234Treatment of degenerated aortic bioprostheses: a comparison between conventional reoperation and valve-in-valve transfemoral transcatheter aortic valve replacement

Abstract

Introduction: Bioprosthetic valves are currently preferred over mechanical ones, but the major drawback is their deterioration and failure over time, requiring redo surgery. Conventional surgical replacement of these valves (re‐SAVR) has been standard of care; however, valve‐in‐valve deployment of transcatheter aortic valve prostheses (VinV) has been used increasingly in the last years. Data comparing both methods are scarce. We sought to evaluate 30‐day‐and 1‐year‐mortality and the incidence of adverse outcomes in patients receiving either VinV or re‐SAVR. Methods: All patients (pts) receiving either VinV or re‐SAVR for a degenerated bioprosthesis in aortic position (excluding pts with active endocarditis) between 01/2006 and 05/2016 were included in the analysis. We defined all‐cause‐30‐day and 1‐year mortality as the primary end points. Continuous variables are displayed median (IQR). All end point definitions were according to the Valve Academic Research Consortium‐2 definitions. Results: Overall, 229 pts were treated for aortic valve prosthesis failure, with 121 receiving VinV and 108 receiving re‐SAVR. Pts receiving VinV had a higher STSPROM (7.1% (4.2; 9.9) vs. 2.0% (1.5; 3.3), p<0.01), were older (78 years (73; 82) vs 60 years (47; 71), p<0.01) and had lower baseline GFR (62.2 ml/min/1.73m2 (46.9; 73.6) vs. 77.9 ml/min/1.73m2 (62.2; 89.9), p<0.05). Baseline ejection fraction did not differ significantly between groups (58% (46.0; 65.0) vs. 60% (50.3; 65.0), p=0.308). Duration of postoperative stay was shorter for pts receiving VinV (10 days (8; 13) vs. 12 days (9; 15), p<0.01). Compared to VinV, re‐SAVR had a higher rate of life threatening bleeding events (5.5% vs 13.9%, p<0.05) and a significantly higher incidence of renal failure (any renal failure: 13.2% vs 32.4%, p<0.01; stage 3 renal failure: 4.1% vs 12.0%, p<0.05). There were no significant differences regarding periprocedural myocardial infarction (4.2% vs 1.9%, p=0.45), stroke (any stroke: 7.5% vs 7.5%, p=1.0) or need for permanent pacemaker implantation (17.4% vs 13.9%, p=0.586). 30‐day mortality (4.1% (VinV) vs. 4.6% (re‐SAVR), p=0.882) and 1‐year mortality (7.4% (VinV) vs. 12.0% (re‐SAVR), p=0.182) did not differ significantly between groups. Conclusion: Despite a higher risk profile in VinV, 30‐day and 1‐year mortality rates were not different compared to re‐SAVR which might be explained by a higher complication rate in re‐SAVR. VinV seems to be a safe and feasible therapeutic option for patients with degenerated aortic bioprosthesis. A randomized controlled clinical trial is necessary to elucidate the impact of VinV vs. re‐SAVR on functional outcome and mortality.