Serum Albumin Modifies the Prognostic Meaning of BNP in Acute Decompensated Heart Failure.

摘要

BACKGROUND: BNP is widely used for risk stratification in acute decompensated heart failure (ADHF), but its prognostic meaning may differ across clinical phenotypes, particularly cardiorenal syndrome type 1 (CRS1). HYPOTHESIS: We hypothesized that serum albumin modifies the association between BNP and adverse in-hospital course in ADHF. METHODS: In this retrospective cohort study, adults hospitalized with ADHF were classified as CRS1 or control. The primary endpoint was adverse in-hospital course, defined as in-hospital death and/or prolonged length of stay (≥ 75th percentile). BNP was analyzed on the natural log scale. Multivariable logistic regression included albumin, ln(BNP), and an albumin × ln(BNP) interaction term, adjusted for age, sex, systolic blood pressure, eGFR, ejection fraction, CRS status, body mass index, and diabetes mellitus. RESULTS: In the complete-case cohort (N = 292; events = 77), albumin significantly modified the BNP-risk association (interaction β = -0.236 per 5 g/L; Wald p = 0.045; likelihood-ratio test p = 0.050). The adjusted odds ratio for a 1-unit increase in ln(BNP) was 1.81 at albumin 30 g/L, 1.42 at 35 g/L, and 1.12 at 40 g/L. Discrimination improved modestly with the interaction (AUC 0.714 vs. 0.726), and the high-BNP/low-albumin stratum showed the highest observed event risk. CONCLUSIONS: Serum albumin modifies the prognostic meaning of BNP in ADHF. Integrating albumin into BNP interpretation identifies a high-risk "congestion plus vulnerability" phenotype and may support pragmatic triage and discharge planning in CRS-enriched ADHF populations.