Abstract Body (Do not enter title and authors here): Introduction: Achieving effective decongestion is pivotal in the management of acute decompensated heart failure (ADHF), and body weight (BW) change is routinely tracked as an index of fluid off-loading, but it integrates alterations in both intravascular and interstitial compartments. Conversely, reductions in the natriuretic peptides directly reflect falling ventricular wall stress and are endorsed by guidelines as biochemical surrogates of decongestion. Using a contemporary, multicenter ADHF registry, we evaluated the individual associations of in-hospital BW reduction and natriuretic peptide decline with one-year mortality or heart failure rehospitalization.
Methods: Data was extracted from a multicenter, prospective cohort study enrolling consecutively patients hospitalized with ADHF at 11 tertiary care hospitals in the Tokyo area (2018–2024). Patients who required intravenous inotropes, mechanical circulatory support, dialysis, and those who died during hospitalization were excluded. Multivariable Cox models were used to examine associations between the 1-year composite outcome (all-cause death or rehospitalization for heart failure) and changes in BW and BNP/NT-proBNP levels. Interaction between BW and BNP/NT-proBNP changes was also evaluated. Restricted cubic splines assessed potential non-linear relationships.
Results: Among 4,095 patients (mean age 76.6 years, 43.1% female, BW 60.7 kg, BNP 934 pg/mL), the median in-hospital BW change was −4.5 kg (IQR −7.4 to −2.4), and BNP/NT-proBNP change was −63.6% (IQR −79.8 to −37.2%). After multivariable adjustment, the risk of all-cause death or heart failure rehospitalization at one year declined across successive tertiles of weight loss: second tertile, HR 0.80 (95% CI 0.67–0.96); third tertile, HR 0.78 (95% CI 0.65–0.93), versus the first tertile. Restricted cubic spline modelling confirmed a non-linear association. In contrast, BNP/NT-proBNP change showed a linear association with the outcome, with each 10-percent decrement conferring an HR 0.93 (95% CI 0.90–0.95; p < 0.001). The interaction between BW and BNP/NT-proBNP changes was non-significant (p = 0.47), indicating that each metric provides independent prognostic information.
Conclusion: These findings demonstrate the complementary and independent value of tracking both BW and BNP/NT-proBNP trajectories, suggesting that their combined assessment may further refine evaluation of decongestion and improve post-discharge risk stratification in ADHF.