Introduction: Transcatheter aortic valve replacement (TAVR) has become an efficient and safe alternative to surgical aortic valve replacement (SAVR). While severe aortic stenosis as well as severe aortic regurgitation (AR) are known to negatively impact left ventricular ejection fraction (LVEF), prior studies have shown that TAVR can lead to an improvement in LVEF. Thus far, little is known about the prognostic implication of LVEF improvement as a sole predictor of outcomes. Therefore, the aim of this study was to assess the prognostic impact of LVEF impairment before TAVR, as well as early LVEF improvement in patients undergoing TAVR. Materials and Methods: Patients undergoing TAVR in a large tertiary university hospital were consecutively included in a prospective registry. Transthoracic echocardiography (TTE) was performed at baseline, after 1 month and annually thereafter. Significant LVEF improvement was defined as a relative increase of ≥10% in LVEF at 30 days compared to baseline LVEF. The primary outcome was all-cause mortality at 1 year. Secondary outcomes were major adverse cardiovascular events (MACEs) including cardiovascular death, non-fatal myocardial infarction, stroke, bleeding and unplanned re-interventions of the aortic valve at 5 years. Results: Among 1655 patients who underwent TAVR between September 2011 and April 2024, the LVEF at baseline was available for 1556 patients. Of these, 1031 patients (66.2%) had preserved LVEF at baseline (LVEF ≥ 53%), whereas 303 patients (19.5%) had moderately reduced LVEF (40-52%) and 222 patients (14.3%) had severely reduced LVEF (<40%). Out of the patients with impaired LVEF, 155 (40.4%) patients showed a significant improvement in LVEF ≥10% after 30 days, while 229 (60.6%) patients showed no significant LVEF improvement (<10%). Patients with preserved LVEF at baseline had significantly better mortality outcomes than those with severely reduced LVEF (p < 0.001). LVEF improvement was associated with a survival benefit after 1 year (p = 0.009, HR 2.68, 0.95 CI 1.23-5.85) which diminished after 5 years (p = 0.058), but patients with LVEF improvement showed lower MACE rates at 5 years (p < 0.001). Conclusions: Preserved LVEF before TAVR is an independent predictor for improved outcomes. Additionally, early improvement in LVEF is associated with beneficial outcomes in patients undergoing TAVR.