Geometry parameterization and multidisciplinary constrained optimization of coronary stents.

TL;DR: This study proposes a technique to parameterize stent geometry, by varying the shape of circumferential rings and the links, and assess performance by modelling the processes of balloon expansion and drug diffusion, and shows that a larger amplitude of the links with minimum curved regions is desirable for improved flexibility, average stresses and drug delivery.

Abstract: Coronary stents are tubular type scaffolds that are deployed, using an inflatable balloon on a catheter, most commonly to recover the lumen size of narrowed (diseased) arterial segments. A common differentiating factor between the numerous stents used in clinical practice today is their geometric design. An ideal stent should have high radial strength to provide good arterial support post-expansion, have high flexibility for easy manoeuvrability during deployment, cause minimal injury to the artery when being expanded and, for drug eluting stents, should provide adequate drug in the arterial tissue. Often, with any stent design, these objectives are in competition such that improvement in one objective is a result of trade-off in others. This study proposes a technique to parameterize stent geometry, by varying the shape of circumferential rings and the links, and assess performance by modelling the processes of balloon expansion and drug diffusion. Finite element analysis is used to expand each stent (through balloon inflation) into contact with a representative diseased coronary artery model, followed by a drug release simulation. Also, a separate model is constructed to measure stent flexibility. Since the computational simulation time for each design is very high (approximately 24 h), a Gaussian process modelling approach is used to analyse the design space corresponding to the proposed parameterization. Four objectives to assess recoil, stress distribution, drug distribution and flexibility are set up to perform optimization studies. In particular, single objective constrained optimization problems are set up to improve the design relative to the baseline geometry—i.e. to improve one objective without compromising the others. Improvements of 8, 6 and 15% are obtained individually for stress, drug and flexibility metrics, respectively. The relative influence of the design features on each objective is quantified in terms of main effects, thereby suggesting the design features which could be altered to improve stent performance. In particular, it is shown that large values of strut width combined with smaller axial lengths of circumferential rings are optimal in terms of minimizing average stresses and maximizing drug delivery. Furthermore, it is shown that a larger amplitude of the links with minimum curved regions is desirable for improved flexibility, average stresses and drug delivery.