Heart transplantation outcomes in patients with continuous-flow left ventricular assist devices

Abstract

Introduction Durable continuous-flow left ventricular assist devices (CF-LVAD) are used increasingly as both destination and bridge to heat transplant (HTx), in end stage heart failure. Data on long-term survival post-HTx in patients supported with CF-LVAD compared to those undergoing de novo heart transplantation are inconsistent. Some studies show no difference and others indicate increased long-term mortality in patients on mechanical circulatory support (MCS) pre-transplant.1,2,3 Methods We retrospectively reviewed all 153 patients who had OHT in the last 5 years in a tertiary cardiac hospital. Patients were divided into two groups (1) those without mechanical circulatory support pre-transplant (n = 90), (2) and those who were bridged with implantable long-term CF-LVAD (n = 34), all other patients were excluded. Pre-transplant and peri-operative factors, as well as early and late outcomes were then compared between groups. Results Baseline pre-transplant factors were the same between groups, though patients bridged with MCS were more likely to have had a previous stroke (8.9% vs 29.4%, p = 0.008). The time spent on cardiopulmonary bypass was longer (168 vs 202 minutes, p = 0.004) in MCS group. Intraoperative haemoglobin was significantly lower in the MCS group (85g/l vs 79g/l, p = 0.01), and intraoperative lactate was significantly higher (7.9 vs 10.6, p = 0.002). The initial post-operative LVEF was lower (p = 0.004), and there was a trend towards developing severe PGD (18.9 % vs 32.3%, p = 0.054) in the MCS group. Early outcomes were similar with an equal need for renal replacement therapy (p = 0.19) and similar length of stay in ITU (p = 0.17). Despite similar 30-day mortality (p = 0.213), 60-day (p = 0.058) and 1-year mortality (Non-MCS 11.1% versus MCS 32.4%, p = 0.008)(Log-rank (Mantel-Cox) p = 0.11) were significantly higher in the MCS group. Kaplan Mayer curves show significant difference in estimated survival between the groups (Fig 1). Discussion Our results in agreement with present literature show comparable 30-d mortality in both subpopulations, however, indicate increased risk of death in 1-year follow-up.