Timing matters: early furosemide administration and mortality in acute heart failure

Introduction Timely therapy is central in acute heart failure, but the real-world impact of delays in intravenous (IV) diuretics remains unclear. This study examined whether time from emergency department (ED) arrival to first IV furosemide was associated with admission, in-hospital mortality, and 1-year outcomes. Methods We performed a retrospective single-center study of patients that went to the ED in 2023 with acute heart failure and received IV furosemide. Demographic, clinical, laboratory, and treatment variables were extracted from electronic records. Outcomes included admission, length of stay, in-hospital mortality, and 1-year mortality. Group comparisons used Mann–Whitney U or χ² tests; logistic regression assessed associations between delay and outcomes. ROC analysis identified an optimal delay threshold, applied in Kaplan–Meier and Cox models. Results Among 305 patients (mean age 81 ± 11 years; 63% female), comorbidities were common: hypertension in 82%, atrial fibrillation in 45%, ischemic heart disease in 22% and diabetes in 38%. Median NT-proBNP was 12,683 pg/mL; mean estimated glomerular filtration rate, 47 ± 22 mL/min; mean LVEF, 51±13%. First furosemide dose averaged 46 ± 20 mg, with a mean administration delay of 10 ± 14 hours. Overall, 166 patients (55%) were admitted (median stay 9 days); 30 (10%) died in-hospital and 71 (23%) died within one year. Admitted patients had longer delays than discharged patients (6.8 vs 5.0 h; p = 0.021), and each additional hour increased odds of admission by 2.3% (OR 1.023; p = 0.029). In-hospital deaths had substantially longer delays (9.7 vs 5.3 h; p = 0.029); each hour increased odds of death by 3.4% (OR 1.034; p < 0.001). ROC analysis identified 10 hours as the optimal cut-off for predicting early mortality. Patients treated after ≥10 hours had higher early mortality (HR 2.38; 95% CI 1.16–4.87; p = 0.017), with early and persistent Kaplan–Meier separation. Delay was initially linked to 1-year mortality (OR 1.022; p = 0.012), but this association disappeared after excluding in-hospital deaths, indicating the long-term effect was driven by early mortality. Conclusion Delayed IV furosemide, especially beyond 10 hours, was associated with increased admission and markedly higher in-hospital and short-term mortality. The effect did not persist after discharge, suggesting the harm of delayed treatment occurs early. Timely furosemide administration in the ED is critical to improve outcomes.