Is anti-platelet therapy needed in continuous flow left ventricular assist device patients? A single-centre experience

Objectives: We report our 5-year experience of continuous flow left ventricular assist device (LVAD) implantation without the use of anti-platelet therapy.

Methods: Between February 2006 and September 2011, 27 patients (26 men; 1 woman) were implanted with a continuous flow LVAD (HeartMate II, Thoratec Corporation, Pleasanton, CA, USA). The mean age was 55.7 ± 9.9 years. The mean duration of support was 479 ± 436 (1-1555) days with 35.4 patient-years on support. Twenty-one patients were implanted as a bridge to transplantation and 6 for destination therapy. The anticoagulation regimen was fluindione for all patients, with aspirin for only 4 patients. At the beginning of our experience, aspirin was administered to 4 patients for 6, 15, 60 and 460 days. Due to gastrointestinal (GI) bleeding and epistaxis, aspirin was discontinued, and since August 2006, no patients have received anti-platelet therapy.

Results: At 3 years, the survival rate during support was 76%. The most common postoperative adverse event was GI bleeding (19%) and epistaxis (30%) (median time: 26 days) for patients receiving fluindione and aspirin. The mean International Normalized Ratio (INR) was 2.58 ± 0.74 during support. Fifteen patients have been tested for acquired Von Willebrand disease. A diminished ratio of collagen-binding capacity and ristocetin cofactor activity to Von Willebrand factor antigen was observed in 7 patients. In the postoperative period, 2 patients presented with ischaemic stroke at 1 and 8 months. One of these 2 patients had a previous history of carotid stenosis with ischaemic stroke. There were no patients with haemorrhagic stroke, transient ischaemic attack or pump thrombosis. The event rate of stroke (ischaemic and haemorrhagic) per patient-year was 0.059 among the patients without aspirin with fluindione regimen only.

Conclusions: A fluindione regimen without aspirin in long-duration LVAD support appears to not increase thromboembolic events and could lead to a diminished risk of haemorrhagic stroke.