Abstract 11868: A Reduced Anticoagulation Strategy Among Patients Supported With Ventricular Assist Device is Not Associated With Increased Mortality

Abstract

Introduction: Patients with left ventricular assist device (VAD) are typically treated with aspirin and warfarin. However, some patients are unable to tolerate full anticoagulation (AC). There is limited information available regarding intermediate-term outcomes in patients on reduced or no anticoagulation. Methods: Patients who underwent VAD implantation at California Pacific Medical Center from January 2015 to January 2020 were studied retrospectively. Patients received aspirin 81 mg daily and warfarin with a target INR of 2-3, and anticoagulation was gradually de-intensified or withdrawn in response to recurrent mucosal bleeding. Pre-implantation characteristics, thrombolytic events (hemolysis, pump thrombosis requiring exchange, TIA, or ischemic CVA), and survival post-implant were compared among patients treated with full AC vs. reduced or no AC. We performed nonparametric survival analysis to obtain Kaplan-Meier survival estimates as of 9/15/2020. Cox proportional hazards regression was used to adjust for confounders measured at baseline. Results: 91 patients underwent VAD implant (80 HeartMate 3 and 11 HeartMate II). The median age was 65, and the majority of patients (79%) were treated as destination therapy. Reduced AC was used in 43 patients (47%), including no anticoagulation (12 patients), aspirin monotherapy (2 patients), warfarin monotherapy (16), warfarin monotherapy with a reduced INR target (10), and aspirin plus warfarin with a reduced INR target (3). During a median follow-up of 22 months, 15 thrombotic events were observed (12 in the reduced AC group and 3 in the full AC group, P=0.19). Two patients underwent pump exchange. 67% of thrombotic events and both pump exchanges occurred in HMII recipients. Survival was similar in both groups (Figure). Conclusions: The use of a reduced or no anticoagulation strategy after VAD implant is not uncommon and is associated with acceptable outcomes, particularly among HeartMate 3 recipients.