Resumen de Experimental and numerical framework for modelling vascular diseases and medical devices
The main objective of the present thesis is to advance towards the develop of a numerical and experimental methodology for modelling vascular pathologies and their treatments by means of intravascular devices. The studied pathologies have been the cerebrovascular accidents, pulmonary thromboembolism and atherosclerosis. While the modelled intravascular devices have been the Ni-Ti self-expanding stents and the antithrombi inferior vena cava filters.
Objectives The main objective of this thesis is the development of a methodology to study different types of pathologies in the cardiovascular system. Several experimental and computational finite element techniques have been carried out. With this purpose, the main work and their partial objectives presented in this thesis are itemized below: 1. To study the differences on stiffness along the swine carotid artery by means of a mechanical characterization (at macro level). For that purpose, cyclic uniaxial tension tests were developed, including histological quantification of collagen, elastin and smooth muscle cells for correlation between (at micro level) the macroscopic obtained results and the microscopic findings.
2. To study the microstructural orientation of collagen fibers in swine carotid arteries by means of polarization light microscopy techniques, as an additional information for mechanical characterization in terms of strain energy density function added to uniaxial tension tests.
3. To improve the understanding of the role of mechanical factors in the appearance of atheroma plaque in left coronary bifurcations. The stiffness of the vessel wall due to inner arterial pressure and heart motion was studied, considered as an additional factor in plaque location and growth by means of the finite element method. Eight patient specific left coronaries were developed in CAD models. Furthermore, fibred hyperelastic model was used to simulate the behavior of coronaries, and finally, the heart motion of the patients has been obtained from medical images by the clinicians and included in the models for a more reliable results.
4. To improve the understanding of the stability and fixation process of inferior vena cava filters due to different mechanical factors, finite element simulations of filter migration process based on experimental tests were carried out.
5. To advance towards a new design of self-expanding stent with variable radial force that decreases the expansion pressure in healthy areas of the vessel wall where exists interaction between the stent and the tissue. Meanwhile, the stent has to maintain the original expansion forces in the atheroma plaque site. Thereby, constant versus variable radial force stents were compared studying their interactions with blood vessels in finite element simulations.
Conclusions In order to a better understanding, the conclusions of this thesis are divided into three main categories: Experimental, numerical and clinical remarks.
Experimental remarks 1. Respect to the mechanical behavior of the swine carotid arteries, there are some significant differences between the mechanical behavior in the circumferential directions from proximal to distal locations. The distal specimens show a stiffer response in uniaxial tests compared with the specimens close to the aorta. Nevertheless, in the case of tensile tests in longitudinal direction, there is no significant difference between proximal and distal locations.
2. Also, a relationship was found between the microstructural composition of the samples observed in the histological analysis and the mechanical behavior of the tissue. There is a clear difference in the strain-stress curves for the circumferential oriented samples depending on the longitudinal position, stiffer for the distal ones, than could be correlated with a significant variation in the areas percentage of the microconstituents, lower proportion of elastin and higher of SMC for the distal samples.
3. Respect to the histological analysis developed to quantify the orientation and dispersion of collagen fibers in swine carotid arteries, a purely circumferential direction was found. Picrosirius Red stained histologies, in combination with double polarization techniques with U-stage, represents a good approach to measure the orientation of the collagen fibers in azimuth and elevation directions.
Computational remarks 1. The stress-stretch curves of the uniaxial tensile tests developed on swine carotids were fitted with a hyperelastic anisotropic model, and provides a good approximation of the experimental data. The model is well suited to the elastic behavior independently of the proximal and distal location samples.
2. Respect to the microstructural quantification of collagen fibers to macroscopic modelling of the mechanical behavior of swine carotid, it is worth mentioning that the use of a Bingham ODF shows a good agreement with the measured data, reproducing the elliptic distribution of the experimental results. On the contrary, the capability of the microsphere-based models together with the Bingham distribution should be further investigated to solve the poor fitting in cases where longitudinal and circumferential responses are similar. The purely circumferential direction of the collagen measured experimentally provokes a non precise fitting of the experimental tension tests in longitudinal direction. These results suggest that the rest of the subconstituent of the carotid artery play a relevant role in the anisotropy of the tissue.
3. The presented study of the influence of the myocardial contraction in the growing process of atheroma plaque in coronary bifurcations is the first attempt to investigate the contribution of this factor including arterial pressure and fibred hyperelastic formulation with in vivo patient specific data. A significant correlation between spatial distribution of luminal arterial wall stiffness and early plaque locations is highlighted in human left main coronary bifurcations.
4. The presented study of simulation of vena cava filter migration constitutes the first attempt to model this process. This is an important step towards an accurate modeling on filter design using the finite element method. The obtained results suggest that the filter becomes much more stable for small vena radius and higher strut thickness, but it could provoke a higher reaction in the cava endothelium difficulting the retrieval operation.
5. The parametric study developed by finite elements on Nitinol stents reveals a slightly higher influence of the circumferential thickness in the radial stent force, being possible a better control of the radial force changing the value of radial thickness. The effect of changing the initial diameter provokes similar decrease of the radial pressure independently of the strut section, showing less variations with higher initial diameters. Therefore, strut dimensions and shapes design play a critical role in the effect on an arterial tissue.
6. According to the results of the interaction stent-carotid artery, it is necessary to reach a compromise between the tissue stress, stent flexibility and radial force. Smaller tissue stress and better flexibility can be achieved decreasing the strut width (Ct) or radial thickness (Rt), but it may cause a negative effect in terms of the radial strength necessary to restore the lumen.
Clinical remarks 1. As a direct clinical application of the current results in stress-stretch curves of the tensile tests in swine carotid arteries, the study has shown that the specific stent deployment location along the length of the porcine carotid in animal trials can affect the final results. However, there are many other parameters that also have a great influence, such as plaque composition, vessel diameter or specimen age.
2. Identification of high-risk zones for atherosclerosis will lead to future advances in vulnerable plaque detection technology and to locally oriented preventive strategies. However, further studies are needed to extend the current work. Also suggest that the wall strain and stiffening resulting from the nonlinear mechanical behavior (due to myocardial contraction) can be an additional factor that contributes, in addition to low endothelial shear stress (ESS) and cyclic stretch, to the initiation and growing of atherosclerotic disease.
3. A clinical remark of the filter migration study is that both aspects, filter stability and fibrotic reaction of the vena cava have to be considered in the design of an ideal filter. The filter reacts differently depending on the vena diameter and the leg thickness, and specifically on the degree of contact zone of the hook and leg with the wall. In fact, for low diameters the whole hook and part of the filter legs could penetrate into the vessel wall due to fibrosis.
4. The results of the study of stent-carotid interaction show that the development of variable radial stiffness stent would improve the stenting procedure in clinical practise, and reduce the negative effect in healthy areas of the vessel.
