329 Patient selection and learning curve influence the results of transapical aortic valve implantation

Abstract

What influence do patient selection and learning curve have on the results of transapical aortic valve implantation (TAVI)? From September 2007 to January 2009, 33 patients treated with TAVI for severe symptomatic aortic stenosis were included in a prospective registry. According to their risk profile, 2 patient groups were distinguished. In Group 1 (n=25) the high surgical risk was related to technical difficulties for conventional aortic valve replacement (porcelain aorta, redo surgery with patent coronary artery bypass grafts, hi-dose mediastinal radiotherapy). In Group 2 (n=8) the high risk was due to severe comorbidities. TAVI was successfully performed in all patients without per-procedural death or stroke. “Valve in valve” reimplantation was necessary in 2 patients due to persistent leaks. Mean follow-up was 10.8 months (1-17). Overall mortality was 16% in Group 1and 37.5% in Group 2 (p=0.03). Predictive factors of hospital mortality were: NYHA class (p=0.03), LV dysfunction (p=0.009) and pulmonary hypertension (p<0.001). No reintervention, prosthetic valve deterioration or haemolysis was recorded. In-hospital mortality decreased from 15.2% in the first 17 patients to 6.1% in the last 16 patients implanted. TAVI gives superior results in patients with technical difficulties for conventional surgery than in patients with severe comorbidities. These results could be improved by accumulating experience and better patient selection.