Vol1PartIch1-5.

qxp 8/29/2005 3:50 PM Page 3

CHAPTER 1
Diagnosis of Anemia
Sherrie Perkins, MD, PhD

he bone marrow contains pluripotential stem cells that have the capacity for self-renewal

T as well as differentiation into mature cells, including red blood cells. Erythropoietin stim-
ulates primitive stem cells to undergo four rounds of cellular divisionover six to seven
daysand differentiation to form pronormoblasts and normoblasts. During the final step of
differentiation the nucleus is extruded from the red blood cell precursor to form a reticulo-
cyte, that still retains cytoplasmic RNA. Reticulocytes enter the peripheral circulation, appear-
ing as faintly blue-gray polychromatophilic cells in Wright stained blood smears that persist
for about 24 hours. Upon the loss of cytoplasmic RNA, a mature red blood cell is formed. A
red blood cell remains in the peripheral circulation for 120 days before being removed via
hemolysis by the spleen and reticuloendothelial system.
Usually red blood cell numbers remain relatively constant with removal from the
peripheral circulation balanced by proliferation and differentiation within the marrow.
However, many disease states will disrupt red blood cell homeostasis, leading to possible
alterations in red blood cell number, appearance or hemoglobin concentration. The most
common pathologic process associated with red blood cells is the decrease in overall red
blood cell numbers. Morphologic characteristics and the complete blood count (CBC) per-
formed by an automated hematology analyzer, will provide important information that will
direct further laboratory testing to allow diagnosis and characterization of a red blood cell
disorder.

1.1 Normal and Pathologic Red Blood Cells

The normal red blood cell is a biconcave disk, 6 to 9 m in diameter and 1.5 to 2.5 m thick.
In the peripheral smear, red blood cells are anucleate and contain predominantly hemoglobin
that is distributed to form a dense outer rim with a paler center that occupies approximately
one third of the diameter of the cell. The hemoglobin imparts a uniform pink to orange-red
color to the cytoplasm that is typically without inclusions. Normally all red blood cells are rel-
atively uniform in size and shape. Numerous disease states affect the size, shape and hemo-
globin content of red cells. Variation in size is referred to as anisocytosis, and variation in
shape is termed poikilocytosis. Pathologic red cells may be larger or smaller than normal,
may be abnormally shaped, or may contain inclusions. Knowledge of the disease states asso-
ciated with specific appearances of red blood cells will provide important diagnostic clues that
aid in the diagnosis of hematologic disorders. t1.1 presents a summary of morphologic red
blood cell abnormalities that may be seen in blood smears, including those associated with
specific disease states.
3
Vol1PartIch1-5.qxp 8/29/2005 3:50 PM Page 4

I Anemias

t1.1 Pathologic Red Blood Cells in Blood Smears

RBC Type Description Underlying Change Disease States
Acanthocyte (spur Irregularly speculated cells Altered cell membrane lipids Abetalipoproteinemia,
cell) with projections of varying parenchymal
length and dense center
Basophilic stippling Punctate basophilic Precipitated ribosomes (RNA) Coarse stippling; lead
inclusions intoxication, thalassemia;
fine stippling; many
anemias
Bite Cell Smooth semicircle taken Heinz body pitting by G6PD deficiency, drug-
(degmacyte) from one edge spleen induced oxidant hemolysis

Burr cell Cells with short, evenly May be associated with Usually artifactual; seen in
(echinocyte), or spaced spicules and altered membrane lipids uremia, bleeding ulcers,
crenated cell preserved central pallor gastric carcinoma, artifact
Howell-Jolly bodies Small, discrete basophilic Nuclear remnant Postsplenectomy, hemolytic
dense inclusions; usually anemia, megaloblastic
single anemia
Hypochromic cell Prominent central pallor Diminished hemoglobin Iron deficiency anemia,
synthesis thalassemia, sideroblastic
anemia
Macrocyte Cells larger than normal Reticulocytes, abnormal cell Increased erythropoiesis,
(>8.5 m), well-filled DNA maturation oval macrocytes in megalo-
hemoglobin blastic anemia, round
macrocytes in liver disease
Microcyte Cells smaller than normal Abnormal hemoglobin Iron deficiency anemia,
(<7 m) production thalassemia, sideroblastic
anemia
Ovalocyte Ellipticall-shaped cell Abnormal cytoskeletal Hereditary elliptocytosis
(elliptocyte) proteins

Pappenheimer Small, dense basophilic Iron-containing Sideroblastic anemia,

bodies granules mitochondrial remnant or postsplenectomy
siderosome
Polychromatophilia Gray or blue hue frequently Ribosomal material Reticulocytosis, premature
seen in reticulocytes marrow release of red
blood cells
Rouleaux Cell aggregates resembling Cell slumping by red cell Paraproteinemia, artifact
stack of coins interactions with
paraprotein
Schistocyte Distorted, fragmented cell, Mechanical destruction micro- Microangiopathic hemolytic
two or three pointed ends vasculature by fibrin strands anemia (DIC, TTP),
mechanical damage or prosthetic heart valves,
prosthetic heart valve severe burns
Sickle Cell Bipolar, speculated forms, Molecular aggregation of Sickle cell disorders excludes
(drepanocyte) sickle-shaped, pointed at hemoglobin S S trait
both ends
Spherocyte Spherical cell with dense Decreased membrane Hereditary spherocytosis,
hemoglobin and absent redundancy immunohemolytic anemia,
central pallor; usually transfusion, artifact
decreased in diameter
Stomatocyte Mouth- or cuplike deformity Membrane defect with Hereditary stomatocytosis,
abnormal cation immunohemolytic anemia
permeability
Target cell Target-like appearance Increased redundancy of cell Liver disease,
(codocyte) hypochromic with central membrane postsplenectomy,
hemoglobin thalassemia, hemoglobin C
disease, iron deficiency
Teardrop cell Distorted, drop-shaped cell Mechanical distortion of red Myelofibrosis, myelophthisic
(dacrocyte) cell anemia

4
Vol1PartIch1-5.qxp 8/29/2005 3:50 PM Page 5

Diagnosis of Anemia

1.2 Automated Hematology

In both office and hospital settings, most patients blood is evaluated with an automated
electronic blood cell counter. Most instruments analyze individually and combine data to
characterize the entire cell population. Analysis of cell characteristics is achieved by various
methods including voltage pulse-impedance analysis and low- or high-angle light scatter
from a coherent or laser light source. In an impedance counter, the passage of a particle
through an orifice of standard size and volume displaces conductive electrolyte solution
within the orifice. If an electric current is applied across the orifice, a change in resistance
and conductivity of the electrolyte solution occurs as the particle passes through it. A
detector notes a pulse when the particle passes through the orifice; this pulse is propor-
tional to the volume of the electrolyte solution displaced by the particle. Thus, the count-
er counts and sizes particles simultaneously. In a light scatter counter, interruption of the
light beam by a particle produces an electronic pulse. The angle of light scatter and the
intensity of the light scattered at a particular angle delineates several physical properties
of the cell, including cell size, volume, shape, and internal complexity. Many modern
hematology analyzers utilize a combination of these two approaches to provide accurate
data about red blood cells.
With the physical data collected, a hematology analyzer can generate a histogram of size dis-
tribution on the x-axis and relative number of particles on the y-axis. From these data, the red
cell number (RBC) and mean corpuscular volume (MCV) can be determined, and other indices,
such as mean corpuscular hemoglobin concentration (MCHC), can be calculated. Newer instru-
ments also generate an index that provides the degree of dispersion of red blood cell sizes (aniso-
cytosis) compared with a normal size distribution histogram, referred to as a red cell distribu-
tion width (RDW).

1.3 Evaluation of the Blood Smear

Examination of the blood smear by a physician who is aware of the patients clinical condi-
tion is extremely useful in evaluating the patient with anemia, as some red cell disorders may
have subtle changes that are easily overlooked, such as minimal hypersegmentation of neu-
trophils in patients with combined folate and iron deficiency (masked macrocytosis) or
basophilic stippling in a patient with thalassemia and other complicating causes of anemia.
Electronically derived red blood cell indices, although useful, are simply representations of the
mean and overall degree of dispersion of the cellular population and provide little information
about specific red blood cell shapes and the presence or absence of minor populations of
abnormal red cells. Examination of the blood smear for specific shape variations, such as those
listed in Table 1-1, can provide valuable information to aid in the diagnosis of a patients
underlying disease.

1.4 Anemia
The primary function of the red blood cell is to deliver oxygen to the tissues. Anemia is
defined as a reduction in the total number of red blood cells, amount of hemoglobin in
the circulation, or circulating red blood cell mass. This results in impaired oxygen delivery
to tissues, giving rise to physiologic consequences of tissue hypoxia as well as compensa-
tory mechanisms initiated by the organism to correct anoxia. Signs and symptoms of ane-
mia include fatigue, syncope, dyspnea, or impairment of organ function due to decreased
oxygen; pallor or postural hypotension due to decreased blood volume; and palpitations,
5
Vol1PartIch1-5.qxp 8/29/2005 3:50 PM Page 6

I Anemias

onset of heart murmurs, or congestive heart failure due to increased cardiac output.
Anemia is not a diagnosis, but a sign of underlying disease. Hence, the evaluation of a
patient with anemia is directed at elucidating the causes for the patients decreased red
blood cell mass. A thorough history and physical examination are crucial for an intelligent,
directed approach to the differential diagnosis of anemia. t1.2 and t1.3 show important
features in a patients history and physical examination that can yield diagnostic clues as
to the cause of anemia, and efficiently direct further laboratory testing.

t1.2 Patient History in the Diagnosis of Anemia

Historical Information Possible Causes of Anemia
Age of onset Inherited or acquired disorder, continuous or recent onset.
Duration of illness Results of previous examinations and blood counts
Prior therapy for anemia Vitamin B12, iron supplementation, and how long ago.
Suddenness or severity of anemia Symptoms of dyspnea, palpitations, dizziness, fatigue, postural
hypotension.
Chronic blood loss Menstrual and pregnancy history, gastrointestinal symptoms, black or
bloody stools.
Hemolytic episodes Episode of weakness with icterus and dark urine.
Toxic exposures Drugs, hobbies, and occupational exposures.
Dietary history Alcohol use, unusual diet, prolonged milk ingestion in infants.
Family history and racial background Possible inherited disorder: family members with anemia, gallbladder
disease, splenomegaly, splenectomy.
Underlying diseases Uremia, chronic liver disease, hypothyroidism

t1.3 Physical Signs in the Diagnosis of Anemia

Physical Sign Associated Disease
Skin and mucous membranes
Pallor Any anemia
Scleral icterus Hemolytic anemia
Smooth tongue Pernicious anemia, severe iron deficiency
Petechiae Thrombocytopenia and bone marrow replacement or aplastic anemia
Ulcers Sickle cell disease

Lymph nodes
Lymphadenopathy Infectious mononucleosis, lymphoma, leukemia

Heart
Cardiac dilatation, tachycardia, Severe anemia
loud murmur
Soft murmurs Anemia, usually mild

Abdomen
Splenomegaly, Infectious mononucleosis, leukemia, lymphoma, hypersplenism
Massive splenomegaly Chronic myelogenous leukemia, myelofibrosis
Hepatosplenomegaly with ascites Liver disease

Central nervous system

Subacute combined Pernicious anemia (vitamin B12 deficiency)
degeneration of spinal cord
Delayed Achilles tendon reflex Hypothyroidism

6
Vol1PartIch1-5.qxp 8/29/2005 3:50 PM Page 7

Diagnosis of Anemia

1.4.1 Examination of the Blood

Anemia has been classified by several different approaches, none of which is completely
satisfactory. For practical purposes, an initial morphologic classification of anemia with inte-
gration of red blood cell indices and morphologic characteristics is probably most useful. With
use of the MCV and the RDW or red cell morphologic index (RCMI), anemia may be classified
into six categories t1.4. The anemia may be characterized by cell size as microcytic, normo-
cytic, or macrocytic. The absence or presence of anisocytosis (as measured by RDW) further
subdivides these three size categories. In general, anemia caused by nutritional deficiencies
(such as iron, folate, or vitamin B12) tend to have a greater degree of anisocytosis than ane-
mia caused by genetic defects or primary bone marrow disorders. However, difficulties arise in
classification using this scheme, particularly with regard to anemia of chronic disease.

t1.4 Classification of Anemia Based on Red Blood Cell Size and Distribution Width
Cell Size Normal RDW High RDW
Microcytosis Thalassemia minor, anemia of chronic Iron deficiency, hemoglobin H disease, some
(MCV <70 m3 disease, some hemoglobinopathy traits anemia of chronic disease, some
[70 fL]) thalassemia minor, fragmentation
hemolysis
Normocytosis Anemia of chronic disease, hereditary Early or partially treated iron or vitamin
spherocytosis, some hemoglobinopathy deficiency, sickle cell disease
traits, acute bleeding
Macrocytosis Aplastic anemia, some myelodysplasias Vitamin B12 or folate deficiency,
(MCV >100 m3 autoimmune hemolytic anemia, cold
[100 fL]) agglutinin disease, some myelodysplasias,
liver disease, thyroid disease, alcohol

RDW = red cell distribution width; MCV = mean corpuscular volume

In addition to pure morphologic criteria, anemia also may be classified by the degree of bone
marrow response or peripheral blood reticulocytosis as hyperproliferative, normoproliferative, or
hypoproliferative. This often provides insights into the pathogenesis of the process. Thus, patients
with defects in red blood cell proliferation or maturation tend to have little or no increase in retic-
ulocytes, reflecting the inability of the bone marrow to increase red blood cell production in
response to the anemia (hypoproliferative anemia). In contrast, patients with anemia caused by
decreased survival of red blood cells with a normal bone marrow proliferative response often
exhibit increased peripheral blood reticulocytes (normoproliferative or hyperproliferative anemia)
(see f1.1). If the degree of reticulocytosis is adequate to replace the loss of red blood cells, the
anemia is termed compensated. If the bone marrow response is inadequate, the anemia will pro-
gressively worsen.
Finally, anemia caused by decreased red blood cell survival are often subdivided by patho-
genetic mechanism into those caused by intrinsic or inherited defects and those that are acquired
or caused by extrinsic factors. This classification is often useful in understanding the underlying
disease processes and may facilitate the evaluation and diagnosis of anemia that arises secondary
to extrinsic processes.

1.4.2 Differential Diagnosis of Anemia

Anemia may be either relative (due to increased plasma volume with a normal red blood cell
mass) or absolute (due to a decreased red blood cell mass). It is important to rule out causes of
relative anemia, such as pregnancy, excessive hydration or macroglobulinemia, as they represent
disturbances in plasma volume rather than a true decrease in red blood cell mass. Similarly,
7
Vol1PartIch1-5.qxp 8/29/2005 3:50 PM Page 8

I Anemias

f1.1 Classification of macrocytic anemia by reticulocyte count

Macrocytic anemia
(MCV >100m3)

Reticulocyte count
(MCV >100m3)

Normal or decreased Increased

Round macrocytes, no Hemolytic disorder, hemorrhage,

hypersegmentation on blood smear treated B12 or folate deficiency
(see f1.2 for additional tests)

Bone marrow
examination Megaloblastic changes

Nonmegaloblastic
Suspect treatable megaloblastic disorder
rule out myelodysplasia, alcohol, drugs,
serum B12/folate levels, red cell folate
toxins, liver disease, aplastic anemia
(see t1.5 for additional tests)
(see t1.5 for additional tests)

decreased plasma volume, caused by dehydration, may mask a real decrease in circulating red
blood cell mass.
Use of a morphologic classification scheme in combination with red blood cell indices and
the reticulocyte count allows for practical classification of anemia into broad groups. This will
facilitate selection of additional laboratory tests to determine the underlying cause of the anemia.

1.4.3 Macrocytic Anemia

Macrocytic anemia (MCV >100 m3 [>100 fL]) are less common than normocytic or micro-
cytic anemia. Macrocytic anemia may be subdivided into those with a normal RDW (principally
those caused by bone marrow failure, such as aplastic anemia and myelodysplasia), and those with
a high RDW (caused by either deficiencies of vitamin B12 or folic acid; by autoimmune hemolysis
or cold agglutinins). However, many exceptions to this general classification scheme exist. For
example, a mild degree of macrocytosis (MCV between 102 and 105 m3 [102 and 105 fL]) with a
normal RDW is relatively common as a direct toxic effect of alcohol. Similarly, some cases of
myelodysplasia may have a high RDW.
Further classification of a macrocytic anemia based on the presence or absence of a reticulo-
cyte response is also helpful (see f1.1). Hemolytic anemia, blood loss, and partially treated vita-
min B12 or folic acid deficiencies will demonstrate an increased reticulocyte count. Normal to
increased reticulocyte counts are more likely to be associated with autoimmune hemolysis, disor-
ders of membrane structural proteins (eg, elliptocytosis or spherocytosis), paroxysmal nocturnal
hemoglobinuria, and fragmentation hemolysis f1.2. For those patients with a normal or decreased
corrected reticulocyte count, disorders associated with decreased bone marrow functioninclud-
ing untreated vitamin deficiency, drugs, toxins, liver and thyroid disease, or primary bone marrow
8
Vol1PartIch1-5.qxp 8/29/2005 3:50 PM Page 9

Diagnosis of Anemia

f1.2 Classification of normocytic or megaloblastic anemias with elevated

reticulocyte counts
Search for blood loss
(suspect gastrointestinal or menstrual)

Evidence of occult blood loss No evidence of occult blood loss

Rule out ulcer disease,

Blood smear review,
colitis, occult carcinoma,
direct antibody test (Coombs)
hiatal hernia, parasites

Coombs positive for antibody on

Coombs negative
red blood cells, spherocytes

Nonautoimmune
Autoimmune hemolytic anemia
hemolytic anemia
(see t1.7)

failureshould be suspected. Blood smears that show morphologic features compatible with
megaloblastic anemia (oval macrocytes and hypersegmented neutrophils) may warrant further
evaluation with vitamin assays but not bone marrow examination. When megaloblastic changes
are present without signs of vitamin B12 or folate deficiency, bone marrow examination and addi-
tional testing t1.5 may be needed. A more extensive discussion of the diagnosis and evaluation
of macrocytic anemia is found in Chapter 5.

t1.5 Ancillary Tests for Macrocytic Anemia Without Increased Reticulocyte Response
Megaloblastic bone marrow changes present only in erythroid line:
Thyroid function tests
Assess iron storesserum iron, iron-binding capacity, ferritin
Cytogenetic analysisevaluate for myelodysplasias

Megaloblastic bone marrow changes present in more than one cell line:
Dietary and drug history
Malabsorption studies
Schilling test if vitamin B12 deficiency

1.4.4 Microcytic Anemia

The three most common causes of microcytic anemia (MCV <75 m3 [<75 fL]) are iron defi-
ciency, thalassemia minor, and anemia of chronic disease f1.3. The RDW is useful in distinguish-
ing thalassemia, which generally (but not invariably) produces elevated red blood cell counts and
a lower RDW than would be expected for the degree of anemia. Iron deficiency is almost always
associated with a high RDW. The values seen in anemia of chronic disease are extremely variable
with some being normocytic, while others are microcytic (particularly in patients with renal dis-
ease). By using additional laboratory testing to determine body iron stores, such as serum iron and
9
Vol1PartIch1-5.qxp 8/29/2005 3:51 PM Page 10

I Anemias

f1.3 Classification of microcytic anemia

Smear review
No diagnostic changes Abnormal RBC morphology

High (RDW >16)

RDW
Sickling, target cell
Suspect iron HbSS, HbS, thalassemia
Normal RDW deficiency

RBC count
Target cells, stippling
Thalassemia minor
Normal or high Low RBC number
RBC number Ferritin
level
Many target cells
HbE, HbC, obstructive
Suspect anemia of liver disease
chronic disease
Suspect early iron
deficiency,
RBC fragments
thalassemia,
Hemolysis
abnormal
hemoglobin Test for disease as
indicated: marrow
iron and serum Rouleaux
ferritin may be useful Increased globulins,
HbA2 level
decreased albumin

>4% <4%

Beta
Hemoglobin
thalassemia
analysis
minor
High
Low Nondiagnostic
(<22 ng/mL men,
<10 ng/mL women) Bone marrow
examination,
sideroblastic anemia, Iron binding
aplastic anemia, capacity testing
Iron deficiency marrow failure (see f1.4)

10
Vol1PartIch1-5.qxp 8/29/2005 3:51 PM Page 11

Diagnosis of Anemia

f1.4 Distinguishing iron deficiency anemia from anemia of chronic disease

Nondiagnostic ferritin

Iron/total iron binding capacity testing (TIBC) Low TIBC

% transferrin saturation
Anemia of
chronic
High (>15%) 9%-15% Low (<9%) disease likely

Serum soluble Iron deficiency anemia

transferrin receptor Low (<2.8
nmol)
Indeterminate
High
(>2.8 nmol) Bone marrow for iron stores
Absent Present

Anemia of chronic Anemia of chronic

Iron deficiency anemia
disease disease likely

total iron binding studies, iron deficiency anemia and anemia of chronic disease may usually be
distinguished without a bone marrow examination f1.4. A more detailed consideration of the eval-
uation and diagnosis of microcytic anemia is found in Chapter 2.

1.4.5 Normocytic, Normochromic Anemia

Patients with normal or hypoproliferative reticulocyte counts and normocytic, nor-
mochromic anemia generally require bone marrow evaluation. A peripheral blood smear may
provide valuable clues for the differential diagnosis t1.6. Patients with normocytic anemia and
an elevated reticulocyte count should undergo the same general evaluation as patients with
macrocytosis and an elevated reticulocyte count (see f1.2). Normocytic, normochromic anemia
with an elevated reticulocyte count can be divided into those with positive direct antiglobulin
test results ( DAT or Coombs test) and those lacking evidence of red blood cell-bound

t1.6 Normochromic, Normocytic Anemia Without High Reticulocyte Response

Findings on Peripheral Further Workup
Blood Smear
Leukoerythroblastosis Suspect myelophthisic processbone marrow examination for space-
occupying lesion (metastatic tumor, lymphoma, myelofibrosis) in
children suspect infection
Abnormal white blood cells Suspect leukemia, lymphomabone marrow examination
Rouleaux Suspect myelomaserum and urine electrophoresis, radiographs to look
for lytic lesions, bone marrow examination
No abnormal cells Suspect anemia of chronic disease or sideroblastic anemiabone
marrow examination; rule out chronic disease processes, ferritin, total
iron-binding capacity, and percent transferrin saturation as indicated

11
Vol1PartIch1-5.qxp 8/29/2005 3:51 PM Page 12

I Anemias

antibodies. Coombs-negative hemolytic anemia is a heterogeneous group of disorders. The

blood smear and patient history often suggest possible causes for the anemia t1.7. A more
detailed discussion of non-hemolytic normocytic anemia may be found in Chapter 3, and dis-
cussion of different causes of hemolytic anemia are presented in Chapters 6 through 15.

t1.7 Workup of Coombs-Negative Anemia

Feature of Anemia Possible Process Tests
Episodic anemia Enzyme deficiency G6PD, other RBC enzymes
Paroxysmal nocturnal Sucrose hemolysis, Hams test,
hemoglobinuria flow cytometry

RBC fragmentation DIC, TTP, HUS Coagulation tests, serum

haptoglobin levels

Abnormal RBC stippling Lead poisoning Lead levels

Thalassemia Hemoglobin electrophoresis

Abnormal shapes or Hemoglobinopathy, thalassemia Hemoglobin analysis

increased target cells

Spherocytes Hereditary spherocytosis Osmotic fragility test

G6PD = glucose-6-phosphate dehydrogenase; DIC = disseminated intravascular coagulation; TTP = thrombotic
thrombocytopenic purpura; HUS = hemolytic-uremic syndrome

The differential diagnosis of anemia is often tempered or modified by knowledge of other

patient data. All algorithmic classification schemes should be qualified by the pragmatic
knowledge of the physician in considering the probable causes of anemia in an individual
patient or patient population. For example, as 98% or more of anemia in children under the
age of 4 years is caused by iron deficiency, many pediatricians simply treat all children with
this type of anemia with iron supplementation and perform workups only for those who fail
to respond to this therapy. In many situations, clinical knowledge can suggest several possible
causes of anemia. Thus, the provided algorithms are suggested pathways for physicians to use
for determining test utilization and should not be considered required clinical work-ups.

1.5 Laboratory Tests

Test 1.5.1 Manual Reticulocyte Count

Purpose. This test enumerates the number of reticulocytes, indicating bone marrow production of
new red blood cells.
Principle. Residual RNA in immature red blood cells is precipitated and stained with a supravital dye.
Specimen. Venous or capillary blood may be used for this test.
Procedure. A blood smear is made, and the red blood cells are stained with brilliant cresyl blue or
methylene blue. Cells containing stained reticular material are enumerated per 1000 red blood
cells and expressed as percent reticulocytes (absolute number per 100 red cells). Many automat-
ed hematology analyzers now analyze reticulocyte counts based on staining and light-scatter
properties as an optional function of the complete blood count.

12
Vol1PartIch1-5.qxp 8/29/2005 3:51 PM Page 13

Diagnosis of Anemia

Interpretation. Reticulocytes are immature red blood cells that contain at least two dots of stainable
reticulin material in their cytoplasm. More immature forms have multiple dots and small net-
works of skeins of bluish-staining material. Intra-observer variation and uneven distribution of
reticulocytes introduce a high analytic variation in manual reticulocyte counting, with interlab-
oratory coefficients of variation often in the range of 20%. Duplicate reticulocyte counts or 3-
day average values may help to reduce the imprecision of the raw reticulocyte count. Automated
reticulocyte counts (see Test 1.2), owing to larger sample analysis and mechanically defined cri-
teria, tend to be more reproducible.
Effective red blood cell production is a dynamic process, and the number of reticulocytes should
be compared with the expected number to be released in a patient without anemia. This is calcu-
lated as 1% of 5 106/mm3 (5 1012/L) red cells daily for an absolute reticulocyte production of
50 103/mm3 (50 109/L). The corrected reticulocyte count takes into account normal red blood
cell proliferation for a specific hematocrit and may be calculated with the following formula:
Corrected Reticulocyte Count = (% Observed Reticulocytes Hematocrit) 45
Another complicating factor in reticulocyte count correction is that patients with anemia may
release reticulocytes prematurely into the circulation. Reticulocytes are usually present in the
blood for 24 hours before they extrude the residual RNA and become erythrocytes. If they are
released early from the bone marrow, immature reticulocytes may persist in peripheral blood for
2 or 3 days. This is most likely to occur when severe anemia causes a marked acceleration in ery-
thropoiesis and release. Some authors have advocated correction of the reticulocyte count for
immature reticulocytes (thought to be the best reflection of bone marrow response to anemia),
called the reticulocyte production index (RPI):
RPI = [(% Reticulocyte Hematocrit Value) 45] [1 Correction Factor]
The correction factor calculation is shown in t1.8.

t1.8 Correction Factor Calculation

Patients Hematocrit Value, % Correction Factor
40-45 1.0
35-39 1.5
25-34 2.0
15-24 2.5
<15 3.0

In cases of low erythropoietin (often seen in patients with renal or hepatic disease), applica-
tion of an RPI correction may mask a failure of bone marrow response, because the shift does not
take place fully or at all. In general, RPI values less than 2 indicate failure of bone marrow red
blood cell production or a hypoproliferative anemia. Reticulocyte production indexes of 3 or
greater indicate marrow hyperproliferation or appropriate response to anemia.

Test 1.5.2 Automated Reticulocyte Count

Purpose. Determination of reticulocyte numbers provides insight into the underlying pathophysiology
of an anemia. Use of automated staining and determination of reticulocytes in a hematology ana-
lyzer provides accurate reticulocyte enumeration by allowing evaluation of many more red blood
cells than can be studied with manual supravital staining and also can provide information as to
time since release from the bone marrow (reticulocyte maturity stage) by the staining patterns.

13
Vol1PartIch1-5.qxp 8/29/2005 3:51 PM Page 14

I Anemias

Principle. Reticulocytes are immature red blood cells in the final stage of differentaition that have
been recently released from the bone marrow and still retain intracellular protein and RNA. They
may be stained with RNA avid dyes that can be detected by fluorescence, light scatter properties
or absorbance characteristics. Specific, proprietary dyes vary between types of hematology ana-
lyzers, but show similar reticulocyte staining patterns. The maturity level of the reticulocytes can
be determined by the amount and intensity of staining, with the most immature reticulocyte
fraction having the highest staining (highest RNA levels). Usually reticulocytes are fractionated
into two to three different populations (immature, intermediate and mature), with the immature
reticulocyte fraction being the most accurate reflection of erythropoietic activity that provides
insight into the bone marrow proliferative response to an anemia.
Specimen. Anti-coagulated whole blood, usually in EDTA.
Procedure. Whole blood is stained with the RNA avid dye and analyzed in the hematology analyzer
using the reticulocyte enumeration program. Data is provided as a percentage of red blood cell
reticulocytes. The instrument will also fractionate the reticulocytes into an immature and mature
fraction, with some instruments providing an intermediate reticulocyte fraction based on levels
of staining.
Problems and Pitfalls. Automated reticulocyte counts may vary widely dependent on the methodology
and instrumentation used, and monitoring should be done using the same methodology over
time. Methods using fluorescence and aragon laser detection may be more sensitive at detecting
low numbers of reticulocytes. There is significant imprecison at very low reticulocyte numbers,
reflecting limitations of analytic sensitivity in the method. Samples with significant numbers of
reticulocytes are usually reproducible on the same instrument over time, both in determination
of total numbers of reticulocytes and the immature fraction.

Test 1.5.3 Bone Marrow Examination

Purpose. Bone marrow examination allows assessment of the cellularity, maturation, and composi-
tion of the hematopoietic elements in the bone marrow, as well as evaluation of iron stores. Some
infections also may be cultured from the bone marrow.
Principle. The cortical bone is penetrated and a sample of the bone marrow is aspirated. In most
cases, a small biopsy specimen of the medullary bone and marrow is obtained. The most common
sites for the procedure are the posterior or anterior iliac crest and the sternum.
Specimen. Bone marrow aspiration and biopsy samples.
Procedure. Bone marrow aspiration and biopsy are innocuous procedures when performed by
experts. Several sites in the skeleton have been used for bone marrow sampling. Because active
hematopoiesis occurs in the long bones of the arms and legs in infants under the age of 8
months, aspiration from the anterior aspect of the tibial tuberosity is useful. For adults, the pos-
terior iliac crest is the recommended site. Patients who are unable to lie on their stomachs may
be approached through the anterior iliac crest or sternum. The sternum is aspirated relatively eas-
ily, but its structure does not allow biopsy. In elderly patients, sternal bone marrow may be most
representative of the patients hematopoietic status and superior to that of the relatively acellu-
lar iliac crest. Sternal aspiration also may be most appropriate for patients who have lesions in
the sternum or ribs.
Notes and Precautions. Processing and interpretation have significant technical variables and require
experienced personnel. Bone marrow examination should be limited to situations in which non-
invasive procedures do not yield clear answers. t1.9 gives the most common indications for bone
marrow examination.

14
Vol1PartIch1-5.qxp 8/29/2005 3:51 PM Page 15

Diagnosis of Anemia

t1.9 Indications for Bone Marrow Examination in Anemia

Abnormalities in blood counts and/or peripheral blood smear
Unexplained cytopenias
Unexplained leukocytosis or abnormal white blood cells
Teardrop cells or leukoerythroblastosis
Rouleaux
No or low reticulocyte response to anemia

Evaluation of systemic disease

Unexplained splenomegaly, hepatomegaly, lymphadenopathy
Tumor staging: solid tumors, lymphomas
Monitoring of chemotherapy effect
Fever of unknown origin (with bone marrow cultures)
Evaluation of trabecular bone in metabolic disease (use undecalcified bone)

Interpretation. When both a Wright-stained aspirate preparation and a histologic core needle biop-
sy are available, optimal evaluation may be performed. The false-negative rate for metastatic car-
cinoma using aspiration alone is about 25%; for lymphomas it seems to be somewhat higher
30% to 40%depending on the cell type. Because of the small nature of the biopsy specimen,
sampling errors still may be a problem, causing false-negative results. Additional testing, such as
iron stains to evaluate iron stores, immunohistochemical staining, flow cytometric analysis, and
cytogenetic analysis, may be performed on aspirated bone marrow specimens to provide addi-
tional information about the disease process.

1.5 Treatment
Treatment of anemia is usually aimed at correcting the underlying abnormality. This may involve
identification of a source of blood loss, iron or vitamin supplementation, or discontinuation of a
drug that predisposes a patient to hemolysis. Acquired anemia associated with hematopoietic
abnormalities (such as myelodysplasia or aplastic anemia) or inherited anemia (such as red blood
cell membrane defects, enzymopathies or hemoglobinopathies) may require transfusions when
symptoms arise due to decreased oxygen delivery to the tissues. The benefit of transfusion ther-
apy must be balanced carefully against the risks of disease transmission and iron overload. Usually
transfusions are not required unless the hemoglobin concentration falls below 7 g/dL (70 g/L),
unless significant cardiac or pulmonary disease is present and hypoxia would be exacerbated by
even modest decreases in oxygen delivery. Long-term transfusion therapy may cause iron over-
load, leading to subsequent organ iron deposition and failure of function as patients are unable
to appropriately decrease gut mucosal iron absorption when iron loading occurs via transfusion.

1.6 References
Aslan D, Gumruk F, Gurgey A, Altay C. Importance of RDW value in differential diagnosis of hypochromic ane-
mias. Am J Hematol. 2002, 69:31-33.
Bain BJ. Bone marrow aspiration. J Clin Pathol. 2001, 54:657-663.
Brugnara C. Iron deficiency and erythropoiesis: new diagnostic approaches. Clin Chem. 2003, 49:1573-1578.
Brugnara C. Use of reticulocyte cellular indices in the diagnosis and treatment of hematological disorders. Int
J Clin Lab Res. 1998, 28:1-11.
Carmel R, Cassileth PA. A focused approach to anemia. Hosp Pract. 1999;34:7191

15
Vol1PartIch1-5.qxp 8/29/2005 3:51 PM Page 16

I Anemias

Cotelingam JD. Bone marrow biopsy: interpretive guidelines for the surgical pathologist. Adv Anat Pathol. 2003,
10:8-26.
Gehrs BC, Friedberg RC. Autoimmune hemolytic anemia. Am J Hematol. 2002, 69:258-271.
Glader, B. Anemia: general aspects. In:Greer JP, Foerster J, Lukens JN, et al, eds. Wintrobes Clinical Hematology.
11th ed. Baltimore, Md: Williams & Wilkins; 2004:947-978.
Irwin JJ, Kirchner JT. Anemia in children. Am Fam Physician. 2001, 64:1379-1386.
Lacombe F, Lacoste L, Vial JP et al. Automated reticulocyte counting and immature reticulocyte fraction meas-
urement. Comparison of ABX PENTRA 120 Retic, Sysmex R-2000, flow cytometry, and manual counts. Am
J Clin Pathol. 1999, 112:677-686.
Oh R, Brown DL. Vitamin B12 deficiency. Am Fam Physician 2003, 67:979-986.
Pierre RV. Reticulocytes. Their usefulness and measurement in peripheral blood. Clin Lab Med. 2002, 22:63-79.
Savage DG, Ogundipe A, Allen RH et al. Etiology and diagnostic evaluation of macrocytosis. Am J Med Sci.
2000, 319:343-352.
Spivak JL. The blood in systemic disorders. Lancet 2000, 355:1707-1712.
Tefferri A. Anemia in adults: a contemporary approach to diagnosis. 2003, Mayo Clin Proc 78:1274-1280.
Wians FH Jr, Urban JE, Keffer JH, Kroft SH. Discriminating between iron deficiency anemia and anemia of
chronic disease using traditional indices of iron status vs transferrin receptor concentration. Am J Clin
Pathol. 2001, 115:112-118.

16