Diuretic dosing and outcomes with torsemide and furosemide following hospitalization for heart failure: the TRANSFORM-HF trial

Abstract Background The TRANSFORM-HF trial found no significant difference in all-cause mortality and all-cause hospitalization following hospitalization for heart failure (HF) with a diuretic strategy of torsemide versus furosemide. However, the impact of diuretic dosing on the trial results and whether the relationship between dosing and outcomes differs between torsemide and furosemide is unclear. Purpose To assess and compare the associations between diuretic dosing and outcomes with torsemide and furosemide. Methods TRANSFORM-HF was an open-label, pragmatic trial that randomized 2859 patients hospitalized for HF in the United States to a diuretic strategy of torsemide versus furosemide after discharge. These post-hoc analyses categorized patients according to investigator-selected discharge dose tertile of torsemide and furosemide, i.e. tertile 1) ≤40 mg, 2) >40-80 mg and 3) >80 mg of furosemide equivalents using a 2:1 conversion with torsemide. Study endpoints were all-cause mortality, all-cause hospitalization, composite all-cause mortality or hospitalization, and change from baseline to 12 months in the Kansas City Cardiomyopathy Questionnaire Clinical Summary Score (KCCQ-CSS). Results Dosing data were available for 2379 (83%) of patients with median age 65 years (IQR: 56, 75), 883 (37.1%) were women, and 812 (34.2%) were Black. When categorized by discharge dose, a total of 1136 (47.8%) patients were included in dose tertile 1, 696 (29.2%) in tertile 2 and 547 (23.0%) in tertile 3. For all-cause mortality, furosemide showed adjusted hazard ratios (aHR) of 1.21 (95% CI: 0.91, 1.59) for tertile 2 and 1.40 (95% CI: 1.04, 1.88) for tertile 3 compared with tertile 1. For torsemide aHR for all-cause mortality were 1.74 (95% CI: 1.32, 2.30) for tertile 2 and 1.58 (95% CI: 1.14, 2.19) for tertile 3 compared with tertile 1 (Table 1). The risk of adverse events increased similarly with dose for both diuretics and no statistically significant interaction between loop diuretic and dose was observed (Pinteraction=0.17). Likewise, for the endpoints of all-cause hospitalization and the composite of all-cause mortality and hospitalization, higher doses of furosemide and torsemide aligned with higher risk of both endpoints, and interaction p-values that did not demonstrate statistical significance by loop diuretic (all-cause hospitalization: Pinteraction= 0.28, all-cause mortality and hospitalization: Pinteraction=0.38) (Table 1). Changes in KCCQ-CSS were similar across the range of dose tertiles from baseline to month 12 for both furosemide and torsemide (Table 2). Conclusions Following hospitalization for HF, no clinical differences were observed across the dosing spectrum in all-cause mortality, all-cause hospitalization, or KCCQ-CSS at 12-mornths between torsemide and furosemide. Irrespective of diuretic agent used, increasing dose of loop diuretics correlates with worse clinical outcomes.