Predicting RV Failure after LVAD Implantation: Utility of Echocardiographic Strain and Defining RV Failure

Abstract

Purpose Post-operative right ventricular failure (RVF) is a major source of morbidity and mortality in left-ventricular assist device (LVAD) recipients. Predicting and defining RVF is challenging. Methods and Materials Prospectively collected pre-op echo and hemodynamic data of LVAD recipients were analyzed including RV mid free wall longitudinal myocardial strain (LMS) by speckle tracking. Four definitions of RVF were used: 1) permanent right ventricular assist device (pRVAD), 2) pRVAD or temporary RVAD (any RVAD), 3) any RVAD or 14 days of post-op inotropic therapy (14d), or 4) any RVAD or 7 days of post-op inotropic therapy (7d). Results Of the 41 pts, 8 (20%) had RVF defined by pRVAD, 14 (34%) by any RVAD, 19 (46%) by 14d, and 32 (78%) by 7d. RVF was associated with lower LMS and cardiac index (CI) and higher right atrial pressure (RAP) in all outcome groups except 7d (table). ROC analysis yielded increasing LMS cutoffs for RVF defined by pRVAD (LMS −9.7%: sensitivity/specificity 75%/85%), any RVAD (LMS −11.4%: 71%/70%), and 14d (−13.2%: 79%/50%). By multivariate analysis, LMS independently predicted RVF defined by pRVAD (HR 1.41, p=0.02), any RVAD (HR 1.28, p=0.03), or 14d (HR 1.24, p=0.04). Conclusions The incidence of RVF differs considerably depending on the definition. Among echo and hemodynamic data, LMS seems most useful to predict RVF in LVAD candidates for most definitions, with progressive cutoffs for more severe RVF. Outcome LMS (%) TAPSE (cm) RV FAC (%) RAP (mm Hg) CI (L/min/m2) pRVAD -7.3±4.1 vs -12.9±4.3 p=0.005 1.5±0.4 vs 1.7±0.4 p=0.4 19±8 vs 24±10 p=0.3 20±12 vs 12±5 p=0.049 1.6±0.5 vs 2.1±0.6 p=0.03 Any RVAD -8.9±4.4 vs -13.4±4.3 p=0.006 1.5±0.5 vs 1.7±0.4 p=0.3 21±8 vs 24±11 p=0.5 18±9 vs 12±5 p=0.03 1.8±0.4 vs 2.2±0.6 p=0.04 14d -10.0±4.8 vs -13.4±4.2 p=0.02 1.5±0.4 vs 1.7±0.4 p=0.1 23±9 vs 23±11 p=0.9 16±9 vs 11±5 p=0.06 1.8±0.5 vs 2.2±0.6 p=0.03 7d -11.0±1.8 vs -14.8±5.4 p=0.14 1.6±0.1 vs 1.8±0.1 p=0.3 22±9 vs 25±12 p=0.6 14±8 vs 11±5 p=0.3 2.0±0.6 vs 2.1±0.4 p=0.7 (+) Outcome vs (-) Outcome