Prediction of 30-day mortality after transcatheter aortic valve implantation: a comparison of logistic EuroSCORE, STS score, and EuroSCORE II.

Background and aim of the study: The logistic EuroSCORE and STS score have been used for the selection of suitable TAVI patients, but their predictive ability is unsatisfactory. The study aim was to evaluate the performance of the EuroSCORE II in predicting 30-day mortality after TAVI in comparison to the logistic EuroSCORE and STS scoring systems. Methods: Between January 2008 and April 2013, a total of 123 consecutive patients underwent TAVI (transapical, n = 85; transfemoral, n = 38) at the authors' institution. Calibration and discriminatory ability was evaluated for three risk scores models (logistic EuroSCORE, STS score, and EuroSCORE II), and compared for the prediction of 30-day mortality using the Hosmer-Lemeshow test for goodness-of-fit and receiver operating characteristics curve analysis. Results: The overall 30-day mortality was 4.1% (5/123). Predicted mortality was 25.0 +/- 15.7% by logistic EuroSCORE, 7.3 +/- 6.9% by STS score, and 7.8 +/- 8.7% by EuroSCORE II. The observed/expected mortality ratio was 0.16 for logistic EuroSCORE, 0.56 for STS score, and 0.52 for EuroSCORE II. The area under the curve was 0.69 (95% CI 0.54-0.84) for the logistic EuroSCORE, 0.60 (95% CI 0.38-0.82) for the STS score, and 0.66 (95% CI 0.46-0.86) for the EuroSCORE II. Conclusion: In the present study, the EuroSCORE II was found to predict 30-day mortality more accurately for the TAVI cohort than did the more established logistic EuroSCORE, and also to compare (at present) on a par with the STS score. However, there were no differences in discriminatory power between the models. It is believed that, in the absence of a more TAVI-oriented risk stratification system, the EuroSCORE II may be a valuable adjunct in the clinical setting.