Reverse left ventricular remodelling after aortic valve replacement for severe aortic insufficiency.

Abstract

The goal of this study was to investigate the long-term outcome of aortic valve replacement (AVR) for severe aortic insufficiency with a focus on pre- and postoperative left ventricular (LV) function to explore predictive factors that influence the recovery of LV function and clinical outcome. A total of 478 patients who underwent AVR for pure severe aortic insufficiency were grouped according to the preoperative echocardiographical LV ejection fraction (EF): low (LO) EF <35% (n = 43), moderate EF 35-50% (n = 150) or normal EF >50% (n = 285). Actuarial survival at 10 years post-AVR was 64% with a LO EF, 92% with a moderate EF and 93% with a normal EF (P = 0.016), whereas 10-year rates of freedom from major adverse cerebral and cardiovascular events were 47%, 79% and 84%, respectively (P < 0.0001). Echocardiography at 1 year post-AVR demonstrated that EF substantially improved in all groups. We noted a significant difference in survival (P = 0.0086) and in freedom from major adverse cerebral and cardiovascular events (P = 0.024) between patients with an EF ≥35% and those with an EF <35% in the LO EF group. The multivariable logistic regression model showed that predictive factors for lack of improvement in EF 1 year post-AVR in the LO EF group included plasma brain natriuretic peptide >365 pg/mL (P = 0.0022) and echocardiographic LV mass index) >193 g/m2 (P = 0.0018). Long-term outcome post-AVR for severe aortic insufficiency was largely influenced by preoperative LV function. Predictive factors of failure to recover ventricular function post-AVR included EF <25%, pre-brain natriuretic peptide >365 pg/mL or LV mass index >193 g/m2.