ASSESSING THE PREDICTIVE PERFORMANCE OF HFA-PEFF AND H₂FPEF SCORES IN PATIENTS WITH SUSPECTED HEART FAILURE AND PRESERVED EJECTION FRACTION

Abstract

Two scoring systems, HFA-PEFF and H₂FPEF, have been developed for the diagnosis of heart failure with preserved ejection fraction (HFpEF) and have also demonstrated prognostic value in individuals with exertional dyspnea. A total of 69 patients suspected of having HFpEF, based on clinical symptoms and transthoracic echocardiography, were included and monitored for 12-18 month. Both HFA-PEFF and H₂FPEF scores were calculated upon enrolment, and patients were classified as having intermediate or low risk based on these scores. During the monitoring period, 27 patients (39% of the cohort) experienced cardiac decompensation.  Receiver operating characteristic (ROC) analysis showed that both scoring systems can predict cardiac decompensation in patients with suspected HFpEF, with an area under the curve (AUC) of 0.730 for the HFA-PEFF score and 0.720 for the H₂FPEF score. Univariate logistic regression analysis revealed that patients with an HFA-PEFF score of ≥ 3.5 had an Odds Ratio (OR) of 7.60 for cardiac decompensation, while those with an H₂FPEF score of ≥ 3.5 had an OR of 6.87 for cardiac decompensation. In the multivariate analysis, both scores remained predictive, with ORs of 5.07 for HFA-PEFF and 4.65 for H₂FPEF for cardiac decompensation. The confusion matrices showed accuracies of approximately 67.74% for HFA-PEFF and 64% for H₂FPEF in predicting the cardiac decompensation. When both scores were combined, their accuracy was also 67.74%. In conclusion, both the HFA-PEFF and H2FPEF scoring systems demonstrated moderate predictive value in assessing the risk of cardiac decompensation in patients suspected of HFpEF, maintaining their significance in multivariate analysis.