Impact of Atrial Fibrillation, Diabetes Mellitus, and Obesity on Outcomes With Aspirin Avoidance and Hemocompatibility With a Left Ventricular Assist Device: An Analysis From the ARIES-HM3 Trial

Background The ARIES-HM3 trial demonstrated the safety and effectiveness of aspirin elimination from the antithrombotic regimen after HeartMate 3 (HM3) left ventricular assist device (LVAD) implantation. We explored the interaction of atrial fibrillation, diabetes mellitus, and obesity (AF/DM/Ob) with aspirin elimination on hemocompatibility-related adverse events at 1-year postimplant. Methods This prospective, double-blind, placebo-controlled trial randomized patients with an HM3 LVAD implant to receive aspirin (100 mg/d) or placebo, in addition to a vitamin K antagonist. The primary endpoint was survival free of nonsurgical (>14 days postimplant) hemocompatibility-related adverse events, including stroke, pump thrombosis, bleeding, and arterial peripheral thromboembolism at 12 months. The composite endpoint and individual components were compared between the arms for those with and without AF/DM/Ob; responsiveness to aspirin was assessed by measurement of thromboxane-B₂ level suppression. Results Among the 589 patients who contributed to the primary endpoint analysis, 1 or more AF/DM/Ob comorbidities were present in 78% (461/589), distributed as AF (259/461), DM (240/461), and Ob (204/461). The presence of 1 or more AF/DM/Ob comorbidities did not influence the effect of aspirin elimination on the primary endpoint outcome (interaction P = .60); patients with all 3 comorbidities (44/589, 7.5%) receiving aspirin had a significantly greater rate of primary endpoint events (difference: 30.6%; 95% confidence interval, 2.2%−59.0%). There was no treatment heterogeneity between the subgroups. Nonsurgical bleeding events were reduced similarly in patients with or without AF/DM/Ob who received placebo versus aspirin with similar reductions in thromboxane-B₂ levels. Conclusion Among ARIES-HM3 trial patients with AF/DM/Ob, no comorbidity, alone or in combination, altered the safety or observed effect size on bleeding reduction with aspirin elimination in patients implanted with the HM3 LVAD.