Anemia and iron deficiency in heart failure

Department ofCardiology, University Medical CenterGroningen, University of Groningen, PO Box 30.001, 9700RB Groningen, The Netherlands(D. J. van Veldhuisen ). Applied CachexiaResearch, Department of Cardiology, Campus Virchow-Klinikum, CharitéUniversitätsmedizin, Augustenburger Platz 1, D-13353 Berlin,Germany (S. D. Anker ). Department of Heart Diseases, Medical University, Military Hospital, Weigla 5, 50-891 Wroclaw,Poland (P . Ponikowski ). Department of Renal Medicine, King’s College Hospital, London SE5 9RS, UK (I. C. Macdougall )Correspondence to: D. J. van Veldhuisen d.j.van.veldhuisen@thorax.umcg.nlAnemia and iron deficiency in heart failure: mechanisms and therapeutic approachesDirk J. van Veldhuisen, Stefan D. Anker, Piotr Ponikowski and Iain C. MacdougallAbstract | Anemia and iron deficiency are common in patients with heart failure (HF), and are associated with worse symptoms and adverse outcomes in this population. Although the two can occur together, anemia in HF is often not caused by iron deficiency, and iron deficiency can be present without causing anemia. Erythropoiesis-stimulating agents have been investigated extensively in the past few years and might be of benefit in patients with HF and anemia. However, concerns have arisen regarding the safety of erythropoiesis-stimulating agents in patients with chronic kidney disease and so the results of a large mortality trial areeagerly awaited to provide information on safety in patients with HF . Iron supplementation or replacement is a much older treatment option for patients with HF and anemia, but questions about the safety of intravenous iron, and absorption problems with oral formulations have prevented its widespread use to date. In the past few years, however, new data on the importance of iron deficiency in HF have become available, and a number of studies with intravenous iron have shown promising results. Therefore, this treatment approach is likely to become an attractive option for patients with HF and iron deficiency, both with and without anemia.van Veldhuisen, D. J. et al. Nat. Rev. Cardiol. 8, 485–493 (2011); published online 31 May 2011; doi:10.1038/nrcardio.2011.77IntroductionAnemia is common in patients with heart failure (HF) and its prevalence increases with the severity of the disease.1–5 The WHO has defined anemia as a hemo g lobin level of <12 g/dl (7.5 mmol/l) in women and <13 g/dl (<8.1 mmol/l) in men, although other criteria are also used. According to the WHO criteria, large registries or meta-analyses have demonstrated that anemia is present in 22–37% of all patients with HF ,2–4 although a much higher prevalence (even >50%) has also been reported.6Anemia has been associated with a marked increase in mortality in patients with HF; in a meta-analysis of more than 150,000 patients with HF, mortality was almost double in patients with anemia compared with patients without this condition.2 This increase in risk remained around 1.5 when adjusted for several potential con-founders, including age and renal dysfunction. Although the majority of patients with both HF and anemia die of cardiovascular causes, data suggest that the incidenceof cancer might also be increased in this population.7The presence of anemia in HF is associated with a reduced exercise capacity and an impaired quality of life,1,3,8 with an increased number of hospitalizations 9 and with elevated levels of natriuretic peptides (B-type natriuretic peptide [BNP] and the N-terminal fragment of pro-BNP).10 In most studies, the relationship between hemoglobin values and outcome in HF shows an inverse and linear relationship, although an increased risk with the highesthemoglobin values, or a U-shaped relationship, have been reported in a few studies.11,12Many factors in HF might contribute to the develop-ment of anemia;1,3,13 therefore, the search for the under-lying mechanism is difficult. Often, more than one etiology is involved. Important factors are renal dysfunction,14–16 hemodilution,15–17 the use of drugs generally used for the treatment of HF (such as angiotensin-converting-enzyme inhibitors,18 angiotensin-receptor blockers,19 and some, but not all, β-blockers 20,21), chronic inflammation,22 bone-m arrow dysfunction,23,24 abnormal steroid metabolism,25 resistance to erythropoietin,26 and hematinic deficiencies, such as vitamin B12, folate and, in particular, iron defi-ciency.5,27,28 Anemia is reported to be more common in women than in men,11 and seems to be similarly common in patients with HF with preserved ejection fraction as in those with reduced ejection fraction.21 Anemia in HF is often normocytic, possibly due to the presence of cytokines and HF-associated deficiencies in vitamin B12 and folate, and patients have a low reticulocyte count.8,9,22 Patients with HF and anemia have been reported to have a shorter leuko-cyte telomere length,29 which might also be associated with bone-marrow dysfunction in this population.Iron deficiency is also common in HF and, in the early 2000s, was reported to occur in up to 20% of patients with HF.1,8,27,28 These early studies focused on the presence of absolute iron deficiency. In a study of 955 patients with HF, in whom the prevalence of anemia was 32%, de Silva et al. reported that 43% of patients with anemia and 15% of those without anemia had at least one test suggest-ing iron and/or folate deficiency.30 In a study of patients with de c ompensated advanced HF published in 2006,Competing interestsAll authors declare associations with the following companies: Amgen and Vifor. I. C. Macdougall also declares associations with the following companies: Affymax, Ortho Biotech, andRoche. See the article online for full details of the relationships.REVIEWSNanas et al. performed bone-marrow aspiration and found that 73% of patients with severe HF and anemia showed evidence of iron deficiency.31 In that study, however, serum ferritin was not a reliable marker of iron deficiency. Therefore, the definition of iron deficiency is not clear cut. Nevertheless, the criteria most-commonly used for detecting this condition are a ferritin level <100 μg/l, or a ferritin level 100–300 μg/l in combination with trans-ferrin saturation <20%.32–34 Using these criteria, a large study published in 2010 showed that iron deficiency was present in 37% of all patients with HF.35 Interestingly, in this study, iron deficiency was detected in one-third of patients with normal hemoglobin levels.35 Iron deficiency can lead to anemia, but it can have clinical consequences even without anemia.36,37Anemia and iron deficiency, therefore, are both common in patients with HF , and partly overlap. Whereas anemia in the setting of HF has received a lot of attention in the past 10 years, and most studies have specifically focused on the efficacy and safety of erythropoietin in the treatment of this condition, less research has focused on iron deficiency in patients with HF . In the past few years, however, there has been an enormous increase in the number of studies on iron deficiency and its treatment in patients with HF. In this Review, we will first discuss the published literature on the treatment of anemia in patients with HF, in which the use of erythropoiesis-stimulating agents (ESAs) has an important place. Subsequently, we will discuss the mecha-nisms of iron metabolism both in general and in the setting of HF, the relevance of iron deficiency, and data on iron supplementation in patients with HF .Erythropoietin therapy for anemiaErythropoietin is produced by the kidney in response to hypoxia.26 Endogenous erythropoietin is disproportionally elevated in patients with HF, and the degree of elevation is associated with outcome.38,39 Similar to natriuretic pep-tides, erythropoietin can thus be considered a biomarker in this respect.Exogenous ESAs are used on a large scale to treat patients with chronic kidney disease and anemia, and patients with cancer-associated anemia, with the aim of increasing hemo-globin.40 In 2000, Silverberg and co-workers from Israel were the first to report the use of these therapeutic agents in patients with HF .41 In their first, retrospective report, they studied 142 patients with anemia and HF , of whom 26 weretreated with subcutaneous erythropoietin and intravenous iron for more than 6 months. The mean hemoglobin level increased significantly, as did the mean left ventricular ejection fraction. Furthermore, patients’ functional class improved, they needed lower doses of diuretics during the study period, and they were hospitalized less often than before. In a randomized, controlled, open study of 32 patients published the following year, the same authors confirmed the earlier favorable findings on the same clini-cal end points.42 In a small, randomized, controlled, single-blind study reported in 2003, Mancini et al. showed that the use of erythropoietin increased hemo g lobin and was associated with an increase in peak oxygen uptake (VO 2 max) and exercise duration in 26 patients with severe HF and anemia.43 Similar, relatively small-scale studies from Italy 44 and Greece,45 which were published in 2006 and 2008, respectively, confirmed the beneficial effect of ESAs in patients with HF, but none of these studies were double-blind.Around this time, Amgen (Thousand Oaks, CA, USA) started its clinical program to examine the efficacy and safety of their ESA, darbepoetin alfa, in HF. In the first study of 41 patients, published in 2007, darbepoetin alfa increased and maintained hemoglobin levels at a target of 13.0–15.0 g/dl, and a trend towards an improvement in exercise duration was observed (+42 s versus –67 s in the placebo group; P = 0.075 between groups).46 Two larger studies were also conducted,47,48 the first of which involved 165 patients and showed that both a weight-adjusted and a fixed dose of darbepoetin alfa maintained hemoglobin levels at 13.4 g/dl over a 6-month period.47 This study also showed that several quality-of-life para m eters increased with this ESA, and a borderline statistically significant increase in 6-min walking distance (+34 ± 7 m versus +11 ± 10 m in the placebo group, P = 0.074 between groups). In the other study, the Study of Anemia in Heart Failure Trial (STAMINA-HeFT),48 319 patients were treated with darbepoetin alfa or placebo for 12 months. This study showed that the drug did not significantly improve exer-cise duration; however, it was well-tolerated and a favor-able trend with regard to its effect on the composite end point of all-cause mortality or first hospitalization for HF was reported (hazard ratio [HR] 0.68, 95% CI 0.43–1.08, P = 0.10). This potentially beneficial effect on outcome was subsequently confirmed in a meta-analysis, in which 363 of the 650 included patients had been treated with erythro-poietin.49 ESA treatment was found to result in a lower risk of hospitalization for HF (risk ratio [RR] 0.59, 95% CI 0.41–0.86, P = 0.006), but had no significant effect on all-cause mortality (RR 0.69, 95% CI 0.39–1.23, P = 0.21). These favorable effects led to the design of the Reduction of Events with Darbepoetin alfa in Heart Failure (RED-HF) trial,50 which will include 2,600 patients with HF and anemia. RED-HF started in 2006 and currently has more than 2,000 patients enrolled.At the end of 2006, considerable concerns were raised by the results of two large trials—Correction in Hemoglobin and Outcomes in Renal Insufficiency (CHOIR)51 and Cardiovascular Risk Reduction in Early Anemia Trial with Epoetin Beta (CREATE)52—in which the effect ofESA in patients with chronic kidney disease was exam-ined. These studies showed that patients with anemia who were treated with an ESA might have an increased risk of cardio v ascular events when their treatment was targeted to higher hemoglobin levels (13.0–15.0 g/dl). Since a pro-portion of patients with chronic kidney disease also have HF, the relevance of these findings was considered for the RED-HF study.50 Given that the populations were very dif-ferent, that no placebo-control groups were included in the CHOIR51 and CREATE52 studies, and that the number of events was rather small, it was decided that there was no reason to extrapolate these data to the RED-HF50 study population, and that the trial could and should continue.53 Additionally, the ESAs used in these studies were differ-ent from darbepoetin alfa, and the latter might, therefore, have a different mechanism of action that could lead to a different safety profile.40,51 The much larger Trial to Reduce Cardiovascular Events With Aranesp® Treatment (TREAT),54 which involved more than 4,000 patients, has provided many more data regarding the safety of darbe-poetin alfa in patients with chronic kidney disease. After TREAT, some concerns about the use of this compound in patients with chronic kidney disease have remained, particularly with regard to an increased risk of stroke and venous thrombo e mbolism, but to a certain extent also for those with a history of malignancy.54 In TREAT, the use of darbepoetin alfa in patients with a history of HF (n = 1,347), however, seemed quite safe.55 A substudy of the CHOIR trial, which specifically focused on comorbidi-ties including HF, was published in 2010.56 Interestingly, this substudy showed that the increased risk associated with targeting the higher hemoglobin level in the CHOIR study51 was not observed in patients with HF at baseline (HR 0.99), but only in those without HF at baseline (HR 1.86, P = 0.004).56Of note, erythropoietin also has nonhematopoietic effects, such as enhancing angiogenesis and inhibiting apoptosis,26,57 the former of which, in particular, might be beneficial in HF58 and the latter of which might be parti-cularly attractive for acute ischemic conditions.64,65 Indeed, ESAs have also been explored in acute ischemic cardio-vascular conditions,59 such as myocardial infarction60,61 and stroke,62,63 where they were investigated primarily as a protective agent. In studies on the use of ESAs in the settings of myocardial infarction and stroke, which have yielded mixed results,60–63 ESAs were only administered during, or shortly after, the acute ischemic event, and hemoglobin level does not increase in this setting. Some have questioned whether increasing the hemoglobin level with an ESA is indeed needed in these circumstances, and have suggested that a low, nonhematopoietic dose might also be effective in patients with HF and remove any potential safety concerns associated with the higher doses of these agents.66 Modified, nonhematopoietic derivates of erythropoietin have been investigated in animal models, and were shown to be cardioprotective, both in models of chronic ischemia67 and in models of HF,68 which sup-ports this hypothesis. Nevertheless, a 2009 meta-analysis of three randomized, double-blind, placebo-controlled trials of darbe p oetin alfa treatment in anemic patientswith kidney disease,70 a setting in which ESA use has been associ a ted with an increased risk of thrombosis3,71 that could potentially lead to hypertension and throm-botic events, and an increased risk of stroke has been reported in patients with renal disease.54 The potential difference in efficacy and safety of these agents might be related to several factors, including differences in rheology and the response to increasing levels of hemoglobin between the two diseases. Furthermore, the well-known nonhemato p oietic effects of ESAs, and their potential benefits,59,67 might have different actions on the heart and kidneys.70 However, no direct head-to-head comparisons between the effect of ESA treatment in these two disease states have been performed, and so all the assumptions discussed above must be considered with caution. Hyporesponsiveness or resistance to ESAs, in terms of increasing hemoglobin level, is an issue in some patients.72 In 2006, Regidor et al.73 demonstrated that the requirement of higher ESA doses was associated with a higher risk of death.73 More recently, similar and more-controlled data from the TREAT trial showed that a poor initial response to darbepoetin alfa was associated with an adverse outcome in patients with chronic kidney disease and in those with type 2 diabetes mellitus.74 Interestingly, a similar associ a tion—that the inability to raise hemo-globin with darbepoetin alfa was associated with a worse prognosis—was also observed in an analysis of the darbe-poetin studies in patients with HF (Figure 1).75 Although several factors have been associated with hyporespon-siveness to ESAs, including inflammation,72 older age, obesity, drug use, and etiology of kidney disease,76 hep-cidin activity (which has an important role in iron defi-ciency in patients with HF77) has also been implicated in ESA hypo r esponsiveness.78 Hepcidin is discussed in more detail below. Another important consideration is that the use of ESAs can lead to iron depletion if there is insufficient supplemental iron during treatment.79 Iron depletion in response to ESA therapy has been associated with platelet activation,80 which could be a factor in the observed increase in risk of thrombosis associ a ted withthe use of ESAs. These associations, however, have not been observed in patients with HF.Overall, the data with ESAs in patients with HF look promising. However, data from large, controlled studies, particularly the results from the RED-HF trial,50 are eagerly awaited.Relevance of iron deficiencyIn addition to its key role in erythropoiesis, iron has an important role in oxygen uptake and transport (as a compo-nent of hemoglobin), oxygen storage (as part of myoglobin), and oxygen metabolism in skeletal and heart muscle (as part of the oxidative enzymes and respiratory chain pro-teins).32,81,82 Moreover, iron is involved in the synthesis and degradation of many other products, such as lipids, carbo-hydrates, DNA, and RNA.83,84 Therefore, the maintenance of normal iron metabolism is crucial for cells with a high mitogenic potential (such as tumor and hematopoietic83–8783,88 The cause of iron deficiency 88,89 Iron 85,90,91 Iron deficiency has 81,85,86,88–90 92 and is associated with impaired exercise 93,94In 2010, however, a large study that used 35 Iron deficiency was P <0.01; Figure 3).35 These data further depletion of iron stores, but it can also be ‘functional’ (or ‘relative’) as a result of impaired iron metabolism, which might be secondary to inflammatory processes.22,90,95 In healthy individuals, ferritin levels are used to define pres-ence of iron deficiency, and a cut-off of <30 μg/l is often used—indeed, the ‘normal’ range in most laboratories is 30–300 μg/l and the geometric mean is 88 μg/l in men, and 49 μg/l in women.96 However, the definition of iron deficiency in HF and other chronic illnesses is prob-lematic, because these conditions are associated with increased activation of both proinflammatory and anti-i nflammatory processes,97,98 including increased levels of the acute phase reactant ferritin. Inflammatory cytokines, such as inter l eukin 6 (IL-6), interfere with iron absorption and iron availability, and cause a state of ‘functional iron deficiency’.22 Functional iron deficiency in HF cannot be diagnosed using standard cut-off values for absolute iron deficiency that were developed for healthy individuals. Indeed, the gold standard to diagnose functional iron defi-ciency is based on assessment of bone-marrow biopsies, asconducted by Nanas and colleagues,31 and the incidence might be much higher when using this technique than when serum ferritin measurements are used. Intestinal absorption of iron is impaired in inflammatory conditions, including HF, but the precise mechanism has not been fully elucidated. Experimental data indicate that expression of duodenal iron transporters are impaired in HF, which might be a mechanism involved in the abnormal iron regulation seen in HF.99 Although the small peptide hormone hepcidin is generally recognized to be important in iron regulation in HF, the exact mechanism of action is still not fully elucidated.100,101 Transgenic or constitutive overexpression of hepcidin results in severe iron-deficient anemia in mice, and inflammation in hepcidin-deficient mice does not lead to iron-deficient anemia.102 Cytokines, in particular IL-6, increase hepcidin release from the liver, which inhibits ferroportin activity in the gut mucosal cells and macrophages. Indeed, the hepcidin–ferroportin axis is the principal regulator of extracellular iron homeo-stasis in health and disease.103 Hepcidin binds to ferro-portin, inactiva t ing this critical iron exporter channel. The hepcidin– f erroportin complex is then internalized and undergoes lysosomal degradation. This process pre-vents iron release from macrophages and Kupffer cells in the liver, and iron is, therefore, trapped in the reticulo-endothelial system, being unavailable for erythropoiesis in the bone marrow (the so-called reticulo-endothelial block).22 The secretion and expression of hepcidin has a crucial role in this process.104To summarize, iron absorption is decreased in HF and iron also accumulates in the reticulo-endothelial stores, which reduces its availability to target tissues and stores.22 In other words, functional iron deficiency can occur in the setting of HF despite adequate iron stores in the body. Use of iron supplementationGiven the prominent unfavorable effects of iron deficiency, attempting to correct this problem by the administration of iron seems logical. However, iron is not absorbed well when administered orally,105 particularly in patients with HF. In one study, oral iron did not increase hemoglobin levels or have a favorable effect on any of the clinical parameters measured.44 A number of studies conducted in patients with kidney disease have shown that intra-venous iron is superior to oral iron in the management of patients with anemia and low iron indices.106–108 However, the intra v enous administration of iron has raised concerns in the past, being associated with increased susceptibility to infection, malignancy, free radical formation, vascu-lar damage, and even cardiomyopathy.109–111 Therefore, as discussed below, several trials have been performed to examine the efficacy and safety of intravenous iron in patients with HF.Intravenous iron has been used to maintain adequate iron replacement as part of a treatment regimen with ESAs in two studies.41,112 However, in these studies, the effect of the combination of ESAs and intravenous iron was exam-ined, and not the effect of intravenous iron alone. The effi-cacy and safety of intravenous iron alone in patients with HF has been examined in five studies to date, but only three of these investigations were randomized, controlled studies (Table 1).In 2006, Bolger et al. were the first to report on the effect of intravenous iron sucrose alone in 16 patients with anemia (hemoglobin ≤12 g/dl) with ferritin levels <400 μg/l and mild-to-moderate HF.113 In an open-label, uncontrolled study, they showed that after intravenous administration of 1,000 mg iron over 2 weeks, hemoglobin increased signifi-cantly after 3 months (from 11.2 ± 0.7 g/dl to 12.6 ± 1.2 g/dl, P = 0.0007), and functional class, quality of life, and walking distance also improved.In another open-label study conducted by Usmanov and colleagues, 32 patients with anemia and iron deficiency were treated with 100 mg iron sucrose three times weekly for 3 weeks, and then once weekly for 23 weeks, for a total dose of 3,200 mg for 26 weeks.114The investigators foundREVIEWSthat treatment with intravenous iron had a favorable effect on left ventricular wall thickness, volumes, and ejection fraction, and that functional class also improved.The first randomized, controlled study of intravenous iron in the setting of HF was conducted by Toblli et al. and the results were published in 2007.115 In total, 40 patients were randomly assigned to 200 mg intravenous iron sucrose or a saline control once weekly, and followed up for 6 months. All patients in this study had anemia (hemo g lobin ≤12.5 g/dl for men and ≤11.5 g/dl for women), iron defi-ciency (ferritin <100 μg/l, or transferrin saturation <20%), and mild renal dysfunction. At the end of the study, patients who had received intravenous iron sucrose (n = 20) had increased hemoglobin levels (from 10.3 g/dl to 11.8 g/dl), better renal function, and lower NT-proBNP levels, as well as improved left ventricular function, quality of life, and 6-min walking distances (all changes P <0.05 versus control group). Interestingly, five patients were hospitalized for HF in the control group, and none in the intravenous iron group.The second randomized, controlled study was the European Effect of Intravenous Ferrous Sucrose on Exercise Capacity in Chronic Heart Failure (FERRIC-HF) trial.116 Okonko et al. randomly assigned 18 patients with anemia and 17 patients without anemia (cut-off value 12.5 g/dl), all of whom had HF and iron deficiency (ferritin <100 μg/l,®33 In this study, 459 patients f errin satura-n = 304) or saline (n = 155).33 116 Treatment P <0.001 versus controls), but no 33 Intravenous ferric carboxymalt-P <0.001), which were the co-primary end points of the study. Interestingly, this beneficial effect was similar in patients both with andin Patients With IRon Deficiency and Chronic Heart Failure (FAIR-HF). Data from Anker, S. D. et al.330.5 1.0 2.0 4.08.0without anemia (Figure 4), and the same clinical effect was observed in all subgroups. Improvements were also seen for 6-min walking distance and quality-of-life parameters. Intravenous iron was generally well-tolerated, and there were no obvious differences in hard clinical end points, such as all-cause death (3.4% versus 5.5%, for intravenous iron versus control, respectively) and first hospitalization (17.7% versus 24.8%, respectively), between the groups (Figure 5). The number of adverse events was also similar in both groups. ConclusionsAnemia and iron deficiency are both common and associ-ated with poor outcomes, including reduced exercise capac-ity, impaired quality of life, and increased hospitaliza t ion, in patients with HF. Identification and subsequent correc-tion of these conditions are, therefore, very important in this patient group.Anemia can be treated with ESAs, together with ironsupplementation, whereas iron deficiency can be effectively treated with iron supplementation alone. Most research on ESAs has been conducted in patients with kidney disease, but in the past 10 years, studies have also been conducted in patients with HF. Although the available data in patients with HF look promising,49 and a large survival trial (RED-HF)50 is ongoing, this field is mainly influenced by the findings from large trials in patients with chronic kidney disease. These trials, and earlier studies, have shown that uptitration of ESA therapy to achieve higher hemoglobin levels might not be appropriate, as this strategy poten-tially exposes patients to an increased risk of stroke andvenous thromboembolism.1. Anand, I. S. Anemia and chronic heart failureimplications and treatment options. J. Am. Coll.Cardiol. 52, 501–511 (2008).2. Groenveld, H. et al. Anemia and mortality inheart failure patients: a systematic review andmeta-analysis. J. Am. Coll. Cardiol. 52, 818–827(2008).3. Tang, Y. D. & Katz, S. D. Anemia in chronic heartfailure: prevalence, etiology, clinical correlates,and treatment options. Circulation113,2454–2461 (2006).4. Go, A. S. et al. Hemoglobin level, chronic kidneydisease, and the risk of death andhospitalization in adults with chronic heartfailure: the Anemia in Chronic Heart FailureOutcomes and Resource Utilization (ANCHOR)Study. Circulation113, 2713–2723 (2006).5. Ezekowitz, J. A., McAlister, F. A. & Armstrong, P. W.Anemia is common in heart failure and isassociated with poor outcomes: insights froma cohort of 12,065 patients with new-onset heartfailure. Circulation 107, 986–994 (2003).6. Vrtovec, B. et al. Significance of anemia inpatients with advanced heart failure receivinglong-term mechanical circulatory support.Eur. J. Heart Fail.11, 1000–1004 (2009).7. Schroten, N. F. et al. High cumulative incidence ofcancer in patients with cardio-renal-anaemiasyndrome. Eur. J. Heart Fail.12, 855–860 (2010).8. Kalra, P. R. et al. Effect of anemia on exercisetolerance in chronic heart failure in men.Am. J. Cardiol.91, 888–891 (2003).10. O’Meara, E. et al. Clinical correlates andconsequences of anemia in a broad spectrumof patients with heart failure: results of theCandesartan in Heart Failure Assessment ofReduction in Mortality and Morbidity (CHARM)Program. Circulation21, 986–994 (2006).10. Hogenhuis, J. et al. Anaemia and renaldysfunction are independently associated withBNP and NT-proBNP levels in patients with heartfailure. Eur. J. Heart Fail. 9, 787–794 (2007).11. Sharma, R. et al. Haemoglobin predicts survivalin patients with chronic heart failure: a substudyof the ELITE Trial. Eur. Heart J.25, 1021–1028(2004).12. Mozaffarian, D., Nye, R. & levy, W. C. Anemiapredicts mortality in severe heart failure. TheProspective Randomized Amlodipine survivalEvaluation. J. Am. Coll. Cardiol. 41, 1933–1939(2003).13. Okonko, D. O. & Anker, S. D. Anemia in chronicheart failure: pathogenetic mechanisms. J. Card.Fail.10, S5–S9 (2004).14. McClellan, W. M., Flanders, W. D.,Langston, R. D., Jurkovitz, C. & Presley, R.Anemia and renal insufficiency are independentrisk factors death among patients withcongestive heart failure admitted to communityhospitals: a population-based study. J. Am. Soc.Nephrol.13, 1928–1936 (2002).15. Anand, I. S. et al. Edema of cardiac origin.Studies of body water and sodium, renalfunction, hemodynamic indexes, and plasmahormones in untreated congestive heart failure.Circulation80, 299–305 (1989).16. Westenbrink, B. D. et al. Anaemia in chronicheart failure is not only related to impaired renalperfusion and blunted erythropoietin production,but to fluid retention as well. Eur. Heart J.28,166–171 (2007).17. Androne, A. S. et al. Hemodilution is common inpatients with advanced heart failure. Circulation107, 226–229 (2003).18. Van der Meer, P. et al. Levels of hematopoiesisinhibitor N-acetyl-seryl-aspartyl-lysyl-prolinepartially explain the occurrence of anemia inheart failure. Circulation112, 1743–1747(2005).19. Anand, I. S. et al. Anemia and change inhaemoglobin over time related to mortality andmorbidity in patients with chronic heart failure:results from Val-HeFT Circulation112,1121–1127 (2005).20. Komajda, M. et al. The impact of new onsetanemia on morbidity and mortality in chronicheart failure: results from COMET. Eur. Heart J.27, 1440–1446 (2006).21. Von Haehling, S. et al. Anaemia among patientswith heart failure and preserved or reducedejection fraction: results from the SENIORSStudy. Eur. J. Heart Fail.13, 656–663 (2011).22. Weiss, G. & Goodnough, L. T. Anemia of chronicdisease. N. Engl. J. Med. 352, 1011–1023 (2005).23. Westenbrink, B. D. et al. Bone marrowdysfunction in chronic heart failure patients.Eur. J. Heart Fail.12, 676–684 (2010).24. Kissel, C. K. et al. Selective functionalexhaustion of haematopoietic progenitor cells inthe bone marrow of patients with postinfarctionheart failure. J. Am. Coll. Cardiol. 49, 2341–2349(2007).The situation for iron replacement is different. In thepast, oral administration of iron was the main treatmentfor patients with anemia, often (later) combined with ESAs.However, iron is not well-absorbed when given orally, andwith the early successful use of ESAs, these drugs wereoften quickly added to or replaced iron supplement a tion.The recent safety concerns about ESAs, the emerging dataon iron deficiency, the increased understanding of its clini-cal importance in the presence and absence of anemia,together with the availability of new intravenous ironformulations have begun to change the picture. Over thepast 5 years, exciting new data have become available inthe field, in particular the results from FAIR-HF,33 whichfurther support a prominent role for iron deficiency inHF. Nevertheless, larger adequately powered trials withmortality end points will be needed to establish the long-term efficacy and safety of intravenous iron in patientswith HF.REVIEWS。