Validation of the HFA-PEFF and the H2FPEF scores for the diagnosis of heart failure with preserved ejection fraction in Japanese patients:a report from the Japanese multicentre registry

Abstract

Abstract Background The standard diagnosis of heart failure (HF) with preserved ejection fraction (HFpEF) is based on the following: 1) symptoms of HF, 2) preserved left ventricular (LV) ejection fraction (LVEF, >50%), and 3) presence of LV diastolic dysfunction confirmed by echocardiography or cardiac catheterisation. However, there are limits to the diagnostic accuracy of individual parameters, and what cut-off values should be applied and how they should be combined remain unclear. Diagnostic algorithms for HFpEF such as the HFA-PEFF algorithm and the H2FPEF score have been proposed; however, previous validation studies were conducted in stable chronic HF and did not include an invasive haemodynamic assessment. Thus, the diagnostic accuracy for HFpEF lacked robustness. Moreover, information on their applicability in the Asian population is limited. Purpose The aim of this study was to investigate these scores' diagnostic validity for HFpEF in Japanese patients recently hospitalised due to acute decompensated HF. Methods We examined patients with HFpEF recently hospitalised with acute decompensated HF whose HFA-PEFF and H2FPEF scores could be calculated at discharge from a nationwide HFpEF-specific multicentre registry (HFpEF group) and control patients who underwent echocardiography to investigate the cause of dyspnoea in our hospital (Non-HFpEF group). We calculated the HFA-PEFF and the H2FPEF scores among the studied population. Receiver operating characteristic (ROC) curves and area under the curve (AUC) were computed to compare the diagnostic accuracy of these scores. Results The studied population included 372 consecutive patients (194 HFpEF group and 178 Non-HFpEF group; HFpEF prevalence, 52%). The HFA-PEFF score classified 155 (42%) of all patients into the high likelihood category (5–6 points) and only 19 (5%) into the low likelihood category (0–1 point). A high HFA-PEFF score could diagnose HFpEF with a high specificity of 84% and a positive predictive value (PPV) of 82%, and a low HFA-PEFF score could rule out HFpEF with a high sensitivity of 99% and a negative predictive value (NPV) of 89%. The H2FPEF score classified 86 (23%) of all patients into the high likelihood category (6–9 points) and 84 (23%) into the low likelihood category (0–1 point). HFpEF could be diagnosed with a high H2FPEF score (specificity, 97%; PPV, 94%) or ruled out with a low H2FPEF score (sensitivity, 97%; NPV, 93%). The diagnostic accuracy for the HFA-PEFF and H2FPEF scores was 0.82 (95% confidence interval [CI] 0.78–0.86) and 0.89 (95% CI 0.86–0.93), respectively, by the AUC of the ROC curve (P=0.004) (Figure 1A). In the HFA-PEFF sub-scores, the functional score showed little diagnostic value, while the morphological and biomarker scores showed moderate diagnostic value (Figure 1B). Conclusions The H2FPEF score may be more useful than the HFA-PEFF score in diagnosing HFpEF in Japanese patients. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (JSPS KAKENHI) Figure 1