A computational and experimental investigation of the biomechanics of nickel-titanium stents for femoropopliteal applications

Self-expanding nickel-titanium stents are the current gold-standard in treating peripheral artery disease in the femoropopliteal artery. Despite this, failure rates in the femoropopliteal artery are much higher than other arterial beds. While the underlying reasons for these negative outcomes remain unclear, some of the primary contributors include stent fractures, arising from the severe level of deformation encountered during limb flexion, and in-stent restenosis, which is caused by altered haemodynamics due to the presence of the implanted stent. Finite element modelling has emerged as a powerful tool to predict the biomechanics of stents, providing detailed insight into functional performance and can even be used to supplement bench testing data for regulatory submissions as part of the certification process. Despite this, there has still been a lack of consensus across the literature on robust methodologies to ensure verification and validation of models of nickel-titanium stents, with detailed guidelines on model credibility for medical devices only recently developed (ASME VV-40, 2018). The objective of this thesis is to investigate the biomechanics of femoropopliteal artery self-expanding nickel-titanium stents using both computational and experimental methods. In particular, the thesis investigates manufacturing processes, in vitro bench performance, and procedural outcomes of these stents through the use of patient-specific modelling and addresses the key issues of model verification and validation through the development of robust experimental comparators. In the first study, this thesis developed an experimentally-validated computational model to predict stenting procedural outcomes in a patient-specific artery model and evaluatesthe effects of stent oversizing (OS), defined as the ratio between stent nominal diameter and vessel diameter, on short- and long-term indicators of clinical performance. It was found that OS < 1.2 resulted in problematic short-term outcomes, with poor lumen gain and significant strut malapposition. Oversizing ratios that were between 1.2 ≤ OS ≤ 1.4 provided the optimum biomechanical performance following implantation, with improved lumen gain, reduced incomplete stent apposition and favourable predicted long-term fatigue performance. Excessive oversizing, OS > 1.4, did not provide any further benefit in outcomes, with limited increases in lumen gain and unfavourable long-term performance, with higher mean strain values predicted from the fatigue analysis. A detailed credibility assessment of finite element modelling of self-expanding Ni-Ti stents was then conducted through verification and validation activities, as per ASME VV-40 standard. In particular, the role of calculation verification, model input sensitivity, and model validation was examined across three contexts of use (COUs), namely radial compression, stent deployment in a vessel, and fatigue estimation. Generally, it was found that the chosen numerical parameters influenced global and local outputs across all COUs considered, although to varying levels of sensitivity. It was found that both global and local quantities were highly sensitive to the mesh discretisation. For explicit analysis, it was concluded that model sensitivity to mass scaling or time incrementation parameters should be investigated independently, irrespective of whether the ratios of kinetic and internal energies are below 5%. Model input sensitivity analysis highlighted the importance of capturing the geometric dimensions of the stent, in particular the strut width, while model results are less sensitive to Ni-Ti material parameters, suggesting that device-level calibration of material behaviour are reasonable. Finally, the validation of stent deployment through the use of in vitro mock vessels offered a simple and accurate validation method when predicting diameter gain, and lumen area, provided that the material of the vessel is appropriately characterised and modelled. A study was then carried out to optimise the manufacture of a flow-enhanced stent by inducing a helical ridge onto laser-cut and wire-braided stents. The process consisted of a shape deformation followed by a shape-setting, whose parameters were thoroughly explored, and their effects assessed in terms of mechanical performance of the device, material transformation temperatures and surface finishing. It was found that the combination of 500°C/30min provided mechanical properties comparable with the original design, and transformation temperatures suitable for stenting applications (Af = 23.5°C). Microscopy analysis confirmed that the manufacturing process did not alter the surface finishing. Deliverability testing showed the helical device could be loaded onto a catheter delivery system and deployed with full recovery of the expanded helical configuration. This demonstrates the feasibility of an additional heat treatment regime to allow for shape forming of laser-cut and wire-braided devices that may be applied to further designs. Based on one of reference laser-cut devices that were developed and manufactured, a computational framework was implemented to investigate its behaviour under realistic deformations of the femoropopliteal environment. Patient-specific data from CT scans of human cadaveric models were obtained at different ranges of knee flexion and were used to inform the loading conditions in the computational model. Using a finite element model, the 3D reconstruction of the baseline artery in the extended position was subjected to the different ranges of knee-flexion deformations, where the artery conformity of and levels of arterial stress were assessed. In the geometry of the artery at each limb-flexion configuration, the Ni-Ti stent was delivered in correspondence to the adductor hiatus through a morphing procedure, and local strain distributions were compared to evaluate the mechanical behaviour of the stent when deployed in extreme arterial configurations. It was found that there was a general increase in the level of arterial stress with increasing severity of leg flexion, although effects were inhomogeneous along the vessel length with deformations localised either in the proximal-mid segment or distally. A comparison of stent deployment at the different limb-flexion configurations allowed to appreciate a non-uniform deformation in the device, with a smooth kink/bend observed at proximal segment in correspondence of the most severe leg flexion position. Interestingly, higher levels of strain in the stent were not associated to the entity of the vessel curvature, but rather to a smaller diameter/flattening of the artery. Together, the outcomes from this thesis have provided detailed insight into the biomechanics of several femoropopliteal artery self-expanding nickel-titanium stents, with a particular focus on understanding and improving their performance in clinically relevant scenarios. Furthermore, the experimental data and computational frameworks developed can form a benchmark for the future development of rigorous and robust approaches to verification and validation for finite element models for self-expanding nickel-titanium stents.

Publisher

NUI Galway

Rights

Attribution-NonCommercial-NoDerivs 3.0 Ireland
CC BY-NC-ND 3.0 IE

cbn

Collections

University of Galway Theses (PhD Theses)