Time out of Therapeutic Range and Relationship with Thrombotic and Bleeding Outcomes amongst Patients with HeartMate 3 Continuous Flow Left Ventricular Assist Device.

We evaluated outcomes in patients with third-generation Heartmate (HM3) left ventricular assist devices (LVAD), stratified by international normalized ratio (INR) values. The objective of our study was to assess thrombotic and hemorrhagic complications with varying anticoagulation intensity. We retrospectively reviewed patients receiving a HM3 LVAD between 2015 and 2020 at a major heart failure center. INR data available on each patient post-implantation was extracted. INR values were categorized by subtherapeutic (INR <2.0), therapeutic (2.0-3.0), or supratherapeutic (>3.0). A linear interpolation was performed between points to establish percent time within each anticoagulation intensity. Thrombotic (ischemic stroke, suspected pump thrombosis) and hemorrhagic (gastrointestinal (GI) bleeding, hemorrhagic stroke) outcomes of interest were compared with these groups. A total of 232 patients received a HM3-CF LVAD. 204 (87.9%) patients were discharged and 184 (79.3%) were alive at last known follow-up to date. A total of 24 thrombotic events occurred: 2 pump thrombosis events and 22 ischemic strokes. A total of 70 hemorrhagic end points was recorded: 63 GI Bleeding events from 43 patients and 7 hemorrhagic strokes (4 intraparenchymal hemorrhage, 2 subarachnoid hemorrhage, 1 subdural hematoma). The percentage of time spent in the subtherapeutic anticoagulation range was not significantly different between patients with and without ischemic strokes. However, patients with hemorrhagic stroke spent a larger portion in the supratherapeutic range. A binary logistic regression was conducted revealing an INR 1.8 to 2.2 had no significant effect on the odds of observing a ischemic stroke or GI bleeding events however a model evaluating therapeutic INR (2-3) with ischemic stroke and GI bleeding were both found to be significant. The ideal anticoagulation intensity for preventing thrombotic and hemorrhagic strokes in HM3 LVAD has not yet been elucidated. Our results suggest that patients with greater time spent in subtherapeutic ranges may experience less adverse thrombotic or hemorrhagic events following implantation. Additional studies are warranted to more clearly understand and define the ideal anticoagulation intensity in this cohort. [ABSTRACT FROM AUTHOR]