Abstract 4366910: Multicenter Model to Predict Right Heart Failure After HeartMate 3 Left Ventricular Assist Device Implantation
Abstract
Background: Right heart failure (RHF) is a serious complication of durable HeartMate 3 (HM3) left ventricular assist device (LVAD) therapy. We aimed to evaluate the incidence, predictors, and prognostic significance of RHF within 3 months following HM3 LVAD implantation among patients in a contemporary multicenter registry, utilizing preoperative imaging, hemodynamics, and medication variables in our analysis. Methods: Patients underwent HM3 LVAD implantation at 6 centers between 2015-2022. A multivariable logistic regression model was developed using least absolute shrinkage and selection operator (LASSO) for the variable selection process. Mechanical Circulatory Support-Academic Research Consortium definitions of early and late RHF were used. Results: Among 596 patients (median age 59 years, 25% female, 48% INTERMACS profiles 1 or 2), 195 (33%) developed RHF within 3 months of LVAD implantation. In multivariable analysis, the following preoperative variables were independently associated with an increased risk of RHF: severe right ventricular systolic dysfunction on echocardiogram (OR 1.78, p=0.036), VA ECMO (OR 3.45, p=0.002), and dobutamine use (OR 1.66, p=0.029). Additionally, standardized lower left ventricular end-diastolic diameter (OR 0.76 per SD increase, p=0.005) and higher mean pulmonary artery pressure (OR 1.23 per SD increase, p=0.032) were significantly associated with increased risk of RHF. Survival, censored for heart transplant and pump exchange, was significantly worse for patients experiencing RHF within 3 months compared to patients who did not develop this complication (75% vs. 90% at 1-year and 63% vs 71% at 4-years, p=0.003; Figure). Conclusions: RHF is common within the first 3 months of HM3 LVAD implantation and is associated with significantly greater mortality during long-term follow-up. We identified preoperative imaging, hemodynamics, and medication variables that can be used during evaluation and selection of candidates for LVAD therapy.
Article Details
Authors (9)
Manyun Zhao
Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania, United States
Michael Genuardi
University of Pennsylvania, Philadelphia, Pennsylvania, United States
Noah Moss
Mount Sinai Medical Center, New York, New York, United States
Arman Kilic
MUSC, Charleston, South Carolina, United States
Omar Saeed
Samit Shah
Northwell, New York, New York, United States
Danyaal Moin
New York University, New York, New York, United States
Jinbo Chen
Himabindu Vidula
University of Pennsylvania, Philadelphia, Pennsylvania, United States