Risk factors for the development of right ventricular failure after left ventricular assist device implantation-a single-centre retrospective with focus on deformation imaging.

Abstract

Right heart failure (RHF) after the implantation of a left ventricular assist device (LVAD) remains a dreaded postoperative complication. Using 2D speckle-tracking echocardiography, it is possible to acquire right ventricular global and regional function. The aim of our study was to assess whether speckle-tracking echocardiography values will better predict the RHF post-continuous-flow LVAD implantation. From January 2014 to January 2016, 54 patients who underwent LVAD implantation were included and retrospectively analysed. Clinical, invasive haemodynamic (right and left heart catheterization), laboratory and transthoracic echocardiography data were reviewed. Multivariable logistic regression was performed using RHF as dependent variable. Thirteen patients (24%) developed RHF. These patients had significantly elevated procalcitonin (P = 0.011), elevated central venous pressure (CVP) pre- and post-LVAD implantation (P = 0.002 and 0.031, respectively), higher right ventricular (RV) and pulmonary systolic pressure (P = 0.016 and 0.013, respectively), higher Michigan Risk Score (P = 0.001) and a lower peak systolic longitudinal strain of the basal RV free wall (P = 0.032). Haemoglobin, procalcitonin, RV systolic pressure, basal right ventricular free wall and pre-CVP entered the final multivariable analysis, only basal right ventricular free wall (P < 0.001) and pre-CVP (P < 0.001) remained significant predictors of RHF. The sensitivity and specificity of the final model were 85.7% and 95.4%, respectively. The negative predictive value reached 94%. 2D strain parameters of the RV free wall seem to be auspicious for RV function and predicting RHF. Moreover, intraoperative CVP should not be neglected since elevated values proved to be highly associated with RHF. Our results represent a valuable supplement to other scores by considering both echocardiography and intraoperative data.