Defining Predictors of RVAD Weaning Post-LVAD Implantation in a Multi-Institutional Retrospective Cohort

Purpose Deciding which patients can be weaned from an RVAD placed after LVAD implantation could help triage advanced heart failure device strategy. Therefore, we set out to determine the predictors of RVAD weaning. Methods A multi-center retrospective study of 826 consecutive patients who received either a HeartMate2 or HVAD between 1/2007 an 12/2007. We identified 91 patients who received an early RVAD insertion during their LVAD implant index admission. A Cox Proportional hazard model was constructed to identify predictors of RVAD weaning. Results There were 91 patients (11%) who required an RVAD after LVAD with 48 (52%) weaned from RVAD support and 43 (48%) not weaned. Patients who were weaned had a lower prevalence of hypertension, higher pre-operative ACE-inhibitor use, and lower pre-operative transpulmonary gradient (TPG), RA-PCWP ratio, and RVSWI; they were also more likely to receive concomitant RVAD at time of LVAD insertion and ultimately transplanted more often (Table 1). Our multivariable model for predictors of RVAD weaning included pre-operative creatinine (HR 0.32 [0.15, 0.70], p = 0.004), PVR (HR 1.23 [1.02, 1.48], p = 0.03), non-white ethnicity (HR 2.56 [1.19, 5.54], p = 0.02), COPD (HR 4.63 [1.32, 16.25], p = 0.02), and concomitant RVAD insertion (HR 2.87 [1.00, 8.21], p-value = 0.05) with a model c-statistic of 0.84. The survival curve demonstrates that patients who are weaned from RVAD have significantly improved 1-year mortality (Figure 1, p Conclusion In a multi-institutional continuous-flow LVAD cohort, we show novel predictors of RVAD weaning that can help triage mechanical support strategy.