Aims
Up to one third of patients with aortic stenosis (AS) undergoing transcatheter aortic valve implantation (TAVI) present with reduced left ventricular ejection fraction (LVEF), which is linked to poor outcomes. Recovery of LVEF after TAVI improves prognosis, but reliable predictors remain limited. This study aimed to assess the prognostic impact of LVEF recovery and identify predictors of recovery or normalization, including non-invasively measured end-systolic elastance (Ees), a surrogate of contractility.
Methods
We analyzed patients with AS and a baseline LVEF of ≤35% at a high-volume centre with available follow-up echocardiographic exams one year post TAVI. Patients were categorized depending on their LVEF one year after TAVI into no recovery, recovery (≥10% increase in LVEF) or normalisation (LVEF ≥50%). The three groups were compared regarding clinical outcomes and a multivariable analysis was conducted to identify predictors for LVEF recovery or normalization. The primary endpoint was defined as all-cause mortality at three years. All patients were additionally screened for eligibility for non-invasive measurement of Ees by Chen’s method and, if suitable, included in a subanalysis to investigate influence of Ees on LVEF recovery or normalisation.
Results
A total of 320 patients (2015 – 2025) were included, with a mean age of 77.7 (± 8.0) years, 71% male, and 63% were treated with a balloon-expandable valve. Of these, 110 (34%) patients did not experience recovery of LVEF, whereas 111 (35%) patients achieved recovery and 99 (31%) patients achieved normalisation. Male sex and prior myocardial infarction were more frequent in the group of patients without LVEF recovery, while other baseline and procedural characteristics were similar. The primary endpoint of all-cause mortality at three years occurred more frequently in patients without recovery (22% [n=25] vs. 14% [n=16] vs. 8% [n=8], p=0.01). In the multivariable analysis, a baseline mean gradient of ≥ 30 mmHg predicted LVEF recovery or normalisation (OR 3.11, 95% CI 1.65–5.83, p<0.001), whereas a high left ventricular end diastolic diameter (LVEDD) was associated with poor recovery (OR 0.94, 95% CI 0.90–0.98, p=0.003). In a subset of 167 eligible patients, Ees was significantly higher in patients that achieved LVEF recovery or normalisation (1.32 [1.04–1.63] mmHg/ml vs. 1.71 [1.44–2.08] mmHg/ml vs. 2.00 [1.54–2.54] mmHg/ml, p<0.001) and emerged as an independent predictor of LVEF recovery or normalisation (OR 6.38, 95% CI 2.67–15.23, p<0.001) in addition to a mean gradient of ≥30 mmHg (OR 2.78, 95% CI 1.19–6.46, p<0.001).
Conclusion
Most patients undergoing TAVI with a LVEF <35% experience recovery or normalization of LVEF, which is associated with improved long-term outcomes. This study refines current prediction of LVEF recovery by introducing Ees as a novel predictor. Association of the latter with LVEF recovery suggests that LVEF impairment may be reversible in patients who retain left ventricular contractility, evidenced by a high mean gradient and Ees, but present with reduced LVEF, likely due to the increased afterload from AS. These findings may enhance decision-making in this specific patient cohort.
