Is EuroSCORE II better than EuroSCORE in predicting mortality after transcatheter aortic valve implantation?

Objectives The purpose of this study was to examine the performance of the European system for cardiac operative risk evaluation II (ESII) in patients undergoing transcatheter aortic valve implantation (TAVI). Background The ESII was developed recently to improve the predictive value of the original logistic EuroSCORE (LES). Methods Between October 2006 and November 2011, 453 consecutive patients with severe aortic stenosis undergoing TAVI with either the Edwards valve or the Corevalve were included in the current analysis. The performance of the ESII, LES, and society of thoracic surgeons predicted risk of mortality score (STS) was evaluated. Results Mean age was 83.1 ± 6.4 years. The Edwards valve was used in 382 patients (84.3%) of the cohort, transfemoral approach (TF) in 55.0%, transapical approach (TA) in 25.2%, transaortic approach (TAo) 17.8%, transsubclabian approach (SC) 2.0%. The observed 30-day mortality was 12.6% (11.2, 18.4, 7.4, and 22.2% for TF, TA, TAo, SC, respectively). The mean LES, STS, and ESII were 22.4 ± 12.1, 8.1 ± 6.0, and 8.1 ± 5.2, respectively. The Hosmer-Lemeshow test showed ESII was inadequately calibrated for 30-day mortality compared with other risk scores (ESII P = 0.09, LES P = 0.84, STS P = 0.34). By using the area under the receiver operating characteristic curve (AUC), ESII better predicted 30-day mortality, albeit poorly, compared to LES and STS (AUC = 0.68, 0.65, and 0.60, respectively). In the TF cohort, ESII was better in predicting 30-day mortality compared to LES and STS (AUC = 0.74, 0.61, 0.60, respectively). Conclusions Although the ESII demonstrated better predictive performance especially in the TF cohort, ESII is still inadequate in predicting mortality after TAVI as are LES and STS.