Finite element analysis and material property modelling of thin-strut polymeric bioresorbable scaffolds

Bioresorbable coronary scaffolds (BRS) offer a potential fourth revolution in interventional cardiology. Despite poor initial results, improvements in pre-processing technologies have facilitated a second generation of thin-strut designs which have shown promise in pre-clinical and early clinical trials. Therefore, the investigation of BRS design via in-silico and in-vitro methods is critical to maximise their clinical efficacy. After a review of the literature concerning the mechanical performance of coronary stents/scaffolds, the following aims were decided upon to; (i) investigate the equivalent plastic strain (PEEQ) within the struts of BRS and its relationship with scaffold design; (ii) achieve an improved consensus upon which material model(s) may be most appropriate for capturing the mechanical response of thin-strut BRS; and (iii) deploy a BRS into coronary artery geometry and design a novel BRS for use in a challenging clinical scenario. Firstly, finite element analysis was conducted of the crimping, expansion and radial crushing of 20 scaffold designs comprising variations in ring length and strut width. Surrogate models were developed to explore the effect of the design variables upon the radial strength, post-expansion diameter, percentage recoil and cell area. The PEEQ distribution was observed along paths in critical locations. PEEQ was up to 2.4 times higher in the wide-strut designs and these also exhibited a twisting behaviour at the scaffold end rings. Whilst the post-expansion scaffold diameter was found to be accurately predicted by the in-silico tests, the radial strength and percentage recoil predictions were inaccurate, necessitating an improved material model. A study was then conducted to explore the effect of different isotropic and anisotropic linear-elastic plastic models in predicting BRS mechanical behaviour. Stress-strain data obtained at different displacement rates, the use of isotropic and transversely isotropic elastic theories, the use of isotropic and anisotropic hardening, as well as the implementation of stress relaxation in the plastic regime of the material, were all explored. A novel anisotropic material model, the Hoddy-Bressloff (HB) model, was developed which achieved prediction of the specific radial strength within 1.1% of the in-vitro result and the scaffold’s post-expansion diameter within 6.7%. The viscoelastic-plastic Bergstrom-Boyce (BB) model was also investigated and this achieved prediction of the post-expansion diameter and radial strength within 2% of the in-vitro result. A multistep crimping process that utilised holding to facilitate stress relaxation and increased ambient temperature were found to improve prediction of the post-crimping scaffold diameter in silico. The baseline BRS successfully revascularised a diseased arterial model after in-silico balloon expansion and exerted both lower maximum and volume averaged stresses on the vessel wall compared to a metallic stent. This investigation highlighted the isotropic limitation of the BB model which resulted in the post-expansion strut configuration being misrepresented and so did not capture certain phenomena predicted by the anisotropic HB material model. Lastly, the novel scaffold for use in the clinically challenging scenario of a left main coronary artery bifurcation was deployed both in silico and in vitro into a patient specific model. The BRS was found to successfully revascularise the diseased part of the vessel but fracture did occur in the proximal rings and strut twisting was also evident, both of which were predicted by the HB model