INTRODUCTION — Anemia is a frequent finding in adults with heart failure (HF), with prevalence varying depending upon the population studied [1]. As an example, in an analysis from the SOLVD trial, 22 percent of patients had a hematocrit ≤39 percent and 4 percent had values below 35 percent [2]. A similar rate of anemia (17 percent) was noted in a population-based cohort of 12,065 patients with newly diagnosed HF [3]. The incidence of anemia appears to increase with worsening functional class (from 9 percent for New York Heart Association class I to 79 percent for class IV in one report) (table 1) [4].

Evaluation, prognosis, and treatment of anemia in patients with HF will be reviewed here [5]. Diagnosis and general approaches to anemia in adults and older adults are presented separately. (See "Diagnostic approach to anemia in adults".)

EVALUATION

Clinical presentation

General considerations — Symptoms related to anemia can result from decreased oxygen delivery to tissues, and, in patients with acute, severe bleeding, the added insult of hypovolemia. Symptoms of anemia such as dyspnea and fatigue may be difficult to distinguish from symptoms of HF. (See "Diagnostic approach to anemia in adults".)

Symptoms from reduced oxygen delivery due to anemia generally occur only with severe anemia, but may occur at less severely reduced hemoglobin levels in patients with HF. Since the extraction of oxygen by tissues can increase from 25 to approximately 60 percent with anemia or hypoperfusion, in individuals with normal hemodynamics, normal oxygen delivery is preserved by extraction alone down to a hemoglobin concentration of approximately 8 g/dL. The hemodynamic effects of chronic anemia were evaluated in a right heart catheterization study [6]. Normal cardiac hemodynamics were maintained in patients with hemoglobin values as low as 7 g/dL; the cardiac output increased at lower hemoglobin concentrations.

In healthy individuals, acute, isovolemic reduction in the hemoglobin concentration induces a variety of compensatory changes including increases in heart rate, stroke volume, and cardiac index along with enhanced tissue oxygen extraction [7]. The net effect is that oxygen delivery can be maintained at rest at a hemoglobin concentration as low as 5 g/dL (equivalent to a hematocrit of 15 percent) if intravascular volume is maintained.

In patients with HF, oxygen delivery may be impaired due to reduced cardiac output, and thus symptoms may arise at higher hemoglobin levels in patients with anemia and HF.

Development of high-output heart failure — Severe anemia is a rare cause of high-output HF. High-output HF in the setting of anemia is more likely to occur in the setting of impaired cardiac reserve associated with an underlying cardiac abnormality such as valvular heart disease or left ventricular dysfunction. Only severe degrees of anemia (ie, hemoglobin <5 g/dL) can produce HF in the absence of underlying heart disease. (See "Causes and pathophysiology of high-output heart failure".)

Evaluation of cause of anemia — Anemia is one of the major signs of disease, and its cause should always be sought. Evaluation of anemia in patients with HF should include consideration of etiologies related to HF as well as other causes. (See "Diagnostic approach to anemia in adults".)

Suggested initial testing includes:

●Complete blood count, including red cell indices, reticulocyte count, and evaluation of the peripheral blood smear

●Iron studies (serum iron, transferrin, iron saturation, ferritin)

●Renal function (eg, serum creatinine, creatinine clearance)

●C-reactive protein and erythrocyte sedimentation rate (may be useful if ferritin is borderline to assess for inflammation as a factor)

●Serum levels of vitamin B12 and folate

A serum ferritin <41 ng/mL or a transferrin saturation (TSAT) <20 percent is strongly suggestive of iron deficiency in patients with comorbidities such as HF. For individuals with values above these thresholds for whom there is a strong suspicion of iron deficiency, additional testing such as soluble transferrin receptor may be appropriate. (See "Causes and diagnosis of iron deficiency and iron deficiency anemia in adults", section on 'Diagnostic evaluation'.)

If preliminary testing does not reveal a specific diagnosis, it may be appropriate to refer the patient to a hematologist for additional evaluations such as bone marrow examination for possible myelodysplastic syndrome. (See "Diagnostic approach to anemia in adults".)

Potential causes of anemia related to heart failure — Factors that may contribute to the development of anemia in patients with HF are discussed here. Most of these factors are suggested by their association with HF, although causal relationships remain largely unproven.

Increased circulating cytokines and the anemia of inflammation — Inflammation may contribute to the anemia seen in chronic HF. This view is supported by the finding of increased levels of circulating cytokines, such as tumor necrosis factor-alpha and interleukin-6 in patients with HF, and increased levels of the acute phase reactants C-reactive protein, erythrocyte sedimentation rate, and serum ferritin [8-14]. Such changes are consistent with those found in patients with the anemia of chronic disease/anemia of inflammation (ACD/AI) as well as in some older adults with unexplained anemia. (See "Anemia of chronic disease/anemia of inflammation", section on 'Cytokine effects'.)

While the regulatory peptide hepcidin plays a central role in the anemia of inflammation, its role in the anemia seen in patients with HF is uncertain [14-16]. (See "Anemia of chronic disease/anemia of inflammation", section on 'Hepcidin (primary regulator of iron homeostasis)'.)

Dilutional anemia — In a report of 37 patients with HF and anemia in whom the plasma volume was directly determined using iodine-131 labeled albumin, 17 had anemia on the basis of hemodilution (ie, an expanded plasma volume) [17]. The patients with hemodilution appeared to have a worse prognosis.

In another study in 99 patients with HF, those with anemia had significantly increased plasma volume and no significant reduction in red cell volume, as measured by a chromium-51 assay [18].

Iron deficiency — Iron deficiency may be more common in HF patients than is suggested by iron studies alone since the ferritin level may not correlate with low iron stores. This has been illustrated in series of patients with HF who have been tested simultaneously with bone marrow iron staining and serum iron studies. As examples:

●In a 2018 series of 42 individuals with HF (mean left ventricular ejection fraction [LVEF] 38 percent, mean age 68 years, 76 percent male), 17 (40 percent) had absent bone marrow iron [19]. Among those with iron deficiency, serum ferritin levels ranged from 44 to 162 ng/mL. Serum iron ≤13 plus TSAT <20 had the best diagnostic performance.

●In a 2006 series of 37 patients with advanced HF (mean LVEF 22 percent; mean New York Heart Association class 3.7) and anemia, there was no stainable bone marrow iron (ie, iron deficiency) in 27 (73 percent) [11]. Serum ferritin levels, however, were reduced in only two of the iron deficient patients, indicating that elevated serum ferritin levels do not reliably exclude iron deficiency in this population.

These studies reinforce the importance of evaluating for iron deficiency in individuals with HF and anemia. (See 'Evaluation of cause of anemia' above.)

Whether these results reflect the prevalence of iron deficiency in populations with less severe HF is not known. In addition, the benefit of iron replacement was not assessed in these studies. Additional details of testing and treatment are discussed separately. (See "Causes and diagnosis of iron deficiency and iron deficiency anemia in adults" and 'Iron' below.)

Use of angiotensin converting enzyme inhibitors — Angiotensin converting enzyme (ACE) inhibitors, which prolong survival in patients with HF, also appear to induce anemia in selected patients. These drugs also reduce the hematocrit following renal transplantation and have been used to treat erythrocytosis in these patients. (See "Kidney transplantation in adults: Posttransplant erythrocytosis".)

The impact of ACE inhibitors was evaluated in a report from the SOLVD trial in which patients with left ventricular dysfunction were randomly assigned to enalapril or placebo [20]. At one year after randomization, the rate of new anemia (hematocrit ≤39 percent in men and ≤36 percent in women) was significantly higher in the enalapril group (11.3 versus 7.9 percent with placebo, adjusted odds ratio 1.56). The difference in hematocrit between the two groups was evident as early as six weeks.

The effect of ACE inhibitors on hematocrit may be mediated by Ac-SDKP (goralatide), a tetrapeptide that inhibits erythropoiesis. Ac-SDKP is metabolized by ACE and would be expected to accumulate in the presence of an ACE inhibitor, thereby inhibiting erythropoiesis.

Reduced kidney function and other contributing factors — When measured, mean serum creatinine levels have been elevated in a number of studies of anemic patients with HF, including those with cardiorenal syndrome [11,21-25]. (See "Cardiorenal syndrome: Definition, prevalence, diagnosis, and pathophysiology" and "Cardiorenal syndrome: Prognosis and treatment".)

Other contributing factors may include poor nutrition due to right-sided HF [1,10].

Given the increased age of many of these patients, the possibility of anemia of chronic disease/anemia of inflammation (ACD/AI), unexplained anemia in the elderly, and/or myelodysplastic syndrome needs to be considered as a cause of the anemia. (See "Diagnostic approach to anemia in adults", section on 'Older adults'.)

TREATMENT — When a specific cause of anemia can be identified (eg, iron, vitamin B12, or folate deficiency), appropriate treatment should be instituted to replace the deficient iron or vitamin. Folate deficiency is increasingly rare in individuals who consume a normal diet, due to routine supplementation of flours and grains in many countries. Data on use of transfusions in patients with HF are limited. Erythropoietic agent trials in patients with HF have found no benefit and an increased risk of thromboembolism.

Transfusion — Based upon low quality evidence, we suggest using a restrictive red blood cell transfusion strategy (eg, trigger hemoglobin threshold of 7 to 8 g/dL) in asymptomatic patients with HF, rather than a more liberal threshold (such as <10 g/dL). In general, transfusion should be considered when the hemoglobin is 7 to 8 g/dL, although some patients may require transfusion at higher levels. Symptomatic anemia should be treated with transfusion in patients with hemoglobin <10 g/dL, provided that the symptoms are severe enough and are clearly related to the anemia rather than the underlying condition. (See "Indications and hemoglobin thresholds for red blood cell transfusion in the adult", section on 'Overview of our approach'.) (Related Pathway(s): Anemia: Indications for red blood cell transfusion in hospitalized adults.)

No large randomized, controlled trials have been published on the effect of a liberal (ie, transfusion for hemoglobins of 10 or less) or restrictive (ie, transfusion for hemoglobins of 8 or less) transfusion regimen in patients with either stable or decompensated HF. A meta-analysis of the available studies in patients with anemia and heart disease has concluded the following [26]:

●Three randomized trials included a small number of patients with stable HF. However, data specific to patients with isolated HF and no coronary heart disease were available in only one trial of patients with HF and hip fracture surgery [27]. This study found no difference in mortality or myocardial infarction outcomes on the use of a liberal or restrictive transfusion regimen.

●Two observational studies in patients with decompensated HF found conflicting results: One showed harm and the other no significant difference [28,29].

●Low-quality evidence from six trials and 26 observational studies suggests that liberal transfusion protocols in patients with anemia and heart disease (coronary heart disease and/or HF) do not improve short-term mortality rates compared with less aggressive transfusion protocols (risk ratio 0.94, 95% CI 0.61-1.42).

When red cell transfusion is required in a patient with HF, careful attention to volume status is recommended, including adjustment of transfusion rate and supplemental diuretics as needed to avoid volume overload. (See "Transfusion-associated circulatory overload (TACO)".)

Our approach is consistent with a systematic review of the literature and practice guideline published by the American College of Physicians (ACP) in 2013 [26,30]. The guideline suggests using a restrictive red blood cell transfusion strategy (trigger hemoglobin threshold of 7 to 8 g/dL), rather than a more liberal threshold, in hospitalized patients with coronary heart disease. We agree with using a restrictive threshold in most patients with heart disease who lack symptomatic anemia, with individualized transfusion decisions based upon clinical judgment, including assessment of whether the patient appears to have symptoms from anemia, rather than the underlying cardiac condition. Links to this and other guidelines are listed below. (See 'Society guideline links' below.)

Iron — As a general rule, iron supplementation is recommended to treat documented iron deficiency or iron deficiency anemia. (See 'Iron deficiency' above.)

Indications for iron and evidence of efficacy — Iron should be administered to individuals with HF who have iron deficiency anemia or iron deficiency without anemia, similar to the general population. As discussed in more detail separately, the cause of iron deficiency should also be evaluated, especially in an older individual, for whom gastrointestinal bleeding or a gastrointestinal condition that interferes with iron absorption (eg, gastrointestinal malignancy or Helicobacter pylori infection, respectively) may be present. (See "Treatment of iron deficiency anemia in adults".)

Evidence to support the administration of iron in individuals with HF and iron deficiency or iron deficiency anemia comes from randomized trials in this population as well as studies in similar populations.

An international randomized trial (Affirm-AHF) found a benefit of intravenous iron in reducing HF hospitalizations in individuals with iron deficiency [31]. The trial randomly assigned 1132 individuals who were hospitalized with acute HF and concomitant iron deficiency (ferritin <100 ng/mL or transferrin saturation [TSAT] <20 percent) to receive intravenous ferric carboxymaltose (FCM) or placebo. Hospitalizations over the course of the 52-week trial were lower in the FCM group (relative risk 0.74, 95% CI 0.58-0.94); cardiovascular mortality was similar between groups. Those assigned to FCM had a median increase in hemoglobin of 0.8 g/dL compared with 0.3 g/dL in the placebo group. Therapy was well-tolerated, with serious adverse events, mostly cardiac, occurring in 45 percent of those assigned to FCM and 51 percent of those assigned to placebo.

Prior to Affirm-AHF, a 2019 meta-analysis of 10 randomized trials comparing iron supplementation versus placebo in 1404 individuals with HF who had iron deficiency or iron deficiency anemia found similar overall mortality, but iron therapy improved several cardiovascular outcomes [32]:

●Reduced hospitalizations for worsening HF (5.3 versus 14.5 percent; odds ratio 3.9, 95% CI 0.19-0.80)

●Improved New York Heart Association (NYHA) class (weighted mean difference [WMD] -0.68, 95% CI -1.13 to -0.24)

●Improved six-minute walk test (WMD 32.7, 95% CI 4.47-60.83)

●Improved left ventricular ejection fraction (LVEF; WMD 3.81, 95% CI 1.11-6.51)

●Improved serum markers of HF and inflammation (N-terminal pro-B-type natriuretic peptide and C-reactive protein, respectively)

Eight of the 10 trials used an intravenous iron formulation; the two trials that evaluated oral iron failed to show a statistically significant benefit, but the trials were small and the duration of follow up was short [33]. (See 'Route of administration (oral versus intravenous)' below.)

A smaller meta-analysis limited to four trials that compared a single formulation of intravenous iron with placebo in 839 individuals with HF and iron deficiency reported similar findings, as would be expected [34].

Some of the trials of iron in HF have used criteria for iron deficiency that were above limits generally considered diagnostic of iron deficiency. However, a slightly higher threshold may be appropriate in this setting due to the chronic inflammatory state in HF (see 'Iron deficiency' above). If more stringent criteria were used, efficacy measures might be more improved, but some individuals would be denied a useful therapy. Criteria for iron deficiency are described above, with further discussion presented separately. (See 'Evaluation of cause of anemia' above and "Causes and diagnosis of iron deficiency and iron deficiency anemia in adults", section on 'Diagnostic evaluation'.)

In addition to individuals with iron deficiency, iron may also be used in individuals with chronic kidney disease and HF who are being treated with an erythropoiesis-stimulating agent and are considered to have functional iron deficiency. (See "Treatment of iron deficiency in nondialysis chronic kidney disease (CKD) patients" and "Treatment of iron deficiency in dialysis patients".)

Iron should be given to replace deficient iron stores and then stopped, as excess iron deposition is cardiotoxic. This subject is discussed separately. (See "Clinical manifestations and diagnosis of hereditary hemochromatosis", section on 'Cardiac iron overload' and "Diagnosis of thalassemia (adults and children)", section on 'Heart failure and arrhythmias'.)

Route of administration (oral versus intravenous) — The best route of iron administration in patients with HF is not known. Both oral and intravenous routes are effective in non-HF patients with normal gastrointestinal absorption.

For patients with HF, some experts prefer intravenous iron replacement given limited available evidence on the efficacy of oral iron supplementation and concerns about possible reduced oral iron absorption in HF. There are no randomized trials comparing these routes of administration in individuals with HF, and small trials in other populations have shown mixed results. (See "Treatment of iron deficiency anemia in adults", section on 'Iron replacement products'.)

The advantages and disadvantages of oral and intravenous preparations differ, as summarized in the table (table 2). The oral route has very low initial costs and very low risk of a serious adverse event; the intravenous route is associated with more rapid correction of deficiency and reduced gastrointestinal side effects, with greater initial costs but possibly lower total costs, depending on the severity of the deficiency and patient adherence to oral therapy.

Evidence for efficacy of specific formulations includes the following:

●Oral – A randomized trial comparing oral iron versus placebo involved 225 individuals with HF with reduced ejection fraction (HFrEF; EF ≤40 percent) and iron deficiency (ferritin <100 ng/mL or ferritin 100 to 299 ng/mL plus TSAT <20 percent) [33]. The mean baseline hemoglobin with 12.6 g/dl. At 16 weeks of observation, there was a slightly greater increase in ferritin level in patients receiving oral iron (18 versus 1 ng/mL). The change in hemoglobin levels was not reported. There was a trend towards an increase in the six-minute walk test that did not reach statistical significance (-13 meters; 95% CI, -32 to 6 meters). Limitations included the relatively high baseline hemoglobin level, the small sample size, the short duration of follow up, and the high dropout rate (22 individuals [10 percent]).

Evidence for the efficacy of oral iron in other populations and information about the available oral iron formulations are presented separately. (See "Treatment of iron deficiency anemia in adults", section on 'Oral iron'.)

●Intravenous – Several randomized trials have compared intravenous iron with placebo in individuals with HF and iron deficiency, as summarized above. (See 'Indications for iron and evidence of efficacy' above.)

Available intravenous iron formulations, dosing, and advantages and disadvantages of different products are discussed separately. (See "Treatment of iron deficiency anemia in adults", section on 'Intravenous iron'.)

A 2017 focused guideline update from the American College of Cardiology and American Heart Association (ACC/AHA) made a weak recommendation for intravenous iron replacement for patients with NYHA class II and III HF and iron deficiency [35]. A 2016 guideline from the European Society of Cardiology made a similar recommendation [36]. Neither of these guidelines includes a recommendation for oral iron. Links to these and other guidelines are presented separately. (See 'Society guideline links' below.)

ESAs (not recommended) — We recommend against the use of erythropoiesis-stimulating agents (ESAs) in patients with mild to moderate anemia and HF. The available evidence does not support the use of erythropoiesis-stimulating agents to treat mild to moderate anemia in patients with HF and suggests an increased risk of venous thromboembolism [26]. The best evidence on the lack of efficacy and risk of complications of such treatment in this population comes from the Reduction of Events by Darbepoetin Alfa in Heart Failure (RED-HF) trial, which randomly assigned 2278 patients with HFrEF to treatment with either darbepoetin alfa (to achieve a hemoglobin target of 13 g/dL) or placebo [37]. Darbepoetin alfa- and placebo-treated groups had similar rates (50.7 and 49.6 percent) of the primary outcome of death from any cause or hospitalization for worsening HF at median 28-month follow-up. Rates of stroke were not significantly different in the two groups (3.7 and 2.7 percent), but thromboembolic adverse events were more frequent in the darbepoetin alfa-treated group (13.5 versus 10 percent).

The finding of a significantly higher thromboembolic event rate along with a nonsignificantly higher stroke rate in the darbepoetin alfa group in the RED-HF trial is similar to the results of the TREAT trial, which compared darbepoetin alfa with placebo therapy in 4038 patients with diabetes, chronic kidney disease, and anemia (one-third with HF) [38,39]. In the TREAT trial, the darbepoetin alfa-treated group had a significantly higher thromboembolic rate as well as a significantly higher stroke rate compared with the placebo group. Possible mechanisms for an adverse effect of erythropoietin therapy include worsening hypertension, increased risk of thrombotic events, and increased release of endothelin [1]. (See "Treatment of anemia in dialysis patients".)

A meta-analysis of 17 trials of ESAs in patients with heart disease (HF or coronary heart disease), including the RED-HF trial, found that ESAs provided no consistent clinical benefit but were associated with an increased risk of thromboembolism [26]. Analysis limited to trials in patients with HF produced similar findings.

Our approach is consistent with the 2013 ACP guideline, which also recommends against the use of ESAs in patients with mild to moderate anemia and HF or coronary heart disease (strong recommendation based on moderate quality evidence; equivalent to Grade 1B) [30].

PROGNOSIS — In patients with HF, anemia is associated with increased mortality, although it remains uncertain whether anemia is an independent predictor of outcomes or reflects more advanced disease and more extensive comorbidities.

A number of retrospective studies have evaluated the relationship between anemia and mortality in patients with HF [1-3,8,20,40-51]. Three studies found a J- or U-shaped relationship between mortality and hemoglobin, with increased mortality associated with hemoglobin levels <13 to 14 or >15 to 17 g/dL [50-52]. Some studies have suggested that anemia independently predicts worse outcomes [1-3,8]. However, the largest observational study that controlled for the greatest number of possible confounding variables found that anemia was not an independent predictor of outcome [49].

The independent prognostic significance of anemia in HF patients was suggested in a review of 1061 patients with New York Heart Association (NYHA) class III or IV HF (table 1) and a left ventricular ejection fraction <40 percent [8]. The following findings were noted:

●Lower hemoglobin concentrations were associated with worse hemodynamics, higher blood urea nitrogen and serum creatinine concentrations, and a lower serum albumin concentration.

●Patients with lower hemoglobin concentrations (<13.6 g/dL) had a significantly greater frequency of NYHA class IV HF and lower peak oxygen consumption.

●On multivariate analysis, a low hemoglobin concentration was an independent predictor of mortality (relative risk 1.13 for each 1 g/dL fall in hemoglobin concentration).

Similarly, in an analysis from the SOLVD trial, a low hemoglobin concentration was an independent predictor of mortality [2]. At a mean follow-up of 33 months, each percentage point reduction in hematocrit was associated with a 3 percent increase in mortality. These effects were seen in patients treated with enalapril or placebo. As noted above, the incidence of anemia was higher in the enalapril-treated group [20]. (See 'Evaluation' above.)

The time course of anemia appears to affect its prognostic significance. In a review of 6159 patients with chronic HF, anemia was present at baseline in 17.2 percent [48]. At six-month follow-up, persistent anemia was associated with higher mortality risk than no anemia (58 versus 31 percent), while transient anemia, which was present in almost one-half of anemic patients, conferred no excess mortality risk.

These analyses, however, do not establish a causal role for anemia in worse outcomes in HF. Among the possible confounding variables is the possibility that advanced HF exacerbates anemia. If this were the case, anemia would be a marker for advanced disease. Two studies evaluated patients with new onset HF, in whom the anemia would be less likely to be due to the HF. Results from these observational studies were conflicting as to whether anemia is [3] or is not [53] an independent predictor of outcome. (See 'Evaluation' above.)

In a larger study, anemia was not an independent predictor of outcomes in over 50,000 HF patients ≥65 years of age who were admitted to the hospital with a principal diagnosis of HF, either new or recurrent [49]. Although anemia was a significant predictor of one-year mortality, it was also associated with a wide range of possible confounding factors, including cardiac and noncardiac comorbidities, indices of HF severity, age, gender, and nursing-home residence. In a multivariate analysis incorporating these variables, there was no difference in one-year mortality between patients with a normal hematocrit (40 to 44 percent) and those with severe anemia (hematocrit ≤24 percent). The authors suggested that prior studies were limited by exclusion criteria, small sample size, and, most importantly, failure to incorporate the broad range of possible confounders.

Endogenous erythropoietin — The low hemoglobin concentrations seen in patients with HF are associated with elevated plasma erythropoietin concentrations in some studies [54-56]. For example, in one report of 108 patients with HF and 45 controls, plasma erythropoietin increased progressively with NYHA class (from 1.4 mU/mL for NYHA class I to 34 mU/mL for NYHA class IV).

Elevated plasma erythropoietin has been correlated with a lower rate of survival. This was illustrated in a study of 74 patients with HF; approximately one-third each were NYHA class II, III, and IV [57]. Two-year mortality was significantly higher for patients with elevated (≥22.6 mU/mL) plasma erythropoietin (32 versus 16 percent). Both the hemoglobin concentration and plasma erythropoietin were independently predictive of survival.

Changes in hemoglobin — The impact of changes in hemoglobin over time was evaluated in a retrospective analysis from the Val-HeFT trial. Independent of the presence or absence of anemia at baseline, changes in hemoglobin over a one-year period were inversely related to morbidity and mortality, as follows [45]:

●Patients in the quartile with the largest average decrease in hemoglobin (14.2 to 12.6 g/dL) had, when compared with the quartile with little change in hemoglobin (13.7 to 13.8 g/dL), a significantly increased risk of morbid events and death (hazard ratio [HR] 1.4 and 1.6, respectively).

●On the other hand, increasing hemoglobin was associated with a significantly lower mortality rate in patients with and without anemia at baseline (HR 0.78 and 0.79, respectively).

Chronic anemia may indicate risk of development of HF in patients with end-stage kidney disease [58,59]. In one study of 432 such patients, each 1 g/dL decrease in hemoglobin was independently associated with left ventricular dilatation, the development of HF, and mortality [58]. (See "Overview of screening and diagnosis of heart disease in patients on dialysis".)

These observations cannot distinguish between a direct effect of the hemoglobin concentration and the effect of systemic factors that affect both outcomes and the hemoglobin concentration in parallel.

Low transferrin saturation — Transferrin saturation (TSAT), which is normally in the range of 20 to 50 percent, can be reduced to <20 percent in both iron deficiency and the anemia of inflammation. In one study of 157 patients with HF, TSAT was <20 percent in 16, 72, and 100 percent of those with NYHA functional class I or II, III, and IV, respectively [60]. A TSAT <20 percent was associated with lower peak oxygen consumption and increased risk of mortality at median two-year follow-up (HR 3.4, 95% CI 1.5-7.7) and predicted mortality independent of hemoglobin.

Thus, disordered iron homeostasis (ie, iron deficiency, anemia of inflammation, or both) in patients with HF is associated with both impaired exercise capacity and survival.

Some have postulated that iron deficiency may directly affect myocardial function. Iron is a necessary component of mitochondrial proteins critical to myocardial energy production. A preliminary study found lower iron content and lower type 1 transferrin receptor mRNA levels in myocardial tissue from six explanted failing hearts compared with tissue from five unused donor hearts [61]. Further study is needed to determine whether there is a relationship between myocardial iron deficiency and HF.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Anemia in adults" and "Society guideline links: Heart failure in adults".)

SUMMARY AND RECOMMENDATIONS

●Anemia is frequently present in patients with heart failure (HF). Multiple factors, including increased levels of circulating cytokines, hemodilution, iron deficiency, use of angiotensin converting enzyme inhibitors, renal insufficiency, and poor nutrition may be operative in an individual patient.

●Anemia is a major sign of disease and its cause should be sought in all patients. (See 'Evaluation' above.)

●We suggest using a restrictive red blood cell transfusion strategy (eg, trigger hemoglobin threshold of 7 to 8 g/dL), rather than a more liberal threshold (such as <10 g/dL) in patients with HF (Grade 2C). Given the low-quality evidence available, we suggest basing individualized transfusion decisions upon clinical judgment, including whether the patient appears to have symptoms from anemia. (See 'Transfusion' above and "Indications and hemoglobin thresholds for red blood cell transfusion in the adult".) (Related Pathway(s): Anemia: Indications for red blood cell transfusion in hospitalized adults.)

●When red blood cell transfusion is required in a patient with HF, careful attention to volume status is recommended, including adjustment of transfusion rate and supplemental diuretics as needed to avoid volume overload. (See "Transfusion-associated circulatory overload (TACO)", section on 'Prevention'.)

●A serum ferritin <41 ng/mL or a transferrin saturation (TSAT) <20 percent is strongly suggestive of iron deficiency in patients with comorbidities such as HF. For individuals with values above these thresholds for whom there is a strong suspicion of iron deficiency, additional testing such as soluble transferrin receptor may be appropriate. (See 'Evaluation of cause of anemia' above and "Causes and diagnosis of iron deficiency and iron deficiency anemia in adults", section on 'Diagnostic evaluation'.)

●Iron deficiency and iron deficiency anemia should be treated with iron replacement, along with an evaluation for the underlying cause of the deficiency. The best route of iron administration in patients with HF is not known. Both oral and intravenous routes are effective in non-HF patients with normal gastrointestinal absorption. For patients with HF, some experts prefer intravenous iron replacement given the limited available evidence on oral iron supplementation and concerns about possible reduced oral iron absorption in HF. (See 'Iron' above and "Treatment of iron deficiency anemia in adults", section on 'Iron replacement products'.)

●We recommend against the use of erythropoiesis-stimulating agents in patients with mild to moderate anemia and HF. (Grade 1B). (See 'ESAs (not recommended)' above.)

●Among patients with HF, it is uncertain whether anemia is an independent predictor of increased mortality or reflects more advanced disease and more extensive comorbidities. (See 'Prognosis' above.)

ACKNOWLEDGMENT — We are saddened by the death of Stanley L Schrier, MD, who passed away in August 2019. The editors at UpToDate gratefully acknowledge Dr. Schrier's role as Section Editor on this topic, his tenure as the founding Editor-in-Chief for UpToDate in Hematology, and his dedicated and longstanding involvement with the UpToDate program.