Implementation of transcatheter aortic valve implantation (TAVI) into clinical daily routine for treatment of severe aortic stenosis in high- and intermediate risk patients has led to an increasing caseload of failing transcatheter heart valves (THV). We herein evaluate early and midterm outcomes after valve-in-valve (ViV) TAVI for failing THV and compare clinical and hemodynamic results with ViV procedures for deteriorated surgical bioprosthetic heart valves (SHV).

 

In-hospital databases were retrospectively screened for patients treated for failing biological aortic valve prosthesis (THV and SHV) by ViV TAVI at our center. Clinical and hemodynamic results, as well as follow-up characteristics were compared between patients undergoing ViV for failing THV and those with failing SHV according to VARC-2 definitions. Comparisons of VARC-2 combined endpoints were adjusted for baseline differences between the groups using logistic regression analysis.

 

Between 08/2008 and 08/2020, 200 patients were treated by ViV TAVI at our center. Of those, 188 patients (94.0%) were treated for failing SHVs and 12 patients (6.0%) underwent ViV for failing THVs. There was no significant difference regarding age between groups (SHV: 77.6±7.1 years, THV: 79.0±4.8 years; p=0.507). EuroSCORE II was significantly higher in the SHV group (10.5±6.4 vs. 6.0±4.2; p=0.017). Interval to index procedure was significantly shorter in patients with failing THV (9.4±5.0 vs. 2.7±2.5 years; p<0.001). In the SHV group, most frequent mode of deterioration was restenosis (31.9%, n=60), whereas patients of the THV group were mainly treated for regurgitation (50.0%, n=6). These were valvular and paravalvular in 50% (n=3), respectively. There was no significant difference in 30-day (SHV: 3.7% (n=7), THV: 8.3% (n=1); p=0.429) or 6-month mortality (SHV: 13.8% (n=13), THV: 28.6% (n=2); p=0.290). Furthermore, there was no significant difference in the rate of disabling stroke (SHV: 0.5% (n=1), THV: 0.0% (n=0); p=0.800) or acute kidney injury stage II or III (SHV: 2.1% (n=4), THV: 0.0% (n=0); p=0.610). Postprocedural mean transvalvular gradient was significantly higher in the SHV group (16.2±7.4 vs. 10.7±4.4; p=0.022), whereas there was no significant difference in the frequency of postprocedural moderate/severe paravalvular leakage between groups (SHV: 0.5% (n=1), THV: 0.0% (n=0); p=0.800). Adjusted regression analysis did not show significant differences between groups regarding VARC-2 combined endpoints clinical efficacy (SHV: 66.5% (n=125), THV: 83.3% (n=10); p=0.313), combined safety (SHV: 78.1% (n=146), THV: 83.3% (n=10); p=0.220), or device success (SHV: 80.9% (n=152), THV: 75.0% (n=9); p=0.892).

 

In this single-center cohort, no significant differences were found regarding mortality or stroke rates in patients undergoing ViV treatment for THV or SHV degeneration. However, hemodynamics (postprocedural transvalvular pressure gradient) were superior after THV ViV. Modes of deterioration were different according to valve type. Furthermore, time to deterioration was shorter in the THV group, substantiating the assumption of distinct failure types in THV (i.e. hypoattenuated leaflet thickening, restricted leaflet motion, paravalvular leakage) compared to SHV. In conclusion, ViV for failing THV is a safe and effective treatment with a low complication rate. These findings must be confirmed in larger patient collectives.