Katsuomi Iwakura, Shiro Hoshida, Nobuaki Tanaka, Masato Okada, Masahiro Seo, Takahisa Yamada, Masamichi Yano, Takaharu Hayashi, Yoshio Yasumura, Yusuke Nakagawa, Shunsuke Tamaki, Akito Nakagawa, Katsuki Okada, Yohei Sotomi, Shungo Hikoso, Daisaku Nakatani, Yasushi Sakata, PURUSUIT-HFpEF investigators
In patients hospitalized for HFpEF with persistent AF, LAP classification was associated with HF hospitalization, implying clinical relevance of the 2025 algorithm-based approach in this challenging population.
BACKGROUND: Non-invasive assessment of left atrial pressure (LAP) remains difficult in patients with atrial fibrillation (AF). The American Society of Echocardiography (ASE) updated the Recommendations for the Evaluation of Left Ventricular Diastolic Function and introduced a specific algorithm for LAP estimation in AF. We retrospectively investigated the prognostic value of the new algorithm-based LAP assessment in patients with heart failure with preserved ejection fraction (HFpEF) and persistent AF.
METHODS: This study included 363 patients hospitalized for acutely decompensated HFpEF who had persistent AF at discharge. LAP was categorized as normal, elevated, or indeterminate using echocardiography at discharge, according to the algorithm in the 2025 ASE guidelines. Multivariable Cox proportional hazards models were constructed to assess associations between LAP classification and clinical outcomes such as death, heart failure (HF) hospitalization and stroke during follow-up of 574 days.
RESULTS: LAP was classified as elevated in 93 patients (25.6%) and normal in 165 (45.5%), while it was indeterminate in 105 (28.9%) cases due to missing variables recommended in the 2025 guidelines. Elevated LAP was independently associated with HF hospitalization during follow-up period (HR 1.57, 95% CI 1.05-2.37, p=0.03), whereas it was not identified as an independent predictor for death or stroke after adjustment. This association remained significant after accounting for the competing risk of death and was largely preserved across multiple imputation and sensitivity analyses using different assumptions for indeterminate cases. The incremental value of LAP estimation for predicting HF hospitalization over natriuretic peptides was modest and not statistically significant.
CONCLUSIONS: In patients hospitalized for HFpEF with persistent AF, LAP classification was associated with HF hospitalization, implying clinical relevance of the 2025 algorithm-based approach in this challenging population.