Conventional aortic valve replacement in transcatheter aortic valve implantation candidates: a 5-year experience

Background Patient selection for transcatheter aortic valve implantation (TAVI) remains highly controversial. Some screened patients subsequently undergo conventional aortic valve replacement (AVR) because they are unsuitable TAVI candidates. This study examined the indications and outcomes for these patients, thereby determining the efficacy of the screening process. Methods Between January 2006 and December 2010, 79 consecutive patients (49% men), aged older than 75 years with high surgical risk, were screened for TAVI, but subsequently underwent conventional AVR through a partial or complete sternotomy. The indications, demographics, and outcomes of this cohort were studied. Results Mean age was 80.4 ± 3.6 years. Mean left ventricular ejection fraction was 0.55 ± 0.16, and the mean logistic European System for Cardiac Operative Risk Evaluation was 13% ± 7%. Of the 79 patients, 6 (7.6%) had prior cardiac surgical procedures. Indications for TAVI denial after patient evaluations were a large annulus in 31 (39%), acceptable risk profile for AVR in 24 (30%), need for urgent operation in 11 (14%), and concomitant cardiovascular pathology in 5 (6%). Mean cross-clamp time was 55 ± 14 minutes, and cardiopulmonary bypass time was 81 ± 21 minutes. Concomitant procedures included a Maze in 12 patients (15%). Postoperative morbidity included permanent stroke in 2 (2.5%), respiratory failure in 9 (11%), and pacemaker implantation in 2 (2.5%). Hospital mortality was 1.3% (1 of 79). Cumulative survival at 6, 12, and 36 months was 88.5%, 87.1% and 72.7%, respectively. Conclusions Our existing patient evaluation process accurately defines an acceptable risk cohort for conventional AVR. The late mortality rate reflects the advanced age and comorbidities of this cohort. The data suggest that overzealous widening of TAVI inclusion criteria may be inappropriate.