Mitral valve repair and bioprosthetic replacement without postoperative anticoagulation does not increase the risk of stroke or mortality

Abstract

Objectives: The study aimed to determine if mitral valve repair (MVRR) or bioprosthetic mitral valve replacement (BMVR) without postoperative anticoagulation is associated with a similar risk of thromboembolism and death as anticoagulation. Methods: We retrospectively reviewed our 2004-09 experience in 249 MVRR and bioprosthetic replacement patients (53% female; 63 year mean age). Concurrent procedures principally included antiarrhythmic surgery, aortic valve replacement, tricuspid valve repair and coronary bypass grafting. Warfarin therapy was instituted at the discretion of the surgeon. Thirty-day, a period known to have the highest risk of valve-related thromboembolism, outcomes were compared relying on the incidence of stroke and death as surrogates of thromboembolic complications. Intermediate-term survival was compared between the groups using Cox proportional hazard models. The mean follow-up was 2.9 years. Given the non-randomized warfarin use, a propensity score using patient comorbidities and concurrent procedures was created and added to the Cox models. Results: One hundred and ninety-two (77%) patients were discharged on warfarin and 57 (23%) were discharged without warfarin. Thirty-day mortality in patients discharged from the index hospitalization was 1.2% and was similar for the two groups (P = 0.99). Four ischaemic perioperative strokes were detected; 3 in the warfarin group and 1 in the no warfarin group (P = 0.99). Overall survival was 84%, with 84% survival in the warfarin group and 86% in the no warfarin group (P = 0.79). Bleeding complications were comparable between the two groups (P = 0.72). In a multivariate analysis, warfarin was not related to mortality. Conclusions: Despite current guidelines recommending postoperative anticoagulation following MVRR or bioprosthetic replacement, the avoidance of warfarin does not increase perioperative complications and has no impact on intermediate survival. Accordingly, a prospective randomized study to adjudicate the role of extended warfarin thromboprophylaxis in mitral valve surgery is warranted. © The Author 2012. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery.