In-hospital mortality in propensity-score matched low-risk patients undergoing routine isolated surgical or transfemoral transcatheter aortic valve replacement in 2014 in Germany.

Abstract

Recent randomized trials have documented the superiority of TAVR-particularly via transfemoral access-over SAVR in patients with severe aortic stenosis considered to have a high or intermediate operative risk of death. We sought to assess in-hospital outcomes of patients with severe aortic stenosis and a low risk of operative mortality undergoing routine surgical aortic valve replacement (SAVR) or transcatheter aortic valve replacement (TAVR). We performed a propensity-score matched comparison of all patients undergoing first-time treatment by SAVR or transfemoral TAVR (TF-TAVR) in 2014 in Germany who had a logistic EuroSCORE (logES) ≤ 10%, considered to reflect low surgical risk. The primary endpoint of our analysis was in-hospital mortality. Of 7624 SAVR and 9969 TF-TAVR procedures, 6844 (89.8%) and 2751 patients (27.6%), respectively, were considered low risk with a logES between 1.505 and 10.0%. Matching yielded 805 TF-TAVR/SAVR patient pairs with identical propensity scores and no difference in pertinent baseline characteristics, except for the logES, which was significantly higher in TF-TAVR patients (6.8 ± 1.7 vs. 4.2 ± 1.3% in SAVR patients, P < 0.001). Observed in-hospital mortalities were 1.7% (95% confidence interval, 1.1-3.0%) in SAVR and 2.0% (1.3-3.3%) in TF-TAVR patients (P = 0.85). Our finding of no difference in in-hospital mortality in propensity-score matched low-surgical-risk patients treated by SAVR or TF-TAVR in a routine clinical setting indicates that TF-TAVR can be offered safely to individual patients, despite their operative risk being low. This finding needs to be confirmed in a randomized trial.