Prognostic significance of echocardiographic abnormalities in patients with heart failure with preserved ejection fraction

Abstract Funding Acknowledgements Type of funding sources: None. Background Heart failure with preserved ejection fraction (HFpEF) is recognised as a highly heterogenous condition (1). This extends to its echocardiographic phenotype, which according to the latest ESC diagnostic criteria, can include the presence of left ventricular hypertrophy (LVH), left atrial (LA) enlargement and elevated pulmonary artery systolic pressure (PASP) amongst others (2). The relative prognostic significance of these abnormalities however has not been well-described. Purpose The aim of this study was to evaluate the relative prognostic significance of echocardiographic abnormalities in a community-based cohort of patients with HFpEF and elevated NT-proBNP. Methods We analysed the echocardiograms of patients referred from the community with clinical signs of HF and elevated NTproBNP. We specifically evaluated the presence of LVH, LA enlargement and elevated PASP according to EACVI criteria. We excluded patients with NT-proBNP <125 pg/ml. The primary outcome was time to all-cause mortality or cardiovascular hospitalisation. Results The total cohort included 473 patients (mean age 78 years, 66% female, median NT-proBNP 587 pg/ml). Median follow-up was 27 months. 88 individuals had an occurrence of the primary outcome (18.6%). LA enlargement was the commonest abnormality (48.2%), then LVH (37.2%) and elevated PASP (18.4%). After adjustment for age, sex and NTproBNP, LA enlargement was significantly associated with the primary outcome in patients in sinus rhythm but not atrial fibrillation (sinus: hazard ratio 2.02, 95% CI 1.09–3.73, p = 0.025; AF: hazard ratio 0.68, 95% CI 0.31–1.48, p = 0.20; interaction p value 0.013). Neither elevated PASP (HR 1.50, 95% CI 0.94–2.39, p = 0.09) or LVH (HR 0.99, 95% CI 0.64–1.55, p = 0.98) were independently associated with the primary outcome. There did however appear to be a stepwise increase in risk based on the number of echocardiographic abnormalities present, with individuals with 2 or more abnormalities being significantly more likely to have an incidence of the primary outcome than those with fewer (Figure). Conclusion There may be differences in prognosis of HFpEF patients based on echocardiographic structural and functional parameters. These may therefore require different therapeutic approaches. The number of abnormalities appear to have a cumulative adverse effect on prognosis.