Abstract
Background Several patient-related characteristics have been associated with inferior outcomes following durable left ventricular assist device (LVAD) implantation in patients transitioned from venoarterial extracorporeal membrane oxygenation (VA ECMO). The impact of LVAD pump type used is less well-known. Methods We compared outcomes between patents who received axial and centrifugal flow LVADs following stabilization with VA ECMO. Results From January 2011 to December 2018, we implanted 28 LVADs in patients transitioned from VA ECMO. This included 17 axial flow devices (HeartMate II LVAD, Abbott Laboratories, Chicago, IL) and 11 centrifugal flow pumps (eight HeartWare HVADs; Medtronic, Minneapolis, MN and three HeartMate 3 LVAS pumps; Abbott Laboratories, Chicago, IL). There was no difference in hospital mortality (23.5% vs 18.2%, P = .74) or 1-year survival (P = .31) between the devices. There were no differences in adverse event rates between the two pump types, apart from a higher rate of gastrointestinal bleeding in patients who received centrifugal flow pumps (1.44 events per 100 patient-months vs 14.67 events per 100 patient-months, P = .010). Preimplantation levels of alanine aminotransferase (hazard ratio [HR], 1.001; 95% confidence interval [CI], 1.000 to 1.002; P = .004) and elevated serum creatinine level (HR, 3.480; 95% CI, 1.121-10.807; P = .031) emerged as significant predictors of decreased 1-year survival. Conclusions Preimplantation optimization of end-organ function is the single most important determinant of successful post-LVAD survival in patients transitioned from extracorporeal life support. There is no association of pump type with LVAD outcomes up to 1-year post implantation.