Impact of Combined “CHADS-BLED” Score to Predict Short-Term Outcomes in Transfemoral and Transapical Aortic Valve Replacement

Abstract

Background. High CHA2DS2-VASC and HAS-BLED scores are linked to increased mortality in structural and nonstructural cardiovascular interventions irrespective of the presence of atrial fibrillation (AF) or oral anticoagulation. We aimed to use the aforementioned scores to quantify the risk of 30-day mortality, major vascular and bleeding events (MVASC/BARC), and cerebrovascular insults (CVI) in patients undergoing different access routes in transcatheter aortic valve replacement (TAVR). Methods. Out of 1329 patients, 980 transfemoral (TF) TAVR (73.7%) and 349 transapical (TA) TAVR (26.3%) were included. CHA2DS2-VASC, HAS-BLED, and combined “CHADS-BLED” scores were calculated and compared to the predictive value of the established EuroSCORE and STS score. Results. In all-comers TF TAVR patients, the applied risk models showed only poor association with 30-day mortality while, in patients with concomitant AF, a strong association was observed using the combined CHADS-BLED score (c-index: 0.83; 95% CI: 0.76–0.91; $p$  < 0.0001). Concerning 30-day mortality, only the STS score for TF TAVR (c-index: 0.68; 95% CI: 0.59–0.76; $p$  = 0.001) and EuroSCORE for TA TAVR (c-index: 0.66; 95% CI: 0.56–0.76; $p$  = 0.005) could show some predictive value. High CHADS-BLED was associated with enhanced CVI (3.0% vs. 7.2%; $p={0.0039}^{\ast }$ ) and more frequent MVASC/BARC (3.2% vs. 6.3%; $p$  = 0.0362) in the all-comers TAVR cohort. All risk models failed in the prediction of CVI and MVASC/BARC for TA TAVR patients. Conclusion. The combined CHADS-BLED score was a strong predictor for 30-day mortality in TF TAVR patients with AF. A high CHADS-BLED score showed a good predictive value for major vascular and bleeding events as well as CVI in TF TAVR patients. This study is registered at clinical trials (NCT01805739).