Anemia is a pathologic state resulting in an insufficient number of erythrocytes to deliver oxygen to organs and tissues. Anemia can arise from blood loss, underproduction of erythrocytes, destruction of erythrocytes (hemolysis), or a combination of these factors. Patients with chronic anemia may be entirely asymptomatic or may experience symptoms of dyspnea, decreased exercise tolerance, palpitations, lightheadedness, or fatigue. Symptoms generally reflect the degree of anemia and the rapidity with which anemia develops. Symptoms are also determined by end-organ function and vascular disease. Anemia developing over the same period in a young, otherwise healthy person may be asymptomatic, whereas in someone with critical coronary artery atherosclerosis, it may manifest with severe chest pain.
From a laboratory perspective, the automated complete blood count (CBC) identifies the severity of the anemia. The CBC includes the erythrocyte count, hematocrit level (percentage of total blood volume composed of erythrocytes), hemoglobin concentration in the blood, and erythrocyte indices such as mean corpuscular volume (MCV) and the red cell distribution width (indication of the range of erythrocyte size). Physiologically, men have a higher hemoglobin concentration than women because of the erythropoietic effects of androgens.
Normal pregnancy is associated with a dilutional anemia resulting from an increase in plasma volume that exceeds the increase in erythrocyte mass to deliver oxygen to the developing fetus at low viscosity. Other pathophysiologic states can affect hemoglobin levels. Because erythrocyte production is controlled by erythropoietin synthesized in the kidney in response to hypoxia, kidney disease can lead to anemia. Similarly, patients deficient in iron, cobalamin, or folate can develop anemia because these vitamins and minerals are required for hemoglobin production. Whether anemia is a normal result of the aging process is a subject of debate, although it is clear that associated pathophysiologic states and, likely, nutritional deficiencies contributing to anemia are more common in older adults.
In assessing patients with anemia, reviewing the CBC along with the peripheral blood smear can lead to valuable diagnostic clues. Defining the anemia as microcytic, macrocytic, or normocytic can narrow the differential diagnosis. Furthermore, reviewing erythrocyte morphology microscopically can provide a diagnosis. Table 12 lists common erythrocyte morphologic features and their interpretations. Anemia in adults is commonly caused by iron deficiency, absolute or relative erythropoietin deficiency (kidney disease or the anemia of inflammation, respectively), or acute blood loss. Acquired hemolytic anemia is much less common. Anemia in hospitalized patients most often results from acute blood loss, the anemia of inflammation, or the anemia of kidney disease.
In addition to the CBC, a reticulocyte count provides information on the marrow response to anemia. A normal marrow will produce reticulocytes in response to anemia or hypoxia. In contrast, patients with vitamin B12, folate, or iron deficiency or those with marrow diseases such as myelodysplasia or aplastic anemia cannot make adequate erythrocytes and therefore have a low reticulocyte count for their degree of anemia. The reticulocyte count may be reported as a percentage of the total erythrocyte count or as the absolute number of reticulocytes as determined by flow cytometry.
An algorithm for the evaluation of anemia can be found in Figure 8. In addition to the CBC and review of the peripheral blood smear, a bone marrow aspirate and biopsy can be helpful in the diagnosis of anemia, especially in assessing stem cell disorders such as aplastic anemia, dysmyelopoietic syndrome, and acute leukemia. Anemia combined with other cytopenias increases the likelihood of a primary marrow cause. Prussian blue staining can identify marrow iron stores, although interpretation of serum ferritin levels is often sufficient for that purpose. Anemia should never be considered a final diagnosis; the cause must be identified. Recognizing the underlying cause leads to more focused treatment beyond transfusion to correct the anemia, as well as other prevention or therapeutic opportunities linked to the specific cause.
Iron is an essential component of the hemoglobin molecule. In addition to its critical role in oxygen delivery, iron is also necessary for DNA synthesis and cellular transport. Most of the iron in the body is contained in the erythron; each milliliter of packed red blood cells contains about 1 mg of elemental iron. Hepcidin, the key peptide involved in iron regulation, is produced in the liver and is a negative regulator of iron absorption. Hepcidin production increases with inflammation and decreases in response to hypoxia, anemia, and iron deficiency. Hepcidin production is regulated by several proteins, including human hemochromatosis, hemojuvelin, matriptase-2, and transferrin receptor 1 and 2. These proteins are important in hereditary iron overload states discussed later in this chapter. Hepcidin causes internalization and proteolysis of ferroportin in the enterocyte and macrophage leading to decreased iron absorption from the gut and decreased iron release from macrophages.
Iron is derived from the diet as heme-based iron from red meat, poultry, and fish and non–heme-based iron from green leafy vegetables, lentils, beans, and peas. Vegetable iron has a more limited uptake because of phytates and oxalates that form complexes with iron and limit bioavailability. The typical adult diet contains 5 mg of iron for every 1000 calories. Only 10% to 20% of oral iron is absorbed. Because iron loss cannot efficiently be increased, in a normal physiologic state, iron absorption is about equal to iron loss. The typical adult man loses approximately 1 mg of iron daily from gastrointestinal mucosal turnover, whereas the typical adult woman loses approximately 1.5 mg of iron daily through mucosal turnover and menstrual blood loss.
Iron intake, use, and loss require a fine balance. Iron deficiency is a significant problem because of increased iron requirements during pregnancy and lactation and endemic helminthic infections that increase iron loss. Iron requirements are increased during normal growth and development, making iron deficiency in infants and children a widespread occurrence. Iron deficiency in adults rarely occurs secondary to decreased oral intake but is more commonly secondary to blood loss. For premenopausal women, this is typically secondary to menstrual blood loss, but for men or postmenopausal women, occult gastrointestinal blood loss should be suspected. Numerous lesions in the upper and lower gastrointestinal tracts may cause such bleeding, but identifying an occult colonic neoplasm may be especially important. Iron is absorbed in the proximal small bowel, and patients with celiac disease, inflammatory bowel disease, or surgical resection can experience iron malabsorption. Importantly, iron malabsorption may occur in the absence of diarrhea, steatorrhea, and weight loss, which are common symptoms of malabsorption. Helicobacter pylori infection is associated with iron deficiency because of impaired iron uptake (this mechanism is not well established) and increased iron loss from gastritis or peptic ulcer disease. Common causes of iron deficiency anemia are listed in Table 13.
Patients with iron deficiency may be asymptomatic or may experience fatigue, lack of a sense of well-being, irritability, headache, and decreased exercise tolerance. Pica, the tendency to eat or crave starch, clay, paper, ice, or other crunchy foodstuffs, is sometimes seen in severe iron deficiency. The pathophysiology of pica remains poorly understood.
Physical examination findings may be normal in patients with early anemia from iron deficiency. Symptomatic patients can experience tachycardia, conjunctival rim pallor (inferior conjunctival rim same color as palpebral conjunctiva), glossitis, and stomatitis. Severe iron deficiency can cause spooning of the nails (koilonychia).
A low MCV, elevated red-cell distribution width, and peripheral blood smear showing microcytosis and anisopoikilocytosis (Figure 9) are virtually diagnostic of iron deficiency, especially in premenopausal women; these findings may obviate the need for additional laboratory testing, provided that a follow-up CBC is performed to assess response to iron therapy. A low serum iron level, elevated total iron-binding capacity, low transferrin saturation (iron/total iron-binding capacity × 100), and serum ferritin level less than 14 ng/mL (14 µg/L) confirm the diagnosis of iron deficiency. Although ferritin is an acute phase reactant that increases with inflammation, a ferritin level greater than 100 ng/mL (100 µg/L) virtually excludes iron deficiency. Measuring serum levels of zinc protoporphyrin and soluble transferrin receptor was previously suggested in the evaluation of iron deficiency, but because of considerable overlap in patients with and without iron deficiency, these measurements are no longer recommended. Thrombocytosis occurs frequently with iron deficiency caused by blood loss. Although a bone marrow stained with Prussian blue can detect iron stores, this test is seldom necessary in the diagnosis of iron deficiency.
Iron deficiency is typically treated with oral iron salts. Oral ferrous sulfate is the least expensive preparation, with each 325-mg tablet containing 65 mg of elemental iron. Oral absorption can be increased with supplemental vitamin C; conversely, medications such as antacids and fiber can reduce absorption. It has become increasingly recognized that frequent dosing (two or three times daily) of oral iron can lead to increased hepcidin production, which actually reduces iron absorption. For this reason, a single daily or every-other-day dose of oral iron sulfate may be the best replacement dose. Although oral ferrous fumarate, ferrous gluconate, and other oral iron salts are available, none have proven superior to ferrous sulfate in tolerability, efficacy, or cost. Delayed-release and slow-release preparations may be better tolerated but are also associated with reduced iron absorption because these products can bypass the intestinal sites where iron absorption occurs. Although oral iron is typically well tolerated, symptoms can include gastrointestinal upset, constipation, and abdominal pain. Iron replacement typically results in a reticulocytosis within days. Hemoglobin levels typically increase by approximately 1 g per week. Oral iron replacement typically lasts 3 to 6 months after normalization of hemoglobin to replace iron stores.
For patients undergoing dialysis who have large iron requirements or for patients with celiac disease, inflammatory bowel disease, or those who have undergone resection of the stomach or small bowel, oral iron may not be adequate treatment. These patients typically require parenteral iron. Iron dextran was the principal intravenous iron product used in the past, but with the risk of anaphylaxis and relatively poor bioavailability, newer parenteral iron preparations with a better safety profile and better ferrokinetics are more often used. These include iron sucrose, ferric gluconate, ferumoxytol, and ferric carboxymaltose. Data are inadequate to recommend one newer parenteral agent over another for patients requiring parenteral iron.
For decades, anemia has been recognized in conjunction with either chronic infections such as tuberculosis or osteomyelitis or in conjunction with malignancy. Because these types of infection and malignancy are “chronic diseases,” these anemic states were often referred to as “anemia of chronic disease.” From a pathophysiologic standpoint, “inflammatory anemia” is a more appropriate term, because these anemias are related to increased hepcidin production in response to inflammatory mediators such as interleukin-6. Hepcidin causes decreased iron absorption from the enterocyte through internalization and proteolysis of ferroportin. No FDA-approved assays for hepcidin evaluation are available, although several enzyme-linked immunosorbent assays and spectrometry-based assays are in development. Inflammatory cytokines also blunt the erythropoietin response to anemia.
In patients with inflammatory anemia, the peripheral blood smear typically shows a normochromic, normocytic anemia. Over time, as iron absorption decreases, microcytosis can be seen. Typically, patients with inflammatory anemia have a hemoglobin level of 8 to 10 g/dL (80-100 g/L). Because of impaired erythropoiesis, caused by a blunted response to erythropoietin, the reticulocyte count is typically low for the degree of anemia. The characteristic iron study pattern seen in inflammatory anemia is increased serum ferritin level, low serum iron level, and a reduced total iron-binding capacity. Table 14 lists characteristic laboratory features useful in distinguishing iron deficiency from inflammatory anemia. Although seldom necessary, a bone marrow biopsy specimen would show adequate stainable iron stores.
Patients with diabetes or heart failure can also have an increase in inflammatory cytokines and related inflammatory anemia. In some patients with findings consistent with inflammatory anemia, the inflammatory state is not obvious. Such patients do not require extensive evaluation for the source of inflammation.
Inflammatory anemia seldom requires treatment. Iron replacement is ineffective. Although supplemental erythropoietin can correct anemia in inflammatory states, it can also lead to hypertension and thrombosis and should be used with extreme caution. Patients with cancer who experienced anemia with chemotherapy and who received erythropoietin felt better and had better blood counts, but cancer mortality was increased.
Because erythropoietin is made in the renal cortex in response to anemia and hypoxia, patients with kidney disease can have anemia that is typically normochromic and normocytic. In kidney disease, the reticulocyte count is typically low because of a relative deficiency of erythropoietin. The peripheral blood smear may show burr cells (echinocytes) in patients with features of uremia.
Because kidney disease is also associated with platelet defects and gastrointestinal bleeding from ulcer disease or angiodysplasia, microcytosis in the setting of kidney disease should raise the suspicion for iron deficiency. In some patients, minor elevations in serum creatinine level can be associated with low erythropoietin levels and anemia. Although it is important to first rule out other causes of anemia, measuring the erythropoietin level can be helpful in evaluating anemia in patients with mild kidney disease.
Anemia of kidney disease can be associated with fatigue, depression, dyspnea, and decreased exercise tolerance. Anemia is also associated with increased morbidity and mortality from heart disease and stroke in patients undergoing dialysis. Supplemental use of erythropoiesis-stimulating agents (ESAs) can improve anemia in patients with kidney disease. Guidelines recommend that ESAs be withheld in patients with chronic kidney disease (CKD) not requiring dialysis who have a hemoglobin level greater than 10 g/dL (100 g/L). For patients with CKD with a hemoglobin level less than 10 g/dL (100 g/L), ESA treatment should be individualized based on symptoms, rapidity of hemoglobin decline, and transfusion needs. For patients undergoing dialysis, ESAs should be initiated for hemoglobin levels less than 10 g/dL (100 g/L), but hemoglobin concentrations should not exceed 11.5 g/dL (115 g/L) to avoid adverse effects, including worsening hypertension, volume overload, and thrombotic complications. Patients with CKD require iron replacement because of gastrointestinal blood loss and the need for freely available iron to maintain adequate response to ESAs. Because of high iron requirements in patients taking ESAs, parenteral iron is typically used to maintain a serum ferritin level greater than 100 ng/mL (100 µg/L) with a transferrin saturation of at least 20%. Although more than 95% of patients respond to ESAs, insufficient response can result from iron deficiency, folate deficiency, aluminum toxicity, blood loss, or inflammation.
Cobalamin is necessary for DNA synthesis. Humans cannot synthesize cobalamin but must consume it in their diet; it is found in animal meats, liver, shellfish, and dairy products. Dietary deficiency is an uncommon cause of cobalamin deficiency because body stores are typically available for many years. Instead, cobalamin deficiency is nearly always a result of malabsorption. Dietary cobalamin is more available for absorption in an acid environment and requires binding with intrinsic factor to enhance absorption in the terminal ileum.
Cobalamin deficiency may be the result of decreased bioavailability, which may be the result of age-related gastric achlorhydria or the use of proton pump inhibitors, or both. Cobalamin malabsorption can also occur in more generalized malabsorptive states such as inflammatory bowel disease, pancreatic insufficiency, and bacterial overgrowth.
Pernicious anemia, characterized by autoimmune gastritis and intrinsic factor deficiency, is another cause of cobalamin deficiency. Antibodies to parietal cells are found in 90% of patients with pernicious anemia, whereas antibodies to intrinsic factor are detected in about 70% of patients. Although antibody testing is sometimes used in the diagnosis of pernicious anemia, the varied sensitivity and specificity of these tests limits the utility of such testing. The Schilling test is no longer used in the evaluation of cobalamin deficiency because of the limited availability of radioactive intrinsic factor.
Patients with cobalamin deficiency can present with weight loss, glossitis, and “lemon yellow” skin because of pallor and jaundice resulting from ineffective erythropoiesis. Severe cobalamin deficiency can cause neurologic symptoms, including loss of vibratory sense, loss of proprioception, spastic ataxia, and other dorsal column symptoms. Psychiatric symptoms (megaloblastic mania) can manifest as dementia, hallucinations, and frank psychosis.
In patients with cobalamin deficiency, the peripheral blood smear shows oval macrocytes and hypersegmented neutrophils (Figure 10). Pancytopenia resulting from ineffective hematopoiesis can also be seen. Other laboratory findings are consistent with intramedullary hemolysis caused by ineffective erythropoiesis, including decreased haptoglobin, elevated lactate dehydrogenase, and elevated indirect bilirubin levels. The reticulocyte count is low in patients with cobalamin deficiency.
Although serum cobalamin levels may be low, serum cobalamin, especially in the low-normal range, may not adequately represent tissue cobalamin levels. As such, an elevated concentration of methylmalonic acid is a more sensitive indicator of cobalamin deficiency.
Although supplemental folate can improve the anemia found in cobalamin deficiency, folate does not correct or prevent the associated neuropsychiatric complications. An important distinction to make is that folate deficiency leads to an elevation in homocysteine levels with normal levels of methylmalonic acid, whereas cobalamin deficiency has increases in both metabolites.
Patients with cobalamin deficiency can be treated with oral cobalamin, 1000 to 2000 µg daily, regardless of cause; an adequate amount of this dose will be absorbed, even if intrinsic factor is lacking or malabsorption is ongoing. Parenteral cobalamin is more expensive and more cumbersome to administer. Intranasal and oral gel preparations are also available but are not clearly superior to tablet preparations. When cobalamin is replaced, megaloblastic changes in the marrow improve within hours. Reticulocytosis appears in several days, and hemoglobin level increases by approximately 1 g per week. If the response to cobalamin is inadequate, an alternative diagnosis, such as myelodysplasia, should be considered. Neurologic changes may not be reversible with replacement.
Folate is a common component of most diets in the United States; it is found in green leafy vegetables, bananas, lemons, melons, and most other fruits. Supplemental folate has been added to grains in the United States for many years to prevent birth defects. As such, dietary folate deficiency is uncommon except in patients with malnutrition. Persons who consume excess alcohol are apt to ingest inadequate amounts of folate as well as having impaired absorption. Folate is poorly stored and deficiency can develop in weeks to months in patients with insufficient folate ingestion. Patients with disease states characterized by rapid cell turnover, such as pregnancy, hemolysis, or desquamating skin disorders (psoriasis), have increased folate requirements.
Drugs such as triamterene, phenytoin, or methotrexate can lead to folate deficiency by either inhibiting folate absorption or conversion to its active form. Because folate is absorbed in the jejunum, small bowel diseases such as amyloidosis, celiac disease, or inflammatory bowel disease can also inhibit folate absorption.
The peripheral blood smear in folate deficiency is identical to that seen with cobalamin deficiency. Serum folate measurement, if very low, helps establish the diagnosis; however, a normal level may be unreliable because a single meal can normalize levels. Although folate levels in erythrocytes may better reflect chronic folate balance, an elevated serum homocysteine level has a sensitivity and specificity of greater than 90% in diagnosing folate deficiency and is the preferred test when deficiency is suspected despite a normal serum folate level.
After cobalamin deficiency is excluded, patients with folate deficiency should receive oral folate, 1 to 5 mg/d.
Thalassemia is associated with erythrocyte underproduction and ineffective erythropoiesis and can be considered a disorder of erythrocyte production. Hemoglobin is a tetramer containing two α chains, two β chains (α2β2), and the iron-containing tetrapyrrole heme moiety. Production of α and β chains normally occurs in a balanced fashion from genes located on chromosomes 16 and 11, respectively. Mismatched production of either α or β chains results in impaired production of hemoglobin and ineffective erythropoiesis. The worldwide incidence of impaired β-chain synthesis is 1% to 5%, with impaired α-chain synthesis being even higher. Thalassemia is common in African and Mediterranean countries, the Middle East, and Southeast Asia. Homozygous α-thalassemia leads to intrauterine demise, and homozygous β-thalassemia causes severe symptomatic anemia invariably diagnosed at an early age. Internists are likely to manage patients with heterozygous forms of thalassemia.
The peripheral blood smear typically shows microcytosis, nucleated erythrocytes, and target cells in patients with thalassemia (Figure 11). Unlike other underproduction anemias, patients with heterozygous thalassemia typically have a preserved or even an increased erythrocyte count associated with a decreased MCV. Although patients with iron deficiency typically have considerable variation in cell shape and size (anisopoikilocytosis), leading to elevation in red cell distribution width, the microcytic cells in thalassemia are more uniform, and the red cell distribution width is normal. Because of ineffective erythropoiesis, thalassemia is associated with increased lactate dehydrogenase, increased unconjugated bilirubin, and decreased haptoglobin levels.
The α-globin gene is duplicated on chromosome 16 leading to several genotypes. Patients with a single α gene mutation are silent carriers and are clinically healthy. A two-gene mutation results in α-thalassemia trait characterized by a hemoglobin level of approximately 10 g/dL (100 g/L) with microcytosis. Unlike patients with iron deficiency, patients with α-thalassemia trait have normal or elevated iron stores. Diagnosis is usually achieved by excluding other causes of hypochromic microcytic anemia. Select reference laboratories can establish the diagnosis through direct sequencing of the globin genes. Patients with a three-gene mutation have more severe anemia and make a tetramer of β globin called hemoglobin H that can be identified on electrophoresis. In contrast, hemoglobin electrophoresis is normal for α-thalassemia trait.
Patients with α-thalassemia trait should receive supplemental folate and genetic counseling before starting a family. Patients with hemoglobin H disease typically have hemoglobin concentrations of approximately 7 to 8 g/dL (70-80 g/L) and are seldom transfusion dependent. Care should be taken to avoid supplemental iron in patients with thalassemia because these patients absorb iron well; transfusion should also be avoided. In such circumstances, these patients may develop iron overload with resultant heart and liver failure requiring iron chelation therapy with parenteral desferrioxamine or oral iron chelators (deferasirox or deferiprone).
More than 250 mutations have been described in the β-globin gene resulting in a spectrum of diseases from mild reduction in β-chain synthesis (β+-thalassemia) to complete absence of β-chain synthesis (β0-thalassemia). As such, the clinical spectrum of disease in β-thalassemia includes β-thalassemia minor, β-thalassemia intermedia, and β-thalassemia major. β-Thalassemia minor (trait) is characterized by a microcytic anemia (MCV, 60-70 fL) with a hemoglobin level of 10 to 12 g/dL (100-120 g/L). Unlike α-thalassemia trait, β-thalassemia minor produces an abnormal hemoglobin electrophoresis with an increase in hemoglobin A2 (α2δ2) because of a substitution of δ globin for β globin. Hemoglobin F may also be increased depending on the specific mutation. Patients with β-thalassemia intermedia have hemoglobin levels of 7 g/dL (70 g/L) without need for transfusion.
Patients with β-thalassemia should be treated with folate; as with α-thalassemia, supplemental iron should be avoided.
Hemolysis occurs when excessive erythrocyte destruction occurs, either through ineffective erythropoiesis; clearance in the reticuloendothelial system of the spleen or liver; immune injury mediated by immunoglobulins or complement; or physical destruction by fibrin, valves, or other intracirculatory devices.
Hemolysis may be accompanied by a bone marrow response (reticulocytosis) and is invariably accompanied by findings consistent with erythrocyte destruction, such as an increase in unconjugated bilirubin, increased lactate dehydrogenase, increased serum free hemoglobin, and hemoglobinuria. Symptoms of hemolysis depend on the degree of anemia and its chronicity. Pigmented gallstones from insoluble calcium bilirubinate are common in chronic hemolytic disorders.
Hemolysis can be broken down into congenital and acquired causes. Congenital disorders segregate into hemoglobinopathies (sickle cell), disorders of the erythrocyte membrane (hereditary spherocytosis), enzyme defects (glucose-6-phosphate dehydrogenase deficiency), and the thalassemia syndromes.
Acquired hemolysis can occur secondary to medications (fludarabine, bendamustine, quinine, penicillins, α-methyldopa); can be immune in nature; or can occur secondary to micro- or macroangiopathic processes, infections, or physical agents.
Hereditary spherocytosis (HS) is typically an autosomal dominant disorder more common in people of Northern European descent. HS is not confined to a single mutation; rather, mutations in α spectrin, β spectrin, ankyrin, band 3, and protein 4.2 have been described in patients with HS. Mutations in these scaffolding proteins adversely affect the interaction between the lipid bilayer and cytoskeleton in the erythrocyte cell wall resulting in a spherocyte (Figure 12) characterized by reduced surface-to-volume ratio, osmotic fragility, and splenic sequestration.
Symptoms of HS can be quite variable, and even though HS is a congenital disorder, patients may adapt to mild anemia without symptoms until well into adulthood. An acute episode of bone marrow suppression, typically associated with infection, can lead to new anemia symptoms in patients who were previously able to compensate for hemolysis. In addition to anemia, symptomatic patients typically present with splenomegaly or, more rarely, splenic infarction or rupture, and calcium bilirubinate (pigmented) gallstones are common.
Laboratory findings in HS include spherocytosis on the peripheral blood smear and varying degrees of anemia, reticulocytosis, and hyperbilirubinemia. The mean corpuscular hemoglobin concentration is elevated in patients with HS. Erythrocytes in these patients have increased osmotic fragility when exposed to fluids with high osmolarity. This is the basis of the osmotic fragility test sometimes used to diagnose HS. Flow cytometry can identify characteristic cell surface abnormalities that establish the diagnosis.
Patients with HS, as with other hemolytic states, have increased folate requirements and should receive supplemental folate. For severe hemolysis, splenectomy is effective, and partial splenectomy has proven effective in young children whose immune function is improved with a functioning spleen. As with other conditions requiring splenectomy, vaccination for Streptococcus pneumoniae, Haemophilus influenzae, and Neisseria meningitides is important before the procedure.
Other congenital disorders of the erythrocyte membrane that can lead to hemolysis include hereditary elliptocytosis and hereditary pyropoikilocytosis. Hereditary elliptocytosis is less commonly associated with anemia.
The gene for glucose-6-phosphate dehydrogenase (G6PD) is located on the X chromosome, and as such G6PD deficiency primarily affects men. It can affect women who are homozygous for G6PD disease, through the process of lyonization (inactivation of one of the two X chromosomes) with preference of expression for the defective gene, or who have Turner syndrome (XO karyotype). G6PD deficiency is the most common enzyme deficiency in humans and has been associated with more than 160 gene mutations. G6PD is important in the pentose phosphate pathway, allowing for reduction of nicotinamide adenine dinucleotide phosphate (NADP to NADPH). NADPH is necessary to reduce oxidative stress in response to drugs (chloroquine), infection, or toxins. Two G6PD variants, G6PD A, which produces mild disease that is often asymptomatic, and G6PD Mediterranean are most commonly seen. G6PD Mediterranean is associated with favism, or hemolysis in people who consume fava beans. G6PD variants are thought to provide partial protection for malarial infections. G6PD deficiency occurs most commonly in people of African, Asian, Mediterranean, and Middle Eastern descent and among Kurdish Jews.
G6PD deficiency has a varied presentation. It can be associated with neonatal jaundice and, in later life, with acute hemolysis typically occurring within 1 to 3 days after exposure to oxidative stress. Typical triggers include drugs such as chloroquine, sulfonamides, rasburicase, dapsone, and phenazopyridine, and environmental toxins such as naphthalene from mothballs. The peripheral blood smear may show bite cells (Figure 13) and Heinz bodies (denatured hemoglobin) visible on supravital stain. Although very uncommon, severe hemolysis can lead to kidney injury and dialysis.
G6PD evaluation involves a fluorescent spot test used to detect NADPH. A positive result shows lack of fluorescence. Evaluation should not be performed during an acute hemolytic episode because, with hemolysis, the older, more enzyme-deficient erythrocytes are preferentially destroyed leaving younger erythrocytes with higher G6PD levels in the circulation; this often leads to a false-negative result. Qualitative G6PD enzyme activity and subtyping can provide a more specific diagnosis.
Typical management of G6PD deficiency includes recognition and diagnosis of the disorder and avoidance of offending agents. During hemolytic episodes, treatment is supportive, with transfusion reserved only for severely symptomatic patients.
The onset of acute hemolysis after treatment with TMP-SMX, negative direct and indirect anti-globulin (Coombs’) test, and the presence of “bite cells” on peripheral smear is consistent with a diagnosis of glucose-6-phosphate dehydrogenase (G6PD) deficiency.
G6PD deficiency, an X-linked disorder, is the most common enzymatic disorder of erythrocytes and is seen in the African American and Mediterranean populations. Oxidative stress is commonly induced by:
Typically, patients present two to four days after an oxidatively stressful event with evidence of hemolysis (e.g., 3-4 g/dL drop in hemoglobin concentration) with a concomitant rise in reticulocytes. In this process, the older deficient RBCs are replaced by the younger cells, which have sufficient G6PD enzyme activity to sustain further oxidative damage without hemolysis. Additional laboratory findings include elevated lactate dehydrogenase and indirect bilirubin, and decreased haptoglobin. The peripheral smear also shows "bite" cells (black arrows in above figure). Since reticulocytes can have normal G6PD levels, measuring G6PD levels during an acute episode can produce normal results. If a false negative is suspected, it is recommended to recheck the G6PD level 2 to 3 months after the hemolytic episode.
There is perhaps no hemolytic anemia as clinically dramatic as sickle cell anemia. Affecting nearly every organ system and characterized by extraordinarily painful vaso-occlusive crises, sickle cell disease (SCD) is prevalent among Black patients. SCD is characterized by homozygosity for a single point mutation in the sixth position of the β-globin gene resulting in abnormal hemoglobin S (Hb S) that polymerizes under hypoxic conditions, leading to deformed erythrocytes that can adhere to the endothelium of capillaries throughout the circulation. Because of hemolysis, free hemoglobin scavenges nitric oxide, and hemolyzed erythrocytes release arginase, depleting the body of arginine, a necessary precursor to nitric oxide. As a result, patients with SCD experience vasoconstriction and platelet activation, complicating their clinical course.
Hb S is more common in patients of African origin, and most patients homozygous for Hb S (Hb SS) are Black, with about 8% of Black persons in the United States having sickle cell trait (Hb S). Hb S can be coinherited with hemoglobin C (Hb SC), β-thalassemia (Sβ+-thalassemia), or other hemoglobins, which tends to lead to milder disease than Hb SS. Characteristic electrophoretic patterns for sickling disorders are shown in Table 15.
Chronic hemolysis, vaso-occlusive disease with acute and chronic end-organ damage, nitric oxide depletion, and immune compromise from functional asplenia lead to the myriad of problems faced by patients with SCD. The only potential “cure” for SCD is allogeneic bone marrow transplantation, but the timing and appropriate population for transplantation have not been adequately determined. Hydroxyurea modulates many of the complications of SCD and even prolongs life expectancy because of its ability to increase fetal hemoglobin levels and generate nitric oxide. Hydroxyurea is not used often enough and is not always used at the appropriate dose. It should be prescribed for all patients with frequent pain crises, severe symptomatic anemia, or previous acute chest syndrome or stroke. Managing SCD requires a team of providers and must include preventive, acute, and chronic treatment goals (Table 16).
Painful vaso-occlusive events are the hallmark of SCD. The frequency and severity of acute painful events varies considerably among patients with SCD, the reasons for which remain poorly understood. In most communities, a few patients with recurrent acute painful events account for most hospitalizations. Research has shown that many practitioners do not properly assess and manage pain in this population, often leaving pain undertreated. Painful events are associated with morbidity and mortality in patients with SCD, and the number of painful crises is inversely related to life expectancy. For these reasons, recognition and treatment of painful events should be immediate. Opiates are the preferred analgesic, with the exception of meperidine, which can cause seizures. Chronic pain is more difficult to treat in SCD, and the pathophysiology is poorly understood. Chronic pain is exacerbated by emotional stress, anxiety, insomnia, and depression and is perhaps best treated with a holistic approach emphasizing nonopioid options such as NSAIDs, relaxation techniques, massage, and biofeedback. A few patients with SCD develop opioid dependence, although the prevalence is no greater than in a comparable age- and illness-matched population.
Two new treatments to prevent painful crisis have recently been approved. Oral L-glutamine, a precursor of nicotinamide adenine dinucleotide (NAD), is necessary to form the antioxidant NADH. It is believed that sickle cell crises are partially related to oxidant stress. A randomized trial in patients receiving L-glutamine showed benefit with reduction in the number of sickle cell crises. Similarly, the recently approved P-selectin inhibitor, crizanlizumab, has been shown to reduce painful crises by preventing cellular adhesion.
Pulmonary complications are a common source of morbidity and mortality in patients with SCD. Acute chest syndrome (ACS) is a clinical diagnosis based on the constellation of fever (>38.6 °C [101.5 °F]), tachypnea, hypoxia, cough, shortness of breath, and a new pulmonary infiltrate. ACS can be the result of infection, in situ thrombosis, thromboembolism, fat embolism, or a combination of these factors. In addition to supportive measures, including oxygen to treat hypoxia and the use of empiric antibiotics to treat infection, patients with ACS require erythrocyte transfusion, either with packed red blood cells (PRBCs) or as exchange transfusion. In addition to ACS, pulmonary hypertension is a recognized complication of SCD found in up to 30% of patients, typically presenting as worsening right heart failure. Although associated with increased mortality, the appropriate management of pulmonary hypertension in patients with SCD has not been determined. Phosphodiesterase inhibitors, such as sildenafil, have been associated with increased painful crises.
Stroke and other central nervous system disease remain major complications of SCD. Approximately 30% of patients with Hb SS, Hb SC, or Sβ+-thalassemia can have silent or symptomatic strokes. Monthly transfusion begun after a stroke can reduce the incidence of subsequent stroke by 50% but carries the risk of iron overload necessitating iron chelation. Hydroxyurea may also help in stroke prevention. In addition to thrombotic stroke, patients with SCD can have moyamoya disease (irregular perforating vascular networks near occluded or stenotic vessels in the region corresponding to lenticulostriate and thalamoperforate arteries) predisposing to cerebral bleeding in the third and fourth decade of life. Adults with SCD have lower cognitive function than healthy adults, perhaps secondary to silent ischemia or chronic anemia. It is unclear whether such cognitive decline can be prevented.
Recently, the National Institutes of Health and the U.S. Department of Health and Human Services published evidence-based guidelines for the treatment of SCD. Strong recommendations from this report are listed in Table 17.
Transfusion management in SCD is complicated. Inappropriate transfusion can lead to alloimmunization, iron overload, and infectious complications. Antibody formation can lead to hyperhemolysis, a presumed immune response leading to hemolysis of nearly all transfused blood in addition to native blood. Simple transfusions with PRBCs are used to increase oxygen delivery and should not be used for uncomplicated pain crises or to treat chronic anemia. Patients undergoing surgical procedures requiring general anesthesia should receive PRBC transfusion for a target hemoglobin level of 10 g/dL (100 g/L) to avoid surgical complications. Monthly transfusions (hypertransfusion) in patients with an initial stroke can reduce the incidence of subsequent stroke and reduce stroke occurrence in patients at risk for stroke based on cranial Doppler arterial velocity. Exchange transfusion is often used in the treatment of acute stroke, complicated acute chest syndrome, and acute retinal artery occlusion. With any transfusion, care should be taken to keep hemoglobin concentrations less than 10 g/dL (100 g/L) to avoid problems with blood hyperviscosity.
More than 1000 mutations have been identified in the α- or β-globin gene, including hemoglobin C, hemoglobin D, and hemoglobin E. Most of these are only clinically relevant if coinherited with Hb S.
Immune-mediated hemolysis is characterized by antibody binding to erythrocytes causing complement-mediated and phagocyte-mediated destruction. Antibodies responsible for immune-mediated hemolysis are divided into warm antibodies, typically IgG, that react at normal body temperature and cold agglutinins, usually IgM, that bind at cooler temperatures. The laboratory hallmark of immune-mediated hemolysis is a positive direct antiglobulin (Coombs) test that detects either IgG or complement (C3) on the erythrocyte surface. Characteristics of immune-mediated hemolysis are shown in Table 18.
In warm autoimmune hemolytic anemia (WAIHA), pathogenic IgG antibodies are directed against Rh-type antigens on the erythrocyte surface. IgG-coated erythrocytes can be completely phagocytized by splenic macrophages via the Fc receptor and are cleared from the circulation. Warm IgG antibodies can bind and activate complement in a portion of these patients. Partial phagocytosis of the erythrocyte surface area results in spherocytes, which can be viewed on the peripheral blood smear. The direct antiglobulin test result is positive for IgG and is also sometimes positive for C3. Although WAIHA can be a primary disorder, it can also occur secondary to drugs (penicillins or α-methyldopa); lymphoproliferative disorders, such as chronic lymphocytic leukemia; or diseases with disordered immune regulation, such as systemic lupus erythematosus.
Treatment of symptomatic WAIHA involves alleviating immune destruction of erythrocytes using glucocorticoids, intravenous immune globulin, or rituximab. Splenectomy is also effective in nearly 70% of patients. Patients with hemolysis and life-threatening anemia may require blood transfusion (see Transfusion). Although donor blood will also be destroyed at an accelerated rate, improved hemoglobin levels may be sustained while waiting for therapy directed at eliminating the autoantibody to take effect.
In cold agglutinin disease, pathogenic IgM antibodies are directed against erythrocyte glycoprotein antigens (I or i antigen). Cold agglutinins bind at temperatures lower than normal body temperature (for example, the temperature of distal extremities during cold exposure). Cold agglutinins cause complement fixation, and C3-coated cells are both cleared by hepatic Kupffer cells and destroyed in circulation. The peripheral blood smear shows erythrocyte agglutination leading to a markedly elevated MCV on laboratory testing (Figure 14). In cold agglutinin disease, the direct antiglobulin test result is positive for C3. Cold agglutinin disease should always raise suspicion for an underlying lymphoproliferative abnormality, and it can also be seen in Mycoplasma and Epstein-Barr virus infections.
Treatment of cold agglutinin disease is primarily directed at avoidance of cold exposure, including warming of all infusates in hospitalized patients. Treatment with glucocorticoids, intravenous immune globulin, or splenectomy is seldom effective. Rituximab, fludarabine, or a combination of both has demonstrated activity in case series.
Microangiopathic hemolytic anemia is characterized by the presence of fragmented erythrocytes (schistocytes) on the peripheral blood smear (Figure 15) and is typically caused by erythrocyte destruction resulting from shearing as erythrocytes circulate through fibrin strands. Microangiopathy can be caused by thrombotic thrombocytopenic purpura or the hemolytic uremic syndrome (see Platelet Disorders); malignancy; disseminated intravascular coagulation; hypertensive crisis; drugs such as cyclosporine, mitomycin, or gemcitabine; or eclampsia. Microangiopathy is often accompanied by thrombocytopenia, kidney injury, and central nervous system disturbance. Treatment is directed toward the underlying disease process.
Patients with artificial valves or those with left ventricular assist devices can develop anemia resulting from erythrocyte fragmentation known as macroangiopathic hemolysis. Similarly, patients with giant hemangiomas, such as those with Kasabach-Merritt syndrome, can develop erythrocyte fragmentation from localized intravascular coagulation.
Patients with PNH lack proteins on the erythrocyte surface that are anchored by glycophosphatidylinositol due to acquired mutations in the PIGA gene that persist in bone marrow stem cells. Two of these proteins, CD55 and CD59, protect the erythrocyte from complement-mediated destruction. Patients lacking these proteins develop episodic hemolysis, marrow aplasia, and thrombosis. The cause of thrombosis is unclear. Patients with PNH are at higher risk for leukemia or myelodysplasia. Diagnosis involves demonstrating absence of glycophosphatidylinositol-linked proteins such as CD55 and CD59 on leukocytes by flow cytometry. PNH therapy includes folate supplementation, glucocorticoids, and a novel monoclonal antibody to C5, eculizumab, which inhibits activation of the terminal complement cascade, decreases hemolysis, reduces thrombotic complications, and improves quality of life. Eculizumab is associated with Neisseria infections, so patients should receive meningococcal vaccination before use.
March hemoglobinuria develops after physical erythrocyte destruction in the soles of the feet in response to long-distance running or marching. March hemoglobinuria has also been described in hand drummers and karate enthusiasts.
Arsenic or arsine gas exposure, elevated serum copper levels, the bite of the brown recluse spider, and severe burns are rare causes of hemolysis.
Malaria, babesiosis, clostridia, and Bartonella (Oroya fever) are all associated with hemolysis.