Predictors of left ventricular ejection fraction recovery after transcatheter aortic valve replacement (TAVR)

ABSTRACT

Introduction: Transcatheter aortic valve replacement (TAVR) has proven benefits in patients with reduced left ventricular ejection fraction (LVEF). A significant proportion of them shows recovery of systolic function Objective: To analyse the main baseline, electrocardiographic and echocardiographic characteristics that may predict LVEF recovery after TAVR. Methods: A cohort study was conducted. Consecutive patients undergoing TAVR in our center from January 2012 to December 2020 were included. Baseline clinical profile, electrocardiographic (EKG), echocardiographic (ECH) parameters were recorded, as well as MACE during followup (major adverse cardiovascular events including all-cause mortality, myocardial infarction, cerebrovascular accident and heart failure hospitalization). Reduced systolic function was defined as LVEF <50%. We considered recovery of systolic function as LVEF ≥50% at follow-up. Results: A total of 292 patients were included. 48% were women and the median age was 81.07 years (77.63-86.22). 22.6% (66 patients) had reduced LVEF at baseline. Half of them showed recovered systolic function during follow-up. Patients who did not recovered LVEF had a higher prevalence of dyslipidemia and peripheral artery disease. History of cardiac surgery was more frequently found in this group, and they showed a higher surgical risk estimated by EuroScore II. They had lower LVEF and aortic valve mean gradient, and more frequently presented non-synus rhythm (NSR), left bundle branch block and right ventricular dysfunction (RVD). These characteristics are shown in figure 1. In univariate analysis lower Euroscore II, presence of synus rhythm, absence of LBBB and RVD, as well as higher aortic valve mean gradient were predictors of LVEF recovery. In multivariate analysis RVD and mean aortic gradient were independent predictors. Among all patients included in our study, those presenting with RV dysfunction were significantly associated with lower LVEF mean values (46,0% vs 57,2%;p<0,01) After a median follow-up of 21.3 (8.52-38.94) months, MACE were lower in recovered LVEF group (HR 0.25 95% CI 0.05-1.21). There were no statistically significant differences in all-cause mortality, nevertheless there was a trend towards a higher non-cardiovascular mortality in this group, essentially at the expense of deaths from malignant neoplasms and SARS-COV- 2 infections. Survival curves for MACE are represented in figure 2. Conclusion: In our study, half of the patients with impaired ventricular function undergoing TAVR showed recovery of ejection fraction. Right ventricular function and aortic valve mean gradient at baseline were independent predictors of recovery. Identifying predictors of LVEF recovery is fundamental in the evaluation of potential candidates for TAVR, and can help clinicians assess risks and benefits, as well as long-term prognosis of these patients.