Simulation of TAVI in Patient-Specific Testing Inserts at Physiologic Flow Conditions

Objectives: Transcatheter aortic valve implantation (TAVI) evolved to be the standard treatment in high-risk patients with aortic stenosis. The prosthesis landing zone, characterized by the anatomy of the aortic root and the calcifications of the aortic valve, is associated to the prevalence of complications during TAVI. This study aims to develop a process that allows the simulation of TAVI in patient-specific anatomies.

Methods: A total of 100 computed tomography (CT) datasets of patients who underwent a TAVI in our clinic have been randomly selected, rendered anonymous and were digitally analyzed (3-Mensio) as follows. Classifications of the total aortic root calcium as well as calcium distribution on the aortic leaflets were conducted. Findings were correlated to the occurrence of complications (e.g., paravalvular leaks and atrioventricular blocks). Based on these results, specific anatomies were selected for 3D printing. 3D models of the aortic valve and the aortic valve calcium were generated based on the CT data. A commercial polymer printer was used to produce flexible aortic roots. These were used for in vitro simulations of TAVI procedures in a newly developed pulsatile flow system under physiologic conditions.

Results: The evaluation of the CT datasets allowed a classification of the patients based on calcification volume and distribution. We were further able to create anatomical models of a representative of each group based on the patients’ CT data. These models were successfully printed and inserted into our testing device to simulate the TAVI procedure under physiologic flow and pressure conditions. We were finally able to correlate the simulation data to the original outcome.

Conclusion: In this study, we successfully analyzed and correlated CT scans to TAVI-related complications. Based on these results a set of 3D-printed aortic roots were manufactured for in vitro testing. TAVI prosthesis evaluation under pulsatile flow conditions was possible and enabled good insights in patho-mechanical causes of TAVI-related complications. The similar behavior of our simulation compared to the actual outcome of the procedure validates our process. The potential of this process, including simulation of patient-specific TAVI in complex cases with different implants, will have to be proven with additional experiments.