28 Anaemia

In patients with chronic heart failure (HF), anaemia is frequently diagnosed. The prevalence of anaemia depends on the severity of heart failure and the diagnostic criteria used to define it. The criteria of the World Health Organization (haemoglobin 〈12 g/dL (7.5 mmol/L) in women, haemoglobin 〈13 g/dL (8.1 mmol/L) in men) are most commonly used in the majority of studies appearing on the topic. Recently a study was performed in 2653 patients randomized in the CHARM Program.¹ This substudy revealed that anaemia was equally common in patients with CHF and preserved (27%) or reduced (25%) left ventricular ejection fraction (LVEF). The presence of anaemia was associated with diabetes, low body mass index, higher systolic and lower diastolic blood pressure, and recent heart failure (HF) hospitalization. In the study, more than one-half of the anaemic patients had impaired renal function (glomerular filtration rate 〈60 mL/min), compared with less than 30% of the nonanaemic chronic HF patients. In 2008, a meta-analysis was conducted to address the relationship between anaemia and mortality in patients with chronic HF. In this analysis, examining more than 150 000 patients, anaemia was frequently observed, being found in over one-third of the chronic HF patients. The presence of anaemia was associated with a doubled mortality risk in patients with systolic as well as diastolic HF (Fig. 28.1). When assessing the mortality risk by using multivariate analyses, anaemia remained an independent risk factor for mortality in chronic HF patients.²

The incidence of anaemia is much more difficult to determine than the prevalence. Data from the OPTIMAAL trial suggest, though, that in a population without anaemia at baseline who had a heart attack complicated by the development of HF, the incidence of anaemia was 10% in the first year.³

Chronic HF and renal failure are two entities that are often seen together, with prevalences ranging from 20% to 40%.⁴ Renal failure plays an important role in the aetiology of anaemia in chronic HF patients (Table 28.1).^5,6 Renal function has been shown to be a useful predictor of morbidity and mortality in patients with HF, both in those with impaired and in those with preserved ejection fraction.⁷ The interaction between mortality, renal failure, chronic HF, and anaemia was analysed by using a registry on more than 1 million patients. The study showed that renal failure, chronic HF, and anaemia are additive in their effects on increasing mortality.^8,9 The annual mortality rate increased from 4% in patients without anaemia, HF, and renal failure to 23% in patients with the combination of renal failure, CHF, and anaemia (Fig. 28.2).

The relationship between anaemia and HF is complex. Anaemia may lead to increased cardiac workload, resulting from an increased heart rate and stroke volume. As a response to the increased workload, the heart may undergo remodelling, marked by left ventricular hypertrophy and dilation, eventually leading to chronic HF. Renal failure itself may cause HF through hypertension and accelerated coronary atheroma. In established chronic HF, anaemia will worsen cardiac function with an increased mortality risk. On the other hand, chronic HF can also cause renal failure due to decreased cardiac output and blood pressure, and to a lesser extent, to venous congestion which subsequently reduces renal perfusion. This impaired renal perfusion may lead to chronic renal ischaemia, resulting in lower erythropoietin (EPO) levels and ultimately inducing anaemia, leading to an increased cardiac workload completing the vicious circle. Recently the cycle has been called ‘cardiorenal anaemia syndrome’.¹⁰

A novel observation has been the relation between red cell distribution width (RDW) and outcome in patients with CHF.¹¹ RDW is a measure of variation in red blood cell size. In a single centre study, increasing RDW was an independent predictor of adverse outcome, even in models containing natriuretic peptide levels and haemoglobin.¹² Why RDW should predict outcome is unclear. It is independent of anaemia as a marker and it may reflect processes including inflammation and nutritional deficiencies.

Since only a minority of chronic HF patients have severe renal failure, other factors play a role in the aetiology of anaemia. Several studies showed that iron, folic acid, or vitamin B₁₂ deficiencies are commonly observed in chronic HF patients.¹³ In a study of 173 ambulatory patients with chronic HF, just fewer than 20% of patients were anaemic: 6% were vitamin B₁₂ deficient, 13% iron deficient, and 8% folate deficient.¹⁴ The cause of these deficiencies may be related to a reduced uptake of iron and vitamins, and related to poor nutrition, malabsorption, and cardiac cachexia.¹⁵ Furthermore, the use of aspirin and oral anticoagulation can lead to microscopic amounts of gastrointestinal blood loss, contributing to the iron-deficiency anaemia. However, measuring iron status in chronic HF patients is difficult. Several markers including ferritin and transferrin are acute-phase proteins and they may be elevated in inflammatory conditions such as HF. The difficulty of identifying patients with iron deficiency by serum markers was emphasized by a Greek study. In this small study, 39 anaemic patients with severe chronic HF were included. Bone marrow biopsies were performed in all patients and the authors measured iron content in the bone marrow. They found that as many as 73% of the patients were iron deficient, despite normal serum iron and ferritin levels.¹⁶ However, it is unrealistic to routinely perform bone marrow biopsies to detect iron deficiency in anaemic chronic HF patients, therefore newer markers including hepcidin and soluble transferrin receptor may be of value since they are less influenced by inflammation.

Hepcidin is a hormone released by the liver that prevents iron leaving the plasma pool. It is reduced in anaemia and hypoxia (encouraging iron uptake and hence haemoglobin synthesis) and increased in iron overload. Crucially, it is induced by inflammation, and appears important in the generation of the anaemic of chronic inflammation. Its role in chronic HF has only recently started to be explored, and its importance is not yet clear.¹⁷

In chronic HF, the presence of haemodilution has been investigated in several small mechanistic studies. Impaired renal perfusion in chronic HF causes activation of the renin–angiotensin system (RAS), resulting in salt and fluid retention, and consequently increases extracellular volume (ECV). Anaemic HF patients have an elevated ECV, which is significantly correlated to lower haemoglobin levels.^17,18 The increased ECV in chronic HF causes haemodilution, which may result in pseudoanaemia. In these small studies, the anaemic HF patients received higher doses of diuretics but nevertheless had an elevated ECV. Importantly, although fluid retention was related to anaemia, signs of fluid retention with physical examination were absent which may indicate that haemodilution is present before clinical manifestation of fluid retention can be observed.

EPO regulates erythroid-cell proliferation in the bone marrow. EPO expression is inversely related to tissue oxygenation and haemoglobin levels, and there is a semilogarithmic relation between the EPO response (log) and the degree of anaemia (linear). EPO responses in the anaemia of chronic disease are inadequate for the degree of anaemia. Such a response has been seen in some studies of patients with chronic HF, but others have found that endogenous EPO levels are elevated in HF, proportional to the severity of symptoms, and are independent predictors of survival.¹⁹ A significant proportion (one-third) of anaemic HF patients had higher EPO levels than one would expect based on the severity of anaemia.²⁰ These relatively high endogenous EPO levels suggest resistance of the bone marrow to the hormone in a proportion of anaemic HF patients. In contrast, most patients with kidney disease show relatively low EPO levels, and although endogenous EPO levels are still elevated in such patients, they are inappropriately low for the degree of anaemia. It has been observed that patients with chronic HF express elevated levels of TNFα,²¹ which in turn may reduce the hematopoietic proliferation. Experimental data support this hypothesis and the proliferative capacity of the bone marrow progenitor cells in mice with HF was shown to be reduced to approximately 50% of control mice.²²

The use of angiotensin converting enzyme (ACE) inhibitors and β-blockers may also lead to anaemia. It has been shown that intervention in the RAS results in reduced haematopoietic activity. A substudy of SOLVD showed that patients randomized to enalapril significantly had a 56% higher risk of developing anaemia.²³ In order to elucidate the mechanism involved, it has been shown that serum of anaemic chronic HF patients inhibited in vitro the proliferation of bone marrow derived erythropoietic progenitor cells of healthy donors by almost one-fifth.²⁴ The hormone responsible for the inhibition of haematopoiesis is N-acetyl-seryl-aspartyl-lysyl-proline (Ac-SDKP). Levels of this haematopoiesis inhibitor, which is almost exclusively degraded by ACE, were significantly higher in

anaemic chronic HF patients. There was a clear correlation between Ac-SDKP levels and proliferation of erythroid progenitor cells, thereby linking haematopoiesis to the RAS. However, it must be emphasized that there is no question about the importance of ACE inhibitors in chronic HF and the potential side effects on haematopoiesis are no contraindication to prescribe this drug in anaemic chronic HF patients. The relation between β-blockade and hematopoiesis is less well known. In a large β-blocker trial in chronic HF, there has been the suggestion that the use of β-blockers is associated with occurrence of anaemia.²⁵

Considering the increased mortality risk associated with anaemia, trials have been designed to increase haemoglobin levels. Because the pathophysiology underlying anaemia is complex, different approaches have been used to increase haemoglobin levels. Of these, treatment with EPO, intravenous iron, or the combination of both has been studied extensively.

The use of iron has been investigated in several small trials with chronic HF patients. Three recent trials showed beneficial effects in anaemic and nonanaemic HF patients.^26,–28 Intravenous iron therapy without EPO resulted in a clear increase in haemoglobin levels. In addition, iron therapy reduced NT-proBNP levels in anaemic patients with chronic HF and moderate renal failure.²⁶ This effect was accompanied by an improved quality of life, exercise capacity, and cardiac function. In another study almost one-half of the patients were considered iron deficient, and they showed, as expected, the highest haemoglobin response to iron therapy. Remarkably, even in the non-iron-deficient group, iron therapy resulted in a modest increase in haemoglobin levels.²⁷

In the largest trial of intravenous iron, FAIR-HF, 459 patients with CHF defined as having iron deficiency on the basis of a ferritin level less than 100 μg/L, or between 100 and 299 μg/L if the transferrin saturation was less than 20%, were randomized to receive intravenous iron to achieve repletion or placebo.²⁹ Although there were no differences in mortality between the groups, iron therapy was associated with marked improvement in symptoms and in six-minute walk test distance (Fig. 28.3). The benefits were seen regardless of whether the patients were actually anaemic at baseline or not.

For decades, EPO has been used to treat anaemia in patients with end stage renal disease. More recently, EPO has also been used in the treatment of anaemia in other conditions, such as malignancies and chronic HF. The first studies in chronic HF patients were published almost 10 years ago and showed a beneficial effect on surrogate cardiovascular endpoints, including cardiac performance, exercise time, and renal function.^30,31 Recently, the effects of darbepoetin alfa (a longer-acting erythropoietin analogue) have been investigated in several larger trials in patients with chronic HF.^32,–34 These studies showed that treatment with darbepoetin alfa was safe and effectively raised haemoglobin levels, and that exercise capacity and quality of life improved. To investigate the role of EPO in anaemic chronic HF patients further, a meta-analysis was performed³⁵ which included a total of 650 patients from 7 studies. EPO treatment was associated with a highly significant 41% lower risk of HF hospitalization (Fig. 28.4). No significant difference in the mortality risk between the two groups was observed. In addition, the prevalence of hypertension or venous thrombosis was similar in both groups. Taking all the data from these small clinical trials together, it seems that in chronic HF, treatment with EPO is not associated with a higher mortality rate or more adverse events, whereas a possible beneficial effect on HF hospitalization is seen. However, these findings need to be confirmed in a large clinical trial. Currently, a phase III morbidity and mortality trial, the RED-HF (Reduction of Events with Darbepoetin alfa in Heart Failure) trial is being conducted and will provide important data on safety and efficacy.³⁶

Recently, concerns about the cardiovascular safety of erythropoietin have been raised. Two separate studies, Correction of Hemoglobin and Outcomes in Renal Insufficiency (CHOIR)³⁷ and Cardiovascular Risk reduction by Early Anemia Treatment with Epoetin beta (CREATE),³⁸ both studied two different erythropoietin regimens, to examine the optimal target level of haemoglobin in patients with chronic kidney disease. Both studies revealed that higher haemoglobin levels were associated with a worse outcome in anaemic patients with severe renal failure. The two trials formed also the backbone of a meta-analysis examining the effect of EPO on raising haemoglobin to different targets in patients with kidney disease.³⁹ This analysis revealed a 17% higher mortality rate in patients randomized to a higher haemoglobin target. In addition, the latter patients also had a higher rate of poorly controlled blood pressure and arteriovenous access thrombosis. On the basis of these studies, the current guidelines recommend target haemoglobin levels of 10–12 g/dL (6.3–7.5 mmol/L).

However, it is important to notice that in all these studies, no placebo group was included. The first placebo-controlled trial was performed by Pfeffer and colleagues. They included in TREAT (Trial to Reduce Cardiovascular Events with Aranesp Therapy) more than 4000 patients with type II diabetes and renal failure not requiring dialysis.⁴⁰ The median achieved haemoglobin concentrations were 12.5 g/dL (7.8 mmol/L) in the patients who received darbepoetin alfa and 10.6 g/dL (6.6 mmol/L) in the patients in the placebo group. The authors report no differences between the two groups in the overall rates of death or combined cardiovascular endpoints. Importantly, of the patients assigned to darbepoetin alfa, 101 patients experienced a stroke, compared to 53 patients randomized to placebo (hazard ratio 1.92), which is statistically significantly different. There was only a marginal difference in quality of life. Therefore, in patients with chronic renal failure who do not require dialysis, the benefit of darbepoetin alfa to correct anaemia do not outweigh the risk of stroke for most patients.

These findings in renal disease are remarkable, and seem to be in contrast with the data in patients with chronic HF.⁵ It needs to be emphasized that in these studies in chronic kidney disease, only a minority of patients suffered from HF and the aetiology of anaemia in both conditions is different. Furthermore, renal function, although impaired, was on average considerably less impaired in the chronic HF studies than in the kidney trials. Taking these factors into consideration, one must conclude that more data on the treatment of anaemia in chronic HF is needed before recommendations can be made as discussed above.

Besides haematopoietic effects, EPO is also involved the protection against ischaemic injury and stimulation of angiogenesis. In recent years, a functional EPO–EPO-receptor system has been demonstrated in nonhaematopoietic cells including the brain and the heart. This suggests that besides stimulating haematopoiesis, EPO also plays a role as a pleiotropic survival and growth factor.

In animal models, EPO administration during cardiac ischaemia/reperfusion injury reduced infarct size and improved haemodynamics.⁴¹ The acute effect of EPO on infarct size are at least partly related to an antiapoptotic effects.^42,43 The beneficial findings of EPO were not only demonstrated in mice and rats, but also in rabbits and canines. Therefore, as a next step, a small safety study in patients with acute myocardial infarction undergoing percutaneous intervention was performed. Intravenous single high-dose darbepoetin alfa in acute myocardial infarction was both safe and well tolerated.⁴⁴ Furthermore, darbepoetin treatment stimulated the mobilization of endothelial progenitor cells (EPCs). Currently, several trials are recruiting patients to investigate the effect of EPO on infarct size and left ventricular function in patients presenting with acute myocardial infarction.

Besides acute cardioprotection, through inhibition of apoptosis, EPO has also shown to induce the formation of new capillaries in the heart after ischaemic injury. In a rat model of chronic HF, EPO administration 3 weeks after myocardial infarction improved cardiac performance which was associated with a significant increase in capillary density.⁴⁵ The mechanism of neovascularization is related to both stimulation of endothelial cell proliferation and mobilization of EPCs from the bone marrow.⁴⁶ Being aware of the pleiotropic effects of EPO, it is tempting to speculate that the beneficial effects of EPO in anaemic patients with chronic HF might be related to neovascularization. However, further studies are needed to investigate whether nonhematopoietic effects of EPO play a role in patients with chronic HF.

Anaemia has been identified as a common comorbidity in patients with HF; approximately one-third of all chronic HF patients have anaemia, according to a recent meta-analysis. It has become clear that morbidity and mortality is significantly higher in patients suffering from anaemia. The aetiology of anaemia in chronic HF is diverse including renal failure, haematinic deficiencies, pharmacological intervention, haemodilution, and anaemia of chronic disease with EPO resistance of the bone marrow. Several small clinical trials have shown a beneficial effect of EPO on exercise capacity and quality of life in anaemic patients with chronic HF. Currently, a large morbidity and mortality trial evaluating the effects of correction of anaemia with darbepoietin is being conducted. Until the results of this trial are available, the treatment of anaemic chronic HF patients should only be assessed in clinical trials, particularly in light of recent developments in patients with chronic renal failure.