Increased Right Ventricular Assist Device (RVAD) Utilization in Patients with HeartMate 3 Left Ventricular Assist Device (LVAD) in the Post-Approval Era

Abstract

MOMENTUM 3 trial excluded patients with planned biventricular support prior to enrollment and 4.2% right ventricular failure requiring RVAD in Heartmate 3 (HM 3) patients. The aim of this study was to assess the use of RVAD in the real-world post-approval era. HM3 implantation at a major academic center were categorized into two era: HM 3 implanted 1) before FDA approval (8/23/2017) [N=67] and 2) after FDA approval [N=107]. Baseline patient characteristics, RVAD utilization patterns, and LVAD complications and mortality were analyzed in the pre- vs. post-approval era. In the post-approval era, HM3 patients were younger (57.5±14.1 vs 61.7±11.2 yrs, p=0.031), more likely to have ischemic etiology of heart failure (55.6% vs 34.3%, p=0.006), to be INTERMACS Class I (21.3% vs. 1.5%, p=0.002), and to have VA-ECMO or paracorporeal VAD support (12.0% vs 0.0%, p=0.012) prior to HM3 implant. 8 (11.9%) patients out of 67 required RVAD in the pre-approval era as opposed to 47 patients out of 108 (43.5%) in the post-approval era (p<0.001). All patients in the pre-approval era received unplanned RVAD (8, 100%) compared to 16 patients (34.0%) in the post-approval era (p<0.001). Despite increased utilization of RVADs, the risk of CVVHD (6.0% vs 10.2%, p=0.413), post-LVAD length of stay (34.3±32.9 vs 41.7±40.0 days), and in-hospital mortality (4.5% vs 6.9%, p=0.741) were comparable in pre- vs. post-approval era. There was no difference in post-implant survival (89.7% vs. 84.7%, log-rank p=0.357), freedom from gastrointestinal bleeding (70.0% vs. 70.4%, log-rank p =0.831) or freedom from stroke (93.2% vs. 92.2% log-rank p = 0.547) at 2-year follow-up in the pre- vs. post- approval period. Level of sickness at the time of device implantation as well as need for RVAD utilization increased substantially in HM3 patients following the FDA approval. However, the change in practice did not adversely affect post-LVAD morbidity or mortality.