Hemodynamic Changes Associated with Sub-Optimal Inflow Cannula Angle in the Heartware HVAD - A Hemostatic Model

Abstract

Pump thrombosis is the epitome of left ventricular assist device dysfunction for end-stage heart failure patients. With the increased utilization of implantable, long-term, left ventricular assist devices (LVADs), understanding the implications associated with device orientation and interaction with the body is exceedingly important. Components associated with pump thrombosis in the Thoratec© HeartMate II™ (HMII) and the HeartWare© HVAD® devices include the inflow cannula, the outflow graft, and the pump elements as well as pump pocket depth for the HMII specifically. Several studies have been conducted to analyze these interactions with the HMII, however there is minimal to no data available analyzing how the device orientation of the HeartWare HVAD affects hemodynamics and a patient’s risk for developing pump thrombosis. Therefore, the purpose of this pilot study is the simulate the hemodynamic implications associated with Sub-optimal cannula angulation of the HVAD. Using Solidworks 2016 Ed., a simplified, hemo-static model of the left side of the heart was created. Dimensions for the atria, ventricle, and mitral valve were determined through the combination of Trans-Esophageal Echo cardiogram data as well as literature references. Three different inflow cannula angle scenarios were developed including a Control, a Clinically Optimal, and a Sub-optimal. Assumptions included body temperature, no accumulation within the ventricle, and no ejection or contraction. The model consists of static continuous flow set to 5 liters per minute with the assumption that the HeartWare HVAD is completely supporting the left ventricle. The results include both qualitative and quantitative data. Flow trajectory plots for each cannula scenario depict the hemodynamic flow patterns for different time points. Results show visible changes in the Sub-optimal orientation when compared to both the Control and the Clinically Optimal scenario. Additionally, it was determined that there were no statistically significant differences in the velocity vectors for any of the scenarios however, the shear stress values were determined to be significantly different for all time points, p < 0.001 for all scenarios when compared to Control. Though there are several limitations of this study, with sub-optimal inflow cannula angulation, there is a potential increased risk of hemolysis due to increased shear stress.