ABSTRACT
Thromboembolic complications after cardiac valve replacement are due to a complex interplay between patients' characteristics, device features and anticoagulation intensity. Subtle design and material differences in available prostheses may thrombosis. We conducted a post-hoc sub-analysis of the LOWERING-IT database to test the safety and feasibility of a low-level oral anticoagulant regime in low-risk patients with aortic LivaNova prosthetic valve replacement. The study population included 148 patients randomized to a low INR target (1.5-2.5; LOW-INR group), and 144 patients to the standard INR (2.0-3.0; CONVENTIONAL-INR group). The non-inferiority of thromboembolic events between LOW-INR and CONVENTIONAL-INR groups was tested. Cumulative follow-up reached 1,545 patient/years. The mean INR was 1.91 ± 0.23 in the LOW-INR group, and 2.59 ± 0.26 in the CONVENTIONAL-INR group (P < 0.001). There were 3 thromboembolic events, all in the CONVENTIONAL-INR group. Comparison of thromboembolic events was not significant. The 1-sided 97.5% exact CI for the difference in primary event proportion was 0.54%, satisfying criteria non-inferiority. Bleeding events were significantly different: 6.61 per 1,000 patient-year in LOW-INR group vs 18.65 per 1,000 patient-year in CONVENTIONAL-INR group (p < 0.045, RR 0.37). In conclusions these data suggest that low-dose anticoagulation is safe in selected patients after aortic LivaNova Bicarbon prosthesis implantation.
