Mitigating Post Transcatheter Aortic Valve Replacement Thrombogenic Risk: Design And Optimization Of Novel Aortic Valve Devices

Introduction:: Transcatheter aortic valve replacement (TAVR) is a minimally invasive procedure used to treat severe aortic stenosis. The number of TAVR procedures is constantly increasing, surpassing surgical aortic valve replacement (SAVR). TAVR is expanding to younger, lower-risk patients including bicuspid aortic valve (BAV) patients [1,2]. However, TAVR is associated with several clinical complications, including long-term, life-threatening thrombotic events that may be a consequence of prolonged exposure of platelets to elevated cyclic stresses. Flow-induced thrombogenicity increases the risk of platelet activation, and microthromboemboli formation, which can lead to prosthetic leaflet thrombosis. The rapid expansion of TAVR into younger and lower-risk patients highlights the importance of developing alternatives to current tissue-based TAVR devices that may suffer from structural deterioration and thrombus formation – as the device must remain functional and safe for the long-term. New polymer technologies facilitate the creation of more durable TAVR valves and mitigate adverse complications [3,4]. Optimization of polymeric TAVR leaflets can reduce the risk of flow-induced thrombogenicity and mitigate the need for post-TAVR anticoagulation therapies. Optimizing valve design to improve hemodynamics and minimize the associated potential thrombotic events is the preferred approach. The goal of this study is to use advanced computational modeling approaches to assess the thrombogenic potential and hemodynamic performance of current commercial TAVR valves. We demonstrate how these devices compare to our dedicated polymeric TAVR leaflet designs for both trileaflet and bicuspid aortic valve patients, which are optimized in multiple steps involving various design features.


Materials and Methods:: Commercial TAVR models were generated for older, self-expandable 26-mm CoreValve and newer generation 26-mm Evolut R devices, in addition to the balloon-expandable 26-mm SAPIEN. The valve models were generated along with their stents as described in our previous work [5]. With the constructed leaflets and stents (Fig. 1a), TAVR deployments were simulated in ABAQUS Explicit v2019 (Dassault Systemes Simulia Corp., Johnston, RI) and the resulting post-deployment geometries were taken into ANSYS LS-DYNA to perform FSI analyses for two beat cycles (Fig. 1b). To validate the constructed TAVR valves, effective orifice area (EOA) calculations and velocity profiles were compared to results in the literature [6]. Following the validation, each FSI simulation was post-processed using ANSYS EnSight in which ~50,000 platelets were seeded in the systolic flow via the implementation of Lagrangian particles. Stress accumulation (SA) was calculated and tracked for all the particles seeded in the fluid domain. With this data, the probability density function (PDF) of the accumulated stresses of the large population of flowing platelets was computed for each valve type (Fig. 1c), which represents the “thrombogenic footprint” for each respective device [7]. To analyze the thrombogenic potential of different leaflet designs, the PDFs of the various commercial devices were compared amongst each other and to multiple parametric trileaflet valve geometries that have varying belly and attachment profiles. Additionally, the thrombogenic potential of our BAV-dedicated TAVR design, PolyV-B, was assessed.


Results, Conclusions, and Discussions:: The FSI results for the commercial valves showed peak systolic jet velocities in the range of 2.5 to 3.0 m/s, comparable to particle image velocimetry data found in the literature [6]. The thrombogenic risk simulations were performed for the commercial valves, followed by the polymeric TAV valves and PolyV-B. Upon comparison of the PDFs for the polymeric TAV valves, different degrees of stress accumulation were highlighted. Valve designs #3-5 showed more stress accumulation on a much larger number of platelets than valve designs #1-2. Valves designed with attachment regions with large tapering angles (least-gradual curvature), and comparatively large surface areas tended to show higher thrombogenic risk due to excessive leaflet fluttering, irregular opening shapes, and other characteristics that would likely activate platelets over time (Fig. 1d). For the commercial valves, the 26-mm SAPIEN and CoreValve devices outperformed the Evolut R. The various commercial valves demonstrated higher thrombogenic risk when compared to valve design #1, thereby highlighting that the commercial leaflets can be better optimized for mitigating flow-induced thrombogenicity.


This study demonstrates the ability to analyze post-TAVR thrombogenic risk by studying hemodynamics of prosthetic devices via a fully integrated TAVR FSI approach. The devices studied here include both self-expandable and balloon-expandable commercial valves currently on the market, old and new generation devices, novel polymeric TAV devices, and our BAV-specific device, PolyV-B. More specifically, we can analyze the impact of TAVR leaflet designs on hemodynamics and thrombogenicity to develop novel devices that are safe and durable with reduced thrombogenicity for younger patient cohorts. Limitations of this study include idealized anatomies for TAVR deployments. Nevertheless, this study points to key parameters of TAVR leaflet design critical for the mitigation of thrombogenic risk and other essential hemodynamic criteria.


Optimized TAVR devices of the future may finally reduce the need for post-TAVR antithrombotic therapies. Future work will include analyzing the thrombogenicity of these novel devices in patient-specific anatomies. Additionally, we plan to perform our own in-vitro hemodynamics testing (ViVitro) with TAVR devices that we are developing.


Acknowledgements (Optional): : Funding provided by NIH-NIBIB Phase II-U01EB012487 (DB); NIH-NIBIB 1U01EB026414-01 (DB). Industry Partners: ANSYS, Dassault Systemes.

References (Optional): : [1] Travaglino, S et al., J Biomech, 142(1) (2020).
[2] Li, K et al., Int J Num Meth Biomed Eng, 33(3): e02814 (2017)
[3] Abbasi, M et al., J Mech Behav Biomed Mat, 102: 103491 (2020)
[4] Kovarovic, B et al., J Biomech Eng, 144(6) (2022)
[5] Helbock, R et al., Annals of Biomed Eng, (2022): 1-13
[6] Hatoum, H et al., JTCVS Open, 9 (2022): 43-56
[7] Xenos, M et al., J Biomech, 43(12): 2400-2409 (2010)