https://orcid.org/0000-0003-1828-2387
Abstract
Current understanding of iron-deficient heart failure is based on blood tests that are thought to reflect systemic iron stores, but the available evidence suggests greater complexity. The entry and egress of circulating iron is controlled by erythroblasts, which (in severe iron deficiency) will sacrifice erythropoiesis to supply iron to other organs, e.g. the heart. Marked hypoferraemia (typically with anaemia) can drive the depletion of cardiomyocyte iron, impairing contractile performance and explaining why a transferrin saturation < ≈15%–16% predicts the ability of intravenous iron to reduce the risk of major heart failure events in long-term trials (Type 1 iron-deficient heart failure). However, heart failure may be accompanied by intracellular iron depletion within skeletal muscle and cardiomyocytes, which is disproportionate to the findings of systemic iron biomarkers. Inflammation- and deconditioning-mediated skeletal muscle dysfunction—a primary cause of dyspnoea and exercise intolerance in patients with heart failure—is accompanied by intracellular skeletal myocyte iron depletion, which can be exacerbated by even mild hypoferraemia, explaining why symptoms and functional capacity improve following intravenous iron, regardless of baseline haemoglobin or changes in haemoglobin (Type 2 iron-deficient heart failure). Additionally, patients with advanced heart failure show myocardial iron depletion due to both diminished entry into and enhanced egress of iron from the myocardium; the changes in iron proteins in the cardiomyocytes of these patients are opposite to those expected from systemic iron deficiency. Nevertheless, iron supplementation can prevent ventricular remodelling and cardiomyopathy produced by experimental injury in the absence of systemic iron deficiency (Type 3 iron-deficient heart failure). These observations, taken collectively, support the possibility of three different mechanistic pathways for the development of iron-deficient heart failure: one that is driven through systemic iron depletion and impaired erythropoiesis and two that are characterized by disproportionate depletion of intracellular iron in skeletal and cardiac muscle. These mechanisms are not mutually exclusive, and all pathways may be operative at the same time or may occur sequentially in the same patients.
Three potential mechanistic pathways of iron-deficient heart failure. In Type 1 iron deficiency, meaningful systemic iron deficiency (typically TSAT <≈15%–16% with anaemia) results in cardiomyocyte iron deficiency and dysfunction, even though cardiomyocytes demonstrate compensatory increases in iron influx proteins and decreases in iron efflux proteins. In Type 2 iron deficiency, systemic or local inflammation and/or deconditioning cause cytosolic iron depletion within skeletal myocytes as a result of decreases in iron influx proteins and increases in iron efflux proteins; the superimposition of modest degrees of hypoferraemia (TSAT < ≈20%) can trigger significant skeletal muscle dysfunction, leading to worsening dyspnoea and exercise intolerance. In Type 3 iron deficiency, neurohormonal activation and still-to-be-defined mechanisms of subcellular iron homeostatic dysregulation cause cytosolic iron depletion as a result of decreases in iron influx proteins and increases in iron efflux proteins; under these conditions, the superimposition of modest degrees of hypoferraemia (TSAT < ≈20%) can possibly exacerbate cardiomyocyte dysfunction and the progression of heart failure. TSAT, transferrin saturation.
About 50% of patients with heart failure with a reduced ejection fraction meet guideline criteria for an iron deficiency state,1 but these criteria are based on the assessment of blood-borne iron biomarkers, which were developed as measures of systemic iron stores, in order to identify iron deficiency in erythroid precursors. The bone marrow requires iron for the synthesis of haemoglobin, and in patients with chronic kidney disease or inflammatory bowel disease, intravenous iron supplementation is given with a principal goal of alleviating anaemia.2
In contrast, a primary reason to identify iron deficiency in patients with heart failure is to alleviate a deficiency of iron within cardiomyocytes or skeletal muscle, which require reactive ferrous (Fe2+) iron for the synthesis of heme and iron–sulfur clusters that are needed for the production of ATP and the support of contractile function.3–6 Interestingly, the improvement in symptoms and functional capacity and the reduction in the risk of hospitalization for heart failure in response to intravenous iron in patients with heart failure may not be limited to those with anaemia.7 This finding, taken together with the fact that the goals of treating iron deficiency in heart failure differ from those in other chronic diseases, suggests the need to examine whether the determinants of iron homeostasis in the heart and skeletal muscle may differ from those in the bone marrow.
Recent work has shown that iron homeostasis occurs not only at the systemic level but also at a cellular level, allowing individual organs to regulate the entry, intracellular trafficking, and efflux of iron to adapt to their microenvironments.8–10 Based on the findings of mechanistic studies in skeletal and cardiac muscle cells, we propose a novel conceptual framework that describes three different potential pathways for the development of clinically relevant iron deficiency in heart failure (Graphical Abstract). Importantly, the mechanisms that underlie these pathways are not mutually exclusive, and several may be operative at the same time or may occur sequentially in the same patient.
Circulating iron is controlled by erythroblasts and is preferentially directed towards haemoglobin synthesis and erythropoiesis
The entry of iron into the bloodstream is regulated by the hormonal actions of hepcidin, which inhibits the absorption of dietary iron and the release of iron stored in hepatocytes and within the reticuloendothelial system.11 When erythropoiesis is stimulated, erythroblasts release erythroferrone, which (by suppressing the synthesis of hepcidin by the liver) allows the enhanced entry of iron into the bloodstream to support red blood cell production.12 Iron is transported by the circulation in the ferric (Fe3+) state by transferrin, which forms a complex with transferrin receptor protein 1 (TfR1) to allow for endocytosis into erythroid precursors. TfR1 can be shed from cell membranes and is measured as soluble transferrin receptor (sTfR).13 Circulating iron that is not bound to transferrin enters erythroid precursors through the actions of divalent metal transporter 1 (DMT1) and other divalent transporters, but these convey a tiny fraction of circulating iron, except when large intravenous doses of iron overwhelm the iron-carrying capacity of transferrin.14
Erythroblasts are the command centre for circulating iron
Following endocytosis of transferrin–TfR1 complex by erythroblasts, Fe3+ is reduced to Fe2+ and is then transported across the endosomal membrane into the cytosol by DMT1.15 The Fe2+ that resides in the cytosol (often referred to as ‘catalytic iron’) is highly reactive and is preferentially directed to mitochondria (by mitoferrin and DMT1) for the synthesis of heme and iron–sulfur clusters used for the production of haemoglobin and ATP.16,17
In the presence of erythropoietin, the level of cytosolic Fe2+ iron is the primary determinant of erythroblast proliferation. If iron uptake by erythroblasts were to exceed their immediate capacity for haemoglobin synthesis, highly reactive cytosolic Fe2+ is sequestered intracellularly as nonreactive Fe3+ within a ferritin nanocage to prevent cytosolic Fe2+ from rising to levels that can induce oxidative stress.17–19 If erythroblast ferritin is genetically suppressed, the cytosolic labile pool of Fe2+ expands, oxidative stress increases dramatically, and proliferation is transiently stimulated.20 However, continuation of oxidative stress induces injury to cell membranes and an iron-dependent form of cell death (ferroptosis), which impairs erythropoiesis.21,22 Conversely, if ferritin is overexpressed in erythroid precursors, the cytosolic labile pool of reactive Fe2+ becomes depleted, and erythropoiesis is impaired.23
Erythroblasts represent a highly privileged destination for circulating iron. When intracellular stores of iron within erythroid precursors are deficient, TfR1 and DMT1 are markedly upregulated17; most of the sTfR in the bloodstream is derived from TfR1 shed from erythroblasts and reticulocytes.13 Erythroblasts in the bone marrow consume >90% of the iron that is transported by transferrin, and erythrocyte haemoglobin contains two-thirds of the body’s iron.18 During the 120-day lifespan of mature erythrocytes, their iron-rich haemoglobin is subjected to almost 200 000 cycles of redox stress. The extraordinary capacity of red blood cells to sustain this stress is exemplified by their surprisingly low turnover rate, i.e. only 1% of the erythrocyte mass is replaced daily.
Taken collectively, these observations indicate that erythroblasts serve as the central command centre for systemic iron homeostasis. By virtue of their synthesis of erythroferrone and their expression of TfR1, erythroid precursors govern the entry and egress of iron in and out of the circulation. These two control mechanisms are critically important, since—unless erythropoietin is deficient or ferroptosis intervenes as a result of iron overload—the level of cytosolic iron in erythroblasts is the primary determinant of erythropoiesis.
Type 1 iron-deficient heart failure: reduced systemic iron availability contributes importantly to worsening heart failure
The dominance of erythroblasts in directing systemic iron metabolism does not explain how iron can be delivered to other nonerythroid organs that require it. It is therefore noteworthy that erythroblasts possess a critically important capacity to export their iron through ferroportin, an efflux transporter, and this export is essential for maintaining circulating levels of iron.24 If ferroportin signalling within erythroblasts is impaired, increases in cytosolic iron promote haemoglobin synthesis25; if ferroportin is silenced, erythroblasts suffer from iron overload-induced oxidative stress, and simultaneously, serum iron and transferrin saturation (TSAT) decline.25,26 The activity of ferroportin in erythroblasts is controlled by both systemic and intracellular levels of hepcidin, which acts to promote the degradation of ferroportin.27,28 Accordingly, in states of absolute iron deficiency, the suppression of hepcidin not only enhances gastrointestinal absorption and macrophage release but it also augments the export of iron out of erythroblasts.28,29 This egress forces erythroblasts to reduce their synthesis of haemoglobin, but it increases iron availability to other tissues26; i.e. erythropoietic activity is sacrificed to prevent iron deficiency in the heart and skeletal muscles. As a result, anaemia is generally the first indicator of a meaningful systemic iron deficiency state.24,25
Cardiomyocyte response to scarce systemic iron availability
Even though the heart is not prioritized for the delivery of circulating iron, systemic iron availability is a critically important determinant of cardiomyocyte iron. Circulating iron enters cardiomyocytes through TfR1, and cardiac-specific silencing of TfR1 causes cytosolic iron depletion and cardiomyopathy.29 In experimental states of reduced tissue iron delivery, iron influx into the heart is supported by upregulation of TfR1 and DMT1 in cardiomyocytes,3,8,30,31 and iron entering the heart is preferentially retained, because upregulation of cardiac hepcidin leads to suppression of cardiac ferroportin, blocking the egress of scarce intracellular iron out of cardiomyocytes.8,31,32 These compensatory mechanisms maintain critical levels of mitochondrial iron, and thus, ventricular function is typically not compromised in iron deficiency states even with anaemia.8,33 However, experimentally, if the reduction in systemic iron delivery is severe and prolonged, cardiomyocyte iron levels are reduced, thereby impairing calcium dynamics, mitochondrial function, and contractile performance in otherwise healthy hearts.3,32,34–36 Myocardial injury may heighten cardiac vulnerability to meaningful degrees of hypoferraemia.37
Therefore, under conditions of marked systemic iron deficiency, blood-borne iron availability (as reflected by circulating biomarkers) can drive the severity of cardiac iron depletion and functional abnormalities. We refer to this pathway as Type 1 iron-deficient heart failure (Figure 1, Table 1, and Graphical Abstract).4 Accordingly, intravenous iron leads to myocardial iron repletion (the magnitude of which is related to baseline values of TSAT) and an improvement in ventricular function.5,6,38
Mechanism characteristic of Type 1 iron-deficient heart failure. The delivery of circulating iron is prioritized to erythroblasts in the bone marrow. When transferrin saturations are low (due to either absolute or functional iron deficiency), iron is released for utilization by other organs (i.e. the heart), even though such release impairs erythropoiesis, leading to anaemia. Cardiomyocytes upregulate the entry of iron and suppress the egress of iron (and iron is also released from the ferritin nanocage); these actions support cytosolic iron levels within the heart. Given the relationship between iron in the circulation and in cardiomyocytes, the measurement of transferrin saturation represents a reasonable metric to identify patients most likely to experience a reduced risk of major heart failure events in large-scale trials
Three proposed mechanistic pathways for the development of iron-deficient heart failure
| Type 1 iron-deficient heart failure | Type 2 iron-deficient heart failure | Type 3 iron-deficient heart failure | |
|---|---|---|---|
| Patient characteristics | TSAT < ≈15%–16% with heart failure and a reduced ejection fraction, typically with anaemia | TSAT < ≈20% with heart failure and meaningful functional disability, typically NYHA class III symptoms, with no or mild anaemia | TSAT < ≈20% with advanced heart failure with severely impaired ejection fraction, with no or mild anaemia |
| Primary feature of pathogenesis | Moderate-to-severe hypoferraemia leading to cardiomyocyte cytosolic iron depletion | Inflammation- and deconditioning-related skeletal muscle dysfunction and intracellular iron depletion exacerbated by mild hypoferraemia | Advanced cardiomyopathy with cardiomyocyte cytosolic iron depletion exacerbated by mild hypoferraemia |
| Cause of iron deficiency state | Absolute deficiency (occult gastrointestinal losses) or functional deficiency (macrophage trapping) | Inflammation-related functional deficiency (macrophage trapping), but also absolute deficiency (occult gastrointestinal losses) | Absolute deficiency (occult gastrointestinal losses) or functional deficiency (macrophage trapping) |
| Effect on erythropoiesis | Hypochromic, microcytic anaemia is typically present | Haemoglobin may be normal. Anaemia (if present) is typically normochromic and normocytic | Haemoglobin may be normal. Anaemia (if present) is typically normochromic and normocytic |
| Systemic iron biomarkers | Transferrin saturation < ≈15%–20%, serum ferritin usually in normal range (20–300 μg/L) | Transferrin saturation < ≈20%, serum ferritin usually in normal range (20–300 μg/L) | Transferrin saturation < ≈20%, serum ferritin usually in normal range (20–300 μg/L) |
| Changes in iron proteins in skeletal myocytes and cardiomyocytes | Cardiomyocytes: increased influx (TfR1 and DMT1). Decreased efflux (increased hepcidin and decreased ferroportin) | Skeletal myocytes: decreased influx (TfR1 and DMT1). Increased efflux (decreased hepcidin and increased ferroportin) | Cardiomyocytes: decreased influx (TfR1 and DMT1). Increased efflux (decreased hepcidin and increased ferroportin) |
| Relationship of serum iron biomarkers to cytosolic iron levels | Cardiomyocyte cytosolic iron deficiency likely related to level of hypoferraemia | Skeletal muscle cytosolic iron depletion disproportionate to hypoferraemia | Cardiomyocyte cytosolic iron depletion disproportionate to hypoferraemia |
| Consequences of iron deficiency | Impaired haemoglobin synthesis and cardiomyocyte mitochondrial function and ATP production | Deficits in skeletal myocyte ATP and myoglobin synthesis; structural and functional derangements in skeletal muscle | Impaired cardiomyocyte mitochondrial function and ATP production |
| Response to intravenous iron supplementation | Decreased risk of major heart failure outcomes | Rapid improvement in symptoms, exercise tolerance, and functional capacity | Iron repletion produces cellular benefits. Clinical response not yet investigated |
| Type 1 iron-deficient heart failure | Type 2 iron-deficient heart failure | Type 3 iron-deficient heart failure | |
|---|---|---|---|
| Patient characteristics | TSAT < ≈15%–16% with heart failure and a reduced ejection fraction, typically with anaemia | TSAT < ≈20% with heart failure and meaningful functional disability, typically NYHA class III symptoms, with no or mild anaemia | TSAT < ≈20% with advanced heart failure with severely impaired ejection fraction, with no or mild anaemia |
| Primary feature of pathogenesis | Moderate-to-severe hypoferraemia leading to cardiomyocyte cytosolic iron depletion | Inflammation- and deconditioning-related skeletal muscle dysfunction and intracellular iron depletion exacerbated by mild hypoferraemia | Advanced cardiomyopathy with cardiomyocyte cytosolic iron depletion exacerbated by mild hypoferraemia |
| Cause of iron deficiency state | Absolute deficiency (occult gastrointestinal losses) or functional deficiency (macrophage trapping) | Inflammation-related functional deficiency (macrophage trapping), but also absolute deficiency (occult gastrointestinal losses) | Absolute deficiency (occult gastrointestinal losses) or functional deficiency (macrophage trapping) |
| Effect on erythropoiesis | Hypochromic, microcytic anaemia is typically present | Haemoglobin may be normal. Anaemia (if present) is typically normochromic and normocytic | Haemoglobin may be normal. Anaemia (if present) is typically normochromic and normocytic |
| Systemic iron biomarkers | Transferrin saturation < ≈15%–20%, serum ferritin usually in normal range (20–300 μg/L) | Transferrin saturation < ≈20%, serum ferritin usually in normal range (20–300 μg/L) | Transferrin saturation < ≈20%, serum ferritin usually in normal range (20–300 μg/L) |
| Changes in iron proteins in skeletal myocytes and cardiomyocytes | Cardiomyocytes: increased influx (TfR1 and DMT1). Decreased efflux (increased hepcidin and decreased ferroportin) | Skeletal myocytes: decreased influx (TfR1 and DMT1). Increased efflux (decreased hepcidin and increased ferroportin) | Cardiomyocytes: decreased influx (TfR1 and DMT1). Increased efflux (decreased hepcidin and increased ferroportin) |
| Relationship of serum iron biomarkers to cytosolic iron levels | Cardiomyocyte cytosolic iron deficiency likely related to level of hypoferraemia | Skeletal muscle cytosolic iron depletion disproportionate to hypoferraemia | Cardiomyocyte cytosolic iron depletion disproportionate to hypoferraemia |
| Consequences of iron deficiency | Impaired haemoglobin synthesis and cardiomyocyte mitochondrial function and ATP production | Deficits in skeletal myocyte ATP and myoglobin synthesis; structural and functional derangements in skeletal muscle | Impaired cardiomyocyte mitochondrial function and ATP production |
| Response to intravenous iron supplementation | Decreased risk of major heart failure outcomes | Rapid improvement in symptoms, exercise tolerance, and functional capacity | Iron repletion produces cellular benefits. Clinical response not yet investigated |
DMT1, divalent metal transporter 1; NYHA, New York Heart Association; TSAT, transferrin saturation; TfR1, transferrin receptor protein 1.
Three proposed mechanistic pathways for the development of iron-deficient heart failure
| Type 1 iron-deficient heart failure | Type 2 iron-deficient heart failure | Type 3 iron-deficient heart failure | |
|---|---|---|---|
| Patient characteristics | TSAT < ≈15%–16% with heart failure and a reduced ejection fraction, typically with anaemia | TSAT < ≈20% with heart failure and meaningful functional disability, typically NYHA class III symptoms, with no or mild anaemia | TSAT < ≈20% with advanced heart failure with severely impaired ejection fraction, with no or mild anaemia |
| Primary feature of pathogenesis | Moderate-to-severe hypoferraemia leading to cardiomyocyte cytosolic iron depletion | Inflammation- and deconditioning-related skeletal muscle dysfunction and intracellular iron depletion exacerbated by mild hypoferraemia | Advanced cardiomyopathy with cardiomyocyte cytosolic iron depletion exacerbated by mild hypoferraemia |
| Cause of iron deficiency state | Absolute deficiency (occult gastrointestinal losses) or functional deficiency (macrophage trapping) | Inflammation-related functional deficiency (macrophage trapping), but also absolute deficiency (occult gastrointestinal losses) | Absolute deficiency (occult gastrointestinal losses) or functional deficiency (macrophage trapping) |
| Effect on erythropoiesis | Hypochromic, microcytic anaemia is typically present | Haemoglobin may be normal. Anaemia (if present) is typically normochromic and normocytic | Haemoglobin may be normal. Anaemia (if present) is typically normochromic and normocytic |
| Systemic iron biomarkers | Transferrin saturation < ≈15%–20%, serum ferritin usually in normal range (20–300 μg/L) | Transferrin saturation < ≈20%, serum ferritin usually in normal range (20–300 μg/L) | Transferrin saturation < ≈20%, serum ferritin usually in normal range (20–300 μg/L) |
| Changes in iron proteins in skeletal myocytes and cardiomyocytes | Cardiomyocytes: increased influx (TfR1 and DMT1). Decreased efflux (increased hepcidin and decreased ferroportin) | Skeletal myocytes: decreased influx (TfR1 and DMT1). Increased efflux (decreased hepcidin and increased ferroportin) | Cardiomyocytes: decreased influx (TfR1 and DMT1). Increased efflux (decreased hepcidin and increased ferroportin) |
| Relationship of serum iron biomarkers to cytosolic iron levels | Cardiomyocyte cytosolic iron deficiency likely related to level of hypoferraemia | Skeletal muscle cytosolic iron depletion disproportionate to hypoferraemia | Cardiomyocyte cytosolic iron depletion disproportionate to hypoferraemia |
| Consequences of iron deficiency | Impaired haemoglobin synthesis and cardiomyocyte mitochondrial function and ATP production | Deficits in skeletal myocyte ATP and myoglobin synthesis; structural and functional derangements in skeletal muscle | Impaired cardiomyocyte mitochondrial function and ATP production |
| Response to intravenous iron supplementation | Decreased risk of major heart failure outcomes | Rapid improvement in symptoms, exercise tolerance, and functional capacity | Iron repletion produces cellular benefits. Clinical response not yet investigated |
| Type 1 iron-deficient heart failure | Type 2 iron-deficient heart failure | Type 3 iron-deficient heart failure | |
|---|---|---|---|
| Patient characteristics | TSAT < ≈15%–16% with heart failure and a reduced ejection fraction, typically with anaemia | TSAT < ≈20% with heart failure and meaningful functional disability, typically NYHA class III symptoms, with no or mild anaemia | TSAT < ≈20% with advanced heart failure with severely impaired ejection fraction, with no or mild anaemia |
| Primary feature of pathogenesis | Moderate-to-severe hypoferraemia leading to cardiomyocyte cytosolic iron depletion | Inflammation- and deconditioning-related skeletal muscle dysfunction and intracellular iron depletion exacerbated by mild hypoferraemia | Advanced cardiomyopathy with cardiomyocyte cytosolic iron depletion exacerbated by mild hypoferraemia |
| Cause of iron deficiency state | Absolute deficiency (occult gastrointestinal losses) or functional deficiency (macrophage trapping) | Inflammation-related functional deficiency (macrophage trapping), but also absolute deficiency (occult gastrointestinal losses) | Absolute deficiency (occult gastrointestinal losses) or functional deficiency (macrophage trapping) |
| Effect on erythropoiesis | Hypochromic, microcytic anaemia is typically present | Haemoglobin may be normal. Anaemia (if present) is typically normochromic and normocytic | Haemoglobin may be normal. Anaemia (if present) is typically normochromic and normocytic |
| Systemic iron biomarkers | Transferrin saturation < ≈15%–20%, serum ferritin usually in normal range (20–300 μg/L) | Transferrin saturation < ≈20%, serum ferritin usually in normal range (20–300 μg/L) | Transferrin saturation < ≈20%, serum ferritin usually in normal range (20–300 μg/L) |
| Changes in iron proteins in skeletal myocytes and cardiomyocytes | Cardiomyocytes: increased influx (TfR1 and DMT1). Decreased efflux (increased hepcidin and decreased ferroportin) | Skeletal myocytes: decreased influx (TfR1 and DMT1). Increased efflux (decreased hepcidin and increased ferroportin) | Cardiomyocytes: decreased influx (TfR1 and DMT1). Increased efflux (decreased hepcidin and increased ferroportin) |
| Relationship of serum iron biomarkers to cytosolic iron levels | Cardiomyocyte cytosolic iron deficiency likely related to level of hypoferraemia | Skeletal muscle cytosolic iron depletion disproportionate to hypoferraemia | Cardiomyocyte cytosolic iron depletion disproportionate to hypoferraemia |
| Consequences of iron deficiency | Impaired haemoglobin synthesis and cardiomyocyte mitochondrial function and ATP production | Deficits in skeletal myocyte ATP and myoglobin synthesis; structural and functional derangements in skeletal muscle | Impaired cardiomyocyte mitochondrial function and ATP production |
| Response to intravenous iron supplementation | Decreased risk of major heart failure outcomes | Rapid improvement in symptoms, exercise tolerance, and functional capacity | Iron repletion produces cellular benefits. Clinical response not yet investigated |
DMT1, divalent metal transporter 1; NYHA, New York Heart Association; TSAT, transferrin saturation; TfR1, transferrin receptor protein 1.
Transferrin saturation, not serum ferritin, determines ability of intravenous iron to reduce the risk of hospitalizations for heart failure in patients with heart failure
In this review, we define a systemic iron deficiency state by a serum TSAT < ≈20%.39,40 Only iron-carrying transferrin is captured by TfR1,14 and it is the endocytosis of the complex and the release of Fe2+ into the cytosol that controls the synthesis of hepcidin in the liver and, thus, the efflux of iron from erythroblasts, duodenal enterocytes, and macrophages for utilization by the heart.41 A TSAT >20% typically does not restrict erythropoiesis.42 Furthermore, it is the endocytosis of the iron-carrying transferrin–TfR1 complex that represents the rate-limiting step for maintaining the level of cytosolic iron in cardiomyocytes.3,29,32 A TSAT < ≈15%–16% corresponds to absolute depletion of total body iron stores and clinically meaningful hypoferraemia (serum iron < ≈12 μmol/L).43,44
This conceptual framework may explain why TSAT is the most consistent predictor of the effect of intravenous iron to reduce the risk of hospitalization for heart failure or cardiovascular death in patients with heart failure. In two patient-level meta-analyses,45,46 a low baseline TSAT (<≈15%–16%) identified patients with the largest risk reduction with intravenous iron, whereas no benefit was observed in patients with a TSAT >20%. We recognize that current guidelines focus on serum ferritin (not TSAT) to identify an iron deficiency state, i.e. iron deficiency is defined by a serum ferritin level <100 μg/L (regardless of TSAT), and a TSAT <20% is considered meaningful only if the serum ferritin level is 100–299 μg/L. Although serum ferritin levels <15–20 μg/L are indicative of the absent bone marrow iron stores,43 the belief that serum ferritin levels reflect the labile iron pool assumes that catalytic iron is the primary determinant of the production and secretion of iron-rich ferritin. However, ferritin synthesis and extrusion are augmented by co-existing inflammation and oxidative stress,47,48 and in heart failure, these mechanisms drive serum ferritin levels into the normal range (20–300 μg/L), even in iron-deficient patients—thus eliminating ferritin’s diagnostic utility. Nearly half of patients with heart failure who have a serum ferritin level <100 μg/L do not have a TSAT <20% and are not hypoferremic.39,43,44 Importantly, in the HEART-FID and IRONMAN trials, intravenous iron did not improve heart failure outcomes in patients who had a serum ferritin level <100 μg/L or who qualified solely because of a serum ferritin level <100 μg/L.49,50
Most patients with heart failure with a marked systemic iron deficiency have anaemia,40,43,49 since erythroblasts sacrifice erythropoiesis to allow iron delivery to other organs.24–26 Therefore, patients with Type 1 iron-deficient heart failure and a TSAT < ≈15%–16%—in whom a marked reduction in systemic iron availability contributes importantly to worsening heart failure—typically have a hypochromic microcytic anaemia (Table 1 and Graphical Abstract). In trials with intravenous iron, patients with anaemia showed the largest reduction in the risk of major heart failure events.50,51
Type 2 iron-deficient heart failure: heart failure-related inflammation and deconditioning increases vulnerability of skeletal muscle to iron depletion, leading to worsening symptoms and exercise tolerance
Exertional dyspnoea, impaired functional capacity, and effort intolerance in patients with chronic heart failure are not closely related to changes in cardiac function or peripheral perfusion, but these limitations are strongly linked to abnormalities of skeletal muscle function.52 In both experimental and clinical heart failure, skeletal muscles demonstrate disrupted oxidative metabolism, abnormal energy generation, proteolysis and ultrastructural abnormalities, and decreased muscle mass and strength, as a result of local and systemic and inflammatory stress53,54 that interferes with normal mechanoreceptor, chemoreceptor, and metaboreceptor signalling.55–61
Numerous pathophysiological factors in heart failure contribute to these skeletal muscle abnormalities. Overactivation of the sympathetic nervous system, renin–angiotensin system, aldosterone, and neprilysin can contribute meaningfully to adverse changes in skeletal muscle metabolism and to the loss of muscle mass.62–65 The systemic inflammatory state that is commonly seen in heart failure can cause deficits in skeletal muscle energy metabolism that contribute to dyspnoea and exercise intolerance.53,54,66 Both experimentally and clinically, skeletal muscle inflammation and deconditioning are strongly associated with dyspnoea and exercise intolerance, independent of left ventricular dysfunction.53,54,66,67 Accordingly, skeletal muscle training can reduce cellular stress, ameliorate local inflammation, improve oxidative metabolism, enhance tissue mass, and improve breathlessness and fatigue.56,67,68–70 Skeletal muscle deconditioning, inflammation, and neurohormonal dysregulation can drive symptoms of exertional dyspnoea whether or not inspiratory muscles are involved.66–73
Vulnerability of skeletal muscle health to hypoferraemia in heart failure
Iron plays a critical role in skeletal muscle, not only in the generation of heme and iron–sulfur clusters required for oxidative phosphorylation and ATP production but also in the synthesis of myoglobin, which acts to retain oxygen for utilization during periods of stress.74,75 Experimentally, iron deficiency can cause skeletal muscle metabolic derangements, diminished oxidative metabolism, decreased myoglobin synthesis, impaired regenerative capacity and atrophy, inspiratory muscle weakness, and reduced physical endurance73,75–82—often as a direct tissue effect76,77 and in the absence of anaemia.80 Clinically, hypoferraemia is associated with disordered enzyme activity and reduced mass and strength of skeletal muscle.73,81,82 Skeletal muscle repair is dependent on local iron recycling, but it is independent of systemic iron homeostasis.83
Importantly, independent of changes in systemic iron availability, heart failure and aging can lead to changes in iron regulatory proteins that markedly increase vulnerability of skeletal muscle to even modest degrees of hypoferraemia. The reduced mobilization of skeletal muscles in the lower limbs (as is expected in elderly patients and in heart failure) leads to a reduction in oxidative phosphorylation and ATP production84–86 and, thus, a decline in iron utilization by skeletal myocytes.87 In deconditioned skeletal myocytes, the expression of TfR1 and DMT1 is decreased, whereas ferroportin is increased.77,87–89 Ferroportin upregulation (and iron export) may be particularly important if the need for iron by erythroblasts is increased.90 The pattern of these changes is intriguing; a systemic iron deficiency would be expected to cause upregulation of TfR1 and decreased ferroportin functionality in target tissues,91 but immobilization causes the opposite effects,77,87,88 promoting intracellular iron depletion and, thereby, contributing to the development of deconditioning-related skeletal muscle dysfunction and atrophy. Similarly, the progressive loss of skeletal muscle mass in iron-deficient subjects with cachexia due to cancer is accompanied by a decrease in TfR1 in skeletal muscle.92 However, studies of TfR1 in the skeletal muscles of patients with heart failure are lacking.
Nevertheless, the available evidence indicates that patients with heart failure are vulnerable to changes in iron proteins in skeletal muscle that can promote a meaningful depletion of intracellular iron. This tissue-specific iron dysregulation can contribute importantly to the skeletal muscle derangements that are responsible for the impaired functional capacity and effort intolerance in heart failure. This susceptibility to skeletal myocyte iron depletion may become particularly apparent if systemic iron availability were even mildly impaired, since hypoferraemia—by itself—can cause skeletal muscle energetic abnormalities, impaired strength, and atrophy.73,81,82,93 We refer to this pathway as Type 2 iron-deficient heart failure (Figure 2, Table 1, and Graphical Abstract).
Mechanism characteristic of Type 2 iron-deficient heart failure. The development of heart failure leads to inflammation- and deconditioning-mediated dysfunction and atrophy of skeletal muscle along with changes in skeletal myocytes that promote the depletion of cytosolic iron through (i) upregulation of ferritin, thereby enhancing intracellular sequestration of cytosolic iron, and (ii) upregulation of ferroportin, promoting egress of iron through ferroportin. The depletion of cytosolic iron enhances the vulnerability of skeletal myocytes to modest decreases in serum iron availability; such depletion impairs the synthesis of ATP and the functions of myoglobin. These derangements in skeletal muscle iron homeostasis can occur without anaemia and without marked decreases in transferrin saturation (as is typically seen in patients with functional iron deficiency). Skeletal muscle inflammation and cellular dysfunction are critical determinants of dyspnoea, exercise tolerance, and functional capacity in patients with chronic heart failure
Inflammation-related hypoferraemia exacerbates skeletal muscle iron depletion and can contribute to skeletal muscle abnormalities and exercise intolerance in heart failure
Given these vulnerable conditions, it is noteworthy that the systemic inflammatory state in heart failure cannot only interfere with skeletal muscle function, but it can also impair the ability of existing total iron body stores to be mobilized, leading to a state of functional iron deficiency.94 Functional iron deficiency is characterized by heightened expression of systemic hepcidin,28 and an increase in circulating hepcidin impairs the action of ferroportin to allow for iron export out of duodenal enterocytes, hepatocytes, and the reticuloendothelial system. Impairment of the release of iron from macrophages creates a particularly problematic state, since the recycling of iron from senescent erythrocytes following macrophage engulfment provides >90% of the daily iron requirements for haemoglobin synthesis and erythropoiesis.95 If iron is trapped in the reticuloendothelial system, circulating iron declines, but increased hepcidin inhibits the export of iron from erythroblasts, thus preserving erythropoiesis.
As a result, functional iron deficiency is characterized by a decrease in TSAT (<≈20%), which is often accompanied by a mild normochromic and normocytic anaemia or occurs without anaemia.94,96,97 Many patients with heart failure have a TSAT <20%, elevated serum hepcidin and proinflammatory biomarkers, macrophage iron trapping in the bone marrow, and haemoglobin and serum ferritin levels in the normal range.1,98–106 Elevated levels of hepcidin explain why—following oral iron supplementation—patients with heart failure do not show an improvement in exercise capacity,107–109 and only a minority (those without systemic inflammation and preserved gastrointestinal absorption110) show changes in TSAT or serum ferritin.107,109,111 Furthermore, higher serum ferritin levels (which are closely related to skeletal muscle ferritin but inversely related to skeletal muscle TfR191,112) cause intracellular sequestration and further depletion of cytosolic Fe2+. Therefore, in patients with heart failure, modest hypoferraemia—insufficient to cause anaemia—may trigger a disproportionate depletion of intracellular iron within skeletal myocytes,113 especially when inflammation or immobilization (acting in concert) have already drained the cytosolic iron pool. Interestingly, experimentally and clinically induced muscle immobilization—perhaps by promoting inflammation—may lead to increased hepcidin and hypoferraemia.114 It is therefore noteworthy that there is a consistent association of systemic inflammation and iron deficiency in prospectively defined cohorts of patients with heart failure, particularly in those with the most marked degrees of exercise intolerance.98,100,115
Iron repletion restores skeletal muscle function and improves exercise capacity in patients with heart failure
Two double-blind, placebo-controlled trials—FAIR-HF and CONFIRM-HF116,117—evaluated the effect of ferric carboxymaltose on symptoms, functional capacity, and health status after 24 and 52 weeks, respectively. Prior to treatment, the patients had marked impairment of 6-min walk distance (mean <300 m) and Kansas City Cardiomyopathy Questionnaire scores (mean <60), and two-thirds had New York Heart Association (NYHA) functional class III symptoms. This degree of disability is worse than that typically seen in large-scale trials of patients with heart failure and a reduced ejection fraction. However, patients were not meaningfully anaemic, i.e. the mean baseline haemoglobin level was 12.0–12.5 g/dL.
In the two trials, intravenous iron produced a remarkable improvement in 6-min walk distance (mean 30–40 m) and health status (mean 5–6 points), along with a benefit on patient and physician global assessments.116–120 An increase in skeletal muscle energetics was apparent within 2 weeks,121 and an alleviation of dyspnoea and enhanced exercise capacity was observed within 4 weeks and was sustained up to 4–6 months.116,117 Similar benefits were reported in two open-label trials (FERRIC-HF and EFFECT-HF),122,123 and an improvement of quality-of-life at 4–6 months was also observed in the AFFIRM-AHF and IRONMAN trials.49,124 These benefits on symptoms, functional capacity, and exercise tolerance were seen in both anaemic and nonanaemic patients, and there were no significant iron-induced changes in haemoglobin or ejection fraction.116,117 The mean baseline TSATs in the FAIR-HF and CONFIRM-HF trials ranged from 17% to 20%, and two-thirds of the patients had a TSAT ≤20%, but it is not known if patients with TSATs ≤20% experienced a greater benefit on functional capacity.116,117 However, it is noteworthy that intravenous iron exerts similar benefits in hypoferremic patients with other conditions that limit exercise tolerance but are associated with normal left ventricular function (i.e. chronic obstructive pulmonary disease and cancer).92,125,126
In summary, the depletion of iron within skeletal muscle related to changes in inflammation and deconditioning is poised to contribute importantly to exertional dyspnoea and the functional limitation of patients with heart failure, independent of changes in iron homeostasis within erythroblasts or cardiomyocytes. Under these conditions, the superimposition of mild degrees of hypoferraemia—due to absolute or functional iron deficiency—may exert a disproportionate effect on cytosolic iron levels within skeletal myocytes, leading to worsening symptoms and exercise intolerance, which can be improved with intravenous iron. These are the features of Type 2 iron-deficient heart failure (Figure 2, Table 1, and Graphical Abstract).
Type 3 iron-deficient heart failure: disproportionate cardiomyocyte iron depletion with the progression of cardiomyopathy
If inflammation and neurohormonal activation can cause skeletal muscle dysfunction and intracellular iron depletion, thereby increasing the susceptibility of skeletal myocytes to modest hypoferraemia, it is possible that a similar process could take place in cardiomyocytes. In most patients with heart failure and without anaemia, the reduction of circulating iron and TSAT in patients with heart failure without anaemia is modest,9,40,42 and acting alone in a healthy heart, such changes in systemic iron homeostasis are unlikely to impair ATP production by cardiomyocytes, especially since increases in cardiac TfR1 and hepcidin and decreases in ferroportin would be expected to help cardiomyocytes absorb and retain iron if delivery of iron to the heart were impaired.30–33
However, the progression of heart failure may itself cause deficits in intracellular iron homeostasis within cardiomyocytes, the severity of which is independent of abnormalities in systemic iron metabolism.10 Experimentally, prolonged neuroendocrine stimulation causes intracellular iron deficiency and mitochondrial dysfunction in cardiomyocytes,127 and increased plasma norepinephrine and lack of use of beta-blockers are associated with systemic and cardiac iron depletion in patients with heart failure.128–130 Although the expression of hepcidin within cardiomyocytes is elevated in response to cardiac injury and stress,31,131–133 these increases disappear and are reversed, as heart failure evolves and progresses.134 Additional still-undefined mechanisms produce meaningful subcellular iron dysregulation in the failing heart, both clinically and experimentally.135,136
As a result, the left ventricles of patients with heart failure show decreased cardiac iron and reduced mitochondrial activity, which is most marked in those with the most advanced disease, but is unrelated to systemic iron biomarkers.131–134 Myocardial iron declines progressively as patients advance towards end-stage heart failure, independently of TSAT,129,130,134–138 and the failing heart shows decreased TfR1, diminished DMT1 and iron regulatory protein 1, impaired endosomal ferrireduction, and reduced transport of iron into the cytosol, along with decreased hepcidin and increased ferroportin.129,135–141 These changes indicate that both diminished entry into and enhanced egress of iron from cardiomyocytes may drive the development of cardiac cytosolic iron deficiency in severe heart failure. Importantly, this pattern of changes in iron proteins is distinguished by being precisely opposite that seen in cardiomyocytes in response to a systemic or extracellular iron deficiency state3,8,30–32 (TfR1 and hepcidin profiles may differ and may be better aligned with circulating iron biomarkers in patients who do not have advanced heart failure, as reflected by Type 1 iron deficiency heart failure91). Nevertheless, the diminished ATP production in iron-deficient hearts is not related to the presence or absence of anaemia,4 and iron supplementation can prevent the development of ventricular remodelling and cardiomyopathy produced by experimental injury, even in the absence of a systemic iron deficiency state.8,34,134,135,142 Interestingly, intravenous iron may cause repletion of myocardial iron without measurable changes in bone marrow iron.6
Exaggerated vulnerability of advanced cardiomyopathy to modest hypoferraemia
If cytosolic iron levels within failing cardiomyocytes are already compromised, it is reasonable to hypothesize that patients who have severe left ventricular dysfunction may be exceptionally vulnerable to even modest decreases in TSAT, as is seen in patients with functional iron deficiency even without anaemia or with mild absolute iron deficiency. Modest degrees of hypoferraemia might be sufficient to trigger a disproportionate depletion of intracellular iron within cardiomyocytes. We refer to this pathway as Type 3 iron-deficient heart failure (Figure 3, Table 1, and Graphical Abstract).4
Mechanism characteristic of Type 3 iron-deficient heart failure. With the progression of cardiomyopathy and the development of advanced heart failure, cytosolic iron levels in cardiomyocytes are threatened by (i) the suppression of iron entry (as a result of downregulation of transferrin receptor protein 1, divalent metal transporter 1, and other endosomal transport mechanisms) and by (ii) augmentation of iron egress (as a result of decreased cardiac hepcidin and increased ferroportin). The resulting depletion of cytosolic iron enhances the vulnerability of failing cardiomyocytes to modest decreases in serum iron availability. These derangements in cardiomyocyte cytosolic iron can occur without anaemia and without marked decreases in transferrin saturation
Among large-scale clinical trials of iron supplementation, the AFFIRM-AHF trial studied patients with decompensated heart failure, and those with iron deficiency without anaemia (who had lower left ventricular ejection fractions) appeared to respond particularly well to intravenous iron,7 even though they had only a modestly decreased TSAT and higher serum ferritin, findings consistent with a functional iron-deficient state. Furthermore, nonanaemic patients exhibited a more favourable response with respect to the reduction in heart failure outcomes, even though these patients showed minimal erythrocytosis during treatment and received less iron supplementation.7 The lack of change in haemoglobin in the placebo group in AFFIRM-AHF and the absence of an iron-induced erythrocytosis in nonanaemic patients suggest that haemoglobin levels reflected red blood cell mass and were not related to the haemoconcentration and haemodilution that can be seen in acutely decompensated or advanced heart failure.143,144 These observations, taken collectively, support the possibility of a meaningful dissociation between systemic and cardiomyocyte iron homeostasis in certain patients with severe cardiomyopathy, in concert with the evidence from experimental studies.8,31,134,135,139,142
Conclusions and implications for future trials
Based on the totality of evidence, we propose three possible mechanistic pathways of iron deficiency in heart failure (Table 1 and Graphical Abstract). In Type 1, a reduction in serum iron (typically marked) can directly exacerbate cardiac dysfunction, so that circulating iron biomarkers likely inform the status of cytosolic iron within erythroblasts and cardiomyocytes, explaining why TSAT and anaemia predict the reduction in major heart failure outcomes with intravenous iron. In Type 2, the symptoms and exercise intolerance of patients with heart failure are related to abnormalities in skeletal muscle, which are caused by inflammation and deconditioning leading to intracellular iron depletion. These derangements may be exacerbated by modest degrees of hypoferraemia and can improve following intravenous iron, regardless of baseline haemoglobin or changes in haemoglobin. In Type 3, independently of systemic iron homeostasis, the progression of heart failure leads to dysregulation of iron homeostasis within cardiomyocytes, causing cytosolic iron depletion (which can be disproportionately aggravated by low serum iron availability), contributing to worsening contractile function in patients with advanced heart failure (Type 3). In Type 2 and Type 3 iron deficiency, the severity of cytosolic iron depletion cannot reliably be assessed by systemic iron biomarkers, and patients may respond favourably to intravenous iron, even though they have mild hypoferraemia and no anaemia. In all three types of iron deficiency, the decrease in TSAT may be absolute or functional.
It should be noted that this paper describes mechanistic pathways and not mutually exclusive clinical courses. The mechanism that is operative in one pathway may coexist the mechanism that drives a different pathway simultaneously in the same patient, and furthermore, the mechanisms may occur sequentially over time. It is likely that many patients with heart failure have both Type 1 and Type 2 iron deficiency states, and that patients with Type 1 iron deficiency may evolve into Type 3 iron deficiency as heart failure progresses over time. Finally, there is considerable uncertainty as to how these types can be reliably identified in the clinical setting, especially since patients with a low TSAT but a serum ferritin >300 μg/L have not been extensively studied in clinical trials. Conversely, the clinical significance of a serum ferritin in the normal range (20 to 300 μg/L) in the absence of hypoferraemia—if any—needs to be further elucidated.
In general, systemic iron biomarkers are used to identify patients in clinical trials with an iron deficiency state before randomization and to guide repeat dosing with intravenous iron following randomization. Additional doses of intravenous iron are generally withheld if systemic iron biomarkers suggest an iron-replete state or if haemoglobin is in the normal range.50 The validity of this long-term dosing strategy is not known. Yet, it is possible that the greater risk reduction seen in patients with the lowest TSAT was related to greater long-term iron repletion7 rather than to the ability of circulating iron biomarkers to reflect the magnitude of cytosolic iron depletion in cardiomyocytes prior to treatment.
Clinically, evidence from randomized controlled trials supports the identification and treatment of Type 1 and Type 2 iron-deficient heart failure, but additional trials are needed to evaluate the relevance and management of Type 3 iron-deficient heart failure. Because of markedly dysregulated iron homeostasis within stressed cardiomyocytes, iron levels become progressively depleted in failing human hearts as cardiomyopathy advances, and experimentally, iron supplementation can prevent post-injury ventricular remodelling and heart failure, even without systemic iron deficiency.134,142 In patients with acutely decompensated heart failure in the AFFIRM-AHF trial, patients without anaemia (and only modest hypoferraemia) showed large reduction in major heart failure outcomes with intravenous iron supplementation.7 Therefore, in advanced heart failure, modest decreases in serum iron availability may trigger decompensation; however, the critical threshold for values for TSAT that would constitute physiologically meaningful hypoferraemia have yet to be defined. Of note, in the FAIR-HF2 trial,145 post-randomization iron supplementation is maintained even in patients whose systemic iron biomarkers no longer show evidence of an iron deficiency state; thus, the trial is poised to test the hypothesis that heart failure-induced—rather than hypoferraemia-induced—cardiomyocyte iron depletion may represent a therapeutic target.134,142 Patients with severe left ventricular dysfunction (who may have the most marked discordance between systemic and cardiac iron stores) might benefit from ongoing iron supplementation. Trials that are designed to evaluate the efficacy and safety of iron supplementation in patients with advanced heart failure with minimal or no hypoferraemia are needed.
Supplementary data
Supplementary data are not available at European Heart Journal online.
Declarations
Disclosure of Interest
M.P. reports personal fees for consulting from 89bio, AbbVie, Actavis, Altimmune, Alnylam, Amarin, Amgen, Ardelyx, AstraZeneca, Attralus, Biopeutics, Boehringer Ingelheim, Caladrius, Casana, CSL Behring, Cytokinetics, Imara, Lilly, Medtronic, Moderna, Novartis, Pharmacosmos, Reata, Relypsa, and Salamandra, all outside the submitted work. S.D.A. reports grants from Vifor and Abbott Vascular and personal fees for consultancies, trial committee work, and/or lectures from Vifor, Abbott Vascular, Actimed, Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Brahms, Cardiac Dimensions, Cardior, Cordio, CVRx, Cytokinetics, Edwards, Faraday Pharmaceuticals, GlaxoSmithKline, HeartKinetics, Impulse Dynamics, Occlutech, Pfizer, Regeneron, Repairon, Scirent, Sensible Medical, Servier, Vectorious, and V-Wave and is named co-inventor of two patent applications regarding MR-proANP (DE 102007010834 and DE 102007022367), but does not benefit personally from the patents, all outside the submitted work. J.B. reports personal consulting fees from Abbott, American Regent, Amgen, Applied Therapeutic, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol Myers Squibb, Cardiac Dimension, Cardior, CVRx, Cytokinetics, Janssen, Daxor Edwards, Element Science, Eli Lilly, Innolife, Impulse Dynamics, Imbria, Inventiva, Lexicon, LivaNova, Medscape, Medtronic, Merck, Occlutech, Novartis, Novo Nordisk, Pfizer, Pharmacosmos, PharmaIN, Roche, Secretome, Sequana, SQ Innovation, Tenex, Tricoq, and Vifor and honoraria from Novartis, Boehringer Ingelheim-Lilly, AstraZeneca, Impulse Dynamics, and Vifor, all outside the submitted work. R.J.M. reports grants from American Regent, AstraZeneca, Amgen, Bayer, Merck, Novartis, Zoll, and Cytokinetics and personal consulting fees from Pharmacosmos, Vifor, Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Merck, Novartis, Abbott, Medtronic, Zoll, Boston Scientific, Cytokinetics, Respicardia, Roche, Vifor, Sanofi, and Windtree, all outside the submitted work. J.G.F.C. reports grants from Bristol Myers Squibb, British Heart Foundation, Medtronic, Pharma Nord, Vifor, and Pharmacosmos; personal consulting fees from Abbott, Biopeutics, Innolife, NI Medical, Novartis, and Servier; honoraria for committee or advisory boards from Idorsia and Medtronic; honoraria for lectures from AstraZeneca and Boehringer Ingelheim; and stock options or holdings in Heartfelt Limited and Viscardia, all outside the submitted work. P.R.K. reports grants from Pharmacosmos and the British Heart Foundation; personal consulting fees from Amgen, Boehringer Ingelheim, Pharmacosmos, Servier, and CSL Vifor; and honoraria for lectures from AstraZeneca, Bayer, Novartis, Pfizer, Pharmacosmos, CLS Vifor, and Amgen, all outside the submitted work. P.P. reports personal fees for consultancies, trial committee work, and/or lectures from AstraZeneca, Bayer, Boehringer Ingelheim, Pfizer, Vifor Pharma, Amgen, Servier, Novartis, Novo Nordisk, Pharmacosmos, Abbott Vascular, Radcliffe Group, and Charité University, all outside the submitted work.
Data Availability
There are no original data in this manuscript.
Funding
All authors declare no funding for this contribution.
