Objective: Thromboembolic events can occur in up to 20% of patients with a left ventricular assist device. The aggressive use of anticoagulation with newer continuous-flow devices has potentially increased the risk of postoperative bleeding. The predecessor of the HeartMate II left ventricular assist device, the HeartMate XVE (Thoratec Corp, Pleasanton, Calif), was associated with an extremely low thromboembolic risk, even without anticoagulation, because of its unique textured surfaces. Even though several areas of the HeartMate II are textured, a protocol was adopted for this new axial flow pump requiring long-term anticoagulation with warfarin. In our study, we investigated whether the HeartMate II left ventricular assist device is associated with a similarly low thromboembolic risk as the HeartMate XVE. Methods: At our institution, 45 patients (mean age, 57.24 ± 14.2 years) underwent implantation of the HeartMate II; 30 underwent bridge-to-transplantation therapy, 7 underwent destination therapy, and 8 underwent left ventricular assist device exchange for a failed XVE left ventricular assist device. Total duration of HeartMate II support was 352.13 patient-months (mean duration, 7.2 ± 5.2 months). All 45 patients were treated postoperatively with warfarin and aspirin. We recorded use of these 2 medications and monthly international normalized ratios. Prospectively, we also monitored patients for any clinical thromboembolic events and for pump thrombus. Results: Of our 45 study patients, 41 had a mean international normalized ratio of less than 2.0; of those 41 patients, 21 had a mean international normalized ratio of less than 1.6. Because of recurrent gastrointestinal bleeding episodes, 7 patients discontinued warfarin for a total duration of 39.1 patient-months. During the entire period of HeartMate II support, we noted 1 thromboembolic event. In addition, another patient had a suspected left ventricular assist device pump thrombus that resolved with a high-intensity heparin anticoagulation protocol (international normalized ratio, 1.3). Conclusions: Our preliminary single-center analysis suggests that the HeartMate II is associated with an extremely low thromboembolic risk and with less stringent requirements for anticoagulation. Selected patients at high risk for bleeding can be safely followed with either no or extremely low anticoagulation requirements for prolonged periods.
Bibliographical noteFunding Information:
This study was part of the multicenter study supported by Thoratec Corp, Pleasanton, Calif.