Receiver operating curves (ROC) were obtained to identify the best cut-off of BNP, global longitudinal strain (GLS), E/E' and left atrial volume index (LAVI) for the identification of patients with HFpEF whenever compared with those without heart failure.We analyze data from 195 age-matched and sex-matched patients: 65 patients with heart failure with reduced ejection fraction, 65 patients with HFpEF and 65 stroke patients.
Of 1174 patients identified (537 [46%] female; median [interquartile range {IQR}] age, 56 [44-66] years), 250 were "at risk" for HFpEF given elevated brain-type natriuretic peptide (BNP) level; 160 had HFpEF by documented clinical diagnosis, and 745 did not have HFpEF.
In the near future, trials should target HFrEF patients using exclusion criteria sourced from landmark trials (e.g. severe renal impairment), select more homogeneous HFpEF populations (e.g. with elevated BNP and structural abnormalities on echocardiography), and determine which patients are likely to benefit from MRAs (e.g. according to prespecified biomarkers).