major organ failure), aortic aneurysm and rupture. The severity of aortic dissection and the resulting complications leads to a high mortality rate. Stanford Type A dissection has a mortality rate of 1% per hour from onset with a rate of 50% by day 3, while Stanford Type B has a lower, but still significant, 30-day rate of 10% (and up to 70% for the highest-risk groups) (Frank J. Criado, 2011). Typically, a Type A dissection requires immediate surgical treatment, while Type B dissections are normally treated using either endovascular or medical therapies, with the decision for which dependent on other complications the patient may be exhibiting (Salameh and Ratchford, 2016).
Given the importance of thrombosis to patient prognosis, it is highly desirable to be able to predict the thrombosis process when treatment is administered. A mathematical thermodynamics-based model (Menichini and Xu, 2016) was developed in idealized models to predict the thrombosis process in Type B dissections. It has since been applied to patient specific geometries to investigate the accuracy of the model (Menichini et al., 2016), and produced promising results with model outputs being comparable to in vivo data collected from the patients. Looking at the second year and onwards of the current PhD, one of the objectives will be to critically analyse this model. This will include carrying out an in-depth sensitivity analysis of the assumptions made, which may lead to the expansion of the model - for example, one desirable factor to investigate and potentially include is fluid-structure interactions to simulate the non-rigid wall of the aorta. There have also been numerous kinetic-based models produced, incorporating the complex biological and chemical mechanisms involved in thrombosis formation (Anand et al., 2006; Biasetti et al., 2012; ?; Tolenaar et al., 2014; Filipovic et al., 2008; Leiderman and Fogelson, 2011). An in-depth review of these models and comparison to the previously discussed hemodynamic model will be conducted to determine key differences in results produced by all models.