(P181) DEVELOPMENT OF A HIGH-FIDELITY SIMULATOR TO PREDICT THE RISK OF CORONARY OCCLUSION IN HIGH-RISK VALVE-IN-VALVE PROCEDURES

Background: The historical trend towards the use of bioprosthetic aortic heart valves led to an increasing number of bioprosthetic valve failures. In patients at high-risk for re-operation, transcatheter aortic valve-in-valve (ViV) has become an alternative therapeutic option, but can be limited by the risk of coronary occlusion. The objective is to develop a high-fidelity simulator based on 3D printing of the patient's specific cardiac anatomy to accurately predict the risk of coronary occlusion post Valve-in-Valve procedures.

METHODS AND RESULTS: Patients judged to be at high risk of coronary occlusion defined by a transcathether valve to coronary ostia distance (VTC) < 4mm, are included in the study. The patients' cardiac anatomies are modeled by deep learning and then printed in 3D (BIOMODEX, Paris, France). Optimization algorithms are applied to give the model true haptic tactile feedback and tissue characteristics similar to those of cardiac structures allowing optimal imaging of coronary arteries in fluoroscopy. The model is coupled to a pulsatile pump generating physiologic systolo-diastolic flow circulation in the aorta and coronary arteries. The patency of the coronary arteries is evaluated with an aortography and with a hemodynamic study of the coronary flow using a high-fidelity pressure guide placed in each coronary to measure the Instantaneous Wave-Free Ratio (IFR) before and after the deployment of the prosthesis. A 75-year-old patient having had an aortic valve replacement 6 years prior with a 21 mm Mitroflow prosthesis presented with symptomatic prosthesis degeneration. The CT scan study found a VTC of 2.8 mm making the patient theoretically ineligible for a ViV procedure. The patient's cardiac anatomy was modeled, printed in 3D and then installed in the simulator. The deployment of an Evolut Pro+ prosthesis of size 23 mm was then carried out. The IFR remained stable before and after deployment and the aortography showed patency of both coronary arteries. Based on the encouraging results of the simulation, the procedure was performed the next day on the real patient with results similar to those observed during the simulation session both for the IFRs and for the patency of the coronary arteries on aortography. The final mean gradient was 12mmHg and only mild paravalvular leak was present.

Conclusion: The development of patient-specific simulators can help to identify sub-groups of patients at high-risk for coronary obstruction in whom ViV might be feasible and safe. New human-grade simulation platforms may have an important role in structural heart intervention planning.