A propensity score matched analysis of transcatheter aortic valve replacement in failed bioprosthetic valves

Abstract Introduction While recently published data indicate similar long-term results of transcatheter aortic valve replacement (TAVR) compared to surgical procedures, surgical aortic valve replacement (SAVR) remains the standard of care in low-risk patients. However, since life expectancy increases, a non-negligible number of patients experience bioprosthetic valve dysfunction with the need for reintervention, but long-term data of TAVR in failed bioprosthetic valves is scarce. Purpose The aim of the study was to analyse long-term data of patients with TAVR in failed bioprosthetic valves compared to patients with native valve TAVR. The primary endpoint was 6-year all-cause mortality and the main secondary endpoint was echocardiographic valve function. Methods Patients undergoing TAVR for severe aortic valve disease between December 2012 and December 2020 were retrospectively analysed. Prior TAVR or aortic valve homografts led to exclusion from the study. A 2:1 propensity score matching by the optimal neighbor matching algorithm was performed to overcome differences in baseline characteristics. Results Between December 2012 and December 2020, 3,423 patients received TAVR for severe aortic valve disease and met inclusion criteria for the study. Of those, 3,287 patients underwent native valve TAVR, whereas 136 patients experienced bioprosthetic valve dysfunction and were treated with TAVR-in-SAV. After 2:1 propensity score matching, baseline characteristics without any relation to prior SAVR were similar between both groups. Concerning postprocedural echocardiographic function, a less pronounced gradient reduction (57.0 vs. 71.0 %, p < 0.01, Figure 1) and a higher number of patients with elevated mean pressure gradient > 20 mmHg could be found after prior SAVR (19.4 vs. 3.0 %, p < 0.01). However, there was no difference regarding 6-year all-cause mortality between both groups (log rank test p = 0.8, Figure 2). Conclusions Besides a less pronounced gradient reduction and higher mean aortic pressure gradients after TAVR-in-SAV compared to patients with native valve TAVR, no differences in 6-year all-cause mortality could be observed in a propensity score matched analysis.Reduction of mean pressure gradient6-year all-cause mortality