Early antithrombotic therapy for aortic valve bioprostheses: is there an indication for routine use?

BACKGROUND

Consensus reports over the past 10 years from the United States, Europe, United Kingdom, and Canada have not provided consistent guidelines for antithrombotic therapy of aortic valve bioprostheses for the three-month period after surgery. This study was conducted to determine if antithrombotic therapy was protective against TE with aortic bioprostheses 30 days or less after aortic valve replacement (AVR).

METHODS

From 1994 to 2000, 1,372 patients implanted with three currently marketed aortic bioprostheses, Medtronic Mosaic (Medtronic, Inc, Minneapolis, MN) (415 patients), Carpentier-Edwards SAV (462), and Carpentier-Edwards PERIMOUNT (495) (Edwards Lifesciences, Irvine, CA), with a mean age of greater than 70 years were evaluated. Patient populations were comparable, inclusive of concomitant coronary artery bypass grafting (CABG) for the overall populations and for patients greater than 70 years.

RESULTS

There were 37 thromboembolic (TE) events: major TE, 14; reversible ischemic neurologic deficit (RIND), 12; and minor TE, 11. There were 4 TE deaths. Multivariate (stepwise logistic regression) analysis revealed no predictive risk factors for overall TE. For the combination of major TE plus RIND there were two predictive risk factors with analysis of 12 risk variables: preoperative cerebrovascular accident (odds ratio [OR] 4.45, 95% confidence interval [CI] 1.17 to 16.87, p = 0.028); and concomitant CABG (OR 3.19, 95% CI 1.16 to 8.76, p = 0.025). Neither anticoagulant nor antiplatelet therapies gave significant protection.

CONCLUSIONS

There does not appear to be an indication for routine antithrombotic management. The study supports the potential use of antithrombotic therapy for comorbidities of preoperative cerebrovascular accident and concomitant CABG but not atrial fibrillation, left ventricular dysfunction, or elderly age greater than 70 years. Vascular burden and advanced age are likely contributing factors to these independent predictors. There may still be a need for, or at least consideration of, a randomized trial for AVR with bioprostheses.