Clinical valve thrombosis after transcatheter aortic valve-in-valve implantation

BACKGROUND: Limited data exist on clinical valve thrombosis after transcatheter aortic valve-in-valve (ViV) implantation. Our objective was to determine the incidence, timing, clinical characteristics, and treatment outcomes of patients diagnosed with clinical ViV thrombosis. METHODS AND RESULTS: Centers participating in the Valve-in-Valve International Data Registry were surveyed for thrombosis cases, and clinical valve thrombosis was defined based on a combination of the presence of new valve dysfunction and an imaging evidence of leaflet thrombosis. Three hundred ViV implantations were included. The surgical valve was stented in 86.3% and stentless in 13.7% of cases; and the transcatheter heart valve was self-expanding in 50%, balloon-expandable in 49%, and mechanically expanding in 1.0%. The incidence of clinical valve thrombosis was 7.6% (n=23), diagnosed at a median time of 101 days (interquartile range, 21-226) after the procedure. Fifteen patients (65%) presented with worsening symptoms and 21 (91%) with transvalvular mean gradient elevation. The mean gradient at the time of diagnosis (median 39 mmHg; interquartile range, 30-44) was significantly higher than immediately post-ViV (13 mmHg; interquartile range, 8-20.5; P<0.001) and was significantly reduced after oral anticoagulation therapy (17.5 mmHg; interquartile range, 11-20.5; P<0.001). There were no deaths or strokes related to valve thrombosis. Factors associated with valve thrombosis were oral anticoagulation (odds ratio [95% confidence limits]: 0.067 [0.008-0.543], P=0.011), surgical valve true internal diameter indexed to body surface area (0.537 [0.331-0.873], P=0.012), and Mosaic or Hancock II stented porcine bioprostheses (4.01 [1.287-12.485], P=0.017). CONCLUSIONS: Clinical valve thrombosis after transcatheter aortic ViV implantation is common, especially in patients not on oral anticoagulation. Although aortic ViV is commonly associated with elevated gradients, valve thrombosis should be ruled out if gradients increase compared with early postprocedural values. A higher incidence was observed after treatment of certain stented porcine surgical valve types, suggesting a specific adjustment of the adjunctive antithrombotic therapy in this subset of ViV patients.