Estimated prevalence (%) of the disorder
Infections (acute and chronic): viral infections, including HIV; bacterial; parasitic; fungal
Cancer: hematological and solid tumors
Systemic lupus erythematosus
Inflammatory bowel disease (IBD)
Chronic rejection after solid-organ transplantation
Chronic kidney disease and inflammation
Pathogenic mechanisms vary within each class of disorders listed in Table 4.1. However, each of the individual factors to be discussed subsequently plays some role in the eventual cause of anemia. One part of the pathophysiological process reflects one overarching goal : deprive the invading cells of iron, whether these cells are cancer cells or external pathogens. Iron is an important nutrient for proliferation of mammalian cells as well as infectious agents. However, it is the red blood cell which becomes the innocent bystander. To truly understand the processes that lead to anemia, the reader is referred to Chap. 2 on Iron-Deficiency Anemia in which the crucial role of normal iron metabolism in erythropoiesis is described. However, in order to fully understand ACD, the effects of inflammation on response to EPO must be understood as well. Disorders in EPO secretion and action and shortening of red cell lifespan also play a role in the pathogenesis of ACD. The main therapy is treatment of the underlying disorder and red cell transfusions in severe anemia. In more severe (protracted) anemia that leads to impaired quality of life and has an impact on the mortality and survival rate, recombinant erythropoiesis-stimulating agents (ESAs) are used.
Importance of Iron
Adequate hemoglobization of red blood cells, an iron-requiring process, is essential for normal erythropoiesis. Total body iron is 50 mg/kg body weight, or approximately 3500 mg for a 70-kg man . Sixty-five percent this iron is distributed within red cells as hemoglobin (Hb); 10% as myoglobin (Mb), cytochromes, and enzymes; and the remainder in the RES, liver, and bone marrow. To meet the daily requirement of producing 300 billion new erythrocytes, differentiating erythroblasts require approximately 20–30 mg/d iron, most of which is obtained from the recycling of senescent red blood cells (RBCs) by phagocytic macrophages of the reticuloendothelial system (RES) . Heme from these cells is metabolized by heme oxygenase, and the Fe2+ released is sequestered by ferritin .
Only 10% of the dietary iron intake of 15–20 mg/dL is present as relatively bioavailable heme compounds , which are readily absorbed into enterocytes and degraded by heme oxygenase to release Fe2+. The remaining non-heme iron exists in the relatively unavailable Fe3+ state and must be reduced to Fe2+ by ferrireductase in conjunction with ascorbic acid; iron is then transported into the enterocyte by divalent metal transporter (DMT)-1 . Cytosolic iron, whether present in duodenal enterocytes, macrophages, or hepatocytes, moves to circulating transferrin via an exporter, ferroportin (FP)-1 , during which Fe2+ is oxidized to Fe3+, and bound to plasma transferrin . In order to provide sufficient iron for reticulocyte production, transferrin-bound iron must be recycled 6–7 times daily . Iron enters the erythroblast when two transferrin molecules bind TfR-1, which then undergoes endocytosis into a clathrin-coated siderosome . The siderosome is acidified, releasing Fe3+ that is again reduced to Fe2+ by ferrireductase and exported to the cytoplasm via DMT-1. The apotransferrin:TfR-1 complex is recycled to the cell membrane and released into the circulation [8, 11]. Meanwhile, cytosolic iron enters the mitochondrion, where ferrochelatase catalyzes its insertion into protoporphyrin IX to form heme, the critical component of Hb [8, 12].
Two models of iron homeostasis have been described. One model postulates that plasma iron is sensed by duodenal crypt enterocytes via the TfR . When iron is scarce, low cytosolic iron induces the transcription of TfR-1, DMT-1, and FP-1 mRNA, all of which stimulate iron absorption. A second, compatible regulatory model proposes that iron absorption is downregulated by hepcidin , a 25-AA polypeptide produced by hepatocytes when iron is abundant. Hepcidin binds to FP-1 in enterocytes, macrophages, and hepatocytes themselves, promoting the JAK-2-mediated tyrosine phosphorylation , internalization, and degradation of FP-1; thus, hepcidin inhibits both the efflux of iron from the duodenum into the plasma as well as the mobilization of iron from the RES . Hepcidin expression is itself regulated at critical points in the homeostatic loop. Specifically, hepcidin transcription is inhibited by HIF during tissue hypoxia, by soluble hemojuvelin during iron deficiency, and by EPO, GDF-15, and twisted gastrulation (TWSG-1) during erythroblast maturation. Conversely, hepcidin transcription is stimulated by iron-mediated production of bone morphogenetic proteins (BMPs) , lipopolysaccharide (LPS) , and IL-6 in states of systemic inflammation [8, 9, 14, 15].
In summary, iron metabolism is balanced by two regulatory systems. One functions systemically and relies on the hormone hepcidin and the iron exporter ferroportin. The other predominantly controls cellular iron metabolism through iron-regulatory proteins that bind iron-responsive elements in regulated messenger RNAs. To optimize iron delivery to cells, the two systems must “tango” together in a coordinated manner , while at the same time avoiding iron deficits or excess [16, 17]. Tight regulation of iron is necessary because iron is highly toxic and human beings can only excrete small amounts through sweat, skin and enterocyte sloughing, and fecal and menstrual blood loss .
Importance of Erythropoietin
The other important component for red cell production is the effect of the hormone EPO. In adults, 90% of the EPO produced in response to anemia originates from the kidney, specifically from a population of interstitial peritubular fibroblasts with neuron-like morphology (pericytes) in the inner cortex and outer medulla [19, 20]. In adults, the population of interstitial peritubular fibroblasts in the inner cortex and outer medulla [19, 21] regions tends to be especially susceptible to hypoxia [22, 23]. EPO expression in other cell types is normally suppressed by transcription factors that bind the G-A-T-A tetranucleotide sequence in the core promoter region of the gene (GATA transcription factors) . Mild hypoxia, rather than increasing the transcription of EPO mRNA within each cell, actually stimulates the brisk recruitment of previously quiescent clusters of cells within the kidney, each of which then generates EPO at a fixed rate . Moderate hypoxia, however, fuels additional EPO production by the liver, mainly from hepatocytes near the central veins and, to a lesser extent, from stellate or Ito cells [19, 21].
RBC production (erythropoiesis) takes place in the bone marrow. It begins with the EPO-independent differentiation of the multipotent hematopoietic stem cell into multipotential colonies containing erythroid cells, megakaryocytes, neutrophilic and eosinophilic granulocytes, and monocyte-macrophages (CFU-GEMM). Some of these colonies in turn develop into the burst-forming unit-erythroid (BFU-E) . The BFU-E is the first cell type in the erythroid lineage to express the EPO-receptor (EPO-R) , and EPO is required for its survival and subsequent proliferation into several colony-forming units-erythroid (CFU-E) , a process which requires 10–13 days . Of all the erythroid precursors, the CFU-E has the highest membrane density of EPO-R and additionally expresses TfR and GATA-1 . In the presence of EPO, GATA-1 promotes transcription of an anti-apoptotic protein allowing the CFU-E to multiply. Conversely, in the absence of EPO, pro-apoptotic caspases are activated, and the CFU-E dies. If the CFU-E survives, it first differentiates into the proerythroblast, which again requires exposure to EPO to escape apoptosis. The proerythroblast has a large nucleus and expresses numerous membrane adhesion molecules . During maturation, the nucleus condenses and many of these adhesion molecules are lost. Thus in this long almost 2-week period in which stem cells differentiate and proliferate into erythroblast, EPO is the main biological driver for making enough cells and different EPO-dependent signals are used as the cells divide, differentiate, and avoid apoptosis. Very little Hb is actually present within any of these cells.
Terminal erythroid differentiation occurs as the erythroblast progresses through the basophilic, polychromatic, and orthochromatic normoblast stages. Thirty-two daughter cells arise from each erythroblast and ultimately become reticulocytes. During this sequence, the cell rapidly acquires Hb and various proteins that promote membrane elasticity and stability . The orthochromatic normoblast extrudes its pyknotic nucleus, which is ingested by a macrophage, and after 2–3 days becomes a reticulocyte . This enucleated, multi-lobed reticulocyte is transformed into a biconcave erythrocyte during the next 2–3 days, initially in the bone marrow and then in the circulation. At the erythroblast stage, the cells send a signal, erythroferrone, to the liver to suppress hepcidin, thus permitting iron to be absorbed from the gut and released from macrophage stores for incorporation into heme in the developing RBCs .
Although the pathophysiological processes operative in ACD are still incompletely understood, they are thought to be mediated through the actions of various cytokines especially tumor necrosis factor (TNF) , interleukins (IL)-1 and −6 , and interferon (IFN) . These cytokines, as well as the acute phase protein, hepcidin released by the liver, inhibit iron release from the marrow macrophages for incorporation into heme by the post erythroblast stage of erythropoiesis. The cytokines also directly induce the modulation of translation/transcriptions of genes involved in iron homeostasis, either directly or via production of labile radicals .
The diagnosis of ACD requires a good knowledge of the processes that affect production of the key molecules and of course interpretation of the results of the circulating iron status parameters as well as ferritin and hepcidin assays. An important step in ACD diagnosis is distinguishing ACD from iron-deficiency anemia . A diagnosis of ACD does not suffice for anemia observed in all chronically ill patients , because of the temptation to label all ill patients in whom a definite cause of anemia cannot be discerned as ACD. A diligent effort should be made to unravel the underlying cause of anemia in all patients as well as to rule out iron deficiency. The success of management of either condition is therefore reliant on making correct diagnosis , as each would require different treatment protocols [31, 32]. In addition to treating the underlying condition, targeting other inflammatory pathways may be beneficial to achieve rapid resolution of anemia [33, 34].
Pathogenesis of Anemia of Chronic Disease
The pathogenic mechanisms vary  as each of the pathways depends largely on the etiological process ongoing in the patient (see Table 4.1). However, each of the individual factors plays some role in the eventual cause of anemia. These processes include bone marrow invasion by tumors or infective agents, alteration of iron metabolism and diversion of body iron, hemophagocytosis, reduction in erythropoiesis secretion, and diminished response to EPO stimulation [1, 6, 7]. In addition, red blood cell survival in the circulation decreases, magnifying the effect of underproduction of red blood cells.
A list of the major cytokines involved in ACD and their mechanisms of effecting these actions is given in Table 4.2.
Effects of cytokines on iron metabolism and erythropoiesis
Effects on iron or erythropoiesis
Inhibits EPO production
Stimulates ferritin synthesis
Enhances degradation and phagocytosis of effete red cells
Direct inhibitor of erythropoiesis
Inhibits production of EPO
Increases intracellular iron (via DMT1) and inhibits ferroportin
Increased NO production and iNOS mRNA expression
Increases iron uptake via DMT-1 activation
Reduces transferrin receptor by decreasing TfR RNA expression
Downregulates expression of SLC-4a1 in erythroid precursors
Interleukin 4 and 10
Increases ferritin via action on IRE/IRP
Influences hepcidin production
Iron Dysregulation (Reticuloendothelial Iron Blockade
This is a major causal pathway in ACD, eventuating in the presence of a hypochromic microcytic red cell picture similar to what is found in iron deficiency. Microbial invasion, malignancies and autoimmune disorders cause activation of CD3 T lymphocytes and macrophages which release cytokines including IFN-γ (from T cells), tumor necrosis factor- (TNF-)α, interleukin- (IL-)1, and IL-6 (from monocytes) [35, 36]. TNF-α is secreted also by neutrophils, macrophages, T-cells and natural killer cells in response to stimulation by IL-2, granulocyte-monocyte colony stimulating factor (GM-CSF) , and platelet associated factor (PAF) , and is inhibited by IL-6, transforming growth factor and prostaglandin E2 . Bacterial LPS and IL-6 induce the hepatic cells to release hepcidin , which enhances breakdown of ferroportin, leading to blockade of the duodenal iron transfer [13, 39, 40]. Both molecules also upregulate the expression of divalent metal transporter (DMT-1) , a trans-membrane protein involved in the uptake of iron by enterocytes [40, 41] and by macrophages . They also inhibit the expression of ferroportin 1, the only iron exporting system in mammalian cells , which reduces export of iron from the macrophages to the growing erythroid precursors; this action is influenced by hepcidin (vide infra) [14, 42, 43].
As part of the “acute phase reaction” IFN-γ stimulates ferritin transcription while simultaneously inhibiting transferrin receptor (TfR) mRNA expression via an IRE/IRP-independent process [44, 45]. It also increases expression of DMT-1, which is involved in the active transport of ferrous molecules from the lumen to the duodenal endothelial cell cytoplasm . As a result of these effects, absorbed iron is retained in the intestinal endothelial cells which are eventually shed while serum ferritin levels remain elevated despite an apparent scarcity of iron. Within the RES and especially within marrow macrophages cytoplasmic iron is retained whereas the post-normoblast proliferating erythroid precursors remain iron-deprived.
Hepcidin , as described earlier, is a molecule produced by the liver that inhibits iron absorption from the duodenum as well as iron release by the bone marrow macrophages [38, 42, 44, 45]. This action to regulate intestinal iron absorption as well as plasma and tissue iron concentration results from its ability to bind to and lead to internal degradation of its receptor—ferroportin [42, 45, 47–51]. Hepcidin, by simultaneously affecting both the influx of iron into plasma from duodenal enterocytes and efflux of iron into macrophages, lowers serum transferrin-bound iron more than the effect of either action alone. This response can occur quite quickly. A mechanism of regulation termed the “mucosal block” phenomenon is the ability of an initial dose of ingested iron to block absorption of a second dose given 2–4 h later. Studies indicate that mucosal block is a fast-response endocytic mechanism destined to decrease intestinal iron absorption during a high ingest of iron and is mediated by changes in serum iron signals via hepcidin .
IL-22 is yet another cytokine that also influences hepcidin liver production. Injection of mice with exogenous mouse IgG1 Fc fused to the N-terminus of mouse IL-22 (Fc-IL-22), an IL-22R agonist with prolonged and enhanced functional potency, induced hepcidin production . This response was independent of IL-6 and was attenuated in the absence of the IL-22R-associated signaling kinase, Tyk2 . Antibody-mediated blockade of hepcidin partially reversed the effects on iron biology caused by IL-22R stimulation. Taken together, these data suggest that exogenous IL-22 also regulates hepcidin production to physiologically influence iron usage. Further support for the physiological role of IL-22 comes from studies of IL-22 knockout mice subjected to an acute inflammatory stimulus via administration of LPS. In the absence of IL-22, there was a response of hepcidin (probably via IL-6) but the hypoferremic response to LPS was blunted .
In summary, hepcidin downregulates intestinal trans-epithelial iron transport by causing an ubiquitin-dependent proteosome degradation of intestinal DMT-1 . The effect on DMT-1 and on ferroportin occurs by internalizing and degrading these membrane receptors and thus inhibiting release of iron by the macrophages while reducing absorption of iron by the intestinal mucosal cells [1, 55, 56]. This is actually a defense mechanism that the body adopts to produce a hypoferremic state that denies bacterial and cancer cells their much-needed iron. This, as well as the pathway involving IL-6, and to a lesser extent IL-22, are considered to be the major contributors for the development of ACD. In patients on hemodialysis (HD) , a significant reduction of hepcidin levels follow administration of ESAs but not intravenous iron . This effect of ESA indicates that despite inflammation, the feedback loop between increased bone marrow erythropoiesis and decreased liver hepcidin production remains intact.
Abnormal hepcidin metabolism is found in virtually all inflammatory conditions. However, some conditions such as uremia or obesity are not usually considered to be inflammatory in nature. In obesity, chronic low-grade inflammation exists and enhances hepcidin production. Adipose tissue is known to secret interleukin-6 and leptin that triggers hepcidin production . It was found that adipose tissue also expresses hepcidin and hemojuvelin, a regulator of hepcidin production . These recent findings suggest that adipose tissue may have an important role in erythropoiesis particularly in obesity that is still poorly clarified.
As important as iron restriction is in the genesis of ACD, other mechanisms contribute as well since anemia of inflammation is usually normocytic and normochromic, while diseases associated with overexpression of hepcidin, alone, are often microcytic and hypochromic. These differences in erythrocyte parameters suggest that anemia in many inflammatory states is not fully explained by hepcidin-mediated iron sequestration . Studies suggest that chronic anemia associated with inflammation may benefit from interventions protecting the number of erythrocytes produced in addition to anti-hepcidin interventions aimed at enhancing iron availability.
Red blood cell formation is inhibited by several cytokines including IFN-γ, transforming growth factor (TGF-β), and TNF-α. The action of TGF-β and TNF-α is mediated via the p38 mitogen-activated protein kinase (MAPK) pathway , whereas IFN-γ acts via the Janus-associated kinase (JAK/STAT) pathway . Activation of this pathway leads to production of intracellular factors which enhance apoptosis eventually resulting in myelosuppression. Other cytokines affecting red cell development include IL-1 and IL-6.
IFN-γ suppresses red cell development, IFN-γ induces apoptosis in erythroid precursors by increasing nitric oxide production and inducible nitric oxide synthase mRNA production . IFN-γ, and to a lesser extent IFN-α and β, have been observed to induce apoptosis of the erythroid burst-forming (BFU-E) and colony-forming units (CFU-E) . This action is mediated in part via the action of ceramide and due to reduction in EPO receptors in precursor erythroid cells . Other modalities of IFN action include reduction in quantity and activity of EPO as well as reduced expression of other growth factors such as stem cell factor.
In an experimental model of ACD, anemia was associated with a 50% reduction in EPO-stimulated differentiation of EPO-R+ cells (pre erythroblast) to erythroblasts . This suppression required accessory cells, including antigen-presenting cells, which activated other cells to produce pro-inflammatory cytokines. In vitro neutralization of IFN-γ, but not IL-12, TNF-α, IFN-α, IL-1α, or IL-1β, abrogated the erythropoietic suppression induced by inflammation. The anemia observed was also associated with reduced RBC survival in vivo, as demonstrated by a seven- to eightfold higher turnover of biotinylated RBCs as compared to that in control animals . In vivo IFN-γ neutralization confirmed that IFN-γ contributed to erythropoietic suppression but not to reduced RBC survival. In a totally different model of anemia, that of acute blood-stage malaria, a role for IFN-γ and IL-4 were found in STAT6-induced erythropoietic suppression .
A role for IFN-γ had been proposed in the macrophage activation syndrome (MAS) , a devastating cytokine storm syndrome complicating many inflammatory diseases, characterized by fever, pancytopenia, and systemic inflammation. Murine models of MAS indicated that IFN-γ was the driving stimulus for hemophagocytosis and immunopathology. A study investigated the inflammatory contributors to a murine model of Toll-like receptor 9 (TLR-9)-induced fulminant MAS produced by IL-10 receptor blocking antibody and a TLR-9 agonist. IFN-γ-knockout mice developed immunopathology and hemophagocytosis comparable to that seen in wild-type mice . These results showed that both fulminant MAS and hemophagocytosis can arise independently of IFN-γ, IL-12, or type I IFNs and that IFN-γ-mediated dyserythropoiesis, not hemophagocytosis, is the dominant cause of anemia in this model of MAS.
TNF-α levels are often increased in ACD. In rheumatoid arthritis (RA), inflammatory cytokines, particularly TNF-α, IL-1, and IL-6, are thought to contribute to the pathogenesis of ACD. The role of TNF-γ was examined in vivo using a chimeric monoclonal antibody to block its action. In RA patients with ACD, administration of the blocking antibody led to a dose-dependent increase in Hb levels compared to placebo and these changes were accompanied by a reduction in both EPO and IL-6 levels , supporting the role of TNF-α in the causation of ACD. TNF-α appears to act directly on bone marrow red cell precursors rather than on suppression of EPO production. In addition to systemic levels, in patients with ACD, increased local production of cytokines including TNF-α by marrow T-lymphocytes may also occur .
IL-6 is one of the more important cytokines mediating the pathogenesis of ACD [36, 56]. It is a potent inhibitor of TNF-α and induces the transcription of ferritin that leads to increased retention and iron storage within the reticuloendothelial system. As stated previously, IL-6 inhibits erythropoiesis through the inhibition of absorption and uptake of iron . Serum IL-6 is elevated in ACD and it correlated well with parameters of disease activity such as erythrocyte sedimentation rate and C-reactive protein (CRP) . Previous studies in chronically ill animal species, however, had suggested the existence of other pathways of anemia induction in which suppression of TNF elevation did not prevent occurrence of anemia [33, 64, 65], and that the IL-6 pathways for anemia were not hepcidin-dependent , thus suggesting either direct inhibition of erythropoiesis or the existence of other yet to be established pathways. In patients with RA, growth of the erythroid colony growth burst-forming units of erythroblasts (BFU-E) is impaired in those with ACD but not in non-anemic RA controls . Furthermore, studies indicate that local production of pro-inflammatory cytokines in the bone marrow, not merely circulating IL-6 and TNF-α may be associated with the development of ACD in RA . IL-6 downregulates the expression of the SLC4a1 gene in late erythroid precursors, and thereby reduces Hb synthesis . It also reduces the mitochondrial mass and function in the developing erythroid progenitors .
Patients surviving sepsis develop persistent anemia but the molecular mechanisms have been unknown until recently. The role of a ubiquitous nuclear protein , high mobility group box 1 (HMGB1) , that is released by activated macrophages/monocytes, and functions as a late mediator of sepsis has gotten recent attention . Circulating HMGB1 levels are elevated in a delayed fashion (after 16–32 h) in septic animals. Administration of recombinant HMGB1 to mice recapitulates many clinical signs of sepsis, including fever, derangement of intestinal barrier function, lung injury, and lethal multiple organ failure . Administration of anti-HMGB1 antibodies or inhibitors (e.g., ethyl pyruvate, nicotine, stearoyl lysophosphatidylcholine, and Chinese herbs such as Angelica sinensis) protects mice against lethal endotoxemia, and rescues mice from lethal experimental sepsis even when the first doses are given 24 h after onset of sepsis . In mice that survive polymicrobial gram-negative sepsis, a hypochromic, microcytic anemia with reticulocytosis develops. The bone marrow of sepsis survivors accumulates polychromatophilic and orthochromatic erythroblasts. Circulating TNF-α and IL-6 are elevated for 5 days after the onset of sepsis, and serum HMGB1 levels are increased from day 7 until at least day 28. Administration of recombinant HMGB1 to healthy mice produces anemia which is ameliorated by administration of anti-HMGB1 monoclonal antibodies after onset of sepsis (hematocrit 48.5 ± 9.0% vs. 37.4 ± 6.1%, p < 0.01, Hb 14.0 ± 1.7 g/dL vs. 11.7 ± 1.2 g/dL, p < 0.01). Together, these results indicate that HMGB1 mediates anemia by interfering with erythropoiesis, suggesting a potential therapeutic strategy for anemia in sepsis.
Activin B production by hepatic cells is markedly increased during inflammation. Activin B binds to the bone morphogenetic protein (BMP) receptor type 1; receptor activation acting via the Smad and JAK-STAT transmembrane proteins to cause upregulation of hepcidin expression .
Both IL-6 and hepatic hepcidin expression have been found to be significantly increased in various malignancies . Thus the iron sequestration from hepcidin, which inhibits the export of iron from the enterocytes , hepatocytes, and marrow macrophages [30, 55], is an important component in the anemia associated with cancer.
Diminished Response to Erythropoietin
As described under erythropoiesis, in some cases of ACD, the erythropoietic response (Hb achieved to exogenous administered EPO or ESA levels) is not commensurate to the degree of anemia; this phenomenon is termed “blunted EPO response, ESA hyporesponse or resistance,” and was first observed in sickle cell patients with chronic kidney disease (CKD) , where ESAs even at very high dose did not correct the anemia. Similar but less severe hyporesponse occurs in those with thalassemia minor. In both, the hyporesponse may be contributed to by hemolysis. An association between EPO-resistant anemia and inflammation  as manifested by increased levels of inflammation markers , such as CRP, IL-6, IFN-γ, and TNF-α in patients with CKD has also been noted.
Hypoferremia, Reduced Erythrocyte Survival, and Hypoxia
IFN-γ and bacterial LPS upregulate the expression of DMT-1 in a dose-dependent manner , thereby enhancing the uptake of unbound iron into the enterocytes and the monocyte/macrophages. Within the monocytes, iron retention occurs as expression of ferroportin mRNA is downregulated by hepcidin. In chronically ill patients, higher levels of both TNF-α and IL-6 correlate with lower levels of serum iron , thus creating a prevailing atmosphere of hypoferremia . At the same time that iron is limited for Hb synthesis within erythrocyte precursors, the rate of erythrophagocytosis increases. The latter is a process designed to remove senescent and damaged red cells in normal situations; in ACD cellular damage is caused by cytokines, endotoxins and reactive oxygen species. Some animal experiments have revealed that sublethal doses of TNF-α may cause phagocytosis of erythrocytes by macrophages. TNF-α rapidly primes human monocytes for enhanced release of O-(2) and erythrophagocytosis and suggest that TNF-α activates monocytes through autocrine or paracrine mechanisms at the inflammatory sites inasmuch as TNF-α is primarily produced by activated monocytes/macrophages [72–74].
Red blood cells (RBCs) often have a short circulating survival in inflammatory conditions, especially in those on HD since ongoing blood losses in the dialysis circuit account to the yearly loss of approximately 5–6 L of blood at a hematocrit of 33%. Fluctuations in EPO levels from exogenous ESAs contribute to shortened RBC lifespan because a decline in the level of EPO triggers the preferential destruction of newly formed RBC, a process termed neocytolysis . Hypochromic RBCs are vulnerable to more rapid turnover in all forms of IDA because iron deficiency increases exposure of the phagocytic signaling molecule phosphatidylserine, loss of deformability , and increased oxidative stress . Additionally, IFN-γ further drives the development of anemia by inhibiting not only erythroid colony growth, but also shortening the lifespan of erythrocytes via increased turnover in the spleen .
Another mechanism for reduced survival of RBCs involves effects of hypoxia per se. Hypoxia leads to increased transcription of hepcidin mRNA . This process is thought to be mediated through platelet-derived growth factor (PDGF) . Furthermore, the production of free oxygen radicals in inflammation causes release of pro-inflammatory cytokines as discussed previously which leads to increased hepcidin. Other pro-inflammatory cytokines like IFN-gamma cause increased expression of inducible nitric oxide mRNA and subsequent production of NO . This in conjunction with the production of superoxide in inflammation may cause reversal of the effects of hypoxia on hepcidin production . This induces nitric oxide-mediated apoptosis of red cell precursors and thus worsens anemia.
Specific Inflammatory Entities
Chronic inflammation is a common feature of end-stage renal disease (ESRD) that is gaining increasing attention as a major cause of morbidity and mortality . Hyporesponse of some degree is most commonly seen in those with CKD, particularly with advanced disease. Even in patients not on dialysis, cytokine levels are often increased. In one study of 50 anemic patients, 23 received ESA treatments. Levels of TNF-α were found to be significantly higher and serum albumin significantly lower with higher IL-6 and IL-8 in anemic compared to non-anemic patients. Further analysis by multiple logistic regression found that anemic patients treated with ESAs had significantly higher odds for being in the upper two quartiles for IL-6, IL-8, and TNF-α compared to non-anemic patients .
Blunted erythropoiesis to exogenous ESA can be partly explained by the fact that the cytokines, bacterial LPS and IFN-γ , induce formation of nitric oxide (NO) and oxygen-free radicals, which directly inhibit expression of EPO in vitro. These reactive oxygen species (ROS) are thought to inhibit the EPO-inducing transcription factors as well as possibly damage EPO-producing cells. Silymarin , which modulates immune cells by inhibiting prostaglandin and prostacyclins production as well as neutrophil and monocyte activation and mobilization, has the capacity to reverse this trend .
Recently our understanding of iron metabolism has indicated that “iron” itself can be a source of oxidative stress. As discussed earlier, living organisms have evolved sophisticated mechanisms to maintain appropriate iron levels within cells and within their body. Labile plasma iron (LPI) represents a component of non-transferrin-bound iron (NTBI) that is both redox active and chelatable, capable of permeating into organs and inducing tissue iron overload with ROS. HD patients are particularly susceptible to the effects of parenteral iron since they get intravenous iron as part of repletion or maintenance iron therapy to prevent hyporesponse to ESA . The LPI measures the iron-specific capacity of a given sample to produce reactive oxygen species; a test result of >0.6 units of LPI in a sample of blood indicates a potential for iron-mediated production of ROS. HD patients with LPI units ≥0.6 have higher serum iron, ESA dose, ferritin, high-sensitivity CRP (hsCRP), hepcidin, and lower hemojuvelin levels. In these HD patients , NTBI correlated with direct markers of inflammation, hsCRP (r = 0.37, p < 0.01), IL-6 (r = 0.43, p < 0.001), and with ferritin (r = 0.41, p < 0.001) .
In normal adults, weekly production of endogenous EPO is ~700 IU which can increase acutely with high altitude exposure, 1.8-fold (at 3000 m) and 3.0-fold (at 4000 m) . Uremia is characterized by an inflammatory state . Not surprisingly then, in adults with advanced CKD/ESRD, exogenous EPO dose varies from 1000 to more than 40,000 IU/week when administered thrice weekly, yet normal hematocrits are not attained . Attaining the latter requires a further two- to threefold increase in EPO dosage . The frequency of moderate anemia (Hb < 12 g/dL in women and <13 g/dL, WHO definition)  increases with severity of kidney disease [90, 91]. The frequency and dose of exogenous ESA also increase with CKD severity  although the starting doses and maintenance doses of ESA are 40–50% lower than those used a decade ago in response to the alert that higher doses producing harm [93, 94].
The source of inflammation is multiple. Currently, a basal level of inflammation is believed to originate primarily from abnormal bacterial function in the gut . Intestinal dysbiosis , alteration in barrier function, and bacterial translocation seem to account for CKD-related systemic inflammation . CKD in animals is manifested by systemic inflammation, including increased plasma levels of pentraxin-2 and activated antigen-presenting cells, CD4 and CD8 T cells, and Th17- or IFN-γ-producing T cells in the spleen as well as regulatory T-cell suppression. CKD-related systemic inflammation in these animals is associated with intestinal dysbiosis of proteobacterial blooms, translocation of living bacteria across the intestinal barrier into the liver, and increased serum levels of bacterial endotoxin . By fermenting undigested products that reach the colon, the intestinal microflora produce indoles, phenols and amines, among others, that are absorbed by the host, accumulate in CKD and have harmful effects. These gut-derived uremic toxins and the increased permeability of the intestinal barrier in CKD are associated with increased inflammation and oxidative stress and may be involved in various CKD-related complications, including cardiovascular disease, anemia, mineral metabolism disorders, or the progression of CKD . Eradication of facultative anaerobic microbiota with antibiotics prevented bacterial translocation, significantly reduced serum endotoxin levels, and fully reversed all markers of systemic inflammation to the level of nonuremic controls .
Much of the enteric effect in uremia is mediated by the gasotransmitter hydrogen sulfide (H2S), which via adenosine monophosphate-activated kinase (AMPK) normally suppresses the inflammatory activation of signal transducer and activator of transcription 3 (STAT3) and thereby hepcidin via AMPK. Pharmacological and genetic activation of AMPK ameliorated hepcidin production, corrected iron dysregulation, and relieved hypoferremia and anemia in both acute and chronic models of inflammation . AMPK suppresses STAT3/hepcidin activation by promoting proteasome-mediated JAK2 degradation, which is dependent on the intact function of suppressor of cytokine signaling 1 (SOCS1) and increased interactions between SOCS1 and JAK2. Importantly, metformin , an AMPK activator, is associated with decreased serum hepcidin content and anemia morbidity in Chinese type 2 diabetes mellitus patients  and perhaps should be carefully investigated in CKD in those with inflammatory component to their anemia.
Although “general state of uremic inflammation in CKD” with effects on muscle (via toll-like receptors)  affects symptomatology and quality of life, the inflammation associated with microbial infection in dialysis patients  is of special importance since CKD patients are more susceptible to such than those without kidney disease . As part of the host-defense system , macrophages generate DMT-1 and neutrophils release apolactoferrin and neutrophil gelatinase-associated lipocalin (NGAL) , all of which remove iron from the circulation and decrease its availability to iron-dependent microorganisms . NGAL is a 25 kDa protein which, similarly to hepcidin, sequesters iron during the acute phase response and binds siderophores, high-affinity iron chelators produced by bacteria . In addition, cytokines such as IL-1β and TNF-α also stimulate the translation of presynthesized ferritin subunit transcripts during the acute phase response . Presumably, this ferritin-mediated iron trapping is protective in states of acute inflammation, such as bacterial infection, but maladaptive in states of chronic inflammation, such as CKD, congestive heart failure (CHF) , and anemia related to inflammation.
A contributing effect to diminished response in HD patients is the bacteriological quality of the water used to create the dialysate. The switch from conventional to online-produced ultrapure dialysate resulted in a lower bacterial contamination with a significant decrease of CRP and IL-6 blood levels . These reductions are accompanied by a significant and sustained reduction of the exogenous EPO dosage required to correct anemia. Using multiple regression analysis , IL-6 levels have a strong predictive value for exogenous EPO dosage . HD patients who are exposed to endotoxin, impure dialysate, and bioincompatible membranes may develop refractory anemia.
In advanced CKD, Hb levels correlate inversely with mortality particularly in the presence of hyporesponsiveness or refractiveness to ESA therapy . Achieved Hb is a surrogate for outcomes with those less than 11 g/dL faring less well once they reach dialysis dependency . Hyporesponse to ESA is also costly. The average per patient cost of anemia in CKD management increases fivefold when comparing patients with Hb >12 g/dL to those patients with Hb <10 g/dL . Additional costs in the latter are driven by comorbidities and the heavier resource utilization by such patients.
Periodontal disease is rampant among dialysis patients and this state produces higher erythrocyte sedimentation rate (ESR) and higher CRP levels which associate with lower Hb levels refractory to dose escalation of ESA. Again this relative hyporesponse and the inflammatory parameters improve with dental treatment of disease . HD patients who are underdialyzed (Kt/V < 1.3) show an association between “dialysis adequacy” and Hb, suggesting that the elimination of uremic toxins is required to sustain adequate erythropoiesis [108, 109].
Inflammation has recently been associated with atherosclerosis and malnutrition in ESRD, and this link has led to the development of the malnutrition, inflammation, atherosclerosis (MIA) hypothesis . This describes a syndrome whereby raised levels of pro-inflammatory cytokines (such as IL-1, IL-6, and TNF-α) are a common link between malnutrition, inflammation, and atherosclerosis . Several inflammatory proteins as well as the inflammatory cytokine, especially IL-6 have been linked with this diminishing response to exogenous EPO. Anemia appears to be an important element linking elevated cytokine levels with poor patient outcomes .
Several mechanisms for cytokine-induced anemia have been proposed, including intestinal bleeding, impaired iron metabolism, and suppression of bone marrow erythropoiesis and EPO production. These latter effects suggest that pro-inflammatory cytokines may be an important cause of lack of response to recombinant human EPO therapy. Statistically significant differences were found between responders and nonresponders to recombinant EPO therapy for total lymphocyte and CD4+ T-lymphocyte counts, albumin (lower in nonresponders) and CRP (higher in nonresponders) levels . Such EPO-resistant anemia is associated with the increased risk of ESRD, cardiovascular events, and death [112, 113].
The issue has been what is a safe dose of exogenous ESA to administer and to what Hb level. Randomized clinical trials clearly showed there was a danger in taking all comers to higher Hb levels and all post hoc analyses of these trials identified hyporesponsiveness as the culprit. Now some additional analyses point to an additional effect of high ESA dose independent of “just” hyporesponsiveness.
Cancer: Bone Marrow Infiltration
Anemia secondary to cancers occur via three basic mechanisms: (1) reduced production of red cells (either by tumor invasions, effect of cytotoxic drugs, suboptimal nutrition or cytokine-based inhibition); (2) increased red cell loss (hemolysis or hemorrhage); and (3) miscellaneous etiologies including nutritional deficiencies of Vitamin B12 and folate . In most cases these mechanisms overlap; however, the major mechanism is cancer-driven inflammation . Marrow invasion by malignant cells leading to physical obstruction and destruction of the bone marrow micro-environment seems to ultimately occur in most cancers. However, in some cases of malignancies, significant anemia is observed in the absence of marrow invasion or scarcity of vital nutrients. This signifies that other pathways may be important in the pathogenic processes leading to anemia in cancers. Secretion by the tumor of cyclooxygenase (COX- 2) , as well as of vascular endothelial growth factor (VEGF) , GM-CSF, IL-6, and TNF-α, which lead to cancer cachexia and anemia , may all play a role. Celecoxib , a COX-2 inhibitor, has also been noted to counteract the anemia and cachexia associated with ACD . The point of action of several cytokines involved in ACD is shown in Table 4.2.
Growth differentiation factor 15 (GDF-15) , a member of the transforming growth factor beta super family , is an inhibitor of leukocyte integrin. GDF-15 is strongly upregulated by stimuli that deplete cells of iron and this response is specifically antagonized by the reprovision of iron. GDF-15 exhibits greater sensitivity to iron depletion than hypoxia, and responses to hypoxia and iron depletion are independent of HIF and IRP activation, suggesting a novel mechanism of regulation . GDF-15 has been observed to have a direct relationship with serum hepcidin levels in multiple myeloma cancer patients with anemia . GDF-15, interleukin-6, and EPO in multiple myeloma patients all increase significantly when these patients are anemic with levels decreasing markedly following successful chemotherapy .
Hyporesponsiveness to ESAs is a major limitation to the treatment of cancer-related anemia. Enhanced GDF-15 levels contribute to cancer progression and metastasis. Paradoxically, GDF-15 has been found to suppress hepcidin expression  and is further associated with angiogenesis, disease progression, and hematopoiesis [118, 119]. Tumor progression in turn results in even more GDF-15 secretion, which, by downregulating hepcidin expression, results in iron overload in some cancer patients, a phenomenon also found in some patients with sideropenic anemia due to chronic blood loss. This has been proposed to be a major mechanism of anemia in cancer-driven inflammation, as the serum level of GDF-15 correlates to a large extent with the degree of anemia in cancer patients [118, 119]. In the case of infective agents like Malaria and HIV, toxic products of these parasites directly suppress erythropoiesis. These organisms and malignant cells also competitively deprive the erythroid precursors of available iron. The invading microbial cells require iron as an important component of several iron-containing enzymes needed for protein synthesis and proliferation.
Decompensated CHF in CKD patients is associated with increased levels of IL-1, IL-6, IL-18, TNF-α, endotoxins, aldosterone, angiotensin II, soluble adhesion molecules, and the soluble receptors TNFR-1, TNFR-2, IL-6R, and gp130, as well as volume overload, all of which contribute to an “inflammatory” state and worsening anemia [111, 113]. Even without CKD, cytokines contribute to the anemia present in CHF . In models of ischemia–reperfusion injury , pathological changes in the myocardial tissue were associated with increased expression levels of TNF-α, IL-6, and IL-1β in the myocardium, and with increased serum levels of these mediators .
Mechanical circulatory assist devices (MCADs) are increasingly utilized independently of cardiac transplantation in the management of heart failure. MCAD support has been correlated with elevated plasma levels of inflammatory cytokines TNF-α, IL-1β, and IL-6, which have separately been found to inhibit EPO-induced erythrocyte (RBC) maturation. Previous analysis of hematological parameters for MCAD-supported patients concluded that an amplified inflammatory response impedes RBC proliferation and recovery from hemolytic anemia. In a study of 78 MCAD-supported patients implanted for greater than 30 days , Hb, RBC distribution width (RDW) , mean corpuscular volume (MCV) , and cardiac index were retrospectively analyzed. Hb, a conventional marker for anemia, declined with MCAD placement and remains below the clinically defined, minimum normal value whereas in an inverse fashion, the RDW rose above maximum normal measure, signifying an increased fraction of juvenile RBCs . Thus, a response of erythropoiesis does occurs in reaction to the onset of anemia manifested by an increased production of immature RBCs, but the patient’s inflammatory cytokine response to the implanted device does not enable full compensation of the MCAD-induced anemia, perhaps by inhibition of full EPO effects.
Idiopathic pulmonary fibrosis (IPF) is a chronic inflammatory process characterized by severe derangement of gas exchange in the advanced stages of disease. However, erythrocytosis is infrequent in IPF, unlike in other pulmonary diseases producing hypoxemia. In a small study, nine patients (six men and three women) with IPF and profound hypoxemia (pO2 < 65 mmHg) were sex- and age-matched to 34 healthy volunteers . Hb was comparable between the two studied groups. By contrast, serum TNF-α, IL-6, and IL-8 values were significantly higher in patients with IPF. Sera from IPF patients induced a significant growth inhibition of BFU-E arising from mononuclear cells of either patients or control subjects . Overall, the findings suggested that there are an increased number of primitive erythroid progenitors in IPF. These, however, fail to proliferate and differentiate in vivo, suggesting ineffective erythropoiesis, and consequently Hb levels do not rise in proportion to the severity of hypoxemia. Perhaps, cytokines released from alveolar macrophages have not only local but also systemic effects, since the serum of these patients directly suppressed erythropoiesis . Whether the suboptimal erythropoietic response to hypoxia is entirely attributed to this suppression is unknown since several other factors could synergistically or additively interfere with erythropoiesis.
Erythroid colony growth, using BFU-E as a parameter, is impaired in anemic RA patients but not in nonanemic RA controls . Levels of IL-6 and TNF-α are also significantly higher in the supernatant of bone marrow cultures of RA patients with ACD compared to controls indicating that local pro-inflammatory cytokine production in the bone marrow may be as important as systemic levels in the development of ACD in RA . Both systemic and local cytokines may contribute to hyporesponse to EPO since systemic levels of EPO may be “appropriately” elevated for the degree of anemia .
Inflammatory Bowel Disease (IBD
Anemia is a frequent complication in inflammatory bowel disease (IBD) and severely impairs the quality of life of affected patients. The etiology of anemia in IBD patients often involves a combination of iron deficiency (ID) and ACD with elements of other nutritional deficiencies . Despite guidelines recommending screening for and treatment of anemia in IBD patients, current data suggest that anemia remains underdiagnosed and undertreated. Surprisingly, anemia was not diagnosed even though the median Hb value was <11 g/dL in both ulcerative colitis (UC) and Crohn’s disease (CD) in one study . This study showed that lab results allowed further differentiation of the type of anemia in 70% of anemic patients. At the time of first diagnosis, an iron-deficiency anemia was diagnosed in 26 of 68 patients with anemia [20 CD, 4 UC, and 2 indeterminate colitis (IC)], but only nine patients received subsequent iron therapy . After 1 year, 27 patients were identified to have an iron-deficiency anemia (19 CD, 8 UC), 20 of them were treated with iron (71.4%). Of nine patients with proven iron-deficiency anemia at time of first diagnosis and subsequent administration of iron, five (55.5%) continued to have iron-deficiency anemia despite treatment for 1 year. In total, 38 patients (54.3%) did not receive any iron substitution at all despite proven iron-deficiency anemia, and only 13 patients of 74 (17.6%) patients were treated with intravenous iron . Even patients with diagnosed iron-deficiency anemia were infrequently and inconsequently treated with iron preparations, despite the high impact on the quality of life .
Besides basic laboratory parameters (see section “Laboratory Diagnosis”), the concentration of soluble TfR (sTfR) and novel parameters such as the sTfR/log ferritin index can help to differentiate ID from ACD . Once identified , causes of anemia should be treated accordingly.
Anemia is one of the characteristics of the frailty phenotype and is often observed in elderly patients. In the elderly, anemia is usually of multifactorial origin, including chronic inflammation, CKD, nutrient deficiencies, and iron deficiency (approximately two-thirds of all cases), but in some cases no identifiable cause is found. In the elderly, the classic diagnosis of anemia based on the mean corpuscular volume associated with a low Hb level might be inaccurate. In frail elderly patients, all investigations should be carefully considered, invasive examinations undertaken only where justified to uncover the underlying cause, and treat it whenever possible . The aging process itself might be an intrinsic factor in the development of anemia, possibly through the age-related dysregulation of certain pro-inflammatory cytokines such as IL-6 which then directly inhibits EPO production or possibly decreases response by an interaction with the EPO receptor . In such patients, the anemia is often mild, with a Hb level > 10 g/dL, and well tolerated . However, a substantial fraction of the anemia found in the elderly appears to occur in the absence of iron deficiency or elevated hepcidin levels .
Initial assessment of ACD involves a good history of the illness [30, 31] as well as general investigations to rule out other causes of anemia. Traditionally, the distinction between different causes of anemia is based on a hematological algorithm starting with the interpretation of the mean corpuscular volume (MCV) . Accordingly, micro-, macro-, or normocytic erythrocyte conditions may hint at different causes for anemia. Review of the morphology of the blood film as well as red blood cell indices (MCV, mean cell Hb, mean cell Hb concentration) and bone marrow, reticulocyte count, stool analysis, serum bilirubin and lactate dehydrogenase assay, and assessment of renal function, are required. Important indices of iron status such as serum iron, total iron binding capacity (TIBC) , transferrin saturation (TSAT) , ferritin , and more recently the ratio of TfR to log of ferritin are important in differentiating ACD from IDA and other causes of hypochromic, microcytic anemia such as the thalassemias. This requires an in-depth understanding of the stimuli and regulatory pathways of production of the various molecules that are usually assayed (see pathogenesis of anemia and Table 4.2). Some writers have also emphasized a key potential role for early markers of impaired erythropoiesis markers, such as the Hb content in reticulocytes (CHr) or reticulocyte Hb equivalent, measures that are not available as part of conventional laboratory tests .
Blood Smear Morphology
Even though the initial red cell morphology shows a normochromic and normocytic picture, with time this evolves into one that is hypochromic and microcytic. Severity of anemia in ACD is usually moderate with Hb usually 8–11 g/dL, rarely decreasing to Hb less than 7 g/dL. In these patients with suspected ACD, other causes of external blood loss or destruction should be looked for. The reticulocyte count (or better still the reticulocyte index) is usually reduced in ACD as well as in IDA. The blood smear may provide information on the underlying cause of ACD; thrombocytosis in cases of chronic hemorrhage, toxic granules in neutrophils in severe sepsis, hyper-segmented neutrophils in mixed nutritional deficiency or folate/Vitamin B12 deficiency found in malignant conditions.