Published: August 2014
Generally defined, anemia is present when the hemoglobin concentration is below a normal value based on the reference population. The mean normal value of hemoglobin is dependent on age, gender, race, and altitude. According to World Health Organization criteria, established over 40 years ago, the lower limit of normal in adults is 13 g/dL in men and 12 g/dL in women. These commonly applied limits have changed, especially when race is taken into account since African-Americans have physiologic levels of hemoglobin and hematocrit that are lower than this range.1 The blood hemoglobin concentration is believed to reflect more accurately the total red cell mass or status of the erythron (erythroid precursors of the marrow and circulating mature red cells) compared with the hematocrit. A drop in the hemoglobin level is observed in older men that may be the result of reduced androgen levels. This assumption does not, however, obviate the need for evaluation, especially if the patient is known to have had normal values in the recent past; the detection of a slight decrease in the hemoglobin level is often a signal of underlying disease, such as myelodysplastic syndrome. Other features that should prompt investigation include microcytic or macrocytic indices, elevated reticulocyte count (signifying hemolysis), and leukocyte or platelet abnormalities.
It's been estimated that 3.4 million Americans have anemia. Approximately 20% to 30% of hospitalized patients have some degree of anemia, with the highest percentage being found in intensive care units. The most common causes of anemia seen in general practice are neoplasia, inflammation "chronic disease," iron deficiency, maturation disorders, hemolytic anemias, acute bleeding, and marrow damage.2 In the elderly, however, upwards of 30% to 45% of anemia remains unexplained even after thorough hematologic evaluation. Theories of the mechanism of this hypoproliferative anemia range from low testosterone level, occult myelodysplastic syndrome, to defects in O2 sensing by erythropoietin-producing cells. However, no unifying cause has been identified.3
Anemia can also be defined physiologically by the degree of impairment of tissue oxygenation. Oxygen supply to tissues is controlled by a well-balanced mechanism that depends on the relative rate of oxygen supply and demand. Tissue oxygen delivery is dependent on the hemoglobin concentration, oxygen saturation and oxygen affinity, the degree and rate of change in blood volume, and the capacity for the cardiovascular and pulmonary systems to compensate. These, in turn, determine the clinical manifestations of anemia, on which the decision to transfuse should ultimately be based. Tissue oxygen delivery is also the major controlling factor of erythropoiesis through the synthesis and release of erythropoietin (EPO) by the proximal tubular cells or the peritubular interstitial cells in the kidney. EPO synthesis is governed by the activation of hypoxia inducible factor-1 (HIF-1), which controls the metabolic responses of multiple gene products to hypoxia. HIF-1 binds and activates the hypoxia-responsive transcriptional enhancer in the EPO gene regulatory region that upregulates EPO expression.4 EPO stimulates erythroid precursor cells (CFU-E [colony-forming units—erythroid]), leading to increased proliferation and shortening of their maturation time. The marrow responds to increased EPO maximally in 4 to 7 days if enough iron is available. Erythropoiesis can be increased by as much as a factor of 8. Typical of an endocrine loop feedback mechanism, there is an inverse relation between the hemoglobin and EPO levels measured in the blood (Figure 1). Although this relation holds true in simple iron deficiency, it is somewhat distorted in the anemia associated with inflammation or chronic disease, in which there may be a blunted EPO response. This has made prediction of the hemoglobin response to treatment with exogenous EPO unpredictable, except in limited circumstances (see below).
The clinical manifestations of anemia depend on the individual's ability to compensate for a loss in oxygen-carrying capacity. The more abrupt the onset of the anemia, the more dramatic the presentation. A sudden loss of more than one third of a patient's blood volume, for example, usually results in hypotension, respiratory distress, and acute mental status change, even in a young, previously healthy patient. With the more typical chronic development of anemia, the clinical changes are subtler and depend on the patient's age and comorbid conditions. The most familiar of these changes is an increase in cardiac output causing symptoms of palpitations and tachycardia, breathlessness, especially on exertion, and dizziness or lightheadedness. The patient may also complain of noise in the ears. This is not true tinnitus, but rather a roaring sound caused by accelerated blood flow through the ear. Some patients develop a feeling of profound generalized fatigue that can be accompanied by a loss of mental acuity, resulting in reduced ability to perform simple tasks such as reading a newspaper. These chronic symptoms are made worse by underlying coronary artery disease, congestive heart failure, and intrinsic pulmonary or cerebrovascular disease.
The evaluation of anemia can be a complex and difficult endeavor that may not yield a definitive diagnosis, even after exhaustive testing, including a bone marrow biopsy. Often, anemia exists in a milieu of chronic organ dysfunction or medical conditions that cloud the diagnosis because of their effect on erythropoiesis or red cell survival. These conditions may also create discrepancies in laboratory results, leading to further obfuscation in the differential diagnosis. Even bone marrow findings can be so subtle as to be nondefinitive, and only after long-term follow-up does the diagnosis become apparent. These difficulties arise in two of the most common types of anemia, iron deficiency anemia and the anemia of chronic disease (ACD). This review will emphasize the pathogenesis and diagnosis of these two types of anemias, how they are differentiated from each other, and how EPO is used therapeutically in select patients to avoid or lessen the requirement for transfusion.
|Red blood cell loss (bleeding)|
|Red blood cell destruction (hemolysis)|
|Mechanical (march hemoglobinuria, artificial heart valves)|
|Microangiopathic (disseminated intravascular hemolysis, thrombotic thrombocytopenic purpura, vasculitis)|
|Parasites and microorganisms (e.g., malaria, bartonellosis, babesiosis, clostridium perfringens )|
|Red cell membrane disorders|
|Chemical injury and complex chemicals (arsenic, copper, chlorate, spider, scorpion, and snake venoms)|
|Physical injury (heat, oxygen, radiation)|
|Hemoglobinopathies (sickle cell disease, unstable hemoglobins)|
|Red cell membrane disorders|
|Red cell enzyme defects (pyruvate kinase, 5' nucleotidase, glucose-6-phosphate dehydrogenase deficiencies)|
|Porphyrias (congenital erythropoietic and hepatoerythropoietic porphyrias, rarely congenital erythropoietic protoporphyria)|
|Red blood cell underproduction|
|Pluripotent stem cell failure|
|Erythroid progenitor cell failure|
|Functional impairment of erythroid and other progenitors due to nutritional and other causes|
|Pluripotent stem cell failure (Fanconi, Shwachman and dyskeratosis congenital syndromes)|
|Erythroid progenitor cell failure (Diamond-Blackfan, congenital dyserythropoietic syndromes)|
|Functional impairment of erythroid and other progenitors due to nutritional and other causes|
(Reprinted with permission of McGraw-Hill Companies, Inc. Prchal JT: Clinical manifestations and classification of erythrocyte disorders. In: Lichtman MA, Kipps TJ, Seligsohn U, et al, eds. Williams Hematology. 8th ed. McGraw-Hill Companies; 2010. Copyright ©2010 The McGraw-Hill Companies, Inc.).
Once anemia has been identified, classification of the physiologic mechanism is the most useful first step (Table 1). This kinetic classification is a useful first approach for any type of cytopenia. In this schema, red blood cells are being lost (blood loss), destroyed (hemolysis), or underproduced. Three patterns of erythropoiesis are identified based on the reticulocyte count, appearance of the bone marrow, and tests for hemolysis, such as the indirect bilirubin assay (Table 22). More than one mechanism can exist, or one category can evolve from another. The last example is best illustrated by iron deficiency anemia, caused first by blood loss and then by underproduction of red cells caused by lack of sufficient iron. Although the mechanism of anemia may be considered in kinetic terms, the mean corpuscular volume (MCV), measured directly by automated cell counters, is used to classify and diagnose the anemia further and guide the rest of the laboratory workup.
|Hemolytic anemia||↑ (appropriate for Hb)||Erythroid hyperplasia||↑|
|Hypoproliferative anemia||Inappropriately low for Hb||Erythroid hypoplasia||↓|
|Hypoproliferative megaloblastic anemia (ie., ineffective erythropoiesis)||Inappropriately low for Hb||Erythroid hyperplasia||↑|
Iron is an essential nutrient found mostly in heme proteins (e.g., hemoglobin, myoglobin), but also in a host of enzymes of intermediary metabolism. About two-thirds of the body's iron is incorporated into the hemoglobin of the erythron. The distribution of iron between the erythron, the plasma, and reticuloendothelial storage sites (e.g., hepatocytes, macrophages) is a tightly controlled process; both iron deficiency and iron overload are serious disorders. There has been a rapid increase in our understanding of the regulation of iron distribution. More than 15 newly identified proteins have been described in recent years. These proteins regulate the absorption of iron from the gastrointestinal (GI) tract, transport to the erythron, and storage of iron in reticuloendothelial cells.
Iron deficiency is one of the most common causes of anemia in the United States and worldwide. The three stages of iron deficiency are iron depletion (reduced stores), early iron deficiency anemia (depleted stores, normal MCV, and red cell morphology), and advanced iron deficiency anemia. Bleeding is the most common cause, typically GI bleeding or menstruation, although other causes of blood loss (e.g., pulmonary, urinary, and even factitious) occasionally present themselves (Table 3).5 A rare cause of iron deficiency is paroxysmal nocturnal hemoglobinuria. Although dietary lack of iron alone is rarely a cause of iron deficiency, it does contribute to the loss of iron brought about by otherwise minor bleeding, such as with normal menstruation. Iron malabsorption is rarely the cause of iron deficiency. Malabsorption of iron may result from achlorhydria (e.g., that seen in vitamin B12 deficiency), gastric bypass surgery, and celiac disease. Gastric resection alone would not be expected to cause iron deficiency unless the upper duodenum, the major site of iron absorption, was also removed or bypassed. About 50% of patients who have undergone a subtotal gastric resection will have impaired food iron absorption, but will still absorb exogenous iron. Inflammatory bowel disease involving the upper jejunum and duodenum may also cause malabsorption of iron. It has been estimated that approximately two-thirds of patients with iron deficiency anemia have GI lesions that can be detected by endoscopy, and 10% to 15% have a GI malignancy.6
|Antacid therapy or high gastric pH|
|Excess dietary bran, tannin, phytates, or starch|
|Competition from other metals (e.g., copper, lead)|
|Loss or dysfunction of absorptive enterocytes|
|Inflammatory bowel disease|
|Intrinsic enterocyte defects|
|Gastrointestinal blood loss: epistaxis, varices, gastritis, ulcer, tumor, Meckel's diverticulum, parasitosis, milk-induced enteropathy of early childhood, arterio-venous malformations, inflammatory bowel disease, diverticulosis, hemorrhoids|
|Genitourinary blood loss: menorrhagia, cancer, chronic infection|
|Pulmonary blood loss: pulmonary hemosiderosis, infection|
|Other blood loss: trauma, excessive phlebotomy, large vascular malformations|
(Reprinted with permission from Massachusetts Medical Society. Andrews NC. Disorders of iron metabolism. N Engl J Med 1999; 341:1986–1995. Copyright ©1999 Massachusetts Medical Society.).
The clinical manifestations of iron deficiency consist of those related to anemia, but there also appear to be effects on the central nervous system causing neuromuscular abnormalities and cognitive defects, especially in children. Pica occurs in children and adults. Effects on epithelial tissues, such as the nails (koilonychia), oropharyngeal mucosa (glossitis, angular stomatitis, and mucosal webs [Plummer-Vinson syndrome]), are rarely seen today because of earlier diagnosis and treatment.
The pure case of iron deficiency anemia is straightforward to diagnose. By the time red cell production is slowed and microcytic anemia occurs, iron stores are depleted, as determined by a low serum ferritin level, the major iron storage protein. The familiar laboratory tests of determination of the serum iron level, total iron-binding capacity (TIBC; transferrin), transferrin saturation, and ferritin level accurately reflect body iron stores, and obviate the need for a bone marrow biopsy in most cases. Although an MCV lower than 80 fL and a serum ferritin lower than 20 ng/mL are traditionally used to diagnose iron deficiency anemia, an MCV of 95 fL or a serum ferritin lower than 45 ng/mL were found to be predictive of iron deficiency secondary to serious GI lesions in one prospective study.7
After searching for a source of blood loss, iron replacement is begun, starting with ferrous sulfate, 325 mg three times daily, preferably taken 1 hour before meals. Both patient and physician must realize that several months of therapy are needed to replenish body iron stores. Although the most frequent cause of failure of iron therapy is patient noncompliance, it must be recognized that many patients cannot tolerate oral iron. This may be circumvented by switching to other iron salts (e.g., ferrous gluconate, ferrous fumarate) or to an oral iron-polysaccharide complex that may produce fewer side effects. Patients may tolerate delayed-release products better, but iron absorption could be impaired because of bypass of the major iron absorption sites in the upper jejunum and duodenum. Some patients tolerate oral iron yet do not respond to therapy. This could result from defects in the iron transport abnormalities in the intestinal epithelium as well as from the more common causes of malabsorption. These patients may benefit from parenteral iron replacement, with iron dextran, sodium ferric gluconate (Ferrlecit®), iron sucrose (Venofer®), or the latest addition to the collection of intravenous irons, ferumoxytol, a carbohydrate-coated iron oxide (Feraheme®). If using iron dextran, test doses should be given first because anaphylactic reactions could occur.
Anemia of Chronic Disease (ACD), or the anemia of inflammation, is a mild to moderate anemia accompanying infectious, inflammatory, or neoplastic disease that is characterized by abundant reticuloendothelial iron unavailable to bone marrow erythroid precursors.8 Iron-restricted erythropoiesis results from this defect in iron recycling, triggered by cytokines (e.g., tumor necrosis factor-α interleukin-1 or interleukin-6) that lower the serum iron level and affect macrophage iron storage so as to prohibit iron uptake. Hepcidin is a major regulator of iron absorption and iron efflux from macrophages. It plays a major role in the pathogenesis of ACD in that it decreases iron absorption in the small intestine and inhibits iron release from macrophages through its interaction with ferroportin. Hepcidin is also inducible by IL-1 and IL-6 and has been found to cause hypoferremia and anemia.9,10 Other mechanisms in ACD that limit erythropoiesis include inappropriately low EPO secretion and diminished EPO responsiveness. This same pattern of iron diversion and altered response to EPO can be seen in acutely ill patients in the intensive care unit with multiorgan dysfunction or sepsis. Although typically found in inflammatory conditions or malignancy, ACD is also associated with noninflammatory disorders, such as congestive heart failure, chronic obstructive pulmonary disease (COPD), alcoholic liver disease, and chronic kidney disease (CKD). In diabetics, for example, the EPO response to anemia is blunted, even in those patients without renal insufficiency.11 A similar blunted EPO response is seen in the unexplained anemia of the elderly in which there is no obvious inflammatory or chronic medical condition.12 In anemic COPD patients, moderately elevated EPO levels have been described, suggesting relative EPO resistance.13 Anemia develops 1 to 2 months after the onset of illness, does not progress, and parallels the severity of the underlying condition. Iron therapy is ineffective because of limited iron absorption and trapping of iron in macrophage storage sites.
The diagnosis of ACD can be difficult, especially if it coexists with iron deficiency. Other contributing causes of anemia should be ruled out, including blood loss, malnutrition, folate or vitamin B12 deficiency, and hemolysis. Myelodysplastic syndrome frequently masquerades as ACD until it progresses, and the threshold for performing a bone marrow biopsy should be low, particularly for those in older age groups or in the absence of an obvious chronic disease. This was recently demonstrated in a study of anemia in patients with rheumatoid arthritis, a prototypical ACD. A large percentage (45%) of the anemic patients, however, were found to be iron deficient. Of these approximately 10% were found to have a malignant or premalignant lesion of the GI tract.14
Laboratory findings of ACD include a mild to moderate anemia with normocytic or slightly microcytic indices, reduced serum iron and TIBC (transferrin) levels, and increased ferritin level, reflecting the increase in iron stores (Table 4). Although the serum ferritin level is useful when there is no accompanying inflammation, it is influenced by acute phase responses. The soluble transferrin receptor (TFR) is present in human plasma and its concentration is determined by marrow erythroid activity and iron status. Its synthesis is induced by iron deprivation, so in iron-deficient states the level of TFR rises. The TFR concentration is not increased with infection or inflammation, providing a way of differentiating iron deficiency from ACD. Patients who have a combination of iron deficiency and an infectious, inflammatory, or malignant disorder may be more accurately diagnosed by the TFR-ferritin index (TFR concentration divided by the log of the ferritin concentration; see Table 4). This index has been found to be a useful tool in identifying iron deficiency in patients with rheumatoid arthritis.14 Theoretically, hepcidin levels should be elevated in ACD, and correlate with elevations in proinflammatory markers. Some but not all studies have demonstrated such a correlation.15–17 Further investigation will be required before we have a sensitive and specific marker for the diagnosis of chronic disease anemia.
|Parameter||Chronic Disease Anemia||Iron Deficiency|
|Iron level||↓ to N||↓|
|Transferrin level||↓ to N||↑|
|Transferrin saturation||↓ to N||↓|
|Ferritin level||N to ↑||↓|
|TFR/log ferritin||Low (<1)||High (>4)|
TFR, transferrin receptor.
Whereas ACD will improve with recovery from the chronic disorder, this is usually not possible, although the anemia will wax and wane with the activity of an associated inflammatory process, and may resolve altogether with successful treatment of an underlying infection. Because the anemia is relatively mild and the time course for its development is prolonged, allowing adequate compensatory mechanisms to come into play, patients are usually asymptomatic from the anemia itself and do not require treatment. Patients with compromised cardiac function and other chronic diseases, however, may have more exaggerated anemia-related symptoms. Transfusions result in immediate correction of anemia, and may be useful in differentiating anemic symptoms from those of the underlying disease process. Chronic blood replacement, however, is not recommended because of the potential complications of iron overload, alloimmunization, delayed and immediate transfusion reactions, and potential viral transmission.
The major objectives for the use of recombinant human EPO therapy are to reduce transfusion requirements and to improve quality of life. Initiation of therapy should be based on the same rationale as one would use in deciding whether a transfusion is indicated—that is, whether or not the patient has clinical signs and symptoms of anemia that would be resolved or lessened by an increase in oxygen delivery. The two erythropoiesis-stimulating agents (ESAs) approved for use in the United States are epoetin alfa (Epogen®, Procrit®) and darbepoetin alfa (Aranesp®). The latter is a recombinant EPO that has been modified to increase the plasma half-life. It yields an equivalent erythropoietic response but requires less frequent dosing.18 The indications for EPO therapy have been refined and narrowed with the recognition of potential harmful toxicities, mainly those of thromboembolic complications and promotion of tumor growth in cancer patients. These led to the establishment of black-box warnings in the product inserts as well as the development of a Risk Evaluation and Mitigation Strategy for prescribing physicians. Guidelines have been developed for the treatment of the anemia of cancer19 and for anemia of CKD.20 These have been updated as new studies have been published regarding risks and benefits.21 EPO therapy has also been successful in patients various types of chronic disease anemia, especially rheumatoid arthritis, in whom improvements in disease activity as well as anemia have been reported.22,23 Hepatitis C patients also benefit from ESA support after developing anemia secondary to treatment with ribavirin.24
In patients with cancer, CKD, and chronic inflammatory diseases, EPO levels may be inappropriately low for the degree of anemia, suggesting that measuring EPO levels would be useful in predicting responses to therapy. In myelodysplastic syndromes, for example, an EPO level higher than 200 mU/L predicted a lack of response.25 Those with a level lower than 100 mU/L are said to be most likely to respond. This was substantiated in patients with advanced cancer, but only when an initial response (within 2 weeks) to epoetin was also observed.26,27 For other malignancies, and in most other chronic diseases, there is insufficient evidence to show that measuring serum EPO levels alone is clinically useful for predicting treatment responses,28 and current guidelines do not recommend it because it is unlikely to guide clinical decision making.20
While anemia management has provided much benefit in avoiding chronic transfusion in CKD, questions remain regarding the efficacy and safety of ESAs in renal failure patients. Thrombosis, stroke and cardiac events were observed during ESA therapy in CKD.29 Multiple randomized trials have now demonstrated that targeting a high hemoglobin level is associated with poorer blood pressure control and higher mortality compared with a lower hemoglobin target.30 Upon further analyses of these trials it was found that the ESA dose and degree of responsiveness were more important than the actual target hemoglobin in the development of adverse cardiovascular events in patients treated to a hemoglobin target of >12 g/dL. In CKD patients an epoetin dose of >20,000 U/week was found to be an independent risk factor for cardiovascular complications and death even in those who did not reach the target Hb.31 A revision of the guidelines in CKD recommends considering ESA therapy in patients with a hemoglobin of ≤11 g/dL, with the goal of maintaining the hemoglobin between 11 and 12 g/dL.21
In cancer patients, randomized, placebo-controlled studies have demonstrated that ESA therapy improves anemia, reduces transfusion requirements, and improves quality-of-life scores. As in CKD, however, ESAs have come under much scrutiny, not only because of the fear of increased cardiovascular events, but from evidence of a growth-promoting effect in certain tumor types. Randomized clinical trials seemed to support an increased mortality with ESA therapy, although more recent ones have not. In a Japanese study of epoetin in lung cancer and lymphoma patients, there was a decreased transfusion requirement and increased quality of life score. No thromboembolic events were observed.32 A second randomized study of epoetin in head and neck cancer showed no effect on survival in patients receiving radiation therapy.33 It should be recognized that these were relatively small studies, and an increased mortality risk from ESAs cannot be ruled out. The current recommendations of the American Society of Hematology (ASH)/American Society of Clinical Oncology (ASCO) Update Committee include discussing the potential risks (thromboembolism, increased mortality) and benefits (reduced transfusion requirement, improved quality of life) with each individual cancer patient. Use of an ESA is advised only in patients with a Hb <10 g/dL who are undergoing myelosuppressive chemotherapy.19 Epoetin and darbepoetin are held to be equivalent in efficacy and safety. The Food and Drug Administration label further stipulates that the chemotherapy be for palliative and not curative intent, although the latter is yet to be supported by appropriate clinical trials.
Recommended starting epoetin doses in cancer patients are higher than those in CKD: 150 U/kg 3 times weekly or 40,000 U weekly subcutaneously, and of darbepoetin, 2.25 µg/kg weekly or 500 µg every 3 weeks subcutaneously. If there is no response (increase in hemoglobin by ≥1 g/dL) after 4 weeks, the dose can be titrated upward to 60,000 U weekly. With darbepoetin, the dose can be increased to as high as 4.5 µg/kg weekly if there has been no response. The ASH/ASCO Update Committee continues to recommend that ESAs not be used in cancer patients who are not undergoing chemotherapy because of a prior randomized study of patients with various solid tumors demonstrating an increase in mortality and no decrease in transfusion rate. The only exception to this is in patients with low risk myelodysplastic syndrome where high-dose epoetin therapy (40,000 U twice/week) has been well tolerated and quite effective in correcting anemia and lessening transfusion dependence.34,35 In all cases, treating physicians are advised to follow the hemoglobin level closely when starting therapy and adjust the dose of EPO to maintain the lowest level required to avoid transfusion.
An uncommon complication of EPO is the development of neutralizing anti-EPO antibodies and pure red cell aplasia. Most of these cases have been described in Europe in patients treated with either the Eprex® brand of epoetin alfa or, to a lesser extent, epoetin beta (NeoRecormon), agents not in use in the United States.36 This has been attributed to those formulations that contain the stabilizer polysorbate 80 as opposed to human serum albumin. It has also been linked to subcutaneous versus IV administration of Eprex®.37 The frequency of EPO-induced red cell aplasia has dropped dramatically after recognition of this phenomenon and institution of changes in handling and administration. In general, however, EPO therapy is safe and, in many cases, results in the resolution of anemic symptoms, allowing responding patients to avoid the detrimental effects of chronic red cell transfusions.
In both renal and cancer patients, special attention should be paid to iron status while administrating erythopoietin. Iron requirements are increased during ESA therapy. If there is no increase in the hemoglobin level or a patient stops responding during therapy, iron deficiency should be excluded, because it will limit the erythropoietic response. Recommendations for iron replacement, however, are variable, and there is debate as to the optimal route, dose, and indication for initiating iron supplements in patients receiving ESAs. Several randomized trials in chemotherapy-induced anemia have demonstrated an increased hematopoietic response and lower transfusion requirement when ESA therapy was supplemented by intravenous iron. However, there was no control arm comparing oral to IV iron.38 The ASH/ASCO Update Committee felt there was still insufficient evidence to recommend intravenous iron as the standard of care. In end-stage renal disease, oral iron supplements have been largely abandoned and replaced by newer generation IV preparations; these are believed to be superior in achieving more rapid repletion of iron stores in patients known to have increased iron losses and possibly poor iron absorption. Rapid responses to ESAs and IV iron are observed. However, there are contrasting results among the few published randomized trials. In predialysis patients, for example, a randomized comparison in ESA-treated patients between oral ferrous sulfate and IV iron sucrose showed no differences in hemoglobin response or required ESA doses.39 Opposite results were seen in chemotherapy-induced anemia. One randomized controlled study has suggested that patients treated with IV iron have a higher hematopoietic response rate to ESAs than those given oral iron.40 Iron dextran preparations have a low but significant incidence of serious adverse effects, including anaphylactic reactions. Non-dextran iron preparations, such as ferric sodium gluconate (Ferrlecit) and iron sucrose (Venofer), are associated with a lower incidence of immediate toxicities. The most recent practice guidelines in CKD anemia recommend intravenous iron in hemodialysis patients, but either oral or IV in either peritoneal dialysis or non-dialysis CKD patients receiving ESA therapy. It should be kept in mind, however, that in these and other patient populations, there is concern that IV nontransferrin bound iron may have long-term effects through the production of excess free radicals, exacerbation of atherosclerosis, and risk of infection. The potential for long-term iron toxicity with IV iron supplementation remains to be determined.