High furosemide doses and survival in heart failure patients: A propensity score study

Abstract

Furosemide is widely used in heart failure (HF) patients, but so far no randomized studies have evaluated their long-term safety profile on HF patients survival. To determine if high furosemide doses (HFD) are associated with impaired survival in heart failure patients. A bi-centered HF registry of patients discharged after acute HF hospitalization was used to retrospectively analyze the two-year overall survival as the main end-point. High furosemide doses (HFD) were defined as daily dose at discharge > 80 mg/day (median dose). A propensity score (PS) was built with logistic regression, based on age, sex, left ventricular ejection fraction, atrial fibrillation (AF), coronary artery disease (CAD), anaemia, chronic obstructive pulmonary disease (COPD), chronic kidney disease (CKD) and cardio protective treatments: renin system antagonists (RSA) drugs and beta-blockers (BB). A total of 392 patients were used to build the PS ( Table 1 ). Mean age was 77.4 years, 37% were men. Mean comorbidity number was 3.9 (± 1.9) and two-year overall mortality was 42%, similar to other registries. In univariate analysis (X 2 ), HFD were associated with poorer survival compaired with low furosemide doses: 39% versus 69%, with odd-ratio (OR) of 3.5 [2.3–5.4] P < 10 −8 ). Other factors impacting survival were CAD (OR 1.7 [1.1–2.6], P < 0.01), COPD (OR 1.9 [1.1–3.1] P < 0.02), anemia (OR 1.6 [1.1–2.4] P < 0.03), SRA use (OR 0.6 [0.4–0.9] P < 0.03), hyponatremia (OR 4.5 [1.4–14] P = 0.008) and CKD (OR 2.1 [1.4–3.3] P < 10 −3 ). After PS quintiles stratification and log-rank test, mortality was higher in patients with HFD in all quintiles, reaching statistical significance in quintiles 2, 3, 4 and 5 (respectively P = 0.044, P = 0.018, P = 0.008 and P = 0.023) ( Fig. 1 ). HFD are associated with poorer survival in patients discharged after acute HF episode. Larger scale studies with PS matching are needed to confirm these results.