CARD19: In vitro and In vivo Testing of the Pulsatile CorWave Membrane LVAD

Abstract

Background: Lack of physiologic pulsatility with the use of continuous flow (CF) LVADs has been associated with serious adverse events including stroke, bleeding, and aortic insufficiency. CorWave is developing a unique LVAD employing an undulating membrane to generate a physiologic pulse while improving hemocompatibility. The present studies focused on in vitro and in vivo testing of the pump control algorithms for hemodynamic performance and hemocompatibility. Methods: The pump actuator, membrane, and blood flow path were designed using respectively electromagnetic simulations (MAGNET), Fluid-Structure Interaction (COMSOL), and Computational Fluid Dynamics. Algorithms were developed for actuator control, synchronization with the native left ventricle (LV), suction detection, and adaptive physiologic control. They were tested in a mock circulation loop (MCL) with INTERMACS 1-2, 3-4, and 5-6 heart failure (HF) profiles using LV pressure-volume loops to examine performance. The algorithms were then tested in chronic implants in healthy sheep for up to 3 months, and in acute implants in sheep with induced HF (ejection fraction < 40%). Results: During MCL tests, in synchronous pulsation (SynchPuls) mode, the membrane pump generated a more physiologic PV loop, with lower mean left atrial pressures (LAP), increased native LV stroke volume, and 1:1 opening of the aortic valve. CF pumps generated a characteristic, non-physiologic narrow PV loop with elevated mean LAP and aortic valve opening once per 3-5 beats. These results demonstrated the SynchPuls algorithm could respond to changes in pre-load to better unload the LV, restore more physiologic hemodynamics, and ensure aortic valve opening, so we proceeded to animal trials, which present more dynamic and challenging conditions. During acute implants in sheep with HF, using SynchPuls, the membrane pump increased cardiac output from ~2 to ~4 LPM. The physiologic control algorithm adjusted pump output to prevent LV suction and retrograde flow during hemodynamic manipulations, including epinephrine administration, and head up or down tilt. The algorithms switched to continuous mode, as intended, when arrhythmias or tachycardia were detected. In chronic implants, the SynchPuls algorithm successfully synchronized with the LV for > 97% of heartbeats and adapted to circadian cycles. Biomarkers of end-organ function returned to baseline within 2 weeks of implant, hemolysis was < 20 mg/dL, and VWF activity was preserved during the chronic implants. Conclusion: These studies showed that the membrane pump can provide physiologic pulsatility and restore cardiac output in HF models. The algorithms' robustness was confirmed in chronic implants. Overall, these successful tests demonstrated that the CorWave membrane pump can deliver the hemocompatibility, pulsatility, and adaptability required to advance the LVAD field. CARD20

Mock circulatory loop applications for testing cardiovascular assist devices and in vitro studies

“…However, with MCL, cheaper tests on the performance of VADs and their control strategies can be conducted in various CVS conditions. The LVAD performance tests are usually conducted in vitro before the in vivo tests, which is more dynamic and challenging ( Snyder et al, 2022 ). Related studies tested the LVAD hydraulic performance in MCL for multiple operating conditions, and meanwhile in vivo tests were conducted for specific operating conditions to evaluate the overall performance ( Oh et al, 2005 ; Jahanmir et al, 2008 ; Murakami et al, 2008 ).…”