Aims: Loop diuretics are recommended to treat congestion in heart failure (HF), despite limited quality evidence. High-dose (HD) loop diuretics seem to worsen outcomes in chronic HF, though; data for acute HF are scarce, with equivocal results.
Methods and results: The ALARM-HF study recorded in-hospital HF therapy in 4953 patients from nine countries. A post-hoc analysis was performed to determine if there was an interaction between intravenous (iv) bolus diuretic dosing and outcomes. Patients were classified as receiving high- or low-dose iv furosemide if their total initial 24 h dose was above (HD) or below [low dose (LD)] 1 mg/kg. Propensity scoring, matching an extensive list of variables, was performed. High-dose and LD patients were matched by propensity scores and outcomes determined. We identified 2460 LD and 848 HD patients, with overall in-hospital mortality of 9 and 13% (P= 0.002), respectively. After propensity matching, there were 506 patients in each subgroup, with the matched LD and HD cohorts having similar mortality (13 vs. 15%; P= 0.4). We further investigated in which subgroups of patients HD diuretics influenced mortality. Before matching, HD diuretics were associated with a greater risk of in-hospital death in some subgroups, including patients aged >80 years, those with an acute coronary syndrome, or with a left ventricular ejection fraction <40%. However, after propensity score matching, no association was found between diuretic dosing and death in any of the studied subgroups.
Conclusions: In the initial management of acute HF, HD iv diuretics, per se, do not influence short-term mortality.