CONGRESS ABSTRACT

Background. Haemodynamics of patients with heart failure (HF) with reduced ejection fraction (HFrEF) is often characterised by increased left ventricular (LV) filling pressures. Non-invasive estimation of pulmonary artery wedge pressure (PAWP) is useful in the clinical management of these patients. The left atrial (LA) filling index could help estimate PAWP in this scenario. Methods. Patients undergoing right heart catheterisation (RHC) and standard and advanced echocardiography were prospectively enrolled. Patients with pulmonary hypertension underwent a reversibility test and were re-evaluated. LA filling index was defined as mitral E wave/peak LA longitudinal strain. We assessed the accuracy of this parameter in detecting patients with increased PAWP at RHC (i.e. PAWP >15 mmHg or 20 mmHg), its clinical associations, and the prognostic implications. The results were validated considering non-invasive PAWP estimation according to guidelines. The primary outcome was death/heart transplant/LV assist device implantation/HF hospitalisation. Results. One-hundred-eight patients were enrolled. Mean age was 52±12 years, 37% had a history of atrial fibrillation, mean LV ejection fraction was 29±11%, and median NTproBNP was 1538 (862-3144) pg/ml. Patients with higher LA filling index had more signs of severe disease and worse haemodynamics at invasive assessment. LA filling index showed a significant association with PAWP (p<0.001, R=0.407) and was the best non-invasive parameter in detecting PAWP > 20 mmHg (AUC 0.741, p<0.001). When validated on increased LA pressure defined by non-invasive assessment according to current recommendations, the accuracy was lower (AUC 0.614, p=0.056). In patients undergoing the reversibility test, we observed a reduction in invasive PAWP and an increase in LA filling index. The changes in LA filling index and PAWP had a strong correlation (p<0.001, R=0.740). Patients with LA filling index above 6.7 cm/sec x %-1 had an increased risk of death/heart transplant/LV assist device implantation/HF hospitalisation (HR each 5 points 1.19 (95% C.I. 1.02 – 1.41), p=0.043). Conclusions. LA filling index in patients with HFrEF is a valid non-invasive parameter to identify patients with increased PAWP. The correlation with non-invasive estimation of increased PAWP was lower. The improvement of haemodynamic conditions led to a reduction in PAWP and in LA filling index. PAWP and LA filling index changes are strongly correlated.