« A Novel model accounting for collateral flow to predict the true severity of a left main coronary stenosis with downstream lesions and a CFD analysis of the impact of malapposed and overlapping stents on hemodynamics. »

Place : Amphithéâtre IAB, Site Santé, La Tronche
 

Jury & Supervision :

• Manuel LAGACHE, Maître de Conférences, Université Savoie Mont-Blanc, Polytech Annecy-Chambéry, Le Bourget du Lac, Supervisor
• Jacques OHAYON, Professeur, Université Savoie Mont-Blanc, Polytech Annecy-Chambéry, Le Bourget du Lac et Laboratoire TIMC-IMAG, UGA, CNRS UMR 5525, Grenoble,  France, Co-supervisor

Jury :

• Anne-Virginie SALSAC, Directrice de Recherche, Université de Technologie de Compiègne, Sorbonne Université, Reporter
• Valérie DEPLANO, Directrice de Recherche, IRPHE UMR 7342, Reporter
• Gérard FINET, Professeur des Universités - Praticien Hospitalier, HCL, Université Claude Bernard Lyon 1, INSERM Unit 886, Examiner
• Laurence Chèze, Professeure des Universités, LBMC UMR T9406, Laboratoire de Biomécanique et Mécanique des Chocs, Université Lyon 1 - IFSTTAR, Examiner

Abstract:

Cardiovascular diseases are the main cause of mortality in developed countries, representing one third of all deaths. Coronary artery disease (CAD) is the most common cardiovascular pathology. Due to the complexity of CAD, it has become the field of study of researchers among various disciplines. The present thesis focuses on cardiovascular biomechanics and aims to assist interventional cardiologists who are dealing with two clinical situations: 1) the assessment of coronary lesions in complex configurations (i.e. bifurcations with multiple stenoses and collateral flow) and 2) stent-induced hemodynamic disturbances and their link with poor clinical outcomes.
The first part of my PhD research was dedicated to the study of coronary lesions at left main arterial bifurcations with downstream stenoses at the daughter branches. It is known that the apparent severity of a left main stenosis is affected by the presence of concomitant lesions. Moreover, since stenoses reduce blood flow to certain areas of the myocardium, collateral arterial connections tend to develop as alternative paths to irrigate the affected heart tissue. Fractional flow reserve (FFR) the standard clinical technique to evaluate the severity of a coronary stenosis, is only valid for isolated lesions. Therefore, an equation was derived to adapt this technique to multi stenosis configurations and improve the severity assessment of the left main stenosis. This could help interventional cardiologist choose the best treatment for each patient. Previous FFR prediction models were extended by considering: 1) up to two downstream lesions (i.e. one at each branch) and 2) the presence of collaterals. Then, the mentioned equation was validated using an in vitro model (developed in-house as part of this work) able to mimic the pulsatile flow conditions in coronary bifurcations. This experimental platform allowed us to adjust the severity of downstream lesions and the degree of collateral circulation. The proposed model significantly improved the FFR prediction compared to previous approaches (i.e. improvement factor of 2).

In the second part of my PhD research, I focus on percutaneous coronary intervention and stent implantation, the most common treatment for blocked coronary arteries. Stent placement is not always ideal, it might not be fully in contact with the arterial wall (malapposition) or, depending on the lesion configuration, two stents may overlap. The impact of different degrees of stent misalignment issues (i.e. malapposition and overlapping) on hemodynamics and their relationship with complications like restenosis and thrombosis were numerically studied. We found that previous computational fluid dynamics (CFD) studies on this topic worked either with complex models (patient specific 3D geometries) or too simplified ones (2D geometries, static analysis). Both groups have successfully identified regions with malapposed and overlapped stent struts as dangerous due to the appearance of abnormal shear stress on the vascular walls and highly disturbed flow. However, to our understanding, a complete parametric study designed to highlight critical malapposition and overlapping configurations, for which flow disturbance over a cardiac cycle becomes significant, has never been conducted. This is why we proposed to use parametric axisymmetric models simpler than 3D ones to let us compute several cases but also using more realistic conditions than previous 2D studies (i.e. pulsatile flow, non-Newtonian behavior for blood and the most used flow-related indices to assess the hemodynamic impact on the vessel wall over a cardiac cycle). This biomechanical study provided valuable information that could be used to improve stent deployment protocols and stent design.