Diagnosis of heart failure with preserved ejection fraction in patients with dyspnea and paroxysmal atrial fibrillation: a role of left atrial strain
A Katbeh,T De Potter,P Geelen,G Di Gioia,M Kodeboina,Z Balogh,M Albano,M Vanderheyden,J Bartunek,E Barbato,G Van Camp,M Penicka
DOI: https://doi.org/10.1093/ehjci/ehaa946.0096
IF: 39.3
2020-11-01
European Heart Journal
Abstract:Abstract Background Diagnosis of heart failure with preserved ejection fraction (HFpEF) in patients with dyspnea and paroxysmal atrial fibrillation (AF) is challenging. Speckle tracking-derived left atrial strain (LAS) provides an accurate estimate of left ventricular filling pressures and left atrial phasic function. However, data on clinical utility of LAS in patients with dyspnea and AF are scarce. Objective To assess relationship between LAS and probability of HFpEF in patients with dyspnea and paroxysmal AF. Methods The study included 205 consecutive patients (62±10 years, 58% males) with limiting dyspnea (NYHA ≥ II), paroxysmal AF and preserved LVEF (≥50%), who underwent speckle tracking echocardiography and natriuretic peptide (NT-proBNP) assessment during sinus rhythm. Patients with manifest ischemic heart or valve disease, and cardiomyopathy were excluded. Probability of HFpEF was estimated using H2FPEF and HFA-PEFF scores, which combine clinical characteristics, echocardiographic parameters and natriuretic peptides. Results A total of 61 (30%), 115 (56%) and 29 (14%) had respectively high, intermediate and low probability of HFpEF. Patients with high probability of HFpEF were significantly older, had higher body mass index, NT-proBNP, E/e', pulmonary artery pressure and larger LA volume index than patients in low-to-intermediate probability groups (all p<0.05). Two distinct patterns of LA phasic function were observed. Firstly, reservoir LAS showed close inverse association with increasing probability of HFpEF. Secondly, contractile LAS showed initial decrease with subsequent compensatory increase in intermediate probability category with final decrease in patients with high HFpEF probability. In contrast, LV global longitudinal strain was similar between groups (NS). In multivariable regression analysis, reservoir LAS emerged as the strongest independent predictor of HFpEF defined by using both scores. Reservoir LAS with optimal cut off value of 24% showed sensitivity of 86% and specificity of 70% to identify high probability of HFpEF. Combination of LAS with NT-proBNP did not increase the accuracy of each parameter alone. Conclusions Reservoir LAS shows a strong independent association with probability of HFpEF in patients with dyspnea and paroxysmal AF. This advocates for more liberal use of LAS assessment to distinguish cardiac from non-cardiac dyspnea in patients with history of AF. Funding Acknowledgement Type of funding source: Public Institution(s). Main funding source(s): International PhD programme in Cardiovascular Pathophysiology and Therapeutics (CardioPaTh).
cardiac & cardiovascular systems
What problem does this paper attempt to address?