Abstract Background Transcatheter aortic valve implantation (TAVI) has become a standard treatment option in severe symptomatic aortic stenosis (AS). Extra-valvular cardiac damage affects prognosis after aortic valve replacement (AVR). Knowledge of cardiac remodeling factors that predict a poor outcome after TAVI is scarce. Purpose The study aim was to assess if echographic cardiac remodeling parameters could predict mid-term outcome after transfemoral TAVI with balloon expandable valves. Methods This study included 157 consecutive patients (76.4±7.3 yrs., 82 men) with severe AS at increased risk for surgical AVR, undergoing TAVI between October 2017 and December 2019. Patients underwent echocardiography before and 30 days after the procedure and clinical follow-up 3 years after inclusion. Pulmonary hypertension (PH) was defined as systolic pulmonary artery pressure (sPAP) >35 mmHg. Longitudinal LA strain was assessed in apical views. Peak values of global longitudinal LA strain (LAε), systolic strain rate (SSr) and early diastolic strain rate (ESr) were measured in all patients. Contractile LA function (late diastolic strain rate, ASr) was assessed in patients in sinus rhythm. Primary outcome was all-cause mortality. Results Three-year follow-up was available for all patients. Mean follow-up was 52±9 months (39-73). During follow-up 45 pts (76.8±7.3 yrs., 24 men) died, mid-term mortality was 29%. Between survivors and non-survivors, we found no differences in age and sex (p=0.3 and 0.8), LV ejection fraction, global longitudinal strain and AS severity. The presence of syncope, angina and significant coronary lesions before TAVI was similar between survivors and non-survivors. There was a significant reduction in sPAP after TAVI (33±11 vs 40±14, p<0.001), still PH was present 30 days after the procedure in 51 pts (32%). Patients with persistent PH at 30 days after TAVI had a significantly higher mid-term mortality (41.6% vs 22.4%, p=0.02), worse global LA strain (-9.2±4.8 vs -13.7±7.1%, p<0.001) and worse LA contractile function (ASr: -1.0±0.6 vs -1.3±0.6, p=0.002). Using binary logistic regression, atrial fibrillation (p=0.004), increased indexed LA volume (baseline and 30 days after TAVI, p=0.006 and p=0.021), impaired peak global LA strain after TAVI (p=0.04), baseline sPAP (p=0.03) and persistent PH (p=0.02) were significant univariate correlates of all-cause mortality. In the multivariable regression model, increased LA volume at baseline (OR 1.01, p=0.012) emerged as an independent predictor of death. Conclusions The mid-term follow-up revealed a mortality of 29% in high-risk patients with severe AS. No echocardiographic parameters of LV or RV systolic function or AS severity were associated with an increased risk of death. PH (at baseline and persistent), LA dilation and impaired LA strain after TAVI were associated with mortality. Increased LA volume before the procedure was an independent predictor of mid-term mortality after TAVI.
Read full abstract