Computational investigation of the influence of LVAD inflow cannula on left ventricle hemodynamic in advanced heart failure

In the last three decades, LVADs have improved hemocompatibility, making them a viable long-term treatment option for individuals with advanced heart failure. Despite reduced complications, the rate of stroke remains high. The LVAD inflow cannula placement impacts the hemodynamics in the Left Ventricle (LV), potentially creating thrombogenic flow patterns. The objective of this research is to analyse, via CFD simulations, the performance of two LVAD cannulae: one deeply implanted into the LV (HeartMate3) and the other flushed with the myocardium wall (Evaheart2). Both cannulae were virtually implanted on the apex of a patient-specific LV. The aortic valve was closed, as is common in LVAD patients, so flow enters the ventricle at the mitral valve and exits through the cannula. Two operating conditions are explored: constant 5 l/min flow rate and pulsatile waveforms from a lumped parameter model that includes the pumps' characteristic curves. Analysis focuses on Eulerian metrics to estimate LV washout and quantify blood stasis. Favourable hemodynamics are associated with a flushed cannula design (Evaheart2) compared to a protruding cannula. This is coupled with the EH2 high flowrate amplitude, that improves LV washout and reduces recirculation at the LV apex. This Eulerian approach will be complemented by a Lagrangian platelet tracking approach, via massless flow tracers, which will compute actual residence time and shear stress history, and help model prothrombotic platelet behaviour.