Transcatheter Aortic Valve-in-Valve Implantation: What Have We Learned?

Objectives: As bioprosthetic valves are prone to structural deterioration, valve-in-valve (ViV) therapy is evolving into an alternative option to reoperation in selected high-risk patients. We present a comparative analysis of immediate and 30-day outcome of ViV TAVR using different types of THV.

Methods: 80 high-risk patients (Age 77.2 ± 7.7 years, 47.5% male, log EuroSCORE I 28.6 ± 18.2 / STS score 7.4 ± 6.3) with symptomatic deterioration of stented aortic bioprostheses undergoing ViV therapy between 2008 and 2017 were prospectively computed into a dedicated database and retrospectively analyzed.

Results: Types of THV and numbers of cases were: Medtronic CoreValve (Evolut R) (group 1) n = 56 (70.0%), Edwards Sapien (XT / S3) (group 2) n = 19 (23.8%) and SJM Portico n = 5 (4.8%). Access was transfemoral in 77.5% (n = 62), transapical in 18.8% (n = 15) and transsubclavian/-axillary in 3.8% (n = 3). Cerebral protection devices were used in 40.2% with increasing frequency in later years. All-cause and cardiovascular mortality rates at 30 days were 4.4 and 1.4% respectively. One case of myocardial infarction occurred. There were no strokes or conversions to open surgery. At discharge, mean pressure gradient by echo was 19.2 ± 7.1 mmHg and effective orifice area was 1.2 ± 0.2 cm². The mean gradient did not significantly differ between group 1 and group 2 (19.0 ± 7.3 vs. 20.2 ± 6.9 mm Hg; p = 0.552). After implantations in deteriorated bioprostheses ≤23 mm the gradient was 20.3 ± 7.1 mm Hg in group 1 and 24.5 ± 6.3 mm Hg in group 2 (p = 0.176). A slight tendency in favor of group1 prostheses was found after implantations in prostheses >23 mm (12.8 ± 4.8 mm Hg [group 1] vs. 17.8 ± 6.3 mm Hg (group 2) (p = 0.066)). Paravalvular leakage was none/trace in 81.1% and mild in 18.9%. The VARC device success criteria were met in 76.3% (81.8% group 1; 50.0% group 2; p = 0.034). Failure to reach this endpoint was mainly caused by residual mean pressure gradients >20 mm Hg. After implantations in deteriorated bioprostheses >23 mm the gradient was >20 mm Hg in 18.2%.

Conclusion: This data demonstrates an adequate safety profile of ViV therapy in high-risk patients. The theoretical advantage of supra-annular versus intra-anular THV could not be proven by the presented hemodynamic data, but was shown through the VARC device success rate. As a relatively high proportion of prostheses show gradients >20 mm Hg, the unfavorable hemodynamic results of ViV therapy should be weighed up against the risk of a re-operation.