Abstract
AIMS: In acute heart failure (AHF), precise assessment of congestion is critical to guide therapy. Urinary sodium (uNa) and urinary chloride (uCl) have emerged as potential biomarkers to monitor decongestion, but their comparative trajectories and links to residual congestion remain unclear. This study examined urinary uCl and uNa trajectories during AHF hospitalization in elderly patients with preserved ejection fraction (HFpEF) and their association with fluid overload. METHODS AND RESULTS: This prospective, single-centre study enrolled 70 patients hospitalized for AHF with HFpEF. All received intravenous furosemide for ≥72 h. Serial measurements of uNa, uCl, clinical congestion score (CCS), portal vein pulsatility and estimated plasma volume status (ePVS) were performed. Linear mixed-effects models analysed electrolyte trajectories in relation to residual congestion (CCS ≥ 2, portal vein pulsatility ≥ 30% and ePVS > 5.5 mL/g). The median age was 88 years (IQR: 85-91), and 72.8% were women. Baseline median uCl and uNa were 94 mmol/L (IQR: 68-116) and 81 mmol/L (IQR: 58-97), respectively. uCl declined significantly by 48 h (P = 0.029) and 72 h (P < 0.001). Higher uCl levels at 72 h were associated with CCS ≥ 2 (P for interaction = 0.039), portal vein pulsatility ≥ 30% (P for interaction = 0.018), and ePVS > 5.5 mL/g (P for interaction = 0.035). uNa trajectories differed significantly only across ePVS (P for interaction = 0.015). ROC AUC for predicting residual congestion was slightly higher for uCl (0.819) than uNa (0.790). The optimal cutoff value for uCl to identify residual congestion at 72 h was 61 mmol/L. CONCLUSIONS: In a cohort of elderly patients hospitalized for AHF, persistently elevated urinary chloride at 72 h of admission was associated with residual congestion. Urinary chloride may serve as a promising tool to guide the transition to oral medication once euvolaemia has been achieved.