Right Ventricular Systolic and Diastolic Function as Assessed by Speckle-Tracking Echocardiography Improve with Long-Term Left Ventricular Assist Device Support

Abstract

Right ventricular (RV) failure is a major cause of morbidity and mortality after left ventricular assist device (LVAD) implantation. As a LVAD primarily supports the left ventricle, it is unknown whether RV pathology progresses over time leading to deterioration in RV function among patients chronically supported with a LVAD. RV longitudinal strain and strain rate as measured by speckle-tracking echocardiography provide a more sensitive, reproducible, and quantitative assessment of RV systolic and diastolic function than conventional echocardiography. Since LVADs decrease left sided filling pressure, reduce pulmonary pressure, and reduce RV afterload we hypothesized that RV systolic and diastolic function would remain stable or even improve after prolonged continuous-flow LVAD support. Echocardiograms were retrospectively reviewed from before and after implantation of a HeartMate II LVAD between 2006 and 2013 at a single center. Patients who had a baseline echocardiogram before LVAD that was performed off inotropes and a follow up echocardiogram after at least 2 months of LVAD support were included in the analysis. Patients were excluded if they required a RV assist device, had a LVAD malfunction, or had poor echocardiographic visualization of the RV endocardium. The speckle-tracking analysis software Velocity Vector Imaging (Siemens Medical) was used to calculate global RV longitudinal systolic strain, strain rate, and diastolic strain rate. Paired t-tests compared measures at baseline before LVAD to the time of maximal follow up after LVAD implantation. Of the 75 LVAD patients screened, 32 had a baseline echocardiogram off inotropes before LVAD along with a paired echocardiogram at least 2 months after LVAD implantation. 15 patients met exclusion criteria, leaving an analysis cohort of 17 patients with a mean duration of LVAD support at the time of the follow up echocardiogram of 271±129 days. Global RV longitudinal strain improved in 12 of 17 patients (70%) with the average RV strain decreasing from -7.9±2.3% before to -9.8±3.4% after LVAD (p=0.034). Similarly, RV longitudinal systolic strain rate improved in 11 patients (64%) with the average RV systolic strain rate decreasing from -0.67±0.24%/sec before to -0.99±0.35%/sec after LVAD (p=0.002). Finally RV longitudinal diastolic strain rate improved in 14 patients (82%) with the average RV diastolic strain rate increasing from 0.77±0.33%/sec before to 0.99±0.33%/sec after LVAD (p=0.029). Among LVAD patients who did not require a RV assist device, RV systolic and diastolic function as determined by speckle-tracking echocardiography improved in the majority of patients during chronic LVAD support. This suggests that LVADs prevent progression of RV dysfunction in this group of patients and emphasizes the need for strategies for temporary support of the failing RV in the initial period after LVAD implantation to allow for long term stabilization and improvement in RV function.