Abstract
INTRODUCTION: For heart failure (HF) with iron deficiency (ID), the benefits of intravenous iron might differ according to kidney function. METHODS: IRONMAN was a randomised, open-label trial of intravenous ferric derisomaltose (FDI) versus usual care (UC) in patients with HF, left ventricular ejection fraction ≤45%, and ID (transferrin saturation <20% and/or ferritin <100 µg/L). The primary composite endpoint of recurrent hospitalisation for HF and cardiovascular (CV) death was lower in those assigned to FDI. Analysis according to baseline estimated glomerular filtration rate (eGFR) is now reported, with outcomes assessed in 3 categories of eGFR. RESULTS: Of 1,137 patients randomised, eGFR was <45 mL/min/1.73 m2 for 435 (38%), 45-59 mL/min/1.73 m2 for 295 (26%), and ≥60 mL/min/1.73 m2 for 407 (36%). Patients with eGFR <45 mL/min/1.73 m2 were older and had more severe HF and more events. For the primary outcome, the primary endpoint rates per 100 patient-years for FDI versus UC across eGFR categories were 164 and 213 (rate ratio [RR]: 0.77 [95% CI: 0.57, 1.03]), 84 and 105 (RR: 0.79 [95% CI: 0.51, 1.22]), 88 and 93 (RR: 0.98 [95% CI: 0.62, 1.54]), respectively, but no statistically significant interaction between eGFR category and treatment effect was observed (p(interaction) = 0.67). When eGFR was <45 mL/min/1.73 m2, FDI was associated with more favourable effects on Minnesota Living with Heart Failure score at 4 months (p < 0.001; p(interaction) = 0.01 by eGFR class) and trends to greater reductions in first hospitalisation for HF or CV death (hazard ratio [HR]: 0.76 [95% CI: 0.58, 0.99]; p(interaction) = 0.53) and first hospitalisation for myocardial infarction, stroke or HF, or CV death (HR: 0.71 [95% CI: 0.55, 0.92]; p(interaction) = 0.29), although tests for interaction by eGFR class were not significant. CONCLUSION: For patients with HF and ID, those with eGFR <45 mL/min/1.73 m2 are more symptomatic, have worse outcomes, and might receive greater benefit from FDI. Analysis of other randomised trials, ideally an individual patient data meta-analysis, are required to confirm these findings.