�Most of the references quoted are from the work of researchfellows over the past 30 yr. This topic was chosen in order to

acknowledge their outstanding contributions and to express mypersonal gratitude for the privilege of working with them.

Blood, Vol.60, No. 6 (December), 1982 1241

BLOOD The American

The Journal of

Society of Hematology

VOL. 60, NO. 6 DECEMBER 1982

REVIEW

Erythropoiesis, Erythropoietin, and Iron

By Clement A. Finch

The basic components of erythropoiesis are stem cellprecursors, their stimulation by erythropoietin, and anadequate supply of iron from which to make hemoglobin.

A NEMIA HAS LOST much of its mystique in the

past 40 yr as a better understanding of erythro-

poiesis and red cell destruction has made diagnosis and

management a more rational process. This brief dis-

cussion* will attempt to summarize certain physiologic

concepts useful to the clinician.

The importance of erythrocyte size as an indicator

of hemoglobin synthesis became evident with the intro-

duction of the hematocrit over 50 yr ago,1 and anemias

were classified as microcytic, normocytic, and macro-

cytic. Microcytosis could be equated with a decrease in

hemoglobin synthesis, attributable to either iron defi-

ciency, a mitochondrial abnormality affecting heme

synthesis, or impaired globin synthesis. Macrocytosis

is the result of a defect in nuclear maturation or of

“stimulated erythropoiesis.” The smear is essential in

differentiating these two causes of red cell enlarge-

ment. True macrocytes are the result of a disruption of

the usual mitotic sequence in the marrow (nuclear

maturation defect). Cells that skip a mitosis appear as

mature “red cell giants” in the circulating blood. A

second type of macrocytosis is caused by increased

erythropoietin stimulation in the presence of an ade-

quate iron supply. In this normal physiologic response,

hemoglobin synthesis within the immature cell is

increased,2 and reticulocytes are released prematurely,

appearing macrocytic, basophilic, and slightly hypo-

chromic on blood smear.3 Automated equipment has

increased the dependability of the mean corpuscular

volume, but the interpretation of macrocytosis still

rests with the smear differentiation of these two types

This article discusses the effects of each of these and thesimple tests by which they may be evaluated clinically.

of macrocytosis. Other calculations of red cell indices,

i.e., mean corpuscular hemoglobin and hemoglobin

concentration, add little.

There are limitations to this “cell size” approach.

Changes take a long time to develop due to the slow

turnover ofcirculating red cells, so that one can usually

detect maturation abnormalities only after the condi-

tion has been present for months. Also, the relationship

between iron supply and erythroid proliferation affects

the degree of macrocytosis (Fig. 1 ). However, the real

problem is that most anemias are normocytic and are

therefore not amenable to this type of analysis.

The heterogeneous group of normocytic anemias

can be best approached by measurements of produc-

tion and destruction.5 Production is most simply quan-

titated by counting circulating reticulocytes as long as

two corrections are made. The first is to convert the

percent of circulating reticulocytes into their absolute

number. The second adjusts for the effect of erythro-

poietin on the release of reticulocytes from the mar-

row.6 When plasma erythropoietin levels are increased,

the usual reticulocyte maturation period in the marrow

of about 3 days is shortened and the time in circulation

is correspondingly lengthened (Fig. 2). The reticulo-

From the Division of Hematology, Department of Medicine,

University of Washington, Seattle, Wash.

Supported in part by Research Grant HL 06242 from the

National Heart, Lung and Blood Institute. National Institutes of

Health, and Training Grant AM 07237from the National Institute

ofArthritis, Diabetes, and Digestive and Kidney Diseases, NIH.

Submitted June 14, 1982; accepted July 21, 1982.

Address reprint requests to Clement A. Finch, M.D., Division of

Hematology, Department of Medicine, University of Washington,

Seattle, Wash.

© I 982 by Grune & Stratton, Inc.

0006-4971/82/6006-0001$O1.OO/O

For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom

. 125MeanErythrocyticVolume (f I)

100

Fig. 1 . Effect of erythropoietin stimulation and iron supply onerythrocyte size. Curves represent proposed changes in red cellvolume at transferrin iron saturations from 1 0% to 1 00%. Normalmarrow function is assumed. Increased erythropoietin stimulationin the presence of increased iron supply increases red cell hemo-globin2 and red cell size.4 As iron supply decreases. so will

hemoglobin synthesis and red cell size. This relationship is also

affected by abnormalities in hemoglobin synthesis, such as in

thalassemia and by nuclear maturation abnormalities, which exag-

gerate the increase in hemoglobin synthesis by the iron-replete,

stimulated erythroid precursor.

010 20 30

Hematocrit,40%

Fig. 2. Effect of anemia on the reticulocyte maturation time inmarrow and in blood.

1242 CLEMENT A. FINCH

75

Hematocrit (%)

4,5 30 1�

100%

0 10 100 1000

Erythropoietin (units/mI)

cyte index derived from these two corrections relates

the rate of entry into circulation of the patient’s

reticulocytes to basal production. The expected pro-

duction capacity of the challenged erythroid marrow is

about 5 times basal, an increase that can be achieved

within a week under appropriate conditions.7’8 For

practical purposes, a reticulocyte index of greater than

3 times basal is taken to indicate an adequate marrow

response, while an index of less than 2 times basal is

assumed to represent impaired production.

Production failure may be due to either inadequate

erythroid proliferation or a maturation abnormality

resulting in cell death during development (ineffective

erythropoiesis).9 Although reticulocytes are low in

each, the smear differentiates the two conditions (Fig.

3). A hypoproliferative marrow is usually associated

with a fairly normal red cell population (a “quiet”

smear), a low bilirubin, and a normal lactate dehydro-

genase. Ineffective erythropoiesis will usually be

detected by changes in mean corpuscular volume (in-

creased with nuclear and decreased with cytoplasmic

abnormalities), but is also associated with red cell

fragmentation, a normal-to-high bilirubin, and an

increased plasma lactate dehydrogenase. The ery-

throid:granulocyte (EG) ratio of the aspirated marrow

or the density of erythroid forms in marrow section

gives direct evidence of the degree of erythroid prolif-

eration. Ferrokinetic measurements provide an even

more accurate quantitation of the erythroid precursor

number’#{176} and are particularly useful when leukocyte

abnormalities invalidate the EG ratio and when there

is difficulty obtaining a representative marrow aliquot.

This separation of anemias into hypoproliferative

states, ineffective erythropoiesis, and hemolytic cate-

gories can serve as a scaffold for further differential

diagnosis (Fig. 4),5 Such cataloging makes it clear that

the vast majority of anemic patients have a hypoproli-

ferative process in which erythropoietin and iron play

prominent causative roles.

ERVTHROPOIETIN

Erythropoietin output is designed to modulate

erythropoiesis so as to optimize blood hemoglobin

concentration. This regulation is illustrated by the

changes that follow altitude exposure, where the circu-

lating hemoglobin concentration increases I g for each

3%-4% decrease in arterial oxygen concentration.”

However, regulation of oxygen supply is more complex

than this. It has been observed that the hemoglobin

concentration in patients with hemoglobinopathies

correlated closely with the affinity of the abnormal

hemoglobin for oxygen.’2 Less complicated is the

increase in hemoglobin concentration observed in

patients with stable hemoglobin mutants having

increased affinity for oxygen.’3 These various observa-

tions indicate that the combination of 02 content and

50 its availability rather than oxygen content of arterial

blood alone is the basis for erythropoietin regulation.

The erythrocyte has an intrinsic mechanism for

modifying oxygen release, based on its content of

For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom

ERVTHROPOIESIS. ERYTHROPOIETIN, AND IRON 1243

Fig. 4. Relative frequency ofdifferent types of anemia.

Fig. 3. Erythrocyte size andshape changes. (A) A normal cell

population; (B) the shift macro-cytes of stimulated erythropoie-sis; (C) microcytosis and cell frag-mentation (D) the macrocytes ofmegaloblastic anemia.

intraerythrocytic 2,3-diphosphoglycerate 14 As

long as there is adequate phosphate substrate, the

amount of erythrocyte DPG will be altered not only by

changes in oxygen availability produced by an altered

blood pH but also according to the proportion of

oxygenated versus reduced hemoglobin in circulation.

The compensatory nature of this response to hypoxia is

shown by increases in DPG occurring in cardiac failure

and anemia, making increased amounts of oxygen

available at any given oxygen tension.’5”6 Thus, two

interlocking regulatory mechanisms involving the cry-

thron exist, designed to ensure an adequate oxygen

supply. DPG is the more rapidly responsive, acting

within hours to improve available oxygen by increasing

the amount release at any given oxygen tension. This

“available oxygen,” in turn, is monitored by the

erythropoietin-regulating apparatus, and further ad-

justments in erythropoietin and erythropoiesis over a

period of weeks are required to achieve an appropriate

concentration of circulating hemoglobin.

The response of the erythroid marrow to anemia is

initiated by the action of erythropoietin on committed

stem cells (CFU-E). While it would be often helpful to

know the level of stimulation in a patient, valid quanti-

7HYPOPROLI FERATIVE

1. Iron DefIcient Erythropolesis

a. relative (bleeding)b. iron deficiencyC. inflanvnation

2. Erythropoletin Lacka. renalb. nutritional S endocrine

C. �O2 release by Hb

3. Stem Cell Dysfunctiona. aplastic anemiab. rbc aplasia

INEFFECTIVE

1. Plegaloblastlc

a. Bl2b. folate

C. refractory

2. Microcytlca. thalassemlab. sideroblastic

3. Norliocytica. stromal dIseaseb. dins,rphic

HEMOLYTI C

1. Inununologlc

a. idiopathicb. secondary to

drug, tunor or

2. lnt9I�fl�a. metabolicb. Hb

3. ExtrInsica. mechanicalb. lytlc subst.

For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom

Tf Fe Receptor�‘ �‘ Tissues

1244 CLEMENT A. FINCH

tative measurements of erythropoietin have been lim-

ited to the research laboratory and have been virtually

unavailable to the clinician. Recently, radioimmune

assays have been developed that have biologic signifi-

cance and may become generally available in the near

future.’7 However, a simple screen is the evaluation of

the blood smear for the presence of “shift cells,” since

the premature release of these young reticulocytes is

erythropoietin-dependent.6”9 In using this, one must be

alert to stromal disorders that produce shift not

because of increased erythropoietin, but rather

because of disruption of marrow architecture.

Altered erythropoietin regulation occurs in patho-

logic states for different reasons. Production of

erythropoietin is decreased by a variety of disorders

impairing renal function,2#{176} but differences between

excretory and endocrine function of the kidney do

occur. For example, in the hemolytic uremic syndrome,

there is active stimulation of erythropoiesis despite

nitrogen retention. A decrease in erythropoietin is seen

with certain endocrine disorders and with protein-

calorie malnutrition.2’ In these situations, the erythro-

poietin-producing apparatus is intact, but output of the

hormone is decreased, presumably because less oxygen

is required for the altered metabolic state of the body.22

If erythropoietin is considered to be the regulator of

red cell proliferation, it might seem surprising that the

plasma erythropoietin concentration per se bears little

relationship to the rate of erythropoiesis in most

anemic patients. Near-normal levels are seen despite

accelerated erythropoiesis in hereditary spherocytosis.

From this we must assume that very little increase in

erythropoietin level above basal is required to sustain

the almost complete compensation in this hemolytic

anemia. On the other hand, high levels of erythropoie-

tin are usually found when production is depressed by

some other abnormality, the highest level being present

in aplastic anemia where stem cells are reduced or

cannot proliferate.20 The usual cause of a discrepancy

between marrow stimulation and response is an made-

quate supply of iron.

IRON

Iron is distributed to body tissues by plasma trans-

ferrin. Recent studies show that the transferrin iron

pool is not homogeneous but is composed of monoferric

and diferric transferrin, the latter having a far greater

capacity to dispense iron to tissue (Fig. 5)#{149}23Since iron

loading on transferrin is random, the amount of difer-

nc iron is a function of the transferrin iron saturation.

It was shown that this was the most meaningful

expression of available iron, and further, that basal

erythropoiesis could not be sustained when the propor-

tion of diferric transferrin iron fell below 16%.24 A

Fig. 5. Transferrin-mediated iron exchange. The transferrinmolecule receives iron, one iron molecule at a time, from iron-

donating tissues. Either monoferric or diferric transferrin canreact with membrane receptors with capture of all iron by the cell

and release of apotransferrin back into the plasma.

limitation of iron supply is characterized by an

increase in the extraction of transferrin iron by the

erythroid marrow from basal values of 5%-lO% to a

maximum of about 30% and a corresponding reduction

in the 80 ± 20 mm T’/2 radioiron clearance to about 20

mm.25

There are two outstanding abnormalities that char-

acterize iron-deficient erythropoiesis. The impairment

of hemoglobin synthesis, which eventually results in a

hypochromic microcytic red cell, reflects the obliga-

tory requirement of hemoglobin for iron. An impair-

ment in viability of developing and mature red cells in

iron deficiency has been suggested by iron kinetic

studies and morphological evidence of cell fragmenta-

tion, but it is likely that the amount of cell wastage has

been overstated. Of more relevance in explaining the

anemia is the effect of iron deficiency on cellular

proliferation. Transferrin iron has been shown to be

essential for proliferation in in vitro cell cultures.26’27

Fetal tissues cannot develop in utero when maternal

iron supply becomes severely restricted.23 Similarly,

the proliferation of the erythroid marrow is restricted

to near-basal levels when iron supply fails. The degree

of disparity between plasma iron supply and erythro-

poietin-imposed demand is reflected in the degree of

microcytosis and hypochromia of circulating red cells.

For this reason, little microcytosis is seen when the

patient with renal failure is iron deficient because

erythropoietin levels are also suppressed.

Considerations of iron-deficient erythropoiesis have

usually related the effect of a decreased iron supply to

basal tissue iron requirements. The term “relative iron

For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom

>‘

Co#{149}0

�- �0cDO>0

�0C .�

0�

‘D

E0ci’E

0 20 40 60 80 100

Transferrin saturation (%)

ERYTHROPOIESIS. ERYTHROPOIETIN. AND IRON 1245

deficiency” designates an iron supply that would be

adequate to meet the needs of basal erythropoiesis but

which is not adequate to meet the needs ofan expanded

erythroid marrow. This relationship can be accurately

monitored by measuring red cell protoporphyrin. An

increased red cell protoporphyrin is found in patients

with simple iron deficiency when transferrin saturation

drops below l6%.29 In hemolytic states, an increase in

protoporphyrin will be found at a normal transferrin

saturation of 35% or at even higher saturations because

of the increased marrow iron requirement.3#{176} Previous-

ly, reticulocytes were considered to be high in proto-

porphyrin, but it is now appreciated that any increase

observed represented a relative iron deficiency.3’ It is

not uncommon to find an inadequate iron supply (high

red cell protoporphyrin) in hemolytic anemia with

production rates greater than 5 times basal.

At these high levels of erythropoiesis, transferrin

procurement of iron becomes critical. Iron is made

available to transferrin through recycling of red cell

iron, through mobilization of iron stores, and through

absorption. Little is known of the regulation of plasma

iron supply except that the very act ofgiving up iron by

transferrin appears to enhance the procurement of iron

by transferrin.32 From existing sources, the normal

individual has difficulty providing sufficient iron to

support rates of erythropoiesis of greater than 3 times

basal.533 Clinicians have learned to interpret high

reticulocyte counts as indicating hemolysis. However,

the only reason for this association is the large amount

of iron made available in hemolytic anemia through

the catabolism of damaged red cells and the corre-

sponding increase in reticulocytes due to the effective

erythropoiesis characteristic of hemolytic states. Simi-

lar degrees of reticulocytosis can be seen when blood

loss occurs in an individual with parenchymal iron

overload34 or when a near-saturated transferrin is

maintained by large doses of oral and/or parenteral

iron.7’8

The critical nature of the plasma iron concentration

in patients with hemolytic anemia may be demon-

strated by ferrokinetic studies (Fig. 6). A decrease in

transferrin saturation from 60% to 30%, as might be

caused by inflammation, may halve the production of

red cells. This is almost certainly due to a marrow iron

flow insufficient to saturate erythroid membrane

receptors for the transferrin iron complex. In the

severely anemic patient, compensatory mechanisms

are at work to maximize iron supply. Thus, iron

absorption has been shown to be increased in thalas-

semia35 and there is an increased release of catabolized

red cell iron from, rather than storage in, the macro-

phage.36 This results in a high plasma iron concentra-

tion, leading in turn to augmented erythropoiesis.

Fig. 6. Effect of transferrin saturation on iron delivery to bodytissues. Plasma iron turnover in the normal subject is doubled astransferrin saturation increases from low to high levels. In con-trast, a sixfold increase was found in six subjects with hemolyticanemia (mean ± 1 SD plotted).

While this homeostatic mechanism may permit sur-

vival of the untransfused patient with thalassemia, it

results in parenchymal iron overload in later years.

THE ERYTHROID MARROW

The target of both erythropoietin and iron is the

committed stem cell population of the marrow. Experi-

ence with marrow transplant suggests that less than

5% of the total stem cell population is required to

establish basal erythropoiesis, so that the potential

proliferative capacity of the erythron is enormous. In

considering the limitations imposed on the response to

anemia, a major one would appear to be the bony

confinement of the medullary cavity. Some additional

space is created initially by the shift of marrow reticu-

locytes into the blood. Further increases can probably

only be achieved through the gradual extension of

active marrow into inactive medullary cavities. Such

an expanded marrow with its augmented blood supply

is seen in thalassemia, where erythroid proliferation

may reach 10- 1 5 times basal as measured by our

current ferrokinetic techniques.37 This particular dis-

order has the advantage of small red cell precursors

minimizing space problems and a decreased hemo-

globin synthesis in individual cells reducing iron re-

quirements. Nevertheless, there is evidence that so

hyperplastic a marrow may have internal difficulties

involving the disruption of sinusoidal walls with inter-

mittent release of marrow into the blood and emboliza-

tion to the lung.38 This level of marrow activity is also

For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom

1246 CLEMENT A. FINCH

poorly tolerated by the skeleton.3739 There is severe

osteopenia, and medullary cavities expand, at times

allowing marrow to rupture through the bony encase-

ment into the mediastinum. Thus, inescapable liabili-

ties arc associated with extreme marrow proliferation.

A host of other factors are involved in the intricate

process of red cell production and maintenance, includ-

ing structural features of the marrow bed, which

permit orderly maturation and transsinusoidal migra-

tion, other aspects of stem cell activation, the provision

of nutrients such as B,2, folic acid, and pyridoxine so

essential to red cell maturation, and various determi-

nants of red cell survival in circulation. However, iron,

erythropoietin, and committed erythroid stem cells

remain a central triad in the determination of red cell

production. An appreciation of the workings of these

through relatively simply laboratory methods provides

the clinician with a physiologic basis for understanding

most anemias and for anticipating the secondary prob-

lems associated with disturbances in erythropoiesis.

REFERENCES

I . Wintrobe MM: Erythrocyte in man. Medicine 9:195, 1930

2. Goldwasser E: Erythropoictin and the differentiation of redblood cells. Fed Proc 34:2285, 1975

3. Blood Cells 1:409-674 (Stress Erythropoiesis issue), 19764. Jacobs P. Finch CA: Iron for erythropoiesis. Blood 37:220,

1971S. Hillman RS, Finch CA: Red Cell Manual (ed 4). Philadelphia,

FA Davis, 1974

6. Labardini JT, Papyannopoulou T, Cook JD, et al: Marrow

radioiron kinetics. Haematology 7:301 , 1973

7. Hillman RS: Characteristics of marrow production and reticu-

Iocyte maturation in normal man in response to anemia. J Clin Invest

48:443, 1969

8. Hamstra RD. Block MH: Erythropoiesis in response to blood

loss in man. J AppI Physiol 27:503, 1969

9. Giblett ER, Coleman DH, Pirzio-Biroli G, Donohue DM,

Motulsky AG, Finch CA: Erythrokinetics: Quantitative measure-

ments of red cell production and destruction in normal subjects and

patients with anemia. Blood 11:291, 195610. Bothwell TH, Charlton RW, Cook JD, Finch CA: Iron

Metabolism in Man. Oxford, Blackwell, 1979I I . Finch CA: Oxygen transport, in Transactions of the Associa-

tion of American Physicians, 87:7 I , 1974I 2. Huehns ER, Bellingham AJ. Disease of function and stability

ofhaemoglobin. Bri Haematol 17:1, 1969

I 3. Adamson JW, Finch CA: Hemoglobin function, oxygen affin-ity. and erythropoietin. Annu Rev Physiol 37:351, 1975

14. Garby L, Meldon J: The Respiratory Functions of Blood.

London, Plenum Medical, 1977

I 5. Torrance J, Jacobs P. Restrepo A, Eschbach J, Lenfant C,

Finch CA: lntraerythrocytic adaptation to anemia. N EngI J Med

283:165, 1970

16. Woodson RD. Torrance JD, Shappell SD, Lenfant C: The

effect of cardiac disease on hemoglobin-oxygen binding. J Clin

Invest 49:1349, 1970

17. Cotes PM: Immunoreactive erythropoietin in serum. Br J

Haematol 50:427, 1982

18. Perrotta AL, Finch CA: The polychromatophilic erythrocyte.Am J Clin Pathol 57:471, 1972

19. Adamson JW, Eschbach J, Finch CA: The kidney and

erythropoiesis. Am J Med 44:725, 1968

20. Erslev AJ, Caro J, Miller 0, Silver R: Plasma erythropoietinin health and disease. Ann Clin Lab Sci 10:250, 1980

2 1 . Finch CA: Erythropoiesis in protein calorie malnutrition, in

Olson RE (ed): Protein-Calorie Malnutrition. New York, Academ-

ic, 19’75,p24’7

22. Caro J, Silver R, Erslev AJ, Miller OP. Birgegard G:

Erythropoietin production in fasted rats. Effects of thyroid hor-mones and glucose supplementation. J Lab Clin Med 98:860, 1981

23. Huebers H, Josephson B, Huebers E., Csiba E, Finch CA:Uptake and release of iron from human transferrin. Proc NatI Acad

Sci USA 78:2572, 1981

24. Bainton DF, Finch CA: The diagnosis of iron deficiency

anemia. Am J Med 37:62, 196425. Finch CA, Deubelbeiss K, Cook JD, Eschbach JW, Harker

LA, Funk DD, Marsaglia G, Hillman RS, Slichter 5, Adamson JW,

Ganzoni A. Giblett ER: Ferrokinetics in man. Medicine 49:17,

I 97026. Rich IN, Sawatzki G, Kubanek B: Specific enhancement of

mouse CFU-E by mouse transferrin. Br J Haematol 49:567, 1981

27. Trowbridge IS, Lopez F: Monoclonal antibody to transferrin

receptor blocks transferrin binding and inhibits human tumor cell

growth in vitro. Proc NatI Acad Sci 79:1175, 1982

28. Shepard TH, Mackler B, Finch CA: Reproductive studies inthe iron-deficient rat. Teratology 22:329, 1980

29. Cook JD, Finch CA, Smith NJ: Evaluation of the iron statusofa population. Blood 48:449, 1976

30. Pootrakul P: Unpublished data

31. Langer EE, Haining RG, Labbe RF, Jacobs P. Crosby EF,

Finch CA: Erythrocyte protoporphyrin. Blood 40:1 12, 1972

32. Finch CA, Huebers H, Eng M, Miller L: Effect of transfusedreticulocytes on iron exchange. Blood 59:364, 1982

33. Coleman DH, Stevens AR, Jr, Dodge I-IT, Finch CA: Rate ofblood regeneration after blood loss. Arch intern Med 92:341, 1953

34. Hillman RS, Giblett ER: Red cell membrane alteration

associated with marrow stress. J Clin Invest 44:1730, 1965

35. Cavill I, Ricketts C, Jacobs A, Letsky E: Erythropoiesis and

the effect of transfusion in homozygous B-thalassemia. N EngI J

Med 298:776, 1978

36. Fillet G: Le Fer dans l’Organisme. Metabolisme et Reutilisa-

tion. Paris, Masson, 1977

37. Pootrakul P. Hungsprenges S. Fucharoen S. Baylink D,Thompson E, English E, Lee M, Burnell J, Finch CA: Relationship

between erythropoiesis and bone metabolism in thalassemia. N EngI

J Med 304:1470, 1981

38. Sonakul D, Pacharee P. Laohapand 1, Fucharoen S. Wasi P:

Pulmonary artery obstruction in thalassaemia. SE Asian J Trop

Med PubI Health I 1:516, 1980

39. Gratwick GM, Bullough PG, Bohne WHO, Markenson AL,

Peterson CM: Thalassemic osteoarthropathy, Ann Intern Med

88:494, 1978

For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom

1982 60: 1241-1246  

CA Finch Erythropoiesis, erythropoietin, and iron 

http://www.bloodjournal.org/content/60/6/1241.citation.full.htmlUpdated information and services can be found at:

Articles on similar topics can be found in the following Blood collections

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

  Copyright 2011 by The American Society of Hematology; all rights reserved.Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of

For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom