Patients’ characteristics and mortality in urgent/emergent/salvage transcatheter aortic valve replacement: insight from the OCEAN-TAVI registry
Yusuke Enta, Masaki Miyasaka, Masataka Taguri, Norio Tada, Masaki Hata, Yusuke Watanabe, Toru Naganuma, Masahiro Yamawaki, Futoshi Yamanaka, Shinichi Shirai, Hiroshi Ueno, Kazuki Mizutani, Minoru Tabata, Kensuke Takagi, Masanori Yamamoto, Kentaro Hayashida
Abstract
OBJECTIVES: Patients' backgrounds and clinical outcomes in urgent/emergent/salvage transcatheter aortic valve replacement (Em-TAVR) remain unclear. We investigated patient characteristics and the mortality in Em-TAVR and the predictors for the need for Em-TAVR. METHODS: We consecutively enrolled 1613 patients undergoing TAVR for severe aortic stenosis between October 2013 and July 2016 from the Optimised transCathEter vAlvular interventioN (OCEAN)-transcatheter aortic valve implantation (TAVI) registry. The urgency was based on the European System for Cardiac Operative Risk Evaluation II. Urgent, emergent or salvage were included with the Em-TAVR group and elective with the El-TAVR group. RESULTS: Em-TAVR was observed in 87 (5.4%) patients. A higher Clinical Frailty Scale (CFS), peripheral artery disease (PAD), hypoalbuminaemia, reduced left ventricular ejection fraction (LVEF) and preoperative at least moderate mitral regurgitation (MR) predicted the need for the Em-TAVR by the multivariate logistic regression analysis. The Em-TAVR group had the higher Society of Thoracic Surgeons Score (13.7 (IQR 8.2-21.0) vs 6.5 (IQR 4.6-9.2); p<0.001) and higher 30-day mortality (9.2% vs 1.3%; p<0.001) than the El-TAVR group. Accordingly, Kaplan-Meier analysis showed that the cumulative mortality was higher in the Em-TAVR group than that in the El-TAVR group (log-rank; p<0.001). However, Em-TAVR did not predict mortality in the multivariate Cox regression analysis. CONCLUSIONS: Em-TAVR was performed in 5.4% of patients. Higher CFS, PAD, hypoalbuminaemia, reduced LVEF and preprocedural MR predicted the need for Em-TAVR. Em-TAVR was not a predictor for mortality in the multivariate analysis, suggesting that it is a reasonable treatment option.