J Am Coll Cardiol. 2026 Jan 7:S0735-1097(25)10279-9. doi: 10.1016/j.jacc.2025.11.024. Online ahead of print.
ABSTRACT
BACKGROUND: Echocardiographic grading of left ventricular diastolic function is recommended to guide diagnostic evaluation of heart failure with preserved ejection fraction (HFpEF). A new algorithm for diastolic function interpretation has been proposed, but it has not yet been systematically evaluated in HFpEF.
OBJECTIVES: The purpose of this study was to determine the false-negative rate of the 2025 American Society of Echocardiography (ASE) algorithm among invasively confirmed ambulatory HFpEF, assess temporal changes in diastolic grades between decompensated and recompensated hospitalized HFpEF, and, secondarily, to compare diagnostic discrimination with existing HFpEF algorithms and prognostic associations.
METHODS: Echocardiography was performed in 2 HFpEF cohorts: 1) ambulatory, compensated patients undergoing invasive hemodynamic exercise testing as part of a prospective cohort study, with an external validation cohort; and 2) hospitalized/decompensated patients both acutely and following recompensation. For secondary analyses, we included noncardiac dyspnea controls and compared performance with existing algorithms.
RESULTS: In the ambulatory/compensated HFpEF cohort, 248 of 756 (32.8%) were graded normal, 263 of 756 (34.8%) had Grade 1 diastolic dysfunction, 219 of 756 (30.0%) had Grades 2 to 3, and 26 of 756 (3.4%) were indeterminate. Among those labeled normal or Grade 1, >60% had resting pulmonary artery wedge pressure ≥15 mm Hg at catheterization. In decompensated HFpEF, 22 of 88 (25.0%) showed normal or Grade 1, and this proportion increased to 45 of 88 (51.1%) after recompensation. In HFpEF with Grade 1 undergoing simultaneous hemodynamic exercise testing with stress imaging, only 11 of 116 (9.5%) met the ASE-recommended stress criteria, resulting in a 90.5% false-negative rate. Similar findings were observed in the external validation cohort. The 2025 ASE algorithm poorly discriminated HFpEF from noncardiac dyspnea (AUC: 0.61). Patients with HFpEF labeled as normal or Grade 1 had 5-fold higher risk for all-cause death or heart failure hospitalization compared with controls (HR: 5.37; 95% CI: 1.27-22.6).
CONCLUSIONS: Among patients with invasively proven HFpEF, the 2025 ASE algorithm frequently assigns normal or low diastolic grades, and the recommended stress criteria detect only a minority of cases. Although echocardiography remains essential to guide HFpEF evaluation, current algorithms proposed have inadequate sensitivity. Diastolic function grades must be interpreted in the context of pretest probability and HFpEF-specific, evidence-based frameworks, rather than used in isolation to exclude disease.
PMID:41532943 | DOI:10.1016/j.jacc.2025.11.024