Biomarker-defined congestion profiles in very old adults with acute heart failure: beyond ejection fraction classification
摘要
Congestion is the main driver of worsening acute heart failure, yet its manifestation in very old adults and its relationship with prognosis remains incompletely understood. Traditional classification based on left ventricular ejection fraction (LVEF) may not adequately capture the biological heterogeneity of congestion in this population. We aimed to characterize patient subgroups defined by BNP and CA125 constellations, to characterize their clinical and echocardiographic features, and to evaluate their prognostic implications. As a secondary objective, we explored their relationship with conventional LVEF categories.
MethodsWe conducted a retrospective cohort study including 830 consecutive patients hospitalized with acute heart failure. Congestion was assessed using clinical signs, echocardiography, biomarkers (BNP and CA125), and point-of-care ultrasound. Outcomes included heart failure rehospitalization and all-cause mortality over a median follow-up of 310 days (IQR 62–543). Multivariable Cox regression models were used.
ResultsThe median age was 87 years, and 65.6% were women. Four patient subgroups were defined by BNP and CA125 constellations, showing distinct clinical and echocardiographic profiles. Mortality increased progressively across phenotypes, with combined elevation of BNP and CA125 independently associated with higher risk of death (HR 1.48; p = 0.008). Isolated CA125 elevation was associated with increased rehospitalization risk (HR 1.45; p = 0.031). In contrast, LVEF categories showed limited prognostic discrimination.
ConclusionsIn very old adults hospitalized with acute heart failure, combined elevation of BNP and CA125 was independently associated with increased mortality, while isolated CA125 elevation was associated with higher rehospitalization risk. LVEF was not independently associated with either outcome. These findings suggest that integrating BNP and CA125 may complement conventional LVEF classification for risk stratification in this population.