Abstract
BACKGROUND: Though commonly noted in clinical practice, it is unknown if decongestion in acute heart failure (AHF) results in increased serum bicarbonate. METHODS AND RESULTS: For 678 AHF patients in the DOSE-AHF, CARRESS-HF, and ROSE-AHF trials, we assessed change in bicarbonate (baseline to 72-96 hours) according to decongestion strategy, and the relationship between bicarbonate change and protocol-defined decongestion. Median baseline bicarbonate was 28 mEq/L. Patients with baseline bicarbonate ≥28 mEq/L had lower ejection fraction, worse renal function and higher N-terminal pro-B-type natriuretic peptide than those with baseline bicarbonate <28 mEq/L. There were no differences in bicarbonate change between treatment groups in DOSE-AHF or ROSE-AHF (all P > .1). In CARRESS-HF, bicarbonate increased with pharmacologic care but decreased with ultrafiltration (median +3.3 vs -0.9 mEq/L, respectively; P < .001). Bicarbonate change was not associated with successful decongestion (P > .2 for all trials). CONCLUSIONS: In AHF, serum bicarbonate is most commonly elevated in patients with more severe heart failure. Despite being used in clinical practice as an indicator for decongestion, change in serum bicarbonate was not associated with significant decongestion.