Editors: Hillman, Robert S.; Ault, Kenneth A.; Rinder, Henry M.

Title: Hematology in Clinical Practice, 4th Edition Copyright 2005 McGraw-Hill

> Table of Contents > Part I - Red Blood Cell Disorders > Chapter 1 - Normal Erythropoiesis

Chapter 1 Normal Erythropoiesis

The oxygen required by tissues for aerobic metabolism is supplied by the circulating mass of mature erythrocytes (red blood cells). The circulating red blood cell population is continually renewed by the erythroid precursor cells in the marrow, under the control of both humoral and cellular growth factors. This cycle of normal erythropoiesis is a carefully regulated process. Oxygen sensors within the kidney detect minute changes in the amount of oxygen available to tissue and by releasing erythropoietin are able to adjust erythropoiesis to match tissue requirements. Thus, normal erythropoiesis is best described according to its major components, including red blood cell structure, function, and turnover; the capacity of the erythroid marrow to produce new red blood cells; and growth factor regulation.

STRUCTURE OF THE RED BLOOD CELL

The mature red blood cell is easily recognized because of its unique morphology. At rest, the red blood cell takes the shape of a biconcave disc with a mean diameter of 8 m, a thickness of 2 m, and a volume of 90 fL. It lacks a nucleus or mitochondria, and 33% of its contents is made up of a single protein, hemoglobin. Intracellular energy requirements are largely supplied by glucose metabolism, which is targeted at maintaining hemoglobin in a soluble, reduced state, providing appropriate amounts of 2,3-diphosphoglycerate (2,3-DPG), and generating adenosine triphosphate (ATP) to support membrane function. Without a nucleus or protein metabolic pathway, the cell has a limited lifespan of 100120 days. However, the unique structure of the adult red blood cell is perfect for its function, providing maximum flexibility as the cell travels through the microvasculature (Figure 1-1).

Membrane

A. INNER AND OUTER LAYERS

The shape, pliability, and resiliency of the red blood cell is largely determined by its membrane. The structure of this membrane is illustrated in Figure 1-2. It is a lipid sheath, just two molecules thick, that consists of closely packed phospholipid molecules. The external surface of the membrane is rich in phosphatidylcholine, sphingomyelin, and glycolipid, whereas the inner layer is largely phosphatidylserine, phosphatidylethanolamine, and phosphatidylinositol. This asymmetry is maintained by two transportersflipase, an ATP-dependent aminophospholipid translocase that rapidly transports phosphatidylserine and ethanolamine from the outer to the inner membrane, and flopase, which moves phospholipids more slowly in the opposite direction. The normal asymmetric distribution of membrane phospholipids can also be rapidly disrupted by a calcium-activated scramblase present in the membrane. Interference with these transporters results in a relocation of phosphatidylserine to the cell surface with a resulting increase in the thrombogenic potential of the cell surface. Moreover, accumulation of excess phosphatidylserine on the red cell surface plays a role in macrophage destruction. Approximately 50% of the red blood cell membrane is made up of cholesterol that is in equilibrium with the unesterified cholesterol in the plasma. Because of this, the cholesterol content of the membrane is influenced by plasma cholesterol levels, as well as by the activity of the enzyme lecithin cholesterol acyltransferase (LCAT) and bile acids. Patients with liver disease who have impaired LCAT activity accumulate excess cholesterol on the red blood cell membrane, which results in abnormal red blood cell morphology (targeting) and at times a shortened survival.

Figure 1-1. Red blood cell morphology. A: Adult red blood cells are characterized by their lack of a nucleus, and biconcave disc shape. Each red blood cell has a diameter of approximately 8 m and a width of 2 m. Red blood cells are extremely pliable as they pass through small vessels and sinusoids. B: The section of a small blood vessel demonstrates the ability of red blood cells to undergo major shape distortions. (Scanning EM photographs used with the permission of Dennis Knuckel, PhD, copyright 1993, American Society of Hematology National Slide Bank.)

P.2

B. RETICULAR PROTEIN NETWORK

The outer lipid membrane layer is affixed to a reticular protein network consisting of spectrin and actin. As shown in Figure 1-2, the integral proteins glycophorin C and Band 3, which function as anion exchangers, extend vertically from the spectrin lattice work through the lipid layer to make contact with the cell surface. Spectrin heterodimers interact horizontally with protein 4.1 and complementary spectrin heterodimers to form a hexagonal lattice framework under the lipid bilayer. Defects in the vertical structure of the membrane (deficiency of spectrin, ankyrin, or band 3, or loss of lipid) result in spherocyte formation. Damage to the horizontal spectrin framework results in severe red cell fragmentation or mild elliptocytosis. The integral proteins and surface glycosphingolipids are also responsible for the cell's antigenic structure. More than 300 red blood cell antigens have now been classified with the ABO and Rh blood group antigens being of primary importance in typing blood for transfusion (see Chapter 37). Autoantibodies against minor blood group antigens can result in increased red blood cell destruction by the reticuloendothelial cells.

Hemoglobin
The red blood cell is, basically, a container for hemoglobina 64,500 dalton protein made up of 4 polypeptide chains, each containing an active heme group. Each heme group is capable of binding to an oxygen

molecule. The respiratory motion of hemoglobin, that is, the uptake and release of oxygen to tissues, involves a specific change in molecular structure (Figure 1-3). As hemoglobin shuttles from its deoxyhemoglobin to its oxyhemoglobin form, carbon dioxide (CO 2 ) and 2,3-DPG are expelled from their position between the -globin chains, opening the molecule to receive oxygen. Furthermore, oxygen binding by one of the heme groups increases the affinity of the other groups to oxygen loading. This interaction is responsible for the sigmoid shape of the oxygen dissociation curve. Inherited defects in hemoglobin structure can interfere with this respiratory motion. Most defects are substitutions of a single amino acid in either the - or -globin chains. Some interfere with molecular movement, restricting the molecule to either a low- or high-affinity state, whereas others either change the valency P.3 of heme iron from ferrous to ferric or reduce the solubility of the hemoglobin molecule. Hemoglobin S (sickle cell disease) is an example of a single amino acid substitution that results in a profound effect on solubility.

Figure 1-2. Red blood cell membrane structure. The red blood cell membrane consists of a two-molecule-thick lipid sheath fixed to an intracellular protein network. The outer lipid layer is rich in phosphatidylcholine, sphingomyelin, and glycolipid; the inner layer is made up of the phosphatides of serine, ethanolamine, and inositol. Almost half of the lipid layer is cholesterol. The membrane proteins, glycophorin C and band 3, penetrate the lipid sheath and make vertical contact with the reticuloproteins, spectrin, protein 4.1, actin, and in the case of band 3, ankyrin. Spectrin heterodimers provide a horizontal framework by bridging protein 4.1 to complementary spectrin dimers.

The normal red blood cell contains approximately 32 pg of hemoglobin [mean cell hemoglobin (MCH) = 32 2 pg]. Normal hemoglobin synthesis requires an adequate supply of iron and normal production of both protoporphyrin and globin (Figure 1-4). Protoporphyrin synthesis is initiated in the mitochondria with the formation of delta aminolevulinic

acid from glycine and succinyl-CoA. Synthesis then moves to the cell cytoplasm for the formation of porphobilinogen, uroporphyrin, and coproporphyrin. The final assembly of the protoporphyrin ring is carried out by the mitochondria, after which iron is incorporated under the control of the cytoplasmic enzyme, ferrochelatase, to form heme. Globinchains are assembled by the cytoplasmic ribosomes under the control of two clusters of closely linked genes on chromosomes 11 and 16. The final globin molecule is a tetramer of two -globin and two non -globin chains. In the adult, 9697% of the hemoglobin is made up of two -globin and two -globin chains (hemoglobin A) with minor components of hemoglobin F and A 2 . The final assembly of the hemoglobin molecule occurs in the cell cytoplasm. Small amounts of iron, protoporphyrin, and free globin chains remain after hemoglobin synthesis is complete. The iron is stored as ferritin, whereas the excess porphyrin is complexed to zinc.

Figure 1-3. Hemoglobin-oxygen dissociation curve. Hemoglobin is

capable of a respiratory motion where oxygen loaded at the lung is unloaded at the tissue level. To accept oxygen, 2,3-DPG and carbon dioxide are expelled, salt bridges are ruptured, and each of the four heme groups opens to receive a molecule of oxygen. Oxygen release to tissues reverses the process; salt bridges are reestablished and both 2,3-DPG and carbon dioxide are accepted. The complex interaction of the four heme groups is responsible for the sigmoid shape of the hemoglobin-oxygen dissociation curve.

Figure 1-4. Hemoglobin synthesis. Normal hemoglobin synthesis requires an adequate supply of iron, amino acids, and pyridoxine (vitamin B 6 ). Porphyrin production is the responsibility of the mitochondria, whereas globin production is controlled by ribosomal RNA. The formation of the complete hemoglobin molecule involves the assembly of heme from protoporphyrin and iron and the union of a heme molecule with the two - and two -globin chains that comprise the globin component.

P.4 P.5 This complex series of reactions is triggered by erythropoietin stimulation of red cell progenitors. With precursor differentiation, there is a coordinated transcriptional induction of heme biosynthesis, globin synthesis and transferrin receptor expression, which is required for iron

transport (see Chapter 5). The rate of hemoglobin synthesis is determined by the availability of transferrin iron and level of intracellular heme. Hemoglobin synthesis is maximal in more mature marrow erythroblasts but persists to a lesser degree in the marrow reticulocytes. The cessation of heme synthesis is heralded by a decrease in membrane transferrin receptor expression, followed by a downregulation of heme and globin synthesis.

Cellular Metabolism
The stability of the red blood cell membrane and the solubility of intracellular hemoglobin depend on four glucose-supported metabolic pathways (Figure 1-5).

Figure 1-5. Red blood cell metabolic pathways. The red blood cell depends on four metabolic pathways to keep hemoglobin in solution and maintain membrane integrity. The Embden-Meyerhoff

pathway is responsible for the generation of high energy phosphate (ATP) for membrane maintenance, whereas the other pathways support hemoglobin function. The methemoglobin reductase (NADHdiaphorase) pathway is required to maintain hemoglobin in a reduced state. The phosphogluconate pathway helps counteract environmental oxidants and the Luebering-Rapaport pathway generates intracellular 2,3-DPG.

A. EMBDEN-MEYERHOFF PATHWAY
The Embden-Meyerhoff pathway (nonoxidative or anaerobic pathway) is responsible for the generation of the ATP necessary for membrane function and the maintenance of cell shape and pliability. Defects in anaerobic glycolysis are associated with increased cell rigidity and decreased survival, which produces a hemolytic anemia. The Embden-Meyerhoff pathway also plays a role in supporting the methemoglobin reductase, phosphogluconate, and Luebering-Rapaport pathways.

B. METHEMOGLOBIN REDUCTASE PATHWAY

The methemoglobin reductase pathway uses the pyridine nucleotideNADH generated from anaerobic glycolysis to maintain heme iron in its ferrous state. An inherited mutation of the methemoglobin reductase enzyme (also referred to as NADH-diaphorase or cytochrome P.6 b 5 reductase) results in an inability to counteract oxidation of hemoglobin to methemoglobin, the ferric form of hemoglobin that will not transport oxygen. Patients with Type I NADH-diaphorase deficiency accumulate small amounts of methemoglobin in circulating red cells, whereas Type II patients have severe cyanosis and mental retardation.

C. PHOSPHOGLUCONATE PATHWAY
In a similar fashion, the phosphogluconate pathway couples oxidative metabolism with NADP and glutathione reduction. It counteracts environmental oxidants and prevents globin denaturation. When patients lack either of the two key enzymes, glucose 6 phosphate dehydrogenase (G6PD) or glutathione reductase (GSH), denatured hemoglobin precipitates on the inner surface of the red blood cell membrane, resulting in membrane damage and hemolysis.

D. LUEBERING-RAPAPORT PATHWAY
Finally, the Luebering-Rapaport pathway is responsible for the production of 2,3-DPG. It is tied to the rate of anaerobic glycolysis and the action of the pH-sensitive enzyme phosphofructokinase. The 2,3DPG response is also influenced by the supply of phosphate to the cell. Severe phosphate depletion in patients with diabetic ketoacidosis or nutritional deficiency can result in a reduced 2,3-DPG production response.

REGULATION OF OXYGEN TRANSPORT

Red blood cells play a central role in oxygen transport. At the cellular level, oxygen supply is a function of red blood cells perfusing the tissue and their hemoglobin oxygen-carrying capacity. The unique physiology of the hemoglobin-oxygen dissociation curve allows an onsite adjustment of oxygen delivery to match tissue metabolism. At the same time, components such as pulmonary function, cardiac output, blood volume, blood viscosity, and adjustments of regional blood flow are also important contributors to oxygen transport.

Hemoglobin-Oxygen Dissociation Curve

Under normal conditions, arterial blood enters tissues with an oxygen tension of 95 mm Hg and a hemoglobin saturation of better than 97%. Pooled venous blood returning from tissues has an oxygen tension of 40 mm Hg and a saturation of 7580%. Thus, only the top portion of the hemoglobin-oxygen dissociation curve is used in the basal state (Figure 1-6). This provides a considerable excess capacity for increased oxygen delivery to support increased oxygen requirements. The sigmoid shape of the hemoglobin-oxygen dissociation curve also helps in this regard by releasing oxygen more easily as the tissue PO 2 falls below 40 mm Hg.

Figure 1-6. pH and hemoglobin-oxygen affinity. Oxygen delivery responds to tissue metabolism and blood pHthe Bohr effect. When acid products are released at the tissue level, the hemoglobin-oxygen dissociation curve of red blood cells in the vicinity immediately responds with a shift of the curve to the right. This shift has the effect of releasing more oxygen to tissues and opening hemoglobin to receive additional amounts of CO 2 . Alkalosis has the opposite effect. It shifts the hemoglobinoxygen dissociation curve to the left and effectively reduces the amount of oxygen released to tissue.

P.7 The affinity of hemoglobin for oxygen is also influenced by temperature, pH, CO 2 concentration, and by the level of red cell 2,3-DPG. As shown in Figure 1-6, the position of the hemoglobin-oxygen dissociation curve is affected by the rate of tissue metabolism, CO 2 production, and blood pH (the Bohr effect). When a tissue generates increasing amounts of CO 2 and acid metabolites, the resulting acidosis shifts the dissociation curve to the right. This shift permits the release of more oxygen for the level of tissue PO 2 . The reverse is also true. With an increase in pH such as with an acute respiratory alkalosis, the hemoglobin-oxygen dissociation curve shifts to the left, reducing the amount of oxygen available at any tissue PO 2 . The Bohr effect is instantaneous and can be highly localized to a specific site. For example, the blood perfusing an exercising muscle will be able to deliver 75% or more of its oxygen because of the low tissue PO 2 and the acidosis-induced Bohr effect. Oxygen unloading simultaneously lowers the CO 2 tension in the red cells (Haldane effect), thereby facilitating its diffusion from metabolizing tissues. This reciprocal interaction promotes optimal exchange of oxygen and carbon dioxide during exercise.

When the amount of oxygen removed by tissues continues at a high level (widened arterial-venous difference), the resulting increase in deoxyhemoglobin in the cell stimulates an increased production of 2,3DPG. This situation will be true regardless of whether the cause of the hemoglobin desaturation is hypoxia, cardiac failure, or anemia. The rise in intracellular 2,3-DPG sustains the shift of the dissociation curve to the right and provides significant compensation for a chronic anemia or hypoxia. 2,3-DPG metabolism also responds to systemic acidosis or alkalosis. The initial shift of the curve to the right in a patient with acidosis will be corrected over the next 1236 h by a compensatory reduction in the 2,3-DPG level. The Bohr effect is reversed by the lower 2,3-DPG and the curve shifts back to normal. Although this readjusts the level of oxygen delivery to match tissue requirements, it can create a problem if the acidosis is suddenly corrected. Because it takes a number of hours to replace the intracellular 2,3-DPG, a sudden return to a normal pH will shift the oxygen dissociation curve to the left owing to the lower than normal 2,3-DPG level.

Hemodynamic Factors
The self-regulating capacity of the oxygen dissociation curve takes care of most of the variation in tissue oxygen requirements in the basal state. With maximal exercise, the untrained subject will reach a limit determined not by oxygen loading but by a low maximal cardiac output resulting in poor oxygen delivery to tissues. In contrast, highly trained athletes have a greatly increased cardiac output, so that pulmonary loading and peripheral transport determine their limits. To maximize performance, they must tolerate both arterial hypoxemia and marked metabolic acidosis.

A. ANEMIA
The oxygen dissociation curve will also compensate for an anemia of moderate severity. However, once the hemoglobin falls below 910 g/dL, components such as changes in blood volume, cardiac output, and regional blood flow come into play. Both the pulse rate and stroke volume increase in patients with severe anemia and there is a redirection of blood flow to vital organs. These hemodynamic changes are often appreciated by patients. As their anemia worsens, they are increasingly aware of the force of ventricular contraction and often complain of pounding headaches, especially with physical exertion.

B. OXYGEN SUPPLY

Impairments in lung function also affect oxygen supply. Although the sigmoid shape of the hemoglobin-oxygen dissociation curve does counterbalance reductions in alveolar PO 2 , there is a limit to this compensation. Moreover, desaturation of hemoglobin results whenever unsaturated venous blood is shunted through areas of damaged lung tissue. The physiologic response to a decreased oxygen tension in ambient air, as for example the oxygen tension at moderately high altitudes (30004000 m), is an increase in 2,3-diphosphoglycerate to raise the P 50 , that is, shift the oxygen dissociation curve to the right. Moderate exercise will still further elevate the P 50 via the Bohr effect to maintain oxygen delivery to tissues. Under conditions of more marked hypoxia (altitudes > 4000 m), reflex hyperventilation results in a reduced PCO 2 and respiratory alkalosis. The latter shifts the oxygen dissociation curve to the left with a reduction in oxygen delivery to tissues. Still, high hemoglobin affinity for oxygen provides a physiological advantage for acclimatization to high altitudes. Subjects born with a high-affinity hemoglobin such as hemoglobin Andrew-Minneapolis (P 50 17 mm Hg) demonstrate normal arterial oxygen saturations at altitudes up to 4000 m, smaller increases in heart rate, and little or no increases in erythropoietin when compared with normal individuals. Animals that normally live at high altitudes also have high-affinity hemoglobins.

C. BLOOD VISCOSITY
Sustained hypoxia usually results in a compensatory rise in the red blood cell mass and hematocrit. Although this increases the oxygen-carrying capacity of blood, it also increases blood viscosity. The interaction of the hematocrit P.8 level and blood viscosity is discussed extensively in Chapter 12. Tissue oxygen delivery theoretically is maximal at a hematocrit of 3336% (hemoglobin of 1112 g/dL), assuming no changes in cardiac output or regional blood flow. Above this level, an increase in viscosity will tend to slow blood flow and decrease oxygen delivery. This effect is relatively minor until the hematocrit exceeds 50%, at which time blood flow to key organs such as the brain can be significantly reduced.

REGULATION OF ERYTHROPOIESIS Red Blood Cell Production

The rate of new red blood cell (RBC) production varies according to the rate of red blood cell destruction and tissue oxygen requirements. Changes in the oxygen delivery to tissue are sensed by peritubular

interstitial, fibroblast-like cells in the kidney. A decrease in the oxygen content of hemoglobin (pulmonary dysfunction), the hemoglobin level (anemia), or the hemoglobin affinity for oxygen (shift in the oxygen dissociation curve) will stimulate an increased production of erythropoietin by renal interstitial cells. This is accomplished by recruitment of new cells to initiate transcription of erythropoietin messenger ribonucleic acid (mRNA) by a single gene on chromosome 7. The mechanism of regulation involves the sensing of oxygen tension by a flavoheme protein that controls the level of hypoxia inducible factor (HIF-1). The latter interacts with response elements in nuclear DNA to activate erythropoietin gene expression. Erythropoietin then travels to the marrow, where it binds to a specific receptor (EPOR) on the surface of committed erythroid precursors. This receptor is a 508amino acid glycoprotein coded by a gene on chromosome 19. Within hours, there is a detectable increase in deoxyribonucleic acid (DNA) synthesis. This is followed by proliferation and maturation of committed stem cells to produce an increased number of new red blood cells. Erythroid progenitor apoptosis is also inhibited. The full marrow response takes several days. Given a sustained increase in erythropoietin stimulation, a rise in the reticulocyte index will not occur for 45 days and a detectable increase in hematocrit will take a week or more.

Figure 1-7. Erythropoietin production and anemia. Once the hemoglobin level falls below 12 g/dL, the plasma erythropoietin level increases logarithmically. Patients with renal disease or the anemia associated with chronic inflammation show a lower than predicted response for their degree of anemia.

A. MEASURING THE ERYTHROPOIETIN RESPONSE

The erythropoietin response to anemia can be directly measured by assaying the serum erythropoietin level (Figure 1-7). Once the hemoglobin level falls below 12 g/dL, there is a logarithmic increase in the serum erythropoietin level. At the same time, it is important to note that with mild anemia (a hemoglobin level greater than 12 g/dL), the erythropoietin level is not increased. This probably reflects the compensation of the 2,3-DPGinduced shift in the hemoglobin-oxygen dissociation curve combined with the sensitivity level of the renal sensor. Physiologically, it may reflect that a hemoglobin level of 12 g/dL is best for maximum tissue oxygen delivery.

B. OTHER FACTORS INFLUENCING ERYTHROPOIETIN LEVEL

Although the erythropoietin response is primarily a function of the severity of anemia or hypoxia, other factors, such as the erythroid marrow mass and levels of inflammatory cytokines, will influence the serum erythropoietin level. Erythropoietin binds avidly to erythroid P.9 progenitors and is removed from circulation. Therefore, with aplastic anemia, extremely high levels of serum erythropoietin reflect both an increased production and a decreased clearance. In contrast, with chronic hemolytic anemias, the expansion of marrow erythroid precursors results in a more rapid clearance of erythropoietin from circulation and, therefore, a lower serum level. Inflammatory cytokines, including interleukin-1, tissue necrosis factor (TNF-), and transforming growth factor , also play a role in regulating erythropoietin production and erythroid progenitor proliferation. They are responsible for the lower than normal erythropoietin response in patients with inflammatory disease states (see Chapter 4). Finally, direct suppression of the erythropoietin response is seen in patients receiving certain drugs (chemotherapeutic agents, cyclosporin A, and theophylline) or who are infected with human immunodeficiency virus (HIV).

Two other factors, angiotensin II and insulin-like growth factor-1 (IGF-1), may also play an erythropoietin-like role in certain settings. The erythropoietin-independent growth of erythroid progenitors in polycythemia vera may involve a hypersensitivity to IGF-1, whereas hypoxia has been shown to induce IGF-1 binding protein. Evidence for a role for angiotensin II is indirect. Post renal transplant erythrocytosis can be reversed by the administration of angiotensin converting enzyme inhibitors, without changing the serum erythropoietin level.

ERYTHROID MARROW PRODUCTION

Erythroid marrow production can be defined for the basal state and in terms of its capacity to increase in response to anemia. Patients who experience acute blood loss or sudden hemolysis of circulating red blood cells can show increases in new red blood cell production of 2 to 3 times normal; that is, the release of 4060 mL of new red blood cells per day. With a chronic hemolytic anemia, even higher production levels can be attained. This capacity to compensate for anemia is as much a normal characteristic of the erythroid marrow as its steady-state characteristics.

Measurement
The level of production can be assessed from several measurements of red blood cell production and destruction (Table 1-1). Clinically, the marrow E/G ratio and reticulocyte index are of the greatest value. The marrow E/G ratio (the ratio of erythroid to granulocytic precursors) is determined by inspecting a stained smear of aspirated marrow particles. As long as the granulocyte production of the marrow is normal, it is possible to estimate the proliferation of erythroid precursors. In the basal state, there will be approximately one erythroid precursor for every 34 granulocytic (myelocytic) precursors. With anemia and high levels of erythropoietin stimulation, the number of erythroid precursors increases dramatically to give ratios of 1:1 or greater. The morphology of the precursors is also important. Normal proliferation shows a balanced increase in erythroid precursors at all stages of maturation. If the number is skewed toward a younger population, especially a population with abnormal morphology, this suggests a defect in DNA synthesis or cytoplasmic maturation. These defects can result in a failure of cells to mature and early death in the marrow, so-called ineffective erythropoiesis. Effective red blood cell production is measured clinically by counting the number of reticulocytes (new red blood cells containing increased amounts of RNA) entering the circulation. Although both the E/G ratio and reticulocyte count are at best semiquantitative, they do provide sufficient information for clinical diagnosis. A measurement of radioiron

incorporation into red blood cells (erythron iron turnover) can provide a more accurate measurement of red blood cell production. This technique was used originally to define and classify red blood cell disorders as defects in either marrow proliferation (hypoproliferative anemias), precursor maturation (ineffective erythropoiesis), or red blood cell destruction (hemorrhagic and hemolytic anemias). The performance of the erythroid marrow can also be extrapolated from studies of red blood cell destruction. Clinical indicators of red blood cell destruction include the serum lactic dehydrogenase level, the indirect bilirubin, and observation of the rate of rise or fall of the hematocrit over time. Research measurements that are more accurate in defining levels of red blood cell destruction include carbon monoxide (CO) excretion, stool urobilinogen, and a direct measurement of radiolabeled red blood cell survival ( 51 Cr red blood cell P.10 survival). The latter has been used clinically to define both the rate and the site of destruction, whether in spleen or liver. The other measurements are not as practical.

Table 1-1. Measurements of red blood cell production and destruction.

Production Marrow E/G ratio Destruction Change in hematocrit

Reticulocyte index

Indirect bilirubin

Erythron iron turnover

Lactic dehydrogenase (LDH)

51 Crred

blood cell survival

CO excretion/stool urobilinogen

Basal and Stimulated Erythropoiesis

The ability of the erythroid marrow to increase red blood cell production

in response to anemia or hypoxia is a basic characteristic of the normal erythron. Thus, normal erythropoiesis is defined not only for the basal state but also for acute and chronic anemia (Table 1-2). A normal 70-kg adult has a circulating red blood cell mass of approximately 2000 mL (300 10 9 red blood cells per kg). Since red blood cells have a lifespan of 100120 days, 1% of the red blood cell mass, approximately 20 mL of red blood cells, is destroyed daily and replaced by new red blood cell production. This steady state is clinically appreciated from the E/G ratio of 1:3 and the reticulocyte index (the reticulocyte count corrected for hematocrit and reticulocyte shift; see Chapter 2). With an acute anemia secondary to hemorrhage or hemolysis, the marrow will respond with a threefold increase in cell production within 710 days. This can be detected from the increase in the E/G ratio to 1:1 or higher and a rise in the reticulocyte index to three times normal. With a chronic hemolytic anemia, red blood cell production can increase further, reaching levels of five to eight times normal. These patients show E/G ratios greater than 1:1 and reticulocyte indices greater than five times normal. The highest levels of red blood cell production in patients with hemolytic anemias require an expansion of the erythroid marrow mass to new areas of the marrow cavity. This process takes time and is most prominent in patients who have congenital, lifelong hemolytic anemias. Several factors play important roles in defining the marrow's response to anemia or hypoxia. Obviously, the severity of the anemia or hypoxia and the adequacy of the erythropoietin response are extremely important in setting a level of expectation. A chronic hypoproliferative anemia develops, for example, when a patient cannot produce increased amounts of erythropoietin because of renal damage.

Table 1-2. Normal response to anemia.

Anemia (Hb 8 g/dL) Basal (Hgb > 13 g/dL) MarrowE/G ratio 1:3

Acute 1:1

Chronic 1:1

Reticulocyte index

1.0

23

38

Factors that determine the marrow's responsiveness include its anatomical structure, the presence of a normal pool of stem cells, and the supply of essential nutrients. The anatomical structure of the marrow is organized to provide a nurturing environment for cell development. Erythroid precursor cells are maintained in a network of reticular cells and fibers in close proximity to vascular sinusoids. The marrow syncytium is designed to sustain the developing cells in a nutrient-rich environment while they proliferate and mature. Cells lining the sinusoids have the ability to regulate the exit of cells from the marrow into circulation, allowing only those cells that have completed maturation to leave. The importance of these marrow characteristics cannot be overemphasized. An abnormality in marrow structure, as seen with radiation damage or myelofibrosis, significantly impairs new red blood cell production. Overgrowth of other cellular components, as with myeloid leukemias or infiltrating tumors or lymphomas, will decrease red blood cell production by occupying the space required for red blood cell precursor growth. The supply of nutrients to the marrow is also important. The most important nutrient is the iron required for hemoglobin formation. The level of the normal marrow's response to a hemorrhagic or hemolytic anemia is essentially a reflection of iron supply (Figure 1-8). In response to a hemorrhagic anemia, a normal individual with normal iron stores will be able to maintain a serum iron level sufficient to support a production increase of up to 3 times normal. As shown in the figure, this level of production is attained as the hematocrit falls to levels between 20% and 30%. More severe anemia with a greater erythropoietin response does not result in a greater marrow production response. The cause of this plateau is the limitation of iron delivery from normal stores. Figure 1-8 also shows the effect of variations in iron supply. With iron deficiency, the erythroid marrow will be unable to respond despite a high level of erythropoietin stimulation. The patient with iron deficiency appears to have a hypoproliferative anemia even though the erythropoietin level is increased and the marrow morphology appears to be normal. In contrast, patients P.11 who have hemolytic anemias, in which the destruction of adult red cells provides a major source of iron for recycling to the marrow, can have

marrow production that increases to levels well above three times normal. Chronically, these patients can achieve production levels in excess of 5 times normal.

Figure 1-8. Red blood cell production and iron supply. The ability of the erythroid marrow to increase production is a direct reflection of iron supply. With worsening anemia, a normal individual with 5001000 mg of reticuloendothelial iron stores can increase red blood cell production twofold to threefold. An individual with iron deficiency will be unable to increase production above basal levels. In contrast, patients with chronic hemolytic anemias show production levels in excess of 35 times normal, with moderate to marked anemia.

BIBLIOGRAPHY

Beutler E et al: Hematology, 6th ed. McGraw-Hill, 2001.

Goodnough LT, Skikne B, Brugnara C: Erythropoietin, iron, and erythropoiesis. Blood 2000;96:823.

Hillman RS, Finch CA: Red Cell Manual, 7th ed. FA Davis, 1997.

Hsia CCW: Respiratory function of hemoglobin. N Engl J Med 1998;338:239.

Ponka P: Tissue-specific regulation of iron metabolism and heme synthesis: Distinct control mechanisms in erythroid cells. Blood 1997;89:1.

Zwaal RFA, Schroit AJ: Pathophysiologic implications of membrane phospholipid asymmetry in blood cells. Blood 1997; 89:1121.