| First author [ref.] | Iron therapy | Patient group | Placebo | Duration | Primary effects | Secondary effects |
| Bolger [16] | ≤1 g iron–sucrose | Anaemic heart failure (n=16) | No | ≤3 months | Hb improvements | Improved symptoms and 6MWD (average change >40 m), no adverse events |
| Toblli [17] | 5 weekly 200-mg doses iron–sucrose | Anaemic CHF (n=40) | 1:1 randomisation | 6 months | Reduced NT-proBNP and CRP | Improved symptoms, LVEF and exercise capacity; all 5 hospitalisations were in placebo group |
| Okonko [18] | 200 mg·week−1 until ferritin >500 μg·L−1, then fortnightly (total mean±sd 1433±365 mg) | CHF (n=35) | 2:1 randomisation (iron:placebo) | 12 weeks | Peak O2 consumption only raised in anaemia | Improved NYHA class, adverse events similar |
| Usmanov [19] | 100 mg three times a week for 3 weeks and then once weekly for 26 weeks (total dose 3200 mg) | CHF (NYHA class III/IV) and persistent anaemia (n=32) | No | 26 weeks | Significant improvements in LV function and structure | Half of class III patients improved to class II, no improvement from class IV |
| Anker [20] | Fortnightly 200 mg ferric carboxymaltose until Ganzoni iron deficit replenished | CHF (NYHA class II/III) with iron deficiency and Hb 95–135 g·L−1 (n=459) | 2:1 randomisation (iron:placebo) | 24 weeks | Significant improvements in patient self-reported global assessment and NYHA class | 6MWD increased ∼40 m versus placebo; quality of life improved Improvements irrespective of anaemia; adverse events similar |
CHF: chronic heart failure; NYHA: New York Heart Association; Hb: haemoglobin; NT-proBNP: N-terminal pro-brain natriuretic peptide; CRP: C-reactive protein; LV: left ventricle; 6MWD: 6-min walk distance; LVEF: LV ejection fraction.