Background: Device indications have changed for placement of continuous-flow left ventricular assist devices (CF-LVADs). We performed a multicenter analysis evaluating temporal variations in outcomes after CF-LVAD implantation.
Methods & results: We retrospectively defined 3 time intervals to reflect changes in CF-LVAD technology (period 1, 2004-2009; period 2, 2010-2012; and period 3, 2012-2014). A total of 1,064 patients (Heartmate II [HMII] = 835; Heartware [HVAD] = 229) underwent CF-LVAD implantation from May 2004 to October 2014. Device utilization was different between periods: period 1: HMII = 134 (100%); period 2: HMII = 480 (88%) and HW = 63 (12%); and period 3: HMII = 221 (57%) and HW = 166 (43%); P < .001. Despite few baseline group differences, adjusted survivals were similar among the time periods (P = .96). Adjusted multivariable analysis revealed age (per 10-year increase) and Interagency Registry for Mechanically Assisted Circulatory Support category (1 vs all others) as the only independent predictors of mortality: P < .001 and P = .008, respectively. Furthermore, it also showed the later periods to be at an increased risk of adverse events: 1) pump thrombosis (periods 2 and 3); and 2) gastrointestinal bleeding (period 3).
Conclusions: Despite significant differences in device types, indications, and patient characteristics, post-implantation survivals were similar across time intervals. The most recent cohort seems to be at an increased risk of gastrointestinal bleeding and pump thrombosis.
Keywords: Ventricular assist devices; continuous-flow; outcomes; temporal analysis.
Published by Elsevier Inc.