Abstract 11590: Body Mass-Corrected Estimated Plasma Volume at Discharge Predicts Mortality of Heart Failure Patients With Non-Normal and Normal Left Ventricular Ejection Fraction

Abstract

Introduction: Estimated plasma volume(ePV) is a parameter of body fluid volume and an indicator for residual congestion. Even in the era of novel cardioprotective drugs, achieving adequate body fluid volume with heart failure is sometimes challenging. We investigated the prognosis of heart failure patients with a one-time evaluation of ePV at discharge corrected by body mass index (BMI). Since current heart failure treatment algorithms follow different paths with left ventricular ejection fraction (LVEF), we also examined whether these groups have a difference in prognosis prediction. Methods: This is a single-center, prospective observational study. A total of 1427 consecutive heart failure hospitalizations, excluding acute coronary syndromes, were included from January 2018 to June 2022. We evaluated 858 patients, excluding repeat hospitalizations of the same patient, in-hospital deaths, transfers for cardiac surgery and forced discharges, and missing data. A prognosis investigation was conducted for all patients. BMI and ePV (ml, Hakim's formula) was calculated from height, weight, and hematocrit at discharge. Survival analysis was performed with ePV divided by BMI (ePV/BMI). The primary endpoint was all-cause death. Results: Median of 527 days of post-discharge observation, 286 deaths occurred. BMI and ePV at discharge show a positive significant correlation in both normal (≧50%) and non-normal LVEF (<50%) groups. Among the mortality group during follow-up, ePV/BMI at discharge was significantly higher than the non-mortality group. Survival analysis with Kaplan-Meier curve by ePV/BMI three tertiles (Low, Medium, High) in normal (66.8% vs 50.1% vs 44.0%, Logrank P<0.001), non-normal LVEF (72.2% vs 69.3% vs 49.6%, Logrank P<0.001) and overall (70.6% vs 55.5% vs 47.4%, Logrank P<0.001) population had a significantly worse three-year prognosis in the high ePV/BMI group. Multivariable Cox proportional hazard model with covariant univariable predictive factors shows higher ePV/BMI tertile, as well as a continuous variable of ePV/BMI, was predictive factor of death in normal, non-normal LVEF and overall population. Conclusions: Even after therapeutic intervention, ePV/BMI at discharge is a useful predictor of heart failure prognosis.